Abstract Accurately calculating the momentum exchange between atmosphere and ocean is essential for global weather forecasting. However, current widely used parameterization schemes face challenges in accurately estimating sea surface roughness . Deep learning (DL) technology offers new opportunities to address this problem. Here, we present a DL‐based scheme roughness, which have integrated into forecasting system. The demonstrates advantages over four traditional offline online tests. In...

In recent years, the study of numerical weather prediction (NWP) in complex orographic areas has attracted a great deal attention. Complex orography plays an important role occurrence and development extreme rainfall events. this study, Yin–He Global Spectrum Model (YHGSM) was used, wave number truncation method employed to decompose data different resolutions. The obtained with resolutions were used simulate Zhengzhou, Henan Province, China, discuss degree influence mechanism on rainfall....

Summary Among the numerical weather prediction models proposed, Yi‐he global spectral model (YHGSM) developed by National University of Defense Technology provides two two‐sided communication schemes that perform parallel computing for semi‐Lagrange convection. One is maximum wind scheme, and other on‐demand scheme. Although latter can greatly reduce volume, it needs to determine pattern interact with adjacent processes every time step, resulting in a limited performance improvement. This...

Complex orography is still a big challenge for all numerical weather prediction (NWP) models. Orography an important factor that affects the NWP results. The in mainly main accuracy of results through two aspects: orographic representation models dynamics and orography-related parameterization schemes physical processes. To ensure results, it necessary to have comprehensive understanding application data NWP. This paper summarized influence on weather, accuracy, parametrization drag Finally,...

Abstract Studying the influences of meteorological factors on air pollutants (including SO 2 , NO O 3 and PM 10 ) quantifying relationships between them should enable further investigation variation in a region, especially their temporal-spatial distribution characteristics. We combined observed data from ground-based monitoring sub-stations relevant Pearl River Delta region to investigate various factors, then established regression models for each sub-station through principal component...

Abstract In a typical non‐hydrostatic spectral dynamic numerical weather prediction (NWP) kernel, all forecast variables are transformed between grid point and spaces to compute their gradients solve the implicit problem. This kernel requires numerous transformations, which depend heavily on extensive global communication significantly hinder parallel computing efficiency. paper introduces an innovative that incorporates finite‐volume method within framework. We have developed horizontal...

Abstract The Yi‐he global spectral model (YHGSM) is a numerical weather prediction (NWP) developed by the National University of Defense Technology, China. YHGSM has two optimization techniques that have recently been adopted for on‐demand communication version semi‐Lagrangian scheme: one‐sided and an adaptive strategy maximum wind speed. However, limitation original computation cannot be performed simultaneously, resulting in waste resources. This article addresses this dividing array to...

The Yin-He global spectral model (YHGSM), embodies a parallel semi-Lagrangian solver and has two schemes implemented: maximum wind speed scheme on-demand communication scheme. Maximum adopts single fixed data structure, which large overhead. Although the overhead of is reduced, it still pretty huge. In this paper, novel adaptable approach proposed in monthly used YHGSM. This reduces difference between actual model; turn, trajectory computation further reduced. Experiments show that schemes,...

Numerical weather prediction (NWP) has become an important method of predicting extreme events, but orography is one the key factors affecting performance NWPs. In this paper, based on Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) elevation data, a for constructing global orographic dataset suitable NWP spectral models investigated. The Yin-He spectrum model (YHGSM) used to simulate early peak periods rainfall event 20 July 2021 in Henan Province, China, heavy rain...

The forecasting of disaster weather using spatial data mining technique is still at an initial stage presently. Recent researches have indicated that intensive precipitation in the Yangtze River Basin China closely related to activity Mesoscale Convective Systems (MCS) moving out Tibetan Plateau China, however, factors influence MCS trajectories are very complex. To discover and features environmental physical field favoring origination development over Plateau, this paper, automatically...

In the Yin-he Global Spectral Model (YHGSM) developed by National University of Defense Technology, an issue to be studied is that how determine parameters and achieve ideal execution time in model. Based on multiple regression analysis, we firstly construct a quantitative relationship between model giving set random its advance. Then establish equation them. this way, finally parameter evaluation (MPE) new model, more parameters, can predict their corresponding time. On basis, given...