Coupling GeoAI with Big Data Analytics to Enhance Urban Flood Resilience

Keywords: Flood Resilience, GeoAI, Big Data, Citizen Science, Urban Areas, Green Infrastructure, Informal Settlements.
Abstract Type: Paper Abstract

Abstract

The increasing intensity of rainfall events due to Climate change is causing unprecedented catastrophic flooding and livelihood challenges in urban areas worldwide. These problems are not limited to urban areas in developing countries in the Global South but include major cities in developed countries. These urban areas depend on neighboring peri-urban or distant rural areas for their water supply and flood mitigation. Still, they lack green spaces (or green infrastructure) within their immediate environs for flood resilience. We are interested in flood resilience in New York City and the vulnerability of informal settlements in Nairobi (Kenya) and Kampala (Uganda). Improving the flood ecosystem provisioning benefits of urban green infrastructure requires in-depth analytics of high-resolution Geospatial “Big” data. Machine learning algorithms are effective in pattern recognition in Big Geospatial data. This presentation will utilize case studies of Rain Gardens for flood resilience in New York City, as well as Environmental Inequality of flood vulnerability in Informal Settlements in Nairobi and Kampala. It will present assessment results that highlight how flood ecosystem services provisioning benefits from synergies between Machine Learning and Geospatial "Big" Data Analytics. We will also discuss the role of Citizen Science data collection initiatives to address a key limitation to utilizing Big Data Analytics, mainly the availability and accuracy of disparate multisensor and multiscale geospatial data.