Spatial Distribution Of Forest Carbon Storage In Heilongjiang Province

Land vegetation plays a key role in global carbon cycle on fixing carbon dioxide (CO2) from the atmosphere into organic matter and providing the fundamental matter and energy sources for people’s life and production, so slowed the impacts of global warming. The forest is the main body of terrestrial vegetation, which play an important role in this process, so the ability of carbon sequestration of forest has become a focal point in the international societies and different study areas. To date, most research of forest carbon storage has been focused on forest biomass and carbon storage which has brought with it good results and benefits, but few on geo-spatial distribution of forest carbon storage. Some heterogeneity has been observed on the geo-spatial distribution of forest carbon storage, and forest management has a big influence on the carbon storage and density, so it is important to access to timely, accurate distribute information of forest carbon storage.This paper explore the spatial patterns based on fix plots set up in Heilongjiang, in2005to2010, and the data from59meteorological stations located in Heilongjiang, Jinlin and Inner Mongolia in that same year. Established three different models of distribution of forest carbon storage analysis both the advantages and disadvantage of different models on solving the problems of the spatial. Considered forest carbon storage is changing with time, established dynamic model to discuss the dynamic of forest carbon storage.Global and local Moran’s I, local statistics (local mean and local standard deviation) was used to explore the spatial patterns, spatial variations and spatial autocorrelations of forest carbon storage in Heilongjiang province using four bandwidths of25km,50km,100km and150km. The results showed that Moran’s I of forest carbon storage had statistically significant and positive correlation, which indicated that the changes of carbon storage tended to be similar with their nearby neighbors without a non-random manner. By local statistics, forest carbon storage was affected by environmental factors. Spatial heterogeneity of changes strongly existed in the study area with a large variation. The spatial distribution of forest carbon storage was significantly different between2005and2010with increased changes. Local statistics are useful tools for characterizing forest carbon storage changes across time and space, which are visualized by ArcGIS.We attempted to fit two global models, ordinary least squares model (OLS) and linear mixed model (LMM), and a local model, geographically weighted regression model (GWR), for the relationship between forest carbon content and s diameter of living trees (DBH), number of trees per hectare (TPH), elevation (Elev), slope (Slope), and the product of precipitation and temperature (Rain Temp). Global Moran’s I was computed for describing overall spatial autocorrelations of model results at different spatial scales using a range of bandwidth (from10to90km by a10km interval). Local Moran’s I was calculated at the optimal bandwidth (25km) to show the "hot spots" and "cold spots" of residual clusters. To quantify the spatial heterogeneity of model residuals, intra-block spatial variances (Sintra) were computed for the residuals using variable block sizes,ranging from5to30km by a5km interval. The results showed that the distribution of forest carbon storage in Heilongjiang had spatial effect. The distribution of forest carbon storage was influenced by stand, environmental and meteorological factors, especially average diameter of living trees (DBH). The GWR model outperformed the two global models in both model fitting and prediction because it successfully reduced both spatial autocorrelation and heterogeneity in model residuals. More importantly, the GWR model provided localized model coefficients for each location in the study area, which allowed us to evaluate the influences of local stand conditions and topographic features on tree/stand growth and forest carbon stock. It also helped us to better understand the impacts of silvicultural and management activities on the amount and changes of forest carbon storage across the province.Established dynamic model of DBH used re-surveying data of fix plots in2005and2010from all data. Predicted TPH used the established density of stocking index (SDI) dynamic estimate model. Plug prediction DBH and prediction TPH in2010into the GWR model which established in2005to predict forest carbon storage in2010. Compared with measured values of forest carbon storage in2010, found that the differences in the prediction forest carbon storage after five years and true values is relatively low. This method can effectively solve dynamic of forest carbon storage. So, calculated prediction DBH and prediction TPH in2015with this method, and plug in the GWR model which established in2010, and got the distribution of forest carbon storage in2015. Form predicted result, the distribution of forest carbon storage has not changed in five years. In2015, the minimum value of forest carbon storage still distributed in the areas of western regions of Heilongjiang province (Songnen plain) and small part of the Greater Hinggan mountains; the maximum value of forest carbon storage distributed in Zhangguangcai mountains of southeast, part of Small Hinggan mountains of north and part of Wanda mountains.