Climate change impacts on tropical cyclones of the Arabian Sea: Projections and uncertainty investigations. (25th January 2022)

• Record Type: Journal Article
• Authors: Ranji, Zahra; Zarifsanayei, Amin Reza; Cartwright, Nick; Soltanpour, Mohsen
• Is Part Of: International journal of climatology. Volume 42:Number 10(2022)
• Page Start: 5121
• Page End: 5141
• Publication Date: 2022-01-25
• Subjects: climate models -- pseudo global warming -- tropical cyclones -- WRF; Climatology -- Periodicals; Climat -- Périodiques; Climatologie -- Périodiques; 551.605
• DOI: 10.1002/joc.7523
• Languages: English
• ISSNs: 0899-8418

Abstract

Abstract: This study investigates some of the uncertainties sources associated with the pseudo global warming (PGW) approach which was employed to project future patterns of tropical cyclones (TCs) over the Arabian Sea (AS). First, the climate variables controlling the patterns of tropical cyclones were extracted from reanalysis datasets of ERA5, ERAI, CFSR, and NCEP/NCAR. Then, each dataset was evaluated against long‐term measurements to identify the best‐performing reanalysis dataset. ERA5 showed the best performance for most of the variables. Outputs of 20 CMIP5 global climate models (GCMs) were then evaluated against the ERA5 data resulting in an ensemble of the best performing GCMs. A PGW framework was then used to project the changes in patterns of three significant historical cyclones: Gonu, Phet, and Ashobaa. In doing so, the signals of future climate variables were extracted from the GCMs ensemble to modify the initial and boundary conditions of the WRF model which was previously tuned for reproducing the historical TCs. Different tests were conducted to address the sources of uncertainty in the PGW approach, including the selection of the climate variables contributing to the computation of the signals, the selection of GCMs, and the spatial variation of signals. A considerable sensitivity of the projected track and intensity of TCs to the choice of GCMs was observed, acknowledging the importance of GCMs evaluation before calculating the signals. Moreover, it was found that among all variables, signals of sea surface temperature and air temperature have major effects on the cyclone's track and intensity. Apart from that, when the signals were applied to the domain of the WRF model uniformly, compared to applying spatially varying signals, different tracks and intensities for future TCs were also observed. Overall, the findings of this paper challenge the reliability of the projected changes in TCs patterns obtained from PGW. Abstract : ERA5 and CNRM show the overall best performance to reproduce the climate variables. GCMs with higher resolution do not necessarily show a better performance. GCMs should be evaluated before application of PGW. SST and AT have major effects on the TC tracks and intensities.