Systematic Nature Positive Markets

Despite decades of global commitments, and increasingly urgent warning of environmental instability, the demand for land to support economic production is still increasing. Isolated and disorganized actions will not be enough to avert ecosystem failures. As many developers are already required to compensate for their ecological impacts through restoration, many see markets trading biodiversity credits as a financial mechanism to counteract degradation and drive investment in conservation. The challenge stems from a desire to recognize the multidimensional nature of biodiversity that contributes to ecosystem integrity without making suitable offsets intractable to supply. Instead, most regulators have opted to streamline ecological assessment, and undermine ecological rigour, in favour of promoting offset supply and economic efficiency. As a result, all evidence suggests offset trading programs have so far failed to mitigate losses, let alone support “nature positive” outcomes. To overcome this disconnect, and support more effective and equitable biodiversity markets, we propose credits be defined by the irreplaceability of a site, a metric long-established in the domain of systematic conservation planning. Irreplaceability avoids the limitations of like-for-like trading, reduces costs of offsetting to developers and society, ensures farmers willing to sell are fairly rewarded for loss of earnings, and that sites critical to achieving conservation goals are safeguarded. We developed an ecological-economic model of a biodiversity offset market to demonstrate irreplaceability guarantees no net loss of biodiversity and is the most efficient metric for guiding investment toward the recovery of Nature.


INTRODUCTION
More than 75% of Earth's land is degraded, and this has led to widespread biodiversity loss, undermining the well-being of billions of people and efforts to combat climate change (IPBES, 2019).Current evidence suggests multiple planetary thresholds have been exceeded (Steffen et al., 2015) and business as usual is highly likely to result in catastrophic collapse across many ecosystems (Armstrong McKay et al., 2022).Yet, numerous global commitments to reduce, stop, or even reverse current rates of biodiversity loss have not been met (Díaz et al., 2019;Tittensor et al., 2014).Instead, reversing global terrestrial biodiversity trends will only be achievable if strategic, coordinated, and above all ambitious actions are adopted (Leclère et al., 2020).To achieve a nature-positive future, private-sector funding of biodiversity conservation needs to be increased to complement established publicly funded programs.
The ENACT initiative (Enhancing Nature-based Solutions for an Accelerated Climate Transformation) launched at COP27 calls for the mobilization of private finance to support action on nature and climate-related targets the world over, accompanied by robust environmental and social safeguards (IUCN, 2022).
Environmental markets can be used to mobilize private financing, which can incentivize landholders to undertake more sustainable land management actions (Schmalensee & Stavins, 2017).Such markets create income streams in the form of tradeable credits for landholders in return for undertaking actions to protect or enhance specified environmental goods and services, for example, biodiversity, carbon sequestration, and water quality.Demand (and thus buyer willingness to pay [WTP]) for these credits can be voluntary, arising from a demand from individuals or companies who wish to offset their negative environmental impacts, or they are created through government regulation, for example, through requiring developers to purchase credits to offset new buildings (Needham et al., 2019;Radu et al., 2020).
We considered regulated markets for biodiversity offsets where developers must purchase credits to mitigate impacts on biodiversity as a result of development activities, such as mine construction, housing, road building, and hydroelectric dams.Credits are supplied by landowners who switch their current land management (e.g., arable farming) to a more conservationorientated alternative (e.g., wetland creation).In regulated offset markets, state-sanctioned intermediary bodies, such as offset banks, validate credits and enforce offset requirements placed on developers.By establishing an appropriate rate of exchange between sellers (landowners) and buyers (developers), biodiversity offset markets can, in principle, achieve no net loss of biodiversity or a net gain in some defined area at the lowest overall economic cost to society and are thus potentially economically efficient (Needham et al., 2019).
In regulated and voluntary nature markets, the choice of biodiversity metric plays a pivotal role in determining the market's ecological and economic performance (Simpson et al., 2021).This metric establishes the units in which biodiversity is traded, determining how a regulator or offset bank measures the gains in biodiversity resulting from restoration actions undertaken by landowners, and balances those gains against the expected biodiversity loss due to development.Simple metrics based on a combination of the area and condition of habitat are often preferred by regulators (Bull et al., 2014;zu Ermgassen et al., 2019) because it is easy to identify matching biodiversity units and because they are based on the assumption that habitat classes indirectly capture benefits on other aspects of the ecosystem (Marshall et al., 2020).However, these approaches rarely benefit biodiversity in the manner intended, or they fail to deliver gains in biodiversity in an economically efficient manner (Bull et al., 2014;Maron et al., 2012;zu Ermgassen et al., 2021).
We developed and then applied a new metric for application in biodiversity offset markets, and in environmental markets more broadly, that derives from the systematic conservation planning (SCP) literature (Margules & Pressey, 2000;McIntosh et al., 2017).Tools of SCP are designed to minimize the cost of achieving conservation targets.The importance of any specific site to achieving conservation targets is measured by its irreplaceability.A site that is essential to achieving targets is completely irreplaceable (i.e., its loss cannot be offset), whereas irreplaceability is low for sites that can be easily substituted for many others to contribute to conservation targets.Crucially, irreplaceability combines the importance of a specific site over multiple biodiversity features and thus captures tradeoffs between features and ensures that overarching targets for each feature are achieved at the landscape scale.Irreplaceability represents a step change away from like-for-like compensation regimes in existing biodiversity offset markets (e.g., BBOP, 2009;Natural England, 2022;NSW DPE, 2022).Furthermore, if conservation targets are chosen to exceed their existing availability in a landscape, this embeds net gain as an implicit outcome where it is needed to meet specific targets.
We sought to demonstrate that an offset market steered by a metric derived from irreplaceability ensures the opportunity to achieve conservation targets is always protected and results in a network of conserved sites selected for being more economically efficient than sites obtained using simpler offset metrics.Irreplaceability as a metric thus offers a step change in the design of biodiversity offsets and environmental markets, which is important given the current fast rate of expansion in such nature markets globally.

Irreplaceability recast for biodiversity offset markets
SCP is a rigorous, repeatable, and structured approach to designing protected areas that efficiently meet conservation objectives (Margules & Pressey, 2000).At an analytical level, the task is a classic resource allocation problem that either maximizes conservation outcomes within a given resource budget or minimizes the cost of achieving specified conservation targets (Moilanen et al., 2009).This structure has led to the use of SCP in supporting conservation decisions across the globe (McIntosh et al., 2017).A key strength of SCP is that it can be used to incorporate a wide variety of data types, including attributes of ecosystems at all levels of structural, taxonomic, and functional organization, as well as accounting for social, financial, and political constraints and opportunities (Ban et al., 2013;Knight et al., 2011).Suitable targets are often based on the principle of adequacy, which aims to maintain the viability and persistence of those features (Kukkala & Moilanen, 2013).Species-level targets may be informed by population viability analyses or habitatlevel targets by species-area relationships, and functional targets may be informed by the need for particular services across landscapes (Bryan et al., 2010).
The value of any specific site is based on its marginal contribution to achieving the conservation targets by complementing what features are already secured.A key feature therefore of SCP is that, unlike ranking procedures, properties of reserve systems emerge from the combination of areas either through the complementarity of their composition or their connectivity in space.This suggests a strong potential advantage for using a metric derived from SCP in biodiversity offset markets, where a need exists to be able to compare ecological gains and losses across space between development sites (where biodiversity declines) and offset supply sites (where biodiversity is increased due to the action of the landowner).Moreover, a biodiversity offset metric needs to make sense in the context of an overall policy target of no net loss or net gain in a specific aggregate indicator of biodiversity.This combination of an aggregate target with the need to compare gains and losses across space suggests that a metric derived from SCP could have important advantages over the kinds of metrics investigated so far in the offset markets literature (Simpson et al., 2022).
Provided with data on feature values for all planning units, planning unit costs, and the desired targets for protection, SCP tools can be used to identify which sets of sites deliver conservation targets most efficiently (Moilanen et al., 2009).For convenience, we refer to features and planning units as species and sites, respectively, hereafter.Often targets can be achieved by many different combinations of sites because alternatives exist with similar, or at least complementary, values.The importance of any specific site to achieving conservation targets is measured by its irreplaceability.A site that is essential to achieving targets is irreplaceable (and its loss could not be offset), whereas irreplaceability is low for sites that can be substituted by many others.An exact calculation of irreplaceability rapidly becomes intractable as the number of combinations to test grows exponentially with the number of planning units (Pressey et al., 1993), and alternatives to estimate irreplaceability have been proposed (Ferrier et al., 2000).Most recently, Baisero et al. (2021) proposed a new metric for describing irreplaceability (α) that defines the extent to which a site k is essential for achieving the conservation of species s as where the difference between the total availability of a species in the landscape R ′ s and its target t s indicates how much of that availability a site can contain (R k,s ) before it becomes irreplaceable.Baisero et al. (2021) defined β as the combined irreplaceability of a site by taking the complement of the product of replacement probabilities β = 1 − Π(1 − α k,s ).However, this constrains site irreplaceability to between 0 and 1; consequently, it no longer indicates whether a site was irreplaceable for 1 or many species.To retain this distinction and make comparisons among sites in an offset market equivalent, we use summed α irreplaceability.Ferrier et al. (2000) also summed irreplaceability for a similar reason (albeit with a different formulation for each species); therefore, hereafter we refer to the sum of α irreplaceability ( ∑  k,s ), which we abbreviate it as ∑α.

The biodiversity offset market
The structure of the biodiversity offset market was based on the model developed by Simpson et al. (2021).A single agent controls each land parcel or site in a landscape.Each agent decides to develop the land for housing, generate biodiversity offset credits by adopting a conservation land management practice, or keep the current land use of agriculture.For an agent to develop the land, each hectare acquired for, for example, new housing requires a number of offset credits to be purchased from an offset provider equal to the measured biodiversity value of the site.The developer's maximum WTP for an offset credit is determined by the expected value of land for housing and the need to purchase offset credits.Ranking this WTP from highest to lowest yields a downward-sloping demand curve for offset credits.This WTP varies over space due to variations in house prices and the value of each site for biodiversity.We assumed offset credits are supplied by agents on agricultural land (i.e., farmers).Farmers change their current agricultural land management practices in a way that increases biodiversity by a measured amount at the site.Every hectare given up to benefit biodiversity means 1 less hectare for agricultural production.Furthermore, the farmer may incur restoration costs in creating an offset credit.Therefore, the conversion cost to the farmer consists of the opportunity costs of the foregone agricultural output plus any associated restoration costs.This sum is the minimum price a farmer would be willing to sell an offset credit for, known as the minimum willingness to accept (WTA).Because agricultural productivity and profits vary across space (due, e.g., to variations in soil quality or site elevation), the minimum WTA of farmers to create biodiversity credits will also vary over space.Ranking farmers from lowest WTA to highest WTA generates a supply curve for offsets.Farmers and developers interact in this market to generate an equilibrium, market-clearing price for offsets at which marginal WTP and marginal WTA are equal, that is, where supply for credits equals the demand for credits.

Simulation
To demonstrate the operation of a biodiversity offset market with the ∑α-irreplaceability metric, we simulated the probability of species occurrence in a 64 × 64 patch (or site) landscape.We used the R packages NLMR and landscapetools to control the degree of spatial autocorrelation in the baseline environmental gradient (Sciaini et al., 2018).However, α irreplaceability is determined by the global availability of that species, not its distribution, and the simulation of maps was solely intended to communicate the parallels with field data and empirical models.We simulated 3 communities, each with 200 species, with distributions that were either equally distributed across the environmental gradient or moderately and highly skewed toward one extreme to produce an overall gradient in richness (Leroy et al., 2016).We ran offset market simulations based on subsets of species from each community.The number of species rose from 5 to 50, and each simulation was repeated 10 times.More complex arrangements in response to multiple gradients are easily generated, but we did not consider them here.Likewise, we did not account for time lags or uncertainties in the ability of conservation actions to generate offset credits.
Four further pieces of information were generated for each site.The values of each patch of land for agriculture and for housing development were generated by defining their correlation to the environmental gradient (ranging from 0 to 1), although without a clear rationale for how these costs are expected to covary, both correlation coefficients were set to zero simulations.To reduce the likelihood that market trading stalls when WTA < WTP (and where therefore potential gains from trade still exist), the mean development value was set to double that of agricultural value.Next, each site was assigned to 1 of 3 initial land-use classes: agriculture, conservation, and development in a 70:20:10 split.Finally, a habitat layer was generated to indicate where habitat, and hence species, currently occurs on agricultural land and in conserved sites so as to define the baseline from which gains should be compared.Agricultural land patches without habitat but with suitable environmental conditions for species to occur were treated as areas with restoration potential.The final inputs were the conservation targets for each species.To illustrate a scenario of net gain, rather than no net loss, we set targets in all scenarios to be the equivalent of each species existing availability plus 20% of their restoration potential at agricultural sites.
After each offset trade, the ∑α irreplaceability was recalculated for all sites.An agricultural site that was not irreplaceable for any species (all individual α < 1) and had the greatest WTP per unit loss in the metric (£/∑α) was selected for development.If the development site ∑α was 0 because the site had no species potential or because all species with potential had already achieved their targets, then no offset was required.Otherwise, an offset site with the lowest WTA per unit gain in the metric (£/∑α) was selected and either all or a fraction of species values at that site were assigned to the conservation class.The species values at the developed sites were removed from the global total R ′ s , and the values added by the offset were deducted from the remaining targets (t s ).These steps were repeated until all species conservation targets had been achieved or until there were no mutually beneficial opportunities to trade in biodiversity credits remaining (i.e., a market equilibrium where for all remaining sites WTP < WTA).
To rate the performance of an offset market based on ∑α irreplaceability, we compared the efficiency with which targets were achieved using alternative metrics for the same landscape.First, the R package prioritizr was used to identify the exact optimal combination of sites that achieved all conservation objectives for minimal cost (Hanson et al., 2022).Second, the offset market was rerun with 3 alternative site-based metrics that increasingly reduced the need for the information involved in strategic planning.The first offset metric (OM1) weighted site scores by the inverse of each species range, thereby favoring the rarest taxa in the landscape (Crisp et al., 2001).The OM1 scores were continually updated to reflect changes in global availability due to the market.The OM1 metric would require the same degree of knowledge as for the calculation of ∑α irreplaceability but without setting targets.Updates to planning unit scores reflected species' global availability but were not complementarity to areas already protected.Offset metric 2 (OM2) was equivalent to OM1, but values for each planning unit were not updated over time, meaning weights for each species were fixed at their starting value.This metric required the same initial understanding of species distributions as for OM1 but did not require an updating register of species affected by previous offset transactions.Finally, offset metric 3 (OM3) was based solely on how many species were present in each site, but not which species, meaning only a map of species richness would be required to guide a market.
The code and a full description of the results are in Appendix S1.

Economic efficiency of irreplaceability
Use of ∑α in an offset market resulted in continuous incremental progress toward conservation targets (Figure 1a).The potential economic gains from trade were realized as long as developers WTP exceeded farmers WTA, and this trading allowed conservation targets to be achieved for all species.Economic gains from trade were initially high when trading began (WTP >> WTA) (Figure 1b), but they declined rapidly as more expensive and less irreplaceable offsets were required to meet demand.At each stage, the market favored the greatest gains toward targets at the least cost, making an economically efficient solution more likely.Conversely, ∑α strongly disincentivized development on land with high ∑α scores because the number of offset sites required to replace their loss was typically prohibitive (Figure 1c).
The ∑α irreplaceability did not specifically prioritize sites that contained species rarely found in the landscape; it valued sites based on the difficulty of achieving conservation targets without them.Nonetheless, because there were typically fewer opportunities to conserve rare species (i.e., low replaceability), sites that contained those species tended to have high scores.Once a species target was reached (green line Figure 1d), its contribution to the ∑α of remaining agricultural sites was zero, meaning there was no benefit to its presence in new offsets or cost associated with its occurrence at new development sites.Nonetheless, some species eventually exceeded their targets because they were present at offset sites added later to achieve targets for other species (Figure 1a,d).Because the ∑α contribution of species with targets that had been met was zero, the burden for developers decreased and their WTP for offsets increased at sites that contained species whose targets had been met (red line Figure 1d).

Accounting for more species in the market
The distribution of biodiversity, in particular the degree of spatial overlap among multiple targets, determines the extent to which additional sites are required to protect additional species.The network was specific to the assemblage and how the ecological community correlated spatially with economic land values (Figure 2).Accounting for conservation targets of more species did not in itself increase the cost of conservation solutions, require more trades, or demand more space to meet targets (Figure 2b,c).

Irreplaceability-led offsetting versus optimal prioritization
Site prioritizations generated by SCP were mathematically optimal; they minimized the cost of land needed to achieve all conservation targets (Figure 3).When we assumed developer's WTP was sufficient to support continued trading, the ∑α offset market achieved all targets based on a very different networks of sites than the SCP solution (Figure 3b).In some cases, the ∑α offset market required fewer sites in total (Figure 3a), but the total cost of conserved sites was always equal to or, more likely, greater than SCP solutions.The total cost of sites selected for conservation in an ∑α offset market was only 2-11% greater than that with SCP, but this gap narrowed as the number of species increased (Figure 3c) because the flexibility by which all targets could be achieved was reduced.Conservation solutions selected by the market were more expensive than networks selected by SCP because the latter minimized the total agricultural value of properties included (WTA), but did not consider whether the sites also provided high returns to developers (WTP).Therefore, although irreplaceability ensured all conservation targets were met as efficiently as possible at the time of each trade, deviations from SCP solutions increased in likelihood as the mean WTP increased or as the correlation between WTP and conservation priorities increased.

Irreplaceability versus simpler offset metrics
T Markets where trade was governed by OM1, OM2, and OM3 typically failed to achieve all the targets (2%, 22%, and 1%, respectively), even when property values increased to support continued trading (Figure 4).The OM2 metric (sites weighted by species rarity) was only more successful because targets in all our scenarios were directly proportional to their availability; hence, this was the only situation in which fixed weighting could sometimes be appropriate.Yet, the few occasions when alternative metrics did achieve all targets relied on the subset of species selected to have narrow distributions, which restricted the flexibility of selection.In the minority of scenarios in which alternative metrics were successful, solutions were achieved with a higher number of sites and at greater cost (115-130%) than with ∑α, and none were successful for a large number of species.

DISCUSSION
Land use and land management are central to addressing challenges of global biodiversity conservation, as well as food security, poverty alleviation, and climate change mitigation (Meyfroidt et al., 2022).The failure to coordinate appropriate and effective actions across sectors not only undermines commitments to drive a recovery of nature, but also further risks the sustained well-being of people.We found that if relevant parties engage in trading of biodiversity credits based on a metric derived from ∑α irreplaceability, an offset market can support the most efficient trajectory toward all conservation FIGURE 2 Variation in ∑α-irreplaceability (defined in text and in Figure 1) offset market trading outcomes as the species richness of communities increases.Assemblages were drawn from communities of 200 with (a, b, c) a strong richness gradient or (d, e, f) no richness gradient (WTP, willingness to pay; WTA, willingness to accept).All conservation targets were achieved in each market simulation, and lines of best fit were added based on local polynomial regression.

FIGURE 3
Illustration of the relation between conservation networks selected by ∑α-irreplaceability offset market trading (defined in text and in Figure 1) and optimal systematic conservation planning (SCP) outcomes for the simulated community with a strong richness gradient: (a) ratio of network size, (b) percentage of planning units that overlap the optimal network, and (c) ratio of network cost.targets.That is, designing an offset market with ∑α irreplaceability as its metric delivered a low-cost way of meeting biodiversity targets.
Our approach challenges the current school of thought that states like-for-like trading should be mandatory in a policy design to ensure no net loss (or achieve a net gain in biodiversity) (Bull et al., 2015;zu Ermgassen et al., 2020).As a metric, ∑α irreplaceability relaxes the need for equivalent species in each transaction and instead motivates restoration of species and FIGURE 4 Conservation success and efficiency of offset market solutions based on ∑α irreplaceability (defined in text and in Figure 1) and alternative offset metrics described in text (OM 1-OM 3).Costs are based on the value of farmland assigned to conservation solutions.ecosystems in greatest need (relative to targets) and identifies where that action is most efficient economically.This element of prioritization ensures offsetting conserves the most important sites and species first, irrespective of whether they face direct development pressure.Indeed, the rationale for such prioritization is entirely transparent, and, although many targets are combined to effectively rank each site, this can easily be traced back to its value for different conservation targets.It has been hypothesized that increasing the complexity of offset trading metrics, in a similar vein to ∑α irreplaceability, is likely to reduce the number of trades and hence the economic efficiency of the policy instrument (Needham et al., 2019).In contrast, we found that simpler metrics are unlikely to achieve their primary goal or guide effective progress toward conservation targets.We also found that the economic cost of solutions based on ∑α irreplaceability was not dependent on the number of conservation targets considered.In line with previous research, we found (continue to use past tense for your results) that the location of offset sites and overall cost of conservation actions are dictated by the overlap among ecological targets and ecological and economic heterogeneity across the landscape (Doyle & Yates, 2010;Drechsler, 2021;Kangas & Ollikainen, 2019;Simpson et al., 2022).Finally, when we selected conservation targets that exceeded species' initial availability because we anticipated restoration potential, then net gain, rather than no net loss, was achieved at the market scale.
The adoption of systematic planning tools allows conservation objectives to be achieved efficiently, but rather than relying on new national parks and reserves to stall biodiversity loss, our intention was to recognize the value of effective off-reserve management (Wilson et al., 2007) and engage private finance in conservation.SCP algorithms may define optimal solutions to meet all conservation targets, but in practice these networks are hard to implement when land is privately owned and landowner decisions are based on the relative payoffs from alternative uses (Knight et al., 2011;McIntosh et al., 2017).Although high ∑α values and associated offset costs would incentivize developers to consider alternatives for development to some sites in the SCP network, if WTP is still sufficiently high to be profitable, then the offset market must settle for a more expensive complement of sites to replace non-irreplaceable options.The basis of SCP is that priority sites are not simply the cheapest or the most ecologically diverse; rather, these sites best complement and add to what is already conserved.By introducing regulations requiring developers to offset the predicted impacts of development on biodiversity, a biodiversity offset market generates a positive financial return for farmers investing in conservation that did not exist prior to this market being created.We demonstrated that ∑α irreplaceability is an effective market metric that allows farmers and developers to independently engage in trades, while ensuring an underlying strategic approach is taken to secure the targets deemed critical to biodiversity conservation.
An ongoing problem in the successful implementation of biodiversity offset markets, and environmental markets more broadly, is the lack of regulatory capacity to implement the program with an emphasis on the follow-up monitoring of newly created sites (BenDor et al., 2009;Brownlie et al., 2017;zu Ermgassen et al., 2021).Similarly, a market based on the ∑α metric could result in higher transactions costs.The metric is dynamic because the values of sites would ideally be recomputed after each successful trade.The uncertainty these updates create may lead to lower gains from trade being realized, eroding the ability of the offset market to deliver conservation actions cost-effectively.We did not address these potential costs.

Avoiding previous mistakes
The quality of knowledge of biodiversity is critical to estimating the appropriate allocation of land for conservation and to quantify trade-offs.Rather than rating performance according to the resources or finance committed, ∑α credits provide the greatest reward to landowners able to deliver high marginal gains in ecological outcomes at low financial cost (Pressey et al., 2021).However, to identify the importance of a site in achieving conservation targets, ∑α-irreplaceability credits combine knowledge of how ecological assets are distributed throughout the market's jurisdiction, not just in sites associated with offset trading.Such information is not static and should be updated routinely by the market metric to reflect their changing stocks.The same information would still be required to weight the alternative metrics in Figure 4, but they typically failed to achieve conservation goals because they could not recognize when losses would be regarded as irreplaceable.Given that inadequate monitoring has been cited as a key constraint to global action for many years (Pressey et al., 2021), as well as in the context of prior attempts to organize biodiversity markets (Kujala et al., 2022;Maron et al., 2012;zu Ermgassen et al., 2021), a change in approach is required if biodiversity is to be valued correctly.
First, a key principle underpinning ∑α-irreplaceability market offsets is that losses to development would not be sanctioned if they cannot be replaced.It is key that the market represents as many asset types as possible, even if their distribution is uncertain, to avoid unintentional losses of biodiversity being permitted because those features were absent from ∑α calculation (Popov et al., 2022).In this study, the wide variation in outcomes for different subsets of taxa illustrated the risks associated with conservation policies reliant on small numbers of indicator species whose suitability to represent the conservation needs of biodiversity and ecosystem processes is unknown (Yong et al., 2018).In this context, the value of ecological monitoring data gains new meaning.If understanding of an ecological feature, such as species distribution, is poor, one should err on the side of caution and protect a higher number of sites to be confident the target has been reached (IUCN, Without this prudent approach, land and ecological assets on which society depends may be lost before there is the knowledge to react.If caution due to data shortages leads to an overestimation of the area required to achieve targets, this increases the difficulty of achieving targets and consequently the financial costs of offsetting for developers.It would therefore be in the interests of both market regulators and developers to improve monitoring to minimize the uncertainty of site's ∑α irreplaceability, balancing the cost of further monitoring against expected efficiency gains for the market (Bolam et al., 2019;Eyvindson et al., 2019).In addition, although the cost of monitoring has traditionally been prohibitive, modern tools such as acoustics, molecular methods, automated imaging, and remote surveys with drones and satellites have dramatically increased the ability to monitor many ecological systems at scale (Besson et al., 2022;Keitt & Abelson Eric, 2021).It is beyond the scope of our paper to provide an overview of these methods, but the capacity to efficiently verify restoration outcomes will grow, particularly if sampling design can be strategically adapted to minimize uncertainties in ∑α (Brown et al., 2013).
The biodiversity market is created by a demand for credits.In our simulated market, trading was enforced by a regulator, rather than emerging from a voluntary demand for credits.However, the guarantee that conservation targets will be safeguarded and eventually achieved cannot be made if developer participation in offset trading is voluntary.The market regulator receives updates from monitoring sources to maintain oversight of each asset's progress toward targets at the market scale, thereby determining local site ∑α scores and the credits required for trades (Kujala et al., 2022).The regulator is also able to intervene in the economic efficiency of the market, for example, by subsidizing restoration costs to farms to increase the market supply of ∑α-irreplaceability credits.Although we recognize that defining site ∑α irreplaceability based on the potential recovery of a site (Sutherland, 2022) and forecasting of the time frame and risks (Ladouceur et al., 2023;Laitila et al., 2014) are challenging, these uncertainties are motivations for targeted research, rather than barriers to adoption (Bolam et al., 2019;Eyvindson et al., 2019).
Public support and trust will be strengthened if individuals can transparently understand how local, and potentially highly visible, losses are accompanied by secure gains at the landscape scale designed to benefit society and the economy (Cvitanovic et al., 2021).Landowners with spatial, strategic advantages due to the location of their land may be able to leverage payments from developers that are well in excess of their opportunity costs if their property is key to achieving a conservation target (Lennox et al., 2012).

Beyond biodiversity offset markets
Even with introduction of planning regulation, to avert substantial biodiversity loss and degradation of ecosystem services, society must raise its ambitions to restore ecosystems (Leclère et al., 2020).The resources available for conservation action are woefully inadequate compared with the resources invested in activities that further degrade or destroy nature (Dasgupta, 2021), and yet the expected benefits of conservation investment often far outweigh the costs (Bradbury et al., 2021;DEFRA, 2022).The evidence of an ecological crisis is so serious that any action or investment is seen as positive, but this lack of discrimination weakens the motivation of individuals and companies to support more transformative change.The ∑α-irreplaceability credits can be used to recognize and reward private investment in conservation because they provide a comparable metric of performance in a market, even if 2 sites or actions affect different ecological assets.
In an ∑α-irreplaceability market, investors can anticipate the relative costs of their actions and define the performance of their investments in restoration and conservation for biodiversity in net terms.The ∑α-irreplaceability metric could therefore be key to allowing fair recognition of investors' contributions and be used to build public trust that companies' statements of environmental responsibility match their claims.
The debates associated with pathways to sustainability and a nature-positive recovery are highly value laden, wicked problems (DeFries & Nagendra, 2017;Meyfroidt et al., 2022), but societies cannot expect ecosystem recovery to emerge from a piecemeal approach.Land is finite, and reconciling demands and interactions of complex multisector systems requires strategic oversight to avoid scenarios of ecological, economic, and societal collapse (Shin et al., 2022;Steffen et al., 2015).Ecologists can identify what targets are required as a minimum to sustain species, ecosystem, or process, but targets must ultimately be defined collaboratively with economists, social scientists, health economists, and politicians.Incentivizing outcomes using insights from systematic planning will become increasingly important as the collective benefits of multiple land uses diverge (Jung et al., 2021;Moilanen et al., 2005).Adopting ∑α irreplaceability would enable authorities to identify and minimize the conflict between conservation targets and other land uses, thereby incentivizing greater private sector investment in actions that accelerate the speed with which nature's recovery is achieved.

FIGURE 1
FIGURE 1Example of a simulated offset market based on ∑α irreplaceability (i.e., the combined progress a site would deliver toward conservation goals for all species relative to the opportunity to achieve those goals elsewhere): (a) progress toward conservation targets (dotted line) for each species as new developments requiring offsets are built, (b) decline in the log ratio between willingness to pay (WTP) and willingness to accept (WTA) as representative of gains from trade, (c) distribution of values for purchasing agricultural land in the simulated landscape and the final proportion of those selected for development and conservation offsets, and (d) changes in the allocation of a single species (also identified in panel [a]) among land types as trading progresses.