USGS Science⌕ Search
Home/Archive/USGS

USGS · 70037577

Automated feature extraction and spatial organization of seafloor pockmarks, Belfast Bay, Maine, USA

Brian D. Andrews·Laura L. Brothers·Walter A. Barnhardt
View original source ↗Export metadata ↓

Abstract

Seafloor pockmarks occur worldwide and may represent millions of m 3 of continental shelf erosion, but few numerical analyses of their morphology and spatial distribution of pockmarks exist. We introduce a quantitative definition of pockmark morphology and, based on this definition, propose a three-step geomorphometric method to identify and extract pockmarks from high-resolution swath bathymetry. We apply this GIS-implemented approach to 25 km 2 of bathymetry collected in the Belfast Bay, Maine USA pockmark field. Our model extracted 1767 pockmarks and found a linear pockmark depth-to-diameter ratio for pockmarks field-wide. Mean pockmark depth is 7.6 m and mean diameter is 84.8 m. Pockmark distribution is non-random, and nearly half of the field's pockmarks occur in chains. The most prominent chains are oriented semi-normal to the steepest gradient in Holocene sediment thickness. A descriptive model yields field-wide spatial statistics indicating that pockmarks are distributed in non-random clusters. Results enable quantitative comparison of pockmarks in fields worldwide as well as similar concave features, such as impact craters, dolines, or salt pools.

Explore related subjects

GeomorphologyArticle

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Brian D. Andrews, Laura L. Brothers, Walter A. Barnhardt. 2010. Automated feature extraction and spatial organization of seafloor pockmarks, Belfast Bay, Maine, USA. https://doi.org/10.1016/j.geomorph.2010.08.009

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related USGS reports

USGS2026

The influence of riparian zone grazing on channel stability in an arroyo

The introduction of large herds of domesticated grazing animals by European immigrants in the late 19th century is one primary hypothesis used to explain the initiation of arroyo cutting in the American West. But studies designed to quantify the effects of grazing on watershed hydrology and sedimentation have had conflicting results that are often scale dependent. We used a before-and-after study design to examine effects of changes in cattle grazing practices on channel stability in Muddy Creek, an arroyo in the Colorado River headwaters. We digitized 620 meander loop cutoff geometries in the alluvial bottom of Muddy Creek and used meander loop cutoff rate as a broad measure of channel stability. Poisson regression modeling indicated that the change in grazing practices caused an order-of-magnitude decrease in meander loop cutoff rate that was independent of other forcings. Repeat measurements of channel geometry showed a decline in bankfull channel width of about 40% after the changes were implemented. Increases in riparian vegetation density are deduced to be the primary mechanism for the increase in channel stability. Prior to the changes in grazing practices, at least one meander loop cutoff occurred each year, some of which were large enough to create headcuts that scale with the depth of downcutting in this arroyo system. Our analysis demonstrates the profound effects that different grazing practices have on channel stability in headwater arroyo systems, and the potential for grazing animals to initiate a cascade of sediment yields to receiving stream networks.

Colorado, Utah, Wyoming↗
USGS2026

Revisiting the geochronology of late Quaternary marine terraces and uplift rates in coastal Santa Barbara County, California, USA

In several early studies, central California marine terraces between Santa Barbara and Point Conception were interpreted to record sea-level high stands of the last interglacial complex, ∼80 ka to ∼120 ka (marine isotope stage [MIS] 5). These ages and their elevations (∼20 m to ∼45 m) indicate modest rates of tectonic uplift, similar to those from other localities in southern and central California. A recent study, using a combination of luminescence and radiocarbon dating, has challenged the older age interpretations, implying much younger terrace ages, between ∼40 ka and ∼55 ka (MIS 3). From these new ages and a considerably lower sea level during MIS 3, much higher rates of tectonic uplift are inferred. In the present study, new uranium-series ages of terrace corals and amino acid age estimates of terrace mollusks were determined to test these competing interpretations. With the exception of a low-elevation terrace in Isla Vista (near Santa Barbara) that dates to MIS 3, terraces farther west are interpreted to date to MIS 5 and imply tectonic uplift rates of 0.20–0.34 m/kyr. A compilation of data for the region yields a decreasing rate of late Quaternary uplift from east, near Ventura, to west, near Point Conception. This trend is interpreted to reflect a decreasing influence of the processes of compression and crustal shortening south of the Big Bend in the San Andreas fault.

California↗
USGS2025

Toward an efficient framework for remote sensing of river bathymetry: Comparing sensors and algorithms on an inaccessible proglacial river in Alaska

Remote sensing can provide reliable information on river depths and this approach might be particularly valuable in areas that are difficult to survey via conventional field methods. In this study, we assessed the potential to map the bathymetry of an inaccessible proglacial river in Alaska from both aerial orthophotos and a multispectral satellite image. In addition, we evaluated a variety of depth retrieval algorithms with different input data requirements, including some methods that require field measurements of water depth for calibration and other techniques that can be applied even when such field data are not available. These approaches might enable more efficient use of remote sensing methods by resource management agencies. Our results suggest that bathymetric mapping along the turquoise-colored river we examined was not only feasible but highly accurate ( R 2 up to 0.94) for both types of image data. Algorithms that use paired observations of depth and reflectance to train depth retrieval models were the most accurate, with errors on the order of 15%–20% and little or no bias. Alternative techniques based on hydraulic and statistical concepts also led to strong agreement between predicted and observed depths but were more susceptible to systematic biases toward under- or over-estimation of depth. In contrast to clear-flowing streams, bathymetric mapping in this environment was enabled by a direct relationship between the depth and brightness of the water due to scattering by suspended sediment. In selecting an appropriate depth retrieval method, a compromise might need to be reached between the level of field effort invested and the accuracy of the resulting image-derived bathymetry. Standalone software for implementing these techniques is freely available.

Alaska↗