Temperature control of larval dispersal and the implications for marine ecology, evolution, and conservation

Abstract

Temperature controls the rate of fundamental biochemical processes and thereby regulates organismal attributes including development rate and survival. The increase in metabolic rate with temperature explains substantial among-species variation in life-history traits, population dynamics, and ecosystem processes. Temperature can also cause variability in metabolic rate within species. Here, we compare the effect of temperature on a key component of marine life cycles among a geographically and taxonomically diverse group of marine fish and invertebrates. Although innumerable lab studies document the negative effect of temperature on larval development time, little is known about the generality versus taxon-dependence of this relationship. We present a unified, parameterized model for the temperature dependence of larval development in marine animals. Because the duration of the larval period is known to influence larval dispersal distance and survival, changes in ocean temperature could have a direct and predictable influence on population connectivity, community structure, and regional-to-global scale patterns of biodiversity.

Fig. 1.

Caterpillar plots comparing ranked species-level residuals (random effects) for 72 species along with 95% confidence intervals, for two of the three level-1 parameters. Confidence intervals that do not intersect zero identify species whose species-specific value for that parameter is significantly different from the corresponding population-averaged value. The caterpillar plot graphically identifies those species poorly represented by the population-averaged model (see SI Appendix). (A) Predictions and 95% confidence intervals (black triangles and gray error bars) for the random effect component (u_1i) of the linear scaling parameter β_1i for each species (Eq. 15). Confidence intervals do not include 0 for seven species (red points): L. polyphemus, C. tyrrhena, H. americanus, G. morhua, S. spirorbis, S. balanoides, and L. californianus. After removing the three most-deviant outliers, L. polyphemus, L. californianus, and C. tyrrhena, there is no longer a need for random effects for the linear and quadratic scaling parameters. (B) Caterpillar plot for species-level residuals u_0i. Because the majority (46 of 72) of the confidence intervals fail to include 0, we conclude that the species-specific intercept parameters β_0i are significantly different from the population-averaged value β₀ for most species. No adjustments for multiple testing were made.

Fig. 2.

The relationship between water temperature and PLD based on results from published experimental laboratory studies on the effect of temperature on larval duration for 72 species (six phyla: 6 fish, 66 invertebrates; SI Tables 3 and 4). (A) Mean recorded larval duration at each temperature for each species; two to six data points per species connected by gray lines. Subsequent analyses identified three outliers (black diamonds). (B) Population-averaged (black) and species-specific (gray) trajectories obtained from a multilevel exponential model quadratic in temperature on a log–log scale with random intercepts displayed here on an arithmetic scale. Estimated population-averaged curve: ln(PLD) = 3.17 − 1.34 × ln(T/T_c) − 0.28 × (ln(T/T_c))², which yields the plotted estimated geometric mean curve: PLD = exp(3.17) × (T/T_c)^{(−1.40−0.27×ln(T/Tc))}, T_c = 15°C (SI Appendix). The parameter estimates β₁ = −1.34 and β₂ = −0.28 adequately describe 69 species, whereas β₀ is highly variable among species (see SI Text for model application). Shown here is the population-averaged trajectory for PLD about which individual species-level trajectories are assumed to vary randomly. β₀ = 3.17 is interpretable as the value of ln(PLD) at 15°C. Three outliers were excluded in estimating the model (data not shown); dashed lines represent the 95% confidence band for the population-averaged trajectory.

Fig. 3.

Arrhenius plot of Universal Temperature Dependence model (Eq. 3) for within-species variation in PLD with temperature (n = 72). Temperature (°C) is expressed as its reciprocal adjusted to Kelvin and multiplied by the Boltzmann constant (k). Population-averaged trajectory for the temperature effect within species as estimated from a multilevel model with random slopes and intercepts: ln(PLD) = −22.47 + 0.64 × (1/(k × (T + 273))) for temperature (T) in °C (solid line), or PLD ∝ exp(0.64/(k × (T + 273))). The model-based empirical Bayes trajectories shown here differ from the ordinary least-squares-fitted trajectories that would be obtained from fitting individual temperature-dependence models to each species one species at a time (SI Appendix). Metabolic theory predicts that on average the slope is 0.62 eV (5) (dashed line) and within the range 0.60–0.70 eV (2). As with the linearized power law model, a random slopes and intercepts UTD model is required for this data set of 72 species (SI Table 9).

Fig. 4.

Effect of climate and developmental mode on the temperature dependence of PLD for 69 species. We used mean ln(test temperature) for each species as a proxy for the average temperature in each species' geographic range. The best model among those we examined was one in which the random intercepts model (Eq. 4) was extended to allow ln(PLD) to vary additively with mean ln(test temperature) and developmental mode (SI Table 6). In the multilevel modeling framework, these two species-level variables are considered predictors of the species-specific intercept, β_0i. In the centered level-1 model presented here (SI Table 7), this intercept is interpretable as ln(PLD) at 15°C. The predicted intercepts from a random intercepts multilevel model (Eq. 4) are plotted here against mean ln(test temperature) (Left) and developmental mode (Right). (Left) The lowess (solid curve) and linear trend (dashed line) suggest that larvae tested at colder temperatures tend to have smaller predicted intercepts than do larvae tested at warmer temperatures. (Right) Schematic boxplots, following standard conventions for such graphs, of predicted intercepts for each developmental mode are displayed, with means indicated by asterisks. Lecithotrophs (L, filled circles) tend to have smaller predicted intercepts than do planktrophs (P, open circles).

Fig. 5.

The predicted effects of ocean temperature on two important ecological and evolutionary parameters: larval dispersal distance (A) and larval survival (B). The predicted effect on dispersal distance is based on our population-averaged temperature-PLD model (Fig. 1 B) and on a published model relating PLD to dispersal (25) that used mean current velocity (U) = 0 cm/s and with standard deviation (s) = 15 cm/s to reflect typical near-shore coastal ocean currents. Species-specific projections are shown (gray lines) to convey the range of variability. Confidence band (95%) is for prediction of mean temperature effect on PLD, as in Fig. 1 B. Predicted effects on cumulative survival assume a constant density- and temperature-independent daily mortality rate of 15% (18).