Abstract
Abstract. To bridge the gap between large-scale GCM (global climate model) outputs and regional-scale climate requirements of hydrological models, a spatiotemporally distributed downscaling model (STDDM) was developed. The STDDM was done in three stages: (1) up-sampling grid-observations and GCM simulations for spatially continuous finer grids, (2) creating the mapping relationship between the observations and the simulations differently in space and time, and (3) correcting the simulation and producing downscaled data to a spatially continuous grid scale. We applied the STDDM to precipitation downscaling in the Poyang Lake watershed using the MRI-CGCM3 (Meteorological Research Institute Coupled Ocean–Atmosphere General Circulation Model 3), with an acceptable uncertainty of ≤ 4.9 %. Then we created future precipitation changes from 1998 to 2100 (1998–2012 in the historical scenario and 2013–2100 in the RCP8.5 scenario). The precipitation changes increased heterogeneities in temporal and spatial distribution under future climate warming. In terms of temporal patterns, the wet season become wetter, while the dry season become drier. The frequency of extreme precipitation increased, while that of the moderate precipitation decreased. Total precipitation increased, while rainy days decreased. The maximum continuous dry days and the maximum daily precipitation both increased. In terms of spatial patterns, the dry area exhibited a drier condition during the dry season, and the wet area exhibited a wetter condition during the wet season. Analysis with temperature increment showed precipitation changes can be significantly explained by climate warming, with p<0.05 and R≥0.56. The precipitation changes indicated that the downscaling method is reasonable, and the STDDM could be successfully applied to the basin-scale region based on a GCM. The results implied an increasing risk of floods and droughts under global warming, which were a reference for water balance analysis and water resource planning.
Highlights
Global warming has caused temporal and spatial redistributions of precipitation (Frei et al, 1998; Alexander et al, 2006; Trenberth et al, 2011) and has increased the frequency and intensity of floods and droughts, seriously threatening social systems and ecosystems (Pall et al, 2011; Dai, 2013)
Considering the spatiotemporal heterogeneity of precipitation at the regional scale such as the Poyang Lake watershed, we developed a spatiotemporally distributed downscaling model (STDDM), which is a logical framework based on a specific mathematic algorithm
The spatiotemporally distributed downscaling method (STDDM) showed a better performance than the non-STDDM
Summary
Global warming has caused temporal and spatial redistributions of precipitation (Frei et al, 1998; Alexander et al, 2006; Trenberth et al, 2011) and has increased the frequency and intensity of floods and droughts, seriously threatening social systems and ecosystems (Pall et al, 2011; Dai, 2013). To the fragile ecological and living environments, the state of the future hydrological situation with future global warming is a crucial question for avoiding or reducing damage from climate warming. Global climate models (GCMs) are basic tools for assessing the effects of future climate change and provide an initial source for future climates (Xu, 1999). GCMs have coarse global resolutions, ranging from 1◦ × 1◦ to 4◦ × 4◦, and are not applicable on regional scales, such as watersheds. Downscaling algorithms have been developed to link the global-scale GCM outputs and the regional-scale climate variables, including dynamic (Giorgi, 1990; Teutschbein and Seibert, 2012) and statistic Zhang et al.: Precipitation projection using a new downscaling method
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call