Using landscape fragmentation thresholds to determine ecological process targets in systematic conservation plans

doi:10.1016/j.biocon.2018.03.025

Biological Conservation

Volume 221, May 2018, Pages 257-260

https://doi.org/10.1016/j.biocon.2018.03.025 Get rights and content

Highlights

•
Systematic conservation planning requires targets for biodiversity representation and persistence (ecological processes).
•
Suitable spatial data with which to set quantitative ecological process targets is mostly absent.
•
We propose using landscape ecology theory, specifically fragmentation thresholds, to set ecological process targets.
•
The percolation theory 60% threshold provides a pragmatic top-down ecological process target for landscapes.
•
This target is useful in landscapes where data on individual process components are not known.

Abstract

Systematic conservation planning requires that quantitative targets be set for both biodiversity pattern and processes. While the challenge of setting quantitative representation targets has been well addressed in the literature, guidelines for conceptualising and setting process targets are lacking. Process targets can be defined as the minimum amount of natural habitat that must remain to ensure the long-term survival of the majority of species. While a representation target may represent the majority of species in a landscape, this target often falls far short of conserving processes necessary for the persistence of these species. This paper explores the potential for landscape ecology research to provide useful insights into developing process targets by relating critical thresholds in habitat amount to the probability of population persistence. It is proposed that these thresholds provide a basis for developing generic top-down ecological process targets in conservation planning. The percolation threshold, theoretically defined at 59%, is increasingly used to inform research into ecological state-shifts and ecosystem resilience. This threshold may provide a basis for developing top-down process targets in instances where comprehensive bottom-up spatial data on individual ecological processes is unavailable. In the context of ongoing global habitat loss, this approach provides a pragmatic, but also potentially biologically meaningful, way of incorporating defensible and quantitative ecological process targets or biodiversity persistence goals into conservation plans.

Introduction

Habitat loss is the primary driver of biodiversity loss on the planet today (Baillie et al., 2004). In this context of ongoing habitat loss, immediate conservation choices have to be made regarding how much and which areas need to be set aside in order to conserve representative and persistent examples of the planet's biodiversity. Such choices should be based on sound scientific evidence and reasoning, but data to support this is often unavailable. Thus, a careful balance is needed between scientific certainty and pragmatic decision-making based on sound ecological logic. The well-established concept of Systematic Conservation Planning has been used for many years to assist in making a range of conservation decisions in a spatially explicit manner (Pressey et al., 2007). Conservation planning is “the process of locating, configuring, implementing and maintaining areas that are managed to promote the persistence of biodiversity and other natural values” (Pressey et al., 2007). Systematic conservation planning is based on two underlying principles: 1) the principle of representation aims to conserve a sufficient sample of the full variety of biodiversity, while 2) the principle of persistence aims to conserve the necessary ecological and evolutionary processes that allow biodiversity to persist over time (Margules and Pressey, 2000). In order to achieve these goals, systematic conservation planning requires that quantitative targets be set for both biodiversity pattern (representation) and ecological processes (persistence).

Despite acknowledged limitations of quantitative conservation targets (Carwardine et al., 2009; Pressey et al., 2003), they remain common practice for conservation in general (Aichi biodiversity targets; CBD, 2010) and systematic conservation planning in particular (Carwardine et al., 2009). Representation targets for species and ecosystems have generally been well-researched (Pressey et al., 2007) and standard methods for determining these targets are becoming established (Rondinini and Chiozza, 2010; e.g. Desmet and Cowling, 2004). However, these representation targets will be ineffective in the long-term unless the ecological processes that maintain these patterns are also conserved. Unfortunately, methods for setting quantitative targets for ecological process are still unresolved. Conserving ecological processes means not only conserving the area where a species currently occurs, but also sufficient of its habitat so that it will continue to survive, now and into the future. From a conservation planning perspective, this involves determining the minimum amount and configuration of area required to conserve these ecological processes.

The most widely used approach to incorporating ecological processes into systematic conservation plans is to map and set targets for individual elements of ecological processes (such as corridors, ecotones, climate refugia, migration routes etc.) (e.g. Pressey et al., 2003). However, this comprehensive, ‘bottom-up’ approach requires not only that the relevant ecological processes can be identified and understood, but also that empirical data be available on the spatial components of these processes (Pressey et al., 2007). The reality in most regions of the world is that this type of data is non-existent. Given the time scales and budgets under which conservation planning operates, it is unlikely that the required research will be achieved soon. This lack of data fundamentally limits the incorporation of individual process elements into systematic conservation plans.

This paper presents an alternative approach for incorporating ecological processes into conservation planning. It is proposed that, based on current ecological understanding, it is possible to use a generalised ‘top-down’ target for a whole landscape that will define the minimum amount of natural habitat required to secure a suite of ecological processes. This paper explores the potential for landscape ecology to provide useful insights into developing ecological process targets by relating the amount and structure of remaining habitat to critical thresholds in the probability of population persistence.

Section snippets

Ecological processes and landscape ecology

Ecological processes include both biological (e.g. survival, reproduction, dispersal and interaction) and abiotic (e.g. climate, geomorphological, pedological and hydrological) processes. Landscape ecology is a field of study that focuses on the interactions between spatial pattern and ecological processes (Mayer et al., 2016; Turner, 2005). It investigates how the spatial configuration of a landscape influences the populations and community dynamics of organisms (Collinge, 2001; Turner, 2005).

Fragmentation and critical thresholds

Most fragmentation studies agree that the overall amount of remaining habitat accounts for almost all of the variation in observed population size in fragmented landscapes (Fahrig, 2013; Flather and Bevers, 2002; Villard and Metzger, 2014). The relationship between habitat loss and species persistence is generally negative, with a higher rate of species persistence where the amount of habitat remains high, and a lower rate of persistence where a large amount of habitat has been lost (Fahrig,

Relating critical thresholds to conservation targets

Pragmatic conservation planning requires that generalisations be made across species, ecosystems and ecological processes. If data were not limiting it would be possible to estimate the persistence target for all organisms in a landscape based on an understanding of their meta-population dynamics and habitat requirements. However, high-levels of scientific uncertainty and lack of available data mean that, at present, conservation plans are unable to explicitly include a full range of ecological

Application of a landscape persistence target in conservation planning

Quantitative targets for representing biodiversity pattern and maintaining ecological processes are a foundation underpinning the success and defensibility of applied systematic conservation planning. These targets can be interpreted as quantitative indicators or thresholds of the minimum area required to achieve our conservation goals. While representation targets are fairly well researched, there is very little information available to guide practitioners in setting targets for persistence.

Acknowledgements and declaration

Emily A. Botts was compensated by the author for writing and editing support in preparing this paper for publication. The author gratefully acknowledges the inputs and comments received from Richard Cowling, Mandy Driver, Sumeshni Pillay, Lilly Bovim and Genevieve Pence.

Funding sources

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

References (38)

S.K. Collinge
Spatial ecology and biological conservation
Biol. Conserv.
(2001)
L. Fahrig
How much habitat is enough?
Biol. Conserv.
(2001)
A.J. Huggett
The concept and utility of ‘ecological thresholds’ in biodiversity conservation
Biol. Conserv.
(2005)
M. Parker et al.
Habitat loss and the habitat fragmentation threshold: an experimental evaluation of impacts on richness and total abundances using grassland invertebrates
Biol. Conserv.
(2002)
R.L. Pressey et al.
Formulating conservation targets for biodiversity pattern and process in the Cape Floristic Region, South Africa
Biol. Conserv.
(2003)
R.L. Pressey et al.
Conservation planning in a changing world
Trends Ecol. Evol.
(2007)
C. Rondinini et al.
Quantitative methods for defining percentage area targets for habitat types in conservation planning
Biol. Conserv.
(2010)
L.K. Svancara et al.
Assessing the landscape context and conversion risk of protected areas using satellite data products
Remote Sens. Environ.
(2009)
S.R. Abades et al.
Fire, percolation thresholds and the savanna forest transition: a neutral model approach
J. Ecol.
(2014)
C.R. Allen et al.
REVIEW: quantifying spatial resilience
J. Appl. Ecol.
(2016)

H. Andrén

Effects of habitat fragmentation on birds and mammals in landscapes with different proportions of suitable habitat - a review

Oikos

(1994)

A.D. Barnosky et al.

Approaching a state shift in Earth's biosphere

Nature

(2012)

J. Carwardine et al.

Hitting the target and missing the point: target-based conservation planning in context

Conserv. Lett.

(2009)

CBD

P. Desmet et al.

Using the species-area relationship to set baseline targets for conservation

Ecol. Soc.

(2004)

R.M. Ewers et al.

Confounding factors in the detection of species responses to habitat fragmentation

Biol. Rev.

(2006)

L. Fahrig

Effect of habitat fragmentation on the extinction threshold: a synthesis

Ecol. Appl.

(2002)

Cited by (19)

A development of intergenerational sustainability indicators and thresholds for mobility system provisioning: A socio-ecological framework in the context of strong sustainability
2023, Environmental and Sustainability Indicators
With the concurrent challenges of looming ecological crisis and growing global social inequity, the need to assess the socio-ecological efficiency of provisioning systems to assess progress towards a ‘safe and just space’ has been highlighted in the literature. Mobility systems represent good examples of provisioning systems required to gain access to essential services and for human connectivity. They also act, however, as a significant source of environmental degradation and are subject to significant social inequalities associated with transport accessibility, affordability, and social externalities, such as traffic fatalities and local air pollution. Sustainable mobility indicator and ‘safe and just space’ studies have thus far struggled to capture a strong state of sustainability (as opposed to relative improvement measurements), at a sectoral depth that allows for actionable insight, respectively. This work thus reviewed both literature fields, collecting mobility indicators as well as sectoral ecological and social thresholds, and organized them according to a developed strong sustainability framework. Through this approach, this work accomplishes three goals: 1) assists in characterizing a ‘safe and just space’ for mobility provisioning systems through the development of an indicator system 2) Provides a layered understanding of their use and suggests future use of them 3) Provides an organized basket of indicators for practitioners to make use of. In achieving these goals, this work helps connect social and ecological thresholds for the transport sector at a granular level, assisting practitioners in applying and measuring strong sustainability states for mobility provisioning.
Alternative biome states of African terrestrial vegetation and the potential drivers: A continental-scale study
2021, Science of the Total Environment
The alternative stable state (ASS) theory provides a plausible framework to explain the spatial distribution of biomes and their dynamics. Existing studies to test the alternative biome states (ABSs) mainly focused on tree-dominated biomes. It is still uncertain whether ABSs are present in a wide range of terrestrial biomes. This study was to examine the ABSs in the terrestrial vegetated areas of Africa and the maintaining factors. The potential landscapes were reconstructed separately for forest, grassland and shrubland using the MODIS Vegetation Continuous Field data along the gradients of temperature, precipitation and aridity index (AI, the ratio of precipitation to potential evapotranspiration). The differences of soil organic carbon density (SOCD), fire count and grazing intensity were compared to test the feedback hypothesis to maintain the ABSs. The results showed that AI performed well in detecting the ABSs at the continental scale of Africa. Forest (at the wetter end) and shrubland (at the drier end) were well separated along the AI axis. Forest had three stable states (i.e. closed forest, woody savanna, and savanna) and shrubland had two stable states (i.e. closed shrubland and open shrubland). Grassland had two stable states (i.e. dense grassland and sparse grassland) distributing in a large AI range. The stable states that shared a specific AI range were regarded as the ABSs. Climate aridity greatly determined the distribution of the ABSs but the positive feedbacks between vegetation and SOCD, fire count, and livestock density played potential roles in driving the shifts between the ABSs. Our study indicated that the ABSs commonly existed in varied biomes (both tree-dominated and non-tree-dominated) in the African continent, which provided an enlarged picture of the ABSs of the terrestrial biomes. The findings contribute to a deeper understanding of large scale vegetation patterns and their dynamics and facilitate to macro management of the terrestrial ecosystems in facing the possible regime shifts of the biomes.
Do ecological restoration programs reduce forest fragmentation? Case study of the Three Gorges Reservoir Area, China
2021, Ecological Engineering
Citation Excerpt :
We found that the power-law behaviour of the distribution of core area proportions becomes a positive skewness distribution when forest cover exceeds 30%. The extinction threshold, known in the literature as the 20% rule, usually occurs between 10% and 30% of the original habitat area remains (Desmet, 2018). The presence of power-law behaviour implies that there are only a few core areas in a given space, so the deviation from power-law behaviour reflects that a state of the system away from extinction.
Ecological restoration programs (ERPs) have greatly expanded forested areas in China to mitigate environmental problems. Unfortunately, this general increase in forest cover has not revealed clear changes in the states of fragmentation and loss of connectivity of forested habitats, and few studies have investigated the effects of ERPs on the spatial patterns of forests. Taking the Three Gorges Reservoir Area as a case study, we assess and quantify how the dynamics of the spatial patterns of forest following massive ERPs from 1990 through 2015 by morphological spatial pattern analysis (MSPA). We examine possible landscape structure thresholds in the forest restoration gradient. The results indicate that dramatic decreases in forest fragmentation and patchiness have occurred since ERP implementation. Among the seven MSPA forest types, core forests show the highest increase of 513.91 km²·yr⁻¹. ERPs promote the defragmentation of forests via two pathways: (1) Old forests are preserved and expanded by building new forests around existing patches and closing perforations within forests; and (2) new core forests are gradually created in areas far from older forests. However, isolated forest patches surrounded by cropland production areas formed after the conversion of croplands into forests have appeared across the study area. Furthermore, at least two points of nonlinear changes in the landscape structure with increasing forest cover are identified. These nonlinear trajectories can be used as indicators of critical shifts in ecosystem states, and relevant thresholds support the identification of specific conservation actions. Our findings demonstrate the importance of ERPs for the defragmentation of forest patches and provide a basis to establish a more effective arrangement of patches necessary to mitigate fragmentation effects. Given that landscape structural changes are not expected in ERP design, future ERPs should focus on remaining isolated forest patches and on maintaining the health and sustainability of landscapes.
Taxonomic and functional threshold responses of vertebrate communities in the Atlantic Forest Hotspot
2021, Biological Conservation
Ecological thresholds are an indicator of rapid and non-linear changes along both natural habitat and land-use gradients. Thus, they can be used to quantify biodiversity responses to human-induced environmental change. We investigated multiple thresholds by analyzing eight environmental predictors of both taxonomic and functional composition of amphibian and bird communities in 47 independent sample units, located within a heterogeneous land use matrix of the megadiverse Atlantic Forest hotspot. We demonstrate that current land use practices alter both the taxonomic and functional composition and promote the establishment of non-native taxa resulting in novel assemblages. Whereas anuran communities experienced dramatic compositional shifts as a result of the transformation of closed forest habitats into Eucalyptus monocultures (at less than 10% land cover), bird communities were mainly altered through the loss of old-growth forest (at 20% forest cover loss). In both groups, observed thresholds were lower than previously reported and much lower than the 20% forest set-aside requirement defined by current land-use legislation. We argue that the strategy of halting forest loss through rapid afforestation with non-native trees has converse and potentially detrimental effects on the conservation of native forest biodiversity. Future studies need to investigate how including novel assemblages in current conservation management strategies can enhance biodiversity protection in human-impacted forest landscapes. The identification of taxon-specific thresholds for both taxonomic and functional community shifts is indispensable when formulating common land use practices and designing mitigation measures. Threshold analyses can guide these actions by providing clear and quantifiable break-points for conservation practitioners.
More than just a (red) list: Over a decade of using South Africa's threatened ecosystems in policy and practice
2020, Biological Conservation
Citation Excerpt :
Vegetation types with less remaining area than required to achieve their conservation targets were considered Critically Endangered (CR), as below this target, species representation would be compromised. Thresholds for Endangered (EN) and Vulnerable (VU) were loosely aligned with theoretical ecosystem fragmentation thresholds (Pierce et al., 2005; Desmet, 2018; Appendix A1). This initial criterion is conceptually equivalent to criterion A3 (historical loss of habitat) in the IUCN RLE framework, but the thresholds, though similar, are ecosystem specific.
One of the stated applications of the IUCN Red List of Ecosystems (RLE) is to influence government policy and decision-making. We share 15 years' experience in integrating an independently developed indicator of ecosystem threat status into government policies and practice. South Africa's ecosystem threat status indicator was conceptualised in the early 2000s and progressed from a project-based indicator to listing of threatened ecosystems in terms of national legislation in 2011. We show the range of applications of the indicator, from its use as a headline indicator in the National Biodiversity Assessment to its role as a direct trigger for Environmental Impact Assessment. The strong link between threatened ecosystems and systematic conservation planning in South Africa also enabled ecosystem threat status to inform multi-sectoral development planning and decision-making. We show how bridging products, data availability, persistent mainstreaming and stakeholder engagement have encouraged the use of the indicator in government policy. The advantages and disadvantages of legislative listing are shared. Sound scientific foundations, combined with pragmatism, have provided a policy-relevant tool for focussing management on threatened ecosystems. We make active recommendations that will facilitate the policy uptake of the IUCN RLE in other countries.
Accounting for spatial patterns in deriving sea-level rise thresholds for salt marsh stability: More than just total areas?
2019, Ecological Indicators
Citation Excerpt :
They have been used in the research of biodiversity and habitat analysis (Tscharntke et al., 2012), water quality, hydrological processes (Yuan et al., 2015), urban road network, aesthetics of landscape and its management, planning and monitoring (Schröder, 2006; Uuemaa et al., 2009). In different landscape patterns, total area of habitats and fragmentation or connectivity of landscapes are particular of interests as habitat loss and fragmentation are among the main causes for changes in biodiversity and distribution of organisms, degradation of ecosystem functions such as decreasing biomass and altering nutrient cycles (Taylor et al., 1993; Andrén and Andren, 1994; Baillie et al., 2004; Wu, 2013; Haddad et al., 2015; Niebuhr et al., 2015; Hu et al., 2016; Wilson et al., 2016; Nichol et al., 2017; Desmet, 2018). In order to improve the use of landscape patterns to illuminate ecological processes, mechanistic models like what was applied here are key (Kareiva and Wennergren, 1995) to enhancing our understanding because these models contain key processes that drive landscape change.
Ecological threshold is an important concept to indicate the boundary of alternate states of ecosystems driven by environmental conditions and to facilitate evaluation of ecosystem resilience. Sea-level rise (SLR) thresholds for the stability of salt marshes, if studied in two dimensions, are generally derived based on total areas without systematic accounting for spatial patterns related to edges, shapes, and contagions of patches. As these spatial patterns are potentially important for functions and ecosystem services of salt marshes and they are likely to be impacted by SLR in a different way from the total areas, it is necessary to study SLR thresholds based on these spatial patterns to obtain a more comprehensive understanding of salt marsh resilience to SLR. This research compares the SLR thresholds based on these spatial patterns of salt marshes to those based on total areas alone across different spatial resolutions.
The spatial patterns of salt marshes were quantified by 26 commonly used landscape metrics, predicted from a mechanistic wetland change model. At spatial resolutions of 2–100 m, SLR thresholds were first derived using individual landscape metrics and then the first principal component that explained >80% of total variance of these metrics showing threshold responses to SLR. In order to separate the effect of spatial configuration from composition, a neutral model which simulated the same amount of salt marsh change as the mechanistic model but at the random locations was applied. The SLR thresholds were derived based on the simulations from the neutral model and compared to those from the mechanistic model.
The results show that total area-based SLR thresholds do not comprehensively represent salt marshes’ resilience to SLR. Particularly, I find 1) the derived SLR thresholds vary from 7.29 to 11.12 mm/yr for 2100 based on landscape metrics used, 2) the SLR threshold based on the first principal components (7.99 mm/yr) is smaller than that based on the total area only (8.40 mm/yr), 3) the SLR thresholds are scale dependent, and 4) the spatial configuration’ effect on SLR thresholds is smaller for smaller salt marsh areas compared to larger salt marsh areas.
This study highlights the need to account for different spatial patterns of salt marshes and apply wetland maps with a spatial resolutions of 30 m or finer in deriving SLR thresholds, as using total areas alone or coarser-resolution maps may provide a biased interpretation that salt marshes are more resilient to SLR than they actually are.

View all citing articles on Scopus

View full text

PerspectiveUsing landscape fragmentation thresholds to determine ecological process targets in systematic conservation plans

Highlights

Abstract

Introduction

Section snippets

Ecological processes and landscape ecology

Fragmentation and critical thresholds

Relating critical thresholds to conservation targets

Application of a landscape persistence target in conservation planning

Acknowledgements and declaration

Funding sources

Biol. Conserv.

Biol. Conserv.

Biol. Conserv.

Biol. Conserv.

Biol. Conserv.

Trends Ecol. Evol.

Biol. Conserv.

Remote Sens. Environ.

Fire, percolation thresholds and the savanna forest transition: a neutral model approach

J. Ecol.

REVIEW: quantifying spatial resilience

J. Appl. Ecol.

Effects of habitat fragmentation on birds and mammals in landscapes with different proportions of suitable habitat - a review

Oikos

Approaching a state shift in Earth's biosphere

Nature

Hitting the target and missing the point: target-based conservation planning in context

Conserv. Lett.

Using the species-area relationship to set baseline targets for conservation

Ecol. Soc.

Confounding factors in the detection of species responses to habitat fragmentation

Biol. Rev.

Effect of habitat fragmentation on the extinction threshold: a synthesis

Ecol. Appl.

Perspective
Using landscape fragmentation thresholds to determine ecological process targets in systematic conservation plans