Search

Drought is a natural disaster that occurs globally and can damage the environment, disrupt agricultural production and cause large economic losses. The accurate prediction of drought can effectively reduce the impacts of droughts. Deep learning methods have shown promise in drought prediction, with convolutional neural networks (CNNs) being particularly effective in handling spatial information. In this study, we employed a deep learning approach to predict drought in the Fenhe River (FHR) basin, taking into account the meteorological conditions of surrounding regions. We used the daily SAPEI (Standardized Antecedent Precipitation Evapotranspiration Index) as the drought evaluation index. Our results demonstrate the effectiveness of the CNN model in predicting drought events 1~10 days in advance. We evaluated the predictions made by the model; the average Nash–Sutcliffe efficiency (NSE) between the predicted and true values for the next 10 days was 0.71. While the prediction accuracy slightly decreased with longer prediction lengths, the model remained stable and effective in predicting heavy drought events that are typically difficult to predict. Additionally, key meteorological variables for drought predictions were identified, and we found that training the CNN model with these key variables led to higher prediction accuracy than training it with all variables. This study approves an effective deep learning approach for daily drought prediction, particularly when considering the meteorological conditions of surrounding regions.

The ability of soil moisture (SM) to affect precipitation (P) is a vital part of the water-energy cycles. Accurately quantifying this coupling enhances the ability to predict hydroclimatic extremes like floods and droughts. In this study, the ability of soil moisture to affect precipitation (SM-P) is characterized by two parts: the influence of soil moisture on evapotranspiration (SM-ET), and the influence of evapotranspiration on precipitation (ET-P). We determined localized ET-P by incorporating the coupling between latent heat flux (LH) and LCL height, to optimize the estimation of the SM-P. This approach links SM more closely to P by considering the influence of surface fluxes. The results indicate that CMIP6 models exhibited the anticipated hotspot patterns for the three coupling metrics in transition regions. However, we observed that climate models generally exhibit weaker SM-P coupling compared to reanalysis models. Both SM-ET and SM-P showcase higher values wherein wet climate regions during dry years, and the converse occurs in dry regions. Due to sensitivity to climate change, the ET-P exhibits a more pronounced upward trend in the future. This study helps understand P’s response to SM shifts in climate models, crucial for predicting hydrological extremes and coupled global warming impact.

The machine learning algorithms application in atmospheric sciences along the Earth System Models has the potential of improving prediction, forecast, and reconstruction of missing data. In the current study, a combination of two machine learning techniques namely K-means, and decision tree (C4.5) algorithms, are used to separate observed precipitation into clusters and classified the associated large-scale circulation indices. Observed precipitation from the Chinese Meteorological Agency (CMA) during 1961–2016 for 83 stations in the Poyang Lake basin (PLB) is used. The results from K-Means clusters show two precipitation clusters splitting the PLB precipitation into a northern and southern cluster, with a silhouette coefficient ~0.5. The PLB precipitation leading cluster (C1) contains 48 stations accounting for 58% of the regional station density, while Cluster 2 (C2) covers 35, accounting for 42% of the stations. The interannual variability in precipitation exhibited significant differences for both clusters. The decision tree (C4.5) is employed to explore the large-scale atmospheric indices from National Climate Center (NCC) associated with each cluster during the preceding spring season as a predictor. The C1 precipitation was linked with the location and intensity of subtropical ridgeline position over Northern Africa, whereas the C2 precipitation was suggested to be associated with the Atlantic-European Polar Vortex Area Index. The precipitation anomalies further validated the results of both algorithms. The findings are in accordance with previous studies conducted globally and hence recommend the applications of machine learning techniques in atmospheric science on a sub-regional and sub-seasonal scale. Future studies should explore the dynamics of the K-Means, and C4.5 derived indicators for a better assessment on a regional scale. This research based on machine learning methods may bring a new solution to climate forecast.

Soil moisture is an important parameter in land surface processes, which can control the surface energy and water budgets and thus affect the air temperature. Studying the coupling between soil moisture and air temperature is of vital importance for forecasting climate change. This study evaluates this coupling over China from 1980−2013 by using an energy-based diagnostic method, which represents the momentum, heat, and water conservation equations in the atmosphere, while the contributions of soil moisture are treated as external forcing. The results showed that the soil moisture−temperature coupling is strongest in the transitional climate zones between wet and dry climates, which here includes Northeast China and part of the Tibetan Plateau from a viewpoint of annual average. Furthermore, the soil moisture−temperature coupling was found to be stronger in spring than in the other seasons over China, and over different typical climatic zones, it also varied greatly in different seasons. We conducted two case studies (the heatwaves of 2013 in Southeast China and 2009 in North China) to understand the impact of soil moisture−temperature coupling during heatwaves. The results indicated that over areas with soil moisture deficit and temperature anomalies, the coupling strength intensified. This suggests that soil moisture deficits could lead to enhanced heat anomalies, and thus, result in enhanced soil moisture coupling with temperature. This demonstrates the importance of soil moisture and the need to thoroughly study it and its role within the land−atmosphere interaction and the climate on the whole.

By using a coupled land surface-atmosphere model with initial conditions of varying resolution and ensembles of systematically changed soil moisture, convective-scale simulations of a typical frontal rainstorm in the Yangtze River Basin are collected to investigate: (1) effects of different datasets on the simulated frontal mesoscale convective systems (MCSs); (2) possible linkages between soil moisture, planetary boundary layer (PBL), MCSs and precipitation in this modeled rainstorm. Firstly, initial soil moisture differences can affect the PBL, MCSs and precipitation of this frontal rainstorm. Specially, for a 90 mm precipitation forecast, the Threat score (TS) can increase 6.61% by using the Global Land Data Assimilation System (GLDAS) soil moisture. Secondly, sensitivity experiment results show that the near-surface thermodynamic conditions are more sensitive to dry soil than wet due to the initial moist surface; atmosphere conditions have suppressed the relations between soil and atmosphere; and decreased precipitation can be found over both wet and dry surfaces. Generally, a positive feedback between soil moisture and the near-surface thermodynamic conditions is identified, while the relations between soil moisture and precipitation are quite complicated. This relationship shows a daytime mixing of warm surface soil over dry surfaces and a daytime evaporation of adequate moisture over wet surfaces. The large-scale forcing can affect these relations and finally cause decreased precipitation over both wet and dry surfaces.