Deep learning methods paired with sidescan sonar (SSS) are commonly used in underwater search-and-rescue operations for drowning victims, wrecks, and airplanes. However, these techniques are primarily used to detect mine-like objects and are rarely applied to identifying features in dynamic dredge pit environments. In this study, we present a Sandy Point dredge pit (SPDP) dataset, in which high-resolution SSS data were collected from the west flank of the Mississippi bird-foot delta on the Louisiana inner shelf. This dataset contains a total of 385 SSS images. We then introduce a new Effective Geomorphology Classification model (EGC). Through ablation studies, we analyze the utility of transfer learning on different model architectures and the impact of data augmentations on model performance. This EGC model makes geomorphic feature identification in dredge pit environments, which requires extensive experience and professional knowledge, a quick and efficient task. The combination of SSS images and the EGC model is a cost-effective and valuable toolkit for hazard monitoring in marine dredge pit environments. The SPDP SSS image dataset, especially the feature of pit walls without a rotational slump, is also valuable for other machine learning models.

Due to high sedimentation rate up to ~1 m/year, mud-capped dredge pits (MCDP) are often considered natural laboratories for studying sedimentary processes, slope stability and the impacts of dredging activities on marine environments. Although many studies have been performed on the Louisiana shelf, there is a lack of high spatial resolution research covering the eastern, central and western Louisiana shelf to comprehensively investigate sediment infilling. Eighteen vibracores were collected from the Peveto Channel dredge pit (PC), Raccoon Island dredge pit (RI) and Sandy Point dredge pit (SP), and more than 1300 samples were analyzed to study the spatial variation in surficial sediment using statistical analyses. Our results indicate that the inner Louisiana continental shelf is silt-dominated, and there was no consistent grain size variation when comparing the sediment within the pits with that outside the pits. Skewness emerged as a prominent factor in the RI and SP, while standard deviation was the most influential in the PC. Our analysis shows also that two principal components are confirmed and account for more than 95% of the total grain size variance.

Deep learning methods paired with sidescan sonar (SSS) are commonly used in underwater search and rescue operations for drowning victims, wrecks, and airplanes. However, these techniques are primarily designed to detect mine-like objects and are rarely applied to identifying features in dynamic dredge pit environments, due to a lack of datasets. In this study, we present a Sandy Point Dredge Pit (SPDP) dataset, in which high-resolution SSS data were collected from the west flank of Mississippi bird foot delta of the Louisiana inner shelf. This dataset contains a total of 385 SSS images. We introduce an Effective Geomorphology Classification model (EGC). By ablation studies, we analyze the utility of transfer learning on different model architectures and the impact of data augmentations on model performance. This EGC model will make geomorphic features identification in dredge pit environments quick and efficient which requires extensive experience and professional knowledge. The combination of SSS images and EGC model is a cost-effective and valuable toolkit for hazards monitoring in marine dredge pit environment. The SPDP SSS images dataset, especially the feature of pit wall without rotational slump, is also valuable for other machine learning modelers.