Land-atmosphere carbon dioxide and methane exchange in a temperate salt marsh
| Author(s) | Hill, Andrew C. | |
| Date Accessioned | 2023-05-03T15:44:54Z | |
| Date Available | 2023-05-03T15:44:54Z | |
| Publication Date | 2023 | |
| SWORD Update | 2023-03-22T16:05:17Z | |
| Abstract | Within temperate regions, the terrestrial-aquatic interface is a critically important zone for global carbon cycling. Much of this importance is tied to high rates of carbon transformation and transport across system boundaries within coastal wetlands. In particular, coastal salt marshes act as biogeochemical hot spots where land-atmosphere carbon fluxes of CO2 and CH4 are often disproportionately high and therefore of increasing interest due to high uncertainties for annual budgets, biophysical drivers, and potential responses to environmental change. Unfortunately, these ecosystems have received far less attention compared to terrestrial or oceanic systems, thus we only have limited information about ecosystem characteristics which restricts our ability to create generalizations about important functional relationships governing fluxes and limits inclusion in many large-scale models. Despite the recent uptick in studies focused on tidal wetlands, we still require more studies within salt marshes to help address numerous challenges associated with characterizing vertical carbon fluxes. This dissertation focuses on improving how we monitor, measure, and track changes in ecosystem carbon dynamics to better understand the patterns and processes that influence carbon fluxes within an understudied and underrepresented ecosystem. ☐ The first study in this dissertation addresses challenges associated with spectral tracking and determining key carbon phenology dates with near-remote sensing approaches where there is limited spatial land coverage for traditional space-borne approaches. Several near remote sensing instruments including a phenocam and spectral reflectance sensors, in addition to MODIS products, are used to determine which vegetation indices and spectral bands best track changes in carbon phenology based on ecosystem scale benchmarks measured with the eddy covariance method. Although there is a wealth of valuable information extracted from these sensors, the greenness index from the phenocam was found to perform best for tracking changes in daily carbon phenology and was most accurate and consistent in determining carbon uptake season start and end dates. More traditional indices containing infrared information, such as NDVI and EVI, considerably overestimated season end dates. ☐ The second study examines a variety of different gas flux measurement techniques performed across several spatiotemporal scales to determine the applications and limitations of the data generated from these methods. Measurements of net carbon exchange, productivity, respiration, and CH4 exchange were performed across all phenological stages at the plot scale using small chambers that isolated leaf and sediment fluxes, larger chambers that encompassed the vegetation canopy, and ecosystem scale using the eddy covariance method. While all measurement techniques generated similar drivers, fluxes measured from eddy covariance were more complex as information is integrated across larger expanses of time and space. When measurements were aligned in time, there was good overall agreement, yet models parameterized with chamber data overestimated CO2 exchange and underestimated CH4 exchange which could create inaccurate results for system characterization and annual carbon budgets. ☐ The third study sought to examine CH4 dynamics using empirical dynamic modeling and convergent cross-mapping as a novel nonparametric approach to disentangle the complex variable interactions and interdependencies influencing CH4 fluxes at seasonal and diurnal scales. This method is applied across a large suite of ancillary variables to determine levels of causality, optimum time lags, and the complex interconnections between variables. While many of the most important causal drivers were identified by studies applying linear approaches, these previous results provide less detailed information on true causality and important lagged relationships fail to be identified. The most important causal drivers for seasonal dynamics were temperature components of the air, soil and water, productivity, and dissolved oxygen, while level and associated variables of salinity and dissolved oxygen were most important for diurnal dynamics. Lags and near-synchronous relationships were present at all scales and in several cases the influence of a suspected driving variable was indirect given the causal connections of additional variables. ☐ This dissertation has addressed several important challenges to improve how we monitor, measure, and track changes in carbon dynamics occurring within a temperate salt marsh. This includes reconciling how near remote sensing approaches can be used to provide critical information on carbon phenology, how information extracted from measurements of carbon exchange performed across various spatiotemporal scales compares, and how nonlinear analysis methods can provide important information about causal variables, lagged relationships, and complex interdependencies influencing CH4 exchange. The findings from this collection of research have contributed to our overall body of knowledge on factors related to land-atmosphere carbon exchange within a temperate salt marsh. | |
| Advisor | Vargas, Rodrigo | |
| Degree | Ph.D. | |
| Department | University of Delaware, Department of Plant and Soil Sciences | |
| DOI | https://doi.org/10.58088/eh69-p025 | |
| Unique Identifier | 1378165223 | |
| URL | https://udspace.udel.edu/handle/19716/32707 | |
| Language | en | |
| Publisher | University of Delaware | |
| URI | https://login.udel.idm.oclc.org/login?url=https://www.proquest.com/dissertations-theses/land-atmosphere-carbon-dioxide-methane-exchange/docview/2790110071/se-2?accountid=10457 | |
| Keywords | Carbon dioxide | |
| Keywords | Ecosystem ecology | |
| Keywords | Methane | |
| Keywords | Modeling | |
| Keywords | Remote sensing | |
| Keywords | Salt marsh | |
| Title | Land-atmosphere carbon dioxide and methane exchange in a temperate salt marsh | |
| Type | Thesis |