Why IMD flagged New Delhi's Mungeshpur record-breaking temperature reading as a “possible error”

“While maximum temperatures across most weather stations in New Delhi on May 29 showed an increase of 0.5-1 degree Celsius from the previous day due to heatwaves, the temperature recorded at Mungeshpur showed a 3-degree Celsius spike from the previous day. We are assessing the reasons behind this difference and will conduct a proper assessment of the Mungeshpur weather station,” said Mrutunjay Mohapatra, director general (DG), IMD.

In this particular case, the allegedly erroneous reading of 52 degrees Celsius was flagged as an anomaly. “Such anomalies are not uncommon and are typically identified and corrected through cross-verification with nearby weather stations and historical data trends,” said another IMD official requesting anonymity.

On June 1, Mohapatra acknowledged an error in the functioning of the Mungeshpur weather station. "The maximum temperature reported at Mungeshpur is not correct due to malfunctioning of the sensor. Necessary remedial measures are being taken by IMD to avoid the repetition of such errors. The data from the sensor should not be considered. We will be replacing the sensor in a few days. We will also examine AWSs, course corrections, if any will be carried out on an individual basis," he added.

How IMD records temperature data

The weather bureau records temperature data using two primary methods: manual and automatic observations, Mohapatra explained. Manual weather stations employ traditional thermometers to gauge temperatures. “These stations typically involve human observers who record the data using instruments such as maximum and minimum thermometers, hygrometers, and rain gauges. This method relies on trained personnel to ensure accuracy and consistency in data collection,” he said.

Automatic weather stations (AWS), such as the one at Mungeshpur, use sensors to measure meteorological parameters, including temperature, humidity, wind speed, and direction. These sensors are calibrated regularly to ensure accuracy, but occasional errors can still occur due to technical faults or external factors like local heat sources or infrastructural changes​​.

The main difference between the two methods lies in the data acquisition process, the IMD DG added. “Manual stations require human intervention and are subject to observational errors, while AWS operates autonomously, providing real-time data, which is important for timely weather forecasts and alerts​. However, even for an AWS, there is always a probability for inaccuracy,” added Mohapatra.

Another important aspect is the difference between maximum (day temperatures) and minimum (night temperatures). The maximum temperature indicates the highest temperature reached during the day, while the minimum temperature represents the lowest temperature at night.

Experts weigh in on the Mungeshpur temperature anomaly

Independent weather experts said AWS’ can be a boon for decision-makers, citizens and other stakeholders if installed at appropriate locations and provided their observations pass through high standards of quality control.

“If this doesn’t happen, faulty AWS data can lead to panic, which could be dangerous during heatwaves and heavy rainfall. Mungeshpur’s incident is not unique since similarly unusually high temperatures used to be recorded at Bhira in western Maharashtra a few years ago,” said Akshay Deoras, a researcher at the National Centre for Atmospheric Science, University of Reading, UK.

Another UK-based independent weather expert Athreya Shetty said manual observations are conducted only at main meteorological centres. “For most of India, AWS reading becomes the only source for temperature data, especially outside urban areas. Since India does not have a vast AWS network, it is far from enough to capture the hyperlocal spatial variation. As a result, many unofficial readings go unnoticed,” he said.

The problem with Mungeshpur arose because of uncalibrated AWS readings. The reason for this, Shetty explained, is areas within a few kilometres of each other may record vastly different temperatures based on surrounding factors like buildings, concrete structures, or trees. “But if the official manual observation is only located in one of these areas, then nearby areas that might have recorded different readings will be ignored. The whole incident could have been avoided if the AWS had been regularly calibrated,” Shetty added.

Many of the IMD's AWS across India are either uncalibrated, not maintained regularly or suffer from over or under-reporting temperature, said Shetty. “While independent forecasters can spot this error and offset the values with the closest nearby readings before communicating them, the general public would not be aware of this aspect, and will likely take readings at face value. And by the time IMD publishes their official reading, it's too late. This is what leads to the spread of misinformation like May 29,” he added.

An incident similar to Mungeshpur took place just days after, on May 30, when the temperature sensor at Nagpur city AWS experienced a malfunction, with erroneous readings showing temperatures as high as 56 degrees Celsius. "The team is actively working to fix this issue. It is important to note that electronic sensors can fail under extreme weather conditions, such as the high temperatures currently being experienced in the region," RMC Nagpur clarified.

In June 2021, an AWS in Amritsar recorded an abnormally high temperature of 50 degrees Celsius, significantly higher than surrounding areas. The IMD found that the sensor malfunction was due to dust accumulation. Similarly, in May 2019, an AWS in Kurnool, Andhra Pradesh, reported a sudden spike in temperature readings, registering 49 degrees Celsius. IMD later confirmed this to be due to a calibration error in the AWS, leading to faulty high-temperature readings.

Deoras added that there had been several such reports from other states. “This warrants an audit of the location and data of AWSs across India. Existing standards of quality control must be beefed up before the AWS data is shared in the public domain,” he said.

What's causing high temperatures in Delhi, Northwest India

On May 29, six cities in India recorded their all-time highest temperatures, while four cities recorded their second-highest temperatures.

Prayagraj in Uttar Pradesh registered an all-time high of 48.8 degrees Celsius, surpassing historical data ranging from 1952 to 2023. Fursatganj also saw its highest-ever temperature at 47.2 degrees Celsius within the range from 2006 to 2023. Sultanpur hit 46 degrees Celsius, while Babatpur recorded 47.4 degrees Celsius. Varanasi recorded 46.7 degrees Celsius, marking its second-highest temperature as well.

DS Pai, head of the Climate Service Division at IMD, explained that while the full summer season is typically experienced from April to June, recurrent western disturbances in March and April brought rainfall and snowfall in northern regions, leading to below-normal temperatures and preventing maximum temperatures from increasing.

“From May onwards, four major factors have driven the intense heat: the absence of western disturbances (a low-pressure system originating in the Mediterranean region, which brings sudden winter rain and snow to the northwestern parts of the Indian subcontinent), hot desert winds affecting the entire northwest region, predominantly clear skies resulting in increased land heating, and warmer nights due to sustained land heating and hot surface winds,” said Pai.

Pai added that this situation is a combination of natural variability and multiple weather factors across the Indo-Pacific region, typical of an El Niño year such as 2024. “However, climate change and global warming have undeniably increased the frequency of heat waves, resulting in higher discomfort and heat stress,” Pai said.

Deoras echoed Pai’s views, saying that El Niño is playing an important role in boosting temperatures this year. “While temperature records in different regions may not break every summer, global warming would continue to increase the frequency, intensity and duration of heat waves across different regions of India in the future climate scenario,” he said.

Temperature variation across Delhi

Delhi's diverse geography and urban layout contribute to significant variations in temperature across the city. Areas like Mungeshpur, Najafgarh, and Palam often record higher temperatures compared to central locations such as Safdarjung. This disparity is attributed to several factors, including the Urban Heat Island (UHI) effect, where densely built-up areas with high-rise buildings, concrete structures, and limited vegetation trap heat.

In a study presented by Roxy Koll from the Indian Institute of Tropical Meteorology to the National Disaster Management Authority earlier this year, a significant temperature disparity was highlighted for night temperatures — up to a 20-degree Celsius difference between urban and rural areas in Delhi and surrounding regions during May 2022. "Climate change is exacerbating heat everywhere, but urban heat islands that trap this heat are our own creation. The UHI effect is most noticeable in nighttime temperatures, with cities being several degrees hotter than surrounding rural areas," explained Koll.

KJ Ramesh, former director general of IMD, said, “The heat released from external units of air conditioners further exacerbates the UHI effect, especially in concrete-heavy and industrial areas. Regions near central Delhi (Lutyens'), with better tree cover, experience comparatively lower heat levels (25%-30% less) than regions with fewer green spaces. Additionally, proximity to water bodies like the Yamuna River can result in cooler temperatures. Conversely, regions further from these natural coolants are hotter.”

Despite the correction in the recorded temperature at Mungeshpur, the overall heat wave situation remains critical. What’s worse, the IMD forecasts the severity of heatwaves to return in June. “Albeit minimal, there is an expected temperature drop for northwest India in the first few days of June. However, the weather outlook indicates strong heat waves to continue for Delhi and surrounding regions till the onset of the monsoon. Citizens are advised to take precautions against heat-related illnesses,” said Pai.

At present, India has around 1,500 manual monitoring stations across all 28 states and 8 Union Territories. Presently, Mumbai has the largest weather network with 60 AWS deployed by the IMD followed by Delhi with 15 AWS. Delhi has 5 manual stations (Safdarjung, Palam, Ayanagar, Ridge, and Lodi Road), while Mumbai has two (Colaba and Santacruz).

“We are targeting to expand the AWS network to the block level across India with 7,000 AWS installed by 2030,” said Mohapatra. “Eight more stations have been planned in Delhi-NCR itself by next year. This will allow a robust coverage of the hyperlocal meteorological data and allow us to study and analyse trends in a much better manner.