摘要
选择中国南方地区为研究区域,以土壤含水量为实际农业干旱情况的参考指标,以日光诱导叶绿素荧光(SIF)为主要变量,结合地表温度、降水因子等参数并引入地理加权回归(GWR)技术,建立基于多源数据的GWR农业干旱监测模型,并对模型的监测效果进行验证。结果表明,该模型能很好地估测土壤水分,且GWR模型比普通最小二乘法回归模型能更准确地体现旱情模式的空间非平稳特征。同时,GWR模型回归系数的多年均值可用于旱情预测,预测结果较为准确。对比使用SIF和归一化植被指数(NDVI)建立干旱监测模型,结果显示,SIF比NDVI更适合于拟合土壤水分含量,SIF有替代NDVI应用于大尺度范围农业旱情监测的潜力。
With the soil water content as the reference index of actual agricultural drought,and the solar-induced chlorophyll fluorescence(SIF)as the main variable,combined with the parameters of land surface temperature and precipitation factor,an agricultural drought monitoring model based on multi-source data was established by using geographically weighted regression(GWR)technique,and the monitoring effect of the model was verified.The results show that the model can estimate soil moisture well,and the GWR model can reflect the spatial non-stationary characteristics of drought pattern more accurately than the ordinary least squares model.At the same time,the multi-year mean value of regression coefficient of GWR model can be used for drought forecast,and the results are accurate.Comparison between two drought monitoring models using SIF and normalized differential vegetation index(NDVI)respectively indicates that SIF is more suitable to fit soil moisture than NDVI,and SIF has the potential to replace NDVI in large-scale agricultural drought monitoring.
作者
董恒
韩超然
何思聪
韩江涛
DONG Heng;HAN Chao-ran;HE Si-cong;HAN Jiang-tao(School of Resources and Environmental Engineering,Wuhan University of Technology,Wuhan 430070,China)
出处
《武汉理工大学学报》
CAS
北大核心
2020年第11期39-44,共6页
Journal of Wuhan University of Technology
基金
国家自然科学基金(52079101,41701483)
国家大学生创新创业训练计划(202010497073)
关键词
干旱监测
日光诱导叶绿素荧光
地理加权回归
土壤水分
归一化植被指数
多源数据
drought monitoring
solar-induced chlorophyll fluorescence(SIF)
geographically weighted regression(GWR)
soil moisture
normalized differential vegetation index(NDVI)
multi-source data
