Improved Precipitation Forecasts Research Articles

Scale‐ and Variable‐Dependent Localization for 3DEnVar Data Assimilation in the Rapid Refresh Forecast System

AbstractThis study demonstrates the advantages of scale‐ and variable‐dependent localization (SDL and VDL) on three‐dimensional ensemble variational data assimilation of the hourly‐updated high‐resolution regional forecast system, the Rapid Refresh Forecast System (RRFS). SDL and VDL apply different localization radii for each spatial scale and variable, respectively, by extended control vectors. Single‐observation assimilation tests and cycling experiments with RRFS indicated that SDL can enlarge the localization radius without increasing the sampling error caused by the small ensemble size and decreased associated imbalance of the analysis field, which was effective at decreasing the bias of temperature and humidity forecasts. Moreover, simultaneous assimilation of conventional and radar reflectivity data with VDL, where a smaller localization radius was applied only for hydrometeors and vertical wind, improved precipitation forecasts without introducing noisy analysis increments. Statistical verification showed that these impacts contributed to forecast error reduction, especially for low‐level temperature and heavy precipitation.

Statistical Postprocessing of Week-1 and Week-2 Precipitation Forecasts over Taiwan

Abstract The predictability of precipitation is hindered by finer-scale processes not captured explicitly in global numerical models, such as convective interactions, cloud microphysics, and boundary layer dynamics. However, there is growing demand across various sectors for medium- (3–10-day) and extended-range (10–30-day) quantitative precipitation forecasts (QPFs) and probabilistic QPFs (PQPFs). This study uses a novel statistical postprocessing technique, APPM, that combines analog postprocessing (AP) with probability matching (PM) to produce week-1 and week-2 accumulated precipitation forecasts over Taiwan. AP searches for historical predictions that closely resemble the current forecast and create an AP ensemble using the observed high-resolution precipitation patterns corresponding to these forecast analogs. Frequency counting and PM are then separately applied to the AP ensemble to produce calibrated and downscaled PQPFs and bias-reduced QPFs, respectively. Evaluation over a 22-yr (1999–2020) period shows that raw ensemble forecasts from the GEFS of NOAA/NWS/Environmental Modeling Center, collected for the subseasonal experiment, are underdispersive with a wet bias. In contrast, the AP ensemble spread well represents forecast uncertainty, leading to substantially more reliable and skillful probabilistic forecasts. Furthermore, the AP-based PQPF demonstrates superior discrimination ability and yields notably greater economic benefits for a wider range of users, with the maximum economic value increasing by 30%–50% for the week-2 forecast. Compared to the raw ensemble mean forecast, the calibrated QPF exhibits lower mean absolute error and explains 3–8 times more variance in observations. Overall, the APPM technique significantly improves week-1 and week-2 QPFs and PQPFs over Taiwan. Significance Statement There are two significant challenges in improving precipitation forecasts beyond a few days in Taiwan. First, large-scale numerical models often struggle with accurately predicting precipitation locations, magnitudes, and providing sufficient detail. Second, probabilistic precipitation forecasts have been unreliable, failing to convey accurate uncertainty information to users. In response to these challenges, this study has developed a relatively simple yet effective technique that corrects the spatiotemporal distribution of predicted precipitation and downscales the forecasts from a 1° to 1-km spatial resolution. Our results demonstrate that this technique significantly alleviates these two issues, resulting in more accurate precipitation forecasts and more reliable probabilistic precipitation forecasts within a 2-week timeframe.

Towards replacing precipitation ensemble predictions systems using machine learning

AbstractForecasting precipitation accurately poses significant challenges due to various factors affecting its distribution and intensity, including but not limited to subgrid variability. Although higher resolution simulations are often considered to improve precipitation forecasts, it is crucial to note that simply increasing resolution may not suffice without appropriate adjustments to parameterization schemes or tuning. Traditionally, ensembles of simulations are used to generate uncertainty predictions associated with precipitation forecasts, but this approach can be computationally intensive. As an alternative, there is a growing trend towards leveraging neural networks for precipitation prediction, which offers potential computational advantages. We propose a new approach to generating ensemble weather predictions for high‐resolution precipitation without requiring high‐resolution training data. The method uses generative adversarial networks to learn the complex patterns of precipitation and produce diverse and realistic precipitation fields, allowing to generate realistic precipitation ensemble members using only the available control forecast. We demonstrate the feasibility of generating realistic precipitation ensemble members on unseen higher resolutions. We use evaluation metrics such as RMSE, CRPS, rank histogram and ROC curves to demonstrate that our generated ensemble is almost identical to the ECMWF IFS ensemble, on which our model was trained on.

Effects of Assimilating Ground-Based Microwave Radiometer and FY-3D MWTS-2/MWHS-2 Data in Precipitation Forecasting

This study investigates the impacts of the joint assimilation of ground-based microwave radiometer (MWR) and FY-3D microwave sounder (MWTS-2/MWHS-2) observations on the analyses and forecasts for precipitation forecast. Based on the weather research and forecasting data assimilation (WRFDA) system, four experiments are conducted in this study, concerning a heavy precipitation event in Beijing on 2 July 2021, and 10-day batch experiments were also conducted. The key study findings include the following: (1) Both ground-based microwave radiometer and MWTS-2/MWHS-2 data contribute to improvements in the initial fields of the model, leading to appropriate adjustments in the thermal structure of the model. (2) The forecast fields of the experiments assimilating ground-based microwave radiometer and MWTS-2/MWHS-2 data show temperature and humidity performances closer to the true fields compared with the control experiment. (3) Separate assimilation of two types of microwave radiometer data can improve precipitation forecasts, while joint assimilation provides the most accurate forecasts among all the experiments. In the single-case, compared with the control experiment, the individual and combined assimilation of MWR and MWTS-2/MWHS-2 improves the six-hour cumulative precipitation threat score (TS) at the 25 mm level by 57.1%, 28.9%, and 38.2%, respectively. The combined assimilation also improves the scores at the 50 mm level by 54.4%, whereas individual assimilations show a decrease in performance. In the batch experiments, the MWR_FY experiment’s TS of 24 h precipitation forecast improves 28.5% at 10 mm and 330% at 25 mm based on the CTRL.

Precipitation Microphysics in Tropical Cyclones: A Global Perspective Using the NASA Global Precipitation Measurement Mission Dual‐Frequency Precipitation Radar

AbstractPrecipitation microphysics in tropical cyclones (TCs) are often poorly represented in numerical simulations, which ultimately affects TC structure, evolution, and prediction. This provides a large incentive to better observe and understand the underlying microphysical processes in TCs in order to improve precipitation forecasts and warning operations. 112 TCs from 2014 to 2020 were matched up with overpasses from the NASA Global Precipitation Measurement (GPM) mission Dual‐Frequency Precipitation Radar (DPR) on a global scale to identify cloud properties and associated precipitation processes by quantifying vertical slopes of reflectivity in the liquid and ice phase. Further, vertical profiles of reflectivity were partitioned into different 850–200 hPa shear‐relative quadrants of each storm, different annuli from the storm center, and 5 TC ocean basins globally. Preliminary results showed the highest echo top heights occurred in the Northwest Pacific and Indian Ocean basins, with all basins and quadrants exhibiting median negative slopes of KuPR in the liquid phase. Additionally, all shear‐relative quadrants revealed negative slopes of reflectivity in the ice phase implying ice hydrometeor growth. These findings can potentially be used to improve the representation of cloud properties and the accuracy of the DPR particle size distribution algorithm in TCs.

An explainable two-stage machine learning approach for precipitation forecast

A common post-processing approach to improve precipitation forecasts is to use machine learning models such as artificial neural networks (more specifically, multi-layer perceptrons) as black-box systems. These models utilize different sources of observations or predictors to generate an improved forecast in terms of desired metrics. However, most existing studies employ a single-stage regression model without considering explainability. The small number of studies with two-stage models that combine classification and regression utilize binary classification and still lack explainable artificial intelligence. Therefore, this study proposes a precipitation prediction system which (i) is composed of two stages for better predictions, (ii) compares the utility of binary and multi-class classification over the regression, and (iii) is explainable, unlike prior studies, in that individual predictions of machine learning-based forecasts are interpretable by humans. The proposed two-stage model first estimates the precipitation intensity category using binary or multi-class classification as the first stage and later utilizes precipitation intensity category information in a regression model, which is the second stage, to obtain daily precipitation magnitude. The utilized approach is made humanly interpretable (i.e., explainable) by providing insight into the model-wide importance of predictors and generation processes of the individual predictions (instance-level explanation). The proposed two-stage approach is compared against single-stage and black-box approaches in terms of prediction quality and explainability, where daily station-based observations are used as ground truth datasets. Experiments show that the proposed two-stage approach yields significant improvement (on average, RMSE reduced by 10.50%, and the correlation between numerical precipitation estimates and observed precipitation values increased by 7.5%) compared to the best-performing physical predictor (ECMWF). Analysis of explainability provides insights into the decisions of our two-stage approach, e.g., the usefulness of seasonality-related parameters, multi-class precipitation intensity classification as a first stage, and the predictors for each task (regression or classification).

Improving the CFSv2 Seasonal Precipitation Forecasts across the United States by Combining Weather Regimes and Gaussian Mixture Models

Abstract Although seasonal climate forecasts have major socioeconomic impacts, current forecast products, especially those for precipitation, are not yet reliable for forecasters and decision-makers. Here we developed a novel statistical–dynamical hybrid model for precipitation by applying weather regimes (WRs) and Gaussian mixture models (WR-GMM) to the National Oceanic and Atmospheric Administration’s Climate Forecast System, version 2 (CFSv2), precipitation forecasts across the continental United States. Instead of directly forecasting precipitation, WR-GMM uses observed precipitation from synoptic patterns similar to the future CFSv2 forecast. Traditionally K-means has been used to classify daily synoptic patterns into individual WRs, but the new GMM approach allows multiple WRs to be represented for the same day. The novel WR-GMM forecast model is trained on daily Climate Forecast System Reanalysis (CFSR) geopotential height and observed precipitation data during the 1981–2010 period and is verified for 2011–22. Overall, the WR-GMM method outperforms the CFSv2 ensemble forecast precipitation in terms of root-mean-square error and for Pearson correlation coefficient for lead months 1–4. Previous studies have used global climate models to forecast WRs in the Pacific Ocean and Mediterranean Sea regions, usually with an emphasis on winter months, but the WR-GMM model is the first of its kind that promises great untapped potential to improve precipitation forecasts produced by CFSv2 across the continental United States.

Clustering Technique Suitable for Eulerian Framework to Generate Multiple Scenarios from Ensemble Forecasts

Abstract In operational weather forecasting, it is effective to aggregate information on all members of an ensemble forecast through cluster analysis. The temporal coherence of ensemble members in each cluster is an important piece of information about the robustness of the forecast scenario given by clusters. This information is especially important for forecasts for which the target area is a city or prefecture, that is, an Eulerian framework, because the members that compose each cluster can change over time because of the small size of the target area. This study provided the temporal coherence of members in clusters by performing principal component analysis and cluster analysis on 3-hourly 500-hPa geopotential height forecasts and linking the clustering results in the time direction. The new method provided a consistently well-divided forecast scenario throughout the forecast period for Eulerian frame forecasts, as well as information on the temporal coherency of the members in the clusters, which was demonstrated to be effective through the experiment to preselect a cluster with small errors. The application of the new technique to improve precipitation forecasts was also discussed. Significance Statement Numerical weather forecasts always contain errors. Although the uncertainty of such forecasts cannot be obtained from the forecast itself, ensemble forecasts, which are aggregates of many forecasts, can be used to estimate the uncertainty of the forecast. In this study, a new method was developed to transfer the information contained in many ensemble forecasts into four forecasts by cluster analysis and to provide forecast information suitable for a small forecasting area such as a prefecture. The use of this method for improving precipitation forecasts was also examined.

Ensemble‐Based Data Assimilation of GPM DPR Reflectivity: Cloud Microphysics Parameter Estimation With the Nonhydrostatic Icosahedral Atmospheric Model (NICAM)

AbstractDirect assimilation of Dual‐frequency Precipitation Radar (DPR) data of the Global Precipitation Measurement (GPM) core satellite is challenging mainly due to its long revisiting intervals relative to the time scale of precipitation, and precipitation location errors. This study explores a method for improving precipitation forecasts using GPM DPR through model parameter estimation. We developed a 28 km mesh global atmospheric data assimilation system that integrates the Nonhydrostatic ICosahedral Atmospheric Model (NICAM) and Local Ensemble Transform Kalman Filter (LETKF) coupled with a satellite radar simulator. Using the NICAM‐LETKF and GPM DPR observations, this study estimates a model cloud physics parameter corresponding to snowfall terminal velocity. To overcome the difficulties of long revisiting intervals and precipitation location errors, we propose a parameter estimation method based on a two‐dimensional histogram known as the contoured frequency by temperature diagram (CFTD). Parameter estimation effectively mitigated the gap between simulated and observed CFTD, resulting in improved 6 hr precipitation forecasts.

Improving Precipitation Forecasts with Convolutional Neural Networks

Abstract A machine learning method based on spatial convolution to capture complex spatial precipitation patterns is proposed to identify and reduce biases affecting predictions of a dynamical model. The method is based on a combination of a classification and dual regression model approach using modified U-Net convolutional neural networks (CNN) to post-process daily accumulated precipitation over the US west coast. In this study, we leverage 34 years of high resolution deterministic Western Weather Research and Forecasting (West-WRF) precipitation reforecasts as training data for the U-Net CNN. The data is split such that the test set contains 4 water years of data that encompass characteristic west coast precipitation regimes: El Niño, La Niña, and dry and wet El Niño/Southern Oscillation (ENSO-neutral) water years. On the unseen 4-year data set, the trained CNN yields a 12.9-15.9% reduction in root mean square error (RMSE) and 2.7-3.4% improvement in Pearson correlation (PC) over West-WRF for lead times of 1-4 days. Compared to an adapted model output statistics correction, the CNN reduces RMSE by 7.4-8.9% and improves PC by 3.3-4.2% across all events. Effectively, the CNN adds more than a day of predictive skill when compared to West-WRF. The CNN outperforms the other methods also for the prediction of extreme events, which we define as the top 10% of events with the greatest average daily accumulated precipitation. The improvement over West-WRF’s RMSE/PC for these events is 19.8-21.0%/4.9-5.5% and MOS’s RMSE/PC is 8.8-9.7%/4.2-4.7%. Hence, the proposed U-Net CNN shows significantly improved forecast skill over existing methods, highlighting a promising path forward for improving precipitation forecasts.

Integrating Regression Models and Climatological Data for Improved Precipitation Forecast in Southern India

Integrating Regression Models and Climatological Data for Improved Precipitation Forecast in Southern India

Subseasonal forecasts of precipitation over maritime continent in boreal summer and the sources of predictability

In this study, subseasonal precipitation forecast skills over Maritime Continent in boreal summer are investigated for the ECMWF and CMA models involved in the S2S Project. Results indicate that the ECMWF model shows generally superior forecast performances than CMA, which is characterized by lower errors and higher correlations compared with the observations. Meanwhile, ECMWF tends to produce wet biases with increasing lead times, while the mean errors of CMA are revealed to be approximately constant throughout lead times of 2–4 weeks over most areas. Besides, the temporal correlations between model outputs and observations obviously decrease with growing lead times, with a high-low distribution presented from north to south. In addition, the roles of large-scale drivers like ENSO and BSISO in modulating subseasonal precipitation forecast skills are also assessed in the models. Both ECMWF and CMA can reasonably capture the ENSO related precipitation anomalies for all lead times, while their capabilities of capturing BSISO related precipitation anomalies decrease with growing lead times, which is more obvious in CMA. The enhanced subseasonal precipitation forecast skills mainly respond to the BSISO associated precipitation variability. For most MC areas such as southern Indochina, western Indonesia, Philippines and the eastern ocean, the forecast skills of both ECMWF and CMA can be improved to a great extent by enhancing the capture of BSISO related precipitation anomalies, with the temporal correlations for both ECMWF and CMA increased by about 0.15 for lead times of 3–4 weeks. It provides an opportunity window for the models to improve precipitation forecasts on the subseasonal timescale.

Influence of Assimilating Wind Profiling Radar Observations in Distinct Dynamic Instability Regions on the Analysis and Forecast of an Extreme Rainstorm Event in Southern China

This study quantitatively examines the contribution of assimilating observations in the regions with different dynamic instabilities to the analysis and prediction of an extreme rainstorm event in Fujian Province of China. The wind profiling radar (WPR) observations are classified into two groups, i.e., strong and weak instability areas (SIA and WIA), according to their local dynamic instability identified by the ensemble spread. Their performance of assimilation and prediction in terms of the wind and precipitation are evaluated and compared in detail. The results show that the wind analysis error by assimilating all of the WPR observations can be reduced by about 30%. In particular, the wind analysis errors by only assimilating the observations in the SIA are about 12% lower than those in the WIA. They are related to the existence of the low-level horizontal wind shear with strong instability in the SIA. The case study shows that the assimilation of observations in the SIA can effectively correct the wind fields on the two sides of the wind shear line, producing an improved precipitation forecast compared to observation assimilation in the WIA.

Comparing 3DVAR and hybrid radar data assimilation methods for heavy rain forecast

Radar data are assimilated into numerical prediction models to improve precipitation forecasts using three-dimensional (3D) variational (VAR) and hybrid data assimilation (DA) methods. The effect of radar DA (RDA) and prediction accuracy according to DA methods are evaluated by conducting experiments of three heavy rainfall events and analyzing the spatial pattern of accumulated precipitation, profiles of hydrometeors, and cloud microphysical processes. Increments of the initial analysis revealed that the hybrid method simulated the convective band, wind convergence in front of the cold front and a higher amount of water vapor mixing ratio compared to 3DVAR. Based on numerical experiments, the root mean square error of the total cumulative precipitation in 3DVAR was 9.68 mm lower than that of the no-RDA experiment (CTRL), whereas that of the Hybrid was 11.12 mm lower than that of CTRL. RDA improved precipitation forecasts, and Hybrid showed better precipitation forecast accuracy than 3DVAR. Although RDA indirectly changed the water vapor, rain, snow, and graupel mixing ratios, the change in the water vapor amount was the major factor affecting microphysical processes. The increase (decrease) in the water vapor mixing ratio had the greatest impact on precipitation formation (dissipation), whereas the effect of snow, rain, and graupel mixing ratios was relatively small. The results indicate that it is critical to create an environment in which water vapor can transform into precipitating hydrometeors and the hydrometeors can grow through cloud microphysical processes. The accuracy of heavy rain forecast can be improved by hybrid DA method, which can consider the model error in real time, making it more adaptable to the day-to-day changing weather conditions.

Вплив категорій землекористування на прогноз WRF та його точність

In this research analysis of possibilities and methods of correction or reclassification of land use and land cover data, obtained modeling effects of using more accurate land use and land cover data and estimations of improvements of surface meteorological parameters simulations by mesoscale numerical atmospheric prediction (NWP) model. For the Kyiv region area, it is shown that the default WRF land use and land cover datasets are not ideal and contain various types of inaccuracies. Thus, compared to USGS, MODIS IGBP is characterized by a better spatial resolution, more correctly describing the urban built-up territories, forest areas and the Dnipro riverbed, but MODIS IGBP does not display small settlements. Analysis of other researches had shown that most of the reviewed studies are based on the case study approach, namely on an analysis of a small number of specific simulations. The disadvantage of this approach is that the sample size from which conclusions can be drawn is insignificant, so it is useful to compare similar studies for the consistency of the obtained results. The most consistent conclusions regarding the simulation improvement by the land use and land cover modification in the WRF model are related to the parameters of air temperature and humidity. Thus, the usage of more accurate and up-to-date land use and land cover databases in WRF simulations made it possible to reduce bias of forecasts and improved the accuracy of minimum and maximum daily temperature forecasts. The most significant improvement in the performed forecasts concerns the night temperature. The second parameter that were most often able to predict better is relative humidity. Improvement of simulated wind parameters using another land use and land cover dataset in the WRF model may be possible in urban areas and near the coast, especially as improving description of breeze circulation by the model. Assessments regarding the possibility of improving WRF precipitation simulation by land use and land cover dataset modification are not definite, although some works have noted the influence of the heat island of urbanized areas on redistribution of precipitation. On the one hand, in most of the analyzed scientific works, land use and land cover changes did not allow to significantly improve precipitation forecasts using the WRF model. On the other hand, it can be assumed that the physical and geographical conditions of a simulation domain as well as available moisture in the atmosphere is playing a more significant role in precipitation modeling than land use and land cover.

The economic effects of improved precipitation forecasts in the United States due to better commuting decisions

AbstractEach year the U.S. government makes significant investments in improving weather forecasts from numerical weather prediction models that are run operationally within the National Weather Service. Although the primary purpose is saving lives and property, more accurate forecasts can create substantial efficiency gains when they elicit improved behavioral responses. One example involves commuters, who make daily decisions about when to leave for work based, in part, on expected road conditions. When workers account for potential weather delays, economic losses due to missed work time are reduced. Economists have often looked at such questions with cost‐loss models, a partial equilibrium approach investigating gains to individual agents. We extend this idea to examine economy‐wide effects, embedding a behavioral model of time allocation with improved information into eight regional computable general equilibrium models of the US economy. Our primary mechanism is that reductions in lost work time lead to gains in firm‐level output. This study evaluates the economic impact of improvements to the High‐Resolution Rapid Refresh weather forecast models over three different versions on commuting. Aggregating results from comparisons between old and new versions of the HRRR for 206 metropolitan statistical areas, we find that forecasting improvements lead to smaller losses in work time, creating notable gains to the U.S. economy.

Wintertime precipitation over the Australian Snowy Mountains: Observations from an Intensive Field Campaign 2018

AbstractUnderstanding the key dynamical and microphysical mechanisms driving precipitation in the Snowy Mountains region of southeast Australia, including the role of orography, can help improve precipitation forecasts, which is of great value for efficient water management. An intensive observation campaign was carried out during the 2018 austral winter, providing a comprehensive range of ground-based observations across the Snowy Mountains. We used data from three vertically pointing rain radars, cloud radar, a PARSIVEL disdrometer, and a network of 76 pluviometers. The observations reveal that all of the precipitation events were associated with cold front passages. About half accumulated during the frontal passage associated with deep, fully glaciated cloud tops; while the rest occurred in the post-frontal environment and was associated with clouds with supercooled liquid water (SLW) tops. About three quarters of the accumulated precipitation were observed under blocked conditions, likely associated with blocked stratiform orographic enhancement. Specifically, more than a third of the precipitation resulted from moist cloudless air being lifted over stagnant air, upwind from the barrier, creating SLW-top clouds. These SLW-clouds then produced stratiform precipitation mostly over the upwind slopes and mountain tops, with hydrometeors reaching the mountain tops mostly as rimed snow. Two precipitation events were studied in detail, which showed that during unblocked conditions, orographic convection invigoration and unblocked stratiform enhancement were the two main mechanisms driving the precipitation; with the latter being more prevalent after the frontal passage. During these events, ice particle growth was likely dominated by vapor deposition and aggregation during the frontal periods, while riming dominated during the post-frontal periods.

Effect of the Assimilation Frequency of Radar Reflectivity on Rain Storm Prediction by Using WRF-3DVAR

An attempt was made to evaluate the impact of assimilating Doppler Weather Radar (DWR) reflectivity together with Global Telecommunication System (GTS) data in the three-dimensional variational data assimilation (3DVAR) system of the Weather Research Forecast (WRF) model on rain storm prediction in Daqinghe basin of northern China. The aim of this study was to explore the potential effects of data assimilation frequency and to evaluate the outputs from different domain resolutions in improving the meso-scale NWP rainfall products. In this study, four numerical experiments (no assimilation, 1 and 6 h assimilation time interval with DWR and GTS at 1 km horizontal resolution, 6 h assimilation time interval with radar reflectivity, and GTS data at 3 km horizontal resolution) are carried out to evaluate the impact of data assimilation on prediction of convective rain storms. The results show that the assimilation of radar reflectivity and GTS data collectively enhanced the performance of the WRF-3DVAR system over the Beijing-Tianjin-Hebei region of northern China. It is indicated by the experimental results that the rapid update assimilation has a positive impact on the prediction of the location, tendency, and development of rain storms associated with the study area. In order to explore the influence of data assimilation in the outer domain on the output of the inner domain, the rainfall outputs of 3 and 1 km resolution are compared. The results show that the data assimilation in the outer domain has a positive effect on the output of the inner domain. Since the 3DVAR system is able to analyze certain small-scale and convective-scale features through the incorporation of radar observations, hourly assimilation time interval does not always significantly improve precipitation forecasts because of the inaccurate radar reflectivity observations. Therefore, before data assimilation, the validity of assimilation data should be judged as far as possible in advance, which can not only improve the prediction accuracy, but also improve the assimilation efficiency.

Precipitation Biases in the ECMWF Integrated Forecasting System

AbstractPrecipitation is a key component of the global water cycle and plays a crucial role in flooding, droughts, and water supply. One way to manage its socioeconomic effects is based on precipitation forecasts from numerical weather prediction (NWP) models, and an important step to improve precipitation forecasts is by diagnosing NWP biases. In this study, we investigate the biases in precipitation forecasts from the European Centre for Medium-Range Weather Forecasts Integrated Forecasting System (IFS). Using the IFS control forecast from 12 June 2019 to 11 June 2020 at 5219 stations globally, we show that in each of the boreal winter and summer half years, the IFS 1) has an average global wet bias and 2) displays similar bias patterns for forecasts starting at 0000 and 1200 UTC and across forecast days 1–5. These biases are dependent on observed (climatological) precipitation; stations with low observed precipitation have an IFS wet bias, while stations with high observed precipitation have an IFS dry bias. Southeast Asia has a wet bias of 1.61 mm day−1 (in boreal summer) and over the study period the precipitation is overestimated by 31.0% on forecast day 3. This is the hydrological signature of several hypothesized processes including issues specifying the IFS snowpack over the Tibetan Plateau, which may affect the mei-yu front. These biases have implications for IFS land–atmosphere feedbacks, river discharge, and for ocean circulation in the Southeast Asia region. Reducing these biases could lead to more accurate forecasts of the global water cycle.

Remote sensing data assimilation

ABSTRACT Data assimilation (DA) offers immense potential for uncertainty identification, improving the initial estimates for hydrological and atmospheric modelling. This paper reviews the studies in hydrological DA using Kalman filters. Recent applications of Kalman filters in hydrological and atmospheric DA are summarized. Existing challenges for DA studies are briefly described. In addition, three case study examples are presented highlighting the effects of: (a) soil moisture DA in the Noah land surface model; (b) variational assimilation for improving precipitation forecasts in the WRF (Weather Research Forecast) model; and (c) simulating AMSR-2 (Advanced Microwave Scanning Radiometer-2) radiances towards DA. Although there are many unresolved issues in DA that warrant further research, it has immense potential for predicting variables at a better lead time for hydrometeorology.