Science exposed Exposing fallacies in ecological marine research in northern New Zealand By Dr J Floor Anthoni (2003) www.seafriends.org.nz/issues/cons/science.htm Science is all about debate and discussion of ideas. This section is highly critical of some marine ecological research carried out in northern New Zealand waters, particularly the urchin barrens hypothesis which has been wielded as a weapon in the marine reserves debate. It is even taught at schools! Although this section intends to rebutt some scientific findings, it is also valuable as study material to further one's understanding of the ecology of the rocky shore in northern New Zealand and what has gone wrong in marine ecological studies. Please note that the principles involved, also apply to other places worldwide.

For comments, suggestions and improvements, please e-mail the author, Dr Floor Anthoni.
For best printed results, read instructions for printing.
-- seafriends home -- conservation index -- Rev 20030731,20031211,20050803,20051219,20060926,20070315,20090521,

The following summary describes some of these concerns

• The sea is unbelievably and unintuitively different from the land. If you don't know this, you will be uninformed. See biodiversity/marine and habitat intro and principles of the intertidal rocky shore. Yet many (marine) scientists are insufficiently aware.
• The sea is hostile and difficult of access. Waves and weather are always in control. We can be underwater only for one to a few hours each day. Compare this with tramping the forests, and it is clear that we have very little opportunity to become marine naturalists, but in the warm clear tropical seas this gets better. Remember how much knowledge was gathered by ecologists who were also keen naturalists?
• As a result, the marine ecology is largely unknown. The plankton ecosystems are also largely unknown, leaving wide gaps for new discovery as our revolutionary Dark Decay Assay method shows (2005).
• With their limited general knowledge, scientists do 'controlled' ecological experiments from which they derive far-reaching conclusions, which often leads to nonsense when viewed from a wider perspective.
• It is quite worrisome that the scientific null-hypothesis, in other words proving that nothing else could have caused the effects, is no longer taken serious. It is equally worrisome that experiments are not replicated before publication.
• Marine scientists do not have their own boats and are not free to dive any place any time. They must account for every hour to some budget. They have hardly enough time in the water to finish their own experiments. As a result, very few experienced marine naturalists are found among marine scientists.
• Marine naturalists must be confident in diving, which takes the best of ten years experience. They must learn to see, remember and understand the functions of myriad creatures. After about 20 years one can begin to see how it all works together and how it degrades. To recognise degradation requires even more skills because what is no longer there, cannot be seen.
• Marine research done in the lab is of high quality. But studying the ecology cannot be done there. Controlled experiments are seldom possible. The scientific method as defined by Francis Bacon fails there. Read Science, technology & human nature. A different approach is needed but not done (yet). FAQ.
• Scientists cannot react to sudden events, even when these change the course of their studies. First a budget is needed, and someone must be found to pay for it. Time must be accounted for. As a result, many significant events have not been studied and neither have these been mentioned, even when it affected published results. Degradation has not been observed, measured and published even though it has been around for nearly a century. Degradation of coral reefs is often ignored. Degradation in the areas of study is not reported.
• Scientific study has become a piecemeal approach where the duration of the study is typical of obtaining a doctoral thesis. Continuity is largely absent. Present-day science consists of collecting disjointed facts that cannot be repeated and confirmed elsewhere. Experiments do not result in falsifiable theories. There is no insight.
• Studying degradation is not 'sexy'. It does not earn scientific admiration.
• We have omitted monitoring the sea for its clarity and sedimentation rate. A network of simple sedimentation traps on all wharves and some buoys could have kept us uptodate with the disastrous trends that have been with us for over fifty years, but which are accelerating steeply now. This has not been done. It is not even being considered.
• Many marine scientists are politically motivated, having traded objectivity for their beliefs. Science funded by conservation organisations is bound to give results pleasing them, like that funded by the tobacco industry pleasing its funders. One does not easily bite the hand that feeds. The DoC (Department of Conservation, NZ) funded research cannot be critical of DoC.
• Many scientists now earn their living from marine reserves. It has become a major source of income for many institutions. Only studies related to global warming exceed this.
• A swarm of politically motivated marine 'scientists' is now involved with marine reserves (and now also global warming), quoted as a 'growth industry'. They want marine reserves to work for the environment, science, fisheries, biodiversity and much more but they fail to see when and why these reserves don't work. Many work with computers and models to make their point.
• These 'scientists' do not take the precaution of distinguishing apples and pears, and uncritically apply findings about tropical reefs to the situation in temperate waters. Even there they do not distinguish the special nature of some spots on the coast. The area around Goat Island for instance, has always been a special place without equivalent. To compare other places with it must be done with care. Yet these 'scientists' do not exercise such care. FAQ.
• Nearly all travelling protagonists for marine reserves are not frequent divers, having no personal appreciation. They are not marine naturalists. Yet their influence has swayed many and has been decisive for this Government's flawed policies (NZ is no exception). Consider for instance the long list of marine research leaders at the University of Auckland: Chapman, Morton, Miller, Ballantine, Foster, Bellamy, Taylor, Bergquist, Creese, Rees, Costello, Bowman, Montgomery, none of whom are divers. Is this also the case at other universities?

The main issue we contend is the finding that the sea urchin habitat zone is unnatural and that it disappears in marine reserves which have had sufficient time to acquire stocks of large predators, particularly predatory fish (like snapper) and crayfish (rocklobster). We contend that this is not true and that scientists have taken insufficient precaution in their experiments. They have also taken insufficient precaution in attaching far-reaching conclusions to failed experiments. As counter arguments, we dissect the very research done by these scientists, supplemented by work done by ourselves and by extensive observations. Since first publication (2003), a number of overseas scientific publications have supported us.

• top-down control: the predators keep the grazers in check, preventing them to outgrow their food supply (plants).
• trophic cascades: from the top-down control model, it follows that if people kill the predators, there will be more grazers, and thus less grass. This is essentially where the snapper-urchin-kelp myth comes from. It is a bit akin to saying: "when we shoot the bears, the forest dies."
• bottom-up control: the amount of grass determines the number of grazers which determine the number of predators.
• birth control: there are other ways in which populations control their numbers, and predation (exploitation) is not the only and most decisive one. In the end, populations adapt to their particular circumstances by the process of evolution, particularly on islands where habitat size is limiting.
• economics: every population is exploited by others, and the rate of exploitation depends on how easy it is to find next meal.
• extreme events: random extreme events like storms, floods, fires, droughts, diseases, oil spills have a major and controlling effect on populations.
• gradual events: pollution, eutrophication, changing ocean currents, upwellings, changing climate. Often an unreported gradual event can explain the effects observed. Eutrophication is the most insidious because it is wide-spread, cannot be measured and has engulfed most marine research stations worldwide, while being ignored by many.

The sea in its weirdness adds another dimension to this problem. On land habitat zones (desert, savannah, forest) are so large that those who live from it are also living on it. In addition, the offspring of all terrestrial species are born near their parents. And the number of offspring is limited to a few. Although this is never mentioned in books about ecology, these points are of major importance to understand the sea, where it is often not the case as the habitat forming organisms spawn profusely (millions of offspring) and their offspring is born somewhere else. It implies that the genes of survivors are not expressed where the survivors live. Evolution and adaptation in the sea are thus necessarily weak. In the sea the narrow habitat zones of the rocky shore are frequented by predators who do not live there, like snapper. The very narrow intertidal zone is even visited by grazers who do not live there.

The rocky shore where the urchins live, is but a narrow fringe of the wide open sea with its vast expanse of sea bottom and water above it. The sea bream snapper (Sparus auratus) lives mainly over the sandy and muddy sea bottom, and only very few take residence on the rocky reefs. It has been shown that most fish specialise in a certain diet during their lives, and snapper being quite versatile, can be found as predator, scavenger, crab-and-worm eater and mollusc eater, yet seldom all of the above. The point I wish to make here is that there are high numbers of snapper that could arrive from the open sea to wreak havoc on coastal populations, but very few actually do. And those who could do harm, are not constrained by a marine reserve. The situation is different for rocklobsters who live on the rocky shore, but they scavenge more than predate.

The trophic cascades model (or paradigm as the authors called it) became popular after a publication (1974) by Estes & Duggins [1,2,3 and many more] who saw a correlation between sea otters (who eat urchins) and the giant kelp (which can be eaten by sea urchins). Thus catching sea otters for the fur trade can have a decisive influence on the algal assemblages it was thought. This uncritically became established knowledge and some of the most cited articles in marine ecology. But although a correlation was seen, a causation was never established (see below)! So much for trusting marine science, its practitioners, its peer reviewers and its prestigious journals!

Finally in April 2006 (22 years later!) the trophic cascades paradigm was tested by Carter, VanBlaricom and Allen [4], showing that there is no significant increase in algal growth when sea urchins are removed manually inside the San Juan Marine Reserve, not even over a longer time period of 2 years.  One would have thought that this experiment would settle the matter, but please note that their study happened in a sheltered place (San Juan Channel, Washington State) where storm barren zones cannot occur, while urchin densities were very low (1 per square metre). This means that the available sea urchins had no influence on their environment, and removing them of course had no influence either. The study is thus too weak to disprove the trophic cascades paradigm.
 
 

The Monterey Bay National Marine Sanctuary has been closely monitored for many years. The graph shows how sea otter numbers (red curve) increased from near-zero in 1900 to over 2000 in 2000, proof that they had been hunted to near extinction. However, recently their numbers have been declining but why is not known. The kelp harvested here is mainly the giant kelp Macrocystis pyrifera (and Nereocystis luetkeana) which prefers sheltered waters, but even so is erratic in its year to year densities. They are also easily removed by waves. The blue curve shows total kelp harvest in all of California, climbing as otter numbers increase, suggesting that there is a linkage between the two. But there are inconvenient exceptions. Remember that the influences of market demand and price, regulation, nutrient supplies and degradation can be decisive. The green curve shows kelp harvested inside the sanctuary, declining as otter numbers are increasing. Kelp is harvested by cutting the floating fronds off their stipes. The seaweeds are used to cultivate abalone, and they are exported to Asian countries.  Monterey Bay was famous for its sardine fisheries wich crashed in 1940-1960, presumably due to overfishing and pollution.

In the graph also the commercial harvests of red and pink abalone are shown, both declining steeply around 1970. The green and black abalone are much less common but also show steep decline (not shown here). Data for urchin catches are not available. The curves suggest that the Californian coast and Monterey Bay suffer from severe degradation affecting all species.
 

tethering urchins to confirm predation Predation on urchins is usually studied separately by tethering urchins to the substrate. In the laboratory a string is threaded through the urchin, which causes it to bleed, as well as internal damage. But the urchin is first kept to recover, before transplanting. The experiment then shows that urchins inside protected areas disappear faster than outside, proof that predators remove them. But is this true?

We found that rocklobsters can quite happily walk over an entire cluster of sea urchins without attacking any. However, when we damage some by scratching some spines, only the damaged urchins disappear within days. Rocklobsters are apparently strongly attracted to bleeding and dying organisms, and nobody can guarantee that a tethered urchin does not bleed as it attempts to untether itself. By tethering an urchin, it is not placed where it wants to be and this causes stress.  See the tethering experiment below. The bottom line is that tethering experiments confirm that there are more predators inside than outside protected areas, but not necessarily that these predators (scavengers, really) usually eat sea urchins. Neither does it prove that a decline in urchins is caused by predators.

We discovered that urchin barren zones are created by large storms and that urchins and other grazers just maintain these. Urchin barrens are thus a normal and healthy habitat, not in any way caused by humans. We propose to rename them storm barrens, for this is what they are. Such barrens are defined and restricted by wave propagation laws, which explain where they are found, where they cannot occur, and how deep they extend. One of the deepest storm barrens is found on the northern side of Niue Island where due to extreme water clarity (40-80m viz), the photic zone extends far beyond diving depth. Poorly occupied by day but grazed by numerous fish species,  the night brings out a true city of sea urchins in perhaps two dozen different species. Read our account of Storm barrens in New Zealand, and underwater images from Niue, its ecology, index1 and index2 (for large picures, our CD is required).

[1] Estes, J A & J F Palmisano (1974): Sea otters: their role in structuring near shore communities. Science 185: 10858-1060.
[2] Estes JA, N S Smith & J F Palmisano (1978): Sea otter predation and community organization in the western Aleutian Islands, Alaska. Ecology 59: 822-833.
[3] Estes, J A & D O Duggins (1995): Sea otters and kep forest in Alaska: generality and variation in a community ecological paradigm. Ecol. Monog. 65: 75-100.
[4] Carter, Sarah K & Glenn R VanBlaricom, Brian L Allen (2006): Testing the generality of the trophic cascade paradigm for sea otters: a case study with kelp forests in northern Washington, USA. Hydrobiologia (2007) 579:233-249 (freely available on the Web)

• In reef habitats across the globe - when uncontrolled - sea urchins have devastated reefs, leaving sea-urchin barrens devoid of algae. Sea urchin barrens are increasing along the east Tasmanian coasts reducing the productivity of the reefs by 80-90%.
• Dense aggregations of urchins (or feeding fronts) form along the margins of kelp beds and destructively graze all upright seaweeds, creating urchin barrens of much lower productivity, habitat complexity, and biodiversity.
• Sea urchin overgrazing is associated with the transition of macroalgal-dominated reef habitats to barrens, and such a change results in a 100-fold decrease in primary production (citing Chapman, 1981 [1]).
• Prof. Johnson said the sea urchin had created ‘barren’ areas along the East Coast by overgrazing seaweed and invertebrates which had sustained the abalone and rock lobster fisheries. “It really is the equivalent of taking a bulldozer to a rainforest and clearing everything to bare earth. It’s a major crash in biodiversity and productivity, with a big impact on important commercial fisheries,”
• An urchin barren is an area of the subtidal where the population growth of sea urchins has gone unchecked, causing destructive grazing of kelp beds or kelp forests. (Wikipedia)
• in the early 1970's, grazing by aggregations of sea urchins, Strongylocentrotus droebachiensis, completely destroyed subtidal kelp beds in St. Margaret's Bay, Nova Scotia, leaving barrens devoid of macroalgae.
• Destructive grazing of kelps by advancing urchin fronts results in the formation of barrens, with dramatic losses of productivity and biodiversity
• etc.

• the giant kelp Macrocystis of 0.2 - 2.6 gC/m2/day: (Robert L. Vadas Sr et al. 2004), or 75-950gC/m2/year
• Macrocystis 2kg/m2/year wet weight (), identical to tropical rain forest?
• tropical rain forest quoted generally at 9-11 kg C/m2/year. (Anderson 1992)
• mediterranean 20-50m depth: 0.3-0.7 gC/m2/day (Drew 1968)
• mediterranean mid-littoral up to 10gC/m2/day (Ballesteros 1992)

• the barrens are more shallow and receive more light.
• because they are covered in crustose coralline algae with a light reflective pink colour, plants growing there receive the light twice.
• single celled and filamentous algae grow faster than large kelps because they do not form woody tissue (cellulose cell walls) and can grow as fast as diatoms do.
• green cells are more efficient than brown seaweeds (Phaeophyta).
• filamentous algae are more palatable (edible) than brown seaweeds.
• ALL the filamentous algae are eaten, whereas LITTLE of the brown macro algae.

[1] Chapman, A.R.O. 1981. Stability of sea urchin dominated barren grounds following destructive grazing of kelp in St. Margaret's Bay, eastern Canada. Mar. Biol 62:307-31 1.

Scientists were confused finding kelp death both in the shallows and deeper than 15m but not in between. They could not conceive that the one had been caused by wave damage whereas the other was caused by lack of light. As a result they went about their investigations in a less than thorough way. They did not recognise the importance of such a major experiment conducted by nature and the possible lessons that could be learnt from it, and its possible ramifications. This spurred us to do an extensive survey to establish the precise areas affected and what other lessons could be learnt. It taught us much about the ecology of the sea coast, particularly that of kelp and urchins. You can find this report in www.seafriends.org.nz/enviro/habitat/survey93.htm.

Our main findings relating to urchin barrens:

• urchin barren zones are found only on exposed coasts
• urchin barren zones do not occur in shallow water
• barrens do not need to have urchins
• barrens are first caused by large storms, then maintained by grazers (urchin, Cooks turban snail, paua)

Our studies of the marine ecology in Niue confirms this hypothesis as hurricanes create very deep barrens only on one side of the island (2004, 2005).
• there are more grazers of young kelp than thought, including planktivorous fish. Also confirmed in Niue.
• the disappearance of the urchin barrens at Leigh is caused by the mass dieoff of the kelp forest.

Because the plankton blooms between October and December 1992 were exceptionally dense, a few exploratory dives were made to observe plant life in the dark. At the peak of plankton blooms, visibility was less than 2m at the surface and the light at 20m depth was so poor that the sandy bottom could hardly be distinguished from the dark rock. It was darker than a moon-lit night. In November the kelp turned dark brown and started to look like wet kraft paper on woody stalks. By mid December, their crowns fell apart when touched but in shallower water the kelp appeared entirely unaffected. It was clear that a major event had happened, and we informed the Marine Laboratory accordingly. The photo shows what the dead kelp forest looked like. In the first week of January 1993 tropical cyclone Oli struck, and when diving became possible again, we saw that the whole kelp forest below 15m had disappeared. In the months following, the kelp died back further to effectively disappear where once the forest stood. Some shallow water dieback was also evident but this was caused by the waves of cyclone Oli. The second photo shows what an extensive sea urchin barren looked like (June 1995, Waterfall Reef), with a cluster of sea urchins in the foreground and a leatherjacket (Parika scaber) above. Note that the kelp had not yet invaded the barrens.

In March 1993 we did an exploratory study in order to prepare a more extensive one, which we did in the winter of 1993 (Aug-Sep). The results of this study you can find on this web site in /enviro/habitat/survey93.htm.
 
 
 

The above two maps are important to understand why the urchin barrens theory is a myth. The first shows where the 91/92 kelp deaths occurred, but more importantly, the second shows where the 92/93 kelpbed death occurred and to what extent. The areas marked 4 and 5 lost 80-100% kelp cover, and all these areas recovered the way of Goat Island, with loss of urchin habitats. Yet these are not marine reserves. Here is what happened to their kelp and urchin habitats:
 

This diagram depicts our observations in the years preceding and  following the kelpbed death events. On the left the situation where the kelp bed was removed twice, first partially in 1992, then fully in 1993 (Little Barrier, Simpsons Rock, Arid Island). On right the situation where the kelp was removed once, but completely. In 1991 both areas had a mixed age kelp forest. In 1992 part of the kelp forest disappeared in some areas, followed by rapid recruitment and full regrowth because grazers had plenty of food in the remaining stands of kelp. The urchins stayed in their barrens. By late 1992 the kelp forest had almost complete coverage.  In Dec 1992/ Jan 1993 the kelp died, again because of dense plankton blooms. Recruitment was immediate and intensive in the areas where the kelp once stood but not outside these. However, all regrowth was eaten by the remaining grazers, which included fish. The urchins strayed from their barrens into the deep, leaving their barrens insufficiently grazed. Successive waves of recruitment but much less intensive followed. Kelp settled slowly into the urchin barrens and the areas where the kelp once stood. This time it was successful and it took over the urchin barrens. The full takeover took several years until the kelp reached maximal growth rates. Isolated reefs like Floors Reef and Leigh Reef took until 1998 before they were recovered fully. These have no urchin barrens, yet grazing there was very intensive. We also noticed massive urchin deaths and poor recruitment, which affected the process.

From the scientific research dissected on this page, it can be seen that scientists were not aware of the above, even though they were in possession of our Hauraki Gulf Marine Survey 1993 report.
 

The changes in community structure by top-down control (the urchin barrens theory) was a reverberation of the 1993 kelpbed death. It was caused not by the beneficial effect of no-take protection but by degradation! The urchin barrens theory is a myth.

The map above is a composite of three maps drawn by Land Information (LINZ), which was the Department of Lands and Survey. The data for this map was obtained by a team of divers under leadership of Dr Tony Ayling in 1977. Click on the map for a page-sized one. The depth of the rocky shore ranges from 15m in the west to 23m in the east with a sandy bay about one quarter from west ending in a shelly beach.
The habitat zoning can clearly be seen. The shallow dark brown zone consists of tough bladder weeds (Carpophyllum spp). The grey/blue zone corresponds to the urchin barren zone. Notice the large area of submerged rocky shore around Goat Island, which is unique. Because of its gentle slope, most of the barren zone is also found here. Also unique are Goat Island's areas of shelter preferred by fish, its extensive crayfish habitat and shellfish beds, caves and broken rock (large boulders). To the west of Goat Island the substrate consists of layered mudstone; to the east of (broken) hard greywacke. Note how the shores to the west and east become more representative of the coast in this region, having a narrow rocky substrate. Further west extends the 25km long Pakiri Beach, which makes Goat Island the first rocky outcrop with shelter and currents, like an oasis at the end of a long stretch of desert. There is no place like Goat Island in a vast region around.

Scientists also do not take sufficient care in comparing apples with apples on the choice of their transects and areas of study. They believe too much in the impartiality of dice. Yet transects and areas of study must be chosen carefully to match exposure, aspect (orientation), depth and isolation (sand, crevices, walls). Otherwise too much spurious data will spoil the experiment.

Scientists view snapper (Pagrus auratus) and crayfish (Jasus edwardsii) too much as predators rather than scavengers. These animals are strongly attracted by stressed organisms such as damaged sea urchins (Evechinus chloroticus), shellfish and other. Likewise the common urchin (Evechinus chloroticus) is strongly attracted to stressed kelp.

Scientists make no mention of major oceanographic events happening during their experiments (or just before), events that could have influenced the outcomes of these experiments. Chronology, mentioning the precise start and end of the experimental period is therefore important. In their papers they fail to report the time of study relative to large events like cyclones, kelpbed deaths, large urchin mortalities and the disappearance of crayfish.

Scientists are so keen to show overwhelming benefits arising from marine reserves that they lose objectivity. It has led to nonsensical conclusions as the reader may see for himself. The fact that much research was funded by the Department of Conservation who manages reserves and whose task it is to create more, did not help - one does not bite the hand that feeds. As a result, politics have entered science. Read also our extensive myths and fallacies section.

The baited underwater video camera (BUV) is a highly questionable measuring device since it aims to influence the quantity measured rather than minimising its influence as is required for any other scientific measuring device. Not only does it bait the camera, but it also offers an additional pilchard as food! Such a method grossly overestimates the quantity measured and should never have been attempted as a scientifically acceptable measuring device. But it can be used qualitatively to say that at some place there are more or less fish than at some other place. Scientists have never proved its linearity (as in a voltmeter) for quantitative measurements. [in a linear instrument, there is a one-to-one correspondence between the measured and actual quantities]. For a description of the BUV method, see further on this page and in our Frequently Asked Questions about marine reserves.

Our own research adds a refreshing turn to the issue, filling in the missing bits while providing new insight into the rocky shore ecology. Although it is not our task to do marine research, it is nonetheless the task of the public to expose faults and fallacies where scientific self-policing fails. We are waiting for the scientific community in Leigh to correct its mistakes, by looking at the bloom-affected areas around the outer Hauraki Gulf, in order to arrive at the inescapable conlusions:

In our own study we excluded the areas north of Whangarei after a preliminary survey in March 1993. It allowed us to concentrate more on the areas surrounding the outer Hauraki Gulf. Because of our extensive spatial data, we can say that the bloom and its mortality did not centre on the Goat Island marine reserve. See our map on right. The area around Port Fitzroy on Gt Barrier I is rather patchy, possibly because this harbour has its own plankton ecosystem and may have been spared some of the bloom's ravages. Also unaffected water flows north and out of the Hauraki Gulf past these areas.

It is always surprising how the human factor is reasoned out of the equation so quickly. If something is unique or cyclical, then humans are not at fault, it is reasoned. But what caused the concentrations of nutrients necessary for this plankton bloom build-up? When is a unique event just the first of many to come? When is a deteriorating trend from moderate kelp die-off in 1982 to a large one in 1992, recognised? When are such events admitted to be degradation of the environment?

Fig 1. Kelp density before and after mortality events. Mean density of Ecklonia radiata in the Cape Rodney to Okakari Point Marine Reserve based on five 1m2 quadrats at each depth. Depths are maximum depths for each depth strata (e.g. 3-6m, 6-9m, 9-12m)

These two data points are accidentally the only ones showing the situation before and after. The top two curves of the before situation, show that the kelp forest is found deeper than 9-12m. After the plankton blooms the deep kelp disappeared, leaving the shallow kelp unaffected. Note how kelp diminishes with depth in the bottom diagrams.

Local mortality patterns
The proportion of kelp populations damaged or intact varied considerably among sites within the marine reserve (Fig 4), though all sites had low densities of intact plants. The proportion of intact plants was relatively high (over 50%) at Marimo and Point, but was much lower elsewhere. For example Waterfall and 'D' the mortality process appeared to be still active. The proportion of the populations contributed by recruits was also variable with North Reef and Lookout having the highest levels. Despite the overall variability, highest levels of damage were consistently seen in the deepest levels at each site.
 
 

Fig 4. Local variation in relative kelp mortality, May 1993. Data presented are proportions in each damage category based on video transects and they represent all depth categories.

The authors made a major error of judgement by including recruitment, which happened after the kelpbed death, in their results, rather than dealing with this separately. The pie diagrams should also have included barren areas without kelp. In this map a shallow shore like Marimo seems to have sustained little kelpbed damage whereas in reality it does not go deep enough to have a substantial kelp bed. The map cannot be interpreted without much local knowledge. Had the authors displayed the amount of kelp canopy left as part of the natural kelpbed zone, it would have been very clear that the kelpbed death is very consistent along the shore. The method also becomes useless in areas or zones which have been hit by a kelpbed death in two successive years, see below.

Regional mortality patterns A survey of sites between Cape Karikari in the north and Kawau Island in the south, and offshore to western Great Barrier Island, revealed that in this region kelp mortality was restricted to the outer Hauraki Gulf area. Sites at Whangarei Heads, Okakari Pt to Cape Rodney Marine Reserve, Little Barrier Island and Great Barrier Island were all affected by recent large-scale mortality of Ecklonia. However, other sites at Gt Barrier I were unaffected, as were sites at Takatu Peninsula and Okakari Pt.

This observation agrees with our findings. What confused the authors is that their method could not distinguish between sites that had been devastated twice, such as some of the Gt Barrier sites. The authors have not taken due caution excluding sites with shallow sand bottoms such as at Okakari Point and some at Takatu. The authors did not make a distinction between shallow and deep kelp death.

In our surveys we identified the seemingly contradictory site at Gt Barrier I as a marginal case where the kelp jaundiced but recovered. It was then destroyed in shallow water by the force of Cyclone Oli. The author's method is not suitable for sites with complete loss of kelp and no recruitment, yet many such places could be found. Although the Gt Barrier I site seems inconsistent, it is consistent with nearby sites. See also notes below.

Recruitment
Very few recruits of Ecklonia were detected in surveys of the marine reserve during January. During March numerous recruits were noticed in parts of the reserve and by May high levels of recruitment were evident at most sites. . .

The recruits should have been dealt with separately since they did not exist before the plankton blooms.

Possible causes of mortality
Several scenarios have been proposed in attempts to explain the mass mortality of Ecklonia radiata in the region under consideration. There are three basic hypotheses:

Pathogens
. . The patchy nature of the mortality at other locations such as Gt Barrier I is consistent with this hypothesis, rather than a broad-scale physical phenomenon. . . [viral tests] . . The possibility of viral involvement is still being actively pursued but the presence of agar in algal tissues has hampered progress since it interferes with techniques used as assays for viral activity. . .

Every marine organism with reduced health runs the risk of attack by decomposing organisms like viruses and bacteria. To find these pathogens on dead and dying kelp is akin to finding fungi on rotting apples. What is pathetic is that the authors have been warned before December 1992 that the kelp forest stood dying and their fronds drooping like wet kraft paper. It is notable that after Cyclone Oli not a trace was found of the masses of dead canopies. They seemed to have vaporised. They did not have any food value left in them. Pathogens cannot achieve a kelpbed disappearance within several hours or even days.

Environmental factors
. . . The first general hypothesis is that mortality is simply and directly related to low temperatures, an idea consistent with the gradual onset of the mortality during 91-93, but at odds with the observation that plants in shallow water were unaffected, and also with the lack of mortality at points further north.

Ecklonia is found all around the North and South Islands, tolerating temperatures between 6 and 23 degrees, and mortality was abrupt, rather than having a gradual onset.

The deep kelp death was without any doubt caused by lack of light. Not only have we seen it happen, we have also predicted that it would happen! What the authors don't realise is that the shallow kelp death was caused by wave action. The authors are rather naive about the nature of life and survival after being exposed to a gradient of a lethal agent. The Hiroshima and Nagasaki atom bombs caused immediate deaths but also some forty years later. Even on the way to recovery, an organism remains susceptible to other attacks.

Although the above reasoning is valid, the authors and many before them have never looked at the age composition of a natural kelp forest. There is talk about the number of plants per square metre or the amount of cover but the age composition is nevere mentioned. Thus the transition from a multiple-aged forest in 1991 to a single aged forest in 1994 has been left unnoticed.

Conclusions
Kelp mortality is limited in extent and is seemingly most severe in the outer Hauraki Gulf region. Within the affected area mortality is patchy, but is most common below 10m. Populations have been recruiting rapidly and should recover over the next one to two years unless the cause of mortality (which has not yet been identified) persists.
The effects of the kelp mortality on other organisms on the reef are not readily predictable, but appear to be of considerable potential importance to energy flows, growth rates and behaviour of grazers, and the population characteristics of fishes. Insufficient long-term information exists to judge the magnitude of the effects which may derive from this event, or the significance of the mortality to the reef environment.

The authors give themselves advanced warning that after a major event, major changes could follow. Yet they have not taken heed as some of the following papers show how they interpreted the disappearance of urchin barrens as a beneficial effect of aged marine reserves (due to the presence of large predators), rather than being related to the 1993 kelpbed death (degradation).

1. has the abundance of predators in protected areas increased? YES We agree.
2. has the abundance of invertebrate grazers decreased? YES We agree.
3. are changes in grazer abundance caused by predator abundance? YES We do NOT agree.
4. has the abundance of macroalgae increased? YES But not for the stated reasons and not gradually.
5. are changes in the abundance of macroalgae caused by changes in grazer abundance? YES To a minor degree.

• It conveniently ignores other marine reserves outside the kelpbed death region, taking the only two marine reserves where barren zones disappeared.
• It ignores a previous study by the same author(s) which gave evidence that there is no difference between inside and outside in marine reserves.
• It does not prove the disappearance of whole urchin barren zones.
• The method used is that of a sledgehammer, by taking measurements on incomparable coasts, then shaking the results in a statistical mixer and interpreting these results as proof.
• Since the original data have been buried in the statistical mixer, they cannot be verified.
• We cannot verify how much the Goat Island dataset dominates all results.
• The research has been set up to prove a political objective, not to learn more of the undersea ecology.
• This research was conducted after one before it, comparing five marine reserves, failed. This one worked because both reserves suffered substantial kelpbed death in 1993.

 

The 1977 date is misleading because the situation stayed stable until the large kelp dieback. Not a word about the large-scale kelp forest die-off in 1993. The 1994 date is wrong since kelp recruitment failed that year due to hungry grazers of all kind. The conclusion is not at all substantiated by the research done (see below). Large-scale disappearance of urchin barrens was caused by the 1993 disaster. This aerial photo taken in April 1999, shows the main barren area of the Goat Island marine reserve. Notice how the kelp is beginning to invade, but most of the invasion happened suddenly a year later. Scientists maintain that it was a gradual process that began before 1980, which is proved false by this photo.

This location map shows in dashed lines the countours of the Goat Island marine reserve in the north and the Tawharanui marine park in the middle. The numbered areas are the places where measurements were taken. The shaded circles are the places where lobster abundance was measured.

Methods
The study areas were the two oldest marine reserves in New Zealand, the Cape Rodney to Okakari Point Marine Reserve (hereafter the Leigh marine reserve, 518ha, established in 1975) and Tawharanui Marine Park 15 km to the south (350ha, established 1982). Non-reserve reference sites were located on similar areas of coast adjacent to the reserves.

It is misleading to call these the two oldest marine reserves, even though they are. Why are the Poor Knights (1981) not included? Why is this not explained? Our objection is that north facing coasts (exposed) cannot be compared with south-east facing coasts (sheltered). The outcome of the results of this study could be interpreted as 'north facing slopes have lost their barrens' rather than 'marine reserves did'.
Another problem of the choice of study site is the choice of areas 1, 2, 3, 13, 14 and 15. These shores are too shallow to have urchin habitat zones. They are bound to cause unnecessary noise in the data.

Sampling effort was spread evenly across both reserve and non-reserve areas by dividing them into a number of sites and then sampling haphazardly within them. This was done to avoid biases that might be associated with human activities in the Leigh marine reserve, (e.g. fish feeding, c.f. Cole et al 1990) . . . .

Fish abundance
Estimates of predatory fish abundance and size were made during October and November 1997 using remotely-deployed baited video stations. A vertically oriented video camera was mounted over an enclosed bait consisting of four whole pilchards (Sardinops neopilchardus) with one additional pilchard tied externally to the food container. . . . .

The baited video camera method exaggerates. See comments about the BUV in recurring mistakes.

The timing of these surveys is at odds with the time of the experiment (1998), since the lobsters were counted in 1995 when all sites were still having urchin zones. During the time of the main experiment, this was no longer so. Furthermore, between June and October 1998, 85% of all crayfish walked out of the marine reserve due to a prolonged period of heavy rains and sediments (personal observation). Not a word about this in this paper. Note also that the Little Barrier Island survey was not part of the 1998 study area, although the areas surveyed are more similar to the marine reserves they are compared with. The graphs below must therefore be interpreted with care, also because the results of each reserve are not shown separately. Leigh may for instance swamp the data from Tawharanui. Why was it done this way? Our suspicion is that Tawharanui has a very low crayfish density.

The diagrams show the densities of snapper (left), as counted with the baited camera and that of crayfish (right) counted visually.

As mentioned before, the SE facing coast cannot be compared with the north facing reserves, since it is more sheltered and more shallow. The sites near large beaches are similarly incomparable. More caution should have been exercised in choosing the survey sites.

Temporal changes in the density of kelp and urchins in the Leigh Marine Reserve.
Measurements of the abundance and size of kelp and urchins made in prescribed areas of the Leigh reserve in 1977-78 ("permanent quadrats" Ayling 1978) provided a direct basis for assessing temporal change. These data were obtained in a range of habitat types using haphazard 1m2 quadrats within 26 defined 100m2 areas. The same measurements were repeated as close as possible to nine of these areas in 1994 and fourteen of them in 1996, using the detailed maps provided by Ayling (1978). All of these areas were classified by Ayling (1978) as either kelp forest or rock-flats. Samples from 1994 and 1996 were estimated to be within 20m radius of the original 1978 areas, and were in the same depth range as Ayling's areas, as indicated on his habitat maps (Ayling 1978) . . . .

It is a blessing to have permanent quadrats for monitoring how the environment changes with time. However, Ayling's maps were used and these have proved to be less accurate. It would have been more helpful to establish new 'permanent quadrat' sites with accurate GPS co-ordinates. The dates of these surveys are again not concurrent with the 1998 date of the main survey, and particularly 1994 is questonable in relation to the 1993 kelp death. However, the data reflects what can be seen with the untrained eye.

(The analysis is waiting for a copy of Littler & Arnold's and Novacek's publications). It would have been more helpful to have measured the productivity in-situ, since using other people's data from other places in the world, and from so long ago, is highly questionable. It is not clear precisely what productivity has been measured. Is it the production of algal biomass? That certainly does not follow ecological concepts of productivity. It is often better to estimate the biomass of the grazers because kelp makes much indigestible biomass. Kelp has very few grazers, which includes fish. Its fronds are often found rotting back for lack of grazers. Broken fronds are caught and eaten by sea urchins. One cannot compare the 1997 kelp forest, consisting essentially of one year class, with that of 1978 which consisted of many year classes (personal obs).
The most disturbing aspect of this part of the research is that it is at odds with ecological principles. The productivity of micro algae growing on the crustose coralline is much higher than that of macro algae because of their much smaller size and much higher growth rate. Furthermore, urchin barrens experience more light and the light colour of the CCA reflect light back, thus enhancing productivity further. Even measuring the biomass of grazers is fraught with problems since urchins spend most of their energy producing spawn.
Reader please note that CCA (Pink paint) is like living stone, hard and inedible while growing only very slowly. No surprise then that its 'productivity' is very low. However, the barrens' productivity comes from fast growing green hairy algae growing on top of the ink paint. Yet this was not measured. This is akin to claiming that the productivity of grassland equals that of bare land.

Note that other research has been done on the predation of 'tethered' urchins, see further on this page. The authors do not mention mass urchin mortality in the 1994-1997 period even though they studied this as well. But the most worrisome aspect of this research is the absence of a time study of the populations of the major players, snapper, crayfish, kelp and urchins. Had they done so, they would have seen that the 1993 kelp die-off event was the cause of all instability observed later, including urchin deaths.

Our results also suggest how no-take marine reserves can change benchmarks for environmental and fisheries management. This statement is for the benefit of DoC.
The recovery of kelp populations subsequent to protection indicates that fishing activities on New Zealand's northeastern coast have had ecological impacts far beyond the target species. These effects, though indirect, are just as striking as those caused by fishing practices such as trawling, that remove or destroy conspicuous components of fished habitats (Rothschild et al. 1994, Dayton et al. 1995, Watling & Norse 1998). Perhaps more importantly in the case of kelp forests, these conspicuous components are also major primary producers (Mann 1973), whose contribution to detrital food webs is central to ecosystem function and diversity (Duggins 1980). . . . (sigh)

. . . The corollary of this (higher productivity) is that benthic primary productivity in areas outside of reserves is much lower than it was prior to intensive fishing. . . . Kelp forests in northeastern New Zealand are known to have far higher rates of secondary production than do rock-flats habitats (Taylor 1998), supporting the suggestion that these effects are likely to be broadly felt throughout the ecosystem. (awaiting arrival of Taylor's publication)

These observations indicate that the efficient management of coastal fisheries can no longer ignore the consequences of fishing on the wider ecosystem and attempts are now being made to understand the benthic ecosystems in terms of multi-species fisheries models (McClanahan & Sala 1997). No-take marine reserves represent a large scale ecological tool that can play an important part in the development of such models. Unfished areas provide a way of assessing ecosystem function and avoiding the 'sliding baseline' phenomenon, in which expectations of what is natural are reduced because may of the original components of the system are reduced or absent (Dayton et al. 1998). . .

Far-reaching conclusions for what is in essence an experiment that measured the sliding baseline effect of degradation, triggered by the kelpbed death in 1993, and did not provide significant results! (Sigh again)

Top-down regulation of biological communities has been a focal point in ecological theory (Hairston et al 1960). This is ironic, given the efficiency with which humans have themselves harvested the large-bodied animals which may be responsible for the top-down control of ecosystems, in many cases to extinction (MacPhee 1999).

The authors seem to have missed the point that predation by Homo sapiens has not led to extinctions in the continent where humans evolved. However, whenever humans migrated to new continents and islands, they were able to extirpate large animals that had not evolved either fear of or defences to humans. The same happened with the introduction of alien predators like rats, cats and stoats to remote islands.
Top-down regulation is a myth since populations interact on the basis of economy or effort. See our explanation of the economics of exploitation in the Resource Management chapter.

Trophic cascades are defined as predatory interactions involving three or more trophic levels, whereby primary carnivores indirectly increase plant abundance by suppressing herbivores (Menge 1995). . . (more references) . . However, good empirical examples supporting the existence of such trophic effects are generally lacking (Scheibling 1996). . . . (The authors mention the Californian sea otter - urchin - kelp story. Which may well be flawed like the research done here in NZ) . . In some areas where sea otters do not occur, fish and lobsters have been implicated as important predators of urchins. (more doubt) While the destruction of kelp beds by sea urchins in these areas has been linked to overfishing of both lobsters and fish, the existence of a direct causal linkage has received much debate (Scheibling 1996).
. . . . . (more references) . . However there has not yet been any decline in the extent of urchin barrens in these protected areas (Mediterranean Sea). Other factors such as recruitment, pollution, disease, large-scale oceanographic events, urchin harvesting, food subsidies and availability of shelters may also be important in controlling algal assemblage structures (Sala et al 1998).

This paper deliberatley fails to discuss the most important effects on urchins and kelp, those of recruitment failure, massive kills, pollution and storms, all of which have had a major influence during and before the experiments. Note how other studies could not replicate the decline of urchin barrens in protected areas.

1. An urchin-tethering experiment to test whether relative predation levels on urchins were higher at marine reserve sites and to determine the sources of predation.
2. Comparing the density and demography of urchins in the urchin barrens habitat at multiple sites in two reserves and two non-reserve areas.
3. Experimental removal of urchins to test whether the observed habitat changes in the Leigh reserve were consistent with a reduction in urchin populations.
4. Comparing the distribution of macroalgal communities among reef habitats between reserve and non-reserve areas.

 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

It is deplorable that this study was motivated politically, to show that marine reserves work, rather than searching for more understanding of urchin and kelp populations. The authors chose the only two marine reserves in New Zealand (out of 15 others) where urchin habitat disappeared!! Is this by chance?

Materials and methods Study area: see map.  Predation . . . The tethering technique involved inserting a hypodermic needle (1.2x38mm) through the dorsal and ventral surface of the urchin test, as far away from the oral-aboral axis as possible. Nylon monofilament was then threaded through the needle and tied off. laboratory trials found 100% survival of 80 tethred urchins [ranging from 25 to 75mm test diameter (TD)] after 10 days. Trials in the field found that tether-related mortality could be reduced by holding the urchins in the laboratory for a week prior to experimentation. This procedure also allowed the urchins to heal, minimising the potential effects of coelomic fluid leakage on predation (McClanahan & Muthiga, 1989) . . .  Daily monitoring enabled detection and replacement of urchins that appeared to be dying as a result of tethering. In each experiment only four out of a total of 180 tethered urchins died as a result of tethering. . . .

Fig 1. Location of study sites in the Cape Rodney to Okakari Point Marine Reserve (CROP) and Tawharanui Marine Park. Circles indicate sites where the predation experiment was carried out. Inset shows general location of study area in New Zealand's North island.

This is a well thought-out experiment with adequate precautions in the laboratory. When urchins bleed (coelomic fluid leakage) they attract scavengers of which there are many, and it cannot be guaranteed that the urchins had indeed healed completely. To do an experiment where this does not play a role, requires careful execution. However, the two coasts selected are incomparable. Both reserves are facing north, being much more exposed than the SE facing coast. Three locations to the west of both reserves are located in shallow water where urchin barren zones normally don't occur. Results could be interpreted as "north-facing coasts favour survival of tethered urchins". coscinasterias being the main predator in the non-reserve group, bears this out. Worse still, was the north facing slope of Kawau (SE of Tawharanui)  ignored as control site because it produced inconsistent data in other experiments?

Predation on urchins was significantly higher at reserve sites than at non-reserve sites (F=9.44, P=0.0133), with the relative odds of predation being 6.9 times higher at reserve sites (Fig.2, Table 1). The fate of all small urchins preyed upon at reserve sites was unknown as the tests were completely removed from their tethers (Table 2). This could have been due either to predation by fish, which completely engulf the urchin, or by lobsters breaking up or removing small urchins. (or by wave action?) At reserve sites approximately 45% of preyed individuals in the larger size classes showed patterns of damage characteristic of spiny lobster predation. No urchins showed signs of spiny lobster predation at non-reserve sites. In most cases mortality at non-reserve sites could be attributed to starfish (Coscinasterias muricata) or the gastropod Charonia lampax, both of which are slow-moving predators. (but which remain on-site, having an acute sense of smell for stressed animals) Reader please note that a sleight of hand is presented here, where the researchers talk about odds of predation, which is the inverse of survival. The curves on right show survival rates, with non-reserve sites doing between 30-90% of reserve sites. Leigh approx 30,50,70% and Tawharanui 40, 80, 80%. Note also that mortality is critical in the first few days, after which it becomes negligible. The authors do not attempt to explain this effect.     Table 1 Size class Number preyed Unknown% Lobster% Coscinasterias% Charonia% Reserve 35mm 55mm 75mm 40 23 13 100 56.5 46.2 0.0 43.3 46.2 0.0 0.0 7.7 0.0 0.0 0.0 Non-Reserve 35mm 55mm 75mm 11 7 4 54.5 42.9 0.0 0.0 0.0 0.0 45.5 42.9 50.0 0.0 14.3 50.0

Fig 2. Survival of tethered urchins at reserve (black circles) and non-reserve (open circles) areas. The mean number of tethered urchins surviving in each of the three size classes is given for Leigh and Tawharanui.

We wonder why this experiment was not continued for days 15, 20 and 30, which would have revealed more interesting information. Reader please not that the difference between 98% and 96% survival rate is negligible while this corresponds to mortality rates of 2% and 4%, representing a major difference. Note how survival rates eventually level off.

Urchin density and size structure
Densities of Evechinus in the urchin-grazed habitat varied widely between sites but were generally lower at reserve sites for both areas (Fig 3). the density of exposed urchins (Fig 3A) was significantly lower at marine reserve sites (Tables 1,3 [left out]). Exposed urchins were 1.7 times more abundant overall at non-reserve sites (Table 1 [left out]). There was no difference in urchin density between Leigh and Tawharanui areas but there was a significan interaction between Area and Status (parameters of the statistical mixer).

Fig 3. Mean density of A exposed and B cryptic urchins, and C mean size of exposed urchins
from quadrat sampling (n=10) at all reserve (open circles) and non-reserve sites (filled circles).

This graph shows fewer large urchins in reserves but a comparable number of recruits. Also those inside reserves are larger, although not so for Tawharanui. The difference is more pronounced for Goat Island and almost negligible for Tawharanui. when expressed in grazing capacity, the difference is likely to become even less significant. This difference comes to expression in the bimodal size distribution of urchins inside the Leigh reserve.

Comparisons of urchin populations Populations were more bimodal (having two peaks) at reserve sites, with very low numbers of urchins between 30 and 50mm and they generally remained cryptic (hidden in cracks and under boulders) to a greater size. This pattern was stronger in the Leigh marine reserve.  The authors make no effort explaining the bi-modality, which could well be a symptom of stress, degradation, mortality or recruitment failure, all three symptoms have been observed in the reserve areas. The fact that urchins remained cryptic to a greater size may well be related to the difference in wave exposure between the reserve and non-reserve sites. It urges caution in interpreting the tethered urchin experiments where the small urchins (35mm) and arguably also the 55mm ones were indeed still cryptic for the north facing coasts. If these are omitted from the experiment, only very little difference in survival rate remains to explain the disappearance of the barren zones.

Fig 4. Size frequency distribution of all Evechinus measured during quadrat sampling at each area. Shaded bars indicate poportion of cryptic urchins.

Urchin removal At the commencement of the experiment in January 1998 Evechinus densities did not vary between treatments. Densities of urchins ranged from 1.2 to 2.4 per 0.25m2. Crustose coralline algae (Lithothamnion and Lithophyllum spp.) were dominant, covering 63-99% of the substratum. Articulated coralline turf was the other dominant encrusting form with cover ranging between 0 and 35%. There was no significant difference between either crustose coralline algae or coralline turf between treatments or between plots within treatments. Macroalgae were rare at the start of the eperiment, with Carpophyllum flexuosum which is relatively resilient to urchin grazing (Cole and Haggit 2001), being the only conspicuous large seaweed. . . .  After 1 year the control plots remained as urchin barrens dominated by crustose coralline algae, while the urchin-removal plots had become dominated by coralline turf, with a mixture of large and small brown algae. . .  This experiment was conducted in the SE facing shore near Mathesons Bay, which is characterised by medium exposure and rock fissures creating ideal isolation for experimental plots. Although the experiment removed the urchin as grazer, other grazers such as Cookia sulcata and many fish species still had access. The experiment shows that large seaweeds will not necessarily take over, even after one year of isolation. But the small seaweed species took over (note the difference of an order of magnitude in the scale of graph D! This result provides a strong argument against the kelp win, urchin lose hypothesis! It also provides a strong argument against high kelp productivity since most productivity by far arose from small brown algae.

Fig 7 A-D. Response of macroalgae to urchin removal. The mean abundance of A Carpophyllum flexuosum, B Ecklonia radiata, C Halopteris virgata and D small brown algae in both control (open circles) and urchin-removal plots (full circles) following the commencement of the experiment in January 1998.

Distribution of urchins habitat
The general distribution of habitats (Fig. 8) at non-reserve sites is typical of northeastern New Zealand with shallow fucoid (bladder weed) assemblages, deep kelp forests (Ecklonia radiata) and intermediate depths dominated by urchin barrens (Choat & Schiel 1982). However, at marine reserve sites all depth ranges were dominated by macroalgal habitats. The proportion of urchin barrens habitat was significantly lower than at non-reserve sites (Tables 1,7) [7 left out]). The relative odds ratio for the proportion of urchin barrens at reserve vs non-reserve sites was 0.2 to 1 (Table 1) or inversly, 5.9 times higher at non-reserve sites. This pattern was consistent between both areas but varied significantly with depth (Table 7). Differences . . were greatest in the shallow depth strata (0-3m and 4-6m) where urchin barrens covered approximately 54% of available reef at non-reserve sites and only about 20% of the reef at reserve sites. The deeper strata (7-9m and 10-12m) at both reserve and non-reserve sites were dominated by macroalgal habitat, predominantly Ecklonia forest.

The top two bar charts show that the fucoid zone in Leigh is deeper than elsewhere, wich demonstrates what we alluded to earlier, that the Goat Island coast is more exposed.  Note how the urchin barrens have been invaded by the kelp forest, which is easy to see under water. Had Kawau and/or Little Barrier Island been included, they too would have shown a kelp takeover. Sadly, this has not been done.

1. New Zealand's two oldest marine reserves at Leigh and Tawharanui support higher predator densities than similar unprotected areas. Snapper 5.8-8.7 times; spiny lobster 1.6-3.7 times. True, but this is evident already after five years of no-take protection (see paper on this page). The high snapper density is caused by using the baited video camera method. The low spiny lobster count was caused by the low numbers in the Tawharanui reserve!
2. Relative rates of predation on (tethered) sea urchins were found to be higher in the reserves. True but by far insufficient to explain the kelp takeover.
3. The densities of actively grazing urchins (excluding cryptic urchins) is lower in the reserves. True but so they are at Kawau, Little Barrier, Great Barrier and further.
4. The cover of macroalgal forests is higher in the reserves. True but so they are at Kawau, Little Barrier, Great Barrier and further.
5. These patterns . . . provide experimental evidence of top-down impact of predators on subtidal reef communities. This is a sleight of words. Every predator causes top-down impact. However, the results do not show sufficient impact to substantiate top-down control, which is an entirely different matter. It shows a shift in the economy of exploitation - perhaps. The results are entirely a consequencce of the 1993 kelp death degradation event.
6. The spiny lobster Jasus edwardsii was found to be an important predator of sea urchins at marine reserve sites. True. However, the control sites were ill-chosen. The prickly star Coscinasterias is not common on exposed coasts.

This is noteworthy, considering that spiny lobsters were not previously thought to forage in the urchin barrens habitat (Andrew & Choat 1982, Andrew & MacDiarmid 1991). From personal observations and those of a professional urchin diver, spiny lobsters are attracted to animals in distress, wherever they are located.
More importantly, when excluding the small urchins, the difference in survival rates between reserve and non-reserve are not enough to base far-reaching conclusions on.

Their study was carried out after only 4 years of marine reserve protection. If it were to be repeated now after 25 years of protection and recovery of predator populations, a larger effect may be expected. This myth lies at the heart of so much research done in recent years, blinding researchers from seeing the obvious. Where is the proof?

This study demonstrates the value of marine reserves as experimental tools to test ecosystem-level hypotheses at ecologically relevant scales, previously unfeasible using traditional caging and manipulation experiments. (Andrew & MacDiarmid 1991). Marine reserves have enabled us to measure the top-down role of predators in structuring subtidal reef communities in northeastern New Zealand, as well as the indirect effects of fishing on the trophic structure of reef communities.

This study does not give enough evidence for the above statement. When other areas are taken into account (Kawau, Little Barrier, Great Barrier, Rakitu, Mokohinau) it can be proved that this study is a failed experiment.The map shows how the authors have observed kelp invasions inside two marine reserves (red areas), but failed to observe similar effects elsewhere, particularly in the yellow areas nearby the study area (yellow areas). All areas where we observed kelp invasion of the urchin barrens were located where previously kelpbed death has been observed. This can still be confirmed today (2006).

In essence, the ideal is protection of a portion of an exploited stock, with the expectation that the biomass of targeted species within protected areas will rebuild to approach unfished densities. Density-dependent processes might then cause emigration of adults from the ‘source’ (reserve) population to fished areas, either by passive diffusion (Beverton & Holt 1957) or by displacement of individuals caused by space limitation (Kramer & Chapman 1999). Additionally, spawning activity within the reserve by greater numbers of large individuals should result in greater production of gametes than in similar, unprotected areas.
. . . (the authors discuss computer models in studies) . . . While these effects are generally positive, on a case-by-case basis evidence for substantial recovery in populations is often limited (Jones, Cole & Battershill 1993; Rowley 1994; but see Russ & Alcala 1996a; Wantiez, Thollot & Kulbicki 1997; Edgar & Barrett 1999). There are three main reasons.

1. First, exploitation in any geographical region tends to begin with large predatory species (Pauly et al. 1998) that are less common than species at lower trophic levels and therefore more difficult to monitor with sufficient statistical power (Cole, Ayling & Creese 1990; Paddack & Estes 2000). Moreover, larger predators tend to be slower to grow and reproduce, which means that population-level responses may be slow (Jennings, Reynolds & Mills 1998).
2. Secondly, with few exceptions (Edgar & Barrett 1997, 1999; Wantiez, Thollot & Kulbicki 1997), the design of marine reserve surveys has often been spatially or temporally confounded (or both) so that the results must be interpreted cautiously.
3. Finally, field methods used to assess fish density have sometimes been subject to biases caused by intra- or interspecific behavioural plasticity . . . . .

1. determine the magnitude of differences in snapper density and size between reserve and adjacent fished areas, and
2. quantify seasonal and interannual variability in snapper density and size.

Methods Study areas The three reserves were the Cape Rodney to Okakari Point (Leigh) Marine Reserve, Tawharanui Marine Park and Te Whanganui a Hei (Hahei) Marine Reserve ( Table 1 and Fig. 1a). All three are complete no-take areas administered under different legislation and by separate government departments (the ‘reserves’ are controlled by the Marine Reserves Act 1971 and administered by the Department of Conservation, whereas the ‘park’ is controlled by the Fisheries Act 1983 and administered by regional authorities). The management of the Tawharanui Marine Park has no nearby local community that can provide day-to-day surveillance to assist enforcement because the adjacent land is a regional park.  Incorrect. Local communities have not been involved in managing any of the three reserves. This is done centrally by government departments.

Fig. 1. (a) Map of the Hauraki Gulf and environs, showing the location of the three reserves surveyed in this study. Depth contours are in metres (?). The inset shows the location of the area in the North Island of New Zealand. (b) Details of the three reserves surveyed, showing reserve boundaries (dashed lines) and survey areas. Top: Leigh (1977); centre, Tawharanui (1982); bottom, Hahei (1992).

We have serious doubt about the accuracy of the BUV system as explained in Recurring Mistakes. Just the way snapper develop diver-positive behaviour, they also develop BUV-positive behaviour with pilchards as reward.

The entire method is aimed at attaining the maximum result.

The formula for daily batch fecundity (F) of Zeldis & Francis 1998 is F = 73.9 x W -7793  where W = snapper biomass or weight. For ease of reference, it approximates to F= (W - 105) x 74, where snapper need to weigh over 105g to produce 74 wet-weighted eggs per gram of body weight in each spawning batch. The formula was derived from snappers weighing between 200g and 5kg (65cm).

Results EFFECTS OF MARINE RESERVE PROTECTION ON P. AURATUS DENSITY, BIOMASS AND EGG PRODUCTION  Biomass per BUV deployment and density of legal-size snapper (LEGsna) were higher in the reserve than adjacent non-reserve areas at all three locations and for all four surveys (Figs 2 and 3). In particular, the Leigh reserve recorded the highest value of LEGsna on all four survey occasions, and the highest density of snapper of all sizes (MAXsna) on all but the April 1999 survey, when the Hahei reserve MAXsna was boosted by large numbers of sublegal fish.  The graphs show the number of snapper (above) and their relative biomass (below) for the two years of observation. It shows that the number of snapper in each reserve were comparable but that the ones in Leigh were much larger. It also shows that the numbers and sizes of snapper outside the reserves were similar for all areas. The main conclusion could be either that Leigh is a special area attracting large fish, or that Leigh has larger fish because it is older. For each possibility, proof is still lacking.  It is regrettable that the authors did only two surveys per year, rather than one for each of the four seasons. Because of this, it is not known when maximum and minimum densities occur. Fig 2b shows that the biomass inside two marine reserves is 2-3 times that outside, with Leigh being an exception. . . . (the authors continue explaining statistical niceties) . . .

Fig. 2. Mean reserve (filled symbols) and non-reserve (open symbols) snapper Pagrus auratus relative density at Leigh, Hahei and Tawharanui from November 1997 to April 1999. (a) Total numerical relative density, all size classes (MAXsna); (b) relative biomass.

Fig. 3. Mean reserve (filled symbols) and non-reserve (open symbols) snapper Pagrus auratus numerical relative density at Leigh, Hahei and Tawharanui from November 1997 to April 1999. (a) Fish > minimum legal size (LEGsna); (b) fish < minimum legal size (JUVsna).

Fig. 4. Predicted mean LEGsna by survey area at Leigh, Hahei and Tawharanui. Vertical dashed lines represent reserve boundaries, solid symbols are reserve (R) areas and open symbols are non-reserve (NR) areas. Fig. 5. Predicted mean JUVsna by survey area at Leigh, Hahei and Tawharanui. Vertical dashed lines represent reserve boundaries, solid symbols are reserve areas and open symbols are non-reserve areas.

The above graph (Fig3) shows that legal-sized snapper are more abundant inside reserves during the end of summer, but Leigh has more legal sized-snapper all year round. Juveniles do not make a distinction between reserve or non-reserve.
Fig 4 and 5 show how adult and juvenile snapper are distributed along the reserve's coast. As expected, juveniles are not affected by marine reserves but adults do. Only in Leigh does the density increase towards its centre. It again demonstrates that Goat Island is a special place which cannot be compared to others. It is remarkable that the number of juveniles increases in areas 9-13 outside the Leigh reserve. But this coast is also more sheltered and influenced by currents.
It is important to note that the differences between a reserve of 7 years old (Hahei, 1992) and one of 17 years (Tawharanui, 1982) is negligible. Only Leigh (1977, 22 years old) stands out, but it is an oasis with an island, next to a long beach.

It is regrettable that the experiment does not show four sample points for each year. Yet the need for this could have been foreseen.

This has indeed happened in several surveys.

Goat Island is a special place with a mix of desirable habitats, exposure and shelter, extensive shellfish feeding grounds and sleeping habitat, all of which are important to snapper, particularly large ones. The outside of Goat Island, in the centre also has the clearest water.

This is not borne out by this experiment. The claim of 14 times is caused by the Leigh data set swamping the two other data sets. By all scientific standards, it should have been removed as a spurious dataset biasing the experiment unduly. The remaining bias comes from the BUV counting method, whose linearity has not been proved. It is a method that exaggerates.

The outside areas are all different habitats. To the west under influence of the Pakiri Beach and to the SE they offer more shelter. They cannot really be compared with Goat Island.

This research was conducted with a measuring device (BUV) which fails the scientific requirements for measuring devices and cannot be used for quantitive values. The data obtained from Leigh completely swamps all other data such that the conclusions in their generality are not substantiated. But it is clear that marine reserves do have more snapper and crayfish.

The same ecological theory demands that such changes happen as gradually as large predators establish themselves. The observed disappearance of urchin barrens however, happened rather quickly between 1995 and 1998, in one reserve of 24 years of age (Leigh, 1977) and at the same time also in a younger but nearby one of 19 years (Tawharanui, 1982).

A cause and effect relationship has never been demonstrated. See also the research exposed above. But the decline of urchin populations could readily be observed, as it happened in 1994/95. After the 1993 kelpbed death, the new kelp invaded the urchin habitat because  (i) urchins wandered into the wide kelp zone, (ii) urchins died from mass mortalities, (iii) the kelp plants reached maximal growth all at the same time and (iv) there were no storms to re-create the urchin barrens.

However, the other reserves refute these findings, as do a number of other places around the outer Hauraki Gulf. The use of the word 'significant' is suspect, since it usually means 'just detectable'. Why not quantify the difference such as 10%, 2-3 times or so?

Tuhua is located in the Bay of Plenty where the dense plankton blooms of 1991/92 did not happen. However, in 2002 they suffered mass urchin mortality.

The control sites chosen around Tawharanui and Leigh are facing southeast rather than north, or they are located in shallow water. These places are unsuitable control sites. The Hahei effect also applies to the control sites around Leigh and Tawharanui. The use of the 'statistical mixer' method hides whether the Leigh marine reserve data swamps that of Tawharaunui.

After becoming a fully protected marine reserve in 1999, the number of snapper suddenly increased spectacularly, including large snapper but the islands have only a small rocklobster population, even though these have been protected since 1981. The Mokohinau Islands have experienced extensive kelpbed death and disappearance of their urchin barren zones, thus confounding the picture.

The main point is that this study does not prove anything at all. By any standard, it is a failed experiment from which no conclusions can be drawn. If it takes so much effort to demonstrate the effects of marine reserves, they are perhaps too small to be of any practical value. More monitoring and pre-and post-reserve sampling will not change this.