Past informal seminars

The following is a listing of seminars/lectures that have been given in the past on topics of interest to the ECMWF scientific community.

2020

17 December
at 14:00

Virtual

Intrinsic Uncertainty in the Euro-Atlantic Response to the MJO heating in boreal winter: Results from ECMWF Re-Forecasts

Speaker: David Straus (George Mason University and COLA)

Energy exchange at the snow‐atmosphere interface in winter is important for the evolution of temperature at the surface and within the snow, preconditioning the snowpack for melt during spring. This study illustrates a set of diagnostic tools that are useful for evaluating the energy exchange at the Earth’s surface in an Earth System Model, from a process‐based perspective, using in‐situ observations. In particular, a new way to measure model improvement using the response of the surface temperature and other surface energy budget (SEB) terms to radiative forcing is presented. These process‐oriented diagnostics also provide a measure of the coupling strength between the incoming radiation and the various terms in the SEB, which can be used to ensure that improvements in predictions of user relevant properties, such as 2m temperature, are happening for the right reasons. Correctly capturing such process relationships is a necessary step towards achieving more skilful weather forecasts and climate projections.

These diagnostic techniques are applied to assess the impact of a new multi‐layer snow scheme in the European Centre for Medium‐Range Weather Forecasts’‐Integrated Forecast System at two high‐Arctic sites (Summit, Greenland and Sodankylä, Finland). A previous study showed that it will enhance 2m temperature forecast skill across the northern hemisphere in boreal winter compared to forecasts with the single layer model, reducing a warm bias. In this study we use the diagnostics to show that the bias is improved for the right reasons.

20 October
at 11:00

Virtual

Classification and dynamical precursors of extreme precipitation events over Northern Italy

Speaker: Federico Grazzini

Abstract

In this work, we investigate a vast number of extreme precipitation events (EPEs) occurring between 1979 and 2015 in northern-central Italy. Through the optimal blending of ERA5 dataset and high resolution gridded daily precipitation observational analysis (ARCIS), we classify, via a simple machine learning approach, EPEs into three categories (Cat1, Cat2, Cat3) according to thermo-dynamic conditions over the affected region. The three categories do not only differ locally but also in the evolution of precursor Rossby wave packet (RWP) associated with the trough responsible for the precipitation. The apparent seasonality of the flow cannot solely explain these differences; therefore, the relevant physical processes in the RWP propagation of each case are further investigated. In particular, we show that RWPs associated with the strongest EPEs, namely the ones falling in Cat2, undergo a substantial amplification over the western N. Atlantic due to anomalous ridge-building two days before the event; arguably due to diabatic heating sources. A greater understanding of flow dynamics and water vapour transport pathways leading to EPEs not only helps evaluate the predictability of regional precipitation extremes but could also clarify the response of local hydrogeological cycle to a warming climate.

30 September at 14:15

Using the power of data assimilation to model the COVID-19 pandemic

Speaker: Geir Evensen (NORCE and NERSC Bergen)

Abstract

The onset of the SARS-COV-2 Pandemic led to a complete lockdown of society in many countries. So here I was, an applied mathematician, in confinement, following the news coverage of extreme situations at hospitals in Italy and New York, and at the same time hearing confusing and contradictory statements from politicians and leaders in various countries. This situation motivated me to contribute with knowledge-based information to support decision-makers. So, in my home office, I started developing an epidemic model and gather data on the COVID-19 decease and related hospitalizations and deaths in Norway. I coupled the model to my data-assimilation library, and end of March, after about two weeks of work, I was able to model and predict the SARS-COV-2 pandemic evolution in Norway. I used the data-assimilation system to calibrate model parameters, including the time-dependent effective-reproductive-number, R. I then provided reports with model predictions to the Norwegian authorities to explain the unstable situation and the pandemic's dependency on R.

On April 10th, I invited an international group of colleagues working in data assimilation to use the model system on data from their respective countries. Nearly everyone jumped at the opportunity, and from then, we have worked together as an international team. We modeled the SARS-COV-2 pandemic in the four European countries Norway, England, The Netherlands, and France; the province of Quebec in Canada; the South American countries Argentina and Brazil; and the four US states Alabama, North Carolina, California, and New York. These countries and states all have vastly different developments of the epidemic, and we could accurately model the SARS-CoV-2 outbreak in all of them.

The joint work led to significant learnings regarding the use of data assimilation in epidemic modeling. We compiled the results into a large manuscript and submitted it to the journal "Fundamentals of Data Science," where we are currently finalizing the revisions. We also made the paper available from MedRxiv

(https://www.medrxiv.org/content/10.1101/2020.06.11.20128777v1).

In this seminar, I will present results from this paper. I will introduce and justify the choice of model used, i.e., an SEIR model with age-classes and compartments of sick, hospitalized, and dead. I will introduce and demonstrate the ESMDA data-assimilation technique used, and I will discuss some of the benefits of using this system.

This work demonstrates how it is possible to use iterative ensemble smoothers to estimate an SEIR model's parameters. The data assimilated are the daily numbers of accumulated deaths and the number of hospitalized. Also, it is possible to condition on the number of cases obtained from testing.

We start from a wide prior distribution representing the model parameters; then, the ensemble conditioning leads to a posterior ensemble of estimated parameters leading to model predictions in close agreement with the observations. The updated ensemble of model simulations has predictive capabilities and include uncertainty estimates.

We estimate the effective-reproductive-number as a function of time, and we can assess the impact of different intervention measures. By starting from the updated set of model parameters, we can make accurate short-term predictions of the epidemic development given knowledge of the future effective-reproductive-number. Also, the model system allows for the computation of long-term scenarios of the epidemic under different assumptions.

The model system is freely available from Github (https://github.com/geirev/EnKF_seir).

We realize that more complex models, e.g., with regional compartments, may be desirable, and we suggest that the approach used here should be applicable also for these models. I have recently upgraded the model system to include multiple regions or countries with specified interactions between them. This new model formulation allows for simulating the pandemic development for, e.g., the countries in Europe or the states in the US.

29 September
at 14:00

VIRTUAL

Cubesats for monitoring atmospheric processes (CubeMAP): a constellation mission to study the middle atmosphere

Speaker: Damien Weidmann (Rutherford Appleton Laboratory, UK)

Abstract

Earth is changing at an unprecedented pace. Understanding and quantifying the processes driving the change and particularly the role played by the atmosphere is necessary, and the CubeMAP mission has been designed to address this need. The CubeMAP mission has been selected as a finalist for the new ESA SCOUT mission programme. It consists of a constellation of three identical CubeSats, each equipped with three high resolution spectrometers and one hyperspectral solar disk imager. The overall mission goal is to study, understand, and quantify processes in the tropical Upper Troposphere and Stratosphere (UTS), study its variability, and contribute to trends analysis in its composition and its effects on climate and vice-versa. The UTS composition plays a significant role in controlling the Earth’s climate, with still poorly explored feedbacks within the Earth System. This region of the atmosphere is coupled to the surface and the free troposphere both dynamically and radiatively. Its composition is determined by anthropogenic emissions of greenhouse gases and pollution precursors and is subject to changes via radiative, dynamical, and chemical processes. The ultra-miniature, while high spectral resolution, spectrometers selected for the payload are integrated thermal infrared laser heterodyne spectro-radiometer probing narrow spectral windows in limb solar occultation. For aerosol and oxygen measurements, an innovative, co-registered, solar disk imager using a multispectral CMOS sensor ensures both pointing knowledge accuracy and a large spectral channel diversity in the visible and near infrared, together with a small form factor. The platform selected for the mission is based on the GomSpace 12U CubeSat form factor which can accommodate the payloads while also keeping volume, mass and cost at a manageable level. The system design is to a high degree based on proven GomSpace COTS products with demonstrable flight heritage, hence lowering the mission risk. The mission contributes to developing a novel observation model: it offers a high level of deployment flexibility and system modularity, as well as economy of scale, through the use of identical payloads and platforms functionalized to specific Earth observation goals. The constellation approach also obviates the traditional limited coverage of limb solar occultation mission, whilst benefiting from the high accuracy this sounding mode brings. CubeMAP is also highly complementary to the nadir sounding infrastructure and contributes to enhancing its outputs and exploitation. It also addresses the forthcoming gap in solar occultation sounding capabilities.

24 September
at 10:00

VIRTUAL

Joint ECMWF/UoR seminar on GloFAS-ERA5 river discharge trends

Speaker: Ervin Zsoter

Abstract

The main objective of this study is to analyse the GloFAS-ERA5 river discharge reanalysis for any noticeable change (including gradual trends or discontinuities) in the annual mean time series across the 1979-2018 (40-year) period, and to evaluate how realistic these are compared with available observed river discharge time series.

These variabilities are quantified by linear regression in order to highlight any concerning features in the GloFAS-ERA5 time series.

This work is particularly important for GloFAS, as large trends, discontinuities or other similar features could have a major consequence on the GloFAS flood thresholds in around 50% of catchments, which are based on GloFAS-ERA5, and thus subsequently on the issuing of flood warnings.

In addition, this study also contributes to the understanding of the water cycle variable behaviour in ERA5 (driver of GloFAS-ERA5) and ERA5-Land (higher resolution land reanalysis forced by ERA5, produced offline) by exploring the linear trends in river discharge and related hydrological variables. In exploring the stability of the time series in ERA5, we seek to trigger potential further discussions and research studies, which subsequently should help with the planning and development for the next generation ECMWF reanalysis, ERA6.

18 June
at 10:30

VIRTUAL

“ecPoint” - a Post-processing Tool that improves Forecasts and highlights Systematic Model Errors

Speaker: Timothy Hewson

Abstract

In April 2019 ECMWF introduced a new, experimental, “point rainfall” forecast product onto its ecCharts web display platform, based on the post-processing package “ecPoint”, to give site-specific forecasts for everywhere in the world up to day 10. Prior to this development forecasters had access to just raw ensemble output in ecCharts, which provides gridbox average totals. ecPoint aims to incorporate probabilistically the expected sub-grid variability, and simultaneously apply gridscale bias corrections. Both these adjustments depend critically on “gridbox-weather-type”.

This presentation will describe the meteorology-based calibration rationale that underpins ecPoint, how this is different to pre-existing post-processing methods, and how it can also be applied to other surface variables such as 2m temperature. Numerous benefits will be highlighted.

The conditional verification concepts underpinning the calibration allow one to identify weather-situation-dependent gridscale biases. Examples will illustrate the diagnostic power of this approach, showing where and when rainfall is typically under- and over-forecast, providing pointers for future model improvements. And using an open source GUI one can apply the calibration code to data from other models, and thereby intercompare performance in different weather situations.

The forecast improvements that then arise will be discussed, using both long term global verification up to day 10, and illustrative case studies, with a focus on how extreme localised rainfall, that might lead to flash floods, is better handled. It will be shown how the post-processing can usefully shift the emphasis for warning issue from one region to another, when one compares with raw ensemble output.

There will be brief reference, from collaborative work, to how ecPoint output seems to compare favorably with the post-processed output of convection-resolving limited area ensembles.

The talk will conclude by discussing, in the context of ongoing and potential projects, numerous future applications of ecPoint, such as bias-corrected inputs to hydrological models, point rainfall re-analyses and tests of theories such as city impact on rainfall. Avenues for improving the methodology will also be highlighted.

28 April
at 10:00

VIRTUAL

Can approximations of the CRPS help us better understand why the skill changes?

Martin Leutbecher and Thomas Haiden

Abstract

Ensemble verification of scalar variables often involves the use of the continuous ranked probability score (CRPS). The talk looks at a Gaussian approximation of the CRPS with the aim to improve our understanding of score changes. The approximation permits to express the CRPS of an ensemble as a function of the variance of the error of the ensemble mean, the mean error of the ensemble mean and the ensemble variance. The methodology will be applied to direct model output from medium-range ensemble forecasts as well as postprocessed ensemble forecasts. We will examine how well this approximation works and whether it might be suitable to complement information we see for instance in scorecards.

14 February
at 10:30

Room: LT

Aircraft Weather Observations and their Use (current and future)

Speaker: Steve Stringer (Met Office and EUMETNET) and Siebren de Haan/Jan Sondij (KNMI and EMADDC)

Abstract

Aircraft based weather observations (ABO) have increasingly become a very valuable input into the global weather forecasting process, with some independent assessments showing their impact on NWP to be second only to that of satellite data. An international workshop on Aircraft Based-Observations (ABO) is to be hosted at ECMWF, 12th/13th February 2020, bringing together ABO data users and ABO data providers to share their experience of aircraft data use and to recommend any necessary changes to procedures in the provision or use of aircraft data in the future. The seminar will provide an overview of the current status ABO data provision, plans for the future, together with outcomes and recommendations from the ABO Workshop. An overview of European plans to implement an operational centre and service for the delivery of quality controlled Mode-S derived aircraft observations will be also included. (Mode-S reports are derived from air traffic management messages and provide very high resolution data, particularly winds, over parts of Europe and potentially elsewhere. They are not yet processed at ECMWF.)

11 February
at 14:00

Room: LT

The relationship between the circumglobal teleconnection, the Indian monsoon and European summer weather

Speaker: Jonathan Beverley (University of Reading, University of Exeter)

Abstract

Recent research has led to improvements in European winter seasonal forecasts, however summer forecast skill remains relatively low. One potential source of predictability for Europe is the Indian summer monsoon, which can influence the weather across many parts of the northern hemisphere via a global wave train known as the “circumglobal teleconnection” (CGT). Here I assess the ability of the ECMWF coupled seasonal forecast model to represent this teleconnection mechanism. To understand how errors in its representation are related to errors in summer forecast skill over Europe, results from relaxation and thermal forcing experiments, as well as barotropic model experiments, will be presented. These experiments were designed to identify possible causes of errors in the teleconnection pathway, and to explore the impact of improving the representation of the CGT on European summer forecast skill.

23 January
at 14:00

Room: LT

Can GNSS Polarimetric Radio Occultations (GNSS PRO) contribute to better understanding, monitoring or predicting extreme events?

Hydrological forecasting is now increasingly conceived as a global issue that transcends catchment boundaries and should be studied from a larger perspective. In addition, while the number of gauging stations is decreasing, potential new sources of meteorological and hydrological data are appearing, including the contribution of citizen scientists. It is likely that in the coming years, many other regions of the world will face a situation of co-existing local and global hydrological forecasting systems. How could we make the most out of both the local and global information? One idea would be to use the global ensemble forecasts as an initial field, and then to use the local forecast as additional information to refine the forecasts locally. Another element of interest in the context of global hydrology is the increasing availability of data from citizen science programs. This is a potentially rich source of information that could complement more « traditional » data. Can this data be assimilated in hydrological models? Is there any gain to assimilate data from citizen scientists in addition to traditional data? I would like to discuss these questions and many others, especially in the context of using machine learning tools to assimilate observations into hydrological models. Hopefully, this seminar will lead to fruitful discussions in preparation for my sabbatical (Sept 2020-August 2021), which I intend to spend mostly at the ECMWF.

12 November
at 10:30

Lecture Theatre

The Bureau of Meteorology's use of NWP for the prediction of lightning and severe convective hazards

Speaker: Harald Richter (BOM)

Abstract

Over the last three years the Bureau's operational modelling suites have begun to incorporate a global design ensemble and convection-allowing models (CAMs). These systems, largely based on the UKMO's Unified Model, are now capable of meaningful predictions that relate to a range of convective hazards such large hail, damaging winds and lightning itself.

I will first introduce the current and planned modelling systems at the Bureau of Meteorology, before moving into aspects of the post-processing approaches taken and planned. While the UKMO's IMPROVER framework is intended to provide the bulk of the post-processing capability, the prediction of convective hazards is achieved through a range of different individual algorithms such as Calibrated Thunder for the prediction of cloud-to-ground lightning or storm attributes such as updraft helicity for the prediction of severe convective storms. The focus of my presentation will be on these convectively focused post-processing approaches.

Harald's scientific interests
------------------------------------

Harald is a senior scientist in the Science to Services program of the Australian Bureau of Meteorology. His main focus is the diagnosis and prediction of deep extratropical convection and the associated hazards such as large hail, damaging winds and tornadoes. Over the past seven years his particular interest has been the use of large-scale and convection allowing models to predict thunderstorms and their hazards.

In a previous life, Harald was responsible for training the Bureau's forecasters in the diagnosis, nowcasting and forecasting of severe thunderstorms based on observational systems and NWP which has set up a strongly operational perspective for his research to follow.

Harald also leads a cross-agency project on physical impact prediction which quantitatively combines exposure and vulnerability information with NWP-based wind predictions. Occasionally he has been called upon to act in more managerial roles, which allowed him to adopt a slightly wider view of the Bureau's research and development activities.

6 November
at 10:30

Clouds, Weather and Climate: From the Southern Ocean to Climate Strikes

Speaker: Dr Andrew Gettelman ((NCAR and visiting scientist at the University of Oxford and ECMWF)

Abstract

Clouds are critical for forecasting weather and climate. This presentation will provide an overview of how critical cloud processes affect weather and climate prediction, and how we simulate them in models. Critical cloud processes often not considered include supercooled liquid clouds, shallow clouds over the ocean, and cloud interactions with aerosols in the atmosphere. Interestingly, supercooled liquid clouds are not only important for extreme weather, they are also important for climate. Recent observations and modeling work, taking us literally to the ends of the Earth, is leading to a better understanding of clouds, aerosols and climate, with possible improvements in weather prediction and some surprising and not fully understood implications for future climate change.

30 October
at 14:00-15:00

Room: LCR

Connecting the dots in hydrological ensemble prediction

Speaker: Maria-Helena Ramos (Irstea, France)

Abstract

Connecting people, disciplines and efficient techniques is crucial to develop hydrological ensemble prediction systems and foster partnerships. Probabilistic and scenario-based forecasts are recognized as essential to quantify uncertainties and support decision making in flood forecasting, drought management and water resources planning. Many successful cases have already been implemented, but still more must be done to bridge knowledge gaps, facilitate operational implementations and raise awareness among decision makers of the value of reliable predictions. This presentation will show some examples of work carried out at Irstea in France to connect different aspects of the hydrometeorological forecasting chain.

29 October
at 15:30

Room: LT

Impact of land temperature analysis in NWP and other recent developments in LSDA at the Met Office

Speaker: Breogan Gomez (UKMO)

Abstract

The Met Office Global Land Surface Data Assimilation (LSDA) system consists of a suite of schemes that give the initialisation fields for various variables. Over the last year, the Met Office has developed a system that creates soil moisture and land temperature analyses using the Simplified Extended Kalman Filter (SEKF) with the aim of having consistent analysis increments across all land variables. Our tests at N320 showed a positive impact in all regions and at all lead times when verifying against screen temperature and relative humidity. Applying the same methodology to the regional NWP model, UKV, yields some positive results. In recent years, a substantial effort has been made to improve the land initialisation in the UK regional NWP system. We have developed a soil moisture analysis using the same methodology as the global model with SEKF and a snow amount analysis using an optimal interpolation algorithm. The former has shown a neutral impact in the screen verification, but it has significantly improved the river flow predictions. The latter provides a much better snow cover estimation when compared to independent satellite products.

17 October
at 10:30

Room: Council

Post-processing at the Australian Bureau of Meteorology: precipitation results and IMPROVER collaboration

Speaker: Tom Gale (BMRC)

Abstract

Post-processing of rainfall is a priority for the Australian Bureau of Meteorology, due to the Australian climate and impacts on a range of industries, particularly agriculture. Significant advances have been made in the last few years, such as 2-3 lead days skill improvement since 2016 for 1mm/day rain amount. The result is better forecasts for the general public and specialised users, and reduced need for meteorologist intervention in forecast production. This talk will cover the changes made to rainfall post-processing over the last few years, and verification results from our 3 hourly rainfall post-processing which became operational in August 2019.

The Australian Bureau of Meteorology is collaborating with the UK Met Office on development of the new IMPROVER post-processing system. IMPROVER focuses on using the outputs from ensemble and convection permitting NWP models, is fully probabilistic, allows for verification of each processing step and is seamless from 15 minutes to 15 days. This talk will cover the differences between expected IMPROVER configuration in the UK and Australia, and work to provide compatible software environments at the Met Office and BoM.

30 September at 10:30

Room: LT

Historical images of Earth from space: IFS versus observations (an informal talk)

Speaker: Philippe Lopez

Abstract

As a follow-up to my improvised (virtual) trip to the Mo on back in July, I have extended my comparison of our model with various historical views of the Earth from space and I thought it would be worth giving an informal presentation about its outcome. Do not expect any slides full of groundbreaking equations. On the other hand, if you just wish to escape from our crazy world for a moment by looking at some nice images of our planet from far (far) away in space and time, please feel free to come.

13 September
at 11:30

Room: LT

Exascale resilience strategies for time-dependent solvers

Speaker: Chris Cantwell (Imperial College)

Abstract

Time-dependent partial differential equations (PDEs) arise in a wide range of application areas, for example in fluid dynamics. The high-fidelity resolution of complex flows often requires large-scale computational resources and is one of the drivers towards exascale computing. Energy usage is a major concern for these systems. The power overhead necessary to implement traditional hardware resilience to combat failures at exascale is likely to be substantial. The usefulness and energy efficiency of our computational tools might therefore be more effectively maintained by making the software more tolerant of the frequent hardware failures anticipated to occur on future large-scale systems.
In this talk I will highlight the case for resilience in software and present our latest efforts to address this challenge in the context of time-dependent PDE solvers. We combine user-level failure mitigation (ULFM), a proposed extension to the MPI 4.0 standard, with remote in-memory check-pointing in a minimally intrusive way, in order to augment existing software tools with scalable fault tolerance capabilities. Our resilience approach improves forward-path performance over conventional techniques by avoiding the parallel file system completely, and allows one or more concurrently failed ranks to be rebuilt with spare ranks on-the-fly and independently of other non-failed processes. I will describe the algorithms, implementation and their performance characteristics, and illustrate their application through examples using the Nektar++ spectral/hp element framework.

20 August
at 10:30

Room: LT

Forecast Evaluation of Set-Valued Properties

Speaker: Tobias Fissler (Imperial College, London)

Abstract

In forecast evaluation, one distinguishes two major tasks: forecast validation (or verification) and forecast comparison. While the former one is commonly performed with identification functions (for point forecasts) or other diagnostic tools of calibrations such as the Probability Integral Transform (for probabilistic forecasts), the latter task utilises scoring functions in the case of point forecasts and scoring rules for probabilistic forecasts. It is a widely accepted paradigm that these scoring functions (rules) should be consistent (proper) in that they honour correctly specified forecasts, thus incentivising truthful reporting under risk-neutrality.

The statistical, climatological and financial literature has seen remarkable contributions to the evaluation of real- or vector-valued properties as well as predictive distributions for real- and vector-valued quantities. On the other hand, the case of set-valued properties has received considerably less attention.

Acknowledging the spatial structure of many climatological and meteorological phenomena of interest, we introduce a theoretical framework for sound forecast evaluation of set-valued quantities such as the expected area of precipitation or confidence regions for floods. To this end, we suggest the formal distinction between a selective notion, where forecasters are content with specifying a single point in the set of interest, and an exhaustive notion, where forecasters are far more ambitious and aim at correctly specifying the entire set of interest. We unveil a stark dichotomy between these two notions: A functional can be either selectively or exhaustively elicitable.

We discuss implications of this mutual exclusivity for best practice in forecast evaluation of set-valued quantities, putting an emphasis on applications in meteorology and climatology.

This talk is based on joint work with Jana Hlavinová and Birgit Rudloff.

13 August
at 10:30

Room: LCR

Trends and Challenges in TC NWP for the 2020s

Speaker: M Fiorino (University of Colorado (USA)

Abstract

Perhaps the greatest success story in the history of Numerical Weather Prediction (NWP) is the 90% improvement1 in tropical cyclone (TC) track forecasts from the 1980s to 2018. This dramatic improvement was made possible by the high-resolution, high-quality global modeling of the leading operational centers ECMWF and the UKMO.
The talk firsts reviews the history of TC NWP – from the early days at Penn State where we made the first TC NWP forecast with operational analyses in a ‘full physics’ model (MM0 – the predecessor to WRF) in the late 1970s – to the 2000s, when global models became the clear ‘go-to’ aid for human forecasters. During the 2000s I was the NWP officer at both the Joint Typhoon Warning Center (JTWC) and the National Hurricane Center (NHC) where my main task was to bring good modeling into the human forecast process. It was during my time at NHC that I demonstrated how the deterministic ECMWF HRES forecasts were often superior to the best forecast aid (consensus) and how this significant advance was not because of a horizontal resolution increase but because of (subtle) changes in the physics (see Fiorino 2009: https://www.ecmwf.int/en/elibrary/17493-record-setting-performance-ecmwf...).
More recently, the limited-area model HWRF, during its 12 years in NCEP operations (2007-2018), has failed to add value to the much lower resolution GFS host global model at the medium range (72 h mean position error). Further, the inability of HWRF to make superior short-range track forecast (12- and 24-h) vis-à-vis ECMWF, despite near ‘perfect’ HWRF initial position error, challenges conventional thinking on the way forward in TC NWP.
This review of TC NWP and the current performance of HWRF/GFS suggest some next steps/challenges for the 2020s: 1) vortex analysis – can model forecasts be improved by a more accurate analysis of the TC vortex? e.g., by assimilation of the truly unique and special observation of the TC forecast centers – ‘TC vitals’; and 2) “completing the forecast” by predicting both TC genesis and dissipation. Metrics is common problem for both vortex initialization and genesis. The talk will conclude with a TC ‘forecast error’ metric proposal, that is more consistent with warnings and TC impacts, and some recent work on deterministic TC genesis verification.

31 July at 11:00

Room: CC

Recent trends and future projections of Meteorological droughts

Speaker: Sergio M Vicente-Serrano (Spanish National Research Council)

Abstract

Drought is one of the most difficult hydroclimatic hazards to quantify and monitor, which has lead to strong scientific debate on the temporal behaviour of drought trends over the last decades and possible future projections. Here the different perspectives considered to identify recent drought trends are shown and discussed, focusing on the different role of both precipitation and atmospheric evaporative demand. The role of the statistical properties of drought (mostly their autoregressive character) and land/atmosphere feedbacks related to plant physiology and CO2 fertilization are also a determining factor to understand droughts under future climate scenarios.

31 July
at 10:30

Room: LT

Efforts on Scaling and Optimizing Climate and Weather Forecasting Programs on Sunway TaihuLight

Speakers: Haohuan Fu (Tsinghua University, China) and Wei Xue (High Performance Computing Institute, Tsinghua, China)

Abstract

The Sunway TaihuLight supercomputer is the world's first system with a peak performance greater than 100 PFlops, and a parallel scale of over 10 million cores. In contrast with other existing heterogeneous supercomputers, which include both CPU processors and PCIe-connected many-core accelerators (NVIDIA GPU or Intel MIC), the computing power of TaihuLight is provided by a homegrown many-core SW26010 CPU that includes both the management processing elements (MPEs) and computing processing elements (CPEs) in one chip. This talk reports our efforts on refactoring and optimizing the climate and weather forecasting programs on Sunway TaihuLight. To map the large code base of CAM and WRF to the millions of cores on the Sunway system, we take OpenACC-based refactoring as the major approach, and apply source-to-source translator tools to exploit the most suitable parallelism for the CPE cluster, and to fit the intermediate variable into the limited on-chip fast buffer. For individual kernels, when comparing the original ported version using only MPEs and the refactored version using both the MPE and CPE clusters, we achieve up to 22x speedup for the compute-intensive kernels. For the 25km resolution CAM global model, we manage to scale to 24,000 MPEs, and 1,536,000 CPEs, and achieve a simulation speed of 2.81 model years per day.

Haohuan Fu is a professor in the Ministry of Education Key Laboratory for Earth System Modeling, and Department of Earth System Science in Tsinghua University, where he leads the research group of High Performance Geo-Computing (HPGC). He is also the deputy director of the National Supercomputing Center in Wuxi, leading the research and development division. Fu has a PhD in computing from Imperial College London. His research work focuses on providing both the most efficient simulation platforms and the most intelligent data management and analysis platforms for geoscience applications, leading to two consecutive winning of the ACM Gordon Bell Prizes (nonhydrostatic atmospheric dynamic solver in 2016, and nonlinear earthquake simulation in 2017).

10 July

at 11.00

Room: CC

London as a laboratory to explore the health effects of air pollution

Speaker: Ian Mudway (Kings College, London)

The new historical reanalysis dataset generated by the Physical Sciences Division of NOAA’s Earth System Research Laboratory and the University of Colorado CIRES, the Twentieth Century Reanalysis version 3 (20CRv3), is a comprehensive global atmospheric circulation dataset spanning 1836 to present, assimilating only surface pressure and using monthly Hadley Centre sea ice distributions (HadISST2.3) and an ensemble of daily Simple Ocean Data Assimilation with Sparse Input (SODAsi.3) sea surface temperatures. SODAsi.3 was forced with a previous version of 20CR that itself was forced with a previous SODAsi, allowing these “iteratively-coupled” boundary conditions to be more consistent with the atmospheric reanalysis. 20CRv3 has been made possible with supercomputing resources of the U.S. Department of Energy and a collaboration with GCOS, WCRP, and the ACRE initiative. It is chiefly motivated by a need to provide an observational validation dataset, with quantified uncertainties, for assessments of climate model simulations of the 19th to 21st centuries, with emphasis on the statistics of daily weather. It uses, together with the NCEP global forecast system (GFS) numerical weather prediction (NWP) land/atmosphere model to provide background "first guess" fields, an Ensemble Kalman Filter (EnKF) data assimilation method. This yields a global analysis every 3 hours as the most likely state of the atmosphere, and also yields the uncertainty of that analysis.

20CRv3 has several improvements compared to the previous version 2c. The analysis and the 80 member ensemble are generated with the NCEP GFS at T254 resolution (about 0.75 degrees latitude by longitude) with 64 levels in the vertical, compared to T62 (about 2 degrees latitude by longitude) and 28 vertical levels in the 20CRv2c 56 member ensemble. This gives an improved representation of extreme events, such as hurricanes. Implementation of a “relaxation to prior” covariance inflation algorithm, combined with stochastic parameterizations in the GFS, provides quantitatively better uncertainty estimates than the previous additive inflation of 20CRv2c. An adaptive localization helps to keep the analysis from over-fitting the observations. A variational quality control system retains more observations. An incremental analysis update procedure produces a temporally smoother analysis without spurious spin-up trends seen in 20CRv2c. Millions of additional pressure observations contained in the new International Surface Pressure Databank version 4.7, such as from the citizen science Oldweather.org project, also improve the analyses. These improvements result in 20CR version “3” having comparable or better analyses to version 2c, as suggested by improved 6 hour forecast skill, more realistic uncertainty in near-surface air temperature, and a reduction in spurious centennial trends in the tropical and polar regions. Possibilities for 200 years of reanalysis are also discussed in light of results of test reanalyses of the 1816 “Year without a Summer”.

Gilbert P. Compo1,2, Laura Slivinski1,2, Jeffrey S. Whitaker2, Prashant D. Sardeshmukh1,2, Benjamin S. Giese3, Philip Brohan4, Rob Allan4

1CIRES, University of Colorado, USA, compo@colorado.edu, 2Physical Sciences Division, Earth System Research Laboratory, NOAA, USA., 3Department of Oceanography, Texas A&M University , USA, 4Met Office Hadley Centre, Exeter, UK

14 June
at 10:30

Room: LT

Ocean Waves as a Missing Link Between Atmosphere and Ocean

Speaker: Alex Babanin (University of Melbourne, Australia)

Abstract

Role of the waves as a link between the ocean and atmosphere will be discussed. It is rapidly becoming clear that many large-scale geophysical processes are essentially coupled with the surface waves, and those include weather, tropical cyclones, ice cover in both Hemispheres, climate and other phenomena in the atmosphere, at air/sea, sea/ice and sea/land interface, and many issues of the upper-ocean mixing below the surface. Besides, the wind-wave climate itself experiences large-scale trends and fluctuations, and can serve as an indicator for changes in the weather climate. In the presentation, we will discuss wave influences at scales from turbulence to climate, on the atmospheric and oceanic sides.

At the atmospheric side of the interface, the air-sea coupling is usually described by means of the drag coefficient Cd, which represents the momentum flux in terms of the wind speed, but the scatter of experimental data with respect to such dependences is very significant and has not improved noticeably over some 40 years. It is argued that the scatter is due to multiple mechanisms which contribute into the sea drag, many of them are due to surface waves and cannot be accounted for unless the waves are explicitly known. We also argue that separation of the momentum flux for the components which go to the waves and to the current, is not trivial and depends on a numbers of factors such as sea state, but also on the measurement setup. In this presentation, field data, both at moderate winds and in Tropical Cyclones, and a WBL model are used to investigate such biases. It is shown that near the surface the turbulent fluxes are less than those obtained by extrapolation using the logarithmic-layer assumption, and the mean wind speeds very near the surface are larger.

Among wave-induced influences at the ocean side, the ocean mixing is most important. Until recently, turbulence produced by the orbital motion of surface waves was not accounted for, and this fact limits performance of the models for the upper-ocean circulation and ultimately large-scale air-sea interactions. Theory and practical applications for the wave-induced turbulence will be reviewed in the presentation. These include viscous and instability theories of wave turbulence, direct numerical simulations and laboratory experiments, field and remote sensing observations and validations, and finally implementations in ocean, Tropical Cyclone, ocean and ice models.

14 June
at 13:00

Room: LT

The Bureau of Meteorology Research Program

Speaker: Peter May (BOM, Australia)

Abstract

An overview of the Australian Bureau of Meteorology's research program will be presented. This will include our plans focusing on high resolution numerical prediction, multi-week and seasonal modelling, advanced post processing, climate science and our work to address some fundamental science and societal challenges. For example our work on fire weather ranging from the changing fire risk in Australia, our current and future guidance and work on firestorms including fully coupled fire -high resolution simulations. Finally I will discuss our future plans in the context of a fundamental transformation of the Bureau and the way we will be providing weather and climate services.

30 May
at 10:30

Room: LT

The atmospheric response to increased ocean model resolution in the ECMWF Integrated Forecasting System: a seamless approach

Speaker: Chris Roberts

Abstract

This study uses initialized forecasts and climate integrations to evaluate the wintertime North Atlantic response to an increase of ocean model resolution from ∼100 km (LRO) to ∼25 km (HRO) in the European Centre for Medium-Range Weather Forecasts Integrated Forecasting System (ECMWF-IFS). The presented analysis considers the atmospheric response at lead times of weeks to decades and assesses mean biases, variability, and subseasonal predictability. Importantly, the simulated impacts are highly dependent on lead time such that impacts seen at climate timescales cannot be generalized to initialized forecasts. At subseasonal lead times (weeks 1-4), sea surface temperature (SST) biases in LRO and HRO configurations are similar and partly inherited from ocean initial conditions. At multidecadal timescales, mean differences are dominated by biases in the LRO configuration, which include a North Atlantic cold bias and a breakdown in the

29 May
at 10:00

Room: LT

The axes of the Météo-France scientific strategy

Speaker: Marc Pontaud (Meteo-France)

Abstract

Make progress in the knowledge and anticipation of extreme phenomena and their impacts, in a context of climate change

Anticipating extreme or high-stakes phenomena and their impacts, in metropolitan France and overseas, is a strong societal expectation addressed to Météo-France. Progress in this area is largely based on improvements made to our numerical weather prediction systems, particularly regional ones, which remain a priority research objective at Météo-France. Progress will come from several areas.

A first source of progress will be the assimilation of an increasing number of observations from new meteorological satellites launched during the period, but also through access to new data sources or through better use of existing sources such as meteorological radars. It will also be a question of optimizing the information provided by these data. This will require the development of a new generation of data assimilation algorithms. The realization of forecasts in the form of sets will be generalized in order to improve by a few hours the anticipation of the occurrence of a meteorological hazard, to better predict its intensity and consequences and to be able to offer new services based on a greater capacity to adapt the forecasts to the expectations of users taking into account their challenges.

Improving the prediction of extreme or high-stakes weather phenomena and their impacts also requires progress in understanding the processes at work and in their modelling at different scales. Research investigation resources (measurement campaigns, instrumented sites, meter-scale modelling, etc.) will be oriented and developed as a priority to meet these objectives. It will also evaluate the contribution of very high resolution modelling, i.e. a few hundred metres grid, to the prediction of meteorological hazards on sites or during events with stakes.

For the very short time frames (0-3h), the contribution of artificial intelligence (AI) and other signal processing techniques to the extrapolation of observations and their fusion with numerical forecasts will be explored.

Moreover, knowledge at the regional level of the evolution of the frequency and intensity of extreme events with climate change is essential for adapting to climate change and strengthening territorial resilience measures. Météo-France will contribute to the scientific response to these issues, identified in the French Adaptation Program, by interpreting recent climate trends through the use of observations and climate model results. The aim will be to carry out simulations of past and future climate, in particular with the fine-scale forecasting model, capable of representing the evolution of Mediterranean episodes, tropical cyclones and the urban heat island during heat waves.

Continue the transition to integrated environmental modelling systems shared between forecasting and climate

Operational weather prediction, seasonal climate prediction and climate study require modelling not only the behaviour of the atmosphere but also that of other interacting components of the Earth system (continental surfaces, ocean, waves, cryosphere, chemical composition of the atmosphere) and anthropogenic factors (urbanization, irrigation, dams, anthropogenic emissions, etc.). This evolution towards environmental modelling will result in the construction and experimentation of a regional "Earth system" composite model with kilometric resolution, with the assistance of partners mastering certain components, such as the ocean or sea ice. This regional modelling system will be modular to allow different configurations and coupling levels depending on the forecast or application objectives.

This work will continue to be part of a single modelling system logic, from the global to the local scale, with tools shared between weather forecasting and climate modelling activities.

For numerical weather prediction, including the chemical composition of the atmosphere, a new research axis will be opened, that of data assimilation coupled between the different components (atmosphere, continental surfaces, aerosols and atmospheric chemistry, ocean, sea states), with the prospect of taking better advantage of certain observations at the interfaces and the exploitation of a greater number of data.

Adapt modeling tools to operational requirements on tomorrow's computing architectures

The operational use of the tools developed by the research is at the heart of the institution's strategy. On the one hand, it allows a rapid transfer of innovations from research to all the institution's activities and, on the other hand, a daily comparison of scientific work with reality, thus allowing researchers to benefit from regular feedback on the quality of their models. This community of tools between scientific and operational activities also imposes constraints that must be integrated by research teams (speed of code execution, compatibility of algorithms with computing infrastructures, etc.). In close collaboration with the meteorological services community, the establishment will carry out the scientific work necessary to prepare for future technological developments in intensive computing, including the development of graphics processors or other new or emerging architectures that will have a profound impact on the structure of digital codes.

Enhance weather and climate forecasts in response to the expectations of internal and external beneficiaries

Météo-France's research must also contribute to the promotion of its innovations, both to internal users (particularly forecasters) and to external users.

In particular, the enhancement of modeling data, which is increasingly accurate and informative but also more numerous, rich and complex, will be a major focus of research. This includes supporting the use of ensemble forecasts by defining methods for statistical extraction of relevant signals and post-processing adapted to this new type of data. To cover the needs in this area, scientific expertise in the field of AI will be strengthened. In a logic of support for end-users, the emphasis will be placed on innovation, the transfer of research results to the operational level and the orientation of scientific activities towards the needs of Météo-France's operational departments.

Strengthen the dynamics of national and international cooperation

The orientation of Météo-France's research activities towards environmental modelling on a regional scale will be accompanied by a strengthening and broadening of cooperation with the French academic and scientific community, such as the CNRS, the academic world and major national research organisations, and with international actors engaged in this same path.

The evolution of numerical weather prediction systems towards coupled systems and the consideration of future supercomputer architectures lead to increased cooperation with the meteorological services community with which the modelling tools are co-developed (European consortia Aladin and Hirlam, ECMWF). To be fully effective, this collaboration between meteorological services will have to be based on better shared tools and software convergence with the ECMWF is a reaffirmed priority for the Establishment.

In the field of space observation of the Earth, Météo-France will consolidate its status as a privileged interlocutor with space agencies for meteorological and climate applications, and more broadly environmental applications. This approach will be based on the close relations established with CNES in France, Eumetsat and ESA in Europe as well as with certain international space agencies.

28 May
at 10:30

Room: LT

Improving atmospheric reanalyses for historical extreme events by rescuing lost weather observations

Speaker: Ed Hawkins (University of Reading)

The Parallel Full Approximation Scheme in Space and Time (PFASST) is an iterative approach to parallelization for the integration of ODEs and PDEs in the time direction. I will give an overview of the PFASST algorithm, discuss the advantages and disadvantages of PFASST compared to other popular parallel in time (PinT) approaches, and show some examples of PFASST in applications. I will also explain the construction of a new class of PinT integrators that combine properties of exponential integrators and PFASST, including some preliminary results on the accuracy and parallel performance of the
algorithm.

13 May
at 11:00

Room: LT

Flood Forecasting and Inundation Mapping using the US National Water Model

Speaker: David R Maidment (University of Texas at Austin)

Abstract

The US National Water Model forecasts flows on 5.2 million km of streams and rivers in the continental United States, divided into 2.7 million forecast reaches. A Medium Range Forecast from this model for Hurricane Harvey prepared 3 days before the hurricane made landfall successfully predicted the spatial pattern of inland flooding in Texas. A continental scale inundation map has been developed using the Height Above Nearest Drainage (HAND) method, and an associated cell phone app called Pin2Flood built that enables flood emergency responders to create their own flood inundation maps using the same HAND map base, thus connecting predictive and observational flood inundation mapping.

About the Presenter: David R Maidment is the Hussein M Alharthy Centennial Chair in Civil Engineering at the University of Texas at Austin, where he has served on the faculty since 1981. He received his BE degree from the University of Canterbury in Christchurch, New Zealand, and his MS and PhD degrees from the University of Illinois. In 2016, he was elected to the US National Academy of Engineering for application of geographic information systems to hydrologic processes.

21 March
at 10:30

Room: LT

Constraining Stochastic Parametrization Schemes using High-Resolution Model Simulations

Speaker: Hannah Christensen (Oxford University)

Abstract

Stochastic parametrizations are used in weather and climate models to represent model error. Designing new stochastic schemes has been the target of much innovative research over the last decade, with a focus on developing physically motivated approaches. We present a technique for systematically deriving new stochastic parametrizations or for constraining existing stochastic parametrizations. We take a high-resolution model simulation and coarse-grain it to the desired forecast model resolution. This provides the initial conditions and forcing data needed to drive a Single Column Model (SCM). By comparing the SCM parametrized tendencies with the evolution of the high-resolution model, we can measure the ‘error’ in the SCM tendencies. As a case study, we use this approach to assess the physical basis of the widely used ‘Stochastically Perturbed Parametrization Tendencies’ (SPPT) scheme using the IFS SCM. We provide justification for the multiplicative nature of SPPT, and for the large temporal and spatial scales used in the stochastic perturbations. However, we also identify issues with the SPPT scheme and motivate improvements. In particular, our results indicate that independently perturbing the tendencies associated with different parametrization schemes is justifiable, but that an alternative approach is needed to represent uncertainty in the convection scheme. It is hoped this new coarse-graining technique will improve both holistic and process-based approaches to stochastic parametrization.

20 March
at 10:30

Room: LT

About novel time integration methods for weather and climate simulations

Speaker: Martin Schreiber (Tech University of Munich)

Abstract

Weather and climate simulations face new challenges due to changes in computer architectures caused by physical limitations. From a pure computing perspective, algorithms are required to cope with stagnating or even decreasing per-core speed and increasing on-chip parallelism. Although this leads to an increase in the overall on-chip compute performance, data movement is increasingly becoming the most critical limiting factor. All in all, these trends will continue and already led to research on partly disruptive mathematical and algorithmic reformulations of dynamic cores, e.g. using (additional) parallelism in the time dimension.

This presentation provides an overview and introduction to the variety of newly developed and evaluated time integration methods for dynamical cores, all aimed at improving the ratio of wall clock time to error:

First, I will begin with rational approximations of exponential integrator methods in their various forms: Terry Haut's rational approach of exponential integrators (T-REXI), Cauchy contour integral methods (CI-REXI) on the complex plane and their relationship to Laplace transformations, and diagonalized Butcher's Tableau (B-REXI).

Second, Semi-Lagrangian (SL) methods are often used to overcome limitations on stable time step sizes induced by nonlinear advection. These methods show superior properties in terms of dispersion accuracy, and we have used this property with the Parareal parallel-in-time algorithm. In addition, a combination of SL with REXI is discussed, including the challenges of such a formulation due to Lagrangian formulation.

Third, the multi-level time integration of spectral deferred correction (ML-SDC) will be discussed, focusing on the multi-level induced truncation of nonlinear interactions and the importance of viscosity in this context. Based on this, the "Parallel Full Approximation Scheme in Space and Time" (PFASST) adds a time parallelism that allows even higher accelerations on the time-to-solution compared to ML-SDC and traditional time integration methods.

All studies were mainly conducted based on the shallow water equations (SWE) on the f-plane and the rotating sphere to investigate horizontal aspects of dynamical cores for weather and climate simulation. Overall, our results motivate further investigation and combination of these methods for operational weather/climate systems.

(With contributions and more from Jed Brown, Francois Hamon, Richard Loft, Michael Minion, Matthew Normile, Nathanaël Schaeffer, Andreas Schmitt, Pedro S Peixoto).

12 March
at 11:15

Room: CC

Running serverless HPC workloads on top of Kubernetes and Jupyter notebooks

Speaker: Thomas Lauvaux (LSCE, Saclay, France)

Abstract

The biogenic component of greenhouse gas fluxes remains the primary source of uncertainties in global and regional inversion systems. But recent results suggest that anthropogenic greenhouse gas emissions from fossil fuel use, so far assumed perfect at all scales, represent a larger fraction of the uncertainties in these systems, and can no longer be ignored. Inversion systems capable of reducing fossil fuel uncertainties are discussed in parallel with planned observing systems deployed across the world and in space. The remaining challenges and recent advances are presented to not only infer fossil fuel emissions but to provide support to climate policy makers at national and local scales.

Uncertainty quantification of pollutant source retrieval: comparison of Bayesian methods with application to the Chernobyl and Fukushima Daiichi accidental releases of radionuclides

Speaker: M Bocquet (CEREA, France)

Abstract

Inverse modeling of the emissions of atmospheric species and pollutants has significantly progressed over the past fifteen years. However, in spite of seemingly reliable estimates, the retrievals are rarely accompanied with an objective estimate of their uncertainty, except when Gaussian statistics are assumed for the errors which is often unrealistic. I will describe rigorous techniques meant to compute this uncertainty in the context of the inverse modeling of the time emission rates -- the so-called source term -- of a point-wise atmospheric tracer. Lognormal statistics are used for the positive source term prior and possibly the observation errors, which precludes simple Gaussian statistics-based solutions.

Firstly, through the so-called empirical Bayesian approach, parameters of the error statistics -- the hyperparameters -- are estimated by maximizing the observation likelihood via an expectation-maximization algorithm. This enables the estimation of an objective source term. Then, the uncertainties attached to the total mass estimate and the source rates are estimated using four Monte Carlo techniques: (i) an importance sampling based on a Laplace proposal, (ii) a naive randomize-then-optimize (RTO) sampling approach, (iii) an unbiased RTO sampling approach, (iv) a basic Markov chain Monte Carlo (MCMC) simulation. Secondly, these methods are compared to a full Bayesian hierarchical approach, using an MCMC based on a transdimensional representation of the source term to reduce the computational cost.

I will apply those methods, and improvements thereof, to the estimation of the atmospheric cesium-137 source terms from the Chernobyl nuclear power plant accident in April/May 1986 and Fukushima Daiichi nuclear power plant accident in March 2011.

2018

1 February
at 13:30

Room: LT

Using All-Sky Satellite Infrared Brightness Temperatures for Model Verification and in Ensemble Data Assimilation Systems

Speaker: Jason Otkin (Cooperative Institute for Meteorological Satellite Studies, University of Wisconsin-Madison, USA)

Abstract

Infrared sensors onboard geostationary satellites provide detailed information about the cloud and water vapor fields with high spatial and temporal resolutions that make them very useful for model verification and within data assimilation systems. In the first part of this talk, results will be shown from several recent studies that used GOES infrared brightness temperatures to assess the accuracy of cloud and water vapor forecasts generated by the High Resolution Rapid Refresh (HRRR) model in real-time and as part of longer-term verification studies. The real-time GOES-based verification system provides operational forecasters objective tools to quickly assess the accuracy of current and prior HRRR model forecasts when they are creating or updating their short-range forecasts. For long-term verification, the forecast accuracy is assessed using a variety of statistical methods ranging from standard grid point metrics to neighborhood-based methods such as the Fractions Skill Score to more sophisticated object-based verification tools. Overall, the results show that the simulated brightness temperatures are too warm during the first hour of the forecast, indicating that the HRRR model initialization is deficient in upper-level clouds. This warm bias, however, is quickly replaced by a large cold bias due to the rapid generation of upper-level clouds with the negative bias often lasting for several hours before the excess cloud cover dissipates. The object-based analysis showed that the HRRR initialization contains too many small cloud objects; however, the number of cloud objects becomes too low by forecast hour 2. This behavior is consistent with the changes in the simulated brightness temperatures and indicates that the forecast cloud objects become too large after a few hours.

In the second part of this talk, output from a high-resolution ensemble data assimilation system (KENDA) is used to assess the ability of a nonlinear bias correction (NBC) method that uses a Taylor series polynomial expansion of the observation-minus-background departures to remove linear and nonlinear conditional biases from all-sky SEVIRI infrared brightness temperatures. Univariate and multivariate NBC experiments were performed in which the satellite zenith angle and variables sensitive to clouds and water vapor were used as the bias correction predictors. The results showed that even though the bias of the entire error distribution is equal to zero regardless of the order of the Taylor series expansion, that there are often large conditional biases that vary as a nonlinear function of the predictor value. The linear 1^st order Taylor series term had the largest impact on the entire distribution as measured by reductions in the variance; however, large conditional biases often remained across the distribution. These conditional biases were typically reduced to near zero after the nonlinear 2^nd (quadratic) and 3^rd (cubic) order terms were used. The results showed that variables sensitive to cloud top height are effective bias predictors especially when higher order Taylor series terms are used. Comparison of statistics compiled for clear-sky and cloudy-sky matched observations revealed that nonlinear bias corrections are more important for cloudy-sky observations as signified by the much larger impact of the 2^nd and 3^rd order terms on the conditional biases.

8 February
at 10:30

Room: LT

The potential of satellites and assimilation to quantify climate forcing, feedbacks and prediction in the Earth System: application to atmospheric chemistry and the carbon cycle

Speaker: Kevin Bowman (JPL, Pasadena, USA)

Abstract

Anthropogenic activities since the industrial revolution have led to profound changes in atmospheric composition (e.g., carbon dioxide, methane and tropospheric ozone) and consequently the trajectory of our climate. However, the coupling of these constituents must be quantified in order to assess the efficacy of climate mitigation strategies against the backdrop of natural variability and climate feedbacks. The last decade has witnessed the launch of satellite constellations that measure Earth’s atmosphere, land, and oceans with a concomitant advance in data assimilation approaches to link these data to Earth System processes.

Using these approaches, we have attributed ozone and methane radiative forcing to global emissions at large urban scales. By incorporating both methane emissions and chemical losses, we show that the top 10% of locations with positive net methane RF are responsible for 50% of the global positive RF and the top 10% of locations with negative RF cause 60% of the global negative RF based upon an RCP 6.0 trajectory through 2050.

To understand the role of the carbon cycle in controlling the most important greenhouse gas, the NASA Carbon Monitoring System Flux (CMS-Flux) project was initiated as a coordinated effort between land surface, ocean, fossil fuel, and atmospheric scientists to develop a comprehensive a carbon cycle data assimilation system. Based upon this system, we attribute the historic atmospheric CO2 growth rate during the 2015 El Nino to spatially-resolved fluxes. We show how tropical productivity and respiration processes related to anomalously high climate variability, i.e., “extreme” events, are responsible for this growth rate and their implications for carbon-climate feedbacks.

Emergent constraints have been become an active area of research that use contemporary observations to constrain climate projections. We have developed a Bayesian formulation of this approach that explicitly accounts for the uncertainty in observations and the uncertainty between the future and present state. We explore the potential of this framework for tropospheric ozone radiative forcing and the carbon cycle.

Taken together, these advances in observations, modeling, and the methodologies to link them point to a scientifically rigorous and policy-relevant framework critically needed for the international community to address climate change.

Dr Kevin Bowman is the Principal Investigator of the EOS Aura Tropospheric Emission Spectrometer and the NASA Carbon Monitoring System (CMS-Flux) project. He received a BEE from Auburn University in 1991, a Diplôme de Spécialisation en Traitement et Transmission des Informations at L'Ecole Supérieure d'Electricité (SUPELEC), Metz, FRANCE in 1993, and a Phd in Electrical Engineering from the Georgia Institute of Technology in 1996. He subsequently continued his career at the Jet Propulsion Laboratory in 1997. His research is centered on understanding the processes controlling atmospheric composition and their impact on climate using satellite observations, modeling, and data assimilation techniques. Dr Bowman's broad interests have led to publications in diverse fields including air quality, carbon cycle, chemistry-climate, atmospheric hydrology, and remote sensing science. An avid musician and guitarist, Dr Bowman is a founding member of the JPL Jazz Propulsion Band.

22 February
at 11:00

Room: Council

Forecasting health hazards linked to heatwaves

Speaker: Claudia di Napoli (Reading University)

Abstract

In recent years severe and prolonged episodes of summer heat such as the 2003 European heatwave proved that extreme high temperatures are responsible for excessed mortality in affected areas, and Heat Health Warning Systems (HHWSs) need to be put in place to mitigate the negative impacts caused by hot weather extremes on human health.

A heatwave-associated HHSW is being developed as part of the pan-European multi-hazard early warning system constructed within the HORIZON2020 ANYWHERE project (EnhANcing emergencY management and response to extreme WeatHER and climate Events). The ANYWHERE HHSW is
based on the forecast of the Universal Thermal Climate Index (UTCI), a state-of-the-art biometeorological index representing the heat stress induced by the atmospheric environment on the human body. Using air temperature, humidity, wind and radiation from ECMWF high-resolution and ensemble prediction models, 10-day UTCI forecasts are computed daily via an operational procedure.

In this seminar the potential of UTCI forecast as a tool to predict heat-related health hazard will be explored. Heat stress conditions across Europe will be presented via UTCI maps computed from 38 years of ERA-Interim data. The association between the UTCI and summer mortality data from 17 European countries will be also discussed, and the UTCI’s ability to represent mortality patterns demonstrated for the 2003 European heatwave.

26 February
at 11:00

Room: LT

The Joint Effort for Data assimilation Integration

Speaker: Yannick Tremolet (JCSDA, USA)

Abstract

The Joint Effort for Data assimilation Integration (JEDI) aims at providing a unified data assimilation framework for all partners of the Joint Center for Satellite Data Assimilation (JCSDA) and the data assimilation community in general. The long term objective is to provide a unified framework for research and operational use, for different components of the Earth system, and for different applications, with the objective of reducing or avoiding redundant work within the community and increasing efficiency of research and of the transition from development teams to operations.

One area where this is particularly important is the use of observations. As Earth observing systems are constantly evolving and new systems launched, continuous scientific developments for exploiting the full potential of the data are necessary. Given the cost of new observing systems and their limited lifetime, it is important that this process happens quickly. Reducing duplication of work and increasing collaboration between agencies in this domain can be achieved through Unified Forward Operators (UFO) and a common Interface for Observation Data Access (IODA).

Over the last decade or two, software development technology has advanced significantly, making routine the use of complex software in everyday life. The key concept in modern software development for complex systems is the separation of concerns. In a well-designed architecture, teams can develop different aspects in parallel without interfering with other teams’ work and without breaking the components they are not working on. Scientists can be more efficient focusing on their area of expertise without having to understand all aspects of the system. This is similar to the concept of modularity. However, modern techniques (such as Object Oriented programming) extend this concept and, just as importantly, help enforce it uniformly throughout a code.

JEDI is based on the Object Oriented Prediction System (OOPS), encapsulating models and observations, which will be briefly described. Extensions towards sharing observations operators and observation related operations such as quality control across models using the UFO will also be described.

JEDI is a collaborative project with developers distributed across agencies and in several locations in different time zones. In order to facilitate collaborative work, modern software development tools are used. These tools include version control, bug and feature development tracking, automated regression testing and provide utilities for exchanging this information. The collaborative development process in JEDI will be presented before concluding with the status of the project.

14 March
at 10:30

Room: MZ

Global Surface Observations of Air Quality

Speaker: Martin G Schultz (Jülich Supercomputing Centre, Forschungszentrum Jülich)

Abstract

The Tropospheric Ozone Assessment Report (TOAR) is an international effort to summarise the scientific knowledge on the distribution, trends and impacts of tropospheric ozone. TOAR has generated global metrics for assessing the impact of tropospheric ozone on climate change, human health and crop/ecosystems. To do this required collating ozone measurements from multiple locations, culminating in the largest database of surface ozone measurements. Martin will talk about the challenges in constructing the database, its use in quantifying ozone impacts, as well as future opportunities for Earth system science (e.g. TOAR-II) and information/data science.

20 March
at 10:30

Room: LT

Dynamics and predictability of sudden stratospheric warmings

Speaker: Thomas Birner (LMU)

Australia has experienced marked climate extremes over the first decade of the 21st century. Its streamflow regime can go through prolonged periods of droughts such as the “Millennium drought” that occurred between 1997 and 2009 across eastern Australia. This extreme dry period was followed by back-to-back La Niña years during 2010-11 and 2011-12, when Australia experienced severe flood events. This variability in extremes has a profound impact on the management of water resources in Australia, key drivers being managing community safety from floods on the one hand and managing water scarcity from droughts on the other hand to minimise the risks related to water supply for urban, irrigation and environmental needs. The Bureau is working actively and cooperatively with all key stakeholders and end users to develop, implement and deliver end-to-end seamless water forecasting services to minimise the impacts of climate variability.

The Bureau’s Flood Forecasting and Warning service provides forecasts of expected river heights across Australia. Under the Water Act (2007), the Bureau is working with water managers across Australia to deliver timely, accurate and reliable seamless water availability forecasts across Australia at seven day and seasonal time scales (http://www.bom.gov.au/water/7daystreamflow; http://www.bom.gov.au/water/ssf). A new service for streamflow, sediment and pollutant fluxes to the Great Barrier Reef is currently under development (http://ereefs.org.au/ereefs).

In this seminar the rationale, progress-to-date and challenges involved in developing and delivering operational water forecasts for Australia will be discussed.

Biographical information

Dr Narendra Kumar Tuteja is manager of the Water Forecasting Services at the Bureau of Meteorology in Australia. He is responsible for development and delivery of the water availability forecast services for Australia, delivering 7 day and seasonal streamflow forecasts as well as long-term water availability trends. His work has supported development of policies and decision making in the water sector. He obtained his PhD from the National University of Ireland in 1996. He is the Advisory Working Group member of the World Meteorological Organisation (WMO) Commission for Hydrology. He is involved in many significant water resources information and management projects across industry and academia in Australia and overseas.

10 April
at 10:30

Room: Council

Supercooled liquid clouds over the Southern Ocean: From processes to cloud feedback

Speaker: Andrew Gettelman (NCAR)

Abstract

The Southern Ocean Clouds Radiation Aerosol Transport Experimental Study (SOCRATES) is measuring from Hobart to the south with a combination of aircraft, ship and station data from Hobart in January and February 2018. One of the major goals of SOCRATES is to understand S. Ocean supercooled liquid and mixed phase clouds. These clouds are important for climate, and may also change as the structure of the atmosphere over the S. Ocean evolves due to climate change. Global models have traditionally struggled to represent these clouds. This talk with show some preliminary SOCRATES data from flights over the S. Ocean, and show how we will link it to global model simulations of cloud microphysical processes, mean climate, and even cloud feedbacks over the S. Ocean. Preliminary implications for how we can modify global model cloud microphysics will be discussed.

12 April
at 10:30

Room: LT

Radiosondes and NWP

Speaker: Bruce Ingleby (ECMWF)

Abstract

A brief introduction to radiosondes will cover the measurement techniques, processing and their uncertainties. One significant daytime uncertainty is the effect of solar radiation on measured stratospheric temperatures, well documented by Dirksen et al. (2014).

There is a migration underway from alphanumeric TEMP reports to binary BUFR reports. BUFR allows for the reporting of high vertical resolution data and the position of each level (currently available from over 30% of global radiosonde stations). Accounting for the radiosonde drift in NWP systems improves the upper level fit between radiosondes and model fields and also the forecasts. This will become operational at ECMWF in June 2018.

Recent work at ECMWF looking at observation-minus-background (O-B) statistics has shown some variations in quality between different radiosonde types (manufacturers), more for temperature and humidity than wind, and this is reflected in new observation uncertainties introduced operationally at ECMWF.
There are also clear variations of O-B statistics with latitude, with larger differences in the tropical stratosphere probably related to gravity wave activity.

In the EU Horizon 2020 GAIA-CLIM project ECMWF and the Met Office looked at the use of radiosonde data as a reference - for both Numerical Weather Prediction systems and satellite sounding data. Biases (in both the observations and models) will be briefly discussed and also recent vertical correlation results derived using the Desroziers et al (2005) technique.

24 May
at 13:30

Room: LT

Possible future ensemble configurations

Speakers: M Leutbecher and F Vitart (ECMWF)

Abstract

Both spatial resolution and ensemble size are main factors determining the cost of operational forecasts. Should ECMWF reduce the ensemble size to achieve the ambitious strategic goal of a medium-range ensemble with 5-kilometre horizontal resolution by 2025? Fewer members will imply reduced skill. Will the resolution increase be able to compensate for that? The combination of subsequent ensemble forecasts, i.e. lagged ensembles, has been proposed to mitigate the impact of reducing ensemble size. In addition, the combination of ensembles with different horizontal resolutions is explored. Initial work has looked at dual-resolution configurations with a mix of higher-resolution members and lower-resolution members. What is the optimum mix of higher and lower-resolution members for the probabilistic skill when the computational cost is constrained? The optimum mix of higher-resolution and lower-resolution members is studied for direct combination of raw forecasts, for combinations with optimum weights and for calibrated ensembles using quantile mapping as well as ensembles calibrated with EMOS. Multi-resolution approaches are also starting to be explored in the context of the initialisation of the ensemble. The first talk gives an overview of on-going work with a focus on the medium-range.

ECMWF extended-range forecasts are currently issued in burst mode from a 51-member ensemble which is integrated for 46 days twice a week (every Monday and Thursday). This methodology is not optimal for users who need extended range forecasts on other days than Mondays and Thursdays. Another approach, operational at NCEP, UKMO and CMA, consists in running smaller ensembles every day and issuing the forecasts in lag mode by combining the most recent forecasts with forecasts produced the preceding days. The second talk will discuss the advantages and disadvantages of both methods (lag vs burst ensemble) for extended range forecasts and will estimate what would be the minimum daily ensemble size and optimal lag window in order to obtain a lag ensemble as skilful as the current 51-member ensemble on Mondays and Thursdays. Following this estimation, several possible configurations of the real-time extended-range forecast ensemble will be proposed for the next HPC in Bologna. Other possible changes to the extended range configuration could include running extended-range forecasts separately from medium-range at legB resolution from step 0 (as before 2008), changing the frequency and ensemble size of the re-forecasts.

20 June
at 10:30

Room: LT

A turbulence scheme with two prognostic turbulence energies

Speaker: Ivan Bastak Duran (Goethe University, Frankfurt)

Abstract

A new turbulence scheme with prognostic equations for two turbulence energies is presented. The scheme is an extension of a Turbulence Kinetic Energy (TKE) scheme with an additional prognostic energy, which represents the effects of temperature and moisture variances in compact form. The extension is inspired by the ideas of Zilitinkevich et al (2013), but the two-energies turbulence scheme is valid for the whole stability range and includes the influence of moisture. The additional turbulence prognostic energy is used only for the calculation of the stability parameter. Thus the two-energies scheme is similar to a standard TKE scheme in that the turbulent fluxes are down-gradient and proportional to the local gradients of the diffused variables. However, the energy-dependent stability parameter is not anymore strictly local and obtains a prognostic character. These characteristics enable the scheme to model both turbulence and clouds in the Planetary Boundary Layer (PBL).

The two-energies scheme was implemented in the Integrated Forecast System (IFS) model developed at the European Centre for Medium-Range Weather Forecasts (ECMWF). The scheme was tested in idealized Single Column Model (SCM) simulations and long-term three-dimensional global simulations. Overall, the scheme performs better than the standard TKE schemes and is able to model both turbulence and shallow convection in the PBL. Long-term three-dimensional global simulations show that the turbulence scheme behaves reasonably well in a full atmospheric model. Compared to the operational turbulence and shallow convection parametrization in IFS, the two-energies scheme shows a more continuous behaviour in time and space, but tends to overestimate cloud cover, especially at low levels.

2 July
at 13:30

Room: LT

Ocean data assimilation challenges for both reanalysis and regional operational applications

Speaker: Andrea Storto (CMRE, Italy)

Abstract

We review in this presentation a few aspects of ocean data assimilation emerging from both global reanalysis and regional operational oceanographic applications.

First, we present current challenges in ocean reanalyses and propose methods to overcome them, looking in particular at methods to limit model drifts, enhance global budget consistency and introduce flow-dependent aspects in data assimilation.

In addition, we show the added value of the ensemble spread in the multi-system reanalysis ensemble from the Copernicus Marine Service to quantify the ocean uncertainty and feed an hybrid ensemble-variational analysis scheme.

Second, we introduce sea-trial activities conducted at CMRE and the related modeling and data assimilation activities, focusing in particular on the challenges of high-resolution ocean data assimilation. Preliminary results from ensemble variational data assimilation with a newly developed stochastic physics package and multi-scale data assimilation will be shown and discussed.

10 July
at 11:30

Room: MZ

Hydrometeorological applications in poorly observed regions and tropical basins : combining satellite observations and complementary techniques using the telecommunication network - examples in Central and West Africa with a focus on the Niger river basin

Speaker: Marielle Gosset (IRD, GET, France)

Abstract

As floods are becoming more frequent in many African cities and rural areas, the impact of these events on population and socio-economical activities is a growing concern. To monitor and predict such intense hydrometeorological events, access to quantitative information on rainfall down to the convective scales is a key - especially to analyse the impact of these events on cities. IRD, CNES and local partners are developing pilot studies on hydrometeorological monitoring/prediction combining satellite information, hydrological modeling and also integrating the use of commercial microwave links (CMLs) from mobile telecommunication networks for high resolution rain estimation. The quantitative results obtained in several countries in west/central Africa using this technique will be presented and further development and possible collaboration on this topic in link with ECMWF activities will be discussed.

19 July
at 14:00

Room: MR1

Model biases and seasonal forecasting

Speaker: Tim Woollings (Oxford University)

Abstract

It is hoped that improvements in models will lead to reduced biases compared to observations, and that this will follow on to enhance the skill of seasonal forecasting systems. This talk will give an overview of some model biases relevant to the midlatitude jets and investigate how these biases might be affecting seasonal forecast skill, with examples from both winter and summer hindcasts.

23 July
at 10:30

Room: LT

JAXA precipitation radars

Speakers: T Kubota, K Furukawa (JAXA, Japan), Y Ikuta (JMA), A Geer (ECMWF)

Abstract

1. Overview of JAXA satellite missions

This talk will provide overview of the Japan Aerospace Exploration Agency (JAXA) satellite missions, such as the "SHIZUKU" (GCOM-W) satellite carrying the AMSR2 instrument, Greenhouse gases observing satellite GOSAT "IBUKI", in addition to the Global Precipitation Measurement (GPM) mission. Recently, "SHIKISAI" (GCOM-C) satellite carrying the SGLI instrument was launched on Dec. 2017, and its data will be open to the public soon. Furthermore, the GOSAT-2 mission and EarthCARE with the ESA are planned to be launched in future. Here, various datasets from the JAXA satellite missions will be introduced briefly.

2. GPM/DPR utilization overview

The Global Precipitation Measurement (GPM) mission is an international collaboration to achieve highly accurate and highly frequent global precipitation observations. The GPM mission is composed of a Tropical Rainfall Measuring Mission (TRMM)-like non-sun-synchronous orbit satellite (GPM Core Observatory) jointly developed by U.S. and Japan and constellation of satellites carrying microwave radiometer instruments provided by the GPM partner agencies. The GPM Core Observatory, launched on February 2014, carries the Dual-frequency Precipitation Radar (DPR) by the Japan. The DPR consists of two radars; Ku-band precipitation radar (KuPR) and Ka-band radar (KaPR), and is expected to advance precipitation science by expanding the coverage of observations to higher latitudes than those of the TRMM Precipitation Radar (PR), measuring snow and light rain by the KaPR, and providing drop size distribution information based on the differential attenuation of echoes at two frequencies. This talk will provide recent results of the GPM/DPR utilization.

3. DPR assimilation at JMA

The GPM/DPR can observe three dimensional distribution of reflectivity all over the earth. In numerical weather prediction (NWP) system, a skillful assimilation of such observation data provides improvement of precipitation forecast. Therefore, the Japan Meteorological Agency (JMA) has been developing the space-born radar assimilation, introducing GPM/DPR in the operational regional NWP system in March 2016 as its first operational assimilation of space-borne radar data. In the GPM/DPR assimilation, vertical humidity profiles are retrieved from the DPR reflectivity profiles using Bayesian theory. Then, the 1-dimensional (1D) relative humidity profiles thus obtained are assimilated in 4-dimensional variational (4DVAR) data assimilation system. This method called 1D+4DVAR is also used for assimilation of ground-based weather radar at JMA. In the performance evaluation tests of the GPM/DPR assimilation, improvement of typhoon and precipitation forecast was demonstrated. These results of performance evaluation test and future plans of development for GPM/DPR assimilation will be presented.

4. Plans for DPR assimilation at ECMWF

Assimilation of DPR reflectivities at ECMWF would be a natural next step following the assimilation of all-sky microwave radiances and NEXRAD gauge-radar precipitation composites. It would improve the initialisation of tropical cyclones, frontal precipitation and convection systems in both the topics and extratropics. In addition, height-resolved precipitation observations would help further constrain microphysical assumptions in the forecast model and in the observation operator. Reflectivities would be assimilated directly into 4D-Var using RTTOV-SCATT version 13, which for the first time will have a radar capability. DPR assimilation will also benefit from preparations being made for the assimilation of the EarthCARE cloud radar.

5. DPR follow-on sensor discussions

JAXA completed the Prime mission phase of the GPM/DPR at the end of November 2017 and moved the Extended mission phase. Now discussions of the DPR follow-on sensor are very active in the JAXA. Recently, the US Decadal Survey 2017 recommended the Clouds, Convection and Precipitation (CCP) mission should be one of 5 designated missions (highest priority), and so we already started to talk with the NASA for possible collaboration in the CCP mission. We’re now studying specifications of the DPR follow-on sensor, such as making the observation width twice (500 km), and opinions from users will be important. Here, we’d like to introduce our recent activity and discuss the DPR follow-on sensor with ECMWF.

25 July
at 10:30

Room: LT

The emerging role of the land surface in weather and climate prediction

Speaker: Paul Dirmeyer (COLA, USA)

Abstract

Like the ocean, the land surface is a slow manifold relative to the atmosphere that provides predictability and prediction skill across a range of time scales. Although the peak influence of land surface states is in the “subseasonal” time range between 1-3 weeks, significant impact of land, or errors in its representation, begins in forecasts at the first morning of simulation. The process chains that link soil moisture, vegetation, snow, and other land states through the energy and water cycles manifest through their effects on the growing daytime boundary layer, cloud formation and convection. Thus, the diurnal cycle is key to assessing and improving model performance related to land-atmosphere interactions. Daily, monthly and seasonal mean skill arising from coupled land-atmosphere feedbacks can only improve by improving the diurnal cycle. We show evidence of land surface impacts on prediction skill from a variety of global models and highlight current shortcomings that may inform model development.

24 September
at 10:30

Room: MR1

Middle atmosphere dynamic, needs and future infrastructures for observations

Speaker: Philippe Keckhut (UVSQ/LATMOS)

Abstract

The middle atmosphere has been poorly investigated for a long time, except the stratosphere and related ozone chemistry issues. Atmospheric waves carry energy and momentum from one region to another. While atmospheric dynamics and the vertical wave interactions exhibit unresolved questions and while new numerical simulations cover a larger vertical range, high-resolution observations are highly required to addressed these issues and improve our understanding and models. A key part of this interaction is the understanding of the role of atmospheric oscillations, particularly tides, gravity and planetary waves, in driving this interaction. Critical to understanding and prediction of sudden stratospheric warming events is to understand the location and structure of shear-zones in the mean-flow where planetary waves break. Most small-scale gravity waves are not resolved by typical climate models and only partially resolved by weather forecasting models. Climate models therefore must parameterize gravity waves to ensure an accurate simulation of middle and upper atmosphere mean climate and variability. Many parameters of the gravity wave parametrizations and particularly gravity wave source parameters are uncertain due to a lack of long-term high-resolution observations.

ARISE (Atmospheric dynamics Research Infrastructure in Europe) is an European infrastructure providing multi-instrumental and multi-sites observations from equator to poles. Some complementary instruments are located on super sites, like lidar, radar and spectrometers, while other provide à global coverage like the infrasonic network. Good progress has been obtained during the last decade and can be expected from a sustainable ARISE infrastructure, in atmospheric modelling, weather forecasting and monitoring of extreme events in relation with civil security applications, from its new 3D images of atmospheric state and its spatial and temporal variability.

The amplitudes of the atmospheric tides are large in the middle atmosphere. Tides generated by stratospheric ozone and by the water vapor in the upper troposphere, can induce some systematic differences between non-simultaneous measurements and represent a key issue for satellite validation and long-term variability when times of measurements are changing. Actual instruments are not numerous and suffer from discontinuities. Temperature trends in the lower and middle stratosphere during the last decade are evaluated using ground-based Lidar, GPS Radio Occultation (RO) and Aqua Advanced Microwave Sounding Unit (AMSU). While trends have slowdown around the 2000’s, in the last decade the stratospheric temperature observations show that atmosphere continue to cool over most of the globe with a rate ranging from -0.4 to -0,7 K/decade. New type of observations can be obtained from space by miniaturized instruments. A possible strategy consists of having a constellation of small satellites to insure a global survey. The nano-satellite MARTIC can provide the required global observations of temperature covering also the mesosphere. GOMOS instrument on Envisat has provided a great opportunity to validate this concept.

Recent results within ARISE project, prospective works and data availability will be presented.

23 October
at 10:30

Room: LT

Turbulence encounter by research aircraft HALO –Generation mechanism and predictability

Speaker: Martina Bramberger (DLR)

Abstract

In September/October 2016, the North Atlantic Waveguide and Downstream Impact EXperiment (NAWDEX) campaign took place in Keflavik, Iceland. During this campaign,
on 13 October 2016, the synoptic background conditions in and around Iceland favoured the excitation of mountain waves and their vertical propagation through the troposphere. The research flight lead the High Altitude and Long Range Research Aircraft HALO through this mountain wave field at lower stratospheric flight levels when it encountered strong localized turbulence. We hypothesize that mountain wave breaking is the dominant mechanism leading to the observed turbulence. To test this hypothesis, we present a comprehensive case study in which turbulence-resolving in-situ aircraft measurements are employed to analyze and quantify turbulence in the described region with parameters such as the turbulent kinetic energy and the energy dissipation rate. Furthermore, the in-situ measurements are compared to turbulence forecasts of the graphical turbulence guidance system (GTG) forced by HRES operational forecasts. This analysis is supported by idealized 3D EULAG simulations to determine the involved processes for the generation of turbulence. Complementing, forecasts and operational analyses of the integrated forecast system (IFS) of the European Centre for Medium-Range Weather Forecasts (ECMWF) are used to analyze the meteorological situation.

1 November
at 15:30

Room: LT

Low-level wind shear and surface winds in cloud-topped boundary layers

Speaker: Louise Nuijens (TU, Delft)

Abstract

Thermodynamic and radiative impacts of clouds on circulations receive much attention. But what about the more direct coupling of clouds to winds, e.g., via the transport of momentum?

I will present idealized Large Eddy Simulations that evaluate the influence of low-level wind shear on convection, cloudiness and the character of momentum transport. We focus on two cases: a subtropical trade-wind regime with predominantly shallow cumulus and congestus, and a midlatitude cold air outbreak with stratocumulus transitioning into cumulus. The simulations are run with DALES on domains of 50x50 - 100x100 km^2.

The first order effect of large-scale wind shear is to adjust near-surface winds via momentum transport. With parametrized surface fluxes, this quickly leads to a different thermodynamic development of the boundary layer, and a different transition from stratocumulus to cumulus. With prescribed surface fluxes, differences in the transition are marginal, but trade cumulus convection and the boundary layer are shallower under stronger shear. I will show that shear reduces minima and maxima in vertical velocity in the sub-cloud layer, and limits the aggregation of convection that can lead to deeper clouds. Shear also has a noticeable impact on cloud-base cloud amount by adjusting turbulence and entrainment across the transition layer.

Under most geostrophic wind profiles, momentum transport slows down the cloud layer, and can either lead to faster or slower near-surface winds depending on the direction of shear and wind turning. The total momentum transport induces friction in the sub-cloud layer, but below convective areas momentum transport accelerates the flow. Mesoscale aggregation of convection is accompanied by pronounced mesoscale variability in near-surface winds, on the order of 2-6 m/s. The relative importance of mesoscale circulations versus local up-and downdrafts on the momentum transport is still under study.

Finally, I will show ongoing work investigating the imprint of cloud cover and convection on low-level winds from 10 year of Cabauw data (soon the new Ruisdael Observatory). Compared to clear-sky days, days with cumulus convection moderately increase mixing of wind across the 200 m near-surface layer, in particular by further enhancing winds in the lowest 20 m during midday, and show more pronounced increases of 200 m averaged winds throughout the day. As cloud cover becomes larger than 50%, the mixing will significantly decrease.

7 November
at 10:30

Room: LT

The Diagonal Score

In this talk I will give an introduction to the following three on-going projects in my group: NUMA (an atmosphere dynamical core for NWP), NUMO (an ocean dynamical core) and CliMa (a new atmospheric model for climate). The aim of this talk is to characterize the overlap between all three modeling components. For example, all three models share similar numerical methods such as continuous and discontinuous Galerkin methods and time-integrators (IMEX). All three models are nonhydrostatic (solving compressible Euler or incompressible Navier-Stokes). NUMA has been well-documented and is performant on multi-CPU and multi-GPU computers. NUMO shares the same modeling framework as NUMAso can use all of the communicators (CPU or GPU). CliMa, on the other hand, is an entirely new model being built from scratch in an open source and open development approach and uses many of the lessons learned from the NUMA project. CliMa is a project in collaboration with Caltech and MIT to build components for the atmosphere, ocean (including ice), and land. Although no results are yet available for CliMa, I will describe the general approach being considered for this new model.

16 November
at 10:30

Room: LT

Tropical cyclones in models: Subseasonal forecasts, climatology and new diagnostics

Speaker: Suzana J Camargo (Lamont-Doherty Earth Observatory, Columbia University)

Abstract

In this talk, I’ll discuss the characteristics of tropical cyclones in climate models, which are used in projections of TC activity under anthropogenic climate change. Some characteristics of TC climatology improve with model resolution, but not all do, and the improvement is not uniform across models. Using a large number of climate models, the relationship of standard TC diagnostics with the mean climate state is analyzed. Models with the same resolution can have a very different TC climatology, even if their large-scale environments are very similar. In order to understand these differences, two new diagnostics were developed that can give insight on how to improve models’ TC climatology, as well as the reliability of their projections.

In the second part of this talk the ability of the current generation of models in forecasting tropical cyclones (TCs) – hurricanes, typhoons – weeks in advance will be discussed. There is predictability in these time scales due to the well-known modulation of TC activity by the Madden-Julian Oscillation (MJO), with a higher level of TC activity when the MJO is in its active phase in a region. As the models’ skill in forecasting the MJO has improved in the last few years, the possibility of forecasting TC formation weeks in advance can be examined. The questions we will discuss are: How well do models simulate the MJO-TC relationship? Do models have skill in forecasting the probability of TC formation weeks in advance? Is the model skill dependent on the amplitude of the MJO?

12 December
at 10:30

Room: LT

Dynamics and predictability of North Atlantic Oscillation regimes

Speaker: Franco Molteni

Abstract

Despite observational evidence of a distinct regime behaviour in the variability of the North Atlantic Oscillation (NAO), the theoretical support for the existence of separate NAO regimes has been limited. In this seminar, a dynamical explanation of NAO regimes is given by a theoretical model of the interactions between zonal flow (and associated temperature gradient), planetary waves and surface heat fluxes in the North Atlantic. Using observational diagnostics to guide the choice of empirical parameters, a 3-variable model is developed, which is formally equivalent to the Lorenz (1963) chaotic model for Rayleigh-Benard convection, and possesses two regimes originated by oscillations around unstable stationary states. Following earlier studies, the model can be expanded to a 5-variable system by splitting the zonal-mean wind into its barotropic and thermal component, and including a variable representing the area-averaged amplitude of high-frequency baroclinic eddies. The extended model still displays a chaotic, two-regime behaviour, with additional sub-seasonal variability driven by the energy exchanges associated with high-frequency eddy amplification and decay. Finally, the usefulness of this model as a tool to interpret and diagnose critical aspects of NAO modelling and long-range predictions will be discussed.

13 December
at 10:30

Room: LT

Seasonal variability of the temperature differences between ECMWF IFS and middle atmospheric Rayleigh lidar measurements at Rio Grande (Tierra del Fuego, Argentina) in 2018

Speaker: Sonja Gisinger (DLR)

Abstract

As revealed from satellite observations, the region around the Southern Andes, Drake Passage, and the Antarctic Peninsula constitutes a hotspot of stratospheric gravity wave (GW) activity. It is this region where global circulation models show a lack of GW drag (GWD) resulting in deficiencies in polar vortex dynamics, polar stratospheric temperatures and ozone concentration. So far, the sources of this "missing GWD" have not been uniquely identified. Different mechanisms are proposed in literature: (i) the downwind advection and meridional refraction of orographic GWs from the Southern Andes and the Antarctic Peninsula into the polar night jet (PNJ), (ii) unresolved orographic GWs from small islands, (iii) secondary GWs generated in the breaking region of the primary orographic waves, (iv) non-orographic GWs from sources associated with winter storm tracks over the Southern Ocean, and (v) a zonally uniform distribution of small amplitude waves by non-orographic mechanisms such as spontaneous adjustment and jet instability around the edge of the stratospheric PNJ.

In order to quantify the important sources and
propagation pathways of mid- and high-latitude GWs in the Southern Hemisphere, the Institute of Atmospheric Physics of the German Aerospace Center (DLR) installed the Compact Rayleigh Autonomous Lidar (CORAL) at Rio Grande (53.79S, 67.75W, Tierra del Fuego, Argentina) in October 2017. These ground-based measurements will be complemented by airborne remote-sensing and in-situ measurements onboard the German research aircraft HALO (High Altitude and Long-Range Research Aircraft) during the SOUTHTRAC field campaign in September 2019. The CORAL is specifically designed for the investigation of gravity waves in the middle atmosphere between 25 and 100 km altitude. It runs fully autonomous and one full year of data above Rio Grande is now available. This temperature data is compared to ECMWF analysis and short-term forecasts (cycle 45r1) at Rio Grande. It is found that above 30 km, the largest temperature difference between the CORAL and the ECMWF system occurs in gravity wave active times of the year. The impact of sponge and vertical resolution on this temperature difference is also investigated.

Cumulus

We review some of the forecast performance across the 2016/17 winter where the seasonal forecasts expected a negative NAO state yet this not really played out. Rather there has been significant forecast volatility, with no forecast skill past week 2 for Europe (or in forecasting the NAO). In the southern hemisphere Australia has experienced an extremely hot summer which was also poorly forecast in the subseasonal range. We discuss sources of forecast error particularly associated with underestimating maritime continent convection but also other factors that may have led to poor forecast skill this year such as lack of MJO activity.

We examine what could be predicted with perfect knowledge of the tropics from relaxation experiments. We compare this year to other recent years where skill in predicting the NAO has been higher. Results based on Rossby wave source analysis suggest that tropical teleconnections are too weak in the ECMWF model - this can lead to periods of under confidence in forecasting predictable extratropical signals. We conjecture that current subseasonal models contain too much damped persistence of the initial conditions and not enough forcing from the tropics.

6 March
at 14:00

Room: LT

Water Information in Australia - The Post Millennium Drought Experience

Speaker: Dr Dasarath Jayasuriya (BoM, Australia)

Abstract

Water data and information are needed for planning, managing and meeting the sustainable development goals related to water resources taking into account climate changes, population growth and other drivers impacting supply security. The Australian National Plan for Water Security (2007) postulated that better water information was essential for providing water security during Australia's periodic droughts. After the Millennium Drought which lasted from 1996 to 2007, the Government and the policy bureaucracy understood that they were facing a water crisis, but had no nationally consistent and accessible data to make informed decisions. Water data were held by approximately 200 organisations across state and local governments with no standardisation, transparency, and accessibility. Parochial interests often led to ‘gaming’.

The presentation lays out the Australian experience and its response to the water data and information challenge and shares the Government’s Water Information Program initiative over the last 10 years. Highlights relate to socialising the universal availability of water data, water information products and water forecasting services. Collectively they provide water intelligence to Bureau’s stakeholders including Government, industry and community.

Dr Dasarath Jayasuriya

23 March
at 10:30

Room: MR1

Uncertainties in simulated evapotranspiration from land surface models over a 15-year Mediterranean crop succession

Speaker: Sebastien Garrigues (CEH/UoR/INRA)

Abstract

https://www.ecmwf.int/sites/default/files/elibrary/2017/17047-uncertaint...

11 April
at 14:00

Room: LT

Dataflow for Geocomputing

Speaker: Georgi Gaydadjiev (Maxeler, London, UK)

Abstract

Ever since Von Neuman, predicting the weather has been one of humanities top computing challenges. The challenge can be split into two parts: (1) writing the software to predict the weather and (2) running the software to predict the weather. Running the software includes deciding on discretisation, data sizes, and building custom configurations of computing, memory and storage components.

A key question arises. Should we build computers that are easy to program or should we build computers that are efficient in running the largest models we can conceive. Of course in the initial decades, software, math and models need to be developed. However in the steady state, we will need to transition to optimising the operational aspects of running complex models.

Maxeler Multiscale Dataflow Computing addresses the steady-state challenge of optimising operational efficiency in Space, Time and Value (STV). We re-evaluate choices in discretisation of space, time and value, and build optimal computational arithmetic pipelines to compute PDE solutions in minimal time and with minimal cost.

Maxeler Dataflow computing has been demonstrated in Italy to compute a local weather model 60x faster and in China, to be 14x faster than a Tianhe 1a (dual GPU node) on a node-2-node basis. The UK government purchased a significant Maxeler machine for Daresbury labs. Still programming challenges remain and we are looking forward to continuing the dialogue with the geoscientists on how this new computational paradigm can be deployed with minimal impact on programming and code maintenance.

21 April
at 10:30

Room: LT

SODA3 (Simple Ocean Data Assimilation ocean/sea ice reanalysis) and a step toward a coupled reanalysis?

Speaker: James Carton (Univ. Maryland, USA)

Abstract

Atmospheric reanalyses produce surface fluxes as a residual of their update cycle. These fluxes should be consistent with estimates of ocean heat and freshwater storage and divergence of transports. Here we compare the ERA-Int, MERRA2, and JRA-55 fluxes with the imbalances apparent in the increments produced by the SODA3 ocean reanalysis system during the data-rich eight year period 2007-2014. The heat flux comparison reveals that the regional imbalance falls in the range of 10-30 W/m2 in time mean with even larger imbalances on seasonal time scales. In the vertical the corresponding temperature imbalances are concentrated in the mixed layer. We argue that in the interior gyres these imbalances are the result of seasonal errors in the atmospheric reanalysis representation of surface heat and freshwater fluxes (the problem is the atmosphere). In other regions such as the Gulf Stream we show evidence of substantial ocean model error. Elsewhere errors in momentum fluxes appear to dominate the temperature and salinity increments. We also examine the impact on fluxes of the choice of bulk parameterization used to calculate surface fluxes from atmospheric state variables.

In the second part of the talk we present a strategy for correcting surface fluxes based on the ocean observations and we then discuss the results of experiments testing this strategy. Examination of the results confirms that in the interior gyres our bias-correction strategy reduces the seasonal error in surface heat flux to less than 5 W/m2. We conclude with a discussion of the changes that are needed to address biases at high latitude.

25 April
at 10:30

Room: LT

Developments in ensemble variational data assimilation at Météo-France

Speaker: Benjamin Ménétrier (Météo-France, France)

Abstract

A brief overview of the work that is being carried out at Meteo-France in data assimilation will be presented. Short term developments include a better resolution for the ensemble of global 4DVars, whose size might double. Significant steps are also made towards the operational implementation of an ensemble of 3DVars for the regional model. Long-term plans for both global and regional models are based on a transition from 3D/4DVar to 3D/4DEnVar algorithms. After 3 years of development within the OOPS framework, numerous improvements for the EnVar methods are now available: enhanced localisation, faster distribution of members, hybrid background error covariance matrix, block-Krylov methods, etc. The implementation of all observation operators in the OOPS framework has been recently completed, enabling deeper investigations about the EnVar behaviour in an operational-like setup.

A crucial aspect of ensemble methods, and especially of EnVar algorithms, is the need for an efficient and well-tuned localisation. Affordable methods are now available to diagnose localisation functions objectively, using the ensemble only. A detailed presentation of both mathematical background and practical implementation of these methods will be given, showing consistent results for several atmospheric and oceanic models. Extensions of the theory towards hybrid covariance matrices and 4D localisation have been developed recently. All these diagnostics are publicly available in an open-source, easily operable code that works for any kind of model grid.

25 April
at 14:00

Room: LT

Non-oscillatory Forward-in-Time Unstructured-Mesh Models for Fluid Flows

Speaker: Joanna Szmelter (Loughborough Univ., UK)

Abstract

The presentation will summarize the development of a fully unstructured (and hybrid) mesh class of models for multi-physics applications with emphasis on simulating inertia gravity waves. Global and limited area atmospheric models will be discussed. The methodology employs an edge-based, finite volume discretisation within the non-oscillatory forward-in-time (NFT) framework. The edge-based data structure allows integration of the generic physical form of the governing PDE over arbitrarily-shaped cells. Aspects of unstructured meshes flexibility, such as optimal point distribution, adaptivity and multigrid will be discussed together with different options of applying the general NFT framework based on the unstructured mesh Multidimensional Positive Definite Advection Transport Algorithm (MPDATA) and elliptic (Krylov) solvers. This approach will be evaluated for a range of models from soundproof and compressible nonhydrostatic to engineering flow solvers. Simulations of stratified orographic flows and the associated gravity-wave phenomena in media with uniform and variable dispersive properties will be presented. They verify the developments and demonstrate the efficacy of the implicit large eddy simulation operating on unstructured meshes for study of stratified turbulent flows. Numerical examples for engineering problems will include coastal flows, gas dynamics at Mach numbers from zero to supersonic, and solutions from Arbitrary Lagrangian Eulerian (ALE) codes using conservative MPDATA based remapping techniques.

26 April
at 10:30

Room: LT

Compatible finite element methods for numerical weather prediction

Speaker: Jemma Shipton (Imperial College, UK)

Abstract

We present recent work on developing a compatible finite element model for numerical weather prediction. This work has been motivated by the requirement for numerical discretisations that are stable and accurate on nonorthogonal grids (such as icosahedral or equiangular cubed sphere grids) without sacrificing properties of conservation, balance and wave propagation that are important for accurate atmosphere modelling on the scales relevant to weather and climate (Staniforth et. al. 2012).

Compatible finite element methods are a type of mixed finite element method (where different finite element spaces are used for different fields) where the divergence of the velocity space maps on to the pressure space. This necessitates the use of div-conforming finite element spaces for velocity, such as Raviart-Thomas and Brezzi-Douglas-Marini, and discontinuous finite element spaces for pressure. The main reason for choosing compatible finite element spaces is that they have a discrete Helmholtz decomposition of the velocity space; this means that there is a clean separation between divergence-free and rotational velocity fields. Cotter and Shipton (2012) used this decomposition to demonstrate that compatible finite element discretisations for the linear shallow water equations satisfy the basic conservation, balance and wave propagation properties listed in Staniforth and Thuburn (2012).

In the talk we will show the progress we have made towards extending this approach to the fully 3D equations via a vertical slice model. Many current atmospheric models use a staggered Charney Phillips grid in the vertical to ensure a good representation of hydrostatic balance. In the finite element context the equivalent staggering requires the temperature field to be discontinuous in the horizontal direction but continuous in the vertical. We present a stable and accurate advection scheme for this field. The success of this approach is illustrated by benchmarking results from our model, implemented in Firedrake (www.firedrake.org)

5 May
at 10:30

Room: LT

Improved Climate Forecasting Service of Australian Bureau of Meteorology

Speaker: William Wang (Bureau of Meteorology, Australia)

Abstract

Australia Bureau of Meteorology has started seasonal climate outlook service since 1989 based purely on statistical methods using tropical SST conditions as predictors. In May 2013 the Bureau of Meteorology Climate Information Services has started issuing climate outlooks based on the Predictive Climate Ocean Atmosphere Model for Australia (POAMA), a dynamical climate model developed by the Bureau of Meteorology and CSIRO Marine and Atmospheric research division. Since 2016, we have been developing our next generation of climate forecasting system and products based on UKMO Glosea5 or ACCESS-S. This talk is to give an overall picture of the climate prediction services in Australia through the following aspects:

A very brief introduction of our Bureau’s climate forecast service history; why we did what we did, experiences and lessons we obtained from the past;
The current dynamic climate forecasting system
1. A very brief introduction of (POAMA)
2. The revolutionary transition from statistic to dynamic and how did we make such a CHANGE;
3. What we investigate when we develop a new climate forecasting system and how, including verification metrics, such as percent consistence, LEPS, ROC skill, Brier skill score etc. and some major results as the benchmarks of a system;
4. The current services and the online products and relevant issues.
The development of the next generation of Bureau’s climate forecasting system
The new ACCESS-S/UK Met Office Glosea5 model and why it was chosen;
1. Development of new forecasting system and products– the concept of seamless forecasting from weather to annual climate prediction.
2. Primary results of this development
3. A brief introduction of a new verification metric, the so called weighted percent consistence

15 May
at 13:30

Room: LT

Measuring Large Scale Divergence and Vorticity by Aircraft

Speaker: Bjorn Stevens (MPI, Germany)

Abstract

During NARVAL-2 an extensive array of dropsondes were launched over the tropical Atlantic, near the ITCZ, to test methods for measuring large scale vertical velocity. These measurements have been evaluated and show that as few as 12 dropsondes launched by an aircraft are well suited to estimating divergence on the scale of a flight leg. Analysis of high-resolution (1-2 km) simulations over the tropical Atlantic during the ICON period are used to evaluate the credibility of the measurements, and the richness of the vertical structure observed, as well as to characterize the spatial and temporal correlation of divergence as a function of scale. The analysis and measurements open a new chapter in observing diabatic processes in the tropical atmosphere and underpin a major new field initiative planned for 2020.

16 May
at 10:30

Room: LT

Mesoscale aggregation of shallow marine cumulus convection

Speaker: Christopher Bretherton (University of Washington, USA)

Abstract

Over the oceans, shallow cumulus convection, often mixed with patchy stratocumulus, is a common cloud type. It is usually 'aggregated' into mesoscale patches or polygons of deeper cumuli, with possible consequences for the mean vertical structure of cloud cover and cloud-precipitation-aerosol interaction. Large-eddy simulations (LES) covering domains 50 km or more across also exhibit mesoscale aggregation of shallow cumulus convection, but it is not fundamentally well understood. To further that understanding, we analyze the development of convective aggregation in multiday LES of a 108x108 km doubly periodic domain simulating mean summertime conditions at a location east of Hawaii. The simulated convection aggregates within 12 hours. Vertically resolved heat and moisture budgets on mesoscale subdomains elucidate this process. Shallow cumulus deepen preferentially in more humid regions of the boundary layer, stimulating net moisture convergence into those regions. Sensitivity studies show that the aggregation does not require precipitation. Aggregation is weakened but not prevented if radiative cooling and surface fluxes are horizontally homogenized. A unifying conceptual model explains these findings.

18 May
at 10:30

Room: MR1

"How do I know if I’ve improved my continental scale flood early warning system?"

Speaker: Hannah Cloke (University of Reading, UK)

Abstract

Flood early warning systems mitigate damages and loss of life and are an economically efficient way of enhancing disaster resilience. The use of continental scale flood early warning systems is rapidly growing. The European Flood Awareness System (EFAS) is a pan-European flood early warning system forced by a multi-model ensemble of numerical weather predictions. Responses to scientific and technical changes can be complex in these computationally expensive continental scale systems, and improvements need to be tested by evaluating runs of the whole system. It is demonstrated here that forecast skill is not correlated with the value of warnings. In order to tell if the system has been improved an evaluation strategy is required that considers both forecast skill and warning value. The combination of a multi-forcing ensemble of EFAS flood forecasts is evaluated with a new skill-value strategy. The full multi-forcing ensemble is recommended for operational forecasting, but, there are spatial variations in the optimal forecast combination. Results indicate that optimizing forecasts based on value rather than skill alters the optimal forcing combination and the forecast performance. Also indicated is that model diversity and ensemble size are both important in achieving best overall performance. The use of several evaluation measures that consider both skill and value is strongly recommended when considering improvements to early warning systems.

19 May
at 14:00
Room: LT

Application of a CubeSat-Based Passive Microwave Constellation to Operational Meteorology

Speaker: Al Gasiewski (University of Colorado, USA)

Abstract

In their most recent decadal assessment (Earth Application from Space, 2007) of Earth science space missions the U.S. National Research Council identified the Precipitation and Allweather Temperature and Humidity (PATH) mission as one of ten recommended medium cost missions. Based on the NRC’s outlined goals, PATH would have the unique capability of providing allweather temperature and moisture soundings and cloud and raincell imagery at spatial scales comparable to AMSU-A/B or ATMS, but at sub-hourly temporal resolution. The essential need is to provide the atmospheric penetrability and spatial resolution of operational microwave sensors but with temporal resolution commensurate with the natural rate of evolution of convectively driven weather. This seminar will focus on the merits of a constellation of passive microwave sounding and imaging CubeSats for achieving PATH goals from the multiple viewpoints of calibration accuracy, data assimilation and global sampling, downlink capability and latency, and orbital lifetime and launch availability. Microwave spectral imagery at 50, 118, and 183 GHz with spatial resolution of ~10-30 km and temporal resolution of ~15-60 minutes from such a fleet could be expected to significantly enhance forecasting of mesoscale convective weather and hurricane rain band evolution, along with provide valuable temporal gap-filling data for synoptic weather forecasting. It is argued that from a joint technology, science, and operational standpoint that a cost-effective realization of the PATH goals, but with the additional features of global coverage and improved NWP sensitivity, can be achieved by a low-cost random-orbit constellation of CubeSats supporting the ATMS and 118 GHz bands. The CU PolarCube mission will be discussed as a basis for this fleet concept.

22 May

at 9:30

Room: CC

Statistical postprocessing of multi-model ensemble
precipitation forecasts in the US National Weather Service

Speaker: Tom Hamill (NOAA Earth System Research Lab, USA)

Abstract

Taking concepts developed in programs such as THORPEX/TIGGE, the US NWS is embracing the use of multi-model ensembles and their statistical postprocessing using short training data sets. The intended goal is to produce dramatically improved forecasts of important weather elements such as temperature, winds, and precipitation, both deterministic and probabilistic. Generating reliable and skillful forecasts of precipitation will be the focus of this talk. Postprocessing of precipitation is especially challenging since forecast bias depends on precipitation amount, and a short training sample (here, the previous 60 days) is often insufficient to estimate this conditional, often location-dependent bias.

This talk will discussed a sequential algorithm for precipitation postprocessing over the US for forecast leads of +1 to +10 days using US and Canadian global ensemble data, an algorithm that provides a workaround for limitations imposed by using a short training data set. The steps in the postprocessing include: (a) pre-specification of "supplemental locations" for each model grid point -- i.e., the specification of other grid points with similar precipitation climatologies and terrain characteristics that are likely to have similar forecast biases. (b) generating forecast and analyzed cumulative distribution functions (CDFs) for each grid point using the last 60 days of forecasts, including data from supplemental locations. (c) Quantile mapping of each member forecast at each grid point; (d) Dressing each quantile-mapped forecast with amount-dependent noise to account for the under-spread character of ensemble forecasts; (e) estimation of probabilities from the ensemble relative frequency; and (f) Savitzky-Golay smoothing of forecasts in regions with little variation in terrain height. The resulting forecasts are shown to be highly reliable and skillful.

25 May
at 10:30

Room: LT

20th Century trends in potential predictability and the role of land surface

Abstract

Speaker: Bart van den Hurk (KNMI, The Netherlands)

Climate change does give rise to shifts in patterns of risks of high-impact extreme events. An important means to adapt to extreme events is early warning, based on skilful forecasts. The ability to forecast any signal is inherently limited by the chaotic nature of the system. A change in the variability patterns in the climate system may or may not impose shifts in the inherent potential predictability of phenomena at monthly to seasonal time scales, which may or may not change our ability to prepare to these phenomena.

During a short sabbatical stay I've explored signatures of trends in potential predictability and the role of land surface processes in this. Use was made of earlier 20th century reforecasts by Antje Weisheimer, where at 4 startdates per year an ensemble of 51 4-month forecasts was produced initialized from ERA20C. In order to explore the role of the land surface, a duplicate experiment was performed in which the land surface in each of the ensemble members was initialized from different states, removing the source of predictability arising from this component in the climate system.

The seminar will review some related studies reported earlier in the literature, and will present early results of the experiments carried out at ECMWF. Suggestions for further analyses from the audience are more than welcome.

19 June
at 14:00

Room: MR1

Atmospheric Rivers in Europe: from moisture sources to impacts and future climate scenarios

Speaker: Alexandre Ramos (Instituto Dom Luiz, Portugal)

Abstract

An atmospheric river (AR) detection algorithm is used for the North Atlantic ocean basin, allowing the identification of the major ARs affecting western European coasts between 1979 and 2014. The western coast of Europe was divided into five domains, namely the Iberian Peninsula, France, UK, Southern Scandinavia and the Netherlands, and Northern Scandinavia. Following the identification of the main ARs that made landfall in western Europe, a Lagrangian analysis was then applied in order to identify the main areas where the moisture uptake was anomalous and contributed to the ARs reaching each domain.

There is a strong relationship between extreme precipitation across Europe and ARs. In the particular case of the Iberian Peninsula, the major AR events that affected the region were studied in detailed. It was analyzed if the extreme precipitation days in the Iberian Peninsula are association (or not) with the occurrence of ARs. Results show that the association between ARs and extreme precipitation days in the western river basins is noteworthy, while for the eastern and southern basins the impact of ARs is reduced.

Since the ARs are associated with high impact weather it is of most importance to assess changes in the ARs frequency in future climate changes. Regarding Europe, changes in the vertically integrated horizontal water transport were analyzed using six global climate models. There is an increase in the vertically integrated horizontal water transport which lead to an increase in the ARs frequency for the 2074- 2099 period when compared with the historical simulation (1980-2005). This increase in future ARs frequency is estimated to be more visible in the high emission scenarios (RCP8.5) when it can more than double the annual mean of ARs obtained for the historical simulation, as shown in Figure 2 for the “Southern Scandinavia and The Netherlands” domain.

20 June
at 14:00

Room: MZR

WAVE2NEMO: A regional WAM-NEMO model setup

Speaker: Øyvind Breivik (Norwegian Meteorological Institute, Norway)

Abstract

The WAVE2NEMO project, funded by CMEMS, aims to implement a number of wave effects in a regional NEMO version. The project builds on work started at ECMWF and implements Coriolis-Stokes forcing, wave-induced mixing by breaking waves and wave-modified momentum fluxes in forced mode with wave fields taken from a regional WAM model. A wave-forced model of the North Sea and the Baltic Sea is found to have improved water level representation during two wind storms in the autumn of 2013. The model also shows slightly improved SST biases in the Baltic Sea compared to a control run. Recent work on approximate Stokes drift profiles (Breivik et al, 2016, Li et al, 2017) has led to a new layer-averaged Stokes drift profile which is now being tested in the regional NEMO setup. We hope to collaborate with ECMWF on the implementation of these wave effects in the most recent version of NEMO.

20 June
at 14:00

Room: LT

Porting IFS physics cloud scheme (CLOUDSC) to GPUs with OpenACC

Speaker: Huadong Xiao (ECMWF, UK)

Abstract

A physics cloud scheme is an important component of a numerical weather/climate prediction model which represents the cloud processes in it. The IFS physics cloud scheme (CLOUDSC) solves five prognostic equations for water vapour, cloud liquid water, rain, cloud ice and snow mixing ratios in an implicit numerical framework. It is a computationally intensive part in the ECMWF model. GPU computing has been extensively applied in scientific and engineering computing recently due to the progress of GPU hardware and related programming models such as CUDA/OpenCL and OpenACC. Given the advantages with respect to code portability and simplicity with OpenACC over CUDA we implement an accelerated CLOUDSC with OpenACC. It combines MPI, OpenMP and OpenACC to fully exploit the power of GPUs across a single or multiple nodes, together with various optimizations. The results show that in terms of the total GPU runtime the obtained performance of 4 NVIDIA GK210 GPUs (2 NVIDIA K80) on a single node for a single time-step is comparable to that of a whole node of 2 Intel Xeon Haswell CPU with 24 cores if the workload saturates GPUs; but without taking into account the GPU data transfer and other overheads, the GPU actual calculation time with a single GPU is about 17% less than that of a whole CPU node with 24 cores for the same problem size. In an operational forecast where multiple time-steps are taken data can be resident on GPU and need to be communicated to CPU only during an I/O step. Therefore, based on the results of this implementation, we conclude that a GPU adaptation of the CLOUDSC can be competitive in terms of performance with respect to CPU.

22 June
at 11:15

Room: MZR

A statistical-dynamical ensemble approach to forecasting water hazards

Speaker: Louise Slater (Loughborough University, UK)

Abstract

Traditional hydrologic forecasting approaches are either statistical, relating climate precursors to streamflow/stage, or dynamical, forcing hydrologic models with meteorological forecasts. In contrast, hybrid statistical-dynamical techniques benefit from different strengths: they are computationally efficient, can integrate and ‘learn’ from a broad selection of input data (GCM forecasts, Earth Observation time series, teleconnection patterns), and can benefit from recent progress in machine learning (e.g. multi-model blending, post-processing and ensembling techniques).

Using climate forecasts from the North American Multi Model Ensemble (NMME: CCSM3, CCSM4, CanCM3, CanCM4, GFDL2.1, FLORb01,GEOS5, and CFSv2), we conduct systematic probabilistic forecasting of monthly-seasonal streamflow and water levels in several hundred catchments across the U.S. and U.K, over weekly to decadal timescales. In heavily urbanised or agricultural catchments, we find that changing land use is a key predictor of streamflow and water levels. While our current approach uses various proxies to model the effects of land cover, this work opens up new possibilities for combining Earth Observation time series with GCM forecasts to predict a variety of hazards (incl. low/high water levels and water pollution) from space using data science techniques.

23 June
at 10:30

Room: MR1

Bias correction of tropical cyclone size and structure in the ECMWF global ensemble prediction system

Speaker: Jeff Kepert (Bureau of Meteorology, Australia)

Abstract

Global EPS systems provide very valuable warning of severe weather risk, including tropical cyclones. One limitation arises because the spatial resolution of the EPS is necessarily too coarse to capture the fine details of the tropical cyclone's core circulation, resulting in the predicted storms having systematic biases in structure and intensity. In particular, the cyclones are too weak and the radius to maximum winds is too large, with these biases increasing with forecast length. Ocean waves are a major impact mechanism of tropical cyclones, but these atmospheric biases result in errors in the predicted ocean wave field also. While these biases to not affect users who respond purely to the expected risk of a tropical cyclone in their vicinity, they are important for users who wish to make more quantitative use of the wind or wave predictions.

We have developed a method for bias correcting tropical cyclone structure and intensity in EPS systems, and applied it to cyclones off north west Australia in the ECMWF global EPS. The method diagnoses the properties of forecast storms within the EPS, and replaces them in the surface wind and pressure field with parametric vortices with the corrected intensity and structure. The correction scheme is statistical, based on regression analysis of historical errors. The bias-corrected wind field ensemble is then used to force a wave model ensemble.

This work was supported by Woodside, Shell, Inpex and Chevron, who operate in the waters off north west Australia.

27 June
at 11:00

Room: MR1

A Case Study of the June 2013 Biomass-Burning Haze Event Using WRF-Chem

Speaker: Alan Betts (Atmospheric Research)

Abstract

We quantify the biases in the diurnal cycle of temperature in ERA-Interim for both warm and cold season using hourly climate station data for four stations in Saskatchewan from 1979-2006. The warm season biases increase as opaque cloud cover decreases, and change substantially from April to October. The bias in mean temperature increases almost monotonically from small negative values in April to small positive values in the fall. Under clear skies, the bias in maximum temperature is of the order of -1 ^o C in June and July, and -2 ^o C in spring and fall; while the bias in minimum temperature increases almost monotonically from +1 ^o C in spring to +2.5 o C in October. The bias in the diurnal temperature range falls under clear skies from -2.5 ^o C in spring to -5 ^o C in fall. The cold season biases with surface snow have a different structure. The biases in maximum, mean and minimum temperature with a stable BL reach +1 ^o C, +2.6 o C and +3 ^o C respectively in January under clear skies. The cold season bias in diurnal range increases from about -1.8 ^o C in the fall to positive values in March. These diurnal biases in 2-m temperature and their seasonal trends are consistent with a high bias in both the diurnal and seasonal amplitude of the model ground heat flux, and a warm season daytime bias resulting from the model fixed leaf area index. Our results can be used as bias corrections in agricultural modeling that use these reanalysis data, and also as a framework for understanding model biases.

8 August
at 10:30

Room: MR1

Introducing the new Master of Science Degree in Environmental Meteorology

Speaker: Dino Zardi (University of Trento, Italy)

Abstract

The seminar introduces the project of a new programme, leading to an MSc degree in Environmental Meteorology, jointly offered by the Universities of Trento (Italy) and Innsbruck (Austria). The programme aims at preparing experts with a solid background in atmospheric processes relevant for a range of environmental systems and interactions. Courses cover not only typical meteorological subjects - including observational, theoretical, and numerical modelling approaches - but also environmental topics, such as hydrology, renewable energy resource assessment, agricultural and forest interactions with the atmosphere, air quality measurement and modelling, mountain meteorology, environmental physical-chemistry, biogeochemistry. All the lectures will be given in English, partly in Trento and partly in Innsbruck. The programme is going to start in September 2018. Access will be offered to a limited number of selected candidates. During my seminar, I will present this new educational initiative, and will be very happy to hear comments and suggestions from the audience.

6 September at 10:30

Room: LT

Parametrization of non-orographic gravity-waves: formalism, impact and test against observations

Speaker: François Lott, (LMD, France)

Abstract

A stochastic parameterization of the non orographic gravity waves emitted by convection and fronts is presented. For the front the formalism is based on a spontaneous adjustment theory where potential vorticity anomalies emit gravity waves in the course of their evolution. The introduction of explicit sources and the use of a stochastic multiwave formalism makes that the momentum fluxes predicted by the scheme are highly intermittent. This intermittent character is one of the key properties of the gravity waves measured by constant level balloons during the vorcore and concordiasi campaign. It is shown that the scheme can be realistic in representing it. It is also shown that intermittency in non-orographic gravity waves prediction can be important to predict the final warming in the middle atmosphere of the Southern Hemisphere in early spring. With sources, we can also address how the gravity waves forcing changes when the climate change. The results so far are quite neutral, at least in the middle atmosphere of the LMDz GCM, which probably follows that GWs are strongly filtered by the large scale background winds in the middle atmosphere.

6 September at 15:30

Room: LT

Ensemble Seasonal Coupled Streamflow Forecast for Hydropower in Brazil

Speaker: Alberto Assis dos Reis (CEMIG, Brazil)

Abstract

The Brazilian electric system is essentially hydrothermal, with a great weight of hydraulic generation (near 65%). The streamflow forecast can be very beneficial to allow early response in extreme climactic events and for a more efficient operation of the hydro power plants. At this project we aim at using ensemble weather forecasts and ESP-Ensemble Streamflow Prediction technique as a forcing for the hydrological models to develop a robust forecasting. The developments will be conducted within the FEWS-Cemig platform which is already running in operational mode. First we will deal with the uncertainty of the observed precipitation, combining different source of gridded observed precipitation to obtain a better estimative, test different sampling methods to apply the ESP methodology, run the hydrologic models for a group of basins and apply flow data assimilation to improve the models results. Then we will deal with the uncertainty of short term, intra-seasonal and seasonal precipitation forecast models and apply bias correction technics to improve the flow forecasts. Finally we will couple the observed precipitation; flow data assimilation; short term, intra-seasonal and seasonal precipitation forecasts and the ESP technic to obtain a seamless ensemble seasonal flow forecast with the hydrological models.

7 September at 10:30

Room: LT

Recent Developments in Atmospheric River Science, Prediction and Applications

Speaker: F. Martin Ralph (CW3E, Univ. California/ Scripps Institute of Oceanography, USA)

Abstract

A number of key developments have occurred recently on the subject of atmospheric rivers (AR). These include advances in observations, physical process understanding, predictions, applications and policies. This presentation will highlight a few of these developments.

Detailed observations of 21 ARs using dropsondes from research aircraft show that an average AR transports, as vapor, roughly 2 times as much water as the Amazon River discharges, as liquid.
Extreme precipitation from a series of landfalling ARs led to damage of a spillway of a major flood control dam in February 2017, which triggered an evacuation of 200,000 people in Northern California.
Methods are being developed to evaluate AR landfall forecast skill, partly to support potential use of AR forecasts in future reservoir operations.
New observations focused on ARs are now available from a specialized mesonet in California and an airborne targeting approach using multi-aircraft dropsonde deployments is under development (including use of research and reconnaissance aircraft normally dedicated to hurricane activities in the warm season).
A critical mass of research and experimental forecast capability, with an emphasis on AR science, predictions and applications, has been developed at the Center for Western Weather and Water Extremes.

15 September at 13:30

Room: LT

Prediction of Arctic sea ice on subseasonal to seasonal time scales

Speaker: Zampieri Lorenzo (AWI, Germany)

Abstract

Sea ice forecasts are becoming a demanding need since human activities in the Arctic are constantly increasing and this trend is expected to continue. In this context, the recent availability of the Subseasonal to Season al Prediction Project (S2S) Dataset has a particularly good timing and provides a solid base to make an initial assessment of the predictive skills of probabilistic forecast systems with dynamical sea ice . In this study , we employ different verification metrics to compare the S2S sea ice forecasts with satellite observations and the models ’ own analys es. In particular , the focus is on the sea ice spatial distribution in the Arctic, which is relevant information for potential final users. The verification metrics, specifically chosen to quantify the quality of the forecasted sea ice edge position , are the Integrated Ice Edge Error (IIEE), the Spatial Probability Score (SPS) and the Modified Hausdorff Distance (MHD).

Despite the early development stage of Arctic sea ice predictions on the seasonal time scale , and the fact that the main focus of the S2S systems is mostly not on sea ice per se, our findings reveal that some of the S2S models are promising , exhibiting better predictive skills than the observation -based climatology and persistence . However, the results also point to critical aspects concerning the data assimilation procedure and the tuning of the models , which can strongly affect the forecasts quality . The comparison of different versions of the ECMWF forecast system shows the benefits brought by a coupled dynamical description of the sea ice instead of its prescription based on persistence and climatological records. Moreover, the systematic application of the verification metrics to such a broad pool of forecasts provides useful indications about strengths and limitation of the verification metrics themselves. Given the increasing availability of new and better sea ice observations and the possible improvement s to coupled seasonal forecast systems, the formulation of reliable Arctic sea ice predictions for the subseasonal to seasonal time scales appears to be a realistic target for the scientific community.

27 September at 14:00

Room: MZR

Strengthening the link between meteorological and energy forecasting communities: Datasets and analytics

Speaker: Pierre Pinson (Technical University of Denmark)

Abstract

Renewable energy modelling and forecasting has come a long way since the first work performed and published in the early 80s. However, the underlying analytics has not progressed much, expect for a few breakthrough related to the transition from deterministic to probabilistic forecasting, or to the recent availability of new types of data in large quantity e.g. distributed surface observations, weather radar images, etc. In this talk, we will review and discuss these changes and how they benefit the integration of renewable energy generation in existing power systems and electricity markets. One is left to realize though that much more could be done if high-resolution regional reanalysis data was available for instance, and if developing big-data based approach to the seamless prediction of renewable energy generation at the continental scale. Personal thoughts on which directions such research is to take, and on further collaboration between energy and meteorological communities, will close the talk."

Short bio: Pierre Pinson is a Professor at the Centre for Electric Power and Energy (CEE) of the Technical university of Denmark (DTU, Dept. of Electrical Engineering), also heading a group focusing on Energy Analytics & Markets. He holds a M.Sc. In Applied Mathematics from INSA Toulouse and a Ph.D. In Energy Engineering from Ecole de Mines de Paris (France). He acts (or has acted) as an Editor for the IEEE Transactions on Power Systems, the International Journal of Forecasting and Wind Energy. His main research interests are centered around the proposal and application of mathematical methods for electricity markets and power systems operations, including forecasting. He has published extensively in some of the leading journals in Meteorology, Power Systems Engineering, Statistics and Operations Research. He has been a visiting researcher at the University of Oxford (Mathematical Institute) and the University of Washington in Seattle (Dpt. of Statistics), as well as a scientist at the European Center for Medium-range Weather Forecasts (ECMWF, UK) and a visiting professor at Ecole Normale Superieure (Rennes, France). He is leading a number of initiatives aiming to profoundly rethink electricity markets for future renewable-based power systems and with a more proactive role of consumers. This focus on consumer-centric and community-driven electricity markets translates into proposals for peer-to-peer energy exchange, from mathematical framework to actual demonstration in Denmark.

5 October
at 10:30

Room: MZR

A new framework for cumulus parametrization

Speaker: Christian Jakob (Monash University, Australia)

Abstract

The representation of convection in climate model remains a major Achilles Heel in our pursuit of better predictions of global and regional climate. The basic principle underpinning the parametrization of tropical convection in global weather and climate models is that there exist discernible interactions between the resolved model scale and the parametrized cumulus scale. Furthermore, there must be at least some predictive power in the larger scales for the statistical behaviour on small scales for us to be able to formally close the parametrized equations.

The presentation will discuss a new framework for cumulus parametrization based on the idea of separating the prediction of cloud area from that of velocity. This idea is put into practice by combining an existing multi-scale stochastic cloud model with observations to arrive at the prediction of the area fraction for deep precipitating convection. Using mid-tropospheric humidity and vertical motion as predictors, the model is shown to reproduce the observed behaviour of both mean and variability of deep convective area fraction well. The framework allows for the inclusion of convective organisation and can - in principle - be made resolution-aware or resolution-independent.

When combined with simple assumptions about cloud-base vertical motion the model can be used as a closure assumption in any existing cumulus parametrization. Results of applying this idea in the the ECHAM model indicate significant improvements in the simulation of tropical variability, including but not limited to the MJO.

6 October
at 10:30

Room: MR1

GEOMETRIC: Geometry and Energetics of Ocean Mesoscale Eddies and Their Rectified Impact on Climate

Speaker: David Marshall (Clarendon Laboratory)

19 October
at 10:30

Room: LT

LDAS-Monde global capacity assimilation of satellite-derived vegetation and soil moisture products: analysis impact over North America

Speaker: Clement Albergel (Météo-France, France)

Abstract

In this study LDAS-Monde, a land data assimilation system with global capacity, is applied over North America to increase monitoring accuracy for land surface moisture energy and water states and fluxes, including evapotranspiration and stream flow as well as vegetation growth. LDAS-Monde ingests information from satellite-derived Surface Soil Moisture (SSM) and Leaf Area Index (LAI) estimates to constrain the Interactions between Soil, Biosphere, and Atmosphere (ISBA) land surface model (LSM) coupled with the CNRM (Centre National de Recherches Météorologiques) version of the Total Runoff Integrating Pathways (ISBA-CTRIP) continental hydrological system. LDAS-Monde uses the CO₂-responsive version of ISBA which models leaf-scale physiological processes and plant growth, while transfer of water and heat through the soil rely on a multilayer diffusion scheme. SSM and LAI estimates are assimilated using a Simplified Extended Kalman Filter (SEKF), which uses finite differences from perturbed simulations to generate flow-dependence between the observations and the model control variables (LAI and seven layers of soil: from 1 cm to 100 cm depth).

LDAS-Monde analysis impact over 2007-2016 is assessed over North America using satellite-driven model estimates of land evapotranspiration from the Global Land Evaporation Amsterdam Model (GLEAM) project as well as upscaled ground-based observations of gross primary productivity from the FLUXCOM project. Over the US, in-situ measurements of soil moisture from the USCRN network and of turbulent heat fluxes and GPP from FLUXNET-2015 are used in the evaluation, together with river discharges from the USGS. Those data sets highlight the added value of LDAS-Monde compared to an open-loop simulation (i.e. no assimilation). Finally, it has been shown that LDAS-Monde has a strong ability to monitor agricultural drought, by providing improved initial conditions which persist through time, aiding agricultural drought prediction.

24 October
at 10:30

Room: MR1

GNSS & ATOMMS RO: New Cubesat-Based Radio Occultation Systems for Weather & Climate

Speaker: E Robert Kursinski (Space Sciences & Engineering, Boulder, CO, USA)

Abstract

Reducing the uncertainty of weather and climate predictions requires better understanding of the hydrological and energy cycles and the intimate coupling and complex interplay between atmospheric water, temperature and stability, dynamics and short and long wave radiation. Orbiting radar and lidar measurements have greatly increased our understanding of aerosols, clouds and precipitation and Aeolus promises analogous advances on winds in the near future. Greatly increasing the quantity and information content of radio occultation (RO) profiling promises to provide analogous advances about temperature and water vapor as well as constraints on winds. When taken together, these combined, global, precise and high vertical resolution observations of temperature water vapor, winds, aerosols, clouds, precipitation and energy fluxes would better tie the entire weather-climate system together and yield more observationally constrained global analyses and forecast initialization, better process-related constraints to guide model improvements and reduced uncertainty about GCM realism and predictions.

The ability of GPS RO to profile atmospheric temperature and pressure with very high vertical resolution, precision and accuracy, particularly in the UTLS, is well established. With regard to water vapor, histograms of low latitude specific and relative humidity derived from COSMIC GPS RO data reveal the levels of uncertainty among global analyses and climate models. These results indicate that MERRA water vapor analyses are the best in the middle and upper troposphere while ECWMF (circa 2007) is better in the lower free troposphere. While present RO sampling densities are too sparse to impact global moisture analyses much, greatly increasing the GNSS RO sampling densities via cubesats would more tightly constrain and improve the analyses. Our cubesat-based GNSS RO constellation, designed to provide several tens of thousands of occultations per day with performance approaching that of COSMIC-2, is scheduled to begin launching at the end of 2018.

A significant advance beyond GPS RO is achievable. We have been developing the Active Temperature Ozone and Moisture Microwave Spectrometer (ATOMMS) RO system to probe the atmosphere near the 22 & 183 GHz absorption lines of water and 169 & 184 GHz lines of ozone. By profiling both the speed and absorption of light, ATOMMS will profile water vapor, temperature and pressure simultaneously, which GNSS RO cannot do, from near the surface to the mesopause, with random uncertainties of 1-3% for water vapor and 0.4K for temperature and still better absolute uncertainty over most of the altitude range, with 100 m vertical resolution. Performance in clouds should be within a factor of two of clear air performance. This orbital profiling performance approaches that of radiosondes, but with far better accuracy via RO’s inherent self-calibration.

With funding from NSF, we developed a prototype ATOMMS instrument and demonstrated several key ATOMMS capabilities on the ground including retrieving water vapor to better than 1% in optical depths up to 17 through clear, cloudy and rainy conditions.

While conceived for climate, miniaturization of the ATOMMS instrument and cubesat technology have made NWP sampling densities feasible. A 60 satellite constellation delivering 25,000 ATOMMS and 170,000 GNSS RO occultations each day, providing complete global coverage every 6 hours, could be implemented for a fraction of the cost of a JPSS satellite. Such a constellation could be implemented in stages, beginning perhaps with 4-6 ATOMMS satellites focused on polar science. This would then be expanded over time via additional satellites with the number of ATOMMS occultations increasing in proportion to the square of the number of satellites.

An ATOMMS observation simulation study would help quantify the impact of large numbers of ATOMMS profiles on NWP analyses and forecasts and move this concept forward.

30 October
at 10:30

Room: MR1

On the Nexus between Carbon Cycle and Air Quality: Exploring Multiple Constraints on Anthropogenic Combustion and Fires

Speaker: Ave Arellano
(University of Arizona, LATMOS/UPMC)

Abstract

It is imperative that we provide more accurate and consistent analysis of anthropogenic pollution emissions at scales that is relevant to air quality, energy, and environmental policy.

Here, we present three proof-of-concept studies that explore observational constraints from ground, aircraft, and satellite-derived measurements of atmospheric composition on bulk characteristics of anthropogenic combustion in megacities and fire regions. We focus on jointly analyzing co-emitted combustion products such as CO2, NO2, CO, SO2, and aerosols from GOSAT, OCO-2, OMI, MOPITT, and MODIS retrievals, in conjunction with USEPA AQS and NASA field campaigns. Each of these constituents exhibit distinct atmospheric signatures that depend on fuel type, combustion technology, process, practices and regulatory policies. Our results show that distinguishable patterns and relationships between the increases in concentrations across the megacity or large fire events due to emissions of these constituents enable us to: a) identify trends in combustion activity and efficiency, and b) reconcile discrepancies between state- to country-based emission inventories and modeled concentrations of these constituents. For example, the trends in enhancement ratios of these species reveal combustion emission pathways for China and United States that are not captured by current emission inventories and chemical reanalysis. Analysis of their joint distributions has considerable potential utility in current and future integrated constituent data assimilation and inverse modeling activities like CAMS for monitoring, verifying, and reporting emissions, particularly for regions with few observations and limited information on local combustion processes. Our targeted evaluation of the global forecast and analysis of CAMS CO and CO2 during KORUS-AQ field campaign in May 2016 suggests that CAMS is able to capture the contrast in combustion efficiency between Chinese outflow and Seoul. Analyses of MOPITT and IASI XCO as well as GOSAT XCO2 column retrievals in CAMS better capture the variability in Seoul but not in the Chinese outflow suggesting insufficient constraints over this region. This work also motivates the need for continuous and preferably collocated satellite measurements of atmospheric composition (including CH4) and studies related to improving the applicability and integration of these observations with ground- and aircraft-based measurements.

16 November
at 15:30

Room: LT

Medium-range forecasts with a non-hydrostatic global atmospheric model on a cubed sphere grid

Speaker: Song-You Hong (KIAPS, Seoul, Korea)

Abstract

Korea Institute of Atmospheric Prediction Systems (KIAPS), Seoul, Korea, has embarked a national project in developing a new global forecast system in 2011. The ultimate goal of this 9-year project is to replace the current operational model at Korea Meteorological Administration (KMA), which was adopted from the United Kingdom’s Meteorological Office’s model. Since July 2015, the test version of the Korean Integrated Model (KIM) system that consists of a spectral element non-hydrostatic dynamical core on a cubed sphere and a revised physics package has been running in a real-time testbed. In 2017, an updated KIM with the advanced 4-DEnvar at about 12-km has been launched. Its performance and operational deployment schedule will be presented, together with a special focus on the aerosol indirect effects on the medium-range forecasts.

21 November at 15:30

Room: LT

Evolutions and developments of global data assimilation at Météo-France

Speaker: Loïk Berre and Gérald Desroziers

Abstract

The global data assimilation at Meteo-France is based on the IFS/ARPEGE model. The current and next configurations of the deterministic and ensemble components of this system will be briefly reminded. This includes a specific normalisation of the wavelet correlation model, which has been recently implemented.

As the error covariance specification plays a crucial role in the assimilation system, a diagnostic study of error contributions in data assimilation cycling is being conducted. This is based on a linear expansion of forecast error as function of contributions with specific ages (i.e. either old, or recent) and of different types (observations, model, background). Specific perturbations are then activated in the Ensemble Data Assimilation (EDA) system, in order to diagnose error contributions from old background states, and also from recent observations. The contribution of recent model errors is then diagnosed by comparing information provided respectively by a specific EDA experiment and by innovation-based diagnostics.

The growing role of ensembles also motivates the development of the 4DEnVar formulation, which has received increasing interest during recent years. From the beginning, Meteo-France has made the choice to implement this new 4DEnVar scheme within the OOPS (Object Oriented Prediction System) framework, initiated at ECMWF. In particular, with such a system, the background error covariance matrix is an object that can take different forms.

Unlike most implementations of 4DEnVar, the formulation developed at Meteo-France relies on the use of 4D state control variables instead of so-called alpha control variables. It also enables an observation space control variable to be used. Moreover, the use of the 4D state as control variables makes easier the formulation of hybrid 4DEnVar. In this case, the 4D state control variable remains indeed unchanged, while the background error covariance matrix simply becomes a linear combination of ensemble localised covariances and wavelet-based covariances.

Several developments related to 4DEnVar will be presented, such as the diagnosis of horizontal and vertical localisations, and the advection of localisation in order to account for dynamical effects. With the intention of using a common localisation for the different variables, and also in order to account for anisotropic aspects of the mass/wind cross-covariances, scale transformations are applied to wind and surface pressure. A parallelized version of the EDA component of 4DEnVar will also be presented, together with preliminary results.

22 November at 13:30

Room: LT

Statistical post-processing of ensemble forecasts for wind speed

Speaker: Sándor Baran (University of Debrecen, Hungary)

Abstract

Ensemble forecasts often show an underdispersive character and may also be biased, so that some post-processing is needed to account for these deficiencies. We present different versions of two popular methods for calibration of wind speed forecasts, namely the Bayesian model averaging and the ensemble model output statistics (EMOS). Both approaches provide estimates of the probability density functions of the predictable weather quantities, where model parameters are estimated using forecast ensembles and validating observations from a given rolling training period. The presented models are applied to wind speed forecasts of the fifty-member ECMWF ensemble and the eleven-member ALADIN-HUNEPS ensemble of the Hungarian Meteorological Service. The results indicate improved calibration of probabilistic and accuracy of point forecasts in comparison to the raw ensemble and to the climatological forecasts.

We also show two semi-local methods for estimating the EMOS coefficients where the training data for a specific observation station are augmented with corresponding forecast cases from stations with similar characteristics. Similarities between stations are determined using either distance functions or clustering based on various features of the climatology, forecast errors, and locations of the observation stations. In a case study on wind speed over Europe with forecasts from the Grand Limited Area Model Ensemble Prediction System, the proposed similarity-based semi-local models show significant improvement in predictive performance compared to standard regional and local estimation methods. They further allow for estimating complex models without numerical stability issues and are computationally more efficient than local parametrization.

30 November at 10:30

Room: LT

Dynamically Consistent Parameterization of Mesoscale Eddies

Speaker: Pavel Berloff (Imperial College)

Abstract

This work aims at developing a new approach for parametrizing mesoscale eddy effects for use in non-eddy-resolving ocean circulation models. The idea is to approximate transient eddy flux divergence in a simple way, to find its actual dynamical footprints, and to relate these footprints to large-scale flow properties.

7 December at 10:30

Room: LT

NOAA Center for Weather and Climate: Data, Information and Services

Speaker: Michael Tanner (NOAA Center for Weather and Climate)

Abstract

The NOAA Center for Weather and Climate (CWC) within the National Centers for Environmental Information (NCEI) is the official data management entity for climatological, oceanographic, and geophysical information in the United States.

CWC performs data synthesis, and data and information dissemination to promote the scientific integrity and usefulness of its products and services. As a part of this responsibility, CWC analyses long-term environmental trends through monitoring and assessment at multiple temporal and spatial scales. CWC also ensures scientific data stewardship of its holdings, including both remotely- and directly-sensed climate and weather, oceanographic, and geophysical data records including related and derived environmental information. CWC also performs quality assurance and reanalysis of historical data to establish and update baseline data sets or standards for global and national environmental monitoring.

CWC advances and enables weather and climate science and decision-making through the acquisition, monitoring, analysis, synthesis, and delivery of observations, derived products and assessments, and information and outreach services. CWC engages across the US, international organizations, the private and public sectors to promote awareness, understanding, access and use of its data, information and capabilities to ensure that all products and services meet user requirements.

11 December at 14:00

Room: LT

Improved winter Euro-Atlantic atmospheric blocking in high-resolution simulations with EC-Earth

Speaker: Paolo Davini (ISAC-CNR, Torino, Italy)

Abstract

The numerical simulation of atmospheric blocking, in particular over the Euro-Atlantic region, still represents a main concern for the climate modeling community. Here we discuss winter atmospheric blocking representation in a set of atmosphere-only climate simulations produced with EC-Earth during the Climate SPHINX PRACE project. This includes 30-year long runs at five different horizontal resolutions (from T159 to T1279) with several ensemble members (up to 10).

Results show that the usual negative bias in blocking frequency over Europe becomes negligible at T511 and T799. A combined effect by the more resolved orography and by a change in tropical precipitation is identified as the source of an upper tropospheric planetary wave. At the same time, a weakening of the meridional temperature gradient reduces the upper level baroclinicity and the zonal mean winds. Following these changes, in the high resolution configurations the Atlantic eddy-driven jet stream is weakened favoring the breaking of synoptic Rossby waves over the Atlantic ridge and thus increasing the simulated European blocking frequency.

However, at high-resolution the Atlantic jet stream is too weak and the blocking duration is still underestimated. This suggests that the optimal blocking frequencies are achieved through compensation of errors between eddies found at upper levels (too strong) and eddies at lower levels (too weak).

2016

7 January
at 10:30

Room: MR1

Fire and haze in Indonesia: causes, impacts and predictability

Speaker: Dr Robert Field (NASA, USA)

Abstract

The 2015 fire season in Indonesia began in August, following the start of the dry season in July. By September, much of Sumatra, Kalimantan, Singapore, and parts of Malaysia and Thailand were covered in thick smoke, affecting the respiratory health of millions of people. At its peak, visibility was reduced to less than 10% of normal in places, and large parts of Borneo could not be seen from space. Preliminary estimates suggest that greenhouse gas emissions from the burning exceeded those of Japan’s mean annual fossil fuel emissions. Even after the worst of the fire was over, the remnant pollution stretched halfway around the equator. It was an environmental and public health catastrophe, made all the worse by the fact that we knew of its potential several months earlier.

2015 was a repeat of events that have occurred in Kalimantan since the 1980s and in Sumatra since at least the 1960s. Fires set to clear land and remove agricultural residues escape into degraded peat soil, which can become flammable during stronger-than-normal dry seasons. Once the fire is underground, it is extremely difficult to extinguish, and can burn continuously until the return of the monsoon rains. Research since the 1997/98 haze disaster has given us a reliable understanding of how much fire and haze will occur for a given drought strength, which is strongly influenced by large-scale climate patterns such as El Niño. Ongoing research into the socio-economic drivers of the fire is beginning to identify the roles of landholders large and small. Solving the problem will ultimately involve eliminating fire from degraded peatlands, which, long-term, will need to be re-wetted and re-habilitated, and done in the context of difficult issues surrounding the governance of Indonesia’s natural wealth and the need for economic development.

In the short-term, however, an urgent priority is to develop systematic responses to medium and long-term weather forecasts. Prevention and pre-preparedness measures need to be triggered when forecasts begin locking in on dangerously strong dry seasons, even if the forecasts are issued when it is still wet. This is not unrealistic. Basic fire danger rating systems are now established in Indonesia and are used to monitor for dry conditions. There was also meaningful progress in Indonesia’s operational response this year, including the acceptance of international fire-fighting assistance. The problem is that burning occurs opportunistically, and as soon as conditions are dry enough. The operational response was too late, and would have been far more effective if action were taken in response to the forecast, with a strong focus on prevention, rather than to the crisis. I argue that early warning triggers tied to medium and long-term rainfall forecasts, such as those produced at the ECMWF, are central to reducing severe fire and haze in Indonesia.

About Robert...

Robert Field is a Columbia University Associate Research Scientist at the NASA Goddard Institute for Space Studies. He works with the GISS climate model, specializing in the water cycle, and the cause, fate, and effects of emissions from biomass burning. From 2000-2004, he was part of a joint project between the governments of Indonesia, Malaysia and Canada to develop fire danger rating systems in the region. He has a PhD in atmospheric physics from the University of Toronto.

20 January
at 10:30

Room: LT

NASA Science: Observing Earth, Observing Other Earths

Speaker: Dr Ellen Stofan (NASA, USA)

Abstract

NASA science pushes the boundaries of our knowledge about Earth, the Solar System including our Sun, and the universe. Our studies of Earth are helping us to better monitor and model our changing climate, and to help increase climate resilience around the world. NASA looks beyond Earth to explore the planets of our solar system, expanding our understanding of planetary climates and addressing the question- Are we alone? Telescopes such as Kepler have shown that planetary systems are common; the next generation of telescopes will be used to begin studying the atmospheres of planets around other stars.

21 January
at 10:30

Room: LT

From parameterising to resolving: the role of eddies and resolution in coupled models

Speaker: Helene Hewitt (Met Office Hadley Centre, UK)

Abstract

As the resolution of coupled models has increased, the representation of eddies in the ocean component has changed from being parameterised to resolved. We show results from coupled models that span the eddy regime and discuss why high resolution and explicit representation of eddies might be important across the full range of prediction timescales. In particular, recent results from a version of HadGEM3 with a 1/12 degree ocean suggest that ocean eddies and resolution may play an important role in accurately simulating the climate system. Future plans around traceability and experiment design to assess the role of ocean resolution are discussed.

26 January
at 14:00

Room: LT

Empowering drought management in the Middle East and North Africa through timely data provision

Speaker: Dr Rachael McDonnell (ICBA)

Abstract

Drought is a constant presence in the Middle East and North Africa (MENA) region. Current droughts reveal the gaps and limitations in drought management in the region. Drought conditions in the southern Levant in January and February 2014 once again emphasized the urgent need to support the governments and people in these countries in developing efforts to manage the impacts of these extreme events. This need is further heightened by any analysis of future climate conditions in the region.

The talk will highlight how modeling and data generation will be used to support decision-makers in the region to try to mitigate some of the impacts. The Middle East and North Africa-Regional Drought Management System (MENA RDMS) will harness the enormous innovations from the ongoing data revolution to bring new climate/water/food data as part of monitoring and early warning systems. These are vital components in implementing preparedness and mitigation measures. It will serve the region by:

Establishing a regional drought monitoring and early warning system and associated information delivery systems
Providing assessment of drought vulnerabilities and impacts
Assist officials who are charge with relief efforts by providing “value-added” information during drought events.

2 February
at 10:30

Room: LT

IFS refactoring for OOPS

Speakers: M. Fisher, A. Geer, P. Lean, O. Marsden, D. Salmond, Y. Tremolet & T. Wilhelmsson (ECMWF, UK)

Abstract

Work on OOPS has progressed and related refactoring has taken place in recent months in many parts of the IFS code. We will present the refactoring work and give an overview of the changes in the code that have happened and that can be expected in the coming cycles. We will focus on changes in the existing IFS code rather than OOPS itself.

The following aspects will be presented:

Introduction
Geometry and Model objects, SPAM
Trajectory handling
Jb, background and first guess
ODB interface
GOM and HOP interfaces
Upcoming IFS cycles

8 February
at 10:30

Room: LT

Strategies for seasonal forecasting with equatorial bias correction

Speaker: David Mulholland (University of Reading, UK)

Abstract

Seasonal forecast skill depends crucially on the accurate and balanced initialisation of the ocean, and this is complicated by the need to use a bias correction scheme during ocean data assimilation to deal with errors in wind stress forcing. It has been found that this initialisation method, as used in the current operational seasonal forecast system, leads to an initialisation shock when the bias correction term is removed, which introduces spurious variability into the tropical thermocline and appears to degrade forecasts of sea surface temperature (SST). Alternative initialisation methods have been tested using sets of 7-month forecasts, including one in which the bias correction term is slowly removed during the forecasts, rather than being removed instantaneously at t=0. This essentially removes the initialisation shock, and results in improved SST forecast skill as measured in several key regions.

29 February
at 10:30

Room: MR1

Better observations, better forecasts?

Speaker: Francois Massonnet (BSC, formerly IC3, Barcelona)

8 March
at 14:00

Room: MZR

Spectral/hp element discretisations for Massively Parallel Computing

Speaker: Prof Spencer Sherwin (Imperial College, London)

Abstract

Compact high order approximations using spectral/hp element discretisations combine the desirable approximation properties of spectral methods with the complex geometry capabilities of finite volume or finite element methods. However an additional advantage of these discretisations is the alignment of their computational footprint with the emerging hardware where computational speed is much dictated by memory movement as floating points operations. Over the past ten years we have been developing an open-ware spectral/hp element framework, Nektar++ (www.nektar.info) to broaden the application of these methods and facilitate their utility. Part of the framework has permits us to allowed to compare continuous Galerkin projection (CG) used in classic finite element methods with Discontinuous projections such as the Discontinuous Galerkin (DG) or Flux Reconstruction methods. In this presentation we will outline our studies comparing CG and DG implementations for the type of elliptic operators we adopt in our semi-implicit approximation of the incompressible Navier-Stokes equations. We will then outline our current developments under the EU exascale project, ExaFlow (http://exaflow-project.eu/) , to combine the favourable properties of the CG and DG approximations to adapt the projections to hardware communication hierarchies. We will then outline our current application of these technique to high Reynolds number aerodynamics prediction related to Formula One cars and emerging application such as shallow water formulations. www.sherwinlab.info

10 March
at 10:30

Room: CC

Microphysics Matters? A Global Perspective

Speaker: Dr Richard Forbes (ECMWF, UK)

Abstract

Microphysics deals with molecular scales, driven at the "cloud" scales of turbulent motions and atmospheric dynamics, but with impacts that are global. How do we reconcile these different scales? How do we know how much of the complexity of the real microphysical world we have to observe, understand and represent in our global models? I will discuss these issues in the context of parametrization in GCMs and then focus on one example of cloud over the Southern Ocean in the IFS, where most global NWP and climate models have large systematic errors. Understanding the source of this error in terms of microphysics has implications for utilising data assimilation for model error diagnosis, for development of model parametrizations, and for reducing a regime-dependent systematic error with potentially far-reaching global consequences relating to hemispheric albedo and prediction of cross-equatorial transport.

22 March
at 10:30

Room: LT

Forecast Verification - Decision Support Analytics

Speaker: Dr Betsy Weatherhead (Univ. Colorado, Boulder, USA)

Abstract

Changes in models, observations or even computing approaches are all likely to result in changes to forecasts. When the forecast change is small, identifying if it is an improvement can be challenging and expensive. Decision support statistics can help identify even very small improvements. By making use of paired forecasts, and respecting the day-to-day and even hour-to-hour autocorrelation in weather, forecasts and forecast errors, identification can be made more efficiently and with higher likelihood of long-term success.

Perhaps just as importantly, statistical input can help design evaluation runs to minimize cost while maximizing the power of the results. For instance, the number of runs can be reduced significantly, if forecasting evaluation techniques are determined in advance. These decision support techniques can be used to determine if an added set of observations is significant, whether a computer change is systematically more harmful, or if the impact of small changes to physics packages or model cores will have a long-term positive impact. Techniques will be reviewed and sample results shared.

Dr. Betsy Weatherhead is a senior scientist at U. Colorado at Boulder. She has worked on issues as diverse as climate change, ozone depletion, radiative transfer and indigenous reports of climate change.

Dr. Weatherhead is proud to share a number of awards including the 2007 Nobel Peace Prize for her contributions to the Intergovernmental Panel on Climate Change. She often works on topics in atmospheric science that require high level statistical or technical expertise, including developing techniques for evaluating small differences in forecasting skill. Her work on detecting changes in ozone culminated in the cover story on ozone recovery for Nature in 2006. In the past few years she has been heavily involved in AMS, chairing the AMS Summer Community Meeting in 2011-2014 and founding the AMS Forecast Improvement Group (FIG). She is currently teaching a new graduate course that she developed in environmental statistics.

23 March
at 14:00

Room: MR1

Land surface model and assimilation system at JMA

Speaker: Kengo Miyaoka (JMA, Japan)

Abstract

In this seminar, I present the JMA land surface modelling and data assimilation activities. The Simple Biosphere (Sib) Model is used to represent the land surface processes. An Optimal Interpolation is used to analyse snow depth whereas the soil moisture analysis relies on a climatology. The land surface model and data assimilation systems for snow analysis are described, and activities for updating the model are presented. Ongoing research on ASCAT surface soil moisture data assimilation development is also briefly presented.

4 April
at 15:30

Room: MR1

Statistical Post-Processing of GEFS Ensemble Forecasts for Precipitation Accumulations

Speakers: Michael Scheuerer & Thomas Hamill (NOAA, USA)

Abstract

We present and compare two different methods for statistical post-processing of ensemble precipitation forecasts which are developed and demonstrated with GEFS precipitation reforecasts over the conterminous United States and verified against an 1/8-degree climatology-calibrated precipitation analyses.

The first approach is non-parametric and forms, for each location, an new ensemble from the analyzed precipitation amounts by identifying dates in the past that had reforecasts similar to today’s forecast. A variant of this method is presented that augments the data at each location by finding supplemental locations with similar characteristics (climatology, terrain, etc.). The resulting increase of training data will be shown to be particularly beneficial for the probabilistic prediction of rare events.

As a second approach we consider a parametric method that generates full predictive probability distributions for precipitation accumulations by fitting shifted, left-censored gamma distributions to statistics of the raw ensemble forecasts. This distribution type is shown to be adequate for modeling the distribution of analyzed precipitation accumulations given the ensemble forecasts both in situations with good predictability (e.g. at short lead times) and decreased predictability (e.g. at longer lead times or during summer months).

Probabilistic forecasts by both methods will be verified using common metrics (skill, reliability, and so forth). We study how the different ensemble statistics and non-linear components in the parametric approach contribute to its performance, and we discuss, for both methods, the effect of training sample size on the reliability and resolution of the post-processed predictions

6 May
at 11:15

Room: LT

Using advanced data assimilation to improve operational predictions from the NCEP Global Forecast System (GFS) model

Speaker: Daryl Kleist, (University of Maryland, USA)

Abstract

Data assimilation improvements have been a key component to the increased skill of the operational NCEP GFS model over the past decade. Two significant improvements within the past five years have been related to the introduction of hybrid assimilation, where information from climatological error covariance estimates is combined with flow-dependent ensemble estimates. This talk will focus on the hybrid algorithm developed at NCEP with emphasis on the application of an ensemble-variational (EnVar) algorithm for the GFS including the extension to 4D EnVar. Results derived from both simulated and real observation experiments will be presented, including verification from the current real-time demonstration at NCEP.

Additionally, a discussion of the initialization of tropical cyclones in the operational GFS will be presented. The current operational GDAS/GFS utilizes a combination of vortex relocation, bogus wind assimilation, and assimilation of minimum sea-level pressure information driven by real-time information from operational tropical cyclone forecast agencies. While the mechanical vortex scheme has been demonstrated to reduce track errors in older versions of the GFS, recent sensitivity experiments show that for newer generations of the GFS the mechanical relocation may produce increased errors in track forecasts at longer lead times despite reducing the initial position uncertainty.

Experiments with the full resolution GFS will be utilized to explore the sensitivity on forecasts of Hurricane Joaquin (2015).

10 May
at 10:30

Room: LT

Significance of changes in forecast scores

Speaker: Alan Geer, (ECMWF, UK)

Abstract

The impact of developments in weather forecasting is measured using forecast verification, but many developments, though useful, have impacts of less than 0.5% (about 0.5h) on medium-range forecast scores. Chaotic variability in the quality of forecasts makes it hard to achieve statistical significance when comparing these developments to a control, and surprisingly large sample sizes may be required. By making an independent realisation of the null distribution used in the hypothesis testing, using 1,885 paired forecasts (about 2.5 years of testing), it is possible to experimentally test the validity of the normal statistical framework for forecast scores. This shows that the naive application of Student's-T can generate too many false results. A known issue is temporal autocorrelation but statistical multiplicity also needs to be considered. For example, across three forecast experiments, multiplicity causes a 1 in 2 chance of a false result. The t-test can be reliably used to interpret the significance of changes in forecast scores, but only when these effects are treated correctly. Given the large sample sizes required to achieve statistical significance in the medium-range, there is need to devise testing strategies to make the best use of scare supercomputer resources. Finally, it is sometimes hoped that short-range verification can avoid the need for medium-range verification, by making an assumption that short-range statistics are indeed an predictor of improvements at longer ranges. However, short-range verification has its own set of problems, which will be discussed briefly.

13 May
at 10:00

Room: MR1

Applications of equal area partitions of the unit sphere

Speaker: Paul Leopardi (Bureau of Meteorology, Australia)

Abstract

IFS uses an algorithm since 2007 for partitioning the sphere which was developed by our guest speaker Dr Paul Leopardi.

The influence of the spatial scale of initial-condition errors on atmospheric predictability

Speaker: Prof. Dale Durran (University of Washington, USA)

Abstract

One important limitation on the accuracy of weather forecasts is imposed by unavoidable errors in the specification of the atmosphere's initial state. Much theoretical concern has been focused on the limits to predictability imposed by small-scale errors, potentially even those on the scale of a butterfly. Very modest relative errors at much larger scales may nevertheless pose a more important practical limitation. We demonstrate the importance of large-scale uncertainty by analyzing ensembles of idealized simulations of mesoscale convective systems. We consider several environments with different low-level shears and pairs of ensembles with equal amplitude large- or small-scale perturbations in the surface moisture.

As foreshadowed by results obtained with a simple barotropic model in a largely overlooked section of Lorenz’s classic 1969 paper “The predictability of a flow which possesses many scales of motion,” equal-amplitude initial perturbations at wavelengths of 8 and 128 km produce identical losses of predictability after five hours of simulation. These results imply that minimizing initial errors on scales on the order of 100 km is at least as likely to extend the accuracy of forecasts at lead times longer than 4-5 hours than potentially expensive efforts to minimize initial errors on much smaller scales.

These simulations also demonstrate that convective systems, triggered in a horizontally homogeneous environment with no initial background circulations, can generate a background mesoscale kinetic energy spectrum with a slope proportional to the -5/3 power of the wave number, similar to that observed in the atmosphere. The horizontally divergent and rotational parts of the kinetic energy spectrum are examined along with their relative contributions to the -5/3 spectrum.

3 November
at 10.30

Room: MR1

Weakly or strongly nonlinear mesoscale dynamics?

Speaker: Erik Lindborg (Royal Insitute of Technology, Sweden)

Abstract

It has been shown that wave number spectra at atmospheric mesoscales (length scales from ten to a couple of hundred km) display an approximate minus five-third third power law, just as small scale turbulence. Recently, it has been debated whether the spectra are generated by a strongly nonlinear downscale energy cascade or by weakly nonlinear inertia-gravity waves. To shed light on this issue investigators have used aircraft data to estimate the ratio, R, between divergent and rotational energy. In this talk I will argue that a necessary condition for weak nonlinearity is that the group velocity is much larger than a characteristic fluid velocity. This leads to a condition on the Rossby number which is not consistent with observations of the ratio R. On the other hand, in order for the dynamics to be defined as strongly nonlinear, I suggest that it is sufficient to demonstrate that the kinetic and potential energy spectra,are of Kolmogorov type. Using data from a huge number of aircraft segments from the MOZAIC data set kinetic energy and temperature spectra are calculated in the upper troposphere and stratosphere, using a criterion based on ozone concentration to distinguish between stratospheric and tropospheric segments. We find that all spectra, in particular the stratospheric spectra, display a clean minus five third power law at mesoscales. Making a Helmholtz decomposition, we show that R is slightly larger that unity in the stratosphere while it is smaller than unity in the troposphere at mesoscales. We also show that it is necessary to use of the order of one thousand flight segments to obtain reasonably converged results for R. Based on these observations we argue that the dynamics is not weakly nonlinear, but is governed by strongly nonlinear interactions.

11 November at 10.30

Room: LT

MJO teleconnections in the S2S database

Speaker: Frederic Vitart (ECMWF, UK)

Abstract

The World Weather Research Program (WWRP) and World Climate Research Program (WCRP) launched a joint research initiative in 2013, the Sub-seasonal to Seasonal prediction project (S2S) whose main goal is to improve forecast skill and understanding of the sub-seasonal to seasonal time-scale and to promote its uptake by operational centres and exploitation by the application communities. A main deliverable of this project is the establishment of an extensive database containing sub-seasonal (up to 60 days) near real-time forecasts (3-weeks behind real-time) and re-forecasts. This research database, hosted at ECMWF and CMA, is now available to the research community from the ECMWF and CMA data portals.

This presentation will describe the S2S project and show some preliminary results on the ability of the S2S models to predict the Madden Julian Oscillation and simulate its teleconnections in the Extratropics.

14 November
at 10.45

Room: LT

Lightning at ECMWF

Speaker: Philippe Lopez (ECMWF, UK)

Abstract

Lightning is a severe weather phenomenon known to affect numerous human activities, in particular through air traffic disruptions, power supply outages, damages to building and the triggering of wildfires. Lightning can also occasionally lead to injuries or even fatalities. A parameterization of lightning has been developed to diagnose lightning flash densities from convective hydrometeor contents, CAPE and convective cloud base height output from the ECMWF convection scheme. Its calibration and validation using LIS/OTD satellite lightning climatology and its comparison against existing parameterizations in 10-year-long 80km resolution simulations will be presented. Examples of applications in higher resolution deterministic forecasts as well as in the ensemble prediction system will then be given. Finally, the sensitivities of lightning (over a target region) to meteorological variables 24 hours earlier, as computed using the adjoint of the lightning parameterization, will be discussed. This will pave the way to future work towards the assimilation of lightning observations, especially from the next-generation of geostationary satellites.

16 November at 10.30

Room: LT

On stochastic representations of model uncertainties

Speaker: Martin Leutbecher (ECMWF, UK)

Abstract

Members in ensemble forecasts differ due to the representations of initial uncertainties and model uncertainties. The inclusion of stochastic schemes to represent model uncertainties has improved the probabilistic skill of the ECMWF ensemble by increasing reliability and reducing the error of the ensemble mean. Recent progress, challenges and future plans concerning stochastic representations of model uncertainties at ECMWF are described. The coming years are likely to see a further increase in the use of ensemble methods in forecasts and assimilation. This will put increasing demands on the methods used to perturb the forecast model. An area that is receiving a greater attention than 5 to 10 years ago is the physical consistency of the perturbations. Other areas where future efforts will be directed are the expansion of uncertainty representations to the dynamical core and to other components of the Earth system as well as the overall computational efficiency of representing model uncertainty.

29 November at 14.00

Room: LT

Semi-Lagrangian semi-implicit finite-difference dynamical core of the SLAV model

Speaker: Mikhail Tolstykh (INM/RAS & Hydrometcentre of Russia, Russia)

Abstract

The semi-Lagrangian semi-implicit finite-difference dynamical core of SLAV global atmosphere model is presented. Its distinct features are the use of vertical component of the absolute vorticity and the horizontal divergence as prognostic variables, the unstaggered grid, and high-order finite-differences. The dynamical core can use reduced lat-lon grid and the variable resolution in latitude. The algorithm for wind velocity reconstruction (Tolstykh, Shashkin, J Comput Phys 2012) avoids solving Poisson equations on the sphere.

Currently, the SLAV configuration with approximately 13km horizontal resolution can use up to 9000 cores with efficiency more than 50%. So far, we use Fast Fourier transforms (FFT) in some parts of the dynamical core. Hence data transpositions are needed in parallel implementation. We work on scalable iterative grid-point solvers to elliptic problems. Also, we have successfully tested computation of meridional derivatives at the reduced grid in the grid-point space. These developments will help to avoid the use of FFT in the model code thus moving scalability limit further.

2015

3 February at 10:30

Room: LT

Solar Radiation Forecasting for the Energy Industry: The Australian experience

Speaker: Alberto Troccoli (Commonwealth Scientific and Industrial Organisation (CSIRO))

Abstract

Prediction of solar radiation is key in sectors such as solar power and agriculture; for instance, it can enable more efficient production of energy from solar power plants. Specifically, select ECMWF model output is a key input to the Australian Solar Energy Forecasting System (ASEFS), an operational system run by the the Australian Energy Market Operator.

Abstract

The intensification of tropical cyclones is often poorly forecast by NWP models. Tropical cyclones intensify as the result of the conversion of available potential energy (APE) into kinetic energy. This deceptively simple definition is hard to use in practice, however, because of the longstanding difficulty to define and quantify available potential energy for a moist atmosphere. In this talk, I'll review the origins of this difficulty and propose a new APE framework that circumvents most of the previous difficulties. Its main new attributes are: 1) it is based in terms of the locally-defined positive definite APE density, which allows for the study of local energy budgets; 2) the reference state does not need to be obtained by an adiabatic re-arrangement of the fluid parcels; 3) it defines separate thermodynamic efficiencies for thermal and moist effects; 4) it cleanly separates boundary sources of APE from interior sinks/sources due to irreversible thermodynamic processes. Some of the concepts will be illustrated in the context of a realistic simulation of hurricane MEGI, and a strategy for identifying weaknesses in model simulations outlined.

Tropical cyclones in a hierarchy of high-resolution GCMs

Speaker: Pier Luigi Vidale, (University of Reading)

Abstract

Over the last few years we have been developing global models, based on the HadGEM family, to study the impact of resolution on the simulation of the climate system. As the resolution is increased, our focus has been on the emergence of processes, and their interaction with the rest of the climate system. While our tropical cyclone (TC) climatology indicates that the GCMs simulate these phenomena with increasing accuracy in terms of track densities and variability, TC intensity continues to be underestimated. We will show some recent examples from investigations with cloud-system resolving (CSR) models applied in limited area mode, as well as (coarse) global. Preliminary results indicate how, for the Unified Model, a grid spacing of 4km starts to mitigate the problem of biases in TC intensity.

Equally, there is increasing need for data sets that allow robust assessment of climate model performance, but TC observations are scarce and inhomogeneous. We have been collecting and processing a number of re-analysis products in order to understand uncertainty in TC frequency, track density and landfall. Results indicate that the re-analyses are in better agreement for more recent times, which points to the value of data assimilation.

28 April at
14:00

Room: LT

Tropical Cyclone Predictions from the Met Office Global Model: a Brief History and Recent Developments

Speaker: Julian Heming, (Met Office, Exeter)

Abstract

This presentation will firstly provide an overview of tropical cyclone predictions from the Met Office Global Model with an emphasis on the impact of the initialisation procedure used between 1994 and 2012.

A major upgrade to the Met Office Global Model was implemented in July 2014. This included a major revision to the model’s dynamical core, improved physics and increased horizontal resolution to about 17 km at mid-latitudes. The impact of these changes on both tropical cyclone track and intensity predictions will be presented.

Finally, results will be presented on the impact of a new scheme which became operational in February 2015 to assimilate tropical cyclone warning centre estimates of central pressure in the Met Office Global Model.

8 June at
10:30

Room: MR1

Surface analysis at DWD - present state and future developments

Speaker: Martin Lange, (DWD)

Abstract

The NWP system at DWD is based on three operational models: ICON, COSMO-EU, and COSMO-DE for the respective global, European and German domain, each with an own assimilation cycle. Current development activities include the ensemble data assimilation both for the global and for convective scale, with a hybrid VarEnKF system for ICON and an ensemble Kalman filter (KENDA) for COSMO. Further, the replacement of COSMO-EU by a nested area over Europe is under intense preparation, and a tile model for ICON is scheduled for operational use in the middle of 2015. Surface analysis is an integral part of all these systems and development steps.

At DWD, the surface analysis is build up by a heterogeneous system of assimilation methods for snow, SST, screen level parameter and soil moisture even changing with the different model suites. Snow and SST analysis make use of an old fashioned Cressman method with blending of external data from NCEP, near surface parameter are analysed with an OI in the global, and a nudging scheme in COSMO-EU. For soil moisture analysis the 2D var method after Hess (2001) is used in C-EU, and a parameterized variant in the ICON analysis which circumvents the additional forecast runs required to determine the Jacobians.

One further aspect of future development at DWD is the use of satellite derived soil moisture data which, besides tuning large scale soil water content itself, should also help to improve surface emissivity that is of general importance for the assimilation of satellite data.

The talk will give an overview about the present system and development activities either integrating the existing surface assimilation methods into ensemble data assimilation or using ensemble Filters also for surface related variables.

12 June at 14:00

Room: LT

Predictability of tropical cyclone tracks: a multi-model multi-analysis approach

Speaker: Takeshi Enomoto (Kyoto University/Application
Laboratory, Japan Agency for Marine-Earth Science and Technology, Japan)

Abstract

Track error of the forecast of tropical cyclones of major operational centres have been reduced to approximately 100 km per day on the average in recent years. For a particular case, however, some or all forecasts from the operational centres may still fail to predict the observed tracks. We examine the relative contribution of the analysis error and model uncertainty by conducting forecast experiments from multiple analyses using multiple numerical weather prediction and climate models. Case studies for a few tropical cyclones indicate that the ambient flow near the storm centre is of primary importance although the distribution of diabatic heating may have significant influence.

24 June at
14:00

Room: LT

A New Paradigm of ENSO's Impact on Asia Monsoon

Speaker: Prof Fei-Fei Jin (Department of Meteorology
School of Ocean & Earth Science & Technology
University of Hawaii at Manoa)

Abstract

Several mechanisms have been proposed to explain how ENSO may influence the Asian monsoon. We propose that a significant part of impacts of ENSO on the Asian monsoon is through a previously overlooked ENSO combination mode of Indo-Pacific climate variability generated by the interaction between the Western Pacific warm pool seasonal cycle and ENSO. This combination mode (C-mode) is characterized by near-annual and sub-annual timescales (combination tones) and contributes substantially to the observed low-level atmospheric circulation and precipitation anomalies in West-North Pacific (WNP) region. In fact, this C-mode is a part of more general ENSO frequency cascade as the result of the nonlinear interaction of ENSO with the annual cycle. This cascade effective transfer from the ENSO power in its interannual band to higher frequencies (combination tones) so to generate ENSO responses not only characterized by different timescales but also by unique circulation and rainfall patterns. This new dynamic theory provides insight into ENSO induced WNP circulation variability and thus ENSO's impact on Asian monsoon system. Moreover, the associated subannual/interannual frequency climate variability is deterministic in its nature and hence as potentially predictable as ENSO.

25 June at
10:30

Room: LT

Improving weather forecasts by including land-surface model parameter uncertainty?

René Orth (ETH, Zürich)

Abstract

The land surface forms an important component of any Earth system model in general and therefore also of a numerical weather prediction system. The hydrology of the land surface is extremely complex and highly uncertain. In this talk I will demonstrate the importance of recognizing land surface uncertainty parameterization and its impact on predicting future weather.

Focusing on the land-surface model HTESSEL we present in this study a novel methodology for a comprehensive calibration using multiple observational datasets in Europe. We select 6 poorly constrained parameters which we vary using multiplicative factors. To explore the entire parameter space we perform a large number of uncoupled simulations with different combinations of parameter perturbations. Therefrom we derive the sensitivity of HTESSELs performance to individual parameters and identify several best-performing parameter perturbations. We discuss our findings within the context of the current default parameter set.

Furthermore we carried out several sub-seasonal coupled forecasts using the current ECMWF ensemble prediction system, which includes HTESSEL. In this context we employ the best-performing versus additional, randomly chosen parameter perturbations. We find better temperature and precipitation forecast skill with the best-performing parameter perturbations. This shows that a calibration using uncoupled (and hence less computationally demanding) simulations may lead to improvements in a coupled model. This is remarkable since the sensitivity of the forecast skills to individual HTESSEL parameters differs from the sensitivities computed from the uncoupled simulations.

JMA's New Seasonal Prediction System
(JMA/MRI-CPS2)

Speaker: Yuhei Takaya (JMA, Japan)

Abstract

A new version of JMA’s Seasonal Ensemble Prediction System (JMA/MRI-CPS2) was implemented to JMA's operational suite in June 2015. The system is used to produce three-month, warm/cold season and El Niño predictions. I will present an overview of the JMA/MRI-CPS2 and improvements with changes made in the new version. The changes include improved resolution and physics in atmospheric and oceanic components and the introduction of an interactive sea ice model. Verification of reforecasts shows that JMA/MRI-CPS2 has higher predictive skill than JMA/MRI-CPS1 for the seasonal and El Niño predictions. Some preliminary diagnostics indicate that the new features such as the sea-ice coupling, land initialization, and greenhouse gas forcing are contributed to improve predictive skills. Predictive skills are also compared with those of some systems in the EUROSIP system.

13 November
at 13:00

This talk will summarize some of the findings of the experimental campaigns made by the Atmospheric Boundary Layer (ABL) research team of the University of the Balearic Islands (UIB) since 2009. Campaign design, data analysis and high-resolution modelling are the three main tools described. The effect of the topography and the surface heterogeneities at different scales as observed in several locations (Garonne and Ebro basins, Mallorca, the Pannonian basin or the Cerdanya valley in the Pyrenees) will be summarily described, with special focus on the nocturnal boundary-layer features and the evening and morning transitions. A recurring picture seems to emerge and will be discussed in the seminar. On the other hand, we have a research station at the UIB Campus since 2013, located in a semi-rural area outside the city of Palma, where we measure all the terms of the Surface Energy Budget and most of the variables of the atmospheric surface layer and of the upper part of the soil. Instrumentation can be disposed at the most interesting configuration for any purpose, including initialization or validation of SVAT schemes. Furthermore, during 2016, a display of 10 supplementary energy-budget stations will be displayed in selected areas of the Campus (roughly 1 km^2), to evaluate the sub-kilometer variability of the measured quantities, in the so-called "Subpixel" campaign. Possible cooperation with the ECMWF using this experimental site may be explored.

11 December at 14:00

Room: MZR

Can the end of Moore's law improve resolution in climate modeling and weather prediction?

Speaker: Krishna V. Palem (Rice University, USA)

Abstract

Many claim that the laws of physics dictating exponential growth in transistor scaling or Moore’s Law will end in the next 10 to 20 years. This argument is based, in part, on an analysis that switching devices cannot function deterministically as feature sizes get reduced to the molecular level. Moore’s Law, however, could continue if systems with nondeterministic switches could process information usefully. My research that aims to exploit this principle, dubbed inexact computing, suggests that circuits and computing architectures can be used effectively at molecular scale. Succinctly put, inexact computing systems allow compromises in accuracy to be traded for significant, often disproportionately high gains in performance and energy consumption. Surprisingly, climate modeling and weather prediction provide great opportunities in this regard. For example, we have shown collaboratively, through simulations, that by lowering the precision of the mantissa, the dynamical core of an IGCM application yields savings through inexactness of (almost) a factor of 3 in energy without compromising the quality of the forecast. Given a fixed energy budget, this result implies that we can afford to compute at higher resolution; a back-of-the-envelope calculation shows that a factor of eight in energy savings would allow us to double the model’s resolution. In active collaboration with colleagues at Argonne National labs on the supercomputing front, and Tim Palmer and Peter Duben at Oxford University, we are currently working to demonstrate that existing supercomputers with half-precision instructions can potentially help us achieve resolution gains by exploiting inexactness. A goal of my talk is to invite collaboration from colleagues at ECWMF to join our consortium with the eventual goal of working towards improved resolution in the IFS model.

Krishna V. Palem is the Ken and Audrey Kennedy Professor at Rice University with appointments in Computer Science, in Electrical and Computer Engineering and in Statistics, and a scholar in the Baker Institute for Public Policy. He was a Moore Distinguished Faculty Fellow at Caltech in 2006-2007, and a Schonbrunn Fellow at the Hebrew University of Jerusalem in 1999, where he was recognized for excellence in teaching. In 2002, he pioneered a novel technology entitled Probabilistic CMOS (PCMOS) and laid the foundations of inexact computing. PCMOS and its potential applications were respectively recognized as one of the ten technologies 'likely to change the way we live' by MIT's Technology Review in 2008, and as one of the seven 'emerging world changing technologies' by IEEE as part of its 125th anniversary celebrations in 2009. In 2012, Forbes (India) ranked him second on the list of eighteen scientists who are “some of the finest minds of Indian origin.” He is a Fellow of the IEEE, the ACM and AAAS, and the recipient of the IEEE Computer Society's 2008 W. Wallace McDowell Award. He was named a Guggenheim fellow in 2015 to pursue the research direction suggested by this abstract.

11 December
at 15:30

Room: LT

Forecasts triggering humanitarian action: methods, successes, and challenges

Speaker: Erin Coughlan (Red Cross)

Abstract

Recognizing the potential value of existing weather and climate forecasts to anticipate extreme events, the Red Cross Red Crescent Movement and the World Food Programme are piloting a new approach called "Forecast-based Financing". This approach not only makes funding available for humanitarians to act based on a forecast (before a disaster happens), but develops standard procedures so that people automatically take action based on a predefined forecast threshold for extreme weather.

Come join this talk to hear which forecasts we are using around the world, from cyclone forecasts in Mozambique to GloFAS flood forecasts in Peru, and how we have recently successfully triggered action in advance of a disaster. We hope to solicit your recommendations and suggestions, as well as discuss further collaboration to scale up this concept and automatically use forecasts to prevent future humanitarian disasters around the globe.

2014

15 January 15:30

Large eddy simulation of urban flows in an idealized fractal model city

Sonja Gisinger (Univ Innsbruck)

Abstract

The majority of human activities and their impacts (e.g. pollution emission) takes place in the lowermost part of the atmosphere, especially in the extensively growing cities. Knowledge about the flow field and its modification through urban areas is essential for numerical weather and air quality predictions. The questions, how to describe an urban area, and how to parameterise its impact on the atmosphere in numerical prediction models must be considered.

In our work, we treated a city as a porous medium with a fractal geometry. We created an idealised model city (cf. Fig. 1) out of the Sierpinski triangle, a self-similar fractal with a fractal dimension of ≈1.585, and implemented it in EULAG using the immersed boundary method.

In analogy to Darcy's law, which describes the flow through a porous medium at low Reynolds numbers, we seek the relationship between the mean horizontal flow and the horizontal pressure gradient [1] for our high Reynolds number (turbulent) flow. A series of numerical simulations of the turbulent flow through the Sierpinski city was conducted for different flow speeds and different building heights. They indicate that the modified Darcy's law is valid for the turbulent flow through the fractal model city.

23 January 11:00

MR1

Using satellite observations and models to understand processes in the composition-climate system: Some examples

Apostolos Voulgarakis (Imperial College)

Abstract

During the last decade, a wealth of satellite observations has brought a new era in atmospheric science as it has provided the opportunity for continuous monitoring of the state of the atmosphere on large spatial scales. There is enormous potential in using such data for understanding large-scale processes in the composition-climate system. Here, I will present how we have used such satellite observations in conjunction with global model sensitivity experiments in order to examine the interannual variability of important tropospheric constituents such as ozone, CO, NO2 and aerosols. Particular focus will be placed on the influence of biomass burning emissions. Furthermore, I will discuss the value of using such satellite datasets in combination rather than individually, in order to investigate and evaluate processes in composition-climate models. The focus will be on a study that examines the global correlation of tropospheric ozone and CO, two important constituents that are interrelated in a complex way. Finally, future plans on how such satellite-based analysis can be expanded using more observational datasets will be discussed.

27 January 10:30

Representing model uncertainty in the ECMWF convection scheme

Hannah Arnold (Atmospheric, Oceanic and Planetary Physics, University of Oxford)

Abstract

In order to produce reliable probabilistic weather forecasts, it is important to account for all sources of error in atmospheric models. In the case of weather prediction, the two main sources of error are due to initial condition uncertainty and model uncertainty – this talk will focus on how to represent the latter. Two approaches are considered – the use of a perturbed parameter scheme, and the use of stochastic parametrisations. The problem is first illustrated using idealised experiments in the Lorenz ’96 “toy model” of the atmosphere, before the two approaches are considered in the context of the convection parametrisation scheme in the IFS. For the perturbed parameter approach, four uncertain parameters are identified in the convection parametrisation scheme, and a Bayesian technique is used to estimate the degree of uncertainty in their values. Both a fixed and stochastically varying perturbed parameter scheme are then tested in the IFS. Finally, we test a generalisation to the SPPT stochastic parametrisation scheme, in which the tendencies from different physics schemes are perturbed independently. This results in a large improvement in ensemble forecast spread in the tropics. The talk concludes by considering the limitations of the study, and suggesting future avenues of research.

20 February 10:30

Tropical cyclones' influence on the ocean: from event scale processes to climate scale consequences

Gurvan Madec (Locean-IPSL, Paris and NOC, Southampton)

Abstract

Strong winds associated to Tropical Cyclones (TCs) trigger intense mixing in the upper ocean. While the resulting surface cooling feeds back negatively on TCs intensity, the associated sub-surface warming has been suggested to substantially modify the ocean heat transport. A 1⁄2 ° global ocean model experiment that realistically samples the ocean response to more than 3,000 TCs over the last 30 years is used to first investigate the processes controlling the TC-induced surface cooling at the local scale and then to assess the impact of TCs at the global scale.

Vertical mixing is the dominant process of the cooling occurring locally close to the TC track. It will be shown that the cooling magnitude can be described by combining an index measuring the storm’s power and an index measuring the resistance to surface cooling by upper-ocean stratification. The cooling is very sensitive to the pre-storm upper-ocean stratification, which can modulate its amplitude by up to an order of magnitude for a given storm’s power.

The processes explaining the surface cooling under TCs also participates to modify the mean ocean heat budget. Previous studies have focused on the climatic importance of TC- induced mixing, but cooling is increasingly due to surface heat fluxes as we consider larger space scales. Both heat fluxes and vertical advection associated to TCs are shown to also influence the ocean mean state. Vertical mixing does induce an enhanced ocean heat uptake consistent with previous estimates. However, most of the heat injected into the ocean during TC seasons is re- entrained by the winter mixed layer deepening. As a consequence, we find that the main TCs’ climatological impact is to reduce the amplitude of surface temperature seasonal cycle more than to modify the ocean heat transport.

24 February 16:00

MCR

Predictability of wintertime Euro-Atlantic weather regimes in medium-range forecasts

Mio Matsueda (University Oxford)

Abstract

A weather regime is a persistent and/or recurrent large-scale atmospheric circulation pattern which is associated with specific weather conditions on a regional scale. Accurate simulations of weather regimes are important in weather and climate. The predictability of Euro-Atlantic weather regimes at medium-range timescales (up to 384hr) are investigated for winter (December-February) in the periods 2006/07-2012/13 and 1984/85-2012/13 using the THORPEX Interactive Grand Global Ensemble (TIGGE) and NOAA’s second-generation global medium-range ensemble reforecast datasets, respectively. The TIGGE portals quasi- operationally provide 9 medium-range ensemble forecasts routinely operated at NWP centres. We focus on five of the leading operational NWP centres: CMC, ECMWF, JMA, NCEP, and UKMO. The NOAA’s reforecast data has been produced with a fixed operational NWP model, using the 2012 version of NCEP’s Global Ensemble Forecasting System (GEFS), whereas the TIGGE data has been produced with a various versions of operational NWP model. The positive and negative phases of the NAO (NAO+ and NAO-), Atlantic ridge (ATLR), and Euro- Atlantic blocking (EABL) are detected as weather regimes over the Euro-Atlantic region from the ERA-Interim data. The NWP models have common biases in the frequency of regime transitions, and therefore the models prefer NAO- and ATLR to NAO+ and EABL with lead time, compared with the ERA-Interim. The models show small skill differences regarding probabilistic regime forecasts, suggesting that the skills of regime forecasts strongly depend on atmospheric flows. The models show higher forecast skills when predicting NAO+ and NAO-. The NAO+ and NAO- forecasts have a better skill than its climatological forecasts even at a lead time of 16 days. The persistence of NAO-is the most predictable. In contrast, EABL forecasts from ATLR have the lowest skill, followed by ATLR forecasts from NAO+, ATLR, and EABL.

19 March 10:30

Production of ERA-20CL: An ensemble of 20th-century global land-surface reanalyses with 25km resolution

Takuya Komori (ECMWF)

Abstract

As one of the products of ERA-CLIM reanalysis project, a set of global land surface reanalysis with 25km-resolution (referred to as ERA-20CL) for the period of 1900-2010 is currently in production. The production of ERA-20CL is based on meteorological forcing from the ERA-20C atmospheric reanalysis, involving application of downscaling techniques. Both the 10-member ensemble and the deterministic version of ERA-20CL are being produced following configurations of the forcing data set.

I will present some verification results of ERA-20CL products, and diagnostics to compare the performance with other reanalysis products.

16 April 10:30

Implementation of 4D-EnVar and other improvements in Environment Canada's Global Deterministic Prediction System (GDPS)

Mark Buehner (Environment Canada, Data Assimilation and Satellite Meteorology Research Section)

Abstract

Major changes to the data assimilation component of the GDPS will be described in this presentation. The most significant change is the replacement of 4D-Var with the 4D ensemble-variational data assimilation approach (4D-EnVar). This approach uses background ensembles generated by the global ensemble Kalman filter (EnKF) to estimate 4D flow-dependent background-error covariances. This enables a 4D analysis to be produced without needing the tangent linear and adjoint versions of the forecast model. Consequently, 4D-EnVar is much more computationally efficient and easy to maintain and adapt than 4D-Var. Other important changes include an improved satellite radiance bias correction scheme; assimilation of additional AIRS/IASI channels; improved treatment of radiosonde (4D) and aircraft observations; and the assimilation of ground-based GPS data. Results showing the impact of these changes (including also, the use of incremental analysis update, IAU, and recycling of physics variables) will be presented in preparation for the implementation of these changes in a parallel run.

2 May 14:00

Revisiting the role of Sea Surface Temperature Structure in Tropical Oceanic Rainfall

Richard E. Carbone (Chief Scientist for Strategic Development and Research Earth Observing Laboratory, NCAR, Boulder, CO)

Abstract

Past studies of sea surface temperature (SST) in relation to tropical ocean rainfall have arrived at somewhat divergent conclusions with respect to SST and gradients thereof (e.g. Neelin and Held, 1987; Lindzen and Nigam, 1991). Unlike most studies, Li and Carbone (2012) examined this issue from a mesoscale perspective, nominally at 100km scale, in an attempt to clarify the circumstances under which rainfall is triggered in the tropical western Pacific. They examined a 49 month timeseries of SST and rainfall as estimated from satellite observations. The seminar will begin with a brief summary of results, which shows a high frequency of mesoscale SST gradients and a propensity for onset of rainfall near the mode of the regional SST spectrum. While the statistical inference of dependence on SST is weaker than one might expect, coincidence with the convergent Laplacian of SST is strong.

The current work mainly emphasizes anomalies of and correlations between SST structure and rainfall in the MJO pass-band across the eastern hemisphere. The statistical inferences are, at once, broadly consistent with conventional wisdom on the role of SST in rainfall production and the pivotal role played by the convergent Laplacian of SST. It is concluded that the main role of SST in tropical oceanic rainfall is in production of moist static energy. That is to say, SST bears a decidedly conditional and somewhat indirect relationship to the occurrence of rainfall: i.e. if rain occurs, then more rain is likely over warm anomalies of SST. The SST Laplacian mainly triggers rainfall events through induced vertical air motion with sufficient kinetic energy to overcome convective inhibition in a conditionally unstable troposphere. Its role is directly associated with the frequency of rainfall events, however, it bears little direct relationship to cumulative rainfall beyond its capacity to influence approximately where, when, and how many events occur.

9 May 10:30

MR1

Extreme value estimation with the ACER method

Prof Arvid Naess (Norwegian University of Science and Technology)

Abstract

Some background on the ACER method
The prediction of long return period values based on observed data for design purposes has been the subject of extensive research over many years. However, there does not seem to be a general agreement on the method and extreme value distribution that would serve these purposes in a completely satisfactory manner. Standard methods for estimating extreme values from limited sets of observed data are commonly based on assuming either that the distribution of epochal extreme values converges to a Gumbel (type-I extreme value distribution) or by adopting a peaks-over-threshold (POT) approach, assuming that the exceedances above high thresholds follow a generalized Pareto distribution . A weakness of these approaches is that they depend on adopting asymptotic distributions. However, it is hard to verify the applicability or correctness of such procedures. Here we establish a method for predicting extreme wind speeds and significant wave height that avoid invoking the ultimate asymptotic distributions, but rather try to capture to some extent also the sub-asymptotic behaviour of extreme value data. This has resulted in a method that appears to be more appropriate for the purpose of predicting for example extreme wave heights than the traditional methods based on asymptotic theory. In addition to a procedure for optimizing the fit to data, the method consists of constructing a cascade of conditional distributions that makes it very easy to account for dependence effects in the data time series. The resulting sequence of conditional distributions constitute an increasingly accurate representation of the exact extreme value distribution given by the data. In fact, when this sequence of conditional distributions has converged, an exact representation of the extreme value distribution given by the data has been obtained since then no approximations have effectively been made. Thus, a plot of these conditional distributions will provide a unique insight into the dependence structure of the recorded data time series.

16 May 09:30

LCR

Global Water Data Sharing

David Maidment (Center for Research in Water Resources, University of Texas at Austin) and James Nelson (Brigham Young University)

Abstract

Talk 1: Global Water Data Sharing: Progress and Trends
Speaker: David Maidment, Center for Research in Water Resources, University of Texas at Austin

Scientists are increasingly being asked to share not only publications of findings but the data used to make those discoveries. The Consortium of Universities for the Advancement of Hydrologic Sciences Inc (CUAHSI) in response to developing tools for better management of hydrologic data has been involved with the OGC and WMO in developing the world standard WaterML for the way time series information at points can be shared. Such tools are making the ability to share data a reality and with an increasing amount of remotely sensed data and other applications available for flood, drought, and water resources management governments and others should be increasingly motivated to share information. David will discuss the current status and future possibilities that WaterML and water data services provides.

Talk 2: Applications of Water Data Sharing with Opportunities to Improve Flood Early Warning Systems in Latin America
Speaker: James Nelson, Brigham Young University

Less developed countries of the world lack resources including hardware, software and more importantly the human capacity necessary to manage their water information. Where initiatives have been undertaken to, they often are unable to pay for maintenance. Data and software tools like the CUAHSI HIS system are helping them join larger communities such as the Group on Earth Observations (GEO) that are pooling resources and making better water data management possible. Jim will discuss how the CUAHSI tools and the initiatives of GEO for data sharing are enabling them to use and create systems that will help them better manage their water resources with a particular emphasis on Flood Warning.

28 May 10:30

Some remarkable properties about the energy balance of Earth

Graeme Stephens ( Jet Propulsion Laboratory, University of Reading and Met Office)

There are several major efforts in improving global and regional data assimilation systems at NOAA. From NOAA National Weather Service to NOAA Office of Atmospheric Research, different data assimilation methodologies have been developed and tested collaborating with other research institutes, agencies and universities, such as 3DVAR, 4DVAR, EnKF and hybrid approaches. For lack of strong evidence of 4DVAR superiority, many NOAA data assimilation systems tend to use variations of ensemble approaches, hybrid 3DVAR or 4D-EnVar. In this presentation, we will review these efforts and discuss a possible global multi-scale 4DVAR system for NOAA future data assimilation. Some features of this system, along with supporting numerical experiment results, will be discussed during this presentation.

2 July 10:30

The GOES-R Geostationary Satellite System and NWP User Readiness

Steven Goodman, NOAA/NESDIS GOES-R Program Office

Abstract

NOAA’s Geostationary Operational Environmental Satellites (GOES) are a mainstay of weather forecasts and environmental monitoring in the United States. The next generation of GOES satellites, known as the GOES-R Series, is scheduled for launch in late 2015 and will usher in a new era in geostationary environmental satellites. The GOES-R satellites will provide continuous imagery and atmospheric measurements of Earth’s Western Hemisphere that will foster a host of improved and new environmental products and services.
GOES-R’s primary instrument, the Advanced Baseline Imager (ABI), will provide twice the spatial and three times the spectral resolution while scanning the Earth nearly five times faster than the current GOES. GOES-R will also host a new instrument, the Geostationary Lightning Mapper (GLM) that is designed to continuously map in-cloud and cloud-to-ground lightning with a 8 km spatial resolution and 80 percent detection efficiency over the Western Hemisphere. It will provide information to improve severe storm monitoring and warnings and contribute to improved aircraft safety and efficient flight route planning. GOES-R’s space weather instruments will provide improved observations of the sun and space environment with more timely dissemination and early warning to a diverse user community.
This presentation will provide an overview and status update of the GOES-R program and the activities leading to an operational GOES-R system. The new observations will provide dramatically improved weather, water, and space environmental services in the coming decades, enhancing public safety and providing economic benefits to the U.S. and our international partners.

3 July 14:00

A modification of the IFS semi-Lagrangian trajectory scheme to improve forecasts of Sudden Stratospheric Warming events

Michail Diamantakis, ECMWF

Abstract

Sudden Stratospheric Warming (SSW) events are poorly forecast by IFS. An investigation has shown that this poor performance is strongly linked with semi-Lagrangian numerics and particularly with the time extrapolation method used for computing the departure points. In this talk used algorithms for computing the departure points will be briefly reviewed. Different options for improving IFS forecasts of such events will be explored focusing on a recently proposed simple modification of the existing scheme. It will be shown that this method achieves large improvement of SSW forecasts without increasing computational cost.

7 July 14:00

Evolution of Ensemble Data Assimilation at ECMWF

Massimo Bonavita, ECMWF

Abstract

The Ensemble of Data Assimilations (EDA) has been a key contributor to the improved skill of both the high resolution (HRES) and ensemble forecasting (ENS) systems at ECMWF in recent years. This is mainly due to its ability to provide realistic estimates of the analysis and background error covariances in the data assimilation cycle. In this talk we review recent progress in the use of EDA-based error estimates in the 4D-Var analysis cycle.

On the other hand, computational constraints effectively limit the size and resolution at which the EDA can be run in operations, which limits the fidelity of the error estimates that can be obtained from the EDA and introduces sampling noise. This has led to exploring more efficient and scalable avenues for the cycling of the errors of the data assimilation cycle. Results from this experimentation will be presented and the implications of their adoption for operational use will be discussed.

30 July 10:30

Coupling of diurnal and seasonal climate to clouds, precipitation, land-use and snow cover on the Canadian Prairies

Alan Betts

Abstract

Analysis of the 13 climate stations across the Canadian Prairies with hourly observations from 1953-2011 of temperature, RH, precipitation and snow depth has transformed our understanding of land-surface-climate coupling, because trained observers (following the same protocol for 60 years) recorded in addition hourly opaque/reflective cloud cover in tenths (defined as cloud that obscured the sun moon or stars). These data can be calibrated on daily timescales against radiation measurements to give SWCF and LWCF. We find that cloud forcing of the diurnal climate has distinct warm and cold season behavior. From April to October, when incoming shortwave radiation dominates over longwave cooling, maximum temperature and the diurnal ranges of temperature and relative humidity increase with decreasing opaque cloud cover, while minimum temperature is almost independent of cloud. Anomalies of precipitation and cloud explain 70-80% of the variance of monthly mean temperature and RH in the warm season. A large land-use change from summerfallow to continuous cropping in recent decades has cooled and moistened the summer climate. During the winter period, both maximum and minimum temperature fall with decreasing cloud, as longwave cooling dominates over the net shortwave flux. With lying snow, 2-m temperature drops 10oC. On regional scales 10% fewer days with snow cover gives a winter climate warming of 1.4oC. Snow cover acts as a climate switch: the transition from the warm season unstable BL to the cold season stable BL occurs within 5 days. Papers are available at http://alanbetts.com

4 September 10.30

Gravity waves and water vapor in the stratosphere as deduced from observations and ECMWF data

Peter Preusse & Paul Konopka, Forschungszentrum Juelich, Germany

Abstract

Data of the global assimilation system of ECMWF resolve mesoscale dynamical process in the stratosphere such as gravity waves and transport processes. In a two-part presentation we will give examples for utilizing ECMWF data for stratospheric research.

A) Gravity waves (GWs) couple the different layers of the atmosphere by transporting momentum when propagating from lower to higher altitudes. Usually, general circulation models are too coarse to resolve GWs and therefore need to take into account their effects by submodels called GW parametrisation schemes. The ECMWF model is of
sufficient spatial resolution to resolve a larger part of the GW spectrum. Still a GW parametrisation is required. In the talk we will present how observations can be utilised to guide the parametrisation. In addition, we will discuss the properties of GWs resolved in ECMWF and assess the resolved waves by comparison with observations.

OLYMPEX: A Field Campaign in Complex Terrain for Validation of Precipitation Measurements by the NASA Global Precipitation Measurement (GPM) Satellites

Dr. Lynn McMurdie, Department of Atmospheric Sciences, University of Washington

Abstract

At 18:37 UTC 27 February 2014, the core satellite of the Global Precipitation Measurement (GPM) mission was successfully launched from the Japanese Space Agency (JAXA) Space Center on Tanegashima Island, Japan. Aboard the satellite is the first space-borne Ku/Ka band Dual Frequency Precipitation Radar (DPR) and the GPM passive Microwave Imager (GMI), with channels ranging from 10-183 GHz. The primary objective of the Core satellite is to measure rain and snow globally, determine its 3D structure, and act as the calibration satellite for a constellation of GPM passive microwave satellites in a wide range of precipitation intensities, geographical locations and weather regimes. In order to identify and understand uncertainties in the GPM measurements and to assess how remotely sensed precipitation can be applied to a range of applications (e.g. determining storm structures and monitoring runoff and flooding events, and as input to numerical models), ground validation (GV) field campaigns are crucial. As such, the Olympic Mountains GV Experiment (OLYMPEX) is planned for November 2015 – February 2016.

The Olympic Peninsula in the northwest corner of Washington State is an ideal location to conduct a GV campaign. It is situated within an active mid-latitude winter storm track and receives among the highest annual precipitation amounts in North America, ranging from over 2500 mm on the coast to 4000 mm in the mountainous interior. In one compact area, the Olympic peninsula ranges from ocean to coast to land to mountains. It contains a permanent snowfield and numerous associated river basins. This unique venue will enable the field campaign to monitor both upstream precipitation characteristics and processes over the ocean and their modification over complex terrain.

The scientific goals of the OLYMPEX field campaign include physical validation of satellite algorithms, precipitation mechanisms in complex terrain, hydrological applications, and modeling studies. In order to address these goals, a wide variety of existing and new observations and instrumentation are planned. These include in situ surface observing networks of meteorological stations, rain and snow gauges, surface microphysical measurements, and snowpack surveys. Surface-based remote sensing instrumentation will include the existing and planned radars such as the NASA N-Pol S-Band dual-polarimetric and NASA Dual-Frequency Dual-Polarimetric Doppler (D3R) scanning radars and other mobile vertically-pointing radars. Several instrumented aircraft are likely to participate. The NASA DC-8 will be equipped with a Ka/Ku band dual-frequency radar and passive microwave sensors that simulate those on the GPM Core satellite. The University of North Dakota Citation will measure in situ microphysics. The aircraft measurements will determine upstream thermodynamic and moisture conditions, sample particle types and sizes for comparison with those employed in the satellite algorithm, and act as a proxy for the satellite itself. The ground-based measurements will test how well the satellite proxy measurements determine the rain and snow over complex terrain.

31 October 14:00

Early warning early action: using global weather forecast data trigger humanitarian interventions

Erin Coughlan, Red Cross Red Crescent Climate Centre, USA

Snow cover is a key component of the climate system. Its presence on the ground drastically changes the physical properties of the continental surfaces and thus strongly impacts the energy exchanges between the surface and the atmospheric boundary layer. The most important features of snow cover with respect to weather prediction will be presented in details. Its radiative and thermal properties, including its capacity, its conductivity and its melting latent heat, lead to an efficient cooling of the snow surface and to a stabilization of the atmospheric boundary layer.

Numerical snow models are quite efficient tools to simulate snow cover evolution, though they might suffer from the difficulty to derive turbulent heat fluxes under very stable conditions. A new method will be presented to assess turbulent heat fluxes under very stable conditions. It is based on the use of remote-sensed surface temperature from MODIS over Antarctica. Taking advantage of the long polar night during which short-wave radiation is negligible, the method reveals an overestimation of sensible turbulent heat fluxes in IFS under very stable conditions, leading to a widespread warm bias in ERA-interim at the surface of Antarctica.

18 November 10.30

Normal-mode function representation of the global 3D circulation

Nedjeljka Zagar, University of Ljubljana, Slovenia

Abstract

The goal of the seminar is to present the open-access software MODES, which is now available to atmospheric research community for the analysis of global NWP and climate models, and the software application to the ECMWF reanalysis and ensemble datasets.

The software, developed within the ERC MODES project, allows one to diagnose properties of balanced and inertio-gravity circulation across many scales. In particular, the IG spectrum, which has only recently become observable in global datasets, can be studied simultaneously in the mass field and wind field and considering the whole model depth. The presentation will include theoretical background and basic technical details of the software that been installed and running on both ecgate and c2a computers at ECMWF. Results from the ERA Interim analysis in terms of the normal-mode function (NMF) representation will be presented. In addition, outputs from the NMF analysis of the EPS system operational in 2013 will be discussed in relation to the previously derived properties of time-averaged and time-dependent forecast errors derived from the EDA system.

19 November 10:30

MR1

Improving the Madden-Julian oscillation in the Met Office model: The roles of air-sea coupling and convective entrainment

Dr Nicholas Klingaman, Department of Meteorology, University of Reading

Abstract

The Madden-Julian oscillation (MJO) is the leading mode of sub-seasonal variability in tropical convection and provides a key source of weekly and monthly predictability globally. In a multi-decadal atmosphere-only simulation, the Met Office Unified Model (MetUM GA3.0) produces roughly one-half the observed level of MJO activity, with events often living for fewer than five days.

We conduct a series of initialised, climate-resolution hindcasts of observed MJO events to test a range of parameterisation and model-configuration changes in MetUM GA3.0. Increasing the rate of entrainment and detrainment in the convective parameterisation is the only change that consistently improves the MJO across the hindcast set. With higher entrainment, MetUM GA3.0 produces near-observed levels of MJO variability. Having improved the MJO in the atmosphere-only model, we investigate the roles of air-sea interaction in the MJO at low and high entrainment rates by coupling to a mixed-layer ocean. At lower rates, with poor atmospheric variability, coupling primarily improves the amplitude of the MJO in the model. At higher rates, when the atmosphere has a reasonable MJO, coupling primarily improves MJO propagation without affecting the amplitude. The seminar will conclude with some comments on how coupled-model SST biases affect not only the representation of the MJO, but also the perceived impact of air-sea coupling.

28 November 10:30

MR1

Does the ECMWF IFS convection parametrization with stochastic physics correctly reproduce relationships between convection and the large-scale state?

Peter Watson, University of Oxford

Abstract

A method for estimating return values from ensembles of forecasts at advanced lead times is presented.

Return values of significant wave height in the North-East Atlantic, the Norwegian Sea and the North Sea are computed from archived +240-h forecasts of the ECMWF ensemble prediction system (EPS) from 1999 to 2009.

We make three assumptions: First, each forecast is representative of a six-hour interval and collectively the data set is then comparable to a time period of 226 years. Second, the model climate matches the observed distribution, which we confirm by comparing with buoy data. Third, the ensemble members are sufficiently uncorrelated to be considered independent realizations of the model climate. We find anomaly correlations of 0.20, but peak events (>P97) are entirely uncorrelated. By comparing return values from individual members with return values of subsamples of the data set we also find that the estimates follow the same distribution and appear unaffected by correlations in the ensemble. The annual mean and variance over the 11-year archived period exhibit no significant departures from stationarity compared with a recent reforecast, i.e., there is no spurious trend due to model upgrades.

EPS yields significantly higher return values than ERA-40 and ERA-Interim and is in good agreement with the high-resolution hindcast NORA10, except in the lee of unresolved islands where EPS overestimates and in enclosed seas where it is biased low. Confidence intervals are half the width of those found for ERA-Interim due to the magnitude of the data set.

30 April 10:30

Calibration of medium-range weather forecasts

Prof Tilmann Gneiting, Univ of Heidelberg

9 May 10:30

MR1

Precipitation and snowpack in Hindu-Kush Karakoram Himalaya: Current features and future projections

Silvia Terzago, SAC-Institute of Atmospheric Sciences and Climate, CNR, Torino, Italy

Abstract

The Hindu-Kush Karakoram Himalaya (HKKH) is the largest mountain region in the world and feeds some of the major rivers in Southeast Asia supplying water to more than 1.5 billion people. Despite its importance, the current precipitation, the snowpack dynamics and their possible changes in the coming decades are still a major challenge due to difficult and sporadic surface observations in such remote areas.

In this work we analyze the precipitation and its mechanisms in the Hindu-Kush Karakoram Himalaya(HKKH) region using currently available data sets, including satellite estimates, reanalyses, gridded in situ rain gauge data, and we compare them to the simulations from the global climate model EC-Earth. The results are related to the characteristics of the snowpack investigated using the output of the global climate models and the ERA-Interim/Land reanalysis. Finally we explore the possible effects of climate change in the HKKH region by using the available future projections in different climate change scenarios.

14 May 15:30

The Soil Moisture Active Passive (SMAP) Mission Application Program Pre Launch Efforts

Vanessa Escobar, NASA/GSFC

Abstract

NASA’s Soil Moisture Active Passive (SMAP) mission is due to launch October of 2014. SMAP is a combined active/passive microwave instrument, which aims to produce a series of soil moisture products and soil freeze/thaw products with an accuracy of +/-10%, a nominal resolution between 3 and 40km, and latency between 12 hours and 7 days. These measurements will be used to enhance the understanding of processes that link the water, energy and carbon cycles, and to extend the capabilities of weather and climate prediction models. SMAP is NASA ’s first decadal survey mission to implement a rigorous applications program that seeks to integrate soil moisture derived data into decision making applications fields of weather, climate, drought, flood, fire and human health, prior to launch.

The focus of the SMAP Applications Program is to (1) improve the missions understanding of the SMAP user community requirements, (2) document accuracies and biases of remote sensing derived-soil moisture and communicate the perceived challenges and advantages to the mission scientists (3)facilitate the movement of science into policy and decision making arenas and 4) strive to provide a clear path to access proxy data (both in terms of accessibility and format). Since 2007, the SMAP Applications Program has provided new concepts, definitions and activities to promote SMAP applications. Executing this effort is the SMAP Applications Team, composed of key individuals from within the SMAP project, SMAP Science Definition Team (SDT), and NASA Headquarters.

The driving success of the SMAP applications program has been the joining of mission scientists to thematic end users and leveraging the knowledge base of soil moisture data applications, increasing the ingestion speed of future SMAP data product into critical processes and improving the societal benefits to science. Understanding how users will apply SMAP data, prior to the satellite’s launch, is an important component of SMAP Applied Sciences. Successfully achieving this requires bridging scientific research and thematic user communities’ operational decision making priorities and data requirements.

Through our pre launch applications work with the SMAP community and the SMAP Early Adopter program, we have learned that there are challenges the SMAP mission can address prior to launch, that will increase the applications and utility of post launch data. Because application requirements are different for each user, we focus on understanding the individual resolution, access and accuracy concerns of SMAP data by thematic discipline. This understanding in the pre launch stages of the mission’s development is an integral part of converting the data collected into actionable knowledge that can be used to inform policy and improve societal applications.

13 June 14:00

Flow dependent predictability in the North Atlantic sector

John Methven, Department of Meteorology, University of Reading

Abstract

The predictability of the atmosphere is believed to be flow-dependent in the sense that it is easier to make skillful predictions from some flow configurations than others. However, identifying the configurations with greater predictability and quantifying the enhanced skill in these situations is difficult. A key aspect in identification of flow-dependent predictability is that the definition of flow "configuration" must occur sufficiently frequently that statistics of ensemble forecast spread can be gathered. Two distinct phenomena in the extratropical atmosphere are discussed.

The first relates to the variability of the westerly jet across the Atlantic and medium-range ensemble predictions for the evolution of the jet. It is found that when the jet weakens or splits (in latitude) it enters into a state more sensitive to small differences between ensemble members and is therefore less predictable.

The second concerns the link between the predictability of "weather variables" (precipitation and surface winds) with the occurrence of mesoscale structures within extratropical storms. The predictive skill of mesoscale features, identified in ensemble forecasts using the Hewson method, is explored. Some illustrations are given of phenomena where the skill in finescale forecast variables is linked to the skill of the forecast for mesoscale features.

26 June 14:00

DICast : The Dynamic Integrated Forecast System

Bill Myers, Global Weather Corporation & National Center for Atmospheric Research Boulder, Co

Abstract

In the past 15 years, operational weather forecasting advancements in the post-processing of numerical weather models have significantly improved weather forecasts provided to the public and to industry. The development of consensus forecasting techniques has probably led to the largest reduction in forecast errors since the advent of Model Output Statistics (MOS) in the 1960s. DICast, developed in the late 1990s at NCAR, is a completely automated consensus forecast system that was modeled on the human forecast process. It considers multiple forecast inputs and continually compares its forecasts to observations in a machine learning approach that generates objective forecasts that outperform its ingredient forecasts, including NWS and ECMWF products, by 10-15%. DICast now drives the forecast engines of the largest weather forecast providers in the United States and, to an increasing degree, internationally.

27 June 10:00

MR-NB

The role of atmospheric rivers in British winter floods

David Lavers, University of Iowa

Abstract

Damage from flooding in the winter and autumn seasons has been widespread in the United Kingdom and Western Europe over recent decades. In this seminar the connection between atmospheric rivers (ARs) and the largest winter floods in a range of British basins will be discussed from a hydrological and atmospheric stand-point. Firstly, an analysis of the hydrological time series is used to evaluate atmospheric fields before the largest floods to show the AR-flood link. Secondly, an algorithm is introduced that screens for ARs in climate model output; this is followed by showing the strong connection between the identified ARs and British winter floods. Future changes to ARs under climate change will also be considered.

David has a BSc (Meteorology) from the University of Reading UK, an MSc (Applied Meteorology) from the University of Birmingham UK and a PhD from the Centre of Ecology and Hydrology Wallingford / University of Birmingham UK. He has worked on seasonal climate prediction at Princeton University and has also been a weather forecaster in the Middle East. He recently completed a post-doctoral appointment at the University of Reading on using hydrologically-relevant processes for the assessment of climate model output. He is currently a post-doctoral researcher at the University of Iowa working in the broad area of hydrometeorology.

9 July 14:00

Composition-climate interactions: What can we (already) learn from global models?

Prof Peter Braesicke, Karlsruhe Institute of Technology, Germany & University of Cambridge, UK

Abstract

The talk will introduce the UMUKCA composition-climate modelling system, a joint Met Office/NCAS development. It will discuss the models performance and introduce some sensitivity studies that reveal composition-climate interactions. I will show that the climate system (as represented by the model) can respond quite sensitively to small chemical perturbations in ozone. I will argue that some modelled circulation adjustments in response to a chemical change can occur in reality and will illustrate this point using MIPAS data. I will finish with an outlook towards smaller scale features that are currently not represented in this climate model set-up, even though they are potentially important for composition-climate interactions. My example will be the Borneo vortex (BV), an important source of extreme weather for Malaysia, and a factor in creating a dry, ozone rich layer in the upper troposphere/lower stratosphere.

11 July 14:00

Using Earth-system science at ECMWF

Erland Källén, Director of Research, ECMWF

Abstract

In order to continue to deliver improved forecast performance the ECMWF forecast model and assimilation system will have to include additional parts of the Earth-system. Today we include the atmosphere, the land surface and, for extended range forecasts, also the ocean. A forecast system for atmospheric composition is under development and will be an integral part of the future ECMWF forecast system. Coupling to the ocean will be extended to start from day zero and the interaction between ocean waves, upper ocean mixing and sea-ice will be further developed. Reanalyses will also be extended to include additional parts of the Earth-system, both atmospheric composition, the land surface and the ocean will gradually be integrated to form a fully coupled reanalysis system. Some recent examples of progress in Earth-system modelling at ECMWF will be shown in order to demonstrate the capabilities and future potential of Earth-system modelling.

25 July 10:30

Medium-range probabilistic verification and predictability of tropical cyclogenesis using the ECMWF EPS

Sharan Majumdar, Meteorology and Physical Oceanography (MPO) Rosenstiel School of Marine and Atmospheric Science (RSMAS), University of Miami

Abstract

As part of the NSF PREDICT project, ensemble-based products have been developed with a goal to improve probabilistic predictions and our understanding of the predictability of tropical cyclogenesis. The probabilistic verification of quantities based on ECMWF ensemble forecasts out to 6 days will be presented for the 2010-12 Atlantic Hurricane seasons. Calibrated threshold values of metrics including a low-layer average circulation and a local thickness anomaly are used to determine the onset of cyclogenesis in each ensemble member. Reliability diagrams suggest a mostly increasing relationship between the predicted probability and frequency of actual genesis cases. However, there has been a tendency, particularly in 2012, for the ensemble to under-predict the development of the warm core. Some case illustrations will be presented. Finally, a series of metrics to investigate the predictability of fields associated with genesis such as wind shear and circulation will be introduced.

27 August 10:30

MR1

Dynamics of the Madden-Julian oscillation

Adam Sobel, Columbia University

Abstract

The Madden-Julian oscillation (MJO) is the dominant mode of variability in the tropics on the intraseasonal time scale (say, 20-90 day periods) and one of the most important coherent, quasi-periodic modes of natural variability in the global climate system altogether. Though it was discovered over 40 years ago, we still do not understand the MJO, in the sense of being able to state a simple mathematical model that explains its basic features.

I will present evidence that the MJO is what some of us now call a "moisture mode", best analyzed by examining the budget of moist static energy or moist entropy. I will argue that cloud-radiative feedbacks are important to the maintenance of the MJO, while horizontal advection of moisture is important to its eastward propagation. I will present evidence from observations, theory, general circulation models, and cloud-resolving models to this effect.

The value of curvilinear coordinates is well established in climate and weather prediction. Such coordinate systems can offer clarity in physical insights about meteorological phenomena of interest as well as simplicity of computational approach. This talk will present the case that extension of curvilinear coordinates with fully continuous dynamic grid adaptivity is a natural generalization for the next generation of NWP models. Continuous GA can offer further enhancements such as increased resolution on targeted regions and/or dynamical–moist structures of interest. Locally enhanced resolution can have global impact. Continuous GA also enables a practical way to test interactions between regions characterized by different meteorological processes via the mechanism of adjustable grid resolution that would otherwise be difficult or impossible to make.

This talk will briefly summarize the basic ideas underlying continuous dynamic grid adaptivity in large–scale computation. In general, the desired grid structure is non-uniform and/or moving. A mapping between two distinct spaces is applied, and in the transformed space the structured computational grid is uniform, stationary, and the boundaries “straight” irregardless of the deformations occurring in the physical space. The chosen map must satisfy the typical continuity constraints and also two types of transformation rules. These transformation rules are intrinsically geometrical in nature and vital for maintaining conservation and the parallel transport property.

Examples are presented ranging from idealized regional and global tests to aqua-planet and African climate simulations. They are chosen to illustrate how GA can improve model efficiency; demonstrate advantages from locally improved resolution that produce regional/global as well as local impact; or help to illuminate the role of different types of global dynamics in moist processes.

14 November 10:30

Climate modeling with super-parameterization and beyond

Marat Khairoutdinov (School of Marine & Atmospheric Sciences Stony Brook University, NY)

Abstract

Clouds play a key role in the Earth's climate system. Yet, they are not generally resolved by general circulation models (GCMs), but rather, parameterized using cloud parameterizations. An alternative approach to cloud parameterizations in GCMs is a so-called super-parameterization(SP), which is a small-domain cloud-resolving model inserted into each grid-column of a GCM to replace cloud parameterizations. The first GCM with the SP, the SP-CAM, based on the NCAR's Community Atmosphere Model (CAM), has been used since about 2001 to better simulate phenomena such as the Madden-Julian Oscillation, and to study cloud feedbacks in the climate system. The details of the SP approach and key results will be discussed. Recently, the SP has been implemented in the OpenIFS model. Some preliminary SP-IFS model results will be presented.

26 November 10:30

Arctic sea ice, modelling and predictability

Matthieu Chevallier, CNRM-GAME, Météo-France/CNRS and Mercator-Océan, Toulouse, France

Abstract

This presentation will focus on activities related to Arctic sea ice modelling and forecasting at CNRM-GAME(Météo-France/CNRS) and Mercator-Océan (Toulouse, France).

The GELATO sea ice model will be presented. This state-of-the-art multicategory model is coupled to the NEMO (version 3.2) model and included in CNRM-CM5 coupled atmosphere-ocean model (1°x1° horizontal resolution). CNRM-CM5 has been used for IPCC AR5 climate scenarios and decadal prediction experiments. CNRM-CM5 is also planned to be used as the next system of seasonal forecasting at Météo-France. The skill of CNRM-CM5 in forecasting the Arctic sea ice has already been assessed in 1990-2010 hindcasts initialized from a forced ocean-sea ice reconstruction with NEMO-GELATO.

Skill scores of summer and winter forecasts of the Arctic sea ice extent are promising, and show that at a 5-month lead time, sea ice predictability is mainly an initial value problem. These hindcasts also help to identify some mechanisms responsible for sea ice predictability at such time scales. Plans for improvements (data assimilation, increase in resolution) and future work planned at CNRM-GAME on sea ice modelling and seasonal predictability in the polar regions will be presented.

Part of the presentation will also be about high-resolution ocean-sea ice modelling at Mercator-Océan, and collaborative plans between Météo-France, Mercator-Océan and the Canadian Meteorological Centre (Environment Canada) for the development of a regional Arctic ocean-sea ice model at 1/12° using the NEMO platform and Mercator-Océan's data assimilation system.

26 November 15:30

MR1

DEWFORA Seminar:

How useful are ECMWF precipitation forecasts over Africa for drought applications?

Programme

Presentation of DEWFORA and main results from ECMWF’s involvement - Emanuel Dutra/Fredrik Wetterhall
Forecasting droughts in East Africa - Emmah Mwangi, KMA, Kenya
Exploring the potential of adapting to climate change through seasonal prediction of meteorological drought indicators - Hessel Winsemius, Deltares, The Netherlands
Developing a Pan-African map viewer for drought - Paulo Barbosa, JRC, Italy

6 December 9:30

Classroom

JRC tropical cyclones simulation in support to the Global Disasters Alerts and Coordination System

Alessandro Annunziato and Delilah Al Khudhairy, Joint Research Centre, EC

Abstract

The Joint Research Centre developed, in collaboration with the United Nation Office for Coordination of Humanitarian Affairs (UN-OCHA) the Global Disasters Alerts and Coordination System aimed at providing the humanitarian community of a pre-alert for possible large disasters event which may need the assistance of large international teams on the field. The area of interest includes Earthquakes, Tsunami, Tropical Cyclones and Floods.

For Tropical Cyclones GDACS provide early estimation of winds, rainfall and storm surge effects by using information provided by international typhoon centers and performing online calculations whose results are then published automatically in the GDACS web site (www.gdacs.org). The cyclone calculations are performed by reconstructing the pressure and wind file using the data present in the bulletins and computing, with a Monte Carlo simulation, the Holland parameters to have the whole wind/pressure profile. The method has been so far successful in identifying the major events worldwide, including the Philippines case.

Recently a new activity has been initiated in order to explore the possibility to use ECMWF wind and pressure estimates from the Medium Range deterministic model as boundary conditions to the storm surge calculations. The reason was the announcement and the availability of finer estimates (16 km) respect to the previous one (50 km) and thus enabling a better resolution, as required by the hydraulic calculations. The results of this pilot activity are extremely interesting for several reasons that will be discussed in the presentation: in some cases the agreement with measured storm surge are really very good while in other cases the forcing is not enough to generate sufficient surge as measured. In any case this method has the advantage to make possible the analysis of extra tropical cyclones which are normally not analysed in GDACS.

10 December 10:30

Preconditioning Saddle Point Formulation of 4D-VAR

Selime Gurol, ECMWF

Abstract

In this talk we will address the numerical solution of the saddle point system arising from four dimensional variational (4D-Var) data assimilation with an emphasis on a study of preconditioning with low-rank updates. The saddle point formulation of 4D-Var allows parallelization in the time dimension. Therefore, it represents a crucial step towards improved parallel scalability of 4D-Var. We will present numerical results obtained from the Object Oriented Prediction System (OOPS) developed by ECMWF.

12 December 10:30

Diabatic processes in extratropical cyclones: Dynamics and relevance for forecast accuracy

Heini Wernli, Institute for Atmospheric and Climate Science, ETH Zurich, Switzerland

Abstract

The prediction of the track, intensity and structure of extratropical cyclones is of key importance for accurately forecasting intense winds and precipitation associated with these weather systems. In addition to the primary agents of baroclinic instability, i.e., the horizontal temperature gradient and typically an upper-level cyclonic vorticity disturbance, moist diabatic processes that occur in saturated ascending airstreams are essential for the evolution of cyclones, the associated surface weather, and the downstream flow evolution. A potential vorticity perspective helps explaining how diabatic processes impact on the dynamics of extratropical cyclones and tropopause-level Rossby waves. Results from complementary studies in our group, mainly based upon ECMWF (re)analysis and forecast data, will be shown to illustrate that (i) the diabatic PV production in the low troposphere is essential for the formation of intense extratropical cyclones, (ii) diabatic processes are particularly intense in so-called warm conveyor belts, i.e., in moist ascending airstreams associated with extratropical cyclones, (iii) a major part of precipitation extremes occurs in associating with extratropical cyclones and/or warm conveyor belts, (iv) exceptionally poor global model medium-range weather forecasts in Europe are often associated with errors in the representation of warm conveyor belts, (v) during the previous 10 years, the accuracy in predicting the intensity and location of these moist ascending airstreams has improved for the ECMWF high-resolution model, and (vi) that a combination of different microphysical processes occurring within these conveyor belts is responsible for the overall latent heating and associated potential vorticity modification. The presentation will also emphasise (a) the usefulness of investigating dynamical processes in ensemble predictions, (b) the need for field experiments dedicated to observing diabatic processes in mid-latitude weather systems, and (c) the value of a systematic feature-based forecast validation of weather systems with NWP and climate models.

12 December 11:30

Weather regime project at ETH

Christian Grams, Institute for Atmospheric and Climate Science, ETH Zurich, Switzerland

2012

9 January 14:00

MR1

On the North Pacific reduced skill in near-term climate predictions of sea surface temperatures

Virginie Guemas, IC3 - Institut Català de Ciències del Clima CFU, Climate Forecasting Unit, Barcelona, Spain

Abstract

Near-term climate prediction relies both on the predictability of the internal climate variability, by initializing climate models from estimates of the observed state, and on the externally forced predictability (by greenhouse gases, aerosols, solar activity). This exercise shows that the North Pacific region is where the current generation of climate forecasting systems performs the worst worldwide. In this presentation, we look for the major events missed by the forecasting systems and we investigate the reasons for this failure.

27 January 14:00

Atmospheric composition monitoring from the geostationary orbit: US plans and the international context

David Edwards, NCAR Earth System Laboratory, Deputy Director

30 January 15:30

When will the Arctic become ice free?

Seymour Laxon, Centre for Polar Observation and Modelling, University College London

Abstract

Changes in Arctic sea ice cover are perhaps the most visible evidence for the Earths changing climate. Satellite records have shown a 30% decrease in the September ice extent over the last 30 years and submarines have found a similar decline in thickness over limited regions. However model predictions of the date at which the Arctic might become ice free in summer vary widely. This seminar will discuss the key issues involved and present new data, in particular from the CryoSat mission, which can shed light on this question. We will also describe the fieldwork which is ongoing to validate the CryoSat measurements.

6 March 15:30

Sub-seasonal predictability and dynamical processes: the two-way MJO and NAO interaction

Gilbert Brunet, Meteorological Research Division (MRD), Environment Canada

Abstract

The Madden-Julian Oscillation (MJO) is the dominant mode of intraseasonal variability in the tropics [Madden and Julian, 1971], which has a direct impact on the weather in the tropical region, as it organizes convection and precipitation. It also has a significant influence on the extratropical atmospheric variability, possibly through Rossby wave propagation [Lin et al., 2009], and thus provides an important signal source for the extratropical weather forecasts on subseasonal timescales. A skilful prediction of the MJO is of great importance. The North Atlantic Oscillation (NAO) is one of the most important modes of variability in the Northern Hemisphere extratropical atmosphere [Hurrell, 1996]. In the work of Lin et al. [2009], it was found that there is a significant two-way interaction between the MJO and the NAO. On the one hand, the MJO, through extratropical Rossby wave propagation associated with its anomalous tropical convection and diabatic heating, influences the extratropical circulation and the NAO amplitude. On the other hand, the NAO variability results in changes in the tropical upper zonal wind and the initiation of the MJO. By analyzing the output of an intraseasonal hindcast, Lin et al. [2010] have shown that the MJO has a significant impact on the intraseasonal forecast skill of the NAO. In Lin and Brunet [2011], it was demonstrated that the NAO also has an important influence on the forecast skill of the MJO. Intraseasonal forecasts would benefit from such an interaction if such a process can be resolved in forecasting system.

1.Lin, H., and G. Brunet, 2011: Impact of the North Atlantic Oscillation on the forecast skill of the Madden-Julian Oscillation. Geophys. Res. Lett., VOL. 38, L02802, doi:10.1029/2010GL046131.

2. Lin, H., G. Brunet and J. S. Fontecilla 2010: Impact of the Madden-Julian Oscillation on the intraseasonal forecast skill of the North Atlantic Oscillation. Geophys. Res. Lett., 37, L19803.

3. Lin, H.,G. Brunet, and J. Derome, 2009: An observed connection between the North Atlantic Oscillation and the Madden-Julian Oscillation. J. Climate, 22, 364-380.

4. Madden, R. A., and P. R. Julian, 1971: Description of a 40-50 day oscillation in the zonal wind in the tropical Pacific. J. Atmos. Sci., 28, 702–708.

7 March 10:30

Experimental 4D-Var assimilation of SYNOP rain gauge data at ECMWF

Philippe Lopez, ECMWF

Abstract

The potential benefits of assimilating worldwide SYNOP rain gauge 6-hour rainfall accumulations in both data-sparse reanalysis-like and high-resolution operations-like experiments have been investigated using ECMWF's 4D-Var system.

Results clearly indicate that rain gauge assimilation can lead to significant improvements in global analysis and forecast scores when the coverage in other observation types is sparse (as would be the case in early-20th-century reanalyses). In contrast, when SYNOP rain gauges are assimilated together with the full modern-day coverage of surface, radiosonde and satellite observations, their impact on analysis and forecast quality remains much more modest, as expected.

A description of the method used to assimilate rain gauge measurements will be given, which will cover the issues of screening, error specification and bias correction. Examples of the impact on analysis and forecast scores will then be presented and finally remaining issues as well as the potential for future improvements will be addressed.

14 March 10:30

Aerosol-Cloud-Radiation Interactions and their Impact on ECMWF/MACC forecasts

Jean-Jacques Morcrette, ECMWF

Abstract

Prognostic aerosols were experimentally introduced in the ECMWF Integrated Forecasting System as part of the GEMS project in 2005. Their representation was refined as part of the MACC project, starting in 2009. Here, the MACC aerosol system is used to explore the impact of different levels of interactions between the aerosols and either the radiation and/or the cloud processes on cloudiness, radiation and precipitation fields, and on objective scores.

Ten-day forecasts including fully interactive aerosols are also compared to forecasts with aerosols specified from the analysis and kept constant thereafter.

Whereas the temporal variability of the prognostic aerosols is shown to have strong local effects on surface parameters, the impact on objective scores is much smaller.

21 March 10:30

Challenges of a sustained Climate Observing System

Kevin Trenberth, National Center for Atmospheric Research, Boulder

Abstract

Observations of planet Earth and especially all climate system components and forcings are increasingly needed for planning and decision making related to climate services in the broadest sense. Although significant progress has been made, much more remains to be done before a fully functional climate observing system exists. Observations are needed on all spatial scales from local to global, and all time scales, especially to understand and document changes in extremes. Climate change from human activities adds both a new dimension and an imperative: to acquire climate observations of sufficient quality and coverage, and analyze them into products for multiple purposes to inform decisions for mitigation, adaptation, assessing vulnerability and impacts, geo-engineering, and predicting climate variability and change and their consequences. A major challenge is to adequately deal with the continually changing observing system, especially from satellites and other autonomous platforms such as in the ocean. Even with new computational tools, further challenges remain to provide adequate analysis, processing, meta-data, archival, access, and management of the resulting data and the data products. As volumes of data continue to grow, so do the challenges of distilling information to allow us to understand what is happening and why, and what the implications are for the future.

22 March 10:30

The new Fortran coding standard for IFS

Mike Fisher, ECMWF

29 March 15:30

Project SimulAMV2: Using geostationary imagery from high resolution model simulations to improve the characterization of current Atmospheric Motion Vectors

Angeles Hernandez and Niels Bormann

Abstract

In this seminar we will present the main results of SimulAMV2, a 13-month project funded by EUMETSAT and carried out by the ECMWF and EUMETSAT, with the collaboration of CIMSS.

The overall objective of the project is to improve the characterization of Atmospheric Motion Vectors (AMVs) and their errors to improve the use of AMVs in Numerical Weather Prediction. This study approaches the analysis of AMV errors by using geostationary imagery generated from high resolution NWP model simulations. AMVs are derived from sequences of simulated images, using a derivation system similar to the one used operationally by EUMETSAT. The NWP model provides a "ground truth", which allows a detailed study of AMV errors, bypassing the usual difficulty of the scarcity of collocated observations of cloud and wind.

First, cloud structures from observed and simulated images will be compared; this is an important step, as findings from simulated imagery can be extended to observed imagery only if the cloud structures produced from model simulations are realistic. Then we will present evaluations of AMVs by comparing them to the model truth, first interpreting AMVs as point, single level observations of wind, and then as horizontal and vertical averages. We will also show results regarding horizontal, vertical and temporal correlations of errors, and finally we will discuss the role of clouds, focussing on the impact of vertical cloud profiles and cloud evolution on systematic AMV errors.

17 April 10:30

Ensemble forecasts of spring flood in Sweden

David Gustafsson and Jonas Olsson, SMHI

Abstract

Part 1: Snow-melt runoff predictions assimilating seasonal weather forecasts and ground penetrating radar measurements of snow water equivalent

Uncertainties in snow melt runoff predictions with regard to the amount of snow and timing of melt typically emerge from uncertainties in the meteorological input data. These uncertainties can be reduced by data assimilation to update the simulated snow water storage using snow survey data and remotely sensed snow extent data. The objectives of this study are (1) to evaluate seasonal snow and snow melt runoff simulated by the hydrological model HYPE when forced by the ECMWF seasonal weather forecasts, and (2) to evaluate to what extent the seasonal snow melt predictions can be improved by assimilation of observations of snow water equivalent at the time of peak accumulation and discharge observations during the snow melt period.

Snow and runoff observations were assimilated into the HYPE model using the Ensemble Kalman filter data assimilation method. The study was conducted using data from two mountain basin in northern Sweden over the winters 2007-2011. Distributed snow cover data was sampled using ground-penetrating radar (GPR) from snow mobiles at the time of the maximum snow cover.

Simulations for the years 2007-2011 were performed with meteorological forcing data either based on station data or seasonal weather prediction data, with and without assimilation of the SWE data, respectively. The results show that the assimilation of GPR-data strongly improves the simulated results of snow melt runoff, especially for the simulations driven by the seasonal weather forecasts. It is suggested that the assimilation of the snow and runoff data can be used in the development of the bias corrections of the seasonal weather forecasts.

Part 2: A comparison of different approaches for forecasting spring floods in Sweden and the feasibility of a multi-model forecast system

In this study, three different ensemble forecast approaches to spring flood forecasting were compared with the state-of-the-art operational hydrological model implemented at SMHI. The methods consisted of (1) a reduced historical ensemble approach, where analogue years from the historical dataset, (2) using seasonal forecasts from ECMWF and (3) statically downscaling large-scale circulation variables from ECMWF seasonal forecasts to accumulated discharge using Singular Value Decomposition. The different approaches were evaluated for forecasts issued on 1/1, 1/3 and 1/5 for the spring floods 2000-2010 in three Swedish rivers. The evaluation was mainly performed in terms of the mean absolute error (MAE) of accumulated discharge with the state-of-the-art forecast as a reference. MAE was reduced in two catchments using the different approaches, whereas no effect was seen in the third catchment. However, a combination of all methods gave the largest error reduction on average.

20 April 10:30

LCR

Medium-range and seasonal river forecasting in China

Dr. Liu Zhiyu and Ms. Yue Zhihui, Ministry of Water Resources, China

25 April 10:30

Assessing the impact of SST forcing on the forecast of past MJO events

Eric de Boisseson

Abstract

Several past Madden-Julian Oscillation (MJO) events are simulated using the ECMWF monthly forecasting system. Several ocean-atmosphere coupled experiments and atmosphere only experiments are conducted. The impact of the different strategies on the MJO forecast skill is investigated through diagnostics and additional experiments.

26 April 10:30

MR1

Optimization and utility of TAMDAR aircraft observations for NWP

Neil Jacobs, AirDat LLC, Morrisville, NC

Abstract

Lower and middle-tropospheric observations are disproportionately sparse, both temporally and geographically, when compared to surface observations. The limited density of observations is likely one of the largest constraints in numerical weather prediction. Atmospheric observations collected by a multi-function in-situ atmospheric sensor on aircraft, called the Tropospheric Airborne Meteorological Data Reporting (TAMDAR) sensor, contain measurements of humidity, pressure, temperature, winds aloft, icing, and turbulence, along with the corresponding location, time, and altitude from built-in GPS are relayed via satellite in real-time to a ground-based network operations center. The TAMDAR sensor was originally deployed in December 2004 on a fleet of 63 Saab 340s operated by Mesaba Airlines in the Great Lakes region as a part of the NASA-sponsored Great Lakes Fleet Experiment (GLFE).Over the last eight years, the equipage of the sensors has expanded beyond CONUS to include Hawaii, Alaska, Mexico, UK, and western Europe on 13 fleets of regional airlines. In addition to the standard commercial airline program, a miniaturized version of the sensor has been deployed on several unmanned aerial vehicles (UAVs). Upon completion of the 2012 installations, more than 7000 daily sounding will be produced globally. An overview will be provided on the status of the TAMDAR sensor network deployment and data availability, as well as an update on data quality, error statistics, and operational forecasting utility, both from soundings and various data assimilation techniques. Current data assimilation optimizations include splitting the ascent, descent, and cruise observations into different phases of flight, correcting for the magnetic deviation bias in heading instrumentation, and isolating wind speed versus wind direction errors.

2 May 10:30

Atmospheric predictability at the convective scale

Giovanni Leoncini, Met Office

Abstract

The current availability of computing power allows weather services, to run ensemble forecasts at resolutions that ensure at least a partial representation of convection and thunderstorms (i.e. the convective scale). The representation of severe weather is much more realistic at this scale and there is an increasing interest in this kind of forecast from emergency services, aviation and industry. However, at these scales (~ 1km and smaller) the predictability of the atmosphere is very different from the predictability at larger and better known scales of the order of 100 or 1000 km and it poses several fundamental challenges to the scientific community. The aim of this presentation is to provide an overview of such challenges, describing current research efforts and discussing future ones. The first part of the talk focuses on the basics differences between atmospheric predictability at the convective scale and larger ones. The second part is a quick overview of the current efforts and implementations at the major weather services. Finally I'll discuss future research directions.

9 May 15:30

What do good weather forecasters do - and how can they do it even better?

Anders Persson (SMHI)

Abstract

We are today in a situation where not only operational weather forecasters at the national meteorological services make use of our products but also an increasing number of meteorological consultants and private weather services, hydrologists and other non-meteorologists enthusiastically trying to make the best use of the enormous amount of freely and commercially available meteorological information from a multitude of sources. How can they make the best use of ECMWF products, in particular the Ensemble Prediction System?

The presentation takes off from the "ECMWF User Guide" and the discussion in the autumn "Newsletter". It further develops the notion that skilled weather forecasters, in the heat and frenzy of operational activity, in the name of "experience", make use of "intuitive statistics". A recent "best-seller" by the Nobel Laureate in economics, Daniel Kahneman, devoted to the problem of the use intuitive statistics", might provide valuable insights how a current intuitive work practice can be put on a more reasoned basis, in particular with respect to the statistically geared Ensemble Prediction System.

28 May 10:30

Report on stochastic physics progress

Glenn Shutts

29 May 14:00

Using climatological information in ensemble data assimilation through observations and through stochastic parametrizations

Dr Georg Gottwald, Sydney University, Australia

Abstract

We investigate how to incorporate climatological information in ensemble data assimilation schemes. This can be done either on the level of providing additional observational, or on the level of parametrized forecast models. In a first part, we consider the problem of an ensemble Kalman filter when only partial observations are available. For small ensemble sizes this leads to an overestimation of the error covariances. We show that by incorporating climatic information of the unobserved variables the variance can be controlled and superior analysis skill is obtained. We then employ this Variance Controlling Kalman Filter to control model error when the model is allowed to be void of stabilizing artificial numerical viscosity. In a second part, we consider a deterministic multiscale toy model in which a chaotic fast subsystem triggers rare transitions between slow metastable regimes, akin to weather or climate regimes in the context of climate dynamics. Using homogenization techniques we derive a reduced stochastic model as a stochastic parametrization model for the slow dynamics only. We show that the stochastic reduced model can outperform the full deterministic model as forecast model in an ensemble data assimilation procedure, in particular in the realistic setting when observations are only available for the slow variables. We relate the observation intervals for which skill improvement can be obtained to the time scales of the system. We then set out to explain why stochastic climate models produce superior skill in an ensemble setting. The improvement in skill is due to the finite size of the ensemble, and we show that there is no skill improvement in very large ensembles or when the forecast variance is artificially and unreasonable inflated. We corroborate this with numerical simulations. This is joint work with Lewis Mitchell and Sebastian Reich.

8 June 11:00

MR1

Correcting for conditional bias in hydrometeorological and hydrologic ensemble predictions: a nonparametric approach

James D. Brown, Hydrologic Solutions Limited, Southampton

Abstract

Ensemble forecasts generally contain biases in the mean, spread, and higher moments of their predictive probability distributions. Practical applications of hydrometeorological and hydrologic ensemble forecasts are sensitive to bias. For example, the tendency to overestimate light precipitation or streamflow and underestimate heavy precipitation or streamflow can reduce the value of hydrologic predictions for drought and flood forecasting. Statistical post-processors have traditionally focused on “calibrating” raw predictions, in order to improve reliability or reduce conditional bias given the predicted outcomes (Type-I conditional bias). However, the Type-II conditional biases are equally important in operational forecasting, as the impacts of decisions about weather and water variables are sensitive to observed outcomes (e.g. flooding being observed to occur, as opposed to simply forecast with given probability).

This paper evaluates a non-parametric technique for bias-correction and uncertainty estimation of hydrologic and hydrometeorological predictions. The technique employs Bayesian optimal linear estimation of indicator variables, and is analogous to indicator cokriging (ICK) in geostatistics. The coefficients of the predictors can be chosen to minimize the conditional error variance of the estimator (Brier Score) or, without loss of analytical tractability, a combination of the conditional error variance and the square bias conditional upon the observed outcome, i.e. Type-II conditional bias.

21 June 10:30

MR1

Hidden error variance theory and its potential use in Hybrid data assimilation

Craig Bishop, Naval Research Laboratory, Monterey

Abstract

A conundrum of predictability research is that while the prediction of flow dependent error distributions is one of its main foci, chaos fundamentally hides flow dependent forecast error distributions from empirical observation. Empirical estimation of such error distributions requires that one obtain a large sample of error realizations given the same flow and the same observational network. However, chaotic elements of the flow and the observing network make it practically impossible to observe and collect the conditioned sample of errors required to empirically define such distributions and their variance. These variances are “hidden”. Here, an exposition of the problem is developed from an ensemble Kalman filter data assimilation system applied to a 10 variable non-linear chaotic model and 25,000 replicate models. The output from this system motivates a new analytical model for the distribution of true error variances given an imperfect ensemble variance. This model is defined by 6 parameters that also determine the optimal weights for the static and flow dependent parts of Hybrid error variance models. Two of the 6 parameters are trivial to estimate. The other four parameters are comprised by the variance and minimum of the climatological distribution of true error variances, parameter that defines the mean response of ensemble variances to changes in the true error variance and the climatological minimum of true error variance. These parameters are hidden because they are defined in terms of the flow dependent forecast error variance– which is unobservable in systems exhibiting aperiodic chaos. Six new equations enable these hidden parameters to be accurately estimated from a long time series of (innovation or observation minus forecast, ensemble variance) data pairs. This new-found ability to estimate hidden parameters provides new tools for assessing the quality of ensemble forecasts, tuning Hybrid error variance models and for post-processing ensemble forecasts. Preliminary results from attempts to use the theory to speed the tuning of Hybrid data assimilation schemes will also be presented.

22 June 14:00

Classroom

EFAS and its links to ECMWF research

Florian Pappenberger, ECMWF

4 July 10:30

The Suomi NPP (SNPP) satellite was launched successfully on October 28, 2011 and is a pathfinder for the future US Joint Polar Satellite System (JPSS) operational satellite series. The primary objectives of the SNPP mission provide a continuation of the group of Earth system observations initiated by the Earth Observing System Terra, Aqua, and Aura missions; and prepare the operational forecasting community with pre-operational risk reduction, demonstration, and validation for selected JPSS instruments and ground processing data systems. The SNPP satellite is now flying with the following five instruments:

Visible/Infrared Imager/Radiometer Suite (VIIRS) has multi-band imaging capabilities to support the acquisition of high-resolution atmospheric imagery and generation of a variety of applied products including visible and infrared imaging of hurricanes and detection of fires, smoke, and atmospheric aerosols.
Cross-track Infrared Sounder (CrIS) is the the first in a series of advanced operational sounders that provide more accurate, detailed atmospheric temperature and moisture observations for weather and climate applications.
Advanced Technology Microwave Sounder (ATMS) operates in conjunction with the CrIS to profile atmospheric temperature and moisture. Higher (spatial, temporal and spectral) resolution and more accurate sounding data from CrIS and ATMS support continuing advances in data assimilation systems and NWP models to improve short-to medium-range weather forecasts.
Ozone Mapping and Profiler Suite (OMPS) measures the concentration of ozone in the atmosphere, providing information on how ozone concentration varies with altitude. Data from OMPS continue three decades of climate measurements of this important parameter used in global climate models. The OMPS measurements also fulfill the U.S. treaty obligation to monitor global ozone concentrations with no gaps in coverage.
Cloud and Earth Radiant Energy System (CERES) seeks to develop and improve weather forecast and climate models prediction, to provide measurements of the space and time distribution of the Earth's Radiation Budget components. The observations from CERES are essential to understanding the effect of clouds on the energy balance (energy coming in from the sun and radiating out from the earth), which is one of the largest sources of uncertainty in our modeling of the climate.

The Suomi NPP instruments are now undergoing a period of intensive calval and their performances in orbit are stable and meet or exceed the specifications. The SNPP SDR products are all reaching a provisional level at which users can order the data from NOAA archival and perform in-depth scientific research. NOAA is also operationally using the ATMS data in its global forecast system (GFS) and generating a suite of EDR products from the NPP ground system. During the past 8 months of intensive calval, NPP/JPSS SDR teams have developed many innovative techniques for characterizing the instrument performance (e.g. noise and bias), fixed numerous SDR processing bugs and improved the quality control flags in the data streams. The critical SNPP calval tasks have been completed and the most recent calval results will be reported in this presentation.

30 July 15:30

Fighting Hunger – How ECMWF Forecasting Platforms have been supporting WFP Operations

Thomas Petroliagis, ECMWF

Abstract

The World Food Programme (WFP) is the United Nations frontline agency in the global fight against hunger. WFP aims to get food to where it is needed as quickly as possible, saving the lives of victims of war, civil conflict and natural disasters. Being well prepared to respond to disasters is a top priority for WFP. To accomplish its mission WFP has been depending heavily on all ECMWF forecasting platforms (operational high-resolution forecast–ensemble prediction system–monthly and seasonal systems) and ERA–Interim reanalysis data. The graphical forecasting environment ecCharts is also utilized to fulfil WFP’s daily operational needs and requirements.

29 August 10:30

Bayesian Statistics – the Emperor's New Clothes or the Solution to Everything?

Anders Persson, SMHI

Abstract

For almost 200 years there has been intense controversy surrounding the Rev. Thomas Bayes (1702-61) simple formula for conditional probability

prob(A|B) = [prob(A)·prob(B|A)] / prob(B)

Until quite recently a minority among statisticians have declared themselves devoted “Bayesians” against a majority of fervent Anti-Bayesian "Frequentists". The seminar tries to explore the historical roots of this conflict, which shows sign of abating partly thanks to improved computer technology.

It will also suggest that a lot of Bayesian applications have been concealed in the literature, sometimes because the scientists wanted to avoid the “B-word”, sometimes because they didn’t know that they were applying Bayesian principles. The latter might be the case among meteorologists and hydrologists working with forecast model development.

8 October 14:00

Time-parallel algorithms for weather prediction and climate simulation

Jean Cote, ESCER Center, Earth and Atmospheric Sciences
Universite du Quebec a Montreal (UQAM)

Abstract

The forecast of weather relies on computer models that need to be executed in real-time, meaning that a forecast needs to be disseminated to users well before the time period for which it is made. A challenge in the future will be to succeed in using the computing power available in massively parallel high-performance computers and meet the real-time requirement. Until now weather forecast and related climate simulation models have taken advantage of the parallelism of the computers by dividing the task to be performed in the horizontal space dimensions.

The purpose of this work is to develop algorithms that allow also parallelism in the time dimension. This increased parallelism should allow an acceleration of the execution time of weather and climate models. This acceleration in turn permits an increase in the space accuracy of models while still meeting the real-time requirement. The talk will present our preliminary work on the "Parareal" algorithm that has been developed for that purpose (Lions et al., 2001) and whose applications to date have included among others air quality, but ignored weather forecasting.

Weather forecasting presents a challenge for the method because of the presence of waves and advection. An important question is to examine how the traditional way to accelerate models with the semi-implicit semi-Lagrangian methodology can be advantageously blended with the Parareal approach.

9 October 14:15

Seasonal forecasting of meteorological droughts

Emanuel Dutra. ECMWF

Abstract

Vast parts of Africa rely mainly on the rainy season for livestock and agriculture. Droughts can have a severe impact in these areas which often have a very low resilience and limited capabilities to mitigate their effects. The predictive capabilities of an integrated meteorological drought monitoring and forecasting system based on the Standard precipitation index (SPI) are investigated. The system is firstly constructed by temporally extending near-real time precipitation fields with forecasted fields as provided by the ECMWF seasonal forecasting system and then is evaluated over four basins in Africa: the Blue Nile, Limpopo, Upper Niger, and Upper Zambezi.

This skill and reliability depends strongly on the SPI time-scale, and more skill is observed at larger time-scales. The ECMWF seasonal forecasts have predictive skill which is higher than using climatology for most regions. Poor quality precipitation monitoring products can reduce the potential skill of SPI seasonal forecasts in two to four months lead time.

10 October 10:30

MR2

The Science and Practice of Operational River Forecasting Systems: Lessons Learned from a World Tour

Tom Pagano (CSIRO)

Abstract

Operational river forecasts have been long produced to support water resources management, covering a range of timescales from flash flooding (e.g. minutes to hours ahead) to seasonal (e.g. months ahead). They are generated by a range of statistical (e.g. regression-based) and dynamical (e.g. rainfall-runoff) model based techniques.

The extent of human control over the forecasts, such as by adjusting model inputs and outputs, varies from nearly completely automated systems to those where forecasts are generated after discussion among a group of experts. Historical and real time data availability, the complexity of hydrologic processes, forecast user needs, and forecasting institution support/resource availability (e.g. computing power, money for model maintenance) influence the character and effectiveness of operational forecasting systems. Although it is an active research topic, nearly all operational forecasting systems struggle to make quantitative use of Numerical Weather Prediction model-based precipitation forecasts.

Similarly, global-scale and spatially distributed hydrologic information, informed by remotely sensed measurements (e.g. radar and satellite) are available experimentally but suffer from biases and other limitations. Forecast uncertainty is commonly estimated operationally by using an ensemble of future precipitation scenarios and/or a measure of historical model error. However, probabilistic forecast communication and use remains a stumbling block for many. The flexibility of the water manager’s operating procedures and the size of the reservoir relative to incoming streamflow determine the relevance of forecasts to hydropower operations; moderately-sized reservoirs that can draw down in anticipation of high flows or slow releases in anticipation of drought stand to benefit the most.

11 October 14:15

Using ECMWF long range forecast for health applications: The prototype Malaria Early Warning System of ECMWF and ICTP (MEWS)

F. Di Giuseppe, ECMWF

Abstract

Temperatures determine the life cycle development rates of mosquitos that spread malaria and the malaria parasite itself.

Rainfall is required for the mosquito larvae. Skilful malaria predictions thus require skilful climate predictions weeks to months ahead, to drive malaria models that incorporate both climate and non-climate factors such as population density and migration, host immunity, treatment and interventions.

In the last years ECMWF have developed a latest generation of the monthly and seasonal forecast system that has predictive skill in the tropic weeks ahead. Building of these new tools a new seamless prediction system which concatenates bias corrected outputs from the monthly and seasonal forecasts has therefore been developed for the purposes of driving sectoral applications. ICTP has developed the VECTRI dynamical malaria model that accounts for climate, surface hydrology and population.

Within the framework of the EU-FTP7 projects QWECI, ECMWF and ICTP are working with partners in the Ministry of Health in Malawi and Uganda to evaluate and validate the MEWS for past outbreaks in highlands areas, before rolling out the pilot MEWS system.

A description of the prototype Malaria Early Warning System of ECMWF and ICTP (MEWS) along with a preliminary performance evaluation is presented.

25 October 10:30

Classroom

Soil moisture validation activities at ECMWF

Clement Albergel

Abstract

After a brief overview of the different available soil moisture products, the validation methodology is presented. Then in situ soil moisture measurements for more than 200 stations from five networks across the world (USA, France, Spain, China and Australia) are used to determine the reliability of three soil moisture products:

a revised version of ERA-Interim reanalysis from ECMWF
a revised version of MERRA reanalysis from NASA-GMAO and
a new multi-satellite surface soil moisture dataset for 2007–2010.

Also, global trends in soil moisture from the revised versions of ERA-Interim and MERRA were analysed for the period 1988–2010.

26 October 10:30

Classroom

A comparison of 4D-Var and an Ensemble Kalman Filter method (LETKF) for assimilation of surface pressure observations

Pau Escribà, AEMET and ECMWF

Abstract

ECMWF has recently developed an Ensemble Kalman Filter (EnKF) for experimental purposes. This enables us to do a fairly clean comparison of 4D-Var and EnKF, and to compare background error estimates from the Ensemble of Data Assimilations (EDA) and EnKF. One of the configurations implemented is the Local Ensemble Transform Kalman Filter (LETKF). Several methods for additive and multiplicative inflation and localization of observations have been implemented. Because of the assumption about linearity in the LETKF equations, a six hours assimilation window is used to ensure that the background ensemble propagation is reasonably linear.

In this seminar five of the analysis systems available at ECMWF are compared for assimilations that only use surface pressure observations. This has relevance for the on-going ECMWF 1900-2010 reanalysis among other activities. It also provides a more stringent test of the quality of background error estimates. The five methods are LETKF, 4DVAR with static background error using 12-hour and 24-hour assimilation windows, hybrid 4DVAR-EDA, and hybrid 4DVAR-LETKF. The comparison looks at the quality of the five analysis states, taking as the truth the operational 4DVAR analysis that runs at ECMWF with the full observing system.

Initial results for a Kalman Smoother LETKF extension will be presented. This technique recomputes the analysis at one assimilation on cycle by using the weights computed in the next cycle. Using the Kalman Smoother LETKF assimilation method allows the analysis to benefit from observations in the subsequent 6 hours, in a way so the assumption about linearity is respected.

Using only surface pressure data in the assimilation, along with the fact that the IFS forecast model used is the same for all the analysis schemes, ensures a quite clean comparison between LETKF and the variational techniques. This avoids the problems related to localisation for satellite radiances which is still an issue for EnKF systems.

This work was performed during a six month period in 2012 when Pau was a visiting scientist in the Research Department. The work was done in collaboration with Massimo Bonavita, Mats Hamrud, Lars Isaksen and Paul Poli.

31 October 10:30

MR1

Water resources in France : reanalysis, trends, seasonal forecasting and climate projections

Jean-Philippe Vidal, Irstea, formerly CEMAGREF

Abstract

This talk will present the outputs from a set of recent research projects related to the assessment of water resources, droughts and low flows in France. First, past trends in low flow characteristics for a benchmark network of stations have been studied and linked to large-scale climate indices: NAO, AMO, and weather types. Second, a seasonal predictability study has been set up over the whole of France in order to assess the potential for forecasting soil moisture and river flows in spring and summer with the Safran-Isba-Modcou (SIM) hydrometeorological suite. Third, a spatio-temporal reanalysis and characterisation of meteorological, agricultural and hydrological droughts in France over the last 50 years has been performed with the same SIM suite. The evolution of such spatio-temporal characteristics under downscaled climate projections has then been studied, by considering a range of emissions scenarios and theoretical adaptation scenarios. Some research outlook linked to the R2D2 project, a fully integrated global change project focused on the Durance catchment (Southern Alps), will close the talk.

12 November 14:00

Impact of various observing systems on NWP

Erik Andersson, ECMWF

Abstract

The 5th WMO workshop on the impact of various observing systems on NWP was held 22-25 May 2012 in Sedona, Arizona. The WMO expert team on the evolution of the Global Observing System (ET-EGOS) had proposed topics for impact studies and participants were encouraged to present results on those topics in particular. The workshop was attended by 59 experts on data assimilation and observation impact, representatives from space agencies and managers of observing networks from 13 countries. Several of the global NWP centres presented detailed assessments of the impact on forecast skill of the main meteorological observing systems. In this talk I will present the main results and what they mean in terms of recommendations to WMO, space agencies and other data producers.

14 November 10:30

A novel variable resolution global spectral method on the sphere

Dr S. Janakiraman, Centre for Development of Advanced Computing, (C-DAC), Bangalore, India

15 November 10:30

Classroom

Externalization of observation screening tasks within the context of the COPE project

Tomas Kral

Abstract

Presently, most of the observation screening tasks, including quality control, observation error assignment and redundancy checks, are executed in the time critical path of the first IFS trajectory run. In this talk we will describe a new observation processing system developed within the context of the COPE project and explain its benefits for the future observation related applications. It will be discussed how these changes will simplify the design and coding of observation related software in ECMWF's assimilation system.

16 November 10:30

MR1

Australia's water information initiative and achievements to date

Dr Dasarath (Jaya) Jayasuriya, A/g Deputy Director Climate and Water Bureau of Meteorology, Melbourne, Australia

Abstract

To address the threat of future water scarcity, the Australian Government in 2007 decided to invest $12.9 billion in Water for Future, a 10-year plan to secure long-term water supply to all Australians. The plan will help secure water supplies for Australian households, businesses and farmers, and provide water to restore the health of Australia’s stressed river systems.

Through the Commonwealth Water Act 2007, the Australian Government has given the Bureau of Meteorology responsibility for compiling and delivering comprehensive water information across the water sector. As part of Water for the Future, the Bureau has been allocated $450 million to administer the Improving Water Information Program. To achieve this, the Bureau is working with water managers across Australia to deliver high quality, national water information to government, industry and the community. The Bureau will deliver a range of products designed to meet the needs of users engaged in emergency services, water policy development, planning, operations, public enquiry, education and research including:

Flood warnings and forecasts
Annual national water resources assessments
An annual National Water Account
Real-time water reporting services
Real-time and extended water availability forecasts; and
Online, single-point, free access to water data collected across Australia.

Improved accessibility, integration and use of national water resources information will yield huge benefits. The integration of data sets, their improved quality and currency, and the introduction of new reporting, analysis and forecasting products and services will lead to more informed policy and infrastructure decisions. They will also enable evaluation of water sector reforms, leading to greater confidence in Australia’s water management.

Dr Jayasuriya will be introducing the Bureau’s Water Information Initiative and sharing outcomes to date with the Audience. He will conclude the presentation by sharing future challenges and the approach adopted by the Bureau to meet these head-on.

22 November 15:30

Challenges in designing a multi-purpose Nonhydrostatic Unified Model of the Atmosphere (NUMA)

Francis Giraldo, Department of Applied Mathematics Naval Postgraduate School, Monterey, US

Abstract

In this talk, I would like to discuss the challenges faced by operational centers in designing unified regional/global atmospheric models. In particular, I will describe the numerical methods that our model relies on which include: high order continuous and discontinuous Galerkin methods, implicit-explicit time-integrators, iterative solvers,pre-conditioners, and adaptive mesh refinement.

I would like to discuss the pros and cons of each of these methods and explain why we have selected these methods to build a new model. I will end a talk with a timeline for the model that is currently under development.

For details please visit the project website.

26 November 14:00

Predicting high-impact weather based on the COSMO-DE ensemble prediction system

Dr Petra Friederichs, University of Bonn

Abstract

Extremes in weather such as high wind speeds or heavy precipitation are associated with high impacts on both environment and society. It is thus of great importance to improve the quality of extreme wind and heavy precipitation forecasts. Numerical weather forecast models provide information about the general conditions in which extremes occur. However, strength, timing and location of such extremes are determined by small scale processes that are not, or only partly, resolved even by high-resolution weather forecast models. Thus, probabilistic prediction is likely to be the best choice of forecasting high-impact weather.

The study uses a mesoscale ensemble prediction system based on COSMO-DE (COSMO-DE-EPS) operated by the German Meteorological Service. The sources of uncertainty and predictability-initial and boundary conditions, and physical parameterizations-within the COSMO-DE-EPS are investigated, and the influence of the uncertainty components on the predictive performance is assessed including a scale dependent spread analysis.

We will present ensemble postprocessing that provides area-wide forecasts of extreme wind and heavy precipitation. We show that fairly simple postprocessing for precipitation provides considerable improvement of the predictive performance. Finally, a spatial Bayesian postprocessing for wind gusts is presented that employs extreme value theory and aims at providing a complete spatial statistical model for the wind gust process.

5 December 10:30

Satellite observations of dust and their assimilation at the Met Office

Dr Bruce Ingleby, UK MetOffice, Exeter

Abstract

At the Met Office mineral dust has been included in the operational global forecast model for just over a year-but without any assimilation. We have been testing a system in which Aerosol Optical Depth (AOD) values are obtained from two satellites and assimilated. Retrievals from the SEVIRI instrument on Meteosat Second Generation (MSG) are performed at the Met Office using a method proposed by Brindley and Russell. Retrievals from the MODIS instrument on the Aqua satellite are obtained from GSFC (USA). Only daytime AOD values over land are currently assimilated. An early assimilation experiment omitted low AOD values on the basis that other aerosols might make up a large fraction of the AOD-however this was found to give a high bias to the analysis values, it is better to use most of the SEVIRI AOD values over Africa. We have also modified different aspects of the background error estimates.

Overall the results of data assimilation trials show an improved short range fit to satellite AOD and to the sparse AERONET network-by day five of the forecast the benefit has largely disappeared. Operational implementation is planned for March 2013. The results and the major issues will be discussed as will future development options.

10 December 10:30

Stochastic processors for modelling weather and climate

Dr Peter Dueben, Dept of Physics, Univ Oxford

Abstract

Today’s processors are designed to calculate all operations exactly. If we can tolerate errors of processors it is possible to increase the performance and reduce the power consumption of a given processor significantly. The use of faulty, or stochastic processors, could therefore bring us closer towards global cloud resolving models in weather and climate science.

To explore the use of stochastic processors we emulate processor errors in numerical simulations of toy models, such as Lorenz63, but also in a spectral dynamical core (IGCM). Since a spectral core allows scale separation between large and small scales, we apply the stochastic processors to the small scales that possess inherent uncertainty due to insufficient resolution and parametrization, while we calculate the crucially important large scales on exact processors. Results are promising.

19 December 11:00

A new formulation of the spectral energy budget of the atmosphere with application to two high-resolution general circulation models

Erik Lindborg (Department of Mechanics - KTH, Stockholm, Sweden)

Abstract

A new formulation of the spectral energy budget of kinetic and available potential energies of the atmosphere is derived, with spherical harmonics as base functions. Compared to previous formulations, there are three main improvements:

the topography is taken into account,
the exact three-dimensional advection terms are considered and
the vertical flux is separated from the energy transfer between different spherical harmonics.

Using this formulation, results from two different high resolution GCMs are analyzed: the AFES T639L24 and the ECMWF IFS T1279L91. The spectral fluxes show that the AFES, which reproduces realistic horizontal spectra with a k^{-5/3} inertial range at the mesoscales, simulates a strong downscale energy cascade. In contrast, neither the k^{-5/3} vertically integrated spectra nor the downscale energy cascade are produced by the ECMWF IFS.

2011

16 February 10:30

Land Surface Analysis at ECMWF: soil moisture and snow, some evaluations

Clement Albergel

22 February 10:30

Does higher resolution provide more accurate forecasts?

Nigel Roberts, Met Office, JCMM, Reading

Abstract

The Met Office now routinely runs a Numerical Weather Prediction models with a grid spacings of 4 and 1.5km. At these 'storm-permitting' resolutions convective showers can be explicitly represented on the model grid and there is also the capability to represent small-scale meteorological features such as local convergence lines, narrow frontal rainbands and valley fog.

The forecasts from the 1.5km (UKV) model in particular, are astonishingly realistic, but are they accurate? There are certainly high expectations along with increased computational cost.

This talk will examine whether increased resolution produces more skilful forecasts and consider how this determines the way we should present the output (at all resolutions).

3 March 14:00

Phase 2 of the IBM HPCF (POWER7)

Isabella Weger

Abstract

Phase 2 of the IBM HPCF contract, based on POWER7 technology, will be installed in 2011. A performance upgrade in mid 2012 is part of a contract extension until mid 2014 which was concluded with IBM in late 2010.

The presentation will give an overview over the new POWER7 system, outlining system configuration, technical specifications and time lines for implementation of the Phase 2 system.

4 March 10:30

An investigation of systematic model error using a quasi-geostrophic weak-constraint 4D-Var system

Harri Auvinen

Abstract

Even the most sophisticated numerical weather prediction model is not perfect. Although some errors may be entirely random, many will result from systematic misspecification of parameters, or deficiencies in parametrizations. In this work we estimate the covariance structure of systematic error a quasi-geostrophic (QG) model using the OOPS assimilation framework. We use this estimate in a Long-Window Weak-Constraint 4d-Var analysis system, and demonstrate that method is successful in taking model error into account.

The Surface Water Ocean Topography satellite mission: prospects for climatology and hydrology science

Prof. Paul Bates, Bristol University

Abstract

In 2019 NASA and CNES will launch the Surface Water Ocean Topography satellite mission (see swot.jpl.nasa.gov) which will measure surface water height to centimetric accuracy every 10 days with complete global coverage at 1km resolution over the oceans and for all rivers above 100m wide. Over oceans, current altimeter constellations can only resolve the ocean circulation at resolutions larger than 300 km. Fundamental questions on the dynamics of ocean variability at scales shorter than 300 km, the mesoscale and submesoscale processes, such as the formation, evolution, and dissipation of eddy variability (including narrow currents, fronts, and quasi-geostrophic turbulence) and its role in air-sea interaction, are to be addressed by the new observations. Over land the SWOT mission will provide measurements of water storage changes in terrestrial surface water bodies and will provide estimates of discharge in large (50 m-100 m width) rivers, globally. These data sets will provide a quantum improvement on our current understanding of surface water dynamics. Beyond improved characterization of the water cycle, meeting these measurements will enable numerous applications of great scientific, social, and political importance.

25 May 10:30

The new "User Guide to ECMWF forecast products"

Anders Persson

Abstract

Abstract: Since 1988 the ECMWF has provided a "User Guide" for users interested in an overview of our forecast system, forecast products and how to make best use of them. The 2011 edition is thoroughly revised and the main topics will be presented:

In the introduction users are recommended not to try to compete with the computer but do the o p p o s i t e : not be detailed, avoid forecast "jumpiness" and, above all, make use of uncertainty.
When presented the forecast and data assimilations system users tend to be more interested in the deficiencies of the system rather than to be reminded how good it is. The same applies to the EPS.
The use of the deterministic and probabilistic products on their own and in combination. To what degree are the users able to modify the output?
A know-how of verification and validation methods and, above all, their interpretation, is important not only for specialists, but also for NWP modellers and operational forecasters.
In an Epilogue recommendations are made in an attempt to narrow the gap between mathematical-scientific rigour of NWP modelling with the empirical-intuitive practises of operational forecasting.

2 June 10:30

A new method for estimating analysis and forecast error variance

Zoltan Toth, GSD/ESRL/OAR/NOAA, Boulder, CO. and Malaquias Pena, EMC/NCEP/NWS/NOAA

Abstract

Accurate estimates of error variances in numerical analyses and forecasts are critical for the evaluation of forecasting systems, the tuning of data assimilation systems, and the proper initialization of ensemble forecasts. A number of issues, however, hinder related efforts. A new approach is introduced for the unbiased estimation of analysis and forecast errors. The method is independent of any assumption or tuning parameter used in data assimilation schemes. In a functional analysis, it combines information from differences between forecast and analysis fields (“perceived errors”) with prior knowledge regarding the time evolution of (a) forecast error variance and (b) correlation between errors in analyses and different lead-time forecasts. In a simulated forecast environment, the method is demonstrated to reproduce the true analysis and forecast error within the predicted error bounds. The method is then applied to forecasts from four leading Numerical Weather Prediction centers to assess the performance of their corresponding data assimilation and modeling systems.

7 June 09:15

Classroom

Observation Handling and Monitoring Tutorial

Drasko Vasiljevic, Anne Fouilloux, Manuel Fuentes, Peter Kuchta, Mohamed Dahoui and Sandor Kertesz

5 July 10:30

Classroom

On a Gain Initialization and Optimization of Adaptive Filter by Simultaneous Perturbation Stochastic Approximation

Hong Son Hoang, Service Hydrographique et Oceanographique de la Marine, Toulouse

Abstract

Initialization of the filter gain plays an important role to provide a high performance of the filter. In this work, it is shown that the error covariance matrix (ECM) of the prediction error can be well estimated using simulated samples of the dominant Schur vectors of the system dynamics. This allows to well initialize the filter gain. By an appropriate parametrization, the adaptation can be applied to minimize the prediction error of the system output. It will be demonstrated that one efficient tool to optimize the performance of an AF is the simultaneous perturbation stochastic approximation (SPSA) algorithm. The main results show that the SPSA is capable of yielding the high filter performance similar to that produced by classical optimization algorithms, with better performance for non-linear filtering problems as more and observations are assimilated. The advantage of the SPSA is that at each iteration it requires only two measurements of the objective function to approximate the gradient vector regardless of the dimension of the control vector. The SPSA approach is thus free from the need to develop a discrete adjoint of tangent linear model and offers promising perspectives on developing optimal assimilation systems encountered in the field of data assimilation in meteorology and oceanography. Numerical results on data assimilation based on MICOM and HYCOM ocean models, in academic and realistic configurations, will be given to show the efficiency of the proposed filter.

7 July 10:30

Use of ensemble forecasts in predicting the supply and demand of European energy

Warwick Norton, Cumulus Funds PCE Investors, London

Abstract

The use of ECMWF ensemble forecasts is illustrated from the perspective of an end user who is a risk taker in the European energy market. This involves not only forecasting temperatures which drives consumption of energy, but the growing renewables sector means there is also large weather sensitivity in the supply of energy. This includes forecasting wind, surface solar radiation, precipitation and runoff. In order to calculate future energy prices, it is necessary to have an evolving view, at daily resolution, of these weather variables. This is accomplished by combining medium range forecasts with monthly forecasts and then with reanalysis data. However decision making using ensemble forecasts is not always straightforward because of the day-to-day changes in the forecasts. Some issues in forecasting these variables is discussed as well as the general performance of the EPS system over recent months.

13 July 9:15

Training sessions on C++ object oriented programming techniques: Effective Advanced C++

Michael Wong, IBM

Abstract

This course will focus on effective C++ design based on the last 20 years of collective wisdom of C++ gurus worldwide. It will discuss coding guidelines, as well as how C++ is implemented under the cover to help truly understand what features costs. Topics of interest will include how language features are implemented by the C++ compiler, understanding inlining, code bloat, interface-based programming, and dynamic memory management.

13 July 9:15

Training sessions on C++ object oriented programming techniques: Trends and futures of C++

Michael Wong, IBM

Abstract

Over the last decades, C++ has become one of the most widely used languages supporting object-oriented programming by making abstraction techniques affordable and manageable for mainstream projects. Using C++ as his tool, Prof. Bjarne Stroustrup pioneered the use of object-oriented and generic programming techniques in application areas where efficiency is a premium; examples include general systems programming, switching, simulation, graphics, user-interfaces, embedded systems, and scientific computation. The influence of C++ and the ideas it popularized are clearly visible far beyond the C++ community. Languages including C, C#, Java, and Fortran99 provide features pioneered for mainstream use by C++.

C++ is a general purpose programming language with a bias towards systems programming which supports multiple styles of programming. Since 1998 C++ is an ISO standard. In these days a new version of the language is about to become a new ISO standard. This seminar will start by introducing the new features of the language and showing the challenges of evolving an existing programming language with a huge code base. This will enable students to truly come to understand C++ and the foundation of its design. We will cover some language concepts but also expand to describe aspects of STL, Boost and C++0x, the next version of C++.

19 July 14:00

MR1

Impact of the European Russia drought in 2010 on the Caspian Sea Level

Klaus Arpe

Abstract

The Caspian Sea (CS) basin has no outlet to the oceans. The inflow from rivers, mainly the Volga River is compensated by evaporation over the Caspian Sea itself, imbalances lead to changes in the CS Level (CSL). It is therefore an ideal test area for investigating the water budget of an area using a variety of observational and reanalysis data. The period 1993 to 2010 was investigated with emphasis on summer 2010 when a severe drought developed over European Russia, the catchment of the Volga River. Precipitation and evaporation from the ECMWF interim analysis are compared with Volga River discharge and the CSL observations.

A drop in precipitation over the Volga basin (VB) in July 2010 occurs simultaneously with a decrease in evaporation for the same area, an increase of evaporation over the CS itself and a drop of the CSL. The drop in the precipitation over the VB cannot have led to an instantaneous drop of the CSL because the precipitated water needs some months to reach the CS. However, the evaporation over the CS itself is considered to be responsible for a simultaneous drop of the CSL in July to September 2010. The impact on the CSL from the precipitation deficit over the VB occurs in the months following the drought. The water deficit for June to September 2010 calculated from the anomalous precipitation minus evaporation over the VB would decrease the CSL by 22 cm, of which only 2.5 cm had been observed until end of September (observed Volga River discharge anomaly), 7 cm from October to the end of 2010 and another 5 cm to the end of May 2011. The remaining 7 cm may be compensated by excessive precipitation from October to February. In previous studies the precipitation over the VB has been identified as the main cause for CSL changes, but here a 10 cm drop from beginning of July to end of September can be directly assigned to the enhanced evaporation over the CS itself (6 cm) and due to reduced precipitation over the CS (2cm).
Further periods with strong changes of the CSL are investigated as well which provide some estimates concerning the accuracy of the data.
The consistency between the different components of the water budget over the Caspian Sea basin based on independent data gives a high confidence in the quality of the ERA interim data.

This investigation provides some scope for making forecasts of the CSL few months ahead to allow for mitigating societal impacts.

20 July 10:30

Classroom

Decadal climate variability and change in the Mediterranean region

Annarita Mariotti, NOAA, USA

Abstract

The Mediterranean region is among the “Hot Spots” projected to experience major climatic changes in the twenty-first century as a result of the global increase in greenhouse gas (GHG) concentrations. However the way in which these changes may initially become manifest in the Mediterranean will also depend on internal decadal variability and its impacts on climate in this region. Here, we present an analysis of the main decadal climate variations that have influenced past climatic conditions in the Mediterranean/South Europe region since the mid-nineteenth century. Decadal variability is discussed in the context of forced climatic changes from increased GHG. Results point to significant connections between Mediterranean climate and decadal and multi-decadal variability in the Atlantic. Namely, a significant influence of the North Atlantic Oscillation on Mediterranean precipitation and a relationship between regional temperatures and the Atlantic Multi-decadal Oscillation which may imply a certain degree of decadal regional predictability. CMIP3 projections indicate that in the longer term “forced” regional climatic changes from GHG increases would bring significantly drier conditions over land and major changes in Mediterranean Sea water cycle.

5 September 15:30

Variational data assimilation without nonlinear models

Roel Stappers and Jan Barkmeijer, KNMI

Abstract

The use of tangent linear and, in particular, adjoint models has been very useful in several applications in numerical weather prediction. For example, at ECMWF these linear models play a crucial role in the computation of initial condition perturbations used in the ensemble prediction system and in their 4D-VAR data assimilation system. The greatest limitation to the application of linear models is that the results are useful only when the linear approximation is valid. As a result the usage of tangent linear and adjoint models is restricted to 'short' time spans.

In this presentation we show that by linearizing the tangent linear model around an ensemble of trajectories the nonlinear growth of perturbations can be described exactly by the tangent linear model. We show that in a three level quasi geostrophic model the tangent linear model can be used for more than 200 days (with increments much larger than typical analysis increments, using only a single linearization trajectory).

Based on this result we introduce a new incremental 4D-VAR method that

Does not require the nonlinear model to update the linearization trajectory in the outer loops (these updates are responsible for a significant fraction of the computational cost in the current ECMWF implementation).
Does not modify the innovation vector in the outer loops. To demonstrate the advantage of this we derive the exact equations for adjoint based observation impact including the effect of using multiple outer loops in 4D-VAR.
Does not require the strict distinction between inner/outer loops. The linearization trajectory can be updated in the inner loops at no additional computational cost.

28 September 10:30

Need for Caution in Interpreting Extreme Weather Statistics

Prashant. D. Sardeshmukh, University of Colorado, Boulder and NOAA

11 November 10:30

MR1

Earth's Radiation Imbalance from a Constellation of 66 Iridium Satellites

We investigate the performance of several sound-proof models, including the anelastic Lipps-Hemler and the pseudo-incompressible Durran nonhydrostatic equations. Physics wise, our primary interests are with the dynamics of inertia-gravity waves, an important element of weather and climate. Our numerical developments are based on a class of non-oscillatory forward-in-time methods and are applicable to global and limited area models. Challenging simulations of atmospheric wave phenomena involve structured-grid and unstructured-mesh discretizations.