Area 4: Integration and Forecasting

Objectives Progress Report Publications Show All  

Integration and Forecasting

We use the information collected in Areas 1-3, along with modeling and remote sensing, to A) produce synoptic descriptions of ecosystem properties, B) create a scaled-up disturbance-scape that tracks the temporal and spatial patterns of perturbations and their cumulative effects, and C) investigate relationships between drivers and ecosystem responses.

Research Objectives

A) Produce synoptic descriptions of ecosystem properties

  • 4A.1 - Produce synoptic habitat maps of the GCE domain
  • 4A.2 - Assess scaled-up biomass patterns
  • 4A.3 - Evaluate C stocks and transport from tidal wetlands to the coastal ocean

B) Create a scaled-up disturbance-scape that tracks the temporal and spatial patterns of perturbations and their cumulative effects

  • 4B.1 - Use drones to track disturbances over time
  • 4B.2 - Construct a scaled-up disturbance-scape

C) Investigate relationships between drivers and ecosystem responses

  • 4C.1 - Use drones to track disturbances over time
  • 4C.2 - Construct a scaled-up disturbance-scape

Current Progress Report

Below is an update for each of the Area 4 objectives as reported in the most recent annual report. For a list of all reports click here (Annual Reports).

A) Produce synoptic descriptions of ecosystem properties

  • 4A.1 - Produce synoptic habitat maps of the GCE domain

      Activities:  Aerial photographs (Obj 2B.2) were analyzed using random forest to generate classified maps of habitat distributions for 2017 and 2018.

      Results:  Habitat maps delineating 11 tidal habitats were generated with overall classification accuracies ranging from 0.83 (2017) to 0.88 (2018). These are being used for change detection analyses.

  • 4A.2 - Assess scaled-up biomass patterns

      Activities:  We extended our studies of Spartina biomass (O'Donnell & Schalles 2016) by applying our algorithm to Landsat 8 data and expanding our analysis to the Georgia coast. We are also optimizing our algorithm to generate estimates of below-ground biomass, and continue to collect monthly samples of Spartina above- and below-ground biomass for groundtruthing.

      Results:  Our expanded analysis of long-term trends in Spartina biomass was applied to 7 USGS HUCs along the Georgia coast and showed declines in all but the Altamaha HUC (which has the most freshwater input). These data were presented at several meetings and are being included in a cross-coastal LTER climate synthesis manuscript. A manuscript describing below-ground biomass trends is also in prep.

  • 4A.3 - Evaluate C stocks and transport from tidal wetlands to the coastal ocean

      Activities:  As part of an ROA supplement awarded in 2019, we sampled 33 stations during successive neap and spring tides to obtain high-temporal-resolution analysis of carbonate chemistry. We also continued processing cores from all 11 GCE sites for C content and radioisotope dates.

      Results:  High-frequency monitoring conducted with an ROA supplement showed evidence for DIC export during spring tides due to flushing of tidal nodes, which may explain the net export observed from the Duplin River (Obj.1A4). Spivak et al. (2019) wrote a synthesis paper highlighting the importance of understanding the key biogeochemical mechanisms within the marsh that control decomposition of soil organic matter when evaluating the effects of climate change on coastal wetland C storage.

B) Create a scaled-up disturbance-scape that tracks the temporal and spatial patterns of perturbations and their cumulative effects

  • 4B.1 - Use drones to track disturbances over time

      Activities:  We launched a large-scale effort to track disturbances via regular drone flights. We acquired a Matrice 200 drone with a Micasense RedEdge Altum camera, obtained appropriate permits and FAA licenses, performed tests to optimize flight conditions, began monthly flights over an initial test site, and obtained ground-truth data for field validation of disturbances. We have developed a work flow for processing imagery and are optimizing our algorithm to detect disturbances.

      Results:  Our initial drone imagery is already yielding important insights into patch dynamics in the salt marsh, with evidence of shifting areas of wrack from month to month that leave residual signals on the landscape (Fig. 1).

2019 Area 4 Figure 1

Fig. 1. Comparison of wrack disturbance patches (outlined in purple) identified from monthly drone flights of the airport marsh. Yellow arrows in the September image indicate areas where wrack packets were no longer present. Source: Tyler Lynn.

  • 4B.2 - Construct a scaled-up disturbance-scape

      Activities:  We used Spartina biomass from clip plots to calibrate a biomass algorithm based on drone reflectance, and we are now working to pair these high-resolution observations with satellite imagery (WorldView, Sentinel, Landsat, MODIS).

      Results:  Spartina biomass maps were successfully produced from the drone imagery flown at the Airport Marsh test site (Fig. 2). Spartina biomass estimates were barely affected (<1.3%) when scaled up using MODIS. This will be presented at the first NOAA-SECOORA Drones in the Coastal Zone Workshop in March, 2020.

2019 Area 4 Figure 2

Fig. 2. Patterns of above-ground Spartina alterniflora biomass at the Airport Marsh site on Sapelo Island (left) based on relationship between NDVI from the GCE drone camera and ground-truthed clip plot data (above). Source: John Schalles.

C) Investigate relationships between drivers and ecosystem responses

  • 4C.1 - Characterize driver-response relationships in GCE data

      Activities:  We have analyzed our long-term salinity data from the Altamaha River estuary to evaluate the conditions under which high salinity events occur (Fig. 3).

      Results:  We identified 79 high salinity events, ranging from 1 to 51 days, in the Altamaha River estuary over the last 16 years. These events could be explained by river flow that was objectively low based on the historical record, strong up-estuary winds, or (to a lesser extent) unusually high tides (Sheldon & Alber 2019).

2019 Area 4 Figure 3

Fig. 3. Estimated probability of a high salinity event at GCE 7 as a function of the combined effects of river flow, along-estuary wind, and tidal amplitude (black dots) and as a function of each driver alone (green lines). Combined-effects probabilities for the conditions of each day in the study period are plotted against individual driver variables, so the vertical scatter at any given driver value is due to the effects of the other two drivers. River flows >500 m3 s-1 not shown because event probabilities are zero. Source: Sheldon and Alber 2019. Corresponds to Objective 4C.1: Characterize driver-response relationships in GCE data.

  • 4C.2 - Develop simple mechanistic models to explore disturbance

      Activities:  We plan to use the results from the SALTEx experiments (Obj. 3B.2) as a test case for characterizing ecological responses from field data with a known perturbation (i.e. saltwater intrusion).

  • 4C.3 - Investigate disturbance patterns through model simulations

      Activities:  The hydrologic monitoring data from groundwater wells will be used to investigate links between major disturbances (high rainfall events, very high tides), and vegetation changes.

Area 4 Publications from GCE-IV

Spivak, A.C., Sanderman, J., Bowen, J.L., Canuel, E.A. and Hopkinson, C.S. 2019. Global-change controls on soil-carbon accumulation and loss in coastal vegetated ecosystems. Nature Geoscience. 12:685692. (DOI: https://doi.org/10.1038/s41561-019-0435-2)

O'Donnell, J. and Schalles, J.F. 2016. Examination of Abiotic Drivers and Their Influence on Spartina alterniflora Biomass over a Twenty-Eight Year Period Using Landsat 5 TM Satellite Imagery of the Central Georgia Coast. Special Issue: Remote Sensing in Coastal Environments. Remote Sensing. 8(6):22. (DOI: 10.3390/rs8060477)

 
LTER
NSF

This material is based upon work supported by the National Science Foundation under grants OCE-9982133, OCE-0620959 and OCE-1237140. Any opinions, findings, conclusions, or recommendations expressed in the material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.