SciEPublish

Ecological Modelling of Physico-Chemical Parameters and EPT Macroinvertebrate Abundance in River Kiminini, Trans-Nzoia County, Kenya

Article Open Access

Ecological Modelling of Physico-Chemical Parameters and EPT Macroinvertebrate Abundance in River Kiminini, Trans-Nzoia County, Kenya

Author Information
1
Department of Environmental Sciences, University of Eldoret, Eldoret P.O. Box 1125-30100, Kenya
2
Department of Aquaculture and Fisheries Science, Lilongwe University of Agriculture and Natural Resources, Lilongwe P.O. Box 219, Malawi
*
Authors to whom correspondence should be addressed.

Received: 24 June 2026 Revised: 22 July 2026 Accepted: 21 October 2026 Published: 11 November 2026

Creative Commons

© 2026 The authors. This is an open access article under the Creative Commons Attribution 4.0 International License (https://creativecommons.org/licenses/by/4.0/).

Views:149
Downloads:39
Hydroecol. Eng. 2026, 3(4), 10015; DOI: 10.70322/hee.2026.10015
ABSTRACT: Freshwater macroinvertebrates are widely used in biomonitoring due to their tendency to change in abundance and distribution with changing environmental conditions. Different taxonomic groups vary in response to individual physicochemical variables. This study assessed the association of physicochemical variables and the abundance of Ephemeroptera, Plecoptera, and Trichoptera (EPT) in River Kiminini, Trans-Nzoia County, Kenya. The sampling was done during the wet season from July to September 2023. The field data was collected twice a month for three months across 10 sampling stations, resulting in a total of 60 observations. Ephemeroptera, Plecoptera, and Trichoptera counts were used as response variables, while temperature, total dissolved solids (TDS), electrical conductivity (EC), pH (potential of hydrogen), nitrate, and phosphate were used as predictors. There was a significant overdispersion among the three taxa from a Poisson distribution, with a variance to mean ratio of 16.49, 5.73 and 5.48 for Ephemeroptera, Plecoptera and Trichoptera, respectively. The negative binomial generalized linear mixed models (NB2 GLMMs) with station as a random intercept were fitted, and predictors were standardized to 1 standard deviation. The NB2 (Negative binomial type 2 parameterization) specification was supported by model comparison over NB1 (Negative binomial type 1 parameterization) for all three taxa, especially for the Ephemeroptera. Ephemeroptera abundance was negatively associated with TDS (incidence-rate ratio [IRR] = 0.564, 95% confidence interval (CI): 0.350–0.907) and pH (IRR = 0.679, 95% CI: 0.511–0.902), but positively associated with nitrate (IRR = 1.372, 95% CI: 1.052–1.788). Trichoptera abundance was negatively associated with TDS (IRR = 0.322, 95% CI: 0.119–0.867) and pH (IRR = 0.387, 95% CI: 0.204–0.733). No statistically significant relationships were found for Plecoptera. Simulation-based residual diagnostics indicated no evidence of excess zero inflation for any taxon. Ephemeroptera demonstrated a modest residual-dispersion signal that warrants caution. The findings illustrate taxon-specific associations between physicochemical gradients and EPT abundance and provide a regional application of count-data modelling for freshwater biomonitoring in Kenya.
Keywords: Physico-chemical variables; Freshwater macroinvertebrates; Negative-binomial GLMM; Overdispersion; Biomonitoring; EPT
TOP