Introduction
Human societies are under increasing pressure to halt unprecedented biodiversity loss (Caro et al. 2022; Cowie et al. 2022). Given the magnitude of current environmental impacts, this will require quickly protecting a considerable portion of the Earth’s biomes, moving from the currently protected ~ 17% of the planet surface to 30% before 2030 (Dinerstein et al. 2019; Maxwellet al. 2020). Yet, our understanding of natural phenomena remains at best partial across most of the planet (Hortal et al. 2015; Hughes et al. 2021). There is, therefore, an urgent need to identify general conservation principles that would allow the most effective habitat protection in the face of limited knowledge about particular ecosystems (Currie 2019; Arroyo-Rodríguez et al. 2020; Fahrig et al. 2022).
One of the most widely applied such general principles for biodiversity conservation, first proposed by Diamond (1975), is the SL > SS principle that a single large patch (or a few large patches) has higher conservation value than several small patches. Although the SL > SS principle seems intuitive, both historical (Simberloff & Abele 1976; Quinn & Harrison 1988) and recent (Deane et al. 2020; Fahrig 2020; Hammill & Clements 2020; Riva & Fahrig 2022) evidence support the opposite SS > SL pattern, that a set of many small patches usually harbors more species than a set of few large patches totaling the same area. In addition, many studies suggest high potential of small patches for conservation of rare species, provision of ecosystem services, and optimal reserve design (Bennett & Arcese 2013; Tulloch et al.2016; Hunter et al. 2017; Deane & He 2018; Wintle et al.2019; Valdés et al. 2020; Yan et al. 2021). Nevertheless, the idea that habitat in large patches is disproportionately important for biodiversity is strongly entrenched in conservation research and practice (Wintle et al. 2019; Fahrig et al. 2022).
Persistence of the SL > SS principle over more than four decades despite evidence to the contrary is largely due to the assumption that the SL > SS principle is valid for species of conservation concern, many of which are habitat specialists that require large expanses of continuous habitat. The SL > SS inference is also expected to hold when the matrix separating habitat patches is hostile to those species. Nevertheless, recent tests do not support these ideas. Species richness of habitat specialists and of threatened or declining species is generally higher across several small than few large patches, and the direction of this relationship does not change when the matrix is hostile (Fahrig 2017, 2020; Riva & Fahrig 2022). Thus, it seems reasonable to ask – should the SL > SS principle be reversed ? In other words, when a choice must be made, should the default principle be to protect many small patches, i.e., SS > SL? The evidence to date supports such a shift, but science is conservative, erring on the side of the status quo unless and until the evidence is overwhelming.
Here we bring the evidence to that tipping point by adding analyses of 76 metacommunities, including 4401 species and 1190 patches, from the database compiled by Chase et al. (2019). Implementing a novel resampling algorithm, we synthesize the largest unbiased (rarefied) metacommunity dataset generated to date for the “SLOSS” question: for the same total area, is biodiversity higher in several small patches or in one or a few large patches? Based on the results, we propose a new principle for biodiversity conservation – for a given total area, protecting the largest possible number of patches, including very small ones, will maximize biodiversity for most taxa . This is opposite to current strategies that typically target for protection patches orders of magnitude larger than the smaller patches in the database we analyzed. Immediate recognition of the value of small patches is therefore fundamental to global biodiversity conservation.