What We Publish

Move up or move over: mapping opportunities for climate adaptation in Pakistan's Indus plains

Rafael J. P. Schmitt, Edgar Virgüez, Sara Ashfaq, and Ken Caldeira · Environmental Research Letters 18, 114024 · 2023

Key finding. For the 2022 Indus floods, rebuilding houses elevated and flood-proofed would have protected 13% of affected people and temporary relocation a further 16%, while the remaining 70% faced water shallow enough to address with low-cost measures — but flood-proofed rebuilding alone would raise reconstruction costs by 26 to 63%, or $1.5 to $3.6 billion on top of $5.8 billion.

A composite panel. A scatter plot of flood depth against distance to dry land, coloured by the number of people affected, is divided into zones labelled "move up", "move over" and "relocate". Cumulative curves show how many of the nearly 40 million impacted people fall in each zone, with a small bar chart summarising the four categories. Photographs show a house on stilts and a fortified dwelling. A map of Pakistan's Indus floodplains shades each area by which adaptation option applies, in blue, green and red.
Depth of water and distance to dry ground decide which adaptation is available, and the scatter shows most of the affected population sitting in shallow water close to safety — the cheap-to-address majority. Only the red areas on the map need relocation or hard adaptation. Figure 4 from Schmitt et al. (2023), Environmental Research Letters 18, 114024. Reproduced under CC BY 4.0. Extracted from the published PDF and resized for web display.

What question did this research address?

The 2022 Indus floods put 30 to 40 million people's homes under water and permanently displaced 8 million. Reconstruction after a disaster is the moment when adaptation is cheapest to build in, because the rebuilding is happening anyway.

Knowing which measure to apply where requires flood-stage and demographic data at high resolution, which Pakistan's floodplains largely lack. This paper asked whether a rapid assessment could nonetheless identify where each option would work, and what it would cost.

What did we find?

The method pairs a high-resolution rapid assessment of flood stages with demographic data, so the depth of water each population faced can be matched to the adaptation that would have helped.

The population splits three ways rather than two. Moving up protects 13%, moving over 16%, and the large remainder — 70% — were exposed to water shallow enough that low-cost measures would have sufficed.

That last group is the important one, because it means most of the exposure does not require either expensive elevation or displacement.

The costs are substantial even so. Flood-proofed rebuilding adds $1.5 to $3.6 billion to an estimated $5.8 billion reconstruction, a 26 to 63% increase, before any relocation or adaptation of other infrastructure is counted.

The authors are explicit that missing local flood-stage and socio-economic data is a real source of uncertainty, and present the approach as a prototype for rapid post-disaster assessment rather than a finished plan.

Why does it matter?

It answers a question that arrives with a deadline. Reconstruction begins within weeks of a flood, long before a careful multi-year study could report, so a rapid method that is roughly right is worth more than a precise one that arrives late.

Splitting the affected population by water depth changes what adaptation means here. Most people did not need to move or to be raised — they needed something cheap — and a policy built only around elevation and relocation would spend heavily to protect a minority.

It also states the cost honestly rather than presenting adaptation as free. Adding a quarter to two-thirds to a reconstruction bill is a real fiscal choice for Pakistan, and one made under the same constraint that makes the country vulnerable in the first place.

Citation

Rafael J. P. Schmitt, Edgar Virgüez, Sara Ashfaq, and Ken Caldeira (2023). Move up or move over: mapping opportunities for climate adaptation in Pakistan's Indus plains. Environmental Research Letters 18, 114024.

Related