Modeling Litter Stocks in Planted Forests of Northern Mexico

Abstract

Litter, LS, is the organic material in which locates in the top A soil horizon, playing key ecological roles in forests. Models, in contrast to common allocation factors, must be used in LS assessments as they are currently absent in the scientific literature. Its evaluation assess the mass, input and flux of several bio-geo-chemicals, rainfall interception as one component of the local hydrology, and wildfire regimes, among others, hence its importance in forestry. The aim of this study was to: (i) develop models to assess LS, accumulation, and loss rates; and (ii) assess rainfall interception and fire regimes in 133 northern forest plantations of Mexico. Two developed techniques: the statistical model (SMLS) and the mass balance budget model (MBMLS) tested and validated local and regional LS datasets. Models use basal area, timber, aboveground tree biomass, litter fall, accumulation, and loss sub-models. The best fitting model was used to predict rainfall interception and fire behavior in forest plantations. Results showed the SMLS model predicted and validated LS datasets (p = 0.0001; r2 = 0.82 and p = 0.0001; r2 = 0.79) better than the MBMLS model (p = 0.0001; r2 = 0.32 and p = 0.0001; r2 = 0.66) but the later followed well tendencies of Mexican and World datasets; counts for inputs, stocks, and losses from all processes and revealed decomposition loss may explain ≈40% of the total LS variance. SMLS predicted forest plantations growing in high productivity 40-year-old stands accumulate LS > 30 Mg ha−1 shifting to the new high-severity wildfire regime and intercepting ≈15% of the annual rainfall. SMLS is preliminarily recommended for LS assessments and predicts the need of LS management in forest plantations (>40-year-old) to reduce rainfall interception as well as the risk of high-severity wildfires. The novel, flexible, simple, contrasting and consistent modeling approaches is a piece of scientific information required in forest management.