Mechanisms behind facilitation-competition transition along rainfall gradients

Abstract Woody cover is rapidly changing due to mortality, shrub encroachment, and afforestation, reshaping herbaceous communities and ecosystem functioning worldwide. Often, trees and shrubs promote herb growth in dry sites but suppress it in wetter ones, as predicted by the classical Stress Gradient Hypothesis. However, explanations for the facilitation-to-competition transition remain verbal and contested, lacking a clear link to resource competition theory. Here, we present a mechanistic framework consisting of two submodels: (i) canopy shading that reduces photosynthesis and evapotranspiration, and (ii) root effects, including water uptake and increased moisture via hydraulic redistribution. We elucidate the conditions under which interactions shift from facilitation to competition. The models reproduce this reversal only when water is not the sole limiting factor at high rainfall or when woody density increases with precipitation. Moreover, the reversal can occur across any aridity gradient, including those driven by evaporative demand influenced by temperature and humidity. The two pathways leave distinct signatures: canopy shading produces a hump-shaped pattern with maximum facilitation at intermediate stress, while the root pathway predicts a shift from positive to negative interactions as water availability increases. By translating a classic idea into a quantitative framework, this model enhances ecosystem management in a changing world. Competing Interest Statement The authors have declared no competing interest. Footnotes ↵2 Slope σ: for d = 0 the slope is σ0 = a/(e0m), whereas for d > 0 it is σd = σ0 · (fa(d)/fe(d)). Since fa(d) > fe(d), we have that σd > σ0. Intercept ω: for d = 0 the intercept is ω0 = −(qs/ed)(m/a)γ−1, whereas for d > 0 it is . Since fa(d), fe(d) 2, we have that each of the parenthesis in the expression for ωd is greater than 1, and therefore ωd < ω0 (remember the minus sign in ω0).