Study On Regional Hydrological And Drought Characteristics Based On Multi- Drought Indexes And Multi-hydrological Models

In Heilongjiang,China's economic growth,an important role is being played by the Songhua River because of the river's unique resources and natural conditions.The Songhua River Basin?SRB?serves as a major grain commodity basin and is located in the northeastern region of China.The stream flow in the Songhua River has dropped significantly in recent decades,which poses a serious threat to sustainable development and water management in the context of global warming.Therefore,the fidelity of precipitation extremes is very important for the assessment of natural disasters?i.e.,floods and drought?.Droughts can have long-lasting impacts on ecosystem including stresses on natural system,economic losses and humanitarian disasters.Thus,it is necessary to study the variation in hydrological and meteorological behaviour for historical and future basis to make a comprehensive plan for water resources of the basin.For this Purpose,the current study was carried out using multiple hydrological models forced by point data and bias corrected precipitation from an ensemble of General Circulation Models?GCMs?to assess the hydrological and meteorological behavior of SRB.Furthermore,utilization of multiple drought indexes also carried out to analyze the historical and future drought behavior of the study area.This research used the snow cover data;derived from Moderate Resolution Imaging Spectroradiometer?MODIS?,stream flow and meteorological records?precipitation?PP?and surface air temperature?SAT??.In addition,different GCMs and three RCPs?8.5,4.5 and 2.6?involved in the Coupled Model Intercomparison Project Phase 5?CMIP5?and two Scenarios?SRES-A2,SRES-A1B?for the period of 1961 to 2005 and 2006 to 2099 were used to assess the stream flow variations and trends of climate variables in the SRB.Firstly,the snow cover trend was assessed to identify the impact of snow cover on stream flow.The results showed that the seasonal snow cover trend had no impact in changes in the stream flow.The snow cover was slightly increasing in the entire area,while the stream flow trend results displayed(22.21 m³/sec×year^-1)a significant decline in the basin.Based on trend analysis,the water evaluation and planning tool?WEAP?integrated with PP,the APHRODITE precipitation product and SAT,were used to simulate the monthly stream flow and evaluate the impact of climate change on this stream flow during the 21^stt century.Furthermore,the stream flow drought index?SDI?and Percent of normal index?PNI?were computed to project the drought conditions.The results indicated that in the 21^stt century,the magnitude of SAT over the study area increased for three RCPs of 8.5,4.5,and 2.6 by5.82°C,2.44°C,and 1.12°C,respectively.The application of the WEAP model suggested a maximum decline in stream flow,reaching 24%until the end of the 21st century under future climate change scenarios.The stream flow drought indexes?SDI and PNI?demonstrated that chances of severe to extreme hydrological drought events were highest in 2059,2060 and2085,while in the remaining time period mild to moderate drought events may occurred in the entire study area.The results also exhibited that the mean drought duration,severity and intensity for the period of 2011 to 2099 under all scenarios may be[?A1B,12,-1.55,-0.12?,?A2,12,-1.41,-0.09?,?max.wetting and warming conditions,12,-1.37,-0.11?and?min.wetting and warming conditions,12,-1.69,-0.19?],respectively.Secondly,the evaluation and application of GCMs coupled with hydrological models?NedborAfstromnings Model?NAM?,GRJ4 and WEAP?and drought indexes were used to assess the performance of hydrological models and drought behavior in the SRB.The results exhibited that WEAP and NAM presented better performance compared to GR4J in the SRB.By forcing hydrological models with bias corrected precipitation,the differences in predicting discharge by NAM,GR4J and WEAP were-25%,-29%and+3%respectively,compared to observed discharge.The results suggested that overall performance of WEAP was better than NAM and GRJ4 in the study area.The results also showed that application of multiple hydrological models is necessary to make better planning and management of agriculture and water resources in the region.Moreover,seven indexes of precipitation extreme with original and reformulated drought indexes were selected to monitor the drought conditions as well as to make a robust comparison between the drought indexes.Composite Index?CI?and Reconnaissance Drought Index?RDI?was reformulated with Penman–Monteith equation and pan evaporation.In addition,CMIP5 historical model simulations were used to identify the discrepancies among the GCMs in capturing precipitation extremes and droughts in the SRB.The results exhibited that wet-day precipitation increased at a rate of 1.9 mm/year,while the simple daily intensity index exhibited a non-significant decreasing trend during the whole time period.The number of consecutive dry days showed a positive trend with a slope of 0.33 days/year,while the number of consecutive wet days showed a negative trend.CMIP5 historical model simulations showed a small differences compared to observed extreme precipitation indexes.A comparison among drought indexes displayed that performance of reformulated drought indexes was better than original ones.During July,CI?Pan?indicated positive?0.23?value,while April and September displayed negative values,-0.6 and-0.7,respectively.This showed that these two months are likely to be prone to drought.Thirdly,three drought indexes based on precipitation and evapotranspiration was selected to predict future situation of drought in the SRB.The projection of drought events indicated that under RCP 8.5,all indexes showed maximum drought occurrence in the region compared to other path ways.The results also indicated that the Harbin,Tonghe,Suihua,Jiamusi and the Mudanjiang stations located within the SRB show maximum drought episodes in late stage?2060-2099?.Moreover,in the early?2006-2030?and middle?2031-2060?stage,the indexes indicated a small spread of drought occurrence over the region compared to late stage of the 21st century.Furthermore,based on multiple drought indexes and ensembles of GCMs,the current analysis also suggested that the indexes used reference evapotranspiration capture fewer drought years compared to temperature based indexes.Fourthly,a comparison of gridded precipitation data sets with station based precipitation data was also carried out in the SRB.For this purpose,three gridded data sets?the Global Precipitation Climatology Centre?GPCC?,Climate Research Unit?CRU?and the University of Delaware?UDEL??and in-situ station based precipitation obtained from National Meteorological Administration?NMA?in the selected region of SRB for the period of 1961to 2005.The spatial comparison among data sets,precipitation concentration index?PCI?and drought frequency analysis was adopted to evaluate the performance in the study area.The results exhibited that on spatial scales,gridded data sets showed higher amount of annual precipitation at most of the stations compared to NMA.The spatial correlation between GPCC,UDEL and NMA showed good agreement compared to CRU.The annual and seasonal precipitation concentration indexes results suggested that the gridded and station precipitation products exhibited an irregular pattern even 97%stations showing extremely irregular pattern in the region during the selected time period.The seasonal results of PCI showed that gridded data products displayed all the station have irregular precipitation distribution except summer,while in-situ station indicated that winter and summer most of the stations had uniform pattern.In summer season,the UDEL results were similar with NMA compared to CRU and GPCC.The analysis revealed that 41%to 22%stations based on gridded data sets experienced non-significant negative trend compared to observed data set precipitation.The drought frequency analysis based on in-situ and gridded data sets was comparable.Overall,the results implied that under future climate conditions the risk of drought and stream flow decline might be increased at the end of 21st century.Therefore,government agencies should take more measures to mitigate and improve water resources of the region.In addition,this study also pointed out a new challenge regarding the utilization of gridded data sets due to discrepancies in the results.Therefore,it is very important to minimize these discrepancies by quantifying the uncertainty,raw data systematic errors,and application of satellite based data to improve the quality of data in the region.