Experiments to date (2003)

[This page is incomplete and under development ]

The following map indicates sites of iron-fertilisation experiments to date

The Sites marked above indicate the sites of iron fertilisation experiments, as follows:

• A: IronEx I
• B: IronEx II
• D: SOIREE
• E: EisenEx
• G: SEEDS
• H: SOFeX
• J: Planktos
• K: SERIES

Planned projects, where known, are mentioned under Future Projects below, and a summary of (most) of the experiments is also given.

IronEx I (A)

Objective

Experiment

The experiment consisted of a single addition of iron to a patch of water near the Galapagos Islands, which spread to cover a 64 km patch over a 24-hour period.
The chemical mixture, as listed below, was dispensed into the ship's propeller wash. Horizontal mixing homogenized the surface streaks of iron within one day, and convective overturn of the mixed layer mixed the iron down to 35m in the first 24 hours of the experiment.

The patch was followed for 9 days, with comparisons of its biology and chemistry being made with the surrounding unfertilised waters throughout that period.
During the first five days, the patch maintained its integrity surprisingly well. On the 6th day, ocean currents caused the patch to subduct to a depth of 20 - 40m. Although it was still trackable under the surface, the experiment was compromised because the subduction subjected the phytoplankton to a light-shift, thus introducing a new variable
[BUT- above i said it got mixed to a depth of 35m in the first 24 hours ???
Both figures obtained from http://earth.agu.org/revgeophys/chisho00/node5.html]

Iron-enrichment bottle experiments were done in parallel with the patch experiment, and some showed striking differences between the results from the open sea.
Much of what we believe about primary productivity in the sea has come from bottle experiments. These comparisons are therefore critical to our understanding of the accuracy and bias of the data from both this, and previous, ocean-water experiments.

The ambient iron concentration at the experimental site was 0.1 nM Fe, and the target concentration for the patch was forty times that (4 nM Fe).

Chemicals used

Results

The addition of the iron caused an initial doubling of the amount of phytoplankton, and the rate of growth quadrupled.
However, after only one day, the phytoplankton activity leveled off. This was probably due to the rapid loss of the iron.

The fate of the iron

'Dissolvable iron' was measured regularly. (This is only one of many fractions of iron, but one that can most easily be measured, and that may be the most bioavailable form.)
The highest concentration of iron in the patch was 3.6nM, within 24 hours of fertilisation. This was a reasonable achievement of the target concentration of 4nM. However, the value decreased steadily each day, and dropped below the detection limit of the method (2 nM) by the 4th day.
This apparently reflected a net loss of available iron from the system.

Phytoplanktonic response

All three measurements revealed quite clearly that the phytoplankton in the iron-enriched patch were stimulated when compared with the assemblages outside the patch.
Furthermore, the results indicated that all size categories of the phytoplankton were stimulated by the added iron. This response contrasts strikingly with the results reported from the parallel bottle incubation experiments, in which the response was usually dominated by one particular group, the pennate diatoms.
However; the later experiments did detect a significant increase in large-celled and pennate diatoms in preference to other species and size categories of the phytoplankton.

Although the stimulation in the patch was almost instantaneous, the total increase in chlorophyll was rather modest, and appeared to level off after the first two days.

CO₂ drawdown

Prior to the experiment, calculations were made about how much CO₂ would be drawn down from the atmosphere, assuming that all the naturally-available nitrogen and phosphate was taken up by an increased phytoplankton population.
[Confirm calculations were based on N and P, not on the added Fe]
However, the total drawdown of CO₂ in the iron-enriched patch was only 10% of that expected on this basis. As Dr. Richard Barber, the chief scientist on the cruise, put it: Apparently the phytoplankton in the patch hadn't read the literature.

Conclusions

Whilst the experiment proved conclusively that iron is a limiting factor in phytoplankton growth, it was also considered to be relatively ineffective in reducing the amount of carbon dioxide in the air above the waters where the iron sulfate was added.
Only ten percent of the carbon dioxide was removed, which was very low compared to the expected amount, given the amount of iron used.

A lead researcher, Kenneth H. Coale, claimed: "The conclusion, then is that these waters must be deficient in one or more necessary trace nutrients, and adding the most limiting nutrient to those waters should promote rapid growth of phytoplankton. In theory, at least, rapid growth of phytoplankton should translate into increased removal of carbon dioxide from the atmosphere."

The researchers of the IronEx project are not necessarily supportive of tinkering with the oceans on such a large scale. They only wish to show the validity of John Martin's hypothesis.

return to top

IronEx II (B)

Objectives

Experiment

Although the same quantity of iron (450kg, or half a ton) was used for both IronEx I and IronEx II, the second experiment was significantly different from the first.
Instead of dumping in all of the iron at once, the iron was added in three separate [[servings]] over a week. This eliminated the problem of the iron immediately sinking to the ocean floors.

The iron spread to cover a path of 76 square kilometres (27 square miles)
The iron was in the form of common industrial iron sulphate, which is widely available. It is of low toxicity, and in the concentration used is considered to be non-toxic in the marine environment.

Isotopically-labelled nitrogen was used to establish the difference in nitrogen uptake, comparing 'in-patch' and 'out-patch' areas of ocean. Measuring and trackng absorption of certain forms of nitrogen (such as ammonium urea and nitrate) offers a sensitive and precise method of tracking and quanitify the growth of certain organisms.
This provided a way to distinguish how much of any increased biological activity occurred in bacteria, and how much in phytoplankton. [and how do was the increased biological activity detected - can;t ahve been chlorophyll?? ]
The distinction is important, as phytoplankton are a food-source, but sink out of the surface water much more rapidly than do bacteria. Increased activity by bacteria would thus show increased biological activity, but be of little use in feeding the food web [??? or sequestering CO2??? what happens to the bacteria? why do i care?? ?

Results

Within a week,

• the electric-blue sea turned green with algae and other phytoplankton
• two million pounds (~900,000 kg) of additional phytoplankton had grown
• concentrations of chlorophyll (indicating plankton biomass) were increased by a factor between 30 and 40

The results in this experiment were visable, with the sea turning into a thick, green 'soup', as enormous phytoplankton bloom developed. Once the algae bloomed, fish and macro-zooplankton such as sharks, turtles and squid congregated for a feast. This grazing pressure by zooplankton was also monitored. Interestingly, zooplankton did not perform typical vertical migration behaviour. [Whatever this is.] Instead, they remained with the patch/bloom to feed both day and night.

The measured concentrations of chlorophyll also indicated increased productivity of the phytoplankton. This, in turn, imples that more carbon dioxide had been removed from the air.
At ten days into the experiment, the concentration of carbon dioxide (dissolved in the surface waters) had gone down by 20%.

Phytoplanktonic Response

The additional ~900,000 kg of additional phytoplankton that were produced has been calculated to be equivalent to the biomass of 100 full-grown redwoods.
The isotopically-labelled nitrogen indicated that all of the biological activity resulting from the iron-fertilisation was from increased phytoplankton, with no change in the bacterial abundance being seen.

The phytoplankton increase produced by the iron fertilisation was dominated by diatoms.

Mesozooplanktonic Response

Mesozooplanktonic are organisms between 202 - 2000um, and feed on phytoplankton[- i think ]
Estimates of the carbon consumed by mesozooplankton suggested that at the peak of the patch-bloom, the mesozooplankton grew at a maximum rate for the given temperature.
However, over the same period of this maximum growth, the [[meso]?? abstract says zoopankton?]zooplankton abundance and biomass declined rather than increased.
This premature decline of mesozooplankton may simply be due to an increase in predator activity. However, more ominously, it may be a relfection of the failure of their reproductive output to produce viable young when the food resources had become dominated by diatoms.

CO₂ drawdown

The experimenters calculated that the ~900,000 kg increased biomass had pulled around 2,500 metric tons (2.25  million kg) of CO₂ out of the atmosphere.
From an initial fertilisation of 450kg iron, this offered a ratio of better than 5000:1.

Conclusions

A third planned project, IronEx III, to investigate fertilisation of the Ross Sea did not take place.

return to top

Soiree (D)

Objective

An international team of scientists from 6 different countries worked with NIWA, to investigate the effects of adding iron to the Southern Ocean, south of Australia.

Experiment

The fertilised path of the Southern Ocean was a zone of 50 km².

Around 15,000 litres of 'iron slurry' were poured into the Antarctic waters, where it was mixed by the cruise ship's (R.V.Tangaroa) propeller wash.

Iron

About 2 tonnes of iron in the form of iron sulphate contained in slurry was added to the Antartcic waters, over a 50 km² area.
The inert tracer dye sulphur hexafluoride (SF₆) was also used, to help follow the iron as it spread.

Results

The 'mixed layer' was relatively deep, and winds were considered moderate throughout the experiment.
However, the team on the Tangaroa had to contend with 55 knot (80 k/ph) winds and 10m waves during the experiment. This weather was responsible for spreading the SF₆-marked iron fertiliser from the initial 50 km² seeding area, to cover an areal extent of about double that during the 2 weeks of the experiment. The spreading continued at a similar rate, to over 1000km² by about 4 weeks later, as indicated by satellite data.

The addition of relatively small amounts of dissolved iron over a 50 km² area raised iron levels to 10 times the concentration normally found in these waters. Within two weeks, phytoplankton abundance was 10 times greater than outside the fertilised patch. This increase was mainly accounted for by diatoms.
This enhanced algal growth promoted a decrease in DIC and in the fugacity of CO₂, which after 13 days had decreased by 10% of its initial value.

The growing patch of phytoplankton also produced significant quantities of dimethyl sulfide, a gas known to be important in cloud formation.

Bloom persistence

Remarkable images from a NASA satellite show that six weeks after the fertilisation, the bloom had grown to cover an area of 1100 square km. The persistence of the bloom for such a long time was entirely unexpected, and is thought to be due to the ability of the phytoplankton to release special substances into the water that hold the iron in a form they can use.
This artifically-created bloom lasted for more than 50 days, whereas naturally-occuring blooms in the same region have been reported to last for 15-20 days.

The spread of the initial patch into the large bloom seen in the satellite images vividly demonstrates the sensitivity of the Southern Ocean ecosystem to small increases in iron concentrations.

Caveats

Although there was a spectacular burst of algal growth after the fertilisation, no appreciable removal of carbon from the surface to deeper waters was measured.
It is possible, but by no means certain, that the lack of removal of any carbon was due to the weather spreading the fertilised patch over a large area. By this lateral spreading of algal growth, the aggregation of particles into sufficiently dense accumulations to sink may have been impeded.

It is likely that, once the bloom finished, some of the carbon contained within the algae was eventually released back into the atmosphere.
Large-scale fertilisation would be likely to cause substantial changes to the naturally occurring ecosystems of this pristine environment.

Conclusions

SOIREE showed conclusively that fertilising the Southern Ocean with iron causes the algae to bloom.
While the results of the SOIREE expedition do not support the use of ocean fertilisation as a way of preventing climate change, they do show that iron fertilisation would produce blooms of an intensity never normally seen in the Southern Ocean.

return to top

EisenEx (E)

After the first 'IronEx' project, where 'eisen' is the German for 'iron'.
Sometimes referred to as the cruise "ANT XIII/2 from RV Polarstern"

Objective

The goal of EisenEx was to monitor the biological and chemical responses to fertilisation in a patch of water north of the Antarctic Polar Front (APF) in the Polar Frontal Zone (PFZ).
At the PFZ, excess macro-nutrients that are unused are subducted. Thus, increased primary production in that area would result in subduction of biologically-fixed carbon, instead of these unused nutrients.

Experiment

The experiment took place within an eddy shed by the Antarctic Polar front. Approximately 500 square kilometers of ocean surface within Polar Front were fertilized with iron, and the plankton development was followed for 3 weeks.

Iron sulphate solution was added in 3 batches, each a week apart. During the experiment, the fertilised patch circled within the eddy, while increasing in area from less than 50km² to roughly 500km².

Iron

Iron sulphate solution was added in 3 batches, each a week apart.
(The Ferrous sulphate variety was used, that is available in gardening shops touted as an anti-moss agent!).
The first was marked with the inert tracer SF₆.

Results

The following results are based on 'in-patch out-patch' comparisons; i.e. differences between water samples taken from both within the experimental 'eddy' patch, and outside it. Any differences due to the eddy are thus not highlighted [[as far as i can tell]].

During EisenEx, the first response of the phytoplankton was found 2 days after initial fertilisation.

Over the whole duration of the experiment, chlorophyll concentrations increased fourfold inside the patch, compared with the surrounding water.
The carbohydrate concentration in the particulate fraction had also doubled in this period.
This large increase in biomass and carbohydrates was caused mainly by the growth of diatoms, with large cells. These contributed about 75% of the biomass by the end of the experiment. This group of phytoplankton were clearly seen to out-compete other groups present when sufficient iron was made available.
However, bacterial production, microzooplankton biomass and microbial food web activity also increased on iron addition. Pronounced shifts in the relative species concentrations were noted.

The biomass increase was accompanied by a decrease in (oceanic) CO₂ partial pressure, which was partly replenished by atmospheric CO₂ during the frequent storm events.
However - no increase in carbon export to the deep ocean inside the patch was observed in the 3 weeks of observation. The bloom was in full swing at the time of departure.

While the EisenEx bloom increased chlorophyll 4-fold, the SOIREE project of the previous year had produced a 6-fold increase. The EisenEx bloom developed more slowly, and to lower total biomass, than did the SOIREE bloom. This was probably due to a difference in physical conditions, with the EisenEx patch experiencing more stormy weather, leading to greater spreading and dilution of the iron fertiliser.
This dilution effect is an result of using a small-scale experimental patches. Experiments at 10 times the horizontal scales might be expected to be far less affected by dilution.

It was also discovered that over the duration of the EISENEX experiment, seawater pH increased by over 0.02pH. An increase in pH can enhance the ocean's capability to absorb CO₂ from the atmosphere. However, modifications of seawater pH has both biological and geochemical implications, and it has been shown that changes in pH can adversely affect the physiology of marine plankton [[as discussed in ocean acidity - link if i do this ] ]

The Alfred Wegener Institute plan a follow-up project to the EisenEx project.
This is EIFEX - for European Iron Fertilsation EXperiment, and will be carried out from January 20th to March 26 2004.

return to top

Conclusion

This experiment once again demonstrated that phytoplankton can react strongly to iron enrichment. However the available observation period was too short to determine the extent to which biologically fixed CO₂ was ultimately exported to the deep ocean via sedimentation of organic material.

return to top

SEEDS (G)

Subarctic Pacific Iron Experiment for Ecosystem Dynamics Study

Objective

This was the first iron-enrichment experiment in the northern hemisphere, and was carried out in the Western Subarctic Gyre during the summer of 2001.
The Subarctic Pacific has both biology and 'water structure' markedly different to those of the Equatorial Pacific and Southern Ocean, the sites of previous iron fertilisation experiments.
These different biogeochemical provinces are likely to react in different ways, and result in different phytoplankton response and communities.

A number [[??] ]of experiments were performed, by [[some] ] international teams.

Experiment

The experiment consisted of a single addition of 350kg iron in the form of 1740kg of iron sulphate (FeSO₄.H₂O), along with the inert tracer gas sulfur hexafluoride (SF₆). These were added in a patch of water (8 x 10 km) located in a western subarctic gyre which had a 'mixed layer depth' of 10m.

Iron

350kg iron in the form of 1740kg iron sulphate (FeSO₄.H₂) was added to the ocean.

Results

GeoChemical changes

Prior to the release of the iron, the surface seawater concentrations were extremely low, from <0.05pM - 0.02nM. However, after one day, this had risen to a mean of 1.9nM, thus showing an increase of over 100 times the initial concentration. The maximum value recorded was 6.0nM.
Nitrate did not deplete until the 14th day, and it appeared by the end of the experiment that light was the limiting factor by then, rather than the lack of any nutrient.
Massive biogeochemical response were seen, with large drawdowns of CO₂ and nutrients.

Further experiments looked at the dynamics of dissolved trace metals. These showed, for the first time, that mesoscale iron fertilisation also affects the dynamics of dissolved trace metals such as zinc, copper and cobalt.

Biological and ecological changes

It was seen that the microorganism concentrations changed, resulting in a change in the dominant phytoplankton species. The floristic shift resulted in the dominance of chain-forming large centric diatoms, unlike the equatorial Pacific and the Southern Ocean where iron stimulated the growth of pennate diatoms.

Abundance of salmon and small squid collected using trawl did not change between inside and outside of the iron patch. However, northern mackerel were abundant in the iron patch.

Conclusions

These biological responses showed that the western subarctic Pacific might be the most sensitive to iron enrichment in the world HNLC regions, a factor supported by the summary below which indicates a 40-fold increase in chlorophyll.

return to top

SOFeX (Southern Ocean Iron Experiment) (H)

Objectives

The principle objective of the SoFex program was to compare the results of iron fertilisation of the Antarctic waters on either side of the Antarctic Polar Front Zone (APFZ).
Two iron-enrichments were perfomed during the expedition, one north and one south of the APFZ, at 10° of latitude apart.

Diatoms, which remove most of the carbon dioxide from surface waters, require silicate to make their tests. They may not, therefore, grow north of the front, due to the low levels of silicate there.
Most of the HNLC water that is favoured for the Iron Fertilisation hypothesis is found north of the front. Thus, if the diatoms don't grow north of the front, the Iron Hypothesis may not work.

Experiment

The SOFeX program involved scientists from 17 different institutions

Three ships were involved, each arriving at a different time. In this manner, the iron fertilized patches could be observed for the longest possible time.

R/V Roger Revelle

R/V Melville

The R/V Melville (also SIO) sailed several weeks later to arrive just as the patches were forming.
The Melville's team made detailed measurements of phytoplankton physiology and rate processes. They took samples daily for

• phytoplankton growth rates
• phytoplankton biomass
• soluble and particulate iron, and
• zooplankton biomass

Particle interceptor traps (deployed from the R/V Revelle) were retrieved by the R/V Melville, and used to compare the amount of carbon sinking from inside and outside of the patches.

Polar Star

Finally, the ice breaker Polar Star (US Coast Guard) arrived to assess how much carbon was removed from the iron fertilized patches. The Polar Star team made carbon export estimates using the naturally occurring isotope Thorium-234 that is found throughout the ocean.
These observations were an essential test of the Iron Hypothesis, and helped form a much clearer understanding of the role that iron and phytoplankton may play in regulating the Earth's climate.

Iron

Results

return to top

Planktos (J)

Objectives

Experiment

Iron

Results

SERIES (K)

(Subarctic Ecosystem Response to Iron Enrichment Study)

Objectives

To follow on from the SOIREE experiments around Antarctica, the SERIES project was led by NIWA in collaboration with Canadian scientists in the Gulf of Alaska.

It would therefore provide a comparison between the Southern and Northern HNLC Polar waters.
SERIES also attempted to look at the dynamics of 'climate relevant gases' such as CO₂, halogens, and DMS produced by the plankton bloom.

Experiment

A 64 km² patch of HNLC ocean was labelled with SF₆ tracer and enriched with dissolved iron.

Measurements included CTD, light and fluorescence profiles, protists identification and enumeration, size fractionated (20, 5 and 0.2 m) chlorophyll, primary productivity and particulate DMSP, bacterial diversity (FISH), macronutrients, iron, ligands, POC, PON, DOC, thorium, dissolved DMSP, DMS, halogens, bacterial productivity, size fractionated DMS production and consumption, and macrozooplankton abundance.

Iron

Results

The phytoplankton bloom grew to be large enough (700 km²) to be visible from space and yielded exciting results for both marine and atmospheric scientists. It was possible to follow the production of DMS in the ocean during the plankton bloom, and also its subsequent transfer to the lower atmosphere.

The expected diatom bloom started around day 10-11, and proceeded until day ca. 18.
During the last phase of the experiment, the Kaiyo-Maru captured the beginning of the sinking of the diatom bloom. The phytoplankton succession was accompanied by important changes in water and atmospheric properties

Conclusions

return to top

Future Plans

The Alfred Wegener Institute plan a follow-up project to the EisenEx project.
This is EIFEX - for European Iron Fertilsation EXperiment.
It will be carried out from January 20th to March 26 2004, in approximately the same area as EisenEx.

return to top

Summary

A summary of the above experiments (excluding Planktos) can be found in the US JGOFS (Joint Global Ocean Flux Study) newsletter, October 2002, vol 12, 1.
http://usjgofs.whoi.edu/general_info/newsletter.html -->
http://usjgofs.whoi.edu/general_info/vol121.pdf

The following 'summary of open-ocean iron enrichment experiments to date' is taken from that publication.

return to top