Impact of climate change on rainfall over Mumbai using Distribution-based Scaling of Global Climate Model projections

Abstract Study region The study is carried out for Mumbai (18°58′30″ N, 72°49′33″ E). Study focus Future projections provided by general circulation models (GCMs) suggest the probability of occurrence of intense rainfall will change in the future. However, GCM data generally need to be downscaled and bias-corrected for impact studies. Although the domains covered by Regional Climate Models (RCMs) are increasing, statistical downscaling of GCM results is the main alternative in many regions. We applied a Distribution-based Scaling (DBS) procedure, with 1975–2004 as a reference period, for bias-correcting and downscaling daily rainfall data from nine global climate projections. New hydrological insights for the region The evaluation in the reference period showed that the scaled data are able to represent various key statistics. All GCMs were unable to accurately reproduce the southwest monsoon season. Using the transient DBS processed projection data, a comprehensive evaluation of key rainfall statistics was performed for three periods: near (2010–2040), intermediate (2041–2070) and distant future (2071–2099). There is an increase in the total accumulated annual rainfall, ranging from 300 to 500 mm in the ensemble. Also, a clear seasonal shift and delayed onset of the monsoon season evolves in the projections, with increasing and decreasing rainfall in September and June, respectively. A trend analysis using Student's t and Mann–Kendall tests was performed for 2010–2099. A significant positive trend was found for four of the GCM projections.