摘要
以黑龙江省区域小兴安岭地区遥感影像和130块2005年小兴安岭二类调查数据及土壤数据为基础,选择各个波段的灰度值、不同波段灰度值之间的线性和非线性波段组合、纹理信息以及非生物因子栅格化后所形成的辅助波段等为自变量,选择与碳密度相关性显著的自变量,采用郁闭度碳密度联立方程组模型,对黑龙江省区域小兴安岭森林生态系统的碳密度进行估算、精度评价。结果表明：黑龙江省区域小兴安岭的南部和中部地区,森林碳密度主要集中在200~250 t·hm-2;北部地区大部在250~300 t·hm-2。黑龙江省区域小兴安岭森林生态系统碳密度,具有从西向东、从南向北,逐步升高的趋势。模型平均拟合精度85.1%,均方根误差=31.27 t·hm-2;平均检验精度84.7%,均方根误差=33.61 t·hm-2。
With the remote sensing images of Xiaoxing＇an Mountains and 130 plots of the forest inventory in 2005 and soil data,we selected the variables including the bands of gray value,linear and nonlinear combination of bands between the different bands of gray value,texture information and auxiliary band formed by abiotic factors of the grid and the independent variables significantly related to the carbon density to estimate the forest ecosystem carbon density of Xiaoxing＇an Mountains by using the simultaneous equation models. In the south region of Xiaoxing＇an Mountains,forest carbon density was 200-250 t·hm-2,and 250-300 t·hm-2in the north region with the rising trend from west to east and from south to north.Average model fitting precision was 85. 1% with RMSE of 31. 27 t·hm-2. The average test accuracy of the model was 84.7% with RMSE of 35. 61 t·hm-2.
出处
《东北林业大学学报》
CAS
CSCD
北大核心
2015年第4期127-130,共4页
Journal of Northeast Forestry University
基金
"十二五"国家科技支撑项目(2011BAD37B01)
关键词
森林生态系统
碳密度
遥感估算
小兴安岭
黑龙江省
Forest ecosystem
Carbon density
Remote sensing estimation
Xiaoxing＇an Mountains
Heilongjiang Province
