Research On The Streamflow Responses To Climate Variability And Human Activities In Northeast China

Evaluation of the impacts of climate change and human activities on runoff are important for water resources planning, management and configuration in a catchment. There are temporal and spatial distribution changes for the observed hydro-climatic factors. It was induced by both of the natural variability and human activities including greenhouse gas emissions and the influence of land use/cover changes. Understanding the happened climate change and human activities and their influence on the hydrological process based on observed data is good for future runoff prediction. Selecting seven catchments as the study subjects, this paper estimated the impacts of human activities and climate change on runoff quantitatively and predicted the future runoff considering the influence of human activities. Firstly, the research thoughts of "Reverse verification after preliminary confirmation" were used to determine the "natural period". After that the climatic elastic coefficient method and hydrological simulation method were used to evaluate the runoff responds to climate change and human activities during the last fifty years. The method of downscaling GCM outputs coupled with hydrological simulation was used to future runoff prediction. The main content and results of this study are as follows:(1) Long time gradual trend and step change are important indicators of the hydro-climatic change/variables. Based on long term observed data, the Mann-Kendall test, linear regression test and Mann-Whitney-Pettitt change point analysis method were used to detect and diagnose the temporal and spatial distribution changes of the hydro-climatic factors. Dominant factors of the water cycle were analyzed and impacts of climate change and human activities on runoff were identified qualitatively according to the test results.(2) SWAT model for the seven selected catchments were established with the DEM data, soil data, land use/cover data and hydrometeorological data. Then the multivariable and multistation method was used in model calibration on the basis of water balance. Parameters of the SWAT models were celebrated step by step from upstream to downstream. Evaluation of the SWAT model showed that SWAT model are available and can descript the runoff process reasonably and can be used to evaluate the impacts of climate change and human activities on runoff.(3) Detection of the "natural period" is extremely important for quantization of the runoff response to climate changes and human activities. Focusing on the defects that "natural period" cannot be verified in the previous study, this paper proposed a thought of "reverse verification after preliminary confirmation" to verify the "natural period". Determine the significant impacted period firstly with change point analysis method. Then, use the earliest hydrometeorological date to calibrated hydrological model and reconstruct "natural runoff" There after "rainfall-runoff and "rainfall-natural runoff double accumulation curve can be drew in the same coordinate. According to the separation time of the two double accumulation curve and the information of human activities, rationality of the "natural period" and "natural runoff can be verified at last.(4) After reasonable division of the "natural period" and reconstruction of the "nature runoff, climatic elastic coefficient method and hydrological simulation method were used to evaluate the impacts of climate change and human activities on runoff quantitatively. The results show that, compared to natural period, contributions of climate change and human activities on runoff change calculated by the two methods are consistent. In the selected catchment, climate change and human activities affect runoff commonly, however, human activities play a leading role.(5) Detailed analysis of the climate elasticity of streamflow and human induced runoff change show that:Streamflow is positively related to precipitation while negatively related to temperature; the precipitation-streamflow sensitivity are greater than that of the streamflow-temperature; increasing runoff reductions were found for the catchments studied; inter-annual runoff distribution has been seriously disturbed in the catchments where large reservoirs were built; runoff was little changed in the forest catchment where human activities were weak. Human activities have distinct regional characteristics, and human-induced runoff changes were related to the annual precipitation.(6) Human activities as well as climate change are driving factors that induced runoff change. The impacts of human activities are even greater in some catchments in northeast China. However, the traditional hydrological projection was directly estimated by calibrated hydrological model with the output from GCMs, without considering the impacts of human activities on water resources. In this study a correction coefficient method were presented to predict future runoff considering human activities. The hydrological projection was assessed by the following steps:celebrate the revise coefficient according to the observed precipitation, observed runoff and reconstructed nature runoff; downscaling the outputs of GCMs to regional scale, and take it as input data of hydrological model to simulate future "natural runoff; estimate the future runoff tendency considering human activities according to the correction formula.