Advancing the consideration of ecological connectivity in environmental assessment: Synthesis and next steps forward

1. Landscape change, biodiversity, and the role of ecological connectivity

The COVID-19 pandemic has triggered reflection and debate about an urgent need to re-evaluate how humans relate to the natural world, particularly biodiversity, and how environmental assessment (EA) can be improved to prevent further biodiversity decline (Gannon 2021; Figueiredo Gallardo et al. 2022). This process requires critical inquiry and interdisciplinary scholarship. Ecological connectivity is a key element for safeguarding biodiversity and an essential component of climate change adaptation (Heller and Zavaleta 2009; Samways and Pryke 2016; Timpane-Padgham et al. 2017). Connectivity refers to ‘the degree to which the landscape facilitates or impedes movement among resource patches’ (Taylor et al. 1993: 571) or ‘the ease with which these individuals can move about within the landscape’ (Kindlmann and Burel 2008: 880), but it also includes the movement or flow of abiotic factors such as nutrients and water. Connectivity is an important condition for maintaining ‘the unimpeded movement of species and the flow of natural processes that sustain life on Earth’ (CMS 2020). While habitat loss remains a primary threat to biodiversity, connectivity needs to be maintained within and between habitats in terrestrial and aquatic systems to allow for dispersal, migration, re-colonization, prevention of inbreeding, and the maintenance of many other ecological processes. Furthermore, as climate change forces species to change their distribution in response to shifting environmental conditions within their traditional range (Parmesan 2006), maintaining connectivity could be the linchpin for the persistence of many wildlife populations.

The practice of EA emerged in the 1970s to evaluate the potential environmental consequences of proposed projects and inform decision-making (Noble 2020). The term encompasses environmental impact assessment (EIA, which includes cumulative effect assessment (CEA)) and strategic environmental assessment (SEA). An EIA examines the consequences of a planned project with the purpose of minimizing negative impacts, while SEA assesses impacts resulting from policies, plans, and programs and is used at a higher tier of decision-making than project-level EIA. The practical requirements for maintaining or restoring ecological connectivity within and between protected areas have become a central theme of conservation planning around the world (Hanson et al. 2022). In the past, connectivity has largely been neglected in EA (Gontier et al. 2006), even for projects known to have serious impacts on connectivity, such as linear infrastructure (Laurance et al. 2014), pipelines (Richardson et al. 2017), and hydroelectric developments (Anderson et al. 2018), thereby contributing to increased fragmentation (Figure 1), degraded ecosystems, and the disruption of biological processes (Keyghobadi 2007; Perkin and Gido 2012). Beyond terrestrial environments, human interventions in marine and coastal ecosystems, even when occurring in limited areas, can also affect species dispersal and the flow of nutrients leading to far-reaching impacts on recruitment, genetic diversity, and adaptive capacity (Bishop et al. 2017; Jonsson et al. 2021). However, as in terrestrial systems, seascape connectivity is often neglected in EA (Manel et al. 2019; Jonsson et al. 2021).

Figure 1. The cumulative effects of development projects lead to landscape change, habitat loss and fragmentation, with important consequences for many wildlife populations. The densities of many species decline when their habitats shrink and landscapes become increasingly fragmented, which can result in local extirpations and shifts in species composition. Watercolor illustrations credit: Charla Patterson.

As biodiversity continues to decline globally, the protection of connectivity in terrestrial, aquatic, and marine ecosystems needs to become an important consideration in EA. This special issue is an outcome of a session on ‘Prioritizing landscape connectivity in environmental impact assessment’ at the 2021 International Conference of the IAIA. While many studies on connectivity have been published in recent years, when preparing this session, it became apparent that there is a paucity of studies about ecological connectivity that explicitly link to EA practice. Thus, we decided to assemble a special issue in collaboration with presenters at our IAIA conference session and new contributors.

2. Objectives of this special issue

The main goal of this special issue of Impact Assessment and Project Appraisal (IAPA) is to compile and catalyze emerging areas of research related to the measurement and consideration of ecological connectivity in the EA process. Specifically, we aimed to (1) highlight the need for studies that link EA and connectivity; (2) learn about current challenges and compile approaches for improving the consideration of connectivity in EA; and (3) foster interdisciplinary and cross-sectoral work that evaluates and advances current practices of connectivity consideration. The special issue is released in two parts, including the current issue and the subsequent issue of IAPA to be released in early 2023.

3. Overview about the content of this special issue

The special issue consists of 9 papers that present findings with global to local scope. They include specific studies from 6 countries: Brazil, Canada, France, Spain, United Kingdom (UK), and Sweden. At the beginning of the special issue, Patterson et al. (2022b) present the results of a global survey about current practices and common issues regarding the treatment of connectivity in EA. This paper compiles the experiences of 134 practitioners, regulators, consultants, researchers, and interest groups from all inhabited continents and sets the scene for identifying key obstacles to advancing the consideration of connectivity and for proposing ways forward. The following four contributions present national overviews about the analysis and consideration of connectivity in urban planning and road planning (Karlsson and Bodin 2022; Oliveira Gonçalves et al. 2022) and in the EA process (Kor et al. 2022; Patterson et al. 2022a). Karlsson and Bodin (2022) reviewed 21 connectivity analysis reports for the assessment of environmental impacts in urban planning in Sweden. Their findings imply that an increased use of quantitative methods needs to be combined with improved guidelines and standards and a continuous knowledge exchange between scientists and practitioners. The letter by Oliveira Gonçalves et al. (2022) discusses various ways to overcome the current gaps and obstacles in the consideration of connectivity loss in road EIAs in Brazil and worldwide and suggest several approaches to improve mitigation. Patterson et al. (2022a) examine the extent to which connectivity has been considered in EA in Canada through a systematic evaluation of 14 environmental impact statements (EIS), whereas Kor et al. (2022) determine how often quantitative approaches are used for the assessment of connectivity or fragmentation in EIAs from the UK. The special issue then zooms into particular cases to illustrate different ways of considering connectivity in EA (Cumming and Tavares 2022; Kor et al. 2022) and potential mitigation measures (Clevenot et al. 2022; Sales Rosa et al. 2023). Kor et al. (2022) introduce a novel quantitative method to model habitat connectivity in EIA that was applied in the Heathrow Third Runway Expansion Project in south-east England. Cumming and Tavares (2022) demonstrate the potential of a multi-tiered approach for assessing connectivity, which facilitates collaboration and analysis at scales relevant for conserving and restoring connectivity, and illustrate the implementation of this approach in national parks in Canada. Clevenot et al. (2022) examine the potential contribution of highway stormwater ponds, a common mitigation measure of road construction, to the connectivity of amphibian populations along a highway in France. Sales Rosa et al. (2023) highlight the relevance of connectivity analyses for planning mitigation and compensation measures such as biodiversity offsets. The authors model and compare current and future connectivity (with restoration offsets implemented) for large mammals in a mining region in the Atlantic rainforest in Brazil. The special issue concludes with a comparison and synthesis of a set of case studies (Patterson et al. 2023). Some of them are also published in more detail as papers in this special issue, while others were contributed by expert practitioners for this synthesis paper. Altogether, these findings reveal how connectivity analysis can be improved and better integrated in EA in the future.

4. Synthesis of the findings of the special issue

4.1 Does ecological connectivity have a place in EA?

Both the opinions of EA practitioners about the treatment of connectivity in EA (Patterson et al. 2022b) and the authors contributing to this special issue indicated a relatively high awareness, but fairly poor treatment of connectivity in EA (Figure 2). For example, the use of connectivity analysis in EA is almost exclusively restricted to transport infrastructure projects, and is rarely applied in other sectors, such as mining (Clevenot et al. 2022; Kor et al. 2022; Oliveira Gonçalves et al. 2022; Patterson et al. 2022a, 2022b). In Sweden, Karlsson and Bodin (2022) document an increase in the awareness and acceptance of connectivity among practitioners and decision makers over the last decade. However, this contrasts with a poor integration of connectivity in the practice of EA (Karlsson and Bodin 2022), as shown also in earlier publications from Sweden and the UK (Karlson et al. 2014). In Canada, Patterson et al. (2022a) conclude that connectivity is largely absent from EIAs, and even projects that attempted to consider connectivity lacked the rigor needed to effectively assess impacts on connectivity as required for evidence-based decision making. Likewise, Oliveira Gonçalves et al. (2022) highlight the poor consideration of connectivity loss in EIAs of road projects globally and specifically in the Brazilian context. There remains a significant gap between a rapidly expanding research field with increasingly sophisticated methods (CCSG 2022; Hanson et al. 2022) and the measurement and consideration of connectivity in EA (Scolozzi and Geneletti 2012 and papers in this special issue).

Figure 2. Synthesis of the contributions to this special issue: Challenges, opportunities, and systemic changes needed for adequate consideration of ecological connectivity in environmental assessment. The first column identifies current challenges and gaps in the EA process. The second column indicates opportunities for better integration of connectivity into EA practice as reported in the studies included in this special issue. The third column summarizes identified needs for systemic changes and further research efforts.

4.2 Factors that influence the consideration of ecological connectivity in EA

4.2.1 How connectivity is measured and assessed

EISs usually lack proper quantitative assessments and predictions of connectivity loss (Jaeger and Torres 2021; Patterson et al. 2022b). Although quantitative assessments are strongly recommended over qualitative assessments, qualitative assessments still predominate in connectivity assessments, even when fragmentation is mentioned as an impact (Kor et al. 2022). The availability of methods to measure connectivity has grown in recent decades (Hilty et al. 2020; CCSG 2022; Hanson et al. 2022) and they are increasingly considered in some countries such as Sweden for urban planning thanks to the development of specific tools and methodologies (Karlsson and Bodin 2022), possibly signaling a transition toward increased future use of quantitative approaches. For instance, a graph-theory approach was used by Kor et al. (2022) to assess the connectivity impacts of the Heathrow Third Runway Expansion Project for grass snakes (Natrix ustinis) and soprano pipistrelles (Pipistrellus pygmaeusthe) and by Clevenot et al. (2022) to model the ecological networks of several amphibian species along a highway in France.

Preparing quantitative assessments comes with challenges, such as the selection of target species and the need for multi-species approaches that cover a range of species with different space and habitat requirements and dispersal patterns (Oliveira Gonçalves et al. 2022). For example, Kor et al. (2022) selected species based on legislation about protected species, availability of baseline data, and expert consultation. Moreover, improved data, modelling, and research about thresholds are needed (Cumming and Tavares 2022).

Another important challenge is the choice of appropriate scales for assessment and mitigation. Fragmentation is a landscape-scale process (Fahrig et al. 2019), yet landscape-scale effects are regularly neglected in EIAs of individual projects, which leads to failure at assessing population-level implications and the adequate extent of application of mitigation measures (Karlson et al. 2014; Tarabon et al. 2019; Bergès et al. 2020; Harker et al. 2021). It is crucial to understand and maintain connectivity at a variety of spatial scales that reflect the movement needs of different species (Hilty et al. 2020; Jaeger and Torres 2021). Clevenot et al. (2022) study regional and local connectivity and show that these can differ strongly, and they conclude that assessment of ‘the contribution of these ponds requires a multi-scale analysis that considers the ecological network as a whole, i.e. at a regional scale, and the location of each pond within the local network.’ Cumming and Tavares (2022) demonstrate the advantages of a multi-tiered approach to impact assessments, which combines SEAs and project EIAs to facilitate the analysis of connectivity in national parks and collaboration with partners at scales required to conserve connectivity within and beyond the boundaries of national parks.

4.2.2 Lack of guidelines, best practices, and standards

Experts point out that the use of sophisticated quantitative methods will not automatically lead to more sustainable outcomes (Karlsson and Bodin 2022). It must be supported by guidelines and a continuous science-practice knowledge exchange. Without adequate guidance, connectivity analysis in EA is conducted in many different ways, with considerable variation in quality (Patterson et al. 2022b). While guidelines for road projects from regions like Europe have recommended considering connectivity in EIA for many years (Iuell et al. 2003) and mention the need to maintain ecological corridors (EEA and FOEN 2011), these older guidance documents are not specific enough and do not include up-to-date scientific methods for connectivity analysis. All papers in this special issue echo an urgent call for best practices, guidelines, and reliable standards and continuous science-practice knowledge exchange to improve the treatment of ecological connectivity in the EA process, including (1) identifying in which cases a connectivity analysis is needed, (2) at what scales of analysis, (3) with what kind of information, and (4) what methods are most appropriate.

4.2.3 EA legislation and regulation

Achieving the biodiversity targets, including those for connectivity, of the post-2020 Global Biodiversity Framework will be contingent on the willingness of governments to establish biodiversity as a priority in decision-making processes regarding development projects via legislation (Laurance et al. 2014; Richardson et al. 2017; Anderson et al. 2018; Sánchez-Cuervo et al. 2020). Legislation pertaining to connectivity was rare at national and sub-national levels up until recently (Lausche et al. 2013). Yet, as connectivity gains traction as a priority for protecting ecosystem resilience and biodiversity, objectives for connectivity conservation are increasingly prevalent in policy initiatives (Hilty et al. 2020). For instance, in the United States, connectivity has been integrated into policy at several levels of government ranging from federal to county levels (Breuer et al. 2020).

Including connectivity in EA legislation would help address the lack of policies, standards, and assessment guidelines (Patterson et al. 2022b). At the regional level, non-profit organizations can play an instrumental role in advocating for better consideration of connectivity in legislation by offering training and support for stakeholders in all sectors, including government. For example, in Quebec, l’Initiative québécoise Corridors écologiques was established by the Nature Conservancy of Canada to facilitate the consideration of connectivity by regional and municipal authorities when making land-use decisions. Following strategic planning meetings between 2010 and 2012, the municipality of Austin became the first in the province of Quebec to include connectivity in their municipal regulations, including bylaws affecting zoning and residential development (see https://ecologicalconnectivity.com/ for additional examples).

4.2.4 Political will

Practitioners and authors of this special issue emphasize that the level of engagement with local, state, or national agencies and their support towards considering connectivity depends largely on the presence of interested people, rather than an institutionalized approach to environmental matters. However, biodiversity-inclusive mandates are often a precondition for interested people working at these agencies to contribute efforts during work hours on these issues, becoming key to ensure success (van der Ree et al. 2011). A promising avenue to strengthen the treatment of connectivity in EA is to encourage governments to articulate a vision and high-level strategy for landscape conservation that can be used to inform EA, thereby ensuring that decision-making processes respect sustainable development goals, biodiversity commitments, and climate-change targets. Patterson et al. (2022b) propose a 4-point strategy that entails the creation of guidance, the inclusion of connectivity in legislation and regulation along with a delineation of the roles and responsibilities of participants in the EA process, the development of meaningful CEAs at landscape-level scales, and the integration of connectivity assessment in national and subnational biodiversity and land-use planning processes.

4.3 Determination of appropriate mitigation for fragmentation effects

A substantial share of the literature on connectivity is focused on evaluating the use and effectiveness of mitigation measures to restore wildlife movement opportunities. However, once projects are established, restoring the connectivity that was lost or diminished is often not possible (van der Ree et al. 2011). This is especially true for threatened and rare ecosystems as well as biodiversity hotspots, which have unique biological characteristics and support large numbers of endemic species. Authors of this special issue strongly recommend the application of the mitigation hierarchy, including the adoption of no-net-loss or net-gain targets, and adherence to the principles of adaptive management to increase the effectiveness of mitigation measures (Oliveira Gonçalves et al. 2022). Early consideration of connectivity in the mitigation hierarchy in project planning, in particular impact avoidance, should be critical for project approval. Also in this special issue, Clevenot et al. (2022) examine the potential of stormwater ponds created in road construction to contribute to the connectivity of amphibian populations and raise questions about the capacity of artificial habitats to serve as a refuge for biodiversity.

5. Conclusions and steps ahead

5.1 Joining the science of connectivity and the practice of EA

The research presented in this issue reflects a growing awareness and interest in integrating connectivity analysis in EA. However, studies on connectivity that explicitly make a link to the practice of EA are still rare, and overall, the treatment of connectivity has remained poor. Connectivity effects of new projects remain mostly unquantified at the landscape level. The absence of proper connectivity assessments from the EA process subverts opportunities to assess and mitigate the severity and extent of project impacts, thereby vastly increasing the likelihood that approved projects will contribute to cumulative impacts on connectivity, to biodiversity loss, and to the erosion of ecosystem structure and function at multiple spatial scales. These impacts can cascade into social and economic systems and decrease the well-being of local communities.

Papers in this special issue identify key directions for addressing existing challenges for the consideration of ecological connectivity in the EA process. A feasible next step identified by practitioners (Patterson et al. 2022b) and contributors to this issue is the development of detailed guidance to improve the treatment of ecological connectivity in the EA process, as described in the previous section. However, there is also an urgent need to tackle ambitious systemic changes affecting not only the consideration of ecological connectivity but the EA process more generally (Figure 2).

5.2 Advancing evidence-based EA and making monitoring data accessible

Despite decades of research on connectivity and significant advances in EA practice, there remain severe gaps in the understanding of how to measure and mitigate many impacts on ecosystems (Christie et al. 2020). Closing these gaps requires continued research funding and adopting ways to ensure the uptake of evidence-based knowledge by EA practice. For this purpose, the data collected during the monitoring phase are instrumental to increase the empirical knowledge about impacts and the effectiveness of mitigation measures. Four decades ago, ecologists Hilborn and Walters (1981: 274–275) had already emphasized the need of EA to perform ‘exhaustive surveys of past actions’ in the form of post-development retrospective studies and follow-ups, because ‘we learn by experience, but we fail to document the most useful of all experience, our failures’ and ‘the information from the experience is lost’. Very little progress has been made in this regard. Thus, we argue strongly for publicly accessible reporting of monitoring outcomes according to FAIR (Findable, Accessible, Interoperable and Reusable; Wilkinson et al. 2016) and open practices. Such reporting should be coupled with regulatory tools that promote capacity building to access, apply, and contribute to the latest available knowledge to make evidence-based decisions and will require cooperation between scientists and participants in the EA process (Oliveira Gonçalves et al. 2022).

5.3 Establishing limits to landscape fragmentation

Connectivity loss is one of the most widely recognized effects of human development and thus, targets and limits to control landscape fragmentation are needed (Jaeger and Torres 2021). Targets and limits provide a regulatory ground for administrative action for curtailing fragmentation when the targets are exceeded. For performance review, EAs and national or regional environmental monitoring systems should routinely report on the degree of connectivity to detect fragmentation and stop or reverse undesirable trends. For example, the Biodiversity Monitoring Switzerland (BDM) and the Swiss Monitoring System for Sustainable Development (MONET) use robust landscape fragmentation metrics, and the German Environment Agency has proposed to establish limits to the rate of increase of landscape fragmentation (Penn-Bressel 2005), with the remaining large unfragmented areas being preserved and enlarged where possible.

5.4 Cumulative effects assessment and the shifting baseline syndrome

The incremental loss of habitat and erosion of connectivity over the course of years and decades – bit by bit – caused by the cumulative nature of development projects is cause of major concern. Already four decades ago, ecologist William Odum (1982) observed that ‘much of the current confusion and distress surrounding environmental issues can be traced to decisions that were never consciously made, but simply resulted from a series of small decisions.’ A fairly common, but erroneous argument in EA contends that an ecosystem or resource that is degraded already by some factor would require less concern about subsequent potential impacts. Peterson (1993) called it the ‘Fallacy of stressed systems’. He characterized it as ‘a “passing-the-buck” type of argument, that does not serve the public interest’ (Peterson 1993: 36). On the contrary, he emphasized a greater need for careful identification and consideration of any and all potential additional impacts to allow for adequate protection of degraded ecosystems and their need for restoration. Peterson pointed out an even greater challenge to ecologists to predict effects on wildlife populations that are already under pressure from existing impacts due to the possibility that the impacts might interact in some complex ways.

Using the example of declining caribou populations in Canada, Collard et al. (2020) demonstrate a fundamental contradiction between proliferating legislation to stop wildlife declines and continued government approvals of major resource extraction projects with potential impacts on endangered species like caribou to proceed. Moreover, the approvals include highly questionable mitigation measures such as wolf culls that may be able to keep the populations in a state of suspension for some time (Johnson et al. 2022), rather than addressing habitat degradation and destruction – despite the established need and promise of caribou habitat restoration according to the Canadian Species-at-risk-Act and elaborated caribou recovery plans. Collard and Dempsey (2022) identify the lack of meaningful CEA and the neglect of impacts from earlier projects – a ‘resetting of time’ to current conditions, just prior to project implementation – as the main mechanisms by which the need for habitat restoration is circumvented on a regular basis. Collard and Dempsey (2022: 1557) point out that ‘this is effectively the opposite of cumulative effects assessment’. Setting the baseline in an already-degraded time and place that licenses further degradation, because it is then seen as ‘insignificant’, is a prime example of the shifting baseline syndrome. More generally, the shifting baseline syndrome describes a gradual lowering of the accepted norms for the condition of the natural environment due to lack of experience or knowledge of past conditions (Pauly 1995; Thérivel et al. 2021). It leads to an increased tolerance for progressive environmental degradation and the use of inappropriate baselines for nature conservation and management (Mihoub et al. 2017; Soga and Gaston 2018). The consideration of connectivity in EA is highly susceptible to the shifting baseline syndrome. An increased quantification and consideration of connectivity may seemingly add scientific credibility to the EA process, but this does not necessarily lead to improved consideration of connectivity, as Karlsson and Bodin (2022) point out very clearly. Indeed, connectivity consideration will remain ineffective if baselines are shifted time and again, resulting in the destruction of the remaining potential for recovery, a false sense of success of mitigation measures, in deceiving promises of insignificant residual effects, and in the illusion of a ‘future eco-perfect development’ (Collard and Dempsey 2022: 1558) that supposedly would harmonize diverse and abundant wildlife populations with continued resource extraction and economic growth.

In addition, there is a significant danger that proponents might increasingly argue that a certain amount of habitat loss would be acceptable as long as enough connectivity is maintained to keep populations connected. However, habitat loss remains in fact the most important threat to biodiversity. All habitat patches of all sizes are important for biodiversity, including small and isolated habitat patches (Fahrig et al. 2019; Wintle et al. 2019), and all habitat loss contributes to reductions in connectivity as it includes both within-patch and between-patch connectivity (Spanowicz and Jaeger 2019). Assessments must still recognize the importance of avoiding habitat loss. Therefore, connectivity needs to be considered as an additional priority along with habitat amount, but not instead of habitat amount or to compensate for habitat loss. In other words, the consideration of connectivity should not be misused as an argument for the destruction of small habitat patches (Fahrig et al. 2019).

This special issue demonstrates that the practice of EA – and CEA and SEA in particular – is confronted with a number of obstacles and important unanswered questions and uncertainties. We urge practitioners and decision-makers to carefully consider these challenges.

We hope this special issue will help improve the treatment of ecological connectivity in the EA process and will inspire progress by catalyzing findings and emerging areas of research related to the measurement and integration of connectivity in the practice of EA. Similar to road ecology moving toward larger scales (van der Ree et al. 2011; Jaeger and Torres 2021), EA will need to move toward larger spatial and temporal scales to get it right.

Acknowledgments

We are grateful to all participants of our IAIA 2021 session about ecological connectivity and to all authors of contributed papers to this special issue for their insightful work, for their encouragement to go ahead with this special issue, and their tenacity and patience during the COVID-19 pandemic. We also warmly thank Dr. Thomas Fischer for his support and coordination of the special issue.

Disclosure statement

The authors declare no potential conflict of interest.

Additional information

Funding

A.T. received funding from the European Union’s Horizon 2020 research and innovation program under the Marie Sklodowska-Curie grant agreement no. 846474.