IMD’s forecast conundrum: Fine data, flawed interpretation

As meteorological puzzles persist, the IMD grapples with a paradox: precise data but perplexing forecasts.

On January 7, the India Meteorological Department (IMD) issued a three-day forecast warning of “moderate rainfall across central and northwest India” on January 8 and 9. It issued an orange alert for Haryana, Punjab, Delhi, Madhya Pradesh, and parts of Rajasthan, predicting rain and thunderstorms in these states. January 8th came and went with no rain in most of these states. In addition for January 9 (although, complying with the January 8 disaster, IMD revised its forecast to “a possibility of isolated showers”).

The issue, according to experts, is not technology or models, but rather weathermen’s interpretation of data.

A senior weather department official defended IMD’s forecast, saying that a cyclonic circulation was developing over the region at the time the warning was issued.

However, this was not the first occasion that India’s weather bureau miscalculated its forecast. At least 10 people died in two days due to “extremely heavy” rainfall that inundated Tamil Nadu last month, an event that IMD was unable to predict with any degree of accuracy. In press conferences, state representatives discussed how an agency warning could have led to greater readiness and less damage.

Without mentioning any particular incident, IMD Director General M Mohapatra stated that no weather agency can be perfect all the time and that minor errors in the forecasting of weather developments, like non-seasonal rains, shouldn’t be written off as “mistakes.”

He continued, “Our forecasts have significantly improved over the last few years, and we are striving to make it better in the coming years.”

According to documents made public by IMD, the organization is nearly finished developing a high-density meso network and a high-resolution modelling framework for major cities to use in early weather and air pollution forecasting and warning systems under the auspices of urban meteorological services.

A chain of automated weather stations that monitor the weather within a 10- to 1,000-kilometer radius is called a mesonet or mesoscale network. Compared to the synoptic scale, which uses measurements at ranges greater than 1,000 km, it is finer. Furthermore, according to the documents, the Ministry of Earth Sciences plans to add 33 more Doppler radars to its network by 2025, guaranteeing complete nationwide observational coverage.

The models used by IMD have improved and are now comparable to the technology used in the U.S., UK, and Japan, which are recognised for producing the most accurate weather forecasts worldwide, according to M Rajeevan, a former secretary of the Ministry of Earth Sciences. However, he made the point that weathermen’s interpretation of data and satellite imagery is a crucial component of forecasting and one in which IMD lags.

IMD not only possesses its models but also has access to the latest models available. Over the past few years, IMD’s models’ resolution has also greatly improved. A precise weather prediction relies on two factors: the model and the forecasters’ interpretation of it. Models are merely instruments, and without the ability to analyze numerous satellite photos, and radar readings, and infer clues from the models, our forecasts will remain inaccurate. That appears to be the situation.

Long-term, seasonal forecasts appear to be IMD’s weak point, despite some progress in short- and medium-range forecasting, according to Rajeevan.

Making seasonal forecasts is more challenging. Predictability is a term used to describe the degree of accuracy with which we can forecast. It is higher for three-day and five-day forecasts than it is for seasonal forecasts.

In agreement, Mahesh Palawat, vice-president (meteorology and climate change) at the for-profit forecasting company Skymet stated that training weathermen, advancing technology, and expanding the number of ground stations can all help to improve forecasts.

The IMD official that was first mentioned thinks that comparing the accuracy of forecasts from weather offices in the U.S., UK, and India is pointless. He said that because India has a tropical climate, it is more difficult to forecast the weather there. He acknowledged there was space for improvement but added that the climate crisis is another factor that presents a problem.

Tropical climates are harder to predict because they are more unpredictable than those in the U.S. and the UK, which have more predictable weather patterns. Forecasters’ jobs will become more challenging as a result of climate change because future weather will be unpredictable and deviate from historical patterns. Nevertheless, we still need to adjust to these new challenges.