Multivariate assessment of riverine water quality in the Ouémé Delta (Benin): ecological suitability and adaptive monitoring design
摘要
Riverine water quality in data-limited deltas is a subject that has received little attention, despite their high biodiversity value and rising agricultural and urban pressures. The present study establishes a baseline for the lower Ouémé Delta (Benin), focusing on the Ouémé and Sô rivers, and tests compact indicators for management. Monthly sampling over 24 months was conducted at ten stations. The in situ parameters encompassed temperature, pH, electrical conductivity (EC), salinity, dissolved oxygen (DO), and turbidity. The laboratory variables included total nitrogen (TN), total phosphorus (TP), orthophosphate (PO4-P), nitrate (NO3-N), ammonia (NH4-N), and chlorophyll-a. These measurements were integrated with discharge estimates from an Acoustic Doppler Current Profiler (ADCP). Multivariate analyses comprised partial correlations, regression, and regularised linear discriminant analysis. The Sô River demonstrated chronic wet-season hypoxia (DO < 1 mg/L during > 80% of July–October), whereas the Ouémé generally maintained higher DO (≥ 2 mg/L); turbidity peaked during floods (Ouémé 250 NTU; Sô 100 NTU). Nutrient levels reached 21 mg/L TN (Sô) and 344 µg/L TP (Ouémé). River identity was moderately separable, with DO, turbidity, TP, and salinity emerging as the main discriminating variables. TP was consistently explained by turbidity and PO₄–P (R2 up to 0.92 in the Ouémé; 0.72 in the Sô), whereas TN remained weakly predictable. Overall, biological suitability was primarily constrained by DO deficits, high turbidity, and nutrient pressure, with salinity acting as a seasonal filter in the Sô. A stratified–adaptive monitoring design, combining control nodes along major gradients, a parsimonious marker set and high-risk seasonal intensification, can maximise information yield under budget constraints and support SDG 6.3.2 reporting through station–season classification of ambient water quality.