干旱区研究2026，Vol.43Issue(8)：1575-1588,14.DOI:10.13866/j.azr.2026.08.02
基于多源数据TVPDI干旱指数在新疆的适用性研究
Study on the applicability of TVPDI drought index based on multi-source data in Xinjiang
摘要
Abstract
This study addresses the unique challenges,namely,frequent droughts,complex terrain,and extreme spatiotemporal variability in precipitation,associated with the region of Xinjiang by employing the temperature-vegetation-precipitation drought index(TVPDI).Utilizing multi-source MODIS and GPM remote sensing data from 2001 to 2020,TVPDI integrates vegetation,temperature,and precipitation information into a three-dimen-sional spatial distance model.This model quantifies drought severity by calculating the Euclidean distance from each pixel to a designated wettest reference point within the vegetation condition index(VCI)-thermal condition index(TCI)-precipitation condition index(PCI)feature space.To rigorously assess TVPDI's suitability for Xinji-ang,we systematically compare it with six widely used remote sensing drought indices,including VCI,TVDI,and DSI.The self-calibrated Palmer drought severity index(scPDSI)from the China High-resolution Multi-drought Index Dataset(CHM_Drought)serves as an independent validation benchmark.Results show that TVP-DI exhibits the strongest correlation with scPDSI(Pearson R=-0.3793)and the lowest error metrics(root mean square error,mean absolute error,and Bias),demonstrating superior drought monitoring performance compared to traditional indices and confirming its applicability and potential in Xinjiang.Applying TVPDI to analyze the spatiotemporal characteristics of drought during Xinjiang's growing season(May-September)from 2001 to 2020 reveals a slight overall alleviation of drought conditions over the two decades,despite significant interannual vari-ability.Spatially,a consistent pattern emerges:"lower in the west than in the east,lower in the north than in the south,and lower in mountainous areas than in basins."Furthermore,a future trend analysis using the Hurst expo-nent showed that July had the highest proportion of pixels with persistent drought intensification in the period of May to September,indicating relatively prominent persistence and uncertainty in future drought evolution for this month.This study systematically evaluates TVPDI's applicability in Xinjiang,providing a scientific basis for pre-cise drought monitoring and risk management in the region.关键词
多源数据/TVPDI干旱指数/时空特征/适用性/新疆Key words
multi-source data/TVPDI drought index/spatiotemporal characteristics/applicability/Xinjiang引用本文复制引用
曹孟磊,张青,王勇,伊力达尔江·阿不力米提,韩雪云..基于多源数据TVPDI干旱指数在新疆的适用性研究[J].干旱区研究,2026,43(8):1575-1588,14.基金项目
新疆维吾尔自治区自然科学基金-青年基金(2022D01B199) （2022D01B199）
新疆气象局科技创新发展基金(MS202519) （MS202519）
新疆气象局科技创新发展专项重点项目(ZD202303) （ZD202303）
