Weak correspondence between modelled ecosystem metabolism and macroinvertebrate-based surrogates across land-use gradients in Afrotropical headwater streams

Ecosystem metabolism is a sensitive functional indicator of carbon processing and disturbance in streams. Macroinvertebrate functional feeding group (FFG) ratios have been proposed as low-cost surrogates, but their validity in tropical systems, where assemblage structure and body-size distributions differ markedly from the temperate streams in which production-to-respiration (P/R) thresholds were calibrated, has rarely been tested. We examined the correspondence between modelled ecosystem metabolism and FFG-based surrogates across a land-use gradient in the headwater streams of the Sondu-Miriu River basin, Kenya, during dry and wet seasons. Gross primary production (GPP) and ecosystem respiration (ER) were estimated using Bayesian diel oxygen modelling across 24 streams spanning four land-use types. FFG P/R ratios, derived from numerical abundance and biomass, were compared with modelled GPP:ER ratios. Modelled GPP (1.57–4.71 g O₂ m⁻² d⁻¹) and ER (−5.78 to −13.56 g O₂ m⁻² d⁻¹) fell within tropical stream ranges. Land-use significantly influenced all metabolic parameters, whereas FFG abundance and biomass showed no significant response to land-use or season. Despite being numerically scarce, shredders dominated community biomass through large-bodied freshwater crabs and tipulids, generating strong divergence between abundance- and biomass-based P/R ratios. Nutrients, organic matter, turbidity, temperature and stream size were the primary drivers of both metabolic and FFG metrics. Both approaches consistently classified all streams as heterotrophic, confirming allochthonous energy dependence across the land-use gradient. However, neither FFG metric significantly tracked site-level variation in modelled P/R, reflecting temporal scale mismatches and the inapplicability of temperate P/R thresholds to Afrotropical assemblages where shredder body-size asymmetry biases biomass-based ratios. FFG metrics are best interpreted as complementary indicators of trophic status rather than substitutes for modelled metabolism; region-specific threshold calibration is needed.