2019 Moorea MPA Commercial Reef Fish Count and Length Data
Description
Data was taken to assess the hypotheses that marine protected area boundaries on Moorea, French Polynesia contain a higher abundance of commercial reef fishes and a difference in size structure between populations of commercial reef fishes in MPAs when compared to fishable control sites. A control site was first selected based on location, proximity to the paired MPA, accessibility without aquatic motorized vehicles, and similarity of conditions including coral cover, substrate type, depth, and surface area. All MPAs were surveyed except for MPA Maatea, which is inaccessible without a boat due to the presence of a deep channel. The MPAs are well marked, including signage on the road and on the waterfront. This information was used to determine the location of each MPA in the field. 60 surveys were conducted at each MPA-control pair with 30 surveys at the MPA and 30 at the control. Surveys were each 25 m by 2 m transects and the surveyor covered 5 m per minute. The surveyor observed the space directly below as well as the water column, including a few meters ahead. The survey was repeated at a location far enough away that it was unlikely the same fish from the previous survey were recounted. All surveys were conducted at the barrier reef, since not all MPAs and control sites have a fringing reef to survey, and surveys at the outer slope require SCUBA diving for accurate counts. Additionally, the barrier reef comprises a large portion of the lagoon where many people actively fish, as observed when conducting field work. For each survey, all individuals were identified to genus and the size of each individual was recorded in one of four size classes: 0 – 9 cm, 10 – 19 cm, 20 – 29 cm, and ≥30 cm. All statistical analyses were conducted in R (R Core Team 2019) as implemented in R Studio (R Studio Team 2018). Abundance data was first analyzed with generalized linear mixed models in R package lme4 (Bates et al. 2015) equivalent to a paired t-test where location is a random effect. Surveys represent a random subsample of each site where survey position was randomized and not expected to have an effect on abundance observed. This model was first separated by genus, then used total number observed as a response variable and status as an MPA or a control as the explanatory variable where prohibition of fishing serves as the experimental treatment. A Poisson link function was used in the model. Alpha was set equal to 0.05 for analysis. A full model prohibition of fishing in an MPA serving as a treatment was then compared to a reduced model without an effect of treatment using a chi-squared goodness of fit test in R package lme4. Alpha was set equal to 0.05 for analysis. Size data was separated by genus and summarized for all MPAs. Size data was analyzed with a Chi-Squared goodness of fit test in lme4 to assess differences in distribution of size classes between MPAs and controls. Alpha was set equal to 0.05.