Accurate forecasting of wind speed is essential for daily life and social production. While numerical weather prediction products are widely used, they rely on global data and mathematical models to solve atmospheric dynamics' equations, often failing to capture localized micrometeorological phenomena accurately. Factors such as surface conditions, land‐sea differences, and topography, particularly in coastal areas, further impact the accuracy of wind speed forecasts. This study presents a new method to enhance short‐term wind speed forecasting along China's coast by incorporating local and neighborhood spatiotemporal information. The approach integrates meteorological data from adjacent grid points as new inputs in the LightGBM, CatBoost, and XGBoost algorithms. Stacking ensemble technique is then employed to effectively combine with the aforementioned foundational models. Two sets of experiments are conducted: Experiments 1 exclude surrounding information, while Experiments 2 include it. Each set consists of five experiment groups: annual, spring, summer, autumn, and winter. Within each group, four models are tested: XGBoost, LightGBM, CatBoost, and stacking. Results show that incorporating surrounding site information improves forecast accuracy. In all five groups with added surrounding site information, the stacking model performs best. Compared to ECMWF forecast data, the stacking model improves wind speed forecast accuracy from 53.3%, 50.9%, 55.2%, 53.0%, and 54.0% to 77.2%, 73.1%, 76.7%, 78.2%, and 77.1%, respectively. These findings demonstrate the potential effectiveness of the proposed method for improving short‐term wind speed forecasts in China's coastal areas.

The characteristics of the synoptic situation and the evolution of the organized warm-sector rainfalls (OWSRs) in the coastal region of South China in the pre-summer rainy season were investigated, using a period (2011–2016) of high-resolution observational data and European Centre for Medium-Range Weather Forecasts Re-Analysis Interim (ERA-Interim) data. The results show that a strong southwesterly low-level jet (LLJ) ahead of a trough over southwestern China with a marked boundary-layer jet (BLJ) over the northern South China Sea (synoptic situation SWLLJ) or a prominent, low-level anticyclone over the Yangtze River Basin (synoptic situation ACR) is present when the OWSRs occur in the coastal region of South China. The OWSRs are prone to initiate on the windward side of the coastal mountains, owing to the convergence enhanced by the colliding of the BLJ with the mountains and the coupling of double LLJs near the coast (for SWLLJ), or due to the convergence between northerly and southeasterly winds near the coastal mountains (for ACR). The OWSRs present a long extension when the LLJ axis is nearby. The translation of the LLJ itself also promotes the long extension of the OWSRs. In contrast, the OWSRs show a short extension when the LLJ axis is farther away or ACR occurs. Meanwhile, the OWSRs are directed northeastward in Guangxi Province and more eastward in Guangdong Province, probably owing to the orientation difference of the LLJ in these two provinces. The rainfall systems in the ACR situation tend to move eastward, whereas those in the SWLLJ situation are prone to move eastward when equivalently strong or much-stronger upper-level winds overlay the LLJ, but move northeastward when much weaker upper-level winds couple with the LLJ.

Eliminating the systematic errors of member models is a key step before the variant‐weight multi‐model ensemble forecasting. However, how to reasonably calculate model systematic errors is a problem worthy study. Taking surface temperature as the subject investigated, this study explores this problem through comparative analyses of a series of variant‐weight multi‐model ensemble forecasting experiments. The results showed that eliminating the systematic errors of member models dramatically improves the ensemble forecasting outcomes. The calculation of model systematic errors based on the optimal time lengths regarding the spatiotemporal distribution characteristics of model forecasting errors is a reasonable and effective approach. The surface temperature forecasts with model systematic errors calculated with the optimal time lengths and those beyond the optimal time lengths exhibit higher accuracies. Owing to the approximately uniform distribution of temperature, no significant difference happens between the accuracies of the surface temperature forecasts with model systematic errors calculated using the average and quantile methods. In addition, the updating of the models participating into the ensemble forecasting would weaken the representativeness of the optimal time lengths for model systematic errors calculation that are obtained based on historical period data, and thus, weaken the forecasting skill. The research results provide some train of thought for enhancing the variant‐weight multi‐model ensemble forecasting accuracy.