bccaq.netcdf.wrapper {ClimDown}R Documentation

Wrapper function for the entire BCCAQ downscaling method

Description

BCCAQ is a hybrid downscaling method that combines outputs from Climate Analogues (CA) and quantile mapping at the fine-scale resolution. First, the CA and climate imprint (CI) plus quantile delta mapping (QDM) algorithms are run independently. BCCAQ then combines outputs from the two by taking the daily QDM outputs at each fine-scale grid point and reordering them within a given month according to the daily CA ranks, i.e., using a form of Empirical Copula Coupling.

The combination mitigates some potential issues with the separate algorithms. First, because the optimal weights used to combine the analogues in BCCA are derived on a day-by-day basis, without reference to the full historical data set, the algorithm may fail to reproduce long-term trends from the climate model. Second, the CI/QDM bias correction step fixes precipitation "drizzle" and other residual biases caused by the linear combination of daily fields from CA. Third, reordering data for each fine-scale grid point within a month effectively breaks the overly smooth representation of sub grid-scale spatial variability inherited from CI/QDM, thereby resulting in a more accurate representation of event-scale spatial gradients; this also prevents the downscaled outputs from drifting too far from the climate model's long-term trend.

Usage

bccaq.netcdf.wrapper(gcm.file, obs.file, out.file, varname = "tasmax")

Arguments

gcm.file

Filename of GCM simulations in NetCDF format

obs.file

Filename of high-res gridded historical observations

out.file

The file to create (or overwrite) with the final BCCAQ NetCDF output

varname

Name of the NetCDF variable to downscale (e.g. 'tasmax')

References

Werner, A. T., & Cannon, A. J. (2016). Hydrologic extremes - an intercomparison of multiple gridded statistical downscaling methods. Hydrology and Earth System Sciences, 20(4), 1483-1508. doi: 10.5194/hess-20-1483-2016


[Package ClimDown version 1.0.2 Index]