Images

The 2023 annual report of the U.S. Geological Survey Woods Hole Coastal and Marine Science Center highlights accomplishments of 2023, includes a list of 2023 publications, and summarizes the work of the center, as well as the work of each of its science groups.

A recent study from the U.S. Geological Survey offers a workflow for generating virtual maps of the deep seafloor from archived exploration dives. This work can help advance exploration of the deep-sea frontier and extends into other fields. 

Nautilus dive on the Mona Rift western slope, off the coast of Puerto Rico.

USGS scientists collect core samples from marsh to explore the drives of wetland accretion in the face of rising seas and human impacts impact the evolution on salt marshes.

Nautilus dive on the Mona Rift western slope, off the coast of Puerto Rico.

(Top): Map of dives labeled and shown in red and orange located near the Mona Rift, northwest of Puerto Rico. Contours show depths in meters at 500 m intervals. Square indicates location of dive. Bathymetry from Andrews et al. Yellow curve denotes landslide tsunami source of Lopéz-Venegas et al.

As sea levels rise, salt water extends further up tidal rivers. As sea-level rise accelerates, vegetation transitions are increasingly observed and USGS scientists are busy assessing those changes. USGS scientists take soil cores in the marsh. 

USGS scientists collect core samples from marsh to explore the drives of wetland accretion in the face of rising seas and human impacts impact the evolution on salt marshes.

A collage of images depicts shoreline habitat and wildlife that use the Salton Sea area in southern California.

Video footage of USGS Research Geologist Patrick Barnard in the field, from the short film "Our Beautiful Planet - Saving Our Shores"

A polar bear walking along the Beaufort Sea coast with storm waves breaking behind it, filmed during fieldwork at Barter Island, Alaska.

 A photo collage shows examples of coastal changes at the mouths of small rivers of the world resulting from contributions of new sediment. Imagery from Google Earth.

Research Physical Scientist Donya Frank-Gilchrist visited three classes at Lakeview Fundamental Elementary School for the 2022 Great American Teach-In to discuss wave dynamics, extreme storms, and how these processes alter coastal ecosystems.

Photo of a passive-sampling semipermeable membrane device deployed on a coral reef in West Maui, Hawai'i, to detect organic compounds carried by groundwater discharge.

A map showing study sites at West Maui, Hawai'i, where passive samplers were deployed to detect groundwater contaminants such as pesticides, pharmaceutical compounds, and personal care products that can harm coral reefs.

Geomorphically diverse settings that are experiencing climate-change impacts or are expected to in the coming decades (from upper left to lower right: eroding permafrost coast, river channel, drylands, cryosphere, steep hillslopes, and a recently burned region; photographs by U.S. Geological Survey).

An animation showing tectonic faults across the Eastern Pacific Ocean from space. The San Andreas fault in California is shown in green, and the Cascadia Subduction Zone north of it is shown in yellow. The animation ends at the location of Queen Charlotte-Fairweather fault (in yellow), displaying the latest 30-meter terrain model.

A computed-generated image showing the bathymetry of the Queen Charlotte Fault and the elevation of mountains in the nearshore area, using bathymetry and lidar data.

Our Nation’s coastlines are almost 100,000 miles long. These extensive coastal areas stretch from Maine’s Atlantic shores in the northeast, south along the shores of the Gulf of America and insular areas in the Caribbean.

A diagram that compares hand-digitization versus human-in-the-loop image segmentation workflows. The image (a) is the first in data set F, captured by Landsat 8 on 15 February 2015. The hand-drawn polygons (b) are rasterized to create a label image (c). Subplots (d) and (e) show details from the two regions identified in (c).

Doodler uses a concept called human-in-the-loop machine learning to interactively segment an image. Doodler uses "doodles", or a few marks of examples of each class present in the scene, and machine learning completes the scene.