Evaluating the stability of deep-water sands that provide habitat for Pacific sand lance, a critical forage fish in Puget Sound

Release Date:

Sand waves and ripples in a deep-water channel in Puget Sound are on the move, but they are migrating so slowly that they will continue to provide stable habitat for Pacific sand lance—a forage fish important to young salmon, lingcod, and other fish, marine mammals, and seabirds.

A skinny fish pokes its head out of the coarse bottom sediment in murky water.

Pacific sand lance sticking its head out of coarse-grained sediment.

(Credit: P. Naylor. Public domain.)

This article is part of the June-July 2020 issue of the Sound Waves newsletter.

The article is dedicated to our close friend and co-investigator, Dr. Monty Hampton, who passed away in December 2019. Monty participated in all aspects of this research and was a major contributor to an article we published together in 2017 in the journal Geosciences. Monty and the first author worked on this research as scientists emeriti at the USGS Pacific Coastal and Marine Science Center in Santa Cruz, California, with support from the USGS Bradley Scholar Program.

Sand lance (genus Ammodytes), also known as “candlefish” or “sand eel,” is an ecologically important forage fish along the U.S. east and west coasts. Sand lance are named for their habit of burrowing into the sand, which they do to avoid predation and to overwinter. The slender fish approach the sandy seabed vertically and then dive and wiggle into the sediment head first, shifting to tail down while in the sand. They are commonly found in coastal waters of the northern Pacific Ocean from Japan to northern California, and in the North Atlantic. Pacific sand lance (Ammodytes personatus) are an important food source for young salmon and lingcod. Marine mammals, other fish, and many species of seabirds also forage on them.

Hundreds of long, skinny, small fish swim or float upright just above the sandy bottom illuminated by a light source.

Pacific sand lance rising above a sand wave, observed along a submersible transect of the San Juan Channel sand wave field. Note sharp form of crest in the background of the photo, and the well-sorted medium-grained sand in the foreground. Photo taken from inside the submersible Cyclops 1 by Matthew Baker.

Credit: Matthew Baker, Friday Harbor Marine Laboratory, University of Washington, public domain.

Scientists have conducted numerous biological and ecological studies of Pacific sand lance in Puget Sound, Washington, but most of these have focused on shallow intertidal areas, where the sand is typically mobile and often ephemeral. We investigated a deep-water sand lance habitat—at water depths of 20 to 80 meters (60 to 260 feet)—in the San Juan Channel east of San Juan Island, the largest island in the U.S. San Juan Island group. Strong bottom currents have shaped the sand there into waves with heights of up to 4 meters (13 feet) and crests as much as 100 meters (330 feet) apart. Previous studies have shown Pacific sand lance to be common and abundant in the sand wave field. Our goal was to assess the stability of this important habitat.

Underwater television cameras have documented Pacific sand lance burrowing into the San Juan Channel sand wave field, and researchers have used bottom-grab samplers to collect and count the fish in the sediment. One study by students from the University of Washington’s Friday Harbor Marine Laboratory on San Juan Island estimated the Pacific sand lance population in the sand wave field at more than 60 million fish. More recent studies indicate a much larger population. There is growing concern that rising sea level is reducing the intertidal sandy areas in Puget Sound that sand lance use as spawning grounds, potentially making their deep-water habitats even more important.

From 1994 to 2007, Gary Greene of the Center for Habitat Studies, Moss Landing Marine Laboratories, and Vaughn Barrie of the Geological Survey of Canada, Pacific Division, conducted detailed seafloor-mapping surveys throughout the San Juan Islands region. These surveys revealed the shape of the seafloor and showed how it changed over time. Analysis of the data suggested that bottom currents might be causing some movement of the sand waves. Such movement could disrupt the sand lance habitat, particularly if it became more pronounced or, in an extreme case, dispersed the sand waves altogether. We decided to collect various types of data to determine how stable the sand waves are.

We obtained multibeam echo-sounding sonar (MBES) data through a cooperative arrangement with the Geological Survey of Canada, Canadian Hydrographic Service, and Moss Landing Marine Laboratories, and private funding from the Dickinson Foundation. We and our partners conducted multiple surveys from 1994 through 2010 to study the dynamics of the sand wave field. These surveys provided detailed images of the sediment waves and ripples through a combination of bathymetric data (seafloor depths) and acoustic backscatter (amount of sound reflected from the seafloor, which provides information about seafloor texture and materials). The map, based on the bathymetric data, shows the elongate sand body in San Juan Channel and the striking sand wave field. The largest sand waves are slightly asymmetrical, with steeper slopes toward the north, about 3 to 4 meters (10 to 13 feet) high, and about 50 to 100 meters (160 to 330 feet) long.

Illustrations show sand waves underwater.

Sand waves in the San Juan Channel, with location map.

(Public domain.)

We carried out four one-day research cruises from July 2010 to July 2013 in this region to estimate sediment variability and migration rates of the large sand waves for comparison with the multibeam-mapping data collected earlier. The one-day cruises were undertaken aboard two small vessels operating in the San Juan Islands: the research vessel (R/V) Tombolo and the R/V Centennial. We collected data on bottom sediment textures and bed shape by using grab samplers, high-resolution seafloor photography, and bottom video imagery. To measure the speed and direction of currents near and above the seabed, we used a shipboard-mounted acoustic Doppler current profiler (ADCP). We analyzed these data along with similar data from studies by students at the Friday Harbor Marine Laboratory.

More than 100 sediment samples in the combined data sets show three major grain-size classes: fine to medium sand (mean diameter approx. 0.25 mm) within the outer sections of the sand body where ripples and small sand waves occur; medium sand (mean diameter approx. 0.5 mm) on the crests and upper flanks of the larger sand waves; and coarse sand and gravel (mean diameter greater than 2.0 mm) in the troughs. The coarse material in the troughs consists dominantly of shell hash.

Two photographs show men standing on a boat and a pier.

Above, left:
Dave Cacchione (left) and Monty Hampton on research vessel Tombolo docked at the University of Washington’s Friday Harbor Marine Laboratory pier, San Juan Island.

Above, right:
USGS scientist Hank Chezar on pier next to Tombolo.

(Public domain.)

Three men dig into a sediment sample in a trough on the deck of a boat.

(Left to right) Hank Chezar, Monty Hampton, and Gary Greene dig into recovered sediment sample onboard the research vessel R/V Centennial, with David Dugins, Friday Harbor Marine Laboratory, in the background.

(Public domain.)

Currents measured with the ADCP along seven transects during flood tide were quite strong at 5 meters (16 feet) above the seabed; the highest flood currents, toward the north, were 87 centimeters (34 inches) per second. Near-bottom ebb flows obtained earlier by Friday Harbor Marine Laboratory students had similar maxima. These strong near-bottom tidal flows suggest active sediment transport, especially for the fine to medium sands. We used calculations from analytical models to estimate bedload transport rates. Then we applied these rates to determine migration rates of the large sand waves. The results indicate that the sand waves migrate much less than 1 meter (3 feet) per year. We also used a recently developed technique to calculate sand wave migration on the basis of bedform asymmetry. This method also showed low migration rates—less than 20 centimeters (8 inches) per year. The low migration estimates agree with the results from the repeated bathymetric surveys.

Photographs of sediment from underwater, taken up close to show the various, individual grains of sand.

Bottom photographs of surface sediment obtained with a specialized, ship-lowered underwater camera developed by USGS scientists and nicknamed the Flying Eyeball. Scale shown on photographs. Image at left taken near the crest of a large sand wave; image at right in a trough.

(Public domain.)

The results from our work suggest that sand waves in San Juan Channel are only slowly undergoing net migration. Longer term data on bottom currents and repeated sediment sampling would be useful to improve the accuracy of these estimates. Effects of sea-level rise on tidal flows in the region might also be examined by further studies.

Our findings led to submersible dives in September 2018—funded by the SeaDoc Society and Ocean Gate, Inc.—to observe directly the seafloor processes and Pacific sand lance activity hypothesized in our study. Observations made during the dives confirmed that sediment moves within the San Juan Channel sand wave field. Although unable to maintain a stable position throughout a complete tidal cycle, scientists in the submersible Cyclops 1 observed shell hash movement at the initiation of tidal flow, transitioning to fine sand movement as tidal energy increased, forming sand ripples on the larger sand waves. Additionally, in a video made while the submersible moved slowly up-current over well-developed sand waves, it appears that the largest numbers of Pacific sand lance were exiting from the crests and upper flanks of the waves, consistent with hypotheses that Pacific sand lance prefer to burrow in medium grain-size sediment that is mobile and well-aerated.

Related Content