Effect of baseline meteorological data selection on hydrological modelling of climate change scenarios

- Climate change impacts have compared across three precipitation and ETo datasets each.
- Model parameterisation compensates for input data choices in baseline.
- Subjectivity in model input space selection influences climate change impacts.

SummaryThis study evaluates how differences in hydrological model parameterisation resulting from the choice of gridded global precipitation data sets and reference evapotranspiration (ETo) equations affects simulated climate change impacts, using the north western Himalayan Beas river catchment as a case study. Six combinations of baseline precipitation data (the Tropical Rainfall Measuring Mission (TRMM) and the Asian Precipitation - Highly Resolved Observational Data Integration Towards Evaluation of Water Resources (APHRODITE)) and Reference Evapotranspiration equations of differing complexity and data requirements (Penman-Monteith, Hargreaves-Samani and Priestley-Taylor) were used in the calibration of the HySim model. Although the six validated hydrological models had similar historical model performance (Nash-Sutcliffe model efficiency coefficient (NSE) from 0.64 to 0.70), impact response surfaces derived using a scenario neutral approach demonstrated significant deviations in the models' responses to changes in future annual precipitation and temperature. For example, the change in Q10 varies between −6.5% and −11.5% in the driest and coolest climate change simulation and +79% to +118% in the wettest and hottest climate change simulation among the six models. The results demonstrate that the baseline meteorological data choices made in model construction significantly condition the magnitude of simulated hydrological impacts of climate change, with important implications for impact study design.

154