Reconsidering the Quality and Utility of Downscaling

Abstract

Dynamical downscaling (DDS) is performed using regional climate models (RCMs) with global atmospheric states as the input, but there is no consensus among researchers on how to define and estimate the resolvable scale of the various climatic variables obtained by DDS. Sources of RCM uncertainties, including both internal model and intermodel variability, have been assessed by performing ensemble simulations and model intercomparisons, sometimes under the controversial assumption that model bias is independent of the climatic state. Compared with low-resolution global climate simulations, DDS can add value in several ways. For example, because they consider high-resolution topographic data, RCMs can often capture mesoscale phenomena and can better represent climate dynamics. Another downscaling method, empirical statistical downscaling (ESD), is complementary to DDS because it is based on a different philosophy (i.e., sources of information) and on a mostly different set of assumptions. More collaboration and communication should be encouraged among those who develop models, those who use models and perform downscaling, those who use downscaling data, and those who make decisions

Changes in Winter Warming Events in the Nordic Arctic Region

“…Therefore, the coarse-resolution RCM data need to be statistically processed to local stations using a bias-correction method before carrying out the climate analysis (e.g., Takayabu et al 2015). On one hand, this bias correction unfortunately leads to loss of physical consistency between variables, but the present-day climate is much closer to reality, which for impact studies has very high priority (Sorteberg et al 2014).…”

Decomposition of the large-scale atmospheric state driving downscaling: a perspective on dynamical downscaling for regional climate study

“…However, global simulations are usually too coarse to represent regional-scale atmospheric variability. In order to represent regional-scale variability, dynamical downscaling (DS) with a regional model driven by a large-scale state, i.e., initial and lateral boundary conditions generated from global simulations or reanalyses, has been widely used (e.g., Dickenson et al 1989;Giorgi and Bates 1989;Mearns et al 1995;Rummukainen et al 2015;Takayabu et al 2016). DS has several advantages over coarser global *Correspondence: s-nishizawa@riken.jp 1 RIKEN Advanced Institute for Computational Science, 7-1-26 Minatojima-minami-machi, Chuo-ku, Kobe, Hyogo 650-0047, Japan Full list of author information is available at the end of the article simulations or reanalyses (e.g., Jones et al 1995;Walsh and McGregor 1995;Giorgi and Shields 1999), e.g., better representation of surface conditions such as topography and land use, and utilization of more sophisticated physical process models such as a cloud microphysics model that avoids the need for cumulus parameterization.…”

A classification algorithm for selective dynamical downscaling of precipitation extremes

“…Veljovic et al, 2010). The assumption of stationarity -that predictor-predictand relationships will remain unchanged in a future climate -in RCM parametrizations and statistical downscaling methods may also not be valid (Takayabu et al, 2016), lowering confidence in projections. Statistical and dynamical downscaling both produce climate change signals that are, to varying degrees, influenced by the climate change signal of the parent GCM.…”

“…If the GCM has an incorrect climate-change signal this may be inherited without meaningful modification. Takayabu et al (2016) further discuss different facets of the statistical and dynamical downscaling approaches, additionally explaining that the approaches are complementary and can be combined, rather than being treated as mutually exclusive alternatives.…”