Hydrological Processes Modeling Considering The Effect Of Vegetation

The effect of vegetation characteristics on hydrological processes is seldom considered explicitly in most of conceptual watershed models, which limits their applicability to evaluate the impacts of climate and land use/cover change on hydrological cycles. In this study, a physically-based vegetation-hydrology mechanism was applied to the Xinanjiang conceptual hydrological model, namely the Xinanjiang vegetation-hydrology model. The vegetation-based model was used to explore the hydrological responses under variable environment in the Hanjiang River basin, the source area of water transfer project from south to northern China.The University of Maryland's global 1 km land cover data was used to represent the vegetation cover over the study area. The spatio-temporal distribution of leaf area index was derived from the remote-sensed NDVI (normalized difference vegetation index) data. Parameters for various vegetation types were estimated based on the land data assimilation system and related references. In consequence, a physically-based two-source potential evapotranspiration model was used to calculate potential evapotranspiration over the drainage area of the Hanzhong hydrological station at the upper reach of the Hanjiang River. Results show that vegetation types have remarkable impact on the spatio-temporal distribution of potential evapotranspiration. The Xinanjiang hydrological model coupled with the two-source potential evapotranspiration model was able to simulate daily streamflow at the Hanzhong hydrological station with the Nash-Sutcliffe efficiency coefficient of 0.737 to 0.977.With the aid of geographic information system and remote sensing technique, the Xinanjiang vegetation-hydrology model was established on digital basin, in which the effect of vegetation phonological characteristics (leaf area index), root depth, physiological features (stomatal opening and closure) and surface roughness on evapotranspiration, runoff generation and overland flow processes were explicitly described. As a result, the Xinanjiang vegetation-hydrology model performed well in simulating daily streamflow and flood events at the Hanzhong hydrological station.Several land cover scenarios in the upstream area of the Hanzhong station were developed by the extreme land use/cover method. Based on those land cover scenarios, the Xinanjiang vegetation-hydrology model was used to predict the possible hydrological variations resulted from land cover change. Results also show that afforestation in the Hanjiang River basin might raise evapotranspiration by 2.1-11.1 % and reduce runoff by 1.3-7.5 %, and cut down and delay flood peaks effectively. Conversion from forests to cropland, wooded grassland or grassland covers tends to reduce evapotranspiration by 3.1-22.2 % and increase runoff volume by 2.0-15.0 %. This invariably could make flood peaks increase and appear in advance.A model to assess the impact of climate change on runoff in the Hanjiang River basin was established using the Xinanjiang vegetation-hydrology model with the grid system at a 0.5-degree spatial resolution. This model was further one-way coupled with the PRECIS (Providing Regional Climate for Impacts Studies) regional climate model to evaluate the possible trends in runoff in theHanjiang River basin under the IPCC-SRES (Intergovernmental Panel on Climate Change-Special Report on Emission Scenarios) greenhouse emission scenarios. Model simulation indicates that mean annual runoff averaged from 1991 through 2100 will increase by about 4.7% in the Hanjiang River basin as compared with the baseline years (1961-1990) and represent more remarkable spatial variability, implying a rise in the possibility of flood and drought occurrence in some local area of the Hanjiang River basin.The LPJ (Lund-Potsdam-Jena) dynamic global vegetation model was used to assess the possible vegetation responses to climate change in the Hanjiang River basin. Simulations with various climatic conditions and carbon dioxide concentration show that remarkable vegetation changes occur in terms of net primary productivity, leaf area index, vegetation composition and distribution. Furthermore, the Xinanjiang vegetation-hydrology model coupled with the LPJ model was applied to hydrological simulation considering the vegetation variations caused by climate change. The conclusion here is drawn that vegetation dynamics should be taken into consideration in streamflow simulation over long-term scale.