Macroinvertebrate assemblages and body size distribution in response to land use change in tropical headwater streams

Taxonomic approaches for monitoring the ecological status of aquatic ecosystems face limitations, creating a need for alternative ataxonomic methods that are valid, disturbance-sensitive, and cost-effective for freshwater management. This study evaluated macroinvertebrate size-spectrum metrics as indicators of ecological status in tropical headwater streams. Macroinvertebrate abundance (individuals/m²), wet weight (mg), and physico-chemical variables were measured during dry and wet seasons from 24 headwater streams in the Sondu-Miriu River Basin, Kenya, across four land-use types: natural-forest (NF), tea and tree plantations (TTP), smallholder-tea (SHT), and smallholder-agriculture (SHA). Water quality based on turbidity, total suspended solids (TSS), total dissolved solids (TDS), electrical conductivity (EC), particulate organic matter (POM), and nitrate (NO₃⁻-N) identified NF as least disturbed and SHA as poorest streams. Taxon richness, diversity, and %EPT (Ephemeroptera-Plecoptera-Trichoptera) indices reflected land-use-specific influences on assemblage structure. Redundancy analysis (RDA) showed small-bodied taxa (<8 mg) associated with disturbance indicators, while large-bodied taxa (>32 mg) indicated higher sensitivity. Permutational multivariate analysis of variance (PERMANOVA) revealed significant differences in macroinvertebrate abundance and biomass per size-class among land uses but not between seasons. Slopes of abundance- and biomass-based size-spectra deviated from theoretical steady-state conditions (-2.0 and -1.0), indicating uneven macroinvertebrate distribution across size-classes but no significant variation between land uses or seasons. In contrast, spectrum midpoint heights differed significantly, with highest values at SHT and NF and lower at SHA and TTP. Thus, size-spectrum midpoint height, rather than slope, was the more disturbance-sensitive metric, varying consistently with land-use effects, identifying it as a potential ataxonomic indicator for routine monitoring of ecological status in tropical headwater streams.