Code and data for: Habitat and local factors influence fish biomass recovery in marine protected areas
Data files
May 01, 2025 version files 580.30 KB
-
global_RLS_biomass_environmental_data_new.csv
179.30 KB
-
README.md
7.10 KB
-
Temperate_AUS_RLS_local_factors_data_new.csv
393.90 KB
Abstract
Well-designed and managed marine protected areas (MPAs) can have positive outcomes for reef biodiversity, but their effectiveness for conservation outcomes is also influenced by local environmental and anthropogenic factors. To assess the importance of local factors on MPA effectiveness, we compared field-collected data on total reef fish biomass from 922 sites inside and outside a network of 49 MPAs across temperate Australia using modelled predictions of biomass based on local biogenic habitat, physical environment, and anthropogenic factors. We found fish biomass was 34% greater in fully protected MPAs in temperate Australia than predicted if they were openly fished, whereas biomass in partially protected MPAs was equivalent to fished sites. Local biogenic habitat and physical environmental features significantly shaped shallow reef biomass across large spatial scales but their effects did not differ between fished and fully protected MPA sites, providing reassurance that regional habitat change inside and outside MPAs will not greatly affect relative effect sizes. These findings affirm the role of fishing in shaping fish biomass on shallow reefs across broad spatial scales and underscore the importance of strict protection from fishing. Strategic MPA design and management should consider local conditions to refine expectations, optimize fish biomass recovery, and enhance conservation outcomes.
Code and data for: Habitat and local factors influence fish biomass recovery in marine protected areas
Total reef fish biomass and associated environmental, anthropogenic, and ecological variables for reproducing the analysis in the manuscript "Habitat and local factors influence fish biomass recovery in marine protected areas."
A comprehensive description of the data included in this analysis can be found in the corresponding manuscript. Code for processing the data and reproducing the analysis in the corresponding manuscript are described under Code/Software below.
Description of the data and file structure
GENERAL INFORMATION
- Title of Dataset: Environmental, anthropogenic, and ecological data for globally-distributed shallow reef sites.
- Author Information: withheld in this addition to maintain anonymity of data
- Date of data collection: 2008-2024
- Geographic location of data collection: Global
DATA & FILE OVERVIEW
-
Description of dataset
-
File List:
File 1 Name: global_RLS_biomass_environmental_data_new.csv
File 1 Description: Total reef fish biomass & Bio-Oracle broadscale environmental data for 700+ openly fished Australian Reef Life Survey shallow reef sites.File 2 Name: Temperate_AUS_RLS_local_factors_data_new.csv
File 2 Description: Total reef fish biomass, broadscale environmental information, and local anthropogenic, physical environment, and biogenic habitat data for 922 Reef Life Survey shallow reef sites inside and outside MPAs.
METHODOLOGICAL INFORMATION
In summary:
Fish Biomass and Biogenic Habitat: collected via underwater visual census using a 50m transect. Detailed methods are available on the Reef Life Survey methods page: https://reeflifesurvey.com/methods/ and described in *Reef Life Survey: Establishing the Ecological Basis for Conservation of Shallow Marine Life* (Edgar et al., 2020, Biological Conservation).
Environmental Variables: broad-scale, remotely sensed environmental covariates were obtained from Bio-Oracle (https://www.bio-oracle.org/index.php), representing surface mean values at each site (2000–2020), extracted from rasters with a native resolution of 5-arcmins.
Biogenic Habitat data: biogenic habitat was assessed using photo quadrats and in-situ quadrats. Data was standardized across methods to aggregate into five categories: turfing algae, sessile invertebrates, sand, canopy-forming macroalgae, and understory-forming algae.
Physical Environmental Conditions: Wave exposure, currents, slope, and relief were scored on a scale of 1–4 by divers familiar with each site.
Human Gravity: calculated using the gravity of human impact model described in *Gravity of Human Impacts Mediates Coral Reef Conservation Gains* (Cinner et al., 2018, PNAS).
DATA-SPECIFIC INFORMATION FOR: global_RLS_biomass_environmental_data_new.csv
- Number of variables: 14
- Number of cases/rows: 724
- Variable List:
total_biomass_log: total reef fish biomass recorded using Underwater Visual Census methods at a given Reef Life Survey site on the last available survey (log10 transformed).
latitude: latitude in decimal degrees of reef site
longitude: longitude in decimal degrees of reef site
sst_mean: Mean sea surface temperature (SST) at the site over the two years preceding the date of survey
KDPAR_mean_mean: diffuse attenuation at the surface in meters.
PAR_mean_mean: photosynthetically available radiation (Em2/day).
chl_mean: chlorophyll-a (mmol / m3)
sws_mean: sea water velocity (m/2)
dfe_mean: iron (mmol / m3)
no3_mean: nitrate (mmol / m3
po4_mean: phosphate (mmol / m3)
phyc_mean: phytoplankton (mmol / m3)
so_mean: salinity
si_mean: silicate (mmol / m3)
DATA-SPECIFIC INFORMATION FOR: Temperate_AUS_RLS_local_factors_data_new.csv
- Number of variables: 35
- Number of cases/rows: 922
- Variable List:
site_code: unique Reef Life Survey shallow reef site identifier
survey_id: unique survey identifier
survey_date: date of survey (dd/m/yyyy)
survey_governance: level of protection against fishing at a given site at the time of survey: fished (openly fished, no restrictions on fishing);
restricted (some restrictions on fishing present); no-take (no extractive/ fishing activities permitted).
total_biomass_kg: total reef fish biomass in kilograms recorded using Underwater Visual Census methods at a given Reef Life Survey site on the last available survey.
depth: mean depth in meters of survey transect
canopy: percent coverage of canopy forming-macroalgae along the survey transect
understorey: percent coverage of understorey-forming algae along the survey transect
sand: percent coverage of sand along the survey transect
sessile_invertebrates: percent cover of sessile invertebrates along the transect
turf: percent cover of turfing alga along the transect
wave_exposure: degree of oceanic swell a site receives on average: 1, sheltered; 2, maximum wave height 1-3 m; 3, ocean swell maximum >3 m; 4, open swell from prevailing direction.
relief: mean vertical relief along the transect line: 1, <0.5 m; 2, 0.5 – 1 m; 3, 1 – 2 m; 4, >2 m.
slope: mean gradient along the transect line: 1, <1:10 gradient; 2, 1:10 - 1:4; 3, 1:4 - 1:2; 4, >1:2.
currents: mean relative strength of the currents at a given site: 1, none; 2, weak; 3, moderate; 4, strong.
latitude: latitude in decimal degrees of reef site
longitude: longitude in decimal degrees of reef site
sst_mean: Mean sea surface temperature (SST) at the site over the two years preceding the date of survey
KDPAR_mean_mean: diffuse attenuation at the surface in meters.
PAR_mean_mean: photosynthetically available radiation (Em2/day).
chl_mean: chlorophyll-a (mmol / m3)
sws_mean: sea water velocity (m/2)
dfe_mean: iron (mmol / m3)
no3_mean: nitrate (mmol / m3
po4_mean: phosphate (mmol / m3)
phyc_mean: phytoplankton (mmol / m3)
so_mean: salinity
si_mean: silicate (mmol / m3)
distance: distance from the nearest shoreline to each site in meters
distance_from_shore_m: distance from the nearest shoreline to each site in meters - log10 transformed
total gravity: a measure of how large and far away a human population is to a given reef site, as detailed in Cinner et al., 2018.
grid_cell: 100km by 100km grid cell each site falls inside
year: year of survey (yyyy)
CODE/ Software
Two R script files are provided to reproduce the analysis in the corresponding manuscript. All provided R files are described below.
"1. Random Forest model.R": R source code predicting total reef fish biomass across shallow temperate Australian reef sites of interest using a global random forest (RF) model of total reef fish biomass and broadscale environmental conditions.
"2. Multimodel inference.R": R source code estimating the relative importance of local anthropogenic, biogenic habitat, and physical environmental factors for differences in reef fish biomass using Generalized Linear Mixed Models (GLMMs) in a multi-model inferences framework.
Data include total reef fish biomass in kilograms for globally-distributed shallow reef sites surveyed between 2008 and 2023. Environmental, anthropogenic, and ecological variables characterizing the conditions for each site-year combination are also provided.
