论文部分内容阅读
利用遥感技术,基于时间序列的MODIS数据对2004—2006年东北三省的林区进行森林生长异常监测。首先利用MODIS数据时间分辨率高的特点,采用Savitzky-Golay滤波函数平滑8天合成的EVI,计算生长季面积和年EVI曲线熵值,两指标联合得到3年间变化量大的像素点,定义为森林生长异常点;然后抽取异常点的时间序列曲线进行分析,并结合森林灾害事件进行比较验证。结果表明:异常点曲线的熵值明显大于正常年,生长季峰值低,并且在生长旺季会出现峰值突然持续下降或双峰等异常现象,这与该区域森林生长异常发生时的植被反射率表征一致,说明用该法对森林生长异常进行监测是基本可行的。
Remote sensing technology and time series MODIS data were used to monitor the forest growth abnormalities in the forest regions of the three northeastern provinces of China from 2004 to 2006. First of all, by using the feature of high temporal resolution of MODIS data, the Savitzky-Golay filter function is used to smooth the 8-day composite EVI to calculate the growing season area and annual EVI curve entropy. The abnormal growth point of the forest was extracted. Then the time series curve of the abnormal point was extracted for analysis and compared with the forest disaster events. The results show that the entropy of the abnormal point curve is obviously larger than the normal one, the peak value of the growth season is low, and the peak value of the abnormal peak continues to decrease or double peak in the growing season, which is consistent with the characteristic of the vegetation reflectance when the forest growth abnormality occurs Consistent, indicating that the use of this method to monitor forest anomalies is basically feasible.