The Research On Carbon Sink And Its Impact Factors Of Broad Leaf Mixed Forest Of Wuzhishan City

Terrestrial vegetation plays a key role in the global carbon cycle on fixing carbon dioxide in the atmosphere into the organic matter and transforming the solar to chemical energy, which not only provides the fundamental material and energy but also mitigates the influence of global warming. Forest plays an important part in this process as the main body of land vegetation, so carbon sequestration in forest has become a research hotspot in domestic and overseas. Wuzhishan city, as a city with the highest forest coverage rate, has extremely rich forest resources, which have an immeasurable impact on the whole climate of Hainan Island. So the paper studies on broad leaf mixed forest of Wuzhishan city. On the one hand, it estimates the dynamic change of carbon sink based on five times of Wuzhishan city Forest Resource Inventory data with biomass model and carbon content. On the other hand,using Wuzhishan 2011 city Forest Resource Survey and DEM data, it studies the spatial distribution of carbon sink from canopy density, altitude, slope and distance to water and establishes the spatial model through Geoda and Arcgis software, to analysis in different spatial positions impacts of each factor on carbon sinks,to provide theoretical evidence for making scientific and efficient carbon sink policy.The main results include:1.The carbon stock of broad leaf mixed forest currently is 2898.46t, which compares with 1993, carbon sink function significantly increases. Especially in 1999, Hainan Province looked the forestry as the theme and the core of building an ecological province. In 2008,Wuzhishan city samely seemed the ceo-city as its building goal. The carbon storage and sink increased highly with the forest area from 8.12 ×104hm2 to 9.19 ×104hm2. But from 2008 to now, carbon stock continue to increase with the smaller margin, which leads to carbon sink start to decline. Carbon sink has been positive overall, which indicates that broad leaf mixed forest always play carbon sink function.2.Analysising the distribution characteristic of carbon sink of broad leaf mixed fores from canopy density, altitude, slope and distance to water, the result shows that carbon sink increase gradually as the area and canopy density increase and the moderate and high degree of canopy density are main carbon pools, which constitute the 92.1% of the total carbon sink. Dividing the altitude into six elevation gradients, the spatial distribution characteristic is a parabola with the opening down, which shows the carbon sink first increases then decreases with the gradually increased altitude. From 142m to 1050m, carbon sink has a positive relationship with altitude and carbon sink reaches the maximum value between 142m and 1050m. After that,carbon sink has a negative relationship with altitude. The distribution characteristic of carbon sink in slope is similar with the altitude, the carbon sink is in the order of abrupt、steep slope (15°<slope<35°)> urgent、dangerous slope (slope>35°)> level、gentle slope (slope<15°). The carbon sink is bigger as the distance to water is closer.3.Due that they all exist spatial auto correlation, this paper use spatial regression model and geographically weighted regression model to fit the equation of carbon sink, DBH*Height, canopy density, altitude, slope and the distance to water. For the both spatial regression models, all parameters pass significant test at a=0.05 level, indicating these factors have important influence on carbon storage. Moreover, influence of DBH*Height and canopy density are the largest factors, the value are about 0.80, Secondly are the altitude and the distance to water. The value of slope is the smallest, about-0.2. But these two models belong to global model, which would generate a big model local error due to differences in topography in Wuzhishan. The GWR model provides localized model coefficients for each location in the study area, which successfully reduces both spatial auto correlation and heterogeneity in model residuals. So the model fitting outperforms the spatial regression model. The result shows that the regression equation between carbon sink and the factors in different samples is different, so the factors have different impacts on carbon sink in different samples. Based on the different influence levels, take corresponding measures to forest management in order to enhance carbon sequestration capacity in the region and thereby increase its carbon sinks.