Effect of environmental predictors on zooplankton diversity
As previously observed for the northern GoM, environmental conditions in the southern GoM “were comparable among years, but more variable in space” (Elliott, Pierson, & Roman, 2012). Indeed, similar patterns of zooplankton spatial aggregations over-time allowed us to combine the genetic information of each field expedition in order to explore the spatio-temporal variability in the zooplankton community as a whole in a comprehensive analysis.
Our findings suggest that all tested environmental predictors may promote changes in the structure of the zooplankton community, even with low contribution percentages. The simplest explanation for such a poor relationship is that other factors, which were not taken into consideration (e.g., nutrient variation and irradiance, among others), could also be involved in shaping the structure of the zooplankton community of the study area. Moreover, biotic interactions within a zooplankton community, such as predation or symbiotic associations, may also shape the structure of the community (Gasca, Suárez-Morales, & Haddock, 2007; Hereu et al., 2020; Jennifer E. & Mary N., 2001). However, alternative and/or complementary possibilities may be also proposed. In this regard, it has been previously reported that environmental predictors generally show patchiness (Espinasse, Carlotti, Zhou, & Devenon, 2014; Gluchowska et al., 2017; Trudnowska, Gluchowska, Beszczynska-Möller, Blachowiak-Samolyk, & Kwasniewski, 2016; Usov et al., 2019), which hampers pattern assessments over a continuous space (Helenius, Leskinen, Lehtonen, & Nurminen, 2017; Usov et al., 2019). This may become particularly relevant in studies that consider large-scale geographic areas or the open ocean, in which environmental gradients usually show three-dimensional reorganizations (Usov et al., 2019). Thus, multiple factors like winds, currents, and water mass movements, among others, may disrupt the linearity of hydrological boundaries (Espinasse et al., 2014; Usov et al., 2019). In our study, the discontinuity of hydrological patterns mainly challenged the cluster organization of the sampling stations. Indeed, even stations that were geographically close to one another presented dissimilarities in environmental profiles that ranged from small to large. These divergences mostly concerned stations located in the central portion of the sampling area (Lines C, D, and E), suggesting that this area may be considered a transitional region separating the northern and southern ecoregions.
Finally, the low strength of the correlations between abiotic and biotic factors may be also explained by considering the experimental design. One of the aims of this study was to relate the environmental variables with the structural patterns of the zooplankton community. Accordingly, the environmental data were evaluated in the context of the zooplankton community, which may have hidden taxa-specific ecological niche partitioning. For example, the abundance of a given taxon may increase in response to a specific combination of environmental conditions, but simultaneously, the same conditions may provoke a decrease in the abundance of other taxa (Elliott et al., 2012). Hence, the observed low environmental contribution percentages may have been the result from opposing taxa-specific responses to the same environmental cues. If validated, this supposition may be used to re-evaluate the implications of the observed correlations. In this context, we posit that the inclusion of additional predictors into the model may not necessarily result in a strengthening of the correlations between biotic and abiotic variables, as some taxa will likely respond positively to a new predictor while others may respond negatively. Nonetheless, additional research is needed to support or reject this hypothesis.