地理学报 ›› 2004, Vol. 59 ›› Issue (3): 341-348.doi: 10.11821/xb200403003

• 水资源 • 上一篇    下一篇

无定河流域水量平衡变化的模拟

莫兴国1, 刘苏峡1, 林忠辉1, 陈丹1, 赵卫民2   

  1. 1. 中国科学院地理科学与资源研究所,北京 100101;
    2. 水利部黄河水利委员会水文局,郑州 450003
  • 收稿日期:2003-10-14 修回日期:2003-12-22 出版日期:2004-05-25 发布日期:2004-05-25
  • 作者简介:莫兴国 (1966-), 男, 副教授, 博士, 从事陆地生态系统动力学模拟。E-mail: moxg@igsnrr.ac.cn
  • 基金资助:

    国家自然科学基金项目 (90211007);中国科学院方向性前沿项目 (KZCX2-310)

Simulating the Water Balance of the Wuding River Basin in the Loess Plateau with a Distributed Eco-hydrological Model

MO Xingguo1, LIU Suxia1, LIN Zhonghui1, CHEN Dan1, ZHAO Weimin2   

  1. 1. Institute of Geographic Sciences and Natural Resources Research, CAS, Beijing 100101, China;
    2. Yellow River Water Conservancy Commission, Zhengzhou 450003, China
  • Received:2003-10-14 Revised:2003-12-22 Online:2004-05-25 Published:2004-05-25
  • Supported by:

    National Natural Science Foundation of China, No. 90211007; Knowledge Innovation Project of CAS, No.KZCX2-310

摘要:

利用黄土高原无定河流域1982~1991年的水文气象、土地利用、土壤质地、数字高程和NOAA-AVHRR遥感信息,建立基于土壤-植被-大气传输机理的分布式生态水文模型,模拟流域水量平衡的时空分布。研究结果发现,该流域的年平均植被指数 (NDVI) 的年际变化不明显,但NDVI最大值的年际变化显著。该流域年累计NDVI与降水年总量关系不明显,说明该流域植被的生长并不完全受控于降水总量。模拟的实际蒸散量用无定河及其岔巴沟子流域实测降水与实测径流的差值进行验证,误差小于5%。整个流域模拟时段的平均降水量为372±53 mm yr-1,实际蒸散量为334±33 mm yr-1,其中蒸腾为130±21 mm yr-1,有明显的年际波动。地表径流的年际变化相对较小。蒸散发的季节变化特征与降雨基本一致,即7、8、9月雨季高,其他月份低。降水量和实际蒸散量呈现显著的空间分异性,表现出由东南 (高NDVI) 向西北 (低NDVI) 递减的梯度差异。 地表径流的空间分异亦沿东南-西北梯度变化,但高值分散在中部。以岔巴沟子流域1991年的地表覆被度为基准,发现在全流域都覆盖上某一种植被的情况下,蒸腾和土壤蒸发的变化非常明显,地表径流和实际总蒸散的变化并不显著。只有在全流域都变成荒漠情景下,实际总蒸散才显示出较明显变化 (17%),表明在西北干旱半干旱区,土地利用/覆被变化对水量平衡的影响非常复杂。

关键词: 无定河;分布式水文模型;土地利用/覆被变化;水文效应;遥感植被指数

Abstract:

With an aid of the geographical information system of the hydrological, meteorological, soil and land use data, the Digital Elevation Model and NOAA-AVHRR remote sensing information, a distributed eco-hydrological model based on soil-vegetation-atmosphere transfer theory is developed to simulate the spatial and temporal variation of water balance components over the Wuding River Basin, the Loess Plateau, China from 1982 to 1991. It is found over the ten years, the maximum Normal Difference Vegetation Index (NDVI) changes year by year while the annually averaged NDVI is stable. There is not an obvious linear relationship between the annually averaged NDVI and annual precipitation, indicating that the variation of vegetation is not fully controlled by precipitation in this basin. The simulated evapotranspiration values over the Wuding River Basin and one of its sub-basins (Chabagou) agree well with the observed, the difference between the annual precipitation and runoff, with the relative errors being less than 5%. Over the simulated periods, the annual precipitation (372±53 mm yr-1) is consumed mostly by evapotranspiration (334±33 mm yr-1), of which about one third is transpiration (130±21 mm yr-1). Compared with the high yearly variation of precipitation and evapotranspiration, that of runoff is relatively stable. The seasonal variation pattern of evapotranspiration is similar to that of precipitation and net radiation with peaks in August, while that of runoff is unique with the high peak in March, indicating that freezing proCess should be further included in the model development. By increasing gradually from the northwestern part of the basin where the NDVI is low, the precipitation and evapotranspiration reaches the highest in the southeastern part of the basin where the NDVI is high. The spatial pattern of runoff shows the highest values in the middle part of the basin. Replacing the vegetation patterns in 1991 with each single vegetation type covering the whole of the Chabagou basin, canopy transpiration and soil evapotranspiration all change dramatically, whereas the changes of total evapotranspiration and runoff are small.

Key words: soil-vegetation-atmosphere transfer scheme, distributed hydrological model, land use/cover change, hydrological effect, remote sensing vegetation index, Wuding River