A comparison of general circulation models and their application to temperature change assessments in a high-latitude agricultural area in northeastern China

The two main focuses of this study are a comparison of the general circulation models (GCMs) from Phase 5 of the Coupled Model Inter-Comparison Project (CMIP5) and an assessment of the surface air temperature under multiple climate scenarios in a high middle latitude area of China. In the past 55 years temperatures in this area have shown an obvious upward trend (a rise of 1.50 degrees C), and another important change during this time period was a significant alteration in tillage practices that occurred in 1986. Using methods and tools such as average deviation, the Taylor figure and the space techniques rating (SS), time sequence related coefficient, and the M2 index, a comprehensive spatial-temporal assessment was performed based on the CMIP5 models. The simulations provided by the models had certain common features, but there were also significant differences. The three best models (CanCM4, INMCM4, and IPSL-CM5A-MR) have a common characteristic: the institutions where they were developed are located at latitudes that are similar to or higher than the latitude of the study area. Future climate changes were analyzed by simulating a representative concentration pathway 4.5/8.5 (RCP4.5/RCP8.5) of emission scenarios with a multi-model ensemble. The temperatures under the RCP4.5 and RCP8.5 scenarios have a certain upward trend, with increases of 2.24 and 5.44 degrees C, respectively. From a spatial perspective, the distributions of the temperature change trend showed a southwest to northeast step increase under both scenarios, but the warming trend in the area of each lattice point under the RCP4.5 scenario is much lower than that of the RCP8.5 scenario. There are no obvious changes in the spatial distribution of the accumulated intensity and frequency of the regional air temperature in the three periods (2016-2035, 2036-2065, and 2066-2095) under the two scenarios.