Indirect remote sensing of a cryptic forest understorey invasive species

Remote sensing has successfully been applied to map the distribution of canopy dominating invasive species. Many invaders however, do not dominate the canopy, and remote sensing has so far not been applied to map such species. In this study, an indirect method was used to map the seed production of Chromolaena odorata , one of the world's 100 worst invasive species. The study was executed in lowland Shorea robusta forest in Nepal, where Chromolaena invaded the understorey of a degraded forest. A Landsat ETM+ image processed through a neural network predicted 89 and 81% of forest canopy density and light intensity reaching the understorey, respectively. We inverted these models to predict Chromolaena seed productivity. Light intensity determined 93% of the variation in log10 seed production per plant. Chromolaena failed to produce seed below a light intensity of 6.5 mJ m- 2 day-1. Further analysis revealed that Chromolaena was absent above this light intensity in case of a high biomass of other shrub and herb species, a situation occurring in the absence of grazing. We therefore suggest that other species control Chromolaena through competitive exclusion in the absence of grazing, whereas grazing breaks the dominance of these other species thus creating the conditions for Chromolaena attain canopy dominance. The presence of grazing was related to distance from the forest edge, a variable that together with light intensity allowed us to map 64% of variation in Chromolaena cover. Predicted Chromolaena cover and seed production per plant were combined into a map displaying the total seed production per unit area. Such map displaying seed producing sites could be used to significantly reduce the costs of controlling Chromolaena infestation by providing information on the spatial segregation of source and sink populations, which will support efficient habitat ranking to restore invaded areas and protect non-invaded ecosystems. This may prove particularly valuable when implementing control measures under circumstances of limited capital and manpower