Context:Considering the inadequacies in the weather forecasts,it is time for the India Meteorological Department(IMD)to move away from the two stage forecasts system.

About Two stage observation system in IMD

  • First Update:
    • It is taken in April while looking for chances of drought or any other anomaly.
  • Second update:
    • It is taken in late June with a more granular look at how the monsoon will likely distribute over the country and whether danger signs are imminent. 

Drawbacks in current forecasts system

  • Reliance on the age old statistical forecast system
    • The April forecast is a vestige of the agency’s more reliance on the ‘statistical forecast system’.
    • In the current system of statistical forecast values of selected meteorological parameters are recorded until March 31 and permutations of these are computed and compared to the IMD’s archive of weather data.
  • Conservative predictions:
    • Being a part of a hierarchical government set-up, the agency defaults to being conservative. 
    • For example In April last year, IMD said the monsoon would be ‘near normal’ category.But contrary to this opinion India received excess rains, the highest in a quarter century.

Being an agrarian nation, weather forecasts hold utmost importance in India. If IMD incorporates new technological advances in the form of dynamic models in weather prediction,the whole nation would be benefited.

Two recent key changes introduced by IMD 

1.Change in definition of ‘Normal Rainfall’

  • IMD reduced the definition of ‘normal’ rainfall by 1 cm, to 88 cm
  • ‘Normal’ means India will get 100% of its long period average, with a potential 5% error margin. 

2.Updating monsoon onset and arrival dates for many States

  • This was long due and constituted acknowledgement of the accumulated impact that global warming has been having on monsoon patterns, particularly for cities and States. 
  • The monsoon was arriving later in many places, had long weak spells, and lingered longer.

About the weather forecast in India

Statistical models of forecasts:

  • Until about the 2010, the only method employed by the IMD to forecast the monsoon was statistical models. 
  • These statistical models involved identifying climate parameters linked to the performance of the monsoon — for instance the following
    • The sea surface temperature gradient between North Atlantic and North Pacific, 
    • The volume of warm water in the equatorial Pacific and
    • The Eurasian snow cover etc
  • Based on extrapolation of values:Their values in February and March are correlated to values of actual rainfall over a hundred years and then, using statistical techniques, extrapolated to forecast a particular year’s monsoon. 
  • No accuracy in the prediction:These statistical models have proved wrong and the IMD missed its mark on forecasting major droughts and rain-deficits many times particularly 2002, 2004 and 2006. 

Dynamic model of forecasts:

  • Only around 2015 India started testing a dynamical system. 
  • This simulates the weather at a chosen set of locations on a given day the land and ocean temperature, moisture, wind speeds at various heights, etc .
  • Role of supercomputers:Powerful supercomputers calculate how these weather variables will change over days, weeks, months. 
  • It's able to do this by solving physics equations that show how each of these weather variables is related to each other. 
  • The IMD and several private weather agencies are increasingly relying on more sophisticated and high-resolution computer models to give localised forecasts, or warn farmers of changes in weather 10-15 days ahead. 
  • Benefits of the model:
    • Rather than long-range forecasts that only give a broad tenuous picture of the likely performance of the monsoon, these shorter forecasts are far more reliable and help farmers make decisions about sowing. 
    • These models are also useful for anticipating heat-wave or a cold-wave and therefore useful to urban planners and government. 
    • Climate, as well as technological change, allows new weather variables such as surface temperatures from as remote as the southern Indian Ocean and regular updates from the Pacific Ocean to be mapped. 
  • Still not considered entirely reliable:Though meteorological agencies around the world are shifting to such techniques, they still aren’t considered still entirely reliable for forecasting the monsoon. 


Image Source:IMD