Bioclimatic modeling the spatial distribution of mountain forests in the Qilian Mountains, Northwest of China, using down-scaled climatic models

Study on the Climate-Vegetation Coupling Relationship (CVCR) has been an attractive field in geography and ecology since the late 19th century. Its results are very helpful for decision-making of vegetation ecological restoration. However, it is hard for scientists to know precisely about the quantified relationships between a variety of vegetation types and a set of climatic parameters because of poor matching between spatially limited climatic data and high-resolution vegetation maps. In this article, the authors suggest new high-resolution distributing models that combine temperature and precipitation respectively with altitude, latitude, slope, and aspect, and perform regression analysis to deduce two equations linking temperature and precipitation respectively with readily observed altitude, slope and aspect. Using the two equations, mean annual temperature, mean monthly temperature, and annual precipitation for each cell can be calculated based upon the digital elevation model. The calculated results of models running are consistent with the actual conditions of temperature and precipitation in alpine zones of Qilian Mountains. According to the statistics of the distributing precipitation model, the regional annual precipitation maximum in Qilian Mountains occurs at 4500 m above sea level (a.s.l.). According to interpretation of Landsat TM/ETM+ imagery data, the actual areas of mountain forests in the study area are 422.94 km². The coupling relationship of mountain forests with both temperature and precipitation is achieved by combining the distributing temperature model with distributing precipitation model in the area of mountain forests together. The CVCR analysis indicates the mean annual temperature of -2.7~0.8degC, the mean July temperature of 9.3~13.7degC in the area of mountain forests, and the annual precipitation is more than 360 mm. According to the above-mentioned coupling relationships, a map which illustrates the potential extent of mo- untain forests is created. Statistics shows that under conditions without human-induced interruptions, potential areas of mountain forests would reach up to 6937.12 km². It means that, the present conditions of both types of forests only account for about 6% of the inferred potential areas