A drought results from the combined action of several factors. The continuous progress of remote sensing technology and the rapid development of artificial intelligence technology have enabled the use of multisource remote sensing data and data-driven machine learning (ML) methods to mine drought features from different perspectives. This method improves the generalization ability and accuracy of drought monitoring and prediction models. The present study focused on drought monitoring in southwest China, where drought disasters occur frequently and with a high intensity, especially in areas with limited meteorological station coverage. Several drought indices were calculated based on multisource satellite remote sensing data and weather station observation data. Remote sensing data from multiple sources were combined to build a reconstructed land surface temperature (LST) and drought monitoring method using the two different ML methods of random forest (RF) and eXtreme Gradient Boosting (XGBoost 1.5.1), respectively. A 5-fold cross-validation (CV) method was used for the model’s hyperparameter optimization and accuracy evaluation. The performance of the model was also assessed and validated using several accuracy assessment indicators. The model monitored the results of the spatial and temporal distributions of the drought, drought grades, and influence scope of the drought. These results from the model were compared against historical drought situations and those based on the standardized precipitation evapotranspiration index (SPEI) and the meteorological drought composite index (MCI) values estimated using weather station observation data in southwest China. The results show that the average score of the 5-fold CV for the RF and XGBoost was 0.955 and 0.931, respectively. The root-mean-square error (RMSE) of the LST values reconstructed using the RF model on the training and test sets was 1.172 and 2.236, the mean absolute error (MAE) was 0.847 and 1.719, and the explained variance score (EVS) was 0.901 and 0.858, respectively. Furthermore, the correlation coefficients (CCs) were all greater than 0.9. The RMSE of the monitoring values using the XGBoost model on the training and test sets was 0.135 and 0.435, the MAE was 0.095 and 0.328, the EVS was 0.976 and 0.782, and the CC was 0.982 and 0.868, respectively. The consistency rate between the drought grades identified using SPEI1 (the SPEI values of the 1-month scale) based on the observed data from the 144 meteorological stations and the monitoring values from the XGBoost model was more than 85%. The overall consistency rate between the drought grades identified using the monitoring and MCI values was 67.88%. The aforementioned two different ML methods achieved a high comprehensive performance, accuracy, and applicability. The constructed model can improve the level of dynamic drought monitoring and prediction for regions with complex terrain and topography and formative factors of climate as well as where weather stations are sparsely distributed.

The source area of the Yellow River (SAYR) is one of the world´s largest wetlands containing the greatest diversity of high altitude marshlands. For this reason, its response to climate change is extremely significant. As revealed by different studies, the response of hydrological processes to global warming results in high uncertainties and complexities in the water cycle of the SAYR. Thus, understanding and projecting future runoff changes in this region has become increasingly important. In the present investigation, we used runoff and meteorological data of the SAYR from 1976 to 2014 (historical period). In addition, Digital Elevation Model (DEM), land-use, and soil data for the period 1976 to 2100 were used considering three future SSPs (Shared Socioeconomic Paths) scenarios of 8 models selected from the Coupled Model Intercomparison Project Phase 6 (CMIP6). The Soil and Water Assessment Tool (SWAT) was used to simulate, project, and analyze potential variations and future runoff of the main hydrological stations (Jimai, Maqu, and Tangnaihai) located in the SAYR. The results showed that: 1) The SWAT model displayed good applicability in historical runoff simulation in the SAYR. A small runoff simulation uncertainty was observed as the simulated value was close to the measured value. 2) Under three different 2021–2100 SSPs scenarios, the yearly discharge of the three hydrological stations located in the SAYR showed an increasing trend with respect to the historical period. Future runoff is mainly affected by precipitation. 3) We compared the 1976–2014 average annual runoff with projected values for the periods 2021–2060 and 2061–2100. With respect to 2021–2060, the lowest and highest increases occurred at Tangnaihai and Maqu Stations in the emission scenarios without (SSP585) and with mitigation (SSP126), respectively. However, the highest and lowest increments at Jimai Station were observed in the intermediate emission (SSP245) and SSP126 scenarios, respectively. Moreover, in 2061–2100, the Maqu and Tangnaihai Stations showed the lowest and highest increments in the SSP585 and SSP245 scenarios, correspondingly. In Jimai Station, the lowest increment occurred in SSP126. The yearly average discharge in the near future will be smaller than that in the far future. Overall, this study provides scientific understanding of future hydrological responses to climate changes in the alpine area. This information can also be of help in the selection of actions for macro-control, planning, and management of water resources, and the protection of wetlands in the SAYR.

Under the background of global warming， it is of great application value to study extreme precipitation events in four southwestern provinces and cities in China， which are vulnerable to the ecological environment and sensitive to climate change.Based on the daily temperature and precipitation data of 93 weather stations from 1969 to 2020， by selecting 11 extreme precipitation indices， using linear regression analysis， M-K mutation test， EOF analysis， and other methods， this paper not only analyzed the spatiotemporal variation characteristics of extreme precipitation events in Southwestern China in 52 years， but discussed the relationship between extreme precipitation indices and intense ENSO events by calculating the composite ratio.The results showed that： （1） The overall frequency and intensity of precipitation in Southwestern China increased from 1969 to 2020， and extreme precipitation trend increased.The four regions divided by topography also showed precipitation characteristics of different degrees； （2） In terms of the distribution of the average spatial variation rate at each station in 52 years， the overall number of consecutive dry days at all stations in Southwestern China increased， and the number of consecutive wet days decreased.The precipitation decreased from northwest to southeast， but extreme precipitation events increased.The increase of extreme precipitation events in Hengduan Mountain was the most obvious； （3） The abrupt change age of the extreme precipitation indices was mainly distributed in the mid-1980s and early 21st century.The EOF analysis showed that the growth trend of R99p was decreasing from west to east， and the growth trend of RX5 showed a distribution pattern of decreasing in the middle and increasing on both sides； （4） The extreme precipitation indices in Southwestern China were closely related to intense ENSO events， mainly characterized by more total precipitation， longer precipitation duration and more extreme precipitation events in robust El Niño years， and less total precipitation but higher precipitation intensity and fewer extreme precipitation events in robust La Niña years.The extreme precipitation events would be affected by intense ENSO events.The study results can provide an important scientific reference for relevant departments to more comprehensively understand and deal with extreme precipitation events in Southwestern China and formulate flood control and disaster reduction measures under the background of global warming.

Due to the complexity of drought and the diversity of influencing factors， the accurate monitoring of drought still faces many problems， especially the increasing frequency and aggravation of drought in Southwestern China， and the formation and disaster causing process have certain particularity.However， the traditional drought monitoring methods cannot meet the requirements of regional drought monitoring， so more scientific monitoring methods and means are needed.Since machine learning can comprehensively consider a variety of disaster causing factors to establish a comprehensive drought monitoring model， it undoubtedly provides a new technical means for drought monitoring.Therefore， this paper used multi-source remote sensing data from 2010 -2019 and meteorological station data from 1980-2019 to first construct a random forest monitoring model to reconstruct and supplement the surface temperature in Southwestern China， and then constructed XGBoost monitoring model to monitor， evaluate and validate the drought in Southwestern China.The results showed that： （1） The correlation coefficients between the training set and test set of the random forest model and the actual surface temperature of the stations exceeded 0.9， which reached a significant correlation.The spatial distribution of LST reconstruction values was similar to that of remote sensing monitoring values， and the values were close to the observed values of meteorological stations.（2） The correlation coefficients between the monitoring values of XGBoost model training set and test set and the SPEI calculated values at the stations were more than 0.86， with significant correlation.The overall consistency rate of drought grade between the monitored value and the calculated SPEI values exceeded 85%.（3） The overall consistent rate between the monitored values of the XGBoost model and the MCI values was above 67.88%， which was more consistent.The consistent rate of all months exceeded 58%， with the highest consistent rate of 75.07% in September and the lowest consistent rate of 58.26% in February.（4） The drought in each season monitored by the model was basically consistent with the actual drought， which could better reflect the spatial distribution and drought in Southwestern China.

黄河源区是黄河流域的重要组成部分，其径流变化影响着整个流域的水资源和生态系统安全。本文利用1976—2014年黄河源区径流、气象、数字高程模型DEM（Digital Elevation Model）、土地利用、土壤以及第六次国际耦合模式比较计划CMIP6（6thCoupled Model Inter-comparison Project）中8个模式的3个未来情景（SSP126、SSP245和SSP585）气象数据，基于SWAT（Soil and Water Assessment Tool）水文模型，对黄河源区主要水文站的径流进行了模拟、未来预估和变化分析。研究表明：（1）SWAT模型对黄河源区历史径流模拟的适用性较好，径流模拟的不确定性较小，模拟值较接近于实测值。（2）参数敏感性分析表明27个与水文有关的参数都对径流模拟有一定的影响。其中，土壤蒸发补偿因子、湿润条件II下SCS（Soil Conservation Sevice）径流曲线数、浅层地下水径流系数的敏感性较强，径流受陆面蒸散发、下垫面和降水影响较大。（3）降水是影响未来径流的主要因素。在SSP126和SSP245两种未来情景下，吉迈、玛曲和唐乃亥3个水文站在2021—2100年的两个时期（2021—2060年和2061—2100年）年均流量均呈增加趋势；而在SSP585情景下，2021—2060年呈增加趋势，2061—2100年则呈减少趋势。相对于1976—2014年，未来近期（2021—2060年）唐乃亥和玛曲站年均流量在SSP585情景下增加幅度最低，SSP126情景下增加幅度最高；吉迈站在SSP245情景下增加幅度最高，SSP126情景下增加幅度最低；未来远期（2061—2100年）3个水文站除了吉迈站是在SSP126情景下增加幅度最低外，其余均是在SSP585情景下增加幅度最低，SSP245情景下增加幅度最高。研究结果可为黄河流域水资源管理、防洪蓄水和生态环境保护等提供科学依据与理论支撑。

针对近年来西南地区旱涝灾害频发的问题,利用西南地区1968-2017年129个气象站点逐日气象资料,采用Mann-Kendall检验、EOF分析、相关和小波周期分析等方法,不仅对近50年西南地区干湿演变特征进行时空分析,而且讨论了干湿对ENSO事件的响应.结果 表明,近50年西南地区主要呈现干旱化趋势,且具有明显的周期变化,在20世纪末到21世纪初有明显转折点;极润、重润和轻旱各季节发生频率随时间尺度增加而减小,秋冬季轻润和春冬季重旱发生频率随时间尺度增加而增加,极旱发生频率在各季节随时间尺度增加而增加;西藏西部、云南南部及贵州北部呈现干旱化,西藏东部、四川北部及云南中部呈现湿润化;各季节干湿有明显的空间差异;ENSO事件对西南地区干湿影响具有滞后性,正影响随滞后时间由东向西移动,在滞后3个月时最显著.