Proceedings Article | 3 November 2008
KEYWORDS: Ecosystems, Climatology, Remote sensing, Climate change, Data modeling, Vegetation, Environmental sensing, Satellites, Carbon, Ecology
Arid and semi-arid ecosystems exhibit a spatially complex biogeophysical structure. According to arid western special
climate-vegetation characters, using the remote sensing spectral mixture analysis. The study developed a modified
production efficiency model Net Primary Productivity-Geography Processing Ecology Model (NPP-GPEM) appropriate
for the arid area at regional scale based on the concept of radiation use efficiency. Combined with recent historical period
climate data and two periods of land use data sets from remote sensing data, we test the net primary productivity (NPP)
data sets over last 20 years in XinJiang of the North China modelled by the NPP-GPEM with the satellite data-driven for
detecting the wide spread spatial temporal characteristics of the impacts of climate and land use change impact with the
regional NPP. Our results shows that over past 20 years, the average annual temperature in the research region has
remarkably increased by more than 0.056°C, but over the same period, there has been a 0.635mm increase in annual
precipitation and increase in NPP by an annual rate of 0.142TgC. Taking use of 16 land covers of Xinjiang which was
carried on the calculation of the matrix transfer, the study got the matrix of the NPP changes over past 20 years. The
result found that the quicker trend increasing is the others woodlands (the 5.51gC/m2/yr) and the oasis farmland (4.32
gC/m2/yr). The pasture area is the biggest to about 29.3% of total area, the pasture growth is smaller than others, but its
increase to the contribution rate of the annual NPP in the whole land ecosystem system biggest is pasture and
agricultures to land use covers, that had the total annual NPP increase with 52.2% and 19.9% of the quantity
respectively, the plant cover is only the shrub pant which has a absolute decrease ecosystem system in the type (-
14.39gC/m2); In fine, the paper mainly reveals the dissimilarities and conversion among the pasture, farmland, woodland
and different covered pasture and other types. The correlation analysis between annual NPP and annual precipitation was
highly consistent with plant cover spatially (R2=0.64, P<0.01) in the overall XinJiang, and the northern of XingJiang is
better than the southern of it, and the correlation coefficient changes with the changes of vegetation types and different
regions. The analysis revealed that the climate changes dominate the impacts on the NPP in the whole study region.
However, land use plays a dominative role in the areas with land cover changes. Over past 20 years, the NPP in the
whole study region remarkably increased due to obvious precipitation decrease and temperature rise. Between two
periods of land use (about 10 years), the changes in climate are predicted to cause increases in NPP, and combined
impacts of climate changes and land use to cause to increase in NPP.
