Spatio-Temporal Dynamics of Flood-Induced Anopheles Breeding Sites and Malaria Hotspots in Urban Abia State: A GIS-Based Cohort Study

Background: Flooding exacerbates malaria transmission in urban sub-Saharan Africa by expanding Anopheles breeding sites, yet spatio-temporal linkages remain understudied in Nigeria’s Niger Delta.

Methods: We conducted a GIS-integrated prospective cohort study (May-October of 2024 and 2025) in flood-prone wards of Aba and Umuahia, Abia State, enrolling 600 participants (300 under-5 children, 300 pregnant women) from 500 households. Entomological surveys (n=200 sentinel sites, 48 fortnights) assessed larval/adult densities. Active surveillance detected malaria episodes using RDT/PCR. Sentinel-2-derived indices (Normalised Difference Vegetation Index [NDVI], Modified Normalised Difference Water Index [MNDWI]) informed geospatial analyses, and included Moran’s I for spatial autocorrelation and Getis-Ord Gi*, alongside Generalised Linear Mixed Model [GLMMs]. Agent-based models simulated Larval Source Management [LSM] impacts.

Results: Among 552 retained participants (92% follow-up; 552 person-years at risk [pyar]), 135 episodes of malaria occurred (incidence: 245/1,000 pyar; 95% CI: 206–286), peaking at 43–71% during floods (Plasmodium falciparum, 98%). Larval density peaked at 19.1/dip (r=0.89 with adults). Breeding sites showed moderate spatial autocorrelation (Global Moran’s I=0.32–0.38, z=3.45, p<0.001). Optimised Getis-Ord Gi* analysis detected 70 significant incidence hotspots (Gi* z-score>1.96; 14% households near the Imo River accounting for 62% episodes despite 28% area coverage). GLMMs (pseudo-R²=0.52) linked MNDWI (incidence rate ratio [IRR] 2.18, 95% CI: 1.45–3.28), fortnightly cumulative rainfall (IRR 1.25 per 100 mm increase, 95% CI: 1.12-1.39), and NDVI (IRR 0.45 per 0.1-unit increase) to malaria incidence, and ITN use protected from malaria (IRR 0.70). LSM simulations with Bacillus thuringiensis israelensis [Bti] larviciding targeted at hotspots projected a 42% reduction in malaria incidence (95% CrI: 35-49%).  

Conclusions: Flood metrics generated clustered hotspots in urban Abia and amplified vulnerability among children and low-income groups. Targeted Bti larviciding using GIS could avert up to 42% malaria incidence, and this should inform Nigeria’s malaria elimination programming amid climate risks. Integrated hydrogeomorphic vector surveillance is essential for equity-driven malaria control in Sub-Saharan Africa.