Management regime influences the recovery of subtidal kelp forests following harvesting
Data files
Jul 18, 2025 version files 35.06 KB
-
Alldata_NERC.csv
29.18 KB
-
README.md
5.88 KB
Abstract
Physical disturbances, especially those that reduce or remove foundation species, have complex effects on ecosystem functioning. Evaluating resilience to predict these impacts requires long-term research over large temporal and spatial scales. This study investigates the recovery of the subtidal kelp (Lessonia trabeculata) to simulated kelp harvesting in areas under different fisheries management regimes (Territorial Use Rights for Fisheries (TURF) versus Open Access (OA) sites) along the Northern-Central Chilean coast. The managed TURF sites demonstrated significantly faster recovery of adult kelp densities following harvesting, while the less OA sites exhibited consistently lower kelp densities and delayed recovery. However, the generality of this pattern varied across regions. Morphological traits such as holdfast diameter were larger at TURF sites, where kelp exhibited larger size. Delayed size recovery was evidenced at OA sites. Herbivore densities were higher in OA than in TURF sites, potentially contributing to differences in kelp density, size and recovery. These patterns highlight the interaction between herbivory and management efficacy. Synthesis and applications: Our findings highlight the critical role of local fisheries management in enhancing kelp recovery after harvesting. The faster recovery of kelp at most TURF sites demonstrates how spatially explicit management can buffer ecological impacts of herbivore density and promote kelp recovery. To improve outcomes, managers should prioritize the implementation and enforcement of management plans in the TURF to promote herbivore control measures and invest in long-term monitoring programs to support adaptive strategies that sustain kelp forests ecosystems and the valuable services they provide to coastal communities.
Description of the Data and File Structure
This dataset documents the post-disturbance dynamics of kelp forest recovery across different fisheries management regimes along the northern-central coast of Chile. Simulated harvesting of adult Lessonia trabeculata was conducted at TURF (Territorial Use Rights for Fisheries) and Open Access (OA) sites, and recovery was monitored using ecological and environmental metrics over a 42-month period.
The dataset is stored in a single CSV file named: Alldata_NERC.csv
This represents means of four replicates. This file contains observations from multiple time points across replicate patches within managed and unmanaged kelp forests. It includes data on kelp density and morphology, herbivore abundance (by functional group and species), bite rates, and derived metrics like fish biomass and maximum abundance (Nmax).
The data are structured in long format, with each row representing a unique sampling event at a given patch and time.
Variable Descriptions
| Variable | Description |
|---|---|
| Site | Sampling site name |
| MAE | Months After Experiment (time since harvesting), in months |
| Zone | Biogeographic zone (e.g., Northern-Central) |
| Management | Management regime (TURF or OA) |
| Patch | Replicate patch ID within each site |
| TotalDens | Total kelp density (adults + juveniles), individuals per square meter |
| AdultDens | Density of adult kelp, individuals per square meter |
| JuvenileDens | Density of juvenile kelp, individuals per square meter |
| Scartichthys spp. | Density of herbivorous fish from genus Scartichthys, individuals per square meter |
| Taliepus spp. | Density of herbivorous crab species (e.g., Taliepus), individuals per square meter |
| Tegula spp. | Density of herbivorous snails (Tegula species), individuals per square meter |
| Tetrapygus niger | Density of sea urchins (Tetrapygus niger), individuals per square meter |
| TotalHerb | Total herbivore density (fish, snails, urchins, crabs), individuals per square meter |
| TotalBites | Total observed bite rate (from video), bites per minute |
| HerbBites | Bite rate attributable to herbivorous fish, bites per minute |
| CarnBites | Bite rate attributable to carnivorous fish, bites per minute |
| TotalNMax | Maximum number of fish observed simultaneously (Nmax), total |
| CarnNMax | Maximum number of carnivorous fish observed (Nmax) |
| HerbNmax | Maximum number of herbivorous fish observed (Nmax) |
| TotalBiomass | Estimated total fish biomass in grams (g) |
| CarnBiomass | Estimated carnivorous fish biomass in grams (g) |
| HerbBiomass | Estimated herbivorous fish biomass in grams (g) |
| Holdfast | Average holdfast diameter of kelp, in centimeters (cm) |
| Stipes.number | Average number of stipes per kelp individual |
| TL_stipes | Average total length of kelp including stipes, in centimeters (cm) |
| TL | Total length of kelp, in centimeters (cm) |
| Unnamed: 0 | Row identifier (can be ignored or removed) |
Notes on Missing Data
Cells with 'NA' represent missing values, which were primarily due to environmental or logistical issues that prevented data collection (e.g., adverse weather, equipment malfunction, or diver safety restrictions). These entries indicate 'not available' data points rather than 'not applicable' ones.
Methods Summary
Simulated harvesting consisted of manually removing adult Lessonia trabeculata from designated patches. Recovery monitoring included:
- Visual census surveys and quadrats for kelp density and morphology
- Remote Underwater Video (RUV) analysis for estimating herbivorous and carnivorous fish activity (Nmax, bite rates)
- Identification and enumeration of benthic herbivores (snails, crabs, urchins)
- Regular sampling intervals from time 0 to 42 months post-harvest
- Comparison between management regimes (TURF vs OA) and among the three biogeographic zones
Code and Software
Data processing, analysis, and visualization were performed using R (version 4.x) in RStudio. The following packages were used:
- tidyverse (Wickham et al. 2019): data manipulation and plotting
- mgcv: Generalized Additive Models (GAMs)
- lme4: mixed-effects modeling
- glmmTMB: generalized linear mixed models with overdispersion
- emmeans: post-hoc contrasts and marginal means
- patchwork: plot composition
- readr: data import
- here: reproducible paths
- janitor: data cleaning
Scripts
Different R scripts were used in the manuscript and are available upon request to aperez@bio.puc.cl. Please contact the corresponding author to request access.
