Promising Results Predict Role for Artificial Intelligence in Weather Forecasting

Artificial intelligence(AI)has already demonstrated its proficiency at difficult scientific tasks like predicting how proteins will fold and identifying new astronomical objects in masses of observational data[1].Now,recent results suggest that AI also excels at weather forecasting.For global predictions,GraphCast,an AI system developed by Google subsidiary DeepMind(London,UK),outperforms the state-of-the-art model from the European Centre for Medium-Range Weather Forecasts(ECMWF),providing more accurate projections of variables such as temperature and humidity 90%of the time[2,3].Other AI systems,including Pangu-Weather from the Chinese tech company Huawei(Shenzhen,China)[4],can also match or beat traditional global forecasting models.

Non-crossing Quantile Regression Neural Network as a Calibration Tool for Ensemble Weather Forecasts

Despite the maturity of ensemble numerical weather prediction(NWP),the resulting forecasts are still,more often than not,under-dispersed.As such,forecast calibration tools have become popular.Among those tools,quantile regression(QR)is highly competitive in terms of both flexibility and predictive performance.Nevertheless,a long-standing problem of QR is quantile crossing,which greatly limits the interpretability of QR-calibrated forecasts.On this point,this study proposes a non-crossing quantile regression neural network(NCQRNN),for calibrating ensemble NWP forecasts into a set of reliable quantile forecasts without crossing.The overarching design principle of NCQRNN is to add on top of the conventional QRNN structure another hidden layer,which imposes a non-decreasing mapping between the combined output from nodes of the last hidden layer to the nodes of the output layer,through a triangular weight matrix with positive entries.The empirical part of the work considers a solar irradiance case study,in which four years of ensemble irradiance forecasts at seven locations,issued by the European Centre for Medium-Range Weather Forecasts,are calibrated via NCQRNN,as well as via an eclectic mix of benchmarking models,ranging from the naïve climatology to the state-of-the-art deep-learning and other non-crossing models.Formal and stringent forecast verification suggests that the forecasts post-processed via NCQRNN attain the maximum sharpness subject to calibration,amongst all competitors.Furthermore,the proposed conception to resolve quantile crossing is remarkably simple yet general,and thus has broad applicability as it can be integrated with many shallow-and deep-learning-based neural networks.

A Deep Learning Approach for Forecasting Thunderstorm Gusts in the Beijing–Tianjin–Hebei Region

Thunderstorm gusts are a common form of severe convective weather in the warm season in North China,and it is of great importance to correctly forecast them.At present,the forecasting of thunderstorm gusts is mainly based on traditional subjective methods,which fails to achieve high-resolution and high-frequency gridded forecasts based on multiple observation sources.In this paper,we propose a deep learning method called Thunderstorm Gusts TransU-net(TGTransUnet)to forecast thunderstorm gusts in North China based on multi-source gridded product data from the Institute of Urban Meteorology(IUM)with a lead time of 1 to 6 h.To determine the specific range of thunderstorm gusts,we combine three meteorological variables:radar reflectivity factor,lightning location,and 1-h maximum instantaneous wind speed from automatic weather stations(AWSs),and obtain a reasonable ground truth of thunderstorm gusts.Then,we transform the forecasting problem into an image-to-image problem in deep learning under the TG-TransUnet architecture,which is based on convolutional neural networks and a transformer.The analysis and forecast data of the enriched multi-source gridded comprehensive forecasting system for the period 2021–23 are then used as training,validation,and testing datasets.Finally,the performance of TG-TransUnet is compared with other methods.The results show that TG-TransUnet has the best prediction results at 1–6 h.The IUM is currently using this model to support the forecasting of thunderstorm gusts in North China.

Improving the Short-Range Precipitation Forecast of Numerical Weather Prediction through a Deep Learning-Based Mask Approach

Due to various technical issues,existing numerical weather prediction(NWP)models often perform poorly at forecasting rainfall in the first several hours.To correct the bias of an NWP model and improve the accuracy of short-range precipitation forecasting,we propose a deep learning-based approach called UNet Mask,which combines NWP forecasts with the output of a convolutional neural network called UNet.The UNet Mask involves training the UNet on historical data from the NWP model and gridded rainfall observations for 6-hour precipitation forecasting.The overlap of the UNet output and the NWP forecasts at the same rainfall threshold yields a mask.The UNet Mask blends the UNet output and the NWP forecasts by taking the maximum between them and passing through the mask,which provides the corrected 6-hour rainfall forecasts.We evaluated UNet Mask on a test set and in real-time verification.The results showed that UNet Mask outperforms the NWP model in 6-hour precipitation prediction by reducing the FAR and improving CSI scores.Sensitivity tests also showed that different small rainfall thresholds applied to the UNet and the NWP model have different effects on UNet Mask's forecast performance.This study shows that UNet Mask is a promising approach for improving rainfall forecasting of NWP models.

Artificial Intelligence Based Meteorological Parameter Forecasting for Optimizing Response of Nuclear Emergency Decision Support System

This paper presents a novel artificial intelligence (AI) based approach to predict crucial meteorological parameters such as temperature,pressure,and wind speed,typically calculated from computationally intensive weather research and forecasting (WRF) model.Accurate meteorological data is indispensable for simulating the release of radioactive effluents,especially in dispersion modeling for nuclear emergency decision support systems.Simulation of meteorological conditions during nuclear emergencies using the conventional WRF model is very complex and time-consuming.Therefore,a new artificial neural network (ANN) based technique was proposed as a viable alternative for meteorological prediction.A multi-input multi-output neural network was trained using historical site-specific meteorological data to forecast the meteorological parameters.Comprehensive evaluation of this technique was conducted to test its performance in forecasting various parameters including atmospheric pressure,temperature,and wind speed components in both East-West and North-South directions.The performance of developed network was evaluated on an unknown dataset,and acquired results are within the acceptable range for all meteorological parameters.Results show that ANNs possess the capability to forecast meteorological parameters,such as temperature and pressure,at multiple spatial locations within a grid with high accuracy,utilizing input data from a single station.However,accuracy is slightly compromised when predicting wind speed components.Root mean square error (RMSE) was utilized to report the accuracy of predicted results,with values of 1.453℃for temperature,77 Pa for predicted pressure,1.058 m/s for the wind speed of U-component and 0.959 m/s for the wind speed of V-component.In conclusion,this approach offers a precise,efficient,and wellinformed method for administrative decision-making during nuclear emergencies.

Forecasting solar still performance from conventional weather data variation by machine learning method

Solar stills are considered an effective method to solve the scarcity of drinkable water.However,it is still missing a way to forecast its production.Herein,it is proposed that a convenient forecasting model which just needs to input the conventional weather forecasting data.The model is established by using machine learning methods of random forest and optimized by Bayesian algorithm.The required data to train the model are obtained from daily measurements lasting9 months.To validate the accuracy model,the determination coefficients of two types of solar stills are calculated as 0.935and 0.929,respectively,which are much higher than the value of both multiple linear regression(0.767)and the traditional models(0.829 and 0.847).Moreover,by applying the model,we predicted the freshwater production of four cities in China.The predicted production is approved to be reliable by a high value of correlation(0.868)between the predicted production and the solar insolation.With the help of the forecasting model,it would greatly promote the global application of solar stills.

Improvements in Weather Forecasting Technique Using Cognitive Internet of Things

Forecasting the weather is a challenging task for human beings because of the unpredictable nature of the climate.However,effective forecasting is vital for the general growth of a country due to the significance of weather forecasting in science and technology.The primary motivation behind this work is to achieve a higher level of forecasting accuracy to avoid any damage.Currently,most weather forecasting work is based on initially observed numerical weather data that cannot fully cover the changing essence of the atmosphere.In this work,sensors are used to collect real-time data for a particular location to capture the varying nature of the atmosphere.Our solution can give the anticipated results with the least amount of human engagement by combining human intelligence and machine learning with the help of the cognitive Internet of Things.The Authors identified weatherrelated parameters such as temperature,humidity,wind speed,and rainfall and then applied cognitive data collection methods to train and validate their findings.In addition,the Authors have examined the efficacy of various machine learning algorithms by using them on both data sets i.e.,pre-recorded metrological data sets and live sensor data sets collected from multiple locations.The Authors noticed that the results were superior on the sensor data.The Authors developed ensemble learning model using stacked method that achieved 99.25%accuracy,99%recall,99%precision,and 99%F1-score for Sensor data.It also achieved 85%accuracy,86%recall,85%precision,and 86%F1 score for Australian rainfall data.

A STUDY OF THE INFLUENCE OF MICROPHYSICAL PROCESSES ON TYPHOON NIDA(2016) USING A NEW DOUBLE-MOMENT MICROPHYSICS SCHEME IN THE WEATHER RESEARCH AND FORECASTING MODEL

The basic structure and cloud features of Typhoon Nida(2016) are simulated using a new microphysics scheme(Liuma) within the Weather Research and Forecasting(WRF) model. Typhoon characteristics simulated with the Liuma microphysics scheme are compared with observations and those simulated with a commonly-used microphysics scheme(WSM6). Results show that using different microphysics schemes does not significantly alter the track of the typhoon but does significantly affect the intensity and the cloud structure of the typhoon. Results also show that the vertical distribution of cloud hydrometeors and the horizontal distribution of peripheral rainband are affected by the microphysics scheme. The mixing ratios of rain water and graupel correlate highly with the vertical velocity component and equivalent potential temperature at the typhoon eye-wall region. According to the simulation with WSM 6 scheme,it is likely that the very low typhoon central pressure results from the positive feedback between hydrometeors and typhoon intensity. As the ice-phase hydrometeors are mostly graupel in the Liuma microphysics scheme, further improvement in this aspect is required.

MULTIMODEL CONSENSUS FORECASTING OF LOW TEMPERATURE AND ICY WEATHER OVER CENTRAL AND SOUTHERN CHINA IN EARLY 2008

Based on the daily mean temperature and 24-h accumulated total precipitation over central and southern China, the features and the possible causes of the extreme weather events with low temperature and icing conditions,which occurred in the southern part of China during early 2008, are investigated in this study. In addition, multimodel consensus forecasting experiments are conducted by using the ensemble forecasts of ECMWF, JMA, NCEP and CMA taken from the TIGGE archives. Results show that more than a third of the stations in the southern part of China were covered by the extremely abundant precipitation with a 50-a return period, and extremely low temperature with a 50-a return period occurred in the Guizhou and western Hunan province as well. For the 24- to 216-h surface temperature forecasts, the bias-removed multimodel ensemble mean with running training period(R-BREM) has the highest forecast skill of all individual models and multimodel consensus techniques. Taking the RMSEs of the ECMWF 96-h forecasts as the criterion, the forecast time of the surface temperature may be prolonged to 192 h over the southeastern coast of China by using the R-BREM technique. For the sprinkle forecasts over central and southern China, the R-BREM technique has the best performance in terms of threat scores(TS) for the 24- to 192-h forecasts except for the 72-h forecasts among all individual models and multimodel consensus techniques. For the moderate rain, the forecast skill of the R-BREM technique is superior to those of individual models and multimodel ensemble mean.

An Implementation of Full Cycle Strategy Using Dynamic Blending for Rapid Refresh Short-range Weather Forecasting in China

The partial cycle(PC)strategy has been used in many rapid refresh cycle systems(RRC)for regional short-range weather forecasting.Since the strategy periodically reinitializes the regional model(RM)from the global model(GM)forecasts to correct the large-scale drift,it has replaced the traditional full cycle(FC)strategy in many RRC systems.However,the extra spin-up in the PC strategy increases the computer burden on RRC and generates discontinuous smallscale systems among cycles.This study returns to the FC strategy but with initial fields generated by dynamic blending(DB)and data assimilation(DA).The DB ingests the time-varied large-scale information from the GM to the RM to generate less-biased background fields.Then the DA is performed.We applied the new FC strategy in a series of 7-day batch forecasts with the 3-hour cycle in July 2018,and February,April,and October 2019 over China using a Weather Research and Forecast(WRF)model-based RRC.A comparison shows that the new FC strategy results in less model bias than the PC strategy in most state variables and improves the forecast skills for moderate and light precipitation.The new FC strategy also allows the model to reach a balanced state earlier and gives favorable forecast continuity between adjacent cycles.Hence,this new FC strategy has potential to be applied in RRC forecast systems to replace the currently used PC strategy.

Real-Time Crop Prediction Based on Soil Fertility and Weather Forecast Using IoT and a Machine Learning Algorithm

The aim of this article is to assist farmers in making better crop selection decisions based on soil fertility and weather forecast through the use of IoT and AI (smart farming). To accomplish this, a prototype was developed capable of predicting the best suitable crop for a specific plot of land based on soil fertility and making recommendations based on weather forecast. Random Forest machine learning algorithm was used and trained with Jupyter in the Anaconda framework to achieve an accuracy of about 99%. Based on this process, IoT with the Message Queuing Telemetry Transport (MQTT) protocol, a machine learning algorithm, based on Random Forest, and weather forecast API for crop prediction and recommendations were used. The prototype accepts nitrogen, phosphorus, potassium, humidity, temperature and pH as input parameters from the IoT sensors, as well as the weather API for data forecasting. The approach was tested in a suburban area of Yaounde (Cameroon). Taking into account future meteorological parameters (rainfall, wind and temperature) in this project produced better recommendations and therefore better crop selection. All necessary results can be accessed from anywhere and at any time using the IoT system via a web browser.

Machine Learning of Weather Forecasting Rules from Large Meteorological Data Bases

Discovery of useful forecasting rules from observational weather data is an outstanding interesting topic.The traditional methods of acquiring forecasting knowledge are manual analysis and investigation performed by human scientists.This paper presents the experimental results of an automatic machine learning system which derives forecasting rules from real observational data.We tested the system on the two large real data sets from the areas of centra! China and Victoria of Australia.The experimental results show that the forecasting rules discovered by the system are very competitive to human experts.The forecasting accuracy rates are 86.4% and 78% of the two data sets respectively

Meteorological observations and weather forecasting services of the CHINARE

By 2018, China had conducted 34 scientific explorations in Antarctica spearheaded by the Chinese National Antarctic Research Expedition(CHINARE). Since the first CHINARE over 30 years ago, considerable work has been undertaken to promote the development of techniques for the observation of surface and upper-air meteorological elements, and satellite image and data reception systems at Chinese Antarctic stations and onboard Chinese icebreakers have played critical roles in this endeavor. The upgrade of in situ and remote sensing measurement methods and the improvement of weather forecasting skill have enabled forecasters to achieve reliable on-site weather forecasting for the CHINARE. Nowadays, the routing of icebreakers, navigation of aircraft, and activities at Chinese Antarctic stations all benefit from the accurate weather forecasting service. In this paper, a review of the conventional meteorological measurement and operational weather forecasting services of the CHINARE is presented.

A Methodological Study on Using Weather Research and Forecasting(WRF) Model Outputs to Drive a One-Dimensional Cloud Model

A new method for driving a One-Dimensional Stratiform Cold （1DSC） cloud model with Weather Research and Fore casting （WRF） model outputs was developed by conducting numerical experiments for a typical large-scale stratiform rainfall event that took place on 4-5 July 2004 in Changchun, China. Sensitivity test results suggested that, with hydrometeor pro files extracted from the WRF outputs as the initial input, and with continuous updating of soundings and vertical velocities （including downdraft） derived from the WRF model, the new WRF-driven 1DSC modeling system （WRF-1DSC） was able to successfully reproduce both the generation and dissipation processes of the precipitation event. The simulated rainfall intensity showed a time-lag behind that observed, which could have been caused by simulation errors of soundings, vertical velocities and hydrometeor profiles in the WRF output. Taking into consideration the simulated and observed movement path of the precipitation system, a nearby grid point was found to possess more accurate environmental fields in terms of their similarity to those observed in Changchun Station. Using profiles from this nearby grid point, WRF-1DSC was able to repro duce a realistic precipitation pattern. This study demonstrates that 1D cloud-seeding models do indeed have the potential to predict realistic precipitation patterns when properly driven by accurate atmospheric profiles derived from a regional short range forecasting system, This opens a novel and important approach to developing an ensemble-based rain enhancement prediction and operation system under a probabilistic framework concept.

Clothing Sales Forecast Considering Weather Information: An Empirical Study in Brick-and-Mortar Stores by Machine-Learning

Reliable sales forecasts are important to the garment industry. In recent years, the global climate is warming, the weather changes frequently, and clothing sales are affected by weather fluctuations. The purpose of this study is to investigate whether weather data can improve the accuracy of product sales and to establish a corresponding clothing sales forecasting model. This model uses the basic attributes of clothing product data, historical sales data, and weather data. It is based on a random forest, XGB, and GBDT adopting a stacking strategy. We found that weather information is not useful for basic clothing sales forecasts, but it did improve the accuracy of seasonal clothing sales forecasts. The MSE of the dresses, down jackets, and shirts are reduced by 86.03%, 80.14%, and 41.49% on average. In addition, we found that the stacking strategy model outperformed the voting strategy model, with an average MSE reduction of 49.28%. Clothing managers can use this model to forecast their sales when they make sales plans based on weather information.

Experiments in Forecasting Mesoscale Convective Weather over Changjiang Delta

The real time operational severe convective weather forecast experiment carried out during May to July in 1990 over the Changjiang Delta is briefly described. The heavy rainfall and severe conveetive weather forecast worksheets for the Changjiang Delta have been proposed and used in the daily forecasting. Results show that the ability of 0-12h convective weather prediction has been improved significantly after the development of the forecast methods and the establishment of a mesoscale forecast base at Shanghai Meteorological Center during 1986 to 1990.Three cases of convective weather systems (meso-alpha, meso-beta, meso-gamma) during the experiment period are described and discussed.

The Need of Incorporating Indigenous Knowledge Systems into Modern Weather Forecasting Methods

The study was aimed to examine the need of incorporating traditional weather forecasting renowned indigenous knowledge system (IKS) into modern weather forecasting methods to be used for planning farming activities. In addition, not only gap that is not infused by current weather forecasting system with their advanced studies to understand why it is incorporated into existing technical frameworks was regarded, but also the limitation of advanced weather forecasting approach and strength to be elicited by indigenous knowledge system are crucial. Perspicuously, forms and onsite interrogates have been conducted to assess people’s beliefs, understanding, and attitudes on the indigenous knowledge system significance on weather forecasting. Therefore, atmospheric and biological conditions, astronomic, as well as relief characteristics were used to predict the weather over short and long periods. Usually, in assessing weather conditions, the conduct of animals and insects were listed as essential. Obviously, in order to predict weather particularly from rain within about short period of time, astronomical characteristics were used. Commonly, there are few peers who know conventional weather prediction approaches. This lowers the reliability of conventional weather prediction. The findings revealed some variables that impact meteorological inaccuracy by scientific methods and help to recognize and evaluate the gap that current meteorological technologies do not achieve and new particulars anticipated to be filled with conventional methods to attain accurate weather prediction. Additionally, the study indicated that both modern and conventional processes have certain positive and limitations, which means that they can be coupled to generate more accurate weather prediction reports for end users.

Quantum Atmospheric Biophysics: A Comparison of Four Weather Stations in India on Average Monthly Temperatures Since 1892 and Forecasts to 2150

The identification method revealed asymmetric wavelets of dynamics, as fractal quanta of the behavior of the surface air layer at a height of 2 m, according to the average monthly temperature at four weather stations in India (Srinagar, Jolhpur, New Delhi and Guvahati). For Srinagar station, the maximum for all years is observed in July, for Jolhpur and New Delhi stations it shifts to June, and for Guvahati it shifts to August. With a high correlation coefficient of 0.9659, 0.8640 and 0.8687, a three-factor model of the form was obtained. The altitude, longitude and latitude of the station are given sequentially. The hottest month for Srinagar over a period of 130 years is in July. At the same time, the temperature increased from 23.4 °C to 24.2 °C (by 3.31%). A noticeable decrease in the intensity of heat flows in June occurred at Jolhpur (over 125 years, a decrease from 36.2 °C to 33.3 °C, or by 8.71%) and New Delhi (over 90 years, a decrease from 35.1 °C to 32.4 °C, or by 7.69%). For almost 120 years, Guvahati has experienced complex climate changes: In 1902, the hottest month was July, but in 2021 it has shifted to August. The increase in temperature at various stations is considered. At Srinagar station in 2021, compared to 1892, temperatures increased in June, September and October. Guvahati has a 120-year increase in December, January, March and April. Temperatures have risen in February, March and April at Jolhpur in 125 years, but have risen in February and March at New Delhi Station in 90 years. Despite the presence of tropical evergreen forests, the area around Guvahati Station is expected to experience strong warming.

Evaluating the Robustness of MDSS Maintenance Forecasts Using Connected Vehicle Data

The Indiana Department of Transportation (INDOT) adopted the Maintenance Decision Support System (MDSS) for user-defined plowing segments in the winter of 2008-2009. Since then, many new data sources, including connected vehicle data, enhanced weather data, and fleet telematics, have been integrated into INDOT winter operations activities. The objective of this study was to use these new data sources to conduct a systematic evaluation of the robustness of the MDSS forecasts. During the 2023-2024 winter season, 26 unique MDSS forecast data attributes were collected at 0, 1, 3, 6, 12 and 23-hour intervals from the observed storm time for 6 roadway segments during 13 individual storms. In total, over 888,000 MDSS data points were archived for this evaluation. This study developed novel visualizations to compare MDSS forecasts to multiple other independent data sources, including connected vehicle data, National Oceanic and Atmospheric Administration (NOAA) weather data, road friction data and snowplow telematics. Three Indiana storms, with varying characteristics and severity, were analyzed in detailed case studies. Those storms occurred on January 6th, 2024, January 13th, 2024 and February 16th, 2024. Incorporating these visualizations into winter weather after-action reports increases the robustness of post-storm performance analysis and allows road weather stakeholders to better understand the capabilities of MDSS. The results of this analysis will provide a framework for future MDSS evaluations and implementations as well as training tools for winter operation stakeholders in Indiana and beyond.

IMPACT OF VERTICAL RESOLUTION, MODEL TOP AND DATA ASSIMILATION ON WEATHER FORECASTING——A CASE STUDY

The impacts of stratospheric initial conditions and vertical resolution on the stratosphere by raising the model top,refining the vertical resolution,and the assimilation of operationally available observations,including conventional and satellite observations,on continental U.S.winter short-range weather forecasting,were investigated in this study.The initial and predicted wind and temperature profiles were analyzed against conventional observations.Generally,the initial wind and temperature bias profiles were better adjusted when a higher model top and refined vertical resolution were used.Negative impacts were also observed in both the initial wind and temperature profiles,over the lower troposphere.Different from the results by only raising the model top,the assimilation of operationally available observations led to significant improvements in both the troposphere and stratosphere initial conditions when a higher top was used.Predictions made with the adjusted stratospheric initial conditions and refined vertical resolutions showed generally better forecasting skill.The major improvements caused by raising the model top with refined vertical resolution,as well as those caused by data assimilation,were in both cases located in the tropopause and lower stratosphere.Negative impacts were also observed in the predicted near surface wind and lower-tropospheric temperature.These negative impacts were related to the uncertainties caused by more stratospheric information,as well as to some physical processes.A case study shows that when we raise the model top,put more vertical layers in stratosphere and apply data assimilation,the precipitation scores can be slightly improved.However,more analysis is needed due to uncertainties brought by data assimilation.