Invasive species mapping: Using Sentinel-2 satellite data to assess the abundance of Amur honeysuckle (Lonicera maackii)

Keywords: Amur Honeysuckle (Lonicera maackii), invasive species, understory vegetation, non-native shrub, Remote sensing
Abstract Type: Poster Abstract

Abstract

Invasive plants, such as Amur honeysuckle (Lonicera maackii), threaten many forest ecosystems in North America by altering vegetation structure and species composition. Previous studies indicate that the presence of such invasives reduces species diversity, inhibits tree regeneration, and negatively affects wildlife habitat. Consequently, understanding the spatial distribution of Amur honeysuckle is essential to inform effective forest management decisions. However, detecting and mapping this species on the ground is resource intensive. Remotely sensed data offer a viable alternative, but effective identification of invasive species is challenging in temperate forests and woodlands because forest canopies obscure honeysuckle foliage to different degrees during most of the growing season. Although there have been efforts to address this issue, methods remain limited for deciduous forests. This study focuses on using remotely sensed data to map the presence and abundance of Amur honeysuckle beneath closed canopies in dry-mesic oak-dominated forests in southern Iowa. Leveraging the unique, extended phenology of Amur honeysuckle, we collected Level-2A Sentinel-2 satellite sensor data during April to assess its abundance. We employed partial least squares regression to identify salient bands and derivative indices to map the distribution and abundance of Amur honeysuckle in mature oak-hickory forests. This mapping technique is relevant to broader landscapes, which is crucial for informing regional forest management decisions.