For the last two years, Jennifer Kist has been working for the Corps’ Charleston District. After graduating from the College of Charleston in South Carolina in 2012, her experiences ranged from working with NOAA to the South Carolina Department of Natural Resources in satellite data collection, hydrographic survey, fisheries biology and land mitigation projects, to name a few. However, her primary duties with USACE are performing hydrographic and topographic surveys in coastal areas. The remotely-sensed data Kist and her team collect is used for multiple purposes, including calculating pre- and post-dredge material quantities, channel condition reports and beach re-nourishment estimates.
With a bachelor’s degree in Marine Biology/Geology and Environmental Geosciences, and her work towards a master’s in Geospatial Information Systems, Kist has the education and work experience to investigate and research new ideas and technologies in the field of data collection, processing and analysis.
Working with her ERDC-U mentor, Thad Pratt, and his team in the Field Data Collection and Analysis Branch of ERDC’s Coastal and Hydraulics Laboratory, Kist worked on completing her research on the use of acoustic backscatter data, as well as the use of multifrequency sonars to collect multifrequency bathymetry and multispectral backscatter data. In simple terms, her in-depth understanding of multibeam physics and image processing techniques allowed her to extract backscatter information from multibeam bathymetry data. Backscatter data, which has mostly been used by organizations such as NOAA to perform habitat characterization studies, is an intensity value produced as a by-product of bathymetric surveying. Bathymetry data shows the surveyor where the seafloor bottom is, but backscatter data has the potential to delineate what it is. Advances in technology over the last year have produced multibeam sonar capable of collecting up to five frequencies at once. This development allows one to collect multifrequency multibeam, as well as multifrequency backscatter data, then process the data to create multispectral backscatter images.
Kist’s research efforts at ERDC involved executing a controlled experiment where multifrequency multibeam data over nine distinct sediment samples was collected. The experiment involved the collection of 11 hydrographic surveys from June to July. Data was collected at 90, 200 and 400 kHz simultaneously. Additionally, several objects such as a turtle shells, dummy unexploded ordnance, wood piles and concrete blocks were placed in the test facility and surveyed in the same manner. A conductivity temperature depth sensor was used each day to adjust multibeam data for any changes in water chemistry. Using well-tested processing techniques and quality control methods, bathymetry and backscatter data were collected, extracted and processed from each multibeam survey. A method for extracting each frequency of backscatter data was developed by an outside source in 2016. However, a method to extract each frequency of bathymetry data, so that each backscatter sample could be corrected for the depth value at each frequency, had to be developed for this study. Backscatter (intensity) mosaics were created for each frequency. These mosaics were then banded as red- 400 kHz, green- 200 kHz and blue- 90 kHz with a composite banding tool traditionally used for satellite image processing. The banded images give specific spectral signatures for each sediment type and allow for much better object detection.
Kist hopes that, in the future, she will be able to further develop image processing and classification techniques to quantitatively delineate sediments using multispectral backscatter. Possible implications for the Corps include the ability to remotely perform sand searches using multibeam surveys already being collected for dredge quantity calculations. This study also showed great promise in the use of these techniques for object detection, specifically unburied or partially buried turtle shells which could be useful for pre-dredge clearance surveys. Single frequency backscatter has already proven to be a valuable tool to other organizations in its ability to distinguish hard bottom from soft and delineate bottom types such coral reefs. Further developments in the use of multifrequency bathymetry to detect and measure fluid muds and other low density materials are being investigated. This study has barely scratched the surface of possible time and money savings implications of collecting and processing single and multifrequency bathymetry and backscatter, as well as processing multispectral backscatter. The possibilities and future uses of this previously unexploited data type are countless.
Kist feels that working with the ERDC-U program and ERDC personnel well-versed in her field has given her the latitude and encouragement to do what she needed for her research. She would definitely recommend the program to others -- “take your pick,” she says.
“I’ve received valuable cross-training, obtained a clearer understanding of ERDC’s capabilities, and every day I’ve been with individuals at CHL where I exchanged tons of information and ideas. The ERDC-U program has been a great experience for both my district and, hopefully, for CHL as well.”