Conservation of birds in fragmented landscapes requires protected areas

For successful conservation of biodiversity, it is vital to know whether protected areas in increasingly fragmented landscapes effectively safeguard species. However, how large habitat fragments must be, and what level of protection is required to sustain species, remains poorly known.

We compiled a global data set on almost 2000 bird species in 741 forest fragments varying in size and protection status, and show that protection is associated with higher bird occurrence, especially for threatened species.

Protection becomes increasingly effective with increasing size of forest fragments. For forest fragments >50 ha our results show that strict protection (International Union for Conservation of Nature [IUCN] categories I–IV) is strongly associated with higher bird occurrence, whereas fragments had to be at least 175 ha for moderate protection (IUCN categories V and VI) to have a positive effect.

This meta-analysis quantifies the importance of fragment size, protection status, and their interaction for the conservation of bird species communities, and stresses that protection should not be limited to large pristine areas.

In a nutshell:

We analyzed the combined effects of size and protection status of isolated forest fragments on bird species occurrence in human-modified landscapes

Declines in species occurrence across all feeding guilds with decreasing fragment size underscore the importance of large intact forests for conserving avian diversity

Positive associations between protection and species occurrence suggest that protected areas are effective for maintaining bird species in fragments >50 ha

Conservation of threatened bird species in fragmented landscapes should preferably focus on strict protection of large forest fragments

Home to ~80% of the world’s terrestrial biodiversity (World Bank 2004), forest systems are increasingly under threat from a wide range of anthropogenic pressures (Newbold et al. 2015). Between 2000 and 2012, 2.3 million km2 of forest was lost globally (Hansen et al. 2013), and up to 70% of the remaining forest area worldwide is estimated to be within only a single kilometer of forest edges (Haddad et al. 2015).

Habitat loss and fragmentation are major drivers of biodiversity loss, with disproportionate species declines in small, isolated fragments (Lees and Peres 2006; Bregman et al. 2014; Keinath et al. 2017). Protected areas (PAs) are increasingly being implemented as a tool to conserve species and maintain associated ecosystem services.

As a result, PA coverage has almost doubled over the past 30 years, from ~8.2% of terrestrial land surfaces in 1990 to 15% in 2020 (UNEP-WCMC and IUCN 2020), although coverage still falls short of the Convention on Biological Diversity’s Aichi Biodiversity Target of 17%.

However, to what extent PAs are an effective conservation measure in landscapes where natural habitat types have become fragmented and are under growing pressure of human activities (hereafter “fragmented landscapes”) remains unclear.

Several global studies on PA performance have demonstrated the importance of protection status of relatively large and pristine areas for conservation (Geldmann et al. 2013; Coetzee et al. 2014; Barnes et al. 2016; Gray et al. 2016). The PAs analyzed in these studies do not reflect the PA characteristics of the many small and isolated PAs embedded within fragmented landscapes.

At the same time, PAs across the world are facing increasing isolation, encroachment, and degradation (DeFries et al. 2005; Laurance et al. 2012). This is a cause for concern given that species richness and population sizes tend to decline with reductions in forest fragment size and increasing fragment isolation (Lees and Peres 2006; Bregman et al. 2014; Keinath et al. 2017).

PAs are invested with different degrees of protection, which might affect their ability to conserve species in fragmented landscapes. The International Union for Conservation of Nature (IUCN) recognizes six categories of PAs (designated as categories I–VI) that can be grouped into two broad protection types (WebTable “Strict protection” areas are managed for ecosystem and species conservation (categories I–IV), while “moderate protection” areas encompass culturally modified landscapes (category V) and areas managed primarily for the sustainable use of natural resources (category VI) (Locke and Dearden 2005; Dudley 2008).

Much of the recent increase in PA cover consists of PAs under moderate protection (WebFigure 1), which currently account for ~42% of the global area dedicated to PAs with a designated IUCN category (UNEP-WCMC and IUCN 2020). Hence, understanding whether the effectiveness of species conservation in fragmented landscapes differs among protection types and how large fragments should be to minimize species loss and maintain ecosystem functioning is of great importance.

Our primary objective was to assess the interplay between fragment size and protection status in determining species occurrence.

We test the hypothesis that protection of forest fragments mitigates declines in bird species occurrence in response to decreasing fragment size (WebPanel 1). We focused on birds because they are highly represented in fragmentation studies and provide important ecosystem services, such as control of phytophagous insects, plant pollination, and seed dispersal (Figure 1, below).

Through a meta-analytical approach, we examined how fragment size and protection status are associated with the probability of occurrence of bird species within 46 fragmented landscapes worldwide.

We systematically searched existing literature and compiled a global dataset that draws on 61,716 occurrence records of 1990 bird species across 741 mature forest and savanna woodland fragments ranging in size from ~0.1 ha to over 10,000 ha (WebFigure 2; WebTable 2). Most of these fragments form sharp boundaries with areas of anthropogenic land use.

Results

Fragment size and protection status

We found that the relationship between fragment size and bird species occurrence depends on the protection status of fragments (Table 1; Figure 2a). This indicates that for the conservation of birds, protection becomes increasingly effective with increasing size of forest fragments regardless of protection status.

Threatened species

Differences in species occurrence were much more pronounced when only threatened species were considered (WebTable 3; Figure 2b)…

As with the analysis of the entire bird community, differences among PA types were greatest in the largest fragments. For fragments ≥10,000 ha, the occurrence of threatened species was higher in fragments under strict protection (0.99 [CI: 0.74–1.00]) than in fragments under no protection (0.58 [CI: 0.10–0.95], z-ratio = –3.43, P < 0.01), and marginally higher than in fragments under moderate protection (0.81 [CI: 0.18–0.99], z-ratio = –2.29, P = 0.06).

To read the full article, visit

https://esajournals.onlinelibrary.wiley.com/doi/full/10.1002/fee.2485#support-information-section