Comment on “Demographic dynamics of the smallest marine vertebrates fuel coral reef ecosystem functioning”
Marine Ecology

Comment on “Demographic dynamics of the smallest marine vertebrates fuel coral reef ecosystem functioning”

Abstract

Brandl et al. (Reports, 21 June 2019, p. 1189) report that cryptobenthic fishes underpin coral reef ecosystem function by contributing ~60% of “consumed fish” biomass and ~20% of production. These results are artifacts of their simulation. Using their data and model, we show that cryptobenthic species contribute less than 4% to fish production, calling into question the extent to which they contribute to the high productivity of coral reefs.

A long-standing question in coral reef ecology is “How can reefs be so productive while residing in such nutrient-poor environments?” Brandl et al. (1) estimated the contribution of cryptobenthic fishes, an elusive and underappreciated group, for total coral reef fish production. They provide interesting and novel perspectives on the unique life history characteristics of cryptobenthics, showing that they have disproportionately high reproductive and larval supply rates, but also high mortality, relative to other major reef fishes. On this basis, the authors suggest that cryptobenthics may help to fuel the rapid nutrient- and energy-recycling rates that typify coral reefs (2, 3). Their main finding was that as a result of this high mortality rate, cryptobenthic fishes contribute ~60% of the “consumed fish” biomass in coral reefs (and ~20% of fish production), and thereby represent a large and previously unknown source of energy that supports coral reef fish production.

We applaud the authors for bringing to light these important life history traits, but we argue that their estimates of the contribution of cryptobenthic fish to fish production are an artifact of their analyses. Brandl et al. used a numerical simulation of larval supply and growth to calculate the relative production of cryptobenthic species after a single year. In doing so, they neglect the contributions of fish that live more than 1 year, thus vastly overestimating the contributions of cryptobenthic fishes to fish production on coral reefs.

The stated goal of the study was to estimate the extent to which cryptobenthic fish species “promote internal reef fish biomass production” (i.e., secondary production). Instead of focusing on the contribution of cryptobenthics to secondary production, defined as the biomass accumulated by a population per unit time (4), Brandl et al. focused primarily on quantifying their contribution to “consumed biomass,” which they define as all biomass lost to mortality (i.e., all fish that die are consumed) (1). Consumed biomass is an inherent component of production, as it represents biomass produced in previous time steps that, once eaten, will not continue to produce biomass. But it is not a good proxy for production. Consider a population with a new cohort of three individual recruits that die over time (Fig. 1A). At time t1, the contribution to production by all individuals is the same; only the black individual has died and thus contributes 100% to consumed biomass. By t3, the red individual contributes much more to production because it has not died, but has contributed nothing to consumed biomass. When all fish die, production and consumed biomass are equal. This conceptual figure illustrates two things: (i) Consumption is not a good proxy for production because fish that are not consumed still contribute to secondary production, and (ii) the timing at which estimates are measured is critical to estimates of both production and consumption.

Fig. 1 Conceptual models illustrating the importance of timing for calculating secondary production.

(A) Calculating secondary production (i.e., the production of new biomass) following individual cohorts through to death. X indicates when an individual dies; O indicates that it is alive and included in production. The inset shows the total biomass produced and consumed, as well as the percent contributions of each individual to produced and consumed biomass per time step [Prod/Cons (%)]. This illustrates the importance of allowing all individuals to die before calculating production and highlights that considering only those individuals that die at certain time steps is not a useful proxy for production. (B) Calculating secondary production following multiple cohorts until all individuals from the initial cohort die. At each time step, a new cohort enters the system (represented by different rows, with the size of the circle indicating the number of individuals remaining over time). The colored portions of each circle represent the proportion of individuals belonging to cryptobenthic (blue) and non-cryptobenthic (orange) fishes. The proportion of cryptobenthic fishes (blue) diminishes rapidly because of their high mortality rate. Brandl et al. (1) calculated production for the first year only (blue box). Doing so inflates the production of cryptobenthics because it omits the additional contribution to production by all other cohorts in the community that survive beyond that time period. In essence, their model assumes that the larval recruits are the only fishes on a reef that contribute to production, as if there were no fish present starting at day 0 in their simulation. This has a greater effect on inflating consumed biomass because of the disproportionate amount of cryptobenthic larvae that die shortly after settlement (death is assumed to be by consumption).

” data-hide-link-title=”0″ data-icon-position=”” href=”https://science.sciencemag.org/content/sci/366/6472/eaay9321/F1.large.jpg?width=800&height=600&carousel=1″ rel=”gallery-fragment-images-113096322″ title=”Conceptual models illustrating the importance of timing for calculating secondary production. (A) Calculating secondary production (i.e., the production of new biomass) following individual cohorts through to death. X indicates when an individual dies; O indicates that it is alive and included in production. The inset shows the total biomass produced and consumed, as well as the percent contributions of each individual to produced and consumed biomass per time step [Prod/Cons (%)]. This illustrates the importance of allowing all individuals to die before calculating production and highlights that considering only those individuals that die at certain time steps is not a useful proxy for production. (B) Calculating secondary production following multiple cohorts until all individuals from the initial cohort die. At each time step, a new cohort enters the system (represented by different rows, with the size of the circle indicating the number of individuals remaining over time). The colored portions of each circle represent the proportion of individuals belonging to cryptobenthic (blue) and non-cryptobenthic (orange) fishes. The proportion of cryptobenthic fishes (blue) diminishes rapidly because of their high mortality rate. Brandl et al. (1) calculated production for the first year only (blue box). Doing so inflates the production of cryptobenthics because it omits the additional contribution to production by all other cohorts in the community that survive beyond that time period. In essence, their model assumes that the larval recruits are the only fishes on a reef that contribute to production, as if there were no fish present starting at day 0 in their simulation. This has a greater effect on inflating consumed biomass because of the disproportionate amount of cryptobenthic larvae that die shortly after settlement (death is assumed to be by consumption).”>

Fig. 1 Conceptual models illustrating the importance of timing for calculating secondary production.

(A) Calculating secondary production (i.e., the production of new biomass) following individual cohorts through to death. X indicates when an individual dies; O indicates that it is alive and included in production. The inset shows the t

Here's the Original Article