Spare half, share the rest: A revised planetary boundary for biodiversity intactness and integrity


 Setting clear biodiversity targets is a pervasive challenge1 due to the context-dependent nature of biodiversity that has evaded concise science-based objectives such as the 1.5°C for climate2. Considering the major risk of continued inaction, and further biodiversity loss, it is imperative that the biodiversity community identify similarly operational science-based boundaries supporting the development of targets to safeguard biodiversity and its contributions to human well-being. Confounding goal setting are the two distinct but not mutually exclusive conservation objectives: (1) halting the rampant loss of intact ecosystems, species extinction and population declines, and (2) maintaining biosphere integrity and ecosystem function. We propose a revised planetary boundary for biodiversity: retaining at least half of the area of each terrestrial ecoregion biologically intact to halt the extinction crisis, and maintaining ecosystem integrity across all lands to preserve and regenerate biosphere, ecosystem functions and their contributions to human well-being. We combine four intactness datasets to provide a global assessment and find that 49.6% of the Earth’s land ice-free surface remains intact. While this is promising globally, 69% of 798 unique ecoregions are less than half intact. For ecosystem integrity, we find 18.1% of working lands have ecosystem integrity deficits precluding the provisioning of biosphere and ecosystem functions. Ninety percent of ecoregions have an ecosystem integrity deficit. Globally, intactness and integrity are at boundary limits with degradation of nature critically jeopardizing biosphere capacity to support a safe and just space for humanity. Combined efforts are needed to halt loss of and restore intactness, while regenerating integrity in working lands.

for humanity. Combined efforts are needed to avoid further loss of and restore intactness, 23 while regenerating integrity in working lands. 24

Main text 25
Introduction 26 Biodiversity is in crisis facing catastrophic conversion and losses of ecosystems and the 27 unique species and genetic diversity they harbour 3-6 . This loss increasingly has regional and 28 global scale impacts on earth systems processes spanning from carbon and water cycles, to 29 pollination and pest and disease regulation 7 . The absence of meaningful system-wide indicators 30 hinders global policy by failing to provide politically relevant status measures and targets to 31 align and galvanize conservation action 1 . Without clear, science-based but politically actionable 32 guardrails, we risk losing a pivotal decade of progress in bending the curve on biodiversity loss 8 . 33 The biodiversity research community has thus far largely resisted setting concise 34 scientific targets due to well-founded concerns that species, and associated ecosystem services, 35 are inherently local and that ecology is more complex and more localized than climate, and thus 36 not amenable to system-wide targets 9 . However, considering the acute nature of the on-going 37 biodiversity crisis 6 , setting evidence-based science targets can be an effective means to foster 38 alignment, identify effective actions, and set up processes to track progress across stakeholder 39 groups 10 . As demonstrated by the Paris Climate Convention, and by the EAT-Lancet 40 Commission 11 , global targets can be set that are compatible with local conservation priority 41 setting and that recognize the diversity of context specific conservation actions. More critically, 42 agreed upon targets and system-wide boundary metrics can provide standardized scientific 43 assessments of biodiversity and ecosystem function status and trends that align local action 44 against global goals. 45 We propose biodiversity intactness and ecosystem integrity as two complementary 46 planetary boundary measures of biodiversity (Box 1). Biodiversity intactness and ecosystem 47 integrity are analogous to the first two goals (Goal A and Goal B) proposed by the Convention 48 on Biological Diversity (CBD) 12 . Both biodiversity intactness and ecosystem integrity are 49 evaluated here from spatially gridded datasets. These data can be aggregated over any larger 50 region, such as regions defined by administrative (e.g., country, continent) or ecological (e.g., 51 ecosystem, ecoregion, biome, global) boundaries, or a combination of these (e.g., ecoregions 52 within countries) to facilitate setting ecologically meaningful and politically actionable targets. 53 We provide a global assessment of intactness and integrity to facilitate countries setting 54 ambitious goals for protection, restoration, and regeneration. 55 56 Biodiversity Intactness 57 The conceptual definition of biodiversity intactness describes the state of an ecosystem 58 being unimpaired from post-industrial human alteration (Box 1) 13 . Biodiversity intactness is a 59 measure of biodiversity status measured as the relative abundance of originally present species 60 or level of human pressures 13,14 . Intact areas, by definition have integrity and maintain 61 ecological functions, for example climate mitigation and regulation 14 . Retaining intact nature 62 globally is required to halt the loss of wild biodiversity, and both the earth system, and 63 ecosystem services it provides. While nuances of conservation efforts are important, including 64 fragmentation, habitat size and extent, maintaining at least half of the global land mass as 65 intact has been proposed as required to avert massive species loss 15-18 . Half intact improves on 66 previous articulations of the Biosphere Integrity planetary boundary by adding an area-based 67 complement to intactness, making it easily operational. Half intact is synonymous with the "no 68 net loss or net gain" targets proposed by the CBD under Goal A. We apply the concept of half 69 intact at the ecoregion level to assess the conservation of the unique local biological 70 composition across the globe that require protection. Intactness levels below this boundary 71 increase irreversible extinction risks, a threshold which has been passed in 69% of 72 ecoregions 1,15,18 . Remaining above the intactness boundary requires important sparing of 73 terrestrial and aquatic ecosystems from conversion 19,20 . 74

Ecosystem Integrity 75
Integrity is conceptually defined as an ecosystem's functional capacity to contribute to 76 biosphere processes and to produce ecosystem services (Box 1) 21 . These include both Earth 77 system-scale processes regulated by the biosphere as well as finer scale ecosystem services 78 provided at the basin, landscape, farm, field, or even neighbourhood scale in urban settings 7,22-79 25 . Non-intact area can have integrity if they retain sufficient functional biodiversity to support 80 ecosystem service provision 26-29 irrespective of whether the species or communities they 81 contain are native or not 30 . 82 The spatial extent of the services provided by species often operates at the sub-83 kilometer scale 25,31-40 . While there is significant context specificity in how biodiversity provides 84 ecosystem services 41,42 , the decay of ecosystem services with increasing distance from 85 provisioning habitat is consistently demonstrated in agroecological and ecosystem service 86 studies. For example, the number and abundance of species able to provide pollination or pest 87 control services, rapidly decreases with the decrease in the area of habitat available 33,43-49 and 88 with increasing distance from source habitat 32-34,37-39,47-57 . Particularly in agricultural landscapes, 89 ecosystem service provisioning is related to the farm or field scale such as nutrient delivery (e.g. 90 nitrogen fixation (0.1-1 m), reduction of soil and sediment loss (1-10 m), pollination (10-1000 91 m), pest control (10-1000m)) 25 . Thus, while specific services provided by distinct ecological 92 communities will remain deeply contextual 41,42 , and the subject of both interesting and 93 important research, evidence suggests that natural habitat extent and proximity is a coarse 94 indicator of potential ecosystem service provisioning. More specifically, the absence of semi-95 natural or natural habitat within a few hundred meters of any given area indicates insufficient 96 associated biodiversity to provide ecosystem services to that area, particularly services that 97 support food production 58 . 98 We use the proportion of natural or semi-natural habitat within a landscape as our 99 operational definition of ecosystem integrity (Box 1; see methods for definitions of natural and 100 semi-natural) and propose an integrity boundary of at least 10% natural or semi-natural habitat 101 per square kilometre. This integrity boundary represents the minimum surrounding natural or 102 semi-natural habitat needed to maintain the functional biodiversity providing ecosystem 103 services (see methods). Others have proposed a more conservative 20% boundary for 104 maintaining integrity 44 . We compute integrity levels using 10%, 20% and 30% to estimate 105 sensitivity of threshold selection. 106 Intactness and integrity capture two distinct but complementary conservation priorities: 107 halting extinction loss and maintaining ecosystem and biosphere function. Achieving intactness 108 targets and halting the loss of biodiversity requires reducing and eliminating pressures such as 109 habitat loss and conversion. Restoring intactness is difficult and in the case of species 110 extinction, nearly impossible to fully recover 59 . Integrity, in contrast, is more readily repairable 111 and can be provided by biodiverse ecosystems with low intactness scores but functional 112 composition approaching that of intact ecosystems (e.g., non-native species or novel 113 ecosystems) 26-28 . This is the premise of nature-based solutions offered by agroecology, 114 regenerative agriculture, or integrated landscape management initiatives 45,46,60-62 . Integrity 115 targets are achieved by combined actions sparing currently intact ecosystems from conversion, 116 but also requires sharing space for biodiversity in working lands 20,45,46 . 117 118

Global Status of Biodiversity Intactness 119
We applied an ensemble approach to map intactness by selecting four available 120 datasets of human influence on the global land surface. We applied a threshold to each of the 121 datasets to identify areas with the lowest human impact and highest portion of biologically 122 intact land remaining (Supplemental Fig. 1; Supplemental data table 1). We combined the four 123 datasets 124  Table 3). While 139 restoration can increase intactness in these locations, recovery to preindustrial levels is unlikely 140 (data at https://rpubs.com/afremier/SuppTable_ER_data_2021). to assess ecosystem integrity. We define lands as having integrity if they have at least 10% 148 natural or semi-natural habitat within 1 square km. We propose that the boundary condition 149 needs to be met in all lands because retaining ecosystem integrity underpins the provision and 150 resilience of ecosystem services required for human well-being including for example, food, 151 nutition 66 , climate 62 , and water security 1,67 (see methods: Establishing Ecosystem Integrity). 152 Globally, 18.2% of all lands, and 23% of the human-dominated lands are below the 153 integrity boundary, an equivalent of 120.0 M km 2 (Supplemental Table 5). In these areas 154 insufficient ecosystem integrity exists to maintain ecosystem service provision such as 155 pollination of crops, or regulation of agricultural pests and diseases 43,68 . Increasing the integrity 156 boundary to 20-30% per km 2 in human dominated lands has been proposed to the CBD 10 . We 157 estimate that 32.8% and 40.3% of human dominated lands fail to meet the 20% and 30% 158 integrity boundary respectively (Supplemental Table 5). We find that 741 of 821 ecoregions 159 Our analysis identifies three generalized categories of country-ecoregion status of 186 combined intactness and integrity measures: (1) high integrity, high intactness (Fig. 3 dark  187 green regions (29% surface land)), (2) high integrity, low intactness (Fig. 2 & Fig. 3 blue regions  188 (54%)) and (3) low integrity and intactness (Fig. 2 & Fig. 3 pink and red regions (17%)). Regions  Table  215 4). Intact areas in ecoregions require continued conservation commitments to achieve no net 216 loss 72 . Returning to intactness levels above boundary conditions (i.e. >50%) for all ecoregions 217 would require restoring 24 M km 2 of land and, with no additional loss of existing intact land, 218 result in intact land covering 69% of the Earth's terrestrial land. We find that 314 of 1745 (18%) 219 country ecoregions have <10% intact lands which we consider post or near-extinction and 220 possibly non-restorable (Supplemental Table 4, https://rpubs.com/afremier/CER). Retaining 221 and restoring the intact areas in these endangered eco-regions needs urgent and critical 222 attention (Figs 2-3). 223 In light of the critical contributions of biodiversity to human well-being though their role 224 in providing ecosystem services, or nature's contributions to people, the non-intact half of our 225 planet cannot become an ecological sacrifice zone. This half must retain sufficient ecological 226 integrity, through retention of sufficient natural and semi-natural habitat at small and large 227 scales, to produce food, regulate pests and diseases, provide safe passage to biodiversity, and 228 contribute to gene flow, regulating water cycles, offer spiritual and recreational spaces and 229 mitigating climate change amongst others 73 . We find that 18.      (b) Percent intact land by ecoregion; (c) Percent land above the 10% integrity target by ecoregion. Note: The designations employed and the presentation of the material on this map do not imply the expression of any opinion whatsoever on the part of Research Square concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. This map has been provided by the authors.

Figure 3
Mapped Intactness and Integrity values for each ecoregion; Colored bins are taken from gure 2. Note: The designations employed and the presentation of the material on this map do not imply the expression of any opinion whatsoever on the part of Research Square concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. This map has been provided by the authors.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download. ShareSpareSupplementalTablesXFiguresSubmitted.pdf