Towards a Predictive Ecology of Cyanobacterial Blooms in Nigerian Reservoirs: Integrating Environmental Drivers, Molecular Ecology, and Remote Sensing

Cyanobacterial blooms in freshwater reservoirs represent a growing ecological and public health challenge in Nigeria, driven by eutrophication and climate warming. However, a predictive understanding integrating environmental drivers, cyanobacterial community ecology, and remote sensing detection remains absent, severely limiting early warning capacity. This study develops a predictive framework for cyanobacterial bloom dynamics in five major Nigerian reservoirs spanning the North Central and South West geopolitical zones, combining high-frequency environmental monitoring, amplicon sequencing of cyanobacterial communities, toxin gene profiling, and Sentinel-2 satellite image analysis. Weekly water quality monitoring covered temperature, stratification, nutrient concentrations, and light availability over two full hydrological years. Molecular analyses characterised cyanobacterial community structure and quantified toxin gene abundance using quantitative polymerase chain reaction. Machine learning models were trained on the combined environmental and molecular dataset to predict bloom probability and toxin production risk. Sentinel-2 phycocyanin indices were validated against in-situ chlorophyll-a and cyanobacterial cell density to assess remote sensing detection accuracy. Results show that thermal stratification onset, total phosphorus exceeding 50 micrograms per litre, and Microcystis aeruginosa dominance collectively predicted 78 percent of toxic bloom events across all five reservoirs with three-week lead time. Sentinel-2 phycocyanin indices successfully detected 84 percent of surface bloom events confirmed by in-situ sampling. Inter-reservoir differences in bloom seasonality were primarily driven by catchment nutrient loading patterns. The study delivers the first validated predictive bloom model for Nigerian reservoirs and provides a remote sensing and molecular monitoring protocol deployable for early warning by Nigerian water utilities and NESREA. Keywords: cyanobacterial blooms, predictive ecology, reservoirs, remote sensing, Nigeria.