scholarly journals Spatially coherent postprocessing of cloud cover ensemble forecasts

2021 ◽  
Author(s):  
Y. Dai ◽  
S. Hemri
Keyword(s):  
Cloud Cover ◽  
Weather Prediction ◽  
Generative Adversarial Networks ◽  
Dependence Structure ◽  
Ensemble Forecasts ◽  
Adversarial Networks ◽  
Probabilistic Forecasts ◽  
Ensemble Model Output Statistics ◽  
Model Output Statistics ◽  
Modern Machine

AbstractStatistical postprocessing is commonly applied to reduce location and dispersion errors of probabilistic forecasts provided by numerical weather prediction (NWP) models. If postprocessed forecast scenarios are required, the combination of ensemble model output statistics (EMOS) for univariate postprocessing with ensemble copula coupling (ECC) or the Schaake shuffle (ScS) to retain the dependence structure of the raw ensemble is a state-of-the-art approach. However, modern machine learning methods may lead to both, a better univariate skill and more realistic forecast scenarios. In this study, we postprocess multi-model ensemble forecasts of cloud cover over Switzerland provided by COSMO-E and ECMWF-IFS using (a) EMOS + ECC, (b) EMOS + ScS, (c) dense neural networks (dense NN) + ECC, (d) dense NN + ScS, and (e) conditional generative adversarial networks (cGAN). The different methods are verified using EUMETSAT satellite data. Dense NN shows the best univariate skill, but cGAN performed only slightly worse. Furthermore, cGAN generates realistic forecast scenario maps, while not relying on a dependence template like ECC or ScS, which is particularly favorable in the case of complex topography.

Spatially coherent postprocessing of cloud cover forecasts using generative adversarial networks

2021 ◽  
Author(s):  
Yinghao Dai ◽  
Stephan Hemri
Keyword(s):  
Cloud Cover ◽  
Physical Dependence ◽  
Weather Prediction ◽  
Generative Adversarial Networks ◽  
Dependence Structure ◽  
Ensemble Forecasts ◽  
Generative Adversarial Network ◽  
Adversarial Network ◽  
Ensemble Model Output Statistics ◽  
Model Output Statistics

<p>Despite considerable improvements over the last few decades, numerical weather prediction (NWP) models still tend to exhibit bias and dispersion errors. Statistical postprocessing reduces these errors and allows quantifying predictive uncertainty. However, classical postprocessing approaches such as ensemble model output statistics (EMOS) destroy any physical dependence structure of the NWP raw ensemble forecasts. Ensemble copula coupling (ECC) is a commonly used state-of-the-art method to map the spatio-temporal dependence structure of the raw ensemble to the postprocessed predictive distributions. However, if the variable of interest exhibits many ties, ECC may not be optimal. Here, the variable investigated is hourly cloud cover over Switzerland. The climatological distribution of cloud cover shows considerable point masses at both zero and one, hence ties are a major issue when it comes to applying ECC. </p><p>We compare a variant of ECC, which is tailored to variables with many ties, applied to postprocessed forecast ensembles obtained by either EMOS or a dense neural network (dense NN) with postprocessed scenarios generated by a conditional generative adversarial network (cGAN). In particular, cGANs are appealing as they directly generate maps of postprocessed cloud cover forecast scenarios without the need of any dependence template. We trained the postprocessing models for COSMO-E and ECMWF IFS raw ensemble forecasts against hourly EUMETSAT CM SAF satellite data with a spatial resolution of around 2 km over Switzerland. For all the approaches, EMOS, dense NN, and cGANs, basic setups with a minimal set of raw ensemble predictors already allowed us to obtain a significantly better univariate performance (in terms of continuous ranked probability score) than the raw NWP ensembles. We present and discuss the advantages and drawbacks of EMOS+ECC, dense NN+ECC, and cGANs with respect to both univariate forecast skill and the ability to produce realistic cloud cover forecast scenario maps. </p>

Towards operational postprocessing of cloud cover at MeteoSwiss

2020 ◽  
Author(s):  
Stephan Hemri ◽  
Christoph Spirig ◽  
Jonas Bhend ◽  
Lionel Moret ◽  
Mark Liniger
Keyword(s):  
Cloud Cover ◽  
Prediction Models ◽  
Weather Prediction ◽  
Weather Forecast ◽  
Forecast Errors ◽  
Observation Data ◽  
Model Output ◽  
Ensemble Forecasts ◽  
Direct Model ◽  
Ensemble Model Output Statistics

<p>Over the last decades ensemble approaches have become state-of-the-art for the quantification of weather forecast uncertainty. Despite ongoing improvements, ensemble forecasts issued by numerical weather prediction models (NWPs) still tend to be biased and underdispersed. Statistical postprocessing has proven to be an appropriate tool to correct biases and underdispersion, and hence to improve forecast skill. Here we focus on multi-model postprocessing of cloud cover forecasts in Switzerland. In order to issue postprocessed forecasts at any point in space, ensemble model output statistics (EMOS) models are trained and verified against EUMETSAT CM SAF satellite data with a spatial resolution of around 2 km over Switzerland. Training with a minimal record length of the past 45 days of forecast and observation data already produced an EMOS model improving direct model output (DMO). Training on a 3 years record of the corresponding season further improved the performance. We evaluate how well postprocessing corrects the most severe forecast errors, like missing fog and low level stratus in winter. For such conditions, postprocessing of cloud cover benefits strongly from incorporating additional predictors into the postprocessing suite. A quasi-operational prototype has been set up and was used to explore meteogram-like visualizations of probabilistic cloud cover forecasts.</p>

scholarly journals Calibrated Ensemble Forecasts Using Quantile Regression Forests and Ensemble Model Output Statistics

2016 ◽  
Vol 144 (6) ◽  
pp. 2375-2393 ◽  
Author(s):  
Maxime Taillardat ◽  
Olivier Mestre ◽  
Michaël Zamo ◽  
Philippe Naveau
Keyword(s):  
Quantile Regression ◽  
Weather Forecasting ◽  
Real Data ◽  
Ensemble Model ◽  
Model Output ◽  
General Measure ◽  
Ensemble Forecasts ◽  
Probabilistic Forecasts ◽  
Ensemble Model Output Statistics ◽  
Model Output Statistics

Abstract Ensembles used for probabilistic weather forecasting tend to be biased and underdispersive. This paper proposes a statistical method for postprocessing ensembles based on quantile regression forests (QRF), a generalization of random forests for quantile regression. This method does not fit a parametric probability density function (PDF) like in ensemble model output statistics (EMOS) but provides an estimation of desired quantiles. This is a nonparametric approach that eliminates any assumption on the variable subject to calibration. This method can estimate quantiles using not only members of the ensemble but any predictor available including statistics on other variables. The method is applied to the Météo-France 35-member ensemble forecast (PEARP) for surface temperature and wind speed for available lead times from 3 up to 54 h and compared to EMOS. All postprocessed ensembles are much better calibrated than the PEARP raw ensemble and experiments on real data also show that QRF performs better than EMOS, and can bring a real gain for human forecasters compared to EMOS. QRF provides sharp and reliable probabilistic forecasts. At last, classical scoring rules to verify predictive forecasts are completed by the introduction of entropy as a general measure of reliability.

A global EMOS postprocessing for temperature and precipitation forecasts for any location in Switzerland

2021 ◽  
Author(s):  
Jan Rajczak ◽  
Keller Regula ◽  
Bhend Jonas ◽  
Hemri Stephan ◽  
Moret Lionel ◽  
...  
Keyword(s):  
Weather Prediction ◽  
Model Combination ◽  
Ensemble Forecasts ◽  
Mountainous Regions ◽  
Temperature And Precipitation ◽  
Hourly Temperature ◽  
Forecasting System ◽  
Ensemble Model Output Statistics ◽  
Model Output Statistics ◽  
Horizontal Grid

<p>MeteoSwiss is developing and implementing a post-processing suite of multi-model ensemble forecasts to produce seamless probabilistic calibrated forecasts at arbitrary locations in Switzerland (i.e. also for un-observed locations). With the complex topography of Switzerland, the raw output of the numerical model is subject to particular strong biases and conditional errors. Here, we present results for hourly temperature and precipitation predictions.</p><p>We apply a global ensemble model output statistics (gEMOS) framework. It extends the classical EMOS approach by incorporating static predictor variables describing relevant topographical features and it is trained for all stations together using a 4-year multi model numerical weather prediction (NWP) archive. As NWP sources, we combine data from the COSMO model suites (1.1 and 2.2 km horizontal grid-spacing) and from the ECMWF IFS medium-range forecasting system. Note that the three NWP suites have different forecast horizons.<br>We show that gEMOS is able to improve forecasts for both variables. Depending on selection of predictors, lead-time, hour-of-day and season we find improvements up to 30% in terms of CRPS for both variables with most pronounced improvements in mountainous regions. Particularly for temperature, the multi-model combination further increases the forecast skill compared to postprocessing using high-resolution simulations of COSMO only.</p><p>While locally optimized approaches show better performance in terms of skill at the observing sites, the advantage of gEMOS lies in the ability to generate calibrated predictions for arbitrary locations in a consistent way. Its computational efficiency makes it a particularly attractive method for operationalization in a realtime context.</p>

Comparison of Ensemble-MOS Methods Using GFS Reforecasts

2007 ◽  
Vol 135 (6) ◽  
pp. 2379-2390 ◽  
Author(s):  
Daniel S. Wilks ◽  
Thomas M. Hamill
Keyword(s):  
Lead Time ◽  
Ensemble Model ◽  
Ensemble Forecasts ◽  
Temperature And Precipitation ◽  
Training Samples ◽  
Probabilistic Forecasts ◽  
Medium Range ◽  
Logistic Regressions ◽  
Ensemble Model Output Statistics ◽  
Model Output Statistics

Abstract Three recently proposed and promising methods for postprocessing ensemble forecasts based on their historical error characteristics (i.e., ensemble-model output statistics methods) are compared using a multidecadal reforecast dataset. Logistic regressions and nonhomogeneous Gaussian regressions are generally preferred for daily temperature, and for medium-range (6–10 and 8–14 day) temperature and precipitation forecasts. However, the better sharpness of medium-range ensemble-dressing forecasts sometimes yields the best Brier scores even though their calibration is somewhat worse. Using the long (15 or 25 yr) training samples that are available with these reforecasts improves the accuracy and skill of these probabilistic forecasts to levels that are approximately equivalent to gains of 1 day of lead time, relative to using short (1 or 2 yr) training samples.

scholarly journals How Suitable is Quantile Mapping For Postprocessing GCM Precipitation Forecasts?

2017 ◽  
Vol 30 (9) ◽  
pp. 3185-3196 ◽  
Author(s):  
Tongtiegang Zhao ◽  
James C. Bennett ◽  
Q. J. Wang ◽  
Andrew Schepen ◽  
Andrew W. Wood ◽  
...  
Keyword(s):  
Joint Probability ◽  
Ensemble Model ◽  
Ensemble Spread ◽  
Quantile Mapping ◽  
Ensemble Forecasts ◽  
Probability Modeling ◽  
Satisfactory Method ◽  
Coherent Forecasts ◽  
Ensemble Model Output Statistics ◽  
Model Output Statistics

GCMs are used by many national weather services to produce seasonal outlooks of atmospheric and oceanic conditions and fluxes. Postprocessing is often a necessary step before GCM forecasts can be applied in practice. Quantile mapping (QM) is rapidly becoming the method of choice by operational agencies to postprocess raw GCM outputs. The authors investigate whether QM is appropriate for this task. Ensemble forecast postprocessing methods should aim to 1) correct bias, 2) ensure forecasts are reliable in ensemble spread, and 3) guarantee forecasts are at least as skillful as climatology, a property called “coherence.” This study evaluates the effectiveness of QM in achieving these aims by applying it to precipitation forecasts from the POAMA model. It is shown that while QM is highly effective in correcting bias, it cannot ensure reliability in forecast ensemble spread or guarantee coherence. This is because QM ignores the correlation between raw ensemble forecasts and observations. When raw forecasts are not significantly positively correlated with observations, QM tends to produce negatively skillful forecasts. Even when there is significant positive correlation, QM cannot ensure reliability and coherence for postprocessed forecasts. Therefore, QM is not a fully satisfactory method for postprocessing forecasts where the issues of bias, reliability, and coherence pre-exist. Alternative postprocessing methods based on ensemble model output statistics (EMOS) are available that achieve not only unbiased but also reliable and coherent forecasts. This is shown with one such alternative, the Bayesian joint probability modeling approach.

scholarly journals Forest-Based and Semiparametric Methods for the Postprocessing of Rainfall Ensemble Forecasting

2019 ◽  
Vol 34 (3) ◽  
pp. 617-634 ◽  
Author(s):  
Maxime Taillardat ◽  
Anne-Laure Fougères ◽  
Philippe Naveau ◽  
Olivier Mestre
Keyword(s):  
Heavy Rainfall ◽  
Hybrid Methods ◽  
Ensemble Prediction ◽  
Ensemble Forecasts ◽  
Ensemble Prediction System ◽  
Wide Range ◽  
Selection Step ◽  
Heavy Tailed ◽  
Ensemble Model Output Statistics ◽  
Model Output Statistics

Abstract To satisfy a wide range of end users, rainfall ensemble forecasts have to be skillful for both low precipitation and extreme events. We introduce local statistical postprocessing methods based on quantile regression forests and gradient forests with a semiparametric extension for heavy-tailed distributions. These hybrid methods make use of the forest-based outputs to fit a parametric distribution that is suitable to model jointly low, medium, and heavy rainfall intensities. Our goal is to improve ensemble quality and value for all rainfall intensities. The proposed methods are applied to daily 51-h forecasts of 6-h accumulated precipitation from 2012 to 2015 over France using the Météo-France ensemble prediction system called Prévision d’Ensemble ARPEGE (PEARP). They are verified with a cross-validation strategy and compete favorably with state-of-the-art methods like analog ensemble or ensemble model output statistics. Our methods do not assume any parametric links between the variables to calibrate and possible covariates. They do not require any variable selection step and can make use of more than 60 predictors available such as summary statistics on the raw ensemble, deterministic forecasts of other parameters of interest, or probabilities of convective rainfall. In addition to improvements in overall performance, hybrid forest-based procedures produced the largest skill improvements for forecasting heavy rainfall events.

scholarly journals The Meteorological Development Laboratory’s Aviation Weather Prediction System

2010 ◽  
Vol 25 (4) ◽  
pp. 1027-1051 ◽  
Author(s):  
Judy E. Ghirardelli ◽  
Bob Glahn
Keyword(s):  
Communication Networks ◽  
Weather Prediction ◽  
Virgin Islands ◽  
Prediction System ◽  
Probabilistic Forecasts ◽  
The Future ◽  
Ceiling Height ◽  
Gfs Model ◽  
Model Output Statistics ◽  
Aviation Weather

Abstract The Meteorological Development Laboratory (MDL) has developed and implemented an aviation weather prediction system that runs each hour and produces forecast guidance for each hour into the future out to 25 h covering the major forecast period of the National Weather Service (NWS) Terminal Aerodrome Forecast. The Localized Aviation Model Output Statistics (MOS) Program (LAMP) consists of analyses of observations, simple advective models, and a statistical component that updates the longer-range MOS forecasts from the Global Forecast System (GFS) model. LAMP, being an update to GFS MOS, is shown to be an improvement over it, as well as improving over persistence. LAMP produces probabilistic forecasts for the aviation weather elements of ceiling height, sky cover, visibility, obstruction to vision, precipitation occurrence and type, and thunderstorms. Best-category forecasts are derived from these probabilities and their associated thresholds. The LAMP guidance of sensible weather is available for 1591 stations in the contiguous United States, Alaska, Hawaii, Puerto Rico, and the Virgin Islands. Probabilistic guidance of thunderstorms is also available on a grid. The LAMP guidance is available to the entire weather enterprise via NWS communication networks and the World Wide Web. In the future, all station guidance will be gridded and be made available in a form compatible with the NWS’s National Digital Forecast Database.

Probabilistic Weather Prediction with an Analog Ensemble

2013 ◽  
Vol 141 (10) ◽  
pp. 3498-3516 ◽  
Author(s):  
Luca Delle Monache ◽  
F. Anthony Eckel ◽  
Daran L. Rife ◽  
Badrinath Nagarajan ◽  
Keith Searight
Keyword(s):  
Current Model ◽  
Weather Prediction ◽  
Computational Cost ◽  
Ensemble Prediction ◽  
Ensemble Prediction System ◽  
Forecast Lead Time ◽  
Ensemble Model Output Statistics ◽  
Model Output Statistics ◽  
Statistical Consistency ◽  
State Of The Science

Abstract This study explores an analog-based method to generate an ensemble [analog ensemble (AnEn)] in which the probability distribution of the future state of the atmosphere is estimated with a set of past observations that correspond to the best analogs of a deterministic numerical weather prediction (NWP). An analog for a given location and forecast lead time is defined as a past prediction, from the same model, that has similar values for selected features of the current model forecast. The AnEn is evaluated for 0–48-h probabilistic predictions of 10-m wind speed and 2-m temperature over the contiguous United States and against observations provided by 550 surface stations, over the 23 April–31 July 2011 period. The AnEn is generated from the Environment Canada (EC) deterministic Global Environmental Multiscale (GEM) model and a 12–15-month-long training period of forecasts and observations. The skill and value of AnEn predictions are compared with forecasts from a state-of-the-science NWP ensemble system, the 21-member Regional Ensemble Prediction System (REPS). The AnEn exhibits high statistical consistency and reliability and the ability to capture the flow-dependent behavior of errors, and it has equal or superior skill and value compared to forecasts generated via logistic regression (LR) applied to both the deterministic GEM (as in AnEn) and REPS [ensemble model output statistics (EMOS)]. The real-time computational cost of AnEn and LR is lower than EMOS.

Sign in / Sign up

Export Citation Format

Share Document