On the correspondence between short- and long-timescale systematic errors in the TAMIP and AMIP

The correspondence between short- and long-term systematic errors in climate models from the transpose-AMIP (TAMIP, short-term hindcasts) and AMIP (long-term free running) archives is systematically examined with a focus on the precipitation, clouds and radiation. The data from TAMIP is based on 16 5-day hindcast ensembles from the tamip200907 experiment during YOTC, and the data from AMIP is based on the July-August mean of 1979-2008. Our results suggest that most systematic errors apparent in the long-term climate runs, particularly those associated with moist processes, also appear in the hindcasts in all the climate models (CAM4, CAM5, CNRM5, HadGEN2-A, IPSL, and MIROC5). The errors, especially in CAM4/5, and MIROC5, grow with the hindcast lead time and typically saturate after few days of hindcasts with amplitudes comparable to the climate errors. Examples are excessive precipitation in much of the tropics and overestimate of net shortwave absorbed radiation in the stratocumulus cloud decks over the eastern subtropical ocean and the Southern Ocean at about 60°S. This suggests that these systematic errors likely resulted from model parameterizations since large-scale flows remain close to observations in the first few days of the hindcasts. We will also discuss possible issues of initial spin-up and ensemble members for hindcast experiments in this presentation. (This work performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344.)