Impact of Volcanic Aerosol Forcing on the Decadal Prediction of Air Temperature

The predictability of sea surface temperature (SST), surface air temperature and rainfall in a decadal prediction system (DPS) stem from both the initial-conditions and the external forcing (natural as well as anthropogenic). In the present study we used a DPS based on IITM-ESM to understand its sensitivity to volcanic forcing (VF). In real-world decadal forecasting, the unexpected rapid changes in the forcing such as volcanic eruptions introduce immense uncertainty into DPS as such events can not be predicted in advance. Volcanic eruptions increase the stratospheric aerosol content, leading to a reduction in global mean surface temperature and impact precipitation from months to years. Thus understanding volcanic eruption responses is vital for improving the decadal forecast systems. Only a limited number of studies have addressed the influence of VF on DPS but they have highlighted the significance of VF on the decadal prediction. According to previous research, volcanic aerosol implementation in DP notably enhanced the hindcast skill for near-surface air temperatures across various regions up to five years of prediction. It is important to note that the model initialized around any volcanic events can include VF without much discrepancy.

The influence of major volcanic eruptions on DP is investigated in this study using the IITM-DPS. Atmospheric greenhouse gas concentrations, solar forcing, ozone forcing, and stratospheric aerosol data are implemented in DPS as per CMIP6 protocol. Additionally, OPAC (Optical Properties of Aerosols and Clouds) climatological tropospheric and stratospheric aerosol forcing data is also used. To elucidate and evaluate the volcanic eruptions' potential skill towards climate predictability, recent major events such as El Chichón (1982) and Mount Pinatubo (1991) are considered for sensitivity experiments with model initialized on 1^st November of 1981 and 1990 respectively.

Three experiments are performed: Exp1 in which VF is absent. In Exp2, a coarser resolution VF is incorporated as in Sato et al (1993). Whereas in Exp3, a high resolution stratospheric volcanic aerosol forcing is implemented in accordance with CMIP6 DCPP baseline. The impact of volcanic eruptions is then determined by comparing Exp2 and Exp3 with respect to Exp1.

Results show that the IITM-DPS have successfully simulated the response to VF in both Exp2 and Exp3 for both the volcanic eruptions. Exp3 showed improved skill in simulating global mean temperature and aerosol optical depth at 550nm. Model experiments have successfully simulated global mean temperature warming (cooling) in stratosphere (troposphere) after the volcanic eruption events. Although there are some differences in aerosol spatiotemporal evolution between the Exp2 and Exp3, these sensitivity experiments highlight the significance of VF in the DPS.