This paper was written as part of the 2011 Alaska Oceans Sciences Bowl high school competition. The conclusions in this report are solely those of the student authors.
Effects of melting glaciers on nutrient flux to the Gulf of Alaska
Authors
Team Auto-Eviscerators
Abstract
The Copper River Watershed is host to many glaciers. Runoff from these glaciers flows into the Copper River, which drains into the Gulf of Alaska. As glaciers melt at a faster rate due to climate change, it is projected that an influx of iron to the Gulf of Alaska will result in higher phytoplankton production, and therefore increasing salmon population. Cordova's economy is dependant on fishing in the Gulf of Alaska, and the increased salmon population will likely benefit the Cordovan economy for approximately the next 50 years. However, as glacial runoff declines due to advanced glacial recession, the influx of iron to the Gulf will correspondingly decrease, likely reducing the amount of salmon available to the fishing fleet. This would be detrimental to the economy, and therefore the town. To manage this projected problem, we propose the placement of cannery waste back into tributary streams of the Copper River. By supplementing the natural process of nutrient recycling from spawning salmon, we hope to slow the decrease in iron flux to the Gulf of Alaska.
Glaciers and Climate Change
The latest period of global warming began about 10,000 years ago when the last major ice epoch ("ice age") ended. Based on geologic and human records, it is known that a "little ice age" occurred between the 14th and 19th centuries (Fagan, 1987). This period did not last long enough for major ice sheets to grow, but alpine glaciers did advance in Europe and North America, including Alaska.
Glacial retreat has greatly accelerated since the 1950s (USGS, 2009). Since alpine areas react faster to global climate change, this glacial retreat is much more pronounced than in continental glaciers like those found in Greenland and Antarctica (Arendt, et al., 2002). This has ocean scientists concerned since melting glaciers can raise sea level, change salinity levels, and affect water chemistry in the regions where glaciers occur.
To better understand changes caused by glaciers one must look at the science of melting glaciers. As glaciers, especially alpine glaciers, move across a landscape they have a tremendous erosive force. They carve away mountains faster than any other form of erosion. Sediment carved by a glacier is carried to the terminus, where melt-water streams carry it to the ocean. Glacial streams have a high sediment load, as well as higher rates of dissolved solids than clear water streams (Ahearn, 2002). As glaciers melt faster, these glacial sediments are carried to the ocean at faster rates, thereby decreasing salinity in the immediate vicinity of the glacier, adding sediments both dissolved and suspended, and contributing to the overall rise in sea level (Ahearn, 2002).
Glaciologists measure change in glaciers by measuring mass balance. This looks at the rate glaciers accumulate ice and snow, and compare it to the rate at which it thins and retreats at the terminus. The difference between accumulation and melting is called ablation. The area of the glacier where no melting or accumulation is occurring is called the ablation zone. The USGS began a benchmark study of glaciers in Washington state, British Columbia, Canada and Alaska in 1957 that continues to the present. This study focused on rates of ablation. They reported that between the years 1957-1989 the benchmark glaciers experienced some negative and some positive mass balance years. In 1989-2004, all the glaciers in the benchmark study experienced higher rates of ablation than any year prior to 1989. This shows an overall trend of glaciers reacting to global climate change.
One major climate trend that affects glacial growth and ablation of the glaciers in the benchmark study is called the Pacific Decadal Oscillation or PDO. This is similar to the El Niño cycle in the southern Pacific. The PDO is a pattern of ocean temperatures and atmospheric pressure that changes (oscillates) on a ten-year (decadal) cycle. The PDO warm phase tends to bring more storms to the North Pacific while the PDO cold phase tends to bring storms on a more southerly track. It was noted in the USGS study that the Washington state glaciers grew during the colder phase while the Alaska glaciers retreated, and vice versa. Yet after 1989, this prevailing climate force had no effect, and all of the glaciers retreated, and lost mass (USGS, 2009).
Glaciologists have only a handful of benchmark glaciers that are easily accessible and measurable with standard measuring devices. Since glaciers in Alaska and Western Canada make up 13% of the world's alpine glaciers (Arendt, et al., 2002), it became apparent that more data was necessary to accurately catalogue climate change from this area. Currently more glaciers are being catalogued using historical aerial photography, current altimetry data, and studying landmarks left behind such as moraines, striations and the like. This research is adding to data already in place and shows similar results to the USGS Benchmark studies (Josberger, 2007). However, as the database grows scientists are seeing that the contribution from Alaska's glaciers in the Gulf of Alaska form the largest contribution to rising sea level yet measured (Arendt, et al., 2002). Studies by Arendt have noted that between 1995 and 2001 Alaska's glaciers contributed about eight percent of the increased rate of sea level rise globally, which is two times the rate of contribution from Antarctic ice sheets, and seven times more than previously calculated in the USGS (2009) Benchmark study.
The Copper River Basin in South-central Alaska and Canada is composed of both coastal and terrestrial glaciers, which both contribute to the high silt load of the Copper River. The terrestrial glaciers in the USGS benchmark study have shown increased rates of melting, though not as much as the coastal glaciers (USGS, 2009; Arendt, et al., 2002). The terrestrial glaciers add a different set of nutrients to the melt waters of the Copper River because they originate in the volcanic mountains of the Wrangell and St. Elias Ranges, where the coastal glaciers originate in the sedimentary Chugach Mountains (Ahearn, 2002).
The Copper River and Gulf of Alaska
Twenty percent of the Copper River basin is covered in glaciers (Mundy, 2005), and melting glaciers are creating an increased input of silt into the Copper River. These glaciers reside in the Wrangell and Chugach Mountains, and are at the tributaries of the Copper River. Glacial silt contains high levels of iron, and when glaciers melt, iron is released into the tributary streams of the Copper River (Campbell, 2010). The iron is then carried down to the mouth of the Copper River, where the gray silty water can be seen creating a plume in the Gulf of Alaska (Figure 2). When the Copper River reaches the ocean, it meets waters high in nitrate from upwelling.
There are two types of upwelling that operate at the mouth of the Copper River: coastal and estuarine. Both processes bring up many nutrients, including nitrate, to the mouth of the Copper. Coastal upwelling (Figure 1) occurs along the Alaska Coastal Current. The Alaska Coastal Current flows parallel to the continental shelf of the Gulf of Alaska. It is fed from glacial runoff, snowmelt, and rainfall, and fueled by coastal winds. As wind blows the current along the coast from east to west, surface waters are blown out towards the ocean, and then pushed down under the thermocline. The water displaced by the surface water is replaced by colder, deeper waters from beneath the thermocline. These cold waters are pushed back up the continental shelf and closer to shore, where they are then blown back out by wind and the cycle repeats. The deeper dense water brings up nutrients that were trapped on the continental shelf from decaying matter (Barnes, 1991). This water is rich in nitrate because there is constant mixing at the plate boundary in the Gulf of Alaska (Hales, 2005). Freshwater runoff makes the Alaska Coastal Current stronger; therefore, glaciers melting at a faster rate will strengthen it (Mundy, 2005).
Estuarine upwelling occurs when freshwater meets saltwater, or when a river meets the sea. The freshwater that comes from the river is less dense and floats on top of the saltwater. There is a strong density gradient between the fresh and saltwater layers called the pycnocline. However, the waves, wind, and tide cause mixing in the layers of water and can disrupt the pycnocline. Freshwater carries saltwater from below the pycnocline out to the ocean, and then more saltwater comes up from deep down to replace it. The deep saltwater is nitrate-rich because it has been traveling along the bottom of the ocean, where bacteria breaks down organic matter from decaying organisms. As the deep water rises, many nutrients are transported to the surface water at the head of the estuary. This cycle repeats itself because all the organic matter is trapped in the estuary (Barnes, 1991).
The Copper River Plume and Dissolved Iron in the Gulf of Alaska
Phytoplankton productivity is dependent on the availability of essential nutrients in euphotic surface waters. All growth is eventually limited by a factor such as nutrient availability, space, or predation. In the case of nutrient availability, this is called a limiting nutrient. Phytoplankton populations are commonly limited by nitrate or iron.
The majority of the Gulf of Alaska is a known high nutrient, low chlorophyll (HNLC) region, meaning it is rich in nutrients but has a low phytoplankton count. This apparent paradox is explained by the low quantity of iron, a micronutrient essential for photosynthesis. Iron in oceanic surface waters originates from terrestrial sediment that is transported by wind or rivers. Iron has a residence time of about 80 days in surface water (Ye, et al.); therefore, it acts as a limiting nutrient in areas that do not receive either freshwater runoff or wind-blown dust. Except for the occasional dust storm from the Copper River Delta, the open waters of the Gulf of Alaska do not have a significant source of particulate or dissolved iron.
The Alaskan Coastal Current is nitrate-poor; however, it is rich in iron from freshwater runoff. A major source of this runoff is the Copper River. The edge of the Copper River plume (Figure 2) is where the iron-limited and nitrate-limited waters meet to create a nutrient-rich biological hotspot. Mixing between the subarctic eastern Alaskan Gyre and the Alaskan Coastal Current results in zones of high productivity in the northwestern Gulf of Alaska (Figure 3). (Bruland Research Lab (a))
Projected Scenario
Although greenhouse gases perform the essential function of capturing heat in the atmosphere to keep earth's temperature in a state of equilibrium, human activities such as the burning of fossil fuels are increasing the amount of CO2 emissions—strengthening the greenhouse effect and thus altering earth's heat equilibrium (EPA, 2010). This shift began during the Industrial Revolution, when a sudden and continuous influx of CO2 emissions influenced earth's greenhouse process (NOAA, 2010). The adverse effects of this shift are present in the retreat and melting of glaciers across both Alaska and the globe.
As the amount of CO2 emissions in the atmosphere swells, more heat is trapped. This causes earth's temperature to rise, resulting in the melting and consequent retreat of glaciers. Whereas a snow-covered glacier has albedo with the potential of reflecting up to 80 percent or more of earth's solar radiation, the melting of this snow causes the albedo to dramatically decrease to only 30 percent (NASA, 2008). This decrease in albedo leads to the ground absorption of solar radiation and warming of "earth's surface and lower atmosphere," as the absence of snow fails to reflect such radiation back into space (NASA, 2008), creating a positive feedback loop.
Glacial retreat causes mountainsides to erode, creating iron-rich sediment that is then deposited into nearby streams like those whose melt-water contributes to the Copper River. As glaciers continue to melt as a result of global climate change, iron-rich sediment deposits continue to flow into the Copper River and the Gulf of Alaska. As most of the Gulf of Alaska currently experiences a low quantity of iron (Campbell, 2010), an essential micronutrient for photosynthesis, the region is known for its low phytoplankton concentration. However, the rapid melting of glaciers that supply water and sediment to the Copper River will create an influx of iron into Alaska's Gulf and thus augment conditions necessary for phytoplankton growth.
The size of the Copper River plume is affected by how much freshwater is delivered to the mouth of the river (Mundy, 2005). As glaciers melt at a faster rate due to climate change, there will be more glacial water transported to the mouth of the Copper River, thus bringing more iron down to the plume. Increased glacial melt will also cause a surge of water in the river, resulting in more landslides on the banks (Bowersox). Because increased freshwater runoff strengthens the Alaska Coastal Current, the current will become stronger and transport more nitrate to the plume.
The town of Cordova, Alaska is dependant on the commercial fisheries in the Gulf of Alaska. Fifty percent of Cordova's jobs are directly related to commercial fishing (Ecotrust (a)), the sole basis of the town's economy. In view of the fact that phytoplankton are by far the most important primary producers in oceanic environments, commercially important species, and Cordova's economy, are ultimately dependant on phytoplankton populations.
We predict that climate change will create a domino effect that will likely prove beneficial for Cordova's commercial fishing economy in the short-term, but will ultimately devastate the community's economy over a long period of time. The warming of the planet will cause glaciers in the Copper River Watershed to melt, increasing the amount of iron necessary for phytoplankton production into the Gulf of Alaska. An influx of iron will lead to an increase in both phytoplankton production and salmon population. This growth due to glacial melt will benefit Cordova's economy in the short-term by creating favorable fishing conditions for the commercial economy. However, the problem arises when the glaciers of the Copper River cease to exist and no longer generate sediment. When the fundamental source of iron is eliminated, phytoplankton production and salmon population will considerably diminish—as will the overall flow of money generated from the Copper River fishing fleet.
Cordova's commercial fishing economy is currently experiencing the short-term benefits of the Copper River Plume's iron influx. As shown in Figure 4, the biomass of zooplankton within the Gulf of Alaska increased during the 1980s, an increase that is likely attributed to an increase in iron, the current limiting factor of the region. This occurrence may have also played a role in salmon population within the area. As zooplankton mass grew, Figure 5 shows that the biomass of all salmon species across Alaska, apart from the biomass of coho salmon, experienced growth as well. Of these calculations, pink salmon exhibited the greatest increase statewide, topping the charts at about 150 million pinks in 1998 and experiencing peaks at 130 million pink salmon in 1991, 1994, 2001, and 2003. Seeing as how these numbers represent statewide harvests, there is no direct correlation with how the Copper River Plume's iron influx played a role. However, as depicted in Figure 6, the Prince William Sound's 2010 pink salmon harvest peaked at 24 million fish, nearly quadrupling the region's five-year average. This spike in population can be attributed to the increase in zooplankton biomass as a result of the phytoplankton increase caused by the Copper River Plume's iron influx.
If the glaciers at the tributaries of the Copper River continue to melt rapidly, more iron will flow down the river and congregate at its plume. This will create favorable conditions for phytoplankton and zooplankton production, provisions necessary for expanding Copper River salmon populations, and such harvests like that of the past season will occur more frequently over a short period of time. However, when the imminent extinction of glaciers feeding the Copper River occurs, harvests such as those experienced in 2010 will become mere legend.
In a community where 50 percent of "jobs are directly related to commercial fishing," "an additional 25 percent are indirectly dependent on the industry," and "half of all households have someone working in commercial harvesting or processing" (Ecotrust (a)), the majority of Cordova's economy and residents would be devastated if the salmon population dramatically declined. Currently, there are 343 commercial fishing permit holders living in Cordova (Cordova District Fishermen United), and of these, there are around 80 to 120 drift gillnet permits, which account for catching two thirds of the harvest (Ecotrust Fisheries, 2010), being used in the Area E fisheries within the Prince William Sound (ADFG (b)).
Although these numbers may seem insignificant, the number of dependents reliant on the income of these permits is much greater. When the industries indirectly related to these permit holders' income and the employment they provide are taken into account, the number of citizens affected by the commercial fishing industry heightens and the majority of Cordovan residents may experience financial affliction.
The devastation of the fishing economy due to changes in iron input as a result of glacial reductions or potential absence, would affect permit holders, their dependents, and Cordovan residents alike. Although this projection won't occur for about 50 years or more, commercial fishing embodies the livelihood of Cordova, and if it were brought to an end, the city of Cordova would experience a similar fate.
Following the depletion of iron to the Copper Plume, the salmon run should eventually stabilize, but at a lower level. The current, iron-infused fishery will become reminiscent of fisheries without such an iron source, similar to those in more southern locations, such as British Columbia or Washington State. However, because it is currently unknown where salmon go after leaving their natal streams (University of Washington) it is perfectly possible that such fisheries also benefit from the Copper Plume. If this is the case, and all salmon feed in the North Pacific, we can expect to see a decline in all Pacific salmon stocks.
Before this stabilization occurs, there will be a temporary spike in the salmon population, following the iron spike that should coincide with the rapid melting of the ice sheets that feed the Copper River. In this increase, as well as in the drop that will follow, the current management system, which first allows a healthy escapement, then catches what is unnecessary in the spawning beds (Ulmer and Knapp, 2004), should be sufficient to keep the population intact. However, this form of management for a smaller run will result in a smaller catch, causing the fishermen to lose money. If this cycle continues long enough—or if we drop management and try to maintain a high catch rate—eventually there may not be enough money in fish to keep Cordova alive.
Proposed Management Plan
We propose several forms of management to augment the nutrient supply, and subsequently, the salmon run. While these changes will not be enough to compensate for the nutrients supplied by the Copper Plume, they may help slow the decline and raise the level of eventual stabilization.
One of these solutions is to try to restore natural nutrient balance in the Copper River Watershed. Salmon returning to their spawning streams "provide marine-derived nutrients that support the web of life within the watershed" (Ecotrust (b)). These nutrients include carbon, nitrogen, and most importantly, iron (Helfman and Collette, 1997). While plants and other life in the watershed will take some of these nutrients up, the rest will be swept out to sea. We propose to complement the nutrient supply from salmon that swim upstream and die with waste from the salmon caught at sea. Cannery waste is currently dumped at one location in Cordova, in such high concentrations it releases ammonia as it decomposes (Hansen, 1994) and could create anoxic environmental conditions (Sarmiento, et al., 1988).
If this waste were to be distributed between the tributaries of the Copper River, it would still provide nutrients without sitting in a toxic lump in Orca Inlet. The distribution would be balanced between spawning streams—where the nutrients can be taken up by fry and fish as they die—and streams without a spawning population, where the nutrients will be more likely to be swept downstream to the ocean. Such waste will be dumped, in accordance with current regulations, in the form of small, one half inch chunks (EPA, 1975). However, it may be more efficient to dump unprocessed fish waste. At least one study found that dumping "unground fish heads and carcasses" into an experimental site located in a trench "revealed that the heads and carcasses were rapidly dispersed and incorporated into the food chain." Meanwhile, at the established, regulation dump site, "Benthic samples consisted of decomposed fish waste and noxious sediments that were essentially biological deserts" (Thorne and Bishop, 2010).
Such waste dumping would most likely take place in the summer, during the fishing season, when the canneries are actively producing fish waste in order to avoid the problem of storage. The responsibility of distributing the waste would most feasibly go to Alaska Department of Fish and Game, who could charter planes from private owners in and around Cordova. This would also create jobs for the pilots, helping to alleviate the potential job loss of fishermen.
Like any solution, this method is not without risks. Excessive dumping could trigger anoxia and ammonia overload, the same as dump sites in the ocean. If anything, the effect would be magnified as a result of the smaller size and reduced circulation in a lake. To prevent such conditions, monitoring would have to take place, most feasibly by having the crews that distribute fish waste also take water samples at their sites, and respond appropriately to the information gleaned. Dissolved oxygen and ammonia levels would be important variables to monitor, along with other potentially harmful chemicals and conditions created by the decomposition.
The optimal time of implementation would be just before the projected downturn in nutrient flow, in approximately fifty years. If we choose to make sure the situation is indeed developing before putting our plan into effect, continuous monitoring must take place, both of nutrient levels within the Copper River and of the overall volume of the plume. Satellite images of the plume and estimates of the water level could both serve to estimate size of the plume, while water samples can be tested for nutrient content. Upon detecting a drop, the plan would be put into place.
Even with an increased supply of decomposing salmon, the runs will be depleted. In order to maintain having any run at all, fishing will have to decrease with the population of salmon. In order to keep income at a high enough level to live on the profits, we may have to rely simply on supply and demand. The smaller run size will likely result in increased prices, keeping the income for Cordovan fishermen more level. Even with such an increase in price, however, the population of Cordova, and other fishing towns that depend on salmon supported by the plume will likely experience an economic downturn, followed by a decrease in fleet size and population.
If none of this happens—if the iron supply does not fluctuate at all—then our proposed action will still provide a nutrient increase. Adverse affects are likely to be minimal, if there are any. Since we are effectively restoring, or possibly enhancing the natural balance of nutrients, the nutrient replacement should still promote the salmon run even if the projected changes happen slower, faster, or not at all.
Conclusion
Although an iron influx in the Copper River Plume will provide short-term benefits for Cordova's commercial fishing industry, economic and cultural devastation may occur if the glaciers of the Copper River Watershed significantly decrease in mass. Our proposal to recycle nutrients existent in cannery waste will work to alleviate this potential occurrence. In a town whose livelihood is dependent on the resources available in the Gulf of Alaska, steps must be taken in order to ensure that such a livelihood continues to thrive.
Figures and Tables
Figure 3. This image is a composite from a 6-day period centered around July 30, 2004. This image was generated from the CCAR Global Near Real-Time Ocean Color Data Viewer using Level 3 Aqua-MODIS data.
Source: Bruland Research Lab (b)
References
- Alaska Department of Fish & Game. "Commercial Salmon Harvests in Alaska by Species, 1960-2003." Commercial Fisheries of Alaska Special Publication No. 05-09.
http://www.sf.adfg.state.ak.us/FedAidPDFs/sp05-09.pdf. 24 Nov 2010. - Alaska Department of Fish & Game. "2010 Inseason Alaska Salmon Summary." 17 Sep 2010.
http://www.cf.adfg.state.ak.us/geninfo/finfish/salmon/catchval/blusheet/summary/summary.php#pws. 22 Nov 2010. - Alaska Department of Fish & Game. "2010 Prince William Sound Pink Salmon Harvest Timing."
http://www.cf.adfg.state.ak.us/geninfo/finfish/salmon/catchval/charts/pws_pink.php. 22 Nov 2010. - Ahearn, D. 2002. The Biogeochemistry of glacial and spring-fed streams in the Copper River watershed, south Alaska. http://watershed.ucdavis.edu/copper_river/background/data/Alaska%20Paper.pdf
- Arendt, A.A., K.A. Echelmeyer, W.D. Harrison, C.S. Lingle, and V.B. Valentine. 2002. Rapid Wastage of Alaska Glaciers and Their Contribution to Rising Sea level: Science Vol 297, p. 383-385
- Barnes, R., and K. Mann. 1991. Fundamentals of Aquatic Ecology. 2nd. Oxford: Blackwell Science Ltd. Editorial Offices. 18. Print.
- Bowersox, R. "Hydrology of a Glacial Dominated System, Copper River, Alaska." UCDavis, Web.
- Brodeur, R. D. and D. M. Ware. 1992. Long term variability in zooplankton biomass in the subarctic Pacific Ocean. Fish. Oceanogr. 1:32-38.
- Bruland Research Lab (a). "Eddie transport of metals and nutrients from coast to gyre." University of California-Santa Cruz.
http://www.es.ucsc.edu/~kbruland/Research/GulfAlaska/kwbResProjGulfAk2.html. 15 Nov 2010. - Bruland Research Lab (b). "Satellite images of the Copper River and eddies delineated with the suspended particulate matter from the river plumes."
http://www.es.ucsc.edu/~kbruland/Research/GulfAlaska/kwbResProjGulfAk2.html. - Campbell, R.W. "Sediments and Freshwater Impact Plants and Fish." Delta Sound Connections 2010, Summer: 3.
- Cordova District Fishermen United. "A Historical Narrative of Fishing in the Prince William Sound." http://www.cdfu.org/about.html. 11/24/2010
- Crusius, J. "Impacts of Climate Change and Melting Glaciers on Coastal Ecosystems in the Nearshore Waters of the Gulf of Alaska." Woods Hole Coastal and Marine Science Center.
- Ecotrust (a). "Copper River Knowledge System: Economy."
http://www.inforain.org/copperriver/content/pages/background/economy.htm. 21 Nov 2010. - Ecotrust (b). "Ecology in Copper River Delta."
http://www.inforain.org/copperriver/content/pages/background/ecology.htm 19 Nov. 2010. - Ecotrust Fisheries. "The Copper River Program." 20 Jun 2010.
http://www.ecotrust.org/copperriver/. 24 Nov 2010. - EPA. 2010. Climate Change Science Facts. http://www.epa.gov/climatechange/basicinfo.html. 21 Nov 2010.
- EPA. 1975. Development Document for Interim Final Effluent Limitations Guidelines and Proposed New Source Performance Standards for the Fish Meal, Salmon, Bottom Fish, Sardine, Herring, Clam, Oyster, Scallop, and Abalone Segment of the Canned and Preserved Seafood Processing Point Source Category. U.S. EPA, Washington, D.C.
- Fagan, B. M. 1987. American Geological Institute's Glossary of Geology. 3rd edition. The Little Ice Age: How Climate Made History, 1300-1850.
- Hales, B. et. al. "Irreversible nitrate fluxes due to turbulent mixing in a coastal upwelling
system." Oregon State University Scholar's Archive. Oregon State University, 13/10/2005.
http://ir.library.oregonstate.edu/xmlui/handle/1957/16541. - Hansen, M. Ohio. Ohio Shale Concretions. 1994.
http://www.dnr.state.oh.us/Portals/10/pdf/GeoFacts/geof04.pdf. 16 Nov 2010. - Helfman, G, and B. Collette. 1997. Diversity of Fishes. 7th printing. Malden MA: Blackwell Science, pg 439.
- Josberger, E.G. et. al. 2007. Glacier mass-balance fluctuations in the Pacific Northwest and Alaska. USA: Annals of Glaciology. V. 46. p. 291-296.
- Meier, M.F. et. al. 1971. Combined ice and water balances of Gulkana and Wolverine Glaciers, Alaska, and South Cascade Glacier, Washington, 1965 and 1966. Hydrologic Years: U.S. Geological Survey Professional Paper 715-A, 23 p.
- Mundy, P.R. 2005. The Gulf of Alaska: Biology and Oceanography. Fairbanks, Alaska: Alaska Sea Grant College Program, University of Alaska Fairbanks. 35-46. Print.
- NASA (a). "Benguela Upwelling."
http://disc.sci.gsfc.nasa.gov/oceancolor/additional/science-focus/classic_scenes/00_classics_index.shtml/02_classics_benguela.shtml. - NASA. "Climate Change Resource Reel." 23 Feb 2008.
http://www.nasa.gov/centers/goddard/earthandsun/climate_change.html. 21 Nov 2010. - NOAA. "Greenhouse Gases: Frequently Asked Questions." 23 Feb 2010. http://www.ncdc.noaa.gov/oa/climate/gases.html. 30 Nov 2010.
- Sarmiento, J.L., T.D. Herbert, and J.R. Toggweiler. 1988. "Causes of anoxia in the world ocean." Global Biochemical Cycles. 2(2): 115-128.
- Thorne, R, and M. A. Bishop. 2010. "Impacts of Seafood Waste Discharge in Orca Inlet, PWS." Delta Sound Connections 2010, Summer: 3.
- Ulmer, F, and G. Knapp. 2004. "Challenges in Restructuring Alaska's Salmon Fisheries." Institute of Social and Economic Research University of Alaska Anchorage. pg 4-5.
- University of Washington. 2002. Anadromous salmonid biology Columbia Basin Research, School of Aquatic & Fishery Sciences, University of Washington.
- USGS Fact Sheet. 2009-2046. "Fifty-Year Record of Glacier Change Reveals Shifting Climate in the Pacific Northwest and Alaska, USA."
- Ye Ying et. al. Modeling the role of dust particles in Fe speciation and residence time. Bremerhaven, Germany: Alfred Wegener Institute for Polar and Marine Research.
