摘要
利用青海省22个生态环境监测站在2003和2004年牧草生长季内各月所测的牧草鲜草产量数据,并搜集了该时段内逐日的MODIS遥感数据,通过对图像的预处理,云区识别,植被指数的计算及月最大植被指数的合成,形成了与牧草产量数据相对应的MODISEVI数据。然后按牧草生长季和草地类型建立了牧草鲜草产量与MODIS植被指数的关系模型。结果表明,牧草产量和MODISEVI之间存在较高的相关性,用指数函数建立产量模型效果较好。按牧草生长季建立的牧草产量检测模型比按草地类型建立的模型相关性要高。
Qinghai Province is one of five largest pasturing areas and ranks fourth in China. It is of great importance to learn the growth status of grass in time to coo-environment protection and sustainable development of stock raising. Based on the data of fresh grass weight per square meter from 22 coo-environmental monitoring stations （which were established by Qinghai Meteorological Bureau in 2002） during grass growing season from 2003 to 2004, as well as the same period of MODIS data, by preprocessing, cloud detecting, EVI computing and monthly maximum EVI compositing, MODIS EVI data were obtained corresponding to grass output data. The relationship between grassland output and EVI was established according to the grass growing season and grassland type. The results show that grassland output have a good correlation with MODIS EVI, and it can be better described by the exponential function. Furthermore, the grassland output model based on the grass growing season is better than that of based on grassland type.
出处
《气象》
CSCD
北大核心
2007年第6期102-106,I0002,共6页
Meteorological Monthly
基金
中国气象局推广项目(CMATG2005M42)
科技部科技公益研究专项(2002DIBI0052)共同资助
