Determining the Role of Environmental Covariates on Planktivorous Elasmobranch Population Trends within an Isolated Marine Protected Area

Abstract

Several studies have found predictable relationships between the behavior of planktivores and environmental conditions, suggesting that planktivores may be especially sensitive to environmental change. However, many studies to date are based on limited observations, include few of the many environmental covariates which could influence planktivores, and do not occur over long enough time periods to make inferences about the potential effects of environmental change. As such, long term datasets on planktivores are necessary to disentangle the potential impacts of oceanographic and environmental variability. In this study, to elucidate the relationship between plankivores and environmental variability, we use data obtained over the last 28 years by a small group of divemasters at Cocos Island, Costa Rica, one of the oldest marine reserves in the world. We found that, in general, for planktivorous elasmobranchs, several environmental variables, such as, chlorophyll A, lunar cycle, and salinity have clear influences on their occurrence and relative abundances. We found that in the phases of lower illuminations, there were significant increases in abundance of mobula rays. Specifically, a 0.10 mg/m3 increase in Chlorophyll A correlated with 26% decrease in whale sharks. We found that increases in salinity correlated with increases in mobula abundance but did not correlate with observations of mantas or whale sharks. We also found that omission of environmental covariates can lead to overprediction and underprediction of relative abundances. Our findings highlight the need to take environmental conditions into account when evaluating the efficacy of marine protection.

Assessing the effects of warming and carbonate chemistry parameters on marine microbes in the Gulf of Mexico through basin-scale DNA metabarcoding

“…We used generalized linear models (GLMs) with either Poisson or negative-binomial error distributions to identify potential explanatory variables of major 16S and 18S taxonomic groups in the GOM. GLMs account for multiple predictor variables (factors) and have been applied to ecological count (and proportional) data of higher trophic level marine organisms (60,61). Here, we applied GLMs to microbial metabarcoding data, allowing us to observe predictor variables and their relation to group-specific relative abundance measured spatially in the photic zone (Fig.…”