Dataset: Port Fourchon, LA Species Range Data

Species Range Estimation

To estimate species ranges, we employed the gbif.range R package (Chauvier et al., 2022) for its ability to integrate multiple data sources, including GBIF occurrences and ecoregion boundaries curated by The Nature Conservancy (2012). We accessed the Global Biodiversity Facility (GBIF) database focusing on species occurrences based on both human observations and preserved specimens spanning 2005 to 2022  (GBIF 2024). The downloaded data were subjected to rigorous cleaning, removing any records outside of the western hemisphere or observations that did not align with Atlantic coastal regions. This step ensured that our analyses focused on species relevant to the ecosystem under study.

The gbif.range package offers automated range delineation by grouping observations into ecoregions. We applied these tools to each species observed at Port Fourchon to estimate both their historical and current ranges. Spatial range estimates were refined using polygon overlays, restricting them to the latitudinal gradients from the North to South Poles along the Atlantic coast. We also evaluated the completeness of occurrence data by assessing whether species records covered their known ecological niches.

For species with sparse or conflicting records, range estimates were adjusted manually based on prior literature and known habitat preferences. Final outputs included range maps for each species, with ecoregion-based polygons providing clear visualizations of distribution shifts over time. The complete list of species, with associated GBIF records and curated metadata, is included in the supplementary materials.

Statistical Analysis of Species Ranges

To test for shifts in species distributions, we hypothesized that northward range expansion would result in a southward shift in the latitudinal ranges of species observed at Port Fourchon over time. We utilized PERMANOVA (Permutational Multivariate Analysis of Variance), a robust statistical approach that accounts for non-parametric data distributions and complex multivariate relationships. The PERMANOVA was run on latitude metrics (mean, minimum, and maximum latitudes) for all species observed during each sampling year (2005, 2015, 2022).

In addition to comparing latitudinal range metrics across years, we assessed the dispersion of species assemblages using beta-dispersion tests. These tests evaluated whether differences in latitudinal spread varied significantly across years, ensuring our PERMANOVA met assumptions of homogeneity of dispersion. If significant differences in dispersion were detected, we adjusted interpretations accordingly to avoid Type I errors in our statistical comparisons. Specifically, beta-dispersion in 2014 was higher than in 2005 (p = 0.048), which we accounted for when analyzing year-to-year differences.

Each PERMANOVA model was validated through permutation-based significance testing with 999 iterations to ensure robustness against potential biases due to sampling variability. We report effect sizes and R² values to assess the magnitude of latitudinal shifts across time points. As a further check, we conducted post hoc pairwise comparisons to explore year-specific trends in range shifts, ensuring no significant temporal bias in our dataset. All analyses were performed using R statistical software (version 4.2.2).

---
The data were processed and analyzed using R and python code. The README for Climate-Driven Range Expansion Analysis Code ("README_Species_Range_figure.R" in "Supplemental Files" ) provides details for code and input files included in this dataset.

This project analyzes the range expansion of black mangroves into areas traditionally dominated by smooth cordgrass using quantitative nekton sampling and satellite imagery analysis. The analysis includes species range estimation, satellite imagery processing, and statistical testing on species distributions over two decades.