Towards a framework for flood vulnerability assessment for rural and urban informal settlements in Malawi

Abstract

The most significant strategy to address flood hazards is to conduct a Flood Vulnerability Assessment (FVA) because it informs the disaster risk reduction and preparedness process. FVA provides metrics that can help manage flood risks and disaster events. However, many flood vulnerable regions like Malawi still lack FVA supporting frameworks in all phases (pre, trans, and post-disaster). Partly, this is attributed to a lack of evidence based studies to inform the processes. On one hand, the frameworks that exist in Malawi such as the Unified Beneficiary Register (UBR) and Rapid Assessment (RA) are tools that reflect the aftermath of a disaster. On the other hand, the Participatory Vulnerability Capacity Assessment (PVCA), though used as a tool, cannot be negated that it is an approach rather than a tool because indicators have been not suggested. This informed the need to assess Households’ Flood Vulnerability (HFV) in the Mtandire ward of Lilongwe City (LC) and Traditional Authority (T/A) of Karonga District (KD) to propose an FVA framework for rural and urban informal settlements in Malawi and beyond. A household survey was used to collect data from a sample of 545 households’ participants in June-August 2021. Vulnerability was explored through a combination of Underlying Vulnerability Factors (UVFs) with Vulnerability Components (VCs). The UVFs and VCs were agglomerated using abinomial multiple logit regression model. Variance Inflation Factors (VIFs) were used to check the multicollinearity of variables in the regression model. The analysis was carried out using R software and STATA. Multiple Correspondence Analysis (MCA) and Artificial Neural Network (ANN) were conducted to determine the variability of factors contributing to vulnerability. The results reveal a total average score of high vulnerability (0.62) and moderate vulnerability (0.52) on MCA in T/A Kilupula and Mtandire ward respectively. The results further show that all the indicator variables in Mtandire ward have an inertia value at the expected rate of less than 10% while in T/A Kilupula lack of credit unions (0.103), lack of markets (0.499) and poverty (0.123) display values that deviate from the expected score of <10%. The study concludes that the determinants of households’ flood vulnerability are place settlement, low-risk knowledge, communication accessibility, lack of early warning systems, and limited access to income of household heads. This study provides an FVA framework that could be applied to promote the resilience of communities to mitigate flood risks and support the planning and decision-making process in Malawi and at any region in the world because all the input data is globally available.