基于植被指数的大庆地区草地退化因子遥感定量反演模型的研制  

Establishment of quantitative remote sensing inversion model of grassland degradation factors based on vegetation indices in Daqing

作　　者：臧淑英[1] 那晓东[2] 冯仲科[1,3] 

机构地区：[1]哈尔滨师范大学地理系 [2]中国科学院东北地理与农业生态研究所 [3]北京林业大学测绘与3S技术中心

出　　处：《北京林业大学学报》2008年第S1期98-104,共7页Journal of Beijing Forestry University

基　　金：国家自然科学基金项目(40771195);中国科学院东北地理与农业生态研究所学科前沿领域项目(KZCX3-SW-NA3-39);黑龙江省自然科学基金项目(ZTF2005-32)

摘　　要：草地是维持大庆生态平衡的重要因素,草地退化严重影响了大庆地区的经济发展和生态环境。研制该地区基于植被指数的草地退化因子的定量反演模型,对于大庆地区草地退化的特征和规律的研究尤为重要。为此,基于大庆地区1988、1996、2001年3个时段的TM影像,应用归一化植被指数(NDVI)和土壤调节植被指数(SAVI),采用像元二分模型法和经验模型法建立统一的大庆地区草地植被盖度和可食牧草率的遥感定量反演模型。然后采用不变的地物类对TM历史影像的像元值做辐射配准;最后,将辐射配准后的历史影像带入模型,制作大庆地区各年度草地盖度和可食牧草率的分级图,从而为大庆地区草地退化的研究提供信息和依据。该研究求出了大庆地区草地植被盖度和可食牧草率的反演模型。经实地验证,模型的精度分别达到70.34%和68.27%。Grassland is an important factor keeping eeologic balance in Daqing City.Its degradation hinders the development of economy and improvement of ecological environment.The establishment of quantitative remote sensing inversion model of grassland degradation factors based on vegetation indices is important to study the characteristics and discipline of Daqing grassland vegetation degradation.Based on TM images in 1988,1996, 2001 and two vegetation indices,normalized difference vegetation index and soil adjusted vegetation index,the general remote sensing inversion model of grassland vegetation coverage and the proportion of edible pasture was established.Then,the historical TM image pixel value was radiometrically calibrated based on invariant ground targets.Finally,the model was applied to the calibrated TM image.The classification maps of grassland vegetation coverage and the proportion of edible pasture were drawn to provide information and foundation for the research of grassland degradation.The general remote sensing inversion models of grassland vegetation coverage and the proportion of edible pasture grass in Daqing were validated,and their accuracies were 70.34% and 68.27%.

关 键 词：遥感反演 植被指数 资源型城市 草地盖度 可食牧草率 

分 类 号：S812[农业科学—草业科学]