Search

Accurate drought forecasting is necessary for effective agricultural and water resource management and for early risk warning. Various machine learning models have been developed for drought forecasting. This work developed and tested a fusion-based ensemble model, namely, the stacking (ST) model, that integrates extreme gradient boosting (XGBoost), random forecast (RF), and light gradient boosting machine (LightGBM) for drought forecasting. Additionally, the ST model employs the SHapley Additive exPlanations (SHAP) algorithm to interpret the relationship between variables and forecasting results. Multi-source data that encompass meteorological, vegetation, anthropogenic, landcover, climate teleconnection patterns, and topological characteristics were incorporated in the proposed ST model. The ST model forecasts the one-month lead standardized precipitation evapotranspiration index (SPEI) at a 12 month scale. The proposed ST model was applied and tested in the German federal states of Brandenburg and Berlin. The results show that the ST model outperformed the reference persistence model, XGBboost, RF, and LightGBM, achieving an average coefficient of determination (R2) value of 0.845 in each month in 2018. The spatiotemporal Moran’s I method indicates that the ST model captures non-stationarity in modeling the statistical association between predictors and the meteorological drought index and outperforms the other three models (i.e., XGBoost, RF, and LightGBM). Global sensitivity analysis indicates that the ST model is influenced by a combination of environmental variables, with the most sensitive being the preceding drought indices. The accuracy and versatility of the ST model indicate that this is a promising approach for forecasting drought and other environmental phenomena.

Abstract: Thorough evaluations of satellite precipitation products are necessary for accurately detecting meteorological drought. A comprehensive assessment of 15 state-of-the-art precipitation products (i.e., IMERG_cal, IMERG_uncal, GSMaP-G, CPC-Global, TRMM3B42, CMORPH-CRT, PERSIANN-CDR, PERSIANN, PERSIANN-CCS, SM2RAIN, CHIRPS, ERA5, ERA-Interim, MERRA-2, and GLDAS) is herein conducted for the period 2010–19 giving special attention to their performance in detecting meteorological drought over mainland China at 0.25° spatial resolution. The cited precipitation products are compared against China’s gridded gauge-based Daily Precipitation Analysis (CGDPA) product, derived from 2400 meteorological stations, and their quality is assessed at daily, seasonal, and annual precipitation time scales. Meteorological droughts in the datasets are determined by calculating the standardized precipitation evapotranspiration index (SPEI). The performance of the precipitation products for drought detection with respect to the SPEI is assessed at three time scales (1, 3, and 12 months). The results show that the GSMaP-G outperforms other satellite-based datasets in drought detection and precipitation estimation. The MERRA-2 and the ERA5 are on average closer to the CGDPA reference data than other reanalysis products for precipitation estimation and drought detection. These products capture well the spatial and temporal pattern of the SPEI in southern and eastern China having a probability of detection (POD) above 0.6 and a correlation coefficient (CC) above 0.65. CPC-Global, IMERG, and the ERA5 reanalysis product are ideal candidates for application in western China, especially in the Qinghai–Tibetan Plateau and the Xinjiang Province. Generally, the accuracy of precipitation products for drought detection is improved with longer time scales of the SPEI (i.e., SPEI-12). This study contributes to drought-hazard detection and hydrometeorological applications of satellite precipitation products. Significance Statement: The purpose of this study is to comprehensively evaluate the quality of 15 global satellite-based, gauge-based, and reanalysis precipitation products for meteorological drought detection at 1-, 3-, and 12-month time scales. This work systematically evaluates these products’ capacity to capture precipitation occurrence and intensity in different seasons. This is followed by a comparison of the precipitation products’ performance in drought detection. This work’s findings and evaluation results will improve the ability of those who develop precipitation products in identifying error sources and further improving retrieval algorithms. This paper’s results will serve as a valuable reference for end users seeking to better understand the application of precipitation products to drought detection.

As an important economic crop, tobacco faces the common challenges of continuous cropping due to the limited amount of arable land. Biochar has certain preventive and control effects on continuous cropping obstacles and can improve crop yield and quality. This study aims to explore the regulation of the antioxidant system and hormone signaling in tobacco under continuous cropping conditions by biochar. We studied the impact of biochar (Nicotiana tabacum LJ911 and LY101) on plant physiological functions and transcriptomes of tobacco under continuous cropping conditions. The results showed that biochar regulates photosynthetic gas exchange parameters in both tobacco varieties. Biochar significantly increased the activity of antioxidant enzymes in the leaves of both tobacco varieties under continuous-cropping conditions. Additionally, biochar up-regulated endogenous abscisic acid, auxin, and jasmonic acid in the leaves of both tobacco varieties, with higher levels observed in LY101 than in LJ911. Through Weighted Gene Co-expression Network Analysis (WGCNA), it was found that the tobacco antioxidant system and hormones showed a significant response. Biochar improved photosynthetic efficiency, enhanced antioxidant capacity, and promoted tobacco growth through hormone synthesis and signaling. In summary, biochar application can serve as a promising strategy to mitigate the challenges associated with continuous cropping in tobacco cultivation.