Invasive Species Compendium

Detailed coverage of invasive species threatening livelihoods and the environment worldwide

Abstract

Negative binomial hurdle models to estimate flower production for native and nonnative northeastern shrub taxa.

Abstract

Although allometric equations have been developed to predict aboveground biomass components for several shrub species, few efforts have been made to predict flower or fruit production. As numerous invasive Eurasian shrubs establish throughout eastern deciduous forests, it is becoming important to understand the ecosystem consequences of shifting dominance from native to nonnative shrub species, particularly with regard to fruit and nectar production. In this study, we created negative binomial hurdle models to predict flower production for Japanese barberry (Berberis thunbergii DC.), dogwoods (Cornus amomum Mill. and Cornus racemosa Lam.), border privet (Ligustrum obtusifolium Siebold & Zucc.), honeysuckle (Lonicera morrowii A. Gray, Lonicera Ă— bella Zabel, and Lonicera tatarica L.), multiflora rose (Rosa multiflora Thunb.), and brambles (Rubus allegheniensis Porter, Rubus occidentalis L., and Rubus phoenicolasius Maxim.). The models effectively predicted flower production per ramet for 97-100% of the shrubs found at our study sites, but exhibited high predictive variability for the few stems in the largest diameter classes. This novel modeling approach can be used in a variety of ways to assist land managers in prioritizing invasive shrubs or trees for removal, determining a minimum replanting size for native plants to have reproducing individuals, estimating landscape-level sugar production, and better understanding shrub demography.