Montrends: a Google Earth Engine application for analysing species habitat suitability over time
Year: 2024
Type: conference-paper
Abstract: We present a new biodiversity monitoring tool in Google Earth Engine (GEE) to measure trends in species habitat suitability over time using ecological niche models (ENMs) with a time series of satellite products. Focusing on Montesinho Natural Park (MNP) in northeastern Portugal, the application uses MaxEnt to calculate species distribution models for amphibians, birds, mammals, vascular plants, and reptiles with data from six Moderate Resolution Imaging Spectroradiometer (MODIS) products from 2001 to 2023. Habitat suitability trends are estimated with the Mann-Kendall test. The main result is a map for each modelled species showing positive, negative, or null trends over time; a negative trend indicates a monotonic decrease in habitat suitability. The application allows users to select species, the temporal period, the number of model replicates, and the proportion of training and test records. Analyses run intuitively in less than a minute, displaying several results: the mean MaxEnt model over time, Mann-Kendall trends for the study area, species presences, significant pixels, and species presences in significant pixels, main MaxEnt outputs, including Area Under the Curve (AUC) values, variable contributions, plots of global predictor variable contributions over time, average trend values, and MaxEnt parameters. Decision-makers and conservation planners can use this application as a complementary tool for biodiversity monitoring and conservation.
DOI: 10.1117/12.3030967
Language: English
Subfield: Ecological Modeling
Field: Environmental Science
Domain: Physical Sciences
SDGs: Life on land, Climate action
Open Access status: green
OpenAlex ID: W4402511391