Quantifying The Impacts Of Climate Change And Human Activities On Water Resources Of The Yangtze River Basin

In recent years,temporal and spatial changes of water cycle and water resources become more intensive under the mutual forcing of climate change and human activities.Responses of hydrology and water resources to climate change and human activities become a hot topic in the metero-hydrological relevant fields.Yangtze river basin,as the largest river basin and one of the most populous and prosperous regions of China,has doubtlessly undergone intensive effects of climate change and human activities,which resulted in the redistribution of the water resources at the temporal and spatial scales over the region.Till now,however,very few exhaustive and systematic studies on the temporal and spatial change of water resources in the Yangtze river basin can be found in the literatures,so it is difficult to answer how much variation of water cycle and water resources under the mutual forcing of climate change and human activities in recent decades.Hence,this study took the basin controlled by Datong hydrometric station as the study watershed which accounts for 97%of the whole Yangtze river basin,to explore the temporal and spatial variations of the climate and land use,quantify the influences and contribution of climate change and human activities(including underlying surface change and reservoirs operations)on variations of runoff and blue and green water,reconstruct terrestrial water storage change(TWSC)time series based on Gravity Recovery and Climate Change(GRACE)satellites observations and hydrological model simulation,and examin the relationship between TWSC and El Nino-Southern Oscillation(ENSO)during the period of 1988-2012.The major conclusions obtained through this study were summarized as follows:(1).Based on the air temperature and precipitation data during the last 30 years collected from 173 meteorological stations,climatic variation trends at each station were analyzed by the linear regression analysis and Mann-Kendall nonparametric test method.The results suggested that,increasing trends exhibited clearly on annual mean air temperature at 171 meteorological stations among which 158 meteorological stations showed obvious warming trends with the significant level of α<0.1.Annual precipitation tended to increase in the north and southeast of the study watershed while decreasing trends were found in other parts of the region,but the variations were found not obvious in most regions.(2).Statistical analysis on land use changes for the study watershed were systematically conducted based on land use remote sensing data in the late 1980s(1990),1995,2000,2005 and 2010.Land use conversion matrixes were obtained to reveal the conversion between different land use types.The results suggested that,intensive conversions among different land use types appeared from 1990 to 2010,but no obvious change on total area of each land use type was found in the study region.The most obvious land use changes reflected in the study region were the decrease of plough land and the continuous increase of urban and rural construction land.The decreased plough land was mainly converted to urban and rural construction land.(3).Based on the Soil and Water Assessment Tool(SWAT)model,watershed delineation for watershed boundary and stream networks were done first of all on the DEM and the hydrologic response units were defined.According to the land use data and human activities conditions,five periods of study duration,i.e.1988-1992,1993-1997,1998-2002,2003-2007 and 2008-2012,were subdivided.The observed discharge data from 6 hydrological stations and evapotranspiration data derived from Global Land Data Assimilation System(GLDAS)CLM and Noah models were used to calibrate and validate the SWAT model.The results suggested that,after calibration and validation,SWAT model can fully meet the requirements for exploring the water resource variations in the study region.(4).By means of SWAT model,based on different simulation scenarios,impacts of climate change and human activities(including underlying surface change and reservoirs operations)on runoff change between different periods were quantitatively evaluated.For the outflow of the whole basin,runoff variation caused by climate change ranged from 0.7%to 14.8%.However,runoff variations resulted from underlying surface change and reservoirs operations during all periods were less than 5.0%and 0.5%,respectively.The contribution rate of climate change to runoff change was above 65%when the runoff varied violently.Contribution rates of underlying surface change to runoff variation would be great only when the runoff slightly changed.Contribution rates of reservoirs operations to runoff variation were very small in the statistical time step(5years).In addition,contribution rates of climate change to outflows variations of most sub-basins were above 60%during all periods.Climate change plays a predominate role in runoff variations for the study region.(5).Comparisons of discharge at Datong hydrological station in the scenarios of reservoirs operations and no reservoirs operations suggested that reservoirs operations posed strong affections on seasonal discharge variations of the watershed.For the periods of 1993-1997 and 1998-2002,the maximum compensation and reduction of discharge at Datong station caused by reservoirs operations reached 5.4%and-3.2%respectively(relative to the discharge in the same month).Because of the operation of the Three Gorges Reservoir after 2003,influence of reservoirs operations on discharge change during 2003-2007 was strengthened,and the discharge was compensated from January to June and reduced from July to November.The maximum compensation and reduction of discharge at Datong station due to reservoirs operations reached 19.6%and-20.4%respectively.During 2008-2012,the discharge was compensated from January to May and reduced from June to December,and the maximum compensation and reduction of runoff at Datong station due to reservoirs operations reached 33.3%and-23.5%respectively.(6).Temporal and spatial variations of blue water flow(i.e.the sum of surface runoff,lateral flow,and groundwater recharge generated in soils),green water flow(i.e.evapotranspiration)and green water coefficient during 1988-2012 in the study area were analyzed.The results suggested that blue and green water gradually increased from the source region to the downstream.The average annual blue and green water were less than 100mm/a and 300mm/a respectively in the source region,and greater than 1000mm/a and 800mm/a respectively in the downstream.Green water coefficients gradually deceased from the source region to the downstream,and annual mean green water coefficients were greater than 80%in the source region and less than 40%in the downstream.Multi-year basin-level averaged blue water,green water and green water coefficient were 557mm/a,554mm/a and 50%(green water coefficient is less than many other basins in the world).During the period of 1988-2012,basin-level averaged annual blue and green water exhibited non-significant(α>0.1)decreasing trends with speed of-27.5mm/10a and-4.9mm/10a respectively,while annual basin-level averaged green water coefficient showed a non-significant(α>0.1)increasing trend with speed of 0.01/10a.(7).By mean of SWAT model,based on different simulation scenarios,impacts of climate and underlying surface changes on blue and green water variations between different periods were quantitatively evaluated.The maximum variation of blue water caused by climate change can reach 14.4%.The contribution rate of climate change to blue water change was above 65%when the blue water variation was great.Blue water variations caused by underlying surface change during all periods were less than 5.0%.Contribution rates of underlying surface change to blue water change would be great only when the blue water slightly changed.The variations of basin-averaged green water between different periods were slight,and the maximum variation was 3.8%.Contribution rates of underlying surface change to green water variations were greater than that of climate change.Further analysis at sub-basin scale also supported the above mentioned findings,which leads us to conclude that climate change plays a predominate role in blue water variations,while underlying surface change governs green water variations.Moreover,the blue water and green water in most sub-basins featured the opposite variations patterns under the influence of underlying surface changes.(8).Monthly TWSC time series during 2003-2012 were established by means of GRACE satellites and hydrological model,respectively.By comparisons with GRACE-based TWSC,"Pswat-Egldas-Robs"(Pswat is the precipitation simulated by SWAT model;Egldas is evapotranspiration from GLDAS;Robs is the observed discharge)and "Pswat-Eswat-Robs"(Eswat is evapotranspiration from SWAT model)can be used to establish monthly TWSC time series during 1988-2012.Monthly non-seasonal TWSC time series smoothed by a 5-month running mean was found negatively correlated with Southern Oscillation Index(SOI)during 1988-2012 with the significant level of α<0.01,which implied a negative correlation exited between terrestrial water storage(TWS)variation and ENSO at a long time scale.From hydrological points of view of runoff,blue and green water and TWS,this study comprehensively evaluated the impact of climate change and human activities on water resource variations.This research work can provide scientific theory and technological support for the rational development,utilization,management and protection of water resources in the Yangtze river basin.