A new Irish-designed artificial intelligence (AI) model that can accurately predict the extent of future flooding could be used by local authorities and communities such as Middleton to take steps to limit emergency damage.
The project is led by researchers at CeADAR, the Center for AI Applications at University College Dublin (UCD) in Ireland, and aims to warn communities ahead of heavy rain and to help local authorities assess damage to homes and businesses. It said it could soon be used to contain and give time to evacuate. Protect residents and livestock.
Using data collected by the European Space Agency’s Sentinel 1 satellite, the research team used data collected by the European Space Agency’s Sentinel 1 satellite to analyze past flood-prone areas in Ireland, including Midleton in Cork, Limerick, Carrick-on-Shannon in Leitrim and Athlone in Westmeath. Flood events were mapped. These maps were then input into an AI model designed to predict the extent of future flooding in these areas to an accuracy of about 20 meters.
Ireland has experienced extensive flooding in recent months, with Storm Ciaran, Storm Debbie and Storm Babette having a negative impact on residents and businesses, particularly in the west and south-west of the country.
Researchers say flooding in Ireland is “certain to worsen” in the coming years due to climate change, with flooding occurring not only in areas that have never flooded before, but also in historically vulnerable areas. He said it would intensify.
Last year, the Irish Fiscal Advisory Council warned that extreme flooding events caused by climate change could cost the country around €500 million a year by the end of 2010. It also warned that a once-in-a-decade flood in Dublin could cause up to €2.9 billion in damage and put more than 14,500 properties at risk.
Dr Omid Memarian Solkabi, who is leading the development of the model, said the Sentinel-1 satellite was used to monitor Middleton flooding in real time during Storm Babet, which would refine and increase the accuracy of the model. .
“Flooding like the one that hit Middleton during Storm Babette is devastating for the homes and business owners whose properties were the hardest hit. A silver lining is that Sentinel-1 was directly over the area at the time. “As a result, we were able to collect a lot of valuable data that will help us predict the scope of the next event and ensure future damage is limited,” he said.
Dr Solkabi said that while the tool was still in the process of being developed, tested and validated, there was “huge potential” for the model to be made available to local authorities and trained and expanded for use in other parts of the world. He said that there is.
Dr Oisin Boydell, CeADAR’s director of applied research, said that while flood forecasting and mapping has traditionally been based on weather models and low-resolution maps, “this is very data-driven and is based on events over the past 10 years. It’s based on the current situation.” in the designated area. ”
“This project will have a significant impact on communities in areas of Ireland that are at high risk of flooding. This is certain to lead to rapid climate change, with even more intense precipitation events expected in the future.”
The model forms part of a larger project led by University College Dublin to develop the Earth Observation (EO) services sector in Ireland and assess the potential impact of EO data in the areas of climate, agriculture and oceans. The purpose is to investigate.
