When assessing the impacts of climate change in hydro-climatological studies, researchers often compare "GCM future run with historical observed" or "GCM future run with GCM control run" ? Can somebody give some insights on a comparison of these approaches?
Dear Madusanka
the reason for including the historical observed information is that the GCM present day runs will not match observed climate perfectly, and may need some bias correction for climate impacts where the (absolute) baseline state is important. So it is most common to calculate the future climate change as the difference between GCM future and GCM present day, and apply this to the observed present day values. This usually involves simple additive approaches (e.g. for temperature) or mutiplicative approaches (eg. for precipitation). But there is a range of bias correction techniques available. I would suggest you look at these papers
Hempel, S., Frieler, K., Warszawski, L., Schewe, J., and Piontek, F.: A trend-preserving bias correction – the ISI-MIP approach, Earth Syst. Dynam., 4, 219–236, doi: 10.5194/esd-4-219-2013 , 2013.
Ehret, U., Zehe, E., Wulfmeyer, V., Warrach-Sagi, K., and Liebert, J.: HESS Opinions “Should we apply bias correction to global and regional climate model data?”, Hydrol. Earth Syst. Sci., 16, 3391–3404, doi: 10.5194/hess-16-3391-2012
, 2012.
Gosling, S. N. and Arnell, N. W.: Simulating current global river runoff with a global hydrological model: model revisions, validation, and sensitivity analysis, Hydrol. Process., 25, 1129–1145,doi:10.1002/hyp.7727, 2011.
best wishes
Pete
You may fine this paper relevant to answer your question.
https://www.researchgate.net/publication/297725252
Article Comparison of different statistical downscaling methods for ...
If interested in mean changes and want to cover full spectrum of future projections (Cold-Dry, Cold-Wet, Warm-Dry, Warm-Wet), use envelope-based approach (four corners with combinations of 10th and 90th percentiles of precipitation and temperature)
If interested in extremes, then use GCM-history vs. GCM-future. (Based on ETCCDI, e.g., CDD, CSDI, CWD, R99pTOT, WSDI, etc.) --- Rclimdex can be used to calculate ETCCDI
If you want to see whether GCMs can represent observed climate at present and near past then you need to compare GCM-history and Observed-history. Statistical indices can be used for this purpose.
If interested in Bias Correction, CMhyd is a handy tool to do that.