Estimating The Parameters Of Forest Ecosystems’ Carbon Sequestration In Qilian Mountains Based On Remote Sensing

As one of the three global carbon pools, the most complex and disturbed by themankind lie in the terrestrial ecosystem. Forest ecosystem is an important component ofterrestrial ecosystem, and it plays an important role in the ecological protection and theglobal carbon cycle. The forest ecosystem of Qilian mountains region, as a naturalecological barrier of northwestern area and China, has an important strategic position in theresearch of forest ecosystem carbon storage and carbon cycle in China. In this paper, Qilianmountains region is taken as the sample region for the study. The database of parametersof carbon sequestration has been established and analyzed by using GIS and RS techniques,and combined with the field measured data. The main conclusions are drawn as follows:（1） Unused land, which accounted for51.66%of the total area, was the main landscapetype in2010, followed by grassland, forests, wetlands, farmland and cities/towns. From2000to2010, grassland increased continuously, and farmland decreased; because of humanactivities, the area of cities/towns increased slowly; the area of forests and wetlands wasincreased. The evergreen coniferous forests and deciduous broad leaved shrubs forestsaccounted for99.7%of total forest areas in Qilian mountains region.（2） This paper developed an approach that can incorporate both vegetation index RVIand forest canopy through a model in estimating LAI of evergreen coniferous forests.Combination of vegetation index NDVI and measured LAI of deciduous broad leavedshrubs forest.With increasing of altitude, the forests LAI decreased from southeast tonorthwest. From2000to2010, the mean LAI value of the evergreen coniferous forests hadincreased0.13and the mean LAI value of deciduous broad leaved shrubs forests hadincreased0.06. The seasonal variation of LAI was significant, the monthly mean LAIincreased from January to April; the LAI of forest increased sharply in May; from June toAugust, the monthly mean LAI peaked on July; the LAI decreased continuously fromSeptember to December.（3） Builded the model of evergreen coniferous forests MODIS FPAR and RVI:y=0.0018x²0.0381x0.3276.Deciduous broad leaved shrubs forests MODIS FPARand NDVI: y=0.2805x0.2539. With increasing of altitude, the forests FPAR of studyarea are decreased from southeast to the northwest. From2000to2005, the FPAR valuereduced, account for72.2%of evergreen coniferous forests FPAR had not obvious changes,and the FPAR of deciduous broad leaved shrubs forests decreased from0.448to0.443.From2005to2010, the FPAR of evergreen coniferous forests and deciduous broad leaved shrubs forests showed a trend of increasing. The FPAR of forest shows a clear seasonvariation, July and August was the highest, followed by May, June, September and October,from January to April, November and December was the minimum.（4）This study used the improved model to estimate the forest NPP of study area. In theimproved model, the water stress factor in the Camegie-Ames-Stanford Approach （CASA）model has been improved, use the estimates results of FPAR in this study as the inputparameters in the CASA model, and take0.389gC/MJ and0.429gC/MJ as the maximumutilization rate of light energy of evergreen coniferous forests and deciduous broad leavedshrubs forests. From2000to2010, the mean NPP of evergreen coniferous forests decreased.Three years average NPP value was334.48gC·m^-2·a^-1, and the annual total NPP was1.08×10¹⁰gC·a^-1.The NPP of deciduous broad leaved shrubs forests decreased and thenincreased, the mean NPP was241.26gC·a^-1and the annual total NPP was4.97×10⁷gC·a^-1.The NPP changed significantly during a year: NPP raised monthly from January toJune, the peak in July and kept in August, and dropped month by month from September toDecember.The features and innovative of this study has the following three aspects:（1）This study discussed the land cover information, LAI, FPAR and NPP of forestecosystems in Qilian mountains and analyzed their dynamic changes. Compared withprevious that discussed only one index, this study can provide more extensive data and offermore dependable scientific evidences for terrestrial carbon cycles and the build of oasisecological environment in northwest China region.（2）When discuss the LAI, FPAR and NPP of forest ecosystem, forest was divided intodifferent types for quick and correct estimation of the Parameters of Forest Ecosystems’Carbon Sequestration in Qilian Mountains. The research result is more reliable than ever.（3）Based on the method of object oriented classification, the land cover classificationand the change of the land cover classification of Qilian mountains region can be get byeCognition with data of HJ^-1 and TM/ETM. Those processing can save calculation time andhave higher efficiency.