Estimating invasion of non native trees with satellite data in Protected Areas

Antonio T. Monteiro (1), João Gonçalves (1), Rui F. Fernandes (2) , Susana Alves (3), Bruno Marcos (1), Richard Lucas (4), Cláudia Santos (1), Salvador Arenas (1),  Ana Cláudia Teodoro (3,5) and João P. Honrado (1)

(1) InBIO/CIBIO – Research Network in Biodiversity and Evolutionary Biology, Associate Laboratory, Universidade do Porto, Campus Agrário de Vairão, 4485-661 Vairão, Portugal; (2) Department of Ecology and Evolution, University of Lausanne, Biophore, 1015 Lausanne, Switzerland; (3) Department of Geoscience, Environment and Land Planning, Faculty of Sciences, University of Porto, Rua Campo Alegre, 4169-007, Porto, Portugal; (4) Centre for Ecosystem Sciences, School of Biological, Earth and Environmental Science, The University of New South Wales, High Street, Kensington, NSW 2052, Australia; (5) Institute of Earth Sciences (ICT), Faculty of Sciences, University of Porto, Porto, Portugal.

Protected areas in European mountains are locations with well-known boundaries that receive protection as a result of significant and co-existing natural and socio-cultural values. Despite these boundaries, invasion by non-native tree species is one of the pressures threatening the conservation of those values. Managing the extent of species invasion is an important task, and predictive model tools on invasion dynamics are commonly used for this purpose. However, there is a frequent scale mismatch between the outputs (e.g. habitat suitability for invasion) and on-ground conservation and mitigation needs, which limits the applicability to invasion management.

In an era of strong advances in species distribution models (SDMs) and Earth observations (EO), hierarchical scaling approaches combining SDMs and satellite mapping at adequate resolutions can contribute to reduce this scale mismatch. While SDMs predicts the propensity to species invasion, the mapping of invasion within the forecasted areas at a certain time-point can evaluate the success of invasion with reference to the predicted maximum potential range. This fine-scale spatially-explicit information can be used to better understand drivers for invasion success and to identify locations with greater potential to suppress invasions.

Using the Peneda-Gerês National Park (Portugal) as testing site and the Acacia dealbata (Silver wattle) as the target, a process with the onset documented to 1905, the conceptual framework was tested. Initially, SDMs predicted the climatically suitable areas for A. dealbata (200m grid cells). Then, object-based satellite mapping using the random-forests classifier on WorldView-2 very-high resolution imagery determined whether the species was able to colonize those predicted areas (invasion success). This scaling down to the landscape level created two metrics that could be used on the ground management: overall and specific success rate of invasion.

SDMs results indicated that most of the study area (67%) was climatically suitable for A. dealbata and satellite mapping highlighted that 12.6% of the area was colonized. Although invasion has still not reached the maximum potential range, A. dealbata was already detected in 62.5% of areas (grid cells) predicted as suitable (overall success). Moreover, the species was also mapped in 55.6% of areas predicted as climatically unsuitable, highlighting the importance of local conditions for the spread of invasion. The specific success rate of invasion was mostly below 40%, indicating that A. dealbata was not dominant and effective management still be possible.

The approach proposed allowed the scale mismatch between outputs from predictive tools and on-ground conservation management to be reduced. Still, the uncertainty of both model predictions and invasion mapping needs to be considered in the interpretation and practical application of results. Free Sentinel-2 data could be considered for future steps. More details can be found in

Citation: Monteiro A. T., Gonçalves J., Fernandes R. F., Alves S., Marcos B., Lucas R., Teodoro A. C., and Honrado J. P. (2017). Estimating Invasion Success by Non-Native Trees in a National Park Combining WorldView-2 Very High Resolution Satellite Data and Species Distribution Models. Diversity 2017, 9(1), 6; doi:10.3390/d9010006.