Fig 1.
Experimental Temperature Manipulation.
Solid lines represent the mean temperature in experimental mesocosms (red = elevated; black = control; n = 8 for each). The dotted line represents ambient temperature in a nearby eelgrass meadow. Boxes indicated the time periods when plant morphology, productivity, and physiology were analyzed.
Fig 2.
Mean cumulative shoot production by count (a) and biomass (b) over a 15 week experiment. During the acclimation and recovery period, growing conditions did not vary between treatments, but during the warming period, plants in the warm treatment were grown in temperatures elevated by 4.5°C. Means are of 7 genotypes (n = 8 per genotype), and error bars represent standard error.
Table 1.
Results of the mixed model split plot ANOVA exploring the effects of genotype, time (acclimation, treatment, recovery), and interactions on new shoot production in plants in control mesocosms (never exposed to elevated temperature).
Additional response variables (photosynthetic physiology, nutrient uptake and leaf growth shown in Table B in S2 File).
Fig 3.
(a) Cumulative shoot production (counts) for 7 distinct Zostera marina genotypes over time. Dots are means of 8 replicates and error bars represent standard error. All plants were grown under ambient conditions and time periods were separated by 5 weeks. (b) An additional set of plants were grown under ambient conditions for 5 weeks (black bars), under temperature elevated by 4.5°C for 5 weeks (light grey bars), and again at ambient temperature for 5 weeks (dark grey bars). Bars represent the difference in shoot production from controls (always grown under ambient conditions), and error bars are standard error.
Table 2.
Results of the mixed model split plot ANOVA exploring the effects of genotype, temperature, time, and interactions on new shoot production.
Results for additional response variables (plant physiology, morphology, and growth) are in Table C in S2 File. The warming treatment was a simulated 5 week event with temperature elevated by 4.5°C. There were 2 sampling time periods: immediately following the warming event and following a 5 week recovery when both treatments were exposed to ambient temperature.
Fig 4.
Initial growth (between the acclimation period and the warming period from control plants that were not influenced by elevated temperature) is negatively correlated with growth during the recovery phase of previously warmed plants, as measured by the difference in shoot counts (m = -0.5) (a) and by biomass (m = -0.5) (b). Each point is the mean for a particular genotype.
Table 3.
Multiple Step-wise Regression Approach used to assess the relationship between traits of genotypes measured at the beginning of the experiment (acclimation period) and change in shoot production due to elevated temperature.
We estimated new shoot production two ways: (a) difference in number of new shoots between elevated and the mean of control genotypes at the end of the recovery period and (b) difference in number of new shoots produced under elevated temperature at the end of the warming period vs. the end of the recovery period.