Policies and People | How AI is helping India’s poor become climate-resilient

Published on Sep 22, 2022 06:53 PM IST

Heatwaves are becoming frequent and intense in India. SEEDS, a non-profit, and Microsoft have developed an AI model that processes large volumes of data to provide risk information at a hyper-local level to vulnerable communities

Sunny Lives. (@SeedsIndia/Twitter) PREMIUM
Sunny Lives. (@SeedsIndia/Twitter)

Pooja Singh, a 35-year-old domestic help, works almost 12 hours daily to support her four-member family. Her income isn’t enough to meet the expenses, yet Singh, a school dropout, made a significant investment this year. The mother of two school-going children took a loan of 15,000 from a relative to buy an insulation sheet to cover the ubiquitous blue tin roof of her house, located inside a congested slum in Ghaziabad, Uttar Pradesh.

“The summers are getting hotter by the year, and my house becomes a furnace. My old parents struggle to rest. I can’t sleep either after a hard day’s work. I fell sick twice this year, and employers cut my salary. My children can't study. The insulation sheet is a boon. It was worth the money and effort,” says Singh.

In a climate crisis-hit world, people like Singh, who cannot afford cooling devices or are not responsible for global warming, are paying a heavy price for rising temperatures.

Rapid rises in heat gain due to exposure to hotter-than-average conditions compromise the body’s ability to regulate temperature and can result in a cascade of illnesses, including heat cramps, heat exhaustion, heatstroke, and hyperthermia, warns the World Health Organization. In addition, deaths and hospitalisations (another considerable drain on already meagre resources) from heat-related illnesses can occur exceptionally rapidly (same day) or have a lagged effect (several days later) and result in accelerating death or illness.

Global warming also increases work-related heat stress, damaging productivity and causing job and economic losses. The poorest countries will be worst affected, warns a 2019 report from the International Labour Organization.

Battling high temperatures

Several scientific reports have warned that heatwaves will become frequent and intense in India (and the world) in the coming years.

This year has been terrible.

An HT analysis of the global temperature database maintained by NASA shows that the June-July-August period might indeed have been the warmest the world has seen since 1880.

Unfortunately, the discussion around heat's impact on cities and their population does not include a very significant section of the population, the underprivileged, who stay in highly congested slum areas, which have much higher temperatures (especially indoors) than surrounding areas.

“If the temperature of any area in the city is 38°C, be sure it would be several degrees higher inside a tin shack because of the building material used and lack of proper ventilation,” explains Dr Anshu Sharma, co-founder of SEEDS, a New Delhi-based disaster response and preparedness non-profit. “Such a temperature spike affects those who usually don’t go out for work such as women, young children, disabled and old people”.

Unfortunately, the impact of heatwaves is not discussed enough.

“There is a lack of data about heat-related deaths. Most of the deaths don’t get reported as heat-stress deaths. This is because eventually, the person will die of a fever or cardiac arrest or dehydration. So minus granular data, generating a conversation on heat and its severe impacts is very difficult,” adds Sharma.

No localised information

While the government does send out regular heat advisories, they are not localised and are easily decipherable by at-risk communities and other actors involved in disaster response and planning.

To overcome this challenge, SEEDS, in collaboration with Microsoft and technology partner Gramener, has developed “Sunny Lives”, an AI model that bridges this gap so that communities can be better prepared and governments can plan better at a local scale.

The model, developed under Microsoft’s global programme “AI for Humanitarian Action”, processes large volumes of data to provide risk information at a hyper-local (building cluster) level.

The model has generated heatwave risk information for around 125,000 people living in slums in New Delhi and Nagpur. (Screengrabs: SaferWorldComm/Youtube)
The model has generated heatwave risk information for around 125,000 people living in slums in New Delhi and Nagpur. (Screengrabs: SaferWorldComm/Youtube)

“The roof condition of a building (for example, tin sheets heat up much faster compared to other materials) says a lot about the condition inside the house. The AI model uses high-resolution satellite imagery to detect and assign risk scores to buildings according to their roof types,” explains Sharma.

Other location and hazard-specific attributes — built-up density, vegetation, the proximity of the building to a water body, and rooftop material, with its attendant heat-absorbing capacity — are also used in calculating the risk scores of the houses. Then these structures are identified, classified, colour-coded and mapped with their risk score.

The risk map is then overlaid onto a regular map available on a smartphone, which volunteers then use on the field. It helps them figure out where they must go to issue warnings, the chances of water scarcity in an area, or where local administrators should direct resources.

The model has generated heatwave risk information for around 125,000 people living in slums in New Delhi and Nagpur.

Reaching the people

However, generating this granular data and mapping the homes are not enough.

SEEDS’ biggest challenge is to communicate the dangers of a heat wave since, unlike other hazards — floods, earthquakes or cyclones — the impact of a heatwave is not exactly visible. Plus, people need to be aware of how to react to these warnings and told about low-cost solutions such as painting the roof white, growing vegetables on the roof or putting insulation under the roof to reduce the temperature inside the rooms.

To ensure that people make sense of the risk scores and encourage them to take appropriate climate-resilience strategies, SEED holds focused group discussions and quizzes on heat.

“These campaigns are designed to understand people’s perception of heat, difficulties and concerns faced by the communities and the measures and interventions taken by the community members in order to tackle the heat and its impacts,” explains Sharma. The organisation also interviews frontline workers such as Accredited Social Health Activists and Anganwadi workers to understand how the rising temperatures impact the community.

During such discussions, local people have also devised their strategies. For example, Razia, a domestic worker, laid a few jute bags on her bamboo roof for insulation. Vanshika, a student, covered her family’s overhead water tank with polystyrene sheets and jute sacks, which provided just enough insulation to stop the water from heating up.

The future of the AI model

As the model matures, the model’s deployment will be scaled for multiple cities across the country.

“We are also looking to collaborate with various city and state governments, to aid them in developing risk-informed planning. Our vision for the model is to use it for climate change adaptation and disaster management in a way that the hyper-local risk of the communities is understood and pathways for their protection and resilience are put into practice,” says Sharma.

The views expressed are personal

Enjoy unlimited digital access with HT Premium

Subscribe Now to continue reading
freemium
SHARE THIS ARTICLE ON
  • ABOUT THE AUTHOR

    KumKum Dasgupta is with the opinion section of Hindustan Times. She writes on education, environment, gender, urbanisation and civil society. .

SHARE
Story Saved
×
Saved Articles
Following
My Reads
My Offers
Sign out
New Delhi 0C
Thursday, December 08, 2022
Start 15 Days Free Trial Subscribe Now
Register Free and get Exciting Deals