Short term climate variability project. Estimating the potential for prediction of short term climate variations

摘要

Short-term climate variability is usually defined to span time scales from a month up to several years. The work described here focuses on predictability on the longer end of these time scales, i.e. a season up to a few years. Climate variability on these time scales obviously affects human activity and even a modest predictive skill may be of importance. The purpose of the work described here is to analyze and possibly quantify the predictive skill on these time scales. The skill may originate from sources external to the climate system such as an anthropogenic change of the composition of the atmosphere and variations in the solar activity or from climate variations internal to the climate system. Only the last type of variations is considered here. Skill originating from the interactions of the atmosphere with the lower boundary of the atmosphere (mainly related to the large thermal inertia of the oceans) will be considered as well as skill related to variability purely internal to the atmosphere, however, with most emphasis on the former. The work includes analysis of both observed data and of data generated by atmospheric GCMs. (au) 12 refs.