News & Views | Published:


Making marine protected areas work

Nature volume 506, pages 167168 (13 February 2014) | Download Citation

  • A Correction to this article was published on 19 February 2014

Globally consistent surveys of five factors influencing the success of marine protected areas — age, size, isolation, protection and enforcement — reveal that only when all five are present does nature thrive. See Letter p.216

In the past few years, several huge marine protected areas (MPAs) have been created in the Pacific and Indian oceans, totalling more than 1.6 million square kilometres. That might sound like a large area, but even when combined with every other MPA on the planet, still less than around 2% of the world's oceans are fully protected. In response to this conservation shortfall, coastal nations have committed to increasing the amount of protected area in their territorial waters to at least 10% by 2020. However, even if this target is met, the actual conservation value may be limited because MPAs often exist in name only — they do not truly provide protection. On page 216 of this issue, Edgar et al.1 provide key insights into why so many of the world's MPAs are failing to meet their full potential and describe a clear path forward for achieving better conservation outcomes.

The problems with existing protected areas are manifold. 'Paper parks', for which protected-area boundaries exist in principle but are not enforced, have little conservation value. Unfortunately, there are a lot of paper parks2,3, and many other protected areas are only partially protected — activities such as recreational or hook-and-line fishing are allowed, leading to less conservation value than full protection4. Furthermore, nearly half of all MPAs are little bigger than a football field or have only recently been created5, limiting their ability to protect many species.

It might seem as if we know a lot about what leads to MPA success or failure, but the simultaneous assessment of how various factors affect MPA success has been missing from previous studies. In other words, most studies tested one factor at a time, without controlling for the others. This is what makes Edgar and colleagues' work particularly unusual. Using data from 87 MPAs around the world, all sampled with the same methods, the authors compared how the biomass, abundance and diversity of species in MPAs varied with all combinations of five factors: whether or not the zones were fully protected, enforced, old (more than 10 years), large (more than 100 km2) and isolated (for example, a remote island or a patch of reef surrounded by large sandy areas).

Their findings are remarkable and unsettling. Instead of each factor being important, such that benefits accrued proportionally to the number of factors present, the authors found that conservation value was nearly absent in MPAs with fewer than three of these factors, and then increased exponentially from three to five factors (Fig. 1). This result is worrying because it is difficult for managers to achieve all five factors.

Figure 1: Five factors for effective protection.
Figure 1

Edgar et al.1 tested how five features — full protection, enforcement of protection, age greater than 10 years, size more than 100 km2 and isolation — affect conservation benefits of marine protected areas (MPAs). They find that, compared with unprotected areas, there is no benefit gained from an MPA with none of these features (for example (a), a young MPA that is still fished by recreational and small-scale commercial fishermen, is too small to encompass the reef and in which protection is not well enforced), or even from an MPA with one or two features. But the conservation value of an MPA rises exponentially as the number of features increases from three (for example (b), an older MPA that protects an isolated reef and is enforced, but that is small and not fully protected) to all five (c).

The authors also report a secondary, but equally interesting, result. As a by-product of the methods used to determine MPA effectiveness, Edgar et al. were able to estimate that fish biomass in coastal regions around the world has, on average, been reduced by about two-thirds relative to pristine conditions. Given the difficulty in assessing near-shore, small-scale fisheries in most parts of the world, this result and its underlying methods should help to improve those stock assessments and thus fisheries management, especially in developing nations.

A key concern about Edgar and colleagues' study is that only 4.6% of the assessed MPAs had all five features, and only an additional 5.7% had four. This is a limitation of the real world — only a few MPAs exist that achieve more than three of the factors — but, nonetheless, it creates a potential bias in their results. Clearly, more research is needed to better understand the generality of the authors' results, and how other factors could influence conservation success. In particular, the past decade has seen advances in the assessment and design of networks of MPAs6, and such research suggests that the connectivity provided by networks might compensate for other missing factors, especially small size and lack of isolation. Most managers do not have the option of placing protected areas in isolated locations (the most influential factor, according to this study) or creating large reserves, such that the menu of design options is constrained to a set of variables that can at best achieve suboptimal results. Future work needs to identify how network connectivity and other MPA attributes might help to achieve maximum conservation success.

Limitations aside, this study is an important advance for several reasons. It is global in extent and consistent in methodology, mitigating the common critique that large-scale studies are not relevant to specific sites or comparable from location to location. The value of this aspect of the study cannot be overstated. The findings also help to calibrate expectations for existing and future MPAs. It is clear that designating and enforcing park boundaries, although necessary, is not sufficient to gain full conservation benefits, and that protected areas without all five features should not be expected to produce such benefits. The authors frame these lessons in the context of international conservation objectives, but they are arguably more important for local managers and stakeholders.


  1. 1.

    et al. Nature 506, 216–220 (2014).

  2. 2.

    et al. Science 312, 750–751 (2006).

  3. 3.

    et al. Mar. Pollut. Bull. 44, 1177–1183 (2002).

  4. 4.

    & Mar. Ecol. Prog. Ser. 367, 49–56 (2008).

  5. 5.

    et al. Oryx 42, 340–351 (2008).

  6. 6.

    et al. Proc. Natl Acad. Sci. USA 107, 18251–18255 (2010).

Download references

Author information


  1. Benjamin S. Halpern is in the Bren School of Environmental Science & Management, University of California, Santa Barbara, Santa Barbara, California 93106, USA, and in the Faculty of Natural Sciences, Imperial College London, UK.

    • Benjamin S. Halpern


  1. Search for Benjamin S. Halpern in:

Corresponding author

Correspondence to Benjamin S. Halpern.

About this article

Publication history



Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Newsletter Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing