• 方法与技术 •

### 基于实测光谱的大兴安岭地区典型森林枯落物含水率估测模型

1. (1东北林业大学林学院， 哈尔滨 150040； 2东北盐碱植被恢复与重建教育部重点实验室， 东北林业大学盐碱地生物资源环境研究中心， 哈尔滨 150040)
• 出版日期:2017-11-10 发布日期:2017-11-10

### Estimating model of typical forest litter moisture content based on the field spectrum in Daxing’anling of China.

XIE Zi-xi1, HU Hai-qing1, YANG Xi-guang2, SUN Long1*#br#

1. (1School of Forestry, Northeast Forestry University, Harbin 150040, China; 2Key Laboratory of SalineAlkali Vegetation Ecology Restoration, Ministry of Education/Alkali Soil Natural Environmental Science Center, Northeast Forestry University, Harbin 150040, China).
• Online:2017-11-10 Published:2017-11-10

Abstract: Daxing’anling is a highly frequent forest fire disaster area in China. Accurate prediction of forest fuel moisture content is of great significance to improve the accuracy of forest fire forecast in this area. In this study, first derivative transformation and continuum removal methods were used to identify the bands sensitive to forest litter moisture. And the correlation coefficients were calculated between measured forest litter moisture and four spectral variables, including the original reflectance, the first derivative reflectance, the continuumremoval reflectance and first derivate of the continuumremoval reflectance. Then the highly relevant bands were selected as independent candidate variables and the forest litter moisture content estimation model was established by using a stepwise regression method. The determination coefficient (R2), mean relative error (MRE) and root mean square error (RMSE) of the model were calculated for evaluating the model. The results showed that there was a higher correlation between forest litter moisture content and first derivate of the continuumremoval reflectance compared to other spectral variables. The more sensitive bands were 398-668, 768-1068, 1098, 1278, 1388-1438, 1458-1538, 1868-1898, 1988-2088, 2198-2208, 2228-2238 nm (P<0.05). The extreme values of the correction coefficients were -0.653 and 0.610 at 2008  and 1888 nm, respectively. The model used to estimate forest litter moisture content was established by using multilinear stepwise regression method. The R2, MRE and RMSE of the model were 0.537, 0.303 and 0.499, respectively. This study provides reference in fast estimating forest litter moisture content by using remote sensing technology.