Scholarlink

<p>In Sub-Saharan Africa, approximately 25% of the urban population resides in slums. Yet, studies examining the spatiotemporal development and flood vulnerability of slums remain scarce in East African cities. This study aims to (i) analyse spatio-temporal dynamics in the Kibera (Kenya) and Katanga (Uganda) slums, comparing them with Bannyahe (Rwanda), where residents were successfully relocated to the Busanza model village, and (ii) to quantitatively assess flood exposure across various land use and land cover (LULC) categories and exposed populations. We utilised the Random Forest (RF) algorithm with Landsat 7, 8, and 9 data (2012–2024) for multi-temporal LULC mapping, alongside Synthetic Aperture Radar (SAR)-based flood detection using Sentinel-1. The model was evaluated via confusion matrices, achieving 92.75% accuracy for RF and 89% for SAR. The results show that built-up areas increased significantly in Kibera (86.67%) and Katanga (77.52%), while Bannyahe experienced a 32.22% decline after 2021 due to the relocation project. Overlay analysis revealed that flooded built-up areas in Katanga increased from 2.29 hectares in November 2023 to 4.0 hectares in March 2025, while in Kibera, more than 10,000 residents were exposed to flooding during each flood event. Importantly, our algorithms generate flood maps for affected areas and estimate exposed populations within minutes, making them readily applicable to other regions. These findings provide evidence that current slums are increasingly unsustainable; therefore, we recommend that urban authorities in Kenya and Uganda adopt proactive resettlement policies—similar to the Bannyahe slum approach in Rwanda—to relocate residents from high-risk zones to planned model villages, such as Busanza, to enhance urban climate resilience.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Spatio-temporal analysis of slum growth and flood inundation in east african cities using machine learning

  • 期刊论文

  • 发表日期：

    2026年2月19日

查看全文
  • Katabarwa Murenzi Gilbert,
  • Qian Shi

摘要

In Sub-Saharan Africa, approximately 25% of the urban population resides in slums. Yet, studies examining the spatiotemporal development and flood vulnerability of slums remain scarce in East African cities. This study aims to (i) analyse spatio-temporal dynamics in the Kibera (Kenya) and Katanga (Uganda) slums, comparing them with Bannyahe (Rwanda), where residents were successfully relocated to the Busanza model village, and (ii) to quantitatively assess flood exposure across various land use and land cover (LULC) categories and exposed populations. We utilised the Random Forest (RF) algorithm with Landsat 7, 8, and 9 data (2012–2024) for multi-temporal LULC mapping, alongside Synthetic Aperture Radar (SAR)-based flood detection using Sentinel-1. The model was evaluated via confusion matrices, achieving 92.75% accuracy for RF and 89% for SAR. The results show that built-up areas increased significantly in Kibera (86.67%) and Katanga (77.52%), while Bannyahe experienced a 32.22% decline after 2021 due to the relocation project. Overlay analysis revealed that flooded built-up areas in Katanga increased from 2.29 hectares in November 2023 to 4.0 hectares in March 2025, while in Kibera, more than 10,000 residents were exposed to flooding during each flood event. Importantly, our algorithms generate flood maps for affected areas and estimate exposed populations within minutes, making them readily applicable to other regions. These findings provide evidence that current slums are increasingly unsustainable; therefore, we recommend that urban authorities in Kenya and Uganda adopt proactive resettlement policies—similar to the Bannyahe slum approach in Rwanda—to relocate residents from high-risk zones to planned model villages, such as Busanza, to enhance urban climate resilience.