No abstract

No abstract

Moso bamboo extensively distributes in southeast and south Asia, and plays an important role in global carbon budget. However, its spatial distribution and heterogeneity are poorly understood. This research uses geostatistics theory to examine the spatial heterogeneity of aboveground biomass (AGB) of moso bamboo, and uses a point kriging interpolation method to estimate and map its spatial distribution. Results showed that (1) spatial heterogeneity and spatial pattern of moso bamboo's AGB can be revealed by an exponential semivariance model. The analysis of the model structure indicating that the AGB spatial heterogeneity is mainly composed of spatial autocorrelation components, and spatial autocorrelation range is from 360 to 41,220 m; (2) kriging standard deviation map showing the level of the model errors indicates that the AGB spatial distribution by point kriging interpolation method is reliable; (3) the average AGB of moso bamboo in Anji County is 44.228 Mg hm -2 , and carbon density is 20.297 Mg C hm -2 . The total AGB of moso bamboo accounts for 16.97% of the total forest-stand biomass in Zhejiang province. The total carbon storage of moso bamboo in China is 68.3993 Tg C, accounting for 1.6286% of the total forest carbon storage. This implies the important contribution of moso bamboo in regional or national carbon budget.

Recently, convolutional neural networks (CNNs) showed excellent performance in many tasks, such as computer vision and remote sensing semantic segmentation. Especially, the ability to learn high-representation features of CNN draws much attention. And random forest (RF) algorithm, on the other hand, is widely applied for variables selection, classification, and regression. Based on the previous fusion models that fused CNN with the other models, such as conditional random fields (CRFs), support vector machine (SVM), and RF, this article tested a method based on the fusion of an RF classifier and the CNN for a very high resolution remote sensing (VHRRS) based forests mapping. The study area is located in the south of China and the main purpose was to precisely distinguish Lei bamboo forests from the other subtropical forests. The main novelties of this article are as follows. First, a test was conducted to confirm if a fusion of CNN and RF make an improvement in the VHRRS information extraction. Second, based on RF, variables with high importance were selected. Then, a test was again conducted to confirm if the learning from the selected variables will further give better results.