Abstract
A comprehensive performance assessment of the empirical–statistical downscaling (ESD) technique named EPISODES is presented. Pertaining evaluation analyses consist of multifarious validation experiments as well as various comparisons of EPISODES’ projections with those of three RCMs and two ESD methods based on the same GCM scenarios driven by two distinct representative concentration pathways (RCPs). EPISODES combines the downscaling of GCM simulations with a follow-up production of synthetic local time series. EPISODES is a comparably simple, computationally rather inexpensive technique, providing multi-variable and multi-site data that are suitable for being merged in an ensemble of RCM projections. This allows (e.g. for different RCPs) the compilation of large multi member ensembles derived from various GCM simulations via both main downscaling strategies (ESD and RCMs). Evaluation experiments reveal satisfying degrees of compliance between various results generated by EPISODES and observations. The grid cell bias for yearly values, for instance, is mostly less than 0.1,^circC for temperature and 10% for precipitation totals. Recorded temperature values and precipitation totals corresponding to their 1st and the 99th percentiles are well represented by EPISODES too. Comparisons of various climate change signals derived by EPISODES and other downscaling approaches, present high levels of agreement as well. Many more findings referring to evaluation experiments and climate change projections are to be found in the paper as well as throughout the “Appendix”.
Highlights
IntroductionClimate change across the globe is driven by changing forcings (e.g. solar irradiance, chemical compositions of the atmosphere, volcanic outbreaks) and shaped by versatile processes in and in-between the spheres of the Earth system on various spatial-temporal scales
Climate change across the globe is driven by changing forcings and shaped by versatile processes in and in-between the spheres of the Earth system on various spatial-temporal scales.Ever since it has been proven that humankind substantially affects the Earth’s climate (Lockwood et al 1991; IPCC 2007, 2013) the necessity of estimating impacts associated with potential future pathways of manhood (describedGlobal Climate Models (GCMs, see e.g. Edwards 2010; von Storch 2010; Müller 2010; Edwards 2011; Taylor et al 2012), which simulate the evolution of climate states in dependence on given forcings, constitute the main tool to analyze potential changes in the climate system
There are two main branches of downscaling techniques: (1) empirical-statistical downscaling (ESD, Fowler et al 2007; von Storch et al 1993; Benestad et al 2008), which relies on transfer-functions derived from observations on the coarse scale of GCMs and regional-scale records as well as (2) dynamical downscaling (DD) making use of Regional Climate Models (RCMs, Rummukainen 2010; Sánchez et al 2015; Nikulin et al 2012; Tang et al 2016; Ozturk et al 2017; Kotlarski et al 2014), which are driven by GCM output at the borders of a limited area (e.g. Europe) and calculate atmospheric processes within this area at a much finer grid
Summary
Climate change across the globe is driven by changing forcings (e.g. solar irradiance, chemical compositions of the atmosphere, volcanic outbreaks) and shaped by versatile processes in and in-between the spheres of the Earth system on various spatial-temporal scales. There are two main branches of downscaling techniques: (1) empirical-statistical downscaling (ESD, Fowler et al 2007; von Storch et al 1993; Benestad et al 2008), which relies on transfer-functions derived from observations on the coarse scale of GCMs and regional-scale records as well as (2) dynamical downscaling (DD) making use of Regional Climate Models (RCMs, Rummukainen 2010; Sánchez et al 2015; Nikulin et al 2012; Tang et al 2016; Ozturk et al 2017; Kotlarski et al 2014), which are driven by GCM output at the borders of a limited area (e.g. Europe) and calculate atmospheric processes within this area at a much finer grid. GCMs’ output, driven by different RCPs are entered into EPISODES in order to derive consistent regional-scale scenarios These scenarios are to be compared to corresponding climate change projections generated via RCMs as well as by other ESD methods
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
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
