What happens when the storm arrives but safety is miles away? In the cyclone-prone villages of coastal West Bengal, this isn’t a hypothetical question, it’s daily reality. As the Bay of Bengal heats up and cyclones grow fiercer, the need for strong disaster infrastructure has never been greater. Yet a new study reveals that shelters, roads, and support systems remain unevenly built and often too far from the people who need them most. Only one in five families in Purba Medinipur district can reach a cyclone shelter within a reasonable distance. For the rest, safety often comes down to chance, a neighbour’s pucca house, an overcrowded school, or simply staying put in a fragile hut. In the age of climate change, survival shouldn’t depend on luck. And yet, in Bengal’s coastal belt, it still does.
A new study published in Regional Science Policy & Practice by Sripurna Kanjilal and Gupinath Bhandari delivers a hard-hitting assessment of just how far India’s preparedness has come, and how far it still needs to go. Focusing on five cyclone-prone blocks of Purba Medinipur, Ramnagar I, Ramnagar II, Contai I, Deshapran, and Khejuri II, the researchers offer a detailed analysis of cyclone shelters, road networks, healthcare centres, schools and administrative buildings. Their conclusion is sobering: the distribution of life-saving infrastructure is not only sparse but also spatially misaligned with population centres, putting thousands at risk.
“Only 21% of the vulnerable population have reasonable access to multipurpose cyclone shelters (MPCS) during disaster periods, leaving the remaining 79% more susceptible to cyclone hazards,” the study states.
In disaster planning, location is everything. In their study, “Physical infrastructure development in cyclone preparedness strategy: An assessment in the cyclone prone blocks of West Bengal, India,” the authors used advanced spatial analysis tools like Nearest Neighbour Analysis (NNA) and population-weighted mean centres to determine how well public infrastructure is aligned with actual community needs. The findings are stark.
Ramnagar I fared best, with cyclone shelters exhibiting a geometrically uniform distribution (Rn = 1.01), and shelters placed reasonably close to population centres. But things deteriorate quickly as one moves east. Khejuri II’s shelters, for instance, are tightly clustered in just two pockets in the southern part of the block — far from where most people live. Its Rn score of 0.45, coupled with a z-score of −10.65, indicates severe clustering — a worst-case scenario for equitable disaster protection.
“An infrastructure is utilised utmost if it is located in the vicinity of the mean centre of population,” note the authors. Yet in Khejuri II, none of the shelters are anywhere near the population mean — a finding that highlights critical planning failures.
During cyclone warnings, time and distance can determine survival. UNDP guidelines advise that vulnerable populations should be able to reach a safe shelter within 1.5 km — or a 30-minute walk at most. In a region where many families still live in mud-and-thatch huts within 200 metres of the high tide line, this is not a luxury. It is a necessity.
In practice, however, only one in five households surveyed live within 1 km of an MPCS. And even high schools, which often double as makeshift shelters, are accessible to only 26% of respondents. Access to metal roads that are vital for evacuation is similarly poor with only 31% of surveyed households within half a kilometre.
Healthcare facilities and administrative buildings were more evenly distributed, with 83% and 55% access respectively, but this offers limited comfort. Hospitals, the report reminds, cannot be used as mass shelters, and Gram Panchayat offices are often too small to accommodate more than a few families at a time.
These findings are not just academic. They echo in the lived experiences of the people of Purba Medinipur — especially during disasters like Cyclone Amphan in May 2020. As the storm battered the coast, many were forced to scramble for safety. Shelters quickly became overcrowded, and social distancing — mandated during the peak of the COVID-19 pandemic — became impossible.
“Some took refuge in nearby pucca (concrete) houses, while others had no choice but to stay back in their huts with livestock,” the authors recount. The chaos highlighted the desperate need for more shelters, better transport links, and clearer public awareness.
Indeed, the researchers found that many villagers didn’t even know where the nearest shelter was. Some feared theft if they left their homes unattended. Elderly people often hesitated to relocate unless shelters were within walking distance. Women, particularly those pregnant or with children, were reluctant to leave their families behind and move alone to distant facilities.
The study goes beyond counting shelters. It examines how different types of infrastructure work together — or fail to. For example, a well-placed MPCS is useless if the road to reach it is flooded or broken. During and after a cyclone, roads are vital for supplying food, water, medicine, and emergency personnel.
The survey found that in several blocks, road accessibility remains a major bottleneck. Poor road conditions can trap entire communities in the aftermath of a storm, turning a temporary refuge into an isolated island. Backup electricity, clean water, and emergency supplies are often disrupted as well.
“The roads are the lifeline of any area and connect to different infrastructure,” the authors stress. “If the connecting roads become inundated and inaccessible, the local inhabitants taking shelter are literally stuck in a deserted island.”
West Bengal is no stranger to cyclones. The frequency and intensity of tropical cyclones in the Bay of Bengal are projected to increase due to climate change. As the sea warms and monsoons grow erratic, millions in India’s eastern coastal belt face an escalating threat.
Yet, as the study points out, these risks are not equally distributed. Vulnerability is shaped by poverty, poor housing, lack of awareness, and political neglect. A well-planned cyclone shelter can make the difference between a temporary disruption and a tragedy. But when shelters are built too far from where people live — or without considering topography and road access — they become monuments of failure rather than safety.
“This study deciphers the lacunae of the existing infrastructure in disaster risk reduction,” the authors write. “Addressing these to ensure a more equitable distribution of infrastructure may lead to a more holistic disaster preparedness planning of the community.”
Based on the findings, the study offers clear recommendations:
Recalibrate shelter locations: New MPCS must be strategically placed closer to the mean centres of population — especially in under-served blocks like Khejuri II and Ramnagar II.
Integrate infrastructure planning: Roads, schools, health centres, and administrative buildings must be planned together as an interconnected system, not in silos.
Public awareness campaigns: People need to know where to go, what to carry, and whom to call. Awareness is as crucial as cement.
Gender-responsive planning: Women’s safety and autonomy must be prioritised, with provisions for separate spaces and transportation for vulnerable individuals.
Use data, not political pull: Site selection should be driven by population data and GIS analysis — not political convenience.
Physical infrastructure is only part of the story. What’s also needed is institutional trust and community engagement. Disaster preparedness is not simply a technical challenge; it is a human one. As climate change intensifies, West Bengal — like many regions worldwide — must make difficult decisions about how, where, and for whom to build its safety net.
The lessons from Purba Medinipur are clear: Resilience begins long before the cyclone makes landfall. And it begins not with concrete alone, but with compassion, clarity, and coordination. When the winds rise and the sea surges, the difference between life and death in Bengal’s coastal villages is measured not in megawatts or masonry, but in access, trust, and dignity.
Cyclone shelters are not just buildings; they are promises of protection, equality, and care. But when those promises are too far, too few, or too fragile, the most vulnerable — women, children, the elderly, and the poor, are left to face the storm alone. True resilience will come only when planning listens to people’s lived realities, not just to maps and models. A just cyclone preparedness system must place human beings at its centre — their safety, their agency, and their right to survive with dignity. Until then, the storms of inequality will continue to batter these shores long after the winds have passed.