基于MOD09GA产品的草地生物量遥感估算模型

利用忖南州2006-2008年的草地地上生物量外业实测数据和EOS Terra卫星中分辨率成像光谱仪MODIS每日地表反射率产品MOD09GA,提取了年实测点上的7个通道的光谱反射率值,计算了17种植被指数,综合研究了植被指数与生物量之间的相关性及草地地上生物量遥感反演模型,采用植被指数最大合成算法,计算了2005-2007年各旬、月及年最大植被指数及草地地上生物量密度时空变化动态特征.研究表明适宜甘南州草地地上生物量估产的植被指数EVI,其估产模型是乘幂模型(y=13583x<1.6652>,R=0.798),模型平均测产精度达76.72%.甘南州草地地上生物量密度除了2005年在7月上旬达到最大值外,其他年份均在7月下旬达到最大值.从月度变化可以看出,除了2005年温性草甸草原、2006年低平地草甸、2007年暖性草丛的月生物量密度分别在6,9,6月达到最大值外,甘南州不同类型草地基本上在7月达到最大值.2006年和2007年甘南草地地上生物量密度的年变化比较相近,但2005年的草地地上生物量密度与其他两年相比差异较大,且多高于三年的平均值.

Modeling of aboveground biomass of grassland using remotely sensed MOD09GA data

The aboveground biomass of grassland in Gannan Prefecture, Gansu Province, China from 2006 to 2008 was measured and 17 vegetation indices were calculated from 7 bands of MOD09GA. The correlation between the vegetation indices and grassland aboveground biomass were analyzed, and the estimation models were established. The maximum vegetation composite method was used to calculate the ten-day, monthly and yearly maximum vegetation index maps from 2005 to 2007, and the spatio-temporal dynamic characteristics of grassland aboveground biomass were analyzed. The results show that EVI has the best correlation with the grassland aboveground biomass in Gannan Prefecture, power regression (y= 13583x~(1.6652), R= 0.798) is suitable for monitoring grassland aboveground biomass in Gannan Prefecture, and the average estimation precision of the model is 76.72%. The grassland aboveground biomass in 2005 reached the maximum in early July, and it reached the maximum in late July for the other two years. For the monthly changes, it reached the maximum of aboveground biomass in July, except the temperate meadow steppe, the lowland meadow and warm-temperate shrubs and herbs reached the maximum values in June of 2005, in September of 2006 and in June of 2007, respectively. For the inter-annual changes, the grassland aboveground biomass in 2006 was similar to that of 2007, but the grassland aboveground biomass in 2005, compared with the other two years, had a large difference, and was higher than the average value of the 3 years.