Abstract
Seagrass meadows are valued globally for their ecosystem services, including their role as a ‘blue carbon’ sink due to high rates of primary production and carbon burial in the sediment. However, seagrasses are threatened by climate change and other natural and anthropogenic stressors, and their ecosystem services risk being lost as well. In some cases, seagrass declines have led to the release of previously stored carbon in the form of greenhouse gases, creating a positive feedback on the climate system. To ensure the success of seagrass conservation and restoration and avoid adverse effects on the climate system, it is essential to better understand the role of seagrass meadows in the global carbon cycle, what drives their metabolism, and their resilience to climate change (e.g. globally increasing temperatures, extreme heating events, sea level rise). These questions have traditionally been addressed via lab, mesocosm, or in situ experiments that do not capture the full range of environmental fluctuations and are difficult to translate to ecosystem-scale dynamics. In this dissertation, I used the relatively novel non-invasive aquatic eddy covariance (AEC) technique in a restored seagrass (Zostera marina) meadow at the Virginia Coast Reserve (VCR) to measure seagrass ecosystem metabolism under naturally varying environmental conditions. These measurements have been ongoing since 2007, and captured an eelgrass die-off event during summer 2015, followed by a slow recovery in the following years. This long-term, high-quality AEC dataset presented a unique opportunity to study the long-term trends and resilience of seagrass metabolism, as well as quantify the response of a seagrass ecosystem to extreme heat stress and evaluate its ability to adapt to future changes in water temperatures and light availability.
I found that seagrass metabolism was highly dynamic and variable, with as much variation within a single month as during an entire year, thus indicating rapid internal carbon cycling. Over the 11-year period, however, seagrass metabolism was generally balanced, despite shifts in trophic status during and after the die-off event, with the meadow becoming heterotrophic the summer of the die-off, and autotrophic during its recovery. Long-term water temperature records indicated a warmer growing season in 2015 compared to other years. Using a modeling approach on hourly AEC fluxes, I found that water temperatures above a threshold of 28.6°C negatively impact eelgrass metabolism, causing a 50% decrease in daytime oxygen fluxes when only exceeded by ~2°C. This was the first time this threshold was determined based on in situ data. Based on this threshold and in situ water temperature measurements from 2016–2019, I developed two metrics to quantify thermal stress in seagrass meadows, providing a framework for understanding thresholds for seagrass survival. These metrics were the cumulative heat stress (as heating degree-hours, HDHs) and thermal stress relief (as cooling degree-hours, CDHs), both relative to the 28.6°C eelgrass thermal tolerance threshold. I compared these metrics to spatiotemporal patterns in summertime seagrass shoot density and length, and found that the healthiest parts of the meadow benefited from greater thermal stress relief (2–3x) from tidal cooling (inputs of cooler ocean water) during periods of heat stress, leading to ~65% higher shoot densities compared to the center of the meadow, which experienced higher heat stress (1.8x) with less relief. I also found that the eelgrass die-off event in 2015 was triggered by heat stress cumulating ~100–200°C-hours in June. This heat stress was later amplified by the effects of sulfide toxicity into seagrass tissues, as indicated by the sulfur isotope signatures of seagrass samples that indicated higher sulfide intrusion into seagrass tissue in 2015 compared to other years. Finally, I estimated the light-use efficiency (LUE) of the eelgrass meadow based on AEC data and in situ light measurements—providing the first ecosystem-scale and in-situ based LUE estimate for eelgrass. LUE was low compared to previous estimates, averaging 0.004–0.005 O2 photon-1 and potentially reflecting nutrient-limited primary production. Hourly pattern in LUE, however, suggested eelgrasses can regulate photosynthesis in response to changes in light availability, which might aid in their resilience to future environmental change.
Overall, this dissertation leveraged the benefits of long-term AEC measurements to offer new insights on assessing the trophic status of seagrass ecosystems and how it may change in response to disturbance events. We provided the first in situ, whole-ecosystem-based estimates for eelgrass LUE and thermal stress thresholds, further advancing scientific research on ecosystem responses to future climate change scenarios.
