Seasonal Forecasting in South Asia: A Review of the Current Status

South Asia has a diverse climate which is dominated by a monsoon system. Rainfall associated with the South Asian monsoon experiences considerable spatiotemporal variability, which has profound socioeconomic impacts on the region. Understanding the factors that drive this variability is critical to advancing seasonal forecasting in South Asia.

Slowly-evolving climate drivers provide a source of predictability on seasonal timescales. Identifying how these drivers affect the South Asian climate is critical to advancing its predictability. Various metrics have been established to define the relationship between summer monsoon rainfall and the large-scale circulation patterns driving the monsoon system, as well as the timing of the southwest monsoon onset.

The key seasonal drivers of the South Asian climate are the El Niño Southern Oscillation (ENSO) and the Indian Ocean Dipole (IOD). These are irregular coupled atmosphere-ocean oscillations which drive variability in monsoon rainfall from year-to-year. In general, the warm (cool) phase of ENSO results in a drier (wetter) monsoon season, whilst a positive (negative) phase of the IOD results in wetter (drier) monsoon season. Variability also exists within the season, driven by intraseasonal oscillations such as the Madden-Julian Oscillation.

Seasonal forecasts predict how average temperature or precipitation in an upcoming season will compare to the past climatology of a region. For example, a forecast can give you information on whether the season is likely to be wetter or drier than the climatological average.

Seasonal forecasts can be made using statistical or dynamical approaches, or a combination of the two. Statistical methods use observational records to identify relationships between local weather patterns and key climate drivers. Whereas dynamical methods use coupled atmosphere-ocean models to represent all complex interactions within the climate system.

Higher skill has been demonstrated using coupled model systems, which exceed the skill of statistical methods. However, further development is required to improve the skill of dynamical models at seasonal forecast timescales. For example, predicting the specific date of the monsoon onset is not currently possible for seasonal dynamical models.

Seasonal forecast experts produce a consensus-based seasonal forecast for the region during the South Asian Climate Outlook Forum (SASCOF). The forum is also an opportunity for interaction between national, regional and international users and providers of climate services. The concept of SASCOF is then extended to the national level through National Climate Outlook Forums.