人为扰动对陆面水分能量的影响——以沩水河流域为例

以沩水河流域为例，基于陆面模式CLM4.5，建立了综合考虑作物种植、地下水开采及灌溉等人类活动的流域陆面水文模型。利用所发展模型，针对1981~2012年，取500 m空间分辨率，探讨人为扰动对陆面过程的影响。研究表明:1）地下水侧向流使得中下游地区地下水位有所提高，平原地下水埋深分布在4 m左右，山区埋深可达到几十米；模拟的叶面积指数较静态MODIS叶面积指数偏大1左右，由此使得种植区月蒸腾量提高约10 mm，土壤蒸发和地表产流有所减少；在灌溉作用下，作物叶面积指数略增长，蒸散发稍有提高，而在假设水稻采用漫灌的情况下，水库灌溉补偿了作物生长产生的水消耗，提高了该区域土壤湿度，增加潜热通量；研究区地下水开采存在但其水文效应并不显著。2）土地覆盖变化自1990年有较大变动，1990~2000年以林地为主，2000年后以耕地为主，其中，1990~2000年土地覆盖类型变化不明显，2001~2012年耕地面积呈先减少再增加又减少的趋势，林地面积则先增加再减少又增加，耕地与林地在2012年所占比例基本持平；同一土地类型内，植被类型变化较为明显，导致陆面水文模拟结果差异较大。

Impacts of Human Activities on Land Surface Water and Energy—A Case Study in Weishui River Watershed

A typical watershed in the subtropical monsoon region was selected for a case study. A land surface hydrological model for Weishui River Basin was established by coupling the schemes of crop growth, empirical reservoir irrigation and groundwater lateral flow into with a land surface model. First, the model with a constant land cover dataset was used to quantify the impacts of groundwater lateral flow, crop cultivation, groundwater exploitation and reservoir irrigation on land surface water and energy. The land cover change and its effects on the watershed were then investigated using several remote-sensing images and modeling studies. The results show that:1) The groundwater lateral flow makes the simulation more reasonable with lower depth in downstream areas (lower than four meters) and deeper table in mountainous areas (deeper than several decameters); the crop model yields a larger leaf area index than the fixed one, and thus increases the transpiration in cropland, which consumes more water in the watershed and lessens the latent heat flux; irrigation then offsets the water loss by utilizing the surface water and increases the latent heat flux. In addition, it is found that the effect of groundwater pumping is not significant. 2) The land cover changed a lot from 1990 to 2012 with small changes during the 1990-2000 period and large changes during the 2001-2012 period. From 2001 to 2012, the area of cropland first decreased, then increased, and then decreased again (the trend of woodland is opposite); the changes of vegetation types that belong to the same land use type are significant, which leads to large differences in land surface modeling.