Friday, February 15, 2008

Oregon ocean dead zone linked to climate change

It ain't natural, say scientists studying Oregon's ocean dead zones. Climate change seems to be causing the ugly die-off of ocean animals.

It's a scary story, stronger winds push this natural ocean breadbasket over the edge into a low oxygen killing field. This type of unpredictable change has ocean scientists chewing their fingernails and losing sleep at night.

The Oregon coast is a place of bountiful food production, a natural gift to salmon, orcas, and people. Upwelling of nutrients (fertilizer) is triggered by routine strong winds in the spring. The bloom of small plankton that follows creates a feeding frenzy of ocean animals and people alike.

But now climate change seems to have strengthened the winds and spoiled the party. It's too much of a good thing, something most of us can relate to every year the morning of January 1st.

We've heard for years that climate change is likely to have surprising effects, and now the Oregon dead zone is emerging as perhaps an early example. Yikes.


Anonymous said...

I read the paper and watched the video today. I think it's worrisome that they are starting to have issues with hypoxia and anoxia, but I'm wondering if they invoked global warming a little too quickly. Any idea if there is any coring data from the affected region that would provide a longer term perspective on what the history of dissolved oxygen is in the region? I realize their data goes back to 1950, but I'd be curious to know how frequently they were sampling back in the 1950s, 60s, and 70s. Looking at the number of casts taken in their figure 1, you can see a definite increase in the frequency of sampling over time and I'm wondering if they simply missed previous hypoxia/anoxia episodes. Additionally, is DO = 0 really that much worse than a DO = 1 (which happened at least occasionally between 1950 and 1999)? I would think that either case is going to kill pretty much any large macroinvertebrate in the area. It seems like the decrease in DO could also be due to an increase in anthropogenic nutrient loading.

Mark Powell said...

Climate change link seems compelling to me. Human nutrient loading is a non-issue in this open ocean area.

Historical data are imperfect, but adequate to support the climate change link IMHO. And for oxygen, 0 is worse than 1ppm. Many animals can probably tolerate 1 longer than zero, and perhaps flee or wait out the problem.

Anonymous said...

I don't know if your argument about it being open ocean makes sense. They talk about the hypoxia/anoxia occurring on the continental shelf in waters 60 m deep and within 2 km of the surf zone. The hypoxic area in the Northern Gulf of Mexico extends 50 to 100 km offshore and the plume of nutrients from the Mississippi River extends at least that far offshore and most of the distance along the Louisana coast into the waters of Texas. I would think that coastal runoff in Oregon would be able to extend at least 10 km offshore. I'm forgetting some of my physical oceanography, but wouldn't the Oregon coast be getting some of the nutrient load from the Fraser River, Columbia River, and the Puget Sound (or does the current go in the opposite direction)?

As for the fleeing, most of the animals that are capable of fleeing would start fleeing at DOs greater than 1.5 and maybe as high as 3-5 mg/L. Many would already be dead by the time DO dropped down to 1 mg/l. I think if there was any difference, it would be for the infauna that aren't able to leave.

Mark Powell said...

Bit of speculation here...the Oregon coast has fast currents, strong winds and rough, turbulent water, and I suspect the residence time of coastal waters on the shelf can be fairly short. Also, the watershed area contributing nutrient runoff is smaller than the Mississippi basin. I think upwelling dominates nutrient levels, not runoff, but I'm not sure. The Oregon coast ain't the Gulf of Mexico. Why do you doubt the climate change link?

Anonymous said...

I doubt the global warming link (but don't totally reject it as a possibility) because they don't appear to consider any other possible causes for the presence of hypoxia and anoxia on the Oregon shelf. Thus, they haven't eliminated other causes of hypoxia and anoxia as likely causes. They also haven't proven that it is a lack of sampling which prevented us from seeing this in the past.

Looking at the figures, it is fairly clear that while they have several thousand data points, many of those data points are coming from samples that were taken along transects and are therefore likely to be autocorrelated both spatially and temporally. This together with the fact that the transect data is likely collected over a short period of time leads me to think that the intensity of the sampling between 1950 and 1999 is not as intense as one might think after a quick glance at the figure. So in summary, this leads me to wonder if hypoxia wasn't noticed in the past simply because of a lack of sampling. Notice that in panel 1a there are several points that come close to their definition of hypoxia, 0.5 mg/l. The average number of samples collected per year increases from 64/year for 1950-1999 to 167/year for 2000-2005 which is nearly triple. They could potentially counter my sampling claim by collecting sediment cores and looking for a DO signal in the cores.

As you can probably guess, I think an increase in anthropogenic nutrient loads is more likely. I would base this argument on the rapid growth in population in the area which would increase the load from sewage. Additionally, if there's a lot of construction and land clearing, along with removal or modification of wetlands this could increase the nutrient loads getting into coastal waters. This argument could be tested by trying to look at longterm trends in nutrient concentrations in the coastal waters.

Maybe someone with more knowledge of physical oceanography than me could answer this... When there is strong upwelling, does this slow the mixing of water from coastal runoff with ocean water? I would think that this might trap the freshwater nearshore. It might also result in very high levels of nutrients with nutrients from both upwelling and runoff. If the water is being retained for awhile, it may also be warmer, allowing for better phytoplankton growth and thus more intense hypoxia.

Anonymous said...

Just as a quick follow-up comment. I believe you are right that upwelling dominates runoff for the nutrient loads on the Oregon coast.

Using the operational definition of hypoxia in the Gulf of Mexico and Chesapeake Bay (DO < 2 mg/l) the Oregon shelf has been experiencing hypoxia for quite awhile. DO<1 mg/l is not unheard on the shelf based on Fig. 1A. I wonder how much of an increase in nutrient loads would be required during a year with strong upwelling to cause a further 0.5 to 0.8 mg/l drop in DO? I would guess that the anthropogenic nutrient load could do it.

Mark Powell said...

Aaron, Have you ever seen the coast in this area during heavy wind/upwelling periods? It's rough and the ocean is in motion. Great for windsurfing.

Upwelling is so dominant that changes in wind and upwelling are much more likely to affect productivity and hypoxia than anthropogenic nutrient runoff. Wind and upwelling change currents and mixing also, so you can't just assume that river plumes add in a simple way to nutrients transported into the system from upwelling.

As for oxygen levels and animal responses, that's what I studied way back when I was a marine biologist, and I think 0 is a lot worse than 1 ppm for animals adapted to this upwelling-driven system. They can probably tolerate low oxygen fairly well, but not zero oxygen. Some pelagic crustaceans have well-developed adaptations to exploit the oxygen minimum zone.

tonyd said...

As someone who works in the Oregon upwelling system, I think I can fill in some of the gaps from the questions posted here. I can confirm that anthropogenic nutrients are not an issue off the Oregon coast during the upwelling period. First, the rivers here are much smaller (the Columbia River may be the lone exception) and do not have a watershed anywhere near the size of the Mississippi. In addition, rainfall and riverflows are highly seasonal...high in the Nov-April period and low in the summer when upwelling occurs. In fact, riverflow is so low during the upwelling season that colleagues of mine have documented upwelling-derived nutrients 10-15km upriver due to tidal transport.
Mark-your point about the speed of the currents is also accurate...the areas that hypoxia was first observed by the research group that published the paper (early 2000's) was in areas that physical oceanographers showed had longer residence time due to complex topography off the coast. In other areas, particularly those with narrow shelves, the currents can be quite fast during upwelling resulting in low residence times reducing the opportunity for hypoxia to develop.

Personally, I feel that invoking climate change is still premature. I have looked at a bunch of the old reports and similar low oxygen measurements were found on research cruises in the 1970's. Also - and this will show my bias as a benthic ecologist - they have little to no data on the role of the seafloor in drawing down oxygen levels. There is little accumulation of the plankton production in shelf sediments which are almost pure sand out to 90-100m depth. I did a series of cruises in 2003-2005 looking at benthic communities in some regions where hypoxia is thought to occur and saw little evidence that benthic invertebrates were impacted by low oxygen...although lower oxygen in the bottom water is common in that area of the shelf (~1/2 saturation in surface waters). The organisms we identified were large, long-lived species more common in undisturbed shelf benthos. A caveat here is that much of the measured regions of hypoxia/anoxia were closer in to shore where rocky reefs and associated crab and rockfish communities occur.

As far as the density of sampling, that has definitely changed in the late 90's when a large study was funded to look at the Oregon upwelling system followed by the PISCO project (awarded to the Lubchenco/Menge group as lead investigators) which involved high frequency cruises and shallow water moorings which increased the data density during the time periods when hypoxia occurs on the shelf.

I hope this info clears up some questions and adds to the discussion!

Mark Powell said...

Thanks Tony, good explanation and your doubt of climate change links is credible. Obviously, more needs to be done to nail down causes. The Science paper hypothesis is useful and credible, and it gives everyone something to debate and evaluate. We'll get to answers on climate change links for the Oregon dead zone.

Anonymous said...

Tonyd maybe you can answer this...

Is there vertical stratification in the area where the hypoxia is occurring?

From Mark's description of the waters being rough and there being lots of wave, I would get the impression that the water column is well mixed. This mixing would be expected to prevent hypoxia and anoxia from occurring in the area as the system would be re-aerated by the wave action.

tonyd said...


There is some vertical stratification as colder (and sometimes more saline) water is brought up from the bottom to surface waters during upwelling. The vertical stratification is one of the major issues with the Gulf of Mexico hypoxic zone as the less dense Mississippi river water lays on top of the more saline, more dense gulf waters. In contrast, the density difference is not as dramatic off of Oregon and upwelling is NOT constant. We have upwelling favorable winds from ~April through September but winds do not always come from the north. The winds often die down or even reverse in events called relaxation events. During these events the stratification is much reduced and exchange between surface and bottom waters can occur. The typical frequency of relaxation events is 7-10 days so there tends to be pulses of upwelling-stimulated phytoplankton blooms followed by periods with lower primary production. The issue is that on years where winds are more intense (here is the key climate change issue) and continuous than is typical then the production and stratification are both increased, reducing vertical mixing, and isolating more water column production in deep water where metazoan and microbial respiration can draw down the oxygen levels.

Anonymous said...


George said...

I am interested in if anyone has heard anything concerning this Oregon Dead Zone for 2007 or 2008. All of the information on the net stops after the 2006 event.

Unknown said...


Hypoxia did return in 2007. Here is a local article on the extent (Oregonian):

Information on Hypoxia monitoring off Oregon is publicly available at the Oregon Coastal Ocean Observing System website ( which has realtime ocean conditions and their hypoxia page ( which has information and a calendar on current sampling. It is also a site that researchers studying OR/WA hypoxia share information. Another resource would be the Partnership for Interdisciplinary Studies of Coastal Oceans (PISCO)website that has been doing most of the monitoring (

George said...

Thanks Tony!