Sapelo Research Application Form
Research Application ID: GCE-152-2026 (submitted: 02/25/2026, status: approved)
Provide a brief title for web display
Influence of mussel mounds on trapped sediment volume on Little Sapelo Island.
Investigator Information
On Island Sponsor: GCE SINERR UGAMI GADNR
| Principal Investigator: | Alberto Canestrelli | ||
| Home Institution: | University of Florida | ||
| Award Information: | NSF CAREER $540k | ||
| Mailing Address: | 1949 Stadium Road | Phone Number: | |
| PO Box 116580 | E-mail Address: | alberto.canestrelli@essie.ufl.edu | |
| Gainesville, FL 32611 | |||
| Co-investigators: | |||
Briefly describe the project goals and methodology
Objective 1 (O1): For each granulometric class, quantify the volume of sediments trapped by a marsh system populated by mussel mounds. My working hypothesis (H1) is that mussel mounds trap a larger volume per unit area than surrounding vegetated areas where mussel mounds are not present.
Objective 2 (O2): Understand if the presence of mussel mounds increases the volume of sediments trapped at the marsh scale. My working hypothesis (H2) is that, although the mounds will increase friction, decrease tidal prism, and decrease the total sediment mass entering the marsh over a tidal cycle, the volume of sediment deposited will still be larger than if mounds were not present. This objective will be carried out with a numerical model (mounds will not be removed in the field, but in the model).
To achieve the scientific objectives, I will use a combination of fieldwork and modeling with Delft3D, which I recently expanded (Crotty et al., 2023) by including a module that computes the volume of sediments filtered by mussel mounds and modifies the morphology accordingly (Delft3D-BIVALVES). See details below on the equipment deployed on the field.
Methods
Ground control point deployment. In the first quarter of Year 1 (1 day in the field), I will position and survey 27 ground control points (GCPs) in the study domain. The survey will be performed using an RTK-GPS. GCPs will be made of a wooden 30 cm×30 cm square target placed on a 2 m tall T-post and painted with red and black paint to make them visible in UAV (Unmanned Aerial Vehicle) images. GCPs will be removed after 1 year.
Vegetation, bathymetry, and topography. They will be estimated from a point cloud produced by a UAV-LiDAR and RGB system flown at a 40 m altitude using the procedure proposed in Pinton et al., 2020b (quarterly in Year 1 to capture seasonal variations, 1 day in the field). Thirty-five ground control points (GCPs) and (optionally) 5 lightweight bottomless pyramids will be positioned across the site to calibrate the lidar data. The GCPs will be made of a wooden 30 cm×30 cm square target placed on a 2 m tall T-post and painted with red and black paint to make them visible in UAV (Unmanned Aerial Vehicle) images. will be deployed by walking, they will be in the marsh for 1 year and then removed. Pyramids Twenty-five GCPs will be placed in the soil on the marsh edge and on the creek bank. Access from the former will be from a boat, access to the latter by walking in the creek to minimize trampling of the marsh. 10 GCPs will be setup by walking on the marsh. Optionally, 5 lightweight bottomless pyramids will be positioned on the edge from the boat on the marsh edge and remain in place for about half a day and will then be removed.
2D Flow field from remote sensing. In the first quarter of Year 1 (three days in the field), I will evaluate the distribution of surface flow velocity in my salt marsh with the method proposed in Pinton et al. (2020a), which uses two UAVs (Parrot Anafi). The first one will fly at a 105 m altitude and release a non-toxic and biodegradable tracing dye (Cole-Parmer, fluorescent yellow/green dye tracer). Dye motion will be tracked by using a series of RGB images collected by a second UAV flying at a 10 m altitude.
Water level, current, and water quality data. To validate the water flow velocity data and to calibrate my numerical model, an upward-looking Nortek Signature 1000 ADCP will be positioned at the bottom of a tidal creek. Water level will be recorded using a HOBO MX2001 pressure sensor, placed at the Sapelo Island ferry dock in a hollow plastic tube (this instrument might not be needed if water level is still being recorded by LTER at the dock for the next couple of years). At the same location, water salinity and temperature data will be collected with a HOBO U24-002-C salinity logger, inserted in a hollow plastic tube. Data will be used to account for the effect of seasonal variations in water salinity and temperature on the filtration rates of mussels. These instruments will be deployed over 1 day in the first quarter of Year 1 and will collect data for two years.
Suspended sediments. To compute particle size distribution, I will use a Sequoia LISST-200X submersible laser diffraction particle size analyzer, placed at the mouth of the creek where the ADCP is located (Figure 1). Data will be continuously collected for 2 years, starting on the first quarter of Year 1. Water samples will also be collected once per quarter at three different phases (flood, ebb, high water slack) of a spring tide large enough to flood the marsh. At each instant and for each of the eight major creeks of the salt marsh, I will collect five water samples: one at the creek mouth, one at the creekhead, one on the adjacent marsh platform, and one above an adjacent mussel mound. One sample will also be collected above the ADCP and LISST. Concentration for each class will be multiplied by the velocities obtained by the drone and the depth computed by the pressure sensor to estimate sediment transport at different instants of a tidal cycle.
Mussel mounds distribution and shape. At the marsh scale, I will apply the classifier proposed in Pinton et al. (2023a) to the LiDAR point cloud. During a field survey in the first quarter of Year 1 (1 day in the field), I will select 24 mounds and count the number of mussels in each mound. This information will be used to determine a relationship that links number of individuals with filtration rates of each mound.
Sedimentation on Marsh and Mounds. I will quantify sediment deposition on 24 selected mussel mounds and the adjacent marsh platform using 24 paired circular plastic sediment traps secured with steel claws (Tognin et al., 2021). Trap material will be collected quarterly for one year starting in the last quarter of Year 1, with up to two additional event based collections after major storms if needed. Laboratory analyses will quantify deposition rates and convert them to millimeters per year using measured bulk density, organic and inorganic fractions, and particle size distributions. To quantify surface accretion and distinguish it from elevation change, I will co locate feldspar marker horizons and five rod based Surface Elevation Tables (rSETs), following Crotty et al. (2023). Marker horizon measurements will be collected quarterly for 2 years. rSET measurements will be collected quarterly for the five-year project to support comparison with sea level rise rates.
Contribution of Vegetation Biomass on Elevation. We will quantify how vegetation biomass contributes to marsh elevation and vegetation change using two cross sections per site, with three sampling locations along each cross section from creek edge to marsh interior. At these locations, we will measure hydrology and biogeochemistry, including quarterly creek water sampling for nutrients, salinity, and temperature, shallow piezometer wells with pressure and temperature sensors to resolve hydroperiod and water table elevation, and redox and dissolved oxygen in wells and core holes. Minirhizotron imaging will be collected only at the center location of each cross section to quantify root production and turnover.
To estimate how much biomass decomposes versus how much is retained in the soil, we will deploy litter bags retrieved quarterly and cotton strips, and collect destructive root cores and sediment cores to measure live and dead belowground biomass, organic content, and bulk density. We will also collect plant material to quantify the fraction of labile versus more refractory biomass. These data will allow us to link biomass production, decomposition, and biomass retention to marsh elevation change and vegetation shifts.
Where will the project be located?
Little Sapelo Island - see attached map for approximate locations.
How will you provide GPS coordinates for study sites?
I will provide a provisional map and arrange with my sponsor to collect and register GPS coordinates
What are the expected start and end dates of the project?
Start Date: 06/01/2026 End Date: 07/01/2028
How many people will access the site and at what frequency?
Planning to have 4-8 people when we deploy instruments in June 2026, for about 12 days. Having 2-4 people for quarterly data collection for about 1-3 days.
Keywords that describe your project
What equipment will be deployed in the field?
• Up to 35 ground control points (wooden 30 cm by 30 cm targets on 2 m T posts) will be installed for UAV surveys and removed after about 1 year; an optional 5 lightweight bottomless pyramids may be placed by boat for lidar calibration for about half a day and then removed.
• One ADCP (creek bottom), one pressure sensor, one salinity and temperature logger, and one LISST 200X at the creek mouth will be deployed in Year 1, serviced quarterly, and removed after about 2 years.
• 24 paired circular sediment traps (on mussel mounds and adjacent marsh) secured with steel claws will be deployed, sampled quarterly for 1 year (plus storm checks if needed), and then removed.
• Five rod based rSET stations and co located feldspar marker horizons will be installed to track accretion and elevation change; marker horizons will be monitored for 2 years, and rSETs will remain for up to 5 years.
• Along two cross sections with three sampling locations each for 2 years we will install these sensors (list is per each of the 6 locations): we will install 1 shallow (<60 cm) piezometer well (5 cm diameter) and redox electrodes (approximately 1 cm diameter) and soil moisture sensors (approximately 2-3 cm diameter), both inserted to a maximum depth of about 40 cm. One minirhizotron tube (clear acrylic, approximately 4 cm diameter and extending to about 40 cm depth) will be installed at the center location of each cross section for continuous root imaging
Will plants or animals be collected as part of this study?
We will collect cores to estimate biomass contribution to elevation, see above for details.
What are the likely impacts of the project on the site?
Small holes for will probably close quickly after retrieval. Minor trampling will be present at the sampling location for probably 1 year.
Will the project design include boardwalks? If not, explain why not.
How long will impacts persist after the research is concluded?
Minor trampling will be present at the sampling location for probably 1-3 years. No long term impact. I will monitor and report toward the end of my 5th year of the award if needed.
Site Photographs:
Files attached to this application
GCE-152-2026_Maps_Canestrelli_proposed_study_location.png (PNG image, 3006.58 kb, submitted 02/25/2026)
GCE-152-2026_Photos_Canestrelli_equipment.png (PNG image, 61.63 kb, submitted 02/25/2026)
GCE-152-2026_Photos_Canestrelli_equipment_GCP.png (PNG image, 28.05 kb, submitted 02/25/2026)
