Mozambique: ‘We take thousands of aerial photos and run them through a big computer’

Source(s): World Food Programme

A flood-hazard model created by WFP and partners is helping people in Mozambique prepare for climate shocks

By Jenny Wilson

“There really is a big technological challenge to fighting world hunger,” says Michael Manalili, a Geospatial Officer for the World Food Programme (WFP).

Michael works with cutting-edge technology to visualize humanitarian emergencies. For his latest project, Michael was part of a WFP team which, collaborating with partners, created a flood-hazard model in Mozambique that calculates where damage from waters is set to occur (such models are created by scientists and techies as a troubleshooting tool).

Now fully up and running, it could change the way communities are affected by flooding. “By already knowing which areas will be flooded we can prepare those areas and limit the damage caused,” he says.

Mozambique is one of the most disaster-prone countries. Last year — for the first time in recorded history — it was hit by two strong tropical cyclones in the same season. Just months later, severe flooding in the north of the country affected 55,000 people. Since the first cyclone, Idai, hit last March, WFP has delivered food assistance to 2.2 million people.

Philippines-born Michael studied Geomatics engineering and joined WFP in 2018, having previously worked for the Food and Agriculture Organization of the United Nations. He describes the new technology in very complicated language. Then, putting it simply, he says: “We fly drones over the flooded areas, take thousands of photos and run them through a big computer.”

In January, as part of the test process for the flood-hazard model, Michael visited two of the worst hit areas — the provinces of Cabo Delgado and Zambezia. He was investigating whether the information gathered by the drones was an accurate depiction of what’s happening on the ground.

“We walked several kilometres and took a boat to reach some of the most cut-off villages,” says Michael. “Many communities were still submerged under water and lots of people hadn’t even left their homes,” he adds. “We asked them about the floods and compared their answers to the information from the drones but found that 30 per cent didn’t match up.”

People, it seems, are reluctant to report flooding — worried that the government could demolish their house in response. “Mozambique will use this information to establish no-build zones so that people aren’t living in the areas almost certain to flood,” says Michael. “Even though it’s for their own safety, the residents don’t want to move. These are their homes”.

Another part of the flood model that Michael has been working on can rapidly count the number of people in a certain area. “Using drones we can detect and count individual humans, even if they are sitting down or walking,” he says. “So far we haven’t had a rapid and consistent way of estimating the number of people in need of assistance. That is an issue in large-scale emergencies. This solves that problem.”

To test the new feature, Michael headed to an evacuation centre where there are large numbers of people in one place. “It was heartbreaking seeing their situation,” he says. “Most of the evacuees said they don’t want to stay in the tents — simply because the material is plastic and it’s very hot inside. Imagine, during the day the temperature can reach 36 degrees. There were about 106 families in this location alone.”

Children outnumbered adults. With emergencies ever more frequent due to complex factors, having the latest technology available is crucial for continuing to save and change lives.

“Investing more time and effort in preparedness is critical. Now that this flood hazard model is up and running, we can implement it in other flood-prone areas where WFP operates to help communities prepare for flooding and prevent disaster,” says Michael.

Explore further

Hazards Flood
Country and region Mozambique
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

Is this page useful?

Yes No Report an issue on this page

Thank you. If you have 2 minutes, we would benefit from additional feedback (link opens in a new window).