The success of weather forecasts depend not just on model accuracy but also on how they are communicated to stakeholders. Forecast skills have to combine with effective dissemination cemented by mutual trust and understanding, consistency, and effective coordination, say experts.

Calculations, especially in earth sciences, should always include an estimate of the possible error, the existence of some error being a basic assumption. According to experts, this applies equally to weather forecasting and considerations of future climates. Weather and climate forecasting is perhaps the most demanding computationally, yet the least certain of scientific trades.

Experts are of the view that binary predictions (just rain or shine) are better avoided, and the related ‘black or white’ science dispensed with. Forecasters need to be explicit about the ‘uncertainty’ of weather, they argue.

“Typically, as human beings, we think in a deterministic way. We may have to start thinking a bit differently,” said Kevin Petty, Director, Science and Forecast Operations at The Weather Company, an IBM Business.

“We’re working on techniques that warn people about uncertainty where we can provide a probability of a particular event happening,” he told BusinessLine in a recent interaction. “To enable decision-making based on a probability, one must know that there’s a certain element of uncertainty built into it, rather than rely on a deterministic outcome.”

Range of variability

Forecasters need to provide a range of probability to make critical decisions. “How we transfer these data to the people is also important. We can have the best models and the best forecasts. But, we also have to look at how do we change people’s behaviour when we tell them about a cyclone and they choose to do nothing about it,” he said.

“Take the example of stubble burning that pollutes the atmosphere in North India. Howe do we change the mindsets of farmers who don’t factor in the extent pollution?” he asked.

India Met Department (IMD) agrees that weather forecasts will be much more beneficial to the public, decision-makers and end users if it contains the uncertainty element in addition to forecasts. The IMD feels that It is advisable for forecasters to reveal the margin of uncertainty since it is better to present an uncertain weather forecast than one of misleading certainty.