Combined dynamical and statistical downscaling for high-resolution projections of multiple climate variables in the Beijing–Tianjin–Hebei region of China

Abstract

High-resolution combined dynamical and statistical downscaling for multivariables (HDM) was performed in the Beijing–Tianjin–Hebei (BTH) region by using observations from China Meteorological Administration Land Data Assimilation System (CLDAS), a regional climate model (RCM), and quantile mapping. This resulted in the production of a daily product with six variables (daily mean, maximum, and minimum temperature; precipitation; relative humidity; and wind speed), five ensemble members, a multidecadal time span (1980–2099), and a high resolution (6.25 km) for climate change projections under the RCP4.5 scenario. The evaluation showed that the HDM output could reproduce well the mean states of all variables and most extreme indices except the consecutive dry and wet days. The biases in the magnitude of interannual variability in HDM were mostly inherited from the RCM. By using the HDM, future projection over BTH was conducted. The results indicated that the annual mean temperature and precipitation as well as extreme heat and heavy precipitation events will increase over most regions. The warming magnitudes over the mountainous and coastal area at the northern BTH and the wetting magnitudes over the Daqinghe River basin (DRB) within BTH will be relatively stronger. The increases in extreme heat events will be much larger in the plain area. More than one-half of regions with the large extreme precipitation increase will be located within DRB. Both the number of models with the same sign of change and the ensemble standard deviation were used to estimate the projection uncertainty. The projected changes and uncertainties over DRB and subregions and Xiong’an city within the basin for each season are also discussed.