Counteracting effect of future Antarctic sea-ice loss on projected increases of summer Monsoon rainfall

South-Asian society and economy are heavily dependent on the seasonal rainfall during the Indian Summer Monsoon (ISM). This rainfall has large spatio-temporal variability, the main pattern of which is expressed in an alternation of "break" and "active" phases. Break phases imply dry conditions over India's main agricultural areas. When sufficiently prolonged or repeated in the course of the season, they are associated with droughts. Food production is highly sensitive to both dry and wet rainfall extremes. Under anthropogenic climate change, model-derived projections generally predict an increase in seasonal-mean rainfall totals along with a reduction in the number of rainy days. However, the associated spatio-temporal patterns of change in mean rainfall are very uncertain. The transition between break and active phases of the ISM corresponds with a large meridional excursion of the precipitation maximum, reflecting a tendency for rainfall to occur over preferred geographical loci: one over the Gulf of Bengal and the Indo-Gangetic plains; and one over the Equatorial Indian Ocean. These map onto oscillations of the zonal-mean Inter-Tropical Convergence Zone, suggesting a link with the global circulation. As a result, the position and intensity of the ITCZ is sensitive to changes in extratropical heat sources. Antarctic sea-ice loss will result in an additional heat source in the SH of about 0.3 PW in austral winter. Hence, we investigate whether a substantial ISM rainfall response may be induced by large-scale interhemispheric heating anomalies. The focus is on future Antarctic sea-ice loss and their impact on the global atmospheric circulation. We are placing a strong emphasis on collaborative research and in particular on both formal and research training with our Indian partners. Our results suggest that the 10 percent drop in Antarctic sea ice in 2016, compared with the mean since 1979 could be due to increased air temperatures near the sea ice edge, strong winds, a shift in the Southern Annular Mode. Bracegirdle et al. (2018) shows that the future changes in the strength of the westerly jet in the Southern Hemisphere for the RCP8.5 scenario could be associated with sea ice area in Antarctica. To test this hypothesis, we have performed and evaluated idealised model simulations with reduced sea-ice. The result partially support our hypothesis. In one of the models we use (ECHAM6), sea-ice reduction results in a shift of the Hadley circulation, with the expected impact on the global ITCZ and on precipitation in the West African Monsoon. However, precipitation in the ISM is not affected in the same way. This is likely related with the activity of Monson depressions and their mid-latitude drivers (Rastogi et al., 2018). Moreover, no significant global circulation response is produced during the ISM season in experiments with the other model (CAM5). We tested the hypothesis that this may depend on seasonality (Day et al., 2018). To do so, in collaboration with Dr. V. Sinclair at the University of Helsinki we compared idealised sensitivity tests with Aquaplanet configurations of CAM and IFS (the dynamical core underlying the ECHAM6 model). These confirm that the sensitivity of the Hadley Cell to perturbations in the storm track is indeed model dependent, in a manner consistent with the Antarctic sea-ice pertrubation experiments (cf. T. Toniazzo's seminar at the Antarktisseminaret in May 2018). We are now evaluating the mechanisms that operate in both models, with a detailed analysis of extratropical storms and their effect on atmospheric transport under different conditions. In parallel to this line of investigation, with our Indian colleagues at TERI we have carried out a detailed analysis of active and break phases of the ISM in observations and in model simulations. We have developed a compositing technique based on equatorial winds for circulation and precipitation anomalies of the ISM region. This work is leading to significant results, which highlight the role of convectively coupled equatorial modes of variability, while the association with Monsoon depressions that is evident in precipitation-based indices is much reduced. Our analysis is thus able to separate two independent mechanisms for the variability in ISM rainfall, and show the effect of teleconections for one of them. This lends a promising interpretative framework to the indipendent results cited above from modelling work, and a potential explanatory argument for inter-model differences in the sensitivity of ISM precipitation response to extratropical teleconnections. This cooperative analysis effort has involved significant competence building with our Indian partners, and has led, among other activities, to a C-ICE Research School which will took place in New Delhi in August 2019, with the participation of all project partners and high-profile Indian and international researchers in lectures and seminars.

Outcomes The project had a strong ambition to achieve a productive research collaboration across international istitutions (Europe and India) and across specialised areas of research (extratropical dynamics and Monsoons). It is has partial succeeded, more so on the international side, including strong dissemination and training aspects, but also on the scientific side. In particular it has demonstrated the importance of future sea-ice changes on the global circulation, while highlighting and quantifying the large uncertainties associated with a numerical modelling framework. Impacts None to date. Society does not seem terribily responsive to science right now.

The potential for an influence of future Antarctic sea-ice loss on the spatio-temporal distribution of rainfall in the South-West Asian Monsoon is investigated. Two main hypothesis drive this proposal. The first is that the adjustment of the Hadley circulation to changes in extratropical heat fluxes results in a tendency for tropical precipitation to be located further south, towards the Equator, in boreal summer. The second is that the general tendency for a southward displacement of tropical precipitation in boreal summer affects rainfall over India. The planned modelling and analysis work is structured around these hypotheses. We will investigate how mid-latitude storms in the southern Hemisphere respond to a poleward retreat of the sea-ice edge; how changes in mid-latitude storm activity affect the Hadley circulation; how such an adjustment impacts the meridional distribution of zonal-mean rainfall; and finally how the zonal-mean signal is reflected in Monsoon rainfall. The first stage in the project will be an analysis of existing model simulations. We will then perform sensitivity tests with different models to isolate the effect of Antarctic sea-ice. Finally, in two separate work-packages we will analyse the mid-latitude and tropical dynamics of the response, and test the findings by means of additional experiments with idealised model configurations. Throughout the project, a strong emphasis will be placed on collaborative research and in particular on both formal and research training with the Indian partners.