It is fundamental and useful to investigate how deep learning forecasting models(DLMs)perform compared to operational oceanography forecast systems(OFSs).However,few studies have intercompared their performances using...It is fundamental and useful to investigate how deep learning forecasting models(DLMs)perform compared to operational oceanography forecast systems(OFSs).However,few studies have intercompared their performances using an identical reference.In this study,three physically reasonable DLMs are implemented for the forecasting of the sea surface temperature(SST),sea level anomaly(SLA),and sea surface velocity in the South China Sea.The DLMs are validated against both the testing dataset and the“OceanPredict”Class 4 dataset.Results show that the DLMs'RMSEs against the latter increase by 44%,245%,302%,and 109%for SST,SLA,current speed,and direction,respectively,compared to those against the former.Therefore,different references have significant influences on the validation,and it is necessary to use an identical and independent reference to intercompare the DLMs and OFSs.Against the Class 4 dataset,the DLMs present significantly better performance for SLA than the OFSs,and slightly better performances for other variables.The error patterns of the DLMs and OFSs show a high degree of similarity,which is reasonable from the viewpoint of predictability,facilitating further applications of the DLMs.For extreme events,the DLMs and OFSs both present large but similar forecast errors for SLA and current speed,while the DLMs are likely to give larger errors for SST and current direction.This study provides an evaluation of the forecast skills of commonly used DLMs and provides an example to objectively intercompare different DLMs.展开更多
The frequent outbreaks of crop diseases pose a serious threat to global agricultural production and food security.Data-driven forecasting models have emerged as an effective approach to support early warning and manag...The frequent outbreaks of crop diseases pose a serious threat to global agricultural production and food security.Data-driven forecasting models have emerged as an effective approach to support early warning and management,yet the lack of user-friendly tools for model development remains a major bottleneck.This study presents the Multi-Scenario Crop Disease Forecasting Modeling System(MSDFS),an open-source platform that enables end-to-end model construction-from multi-source data ingestion and feature engineering to training,evaluation,and deployment-across four representative scenarios:static point-based,static grid-based,dynamic point-based,and dynamic grid-based.Unlike conventional frameworks,MSDFS emphasizes modeling flexibility,allowing users to build,compare,and interpret diverse forecasting approaches within a unified workflow.A notable feature of the system is the integration of a weather scenario generator,which facilitates comprehensive testing of model performance and adaptability under extreme climatic conditions.Case studies corresponding to the four scenarios were used to validate the system,with overall accuracy(OA)ranging from 73%to 93%.By lowering technical barriers,the system is designed to serve plant protection managers and agricultural producers without advanced programming expertise,providing a practical modeling tool that supports the construction of smart plant protection systems.展开更多
[Objectives]To assess the effectiveness of the intelligent small insect monitoring and forecasting system developed by Zhejiang Top Cloud-Agri Technology Co.,Ltd.in monitoring,providing early warnings,and identifying ...[Objectives]To assess the effectiveness of the intelligent small insect monitoring and forecasting system developed by Zhejiang Top Cloud-Agri Technology Co.,Ltd.in monitoring,providing early warnings,and identifying rice planthoppers.[Methods]In 2024,an experiment involving the automatic identification and counting of rice planthoppers was conducted using the intelligent small insect monitoring and forecasting system in the rice production demonstration area of Qingxichang Sub-district,Xiushan Autonomous County,Chongqing City.The results obtained were subsequently compared and analyzed against those derived from manual identification.[Results]The intelligent small insect monitoring and forecasting system achieved recognition accuracy rates of 95.14%,94.25%,and 97.78% for Nilaparvata lugens,Sogatella furcifera,and Laodelphax striatellus,respectively,resulting in an average accuracy rate of 95.72%.The outcomes derived from automatic recognition closely corresponded with those obtained through manual identification.[Conclusions]This research provides a reference for the optimization of the intelligent small insect monitoring and forecasting system.展开更多
Urban air pollution has brought great troubles to physical and mental health,economic development,environmental protection,and other aspects.Predicting the changes and trends of air pollution can provide a scientific ...Urban air pollution has brought great troubles to physical and mental health,economic development,environmental protection,and other aspects.Predicting the changes and trends of air pollution can provide a scientific basis for governance and prevention efforts.In this paper,we propose an interval prediction method that considers the spatio-temporal characteristic information of PM_(2.5)signals from multiple stations.K-nearest neighbor(KNN)algorithm interpolates the lost signals in the process of collection,transmission,and storage to ensure the continuity of data.Graph generative network(GGN)is used to process time-series meteorological data with complex structures.The graph U-Nets framework is introduced into the GGN model to enhance its controllability to the graph generation process,which is beneficial to improve the efficiency and robustness of the model.In addition,sparse Bayesian regression is incorporated to improve the dimensional disaster defect of traditional kernel density estimation(KDE)interval prediction.With the support of sparse strategy,sparse Bayesian regression kernel density estimation(SBR-KDE)is very efficient in processing high-dimensional large-scale data.The PM_(2.5)data of spring,summer,autumn,and winter from 34 air quality monitoring sites in Beijing verified the accuracy,generalization,and superiority of the proposed model in interval prediction.展开更多
The solar cycle(SC),a phenomenon caused by the quasi-periodic regular activities in the Sun,occurs approximately every 11 years.Intense solar activity can disrupt the Earth’s ionosphere,affecting communication and na...The solar cycle(SC),a phenomenon caused by the quasi-periodic regular activities in the Sun,occurs approximately every 11 years.Intense solar activity can disrupt the Earth’s ionosphere,affecting communication and navigation systems.Consequently,accurately predicting the intensity of the SC holds great significance,but predicting the SC involves a long-term time series,and many existing time series forecasting methods have fallen short in terms of accuracy and efficiency.The Time-series Dense Encoder model is a deep learning solution tailored for long time series prediction.Based on a multi-layer perceptron structure,it outperforms the best previously existing models in accuracy,while being efficiently trainable on general datasets.We propose a method based on this model for SC forecasting.Using a trained model,we predict the test set from SC 19 to SC 25 with an average mean absolute percentage error of 32.02,root mean square error of 30.3,mean absolute error of 23.32,and R^(2)(coefficient of determination)of 0.76,outperforming other deep learning models in terms of accuracy and training efficiency on sunspot number datasets.Subsequently,we use it to predict the peaks of SC 25 and SC 26.For SC 25,the peak time has ended,but a stronger peak is predicted for SC 26,of 199.3,within a range of 170.8-221.9,projected to occur during April 2034.展开更多
Accurate short-term electricity price forecasts are essential for market participants to optimize bidding strategies,hedge risk and plan generation schedules.By leveraging advanced data analytics and machine learning ...Accurate short-term electricity price forecasts are essential for market participants to optimize bidding strategies,hedge risk and plan generation schedules.By leveraging advanced data analytics and machine learning methods,accurate and reliable price forecasts can be achieved.This study forecasts day-ahead prices in Türkiye’s electricity market using eXtreme Gradient Boosting(XGBoost).We benchmark XGBoost against four alternatives—Support Vector Machines(SVM),Long Short-Term Memory(LSTM),Random Forest(RF),and Gradient Boosting(GBM)—using 8760 hourly observations from 2023 provided by Energy Exchange Istanbul(EXIST).All models were trained on an identical chronological 80/20 train–test split,with hyperparameters tuned via 5-fold cross-validation on the training set.XGBoost achieved the best performance(Mean Absolute Error(MAE)=144.8 TRY/MWh,Root Mean Square Error(RMSE)=201.8 TRY/MWh,coefficient of determination(R^(2))=0.923)while training in 94 s.To enhance interpretability and identify key drivers,we employed Shapley Additive Explanations(SHAP),which highlighted a strong association between higher prices and increased natural-gas-based generation.The results provide a clear performance benchmark and practical guidance for selecting forecasting approaches in day-ahead electricity markets.展开更多
Deep learning-based methods have become alternatives to traditional numerical weather prediction systems,offering faster computation and the ability to utilize large historical datasets.However,the application of deep...Deep learning-based methods have become alternatives to traditional numerical weather prediction systems,offering faster computation and the ability to utilize large historical datasets.However,the application of deep learning to medium-range regional weather forecasting with limited data remains a significant challenge.In this work,three key solutions are proposed:(1)motivated by the need to improve model performance in data-scarce regional forecasting scenarios,the authors innovatively apply semantic segmentation models,to better capture spatiotemporal features and improve prediction accuracy;(2)recognizing the challenge of overfitting and the inability of traditional noise-based data augmentation methods to effectively enhance model robustness,a novel learnable Gaussian noise mechanism is introduced that allows the model to adaptively optimize perturbations for different locations,ensuring more effective learning;and(3)to address the issue of error accumulation in autoregressive prediction,as well as the challenge of learning difficulty and the lack of intermediate data utilization in one-shot prediction,the authors propose a cascade prediction approach that effectively resolves these problems while significantly improving model forecasting performance.The method achieves a competitive result in The East China Regional AI Medium Range Weather Forecasting Competition.Ablation experiments further validate the effectiveness of each component,highlighting their contributions to enhancing prediction performance.展开更多
作为天气系统的主要组成部分,三维云仿真在军事、航空等领域都起着重要作用.目前主流的边界体积层次结构(Bounding Volume Hierarchy,BVH)在处理形状不均匀且体积较大的云时存在渲染效率低下的问题,为此提出一种基于优化BVH算法的云产...作为天气系统的主要组成部分,三维云仿真在军事、航空等领域都起着重要作用.目前主流的边界体积层次结构(Bounding Volume Hierarchy,BVH)在处理形状不均匀且体积较大的云时存在渲染效率低下的问题,为此提出一种基于优化BVH算法的云产品渲染方法.将WRF(Weather Research and Forecasting,天气研究与预报)模型网格点中的数据作为云基元,利用Z-order Hilbert曲线对其进行空间排序,结合云基元密度优化BVH算法,提高计算效率.提出ONS(Overlapping Node Sets,重叠节点结构)降低数据存取耗时.优化BVH算法能够减少不必要的光线和三角形面之间的相交测试次数,并解决边界体无效重叠问题.仿真实验显示,SAH(Surface Area Heuristic,表面积启发式)成本较同类最优算法可提升15.6%,EPO(Effective Partial Overlap,有效重叠部分)可提升10%,构建时间减少100%以上,在任意云场景中优化BVH算法的计算效率较同类算法都有显著提高,表明其能实现WRF云产品的快速渲染.展开更多
We investigated the impact of tuning the length scale of the background error covariance in the Weather Research and Forecasting （WRF） three-dimensional variational assimilation （3DVAR） system. In particular, we s...We investigated the impact of tuning the length scale of the background error covariance in the Weather Research and Forecasting （WRF） three-dimensional variational assimilation （3DVAR） system. In particular, we studied the effect of this parameter on the assimilation of high-resolution surface data for heavy rainfall forecasts associated with mesoscale convective systems over the Korean Peninsula. In the assimilation of high-resolution surface data, the National Meteorological Center method tended to exaggerate the length scale that determined the shape and extent to which observed information spreads out. In this study, we used the difference between observation and background data to tune the length scale in the assimilation of high-resolution surface data. The resulting assimilation clearly showed that the analysis with the tuned length scale was able to reproduce the small-scale features of the ideal field effectively. We also investigated the effect of a double-iteration method with two different length scales, representing large and small-length scales in the WRF-3DVAR. This method reflected the large and small-scale features of observed information in the model fields. The quantitative accuracy of the precipitation forecast using this double iteration with two different length scales for heavy rainfall was high; results were in good agreement with observations in terms of the maximum rainfall amount and equitable threat scores. The improved forecast in the experiment resulted from the development of well-identified mesoscale convective systems by intensified low-level winds and their consequent convergence near the rainfall area.展开更多
An air pollution forecast system,ARIA Regional,was implemented in 2007–2008 at the Beijing Municipality Environmental Monitoring Center,providing daily forecast of main pollutant concentrations.The chemistry-transpor...An air pollution forecast system,ARIA Regional,was implemented in 2007–2008 at the Beijing Municipality Environmental Monitoring Center,providing daily forecast of main pollutant concentrations.The chemistry-transport model CHIMERE was coupled with the dust emission model MB95 for restituting dust storm events in springtime so as to improve forecast results.Dust storm events were sporadic but could be extremely intense and then control air quality indexes close to the source areas but also far in the Beijing area.A dust episode having occurred at the end of May 2008 was analyzed in this article,and its impact of particulate matter on the Chinese air pollution index （API） was evaluated.Following our estimation,about 23 Tg of dust were emitted from source areas in Mongolia and in the Inner Mongolia of China,transporting towards southeast.This episode of dust storm influenced a large part of North China and East China,and also South Korea.The model result was then evaluated using satellite observations and in situ data.The simulated daily concentrations of total suspended particulate at 6：00 UTC had a similar spatial pattern with respect to OMI satellite aerosol index.Temporal evolution of dust plume was evaluated by comparing dust aerosol optical depth （AOD） calculated from the simulations with AOD derived from MODIS satellite products.Finally,the comparison of reported Chinese API in Beijing with API calculated from the simulation including dust emissions had showed the significant improvement of the model results taking into accountmineral dust correctly.展开更多
The results from a hybrid approach that combines a mesoscale meteorological model with a diagnostic model to produce high-resolution wind fields in complex coastal topography are evaluated.The diagnostic wind model(Ca...The results from a hybrid approach that combines a mesoscale meteorological model with a diagnostic model to produce high-resolution wind fields in complex coastal topography are evaluated.The diagnostic wind model(California Meteorological Model,CALMET) with 100-m horizontal spacing was driven with outputs from the Weather Research and Forecasting(WRF) model to obtain near-surface winds for the 1-year period from 12 September 2003 to 11 September 2004.Results were compared with wind observations at four sites.Traditional statistical scores,including correlation coefficients,standard deviations(SDs) and mean absolute errors(MAEs),indicate that the wind estimates from the WRF/CALMET modeling system are produced reasonably well.The correlation coefficients are relatively large,ranging from 0.5 to 0.7 for the zonal wind component and from 0.75 to 0.85 for the meridional wind component.MAEs for wind speed range from 1.5 to 2.0 m s-1 at 10 meters above ground level(AGL) and from 2.0 to 2.5 m s-1 at 60 m AGL.MAEs for wind direction range from 30 to 40 degrees at both levels.A spectral decomposition of the time series of wind speed shows positive impacts of CALMET in improving the mesoscale winds.Moreover,combining the CALMET model with WRF significantly improves the spatial variability of the simulated wind fields.It can be concluded that the WRF/CALMET modeling system is capable of providing a detailed near-surface wind field,but the physics in the diagnostic CALMET model needs to be further improved.展开更多
为提升低空风切变预报精度,本文综合运用欧洲中期天气预报中心第五代再分析资料[European Centre for Medium-Range Weather Forecasts(ECMWF)fifth-generation reanalysis data,ERA5]和美国国家环境预报中心(National Centers for Envi...为提升低空风切变预报精度,本文综合运用欧洲中期天气预报中心第五代再分析资料[European Centre for Medium-Range Weather Forecasts(ECMWF)fifth-generation reanalysis data,ERA5]和美国国家环境预报中心(National Centers for Environmental Prediction,NCEP)的FNL全球再分析资料(Final Operational Global Analysis)、先进星载热发射和反射辐射仪全球数字高程模型以及兰州中川机场的实况观测资料,采用中尺度数值天气预报模式(Weather Research and Forecasting Model,WRF)、WRF结合计算流体动力学(Computational Fluid Dynamics,CFD)方法、长短期神经网络(Long Short-Term Memory,LSTM)方法,对2021年4月15-16日兰州中川机场的两次风切变过程进行模拟分析。结果表明:(1)在小于1 km的网格中使用大涡模拟,WRF模式在单个站点风速模拟任务中表现更好,但在近地面水平风场风速模拟效果上,不如WRF模式结合计算流体力学模型方案;(2)对于飞机降落过程中遭遇的两次低空风切变的模拟,WRF-LES和WRF-CFD两种模式都可以模拟出第一次低空风切变,而第二次受传入模式的WRF风速数据值较小的影响,两种模式风速差都没有达到阈值,需要在后续工作中进一步验证;(3)低风速条件(6 m·s^(-1))下,基于LSTM的单变量风速预测模型平均绝对误差基本维持在0.59 m·s^(-1),能较好地把握不同地形与环流背景条件下风速变化的非线性关系,虽然受到WRF误差和观测要素不全的限制,多变量风速预测能在保证平均绝对百分比误差小于6.60%的情况下,以更高的计算效率和泛化能力实现风速预测。本文不仅验证了WRF-CFD和WRF-LES耦合方案在风场和低空风切变预报中的差异,还探讨了基于LSTM的风速预测的可行性和准确性,期望为提高风场模拟精度,缩短精细风场模拟时间提供新的视角和方法。展开更多
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 ...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.展开更多
A dual-resolution（DR） version of a regional ensemble Kalman filter（EnKF）-3D ensemble variational（3DEnVar） coupled hybrid data assimilation system is implemented as a prototype for the operational Rapid Refresh f...A dual-resolution（DR） version of a regional ensemble Kalman filter（EnKF）-3D ensemble variational（3DEnVar） coupled hybrid data assimilation system is implemented as a prototype for the operational Rapid Refresh forecasting system. The DR 3DEnVar system combines a high-resolution（HR） deterministic background forecast with lower-resolution（LR） EnKF ensemble perturbations used for flow-dependent background error covariance to produce a HR analysis. The computational cost is substantially reduced by running the ensemble forecasts and EnKF analyses at LR. The DR 3DEnVar system is tested with 3-h cycles over a 9-day period using a 40/13-km grid spacing combination. The HR forecasts from the DR hybrid analyses are compared with forecasts launched from HR Gridpoint Statistical Interpolation（GSI） 3D variational（3DVar）analyses, and single LR hybrid analyses interpolated to the HR grid. With the DR 3DEnVar system, a 90% weight for the ensemble covariance yields the lowest forecast errors and the DR hybrid system clearly outperforms the HR GSI 3DVar.Humidity and wind forecasts are also better than those launched from interpolated LR hybrid analyses, but the temperature forecasts are slightly worse. The humidity forecasts are improved most. For precipitation forecasts, the DR 3DEnVar always outperforms HR GSI 3DVar. It also outperforms the LR 3DEnVar, except for the initial forecast period and lower thresholds.展开更多
In this study,an operational forecasting system of sea dike risk in the southern Zhejiang Province,South China was developed based on a coupled storm-surge and wave model.This forecasting system is important because o...In this study,an operational forecasting system of sea dike risk in the southern Zhejiang Province,South China was developed based on a coupled storm-surge and wave model.This forecasting system is important because of the high cost of storm-surge damage and the need for rapid emergency planning.A comparison with astronomical tides in 2016 and the validation of storm surges and high water marks of 20 typhoons verified that the forecast system has a good simulation ability.The system can forecast relatively realistic water levels and wave heights as shown under the parametric atmospheric forces simulated in a case study;the sea dikes in credible high risk were mainly located in the estuaries,rivers,and around the islands in the southern Zhejiang.Therefore,the forecast system is applicable in the southern Zhejiang with a support to the effective prevention from typhoon storm-surge damage.展开更多
An operational ocean circulation-surface wave coupled forecasting system for the seas off China and adjacent areas（OCFS-C） is developed based on parallelized circulation and wave models. It has been in operation sin...An operational ocean circulation-surface wave coupled forecasting system for the seas off China and adjacent areas（OCFS-C） is developed based on parallelized circulation and wave models. It has been in operation since November 1, 2007. In this paper we comprehensively present the simulation and verification of the system, whose distinguishing feature is that the wave-induced mixing is coupled in the circulation model. In particular, with nested technique the resolution in the China＇s seas has been updated to（1/24）° from the global model with（1/2）°resolution. Besides, daily remote sensing sea surface temperature（SST） data have been assimilated into the model to generate a hot restart field for OCFS-C. Moreover, inter-comparisons between forecasting and independent observational data are performed to evaluate the effectiveness of OCFS-C in upper-ocean quantities predictions, including SST, mixed layer depth（MLD） and subsurface temperature. Except in conventional statistical metrics, non-dimensional skill scores（SS） is also used to evaluate forecast skill. Observations from buoys and Argo profiles are used for lead time and real time validations, which give a large SS value（more than 0.90）. Besides, prediction skill for the seasonal variation of SST is confirmed. Comparisons of subsurface temperatures with Argo profiles data indicate that OCFS-C has low skill in predicting subsurface temperatures between 100 m and 150 m. Nevertheless, inter-comparisons of MLD reveal that the MLD from model is shallower than that from Argo profiles by about 12 m, i.e., OCFS-C is successful and steady in MLD predictions. Validation of 1-d, 2-d and 3-d forecasting SST shows that our operational ocean circulation-surface wave coupled forecasting model has reasonable accuracy in the upper ocean.展开更多
The generally used methods of forecasting coal requirement quantity include the analogy method, the outside push method and the cause effect analysis method. However, the precision of forecasting results using these m...The generally used methods of forecasting coal requirement quantity include the analogy method, the outside push method and the cause effect analysis method. However, the precision of forecasting results using these methods is lower. This paper uses the grey system theory, and sets up grey forecasting model GM (1, 3) to coal requirement quantity. The forecasting result for the Chinese coal requirement quantity coincides with the actual values, and this shows that the model is reliable. Finally, this model are used to forecast Chinese coal requirement quantity in the future ten years.展开更多
The floods in river Mahanadi delta are due to either dam release of Hirakud or due to contribution of intercepted catchment between Hirakud dam and delta. It is seen from post-Hirakud periods (1958) that out of 19 flo...The floods in river Mahanadi delta are due to either dam release of Hirakud or due to contribution of intercepted catchment between Hirakud dam and delta. It is seen from post-Hirakud periods (1958) that out of 19 floods 14 are due to intercepted catchment contribution. The existing flood forecasting systems are mostly for upstream catchment, forecasting the inflow to reservoir, whereas the downstream catchment is devoid of a sound flood forecasting system. Therefore, in this study an attempt has been made to develop a workable forecasting system for downstream catchment. Instead of taking the flow time series concurrent flood peaks of 12 years of base and forecasting stations with its corresponding travel time are considered for analysis. Both statistical method and ANN based approach are considered for finding the peak to reach at delta head with its corresponding travel time. The travel time has been finalized adopting clustering techniques, there by differentiating high, medium and low peaks. The method is simple and it does not take into consideration the rainfall and other factors in the intercepted catchment. A comparison between both methods are tested and it is found that the ANN methods are better beyond the calibration range over statistical method and the efficiency of either methods reduces as the prediction reach is extended. However, it is able to give the peak discharge at delta head before 24 hour to 37 hour for high to low peaks.展开更多
基金supported by the National Natural Science Foundation of China(Grant Nos.42375062 and 42275158)the National Key Scientific and Technological Infrastructure project“Earth System Science Numerical Simulator Facility”(EarthLab)the Natural Science Foundation of Gansu Province(Grant No.22JR5RF1080)。
文摘It is fundamental and useful to investigate how deep learning forecasting models(DLMs)perform compared to operational oceanography forecast systems(OFSs).However,few studies have intercompared their performances using an identical reference.In this study,three physically reasonable DLMs are implemented for the forecasting of the sea surface temperature(SST),sea level anomaly(SLA),and sea surface velocity in the South China Sea.The DLMs are validated against both the testing dataset and the“OceanPredict”Class 4 dataset.Results show that the DLMs'RMSEs against the latter increase by 44%,245%,302%,and 109%for SST,SLA,current speed,and direction,respectively,compared to those against the former.Therefore,different references have significant influences on the validation,and it is necessary to use an identical and independent reference to intercompare the DLMs and OFSs.Against the Class 4 dataset,the DLMs present significantly better performance for SLA than the OFSs,and slightly better performances for other variables.The error patterns of the DLMs and OFSs show a high degree of similarity,which is reasonable from the viewpoint of predictability,facilitating further applications of the DLMs.For extreme events,the DLMs and OFSs both present large but similar forecast errors for SLA and current speed,while the DLMs are likely to give larger errors for SST and current direction.This study provides an evaluation of the forecast skills of commonly used DLMs and provides an example to objectively intercompare different DLMs.
基金supported by Zhejiang Provincial Natural Science Foundation of China(Grant No.LR25D010003)The Zhejiang Provincial Key Research and Development Program(Grant No.2023C02018)National Natural Science Foundation of China(Grant No.42401400).
文摘The frequent outbreaks of crop diseases pose a serious threat to global agricultural production and food security.Data-driven forecasting models have emerged as an effective approach to support early warning and management,yet the lack of user-friendly tools for model development remains a major bottleneck.This study presents the Multi-Scenario Crop Disease Forecasting Modeling System(MSDFS),an open-source platform that enables end-to-end model construction-from multi-source data ingestion and feature engineering to training,evaluation,and deployment-across four representative scenarios:static point-based,static grid-based,dynamic point-based,and dynamic grid-based.Unlike conventional frameworks,MSDFS emphasizes modeling flexibility,allowing users to build,compare,and interpret diverse forecasting approaches within a unified workflow.A notable feature of the system is the integration of a weather scenario generator,which facilitates comprehensive testing of model performance and adaptability under extreme climatic conditions.Case studies corresponding to the four scenarios were used to validate the system,with overall accuracy(OA)ranging from 73%to 93%.By lowering technical barriers,the system is designed to serve plant protection managers and agricultural producers without advanced programming expertise,providing a practical modeling tool that supports the construction of smart plant protection systems.
文摘[Objectives]To assess the effectiveness of the intelligent small insect monitoring and forecasting system developed by Zhejiang Top Cloud-Agri Technology Co.,Ltd.in monitoring,providing early warnings,and identifying rice planthoppers.[Methods]In 2024,an experiment involving the automatic identification and counting of rice planthoppers was conducted using the intelligent small insect monitoring and forecasting system in the rice production demonstration area of Qingxichang Sub-district,Xiushan Autonomous County,Chongqing City.The results obtained were subsequently compared and analyzed against those derived from manual identification.[Results]The intelligent small insect monitoring and forecasting system achieved recognition accuracy rates of 95.14%,94.25%,and 97.78% for Nilaparvata lugens,Sogatella furcifera,and Laodelphax striatellus,respectively,resulting in an average accuracy rate of 95.72%.The outcomes derived from automatic recognition closely corresponded with those obtained through manual identification.[Conclusions]This research provides a reference for the optimization of the intelligent small insect monitoring and forecasting system.
基金Project(2020YFC2008605)supported by the National Key Research and Development Project of ChinaProject(52072412)supported by the National Natural Science Foundation of ChinaProject(2021JJ30359)supported by the Natural Science Foundation of Hunan Province,China。
文摘Urban air pollution has brought great troubles to physical and mental health,economic development,environmental protection,and other aspects.Predicting the changes and trends of air pollution can provide a scientific basis for governance and prevention efforts.In this paper,we propose an interval prediction method that considers the spatio-temporal characteristic information of PM_(2.5)signals from multiple stations.K-nearest neighbor(KNN)algorithm interpolates the lost signals in the process of collection,transmission,and storage to ensure the continuity of data.Graph generative network(GGN)is used to process time-series meteorological data with complex structures.The graph U-Nets framework is introduced into the GGN model to enhance its controllability to the graph generation process,which is beneficial to improve the efficiency and robustness of the model.In addition,sparse Bayesian regression is incorporated to improve the dimensional disaster defect of traditional kernel density estimation(KDE)interval prediction.With the support of sparse strategy,sparse Bayesian regression kernel density estimation(SBR-KDE)is very efficient in processing high-dimensional large-scale data.The PM_(2.5)data of spring,summer,autumn,and winter from 34 air quality monitoring sites in Beijing verified the accuracy,generalization,and superiority of the proposed model in interval prediction.
基金supported by the Academic Research Projects of Beijing Union University(ZK20202204)the National Natural Science Foundation of China(12250005,12073040,12273059,11973056,12003051,11573037,12073041,11427901,11572005,11611530679 and 12473052)+1 种基金the Strategic Priority Research Program of the China Academy of Sciences(XDB0560000,XDA15052200,XDB09040200,XDA15010700,XDB0560301,and XDA15320102)the Chinese Meridian Project(CMP).
文摘The solar cycle(SC),a phenomenon caused by the quasi-periodic regular activities in the Sun,occurs approximately every 11 years.Intense solar activity can disrupt the Earth’s ionosphere,affecting communication and navigation systems.Consequently,accurately predicting the intensity of the SC holds great significance,but predicting the SC involves a long-term time series,and many existing time series forecasting methods have fallen short in terms of accuracy and efficiency.The Time-series Dense Encoder model is a deep learning solution tailored for long time series prediction.Based on a multi-layer perceptron structure,it outperforms the best previously existing models in accuracy,while being efficiently trainable on general datasets.We propose a method based on this model for SC forecasting.Using a trained model,we predict the test set from SC 19 to SC 25 with an average mean absolute percentage error of 32.02,root mean square error of 30.3,mean absolute error of 23.32,and R^(2)(coefficient of determination)of 0.76,outperforming other deep learning models in terms of accuracy and training efficiency on sunspot number datasets.Subsequently,we use it to predict the peaks of SC 25 and SC 26.For SC 25,the peak time has ended,but a stronger peak is predicted for SC 26,of 199.3,within a range of 170.8-221.9,projected to occur during April 2034.
文摘Accurate short-term electricity price forecasts are essential for market participants to optimize bidding strategies,hedge risk and plan generation schedules.By leveraging advanced data analytics and machine learning methods,accurate and reliable price forecasts can be achieved.This study forecasts day-ahead prices in Türkiye’s electricity market using eXtreme Gradient Boosting(XGBoost).We benchmark XGBoost against four alternatives—Support Vector Machines(SVM),Long Short-Term Memory(LSTM),Random Forest(RF),and Gradient Boosting(GBM)—using 8760 hourly observations from 2023 provided by Energy Exchange Istanbul(EXIST).All models were trained on an identical chronological 80/20 train–test split,with hyperparameters tuned via 5-fold cross-validation on the training set.XGBoost achieved the best performance(Mean Absolute Error(MAE)=144.8 TRY/MWh,Root Mean Square Error(RMSE)=201.8 TRY/MWh,coefficient of determination(R^(2))=0.923)while training in 94 s.To enhance interpretability and identify key drivers,we employed Shapley Additive Explanations(SHAP),which highlighted a strong association between higher prices and increased natural-gas-based generation.The results provide a clear performance benchmark and practical guidance for selecting forecasting approaches in day-ahead electricity markets.
基金supported by the National Natural Science Foundation of China[grant number 62376217]the Young Elite Scientists Sponsorship Program by CAST[grant number 2023QNRC001]the Joint Research Project for Meteorological Capacity Improvement[grant number 24NLTSZ003]。
文摘Deep learning-based methods have become alternatives to traditional numerical weather prediction systems,offering faster computation and the ability to utilize large historical datasets.However,the application of deep learning to medium-range regional weather forecasting with limited data remains a significant challenge.In this work,three key solutions are proposed:(1)motivated by the need to improve model performance in data-scarce regional forecasting scenarios,the authors innovatively apply semantic segmentation models,to better capture spatiotemporal features and improve prediction accuracy;(2)recognizing the challenge of overfitting and the inability of traditional noise-based data augmentation methods to effectively enhance model robustness,a novel learnable Gaussian noise mechanism is introduced that allows the model to adaptively optimize perturbations for different locations,ensuring more effective learning;and(3)to address the issue of error accumulation in autoregressive prediction,as well as the challenge of learning difficulty and the lack of intermediate data utilization in one-shot prediction,the authors propose a cascade prediction approach that effectively resolves these problems while significantly improving model forecasting performance.The method achieves a competitive result in The East China Regional AI Medium Range Weather Forecasting Competition.Ablation experiments further validate the effectiveness of each component,highlighting their contributions to enhancing prediction performance.
文摘作为天气系统的主要组成部分,三维云仿真在军事、航空等领域都起着重要作用.目前主流的边界体积层次结构(Bounding Volume Hierarchy,BVH)在处理形状不均匀且体积较大的云时存在渲染效率低下的问题,为此提出一种基于优化BVH算法的云产品渲染方法.将WRF(Weather Research and Forecasting,天气研究与预报)模型网格点中的数据作为云基元,利用Z-order Hilbert曲线对其进行空间排序,结合云基元密度优化BVH算法,提高计算效率.提出ONS(Overlapping Node Sets,重叠节点结构)降低数据存取耗时.优化BVH算法能够减少不必要的光线和三角形面之间的相交测试次数,并解决边界体无效重叠问题.仿真实验显示,SAH(Surface Area Heuristic,表面积启发式)成本较同类最优算法可提升15.6%,EPO(Effective Partial Overlap,有效重叠部分)可提升10%,构建时间减少100%以上,在任意云场景中优化BVH算法的计算效率较同类算法都有显著提高,表明其能实现WRF云产品的快速渲染.
基金supported by International S&T Cooperation Program through the National Research Foundation of Korea(NRF) funded by the Ministry of Education,Science and Technology(MEST)(2011-00265)the BK21 program of the Korean Government Ministry of Education
文摘We investigated the impact of tuning the length scale of the background error covariance in the Weather Research and Forecasting （WRF） three-dimensional variational assimilation （3DVAR） system. In particular, we studied the effect of this parameter on the assimilation of high-resolution surface data for heavy rainfall forecasts associated with mesoscale convective systems over the Korean Peninsula. In the assimilation of high-resolution surface data, the National Meteorological Center method tended to exaggerate the length scale that determined the shape and extent to which observed information spreads out. In this study, we used the difference between observation and background data to tune the length scale in the assimilation of high-resolution surface data. The resulting assimilation clearly showed that the analysis with the tuned length scale was able to reproduce the small-scale features of the ideal field effectively. We also investigated the effect of a double-iteration method with two different length scales, representing large and small-length scales in the WRF-3DVAR. This method reflected the large and small-scale features of observed information in the model fields. The quantitative accuracy of the precipitation forecast using this double iteration with two different length scales for heavy rainfall was high; results were in good agreement with observations in terms of the maximum rainfall amount and equitable threat scores. The improved forecast in the experiment resulted from the development of well-identified mesoscale convective systems by intensified low-level winds and their consequent convergence near the rainfall area.
基金French Ministry of Economy and Finance is acknowledged for their financial support in the framework of the FASEP projectsupported by French ANRT CIFRE grant attributed to ARIA Technologies and LISA laboratories
文摘An air pollution forecast system,ARIA Regional,was implemented in 2007–2008 at the Beijing Municipality Environmental Monitoring Center,providing daily forecast of main pollutant concentrations.The chemistry-transport model CHIMERE was coupled with the dust emission model MB95 for restituting dust storm events in springtime so as to improve forecast results.Dust storm events were sporadic but could be extremely intense and then control air quality indexes close to the source areas but also far in the Beijing area.A dust episode having occurred at the end of May 2008 was analyzed in this article,and its impact of particulate matter on the Chinese air pollution index （API） was evaluated.Following our estimation,about 23 Tg of dust were emitted from source areas in Mongolia and in the Inner Mongolia of China,transporting towards southeast.This episode of dust storm influenced a large part of North China and East China,and also South Korea.The model result was then evaluated using satellite observations and in situ data.The simulated daily concentrations of total suspended particulate at 6：00 UTC had a similar spatial pattern with respect to OMI satellite aerosol index.Temporal evolution of dust plume was evaluated by comparing dust aerosol optical depth （AOD） calculated from the simulations with AOD derived from MODIS satellite products.Finally,the comparison of reported Chinese API in Beijing with API calculated from the simulation including dust emissions had showed the significant improvement of the model results taking into accountmineral dust correctly.
基金National Public Benefit Research Foundation of China (2008416048GYHY201006035)
文摘The results from a hybrid approach that combines a mesoscale meteorological model with a diagnostic model to produce high-resolution wind fields in complex coastal topography are evaluated.The diagnostic wind model(California Meteorological Model,CALMET) with 100-m horizontal spacing was driven with outputs from the Weather Research and Forecasting(WRF) model to obtain near-surface winds for the 1-year period from 12 September 2003 to 11 September 2004.Results were compared with wind observations at four sites.Traditional statistical scores,including correlation coefficients,standard deviations(SDs) and mean absolute errors(MAEs),indicate that the wind estimates from the WRF/CALMET modeling system are produced reasonably well.The correlation coefficients are relatively large,ranging from 0.5 to 0.7 for the zonal wind component and from 0.75 to 0.85 for the meridional wind component.MAEs for wind speed range from 1.5 to 2.0 m s-1 at 10 meters above ground level(AGL) and from 2.0 to 2.5 m s-1 at 60 m AGL.MAEs for wind direction range from 30 to 40 degrees at both levels.A spectral decomposition of the time series of wind speed shows positive impacts of CALMET in improving the mesoscale winds.Moreover,combining the CALMET model with WRF significantly improves the spatial variability of the simulated wind fields.It can be concluded that the WRF/CALMET modeling system is capable of providing a detailed near-surface wind field,but the physics in the diagnostic CALMET model needs to be further improved.
文摘为提升低空风切变预报精度,本文综合运用欧洲中期天气预报中心第五代再分析资料[European Centre for Medium-Range Weather Forecasts(ECMWF)fifth-generation reanalysis data,ERA5]和美国国家环境预报中心(National Centers for Environmental Prediction,NCEP)的FNL全球再分析资料(Final Operational Global Analysis)、先进星载热发射和反射辐射仪全球数字高程模型以及兰州中川机场的实况观测资料,采用中尺度数值天气预报模式(Weather Research and Forecasting Model,WRF)、WRF结合计算流体动力学(Computational Fluid Dynamics,CFD)方法、长短期神经网络(Long Short-Term Memory,LSTM)方法,对2021年4月15-16日兰州中川机场的两次风切变过程进行模拟分析。结果表明:(1)在小于1 km的网格中使用大涡模拟,WRF模式在单个站点风速模拟任务中表现更好,但在近地面水平风场风速模拟效果上,不如WRF模式结合计算流体力学模型方案;(2)对于飞机降落过程中遭遇的两次低空风切变的模拟,WRF-LES和WRF-CFD两种模式都可以模拟出第一次低空风切变,而第二次受传入模式的WRF风速数据值较小的影响,两种模式风速差都没有达到阈值,需要在后续工作中进一步验证;(3)低风速条件(6 m·s^(-1))下,基于LSTM的单变量风速预测模型平均绝对误差基本维持在0.59 m·s^(-1),能较好地把握不同地形与环流背景条件下风速变化的非线性关系,虽然受到WRF误差和观测要素不全的限制,多变量风速预测能在保证平均绝对百分比误差小于6.60%的情况下,以更高的计算效率和泛化能力实现风速预测。本文不仅验证了WRF-CFD和WRF-LES耦合方案在风场和低空风切变预报中的差异,还探讨了基于LSTM的风速预测的可行性和准确性,期望为提高风场模拟精度,缩短精细风场模拟时间提供新的视角和方法。
基金jointly supported by the Main Direction Program of Knowledge Innovation of the Chinese Academy of Sciences(Grant No.KZCX2EW203)the National Key Basic Research Program of China(Grant No.2013CB430105)the National Department of Public Benefit Research Foundation(Grant No.GYHY201006031)
文摘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.
基金supported by the National Natural Science Foundation of China (Grant Nos.41730965,41775099 and 2017YFC1502104)PAPD (the Priority Academic Program Development of Jiangsu Higher Education Institutions)
文摘A dual-resolution（DR） version of a regional ensemble Kalman filter（EnKF）-3D ensemble variational（3DEnVar） coupled hybrid data assimilation system is implemented as a prototype for the operational Rapid Refresh forecasting system. The DR 3DEnVar system combines a high-resolution（HR） deterministic background forecast with lower-resolution（LR） EnKF ensemble perturbations used for flow-dependent background error covariance to produce a HR analysis. The computational cost is substantially reduced by running the ensemble forecasts and EnKF analyses at LR. The DR 3DEnVar system is tested with 3-h cycles over a 9-day period using a 40/13-km grid spacing combination. The HR forecasts from the DR hybrid analyses are compared with forecasts launched from HR Gridpoint Statistical Interpolation（GSI） 3D variational（3DVar）analyses, and single LR hybrid analyses interpolated to the HR grid. With the DR 3DEnVar system, a 90% weight for the ensemble covariance yields the lowest forecast errors and the DR hybrid system clearly outperforms the HR GSI 3DVar.Humidity and wind forecasts are also better than those launched from interpolated LR hybrid analyses, but the temperature forecasts are slightly worse. The humidity forecasts are improved most. For precipitation forecasts, the DR 3DEnVar always outperforms HR GSI 3DVar. It also outperforms the LR 3DEnVar, except for the initial forecast period and lower thresholds.
基金Supported by the National Key Research and Development Program of China(No.2016YFC1402000)
文摘In this study,an operational forecasting system of sea dike risk in the southern Zhejiang Province,South China was developed based on a coupled storm-surge and wave model.This forecasting system is important because of the high cost of storm-surge damage and the need for rapid emergency planning.A comparison with astronomical tides in 2016 and the validation of storm surges and high water marks of 20 typhoons verified that the forecast system has a good simulation ability.The system can forecast relatively realistic water levels and wave heights as shown under the parametric atmospheric forces simulated in a case study;the sea dikes in credible high risk were mainly located in the estuaries,rivers,and around the islands in the southern Zhejiang.Therefore,the forecast system is applicable in the southern Zhejiang with a support to the effective prevention from typhoon storm-surge damage.
基金China-Korea Cooperation Project on the development of oceanic monitoring and prediction system on nuclear safetythe Project of the National Programme on Global Change and Air-sea Interaction under contract No.GASI-03-IPOVAI-05
文摘An operational ocean circulation-surface wave coupled forecasting system for the seas off China and adjacent areas（OCFS-C） is developed based on parallelized circulation and wave models. It has been in operation since November 1, 2007. In this paper we comprehensively present the simulation and verification of the system, whose distinguishing feature is that the wave-induced mixing is coupled in the circulation model. In particular, with nested technique the resolution in the China＇s seas has been updated to（1/24）° from the global model with（1/2）°resolution. Besides, daily remote sensing sea surface temperature（SST） data have been assimilated into the model to generate a hot restart field for OCFS-C. Moreover, inter-comparisons between forecasting and independent observational data are performed to evaluate the effectiveness of OCFS-C in upper-ocean quantities predictions, including SST, mixed layer depth（MLD） and subsurface temperature. Except in conventional statistical metrics, non-dimensional skill scores（SS） is also used to evaluate forecast skill. Observations from buoys and Argo profiles are used for lead time and real time validations, which give a large SS value（more than 0.90）. Besides, prediction skill for the seasonal variation of SST is confirmed. Comparisons of subsurface temperatures with Argo profiles data indicate that OCFS-C has low skill in predicting subsurface temperatures between 100 m and 150 m. Nevertheless, inter-comparisons of MLD reveal that the MLD from model is shallower than that from Argo profiles by about 12 m, i.e., OCFS-C is successful and steady in MLD predictions. Validation of 1-d, 2-d and 3-d forecasting SST shows that our operational ocean circulation-surface wave coupled forecasting model has reasonable accuracy in the upper ocean.
文摘The generally used methods of forecasting coal requirement quantity include the analogy method, the outside push method and the cause effect analysis method. However, the precision of forecasting results using these methods is lower. This paper uses the grey system theory, and sets up grey forecasting model GM (1, 3) to coal requirement quantity. The forecasting result for the Chinese coal requirement quantity coincides with the actual values, and this shows that the model is reliable. Finally, this model are used to forecast Chinese coal requirement quantity in the future ten years.
文摘The floods in river Mahanadi delta are due to either dam release of Hirakud or due to contribution of intercepted catchment between Hirakud dam and delta. It is seen from post-Hirakud periods (1958) that out of 19 floods 14 are due to intercepted catchment contribution. The existing flood forecasting systems are mostly for upstream catchment, forecasting the inflow to reservoir, whereas the downstream catchment is devoid of a sound flood forecasting system. Therefore, in this study an attempt has been made to develop a workable forecasting system for downstream catchment. Instead of taking the flow time series concurrent flood peaks of 12 years of base and forecasting stations with its corresponding travel time are considered for analysis. Both statistical method and ANN based approach are considered for finding the peak to reach at delta head with its corresponding travel time. The travel time has been finalized adopting clustering techniques, there by differentiating high, medium and low peaks. The method is simple and it does not take into consideration the rainfall and other factors in the intercepted catchment. A comparison between both methods are tested and it is found that the ANN methods are better beyond the calibration range over statistical method and the efficiency of either methods reduces as the prediction reach is extended. However, it is able to give the peak discharge at delta head before 24 hour to 37 hour for high to low peaks.
