Streamflow responses of land use and climate change in a watershed of Northern China: implications for adaptive watershed management

It is of great significance to assess and project the respective impacts of land use change (dQ_Landuse) and climate change (dQ_Climate) on streamflow (Q) for water resources management. In this study, we used elasticity differential analysis approach and physical processes based distributed parameter watershed hydrological model to quantify the relative contributions that land use change and climate variability have on the decadal streamflow dynamics of the Chaohe watershed with the area of 4854km2 located in the northern China. Furthermore, the watershed hydrological model was applied to investigate the future hydrographic characteristics driven by downscaled precipitation and temperature projected by General Circulation Models (GCMs) under three emissions scenarios. The result suggested that watershed streamflow, compared with the reference period from 1963-1979, greatly decreased during 1980–1989 and 2000–2008, whilst it slightly changed during 1990–1999. The insignificant streamflow change for 1990–1999 was due to the effects of lower soil water storage capacity than that of other periods on the hydrological impact of land use change. In addition, dQ_Climate for 1980–1989 and 2000–2008 were different between the approaches: dQ_Climate were almost similar to dQ_Landuse for these two periods according to eco-hydrological approach, whilst dQ_Climate from the differential elasticity-based analysis only 33% and 45% and from modelling 51% and 78% for 1980–1989 and 2000–2008, respectively. The future climate exhibits a drier and warmer trend in the summer monsoon period contrasting with other seasons in the watershed. Precipitation will decrease by 47.5–57.2 mm during the summer monsoon period while increasing annually. Future summer streamflow will decrease accordingly driven by increased evapotranspiration due to the rising temperature. An increased dispersion coefficient of streamflow also indicates more dramatic variations in summer than that of other seasons. The annual streamflow magnitude with a 5-year return period increases significantly (p < 0.01), indicating a reduced risk for future water shortages. However, the magnitude of streamflow will decrease with the prolonged return periods (p<0.01). Our study highlights the critical importance to interpret the hydrological impacts by different approaches with great care and to predict the seasonal variability of streamflow characteristics for developing adaptive resource management and hazard relief strategies as the hydrological impacts of land use change and climate change are temporally varied.