Search

Search Results

• 1. [Article] Improving Projections of Tidal Marsh Persistence under Climate Change with Remote Sensing and Site-Specific Data

  Tidal marshes are dynamic ecosystems that are threatened by climate change and sea-level rise. To characterize baseline condition and historic climate sensitivities, and improve projections into the future, ...

  Citation × Citation Citation Abstract Copy Title: Improving Projections of Tidal Marsh Persistence under Climate Change with Remote Sensing and Site-Specific Data Author: Buffington, Kevin John Tidal marshes are dynamic ecosystems that are threatened by climate change and sea-level rise. To characterize baseline condition and historic climate sensitivities, and improve projections into the future, new methods are required that integrate data from the field and remote sensing platforms. Marsh elevation response models can be calibrated with site-specific data to determine the vulnerability of a marsh to sea-level rise and help guide management decisions. Elevation models are sensitive to initial elevation, the rate of accretion, and aboveground biomass. The overarching goal of this dissertation was to develop techniques to improve these important model inputs and evaluate the range of spatial and temporal variation. Light detection and ranging (lidar) is an invaluable tool for collecting elevation data, however dense vegetation prevents the accurate measurement of the tidal marsh surface. In Chapter 2, I describe the development of a technique to calibrate lidar digital elevation models with survey elevation data using readily available multispectral aerial imagery from the National Agricultural Inventory Program (NAIP). Using survey elevation data across 17 Pacific Coast tidal marshes, I demonstrated the utility of the Lidar Elevation Adjustment with NDVI (LEAN) technique to account for the positive bias in lidar due to vegetation. LEAN reduced error from an average of 23.1 cm to 7.2 cm root mean squared error and removed the positive bias caused by vegetation. This increase in accuracy will facilitate more accurate assessments of current and future vulnerability to sea-level rise. The phenology of aboveground biomass in tidal marsh plants in relation to climate variation has not been explored in the Pacific Northwest (PNW). In Chapter 3 I explain how I leveraged the Landsat archive and cloud computing capabilities to assess how Tasseled Cap Greenness (TCG, a proxy for aboveground biomass) in three PNW tidal marshes has responded to recent variation in climate to characterize sensitivity to climate change. Through analysis of over 3700 Landsat images obtained from 1984-2015, I found increased annual precipitation resulted in a higher peak TCG, while warmer May temperatures resulted in an earlier day of peak TCG. These results also demonstrate how time-series analysis of remote sensing data can be used to examine the sensitivity of tidal marsh plants to climate variability and directional change. The range of variation in tidal marsh accretion rates has not been characterized across the PNW. For Chapter 4, I collected and analyzed twenty-two soil cores from eight estuaries to estimate historic accretion rates with radioisotope dating techniques and evaluated the amount and source of variation across estuaries. I found that tidal marshes across the PNW had accretion rates greater than the current rate of sea-level rise, ranging from 2.3 – 7.3 mm yr⁻¹. Using a watershed-scale analysis, I found that long-term average annual fluvial discharge was the top predictor of tidal marsh accretion rates. Additionally, I found that calibrating the Wetland Accretion Rate Model for Ecosystem Resilience (WARMER) with accretion rate data from nearby estuaries can result in uncertainties of up to 41% (50 cm) after 100 years. Finally, in Chapter 5, I demonstrate that a range of 62 cm of error is possible in WARMER models after a 100 year simulation when both uncorrected lidar and non-local accretion rates are used, fundamentally changing the interpretation of the results. Altogether, this dissertation illustrates the importance of collecting site-specific wetland vegetation and elevation data and demonstrates how lidar and multispectral remote sensing data can be leveraged to improve our understanding of how climate variability and change impacts coastal ecosystems. Title: Improving Projections of Tidal Marsh Persistence under Climate Change with Remote Sensing and Site-Specific Data Author: Buffington, Kevin John Tidal marshes are dynamic ecosystems that are threatened by climate change and sea-level rise. To characterize baseline condition and historic climate sensitivities, and improve projections into the future, new methods are required that integrate data from the field and remote sensing platforms. Marsh elevation response models can be calibrated with site-specific data to determine the vulnerability of a marsh to sea-level rise and help guide management decisions. Elevation models are sensitive to initial elevation, the rate of accretion, and aboveground biomass. The overarching goal of this dissertation was to develop techniques to improve these important model inputs and evaluate the range of spatial and temporal variation. Light detection and ranging (lidar) is an invaluable tool for collecting elevation data, however dense vegetation prevents the accurate measurement of the tidal marsh surface. In Chapter 2, I describe the development of a technique to calibrate lidar digital elevation models with survey elevation data using readily available multispectral aerial imagery from the National Agricultural Inventory Program (NAIP). Using survey elevation data across 17 Pacific Coast tidal marshes, I demonstrated the utility of the Lidar Elevation Adjustment with NDVI (LEAN) technique to account for the positive bias in lidar due to vegetation. LEAN reduced error from an average of 23.1 cm to 7.2 cm root mean squared error and removed the positive bias caused by vegetation. This increase in accuracy will facilitate more accurate assessments of current and future vulnerability to sea-level rise. The phenology of aboveground biomass in tidal marsh plants in relation to climate variation has not been explored in the Pacific Northwest (PNW). In Chapter 3 I explain how I leveraged the Landsat archive and cloud computing capabilities to assess how Tasseled Cap Greenness (TCG, a proxy for aboveground biomass) in three PNW tidal marshes has responded to recent variation in climate to characterize sensitivity to climate change. Through analysis of over 3700 Landsat images obtained from 1984-2015, I found increased annual precipitation resulted in a higher peak TCG, while warmer May temperatures resulted in an earlier day of peak TCG. These results also demonstrate how time-series analysis of remote sensing data can be used to examine the sensitivity of tidal marsh plants to climate variability and directional change. The range of variation in tidal marsh accretion rates has not been characterized across the PNW. For Chapter 4, I collected and analyzed twenty-two soil cores from eight estuaries to estimate historic accretion rates with radioisotope dating techniques and evaluated the amount and source of variation across estuaries. I found that tidal marshes across the PNW had accretion rates greater than the current rate of sea-level rise, ranging from 2.3 – 7.3 mm yr⁻¹. Using a watershed-scale analysis, I found that long-term average annual fluvial discharge was the top predictor of tidal marsh accretion rates. Additionally, I found that calibrating the Wetland Accretion Rate Model for Ecosystem Resilience (WARMER) with accretion rate data from nearby estuaries can result in uncertainties of up to 41% (50 cm) after 100 years. Finally, in Chapter 5, I demonstrate that a range of 62 cm of error is possible in WARMER models after a 100 year simulation when both uncorrected lidar and non-local accretion rates are used, fundamentally changing the interpretation of the results. Altogether, this dissertation illustrates the importance of collecting site-specific wetland vegetation and elevation data and demonstrates how lidar and multispectral remote sensing data can be leveraged to improve our understanding of how climate variability and change impacts coastal ecosystems. Close