Invasion in the Niger Delta: Remote Sensing of Mangrove Conversion to Nypa Fruticans from 2015-2020

Invasive species encroachment is a top threat to biodiversity worldwide. In recent years, Nypa fruticans (Nypa Palm) have emerged as a concerning invasive species in the Niger Delta mangrove system - the largest in Africa and the third largest in the world. However, this region is currently experiencing loss due to oil exploration and urbanization, exacerbating the spread of Nypa fruticans. Here, we use machine learning and Google Earth Engine to quantify and monitor the change in extent of Nypa from 2015-2020. With Landsat imagery and random forest classification, we quantify the extent of mangroves in the Niger Delta in 2019. The Nypa extent map was classified within areas of mangrove by testingSentinel 1 SAR, Sentinel 2 MSI, and ALOS PALSAR models. Random forest classifications using both SAR and Optical data were assessed for separating Nypa from native mangroves. A final model using Sentinel 1 SAR had the highest accuracy and was selected to monitor changes in Nypa extent. Forest height and complexity estimates from GEDI data were compared between areas of mangrove and of Nypa. The extent of Nypa grew by ~10% within the study period and accounted for ~40% of the study area by 2020. At its current rate of growth, Nypa could dominate the study area by 2070. Additionally, Nypa was skewed to lower estimates for all measures of structural variability, including top-of-canopy height, vegetation cover, plant area index, and foliage height diversity. Invasion of Nypa has the potential for devastating impacts on the Niger Delta. Compared to Nypa, Mangroves store more carbon, while supporting of biodiverse ecosystems and livelihoods. Additionally, Nypa can disrupt travel routes by blocking waterways. Continued monitoring of Nypa in this region, as well as economic exploitation of this species presents an opportunity for reducing its encroachment. Future work should explore the use of LiDAR structural metrics to further improve the accuracy of Nypa classification.