Observing Arctic ice-edge plankton blooms from space
False-colour satellite image of ice-edge phytoplankton blooms
Ongoing climate-driven changes to the Arctic sea-ice could have a significant impact on the blooming of tiny planktonic plants (phytoplankton) with important implications for the Arctic ecosystem, according to new research conducted by scientists at the UK’s National Oceanography Centre (NOC).
“Ice-edge phytoplankton blooms in the Arctic Ocean provide food for planktonic animals called zooplankton, which are in turn exploited by animals higher up the food chain such as fish,” explained Dr Andrew Yool, one of the team of NOC researchers.
During the Arctic spring and summer, sea-ice melts and breaks up. Freshwater from melting ice forms a blanket over the denser, saltier water below. This stratification of the water column, along with seasonal sunshine, triggers the appearance of phytoplankton blooms, which often form long but narrow (20–100 km) bands along the receding ice-edge.
Arctic ice-edge blooms have in the past been studied largely during research cruises. These studies have often focused on regions such as the Barents Sea between Norway and the Svalbard Archipelago, and the Bering Shelf bordering Alaska, where blooms are thought to account for 50% or more of biological production.
However, advances in modern satellite technology now offer the opportunity to observe and monitor ice-edge blooms at high spatial resolution over large areas and extended periods of time from space.
“Our aim was to use satellite data to get a synoptic view of ice-edge blooms across the whole Arctic region,” said Dr Yool.
To do this, the research team used daily data from the NASA’s SeaWiFs satellite, which was launched in 1997. SeaWiFs continuously observes ocean colour (sea-ice, cloud and fog cover permitting), sampling the whole globe every two days. To provide an alternative estimate of bloom occurrence, and an independent check on their findings, the researchers also used data from the MODIS satellite.
Ice-edge blooms are identified from the spectral signature of the photosynthetic pigment chlorophyll, after correction for contamination by other coloured organic matter in surface waters.
So as better to understand the relationship between phytoplankton blooms and seasonal changes in sea ice, the researchers also used information on sea ice concentrations obtained from the US National Snow and Ice Data Center (NSIDC). Their study covered the period 1998–2007.
They found that ice-edge blooms occurred in all seasonally ice-covered areas and from spring to late summer. They observed ice-edge blooms in 77–89% of locations for which they had adequate data. The blooms usually peaked within 20 days of ice retreat, sometimes forming long belts along the ice edge (greater than 100 km).
“The bloom peak is most often located close to the ice edge,” said Dr Yool, “We observed blooms propagating in a wave-like fashion behind the receding ice edge over hundreds of kilometres and over several months, while others remained stationary.”
Because of the geography of the Arctic Ocean, sea ice does not always retreat northwards. For example, in Baffin Bay and Davis Strait, west of Greenland, ice shrunk both westward and south-eastward from the north in spring and summer, with phytoplankton blooms propagating along the ice edge as it receded.
“Our findings demonstrate strong biophysical linkage between bloom propagation and sea-ice melt back, which is independent of the actual direction of retreat,” said Dr Yool.
These findings are important because they indicate that future change in Arctic sea- ice resulting from climate change could significantly impact the occurrence of phytoplankton blooms as well as the animals further up the food chain that ultimately depend upon them, including fish.
Ice-edge phytoplankton blooms also play an important role in the Arctic carbon cycle. Through photosynthesis, phytoplankton blooms draw large amounts of carbon dioxide down from the atmosphere, some of which is exported to the deep ocean.
What effects future shrinkage in sea-ice will have on the ecology and biogeochemistry of the Arctic Ocean are still largely unclear, as Dr Yool explained:
“It is quite possible that ongoing climate change will lead to ice-free summers in the Arctic within the next few decades. As the melt season becomes longer, ice-edge blooms may propagate over larger distances, stripping out surface nutrients as they go. However, whether the Arctic becomes more or less productive will ultimately depend on complex factors affecting ocean stratification and mixing, and thus the availability of nutrients in sunlit surface waters.”
Dr Yool and his colleagues hope that their findings will contribute to a better conceptual understanding of the ecology of the Arctic Ocean, which should help computer modellers forecast future changes under global warming.
The researchers are MahéPerrette, Andrew Yool, Graham Quartly and Ekaterina Popova of the National Oceanography Centre. The research work began as part of MahéPerrette’s Masters degree undertaken within the University of Southampton’s School of Ocean and Earth Sciences.
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
Elizabeth says, Am I reading this wrong or is the reference to “stripping out surface nutrients” implying a negative consequence of phytoplankton blooms in the Arctic?
As the sea ice melts it produces fresh water which encourages phytoplankton to grow thus starting the great and wonderful food chain process. Phytoplankton is food for small creatures such as krill, small bait fish eat the krill and propagate by the millions, larger fish eat the bait fish such as tuna, salmon and cod. Fisherman catch the fish and thus provide food for us all. This is one of the wonders of nature, we should appreciate it, but scientist instead tend to see gloom and doom in everything they observe.
Sounds to me like, depending upon the shape as well as extent of the sea ice that some “retreat” could actually increase the plankton and fish populations along with everything else up the food chain.
I’m with Jim G! Intuitively the further the ice retreats in the summer the wider the area that would be treated to the blooms. Even if the Arctic were to become ice free at the end of the summer it would remain so for a very short time. Have studies been done yet to see if bloom intensities can be seen in sediment deposits?
Fred H. Haynie says:
March 5, 2011 at 5:32 am
This is another bit of evidence that the freeze/thaw cycle is the factor that controls the annual atmospheric CO2 cycle, rather than trees in the mid latitudes. Will someone quantify which consumes the most CO2 and when?>>>
I will defer to Ferdinand Englebeen since he’s following this thread, but I’d think you’ve asked the wrong question by assuming the biggest factor is one bio process or the other bio process.
The biggest driver is ocean circulation. Cold water absorbs far more CO2 than does warm water. So the arctic regions are CO2 sinks, absorbing from the air, rain, snow and as ocean currents bring that water toward the equator, it warms, can’t hold as much CO2, and so CO2 is outgassed.
My guess is the amount of CO2 circulated in that manner would make how much plankton or rainforest remove fairly insignificant in any given year or cylce. you would have to add up their net effect over decades to measure a meaningfull trend compared to annual uptake and outgas from the oceans.
RE: UK Sceptic and Lawrie Ayres
Pardon my ignorance, but what evidence do you have that there was less sea ice in the 1950s than there is now? And don’t dredge up those tired shots of submarines surfacing through leads in the sea ice as anecdotal proof. Submarines could surface in the high Arctic in the dead of winter if they wanted to, provided they found a lead like those subs did. If you watch satellite images of the Arctic during the winter, you’d know those leads are not hard to find. The best hard data we have suggests that there was significantly MORE sea ice in the Arctic back in the 1950s than there is now.
http://nsidc.org/sotc/sea_ice.html
Travis,
This post by the late, great John Daly has plenty of info regarding your question.
If it’s too much information, at least read the Conclusion.
Fred H. Haynie says:
March 5, 2011 at 8:14 am
I don’t see how you can expect CO2 uptake to decrease in summer as a result of more melting. Those buggers thrive in relatively fresh water with a ready supply of sunlight and CO2. The bloom will more likely grow as more fresher water is produced.
It was your idea that expanding/melting ice sheets are the cause of the seasonal variations in CO2. But that doesn’t hold:
– Sea ice expands and melts at least with the same quantities in Antarctica as in the Arctic. Same response from phytoplankton there (there is a recent alarming investigation from the Peninsula, not the rest of Antarctica), but only a very small seasonal amplitude is observed in the SH.
– The plankton growth and the leaves growth of land plants in spring are similtaneous. Both are lumped together as the biogenic response to temperature. In contrast, the direct sink of CO2 into the oceans is opposite in sign for quantity and quite different in amplitude for d13C compared to the biogenic response.
– The oceans are net sinks in winter, despite expanding ice cover, because the mid-latitudes up to the ice sheets are absorbing CO2. Only the amount absorbed at the THC sink place, going into the deep oceans, might be influenced by the ice sheet expansion. In contrast, algal blooms are only small sources in winter (about 6 GtC total mass), as the mass is small, be it with a huge turnover in summer (most is feed for the rest of the food chain which returns it into CO2). But land vegetation is a huge source in winter (about 60 GtC land vegetation decay).
– The maximum bloom is some 20 days after ice retreat and then is reducing, depending of the availability of nutritients. Thus CO2 aborbing also depends of how much more ice is melting, which diminishes over time until late summer. More important is how much ice in total melts. I suppose that will decrease together with the decrease of total summer/winter area.
For more background, see the (somewhat outdated) NASA diagram:
http://earthobservatory.nasa.gov/Features/CarbonCycle/Images/carbon_cycle_diagram.jpg
I’m with others in thinking that this is just another example of the net positive effects of natural global warming. I cant see the alarmists liking this story much. But then again, I tend to view things in a very straight forward way. Unlike the warmists, who through a convoluted and elaborate narrative believe everything is caused by CO2.
Cheers!
Why do we think it is cute or acceptable that researchers make even token obeisance to any motley CRU or deity other than Science?
Every scientist who bends over for prejudice and bias diminishes the credibility of their work.
RE: Smokey
I’ve seen Daly’s work, and it is partially to what I refer. He, in fact, confirms that there was more ice in the Arctic during the 1950s then there is today, and when it comes down to it, he concludes there “Unfortunately there is no comprehensive ice data from the 1930s” which leads me to repeat my question, perhaps modified now that I’m reminded that Daly established that the extent was greater than today in the 1950s: what hard evidence is there that sea ice extent in the 1930s was less than (or comparable to what) it is today?
Daly’s analysis relied on an extrapolation of Arctic temperature data that is shakier than what goes on at GISS, as well as his interpretation of a series of climate patterns whose interrelationships are frankly much more complex than he admitted in his analysis (or perhaps than he knew at the time).
Daly’s analysis is at least 7 years old, going on 8. How much have we learned about Earth’s climate since his death? How much have we discovered about how the Arctic climate system works and how it is impacted by other global and regional climate cycles? Our understanding of what drives changes in the Arctic, while light years from complete, is far better now than it was in 2004. Sooner or later, it will be time to put that website to rest.
davidmhoffer,
I agree that the Arctic sea ice freeze/thaw is the pump driving that circulation and is a major part of earth’s thermastat. Ferdinand and I disagree as to what is controlling the annual cycle in atmospheric CO2. I have observed that the concentration in the arctic has a very strong statistical relationship with the freeze/thaw cycle. Ferdinand believes the annual cycle is controlled by land based vegitation in the mid latitudes as proposed by Keeling years ago.
Ferdinand,
As you know, the Arctic is water surrounded by land and the Antarctic is land surrounded by water. In the Arctic, when the ice freezes, the area of exposed cold water decreases and shifts the sink further south. In the Antarctic, when the ice freezes the sink is shifted north and area of the sink increases. Surface currents flow into the Arctic through relatively narrow passages. Surface currents flow around Antarctica and are restricted to some degree by the West Antarctica Peninsula.
The subject matter s interesting, but the article seems to just use state of the art satellite and image analysis technology to state the obvious and describe methodologies. It reads more like a grant application than a research paper. What did it find? That plankton blooms happen in spring? Wow!
“The bloom peak is most often located close to the ice edge,” said Dr Yool
Well its not exactly going to occur in the solid ice. Since the melt speed is related to spring-summer insolation, a close relation between blooms and the melting edge is not surprising either. As Alan Mitchell pointed out, defrosted phytoplankton from ice melt can aparently revive and return to growth; this would have been something interesting to talk about, it would be a good explanation for the prompt bloom at the melt edge. But it was probably either not known by the authors or judged to distract from the climate change epilogue (too much interesting genuine science can distract from the political message). BTW it should be possible to so how experimentally if phytoplankton frozen in ice can revive on thawing – that would be a worthwhile project.
Travis,
Take a look at this. http://www.kidswincom.net/arcticseaice.pdf.
Travis writes,
“Pardon my ignorance, but what evidence do you have that there was less sea ice in the 1950s than there is now?”
As you suspect, all the evidence we have points in the opposite direction — there was substantially greater Arctic ice cover during the 1950s. A recent study by Kwok & Rothrock (GRL, 2009) did in fact use data from Cold War submarine observations. They conclude that
“This combined analysis shows a long-term trend of sea ice thinning over submarine and ICESat records that span five decades.”
For example, mean submarine-observed ice thickness at the North Pole over 1958-75 was more than 3.5 meters. By 2007, the best estimates are closer to 1 meter. There’s much more in the article, which for the scientifically-inclined should put to rest the legend that submarine-in-an-ice-lead photos prove there was less Arctic ice in the 50s.
As Travis also notes, leads could open up near the North Pole then and now. They’re not what scientists mean when they discuss a “seasonally ice-free central Arctic.”
Gneiss says:
March 5, 2011 at 2:13 pm
Has it been so long ago that one forgets exactly what was being tossed about as proof of Arctic Doomsday Melt?
The images of the submarines surfacing at the North Pole in 1959 were my idea, I am the one who did that. I just had help in locating them by posters in here. They were in direct address of the panic-stricken claims of open waters in the Arctic as proof of impending AGW calamity. They are normal, and the thickness of the ice varies from year to year, as the images of 1959 and 1961 attest:
http://www.robertb.darkhorizons.org/WhatGlobalWarming.htm
and note the date of March 17 and the open water before the sun has returned to melt (a la Gore’s tipping point of no return).
There is nothing going on in the Arctic right now that is cause for yelling Fire on a Crowded Planet.
Neither is the Antartic going anywhere soon, as was claimed 35 years ago. Instead, it recovered from it’s low point and went to record high ice extents/areas.
So what if Global Sea Ice is down 1 M km2? Well within natural variances.
RE: Fred H. Haynie
I have a number of questions about that analysis, too. Foremost among them is about the supposed 309-year and the 19-year cycle. What physical processes are those supposed to model? It may have a high R-Squared value, but if the model does not adhere to any known physical process, then either it has come up with something scientifically noteworthy, or it’s completely and utterly useless for prediction.
Another question: can I assume that the discrepancy between the number of observations (386) and the degrees of freedom (375) is due to the rejection of the 10 outliers? If so, was that taken into account when the standard errors were calculated? If not, what accounts for the difference? Was the analysis on every parameter done with the same significance level?
The proof is in the pudding, of course, but I highly doubt the conclusion that we’re likely to see a new summer minimum only once every 19 years. Time will tell.
Travis,
This is a statistical model of observations. I have yet to check cause and effect. I suspect the 19 year cycle is related to the multi-decadal changes in equatorial SSTs.
The 309 year cycle may be associated with changes in the deep oceanic currents.
A degree of freedom is used each time you fix a value (coefficients or constants). Four coefficients are produced in the regression analysis. The other degrees of freedom were fixed in finding the constants in the cyclic functions. The spreadsheet program does not consider these constants in calculating the error, but there are enough remaining degrees of freedom that it makes little difference. The extreme values were adjusted to model values and not disregarded. The number of values was not reduced.
I’m not sure how this would effect the error statistics. It improves the fit.
Fred H. Haynie says:
March 5, 2011 at 12:56 pm
davidmhoffer,
I agree that the Arctic sea ice freeze/thaw is the pump driving that circulation and is a major part of earth’s thermastat. Ferdinand and I disagree as to what is controlling the annual cycle in atmospheric CO2. I have observed that the concentration in the arctic has a very strong statistical relationship with the freeze/thaw cycle. Ferdinand believes the annual cycle is controlled by land based vegitation in the mid latitudes as proposed by Keeling years ago.
______
Why would the cause of annual CO2 cycle have to be just one thing? Certainly it is based on the seasonal vegetation cycle, but could also have some statistical relationship to the freeze/thaw cycle. Point of fact, these two cycles (vegetation and freeze/thaw) are of course statistically related, but as we all know, a statistical relationship only indicates a common connection, not causation. The change of seasons (i.e. increasing or decreasing solar insolation) is the connection, and a direct causation of annual CO2 cycles is the increase or decrease in vegetative activity the seasons come and go.
Skeptics are betting that the multi-decadal decline in Arctic Sea ice is part of a longer-term natural cycle, and that soon (yes, it must be very soon, as in the next few years, for them to be right) the year-to-year Arctic Sea ice must begin a long-term incline, as this mysterious (PDO, AMO, solar, or what have you) natural cycle swings the other way. Warmists would posit that GCM’s have it generally correct and that we’ll see a seasonally ice free Arctic in this century– all because of the fact that CO2 levels are at their highest levels in at least 800,000 years.
Gates says:
“Skeptics are betting that the multi-decadal decline in Arctic Sea ice is part of a longer-term natural cycle, and that soon (yes, it must be very soon, as in the next few years, for them to be right) the year-to-year Arctic Sea ice must begin a long-term incline, as this mysterious (PDO, AMO, solar, or what have you) natural cycle swings the other way. Warmists would posit that GCM’s have it generally correct and that we’ll see a seasonally ice free Arctic in this century– all because of the fact that CO2 levels are at their highest levels in at least 800,000 years.”
There is so much wrong with this comment that it begs debunking. That’s not hard to do with an R Gates post:
Skeptics are not ‘betting’ anything. The scientific method, ignored by alarmists except for lip service, requires skepticism. Gates doesn’t qualify as being scientific; he’s just spouting opinion.
Where do skeptics say a ‘longer term cycle’? That’s just a strawman argument. There are many climate cycles, as Gates shows with his comment on the [to him, ‘mysterious’] AMO, PDO, solar, etc. cycles. Rather, Gates attributes the entire 0.7°C warming to CO2. As if.
What Gates always tries to avoid is the climate null hypothesis, which has never been falsified. He avoids it because it is a cornerstone of the scientific method. Kevin Trenberth is so upset that the null hypothesis falsifies his alternative CAGW hypothesis that he demands that it must be changed to Trenberth’s own cherry-picked “null hypothesis.” That clearly shows how angry Trenberth is that both the real world and the scientific method are not cooperating with his lucrative alarmism. Gates is just repeating the talking points.
To falsify the null hypothesis the temperature would have to exceed the parameters of the Holocene. It isn’t even close. The current *mild* 0.7°C warming over the past 150 years is entirely consistent with numerous past warming cycles. Actually, it is smaller than most. As anyone can see, it has been considerably warmer many times over the past ten millennia, which means that the Arctic has melted many times prior to the industrial revolution. It’s a natural occurrence.
Finally, I am still waiting for Gates or anyone else to show us real world evidence of global harm as a result of the increase in the tiny trace gas carbon dioxide. Despite endless predictions of runaway global warming, no one has ever identified global damage resulting from CO2.
Thus Gates’ presumption, based on ignorance, is that the cause of the entire 0.7° warming is due to CO2, is nothing but a WAG, because there is zero empirical evidence to support it. It is the argumentum ad ignorantium fallacy: “Since I can’t think of anything else that would cause the warming cycle, then it must be due to CO2.” How stupid is that?
Debunking complete.☺
R. Gates,
I’m going on what I observe and not on what I think should happen if this is causing that. Place your bets and lets see how it goes. I expect the next 19 year minimum will occur around 2028. A 309 year minimum is expected around 2067. I might not be around for either event, but I could set up a trust for my grand children.
RE: Smokey
Somkey Says:
“To falsify the null hypothesis the temperature would have to exceed the parameters of the Holocene. It isn’t even close. The current *mild* 0.7°C warming over the past 150 years is entirely consistent with numerous past warming cycles. Actually, it is smaller than most. As anyone can see, it has been considerably warmer many times over the past ten millennia, which means that the Arctic has melted many times prior to the industrial revolution. It’s a natural occurrence.”
Your null hypothesis is insufficient. No credible climate scientist would claim that warming episodes have not occurred naturally in the past. But with the existing unknowns in the climate system, a null hypothesis that simply states that the warming is natural is unsatisfactory. The null hypothesis must be that the current warming trend is explained by known natural phenomenon, which any credible “skeptic” must agree is false. It would be a mistake, however, for a skeptic to casually dismiss the alternative that humans are responsible for the current warming. Unless they, too, can come up with a physical process that accurately models recent changes in climate, than their suppositions are just as worthless as they claim “warmist” science to be.
The mistake of a casual AGW proponent, likewise, would be to assume that the logical alternative hypothesis would be that rising levels of GHGs are the sole cause. The current warming trend is a complex issue involving multiple factors, and necessitates many tests and many observations, many of which are unfortunately not being made.
If anyone on either side really wants this to come to any firm conclusion, then we need to press for MORE climate research, not less. It pains me to see people disparage science and scientists and jeer about incompetence when an endeavor fails (like the OCO last year and the GLORY satellite this year). Climate is not an issue that can be resolved politically, and public opinion will not affect whether the “skeptics” or the “warmists” are correct. It just makes one of them look bad in the end.
Travis says,
“The null hypothesis must be that the current warming trend is explained by known natural phenomenon…”
That isn’t correct at all. It is not necessary to explain the mechanism. Otherwise, all mechanisms would have to be explained. For example, gravity is a fact, but the Higgs boson has still not been found despite tens of $billions expended in the search, and for all we know it could be a model error. The mechanism for climate cycles [PDO, AO, AMO, etc.] are even less certain. But that does not affect the null hypothesis.
The null hypothesis is simply the statistical hypothesis that states that there are no differences between observed and expected data. The purported expected data from the alternative hypothesis is that CO2 will cause runaway global warming [CO2=CAGW]. But there are no differences between the observed data and the historical data. Therefore, CO2=CAGW is falsified. QED
Smokey says:
March 6, 2011 at 11:43 am
“Where do skeptics say a ‘longer term cycle’? That’s just a strawman argument.”
_____
The much beloved hero of the skeptics, Joe Bastardi (I actually like the guy myself, though disagree with his sea ice analysis) is constantly speaking about the PDO, AMO, and other natural cycles and how we are seeing them shift back to a their cold cycles and this is going to cause the year-to-year Arctic sea ice to begin to expand again as well, so don’t give me this nonsense about a straw-man argument, because it is just that…nonsense.
Also, in regards to the decline of the Arctic sea ice, and of global warming in general, it is beyond “known” natural variability, and your inability to understand this provides me ample reason to doubt your overall understanding of the topic. I believe that the GCM’s have it generally correct– CO2 is causing both the warming and the decline in sea ice. If they are incorrect, it is one of the most remarkable of coincidences that they could correctly model these general trends and just so happen to have correctly modeled what some unknown natural cycles are causing. The GCM’s represent an Alternative Hypothesis, which of course need not in any way be the opposite of, or negate a null-hypothesis. The GCM’s collectively embody an Alternative Hypothesis and predict the general trends in the the climate that we’ve seen (through back testing) and are continuing to see. In this sense, Trenberth is correct, and the collective Alternative Hypothesis as modeled and thus embodied in the GCM’s (i.e. the 40% increase in CO2 and related feedbacks) now represent a working hypothesis that has predictive power and that some other hypothesis would have to either replace or refute.
Re: Smokey
“That isn’t correct at all. It is not necessary to explain the mechanism. Otherwise, all mechanisms would have to be explained.”
That is taking it to the extreme, perhaps, but I’ll concede the point on those grounds. However, your null hypothesis does not even seem to assess the cause of warming. You state:
“To falsify the null hypothesis the temperature would have to exceed the parameters of the Holocene. It isn’t even close. The current *mild* 0.7°C warming over the past 150 years is entirely consistent with numerous past warming cycles.”
Can I ask you to put that into words as a formal null hypothesis? What you say here doesn’t seem to even assess the CAUSE of warming, just whether it is happening or not. If you are going to compare it the scope of the entire Holocene, then of course it’s not statistically significant. As I’ve said, no respectable climatologist would claim that warming events have not happened in the past without aid of mankind.
Yet in that light, what do you mean by saying that the past 150 years of warming is “consistent” with numerous past warming cycles? Do you mean that it’s the same length? The same magnitude? The same cause? All of the above? A null hypothesis means nothing unless it clearly defines the parameters and population being targeted. So let’s not be nebulous. What is your specific claim?
Suppose we start with “The trend in temperature over the past 150 years is not statistically significant.” I hope we can agree to reject that without argument. I disagree with the premise that since warming has happened before, it necessarily implies that it is entirely due to natural factors now (just as I reject the notion that it is entirely due to mankind’s influence). Climate scientists, in any case, claim to have falsified the null hypothesis that “the current warming trend can be explained solely by known climate cycles and variability” which you seem to dispute in turn, or at least disregard in favor of hypothesized yet heretofore unknown or hypothetical natural factors. Where does that leave us?