The spillover effect

18 04 2010

© everlessaday

The so-called ‘spillover effect’ is a long-standing debate in conservation ecology. The idea is relatively simple – put in a marine reserve (or, no-take zone, park, whatever you wish to call it as long as it restricts blanket over-fishing) and the area around the reserve eventually profits from the nearby over-production of fish (and other taxa). The idea is very attractive because even if you’re thick enough not to understand the absolute necessity of marine reserves in our age of mass, global over-exploitation, at least you might have enough grey matter to appreciate the value of more fish ‘spilling over’ into your favourite fishing area. More proposed marine reserves have been sold to the more Luddite ‘stakeholder’ this way than I care to count.

However, as attractive an idea it was, early on in the marine reserve literature (i.e., the early Devonian 1990s), there was limited (Rowley 1994; Willis et al. 2003) or only circumstantial evidence (Russ & Alcala 1996; Roberts et al. 2005) for the effect. Indeed, many have suggested that the spillover benefit, if present, depends entirely on the size of the reserve and whether adjacent areas are managed at all (Allison et al. 1996; McClanahan & Mangi 2000). Others have even suggested that marine reserves can displace fishing effort into smaller areas and change local community structure enough to facilitate invasion by exotic species (Kellner & Hastings 2009).

It is happier time now that we have more than ample evidence that marine reserves do in fact result in species spillover (e.g.,Roberts et al. 2001; Russ et al. 2004; Abesamis & Russ 2005). So it is not with any great claims of novelty that I highlight Garry Russ & Angel Alcala’s latest paper, Enhanced biodiversity beyond marine reserve boundaries: the cup spilleth-over; rather, it’s how they quantify the long-term evidence, the mechanisms for how spillover occurs and how the community changes that they deserve a mention.

Garry Russ, that denizen of marine spillover research, has produced a fine paper showing how communties change in response to the establishment of marine reserves. Using up to 25 years’ worth of data, Russ & Alcala show convincingly that:

  • species richness increases linearly with time since reserve establishment, outside of the reserve as well as inside
  • the change is not (primarily) due to habitat change
  • the effect tapers off with distance from reserve (as has been shown before)
  • large, predatory fish are more common inside and just outside reserves than farther away
  • the community composition outside the reserves becomes more like that inside over time

It’s these last two points that I find fascinating: 1. The importance of maintaining top predators in guarding ecosystem health cannot be understated (see previous posts here, here and here); and 2. It’s not just an immediate benefit – entire communities slowly spread outward from the area of initial protection like a growing amoeba of recovery.

There are quite a few other titbits to discover in this paper, and certainly a wealth of literature is now available to say once and for all that marine reserves benefit everything within and around them, for fishers and non-fishers alike. Take heed the next time you think you are being relieved of your fishing ‘rights’ when someone proposes a marine reserve in ‘your’ favourite patch. If you want fish in your system for your children and grandchildren to enjoy, embrace such proposals with enthusiasm.

CJA Bradshaw

ResearchBlogging.orgRuss, G., & Alcala, A. (2010). Enhanced biodiversity beyond marine reserve boundaries: the cup spilleth-over Ecological Applications DOI: 10.1890/09-1197

Add to FacebookAdd to DiggAdd to Del.icio.usAdd to StumbleuponAdd to RedditAdd to BlinklistAdd to TwitterAdd to TechnoratiAdd to Yahoo BuzzAdd to Newsvine



14 responses

3 11 2022
Would sustainable exploitation of polar biomes leave us on thin ice? | kdm communications

[…] ecological profits it offers. This vast area represents a massive opportunity for conservation ‘spillover’ and socio-economic gain, such as in ecotourism and for global trade. An open route through the […]


5 02 2019
Hedgerows; the hope for our growing population lies within the undergrowth. – Wandering Westhenry

[…] beyond marine reserve boundaries: the cup spilleth-over Ecological Applications, available at:, Accessed on: […]


27 10 2017
Honey, I’m home – Cities as a reservoir for UK bees | Science Blog - MResBEC

[…] they could become so numerous that they repopulate surrounding landscape types by what is called a spillover effect. Scientists cannot definitively say that a spill over effect will happen but it is far from […]


12 12 2014
Benefits of Marine Reserves in Fiji

[…] Corey Bradshaw (currently at the University of Adelaide in South Australia), wrote an interesting blog about the “spill over effect” and referenced a relevant academic paper called Enhanced […]


19 06 2013
The MPA backlash has officially begun – Sea Monster

[…] Spillover effects could or would prevent the need for fishery displacement – if by locally closing 20% of an […]


23 02 2013
Sea sharing or sea sparing: How should we manage our oceans? | Mola Mola

[…] (widely supported) in situ success of MPAs, but also in their (often positive, but more variable) spillover effects. Responsible fishing, by contrast, requires more extensive areas to be exploited, which may limit […]


18 06 2011
The few, the loud and the factually challenged «

[…] who benefit from the free, public-good resource that they assist in maintaining (see here, here and here). The evidence is clear world-wide: marine parks benefit pretty much everything and […]


5 04 2011
Does the pope wear a funny hat? «

[…] blogged several times on the subject (see Marine protected areas: do they work?, The spillover effect, Interview with a social (conservation) scientist, and Failing on ocean protection), but […]


5 04 2011
Does the pope where a funny hat? «

[…] blogged several times on the subject (see Marine protected areas: do they work?, The spillover effect, Interview with a social (conservation) scientist, and Failing on ocean protection), but […]


21 03 2011
Colin Buxton

Hi Corey – no more metaphors!

As Caleb says, your article is about spillover, a putative benefit of MPA to fisheries.

But setting that aside, where is the science that demonstrates that fisheries is a key threatening process to those biodiversity values you suggest should be protected by MPAs? Furthermore, assuming that other activities such as those you have listed (desalination, shipping transport, human population size and pollution) are threatening processes – where is the science that demonstrates that MPAs can mitigate these effects?

This is a geniune request for more information, because I can’t find it!

To clarify my perspective, I am just as concerned as you are about the health and wellbeing of our marine environment, recognise the need to conserve biodiversity and recognise the need to ensure that our actions in the marine environment are sustainable and do not cause long term damage.

The difference is that I am highly sceptical that MPAs can mitigate the recognised threats. I see absolutely no point in argueing that no-take is an appropriate management response to these threats, especially if fishing in itself is not a demonstrated threat.


5 04 2011


Excuse the pun, but your arguments do not hold water because they are not falsifiable, nor are they clearly objective. I have to state straight up that your positions, as fisheries researchers funded in large part by the fishing industry itself (via the Fisheries Research & Development Corporation), makes your stance highly circumspect.

Second, you fisheries scientists don’t exactly have the best track record worldwide. Almost every single major fishery in the world (Grand Banks, North Sea, Cortes…) has invariably been preceded with predictions by fisheries scientists that all is well; then ‘wham!’ – severe population declines and fishing moratoria.

The main problem here is that you work in a system that is inherently vague, difficult to study and prone to gross inaccuracies because you simply cannot survey target populations effectively – never have, and probably never will. Thus, most models are no more than educated guesses, and they are often wrong.

The straw man of ‘prove we don’t have a problem’ is a tired mantra repeated so many times it ceases to amaze me. With inadequate data and some of the highest ecosystem uncertainty on the planet, ‘proving’ there is a problem is an impossible hypothesis to test. It’s like trying to prove god exists (or doesn’t).

So I come back to my original arguments – if you do not reduce killing fish, you will kill the same amount of fish. The tired and inevitable wrong assumptions about depensation in fisheries (assuming the basic and erroneous Beverton-Holt density feedback model) means that you do not have a magic replenishment of fish just because you happen to be killing them. Look at the literature, guys! It doesn’t work.

I’ll leave you with a post by Hugh Possingham exposing the weakness of your arguments:

“Does Fishing Kill Fish? Do Marine Reserves Work?
By Hugh Possingham

Science has long demonstrated that marine reserves protect marine biodiversity. Rather than answer the same question again, isn’t it about time we started funding research that answers some useful scientific questions?

As marine reserves spread inexorably across the planet, the cry from skeptics and some fishermen is: “Do marine reserves work?” The science is pretty clear but acknowledgement of this by the public is another story. Let me begin with a story of my experience answering this question while communicating to stakeholders the subtleties of marine conservation planning during the rezoning of Moreton Bay.

I was asked by the then-Queensland Environmental Protection Agency to explain to stakeholders the process of marine reserve system design as it applied to the Moreton Bay rezoning. I told the gathering that the rezoning was about conserving a fraction of each mappable biodiversity attribute (species and habitats) for the minimum impact on the livelihood of others.

Things were going well (I thought), but then came the first question: “Prove to us marine reserves work! I don’t think they do.”

When I asked why he felt this way he responded: “Because most of the fish we catch move so much we will catch them when they come out of the reserve”.

I responded (though not quite this eloquently): “Your argument leads to two logical responses – there is no case for compensating fishermen for the new marine reserves in Moreton Bay because you catch the fish when they come out, or the reserves need to be a lot bigger”.

I went on to say: “Furthermore we have proved that marine reserves work in at least 100 studies worldwide. Why wouldn’t they work in Moreton Bay?” But the response from the fishing stakeholders was: “We need it proved that they work in Moreton Bay. Moreton Bay is different.”

Around Australia, as our governments boldly forge ahead rezoning Australia’s waters, the question of whether marine reserves work is asked by many people who do not know the marine literature. As a result a great deal of monitoring and evaluation effort is being spent on testing the null hypothesis: “Fishing does not kill fish”.

The classic Honours student Before-After-Control-Impact (BACI) design is being rolled out across Australia, and it looks a bit like this: we count and measure fish in several places, stop fishing in some of those places, continue to count and measure so we can try to reject the null hypothesis that fishing has no impact.

Sometimes we can’t prove it, because we don’t have enough data. However, an alternative approach is to go to the local primary school and ask a small child: “If we stop killing all the fish in a place do you think there may be more of them and the ones left may be bigger?”

Of course there are many subtleties to the removal of fishing pressure – in more sedentary species territorial adults may exclude juveniles and numbers may drop. For some species reserves will be too small, and if we have reserve size as a covariate then we might get useful information about minimum reserve size.

However, when you establish a protected area, invariably biomass (the combination of numbers and size) will go up, especially in the higher trophic levels. You can’t remove lots of fish without having an impact – to ask this question again and again is like questioning the powers of gravity every day in every place. Next time you fly to New Zealand be careful when you step out of the plane in case gravity failed while you were crossing the Tasman.

Just because policy-makers, stakeholders, managers or politicians ask a question it doesn’t mean that scientists should take funding to answer it. Ecology may be a difficult science with few fundamental laws, but many of its laws and principles are global – like killing animals reduces population size.

There are useful scientific questions we can answer while monitoring marine reserves that will help with future rezoning and policy. The key is that applied monitoring has to pass some tests.

Do we know enough already from other studies? Is the answer already obvious? If we get an answer do we have a management response?

This sort of thinking will be the subject of some of the research of the new ARC Centre of Excellence for Environmental Decisions, which gets underway this year. Watch this space!”

Professor Hugh Possingham is Director of the Applied Environmental Decision Analysis centre at the University of Queensland.


21 03 2011
caleb gardner

A few observations on this interesting discussion….

1. The original article by Bradshaw was about the spillover effect and potential benefits to fisheries that could result from closures. Buxton disputed this claim of a fisheries benefit and Bradshaw responded by saying “your perspective is somewhat tangential to the main issue. SA marine parks were not designed specifically to enhance fisheries”. Let’s stick to topic – this blog IS about the effect of MPAs on fisheries. If the claim of a fisheries benefit is weak then let’s discuss it rather than dodge.

2. I don’t see the logic of promoting MPAs in SA citing research from the Philippines and St Lucia but dismissing research from Tasmania. The obvious reasons for discussing the Tasmanian research is that the species are largely the same and the management / harvest rates similar. The St Lucia and Philippines research was with fisheries that were recruitment overfished, which is different to the SA situation. I doubt there’s any debate about the benefit of reducing harvest rate in the Russ and Alcala scenario (we’d all agree any management is better than none). The more interesting issue is what are the effects of MPAs on fisheries as per those in SA?

3. To be clear about what was reported from our research in Tasmania in the late 1990s and early 2000’s on Tasmanian MPAs (“the Devonian” as Corey refers to it). That research reported apparent spillover from MPAs – you are more likely to catch some large fish adjacent to the boundaries of MPAs. However, this doesn’t answer the bigger question of whether MPAs have reduced or increased biomass and what is the impact of MPAs on overfishing risk? We found that the effect of MPAs was mixed but more commonly led to a net loss of reproductive output which equals increased risk of recruitment overfishing. MPAs created this problem because they didn’t remove catch from the system – they shifted it. This means that the effects of MPAs can be complicated and it’s wrong to assume a simple benefit as Bradshaw’s article seems to suggest. The conclusion from the Tasmanian research is that it’s better to manage overfishing risk by controlling catch rather than shuffle catch with MPAs. The lesson here for SA is to be forceful in controlling catches but tread carefully with MPAs because they can do more harm than good.

4. The discussion argues in places that SA fisheries management is performing badly ( “SA is ramping up fisheries”….”rock lobster declines and quota reductions, tuna quota reductions”) . This point is really critical because MPAs are worthwhile where fisheries management fails (as per the Philippine examples). However – I can’t see the case in SA. The fact that tuna quotas were reduced when stocks fell would normally be taken as a sign of good management. Likewise, we know the decline in rock lobster stocks was not induced by overfishing because larval supply remained high and steady (Linnane et al., 2010; Phillips et al., 2010). Most importantly, the quota was massively reduced in response to stock decline. SA already has a process for responding to overfishing through stock assessments that inform ministerial decision making. If there are overfished stocks then the SA fisheries minister can and should be made to respond.

5. Moving on to biodiversity and the fact that we shouldn’t “wait until it does become seriously compromised and we have lost species before implementing protection”. Well yes, but specifically, what species are at risk from fishing in SA? If we know this then the SA Govn. must apply the Act and adjust management of the offending fishery, as per the process used to manage GHAT effort to protect threatened Australian Sealions (ASL). MPAs are often unhelpful here because they only shift the location of effort rather than control it. In the ASL example from SA, after the GAB MPAs were implemented they were shown to have failed to protect foraging areas and thus entanglements. The South Australia Fisheries Management Act 2007 was applied after the MPAs failed because the Act requires “(a) proper conservation and management measures are to be implemented to protect the aquatic resources of the State from over-exploitation and ensure that those resources are not endangered; (c) aquatic habitats are to be protected and conserved, and aquatic ecosystems and genetic diversity are to be maintained and enhanced.”

The stock response from MPA advocates to this type of example is along the lines that MPAs are supposed to complement not replace fisheries management. However, this misses the point that the GAB MPAs diverted resources and attention from better management of the threat to ASL by applying a blanket and poor solution to biodiversity conservation. So, will MPAs contribute to the recovery of any of the threatened marine species in SA or could resources be better used?


15 03 2011
Colin Buxton

Your article overlooks one very important consideration in advocating the benefit of MPAs to fisheries. The example you are quoting comes from a situation – the Phillipines – where there has been serious overfishing, where dynamiting and cynanide fishing has occurred in the past and although outlawed still occurs today.

Fortunately here in Australia and particularly in South Australia, our fisheries are well managed and blanket overfishing is not the same problem.

Several studies closer to home do not point to the same fishery benefits (Haddon et al 2003) and/or biodiversity improvements (Barrett et al 2007, 2009).

Unfortunately the uncritical reader may be misled by your blog and I fear some may even misuse this information in support of their MPA agendas.

Of course preventing blanket overfishing is likely to have benefits, but you cannot rectify the problem if you do not have it. It is certainly not an issue in SA.

Furthermore, and equally important in areas where there is good fisheries management, you ignore the problem that arises when effort is concentrated in the areas adjacent to the MPA – not a good overall biodiversity outcome.

Preachers of religion do not like to believe there is no god!

BARRETT, N.S., EDGAR, G.J., BUXTON C.D & HADDON, M. 2007. Changes in fish assemblages following 10 years of protection in Tasmanian marine protected areas J. Exp. Mar. Biol. Ecol. 345(2): 141-157.

BARRETT, N. S., BUXTON, C. D. & EDGAR, G. J. 2009. Changes in invertebrate and macroalgal populations within Tasmanian marine reserves in the decade following protection. Journal of Experimental Marine Biology and Ecology 370:104-119.

HADDON, M., BUXTON, C., GARDNER, C & BARRETT, N. 2003. Modelling the effect of introducing MPAs in a commercial fishery: A rock lobster example. In: Beumer, J.P, Grant, & Smith, D.C. Aquatic Protected Areas – What works best and how do we know? Proc. World Congress on Aquatic Protected Areas. Cairns. 428-436


15 03 2011

Thanks, Colin. I appreciate your comments, but you are just as guilty for pushing your perspective. First, your work is based on Tasmanian data, not SA, and despite your contention, there is no evidence whatsoever that SA Fisheries are preventing declines. Consider rock lobster declines and quota reductions, tuna quota reductions, and some evidence for commercial shark declines.

However, your perspective is somewhat tangential to the main issue. SA marine parks were not designed specifically to enhance fisheries – the are based almost entirely on protecting non-commercial marine biodiversity values (e.g., the endemism hotspot of marine algae, giant cuttlefish, blue groper, etc., etc.). Second, even if we accept that all fisheries in SA are fine, thank you very much (a great leap of faith and one I do not accept for all species), providing sanctuary now will ensure some degree of spillover in the future. Why wait until it does become seriously compromised and we have lost species before implementing protection? A ridiculous notion that is not in keeping with what we know from countless cases the world over that it’s nearly impossible to go back once the damage is done.

SA is ramping up everything, from fishing, aquaculture, desalination, shipping transport, human population size and pollution. If we don’t put these up now, there will be nothing to ‘spillover’ in the future.

Therefore, an uncritical reader might actually think your perspective is supported (and do not suggest that I base my conclusions on unsupported rubbish like religion – my blog is based on sound research – I found your metaphor rather insulting).


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: