Aboveground Biomass Of Coniferous Forest Estimation Of Youyang County Using Spectra And Topographic Factors

Forest biomass is an important indicator for assessing the ecosystem productivity,and also the basis for analysis of substance circulation in forest ecosystem.The traditional methods for calculating biomass rely on a considerable amount of measurements that involves extreme time and labour,which is also difficult to update spatial distribution in a large area.However,with the rapid development of remote sensing technique,multi-source remote sensing data has been extensively used for monitoring the forest types,spatial distribution and structural features,which provides a fast,cheap,and convenient way to estimate the forest biomass in large scale,especially in mountainous area.Terrain correction is typically used to adjust surface reflectivity for the influence of slope and aspect when plant biomass is estimated by remote sensing.However,there are very few studies on the different between the pixel area of a remote sensing image and its real surface area.but this difference inevitably influences the difference between the pixel area of a remote sensing image and its real surface area.Taking Youyang County as the study area,forest biomass of sample plots were obtained as the true value through field investigation.At the same time,the image of satellite GF1(GF-1)were processed,and corresponding image and topographic information of biomass were obtained.This paper addresses the effects of the difference between the pixel area of a remote sensing image and the real surface area pf plant biomass obtained with remote sensing,when biomass-terrain correction model was calibrated according to surface-area calculation model and the law of the conservation of mass.Forest biomass has very important practical significance to realize the sustainable development in forest ecosystem and the sustainable development in forest resources,meet the needs of the economic and social development provided by forest and construct the regional decision-making system of sustainable forest management.The main findings are as follows:1.Obtained the real surface area of image through the analysis in Arcgis.The results show that the pixel area was,influenced by undulating topography,was different from the actual surface area for Youyang County.Statistical values for the areas in coniferous forest area was at altmost 965m2 higher afer terrain thanValue in the northeastem and wesem was much greater than that in eastem,the difference in biomass distribution was correspond with the terrain undulation in study area.2.Corresponding band spectral value,combination value of spectral,texture value of image and topographic information of the sample biomass were obtained.The result showed that the textures extracted by Red band of image were at a lower correlation level with forest biomass,and the mean texture extracted by NIR band,NIR band of image,DVI and MSAVI were at most significant correlation with coniferous forest biomass.3.Two classification methods,that are supervised classification method and artificial neural network method,were tested on GF-1 WFV data.Compared with the traditional supervised classification method based on statistical theory,artificial neural network method under the condition of incomplete data information,can still obtain the ideal effect in the respect of pattern recognition.Especially joined the texture information of artificial neural network classification accuracy reached 86.62%,which was higher than traditional supervised classification accuracy.Therefore,in the analysis and processing of WFV image data,using the method of artificial neural network for feature space distribution information is more reliable and more convenient.4.Based on the uncertainty analysis for forest bioass estimation with remote sensing,a model established with the multi-variate linear stepwise regression(MLSR)was used to estimate aboveground biomass of coniferous forest.Rsult showed that biomass was significantly correlated with WFV image and topographic factors,and their correlation coefficients were 0.853.Therefore,these models can be applied to estimating the biomass of coniferous forest.The biomass of coniferous forest increased by 8.54% with terrain correction.We conclude that the impace of terrain cannot be ignored as it was an important factor affecting the precision of biomass conversions.