A new approach to combine climate change projections by ordered weighting averaging operator; applications to northwestern provinces of Iran

Abstract

The projections of the climate change by using General Climate Models (GCMs) are uncertain. Hence, combining the results of GCMs is now an effective solution to tackle this uncertainty. To evaluate the performance of GCMs, a new measure based on the similarity of the projections is defined. In defining this measure the Ordered Weighted Averaging (OWA) approach is used. The relative weights of the GCMs projections in different stations, to be aggregated by the OWA operator, are obtained by regular increasing monotone fuzzy quantifiers, which model the risk preferences of the decision maker. To show the effectiveness of the approach, climate change in the northwestern provinces of Iran is studied by using the data of 15 synoptic stations. The weather generator of LARS-WG is used to downscale the GCMs under three emission scenarios (A2, A1B and B1) for the period 2011 to 2030. The combined results, by using the similarity values, indicate a −0.1°C to +4.5°C change in temperature in the region. Precipitation is expected to increase in summer and fall. Changes in wintry precipitation depend on the location; however the precipitation in spring would have a medium change. The results of this study show the usefulness of OWA operator, which considers the risk attitudes of the decision maker. This approach could help water and environmental managers to tackle the climate uncertainties.