Landscape forest cover, not edge contrast, modulates edge effects on palm diversity
Data files
May 12, 2026 version files 28.46 KB
-
README.md
4.27 KB
-
Trujillo_Miranda_etal_Edge_effects_on_palm_diversity.xlsx
24.18 KB
Abstract
Edge effects are among the most important consequences of habitat fragmentation, altering abiotic conditions and biological communities in remnant forests. However, the magnitude of edge effects is highly variable, and the factors that modulate them, particularly the interplay between local edge type and the landscape context, remain poorly understood. Given ongoing increases in the proportion of the world's remaining forest found near edges, understanding these effects is critical for conservation.
We investigated how edge effects are mediated by type of edge (cacao plantation vs. cattle pasture) and landscape forest cover (%), shaping palm density and diversity across forest fragments in the Ecuadorian Chocó. We established plots (100 m2) at four distances to the edge (0-10 m, 30-40 m, 60-70 m, and 90-100 m) in 30 sites, 15 for each type of edge across a gradient of landscape forest cover. In each plot, we recorded all adult palms, juveniles and seedlings.
The response of palms to distance to the edge and level of landscape forest cover varied across life stages: seedling communities were primarily associated with distance from the edge, showing an increase in density and richness, but a decrease in evenness, towards the forest interior. For juveniles, landscape forest cover was as important an additive predictor as distance to the edge, both increased density and richness, whereas evenness declined with higher surrounding forest cover. Adult density also increased with distance to edge and higher levels of landscape forest cover. Negative edge effects on adult richness and evenness were buffered with an increase in landscape forest cover. Unexpectedly, the type of edge contrast had no significant effect on palm density or diversity at any life stage.
Synthesis and applications: Our findings demonstrate that landscape forest cover plays a critical role in modulating edge effects in plant communities in our study system, but not all woody crops systems provide buffering benefits, as cacao plantations in our study failed to soften edge effects. Consequently, conservation strategies should prioritize maintaining high landscape forest cover to ensure the integrity of remaining forest fragments.
Dataset DOI: 10.5061/dryad.cz8w9gjkh
Description of the data and file structure
Study area
We conducted this study in the Mache-Chindul Ecological Reserve and surrounding areas in northwestern Ecuador, within the Chocó biogeographical region, a global biodiversity hotspot known for its exceptional plant diversity and humid tropical climate. The vegetation is classified as evergreen premontane forest, with uneven to hilly topography, mean annual temperatures of 23–25.5 °C, and annual precipitation generally ranging from 2,500 to 4,000 mm. Although the area has a wet season from December to June, rain, mist, and fog occur throughout the year.
Site selection
We selected study sites by digitizing forest fragments using March 2023 Planet satellite imagery at 3 m resolution in QGIS. To focus on mature forest remnants, we selected fragments with trees >18 m tall using Global Forest Watch tree-cover height data and verified with local farmers that sites had not been fully cleared within the past 20 years. Because most forest interiors in the region were <120 m from an edge, we sampled an edge–interior gradient from 0 to 100 m.
We selected 30 sites across a gradient of landscape forest cover, including 15 cacao edges and 15 pasture edges, separated by at least 500 m to ensure spatial independence. Edge type was verified using drone imagery or field observations, and edge ages were generally at least five years old, except for one younger site. Cacao edges were mostly full-sun CCN-51 plantations, while pasture edges were used for cattle grazing and contained scattered remnant trees and palms.
Palm survey
From September to December 2023, we sampled palms at four distances from the forest edge: 0–10, 30–40, 60–70, and 90–100 m. At each distance, we established three 10 × 10 m plots parallel to the edge, resulting in 358 plots across all sites. Plots were subdivided into 5 × 5 m subplots for data collection. We recorded all adult palms, juveniles, and seedlings. Adults were defined as individuals with current or past signs of reproduction, juveniles as individuals >1 m tall without reproductive signs, and seedlings as individuals ≤1 m tall. For small-statured Geonoma and Chamaedorea species, individuals ≤30 cm were classified as seedlings and taller non-reproductive individuals as juveniles.
Microclimate
To characterize understory microclimate, we measured canopy openness, soil moisture, and soil temperature. Canopy photos were taken at the center of each subplot under overcast conditions and used to estimate Diffuse Non-Interceptance as an indicator of light availability. Soil moisture and temperature were measured during the drier season from September to early December 2023.
Files and variables
File: Trujillo_Miranda_etal_Edge_effects_on_palm_diversity.xlsx
Description:
Variables
- Site_Number: Site ID
- Edge type: Cacao and Pasture
- Distance to the edge: 0-10, 30-40, 60-70, 90-100 m
- Humidity_Soil: Soil Humidity (%)
- Temperature_Soil: Soil temperature (F)
- DIFN: Diffuse Non-Interceptance as an indicator of light availability
- Forest_Cover_600: Landscape forest cover in concentric buffer of 600 m
- Density_Seedlings: number of seedlings individuals per unit area
- Density_Juveniles: number of juveniles individuals per unit area
- Density_Adults: number of adults individuals per unit area
- Richness_Seedlings: number of species of seedlings
- Chao1_ajust_Seedlings: Adjusted Chao1 species richness for seedlings
- Evenness_Seedlings: distribution of individuals among species, E1=1D/0D; (Hill, 1973) for seedlings
- Richness_Juveniles: number of species of juveniles
- Chao1_ajust_Juveniles: Adjusted Chao1 species richness for juveniles
- Evenness_Juveniles: distribution of individuals among species, E1=1D/0D; (Hill, 1973) for juveniles
- Richness_Adults: number of species of adults
- Chao1_ajust_Adults: Chao1 species richness for adults
- Evenness_Adults: distribution of individuals among species, E1=1D/0D; (Hill, 1973) for Adults
Code/software
Excel
