Ensemble species distribution models for 69 marine non-indigenous species in European waters: current and future scenarios.

Résumé fourni par la source

Data and model outputs for ensemble species distribution modeling of 69 marine non-indigenous species across European watersThis repository contains the input data, model outputs, and performance metrics associated with:Pagnier, J., Andermann, T., Andersson, M.G., Obst, M. (2026). The role of genetic observatory networks in the detection and forecasting of marine non-indigenous species. DOI preprint: https://doi.org/10.21203/rs.3.rs-8702791/v1New version Pagnier, J., Andermann, T., Andersson, M.G., Obst, M. (2026). Proactive marine conservation through early detection and forecasting of non-indigenous species using genetic observatory networks (in review)The dataset supports ensemble species distribution modeling analyses of marine non-indigenous species under current and future climate scenarios (SSP1-2.6, SSP2-4.5, SSP5-8.5 for 2100).Associated scripts: https://github.com/JustinePa/ARMS-marine-alien-SDM Repository structure:Raw occurrence downloadsraw_occurrences_downloads/gbif/: Zipped raw GBIF downloads (one per species) containing dataset identifiers for data provenanceraw_occurrences_downloads/obis/: Raw OBIS downloads (one per species) containing dataset identifiers for data provenanceInput datainput_data/final_occurrences_thinned_curated/: Spatially thinned and quality-controlled occurrence records used for modeling (one file per species)input_data/environmental_data/: Environmental predictor layers for current conditions (myExpl_shelf.tif) and future projections (ssp126_shelf_2100.tif, ssp245_shelf_2100.tif, ssp585_shelf_2100.tif)Model outputsraw_outputs/: Zipped folders (one per species) containing species-level ensemble model projections for:EMwmeanByBOYCE_*: Ensemble mean weighted by CBI by evaluation metrics (current, SSP1-2.6, SSP2-4.5, SSP5-8.5)EMwmeanByBOYCE_*_...BOYCEbin: Binarized ensemble mean weighted by evaluation metrics (current, SSP1-2.6, SSP2-4.5, SSP5-8.5)EMcvByBOYCE_*: Ensemble coefficient of variation indicating prediction uncertainty (current, SSP1-2.6, SSP2-4.5, SSP5-8.5)fratio plots: P/E curve use to define the species-specific thresholdthe .ensemble.projection.out file: biomod2 projection objectMulti-species richness patternsaggregations_allScenarios/: Combined richness projections across all 69 modeled species, obtained by summing the binarized individual maps.unmasked.zip: richness computed with all species' full distributionmeow_masked.zip: richness computed with species distribution only in the marine regions where they are considered introducedModel performance assessmentmodel_performance/individual_models_performance/: Evaluation metrics for individual modeling algorithms (one file combining all models generated for all species)model_performance/ensemble_models.zip: Evaluation metrics for ensemble model projections (one file per species)Ballast Water Contigency Area mapping:BWCA/: contains the input neccessary to compute the BWCA:the masked NIS richness layerthe Offshore Wind Farm layer: windfarmspolyPolygon.shpthe Marine Protected Areas layer: wdpa_raster_europe.tifEuropean countries polygons: CNTR_RG_01M_2020_4326.shpA chlorophyll raster reused just as a spatial template/mask for the coastline, used to compute distance-from-coast This dataset enables full reproducibility of the modeling workflow and provides species-level projections for researchers and marine resource managers interested in individual species distributions beyond the multi-species patterns presented in the associated publication.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.