
==== Front
Biol Lett
Biol Lett
rsbl
roybiolett
Biology Letters
1744-9561
1744-957X
The Royal Society

rsbl20240283
10.1098/rsbl.2024.0283
1001100110016069Conservation Biology
Review Articles
Are agricultural commodity production systems at risk from local biodiversity loss?
Are agricultural commodity production systems at risk from local biodiversity loss?
https://orcid.org/0000-0002-6907-8178
Maney Calum 1 2 Conceptualization Data curation Formal analysis Investigation Methodology Visualization Writing – original draft Writing – review and editing calum.maney@unep-wcmc.org

https://orcid.org/0000-0001-8844-7437
Sassen Marieke 1 Supervision Writing – original draft Writing – review and editing marieke.sassen@wur.nl

https://orcid.org/0000-0002-5998-4652
Giller Ken E. 1 Supervision Writing – original draft Writing – review and editing ken.giller@wur.nl

1 Plant Production Systems, Wageningen University, P.O. Box 430 , Wageningen 6700, The Netherlands
2 UN Environment Programme World Conservation Monitoring Centre (UNEP-WCMC), 219 Huntingdon Road , Cambridge CB3 0DL, UK
Electronic supplementary material is available online at https://doi.org/10.6084/m9.figshare.c.7430648.

9 2024
18 9 2024 September 18, 2024
18 9 2024 September 18, 2024
20 9 2024028321 5 2024 May 21, 2024
09 7 2024 July 9, 2024
22 7 2024 July 22, 2024
© 2024 The Author(s).
2024
https://creativecommons.org/licenses/by/4.0/ Published by the Royal Society under the terms of the Creative Commons Attribution License http://creativecommons.org/licenses/by/4.0/, which permits unrestricted use, provided the original author and source are credited.

Compelling evidence for feedbacks between commodity crop production systems and local ecosystems has led to predictions that biodiversity loss could threaten food security. However, for this to happen agricultural production systems must both impact and depend on the same components of biodiversity. Here, we review the evidence for and against the simultaneous impacts and dependencies of eight important commodity crops on biodiversity. We evaluate the risk that pollination, pest control or biodiversity-mediated soil health maintenance services are at risk from local biodiversity loss. We find that for key species groups such as ants, bees and birds, the production of commodities including coffee, cocoa and soya bean is indeed likely to be at risk from local biodiversity loss. However, we also identify several combinations of commodity, ecosystem service and component of biodiversity that are unlikely to lead to reinforcing feedbacks and lose–lose outcomes for biodiversity and agriculture. Furthermore, there are significant gaps in the evidence both for and against a mutualism between biodiversity and agricultural commodity production, highlighting the need for more evaluation of the importance of specific biodiversity groups to agricultural systems globally.

ecosystem services
; agricultural biodiversity
; agricultural commodities
; pollination
; pest control
; soil health
Direktoratet for Utviklingssamarbeid http://dx.doi.org/10.13039/100007843 Economic and Social Research Council http://dx.doi.org/10.13039/501100000269
==== Body
pmc1. Commodity crop production intersects with the threatened biodiversity that underpins ecosystem services

Much of the world’s agricultural land lies within biodiversity-rich areas and is used to produce commodities ‘for export’ at the expense of local biodiversity [1,2]. The expansion and intensification of agricultural production have been rapid and are among the most important drivers of global biodiversity loss [3]. At the same time, biodiversity in agricultural landscapes underpins several ecosystem functions, such as wild pollinator activity and the cycling of soil nutrients [4,5]. These functions contribute to the stability of agricultural systems [6] and their resilience to fluctuations in environmental conditions [7,8]. Human-mediated impacts on ecosystem services are also increasing: indicators designed to measure pollination, climate, freshwater quality and quantity, the regulation of biological processes, and the provision of food and materials all show considerable declines since 1970 [9].

Observable relationships between agricultural biodiversity and desirable outcomes have led to the idea of a ‘biodiversity–production mutualism’ in agricultural landscapes. This theory suggests that the functions associated with biodiversity contribute to agricultural production [10]. Accordingly, if agricultural expansion and intensification exceed a sustainable limit, ecosystem services will be degraded and crop yields reduced through the loss of functional groups of biodiversity [11–13]. For example, increasing demand for commodities that depend on pollination services makes the delivery of pollination services more important to global production landscapes [11]. At the same time, increasing land-use intensity and the expansion of cropland are associated with falling richness and abundance of insect pollinators [14]. Systems that impact ecosystem services such as pollination too heavily could become caught in an ‘intensification trap’, whereby they become increasingly dependent on manufactured inputs to replace lost ecosystem services, reducing the profitability of the system in a lose–lose situation for biodiversity and agriculture [15,16] (figure 1).

Figure 1. Ecosystem service feedbacks may drive a reinforcing feedback loop with negative outcomes if production is dependent on ecosystem services. Arrows represent hypothesized causal relationships.

Ecosystem service feedbacks may drive a reinforcing feedback loop with negative outcomes if production is dependent on ecosystem services.

At the global scale, simultaneous increases in crop system impacts and dependencies on biodiversity could lead to a future where the production of key commodity crops is jeopardized [13]. Conversely, the ‘interdependence’ of agriculture and biodiversity means that nature-based solutions could be valuable to support biodiversity and consequently, food production. Utilizing nature-based solutions to capitalize on the biodiversity–production mutualism may even reduce the need for continued expansion of agriculture into natural areas [17]. Interdependencies are well-known for some species groups such as birds, where context-specific examples of impacts and dependencies are well-described [18,19]. However, the biodiversity–production mutualism is a broad concept, which lacks detailed causal evaluation, and needs contextualization.

2. Projecting the impacts of biodiversity decline on agriculture may overlook context-specificity

Evidence for a biodiversity–production mutualism and the risk of ‘intensification traps’ has fuelled powerful statements about the value of biodiversity to agriculture [20,21]. This evidence also informs recent global analyses of the risks that declining biodiversity may pose to global agriculture and food security [11,13,22,23]. Global-scale projections of the future of biodiversity and agriculture that incorporate interdependencies between the two often predict large shocks to food systems driven by biodiversity loss [13,15]. At the global level, these findings are important: they highlight a pressing concern that warrants policy responses commensurate with the threat posed to people and nature. However, the biodiversity–agriculture interdependencies underpinning these findings are based on, often broad, theoretical assumptions [16,24]. Dependency relationships in particular tend to be simplified at least to crop level; for pollination, this often means using the classification of pollination dependency from Klein et al. [25] alongside estimates of habitat proximity (e.g. [23]) or an assumption of linear decreases in yield proportional to crop dependence on pollination (e.g. [13]). Yet, when considering practical options to improve agronomic and ecological outcomes in individual cropping systems, it is important to consider in context the specifics of which biodiversity is important, and the extent to which it is threatened by agriculture [26].

The relationship between landscape-level natural habitat, functional group biodiversity, ecosystem service provision and crop production has been summarized quantitively across many crops and components of biodiversity. Overall patterns indicate evidence of interdependence between agriculture and biodiversity [27]. However, analyses to estimate interdependence in individual cropping systems or for specific components of biodiversity are limited by the availability of data, and patterns at the global level may not hold in specific contexts.

Realistically, in any given instance, only a small proportion of species are likely to contribute directly to ecosystem services and yield [28]. These are also often the most abundant species within a given functional group [29], which weakens the overall argument for biodiversity–productivity co-benefits. While there is strong evidence that biodiversity–production mutualisms exist, they are not ubiquitous and need to be established in each case [30,31]. Depending on the specific circumstances, priorities to protect ecosystem service provision and actions to protect biodiversity may not always align [28,31]. These circumstances can be related to many properties of the production system, including geographic region, farm scale and cropping system [32–34]. At best, the argument to increase all biodiversity within agricultural systems may be inefficient in supporting production. At worst, holding biodiverse agroecosystems as an ideal could risk threatening food security and local livelihoods. Further, farmers can reduce their dependence on natural systems by introducing non-natural substitutes and enhancements. For example, by using bee hives, farmers may reduce the dependence they would otherwise have on nearby forests to provide pollinators [35]. Smallholder or ‘family’ farmers are often particularly dependent on ecosystem services for crop production as they often have fewer resources to invest in external inputs, especially in the tropics (e.g. [36]). They produce 30% of global food supply [37], highlighting the importance of ensuring tipping points of biodiversity degradation are not crossed.

Crop failure can mean disaster for farmers, so agronomic and ecological research must be thorough and nuanced when evaluating biodiversity as a tool to sustain productivity. If trust and enthusiasm for protecting local biodiversity to prevent food system collapses is to be garnered within agricultural communities, the risk of interdependencies that lead to productivity declines must be established in real-world commodity production systems. Of course, there are many reasons to conserve biodiversity beyond a recognition of its contributions to people—moral, ethical and biocultural values also guide human interactions with nature [38]. Where such other reasons to conserve a component of biodiversity dominate but are not shared by farming communities, incentives and/or compensation for any trade-offs with farming will need to be found.

3. Investigating interdependencies with a case study approach

The goal of this review is to establish, for a range of crops, whether a reinforcing feedback between biodiversity loss and low productivity is likely to exist. We began by combining existing evidence from agronomic and ecological studies. Due to the varied nature of the evidence collected, it was impractical to conduct a quantitative meta-analysis such as has been done for individual services, crops or biodiversity features globally [27,39]. Instead, we opt for a case study and narrative approach, focusing on the findings of specific studies, to identify the balance of evidence either supporting or contradicting the two hypotheses. To do this, we reviewed evidence of how biodiversity supports productivity, comparing it to the evidence that those production systems put the same biodiversity at risk. We focus on eight economically and ecologically significant commodity crops that are predominantly exported and are known to grow in areas of high biodiversity [40,41]. Such a selection limits the findings of our review to a specific set of socio-economic and geographic conditions but should more reliably capture systems where there are material impacts of agriculture of biodiversity and vice-versa. These were cocoa (Theobroma cacao L.), coffee (Coffea arabica L.; also C. canephora P. and others), cotton (Gossypium hirsutum L.), oil palm (Elaeis guineensis Jacq.), rubber (Hevea brasiliensis (Willd. ex A. Juss.) Müll. Arg.), soya bean (Glycine max (L.) Merr), sugar cane (Saccharum officinarum L.) and tea (Camellia sinensis (L.) Kuntze).

Our review is targeted at evaluating the evidence for and against two hypotheses, both of which must be true for reinforcing feedbacks between biodiversity loss and low productivity to occur. The first is that, at the local level, the productivity of commodity cropping systems depends upon specific components of biodiversity. The second is that these same components of biodiversity are impacted negatively by the expansion and intensification of specific cropping systems. Following literature searches using the Web of Science tool (https://www.webofscience.com/) and constructed search terms (electronic supplementary material, table S1), we identified the components of biodiversity implicated in delivering ecosystem services. For each component, we reviewed the evidence that cropping systems benefit from biodiversity, focusing on studies which measured both a component of biodiversity and a measure of productivity. We then reviewed the evidence on the impacts of the cropping system impacting each component. This limited our review to components of biodiversity that are already known or assumed to be useful to production systems.

Our initial searches identified a total of 159 scientific papers containing findings about the impacts and dependencies of commodity production systems on biodiversity. This limited number of findings was surprising, though perhaps less so when the specific requirements for evidence on ‘dependency’ are considered. There are only a small proportion of published articles concerning biodiversity and ecosystem services that directly relate the diversity of a component of biodiversity, through the provision of an ecosystem service, to productivity in crops. For instance, the CropPol database, the largest open database on biodiversity–crop pollinator interactions, covers 202 studies over 32 crops—a much larger range than the scope of this study [42]. Thus, we were satisfied a genuine data gap has been identified. Our review covered nine components of biodiversity, some of which were nested, such as ‘ants’ within ‘insects’. Our follow-up searches for impacts and dependencies of production systems on the components of biodiversity uncovered 46 further sources, including for derivative and prior research linked to sources in the original search, and those in languages other than English. In total, our review covered eight commodity crops, three ecosystem service categories and nine main components of biodiversity. The total number of ‘findings’ was 151 for dependencies of crop systems on biodiversity, and 163 for impacts of crop systems on biodiversity. We summarized these in both a narrative format (electronic supplementary material, table S2) and figures (figures 2 and 3).

Figure 2. The distribution of evidence collected in the review, arranged by crop and ecosystem service. (a) Evidence collected on the hypothesis of dependence. (b) Evidence collected on the hypothesis of impacts. Fill colour represents the balance between evidence broadly supporting hypotheses, where red (the top colour) represents the number of findings not supporting the hypothesis or with only context-specific support for it.

The distribution of evidence collected in the review, arranged by crop and ecosystem service

Figure 3. The evidence bases for ecological impacts and agronomic dependencies of commodity crop production systems on biodiversity. Shaded areas mostly in the top-right quadrant belong to combinations of crop, ecosystem service and biodiversity component, where evidence is likely to support the hypothesis that interdependencies, feedback loops and ‘intensification traps’ could emerge due to crop-induced biodiversity change. The shaded regions represent the 95% percentile mass of the joint posterior distribution of findings’ support for the impact and dependence hypotheses in the reviewed evidence. This therefore represents the likely outcome of seeking both evidence for dependence and evidence for impacts of a given cropping system via one of the three ecosystem services. A complete breakdown of the results by crop is also available for clarity (electronic supplementary material, figure S2).

The evidence bases for ecological impacts and agronomic dependencies of commodity crop production systems on biodiversity.

To illustrate how these findings translate into the likelihood of observing effects of interdependencies in the eight target cropping systems, we identified findings from the context-specific evidence base simply as supporting (Y), mixed (Y and N) or contradictory (N) to the dependency impact hypotheses. We then summarized the evidence base in a simple Bayesian framework [43], with a prior assumption that finding evidence to support or contradict the two hypotheses was equally likely:

Ydependence ∼ Binomial(ndependence,pdependence)

pdependence ∼ Uniform(0,1)

Yimpact ∼ Binomial(nimpact, pimpact)

pimpact ∼ Uniform(0,1)

R=Ydependence ⋅ Yimpact,

where n represents the number of findings related to each of the two hypotheses that were identified for each combination of crop, service and component of biodiversity; the Y variables represent the number of those studies supporting the hypotheses and overall risk, R, represents the probability that two pieces of evidence, one concerning each hypothesis, will reflect impacts and dependencies of a cropping system on a component of biodiversity.

4. The evidence base for impacts and dependencies is distributed differently among biodiversity groups

As reflected in the combined evidence for impacts and dependencies of agricultural systems on wild biodiversity [27], the findings of this review show consistent evidence both for agriculture degrading the diversity of potentially beneficial species groups and for the measurable benefits of those species groups to productivity. However, the evidence base for impacts and dependencies is distributed differently among biodiversity groups. When results are separated by their relevant component of biodiversity, evidence concerning some species groups is weaker or may even indicate a lack of dependence or impact (figure 3; electronic supplementary material, figure S2).

Rather than reiterating the findings of each study contributing to the results of this review (figure 3; electronic supplementary material, table S2), here we discuss illustrative case studies, apparent data gaps and departures from the overall trend that interdependencies between agriculture and biodiversity are present in the reviewed commodity production systems.

The evidence base substantiates pollination service interdependencies across most investigated components of biodiversity. This reinforces the value of broadly applied models of this ecosystem service, which typically focus on bee species [44]. While the evidence for interdependence in the case of other pollinator groups was more mixed, it was still mostly supported. The evidence for interdependencies was strongest in coffee and cotton, though benefits have been found to depend on important interactions between pollinators and other insects.

Oil palm systems present a well-known departure from typical pollination interdependencies. Oil palm pollination depends on the oil palm weevil Rhynchophorus ferrugineus. This species lives within the palm itself and is not known to be affected by the habitat degradation typically associated with oil palm production. Yet periods of heavy rainfall and nematode infections have been associated with pollinator declines and low yield in oil palm [45]. More research is warranted on interdependencies linked to pollination in some of our crops. For example, cocoa is highly dependent on pollination [25], cocoa yields are low in most production systems, especially in West Africa and hand pollination is known to increase yields [46–48]. The abundance of Diptera in cocoa plantations increased with increasing shade and shade complexity in one study [49]. Yet other practices with narrower biodiversity co-benefits, such as adding rotting banana pseudostems to farms, can also benefit pollination [50]. The evidence suggests that enhanced pollination services would benefit cocoa production, but the key target pollinator species and the best ways to promote them remain uncertain.

There are key discrepancies in some findings related to biodiversity dependencies, for instance, in soya bean pollination. One recent article argues that current systems ignore the importance of this service to the crop, and its potential to help maintain global soya bean production while sparing land for biodiversity [17]. Indeed, different studies find that insect pollinators may contribute to soya bean productivity [44,51–55], and the design and management of soya bean production systems mediate the diversity of insect pollinators [54,56,57]. This suggests that soya bean production systems may experience feedback loops related to pollination interdependency. Yet soya bean is a self-fertilizing crop; how this can be reconciled with the observed contributions of pollinators to soya bean yields is debated [34]. Secondary mechanisms, such as the revival of relegated flowers when pollen is delivered to them, have been suggested as an explanation [52], though further research is needed to make a stronger case for biodiversity within the production system from the perspective of soya bean growers. Overall, nature-based solutions to pollination-limitation, such as on-farm habitat provision [58,59], are likely to become more important as pollinator-dependent commodity crops expand and intensify worldwide [11]. Indeed, some coffee and cotton supply chains already integrate knowledge of this interdependence in their policies and strategies. For example, Texas, the largest cotton-producing state in the USA, has a pollinator conservation plan targeting butterflies [60]. Nestlé’s ‘bees for coffee’ project [61] seeks to protect pollinators to benefit their ‘regenerative agriculture’ plans.

For pest control, known to be a more varied and context-specific ecosystem service than pollination [62], the evidence showed a more variable risk of interdependencies. Departures from generalized expectations of both impacts and dependencies on biodiversity were found for bats, birds, non-flying vertebrates and spiders across multiple cropping systems. Most of the studies on the benefits of flying predators did not consider birds and bats separately [63,64]. In the few studies that did separate effects, benefits from bats were lower [65,66], suggesting that birds may have driven most of the service provision in the mixed studies. Studies on invertebrates generally supported both the dependence and impact hypotheses [67,68], though some groups such as spiders were not studied in all systems. A recent review of the economic benefits of pest suppression by bats found that no exclosure experiment to measure the pest control benefits of bats in cotton had been carried out [39]. This may be particularly pertinent in Southeast Asia and the Amazon, which are both global centres of cotton production and bat diversity [69,70]. We did not find any study linking natural enemy diversity to rubber or sugar cane production. Yet, these commodity crops can have significant negative impacts on the diversity of the species groups implicated as natural enemies in other crop production systems [71–73], so research into their role in these systems is warranted. Conversely, while spiders were identified as a potential group around which a feedback could develop in tea, the negative impacts of tea plantations on spider diversity were found to be limited [74,75]. Non-flying mammals were not typically implicated as beneficial natural enemies (e.g. [76]), and though there was evidence for pythons as predators of the common rodent pests in oil palm, python diversity has not been linked to yield [77]. Finally, in agroforestry systems such as those used in cocoa and coffee production, we identified evidence for knock-on benefits to pest control from trees, though more evidence supporting specific species, benefits and potential trade-offs is needed to fully understand interdependencies in these systems [78,79].

Regarding biodiversity-mediated soil health services, evidence for production benefits was sparser than for other services, though generally positive towards both hypotheses. Evidence largely revolved around plant species associated with cocoa, coffee and tea production systems. Trees supporting soil health made up most of the evidence [80–82], though the structure, placement and identity of trees might be more important than tree diversity itself [83,84]. In systems such as cocoa, there are also important trade-offs with tree densities, as shade removal allows for higher cocoa planting densities, and too much shade may enhance the spread of common fungal diseases [85]. For tea, studies so far have found positive results for single-species agroforestry using alder and gingko trees [86,87]. As tea plantations are often considered detrimental to biodiversity at the site and landscape scale [71], further research into more complex mixes, and the impacts they have on biodiversity, is warranted. Identifying the best range and diversity of species to include in agroforestry systems allows for fine-tuning ecosystem service provision and the maximization of benefits [88].

Evidence for biodiversity-mediated benefits to the production of sugar cane and soya bean via soil health services was limited; some evidence on benefits from non-tree plants was found, but this was relatively sparse. Increases in soil organic carbon have been associated with incorporating woody vegetation into soya bean systems [89], but no direct connection to productivity was observed. Continuous sugar cane replanting harms both soil biodiversity and production [90]; therefore, studies into any dependence on biodiversity in sugar cane systems have the potential to reveal interdependencies.

(a) The case of soil microbial diversity

During our search, we identified several studies that implied a role for soil microbial diversity in supporting the focal crop production systems. Most of these studies involve fertility treatments or other soil interventions where microbial diversity is also a response variable [91–94]. Studies based on interventions that increase both fertility and microbial diversity or activity cannot establish causality between them and, therefore, do not provide reliable evidence.

5. Takeaways and lessons for practice and policy

Broadly, the evidence examined in this study supports the notion that negative consequences can emerge through the impacts of agricultural commodity production on specific groups of local biodiversity, as reflected in recent syntheses [27,95]. Still, interdependencies between nature and agriculture are unique to crop, service and biodiversity component combinations. Even within similar combinations of crop, service and component of biodiversity, different patterns emerge. For example, while both wild bees and butterflies were implicated in delivering pollination services in coffee, the evidence base was much stronger for bees. Geographic patterns in crop impacts and dependencies further complicate relationships (e.g. [34]). Studies investigating only broad functional groups make an inadequate case for protecting groups at a finer taxonomic level that do not provide tangible benefits. Maintaining landscape-level habitat and overall ecosystem health in agricultural matrices overall seems likely to provide benefits to biodiversity with feedbacks benefitting the agroecosystem [24,27]. However, it is important for the content of both public policy and private-sector initiatives to reflect a context-specific and nuanced understanding of biodiversity–agriculture interdependencies.

Across the evidence in this review, the most consistent driver of reinforcing feedbacks and lose–lose situations for agriculture and biodiversity was related to insects, which were consistently found to both support production and be impacted by changes to farms and farming landscapes—including pesticide use [96,97]. Tree diversity was also linked to a variety of on-farm benefits as well as knock-on biodiversity and ecosystem service benefits [85,98].

Not all beneficial species are of conservation concern. Many of the useful plants identified by studies in this review are specific species, and not necessarily wild, natural or native to the regions where their benefits have been measured. This was the case for many cover crops, as well as for some trees [86,87,99].

When land-use change or land-use intensification impact biodiversity, there are ‘winners’ and ‘losers’ among groups of species: some are strongly impacted, while others thrive in modified environments [100]. In some cases, the species responsible for providing benefits to cropping systems may actually be the ‘winners’, reducing the risk of biodiversity-mediated feedbacks on agricultural production even as impacts on biodiversity at large grow [28].

Charismatic, unconventional service providers may benefit commodity production systems, as shown by a scattered but significant amount of evidence. These species, including marsupials in soya bean [101], owls in oil palm [102], and reptiles in cocoa and coffee [103,104], are not usually accounted for in large-scale analyses of biodiversity-mediated ecosystem services. However, human–wildlife conflict remains a pertinent barrier to the suitability of some of these species for promotion as nature-based solutions [105]. While charismatic species associated with pest control services may come from groups typically associated with conservation gains [106], decision-makers need to consider potential local-scale trade-offs, such as danger to people, livestock, crops [105] and property [107].

The development of effective solutions and generating trust in them requires avoiding oversimplifications. The chain of causality between interventions, biodiversity responses, ecosystem functions, ecosystem service provision and production outcomes should always be established through context-specific studies. Though interdependencies are likely in many cropping systems, it is important to avoid making projections of agricultural outcomes related directly to biodiversity loss, where evidence is currently limited or not indicative of interdependencies for the relevant species and crop combinations. Continuing to evaluate and publish evidence when biodiversity is found not to support agriculture, or found to thrive in an intensified environment, is crucial to developing and refining reliable interventions. Furthermore, this will help to establish where alternative options to enhancing ecosystem services are most appropriate for different farming systems. Smallholders in particular should have the right to ‘substitute’ for nature when their livelihoods are dependent on optimal decision-making [35].

Many of the studies included in this review were carried out in systems where some kind of nature-based solution or biodiversity-friendly practice was already implemented. This does not match the reality of many commodity crop production systems, which can be extensive and intensified [108]. Biodiversity baselines are therefore needed for many species groups and cropping systems, especially in intensive agriculture. If we are to understand the extent and importance of feedbacks in commodity production systems, we must also survey biodiversity and ecosystem services in more intensive systems and degraded landscapes to assess the extent to which biodiversity and ecosystems are also degraded. The ‘green deserts’ of large-scale, intensified agricultural commodity production systems have an as-yet-unknown potential for feedbacks between their biodiversity and productivity for most of the crops in this study. Further, there was geographic bias in the studies in this review. When the locations where evidence was found were compared to the overall growing regions of the crops, some areas had evidence focused on impacts or dependence only; other regions represented evidence gaps (electronic supplementary material, figure S1).

Policies are now in place attempting to limit the damage done to natural ecosystems by commodity production systems. The EU restoration law has specific elements addressing pollinators and farmland birds, among other important biodiversity components associated with ecosystem services in agriculture [109]. The EU anti-deforestation regulation [110] also attempts to address the demand-driven impacts of globally traded commodities. This regulation covers, among other crops, cocoa, coffee, oil palm, soya bean and rubber, for which we found evidence for reinforcing feedbacks driven by on-farm changes. If countries importing agricultural commodities wish to stop their trading activities from driving further deforestation, collapses in production driven by feedbacks must also be avoided. Furthermore, countries importing large quantities of agricultural commodities from biodiversity-rich areas need to strike a balance between protecting and restoring their own local biodiversity and protecting biodiversity in the overseas landscapes where commodities are produced [111].

Future trends in global trade will undoubtedly drive further expansion and intensification of key agricultural commodity production systems. Understanding the intricate and context-specific relationships between land-use intensity, specific components of biodiversity, ecosystem services and productivity is crucial to determining the hard limits of such intensification and expansion. Developed rigorously, and implemented well in decision-making, this understanding may be able to circumvent a scenario of biodiversity loss, service failure and unsustainable production.

Acknowledgements

We thank Professor Neil Burgess for his initial review and comments on this manuscript, and Eric Rahn and Isabel Otero for their valuable suggestions of non-English language resources.

Ethics

This work did not require ethical approval from a human subject or animal welfare committee.

Data accessibility

A list of the full collection of resources used in this analysis, and all the code needed to produce the figures in this manuscript has been made available (Supplementary Information) [112].

Declaration of AI use

We have not used AI-assisted technologies in creating this article.

Authors’ contributions

C.M.: conceptualization, data curation, formal analysis, investigation, methodology, visualization, writing—original draft, writing—review and editing; M.S.: supervision, writing—original draft, writing—review and editing; K.E.G.: supervision, writing—original draft, writing—review and editing.

All authors gave final approval for publication and agreed to be held accountable for the work performed therein.

Conflict of interest declaration

We declare we have no competing interests.

Funding

This work was supported by the UK Research and Innovation’s Global Challenges Research Fund under the Trade, Development, and the Environment Hub project (ES/S008160/1), and by NORAD under the CocoaSoils project (RAF-17/0009-CocoaSoils).
==== Refs
References

1. Baudron F, Giller KE. 2014 Agriculture and nature: trouble and strife? Biol. Conserv. 170 , 232–245. (10.1016/j.biocon.2013.12.009)
2. Schwarzmueller F, Kastner T. 2022 Agricultural trade and its impacts on cropland use and the global loss of species habitat. Sustain. Sci. 17 , 2363–2377. (10.1007/s11625-022-01138-7)
3. Jaureguiberry P et al . 2022 The direct drivers of recent global anthropogenic biodiversity loss. Sci. Adv. 8 , eabm9982. (10.1126/sciadv.abm9982)36351024
4. Middleton B, Grace J. 2004 Biodiversity and ecosystem functioning: synthesis and perspectives. Restor. Ecol. 12 , 611–612. (10.1111/j.1061-2971.2004.120401.x)
5. Allan E et al . 2015 Land use intensification alters ecosystem multifunctionality via loss of biodiversity and changes to functional composition. Ecol. Lett. 18 , 834–843. (10.1111/ele.12469)26096863
6. Cardinale BJ et al . 2012 Biodiversity loss and its impact on humanity. Nature 486 , 59–67. (10.1038/nature11148)22678280
7. Conway GR. 1987 The properties of agroecosystems. Agric. Syst. 24 , 95–117. (10.1016/0308-521X(87)90056-4)
8. Frison EA, Cherfas J, Hodgkin T. 2011 Agricultural biodiversity is essential for a sustainable improvement in food and nutrition security. Sustainability 3 , 238–253. (10.3390/su3010238)
9. Díaz S et al . 2019 Pervasive human-driven decline of life on earth points to the need for transformative change. Science 366 , eaax3100. (10.1126/science.aax3100)31831642
10. Seppelt R, Arndt C, Beckmann M, Martin EA, Hertel TW. 2020 Deciphering the biodiversity-production mutualism in the global food security debate. Trends Ecol. Evol. 35 , 1011–1020. (10.1016/j.tree.2020.06.012)32943219
11. Aizen MA et al . 2019 Global agricultural productivity is threatened by increasing pollinator dependence without a parallel increase in crop diversification. Glob. Chang. Biol. 25 , 3516–3527. (10.1111/gcb.14736)31293015
12. Synes NW, Brown C, Palmer SCF, Bocedi G, Osborne PE, Watts K, Franklin J, Travis JMJ. 2019 Coupled land use and ecological models reveal emergence and feedbacks in socio‐ecological systems. Ecography 42 , 814–825. (10.1111/ecog.04039)
13. Millard J, Outhwaite CL, Ceaușu S, Carvalheiro LG, da Silva E Silva FD, Dicks LV, Ollerton J, Newbold T. 2023 Key tropical crops at risk from pollinator loss due to climate change and land use. Sci. Adv. 9 , eadh0756. (10.1126/sciadv.adh0756)37824611
14. Millard J et al . 2021 Global effects of land-use intensity on local pollinator biodiversity. Nat. Commun. 12 , 2902. (10.1038/s41467-021-23228-3)34006837
15. Brown C, Seo B, Rounsevell M. 2019 Societal breakdown as an emergent property of large-scale behavioural models of land use change. Earth Syst. Dynam. 10 , 809–845. (10.5194/esd-10-809-2019)
16. Burian A et al . 2024 Biodiversity–production feedback effects lead to intensification traps in agricultural landscapes. Nat. Ecol. Evol. 8 , 752–760. (10.1038/s41559-024-02349-0)38448509
17. Garibaldi LA, Schulte LA, Nabaes Jodar DN, Gomez Carella DS, Kremen C. 2021 Time to integrate pollinator science into soybean production. Trends Ecol. Evol. 36 , 573–575. (10.1016/j.tree.2021.03.013)34034911
18. O’Connor RJ, Shrubb M, Watson D. 1986 Farming and birds. New York, NY: Cambridge University Press.
19. Whelan CJ, Wenny DG, Marquis RJ. 2008 Ecosystem services provided by birds. Ann. NY. Acad. Sci. 1134 , 25–60. (10.1196/annals.1439.003)18566089
20. Giannini TC, Cordeiro GD, Freitas BM, Saraiva AM, Imperatriz-Fonseca VL. 2015 The dependence of crops for pollinators and the economic value of pollination in Brazil. J. Econ. Entomol. 108 , 849–857. (10.1093/jee/tov093)26470203
21. Borges RC, Brito RM, Imperatriz-Fonseca VL, Giannini TC. 2020 The value of crop production and pollination services in the eastern Amazon. Neotrop. Entomol. 49 , 545–556. (10.1007/s13744-020-00791-w)32557403
22. Zabel F, Delzeit R, Schneider JM, Seppelt R, Mauser W, Václavík T. 2019 Global impacts of future cropland expansion and intensification on agricultural markets and biodiversity. Nat. Commun. 10 , 2844. (10.1038/s41467-019-10775-z)31253787
23. Johnson J et al . 2020 Global futures report: a landmark study into the global economic impacts of nature loss. See https://www.wwf.org.uk/globalfutures (accessed 30 August 2022).
24. Garibaldi LA et al . 2021 Working landscapes need at least 20% native habitat. Conserv. Lett. 14 , e12773. (10.1111/conl.12773)
25. Klein AM, Vaissière BE, Cane JH, Steffan-Dewenter I, Cunningham SA, Kremen C, Tscharntke T. 2007 Importance of pollinators in changing landscapes for world crops. Proc. Biol. Sci.Biol. sci 274 , 303–313. (10.1098/rspb.2006.3721)
26. Seufert V, Ramankutty N. 2017 Many shades of gray: the context-dependent performance of organic agriculture. Sci. Adv. 3 , e1602638. (10.1126/sciadv.1602638)28345054
27. Dainese M et al . 2019 A global synthesis reveals biodiversity-mediated benefits for crop production. Sci. Adv. 5 , eaax0121. (10.1126/sciadv.aax0121)31663019
28. Senapathi D, Biesmeijer JC, Breeze TD, Kleijn D, Potts SG, Carvalheiro LG. 2015 Pollinator conservation—the difference between managing for pollination services and preserving pollinator diversity. Curr. Opin. Insect Sci. 12 , 93–101. (10.1016/j.cois.2015.11.002)
29. Winfree R, Fox JW, Williams NM, Reilly JR, Cariveau DP. 2015 Abundance of common species, not species richness, drives delivery of a real-world ecosystem service. Ecol. Lett. (ed. K Shea), 18 , 626–635. (10.1111/ele.12424)25959973
30. Duncan C, Thompson JR, Pettorelli N. 2015 The quest for a mechanistic understanding of biodiversity–ecosystem services relationships. Proc. R. Soc. B Biol. sci. 282 , 20151348. (10.1098/rspb.2015.1348)
31. Kleijn D et al . 2015 Delivery of crop pollination services is an insufficient argument for wild pollinator conservation. Nat. Commun. 6 , 7414. (10.1038/ncomms8414)26079893
32. Tryjanowski P et al . 2011 Conservation of farmland birds faces different challenges in western and central-eastern Europe. Acta Ornithol. 46 , 1–12. (10.3161/000164511X589857)
33. Ricciardi V, Mehrabi Z, Wittman H, James D, Ramankutty N. 2021 Higher yields and more biodiversity on smaller farms. Nat. Sustain. 4 , 651–657. (10.1038/s41893-021-00699-2)
34. Cunha NL da et al . 2023 Soybean dependence on biotic pollination decreases with latitude. Agric. Ecosyst. Environ. 347 , 108376. (10.1016/j.agee.2023.108376)
35. Boreux V, Krishnan S, Cheppudira KG, Ghazoul J. 2013 Impact of forest fragments on bee visits and fruit set in rain-fed and irrigated coffee agro-forests. Agric. Ecosyst. Environ. 172 , 42–48. (10.1016/j.agee.2012.05.003)
36. Waarts Y et al . 2019 Living income for smallholder commodity farmers. Wageningen, The Netherlands: Wageningen Economic Research.
37. Lowder SK, Sánchez MV, Bertini R. 2021 Which farms feed the world and has farmland become more concentrated? World Dev. 142 , 105455. (10.1016/j.worlddev.2021.105455)
38. Hill R et al . 2019 Biocultural approaches to pollinator conservation. Nat. Sustain. 2 , 214–222. (10.1038/s41893-019-0244-z)
39. Tuneu-Corral C, Puig-Montserrat X, Riba-Bertolín D, Russo D, Rebelo H, Cabeza M, López-Baucells A. 2023 Pest suppression by bats and management strategies to favour it: a global review. Biol. Rev. Camb. Philos. Soc. 98 , 1564–1582. (10.1111/brv.12967)37157976
40. Ortiz AMD, Outhwaite CL, Dalin C, Newbold T. 2021 A review of the interactions between biodiversity, agriculture, climate change, and international trade: research and policy priorities. One Earth 4 , 88–101. (10.1016/j.oneear.2020.12.008)
41. Schaafsma M et al . 2023 A framework to understand the social impacts of agricultural trade. Sust. Dev. 31 , 138–150. (10.1002/sd.2379)
42. Allen-Perkins A et al . 2022 CropPol: A dynamic, open and global database on crop pollination. Ecology 103 , e3614. (10.1002/ecy.3614)34921678
43. McElreath R. 2020 Statistical rethinking: a Bayesian course with examples in R and Stan. Boca Raton, FL: CRC Press.
44. Chaplin-Kramer R et al . 2019 Global modeling of nature’s contributions to people. Science 366 , 255–258. (10.1126/science.aaw3372)31601772
45. Woittiez LS, van Wijk MT, Slingerland M, van Noordwijk M, Giller KE. 2017 Yield gaps in oil palm: a quantitative review of contributing factors. Eur. J. Agron. 83 , 57–77. (10.1016/j.eja.2016.11.002)
46. Bos MM, Veddeler D, Bogdanski AK, Klein AM, Tscharntke T, Steffan-Dewenter I, Tylianakis JM. 2007 Caveats to quantifying ecosystem services: fruit abortion blurs benefits from crop pollination. Ecol. Appl. 17 , 1841–1849. (10.1890/06-1763.1)17913145
47. Groeneveld JH, Tscharntke T, Moser G, Clough Y. 2010 Experimental evidence for stronger cacao yield limitation by pollination than by plant resources. Perspect. Plant Ecol. Evol. Syst. 12 , 183–191. (10.1016/j.ppees.2010.02.005)
48. Toledo-Hernández M, Tscharntke T, Tjoa A, Anshary A, Cyio B, Wanger TC. 2020 Hand pollination, not pesticides or fertilizers, increases cocoa yields and farmer income. Agric. Ecosyst. Environ. 304 , 107160. (10.1016/j.agee.2020.107160)
49. Toledo-Hernández M, Tscharntke T, Tjoa A, Anshary A, Cyio B, Wanger TC. 2021 Landscape and farm-level management for conservation of potential pollinators in Indonesian cocoa agroforests. Biol. Conserv. 257 , 109106. (10.1016/j.biocon.2021.109106)
50. Vanhove W, Yao RK, N’Zi JC, N’Guessan Toussaint LA, Kaminski A, Smagghe G, Van Damme P. 2020 Impact of insecticide and pollinator-enhancing substrate applications on cocoa (theobroma cacao) cherelle and pod production in Côte d’Ivoire. Agric. Ecosyst. Environ. 293 , 106855. (10.1016/j.agee.2020.106855)
51. de O. Milfont M, Rocha EEM, Lima AON, Freitas BM. 2013 Higher soybean production using honeybee and wild pollinators, a sustainable alternative to pesticides and autopollination. Environ. Chem. Lett. 11 , 335–341. (10.1007/s10311-013-0412-8)
52. Blettler DC, Fagúndez GA, Caviglia OP. 2018 Contribution of honeybees to soybean yield. Apidologie 49 , 101–111. (10.1007/s13592-017-0532-4)
53. Cunningham-Minnick MJ, Peters VE, Crist TO. 2019 Nesting habitat enhancement for wild bees within soybean fields increases crop production. Apidologie 50 , 833–844. (10.1007/s13592-019-00691-y)
54. Levenson HK, Sharp AE, Tarpy DR. 2022 Evaluating the impact of increased pollinator habitat on bee visitation and yield metrics in soybean crops. Agric. Ecosyst. Environ. 331 , 107901. (10.1016/j.agee.2022.107901)
55. Santone A, Mazzei MP, Vesprini J, Torres C, Amarilla LD, Galetto L. 2022 Pollination service and soybean yields. Acta Oecol. 116 , 103846. (10.1016/j.actao.2022.103846)
56. Schulte LA et al . 2017 Prairie strips improve biodiversity and the delivery of multiple ecosystem services from corn-soybean croplands. Proc. Natl Acad. Sci. USA 114 , 11247–11252. (10.1073/pnas.1620229114)28973922
57. Huais PY, Grilli G, Amarilla LD, Torres C, Fernández L, Galetto L. 2020 Forest fragments influence pollination and yield of soybean crops in Chaco landscapes. Basic Appl. Ecol. 48 , 61–72. (10.1016/j.baae.2020.09.003)
58. Forbes SJ, Northfield TD. 2017 Increased pollinator habitat enhances cacao fruit set and predator conservation. Ecol. Appl. 27 , 887–899. (10.1002/eap.1491)28019052
59. Merle I, Hipólito J, Requier F. 2022 Towards integrated pest and pollinator management in tropical crops. Curr. Opin. Insect Sci. 50 , 100866. (10.1016/j.cois.2021.12.006)34971783
60. Texas Parks and Wildlife Department. 2016 Texas monarch and native pollinator conservation plan. See https://tpwd.texas.gov/publications/pwdpubs/media/pwd_rp_w7000_2070.pdf.
61. Nestlé. 2021 Bees for coffee project update. See https://www.sustainability.nespresso.com/sites/site.prod.sustainability.nespresso.com/files/Ubees_Project_Caldas-May_2021_Status.pdf.
62. Alexandridis N et al . 2021 Models of natural pest control: towards predictions across agricultural landscapes. Biol. Control 163 , 104761. (10.1016/j.biocontrol.2021.104761)
63. Maas B, Clough Y, Tscharntke T. 2013 Bats and birds increase crop yield in tropical agroforestry landscapes. Ecol. Lett. 16 , 1480–1487. (10.1111/ele.12194)24131776
64. Ferreira DF, Jarrett C, Wandji AC, Atagana PJ, Rebelo H, Maas B, Powell LL. 2023 Birds and bats enhance yields in afrotropical cacao agroforests only under high tree-level shade cover. Agric. Ecosyst. Environ. 345 , 108325. (10.1016/j.agee.2022.108325)
65. Karp DS, Daily GC. 2014 Cascading effects of insectivorous birds and bats in tropical coffee plantations. Ecology 95 , 1065–1074. (10.1890/13-1012.1)24933824
66. Gras P, Tscharntke T, Maas B, Tjoa A, Hafsah A, Clough Y. 2016 How ants, birds and bats affect crop yield along shade gradients in tropical cacao agroforestry. J. Appl. Ecol. 53 , 953–963. (10.1111/1365-2664.12625)
67. Costamagna AC, Landis DA, Brewer MJ. 2008 The role of natural enemy guilds in Aphis glycines suppression. Biol. Control 45 , 368–379. (10.1016/j.biocontrol.2008.01.018)
68. Jezeer RE, Santos MJ, Verweij PA, Boot RGA, Clough Y. 2019 Benefits for multiple ecosystem services in peruvian coffee agroforestry systems without reducing yield. Ecosyst. Serv. 40 , 101033. (10.1016/j.ecoser.2019.101033)
69. Alves DMCC, Diniz-Filho JAF, da Silva e Souza K, Gouveia SF, Villalobos F. 2018 Geographic variation in the relationship between large-scale environmental determinants and bat species richness. Basic Appl. Ecol. 27 , 1–8. (10.1016/j.baae.2017.12.002)
70. Food and Agriculture Organization of the United Nations. 2020 FAOSTAT. See http://www.fao.org/faostat/en/#home.
71. Chowdhury A, Samrat A, Devy MS. 2021 Can tea support biodiversity with a few 'nudges' in management: evidence from tea growing landscapes around the world. Glob. Ecol. Conserv. 31 , e01801. (10.1016/j.gecco.2021.e01801)
72. Araújo F de, Lima E de, Costa E, Daud RD. 2022 Influence of natural vegetation conservation on the distribution of mites in rubber tree crops. Syst. Appl. Acar. 27 , 1648–1662. (10.11158/saa.27.8.13)
73. Ramos D, Hartke TR, Buchori D, Dupérré N, Hidayat P, Lia M, Harms D, Scheu S, Drescher J. 2022 Rainforest conversion to rubber and oil palm reduces abundance, biomass and diversity of canopy spiders. PeerJ 10 , e13898. (10.7717/peerj.13898)35990898
74. Liu S, Li Z, Sui Y, Schaefer DA, Alele PO, Chen J, Yang X. 2015 Spider foraging strategies dominate pest suppression in organic tea plantations. Biocontrol 60 , 839–847. (10.1007/s10526-015-9691-2)
75. Rubio GD, Stolar CE, Ohashi DV, Baigorria JE. 2019 Jumping spiders (Araneae: Salticidae) in agroecosystems: a case study to know how friendly some crops can be for native fauna. Stud. on Neotrop. Fauna Environ. 54 , 133–148. (10.1080/01650521.2019.1629174)
76. Gray CL, Lewis OT. 2014 Do riparian forest fragments provide ecosystem services or disservices in surrounding oil palm plantations? Basic Appl. Ecol. 15 , 693–700. (10.1016/j.baae.2014.09.009)
77. Shine R, Madsen T. 1997 Prey abundance and predator reproduction: rats and phytons on a tropical Australian floodplain. Ecology 78 , 1078. (10.2307/2265859)
78. Daghela Bisseleua HB, Fotio D, Missoup AD, Vidal S. 2013 Shade tree diversity, cocoa pest damage, yield compensating inputs and farmers’ net returns in West Africa. PLoS One 8 , e56115. (10.1371/journal.pone.0056115)23520451
79. Rezende MQ, Venzon M, dos Santos PS, Cardoso IM, Janssen A. 2021 Extrafloral nectary-bearing leguminous trees enhance pest control and increase fruit weight in associated coffee plants. Agric. Ecosyst. Environ. 319 , 107538. (10.1016/j.agee.2021.107538)
80. Saha R, Ghosh PK, Mishra VK, Majumdar B, Tomar JMS. 2010 Can agroforestry be a resource conservation tool to maintain soil health in the fragile ecosystem of north-east India? Outlook Agric. 39 , 191–196. (10.5367/oa.2010.0004)
81. Sauvadet M, Saj S, Freschet GT, Essobo J, Enock S, Becquer T, Tixier P, Harmand J. 2020 Cocoa agroforest multifunctionality and soil fertility explained by shade tree litter traits. J. Appl. Ecol. 57 , 476–487. (10.1111/1365-2664.13560)
82. Schmidt JE, DuVal A, Isaac ME, Hohmann P. 2022 At the roots of chocolate: understanding and optimizing the cacao root-associated microbiome for ecosystem services. a review. Agron. Sustain. Dev. 42 , 14. (10.1007/s13593-021-00748-2)
83. Blaser WJ, Oppong J, Yeboah E, Six J. 2017 Shade trees have limited benefits for soil fertility in cocoa agroforests. Agric. Ecosyst. Environ. 243 , 83–91. (10.1016/j.agee.2017.04.007)
84. Wartenberg AC, Blaser WJ, Gattinger A, Roshetko JM, Van Noordwijk M, Six J. 2017 Does shade tree diversity increase soil fertility in cocoa plantations? Agric. Ecosyst. Environ. 248 , 190–199. (10.1016/j.agee.2017.07.033)
85. Niether W, Jacobi J, Blaser WJ, Andres C, Armengot L. 2020 Cocoa agroforestry systems versus monocultures: A multi-dimensional meta-analysis. Environ. Res. Lett. 15 , 104085. (10.1088/1748-9326/abb053)
86. Tian Y, Cao F, Wang G. 2013 Soil microbiological properties and enzyme activity in ginkgo–tea agroforestry compared with monoculture. Agroforest. Syst. 87 , 1201–1210. (10.1007/s10457-013-9630-0)
87. Mortimer PE, Gui H, Xu J, Zhang C, Barrios E, Hyde KD. 2015 Alder trees enhance crop productivity and soil microbial biomass in tea plantations. Appl. Soil Ecol. 96 , 25–32. (10.1016/j.apsoil.2015.05.012)
88. Cerda R, Avelino J, Harvey CA, Gary C, Tixier P, Allinne C. 2020 Coffee agroforestry systems capable of reducing disease-induced yield and economic losses while providing multiple ecosystem services. Crop Prot. 134 , 105149. (10.1016/j.cropro.2020.105149)
89. Salceda M, Udawatta RP, Nelson KA, Mendis SS, Bardhan S. 2022 Spatial and temporal variability of soil organic carbon on a corn–soybean watershed with 23 years of agroforestry. Agron. J. 114 , 440–451. (10.1002/agj2.20948)
90. Pang Z et al . 2021 Continuous sugarcane planting negatively impacts soil microbial community structure, soil fertility, and sugarcane agronomic parameters. Microorganisms 9 , 2008. (10.3390/microorganisms9102008)34683329
91. Zhang F, Gao C, Wang J, Lu Y, Shen Z, Liu T, Chen D, Ran W, Shen Q. 2017 Coupling sugarcane yield to soil nematodes: implications from different fertilization regimes and growth stages. Agric. Ecosyst. Environ. 247 , 157–165. (10.1016/j.agee.2017.06.020)
92. Jiang Y et al . 2019 Restoration of long-term monoculture degraded tea orchard by green and goat manures applications system. Sustainability 11 , 1011. (10.3390/su11041011)
93. Zhou Y, Bastida F, Zhou B, Sun Y, Gu T, Li S, Li Y. 2020 Soil fertility and crop production are fostered by micro-nano bubble irrigation with associated changes in soil bacterial community. Soil Biol. Biochem. 141 , 107663. (10.1016/j.soilbio.2019.107663)
94. Rao D et al . 2021 Changes in soil microbial activity, bacterial community composition and function in a long-term continuous soybean cropping system after corn insertion and fertilization. Front. Microbiol. 12 , 638326. (10.3389/fmicb.2021.638326)33897643
95. IBPES. 2019 Global assessment report on biodiversity and ecosystem services of the Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services. IPBES Secretariat. (10.5281/zenodo.3831673)
96. Asiimwe P, Naranjo SE, Ellsworth PC. 2014 Effects of irrigation levels on interactions among lygus hesperus (Hemiptera: Miridae), insecticides, and predators in cotton. Environ. Entomol. 43 , 263–273. (10.1603/EN12328)24533912
97. Goldenberg MG, Burian A, Seppelt R, Santibañez Ossa FA, Bagnato CE, Satorre EH, Martini GD, Garibaldi LA. 2022 Effects of natural habitat composition and configuration, environment and agricultural input on soybean and maize yields in Argentina. Agric. Ecosyst. Environ. 339 , 108133. (10.1016/j.agee.2022.108133)
98. De Beenhouwer M, Aerts R, Honnay O. 2013 A global meta-analysis of the biodiversity and ecosystem service benefits of coffee and cacao agroforestry. Agric. Ecosyst. Environ. 175 , 1–7. (10.1016/j.agee.2013.05.003)
99. Chen LL, Yuan P, You MS, Pozsgai G, Ma X, Zhu H, Yang G. 2019 Cover crops enhance natural enemies while help suppressing pests in a tea plantation. Ann. Entomol. Soc. Am. 112 , 348–355. (10.1093/aesa/say050)
100. Newbold T et al . 2018 Widespread winners and narrow-ranged losers: land use homogenizes biodiversity in local assemblages worldwide. PLoS Biol. 16 , e2006841. (10.1371/journal.pbio.2006841)30513079
101. de Camargo NF, dos Reis GG, Mendonça AF, Laumann RA, Nardoto GB, de Camargo AJA, Vieira EM. 2022 Native marsupial acts as an in situ biological control agent of the main soybean pest (Euschistus heros) in the neotropics. Eur. J. Wildl. Res. 68 , 62. (10.1007/s10344-022-01609-3)
102. Zainal Abidin CMR, Mohd Noor H, Hamid NH, Ravindran S, Puan CL, Kasim A, Salim H. 2022 Comparison of effectiveness of introduced barn owls, tyto javanica javanica, and rodenticide treatments on rat control in oil palm plantations. J. Pest Sci. 95 , 1009–1022. (10.1007/s10340-021-01423-x)
103. Wanger TC, Saro A, Iskandar DT, Brook BW, Sodhi NS, Clough Y, Tscharntke T. 2009 Conservation value of cacao agroforestry for amphibians and reptiles in south‐east Asia: combining correlative models with follow‐up field experiments. J. Appl. Ecol. 46 , 823–832. (10.1111/j.1365-2664.2009.01663.x)
104. Monagan IV, Morris JR, Davis Rabosky AR, Perfecto I, Vandermeer J. 2017 Anolis lizards as biocontrol agents in mainland and Island agroecosystems. Ecol. Evol. 7 , 2193–2203. (10.1002/ece3.2806)28405283
105. Baudron F, Guerrini L, Chimimba E, Chimusimbe E, Giller KE. 2022 Commodity crops in biodiversity-rich production landscapes: friends or foes? the example of cotton in the mid-Zambezi valley, Zimbabwe. Biol. Conserv. 267 , 109496. (10.1016/j.biocon.2022.109496)
106. Braczkowski AR, O’Bryan CJ, Lessmann C, Rondinini C, Crysell AP, Gilbert S, Stringer M, Gibson L, Biggs D. 2023 The unequal burden of human-wildlife conflict. Commun. Biol. 6 , 182. (10.1038/s42003-023-04493-y)36823291
107. Mackenzie CA, Ahabyona P. 2012 Elephants in the garden: financial and social costs of crop raiding. Ecol. Econ. 75 , 72–82. (10.1016/j.ecolecon.2011.12.018)
108. Waha K, Dietrich JP, Portmann FT, Siebert S, Thornton PK, Bondeau A, Herrero M. 2020 Multiple cropping systems of the world and the potential for increasing cropping intensity. Glob. Environ. Change 64 , 102131. (10.1016/j.gloenvcha.2020.102131)33343102
109. The European Parliament and The Council of the European Union. Document 52022PC0304: Proposal for a Regulation of the European Parliament and of the Council on Nature Restoration. See https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX%3A52022PC0304.
110. The European Parliament and The Council of the European Union. 2023 Regulation (EU) 2023/1115 of the European Parliament and of the Council of 31 May 2023 on the making available on the Union market and the export from the Union of certain commodities and products associated with deforestation and forest degradation and repealing Regulation (EU) No 995/2010. See https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX%3A32023R1115.
111. Bateman I, Balmford A. 2023 Current conservation policies risk accelerating biodiversity loss. Nature 618 , 671–674. (10.1038/d41586-023-01979-x)37344644
112. Maney C, Sassen M, Giller K. 2024 Data from: Are agricultural commodity production systems at risk from local biodiversity loss. Figshare. (10.6084/m9.figshare.c.7430648)
