India's "Invisible' Slums- A challenge to flattening the curve

India's "Invisible' Slums- A challenge to flattening the curve

Photo Credit: Hindustan Times

From op-eds in international publications to local news apps, ’flattening the curve’ is a much used – and arguably abused – phrase in the midst of the current pandemic. Though it is a tangible measure at a time when everything else seems riddled with uncertainty, the term is perhaps used without accounting for the hurdles that lie in the way.

Developing countries, like India, face unique challenges in flattening the curve which go beyond issues of systems and infrastructure. Perhaps one of the most significant of these is the problem of informal settlements where millions of people across the developing world live with poor housing and sanitation infrastructure.

To understand why this poses a grave threat to containing the epidemic and flattening the curve, let’s take a look at some statistics. Over a billion people across the world live in informal settlements or ’slums’. India’s 2011 census estimated that over 65 million people live in informal settlements in India.

This number is worrying, because of the living conditions in these settlements. Over 50% of slum households live in one-room tenements, with Gujarat and Maharashtra topping the list. A 2016 report by FSG titled ‘Informal Housing, Inadequate Property Rights’ pointed out that in cities like Delhi and Pune, over 90% of households in informal settlements use community toilets. In Delhi, only 17% of informal settlement residents had a tap water connection in their house. Under such living conditions, physical distancing with frequent washing of hands, as required to contain the spread of COVID-19 is impossible. Even a few infected residents can lead to an outbreak in these densely populated settlements and can bring the entire state’s or even the country’s medical infrastructure down on its knees.

A precedent of this is the Ebola epidemic in West Africa between 2014 and 2016. At the time, its spread was largely through the vast and densely populated urban slums of Liberia, Guinea, and Sierra Leone.

Thus, it is crucial for public health officials to closely monitor these informal settlements to reduce the potential for an outbreak.

The conundrum of “invisible” slums

Therein lies another challenge. Government officials do not have adequate visibility on the location, and population, of these informal settlements since the official data, is woefully out of date. The 2011 census alone counted 25 million people living in ‘identified slums’, a category of households that do not exist on any government records. Given the dynamic and mostly unregulated nature of slums, coupled with rapid urbanization, this number has probably exploded in the last decade. This change is starkly visible when we look at satellite images of the same locations over the years.

It will be critical for public health officials to identify these slum locations, which could become future hotspots for the pandemic, and implement preventive control measures soon.

Over the years, Omidyar Network India has supported our partners in developing a deeper understanding of urban informal settlements, and leveraging geospatial technology to enhance effectiveness of government policy. Our mission drives us to offer up all these tools in the fight against the pandemic, especially to protect the “Next Half Billion” who are among the most vulnerable members of our society.

Open-source, machine learning algorithms to identify informal settlements

In a study released in 2018, a multi-disciplinary research team from Duke University, Univ. of North Carolina and North Carolina State University used high resolution satellite imagery, combined with ground verification, to build an open-source machine learning algorithm that identifies informal settlements. The study showed that while Bangalore’s official slum settlement count was 597, a machine learning algorithm was able to identify nearly 2000 slum settlements in Bangalore with over 80% accuracy. As long as such informal settlements remain excluded from the official purview, they could potentially be significant blind spots in the administrative machinery’s efforts to control the pandemic outbreak.

These algorithms are available on GitHub for use by the research community and public officials to map urban slums and focus precious resources in these communities in the efforts to flatten the curve. We, at Omidyar Network India strongly encourage the academic and research community to build on these algorithms and expand them to other cities as part of the efforts to control the outbreak of Covid-19.

In the global fight against the COVID-19 pandemic, we hope that technology and resources like this help to support and protect the inhabitants of urban India’s informal settlements and ensure that they do not remain ‘invisible’.