基于BEPS模型模拟地表蒸散的干旱监测技术

Drought monitoring based on simulated surface evapotranspiration by BEPS model

为构建有效的干旱监测与评估指标,利用中分辨率成像光谱仪(MODIS)数据以及NCEP/NCAR再分析资料并基于遥感过程耦合模型BEPS模拟了地表蒸散(evapotranspiration,ET)。同时,利用通量观测网AmeriFlux的站点观测数据验证该模型在ET模拟研究中的可靠性与适应性,在此基础上,构建BMI指数,分析了2007-2009年研究区域干湿状况的时空分布特征,并与其他干旱指数进行对比分析。结果表明:12个站点通量实测值与模型模拟的ET的相关系数为0.8568(p<0.01),表明BEPS模型能够较好的模拟美国陆地生态系统的ET。BMI值的空间分布特征较为明显,与降水量与ET经纬向变化较为一致。而BMI值的时间分布特征表明,其具有较为显著的季节变化。这与地形、气候特征以及植被的分布关系密切。从检验效果来看,受降水量的影响,BMI在月尺度以上干旱监测效果相对较好。BMI与USDM的相关系数表明,本文在BEPS模型模拟ET的基础上构建的指标在干旱监测中是可行的。

To construct effective drought monitoring and evaluation index, the moderate resolution imaging spectrometer （ MODIS ） data and the NCEP/NCAR reanalysis data were utilized, and on the basis of remote sensing process model （BEPS） , the surface evapotranspiration was simulated in this paper. At the same time, using the data of flux observation network AmeriFlux, the reliability and adaptability of this model was validated. On this basis, relative moisture index （BMI） was constructed to analyze the temporal and spatial distribution characteristics of regional dry and wet conditions in 2007 - 2009, and was compared with other drought index. The findings suggest that, BEPS model has a good effectiveness for terrestrial ecosystem evapotranspiration in the United States of America. The correlation coefficient of 12 sites reaches 0. 8568 （p 〈 0. 01 ）. Spatial distribution characteristics of BMI value is obvious, and is consistent with latitudinal change and longitude of the precipitation and evapotranspiration. The temporal distribution characteristic of BMI values shows that, its seasonal variation issignificant too. It is closely linked with the terrain, climatic characteristic and vegetation distribution. Affected by precipitation, good indication effectiveness of BMI on drought monitoring is in the month scale or seasonal scale. The correlation coefficients of BMI and USDM show that, the constructed index based on the BEPS model simulating ET is feasible in drought monitoring