Tracking Eddies that Feed the Sea   Page 2
  by William Crawford and Frank Whitney

Vortices of water, called "eddies," form off the northwestern coast of North America in the winter, and are particularly large during El Niño winters when warm waters along the coast flow northward at greater speed than normal. These eddies carry local nutrient-rich waters far offshore into regions with low ambient nutrient levels. As the eddies transport fresh water and nutrients out into the middle of the Gulf of Alaska, they provide nourishment for phytoplankton—microscopic plants that form the foundation of the marine food chain. (To learn more about these tiny plants, read: What are Phytoplankton?) How important are these eddies for the Gulf’s ecosystem? It is hard to answer that question now but, in one example, scientists say it is possible that variations in the size and frequency of the eddies are one of the factors governing the success of salmon in the open ocean.

  Eddies in the Gulf of Alaska

Recently, Canadian and American scientists teamed up to collect and analyze data from satellite and ship-borne sensors taken over the region. With these data, they set out to determine the properties and behavior of the eddies and measure their impact on the Gulf of Alaska’s ecosystem. The researchers found that the eddies, particularly those created during El Niño years, can last several years. They found that the eddies migrate slowly through the Gulf, moved about by shifting currents, and replenish nutrient-starved regions with iron and nitrate.

"Our concern over the depletion of fish in this region makes satellite altimeter measurements such as TOPEX/Poseidon data particularly important in understanding the formation and movement of these nutrient-rich eddies, and how they influence salmon growth and other fisheries," says William Crawford of Fisheries and Oceans Canada at the Institute of Ocean Sciences.

  Eddies are rotating masses of water in the ocean that typically form along the boundaries of ocean currents. In the Gulf of Alaska, eddies of warm water, filled with nutrients from shallow coastal water, mix with the cold water off the continental shelf. The mixing fertilizes the nutrient-poor water of the gulf, resulting in blooms of phytoplankton (microscopic ocean plants.) This true color image from the Sea-viewing Wide Field-of-view Sensor shows the green spiral of an eddy in bright blue water. Also notice the sediments suspended in the water along the south coast of Alaska. (Image provided by the SeaWiFS Project, NASA/Goddard Space Flight Center, and ORBIMAGE)
Sea Surface Height: July 10, 2000

In 1998, he and colleague Frank Whitney began using TOPEX/Poseidon images produced by the University of Colorado to track the large-scale eddies. The satellite data, along with in situ data collected aboard a Canadian Coast Guard Ship gave Crawford and Whitney unique insight into these eddies as a natural mechanism for nourishing the sea.

next Eddies in the Gulf of Alaska

The data used in this study are available in one or more of NASA's Earth Science Data Centers.

Satellites monitor movement and evolution of eddies continuously. Using radar that sees through clouds, the TOPEX/Poseidon mission and the European Remote-Sensing Satellite-2 (ERS-2) produce maps of sea surface height. Since eddies that are warmer than the surrounding water are higher than the usual sea surface height, they appear on these maps. This image shows the difference from normal sea surface height for the northeastern Pacific. Warm core eddies appear as red circles. (Image courtesy Colorado Center for Astrodynamics Research)

  Eddies in the Gulf of Alaska   Page 1Page 3
  It was in mid-September 1998 that Bill Crawford and Frank Whitney met over coffee to discuss Whitney’s cruise on the Coast Guard Vessel John P. Tully into the Gulf of Alaska. While senior scientist on board ship the previous three weeks, Whitney had found a huge, warm, relatively fresh water mass 200 km wide and more than 1000 m deep, about 600 km west of Vancouver Island.
Eddy Temperature Profile

"Do you see this feature in the TOPEX data?" Whitney asked. TOPEX/Poseidon can measure sea surface height accurate to within 2 centimeters. Crawford and colleague Josef Cherniawsky of the Institute of Ocean Sciences had been processing TOPEX/Poseidon data for several months to look for sea level rise in coastal waters, as part of the El Niño event the previous winter. The idea is that as its temperature increases, sea water expands, and TOPEX/Poseidon can measure the corresponding change in sea surface elevation. Although most water in the eddy is warmer than the surrounding ocean, the waters near the surface are either similar or even slightly cooler than surrounding seas. For this reason, satellites that sense ocean surface temperature seldom find these eddies.

An American-French program launched TOPEX/Poseidon in 1992, and released the first data from it in October that year. The satellite senses sea surface height along a 20-km-wide swath, on an orbital track that repeats about every ten days. Each ten-day sample is denoted a "cycle." "TOPEX" refers to the American dual-frequency radar sensor that is turned on for nine of the ten cycles. "Poseidon" is the French radar unit that samples on every tenth cycle. The satellite can measure sea surface topography accurately to within several centimeters.

Cherniawsky told Crawford that he found a new Web site that posts near-real-time TOPEX images. Crawford signed on and entered the latitude and longitude range of the Gulf of Alaska, and the cruise date, and there it was: a red bulls-eye of water whose core rose 30 cm above the surrounding ocean, at the same place and diameter as Frank’s warm, relatively fresh water mass (see letter "A" in sea surface height image below). He had just found a Web site that posts, for free, the most up-to-date, accurate information on these eddies!

  Temperature (°C) in the Gulf of Alaska as measured in Aug-Sept 1998 from the Canadian Coast Guard Ship John P. Tully. The depressed isotherms near 600 km from the coast are at point A in the sea surface height image below. Canadian scientists have sampled ocean waters from Vancouver Island to Station P at 50°N, 145°W for more than 40 years. (Image courtesy Fisheries and Oceans Canada, Institute of Ocean Sciences)
Eddy SSH 1998 vs. 2000

Robert Leben of the University of Colorado had posted the web site only three weeks before Cherniawsky looked. Leben wanted to enable the public to find their own eddies. He combined TOPEX/Poseidon altimetry data with similar observations by the ERS-2 satellite, launched by the European Space Agency. Leben then applied spatial filters to enhance the display of ocean eddies and suppress large-scale seasonal signals. He developed this tool for his own studies in the Gulf of Mexico, but by putting all the data on his web site, he provided a new "digital eye on the world." To see for yourself, visit his altimeter data viewer site.

Crawford used this web site to plot images of the eddy over the previous seven months, and continues to track the eddy. Its status in June 2000 was ambiguous, but a trace of it might be found at 45.5°N, 142°W. The satellite images revealed that the eddy formed in winter 1997/1998 along the West Coast of the Queen Charlotte Islands. He labeled it Haida-1998, after the First Nations of the region and its year of formation. Crawford and Cherniawsky have completed their own analysis of TOPEX/POSEIDON and ERS-2 data, beginning with processed data provided by Richard Ray and Brian Beckley of NASA Goddard Space Flight Center. Crawford and Cherniawsky have applied their own tidal constants near shore, and have found that the eddy first began to form off the west coast of the Queen Charlotte Islands in November 1997. (These tidal constants are provided by a detailed numerical model of tides in the Gulf of Alaska computed by the team of scientists at the Institute of Ocean Sciences led by Michael Foreman.) They also determined that some of these eddies might be the source of meanders and eddies in the Alaskan Stream (Crawford et al., 2000).

This team is presently using the same models to determine the average seasonal height of the sea surface along the Canadian margin of the Gulf (Cherniawsky et al., submitted; Foreman et al., submitted). Once combined with all satellite altimetry data, they will be able to determine absolute sea surface heights, and use them to compute northward flow of surface currents along our coast.

The Colorado web site showed Haida-1998 to be one of an annual supply of eddies that transport fresh water and nutrients into the Gulf from the Alaskan Panhandle and the Canadian West Coast. The unusually high elevation of the eddy core marks it as one of the largest eddies observed in this region. Haida eddies belong to a class of anticyclonic, coastal-generated eddies first noticed in water property data near Sitka, Alaska at 57 °N (Tabata, 1982), and later in satellite infrared measurements by Thomson and Gower (1998). Crawford and Whitney (1999) identified another region where eddies are typically generated between 51°N and 54°N, off the West Coast of the Queen Charlotte Islands. Over the years 1994 to 1999, they found that three to five large eddies formed along the Alaskan Panhandle and Canadian West Coast in any one winter.

next Inside An Eddy
back Tracking Eddies that Feed the Sea

These false-color images show contours of sea surface height from ERS-2 and TOPEX altimeters, as displayed on the Colorado Center for Astrodynamics Research Global Near Real-Time Altimeter Data Viewer web site.

  Inside an Eddy   Page 1

Whitney’s salinity and temperature measurements in August 1998 showed the waters in Haida-1998 to be to be fresher and warmer than surrounding waters below 100-m depth (see previous page). Above 100 m in depth, both salinity and temperature in the eddy were slightly lower than in surrounding waters. Dynamic height calculations, which use seawater density profiles to determine how high the eddy surface "sits" above the surrounding ocean surface, reveal that sea surface in the core of the eddy was 30 cm higher than outside the eddy. This calculation matches the altimetry measurements from TOPEX/Poseidon. Nutrient levels in its thermocline were substantially higher than in surrounding waters. (Here, "thermocline" refers to the temperature gradient across the width and depth of the eddy.) The ocean water type of this eddy matches that found near the Queen Charlotte Islands in winter (53°N, 133°W).

In February and June 1999, Crawford sent the web-generated images to Whitney at sea on the John P. Tully to direct him to the eddy’s location for sampling. His measurements taken in September ’98, February ’99, and June ’99 show the steady erosion of the nutrient excess in the eddy waters, and a three-fold enhancement of phytoplankton in the September 1998 samples around the perimeter of the eddy. Whitney’s measurements demonstrate that the eddy provided nutrient to a nitrate-starved region of the Gulf of Alaska (Whitney, Wong, and Boyd 1998).

According to Crawford and Whitney, the eddies usually drift westward and disappear within two years in deep waters in the Gulf of Alaska. These rotating masses of water average up to a two hundred kilometers in diameter, and a large eddy can contain up to 5,000 cubic kilometers of water, which is about the volume of Lake Michigan.

  John P. Tully (ship)
The Canadian Coast Guard Offshore Research & Survey Ship, John P. Tully. (Image courtesy Canadian Coast Guard)
  Eddy schematic

Crawford notes that in 1999, colleagues of theirs published a paper showing that Sitka and Haida eddies are frequently created in their computer simulations of wind-driven currents along this coast (Melsom 1999). The researchers believe it is baroclinic instability of the coastal flow that triggers the set up of eddies. ("Baroclinic instability" may occur in a flow in which there are density gradients along surfaces where the pressure is constant. Such instabilities are typically produced in rotating systems where there is ample potential energy being converted into kinetic energy.)

Based on calculations of dynamic heights of the 100-m surface relative to the 1000-m surface, using archived water property data, two of the highest-elevation eddies were Haida-1998, and Haida-1983, both generated in severe El Niño winters. This finding supports the calculations by Melsom et al. (1999), based on their numerical model.

So what’s up with these eddies now? By mid-June 2000, new Haida and Sitka eddies had drifted away from shore, and the final remnants of Haida-1998 were merging into the surrounding seas, as shown on the previous page. The eddies of 1999 were weak and by June 2000, had either disappeared or were barely visible. Whitney is senior scientist on another cruise of the John P. Tully to sample Haida-2000 in June 2000. Its position shown in places it over Bowie Seamount, a potential Canadian Marine Protected Area. "We now have a combined eddy and seamount study, with too little time to sample both," says Whitney. The cruise was set up to examine nitrate and iron concentrations in the eddy, and to map their depletion in time and impacts on surrounding biota. He hopes to examine eddy water "upstream" of the seamount, and than run a quick survey over the seamount on the way home.

The first satellite images solved a previous mystery. Canadian scientists had wondered why Bowie Seamount biota could be so similar to coastal species, when no prevailing currents flowed from shore to the seamount. However, the track of Haida-1998, passing directly over Bowie Seamount, provided the missing link. The eddies had carried coastal species away from shore right to the seamount.


Cherniawsky, J.Y., M.G.G. Foreman and W.R. Crawford, Ocean Tides from TOPEX/ POSEIDON sea level data, submitted to Journal of Atmospheric and Oceanic Technology.

Crawford, W.R., J.Y. Cherniawsky and M.G.G. Foreman, 2000: Multi-year meanders and eddies in Alaskan Stream as observed by TOPEX/Poseidon altimeter, Geophysical Research Letters, 27(7), 1025-1028.

Crawford, W.R. and F. Whitney, 1999: Mesoscale eddies aswirl with data in Gulf of Alaska Ocean, EOS, Transactions of the American Geophysical Union, 80(33), 365, 370.

Foreman, M.G.G., W.R. Crawford, J.F.R. Gower, L. Cuypers and V.A. Ballantyne, 1998: Tidal correction of TOPEX/POSEIDON altimetry for seasonal sea surface elevation and current determination off the Pacific Coast of Canada. J. Geophys. Res. 103:(C12) 27,979-27,998.

Foreman, M.G.G. W.R. Crawford, J.Y. Cherniawsky, R.F. Henry, and M. Tarbotton,: A high-resolution assimilating tidal model for the Northeast Pacific Ocean, submitted to J. Geophys. Res.

Gower, J. F. R., and S. Tabata, 1993: Measurement of eddy motion in the northeast Pacific using the Geosat altimeter, in Satellite Remote Sensing of the Oceanic Environment, edited by I. S. F. Jones, Y. Sugimori and R. W. Stewart, pp 375-382, Seibutsu Kenkyusha, Tokyo.

Melsom, A., S. D. Meyers, H. E. Hurlburt, E. J. Metzger, J. J. O'Brien, 1999: ENSO effects on Gulf of Alaska eddies, Earth Inter., 3, pap. 001, (Available at

Tabata, S., 1982: The anticyclonic, baroclinic eddy off Sitka, Alaska, in the Northeast Pacific Ocean, J. Phys. Oceanogr., 12, 1260-1282.

Thomson, R. E., and J. F. R. Gower, 1998: A basin-scale oceanic instability event in the Gulf of Alaska, J. Geophys. Res., 103, 3033-3040.

Whitney, F. A., C. S. Wong, and P. W. Boyd, 1998: Interannual variability in nitrate supply to surface waters of the Northeast Pacific Ocean, Mar. Ecol. Prog. Ser., 170, 15-23.

back Eddies in the Gulf of Alaska

  This schematic shows an idealized eddy in the Gulf of Alaska. "Isotherms" are lines connecting points of equal temperature, as on a weather map. Warm, nutrient-rich coastal water spirals clockwise, forming the core of the eddy. Phytoplankton grow in the edges of the eddy near the ocean surface, nourished by the nutrient-rich eddy water. (Image by Robert Simmon)