Differential Credibility Assessment (DCA) for Statistical Downscaling
![Results of daily min temperature (Tmin) downscaling in the frequency of (a,b) tropical nights and (c,d) frost days for the contemporary climate from GFDL or MPI. (e–h) Difference in frequency per year in the future (2075–99) minus the current (1979–2005) of these CLIMDEX indices. Symbol size scales linearly with frequency of occurrence [in (a–d)] or the difference [future minus current; in (e–h)]. Colors denote two aspects of DCA (outer & inner circles): red = low, yellow = moderate, green = high.](https://eesm.science.energy.gov/system/files/private/highlights/images/a.png)
Climate science is increasingly using (i) ensembles of climate projections from multiple models derived using different assumptions and/or scenarios and (ii) process-oriented diagnostics of model fidelity. However, while efforts to assign differential credibility to projections from different numerical models are rapidly advancing, DCA has yet to be embraced by the statistical downscaling community. A framework to quantify and depict the credibility of statistically downscaled model output is presented and demonstrated that assigns credibility based on 1) the reproduction of relevant large-scale predictors by the GCMs (i.e., fraction of regression beta weights derived from predictors that are well reproduced) and 2) the degree of variance in the observations reproduced in the downscaled series following application of a new variance inflation technique.
The differential credibility assessment framework demonstrated here is easy to use and flexible. It can be applied as is to inform decision-makers about projection confidence and/or can be extended to include other components of the transfer functions, and/or used to weight members of a statistically downscaled ensemble.
This work also addresses a persistent problem in the statistical downscaling community; under-dispersion in the downscaled series relative to observations. Methods for artificially increasing the variability include randomization and variance inflation. However, variance inflation increases the mean square error between the observations and adjusted predictands, while randomization changes the temporal autocorrelation of the downscaled predictand. A further methodological advance presented in this research ensures the statistical downscaling models more accurately represent variability in the predictand (response variable) and is further designed such that changes in the temporal autocorrelation of predictors are propagated to the downscaled variable.
To aid societal preparedness for climate change it is critical to develop tools to describe projection credibility in different regions and metrics derived using different GCMs. This research is an initial step toward application of DCA to statistical downscaling. It provides input to assessments of the credibility with which we can extrapolate contemporary credibility or realism into scenarios of future climate conditions developed using statistical transfer functions conditioned using data from the contemporary climate. While there is no definitive basis for assigning credibility to such projections process-level DCA offers one approach to selecting which are the best candidates for extrapolation.