Abstract
Interactions between species and detritus in aquatic ecosystems involve unassimilated food, non-predator mortality, and complex trophic relationships, making it challenging to quantify interaction strengths. This study utilized classic and revised Lotka-Volterra equations, combined with the food web of Baiyangdian Lake, to develop methods for measuring interaction strengths in a phytoplankton-based and a detritus food web. The analysis relied on three types of species-detritus interactions and outputs from an Ecopath model (1958-2019). Loop weight and Diagonal strength (S) were employed to assess stability. Lighter loops weight and lower S value indicate higher stability. From 1958 to 2009, the stability of Baiyangdian Lake was limited by a three-link omnivorous loop: Detritus > zooplankton > filter-feeding fish. As the predator-prey biomass ratio (filter-feeding fish/detritus) increased, instability increased, and vice versa. However, the new loop (detritus > zooplankton > phytoplankton) and corresponding new predator-prey biomass ratios (zooplankton/detritus) resulted in stability from 2009 to 2019. It inferred dominant top-down trophic cascade effects changed to dominant bottom-up trophic cascade effects. Besides focusing on the heaviest loop weight, it was necessary to examine the heavier loops that may have a chance of evolving into the heaviest ones following catastrophic or long-term perturbations to the food web. To facilitate management, a geometric mean ratio of predator-to- prey biomass (/)t was proposed as a simplified indicator. This metric correlates with diagonal strength (R (2) = 0.6645) and offers a practical tool for early-warning assessments of food web stability, despite its moderate precision. This study highlights the importance of integrating detritus dynamics into stability analyses and using loop weight analysis to identify critical trophic interactions. The proposed empirical indicators provide a bridge between theoretical models and ecosystem management practices.