【作者】 徐晓桃；

【导师】 颉耀文；

【作者基本信息】 兰州大学 ， 地图学与地理信息系统， 2008， 硕士

【摘要】 随着西部人类活动强度的日益加剧,环境和生态受到的压力将愈来愈大,缺水或水土流失造成大范围的贫困。植被水分利用效率是联系植被生态系统碳循环和水分循环的重要变量,因而具有重要的生态学和水文学意义。为了更加合理高效地利用我国西部地区有限的水资源,需要我们更全面和深入地了解该地区不同植被生态系统水分利用效率特征,从而发现并推广耗水量低而生产力高的干旱区耐旱植被,为水资源的可持续利用打好基础。本文选用CASA模型(Carnegie Ames Stanford Approach),结合MODIS遥感数据产品,气象数据,植被数据和土壤数据,在ENVI环境下实现对黄河源区2001-2005年净初级生产力NPP的计算;根据NPP及潜在蒸散发PET计算植被水分利用效率WUE;分析NPP,WUE的空间分布和时间变化特征;对不同植被类型下的NPP和WUE特征进行分析;分析NPP和WUE与影响因子的相关性。研究结论如下:1、以CASA模型为基础,结合MODIS遥感数据产品、地面气象资料和植被、土壤数据计算NPP,避免了统计模型以点代面的缺点,使NPP的估算更具科学性,且能实时地反映NPP时空变化;验证了CASA模型在黄河流域的可适用性。2、从NPP空间分布来看,NPP总量主要集中在源区中下部位,其次为达日站点周围地区;果洛、久治、达日三站点之间分布较为零散;达日与果洛站点以上部位NPP总量很低,最低值区域为湖泊和无植被区域。3、从NPP年际变化来看,在不同的区域有不同的响应。区域NPP年际变化中,2003年NPP年总量最大;年内变化中,基本上从3月开始快速增长,7月达最大值,9月以后快速减少。4、从不同植被类型NPP来看,年际上有着不同的变化趋势。有的植被NPP一直处于增长趋势;大部分植被NPP在2001-2003年增长,2003-2005年下降;有的植被NPP变化处于波动状态;个别植被NPP在2004年达最大值。5、从WUE年际变化来看,总体上呈上升趋势,2004年有所下降;年内变化中,WUE达到最大值基本在6月,7月下降较多,8月和9月有所回升,之后迅速下降。6、WUE年际变化趋势与NPP年际变化趋势类似。不同植被类型中,具有较高WUE值的有三种沼泽,其他植被类型中温带落叶阔叶林最高,其次为栽培植被,温带草原,盐生草甸,高寒草甸等,WUE值最低的为湖泊和无植被地段。7、分析源区NPP及WUE与NDVI、气温、降水、太阳辐射和海拔的相关性。发现NPP与NDVI的相关性最大,其次为气温,海拔,与降水和太阳辐射相关性较差;WUE是NPP与蒸散发的一个耦合结果,与NDVI的相关性最大,其次为太阳辐射,再次为降水量,气温,相关性最低为海拔。更多还原

【Abstract】 With more and more human activities in West of China, the pressure on environment and ecology is getting greater. Shortage of water or soil-water loss causes poverty. Water Use Efficiency (WUE) is the key factor linking the carbon nitrogen cycle and water cycle in the vegetation ecology system; therefore it has special ecological and hydrological meaning. To use the limited water resource in West of China reasonably and effectively, we should comprehensively and deeply understand the characteristics of WUE in different vegetation ecological systems, find out and popularize the vegetations which are of low-water-consumption but high -production - capability, improve the use efficiency of irrigation water in agricultural ecosystem and prepare for the sustainable use of water resource.In this thesis, based on CASA model (Carnegie Ames Stanford Approach) in ENVI, with MODIS products, meteorological data, vegetation and soil data, NPP (Net Primary Productivity) in yellow river region from 2001 to 2005 is obtained. The spatial distribution and temporal change characters of NPP are analyzed. According to NPP and potential evapotranspiration, WUE in yellow river region from 2001 to 2005 is obtained and also its spatial distribution and temporal change characters are analyzed. The NPP and WUE between different kinds of vegetation are analyzed; the relationship between influence factors and NPP and WUE is analyzed. The research conclusions are as follows:1. Based on CASA model, with MODIS high temporal resolution data products, meteorological data, vegetation and soil data, NPP in Yellow River source region is calculated. In this way, the shortcoming in the traditional statistic model can be avoided, the precise of NPP can be improved and the temporal and spatial variation of NPP can be reflected better; the applicability of CASA model in yellow river source region is validated..2. As to the spatial distribution of NPP, most of the quantity is in the mid lower part of the source region Quantities around Tongde station and Henan station are relatively concentrative and Dari station and its surrounding followed; In contrast, NPP distributed in Guoluo station, Jiuzhi station and Dari station is scattered, the part above Dari station and Guoluo station has rather low quantity, the lowest region is lakes and the bare mountains with a high elevation.3. As to the temporal change of NPP, different regions have different responses. For the whole region, NPP in 2003 reaches maximum; As to the month change, NPP grows rapidly in March, reaches the maximum in July, and reduce rapidly in September.4. As to different kinds of vegetations, they have different tendencies in NPP change between 2001 and 2005. Some vegetations are in the increased tendency all through, some grow between 2001 and 2003 and reduce between 2003 and 2005, others are always fluctuant and several reaches maximumin 2004. 5. As to the temporal change of WUE, it increases between 2001 and 2005 as a whole, but that in 2004 decreased a little. Referring to the mensal mean value in every year, WUE reaches the maximum in June, decreases much in July, and rallies in August and September, and then decreases rapidly.6. As to different kinds of vegetations’ WUE, it has the similar tendency with that of annual change of NPP. Three kinds of swamp are high in WUE. In other kinds, the one of temperate deciduous broad-leaf forest is the highest; others are in turn as follows, cropland vegetation, temperate grassland, salinization meadow, alpine meadow, and so on. Lakes and bare regions have the lowest WUE.7. After analyzing the relationship between NPP, WUE and their influence factors, such as, temperature, precipitation, solar radiation, NDVI, it is easy to find out that NPP is close related to NDVI, and then temperature and elevation, precipitation and solar radiation are poorly related. Vegetation growth in the source region, especially in dry season, mainly depends on unfreezed water from frozen soil, glacier and snow. WUE is a coupling value of NPP and evapotranspiration, which is close related to NDVI, solar radiation followed, which mainly affects evapotranspiration; precipitation and temperature are poorly related to WUE and the worst one is elevation.更多还原