Skill Of Weather Forecasts Research Articles

Improving weather forecast skill of the Korean Integrated Model by assimilating Soil Moisture Active Passive soil moisture anomalies

AbstractThis study investigates the potential benefit of assimilating soil moisture (SM) data retrieved from Soil Moisture Active Passive (SMAP) in improving global SM estimates and enhancing the weather forecast skill of the Korean Integrated Model (KIM). The 36‐km SMAP L2 SM retrievals are assimilated into the Noah land surface model (LSM) using the ensemble Kalman filter scheme through the National Aeronautics and Space Administration Land Information System (LIS) that is weakly coupled to KIM. A suite of cycling experiments of the KIM–LIS system that includes land and atmospheric data assimilation (DA) and five‐day weather forecasts are conducted over the global domain from March to July 2022. In the SMAP SM DA, two different SM bias correction methods, namely the cumulative distribution function matching and anomaly‐based bias correction, are applied to correct systematic biases between SMAP and Noah‐LSM before assimilation. The global triple collocation analysis reveals that compared to the control case (without SM DA), significant improvements in the global SM estimates are achieved by assimilating the SMAP data, especially by employing the anomaly‐based bias correction. Notably, the improved SM initial conditions lead to an improved screen‐level specific humidity and air temperature analysis and forecasts when the results are compared against the European Centre for Medium Range Weather Forecasts‐Integrated Forecasting System analysis. The beneficial impacts of the SMAP DA on the atmospheric variables extend up to an atmospheric level of 700 hPa. Prominent improvements in the KIM forecast skill by the SMAP DA applying the anomaly‐based bias correction are observed in the northern part of Africa and West and Central Asia with stronger impacts for longer forecast lead time. This paper demonstrates the feasibility of assimilating the SMAP data within KIM–LIS to enhance the KIM weather forecasts in the lower atmosphere when a proper SM bias correction method is applied.

Generative Ensemble Deep Learning Severe Weather Prediction from a Deterministic Convection-Allowing Model

Abstract An ensemble postprocessing method is developed for the probabilistic prediction of severe weather (tornadoes, hail, and wind gusts) over the conterminous United States (CONUS). The method combines conditional generative adversarial networks (CGANs), a type of deep generative model, with a convolutional neural network (CNN) to postprocess convection-allowing model (CAM) forecasts. The CGANs are designed to create synthetic ensemble members from deterministic CAM forecasts, and their outputs are processed by the CNN to estimate the probability of severe weather. The method is tested using High-Resolution Rapid Refresh (HRRR) 1–24-h forecasts as inputs and Storm Prediction Center (SPC) severe weather reports as targets. The method produced skillful predictions with up to 20% Brier skill score (BSS) increases compared to other neural-network-based reference methods using a testing dataset of HRRR forecasts in 2021. For the evaluation of uncertainty quantification, the method is overconfident but produces meaningful ensemble spreads that can distinguish good and bad forecasts. The quality of CGAN outputs is also evaluated. Results show that the CGAN outputs behave similarly to a numerical ensemble; they preserved the intervariable correlations and the contribution of influential predictors as in the original HRRR forecasts. This work provides a novel approach to postprocess CAM output using neural networks that can be applied to severe weather prediction. Significance Statement We use a new machine learning (ML) technique to generate probabilistic forecasts of convective weather hazards, such as tornadoes and hailstorms, with the output from high-resolution numerical weather model forecasts. The new ML system generates an ensemble of synthetic forecast fields from a single forecast, which are then used to train ML models for convective hazard prediction. Using this ML-generated ensemble for training leads to improvements of 10%–20% in severe weather forecast skills compared to using other ML algorithms that use only output from the single forecast. This work is unique in that it explores the use of ML methods for producing synthetic forecasts of convective storm events and using these to train ML systems for high-impact convective weather prediction.

Old Dog, New Trick: Reservoir Computing Advances Machine Learning for Climate Modeling

AbstractPhysics‐informed machine learning (ML) applied to geophysical simulation is developing explosively. Recently, graph neural net and vision transformer architectures have shown 1–7 days global weather forecast skill superior to any conventional model with integration times over 1,000 times faster, but longer simulations rapidly degrade. ML that achieves high skill in both weather and climate applications is a tougher goal. This Commentary was inspired by Arcomano et al. (2023, https://doi.org/10.1029/2022GL102649), who show impressive progress toward that goal using hybrid ML, combining reservoir computing (RC) to a coarse‐grid climate model and coupling to a separate data‐driven RC model that interactively predicts sea‐surface temperature. This opens new horizons; where will the next ML breakthrough come from, and is conventional climate modeling about to be disrupted?

A Record‐Breaking Cyclone Over the Southern Ocean in 2022

AbstractIn October 2022, an extreme cyclone developed in the South Pacific Ocean with a sea level pressure of 900 hPa, becoming the strongest extratropical cyclone in the satellite era. Using ERA5 reanalysis data, we investigated its development mechanisms and examined long‐term changes in the occurrence of extreme cyclones over the Southern Ocean. Our findings indicate that the cyclone formed within a low‐pressure anomaly over the South Pacific. Its explosive development was initiated by upper‐level dynamic forcing and driven by low‐level latent heat release, which had been preconditioned by surface heat flux. Extreme cyclones have increased significantly in the Amundsen‐Bellingshausen Seas (ABS) and the South Indian Ocean since 1980. The large‐scale environmental variables in the ABS also showed a consistent trend toward more favorable conditions for cyclone intensification. Understanding these extreme cyclone events will help to overcome the uncertainty in projections of climate change impacts and improve weather forecast skills.

Конфигурация COSMO-Ru2By модели COSMO: успешность и методология оценки численных прогнозов - и γ-мезомасштабных атмосферных процессов

The paper gives а brief description of the COSMO-Ru2By configuration (the grid spacing is 2.2 km) of the COSMO model, which provides numerical weather forecasts for up to 48 hours for the European part of Russia and for the Republic of Belarus, as well the methodology and results of skill of these forecasts. The COSMO-Ru2By was realized in the Hydrometcentre of Russia and operates as an element of the COSMO-Ru operational limited-area numerical weather prediction system on the CRAY XС40-LC supercomputer. The features of the COSMO-Ru2By are: 1) a vast calculation domain with a grid spacing that allows an explicit description of large (over 5-6 km high) convective motions and considering in detail the features of terrain; 2) the “embedded” technology for assimilation of DMRL-C Doppler weather radar data providing more accurate forecasts of rapidly developing weather processes for the next few hours; 3) coupled visualization system providing a great number of maps for different regions with a cascade image detailing. Operational trials in 2020-2021 showed a high skill of forecasts of the basic weather parameters with the COSMO-Ru2By. For comparison with forecasts of highly variable weather parameters (wind gusts, hourly precipitation, parameters in mountain areas, etc.) based on mesoscale models with lower resolution (e.g., COSMO-Ru6ENA, the grid size is 6.6 km), it was proposed to apply expanded trial approaches and criteria: e.g., the estimation over geographically homogeneous areas, the use of special criteria for predicting rare events, the comparison with radar data. Keywords: numerical weather prediction, mesoscale atmospheric modeling, radar data assimilation, weather forecast skill estimation, non-hydrostatic atmosphere models

Systematic Assessment of the Effects of Space Averaging and Time Averaging on Weather Forecast Skill

Intuitively, one would expect a more skillful forecast if predicting weather averaged over one week instead of the weather averaged over one day, and similarly for different spatial averaging areas. However, there are few systematic studies of averaging and forecast skill with modern forecasts, and it is therefore not clear how much improvement in forecast performance is produced via averaging. Here we present a direct investigation of averaging effects, based on data from operational numerical weather forecasts. Data is analyzed for precipitation and surface temperature, for lead times of roughly 1 to 7 days, and for time- and space-averaging diameters of 1 to 7 days and 100 to 4500 km, respectively. For different geographic locations, the effects of time- or space-averaging can be different, and while no clear geographical pattern is seen for precipitation, a clear spatial pattern is seen for temperature. For temperature, in general, time averaging is most effective near coastlines, also effective over land, and least effective over oceans. Based on all locations globally, time averaging was less effective than one might expect. To help understand why time averaging may sometimes be minimally effective, a stochastic model is analyzed as a synthetic weather time series, and analytical formulas are presented for the decorrelation time. In effect, while time averaging creates a time series that is visually smoother, it does not necessarily cause a substantial increase in the predictability of the time series.

Xin’anjiang Nested Experimental Watershed (XAJ-NEW) for Understanding Multiscale Water Cycle: Scientific Objectives and Experimental Design

• Designed a five-tier nested experimental watershed for multi-scale water cycle study. • Hydrological processes exhibit a spatiotemporal scaling effect. • Canopy leaf density has an exponential effect on rainfall interception. • Surface runoff serves as a major component of total runoff in the humid zone. This paper presents the background, scientific objectives, experimental design, and preliminary achievements of the Xin’anjiang nested experimental watershed (XAJ-NEW), implemented in 2017 in eastern China, which has a subtropical humid monsoon climate and a total area of 2674 km 2 . The scientific objectives of the XAJ-NEW include building a comprehensive, multiscale, and nested hydrometeorological monitoring and experimental program, strengthening the observation of the water cycle, discovering the spatiotemporal scaling effects of hydrological processes, and revealing the mechanisms controlling runoff generation and partitioning in a typical humid, hilly area. After two years of operation, preliminary results indicated scale-dependent variability in key hydrometeorological processes and variables such as precipitation, runoff, groundwater, and soil moisture. The effects of canopy interception and runoff partitioning between the surface and subsurface were also identified. Continuous operation of this program can further reveal the mechanisms controlling runoff generation and partitioning, discover the spatiotemporal scaling effects of hydrological processes, and understand the impacts of climate change on hydrological processes. These findings provide new insights into understanding multiscale hydrological processes and their responses to meteorological forcings, improving model parameterization schemes, and enhancing weather and climate forecast skills.

Short-Term Weather Forecast Skill of Artificial Neural Networks

Abstract We evaluate the short-term weather forecast performance of three flavors of artificial neural networks (NNs): feed forward back propagation, radial basis function, and generalized regression. To prepare the application of the NNs to an operational setting, we tune NN hyperparameters using over two years of historical data. Five objective guidance products serve as predictors to the NNs: North American Mesoscale and Global Forecast System model output statistics (MOS) forecasts, the High-Resolution Rapid Refresh (HRRR) model, National Weather Service forecasts, and the National Blend of Models product. We independently test NN performance using 96 real-time forecasts of temperature, wind, and precipitation across 11 U.S. cities made during the WxChallenge, a weather forecasting competition. We demonstrate that all NNs significantly improve short-range weather forecasts relative to the traditional objective guidance aids used to train the networks. For example, 1-day maximum and minimum temperature forecast error is 20%–30% lower than MOS. However, NN improvement over multiple linear regression for short-term forecasts is not significant. We suggest this may be attributed to the small number of training samples, the operational nature of the experiment, and the short forecast lead times. Regardless, our results are consistent with previous work suggesting that applying NNs to model forecasts can have a positive impact on operational forecast skill and will become valuable tools when integrated into the forecast enterprise. Significance Statement We used approximately two years of historical weather data and objective forecasts for a number of cities to tune a series of artificial neural networks (NNs) to forecast 1-day values of maximum and minimum temperature, maximum sustained wind speed, and quantitative precipitation. We compare forecast error against common objective guidance and multiple linear regression. We found that the NNs exhibit about 25% lower error than common objective guidance for temperature forecasting and 50% lower error for wind speed. Our results suggest that NNs will be a valuable contributor to improving weather forecast skill when adopted into the existing forecast enterprise.

The Sensitivity of Downstream Ridge Building Forecasts to Upstream Warm Conveyor Belt Forecast Uncertainty Using MPAS

Abstract One potential way to improve the skill of medium-range weather forecasts is to improve the evolution of Rossby waves, which largely modulate extratropical weather. Recent research has hypothesized that the predictability of downstream Rossby waves may be limited by forecast uncertainty linked to upstream diabatic processes such as latent heat release within the warm conveyor belt (WCB) of extratropical cyclones. This hypothesis is evaluated using Model for Prediction Across Scales (MPAS) ensemble forecasts for two events characterized by highly amplified flow over the North Atlantic associated with cyclogenesis. The source of variability in ridge forecasts is diagnosed using the ensemble-sensitivity technique and a potential vorticity (PV) tendency budget, which quantifies the contribution from individual physical processes toward subsequent ridge amplification. Before the onset of ridge amplitude differences for both events, ensemble forecasts with a more amplified ridge are associated with greater negative PV advection by the irrotational wind. The importance of PV advection by the irrotational wind suggests that PV changes are modulated by diabatic heating, which is confirmed by the sensitivity of ridge amplitude to earlier diabatic heating and lower-tropospheric moisture within an upstream WCB. After the onset of ridge amplitude differences, PV advection by the nondivergent wind becomes the primary driver of downstream forecast differences. Initial condition perturbations within the sensitive areas of the WCB confirm that increasing the initial lower-tropospheric moisture results in a more amplified ridge. This suggests that more accurate initial conditions near the WCB could lead to better downstream forecasts.

Glacier parameterization in SLAV numerical weather prediction model

Abstract In the present paper, we describe a one-dimensional glacier parameterization for use in the numerical weather prediction models. The proposed scheme is implemented into the global atmospheric model SLAV. To avoid inconsistency of surface temperature and turbulent heat fluxes in the lower troposphere, glacier parameterization has been iteratively coupled with both planetary boundary layer and land surface schemes. First results from numerical experiments with the SLAV model show that the introduction of a simplified description of the glacier heat capacity can significantly improve the 2-meter temperature long-range weather forecast skill.

Benefits and challenges of dynamic sea ice for weather forecasts

Abstract. The drive to develop environmental prediction systems that are seamless across both weather and climate timescales has culminated in the development and use of Earth system models, which include a coupled representation of the atmosphere, land, ocean and sea ice, for medium-range weather forecasts. One region where such a coupled Earth system approach has the potential to significantly influence the skill of weather forecasts is in the polar and sub-polar seas, where fluxes of heat, moisture and momentum are strongly influenced by the position of the sea ice edge. In this study we demonstrate that using a dynamically coupled ocean and sea ice model in the European Centre for Medium-Range Weather Forecasts (ECMWF) Integrated Forecasting System results in improved sea ice edge position forecasts in the Northern Hemisphere in the medium range. Further, this improves forecasts of boundary layer temperature and humidity downstream of the sea ice edge in some regions during periods of rapid change in the sea ice, compared to forecasts in which the sea surface temperature anomalies and sea ice concentration do not evolve throughout the forecasts. However, challenges remain, such as large errors in the position of the ice edge in the ocean analysis used to initialise the ocean component of the coupled system, which has an error of approximately 50 % of the total forecast error at day 9, suggesting there is much skill to be gained by improving the ocean analysis at and around the sea ice edge. The importance of the choice of sea ice analysis for verification is also highlighted, with a call for more guidance on the suitability of satellite sea ice products to verify forecasts on daily to weekly timescales and on meso-scales (< 500 km).

The Impact of Data Latency on Operational Global Weather Forecasting

Abstract The impact of low data latency is assessed using observations assimilated into the NCEP Finite-Volume Cubed-Sphere Global Forecast System (FV3GFS). Operationally, a full dataset is used to generate short-term (9-h) forecasts used as the background state for the next cycle, and a limited dataset with fewer observations is used for long-term (16-day) forecasts due to time constraints that exist in an operational setting. In this study, the sensitivity of the global weather forecast skill to the use of the full and limited datasets in both the short- and long-term forecasts (out to 10 days only) is evaluated. The results show that using the full dataset for long-term forecasts yields a slight improvement in forecast skill, while using the limited dataset for short-term forecasts yields a significant degradation. This degradation is primarily attributed to a decrease of in situ observations rather than remotely sensed observations, though no individual observation type captures the amount of degradation noted when all observations are limited. Furthermore, limiting individual types of in situ observations (aircraft, marine, rawinsonde) does not result in the level of degradation noted when limiting all in situ observations, demonstrating the importance of data redundancy in an operational observational system. Significance Statement Millions of observations are used in global models every day to understand the state of the atmosphere. These observations rely on quick transmission from observation source to weather centers for inclusion in operational models. For this study, we test how different groups of observations, which arrive at the model center at different times, impact the model forecast. We find that by not using the observations that take longer to arrive at the weather centers, the forecast is much worse, showing the importance of quick transmission of observations. Direct observations (those measured within the atmosphere) have a greater impact than remote observations (those viewed from afar, such as by satellites). However, no single observation type by itself causes a poor forecast by being limited, showing the importance of using different types of observations to capture the state of the atmosphere.

Impact of Soil Moisture Data Assimilation on Analysis and Medium-Range Forecasts in an Operational Global Data Assimilation and Prediction System

Accurate initial soil moisture conditions are essential for numerical weather prediction models, because they play a major role in land–atmosphere interactions. This study constructed a soil moisture data assimilation system and evaluated its impacts on the Global Data Assimilation and Prediction System based on the Korea Integrated Model (GDAPS-KIM) to improve its weather forecast skill. Soil moisture data retrieved from the Advanced Scatterometer (ASCAT) onboard the Meteorological Operational Satellite was assimilated into GDAPS-KIM using the ensemble Kalman filter method, and its impacts were evaluated for the 2019 boreal summer period. Our results indicated that the soil moisture data assimilation improved the agreement of the observations with the initial conditions of GDAPS-KIM. This led to a statistically significant improvement in the accuracy of the initial fields. A comparison of a five-day forecast against an ERA5 reanalysis and in situ observations revealed a reduction in the dry and warm biases of GDAPS-KIM over the surface and in the lower- and mid-level atmospheres. The temperature bias correction through the initialization of the soil moisture estimates from the data assimilation system was shown in the five-day weather forecast (root mean squared errors reduction of the temperature at 850 hPa by approximately 5% in East Asia).

Studies of FY-3 Observations over the Past 10 Years: A Review

With the rapid advances and abundant observations from Chinese Fengyun-3 (FY-3) meteorological satellites, it is of great interest to summarize a decade of quality assessments of FY-3 observations. The topics covered are noise characterization, bias estimation, striping noise detection and mitigation of striping noise, radio frequency interference detection, geolocation accuracy estimation and improvement, data assimilation cloud detection and quality control for observations from the MicroWave Temperature Sounder (MWTS), the MicroWave Humidity Sounder (MWHS), the MicroWave Radiation Imager (MWRI) and the Hyperspectral Infrared Atmospheric Sounder (HIRAS) instruments on board FY-3A/B/C/D. Whether and how much FY-3 data assimilation could improve the numerical weather forecast skill strongly depends on how well the FY-3 data characteristics and errors listed above are known. This review article shall contribute to promoting internal and national usages of FY-3 observations for weather and climate studies.

Segregation of Forecast Errors in the Planetary Boundary Layer Parameterization Over the State of Odisha and Neighboring Regions in India During Summer Monsoon Season

Planetary boundary layer parametrization (PBL) is a key factor influencing weather forecast skills. This article provides a quantitative illustration of the segregation of forecast errors and their growth arising from different PBLs of the Weather Research and Forecasting (WRF) model. The systematic (root mean square) components of forecast errors arising from four different PBLs of the WRF over the state of Odisha (India) and its surrounding regions are elucidated. Error characterizations are carried out for the forecast lead time up to 96 h (day-4) for two contrasting monsoon seasons, i.e., 2013 (normal) and 2014 (deficit). A total of 1112 simulations are carried out for each initial condition, i.e., May 15 to September 29 for both monsoon seasons using four PBL schemes, i.e., Yonsei University (YSU), Mellor–Yamada–Nakanishi-Niino (MYNN), Asymmetric Convective Model version 2 (ACM2), and medium-range forecast (MRF). The overall results suggest the errors in thermodynamical variables (i.e., temperature and relative humidity) are large compared to the dynamical variable (i.e., wind). For the normal monsoon year (i.e., 2013), the MRF and MYNN exhibit the lowest root mean square error (RMSE) of temperature compared to ACM2 and YSU, whereas MRF (MYNN) shows the lowest (highest) error growth at 24–48 h (72–96 h). The deficit year (2014) has a higher temperature error compared to that of the normal monsoon year (2013), which might be due to frequent monsoon break periods. The spatial distribution of wind exhibits the lowest systematic error for MRF with lead time up to 48 h. The subsequent decrease (increase) of convergence (divergence) of error flux over northern Odisha and increase (decrease) of the same over southern Odisha suggests that the error propagation occurs from north to south. In general, both the convergence and divergence of error energy are found to be weak in MRF and MYNN, attributable to the lower error growth rate and hence the smaller systematic errors compared to ACM2 and YSU for both these monsoon seasons. It is also found that the systematic component of the linear (nonlinear) error growth rate is contributed by the physics (dynamics) components of the model. These findings will provide guidance for the model community and operational agencies to make an optimal choice of PBL parameterizations, particularly for the monsoon forecast.

Assessing the value of seasonal hydrological forecasts for improving water resource management: insights from a pilot application in the UK

Abstract. Improved skill of long-range weather forecasts has motivated an increasing effort towards developing seasonal hydrological forecasting systems across Europe. Among other purposes, such forecasting systems are expected to support better water management decisions. In this paper we evaluate the potential use of a real-time optimization system (RTOS) informed by seasonal forecasts in a water supply system in the UK. For this purpose, we simulate the performances of the RTOS fed by ECMWF seasonal forecasting systems (SEAS5) over the past 10 years, and we compare them to a benchmark operation that mimics the common practices for reservoir operation in the UK. We also attempt to link the improvement of system performances, i.e. the forecast value, to the forecast skill (measured by the mean error and the continuous ranked probability skill score) as well as to the bias correction of the meteorological forcing, the decision maker priorities, the hydrological conditions and the forecast ensemble size. We find that in particular the decision maker priorities and the hydrological conditions exert a strong influence on the forecast skill–value relationship. For the (realistic) scenario where the decision maker prioritizes the water resource availability over energy cost reductions, we identify clear operational benefits from using seasonal forecasts, provided that forecast uncertainty is explicitly considered by optimizing against an ensemble of 25 equiprobable forecasts. These operational benefits are also observed when the ensemble size is reduced up to a certain limit. However, when comparing the use of ECMWF-SEAS5 products to ensemble streamflow prediction (ESP), which is more easily derived from historical weather data, we find that ESP remains a hard-to-beat reference, not only in terms of skill but also in terms of value.

Quantifying Convective Weather Impacts to Airspace Capacity: Framework and Preliminary Results

China is facing ever-increasing constraints in airspace capacity due to a 9.7% average annual increase rate in air traffic over the past decade. This situation underscores the need for the capability to forecast regional air traffic capacity with the operationally critical lead time for the region as well as the whole airspace of China. The end goal is to improve efficiency while maintaining safety and sustaining the growth in civil aviation. Similar to the United States and Europe, weather accounted for over 50% of all the flight delays in China since 2016 and is the biggest detriment to on-time flight performance. This paper reports results aiming to develop and demonstrate a decision support tool concept that provides objective and quantitative operational thresholds for east China regional airspace capacity reductions in the presence of convective weather impacts. By estimating airspace capacities for selected route segments, waypoints, and sectors, we have demonstrated 1) the effectiveness of the flow constraint index-based approach to quantify the airspace capacity in east China, and 2) the capabilities of the forecasted capacity based on the high-resolution nowcast algorithm and numerical weather prediction model compared to the calculated capacity based on observed weather data. The preliminary results indicated that the airspace capacity estimation using the proposed approach is limited by the convective weather forecast skill, which is demonstrated in the overall accuracy of the results. The introduction of the airspace capacity estimate to operational decision making has the potential of providing useful guidance to air traffic flow management.

Observation of Jet Stream Winds during NAWDEX and Characterization of Systematic Meteorological Analysis Errors

Abstract Observations across the North Atlantic jet stream with high vertical resolution are used to explore the structure of the jet stream, including the sharpness of vertical wind shear changes across the tropopause and the wind speed. Data were obtained during the North Atlantic Waveguide and Downstream Impact Experiment (NAWDEX) by an airborne Doppler wind lidar, dropsondes, and a ground-based stratosphere–troposphere radar. During the campaign, small wind speed biases throughout the troposphere and lower stratosphere of only −0.41 and −0.15 m s−1 are found, respectively, in the ECMWF and Met Office analyses and short-term forecasts. However, this study finds large and spatially coherent wind errors up to ±10 m s−1 for individual cases, with the strongest errors occurring above the tropopause in upper-level ridges. ECMWF and Met Office analyses indicate similar spatial structures in wind errors, even though their forecast models and data assimilation schemes differ greatly. The assimilation of operational observational data brings the analyses closer to the independent verifying observations, but it cannot fully compensate for the forecast error. Models tend to underestimate the peak jet stream wind, the vertical wind shear (by a factor of 2–5), and the abruptness of the change in wind shear across the tropopause, which is a major contribution to the meridional potential vorticity gradient. The differences are large enough to influence forecasts of Rossby wave disturbances to the jet stream with an anticipated effect on weather forecast skill even on large scales.

Revised GNSS‐RO observation uncertainties in the Met Office NWP system

AbstractThe diagnostics of Desroziers et al. have been used to estimate the uncertainties of Global Navigation Satellite System Radio Occultation (GNSS‐RO) observations. Compared with the uncertainties that are currently operational, the new uncertainties are smaller and vary with height, latitude, satellite identifier, and other variables. The initial tests allowed the observation uncertainties to vary with latitude, satellite identifier, and the tangent‐point height of the observation. Using these uncertainties for a winter period demonstrated clear improvements in the weather forecast skill. Further tests showed that this improvement derived from the variation of the uncertainties with satellite identifier and from using a smooth variation with latitude—the current method gives a sharp change in the uncertainty at 30° intervals in latitude. The diagnosis method indicated a maximum in the observation uncertainty just above the tropical tropopause (around 20 km). Experiments demonstrated that the high values for the observation uncertainty in this region degrade the forecast performance. It is suggested that this is a sign that this method is identifying some model errors incorrectly as observation errors, which may be a consequence of the known assumptions of the method being violated. The uncertainties diagnosed in this study were calculated over a one‐month period in winter 2017/2018, and the initial tests were conducted using a three‐month period for the same winter. Using uncertainties that vary with latitude, height, and satellite identifier in an experiment for summer 2018 resulted in a much smaller increase in forecast skill than for the winter period. Therefore uncertainties were diagnosed that vary with quantities other than latitude. The most effective of these is to allow the observation uncertainties to vary with the average temperature between the Earth's surface and 20 km. Uncertainties that vary with this and with satellite identifier and height give good forecast skill improvements in both seasons.

Assessing the Benefits of Long‐Run Weather Forecasting for the Rural Poor: Farmer Investments and Worker Migration in a Dynamic Equilibrium Model

The livelihoods of the majority of the world's poor depend on agriculture. They face substantial risk from fluctuations in weather conditions. Better risk, credit and savings markets can improve productivity and welfare in rural areas but entail high administrative costs. We consider a classic public good with benefits that theoretically exceed those of perfect insurance contracts – improving the skill of long-run weather forecasts. We use an equilibrium model of agricultural production and labor migration, and a variety of Indian panel datasets to assess quantitatively the effects of improvements in seasonal forecasts of monsoon weather. We find that in areas where the forecast is accurate (has “skill”) that investment, migration and rural wages respond to forecasts. We calculate that if such skill were pervasive across India, the total value of an accurate forecast for farmers and wage workers is in the tens of billions of rupees.