| Forest productivity plays an important role in terrestrial carbon cycle. Forest productivity,the direct reflection of forest community productivity for a certain natural environment，is thebasis of matter and energy cycle of forest ecosystem. At present, estimation of forest biomass andNPP on regional scale is one of the important scientific issues. Using different methods, estimateson forest biomass and NPP vary highly. This article summarizes monitoring and assessmentmethods of forest productivity, and large-scale forest productivity estimation mainly includesmeasured data space prediction and NPP Model, and simulates spatial distribution of forestproductivity with the geostatistical techniques of GIS. The results indicated that:（1） Based on the investigation data of167forest typical sampling plots collected frompublications，and by using equations between the stand age and biomass and NPP, we chose theexample of Guangdong and Guangxi provinces to estimate stand biomass and NPP under naturalgrowth in2010．In addition，the spatial pattern of forest biomass and NPP for each cell of thestudy region was derived by the spatial interpolation and geostatistics analysis of ArcGIS, and thespatial pattern of forest parts biomass was explored． The average of the biomass was140.35t·hm-2, and the total biomass in this region was up to2234.73Tg（1Tg=×106t） for that time, theaverage NPP was16.22t·hm-2·a-1，the annual net production of forest in the region was222.01Tg（1Tg=×106t）. There were different levels on biomass and NPP at different forest types. Thepattern of forest biomass decreased from north to south area, and the highest value was287.9t·hm-2. The NPP was lower in the center area, the highest value was in the west and south, whereis tropical rainforest. The biomass of the subtropical coniferous forest was the highest, about1786.84×106t. The NPP values of tropical evergreen no-timber woods reached a maximum of17.37t·hm-2·a-1.（2） This paper employed plots data collected, vegetation information systems to simulatespatial distribution of forest biomass and NPP with the geostatistical techniques. There are1959forest plot data distributed in various provinces. Forest biomass variation model showedthat nugget value and the ratio of still value was58.64%. The analysis of spatial autocorrelationof forest biomass in China indicates that the spatial variation of forest biomass appears mediumautocorrelation with partially patchy distribution. Nugget value and the ratio of still value of NPPwas16.45%. The analysis of spatial autocorrelation of forest NPP in China indicated that thespatial variation of forest biomass appeared high autocorrelation. The overall trend of forestbiomass and NPP was lower in central and western part, and gradually increases in eastern part.Forest biomass was one of the highest value appeared in the southwest and the Tianshanmountains in Xinjiang province, because of well hydrothermal conditions, and forest types weremainly original vegetation of tropical rain forests and spruce forest, that coverage rate was higher.In South China and Southwest, forest NPP was higher, mainly due to the higher forest coverage,and subtropical and tropical forests productivity was higher. Overall, the pattern of forest NPP increased from north to south area. Forest biomass was higher in Southwest, and North Chinaarea was the lowest. Forest NPP of the highest distribution was in South China, Northwest areawas the lowest.The results indicated that geostatistics was a powerful tool for spatial data analysis. It solvedusing a small amount of sample plots data for mapping forest biomass and NPP distribution.Future, applying higher precision integrated remote sensing data source and a variety of methodsto improve the accuracy of estimation is research direction of forest productivity. |
PDF Full Text Request