Remote Sensing Estimation Of Siyang Poplar Biomass And Dynamic Analysis

Forest biomass is a basic quantity character of the forest ecological system. Biomass data are foundation of researching many forestry and ecology problems. Based on the data of forest inventory of Jiangsu Province in 2005 and 2009,the poplar forest biomass are estimated at the Siyang county, Suqian city, Jiangsu Province by the method of biomass expansion factor (BEF) which is derived from the relationship between biomass and volume. Inventory data is based on biomass conversion factor to estimate the continuous function poplar biomass, and correspond to the 2005 Landsat TM,2009 The HJ-1B in remote sensing image extraction factor to stepwise multiple regression fitting analysis of the biomass estimates and the establishment of remote sensing estimation models; The paper analyzed the Forest Resource Inventory data in 2009 about the study region, for example, of different age groups, canopy, soil texture on the growth of poplar biomass analysis, and established based on age groups (young forest, middle-aged forest, near mature forest, mature forest, over-mature forest) remote sensing estimation models,providing Siyang county even the larger scale regional management poplar with the basic data and theoretical basis in the future. Results are as follows:(1)Before the classification of the two images, the paper defined the classification category according to Forest Resource Inventory procedures, selected strictly the training area and evaluated their divisibility, then made the classification with the maximum likelihood classification method. The TM image pixel size is 30*30m and the total number of poplar pixels is 330,929, so the poplar forest area is about 29,768.6 ha; The HJ image pixel size is 30*30m and the total number of poplar pixels is 468,925, so the poplar forest area is about 38,848.8 ha, we know the area increased by 9080.2 hectares from 2005 to 2009, an increase of 30.5%.(2) The paper estimated the poplar biomass in 2005 with forest statistical data in Jiangsu; Then extracted the vegetation index,band ratio factors on the TM remote sensing data and analyzed the correlation with the poplar biomass at 0.01 level Significant correlation, which provided support for using remote sensing model to estimate the biomass of poplar. Poplar forest biomass per unit area and NDVI, RVI, B1/B3, B2*B3/B4 scatter distribution have a clear linear relationship, selecting randomly 50 from 66 samples of data to establish biomass remote sensing estimation model with multiple regression, and the correlation coefficient R2 is 0.7011. Make precision test with the remaining 16 samples, the average error is-1.996.(3)The paper estimated the poplar biomass in 2009 with forest survey data in Jiangsu; Then extracted the vegetation index,band ratio factors on the HJ remote sensing data and analyzed the correlation with the poplar biomass at 0.01 level Significant correlation, which provided support for using remote sensing model to estimate the biomass of poplar. Poplar forest biomass per unit area and TNDVI, B1/B4, B2/B4, B3/B4, B3/(B1+B2+B3+B4), B4/(B3+B2), shuguan scatter distribution have a clear linear relationship, selecting randomly 60 from 93 samples of data to establish biomass remote sensing estimation model with multiple regression, and the correlation coefficient R2 is 0.444. Make precision test with the remaining 33 sample data, the average error is-1.608. Established the remote sensing estimation models by age group-age sequence with selecting randomly 40 samples of data from each age group (young forest, middle-aged forest, near mature forest, mature forest) and the model correlation coefficients are 0.3634,0.3696,0.4164,0.3597.(4)The paper extracted the poplar category based on the comparison of classification, then added up the statistics of two poplar biomass data based on remote sensing estimation models. The results are as follows:Siyang county poplar biomass is 1,786,771.2 tons in 2005, and 2,760,600.4 tons in 2009, in the four years, an increase of 973,829.2 tons with an increase of 35.3%, so the growth trend is same.The result shows both national and local government policies improved the promotion of eco-forestry and paid attention to the enormous benefits from the forest.