Studying The Atmosphere-land Interactions In Heihe River Basin Using Regional Climate Model And VIC Hydrological Model

Climate and water resource are essential factors which control the sustainable development of natural environment and human society in arid land in the northwestern China. Heihe river basin, the second largest inland river of China, is suffering from a series of serious environmental problems, such as vegetation degradation, desertification, and water deficit. The studying of land-atmosphere interactions will improve the understanding of water and heat exchange processes between land and atmosphere, comprehend the evolution of climate, ecology, and hydrology systems in the basin, and provide scientific basis for decision making in river basin planning and management.In order to explore the climate responses to vegetation change and the climate impacts on surface hydrology in Heihe river basin, we introduced a regional climate model (RegCM3) and a land-surface hydrological model (VIC) to analyze the land-atmosphere interactions through sensitivity experiments. My work focused on:(1) high-resolution simulation of temperature and precipitation in summer 1998 in Heihe river basin by using RegCM3 model with double-nested method; (2) analysis of vegetation change impacts on temperature and precipitation through three sensitivity experiments (forest restoration in mountainous area, farmland expanding in oasis, and desertification in lower reach); (3) simulation of runoff in the upper Heihe river basin by employing VIC hydrological model coupled routing model; (4) the discussion on the applicability of SDSM statistical downscaling model in upper Heihe river basin, and the projection of runoff change under different future climate conditions according to 15 synthetic climate change scenarios.The following conclusions were obtained:(1) RegCM3 could well capture the basic spatial-temporal pattern of temperature and precipitation at basin scale. In the temperature simulation, however, there was negative bias in the mountainous area, as well as positive bias in oasis and desert area. In the precipitation simulation, model's performance was not as good as that of air temperature. There was less rainfall simulated in oasis and desert area compared with its observation value. (2) Forest restoration in mountainous area might bring temperature decline and precipitation growth in summer. These changes could generate a pleasant climate condition for further vegetation restoration in mountainous area, but they might not affect the desert area in the north. With the expansion of farmland in oasis, temperature might fall in this area. There might be significant increase of precipitation in mountainous area, but slight decrease in some region of the lower reaches. Therefore, farmland expanding could trigger desertification in the edge of oasis or intensify desertification in the lower reaches. Desert encroachment might cause temperature rise and precipitation decrease in whole area. These effects could accelerate desertification and ecological degradation through its positive feedback processes.(3) As a large-scale land surface hydrological model with certain physical foundation, VIC model performed with good capacity in the upper Heihe river basin. VIC model effectively captured the characteristics of monthly runoff change at Yingluoxia station from 1990 to 2006. However, due to the shortage of meteorological forcing data and soil parameters in the model, the simulated runoff was less than the observed value. This negative runoff bias is much significant in dry seasons (winter and spring).(4) Because of the complex terrain conditions in the mountainous area, there was no significant statistical relationship between the climate characteristics in small region and the large-scale climatic field. In our study, statistical downscaling model might not be a suitable method to provide future climate scenarios to VIC model. The results of sensitivity analysis of future climate change impact on hydrology showed that:with the same temperature change, larger precipitation increase might bring more runoff; with the same precipitation change, higher temperature increase might cause less runoff. Compared with temperature, precipitation was the more remarkable factor influencing the runoff change in Heihe river basin. Moreover, climate impacts on runoff were different in a year. In the same precipitation change scenarios, temperature rising could lead to runoff increase in winter and spring, which might be the result of acceleration of snow melting, and runoff reduction in summer and autumn, which might be caused by the enhancement of evapotranspiration.