Upscaling species invasion patterns from local to regional for forest ecosystem management

Almost all remote sensing studies model invasion of species dominating ecosystems canopies, typically in predicting their presence or absence. However, a vast majority of invasive species do not dominate ecosystems canopies. Remote sensing experts have traditionally overlooked such cryptic invaders spatial distribution and productivity pattern within map units for a variety of reasons including spectral response limitations and inadequate quantitative data. Almost none attempted to model and upscale such populations distributed across forest understorey environment. These inclusions reduce the capability of remote sensing and GIS techniques or importance of maps. This is crucial in invasive species management. In this study, we synthesized data from different remote sensing and GIS sources to (1) model the actual and potential area and forest types in Nepal vulnerable to invasion of the Chromolaena odorata, one of the world's worst cryptic invasive species and (2) segregate the reproductive and non-reproductive populations at national scale using local scale information. Results reveals that invasive species models developed at local scale could successfully be up scaled at national scale. The map of current potential distribution of C. odorata shows that out of 75 forest communities of Nepal, 9 are currently infested by the C. odorata. They are: Acacia - Dalbergia, Alder, Hill Shorea, Pine, Pine - broad leaved forest, Riverine broad leaved, Schima - Castanopsis, Terai Shorea and Upper tropical riverine forest. Such information is crucial for land managers to focus their precious funds and efforts to control the spread of this species and so that the control methods are practical.