Abstract: Biomass is an important metric to quantify vegetation productivity. Shrub biomass models are one of the important methods for quickly estimating shrub biomass. This study focused on three common slope-protection shrub species, Amorpha fruticosa, Vitex negundo, and Lespedeza bicolor, on the slopes of expressways in Henan Province. Based on data of leaves, stems and aboveground biomass obtained by field harvesting as well as the estimation of plant height, crown width and basal diameter of each shrub, the estimation models for foliage, stem and aboveground biomass of the three shrub species were established. The optimal prediction models were selected according to the size of the determination coefficient (R²), the size of the standard error (SEE) of the estimated value and the significance level of the regression test (P). The biomass of three shrub species was estimated by using the aboveground biomass optimal models. The results showed that stem/leaf ratio of the three species had little difference, ranging from 2.57-3.35. Plant volume (V), plant height (H) or basal diameter square multiplied by plant height (D₂H) were used as the independent variables for the optimal models, and the optimal models were mainly quadratic or cubic functional equations. The results of the measured values showed that the prediction accuracy of the function models was above 97%, with good prediction accuracy. The aboveground biomass of shrubs on the expressway slopes in Henan Province was 6.86 × 10⁷ kg. The results demonstrated that the established models can be applied to the estimation of the leaves, stems and total aboveground biomass of Amorpha fruticosa, Vitex  negundo, and Lespedeza bicolor on expressway slopes.

Key words: slope, shrub, biomass, prediction model, prediction accuracy