Communities in India flooded by ‘Amphan-scale cyclones’ could triple by 2100

Author

Ayesha Tandon

Source(s)
Carbon Brief
An Indian family standing outside their house after cyclone Fani devastated the neighborhood.
Cylone Fani hit India in 2019.
Rakesh Roul/Shutterstock

Super cyclone Amphan” hit the Bay of Bengal in May 2020, exposing hundreds of thousands of people to flooding.

New research, published in Climate Resilience and Sustainability, finds that if a storm the size of cyclone Amphan hits the Bay of Bengal in 2100, it would expose “vastly” more people to extreme flooding – mainly due to sea level rise. 

The study concludes that in high- and medium-emission scenarios, an Amphan-scale cyclone could triple the percentage of people in India exposed to severe flooding by 2100, compared to the 2020 event. An increase of up to 20% in exposure is expected in Bangladesh, the study adds.

Tropical cyclones are “the most devastating things that can hit south Asia”, the lead author of the study tells Carbon Brief. Despite this, he notes that research and funding into climate extremes in this part of the world is lacking.

The study emphasises the need to develop coastal adaptation strategies to help communities in the region become more resilient to future flooding, a scientist not involved in the research tells Carbon Brief.

Super cyclone Amphan

The Bay of Bengal – bounded by countries including India, Bangladesh and Myanmar – is home to around 500 million people. It has also seen some of the deadliest tropical cyclones in history.

On 20 May 2020, “super cyclone Amphan” hit the Bay. Amphan was the biggest source of displacement in 2020, forcing 2.4 million people in India alone to leave their homes. According to the study, Amphan was also the costliest cyclone to ever make landfall in south Asia:

“Hundreds of thousands of cattle were lost, tens of thousands of buildings completely destroyed and ~$20m damage done to fisheries alone, primarily from the storm surge”.

Meanwhile, there was a notable rise in Covid-19 cases in the area in the weeks following the cyclone, as people crowded into cyclone resistant shelters. This is an example of “compounding crises”, the study says.

Damage from cyclones is predominantly caused by storm surges – a temporary change in sea level caused by a storm.

To model the storm surge caused by cyclone Amphan, the authors use forecasts of water height along the coastline taken from a high-resolution tide-surge model. They find that Amphan created storm surges 2-4m high across coastal regions of the Bay of Bengal.

The authors combine this information with a flood inundation model – which uses a “digital elevation map” to show the height of land in the Bay of Bengal – and predict where the storm surge goes. Gridded population data is used to estimate how many people were affected by the flooding. The study finds that around 740,000 people in Bangladesh and 420,000 in India were exposed to flooding from the storm surge.

The map below shows the flood height driven by cyclone Amphan (top), where pink indicates greater flood heights. In the population exposure map (bottom), darker purple indicates a greater percentage of the population exposed to flooding. The size of the circle in the bottom map indicates the number of people exposed, where a larger circle indicates greater exposure.

Flood inundation and population exposure maps of cyclone Amphan.
Flood inundation and population exposure maps of cyclone Amphan. Source: Mitchell et al (2022).

Sea level rise

As sea levels rise, storm surges will produce higher floods, which are able to reach further inland. This study asks: “If an Amphan-scale storm surge occurred under a world with increased sea levels, how might the population exposed [to flooding] change?”

Dr Dann Mitchell – a professor of climate science at the University of Bristol and lead author on the study – tells Carbon Brief that this paper is an “attribution study” because it aims to identify the fingerprint of climate change in an extreme weather event. However, he notes that it is “a bit different from normal extreme event attribution”.

Most attribution studies – such as those conducted by the World Weather Attribution – ask how much more intense, likely or long-lasting an extreme event was due to climate change. However, this paper looks into the future, estimating both meteorological and societal changes from an event occuring in a warmer world.

To model future sea level change, the authors combine projections from the sixth coupled model intercomparison project (CMIP6), with storm surge estimates produced using a dynamic storm model. They consider three different future scenarios – a low-emission scenario (SSP1-RCP2.6), a “business-as-usual” scenario (SSP2-RCP4.5), and a high-emission scenario (SSP5-RCP8.5).

The plot below shows the projected temperature increase between the 1850-1900 and 2090-2100 averages (left), and sea level rise in the Bay of Bengal between 2020 and 2100 (right). Three emission scenarios are shown – a low-emissions scenario (blue), a “business as usual” scenario (yellow) and an extremely high-emission scenario (red).

Projected temperature increase and sea level rise in the Bay of Bengal over 2020-2100
Projected temperature increase between the 1850-1900 and 2090-2100 averages (left), and sea level rise in the Bay of Bengal over 2020-2100 (right). Source: Mitchell et al (2022).

In the Bay of Bengal, the sea level rise between 2020-2100 could range from 0.32m to 0.84m, depending on the emission scenario, the authors find.

Flood exposure

While sea level rise is an important predictor of flood risk, population growth and migration patterns will also determine how many people are exposed to flooding.

The authors assess a range of shared socioeconomic pathways (SSPs) – scenarios that examine how global society, demographics and economics might change over the next century. SSP1 and SSP5 project the lowest future populations, while SSP2 offers a more “middle-of-the-road” population projection.

The plot below shows the changes in population flood exposure in Bangladesh (purple) and India (yellow), as a percentage change over 2020–2100. Extreme flooding (greater than 3 metres) is shown by the top bars, moderate (greater than one metre) by the middle bars and low (greater than 0.1 metres) by the bottom bars in each subplot.

Results are shown for four scenarios: low emission and population growth (a), medium emission and population growth (b), high emission and low population growth (c) and high emissions, using the 2020 population (d).

Change in population exposure to a cyclone Amphan-scale storm in 2020 and 2100
Percentage change in population exposure to a cyclone Amphan-scale storm in 2020 and 2100, for Bangladesh (purple) and India (yellow). Subplots show the low- (a) mid- (b) and high- (c) emission scenarios. Meanwhile, d shows the high-emission scenario, using the 2020 population. Source: Mitchell et al (2022).

The authors conclude that if a storm the size of cyclone Amphan hits the Bay of Bengal in 2100, more people in India and Bangladesh will be exposed to the resulting storm surge, in the majority of future scenarios, than in 2020. They add that, generally, higher levels of future emissions will drive greater flood exposure in 2100.

However, Mitchell tells Carbon Brief that the finer details are “a mixed bag”. 

The largest increases in flood exposure are estimated in India, where population exposure to extreme flooding is projected to increase by 50-90% in the low-emission scenario, and reach beyond 200% – a tripling of flood exposure – in the high-emission scenario.

Meanwhile, percentage changes in flood exposure in Bangladesh are “considerably lower” than in India for all flood severity levels, the report finds. 

For example, under the SSP5 scenario, the global population is projected to peak and decline in the 21st century. In this scenario, the percentage of people in Bangladesh and India living in urban areas will increase from 35% to 90%, due to migration away from the coast.

The study adds: 

“It is only when we follow a low-emission scenario, consistent with the 2C Paris Agreement goal, that we see no real change in Bangladesh’s storm surge exposure, mainly due to the population and climate signals cancelling each other out.”

South Asian research

The Bay of Bengal is a “hotbed” for tropical cyclones, which – according to Mitchell – are among “the most devastating things that can hit south Asia”. Nevertheless, the opening lines of the study note that “research on climate extremes in the region is substantially lacking compared to other parts of the world”.

For example, Mitchell tells Carbon Brief that errors can be “quite large” in elevation maps around south Asia. Lack of data in Myanmar is why the authors chose to exclude the country from their analysis, Mitchell explains. 

Dr Kevin Walsh – a professor of meteorology at the University of Melbourne, who was not involved in the research – says the study is “technically very thorough and worthwhile”. However, he also notes the limitation in accuracy of the land elevation data in the Bay of Bengal. Furthermore, he says that data on undersea water depth could also have “considerable uncertainties” in the region.

The study “flags the potential impacts of such a storm in the future if no adaptation measures are undertaken,” Walsh adds.

Dr Andra Garner – an assistant professor at Rowan University in New Jersey, who was not involved in the study – says that the study emphasises both the need to “develop coastal adaptation strategies to help communities in the region become more resilient to future flooding” and to mitigate our emissions.

Mitchell also highlights the investment into cyclone research and forecasting in recent years that has already paid off.

For example, he notes that over the past five years, forecasting centres and procedures have been put in place to give people in India and Bangladesh advanced warning of upcoming cyclones. This advanced notice was “a game-changer”, without which deaths would have been “significantly higher”, according to Mitchell.

Share this

Please note: Content is displayed as last posted by a PreventionWeb community member or editor. The views expressed therein are not necessarily those of UNDRR, PreventionWeb, or its sponsors. See our terms of use