Best paper in Hydrology Research 2012
The study was motivated by the fact that climate change could have considerable influence on hydrology and corresponding water management, appropriate climate change inputs should be used to hydrological models for assessing future impact. Although the performance of RCMs has improved over time, systematic model biases still constrain the direct use of RCMs’ output for hydrological impact studies. The so called delta change (DC) approach has therefore been widely used in impact studies for producing input scenarios to hydrological models, but delta change approach produced future scenarios cannot reflect the important statistical properties of the observations.
This study developed a new distribution-based scaling (DBS) approach that adjusts precipitations and temperature from RCMs to better reflect observations. The scaling uses a double gamma distribution for daily precipitation, partitioned into normal and extreme, and a normal distribution for daily temperature. The DBS approach was compared with the delta change (DC) approach in several places in Sweden. The results show that the DBS approach produced a more realistic representation of local hydrology than by using raw RCM output; and the DBS approach preserves much better the future variability produced by regional climate model than the DC approach does.
The approach reported in the paper not only has high scientific value but also improves our operational efficiency and accuracy in simulation of hydrological impact of climate changes.