Designating land as a protected area, such as a national park, is a frequently used strategy to prevent habitat loss and maintain species diversity. In addition to preserving habitats and biodiversity, they are thought to have a role in protecting ecosystem services such as carbon storage and sequestration, water, climate and soil stabilisation, among other functions. However, there is debate over the extent to which protected areas deliver desired conservation outcomes.
To review the evidence on whether terrestrial protected areas maintain natural species populations and prevent habitat loss.
The authors included studies that assessed the effect of establishing a protected area on change in species abundance, or habitat extent or structure, compared with areas outside the protected area or the same area before the protected area was established. They included studies using an inside/outside protected areas (PA) comparison, before/after establishment of PA comparisons as well studies describing drivers and interventions. The authors searched online databases and catalogues (including ISI Web of Knowledge, Science Direct and ProQuest), specialist websites (including the United Nations Development Programme and the World Conservation Monitoring Centre) and Google scholar. Searches were conducted in English between July and August 2010, with additional searches of ISI Web of Knowledge and Google scholar conducted in Spanish and Danish. The authors also consulted with an expert group from the International Union for the Conservation of Nature joint taskforce on Biodiversity and Protected Areas. The authors assessed the quality of studies (based on their own list of criteria), extracted data and wrote a narrative synthesis of the results.
Headline Findings: a summary statement
The authors found generally positive outcomes, particularly for protection of habitat in tropical forested areas, but concluded that there is insufficient evidence to make policy recommendations about protected areas.
The authors included 86 articles, which they divided into 118 studies (based on the number of counterfactual scenarios presented) resulting in 42 studies on species population trends and 76 studies on habitat change. Most of the data (74%) on species trends came from protected areas in tropical regions: twentyfour studies from Sub-Saharan Africa, five from Asia, two from Latin America, five from North America, five from Europe and one from Oceania Of the studies on habitat change, 35 took place in Latin America and the Caribbean, 18 came from Sub-Saharan Africa, 16 from Asia, one from Europe, one from North America, two from Oceania, and four were global studies.'
Implications for policy and practice
In the studies examining species, 31 out of the 42 studies found that species populations fared better in protected areas (whether this was maintaining or increasing the population size, or declining at a slower rate than outside the protected areas). The authors find that the results are generally positive, but insufficient to make policy recommendations.
For habitat change, 60 of the 76 studies found that the rate of habitat loss was lower inside protected areas. The authors conclude that protected areas are an important element of conservation strategies to preserve tropical forests, but there is insufficient data to make conclusions about other areas.
Implications for further research
The authors call for monitoring data from conservation projects to be available, transparent and standardised. They also note the need for more before-after, control-intervention designs linking observed changes to conservation interventions. They also suggest there is a need for studies that provide a more nuanced understanding of why protected areas are effective (or not effective), to guide management of protected areas.
The review authors attempted to synthesise a diverse set of studies. They used standardised methods to search, screen and extract data in a transparent manner.' However, the review has a few limitations. It is not clear whether independent data extraction was done by two reviewers. The authors clearly describe that meta-analysis was not possible due to study heterogeneity, however they use vote counting which can misrepresent the overall weight of evidence. Finally, the reviewers describe a process of critical appraisal of included studies, but they do not report methods or the results of this assessment. This is an important limitation, in particular as the review adopts broad study design inclusion criteria. These limitations are partially mitigated by the lack of any strong policy conclusions due to the overall lack of evidence.'