The geometrical structure of the upper-tropospheric African–Asian jet and its response to global warming in the CMIP5 Models

In this study, a set of models from the phase 5 of the Coupled Model Intercomparison Project (CMIP5) is used to examine the simulation of the upper-tropospheric subtropical African–Asian jet and its response to global warming. The ERA-Interim re-analysis dataset is used here to assess the model biases in representing the seasonal-mean jet features in the historical period (1980–2005). This study analyzes the geometrical parameters of the jet including “latitude”, “speed” and “width” in each season and for two separate sectors of the jet region: “North Africa” and “Southwest Asia”, which is briefly named “African” and “Asian” hereafter.
The main features of the observed seasonal cycle of the jet in the re-analysis data is well captured in ensemble multimodel mean historical simulations: jet latitude increase (decrease) from cold (warm) to warm (cold) season and vice versa are correctly simulated for jet speed and width. In addition, in all seasons, the jet latitude and speed is greater in Asian sector than the African except for springtime jet speed. Despite the large inter-model spread in the historical jet simulations, the models do not show large systematic biases in most cases (seasons). However, systematic biases in each of the geometrical jet indices are found in some seasons: most models exhibit equatorward jet biases in summertime and wintertime of the African sector (about 1.8° and 0.9° of latitude respectively, in multimodel mean), positive biases in jet width in summertime Asia (0.9° in multimodel mean), negative biases in jet speed in summertime Asia and wintertime of the African sector (approximately 2.9 m/s) and positive jet speed biases in autumntime of the African sector (1.8 m/s). There is large spread across the models in the historical jet simulations and finding the sources of this spread and the model biases is a significant challenge that should be addressed in future works.
In almost all seasons and for all of the geometrical jet indices, the multimodel mean jet response to climate change is stronger in RCP8.5 than RCP4.5 integrations. Robustness and the quantitative value of the multimodel mean jet response in each of the jet indices vary among different seasons and sectors. In winter months, we found no robust response in any of the geometrical jet indices in African or Asian sector except for a slight and relatively robust increase in jet width (0.2° of latitude in RCP8.5) in African sector. However, in other seasons, we found robust multimodel mean changes in jet indices between the historical period and the end of twenty first century (2076–2099) in the RCP8.5 scenario: In spring, models predict a robust increase in jet width of about 0.5° and 0.2° of latitude in African and Asian sectors, respectively, and also a robust increase in jet speed of 1.1 m/s for Asian sector. In summer, in the African sector, the jet speed is found to be decreased (0.7 m/s), whereas in the Asian sector, jet speed will increase (0.4 m/s), and it will move equatorward by 0.8° of latitude.