Global whole landscape connectivity to complement protected area connectivity

Global whole landscape connectivity to complement protected area connectivity | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Global whole landscape connectivity to complement protected area connectivity Erin Poor, Kimberly Hall, Jesse Anderson, Melissa Clark, Aaron Jones, and 11 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6457802/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Protecting and maintaining ecosystem connectivity is crucial for stemming the biodiversity crisis, but current tools and indicators assessing connectivity primarily focus on connectivity among protected areas, not across ecosystems. Here, we develop a map of global connectivity across all lands that accounts for the cumulative impacts of human modification. We then parse this continuous connectivity into categories directly relevant for conservation actions (i.e., spatial planning, restoration, and protection). We find that most global lands (66%) retain high levels of connectivity, particularly in tundra, boreal, and conifer forest biomes, which often fall outside of protected areas. Conversely, 29% of global terrestrial areas have low levels of connectivity due to high human modification, which is prevalent in Asia, Europe, and North America, and in temperate and tropical forested biomes. Our results underscore that focusing only on connectivity of protected areas misses ecologically important areas that currently lack protection and may fail to identify important conservation and restoration areas for connectivity that could benefit ecosystems. This work is broadly applicable to inform global conservation goals, such as those enumerated in the Kunming-Montreal Global Biodiversity Framework (GBF). We provide summaries to inform global, national, and sub-national decision making and demonstrate how this whole-landscape connectivity data can be used to support GBF targets. Earth and environmental sciences/Ecology/Conservation biology Earth and environmental sciences/Ecology/Ecological networks Earth and environmental sciences/Ecology/Ecological modelling Connectivity circuit theory Omniscape human modification conservation planning biodiversity targets Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 INTRODUCTION The biodiversity crisis is fueled by rapid losses in the extent and integrity of ecosystems due to human-driven land modification and climate change 1 – 6 . Ecological connectivity, defined as the “unimpeded movement of species and flow of natural processes” 7 , is critical to ecosystem and species conservation and has been established as a key conservation component in the Convention on Biological Diversity (CBD)’s Kunming-Montreal Global Biodiversity Framework (GBF) 8 – 10 . Yet despite the critical importance of connectivity across ecosystems and landscapes, data and guidance to fully support the implementation of connectivity-related targets within the GBF are lacking or incomplete 8 . Most of the work on developing indicators for monitoring connectivity has focused on protected areas (PA), which has helped to raise awareness of the importance of ecological connectivity 11 . However, addressing the coupled biodiversity and climate crises will require a “whole landscape” approach to connectivity, where all lands, regardless of their protection status and ecological condition, are considered and valued appropriately 12 . Under the GBF, 195 countries aim to meet four long-term biodiversity conservation goals by 2050. The first, Goal A, targets that “the integrity, connectivity, and resilience of all ecosystems are maintained, enhanced, or restored” by 2050. Under Goal A, signatories aim to achieve multiple targets by 2030. Targets 1, 2, and 3 are related to spatial planning, restoration, and protection of ecosystems, and GBF guidance provides multiple indicators for measuring progress towards meeting each target. Target 3 contains particular emphasis on connectivity, stating that PAs and other effective conservation measures (OECMs) should be connected through corridors and “integrated into wider landscapes, seascapes and the ocean”. A PA is “a geographically defined area, which is designated or regulated and managed to achieve specific conservation objectives” 8 . The current set of proposed indicators only provide methods for evaluating linkages between PAs and OECMs 8 , yet, cost-effective actions will require integrated approaches that carefully address all three targets. Despite the focus on PAs in Target 3, there are many unprotected areas that must be considered that are potentially relevant for connectivity 8 . While PAs are a necessary conservation tool, the current global PA network is geographically and ecologically biased toward areas that are inherently difficult to access in steeper, higher, or more remote terrain and are less populated 13 – 16 . Focusing indicators solely on corridors between PAs will result in biased assessments of areas deemed important for connectivity. Instead, the importance of connectivity in whole landscapes, independent of the amount or distribution of PAs they contain, is needed to sustain biodiversity and achieve the ecological representativeness proposed in the GBF targets 8 , 14 , 17 – 19 . Identifying areas of connectivity importance outside of the current PA system can inform where future additions to the PA and OECM network could contribute to conservation goals. For example, Brennan et al. ( 20 ) identified globally important areas for the movement of mammals between PAs, but some areas known to harbor high biodiversity (United States Appalachian region, Madagascar, India, and Southeast Asia 21 ) were identified as having low connectivity owing to a dearth of globally registered PAs in these regions. Despite the need, a gap remains in modeling connectivity across ecosystems beyond specific habitat patches or protected areas, in part driven by computational limitations. This gap may be addressed by a variety of well-vetted theoretical frameworks, models, and tools to identify species corridors that are commonly used for habitat and population management at the landscape scale 22 , 23 . Of these, Omniscape, an omnidirectional implementation of circuit-theory modeling, has shown promise to assess whole landscape connectivity at continental scales. In Omniscape, every pixel, not just PAs or habitat patches, can act as a source for potential species movement 24 , 25 . These models can thus be applied to any land cover map, independent of political or geographic boundary and may thus be more ecologically accurate, as it allows modelers to represent gradients in ecological condition and does not infer that species recognize human-designed political or geographic boundaries. The result is a continuous dataset representing potential movement across a study area, which until now has been hindered by computational requirements that limited the geographic scale at which these models can be run 24 – 26 . The outputs of Omniscape typically represent movement potential of species from habitat, or “source” pixels across landscapes, with varying assumed resistance to movement (Table 1 ; 24 , 27 ). Model outputs represent potential movement across a landscape, where high values represent pinch points along pathways and low values represent barriers to movement or a lack of movement sources 24 . Although high movement areas are often interpreted as corridors and treated as conservation priorities, additional translation is needed to turn continuous outputs into meaningful information for prioritization of conservation efforts. Importantly, moderate connectivity values often characterize highly intact natural landscapes where movement is unconstrained, yet these areas are not usually the focus of connectivity outputs. To ease the use and interpretation of connectivity modeling in policy applications, the continuous model outputs must then be translated into products that support decision-making 28 , 29 . One approach for facilitating interpretation of results is to compare potential movement to a null model with uniform resistance, thereby determining how connectivity changes in relation to the resistance data 24 . Represented by the normalized output layer (Table 1 ), this comparison allows modelers to determine where connectivity has been lost, with respect to the resistance data such as human modification, for example. Another approach is to categorize model outputs. To date, categorization has not been a standard post-processing step, though breaking continuous outputs into categories could help users understand and interpret the patterns of movement potential. For example, Cameron et al. ( 30 ) categorized connectivity data across California based on management and land tenure status to prioritize sites for conservation. More recently, Anderson et al. ( 31 ) mapped climate connectivity for the continental United States and categorized flow based on the distribution and variation of the model outputs, and combined those products with other data to delineate an expansion of the US PA network. We build on this body of work and overcome prior computational hurdles to produce a map of connectivity across all terrestrial landscapes globally. These data represent the potential movement across natural vegetation as determined by levels of human modification at 1 km resolution. We capitalized on increased cloud processing capabilities and the high-efficiency language Julia to run Omniscape globally 32 , 33 . We also improved methods for handling water bodies 34 in the model and incorporated updated and tailored global human modification data that comprehensively accounts for industrial human activities expected to impact movement patterns 35 . We developed six categories of connectivity that can be used to evaluate the extent to which potential movement (connectivity) is impeded, channelized, or diffuse (Table 1 ). Our approach allowed us to produce maps that illustrate how the spatial pattern and intensity of human impacts impede or channel movement and indicate natural areas where connectivity is most likely to be unconstrained. This work is broadly applicable to inform global conservation goals, such as those enumerated in the GBF. To demonstrate its utility, we provide summaries to inform global, national, and sub-national decision making and show how this whole landscape connectivity approach can be used to support global biodiversity goals. Table 1 Definitions of terms for inputs, outputs, and categories from the global Omniscape connectivity modeling process, with corresponding color coding used to display connectivity categories in subsequent figures. Model Component Term Definition Inputs Source The presumed capacity of a site to function as a source for species or gene movement (in this study, all natural lands; 36 ). Resistance Landscape impedance to movement potential, derived in this study from a measure of human modification 35 . Outputs Flow, F (current) The representation of species movement (movement potential) or genetic migration probabilities from source areas across a resistance dataset. Potential flow, F p A null model of flow that indicates movement without resistance. Flow potential allows us to ask: given the amount and configuration of source pixels, how much flow would be expected in the absence of barriers 24 ? Normalized flow, F n The ratio of flow to potential flow. Normalized flow enables more accurate identification of the mechanisms underpinning different flow magnitudes such as where barriers are redirecting flow 24 . Output categories Channelized movement area High flow in relatively narrow pathways near distinct barriers of high resistance. Resistance levels in these areas can vary from low to high; channelization indicates increasing importance of these sites as other options have been lost. Diffuse movement area Areas where flow patterns are similar to what would be expected with a homogeneous surface of low resistance. This includes: Intensified areas : med-high flow areas, adjacent to higher resistance areas from which flow has been redirected. Robust areas : moderate flow across low resistance; many redundant options for movement, indicating natural areas with the least disruption of flow. Dampened areas : lower flow areas within a low resistance neighborhood. This reduction in current flow relative to diffuse can indicate a slight increase in resistance, or lower source strengths. Impeded movement area Restriction and redirection of flow away from these locations due to high resistance. This includes: Weak areas : low flow across minimally permeable lands with medium-high to high resistance Obstructed areas : highly restricted flow, reflecting redirection to other areas by high resistance RESULTS To promote connectivity data in decision making and conservation planning, we developed a framework categorizing complex, continuous connectivity data to policy-relevant outputs; we then applied this framework to produce a global map describing potential movement patterns across all lands (Fig. 1 , Fig. S1 ). The categories derive from classifying Omniscape model outputs (the flow ( F ), flow potential ( F p ), normalized flow ( F n ) datasets; Table 1 ) and the resistance data (levels of land modification 35 ) to create six categories of potential movement that describe the unique combinations of these layers and the flow patterns and magnitudes given the human modification resistance layer ( Methods; Table 1 , Table S1 ). In the text below, we condensed the six categories to ‘channelized’, ‘diffuse’ (i.e., ‘dampened’, ‘robust’, and ‘intensified’), and ‘impeded’ (i.e., ‘obstructed’ and ‘weak’) categories for brevity, although figures and tables present all six categories. Land falling within the diffuse category can be considered the goal for connectivity – these are areas with redundancy in movement pathways that afford multiple options for movement and where human pressures have not greatly changed how species may move. Areas with impeded movement potential occur where there is likely a high level of land modification and where potential movement is much lower than expected (Fig. 2 ). Channelized movement potential is the highest movement potential (i.e. “corridors”), likely where surrounding land modification has forced flow through an area with lower human development, resulting in much higher movement potential than expected in the absence of modification. We find that 66% of natural lands globally has diffuse potential movement, 29% has low potential movement (i.e. movement is ‘impeded’), and 5% has channelized movement. When comparing the precent diffuse per country to other global connectivity indicators (PAI 20 , PARC 37 , ProNet 38 , and ProtConn 39 ), as recommended in the GBF and with 2024 updates 15 , we found low correlations across each indicator. Percent diffuse is most correlated with PARC at r = 0.68. Other correlations with percent diffuse range from − 0.42 to -0.11 (Table 2 ). Table 2 Pearson correlation coefficients across connectivity indices. Percent of diffuse connectivity category per country (> 10,000 km 2 ) vs. Protected Area Index 20 (PAI), Protected Area Representativeness and Connectedness 37 (PARC), Protected Areas Network 38 (ProNet), and Protected Connected 39 (ProtConn) connectivity indicators as updated in UNEP-WCMC and IUCN ( 15 ). % Diffuse PAI PARC ProNet ProtConn % Diffuse -0.42 0.68 -0.37 -0.11 PAI -0.42 -0.49 0.16 -0.17 PARC 0.68 -0.49 -0.30 0.21 ProNet -0.37 0.16 -0.30 0.52 ProtConn -0.11 -0.17 0.21 0.52 Connectivity patterns in biomes and continents Globally, temperate broadleaf and mixed forests have the highest proportion of impeded movement potential (have lost the most connectivity), followed by mediterranean forest, woodlands, and scrub (Fig. 3 ; Table S2). Tundra, boreal, and tropical moist forest biomes have the highest proportion of diffuse movement potential areas. Across biomes, Mediterranean forests, woodlands, and scrub have the highest proportion of channelized areas (6%) and the lowest amount of diffuse areas (58%; Fig. 3 ; Table S2), reflecting high levels of land modification. Deserts and xeric shrublands have the least amount of channelized movement areas (93% diffuse), likely because these areas also have lower land modification. Tundra and boreal forests/taiga both have 97% diffuse movement areas, followed by montane grasslands and shrublands. Tundra and boreal areas also have the least amount of impeded areas. In general, across continents Africa has the highest proportion of diffuse movement areas and the lowest proportion of channelized movement areas (91% and 1%, respectively); whereas Asia has the lowest proportion of diffuse movement areas and highest level of channelized movement areas (74% and 4%, respectively). Asia has the highest and Oceania has the lowest proportion of impeded areas (23% and 7%, respectively; Fig. 4 ; Table S3). In addition, the biome-level patterns of low connectivity (indicated by proportion impeded) varied by continent, with Europe and North America exhibiting high levels of impeded areas in temperate grasslands relative to other continents. Connectivity patterns in countries Unsurprisingly, smaller countries with a large proportion of land modification have the greatest percentage of impeded areas (Fig. 4 , Table S4). The Netherlands, Belgium, Moldova, and Denmark all have the highest proportion of impeded movement area (≥ 85%) and the lowest proportion of diffuse movement areas (≤ 14%). In contrast, relatively large countries with low levels of land modification and with expansive natural land cover have the highest proportion of diffuse movement areas. This includes many nations across Africa and some South American countries, such as Peru, Bolivia, and Colombia (Fig. 4 ; Table S4). Applying the framework to support the Global Biodiversity Framework The categorized connectivity data can be combined with other datasets (i.e. species occurrences or habitats, ecosystem services, etc. 41 ), facilitating consideration of connectivity in global and national-level biodiversity commitments. We illustrate conceptually how our connectivity data and categories can be used to inform GBF Goal A Targets 1 (on spatial planning), 2 (on restoration), and 3 (on protection), though there are uses and applications beyond these targets (for example, Target 4, focusing on preventing species extinction 8 ). We do not attempt to identify specific sites for conservation measures, as this process should be done in collaboration with local communities, consider competing land uses and multiple actions, and incorporate local context, land tenure, and inclusion of equity considerations in line with rights-based approaches to conservation. Target 1 (spatial planning) Our data can inform biodiversity-inclusive spatial conservation planning, which recognizes the importance of considering ecological connectivity in the design of conservation networks within landscapes or seascapes 41 . Alternative spatial scenarios can be evaluated based on their ability to improve connectivity, for example by increasing protection of channelized areas that represent the last remaining pathways of connectivity and to increase movement options through targeted restoration as described in Target 2 (see below). These represent areas important for maintaining or enhancing connectivity across whole landscapes – not just between PAs – and can facilitate the consideration of whole landscapes in global and regional spatial connectivity planning. Target 2 (restoration) Identifying where connectivity has been lost, as identified by the weak, obstructed, or channelized areas — in concert with other information such land tenure, competing land uses, biophysical suitability, and feasibility for ecological recovery — can guide site selection and efficient allocation of resources for restoration. Restoring areas close to channelized flow areas can help preserve and enhance connectivity and restoring areas in weak or obstructed flow may help restore connectivity, depending on the cause of loss (Fig. 5 ). Target 3 (protection) : This data can also be used to identify priority sites for future conservation measures, including siting of new PAs and OECMs. In overlaying the categorized connectivity data with global PA locations, we found that countries with greater proportions of diffuse connectivity areas tend to have lower levels of protection, but more concerningly, there are many countries with low levels of protection and high levels of impeded movement areas (Fig. 4 ; Fig. 6 ). These data allow users to better understand the landscape context of a PA. For example, if a country’s PAs primarily fall within diffuse areas, biodiversity may benefit from conservation measures focused in channelized areas. If PAs are surrounded by impeded areas, future conservation efforts may focus on any channelized areas between PAs (Fig. 5 ). For example, Australia has reached approximately 20% protection 15 with the majority of PAs occurring in the desert and xeric shrublands biome and in diffuse areas (i.e. are connected). To advance reaching both Targets 2 and 3, concentrating future conservation in more modified areas may be advantageous from the point of view of improving connectivity and providing potential additional benefit to both people and wildlife, rather than focusing conservation where areas are already well-connected. Future siting of PAs, OECMs, or conservation and management actions could be targeted at remaining channelized areas within modified areas to maintain existing connections. If restoration is feasible, these efforts could be focused near currently protected channelized areas to enhance these connections and increase their resilience. DISCUSSION Nearly all countries agreed to ambitious goals and targets in the GBF that build on decades of science and advocacy. Restoration, protection, and ecosystem connectivity are foundational to Goal A within the GBF. Yet, to date the technical tools and evidence on how best to consider and integrate connectivity across whole landscapes have been lacking. We fill this gap by conducting a global connectivity analysis which considers broad-scale connectivity across entire landscape gradients and providing a framework for analyzing the complex outputs that is scientifically robust and policy relevant. This framework can be applied across scales to meet landscape-level needs with high resolution and country/region-specific data. We find that when whole landscapes are considered, the majority of Earth’s terrestrial areas lie within connected zones (66% of terrestrial lands have diffuse movement potential). Conservation efforts should focus on protecting these well-connected and channelized areas, since many countries with high amounts of diffuse movement areas have low levels of protection (Fig. 6 ). In addition, restoring impeded and channelized areas (areas which have lost connectivity due to human-driven land modifications) should also be a conservation goal. With a growing human population and ongoing development expansion, diffuse movement areas could become channelized or impeded without continued or improved conservation actions. This information should be used in concert with conversion risk mapping (e.g., 42 ) and other connectivity measures. When whole landscapes are not included in connectivity areas, unprotected but biologically important areas could be left out of connectivity planning, leading to reduced species and ecosystem persistence over time 9 . The categorization framework we developed strengthens the ability to communicate the importance of connectivity for conservation planning in two ways. First, it recognizes at least two ways that connectivity can be compromised: it can be impeded, in which movement is reduced or lost compared to unmodified landscapes, or it can be channelized, in which movement in a channel is increased even as it is reduced in other areas of the landscape. Channelized systems have lost redundancy in movement options that provide resilience to a system and increase the risk of complete loss of connectivity in the face of future land use and climate changes. The second advantage of this categorization derives from explicitly evaluating the site-specific resistance values, which enables us to model the impact of site-specific interventions on whole landscape connectivity. The low correlation of diffuse movement areas per country with other connectivity measures suggests that our data provides complementary information (Table 2 ). Indeed, our results differ from prior connectivity analyses in several important ways. First, analyses that model connectivity between PAs miss connections elsewhere. If not used in concert with non-PA based connectivity measures, these results will bias connectivity conservation towards areas of high PA occurrence, the global pattern of which is geographically and ecologically biased in remote areas potentially jeopardizing important but unprotected areas for both nature and people 14 , 15 . By using natural land cover as our source data and modeling connectivity across all lands irrespective of conservation status, our data depict the ecological connectivity regardless of PA distribution. For example, the Appalachian region of the eastern United States has a low density of PAs but has swathes of channelized areas which are remaining connections running along mountain ridges in a north-south orientation (Fig. 2 ). These areas are potentially important for maintaining connectivity for wildlife as climate changes. Yet, these ecologically important areas are not identified in other connectivity assessments 20 , 25 , 43 . In addition, our approach and results remain robust to sociopolitical instability resulting in PA downgrading, degazettement, or degradation or when PAs are unreported 44 . Biased protection is also reflected in the distribution of connectivity categories within biomes. For instance, temperate broadleaf and mixed forests and grasslands, mediterranean forests, woodlands, and scrub, and tropical and subtropical dry broadleaf forests all have high proportions of impeded lands and low proportions of land protection 40 . Our results underscore the need for enhanced conservation measures in these biomes. A lack of PA prevalence in these biomes may direct conservation elsewhere when relying on PA-based connectivity analyses. Lack of protection and PA-centric consideration of connectivity in grasslands, for example, can be seen in the midwestern United States, southern Brazil, and India. These areas have moderate levels of human land pressures but remain important for maintaining connectivity across broad landscapes. At a continental scale, there are areas identified as highly connected when solely examining distance between PAs (e.g., ProtConn 45 ), even though these areas have low proportions of diffuse and high proportions of impeded areas. If we use the percent of diffuse area as a measure of landscape connectivity, Africa ranks the highest globally, followed by Oceania (Fig. 5 ; Table S3). However, most connectivity indicators have been calculated assuming all areas outside of PAs are equally inhospitable (e.g., 15 , 45 ). With this assumption, Europe is the most connected continent, with the Americas ranking second. This difference stems from geographically clustered PAs in the EU and Americas, within landscapes of moderate-to-high levels of human modification, resulting in impeded lands between PAs. PAs may be geographically near one another, but movement between PAs may not be likely given the surrounding landscape. Many countries with low connectivity as indicated by ProtConn 45 were dominated by the diffuse category in our analysis, indicating redundancy in movement pathways throughout most of these countries’ land areas (Mexico, Panamá, Colombia, Ecuador, Madagascar, Papua New Guinea, Nepal; Fig. 1 , Table S4). Many of these countries have large amounts of diffuse connectivity through natural lands that remain unprotected. Our results are similar to the findings of Ward et al. ( 46 ), where Europe had the lowest PA connectivity, when measured by intact lands as defined by levels of human pressure. Our work enables evaluation of the effectiveness of the existing PA networks in ensuring connectivity. Countries with high amounts of diffuse under the 30% target should identify whether there are unprotected channelized areas (bottom right, Fig. 6 ). Countries with low amounts of diffuse areas, moderate protection, and high amounts of channelized areas should ensure diffuse areas are protected to maintain connectivity redundancy (center of the graph, Fig. 6 ). Countries with low amounts of diffuse areas, low proportions of protection, and low amounts of channelized areas likely have high land modification and conservation efforts may be better suited for non-area-based conservation or ecosystem restoration measures, such as many countries in the EU (Fig. 6 ). The categorized connectivity data allows jurisdictions to identify key locations for restoration to bolster existing connectivity. Another major difference between our work from others stems from the theoretical approach to connectivity underpinning our modelling. Our work represents connectivity across natural land cover considering the gradient of anthropogenic land modification, regardless of species or taxa, and may be considered structural connectivity. Where prior connectivity analyses have taken a particular species or taxa into account, results depict areas important for that species or taxa and must be used in concert with other species and ecosystem condition information to inform global biodiversity goals. For example, Brennan et al. ( 20 ) show low connectivity across much of Southeast Asia, likely reflecting the regional-scale loss of large mammals or the lack of large mammal movement data in this region 47 , 48 , rather than this area’s inherent connectivity value. When modeling connectivity across all natural land cover, we found impeded, diffuse, and potentially important channelized areas across South and Southeast Asia, including the Indian subcontinent (Fig. 2 ). India is dominated by impeded areas and the remaining channelized areas could be conserved or managed to maintain the existing connectivity (Table S2). The differences in our work compared with others confirm the complexity and contextuality of measuring ecological connectivity. Multiple measures should be used to assess and track global connectivity in complementarity to capture the multiple aspects of connectivity; both between PAs and OECMs and to the wider landscape 8 . We modeled connectivity between terrestrial natural vegetation, not connectivity for any one species or type of geography. We assumed that human pressures, as mapped in the human modification data and modified here to reflect land cover alterations impacting migration, represent meaningful resistance from anthropogenic barriers to movement for many species 49 , 50 . While we did not attempt to define specific mechanisms of resistance, negative impacts of human modification of land cover and use on animal movement are shown to include decreased habitat quality, increased mortality or predation risk, and associated behavioral responses (avoidance) 51 – 53 . We recognize that for some species, especially generalists and those commonly found in urban environments, this assumption may not be valid. Rather, this modeling effort is intentionally designed to apply to species which are negatively impacted by industrial human pressures on a landscape that can cause destruction, degradation, and impairment of ecological processes (as outlined by the IUCN biodiversity threats classification 54 and adopted by global human modification mapping 35 ). To identify more small-scale connectivity patterns, relevant for short distance dispersers such as herpetofauna and most plants 55 , 56 , our methods and framework can be downscaled using local data since the global results were sensitive to some parameterization decisions. Our global team (see Methods) specifically identified grassland areas with low human modification as sensitive to slight increases in human modification due to pasture/grazing. We subsequently removed the pasture/grazing layer from our analysis, which assumes limited impacts of grazing lands on movement potential. While there are caveats to our model, global products such as ours have broadscale utility (e.g., 57 ), and this work will provide critical context that helps to identify important patterns of variation in connectivity constraints and where to guide more in-depth assessments. By no means is this work meant to replace local or landscape-tailored modeling and planning where fine-scale data are available to answer more targeted questions. Measuring and monitoring connectivity is not only theoretically challenging, but also technologically challenging. Fortunately, improvements in software and computing resources are rapidly enabling major advances to allow this modeling at high resolution. There are several modeling options for continuous circuit-theory connectivity modeling 26 , but the implementation of Omniscape in the Julia language in particular allows relatively efficient global modeling that facilitates iterative parameter testing 32 , 33 , 58 , 59 . The power to consider connectivity across a continuum of landscape conditions without set start and end points (such as PAs) provides a powerful, flexible way to deliver decision-relevant products. We expect that faster runs, strengthened by deeper understanding of model behavior, will accelerate model co-creation with decision-makers and regional experts. Increased processing speed allowed us to iteratively test model parameters and work closely with collaborators to refine the connectivity modeling inputs (e.g., human modification resistance surface), provide guidance on modeling parameterizations (i.e., how water was treated as movement barrier), and iteratively consult with experts on draft model outputs. Combined, these collaborative efforts and advancements refined our approach, improved the reliability of our outputs, thereby promoting higher relevance for the policy applications and goal setting processes. Beyond the GBF, there is global interest and growing momentum for restoration and protection of natural systems at scale for climate benefits, including natural climate solutions (NCS) 60 , 61 . Our map allows identification of sites for restoration and protection that maximize benefits to connectivity for biodiversity conservation in addition to carbon sequestration 62 – 65 . When further paired with land tenure information and with community and local engagement, protection and restoration of natural systems can be powerful tools for durable conservation outcomes 66 . Users of our framework and map should apply best practices in engaging local communities and other critical local stakeholders for advancing GBF goals. Importantly, large-scale projects protecting connectivity on unprotected territories must meet the equity and justice concerns of local communities 67 . Global environmental agendas often lead to ambitious goals that can galvanize environmental advocates, but the impacts on local populations is not uniform. A next step may be to overlay human populations (e.g., 68 ) and land tenure (e.g., 69 ) with our global map to assess potential areas where careful planning and implementation will be critically important. Government commitments for land-based carbon removal projects alone now exceed 1 billion hectares 70 and further demands will likely lead to competing and greater demands on landowners and residents. CONCLUSION Ecosystems, species ranges, and human land uses are shifting, often in unpredictable ways, necessitating more holistic approaches to conservation that extend beyond PA boundaries 71–73 . Protecting and maintaining ecological connectivity is critically important to stem the biodiversity crisis. Indicators that can evaluate connectivity are needed to set targets and measure success towards global conservation goals such as those in the GBF 8 . There has been a lack of sufficient methods for whole landscape connectivity modeling, especially for implementation at the large scales required to meet national and global goals. Global connectivity indicators must consider unprotected areas as well as protected areas and efforts to stem biodiversity loss need to include the restoration and protection of natural lands outside of PAs 46 . By presenting a continuous global connectivity map with a policy-relevant framework, applicable across different landscapes and contexts, we can better inform policy and land-use planning. Such an approach can avoid the pitfalls of relying exclusively on PA-centric connectivity indicators, which can result in biased information and direct conservation focus away from unprotected areas critical for whole landscape connectivity. As landscapes, ecosystems, and climate change, moving beyond the boundary to identify and preserve whole system connectivity is critical for conserving biodiversity in its current and future habitat. METHODS There are many ways to model connectivity. We chose a circuit theory method that allowed us to emphasize how landscape patterns influence broad movement potential at multiple scales, from landscapes to continents. To estimate connectivity at the full global extent at 1 km resolution, we used Omniscape.jl, a moving window-based adaptation of the Julia language implementation of the circuit-theoretic modeling software Circuitscape 27 , 33 , 58 . Circuit-theory models allow considering how movement may spread or be constrained in different parts of a landscape by using the mathematics of electrical circuit theory to describe cumulative patterns that represent traces of ‘random walkers’ as they encounter variable resistance using the analogy of electrical current flow 27 . Resistance weights are assigned to pixels based on hypotheses of the degree to which the conditions in that pixel would support or inhibit movement. The assumption of random movement suggests that, as one example, directional routes reflecting knowledge of the landscape (like annual migrations) are likely not well modeled with this approach 74 . Rather, a key strength of treating movement as a random process is that it allows seeing where many movement routes are possible, or where options are highly constrained or absent. Omniscape is an expansion of Circuitscape, which considers connectivity between focal nodes or patches, by allowing modelers to evaluate connectivity in all directions (“omnidirectional” Circuitscape) to and from all pixels (for full details on the Omniscape approach, see 24 , 33 ). Here, we deployed Omniscape with separate resistance and source layers using a moving window radius of 125 km, globally at 1 km resolution. Resistance data We used an updated version of the global 300 m Human Modification (HM) dataset for 2022, refined specifically for modeling connectivity 35 , and coarsened it to 1 km using average-based resampling for the resistance layer. The HM includes the spatial extent and intensity of cumulative human pressures from settlement, agriculture, forestry, transportation, mining, energy production, electrical infrastructure, dams, pollution and human accessibility. For the connectivity resistance surface, we removed pasture/grazing and air pollution data, because we assumed they have low influence on species movement. For pixels intersecting major roads and highways, we assigned a value of 0.8 (for 300 m pixels) to decrease the muting of linear features when decreasing resolution (to 1 km). In Omniscape, small differences in resistance at the low end of the range, especially in factors that occur in large swaths of a landscape (such as large blocks of agriculture) typically have a large influence on the results. Although HM has been used as resistance surfaces in connectivity modeling 75 , few studies have explored using transformations on the resistance data to understand the model’s sensitivity to the distribution of continuous values 76 . Exponential functions may better describe the relationship of wildlife to landscapes during movement life history periods than non-exponential linear functions 25 , 77 , 78 . Specifically, exponential transformations may better describe the relationship of raw HM to resistance with respect to broad-scale animal movement than the raw HM data itself. This improvement reflects that negligibly to moderately modified areas of the landscape may be readily used for movements, and only the most modified areas are unusable. These transformations are more ecologically sound when considering long-distance movements and allow for pattern exploration in the middle of the HM distribution 77 , 78 . We used the function: $$\:f=100-99\frac{1-exp(-c(1-HM)}{1-exp(-c)}$$ Where f i s the transformed value of the HM and c is a rescaling parameter 77 . With larger c values, the transformation approaches an exponential function of the HM and smaller positive values approach a linear transformation. We tested c = 0.25, 2, 4, and 8 (Table S5) scaling parameters on the HM data during our sensitivity testing. Source data A source layer in Omniscape is used to represent locations from which movement originates, such as a parcel of natural land or a protected area 24 . A common source layer in connectivity modeling is the inverse of the resistance layer, especially in models where human modification is used as the resistance layer 24 , 25 . Instead, we used natural land cover as our source to model a global ecological network. Natural land cover pixels better represent areas that will likely be moved from, and although correlated with HM, provides a clear way to represent ecological networks that connect major ecosystem types. We used the ESA World Cover 10 m v 200 36 product as our source layer because we were interested in creating a species-agnostic connectivity model, with connectivity identified between areas of natural lands and it includes accuracy assessment and an uncertainty layer with an estimated 76% global accuracy. We first labeled each class in the ESA World Cover data as representing natural vegetation (e.g., tree cover, shrubland, grassland, bare/sparse, wetland, mangroves, moss/lichen) or non-natural (agriculture, urban) lands, then rescaled the data to 1 km 2 pixels. We used the proportion of natural lands within 1 km pixels as our source weight data by calculating the average value (0, non-natural; 1, natural) across all 10 m pixels within each rescaled 1 km cell, excluding water pixels. Water The treatment of water in resistance data can have a significant impact on connectivity outputs 79 . We developed a method for addressing water that aligned with the purpose of mapping terrestrial connectivity and our intent of informing conservation strategies that address the influence of human modifications on movement potential. demonstrates Water has been treated in ways that range from “no data” (effectively complete barriers to movement) to low or moderate resistances, to approaches that scale the resistance of water to size of water body or flow rate, to maximum resistance (e.g., 24 , 25 , 31 , 80 – 82 ). Through iterative testing, we developed an approach intended to limit the effect of water on the model outputs 34 . To include water in a “neutral” way that does not prohibit or facilitate resulting flow ( F ), we assigned ocean and inland water pixels within 10 km of land random resistance values, within 2% of the average resistance value of a 40 km surrounding radius. This approach ensured that the resistance values of water would not have higher or lower resistance than neighboring land. We assigned the maximum resistance value to water more than 10 km from land, assuming that few terrestrial species would be able to move in water > 10 km from land. Finally, we masked water from the final output maps. Our map of water was based on the ESA CCI Global Water Bodies dataset (v4.0; 83 ). Model parameters Omniscape allows for ‘blocking’ or grouping squares of pixels and using the center pixel as the source 33 . This method decreases processing time and does not impact model results 25 . In all cases, we used a block size 10% of the moving window radius to decrease model run time while maintaining model integrity 25 . We tested radius sizes of 50 km, 125 km, and 425 km and we chose the radius size of 125 km to reflect potential dispersal distances of most birds and mammals, ranging from < 1 km – 467 km, with medians ranging from < 0.1 km to 87 km 84 . Model estimation We computed all outputs on r6i.4xlarge EC2 virtual machine instances on Amazon Web Services (AWS), each containing 16 vCPUs and 128 GiB of RAM. Models were estimated using Omniscape v0.5.8 running on Julia v1.9.2 with 6 threads. Model run time varied depending on the window radius and ranged from several hours (for a 50 km radius) to approximately one week (for a 425 km radius). Output categories The Omniscape main outputs, F and the F n provide complementary information to convey how the resistance and source data interact to influence the magnitude of flow across landscapes, and the degree to which movement is constrained by barriers 24 . The F emphasizes the most likely places for movement to occur, highlighting key areas of concentration and broad movement zones between the most intact areas. The F n , where the magnitude of flow is scaled to the regional magnitude of sources, identifies places where F has become concentrated relative to flow in the absence of resistance ( F p ), even if the overall flow amount is moderate or low relative to other locations. This allows the model to highlight key movement zones and pinch points (areas where flow is restricted and narrowed) across the full range of resistance levels and geographic distributions of source areas. The categorization was informed by four continuous components which were each categorized: the continuous flow pattern (impeded, diffuse, channelized) from the F n , the F p , F , and the HM resistance data. We categorized each of these four layers based on thresholds we identified which aligned with changes in the distribution of each data layer: Normalized flow (F n , ) : We adopted the thresholds identified by Cameron et al. ( 30 ), where F n > 1.7 is channelized flow, 1.3 ≥ F n ≤ 1.7 is intensified flow, 0.7 ≥ F n <1.3 is diffuse flow, and F n < 0.7 is impeded flow. Flow potential (F p ) To determine the thresholds for the flow potential layer, we calculated the global distribution of flow potential and cumulative current and identified one threshold break for the flow potential data, opting to include the majority of data into one lower category (µ + 0.5 SD) and defining values > µ + 0.5 SD as ‘high’. Flow (F) Because the flow layer is calculated and defined relative to the potential flow layer, we used the same values as the flow potential layer with the addition of a ‘very high’ category. We defined ‘very high’ cumulative flow as > µ Fp + 1.5SD Fp . Human modification layer : We based the thresholds for the HM on those identified in Kennedy et al. ( 49 ) and Riggio et al. ( 85 ). However, because we used the HM as it relates to how species may move across landscapes, we adjusted the thresholds to align with HM values as found within ESA World Cover land cover classes. Because tree cover and other natural classes were most abundant < 0.1 HM and because species can move across sub-optimal habitat, we combined the “very low” and “low” HM classes previously identified 49 , 85 . Our HM classes were thus: 0.7 “high” modification. We reclassified each of the above continuous layers according to the respective identified thresholds and combined all four layers to create one global categorized map with 96 potential unique categories. Only 63 categories occurred globally, and we iteratively identified commonalities in the pattern and occurrence of between the categories to further group them. In doing so, we prioritized normalized flow categorization then grouped these by human modification level and used the difference in flow potential and cumulative flow to further inform groupings. Sensitivity testing We tested multiple model inputs and parameters to determine their effects on model outputs (Tables S5, S6, Fig S2). We ran models with different c values, model search radii, source layers, water treatment methods, slope, and treatment of grazing lands. We compared cumulative flow outputs using a Pearson correlation coefficient, calculated in R package terra v 1.7.71 86 (Table S6). Review process Because we conducted a global analysis, we formed a review team of 14 conservation practitioners across six continents to fine-tune model inputs and ensure model validity. The team (including coauthors J.B.-G., C.R.D., J.F., M.N.-R., and M.F.R.) included conservation scientists with geospatial and/or connectivity modeling backgrounds with knowledge of African, Australian, Asian, North American, and South American landscapes. They were provided a data viewer with the input and output datasets, written materials about our methods, a video of our intent and methods, and a questionnaire guiding their review. After these review sessions, the team reviewed the output datasets with respect to their knowledge of landscapes and other existing connectivity analyses. We then reviewed feedback received and made changes to the input data to address concerns raised by reviewers. Overall, across geographies, reviewers expressed concerns around the inclusion of pasture/grazing lands as a component of the resistance layer, which led us to use a modified version of the HM data 35 , 49 . Data summaries To support country and regional decision making, we summarized movement potential by country, continent, biome, and protected areas boundaries. For summaries by biome, we used the biomes identified by Dinerstein et al. ( 40 ). For protected areas, we used data from the World Database of Protected Areas 87 . We followed Allan et al. ( 88 ) by including China and India PAs using the 2017 WDPA data. We excluded marine protected areas, PAs that were “proposed” or in the “not applicable” protection categories. We created separate layers for each IUCN protection category: I-VI, “not reported”, or “not assigned” and converted these to 1 km 2 cell raster data, where any cell with ≥ 50% overlap was included as protected. We combined these raster layers and removed any overlap, retaining the highest level of protection, where applicable. To compare our data with other connectivity measures, we calculated Pearson correlation coefficients for countries > 10,000 km 2 for the percentage diffuse within a country, protected area index 20 (PAI), Protected Area Representativeness and Connectedness 37 (PARC), Protected Areas Network 38 (ProNet), and Protected Connected 39 (ProtConn) connectivity indicators as updated in UNEP-WCMC and IUCN ( 15 ) using R v 4.4.3 89 . Declarations ACKNOWLEDGEMENTS We are grateful to Brad McRae for the theoretical development of Omniscape and his pioneering work in the application of circuit theory to ecology. Viral Shah, Ranjan Anantharanam, and Vincent Landau continue to maintain Circuitscape and Omniscape and have provided valuable insights during the development of this work. We are grateful to Manuel Agra, Mario Barroso, Hugo Cardenas Rodriguez, Mercedes Ibanez, Wang Longzhu, Nathaniel Robinson, Anne M. Trainor for reviewing preliminary data and improving this work and to Aimee Biegel, Peter W. Ellis, and Preston Welker for supporting this work. We thank the Bezos Earth Fund for making this work possible. AUTHOR CONTRIBUTIONS Erin E. Poor– Conceptualization, methodology, project administration, writing, supervision, funding acquisition; Kim Hall– conceptualization, methodology, writing; Jesse Anderson– data curation, formal analysis, software, visualization; Melissa Clark– methodology, writing; Aaron Jones– conceptualization, visualization, writing, data curation, methodology; Christina M. Kennedy– conceptualization, methodology, writing; Carrie A. Schloss – methodology, writing; David M. Theobald– data curation, methodology, writing; Jaime Burbano-Girón– writing; Susan C. Cook-Patton– writing; C. Ronnie Drever– writing; Joseph Fargione– conceptualization, writing; James Fitzsimmons– writing; Mauricio Nunez-Regueiro– writing; Milena Fermina Rosenfield– writing; Yuta J. Masuda– funding acquisition, writing References Ceballos, G. et al. Accelerated modern human – induced species losses : Entering the sixth mass extinction. 9–13 (2015). Díaz, S. et al. Pervasive human-driven decline of life on Earth points to the need for transformative change. Science (80-. ). 366 , (2019). Jaureguiberry, P. et al. The direct drivers of recent global anthropogenic biodiversity loss. Sci. Adv. 8 , 1–11 (2022). Maxwell, S. L., Fuller, R. A., Brooks, T. M. & Watson, J. E. M. The ravages of guns, nets and bulldozers. Nature 536 , 143–145 (2016). Pecl, G. T. et al. Biodiversity redistribution under climate change: Impacts on ecosystems and human well-being. Science (80-. ). 355 , (2017). IPBES. Global assessment report of the Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services . Debating Nature’s Value (2019). UNEP-CMS. IMPROVING WAYS OF ADDRESSING CONNECTIVITY IN THE CONSERVATION OF MIGRATORY SPECIES . (2020). CBD. Kunming-Montreal Global Biodiversity Framework. 1–15 (2022). Hanski, I. Metapolulation dynamics. Nature 396 , 41–49 (1998). Fahrig, L. Effects of Habitat Fragmentation on Biodiversity. Annu. Rev. Ecol. Evol. Syst. 34 , 487–515 (2003). Saura, S. et al. Global trends in protected area connectivity from 2010 to 2018. Biol. Conserv. 238 , 108183 (2019). Locke, H. et al. Three global conditions for biodiversity conservation and sustainable use: An implementation framework. Natl. Sci. Rev. 6 , 1080–1082 (2019). Joppa, L. N. & Pfaff, A. High and far: Biases in the location of protected areas. PLoS One 4 , 1–6 (2009). Dinerstein, E. et al. A ‘global safety net’ to reverse biodiversity loss and stabilize earth’s climate. Sci. Adv. 6 , 1–14 (2020). UNEP-WCMC & IUCN. Protected Planet Report 2024 . (2024). Laurance, W. F. et al. Averting biodiversity collapse in tropical forest protected areas. Nature (2012) doi:10.1038/nature11318. Maxwell, S. L. et al. Area-based conservation in the twenty-first century. Nature 586 , 217–227 (2020). Wolff, N. H. et al. Prioritizing global land protection for population persistence can double the efficiency of habitat protection for reducing mammal extinction risk. One Earth 6 , 1564–1575 (2023). Mendenhall, C. D., Karp, D. S., Meyer, C. F. J., Hadly, E. A. & Daily, G. C. in agricultural landscapes. (2014) doi:10.1038/nature13139. Brennan, A. et al. Functional connectivity of the world’s protected areas. Science (80-. ). 376 , 1101–1104 (2022). Myers, N., Mittermeier, R. A., Mittermeier, C. G., Fonseca, G. A. B. & Kent, J. Biodiversity hotspots for conservation priorities. 403 , 853–858 (2000). Hilty, J. et al. Guidelines for conserving connectivity through ecological networks and corridors. Best Practice Protected Area Guidelines Series No. 30 . (2020). Dutta, T., Sharma, S., Meyer, N. F. V., Larroque, J. & Balkenhol, N. An overview of computational tools for preparing, constructing and using resistance surfaces in connectivity research. Landsc. Ecol. 37 , 2195–2224 (2022). McRae, B. H. et al. Conserving Nature’s Stage: Mapping Omnidirectional Connectivity for Resilient Terrestrial Landscapes in the Pacific Northwest. http://nature.org/resilienceNW (2016) doi:10.13140/RG.2.1.4158.6166. Belote, R. T., Barnett, K., Zeller, K., Brennan, A. & Gage, J. Examining local and regional ecological connectivity throughout North America. Landsc. Ecol. (2022) doi:10.1007/s10980-022-01530-9. Phillips, P., Clark, M. M. & Koen, E. L. Comparison of methods for estimating omnidirectional landscape connectivity. Landsc. Ecol. 36 , 1647–1661 (2021). McRae, B. H., Dickson, B. G., Keitt, T. H. & Viral B. Shah. Using circuit theory to model connectivity in ecology, evolution, and conservation. Ecology 89 , 2712–2724 (2008). Cook, C. N., Mascia, M. B., Schwartz, M. W., Possingham, H. P. & Fuller, R. A. Achieving conservation science that bridges the knowledge-action boundary. Conserv. Biol. 27 , 669–678 (2013). Beger, M. et al. Demystifying ecological connectivity for actionable spatial conservation planning. Trends Ecol. Evol. 37 , (2022). Cameron, D. R., Morrison, S. A., Schloss, C. A. & Theobald, D. M. A framework to select strategies for conserving and restoring habitat connectivity in complex landscapes. 1–16 (2022) doi:10.1111/csp2.12698. Anderson, M. G. et al. A resilient and connected network of sites to sustain biodiversity under a changing climate. Proc. Natl. Acad. Sci. U. S. A. 120 , 1–9 (2023). Bezanson, J., Edelman, A., Karpinski, S. & Shah, V. B. Julia: A fresh approach to numerical computing. SIAM Rev. 59 , 65–98 (2017). Landau, V., Shah, V., Anantharaman, R. & Hall, K. Omniscape.jl: Software to compute omnidirectional landscape connectivity. J. Open Source Softw. 6 , 2829 (2021). Clark, M., Anderson, J., Anderson, M. G., Hall, K. R. & E., P. E. Wade carefully: Water parameterizaion in structural landscape connectivity modeling. Landsc. Ecol. Theobald, D. M., Oakleaf, J. R., Moncrieff, G., Voigt, M. & Kiesecker, J. Global extent and change in human modification of terrestrial ecosystems from 1990 to 2022. 1–66 (2025). Zanaga, D. et al. ESA WorldCover 10m 2020v 100. (2021). Harwood, T., Ware, C., Hoskins, A. & Ferrier, S. PARC: Protected Area Connectedness Index v2: 30s global layer 2020. (2022) doi:https://doi.org/10.25919/kt3f-2z04. Theobald, D. M., Keeley, A. T. H., Laur, A. & Tabor, G. A simple and practical measure of the connectivity of protected area networks: The ProNet metric. Conserv. Sci. Pract. 4 , 1–13 (2022). Saura, S. et al. Protected area connectivity: Shortfalls in global targets and country-level priorities. Biol. Conserv. 219 , 53–67 (2018). Dinerstein, E. et al. An Ecoregion-Based Approach to Protecting Half the Terrestrial Realm. Bioscience 67 , 534–545 (2017). Grantham, H., Adams, V. & Planning, T. I. W. T. on S. An overview of Participatory, Integrated, and Biodiversity-Inclusive Spatial Planning and Target 1 under the Convention on Biological Diversity Kunming-Montreal Global Biodiversity Framework, IUCN WCPA Issues Paper Series No. 4 . Oakleaf, J. et al. Mapping global land conversion pressure to support conservation planning. Sci. Data 11 , 1–10 (2024). Saura, S. et al. Protected area connectivity: Shortfalls in global targets and country-level priorities. Biol. Conserv. 219 , 53–67 (2018). Golden Kroner, R. E. et al. The uncertain future of protected lands and waters. Science (80-. ). 364 , 881–886 (2019). Saura, S. et al. Protected area connectivity : Shortfalls in global targets and country-level priorities. Biol. Conserv. 219 , 53–67 (2020). Ward, M. et al. Just ten percent of the global terrestrial protected area network is structurally connected via intact land. Nat. Commun. 11 , 1–10 (2020). Edwards, D. P., Tobias, J. A., Sheil, D., Meijaard, E. & Laurance, W. F. Maintaining ecosystem function and services in logged tropical forests. Trends Ecol. Evol. 29 , 511–520 (2014). Wilcove, D. S., Giam, X., Edwards, D. P., Fisher, B. & Koh, L. P. Navjot’s nightmare revisted: logging, agriculture, and biodiersity in Southeast Asia. Trends Ecol. Evol. 28 , 531–540 (2013). Kennedy, C. M., Oakleaf, J. R., Theobald, D. M., Baruch-Mordo, S. & Kiesecker, J. Managing the middle: A shift in conservation priorities based on the global human modification gradient. Glob. Chang. Biol. 25 , 811–826 (2019). Theobald, D. M. et al. Earth transformed: Detailed mapping of global human modification from 1990 to 2017. Earth Syst. Sci. Data 12 , 1953–1972 (2020). Dyer, S. J. ., Neill, J. P. . O. ’, Wasel, S. M. . & Boutin, S. Avoidance of Industrial Development by Woodland Caribou. J. Wildl. Manage. 65 , 531–542 (2001). Gantchoff, M. G. & Belant, J. L. Regional connectivity for recolonizing American black bears (Ursus americanus) in southcentral USA. Biol. Conserv. 214 , 66–75 (2017). Scoyoc, A. Van, Smith, J. A., Gaynor, K. M., Barker, K. & Brashares, J. S. The influence of human activity on predator – ­ prey spatiotemporal overlap. 1124–1134 (2023) doi:10.1111/1365-2656.13892. IUCN. IUCN CMP Unified Classification of Direct Threats v.3.3 . (2022). Cain, M. L., Milligan, B. G. & Strand, A. E. Long-distance seed dispersal in plant populations. Am. J. Bot. 87 , 1217–1227 (2000). Semlitsch, R. D. Differentiating Migration and Dispersal Processes for Pond-Breeding Amphibians. J. Wildl. Manage. 72 , 260–267 (2008). Chaplin-Kramer, R. et al. Conservation needs to integrate knowledge across scales. Nat. Ecol. Evol. 6 , 118–119 (2022). Anantharaman, R., Hall, K., Shah, V. B. & Edelman, A. Circuitscape in Julia: High Performance Connectivity Modelling to Support Conservation Decisions. JuliaCon Proc. 1 , 58 (2020). Hall, K. R. et al. Circuitscape in julia: Empowering dynamic approaches to connectivity assessment. Land 10 , 1–24 (2021). Griscom, B. W. et al. Natural climate solutions. Proc. Natl. Acad. Sci. U. S. A. 114 , 11645–11650 (2017). IPCC. Section 4: Near-Term Responses in a Changing Climate. Clim. Chang. 2023 Synth. Rep. 42–66 (2023) doi:10.59327/IPCC/AR6-9789291691647. Yang, X. et al. Establishment success in a forest biodiversity and ecosystem functioning experiment in subtropical China (BEF-China). Eur. J. For. Res. 132 , 593–606 (2013). Pettorelli, N. et al. Time to integrate global climate change and biodiversity science-policy agendas. J. Appl. Ecol. 58 , 2384–2393 (2021). Ellis, P. W. et al. The principles of natural climate solutions. Nat. Commun. 15 , (2024). Chang, C. H. et al. Global evidence of human well-being and biodiversity impacts of natural climate solutions. Nat. Sustain. 8 , (2025). FILAC., F. and. Forest governance by indigenous and tribal peoples An opportunity for climate action in Latin America and the Caribbean Forest governance by indigenous and tribal peoples An opportunity for climate action in Latin America and . https://doi.org/10.4060/cb2953en (2021). Raymond, C. M. et al. Inclusive conservation and the Post-2020 Global Biodiversity Framework: Tensions and prospects. One Earth 5 , 252–264 (2022). Lebakula, V., Epting, J., Moehl, J., Stipek, C., Adams, D., Reith, A., Kaufman, J., Gonzales, J., Reynolds, B., Basford, S., Martin, A., Buck, W., Faxon, A., Cunningham, A., Roy, A., Barbose, Z., Massaro, JLebakula, V., Epting, J., Moehl, J., Stipek, C., M. LandScan Silver Edition [data set]. (2024) doi:https://doi.org/10.48690/1531770. Garnett, S. T. et al. A spatial overview of the global importance of Indigenous lands for conservation. Nat. Sustain. 1 , 369–374 (2018). Self, A., Burdon, R., Lewis, J., Riggs, P. & Dooley, K. Land Gap Report Briefing Note 2023 Update . (2023). Lenoir, J. et al. Going against the flow: Potential mechanisms for unexpected downslope range shifts in a warming climate. Ecography (Cop.). 33 , 295–303 (2010). Lenoir, J. & Svenning, J. C. Climate-related range shifts - a global multidimensional synthesis and new research directions. Ecography (Cop.). 38 , 15–28 (2015). Tomiolo, S. & Ward, D. Perspectives in Plant Ecology , Evolution and Systematics Species migrations and range shifts : A synthesis of causes and consequences. Perspect. Plant Ecol. Evol. Syst. 33 , 62–77 (2018). Poor, E. E., Loucks, C., Jakes, A. & Urban, D. L. Comparing Habitat Suitability and Connectivity Modeling Methods for Conserving Pronghorn Migrations. PLoS One 7 , (2012). Theobald, D. M., Reed, S. E., Fields, K. & Soul, M. Connecting natural landscapes using a landscape permeability model to prioritize conservation activities in the United States. 5 , 123–133 (2012). Zeller, K. A., Mcgarigal, K. & Whiteley, A. R. Estimating landscape resistance to movement : a review. 777–797 (2012) doi:10.1007/s10980-012-9737-0. Trainor, A. M., Walters, J. R., Morris, W. F., Sexton, J. & Moody, A. Empirical estimation of dispersal resistance surfaces: A case study with red-cockaded woodpeckers. Landsc. Ecol. 28 , 755–767 (2013). Keeley, A. T. H., Beier, P. & Gagnon, J. W. Estimating landscape resistance from habitat suitability: effects of data source and nonlinearities. Landsc. Ecol. 31 , 2151–2162 (2016). Marrec, R. et al. Conceptual framework and uncertainty analysis for large- scale , species-agnostic modelling of landscape connectivity across. Sci. Rep. 1–14 (2020) doi:10.1038/s41598-020-63545-z. Dickson, B. G. et al. Informing Strategic Efforts to Expand and Connect Protected Areas Using a Model of Ecological Flow, with Application to the Western United States. Conserv. Lett. 10 , 564–571 (2017). Schloss, C., Cameron, Ri. C., Ae, B. R. A. D. H. M. C. R., Heobald, D. A. M. T. & Ones, A. A. J. “ No-regrets ” pathways for navigating climate change : planning for connectivity with land use , topography , and climate. 32 , 1–11 (2022). Pither, R., O’Brien, P., Brennan, A., Hirsh-Pearson, K. & Bowman, J. Predicting areas important for ecological connectivity throughout Canada. PLoS One 18 , 1–21 (2023). Lamarche, C. et al. Compilation and Validation of SAR and Optical Data Products for a Complete and Global Map of Inland/Ocean Water Tailored to the Climate Modeling Community. Remote Sens. 9 , (2017). Whitmee, S. & Orme, C. D. L. Predicting dispersal distance in mammals: A trait-based approach. J. Anim. Ecol. 82 , 211–221 (2013). Riggio, J. et al. Global human influence maps reveal clear opportunities in conserving Earth’s remaining intact terrestrial ecosystems. Glob. Chang. Biol. 26 , 4344–4356 (2020). R, H. terra: Spatial Data Analysis. (2025). UNEP-WCMC & IUCN. UNEP-WCMC and IUCN. August 2019. Protected Planet: The World Database on Protected Areas (WDPA), August 2019 version, Cambridge, UK: UNEP-WCMC and IUCN. Protected Planet: The World Database on Protected Areas (WDPA) (2023). Allan, J. R. et al. attention to safeguard biodiversity. 1101 , 1–7 (2022). R Core Team. R: A Language and Environment for Statistical Computing. 2025 (2025). Additional Declarations There is NO Competing Interest. Supplementary Files Supplemental.docx Supplemental Information Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6457802","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":455344718,"identity":"7cc81214-8540-4001-ace1-1b1e2f3ef896","order_by":0,"name":"Erin Poor","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAxklEQVRIiWNgGAWjYFACxgYIzd7/TeIHg009fwPRWngOtkv2MKQlzjhAtG0Sie2zWRgOJ24gpJBf+nDbA8YdNnLmMx+2z2YgRotkX2K7AeOZNGOZ24ltt4Fa6g0IaTE4w9gmwdh2OHGGNERLjeEBAlrs4VokD7ZVA7WkGRDSYsAD0wIkrYHeTyaoRQJkSyLQLxI8iW2WPQZEBDJ/D/sziY/AEJNgP9gm8aPCJrG/gYAWMEiEqyIYXDAASwCjYBSMglEwCrACAG0QQuYpFMl+AAAAAElFTkSuQmCC","orcid":"https://orcid.org/0000-0002-8799-3193","institution":"Tackle Climate Change, The Nature Conservancy","correspondingAuthor":true,"prefix":"","firstName":"Erin","middleName":"","lastName":"Poor","suffix":""},{"id":455344719,"identity":"8dafe771-8b87-490b-afbe-9f2f57f3b2ea","order_by":1,"name":"Kimberly Hall","email":"","orcid":"","institution":"The Nature Conservancy","correspondingAuthor":false,"prefix":"","firstName":"Kimberly","middleName":"","lastName":"Hall","suffix":""},{"id":455344720,"identity":"eb3acb5a-1a56-43e7-84cb-b4a01246e97a","order_by":2,"name":"Jesse Anderson","email":"","orcid":"","institution":"Numbers and Strings","correspondingAuthor":false,"prefix":"","firstName":"Jesse","middleName":"","lastName":"Anderson","suffix":""},{"id":455344721,"identity":"148da016-656e-4b64-b2ff-34932f976f39","order_by":3,"name":"Melissa Clark","email":"","orcid":"","institution":"Center for Resilient Conservation Science, The Nature Conservancy","correspondingAuthor":false,"prefix":"","firstName":"Melissa","middleName":"","lastName":"Clark","suffix":""},{"id":455344722,"identity":"478f10fc-3674-4acb-ac70-bc7fbb798721","order_by":4,"name":"Aaron Jones","email":"","orcid":"https://orcid.org/0009-0004-0250-8032","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Aaron","middleName":"","lastName":"Jones","suffix":""},{"id":455344723,"identity":"9d67873f-b76b-4e59-9e28-4690788ff485","order_by":5,"name":"Christina Kennedy","email":"","orcid":"https://orcid.org/0000-0001-8902-8728","institution":"Nature Conservancy","correspondingAuthor":false,"prefix":"","firstName":"Christina","middleName":"","lastName":"Kennedy","suffix":""},{"id":455344724,"identity":"e6efc893-6218-44b3-a493-1fb19471722b","order_by":6,"name":"Carrie Schloss","email":"","orcid":"","institution":"The Nature Conservancy","correspondingAuthor":false,"prefix":"","firstName":"Carrie","middleName":"","lastName":"Schloss","suffix":""},{"id":455344725,"identity":"a5f1c76f-da62-4e81-be34-49914ee90bf1","order_by":7,"name":"David Theobald","email":"","orcid":"","institution":"Conservation Planning Technologies","correspondingAuthor":false,"prefix":"","firstName":"David","middleName":"","lastName":"Theobald","suffix":""},{"id":455344726,"identity":"12285f5b-9a7c-47e5-8668-ed8066d6dc06","order_by":8,"name":"Jaime Burbano-Giron","email":"","orcid":"","institution":"The Nature Conservancy","correspondingAuthor":false,"prefix":"","firstName":"Jaime","middleName":"","lastName":"Burbano-Giron","suffix":""},{"id":455344727,"identity":"92a279cb-97af-4348-833b-a4b6cfcaa82a","order_by":9,"name":"Susan Cook-Patton","email":"","orcid":"https://orcid.org/0000-0002-7194-4397","institution":"TNC","correspondingAuthor":false,"prefix":"","firstName":"Susan","middleName":"","lastName":"Cook-Patton","suffix":""},{"id":455344728,"identity":"f30aeccf-a99f-430c-98c7-4ae9d7ef74a1","order_by":10,"name":"Ronnie Drever","email":"","orcid":"https://orcid.org/0000-0001-5338-3496","institution":"Nature United","correspondingAuthor":false,"prefix":"","firstName":"Ronnie","middleName":"","lastName":"Drever","suffix":""},{"id":455344729,"identity":"59f0e675-1fa4-4f89-ba1b-e2a2489f4232","order_by":11,"name":"Joseph Fargione","email":"","orcid":"","institution":"The Nature Conservancy","correspondingAuthor":false,"prefix":"","firstName":"Joseph","middleName":"","lastName":"Fargione","suffix":""},{"id":455344730,"identity":"05fd6e10-0d2a-4c62-b36f-b1beceab1ab2","order_by":12,"name":"James Fitzsimons","email":"","orcid":"https://orcid.org/0000-0003-4277-8040","institution":"The Nature Conservancy","correspondingAuthor":false,"prefix":"","firstName":"James","middleName":"","lastName":"Fitzsimons","suffix":""},{"id":455344731,"identity":"3e6049ae-f7f2-4e53-a21c-f62d4a8e81fb","order_by":13,"name":"Mauricio Nunez-Regueiro","email":"","orcid":"","institution":"The Nature Conservancy","correspondingAuthor":false,"prefix":"","firstName":"Mauricio","middleName":"","lastName":"Nunez-Regueiro","suffix":""},{"id":455344732,"identity":"33dee711-8d04-4e93-b165-f774a028339a","order_by":14,"name":"Milena Rosenfield","email":"","orcid":"","institution":"The Nature Conservancy","correspondingAuthor":false,"prefix":"","firstName":"Milena","middleName":"","lastName":"Rosenfield","suffix":""},{"id":455344733,"identity":"88745eb3-d02f-408f-83b8-ad46973a7797","order_by":15,"name":"Yuta Masuda","email":"","orcid":"https://orcid.org/0000-0002-1698-4855","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Yuta","middleName":"","lastName":"Masuda","suffix":""}],"badges":[],"createdAt":"2025-04-15 20:25:09","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6457802/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6457802/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":82680591,"identity":"8f232735-b869-4551-a222-3fcc0fcb9a0a","added_by":"auto","created_at":"2025-05-14 05:39:02","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":7172684,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eCategorized global connectivity between natural vegetation, given human-driven land modification. \u003c/strong\u003eIdeal connectivity can be considered ‘robust diffuse’ and efforts to restore connectivity should focus on channelized or impeded areas, depending on local context. The most common categories are subdivided into low/high classes for increased ease of visualization.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-6457802/v1/15b642f29d8866c229dca29b.png"},{"id":82680598,"identity":"3569a809-63e9-4aca-8baa-cf89ce73f927","added_by":"auto","created_at":"2025-05-14 05:39:03","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":12584017,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eCategorized connectivity in four geographies. \u003c/strong\u003eGlobal categorized connectivity data show areas where there are large amounts of moderate levels of land modification like the midwestern agricultural lands of the United States (a) and in populated areas such as in India (b). Movement potential is higher in areas constrained by land modifications, such as the North American southeast (a) and southeastern Kenya (c). Movement potential is at moderate levels where there are wide areas of natural lands with low levels of land modification, such as in the Amazon (d). The most common categories are subdivided into low/high classes for increased ease of visualization.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-6457802/v1/7e37fa8695c641c64ae7e75b.png"},{"id":82680596,"identity":"f824926c-5701-473b-9365-d2150de739a1","added_by":"auto","created_at":"2025-05-14 05:39:03","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":794145,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eConnectivity categories across global biomes. \u003c/strong\u003eProportionate distribution of connectivity categories by biome\u003csup\u003e40\u003c/sup\u003e, ordered by proportion of impeded movement areas. Global temperate broadleaf \u0026amp; mixed forests and mediterranean forests, woodlands \u0026amp; scrub have low connectivity while tundra has high amounts of diffuse movement potential.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-6457802/v1/cc944006d40671eaaedd7027.png"},{"id":82680602,"identity":"d4ff8618-7eca-4855-aa01-f5acbf765355","added_by":"auto","created_at":"2025-05-14 05:39:03","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":7432254,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eConnectivity categories across countries and level of protection. \u003c/strong\u003eDiffuse movement areas dominate Africa while much of Europe has impeded movement potential and high proportions of channelized movement areas. Within continents, countries are ordered from most to least impeded movement areas, and black lines represent quantiles. Continents: AF: Africa; AS: Asia; EU: Europe; NA: North America; OC: Oceania; SA: South America.\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-6457802/v1/b961492bea736b0cccb5df6f.png"},{"id":82680588,"identity":"42d5e668-6adb-4d9d-b8c3-e37e0b9aad62","added_by":"auto","created_at":"2025-05-14 05:39:00","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":3500374,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eApplication of categorized connectivity data to the Global Biodiversity Framework. \u003c/strong\u003eContributions to Targets 2 and 3 of the Kunming-Montreal Global Biodiversity Framework\u003cstrong\u003e \u003c/strong\u003ecan partially be informed using global categorized connectivity data as shown by example landscapes. Diffuse areas represent redundancies in movement areas, channelized areas represent areas of constrained movement potential, and impeded areas indicate low movement potential.\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-6457802/v1/3e1ff2a5ee150ab315846261.png"},{"id":82681702,"identity":"1792aea5-b874-4fa0-8c41-cb749d019985","added_by":"auto","created_at":"2025-05-14 06:03:03","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":1070723,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eLevel of diffuse and channelized movement areas and level of protection within countries. \u003c/strong\u003eMany large countries have moderate levels of protection and higher levels of diffuse movement areas, where increased protection could be targeted (Brazil, Australia, and China, for example). Small countries with high levels of channelization and moderate levels of protection may target protection to preserve remaining corridors or restore areas.\u003c/p\u003e","description":"","filename":"6.png","url":"https://assets-eu.researchsquare.com/files/rs-6457802/v1/83816cf05ebf2453dac02928.png"},{"id":107482094,"identity":"731abbee-9477-4898-82c1-ff453a90601e","added_by":"auto","created_at":"2026-04-22 02:21:51","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":31180385,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6457802/v1/5b861632-03a8-4231-8e82-2936256a6cae.pdf"},{"id":82680587,"identity":"b0cc8a81-a565-416b-8b5f-139146e385d4","added_by":"auto","created_at":"2025-05-14 05:39:00","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":2596529,"visible":true,"origin":"","legend":"Supplemental Information","description":"","filename":"Supplemental.docx","url":"https://assets-eu.researchsquare.com/files/rs-6457802/v1/644f5003ff5a3c213d6320ab.docx"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Global whole landscape connectivity to complement protected area connectivity","fulltext":[{"header":"INTRODUCTION","content":"\u003cp\u003eThe biodiversity crisis is fueled by rapid losses in the extent and integrity of ecosystems due to human-driven land modification and climate change\u003csup\u003e\u003cspan additionalcitationids=\"CR2 CR3 CR4 CR5\" citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e. Ecological connectivity, defined as the \u0026ldquo;unimpeded movement of species and flow of natural processes\u0026rdquo; \u003csup\u003e\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u003c/sup\u003e, is critical to ecosystem and species conservation and has been established as a key conservation component in the Convention on Biological Diversity (CBD)\u0026rsquo;s Kunming-Montreal Global Biodiversity Framework (GBF)\u003csup\u003e\u003cspan additionalcitationids=\"CR9\" citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e. Yet despite the critical importance of connectivity across ecosystems and landscapes, data and guidance to fully support the implementation of connectivity-related targets within the GBF are lacking or incomplete\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e. Most of the work on developing indicators for monitoring connectivity has focused on protected areas (PA), which has helped to raise awareness of the importance of ecological connectivity\u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u003c/sup\u003e. However, addressing the coupled biodiversity and climate crises will require a \u0026ldquo;whole landscape\u0026rdquo; approach to connectivity, where all lands, regardless of their protection status and ecological condition, are considered and valued appropriately\u003csup\u003e\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eUnder the GBF, 195 countries aim to meet four long-term biodiversity conservation goals by 2050. The first, Goal A, targets that \u0026ldquo;the integrity, connectivity, and resilience of all ecosystems are maintained, enhanced, or restored\u0026rdquo; by 2050. Under Goal A, signatories aim to achieve multiple targets by 2030. Targets 1, 2, and 3 are related to spatial planning, restoration, and protection of ecosystems, and GBF guidance provides multiple indicators for measuring progress towards meeting each target. Target 3 contains particular emphasis on connectivity, stating that PAs and other effective conservation measures (OECMs) should be connected through corridors and \u0026ldquo;integrated into wider landscapes, seascapes and the ocean\u0026rdquo;. A PA is \u0026ldquo;a geographically defined area, which is designated or regulated and managed to achieve specific conservation objectives\u0026rdquo;\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e. The current set of proposed indicators only provide methods for evaluating linkages between PAs and OECMs\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e, yet, cost-effective actions will require integrated approaches that carefully address all three targets. Despite the focus on PAs in Target 3, there are many unprotected areas that must be considered that are potentially relevant for connectivity\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eWhile PAs are a necessary conservation tool, the current global PA network is geographically and ecologically biased toward areas that are inherently difficult to access in steeper, higher, or more remote terrain and are less populated\u003csup\u003e\u003cspan additionalcitationids=\"CR14 CR15\" citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e. Focusing indicators solely on corridors between PAs will result in biased assessments of areas deemed important for connectivity. Instead, the importance of connectivity in whole landscapes, independent of the amount or distribution of PAs they contain, is needed to sustain biodiversity and achieve the ecological representativeness proposed in the GBF targets\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e,\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e,\u003cspan additionalcitationids=\"CR18\" citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u003c/sup\u003e. Identifying areas of connectivity importance outside of the current PA system can inform where future additions to the PA and OECM network could contribute to conservation goals. For example, Brennan et al. (\u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e) identified globally important areas for the movement of mammals between PAs, but some areas known to harbor high biodiversity (United States Appalachian region, Madagascar, India, and Southeast Asia\u003csup\u003e\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e) were identified as having low connectivity owing to a dearth of globally registered PAs in these regions.\u003c/p\u003e \u003cp\u003eDespite the need, a gap remains in modeling connectivity across ecosystems beyond specific habitat patches or protected areas, in part driven by computational limitations. This gap may be addressed by a variety of well-vetted theoretical frameworks, models, and tools to identify species corridors that are commonly used for habitat and population management at the landscape scale\u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e,\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e. Of these, Omniscape, an omnidirectional implementation of circuit-theory modeling, has shown promise to assess whole landscape connectivity at continental scales. In Omniscape, every pixel, not just PAs or habitat patches, can act as a source for potential species movement\u003csup\u003e\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e,\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e. These models can thus be applied to any land cover map, independent of political or geographic boundary and may thus be more ecologically accurate, as it allows modelers to represent gradients in ecological condition and does not infer that species recognize human-designed political or geographic boundaries. The result is a continuous dataset representing potential movement across a study area, which until now has been hindered by computational requirements that limited the geographic scale at which these models can be run\u003csup\u003e\u003cspan additionalcitationids=\"CR25\" citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eThe outputs of Omniscape typically represent movement potential of species from habitat, or \u0026ldquo;source\u0026rdquo; pixels across landscapes, with varying assumed resistance to movement (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e; \u003csup\u003e\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e,\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u003c/sup\u003e). Model outputs represent potential movement across a landscape, where high values represent pinch points along pathways and low values represent barriers to movement or a lack of movement sources\u003csup\u003e\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e. Although high movement areas are often interpreted as corridors and treated as conservation priorities, additional translation is needed to turn continuous outputs into meaningful information for prioritization of conservation efforts. Importantly, moderate connectivity values often characterize highly intact natural landscapes where movement is unconstrained, yet these areas are not usually the focus of connectivity outputs.\u003c/p\u003e \u003cp\u003eTo ease the use and interpretation of connectivity modeling in policy applications, the continuous model outputs must then be translated into products that support decision-making\u003csup\u003e\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e,\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e. One approach for facilitating interpretation of results is to compare potential movement to a null model with uniform resistance, thereby determining how connectivity changes in relation to the resistance data\u003csup\u003e\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e. Represented by the normalized output layer (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e), this comparison allows modelers to determine where connectivity has been lost, with respect to the resistance data such as human modification, for example. Another approach is to categorize model outputs. To date, categorization has not been a standard post-processing step, though breaking continuous outputs into categories could help users understand and interpret the patterns of movement potential. For example, Cameron et al. (\u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e) categorized connectivity data across California based on management and land tenure status to prioritize sites for conservation. More recently, Anderson et al. (\u003csup\u003e\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e) mapped climate connectivity for the continental United States and categorized flow based on the distribution and variation of the model outputs, and combined those products with other data to delineate an expansion of the US PA network.\u003c/p\u003e \u003cp\u003eWe build on this body of work and overcome prior computational hurdles to produce a map of connectivity across all terrestrial landscapes globally. These data represent the potential movement across natural vegetation as determined by levels of human modification at 1 km resolution. We capitalized on increased cloud processing capabilities and the high-efficiency language Julia to run Omniscape globally\u003csup\u003e\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e,\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u003c/sup\u003e. We also improved methods for handling water bodies\u003csup\u003e\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e\u003c/sup\u003e in the model and incorporated updated and tailored global human modification data that comprehensively accounts for industrial human activities expected to impact movement patterns\u003csup\u003e\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u003c/sup\u003e. We developed six categories of connectivity that can be used to evaluate the extent to which potential movement (connectivity) is impeded, channelized, or diffuse (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Our approach allowed us to produce maps that illustrate how the spatial pattern and intensity of human impacts impede or channel movement and indicate natural areas where connectivity is most likely to be unconstrained. This work is broadly applicable to inform global conservation goals, such as those enumerated in the GBF. To demonstrate its utility, we provide summaries to inform global, national, and sub-national decision making and show how this whole landscape connectivity approach can be used to support global biodiversity goals.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eDefinitions of terms for inputs, outputs, and categories from the global Omniscape connectivity modeling process, with corresponding color coding used to display connectivity categories in subsequent figures.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eModel Component\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTerm\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003eDefinition\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"1\" nameend=\"c5\" namest=\"c5\"\u003e\u0026nbsp;\u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eInputs\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSource\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003eThe presumed capacity of a site to function as a source for species or gene movement (in this study, all natural lands; \u003csup\u003e\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u003c/sup\u003e).\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"1\" nameend=\"c5\" namest=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eResistance\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003eLandscape impedance to movement potential, derived in this study from a measure of human modification\u003csup\u003e\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"1\" nameend=\"c5\" namest=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003eOutputs\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eFlow, \u003cem\u003eF\u003c/em\u003e (current)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003eThe representation of species movement (movement potential) or genetic migration probabilities from source areas across a resistance dataset.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"1\" nameend=\"c5\" namest=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePotential flow, \u003cem\u003eF\u003c/em\u003e\u003csub\u003e\u003cem\u003ep\u003c/em\u003e\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003eA null model of flow that indicates movement without resistance. Flow potential allows us to ask: given the amount and configuration of source pixels, how much flow would be expected in the absence of barriers\u003csup\u003e\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e?\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"1\" nameend=\"c5\" namest=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNormalized flow, \u003cem\u003eF\u003c/em\u003e\u003csub\u003e\u003cem\u003en\u003c/em\u003e\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003eThe ratio of flow to potential flow. Normalized flow enables more accurate identification of the mechanisms underpinning different flow magnitudes such as where barriers are redirecting flow\u003csup\u003e\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"1\" nameend=\"c5\" namest=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"7\" rowspan=\"8\"\u003e \u003cp\u003eOutput categories\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eChannelized movement area\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eHigh flow in relatively narrow pathways near distinct barriers of high resistance. Resistance levels in these areas can vary from low to high; channelization indicates increasing importance of these sites as other options have been lost.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\" morerows=\"3\" rowspan=\"4\"\u003e \u003cp\u003eDiffuse movement area\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003eAreas where flow patterns are similar to what would be expected with a homogeneous surface of low resistance. This includes:\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"1\" nameend=\"c5\" namest=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003eIntensified areas\u003c/b\u003e: med-high flow areas, adjacent to higher resistance areas from which flow has been redirected.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003eRobust areas\u003c/b\u003e: moderate flow across low resistance; many redundant options for movement, indicating natural areas with the least disruption of flow.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003eDampened areas\u003c/b\u003e: lower flow areas within a low resistance neighborhood. This reduction in current flow relative to diffuse can indicate a slight increase in resistance, or lower source strengths.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003eImpeded movement area\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003eRestriction and redirection of flow away from these locations due to high resistance. This includes:\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"1\" nameend=\"c5\" namest=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003eWeak areas\u003c/b\u003e: low flow across minimally permeable lands with medium-high to high resistance\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003eObstructed areas\u003c/b\u003e: highly restricted flow, reflecting redirection to other areas by high resistance\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e"},{"header":"RESULTS","content":"\u003cp\u003eTo promote connectivity data in decision making and conservation planning, we developed a framework categorizing complex, continuous connectivity data to policy-relevant outputs; we then applied this framework to produce a global map describing potential movement patterns across all lands (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, Fig. \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e). The categories derive from classifying Omniscape model outputs (the flow (\u003cem\u003eF\u003c/em\u003e), flow potential (\u003cem\u003eF\u003c/em\u003e\u003csub\u003e\u003cem\u003ep\u003c/em\u003e\u003c/sub\u003e), normalized flow (\u003cem\u003eF\u003c/em\u003e\u003csub\u003e\u003cem\u003en\u003c/em\u003e\u003c/sub\u003e) datasets; Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e) and the resistance data (levels of land modification\u003csup\u003e\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u003c/sup\u003e) to create six categories of potential movement that describe the unique combinations of these layers and the flow patterns and magnitudes given the human modification resistance layer (\u003cem\u003eMethods;\u003c/em\u003e Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eIn the text below, we condensed the six categories to \u0026lsquo;channelized\u0026rsquo;, \u0026lsquo;diffuse\u0026rsquo; (i.e., \u0026lsquo;dampened\u0026rsquo;, \u0026lsquo;robust\u0026rsquo;, and \u0026lsquo;intensified\u0026rsquo;), and \u0026lsquo;impeded\u0026rsquo; (i.e., \u0026lsquo;obstructed\u0026rsquo; and \u0026lsquo;weak\u0026rsquo;) categories for brevity, although figures and tables present all six categories. Land falling within the diffuse category can be considered the goal for connectivity \u0026ndash; these are areas with redundancy in movement pathways that afford multiple options for movement and where human pressures have not greatly changed how species may move. Areas with impeded movement potential occur where there is likely a high level of land modification and where potential movement is much lower than expected (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). Channelized movement potential is the highest movement potential (i.e. \u0026ldquo;corridors\u0026rdquo;), likely where surrounding land modification has forced flow through an area with lower human development, resulting in much higher movement potential than expected in the absence of modification.\u003c/p\u003e \u003cp\u003eWe find that 66% of natural lands globally has diffuse potential movement, 29% has low potential movement (i.e. movement is \u0026lsquo;impeded\u0026rsquo;), and 5% has channelized movement. When comparing the precent diffuse per country to other global connectivity indicators (PAI\u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e, PARC\u003csup\u003e\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u003c/sup\u003e, ProNet\u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e, and ProtConn\u003csup\u003e\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e), as recommended in the GBF and with 2024 updates\u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e, we found low correlations across each indicator. Percent diffuse is most correlated with PARC at \u003cem\u003er\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.68. Other correlations with percent diffuse range from \u0026minus;\u0026thinsp;0.42 to -0.11 (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003e\u003cb\u003ePearson correlation coefficients across connectivity indices.\u003c/b\u003e Percent of diffuse connectivity category per country (\u0026gt;\u0026thinsp;10,000 km\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e) vs. Protected Area Index\u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e (PAI), Protected Area Representativeness and Connectedness\u003csup\u003e\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u003c/sup\u003e (PARC), Protected Areas Network\u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e (ProNet), and Protected Connected\u003csup\u003e\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e (ProtConn) connectivity indicators as updated in UNEP-WCMC and IUCN (\u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e).\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"6\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003e% Diffuse\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003ePAI\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003ePARC\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eProNet\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eProtConn\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003e% Diffuse\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-0.42\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.68\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-0.37\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e-0.11\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003ePAI\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-0.42\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-0.49\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e-0.17\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003ePARC\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.68\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-0.49\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-0.30\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.21\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eProNet\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-0.37\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-0.30\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.52\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eProtConn\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-0.11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-0.17\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.21\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.52\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eConnectivity patterns in biomes and continents\u003c/h2\u003e \u003cp\u003eGlobally, temperate broadleaf and mixed forests have the highest proportion of impeded movement potential (have lost the most connectivity), followed by mediterranean forest, woodlands, and scrub (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e; Table S2). Tundra, boreal, and tropical moist forest biomes have the highest proportion of diffuse movement potential areas. Across biomes, Mediterranean forests, woodlands, and scrub have the highest proportion of channelized areas (6%) and the lowest amount of diffuse areas (58%; Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e; Table S2), reflecting high levels of land modification. Deserts and xeric shrublands have the least amount of channelized movement areas (93% diffuse), likely because these areas also have lower land modification. Tundra and boreal forests/taiga both have 97% diffuse movement areas, followed by montane grasslands and shrublands. Tundra and boreal areas also have the least amount of impeded areas.\u003c/p\u003e \u003cp\u003eIn general, across continents Africa has the highest proportion of diffuse movement areas and the lowest proportion of channelized movement areas (91% and 1%, respectively); whereas Asia has the lowest proportion of diffuse movement areas and highest level of channelized movement areas (74% and 4%, respectively). Asia has the highest and Oceania has the lowest proportion of impeded areas (23% and 7%, respectively; Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e; Table S3). In addition, the biome-level patterns of low connectivity (indicated by proportion impeded) varied by continent, with Europe and North America exhibiting high levels of impeded areas in temperate grasslands relative to other continents.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eConnectivity patterns in countries\u003c/h3\u003e\n\u003cp\u003eUnsurprisingly, smaller countries with a large proportion of land modification have the greatest percentage of impeded areas (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e, Table S4). The Netherlands, Belgium, Moldova, and Denmark all have the highest proportion of impeded movement area (\u0026ge;\u0026thinsp;85%) and the lowest proportion of diffuse movement areas (\u0026le;\u0026thinsp;14%). In contrast, relatively large countries with low levels of land modification and with expansive natural land cover have the highest proportion of diffuse movement areas. This includes many nations across Africa and some South American countries, such as Peru, Bolivia, and Colombia (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e; Table S4).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e\n\u003ch3\u003eApplying the framework to support the Global Biodiversity Framework\u003c/h3\u003e\n\u003cp\u003eThe categorized connectivity data can be combined with other datasets (i.e. species occurrences or habitats, ecosystem services, etc.\u003csup\u003e\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e\u003c/sup\u003e), facilitating consideration of connectivity in global and national-level biodiversity commitments. We illustrate conceptually how our connectivity data and categories can be used to inform GBF Goal A Targets 1 (on spatial planning), 2 (on restoration), and 3 (on protection), though there are uses and applications beyond these targets (for example, Target 4, focusing on preventing species extinction\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e). We do not attempt to identify specific sites for conservation measures, as this process should be done in collaboration with local communities, consider competing land uses and multiple actions, and incorporate local context, land tenure, and inclusion of equity considerations in line with rights-based approaches to conservation.\u003c/p\u003e \u003cp\u003e \u003cstrong\u003eTarget 1 (spatial planning)\u003c/strong\u003e \u003cp\u003eOur data can inform biodiversity-inclusive spatial conservation planning, which recognizes the importance of considering ecological connectivity in the design of conservation networks within landscapes or seascapes\u003csup\u003e\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e\u003c/sup\u003e. Alternative spatial scenarios can be evaluated based on their ability to improve connectivity, for example by increasing protection of channelized areas that represent the last remaining pathways of connectivity and to increase movement options through targeted restoration as described in Target 2 (see below). These represent areas important for maintaining or enhancing connectivity across whole landscapes \u0026ndash; not just between PAs \u0026ndash; and can facilitate the consideration of whole landscapes in global and regional spatial connectivity planning.\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003eTarget 2 (restoration)\u003c/strong\u003e \u003cp\u003eIdentifying where connectivity has been lost, as identified by the weak, obstructed, or channelized areas \u0026mdash; in concert with other information such land tenure, competing land uses, biophysical suitability, and feasibility for ecological recovery \u0026mdash; can guide site selection and efficient allocation of resources for restoration. Restoring areas close to channelized flow areas can help preserve and enhance connectivity and restoring areas in weak or obstructed flow may help restore connectivity, depending on the cause of loss (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e).\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003e \u003cem\u003eTarget 3 (protection)\u003c/em\u003e: This data can also be used to identify priority sites for future conservation measures, including siting of new PAs and OECMs. In overlaying the categorized connectivity data with global PA locations, we found that countries with greater proportions of diffuse connectivity areas tend to have lower levels of protection, but more concerningly, there are many countries with low levels of protection and high levels of impeded movement areas (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e; Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e). These data allow users to better understand the landscape context of a PA. For example, if a country\u0026rsquo;s PAs primarily fall within diffuse areas, biodiversity may benefit from conservation measures focused in channelized areas. If PAs are surrounded by impeded areas, future conservation efforts may focus on any channelized areas between PAs (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003cp\u003eFor example, Australia has reached approximately 20% protection\u003csup\u003e15\u003c/sup\u003e with the majority of PAs occurring in the desert and xeric shrublands biome and in diffuse areas (i.e. are connected). To advance reaching both Targets 2 and 3, concentrating future conservation in more modified areas may be advantageous from the point of view of improving connectivity and providing potential additional benefit to both people and wildlife, rather than focusing conservation where areas are already well-connected. Future siting of PAs, OECMs, or conservation and management actions could be targeted at remaining channelized areas within modified areas to maintain existing connections. If restoration is feasible, these efforts could be focused near currently protected channelized areas to enhance these connections and increase their resilience.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e"},{"header":"DISCUSSION","content":"\u003cp\u003eNearly all countries agreed to ambitious goals and targets in the GBF that build on decades of science and advocacy. Restoration, protection, and ecosystem connectivity are foundational to Goal A within the GBF. Yet, to date the technical tools and evidence on how best to consider and integrate connectivity across whole landscapes have been lacking. We fill this gap by conducting a global connectivity analysis which considers broad-scale connectivity across entire landscape gradients and providing a framework for analyzing the complex outputs that is scientifically robust and policy relevant. This framework can be applied across scales to meet landscape-level needs with high resolution and country/region-specific data.\u003c/p\u003e \u003cp\u003eWe find that when whole landscapes are considered, the majority of Earth\u0026rsquo;s terrestrial areas lie within connected zones (66% of terrestrial lands have diffuse movement potential). Conservation efforts should focus on protecting these well-connected and channelized areas, since many countries with high amounts of diffuse movement areas have low levels of protection (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e). In addition, restoring impeded and channelized areas (areas which have lost connectivity due to human-driven land modifications) should also be a conservation goal. With a growing human population and ongoing development expansion, diffuse movement areas could become channelized or impeded without continued or improved conservation actions. This information should be used in concert with conversion risk mapping (e.g.,\u003csup\u003e\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e\u003c/sup\u003e) and other connectivity measures. When whole landscapes are not included in connectivity areas, unprotected but biologically important areas could be left out of connectivity planning, leading to reduced species and ecosystem persistence over time\u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eThe categorization framework we developed strengthens the ability to communicate the importance of connectivity for conservation planning in two ways. First, it recognizes at least two ways that connectivity can be compromised: it can be impeded, in which movement is reduced or lost compared to unmodified landscapes, or it can be channelized, in which movement in a channel is increased even as it is reduced in other areas of the landscape. Channelized systems have lost redundancy in movement options that provide resilience to a system and increase the risk of complete loss of connectivity in the face of future land use and climate changes. The second advantage of this categorization derives from explicitly evaluating the site-specific resistance values, which enables us to model the impact of site-specific interventions on whole landscape connectivity.\u003c/p\u003e \u003cp\u003eThe low correlation of diffuse movement areas per country with other connectivity measures suggests that our data provides complementary information (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). Indeed, our results differ from prior connectivity analyses in several important ways. First, analyses that model connectivity between PAs miss connections elsewhere. If not used in concert with non-PA based connectivity measures, these results will bias connectivity conservation towards areas of high PA occurrence, the global pattern of which is geographically and ecologically biased in remote areas potentially jeopardizing important but unprotected areas for both nature and people \u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e,\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e. By using natural land cover as our source data and modeling connectivity across all lands irrespective of conservation status, our data depict the ecological connectivity regardless of PA distribution. For example, the Appalachian region of the eastern United States has a low density of PAs but has swathes of channelized areas which are remaining connections running along mountain ridges in a north-south orientation (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). These areas are potentially important for maintaining connectivity for wildlife as climate changes. Yet, these ecologically important areas are not identified in other connectivity assessments\u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e,\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e,\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e\u003c/sup\u003e. In addition, our approach and results remain robust to sociopolitical instability resulting in PA downgrading, degazettement, or degradation or when PAs are unreported\u003csup\u003e\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eBiased protection is also reflected in the distribution of connectivity categories within biomes. For instance, temperate broadleaf and mixed forests and grasslands, mediterranean forests, woodlands, and scrub, and tropical and subtropical dry broadleaf forests all have high proportions of impeded lands and low proportions of land protection\u003csup\u003e\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e\u003c/sup\u003e. Our results underscore the need for enhanced conservation measures in these biomes. A lack of PA prevalence in these biomes may direct conservation elsewhere when relying on PA-based connectivity analyses. Lack of protection and PA-centric consideration of connectivity in grasslands, for example, can be seen in the midwestern United States, southern Brazil, and India. These areas have moderate levels of human land pressures but remain important for maintaining connectivity across broad landscapes.\u003c/p\u003e \u003cp\u003eAt a continental scale, there are areas identified as highly connected when solely examining distance between PAs (e.g., ProtConn \u003csup\u003e\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e\u003c/sup\u003e), even though these areas have low proportions of diffuse and high proportions of impeded areas. If we use the percent of diffuse area as a measure of landscape connectivity, Africa ranks the highest globally, followed by Oceania (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e; Table S3). However, most connectivity indicators have been calculated assuming all areas outside of PAs are equally inhospitable (e.g.,\u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e,\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e\u003c/sup\u003e). With this assumption, Europe is the most connected continent, with the Americas ranking second. This difference stems from geographically clustered PAs in the EU and Americas, within landscapes of moderate-to-high levels of human modification, resulting in impeded lands between PAs. PAs may be geographically near one another, but movement between PAs may not be likely given the surrounding landscape. Many countries with low connectivity as indicated by ProtConn\u003csup\u003e\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e\u003c/sup\u003e were dominated by the diffuse category in our analysis, indicating redundancy in movement pathways throughout most of these countries\u0026rsquo; land areas (Mexico, Panam\u0026aacute;, Colombia, Ecuador, Madagascar, Papua New Guinea, Nepal; Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, Table S4). Many of these countries have large amounts of diffuse connectivity through natural lands that remain unprotected. Our results are similar to the findings of Ward et al. (\u003csup\u003e\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e\u003c/sup\u003e), where Europe had the lowest PA connectivity, when measured by intact lands as defined by levels of human pressure.\u003c/p\u003e \u003cp\u003eOur work enables evaluation of the effectiveness of the existing PA networks in ensuring connectivity. Countries with high amounts of diffuse under the 30% target should identify whether there are unprotected channelized areas (bottom right, Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e). Countries with low amounts of diffuse areas, moderate protection, and high amounts of channelized areas should ensure diffuse areas are protected to maintain connectivity redundancy (center of the graph, Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e). Countries with low amounts of diffuse areas, low proportions of protection, and low amounts of channelized areas likely have high land modification and conservation efforts may be better suited for non-area-based conservation or ecosystem restoration measures, such as many countries in the EU (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e). The categorized connectivity data allows jurisdictions to identify key locations for restoration to bolster existing connectivity.\u003c/p\u003e \u003cp\u003eAnother major difference between our work from others stems from the theoretical approach to connectivity underpinning our modelling. Our work represents connectivity across natural land cover considering the gradient of anthropogenic land modification, regardless of species or taxa, and may be considered structural connectivity. Where prior connectivity analyses have taken a particular species or taxa into account, results depict areas important for that species or taxa and must be used in concert with other species and ecosystem condition information to inform global biodiversity goals. For example, Brennan et al. (\u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e) show low connectivity across much of Southeast Asia, likely reflecting the regional-scale loss of large mammals or the lack of large mammal movement data in this region\u003csup\u003e\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e,\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e\u003c/sup\u003e, rather than this area\u0026rsquo;s inherent connectivity value. When modeling connectivity across all natural land cover, we found impeded, diffuse, and potentially important channelized areas across South and Southeast Asia, including the Indian subcontinent (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). India is dominated by impeded areas and the remaining channelized areas could be conserved or managed to maintain the existing connectivity (Table S2).\u003c/p\u003e \u003cp\u003eThe differences in our work compared with others confirm the complexity and contextuality of measuring ecological connectivity. Multiple measures should be used to assess and track global connectivity in complementarity to capture the multiple aspects of connectivity; both between PAs and OECMs and to the wider landscape\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e. We modeled connectivity between terrestrial natural vegetation, not connectivity for any one species or type of geography. We assumed that human pressures, as mapped in the human modification data and modified here to reflect land cover alterations impacting migration, represent meaningful resistance from anthropogenic barriers to movement for many species\u003csup\u003e\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e,\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e\u003c/sup\u003e. While we did not attempt to define specific mechanisms of resistance, negative impacts of human modification of land cover and use on animal movement are shown to include decreased habitat quality, increased mortality or predation risk, and associated behavioral responses (avoidance)\u003csup\u003e\u003cspan additionalcitationids=\"CR52\" citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e\u003c/sup\u003e. We recognize that for some species, especially generalists and those commonly found in urban environments, this assumption may not be valid. Rather, this modeling effort is intentionally designed to apply to species which are negatively impacted by industrial human pressures on a landscape that can cause destruction, degradation, and impairment of ecological processes (as outlined by the IUCN biodiversity threats classification\u003csup\u003e\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e\u003c/sup\u003e and adopted by global human modification mapping\u003csup\u003e\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u003c/sup\u003e).\u003c/p\u003e \u003cp\u003eTo identify more small-scale connectivity patterns, relevant for short distance dispersers such as herpetofauna and most plants\u003csup\u003e\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e,\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e\u003c/sup\u003e, our methods and framework can be downscaled using local data since the global results were sensitive to some parameterization decisions. Our global team (see \u003cem\u003eMethods)\u003c/em\u003e specifically identified grassland areas with low human modification as sensitive to slight increases in human modification due to pasture/grazing. We subsequently removed the pasture/grazing layer from our analysis, which assumes limited impacts of grazing lands on movement potential. While there are caveats to our model, global products such as ours have broadscale utility (e.g., \u003csup\u003e\u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e57\u003c/span\u003e\u003c/sup\u003e), and this work will provide critical context that helps to identify important patterns of variation in connectivity constraints and where to guide more in-depth assessments. By no means is this work meant to replace local or landscape-tailored modeling and planning where fine-scale data are available to answer more targeted questions.\u003c/p\u003e \u003cp\u003eMeasuring and monitoring connectivity is not only theoretically challenging, but also technologically challenging. Fortunately, improvements in software and computing resources are rapidly enabling major advances to allow this modeling at high resolution. There are several modeling options for continuous circuit-theory connectivity modeling\u003csup\u003e\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e, but the implementation of Omniscape in the Julia language in particular allows relatively efficient global modeling that facilitates iterative parameter testing\u003csup\u003e\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e,\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e,\u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e58\u003c/span\u003e,\u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e59\u003c/span\u003e\u003c/sup\u003e. The power to consider connectivity across a continuum of landscape conditions without set start and end points (such as PAs) provides a powerful, flexible way to deliver decision-relevant products. We expect that faster runs, strengthened by deeper understanding of model behavior, will accelerate model co-creation with decision-makers and regional experts. Increased processing speed allowed us to iteratively test model parameters and work closely with collaborators to refine the connectivity modeling inputs (e.g., human modification resistance surface), provide guidance on modeling parameterizations (i.e., how water was treated as movement barrier), and iteratively consult with experts on draft model outputs. Combined, these collaborative efforts and advancements refined our approach, improved the reliability of our outputs, thereby promoting higher relevance for the policy applications and goal setting processes.\u003c/p\u003e \u003cp\u003eBeyond the GBF, there is global interest and growing momentum for restoration and protection of natural systems at scale for climate benefits, including natural climate solutions (NCS)\u003csup\u003e\u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e60\u003c/span\u003e,\u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e61\u003c/span\u003e\u003c/sup\u003e. Our map allows identification of sites for restoration and protection that maximize benefits to connectivity for biodiversity conservation in addition to carbon sequestration\u003csup\u003e\u003cspan additionalcitationids=\"CR63 CR64\" citationid=\"CR62\" class=\"CitationRef\"\u003e62\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e65\u003c/span\u003e\u003c/sup\u003e. When further paired with land tenure information and with community and local engagement, protection and restoration of natural systems can be powerful tools for durable conservation outcomes\u003csup\u003e\u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e66\u003c/span\u003e\u003c/sup\u003e. Users of our framework and map should apply best practices in engaging local communities and other critical local stakeholders for advancing GBF goals. Importantly, large-scale projects protecting connectivity on unprotected territories must meet the equity and justice concerns of local communities\u003csup\u003e\u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e67\u003c/span\u003e\u003c/sup\u003e. Global environmental agendas often lead to ambitious goals that can galvanize environmental advocates, but the impacts on local populations is not uniform. A next step may be to overlay human populations (e.g., \u003csup\u003e\u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e68\u003c/span\u003e\u003c/sup\u003e) and land tenure (e.g., \u003csup\u003e\u003cspan citationid=\"CR69\" class=\"CitationRef\"\u003e69\u003c/span\u003e\u003c/sup\u003e) with our global map to assess potential areas where careful planning and implementation will be critically important. Government commitments for land-based carbon removal projects alone now exceed 1\u0026nbsp;billion hectares\u003csup\u003e\u003cspan citationid=\"CR70\" class=\"CitationRef\"\u003e70\u003c/span\u003e\u003c/sup\u003e and further demands will likely lead to competing and greater demands on landowners and residents.\u003c/p\u003e"},{"header":"CONCLUSION","content":"\u003cp\u003eEcosystems, species ranges, and human land uses are shifting, often in unpredictable ways, necessitating more holistic approaches to conservation that extend beyond PA boundaries\u003csup\u003e71\u0026ndash;73\u003c/sup\u003e. Protecting and maintaining ecological connectivity is critically important to stem the biodiversity crisis. Indicators that can evaluate connectivity are needed to set targets and measure success towards global conservation goals such as those in the GBF\u003csup\u003e8\u003c/sup\u003e. There has been a lack of sufficient methods for whole landscape connectivity modeling, especially for implementation at the large scales required to meet national and global goals. Global connectivity indicators must consider unprotected areas as well as protected areas and efforts to stem biodiversity loss need to include the restoration and protection of natural lands outside of PAs\u003csup\u003e46\u003c/sup\u003e. By presenting a continuous global connectivity map with a policy-relevant framework, applicable across different landscapes and contexts, we can better inform policy and land-use planning. Such an approach can avoid the pitfalls of relying exclusively on PA-centric connectivity indicators, which can result in biased information and direct conservation focus away from unprotected areas critical for whole landscape connectivity. As landscapes, ecosystems, and climate change, moving beyond the boundary to identify and preserve whole system connectivity is critical for conserving biodiversity in its current and future habitat.\u003c/p\u003e"},{"header":"METHODS","content":"\u003cp\u003eThere are many ways to model connectivity. We chose a circuit theory method that allowed us to emphasize how landscape patterns influence broad movement potential at multiple scales, from landscapes to continents. To estimate connectivity at the full global extent at 1 km resolution, we used Omniscape.jl, a moving window-based adaptation of the Julia language implementation of the circuit-theoretic modeling software Circuitscape\u003csup\u003e\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e,\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e,\u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e58\u003c/span\u003e\u003c/sup\u003e. Circuit-theory models allow considering how movement may spread or be constrained in different parts of a landscape by using the mathematics of electrical circuit theory to describe cumulative patterns that represent traces of ‘random walkers’ as they encounter variable resistance using the analogy of electrical current flow\u003csup\u003e\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u003c/sup\u003e. Resistance weights are assigned to pixels based on hypotheses of the degree to which the conditions in that pixel would support or inhibit movement. The assumption of random movement suggests that, as one example, directional routes reflecting knowledge of the landscape (like annual migrations) are likely not well modeled with this approach\u003csup\u003e\u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e74\u003c/span\u003e\u003c/sup\u003e. Rather, a key strength of treating movement as a random process is that it allows seeing where many movement routes are possible, or where options are highly constrained or absent. Omniscape is an expansion of Circuitscape, which considers connectivity between focal nodes or patches, by allowing modelers to evaluate connectivity in all directions (“omnidirectional” Circuitscape) to and from all pixels (for full details on the Omniscape approach, see \u003csup\u003e\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e,\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u003c/sup\u003e). Here, we deployed Omniscape with separate resistance and source layers using a moving window radius of 125 km, globally at 1 km resolution.\u003c/p\u003e\u003ch3\u003eResistance data\u003c/h3\u003e\u003cp\u003eWe used an updated version of the global 300 m Human Modification (HM) dataset for 2022, refined specifically for modeling connectivity\u003csup\u003e\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u003c/sup\u003e, and coarsened it to 1 km using average-based resampling for the resistance layer. The HM includes the spatial extent and intensity of cumulative human pressures from settlement, agriculture, forestry, transportation, mining, energy production, electrical infrastructure, dams, pollution and human accessibility. For the connectivity resistance surface, we removed pasture/grazing and air pollution data, because we assumed they have low influence on species movement. For pixels intersecting major roads and highways, we assigned a value of 0.8 (for 300 m pixels) to decrease the muting of linear features when decreasing resolution (to 1 km).\u003c/p\u003e\u003cp\u003eIn Omniscape, small differences in resistance at the low end of the range, especially in factors that occur in large swaths of a landscape (such as large blocks of agriculture) typically have a large influence on the results. Although HM has been used as resistance surfaces in connectivity modeling\u003csup\u003e\u003cspan citationid=\"CR75\" class=\"CitationRef\"\u003e75\u003c/span\u003e\u003c/sup\u003e, few studies have explored using transformations on the resistance data to understand the model’s sensitivity to the distribution of continuous values\u003csup\u003e\u003cspan citationid=\"CR76\" class=\"CitationRef\"\u003e76\u003c/span\u003e\u003c/sup\u003e. Exponential functions may better describe the relationship of wildlife to landscapes during movement life history periods than non-exponential linear functions\u003csup\u003e\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e,\u003cspan citationid=\"CR77\" class=\"CitationRef\"\u003e77\u003c/span\u003e,\u003cspan citationid=\"CR78\" class=\"CitationRef\"\u003e78\u003c/span\u003e\u003c/sup\u003e. Specifically, exponential transformations may better describe the relationship of raw HM to resistance with respect to broad-scale animal movement than the raw HM data itself. This improvement reflects that negligibly to moderately modified areas of the landscape may be readily used for movements, and only the most modified areas are unusable. These transformations are more ecologically sound when considering long-distance movements and allow for pattern exploration in the middle of the HM distribution\u003csup\u003e\u003cspan citationid=\"CR77\" class=\"CitationRef\"\u003e77\u003c/span\u003e,\u003cspan citationid=\"CR78\" class=\"CitationRef\"\u003e78\u003c/span\u003e\u003c/sup\u003e. We used the function:\u003c/p\u003e\u003cdiv id=\"Equa\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equa\" name=\"EquationSource\"\u003e\n$$\\:f=100-99\\frac{1-exp(-c(1-HM)}{1-exp(-c)}$$\u003c/div\u003e\u003c/div\u003e\u003cp\u003eWhere \u003cem\u003ef i\u003c/em\u003es the transformed value of the HM and \u003cem\u003ec\u003c/em\u003e is a rescaling parameter\u003csup\u003e\u003cspan citationid=\"CR77\" class=\"CitationRef\"\u003e77\u003c/span\u003e\u003c/sup\u003e. With larger \u003cem\u003ec\u003c/em\u003e values, the transformation approaches an exponential function of the HM and smaller positive values approach a linear transformation. We tested \u003cem\u003ec\u003c/em\u003e = 0.25, 2, 4, and 8 (Table S5) scaling parameters on the HM data during our sensitivity testing.\u003c/p\u003e\u003ch3\u003eSource data\u003c/h3\u003e\u003cp\u003eA source layer in Omniscape is used to represent locations from which movement originates, such as a parcel of natural land or a protected area\u003csup\u003e\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e. A common source layer in connectivity modeling is the inverse of the resistance layer, especially in models where human modification is used as the resistance layer\u003csup\u003e\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e,\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e. Instead, we used natural land cover as our source to model a global ecological network. Natural land cover pixels better represent areas that will likely be moved from, and although correlated with HM, provides a clear way to represent ecological networks that connect major ecosystem types.\u003c/p\u003e\u003cp\u003eWe used the ESA World Cover 10 m v 200\u003csup\u003e36\u003c/sup\u003e product as our source layer because we were interested in creating a species-agnostic connectivity model, with connectivity identified between areas of natural lands and it includes accuracy assessment and an uncertainty layer with an estimated 76% global accuracy. We first labeled each class in the ESA World Cover data as representing natural vegetation (e.g., tree cover, shrubland, grassland, bare/sparse, wetland, mangroves, moss/lichen) or non-natural (agriculture, urban) lands, then rescaled the data to 1 km\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e pixels. We used the proportion of natural lands within 1 km pixels as our source weight data by calculating the average value (0, non-natural; 1, natural) across all 10 m pixels within each rescaled 1 km cell, excluding water pixels.\u003c/p\u003e\u003ch2\u003eWater\u003c/h2\u003e\u003cp\u003eThe treatment of water in resistance data can have a significant impact on connectivity outputs\u003csup\u003e\u003cspan citationid=\"CR79\" class=\"CitationRef\"\u003e79\u003c/span\u003e\u003c/sup\u003e. We developed a method for addressing water that aligned with the purpose of mapping terrestrial connectivity and our intent of informing conservation strategies that address the influence of human modifications on movement potential. demonstrates Water has been treated in ways that range from “no data” (effectively complete barriers to movement) to low or moderate resistances, to approaches that scale the resistance of water to size of water body or flow rate, to maximum resistance (e.g., \u003csup\u003e\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e,\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e,\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e,\u003cspan additionalcitationids=\"CR81\" citationid=\"CR80\" class=\"CitationRef\"\u003e80\u003c/span\u003e–\u003cspan citationid=\"CR82\" class=\"CitationRef\"\u003e82\u003c/span\u003e\u003c/sup\u003e). Through iterative testing, we developed an approach intended to limit the effect of water on the model outputs\u003csup\u003e\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e\u003c/sup\u003e. To include water in a “neutral” way that does not prohibit or facilitate resulting flow (\u003cem\u003eF\u003c/em\u003e), we assigned ocean and inland water pixels within 10 km of land random resistance values, within 2% of the average resistance value of a 40 km surrounding radius. This approach ensured that the resistance values of water would not have higher or lower resistance than neighboring land. We assigned the maximum resistance value to water more than 10 km from land, assuming that few terrestrial species would be able to move in water \u0026gt; 10 km from land. Finally, we masked water from the final output maps. Our map of water was based on the ESA CCI Global Water Bodies dataset (v4.0; \u003csup\u003e\u003cspan citationid=\"CR83\" class=\"CitationRef\"\u003e83\u003c/span\u003e\u003c/sup\u003e).\u003c/p\u003e\u003ch2\u003eModel parameters\u003c/h2\u003e\u003cp\u003eOmniscape allows for ‘blocking’ or grouping squares of pixels and using the center pixel as the source\u003csup\u003e\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u003c/sup\u003e. This method decreases processing time and does not impact model results\u003csup\u003e\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e. In all cases, we used a block size 10% of the moving window radius to decrease model run time while maintaining model integrity\u003csup\u003e\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e. We tested radius sizes of 50 km, 125 km, and 425 km and we chose the radius size of 125 km to reflect potential dispersal distances of most birds and mammals, ranging from \u0026lt; 1 km – 467 km, with medians ranging from \u0026lt; 0.1 km to 87 km\u003csup\u003e\u003cspan citationid=\"CR84\" class=\"CitationRef\"\u003e84\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003ch2\u003eModel estimation\u003c/h2\u003e\u003cp\u003eWe computed all outputs on r6i.4xlarge EC2 virtual machine instances on Amazon Web Services (AWS), each containing 16 vCPUs and 128 GiB of RAM. Models were estimated using Omniscape v0.5.8 running on Julia v1.9.2 with 6 threads. Model run time varied depending on the window radius and ranged from several hours (for a 50 km radius) to approximately one week (for a 425 km radius).\u003c/p\u003e\u003ch2\u003eOutput categories\u003c/h2\u003e\u003cp\u003eThe Omniscape main outputs, \u003cem\u003eF\u003c/em\u003e and the \u003cem\u003eF\u003c/em\u003e\u003csub\u003e\u003cem\u003en\u003c/em\u003e\u003c/sub\u003e provide complementary information to convey how the resistance and source data interact to influence the magnitude of flow across landscapes, and the degree to which movement is constrained by barriers\u003csup\u003e\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e. The \u003cem\u003eF\u003c/em\u003e emphasizes the most likely places for movement to occur, highlighting key areas of concentration and broad movement zones between the most intact areas. The \u003cem\u003eF\u003c/em\u003e\u003csub\u003e\u003cem\u003en\u003c/em\u003e\u003c/sub\u003e, where the magnitude of flow is scaled to the regional magnitude of sources, identifies places where \u003cem\u003eF\u003c/em\u003e has become concentrated relative to flow in the absence of resistance (\u003cem\u003eF\u003c/em\u003e\u003csub\u003e\u003cem\u003ep\u003c/em\u003e\u003c/sub\u003e), even if the overall flow amount is moderate or low relative to other locations. This allows the model to highlight key movement zones and pinch points (areas where flow is restricted and narrowed) across the full range of resistance levels and geographic distributions of source areas.\u003c/p\u003e\u003cp\u003eThe categorization was informed by four continuous components which were each categorized: the continuous flow pattern (impeded, diffuse, channelized) from the \u003cem\u003eF\u003c/em\u003e\u003csub\u003e\u003cem\u003en\u003c/em\u003e\u003c/sub\u003e, the \u003cem\u003eF\u003c/em\u003e\u003csub\u003e\u003cem\u003ep\u003c/em\u003e\u003c/sub\u003e, \u003cem\u003eF\u003c/em\u003e, and the HM resistance data. We categorized each of these four layers based on thresholds we identified which aligned with changes in the distribution of each data layer:\u003c/p\u003e\u003cp\u003e \u003cem\u003eNormalized flow (F\u003c/em\u003e \u003csub\u003e \u003cem\u003en\u003c/em\u003e \u003c/sub\u003e,\u003cem\u003e)\u003c/em\u003e: We adopted the thresholds identified by Cameron et al. (\u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e), where \u003cem\u003eF\u003c/em\u003e\u003csub\u003e\u003cem\u003en\u003c/em\u003e\u003c/sub\u003e \u0026gt; 1.7 is channelized flow, 1.3 ≥ \u003cem\u003eF\u003c/em\u003e\u003csub\u003e\u003cem\u003en\u003c/em\u003e\u003c/sub\u003e ≤ 1.7 is intensified flow, 0.7 ≥ \u003cem\u003eF\u003c/em\u003e\u003csub\u003e\u003cem\u003en\u003c/em\u003e\u003c/sub\u003e \u0026lt;1.3 is diffuse flow, and \u003cem\u003eF\u003c/em\u003e\u003csub\u003e\u003cem\u003en\u003c/em\u003e\u003c/sub\u003e \u0026lt; 0.7 is impeded flow.\u003c/p\u003e\u003cp\u003e \u003cstrong\u003eFlow potential (F\u003csub\u003ep\u003c/sub\u003e)\u003c/strong\u003e \u003c/p\u003e\u003cp\u003eTo determine the thresholds for the flow potential layer, we calculated the global distribution of flow potential and cumulative current and identified one threshold break for the flow potential data, opting to include the majority of data into one lower category (µ + 0.5 SD) and defining values \u0026gt; µ + 0.5 SD as ‘high’.\u003c/p\u003e\u003cp\u003e \u003cstrong\u003eFlow (F)\u003c/strong\u003e \u003c/p\u003e\u003cp\u003eBecause the flow layer is calculated and defined relative to the potential flow layer, we used the same values as the flow potential layer with the addition of a ‘very high’ category. We defined ‘very high’ cumulative flow as \u0026gt; µ\u003csub\u003e\u003cem\u003eFp\u003c/em\u003e\u003c/sub\u003e + 1.5SD\u003csub\u003e\u003cem\u003eFp\u003c/em\u003e\u003c/sub\u003e.\u003c/p\u003e\u003cp\u003e \u003cem\u003eHuman modification layer\u003c/em\u003e: We based the thresholds for the HM on those identified in Kennedy et al. (\u003csup\u003e\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e\u003c/sup\u003e) and Riggio et al. (\u003csup\u003e\u003cspan citationid=\"CR85\" class=\"CitationRef\"\u003e85\u003c/span\u003e\u003c/sup\u003e). However, because we used the HM as it relates to how species may move across landscapes, we adjusted the thresholds to align with HM values as found within ESA World Cover land cover classes. Because tree cover and other natural classes were most abundant \u0026lt; 0.1 HM and because species can move across sub-optimal habitat, we combined the “very low” and “low” HM classes previously identified\u003csup\u003e\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e,\u003cspan citationid=\"CR85\" class=\"CitationRef\"\u003e85\u003c/span\u003e\u003c/sup\u003e. Our HM classes were thus: \u0026lt;0.1 “low”, 0.1–0.3 “low-medium”, 0.3–0.7 “medium-high”, and \u0026gt; 0.7 “high” modification.\u003c/p\u003e\u003cp\u003eWe reclassified each of the above continuous layers according to the respective identified thresholds and combined all four layers to create one global categorized map with 96 potential unique categories. Only 63 categories occurred globally, and we iteratively identified commonalities in the pattern and occurrence of between the categories to further group them. In doing so, we prioritized normalized flow categorization then grouped these by human modification level and used the difference in flow potential and cumulative flow to further inform groupings.\u003c/p\u003e\u003ch2\u003eSensitivity testing\u003c/h2\u003e\u003cp\u003eWe tested multiple model inputs and parameters to determine their effects on model outputs (Tables S5, S6, Fig S2). We ran models with different \u003cem\u003ec\u003c/em\u003e values, model search radii, source layers, water treatment methods, slope, and treatment of grazing lands. We compared cumulative flow outputs using a Pearson correlation coefficient, calculated in R package terra v 1.7.71\u003csup\u003e86\u003c/sup\u003e (Table S6).\u003c/p\u003e\u003ch2\u003eReview process\u003c/h2\u003e\u003cp\u003eBecause we conducted a global analysis, we formed a review team of 14 conservation practitioners across six continents to fine-tune model inputs and ensure model validity. The team (including coauthors J.B.-G., C.R.D., J.F., M.N.-R., and M.F.R.) included conservation scientists with geospatial and/or connectivity modeling backgrounds with knowledge of African, Australian, Asian, North American, and South American landscapes. They were provided a data viewer with the input and output datasets, written materials about our methods, a video of our intent and methods, and a questionnaire guiding their review. After these review sessions, the team reviewed the output datasets with respect to their knowledge of landscapes and other existing connectivity analyses. We then reviewed feedback received and made changes to the input data to address concerns raised by reviewers. Overall, across geographies, reviewers expressed concerns around the inclusion of pasture/grazing lands as a component of the resistance layer, which led us to use a modified version of the HM data\u003csup\u003e\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e,\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003ch2\u003eData summaries\u003c/h2\u003e\u003cp\u003eTo support country and regional decision making, we summarized movement potential by country, continent, biome, and protected areas boundaries. For summaries by biome, we used the biomes identified by Dinerstein et al. (\u003csup\u003e\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e\u003c/sup\u003e). For protected areas, we used data from the World Database of Protected Areas\u003csup\u003e\u003cspan citationid=\"CR87\" class=\"CitationRef\"\u003e87\u003c/span\u003e\u003c/sup\u003e. We followed Allan et al. (\u003csup\u003e\u003cspan citationid=\"CR88\" class=\"CitationRef\"\u003e88\u003c/span\u003e\u003c/sup\u003e) by including China and India PAs using the 2017 WDPA data. We excluded marine protected areas, PAs that were “proposed” or in the “not applicable” protection categories. We created separate layers for each IUCN protection category: I-VI, “not reported”, or “not assigned” and converted these to 1 km\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e cell raster data, where any cell with ≥ 50% overlap was included as protected. We combined these raster layers and removed any overlap, retaining the highest level of protection, where applicable.\u003c/p\u003e\u003cp\u003eTo compare our data with other connectivity measures, we calculated Pearson correlation coefficients for countries \u0026gt; 10,000 km\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e for the percentage diffuse within a country, protected area index\u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e (PAI), Protected Area Representativeness and Connectedness\u003csup\u003e\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u003c/sup\u003e (PARC), Protected Areas Network\u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e (ProNet), and Protected Connected\u003csup\u003e\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e (ProtConn) connectivity indicators as updated in UNEP-WCMC and IUCN (\u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e) using R v 4.4.3\u003csup\u003e89\u003c/sup\u003e.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eACKNOWLEDGEMENTS\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe are grateful to Brad McRae for the theoretical development of Omniscape and his pioneering work in the application of circuit theory to ecology. Viral Shah, Ranjan Anantharanam, and Vincent Landau continue to maintain Circuitscape and Omniscape and have provided valuable insights during the development of this work. We are grateful to Manuel Agra, Mario Barroso, Hugo Cardenas Rodriguez, Mercedes Ibanez, Wang Longzhu, Nathaniel Robinson, Anne M. Trainor for reviewing preliminary data and improving this work and to Aimee Biegel, Peter W. Ellis, and Preston Welker for supporting this work. We thank the Bezos Earth Fund for making this work possible.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAUTHOR CONTRIBUTIONS\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eErin E. Poor– Conceptualization, methodology, project administration, writing, supervision, funding acquisition; Kim Hall– conceptualization, methodology, writing; Jesse Anderson– data curation, formal analysis, software, visualization; Melissa Clark– methodology, writing; Aaron Jones– conceptualization, visualization, writing, data curation, methodology; Christina M. Kennedy– conceptualization, methodology, writing; \u0026nbsp; Carrie A. Schloss – methodology, writing; David M. Theobald– data curation, methodology, writing; Jaime Burbano-Girón– writing; Susan C. Cook-Patton– writing; C. Ronnie Drever– writing; Joseph Fargione– conceptualization, writing; James Fitzsimmons– writing; Mauricio Nunez-Regueiro– writing; Milena Fermina Rosenfield– writing; Yuta J. Masuda– funding acquisition, writing\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n \u003cli\u003eCeballos, G. \u003cem\u003eet al.\u003c/em\u003e Accelerated modern human \u0026ndash; induced species losses : Entering the sixth mass extinction. 9\u0026ndash;13 (2015).\u003c/li\u003e\n \u003cli\u003eD\u0026iacute;az, S. \u003cem\u003eet al.\u003c/em\u003e Pervasive human-driven decline of life on Earth points to the need for transformative change. \u003cem\u003eScience (80-. ).\u003c/em\u003e \u003cstrong\u003e366\u003c/strong\u003e, (2019).\u003c/li\u003e\n \u003cli\u003eJaureguiberry, P. \u003cem\u003eet al.\u003c/em\u003e The direct drivers of recent global anthropogenic biodiversity loss. \u003cem\u003eSci. Adv.\u003c/em\u003e \u003cstrong\u003e8\u003c/strong\u003e, 1\u0026ndash;11 (2022).\u003c/li\u003e\n \u003cli\u003eMaxwell, S. L., Fuller, R. A., Brooks, T. M. \u0026amp; Watson, J. E. M. The ravages of guns, nets and bulldozers. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e536\u003c/strong\u003e, 143\u0026ndash;145 (2016).\u003c/li\u003e\n \u003cli\u003ePecl, G. T. \u003cem\u003eet al.\u003c/em\u003e Biodiversity redistribution under climate change: Impacts on ecosystems and human well-being. \u003cem\u003eScience (80-. ).\u003c/em\u003e \u003cstrong\u003e355\u003c/strong\u003e, (2017).\u003c/li\u003e\n \u003cli\u003eIPBES. \u003cem\u003eGlobal assessment report of the Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services\u003c/em\u003e. \u003cem\u003eDebating Nature\u0026rsquo;s Value\u003c/em\u003e (2019).\u003c/li\u003e\n \u003cli\u003eUNEP-CMS. \u003cem\u003eIMPROVING WAYS OF ADDRESSING CONNECTIVITY IN THE CONSERVATION OF MIGRATORY SPECIES\u003c/em\u003e. (2020).\u003c/li\u003e\n \u003cli\u003eCBD. Kunming-Montreal Global Biodiversity Framework. 1\u0026ndash;15 (2022).\u003c/li\u003e\n \u003cli\u003eHanski, I. Metapolulation dynamics. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e396\u003c/strong\u003e, 41\u0026ndash;49 (1998).\u003c/li\u003e\n \u003cli\u003eFahrig, L. Effects of Habitat Fragmentation on Biodiversity. \u003cem\u003eAnnu. Rev. Ecol. Evol. Syst.\u003c/em\u003e \u003cstrong\u003e34\u003c/strong\u003e, 487\u0026ndash;515 (2003).\u003c/li\u003e\n \u003cli\u003eSaura, S. \u003cem\u003eet al.\u003c/em\u003e Global trends in protected area connectivity from 2010 to 2018. \u003cem\u003eBiol. Conserv.\u003c/em\u003e \u003cstrong\u003e238\u003c/strong\u003e, 108183 (2019).\u003c/li\u003e\n \u003cli\u003eLocke, H. \u003cem\u003eet al.\u003c/em\u003e Three global conditions for biodiversity conservation and sustainable use: An implementation framework. \u003cem\u003eNatl. Sci. Rev.\u003c/em\u003e \u003cstrong\u003e6\u003c/strong\u003e, 1080\u0026ndash;1082 (2019).\u003c/li\u003e\n \u003cli\u003eJoppa, L. N. \u0026amp; Pfaff, A. High and far: Biases in the location of protected areas. \u003cem\u003ePLoS One\u003c/em\u003e \u003cstrong\u003e4\u003c/strong\u003e, 1\u0026ndash;6 (2009).\u003c/li\u003e\n \u003cli\u003eDinerstein, E. \u003cem\u003eet al.\u003c/em\u003e A \u0026lsquo;global safety net\u0026rsquo; to reverse biodiversity loss and stabilize earth\u0026rsquo;s climate. \u003cem\u003eSci. Adv.\u003c/em\u003e \u003cstrong\u003e6\u003c/strong\u003e, 1\u0026ndash;14 (2020).\u003c/li\u003e\n \u003cli\u003eUNEP-WCMC \u0026amp; IUCN. \u003cem\u003eProtected Planet Report 2024\u003c/em\u003e. (2024).\u003c/li\u003e\n \u003cli\u003eLaurance, W. F. \u003cem\u003eet al.\u003c/em\u003e Averting biodiversity collapse in tropical forest protected areas. \u003cem\u003eNature\u003c/em\u003e (2012) doi:10.1038/nature11318.\u003c/li\u003e\n \u003cli\u003eMaxwell, S. L. \u003cem\u003eet al.\u003c/em\u003e Area-based conservation in the twenty-first century. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e586\u003c/strong\u003e, 217\u0026ndash;227 (2020).\u003c/li\u003e\n \u003cli\u003eWolff, N. H. \u003cem\u003eet al.\u003c/em\u003e Prioritizing global land protection for population persistence can double the efficiency of habitat protection for reducing mammal extinction risk. \u003cem\u003eOne Earth\u003c/em\u003e \u003cstrong\u003e6\u003c/strong\u003e, 1564\u0026ndash;1575 (2023).\u003c/li\u003e\n \u003cli\u003eMendenhall, C. D., Karp, D. S., Meyer, C. F. J., Hadly, E. A. \u0026amp; Daily, G. C. in agricultural landscapes. (2014) doi:10.1038/nature13139.\u003c/li\u003e\n \u003cli\u003eBrennan, A. \u003cem\u003eet al.\u003c/em\u003e Functional connectivity of the world\u0026rsquo;s protected areas. \u003cem\u003eScience (80-. ).\u003c/em\u003e \u003cstrong\u003e376\u003c/strong\u003e, 1101\u0026ndash;1104 (2022).\u003c/li\u003e\n \u003cli\u003eMyers, N., Mittermeier, R. A., Mittermeier, C. G., Fonseca, G. A. B. \u0026amp; Kent, J. Biodiversity hotspots for conservation priorities. \u003cstrong\u003e403\u003c/strong\u003e, 853\u0026ndash;858 (2000).\u003c/li\u003e\n \u003cli\u003eHilty, J. \u003cem\u003eet al.\u003c/em\u003e \u003cem\u003eGuidelines for conserving connectivity through ecological networks and corridors. Best Practice Protected Area Guidelines Series No. 30\u003c/em\u003e. (2020).\u003c/li\u003e\n \u003cli\u003eDutta, T., Sharma, S., Meyer, N. F. V., Larroque, J. \u0026amp; Balkenhol, N. An overview of computational tools for preparing, constructing and using resistance surfaces in connectivity research. \u003cem\u003eLandsc. Ecol.\u003c/em\u003e \u003cstrong\u003e37\u003c/strong\u003e, 2195\u0026ndash;2224 (2022).\u003c/li\u003e\n \u003cli\u003eMcRae, B. H. \u003cem\u003eet al.\u003c/em\u003e \u003cem\u003eConserving Nature\u0026rsquo;s Stage: Mapping Omnidirectional Connectivity for Resilient Terrestrial Landscapes in the Pacific Northwest.\u003c/em\u003e http://nature.org/resilienceNW (2016) doi:10.13140/RG.2.1.4158.6166.\u003c/li\u003e\n \u003cli\u003eBelote, R. T., Barnett, K., Zeller, K., Brennan, A. \u0026amp; Gage, J. Examining local and regional ecological connectivity throughout North America. \u003cem\u003eLandsc. Ecol.\u003c/em\u003e (2022) doi:10.1007/s10980-022-01530-9.\u003c/li\u003e\n \u003cli\u003ePhillips, P., Clark, M. M. \u0026amp; Koen, E. L. Comparison of methods for estimating omnidirectional landscape connectivity. \u003cem\u003eLandsc. Ecol.\u003c/em\u003e \u003cstrong\u003e36\u003c/strong\u003e, 1647\u0026ndash;1661 (2021).\u003c/li\u003e\n \u003cli\u003eMcRae, B. H., Dickson, B. G., Keitt, T. H. \u0026amp; Viral B. Shah. Using circuit theory to model connectivity in ecology, evolution, and conservation. \u003cem\u003eEcology\u003c/em\u003e \u003cstrong\u003e89\u003c/strong\u003e, 2712\u0026ndash;2724 (2008).\u003c/li\u003e\n \u003cli\u003eCook, C. N., Mascia, M. B., Schwartz, M. W., Possingham, H. P. \u0026amp; Fuller, R. A. Achieving conservation science that bridges the knowledge-action boundary. \u003cem\u003eConserv. Biol.\u003c/em\u003e \u003cstrong\u003e27\u003c/strong\u003e, 669\u0026ndash;678 (2013).\u003c/li\u003e\n \u003cli\u003eBeger, M. \u003cem\u003eet al.\u003c/em\u003e Demystifying ecological connectivity for actionable spatial conservation planning. \u003cem\u003eTrends Ecol. Evol.\u003c/em\u003e \u003cstrong\u003e37\u003c/strong\u003e, (2022).\u003c/li\u003e\n \u003cli\u003eCameron, D. R., Morrison, S. A., Schloss, C. A. \u0026amp; Theobald, D. M. A framework to select strategies for conserving and restoring habitat connectivity in complex landscapes. 1\u0026ndash;16 (2022) doi:10.1111/csp2.12698.\u003c/li\u003e\n \u003cli\u003eAnderson, M. G. \u003cem\u003eet al.\u003c/em\u003e A resilient and connected network of sites to sustain biodiversity under a changing climate. \u003cem\u003eProc. Natl. Acad. Sci. U. S. A.\u003c/em\u003e \u003cstrong\u003e120\u003c/strong\u003e, 1\u0026ndash;9 (2023).\u003c/li\u003e\n \u003cli\u003eBezanson, J., Edelman, A., Karpinski, S. \u0026amp; Shah, V. B. Julia: A fresh approach to numerical computing. \u003cem\u003eSIAM Rev.\u003c/em\u003e \u003cstrong\u003e59\u003c/strong\u003e, 65\u0026ndash;98 (2017).\u003c/li\u003e\n \u003cli\u003eLandau, V., Shah, V., Anantharaman, R. \u0026amp; Hall, K. Omniscape.jl: Software to compute omnidirectional landscape connectivity. \u003cem\u003eJ. Open Source Softw.\u003c/em\u003e \u003cstrong\u003e6\u003c/strong\u003e, 2829 (2021).\u003c/li\u003e\n \u003cli\u003eClark, M., Anderson, J., Anderson, M. G., Hall, K. R. \u0026amp; E., P. E. Wade carefully: Water parameterizaion in structural landscape connectivity modeling. \u003cem\u003eLandsc. Ecol.\u003c/em\u003e\u003c/li\u003e\n \u003cli\u003eTheobald, D. M., Oakleaf, J. R., Moncrieff, G., Voigt, M. \u0026amp; Kiesecker, J. Global extent and change in human modification of terrestrial ecosystems from 1990 to 2022. 1\u0026ndash;66 (2025).\u003c/li\u003e\n \u003cli\u003eZanaga, D. \u003cem\u003eet al.\u003c/em\u003e ESA WorldCover 10m 2020v 100. (2021).\u003c/li\u003e\n \u003cli\u003eHarwood, T., Ware, C., Hoskins, A. \u0026amp; Ferrier, S. PARC: Protected Area Connectedness Index v2: 30s global layer 2020. (2022) doi:https://doi.org/10.25919/kt3f-2z04.\u003c/li\u003e\n \u003cli\u003eTheobald, D. M., Keeley, A. T. H., Laur, A. \u0026amp; Tabor, G. A simple and practical measure of the connectivity of protected area networks: The ProNet metric. \u003cem\u003eConserv. Sci. Pract.\u003c/em\u003e \u003cstrong\u003e4\u003c/strong\u003e, 1\u0026ndash;13 (2022).\u003c/li\u003e\n \u003cli\u003eSaura, S. \u003cem\u003eet al.\u003c/em\u003e Protected area connectivity: Shortfalls in global targets and country-level priorities. \u003cem\u003eBiol. Conserv.\u003c/em\u003e \u003cstrong\u003e219\u003c/strong\u003e, 53\u0026ndash;67 (2018).\u003c/li\u003e\n \u003cli\u003eDinerstein, E. \u003cem\u003eet al.\u003c/em\u003e An Ecoregion-Based Approach to Protecting Half the Terrestrial Realm. \u003cem\u003eBioscience\u003c/em\u003e \u003cstrong\u003e67\u003c/strong\u003e, 534\u0026ndash;545 (2017).\u003c/li\u003e\n \u003cli\u003eGrantham, H., Adams, V. \u0026amp; Planning, T. I. W. T. on S. \u003cem\u003eAn overview of Participatory, Integrated, and Biodiversity-Inclusive Spatial Planning and Target 1 under the Convention on Biological Diversity Kunming-Montreal Global Biodiversity Framework, IUCN WCPA Issues Paper Series No. 4\u003c/em\u003e.\u003c/li\u003e\n \u003cli\u003eOakleaf, J. \u003cem\u003eet al.\u003c/em\u003e Mapping global land conversion pressure to support conservation planning. \u003cem\u003eSci. Data\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, 1\u0026ndash;10 (2024).\u003c/li\u003e\n \u003cli\u003eSaura, S. \u003cem\u003eet al.\u003c/em\u003e Protected area connectivity: Shortfalls in global targets and country-level priorities. \u003cem\u003eBiol. Conserv.\u003c/em\u003e \u003cstrong\u003e219\u003c/strong\u003e, 53\u0026ndash;67 (2018).\u003c/li\u003e\n \u003cli\u003eGolden Kroner, R. E. \u003cem\u003eet al.\u003c/em\u003e The uncertain future of protected lands and waters. \u003cem\u003eScience (80-. ).\u003c/em\u003e \u003cstrong\u003e364\u003c/strong\u003e, 881\u0026ndash;886 (2019).\u003c/li\u003e\n \u003cli\u003eSaura, S. \u003cem\u003eet al.\u003c/em\u003e Protected area connectivity : Shortfalls in global targets and country-level priorities. \u003cem\u003eBiol. Conserv.\u003c/em\u003e \u003cstrong\u003e219\u003c/strong\u003e, 53\u0026ndash;67 (2020).\u003c/li\u003e\n \u003cli\u003eWard, M. \u003cem\u003eet al.\u003c/em\u003e Just ten percent of the global terrestrial protected area network is structurally connected via intact land. \u003cem\u003eNat. Commun.\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, 1\u0026ndash;10 (2020).\u003c/li\u003e\n \u003cli\u003eEdwards, D. P., Tobias, J. A., Sheil, D., Meijaard, E. \u0026amp; Laurance, W. F. Maintaining ecosystem function and services in logged tropical forests. \u003cem\u003eTrends Ecol. Evol.\u003c/em\u003e \u003cstrong\u003e29\u003c/strong\u003e, 511\u0026ndash;520 (2014).\u003c/li\u003e\n \u003cli\u003eWilcove, D. S., Giam, X., Edwards, D. P., Fisher, B. \u0026amp; Koh, L. P. Navjot\u0026rsquo;s nightmare revisted: logging, agriculture, and biodiersity in Southeast Asia. \u003cem\u003eTrends Ecol. Evol.\u003c/em\u003e \u003cstrong\u003e28\u003c/strong\u003e, 531\u0026ndash;540 (2013).\u003c/li\u003e\n \u003cli\u003eKennedy, C. M., Oakleaf, J. R., Theobald, D. M., Baruch-Mordo, S. \u0026amp; Kiesecker, J. Managing the middle: A shift in conservation priorities based on the global human modification gradient. \u003cem\u003eGlob. Chang. Biol.\u003c/em\u003e \u003cstrong\u003e25\u003c/strong\u003e, 811\u0026ndash;826 (2019).\u003c/li\u003e\n \u003cli\u003eTheobald, D. M. \u003cem\u003eet al.\u003c/em\u003e Earth transformed: Detailed mapping of global human modification from 1990 to 2017. \u003cem\u003eEarth Syst. Sci. Data\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, 1953\u0026ndash;1972 (2020).\u003c/li\u003e\n \u003cli\u003eDyer, S. J. ., Neill, J. P. . O. \u0026rsquo;, Wasel, S. M. . \u0026amp; Boutin, S. Avoidance of Industrial Development by Woodland Caribou. \u003cem\u003eJ. Wildl. Manage.\u003c/em\u003e \u003cstrong\u003e65\u003c/strong\u003e, 531\u0026ndash;542 (2001).\u003c/li\u003e\n \u003cli\u003eGantchoff, M. G. \u0026amp; Belant, J. L. Regional connectivity for recolonizing American black bears (Ursus americanus) in southcentral USA. \u003cem\u003eBiol. Conserv.\u003c/em\u003e \u003cstrong\u003e214\u003c/strong\u003e, 66\u0026ndash;75 (2017).\u003c/li\u003e\n \u003cli\u003eScoyoc, A. Van, Smith, J. A., Gaynor, K. M., Barker, K. \u0026amp; Brashares, J. S. The influence of human activity on predator \u0026ndash; \u0026shy; prey spatiotemporal overlap. 1124\u0026ndash;1134 (2023) doi:10.1111/1365-2656.13892.\u003c/li\u003e\n \u003cli\u003eIUCN. \u003cem\u003eIUCN CMP Unified Classification of Direct Threats v.3.3\u003c/em\u003e. (2022).\u003c/li\u003e\n \u003cli\u003eCain, M. L., Milligan, B. G. \u0026amp; Strand, A. E. Long-distance seed dispersal in plant populations. \u003cem\u003eAm. J. Bot.\u003c/em\u003e \u003cstrong\u003e87\u003c/strong\u003e, 1217\u0026ndash;1227 (2000).\u003c/li\u003e\n \u003cli\u003eSemlitsch, R. D. Differentiating Migration and Dispersal Processes for Pond-Breeding Amphibians. \u003cem\u003eJ. Wildl. Manage.\u003c/em\u003e \u003cstrong\u003e72\u003c/strong\u003e, 260\u0026ndash;267 (2008).\u003c/li\u003e\n \u003cli\u003eChaplin-Kramer, R. \u003cem\u003eet al.\u003c/em\u003e Conservation needs to integrate knowledge across scales. \u003cem\u003eNat. Ecol. Evol.\u003c/em\u003e \u003cstrong\u003e6\u003c/strong\u003e, 118\u0026ndash;119 (2022).\u003c/li\u003e\n \u003cli\u003eAnantharaman, R., Hall, K., Shah, V. B. \u0026amp; Edelman, A. Circuitscape in Julia: High Performance Connectivity Modelling to Support Conservation Decisions. \u003cem\u003eJuliaCon Proc.\u003c/em\u003e \u003cstrong\u003e1\u003c/strong\u003e, 58 (2020).\u003c/li\u003e\n \u003cli\u003eHall, K. R. \u003cem\u003eet al.\u003c/em\u003e Circuitscape in julia: Empowering dynamic approaches to connectivity assessment. \u003cem\u003eLand\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, 1\u0026ndash;24 (2021).\u003c/li\u003e\n \u003cli\u003eGriscom, B. W. \u003cem\u003eet al.\u003c/em\u003e Natural climate solutions. \u003cem\u003eProc. Natl. Acad. Sci. U. S. A.\u003c/em\u003e \u003cstrong\u003e114\u003c/strong\u003e, 11645\u0026ndash;11650 (2017).\u003c/li\u003e\n \u003cli\u003eIPCC. Section 4: Near-Term Responses in a Changing Climate. \u003cem\u003eClim. Chang. 2023 Synth. Rep.\u003c/em\u003e 42\u0026ndash;66 (2023) doi:10.59327/IPCC/AR6-9789291691647.\u003c/li\u003e\n \u003cli\u003eYang, X. \u003cem\u003eet al.\u003c/em\u003e Establishment success in a forest biodiversity and ecosystem functioning experiment in subtropical China (BEF-China). \u003cem\u003eEur. J. For. Res.\u003c/em\u003e \u003cstrong\u003e132\u003c/strong\u003e, 593\u0026ndash;606 (2013).\u003c/li\u003e\n \u003cli\u003ePettorelli, N. \u003cem\u003eet al.\u003c/em\u003e Time to integrate global climate change and biodiversity science-policy agendas. \u003cem\u003eJ. Appl. Ecol.\u003c/em\u003e \u003cstrong\u003e58\u003c/strong\u003e, 2384\u0026ndash;2393 (2021).\u003c/li\u003e\n \u003cli\u003eEllis, P. W. \u003cem\u003eet al.\u003c/em\u003e The principles of natural climate solutions. \u003cem\u003eNat. Commun.\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, (2024).\u003c/li\u003e\n \u003cli\u003eChang, C. H. \u003cem\u003eet al.\u003c/em\u003e Global evidence of human well-being and biodiversity impacts of natural climate solutions. \u003cem\u003eNat. Sustain.\u003c/em\u003e \u003cstrong\u003e8\u003c/strong\u003e, (2025).\u003c/li\u003e\n \u003cli\u003eFILAC., F. and. \u003cem\u003eForest governance by indigenous and tribal peoples An opportunity for climate action in Latin America and the Caribbean Forest governance by indigenous and tribal peoples An opportunity for climate action in Latin America and\u003c/em\u003e. https://doi.org/10.4060/cb2953en (2021).\u003c/li\u003e\n \u003cli\u003eRaymond, C. M. \u003cem\u003eet al.\u003c/em\u003e Inclusive conservation and the Post-2020 Global Biodiversity Framework: Tensions and prospects. \u003cem\u003eOne Earth\u003c/em\u003e \u003cstrong\u003e5\u003c/strong\u003e, 252\u0026ndash;264 (2022).\u003c/li\u003e\n \u003cli\u003eLebakula, V., Epting, J., Moehl, J., Stipek, C., Adams, D., Reith, A., Kaufman, J., Gonzales, J., Reynolds, B., Basford, S., Martin, A., Buck, W., Faxon, A., Cunningham, A., Roy, A., Barbose, Z., Massaro, JLebakula, V., Epting, J., Moehl, J., Stipek, C., M. LandScan Silver Edition [data set]. (2024) doi:https://doi.org/10.48690/1531770.\u003c/li\u003e\n \u003cli\u003eGarnett, S. T. \u003cem\u003eet al.\u003c/em\u003e A spatial overview of the global importance of Indigenous lands for conservation. \u003cem\u003eNat. Sustain.\u003c/em\u003e \u003cstrong\u003e1\u003c/strong\u003e, 369\u0026ndash;374 (2018).\u003c/li\u003e\n \u003cli\u003eSelf, A., Burdon, R., Lewis, J., Riggs, P. \u0026amp; Dooley, K. \u003cem\u003eLand Gap Report Briefing Note 2023 Update\u003c/em\u003e. (2023).\u003c/li\u003e\n \u003cli\u003eLenoir, J. \u003cem\u003eet al.\u003c/em\u003e Going against the flow: Potential mechanisms for unexpected downslope range shifts in a warming climate. \u003cem\u003eEcography (Cop.).\u003c/em\u003e \u003cstrong\u003e33\u003c/strong\u003e, 295\u0026ndash;303 (2010).\u003c/li\u003e\n \u003cli\u003eLenoir, J. \u0026amp; Svenning, J. C. Climate-related range shifts - a global multidimensional synthesis and new research directions. \u003cem\u003eEcography (Cop.).\u003c/em\u003e \u003cstrong\u003e38\u003c/strong\u003e, 15\u0026ndash;28 (2015).\u003c/li\u003e\n \u003cli\u003eTomiolo, S. \u0026amp; Ward, D. Perspectives in Plant Ecology , Evolution and Systematics Species migrations and range shifts : A synthesis of causes and consequences. \u003cem\u003ePerspect. Plant Ecol. Evol. Syst.\u003c/em\u003e \u003cstrong\u003e33\u003c/strong\u003e, 62\u0026ndash;77 (2018).\u003c/li\u003e\n \u003cli\u003ePoor, E. E., Loucks, C., Jakes, A. \u0026amp; Urban, D. L. Comparing Habitat Suitability and Connectivity Modeling Methods for Conserving Pronghorn Migrations. \u003cem\u003ePLoS One\u003c/em\u003e \u003cstrong\u003e7\u003c/strong\u003e, (2012).\u003c/li\u003e\n \u003cli\u003eTheobald, D. M., Reed, S. E., Fields, K. \u0026amp; Soul, M. Connecting natural landscapes using a landscape permeability model to prioritize conservation activities in the United States. \u003cstrong\u003e5\u003c/strong\u003e, 123\u0026ndash;133 (2012).\u003c/li\u003e\n \u003cli\u003eZeller, K. A., Mcgarigal, K. \u0026amp; Whiteley, A. R. Estimating landscape resistance to movement : a review. 777\u0026ndash;797 (2012) doi:10.1007/s10980-012-9737-0.\u003c/li\u003e\n \u003cli\u003eTrainor, A. M., Walters, J. R., Morris, W. F., Sexton, J. \u0026amp; Moody, A. Empirical estimation of dispersal resistance surfaces: A case study with red-cockaded woodpeckers. \u003cem\u003eLandsc. Ecol.\u003c/em\u003e \u003cstrong\u003e28\u003c/strong\u003e, 755\u0026ndash;767 (2013).\u003c/li\u003e\n \u003cli\u003eKeeley, A. T. H., Beier, P. \u0026amp; Gagnon, J. W. Estimating landscape resistance from habitat suitability: effects of data source and nonlinearities. \u003cem\u003eLandsc. Ecol.\u003c/em\u003e \u003cstrong\u003e31\u003c/strong\u003e, 2151\u0026ndash;2162 (2016).\u003c/li\u003e\n \u003cli\u003eMarrec, R. \u003cem\u003eet al.\u003c/em\u003e Conceptual framework and uncertainty analysis for large- scale , species-agnostic modelling of landscape connectivity across. \u003cem\u003eSci. Rep.\u003c/em\u003e 1\u0026ndash;14 (2020) doi:10.1038/s41598-020-63545-z.\u003c/li\u003e\n \u003cli\u003eDickson, B. G. \u003cem\u003eet al.\u003c/em\u003e Informing Strategic Efforts to Expand and Connect Protected Areas Using a Model of Ecological Flow, with Application to the Western United States. \u003cem\u003eConserv. Lett.\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, 564\u0026ndash;571 (2017).\u003c/li\u003e\n \u003cli\u003eSchloss, C., Cameron, Ri. C., Ae, B. R. A. D. H. M. C. R., Heobald, D. A. M. T. \u0026amp; Ones, A. A. J. \u0026ldquo; No-regrets \u0026rdquo; pathways for navigating climate change : planning for connectivity with land use , topography , and climate. \u003cstrong\u003e32\u003c/strong\u003e, 1\u0026ndash;11 (2022).\u003c/li\u003e\n \u003cli\u003ePither, R., O\u0026rsquo;Brien, P., Brennan, A., Hirsh-Pearson, K. \u0026amp; Bowman, J. Predicting areas important for ecological connectivity throughout Canada. \u003cem\u003ePLoS One\u003c/em\u003e \u003cstrong\u003e18\u003c/strong\u003e, 1\u0026ndash;21 (2023).\u003c/li\u003e\n \u003cli\u003eLamarche, C. \u003cem\u003eet al.\u003c/em\u003e Compilation and Validation of SAR and Optical Data Products for a Complete and Global Map of Inland/Ocean Water Tailored to the Climate Modeling Community. \u003cem\u003eRemote Sens.\u003c/em\u003e \u003cstrong\u003e9\u003c/strong\u003e, (2017).\u003c/li\u003e\n \u003cli\u003eWhitmee, S. \u0026amp; Orme, C. D. L. Predicting dispersal distance in mammals: A trait-based approach. \u003cem\u003eJ. Anim. Ecol.\u003c/em\u003e \u003cstrong\u003e82\u003c/strong\u003e, 211\u0026ndash;221 (2013).\u003c/li\u003e\n \u003cli\u003eRiggio, J. \u003cem\u003eet al.\u003c/em\u003e Global human influence maps reveal clear opportunities in conserving Earth\u0026rsquo;s remaining intact terrestrial ecosystems. \u003cem\u003eGlob. Chang. Biol.\u003c/em\u003e \u003cstrong\u003e26\u003c/strong\u003e, 4344\u0026ndash;4356 (2020).\u003c/li\u003e\n \u003cli\u003eR, H. terra: Spatial Data Analysis. (2025).\u003c/li\u003e\n \u003cli\u003eUNEP-WCMC \u0026amp; IUCN. \u003cem\u003eUNEP-WCMC and IUCN. August 2019. Protected Planet: The World Database on Protected Areas (WDPA), August 2019 version, Cambridge, UK: UNEP-WCMC and IUCN.\u003c/em\u003e \u003cem\u003eProtected Planet: The World Database on Protected Areas (WDPA)\u003c/em\u003e (2023).\u003c/li\u003e\n \u003cli\u003eAllan, J. R. \u003cem\u003eet al.\u003c/em\u003e attention to safeguard biodiversity. \u003cstrong\u003e1101\u003c/strong\u003e, 1\u0026ndash;7 (2022).\u003c/li\u003e\n \u003cli\u003eR Core Team. R: A Language and Environment for Statistical Computing. 2025 (2025).\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Connectivity, circuit theory, Omniscape, human modification, conservation planning, biodiversity targets","lastPublishedDoi":"10.21203/rs.3.rs-6457802/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6457802/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eProtecting and maintaining ecosystem connectivity is crucial for stemming the biodiversity crisis, but current tools and indicators assessing connectivity primarily focus on connectivity among protected areas, not across ecosystems. Here, we develop a map of global connectivity across all lands that accounts for the cumulative impacts of human modification. We then parse this continuous connectivity into categories directly relevant for conservation actions (i.e., spatial planning, restoration, and protection). We find that most global lands (66%) retain high levels of connectivity, particularly in tundra, boreal, and conifer forest biomes, which often fall outside of protected areas. Conversely, 29% of global terrestrial areas have low levels of connectivity due to high human modification, which is prevalent in Asia, Europe, and North America, and in temperate and tropical forested biomes. Our results underscore that focusing only on connectivity of protected areas misses ecologically important areas that currently lack protection and may fail to identify important conservation and restoration areas for connectivity that could benefit ecosystems. This work is broadly applicable to inform global conservation goals, such as those enumerated in the Kunming-Montreal Global Biodiversity Framework (GBF). We provide summaries to inform global, national, and sub-national decision making and demonstrate how this whole-landscape connectivity data can be used to support GBF targets.\u003c/p\u003e","manuscriptTitle":"Global whole landscape connectivity to complement protected area connectivity","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-05-14 05:38:34","doi":"10.21203/rs.3.rs-6457802/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"5e95da4c-c722-4f6e-9692-6f134cf93dba","owner":[],"postedDate":"May 14th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":48401488,"name":"Earth and environmental sciences/Ecology/Conservation biology"},{"id":48401489,"name":"Earth and environmental sciences/Ecology/Ecological networks"},{"id":48401490,"name":"Earth and environmental sciences/Ecology/Ecological modelling"}],"tags":[],"updatedAt":"2026-04-17T08:35:17+00:00","versionOfRecord":[],"versionCreatedAt":"2025-05-14 05:38:34","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6457802","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6457802","identity":"rs-6457802","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}