For PGR conservationists to meet the breeders' needs requires that PGR conservation maximizes the broadest range of genetic diversity in the minimum number of accessions available to breeders. Tanksley and McCouch (1997) established that the process of crop domestication necessarily involves a significant loss of genetic diversity, because individual populations of a species are domesticated, not the whole species, farmers require uniformity during agricultural cultivation to maximize production and some natural wild traits are unsuitable for inclusion in any crop (e.g., brittle rachis, exploding fruits). Therefore, genetic diversity is lost in the transition between wild species and crop landraces, and further between landraces and cultivars. It is therefore unsurprising, as Tanksley and McCouch (1997) note, that 95% of tomato genetic diversity is found in wild Lycopersicon/Solanum spp. not the crop itself. This can be generalized to demonstrate the relative amounts of genetic diversity in different elements of the crop gene pool, as illustrated for the barley gene pool (Figure 1), where the filled circles indicate the crop, the hatched circles indicate the wild species, and the relative area of the circle indicates the quantity of genetic diversity in that constituent of the crop gene pool. The highest quantity of genetic diversity is found in the primary and secondary CWR and then crop landraces.

If the breeders' focus is moving toward access to greater breadth of diversity, then, as Tanksley and McCouch (1997) conclude, the obvious focus would be CWR taxa. They are wild plant species relatively closely related to crops, including crop's wild ancestors that retain indirect use value as gene donors for crop improvement and high level of genetic diversity. Although initially the CWR conservation and use focus has been on the most closely related CWR to the crops and the highest value crops, in the longer term it is likely, with the easier application of gene editing (Hartung and Schiemann, 2014; Wang et al., 2022) and by applying speed breeding (Watson et al., 2018), CWR usage will be expanded providing the full breadth of CWR diversity is conserved and available to breeders.

Over the last 60 years, as the science of PGR conservation has developed, the gene banks have played a central role in that conservation: (a) to sustainably conserve the broadest range of genetic diversity found in the target species (as many alleles, or as many gene combinations as possible) held as population samples or accessions, (b) to characterize and evaluate the diversity conserved to aid selection for utilization, and (c) to make the conserved accessions available to the user community. Therefore, the role of the gene banks might be defined as being “to maximize the conservation, characterization, documentation and promoting the use of PGR diversity for the benefit of humankind” (Maxted et al., 2020), as summarized in Figure 2. Gene bank-based conservation of population samples as seed is the primary (≈85%) method for conserving diversity, because it is suitable for most species, is relatively inexpensive, facilitates characterization, evaluation and user access, and requires little routine maintenance. However, many recalcitrant species, whose seed cannot be conserved by desiccation and freezing, are very poorly conserved, except for a few major crop species (e.g., potato, banana, coffee, chocolate).

As stated, the traditional role of gene banks in CWR conservation is to sample populations, maintain sample viability, characterization and evaluation, and make the accession available for use. An essential step of which is to multiply/regenerate the sample accessions when quantities or viability decline. However, unlike crop materials where the sampling/multiplication/regeneration protocols are well established, these are highly specific and vary extensively for CWR (Terry et al., 2003; Way, 2003; FAO, 2016), although the Seed Information Database (Bone et al., 2003) does provide help in suggesting protocol for many species. Nevertheless, gene bank maintenance of CWR diversity is more complicated and requires greater knowledge/skills than crop material and is, therefore, more likely to fail, which reinforces the need for complementary in situ conservation action.

Despite the 7.4 million ex situ accessions held in ≈1,750 gene banks globally (FAO, 2010), several authors draw attention to the lack of sufficient access to diversity and note it is restricting plant breeding outcomes (Volbrecht and Sigmon, 2005; Dwivedi et al., 2007; Feuillet et al., 2008; McCouch et al., 2013). While at the same time, the CWR taxa, that contain the bulk of the diversity required are suffering erosion and extinction, 16–35% of CWR are assessed as threatened using the IUCN Red List Categories and Criteria (Bilz et al., 2011; Kell et al., 2012; Goettsch et al., 2021). A global analysis of priority CWR holdings also found that about a third were unconserved, a third were poorly conserved (<10 accessions per CWR taxon) and 95% required additional collections (Castañeda-Álvarez et al., 2016). It is well established that the conservation ideal is to apply ex situ and in situ techniques in a complementary manner (FAO, 1998, 2010). In practice, the application of in situ conservation has almost completely been ignored and as yet only a handful of active genetic reserves, either in protected areas (PA) or Other Effective Area-based Conservation Measures, have been established.

It is this lack of in situ actions which offers such an opportunity now for action. It is estimated that systematically applying in situ conservation measures would at least double the diversity available to users which, in turn, would generate substantial economic gain and further underpin the utilization of genetic resources in contributing to food security.

These recent research projects have, for example, (a) identified global areas of CWR richness (Vincent et al., 2019) (Figure 3A), and (b) hotspots for further CWR collecting for ex situ conservation (Castañeda-Álvarez et al., 2016) (Figure 3B); (c) revised the Vavilov centers of diversity (Vavilov 1926, Maxted and Vincent, 2021) (Figure 3C); and (d) identified the top 170 sites for global in situ CWR conservation (Vincent et al., 2019) (Figure 3D). The Farmer's Pride and the SADC Crop Wild Relatives projects have also particularly developed useful tools to aid conservation planning, including:

Additionally, the Secretariat of the International Treaty on Plant Genetic Resources for Food and Agriculture (ITPGRFA) of the FAO together with worldwide experts, technical staff and national focal points of the Treaty elaborated the Descriptors for Crop Wild Relatives Conserved Under In Situ Conditions (Alercia et al., 2022).

The evidence base has developed significantly, and we currently have a much better idea of where CWR diversity is found, what conservation actions are needed and how they might best be implemented. Recent research has shown the disjunction between IUCN Key Biodiversity Areas and CWR distribution (Saunders et al., in prep.) and highlighted the need for specific CWR conservation initiatives as general biodiversity conservation measures will not effectively address CWR conservation requirements. Therefore, the imperative now is to move from research to the implementation of in situ conservation actions to ensure CWR diversity is available for crop improvement.

It is also increasingly recognized that the application of ex situ conservation techniques alone will not meet the breeders' requirements for diversity to mitigate the impact of climate change on agriculture. Urgent focus is therefore required on (a) actively conserving threatened CWR in globally important hotspots, (b) filling conservation gaps in existing ex situ holdings, (c) filling gaps in conserved germplasm that is unavailable for users, (d) linking with breeders' needs for more diversity to adapt cultigen improvement programmes to climate change, (e) re-focusing PGR conservation activities in a way that regional and national level conservation activities are fully integrated and complementary—possibly building on a European network composed of networks of European national networks, and on (f) meeting policy and legislative obligations (e.g., Sustainable Development Goals,¹⁴ Second Global Plan of Action for Plant Genetic Resources for Food and Agriculture,¹⁵ Convention on Biological Diversity,¹⁶ European Green Deal,¹⁷ including the Biodiversity and Farm to Fork Strategies¹⁸ in Europe).

Most literature on CWR in situ conservation assumes CWR will be most effectively conserved in pre-existing conservation PA sites. This implies there is no fundamental difference between sites managed for CWR and for other wild plant species, though in fact different CWR may require slightly different management regimes just as any wild plant species may vary in their specific conservation management requirements (Maxted et al., 2008a). Sites where CWR populations are conserved in situ are named “genetic reserves” and they are defined as “sites established to manage and monitor the genetic diversity of natural wild populations within defined areas designated for active, long-term conservation” (Maxted et al., 1997). They may be established on private lands, roadsides, in indigenous reserves and community conserved areas, as well as officially recognized protected areas.

Initially the in situ CWR conservation communities have pragmatically focused on establishing genetic reserves within PA, the reasons being: (a) sites already have a generalized long-term conservation ethos, (b) sites are less prone to hasty management changes associated with private lands or roadsides, (c) it is relatively easy to amend the existing site management plan to facilitate CWR genetic conservation, (d) the prohibitive cost of acquiring non-conservation land is avoided, and (e) there is evidence from throughout the world that CWR populations are found in existing PA in significant numbers (Maxted et al., 2008b; Vincent et al., 2019; Magos Brehm et al., 2022). Therefore, often the simplest way forward in economic and political terms is for countries to locate genetic reserves in existing PA, ideally IUCN recognized national parks or heritage sites. However, to ensure the widest range of CWR diversity is conserved and available for use, conserving CWR populations in PAs alone will not suffice. CWR are more often found in pre-climax, anthropogenic environments outside of formal PA networks, primarily field margins, orchards, cultivated terraces, roadsides and weedy fields (Jain, 1975; Jarvis et al., 2015; Fagandini Ruiz et al., 2021).

The requirement for extra-PA site based in situ conservation for CWR taxa was first highlighted by Al-Atawneh et al. (2007), when implementing area-based conservation in the Fertile Crescent of West Asia. In this region, the global hottest spot of CWR diversity (Vincent et al., 2019), large CWR populations are often found in and around cultivated areas, in the weedy crop itself or at its margins in field edges, orchards, terraces, habitat patches or roadsides (Al-Atawneh et al., 2007; Maxted et al., 2008b; Iriondo et al., 2021). This concept of active conservation outside of established PA was also developed by IUCN, who within the IUCN-WCPA established a Task Force on Other Effective Area-based Conservation Measures (OECMs) (IUCN-WCPA Task Force on OECMs, 2019) to complement more formal PA-based conservation. If an OECM site is established as opposed to a PA, there will be no need to compromise site management objectives to facilitate overall PA vs. CWR conservation and the management can focus entirely on the CWR populations. Additionally, it is relatively easy to make the link between CWR and sustainable agriculture for rural communities as they generally see the obvious phenotypic likeness between CWR and the crops they cultivate. Therefore, they perceive the direct benefit to themselves of OECM based CWR conservation and are more likely to support conservation initiatives. As such the involvement of local communities in active in situ conservation of CWR in OECM sites may be easier than in existing PA since the latter are more rarely located near dense human populations' centers. In situ conservation of CWR in OECM sites has, therefore, significant potential to complement and possibly in time exceed PA-based CWR conservation.

Given that any in situ CWR conservation effort is starting from scratch, because it did not evolve ad hoc as happened with ex situ conservation over the last 60 years, we can structure the network in the most appropriate manner. This includes evaluating whether it is preferable to allow each in situ CWR conservation site to be independent or to be linked in some form of network. Whatever the geographic scale, it appears to be preferable to establish a network of collectively managed in situ CWR sites because: (a) it helps ensures systematic coordination and reporting of the sites and their CWR populations (e.g., to the FAO Global Plan of Action); (b) fosters stronger partnerships and mutual support between sites, populations and staff; (c) facilitates integration and complementarity of global, regional and national actions; (d) links local communities of practice with common goals; (e) helps safeguards evolving in situ CWR populations for perpetuity; (f) helps further promote integrated, long-term complementary in situ–ex situ conservation; (g) promotes access to PGR held in protected areas and farmers/farming communities via their linked Genetic Resources Centers (GRC); and (h) ensures land managers, protected areas and farmers/farming communities receive adequate help with implementing the Access and Benefit Sharing (ABS)¹⁹ mechanisms.

The benefits to individual land managers, protected areas and farmers/farming communities of network membership are expected to be numerous, including:

For germplasm users that belong to the network, benefits include:

Just as there are obvious advantages of organizing CWR in situ conservation sites into networks that are managed collectively, so it is also obvious that sites and networks should be geographically and politically integrated. This means that sites are integrated like a series of Russian dolls, with some sites likely containing the highest concentration of CWR populations of multiple CWR taxa being incorporated into the national in situ network, and some of these from multiple countries forming the sub-regional or continental regional in situ network, and some of these from multiple countries and continents forming the global in situ conservation network.

In essence, effective and systematic global in situ conservation of CWR diversity will be achieved via three interrelated geographic or more precisely geopolitical levels of conservation strategy planning: (i) national (Figure 4 in light green) (ii) regional (Figure 4 in blue) and (iii) integrated global (Figure 4 in orange). National governments will be at the core of the establishment of the Network and their support will be essential to its success. Therefore, whether a particular site is provided with national only, national and regional, or national, regional and global designation as part of a CWR network, its inclusion is justified by containing significant CWR populations that have respective national, regional and/or global value, and the national authority alone has precedence over this decision, since it should be the national agency that nominates the sites/populations to join the national, regional or global networks. This is in line with what the CBD stressed that “the conservation of biological diversity is a common concern of humankind” but recognizes “that States have sovereign rights over their own biological resources” (UNEP, 1992). Figure 4 also highlights the key point that CWR conservation must be linked to utilization and germplasm utilization can only be granted via the appropriate nation agency reiterating its national sovereign rights over biological (including CWR) resources.

Priority sites containing CWR diversity of global importance for inclusion in the global network can be identified by national, regional and global authorities and these sites could be recommended to individual countries as sites within their borders where genetic reserves might be established (Figure 4 in dark green). The actual establishment of these regional or globally identified genetic reserves would be at the discretion of individual nations, although support to encourage their establishment could be forthcoming from the international community. The final integrated global CWR in situ network would contain both genetic reserves identified by individual countries (bottom-up) and those initially identified by regional or global research (top-down), but the latter's nomination for inclusion in the integrated global network would be by individual countries and the genetic reserve conservation planning, practical management and monitoring would necessarily be implemented at national level, though potentially with international support. The integrated global CWR in situ network is therefore integrated because it contains both the bottom-up and top-down identified priority sites.

The integrated global CWR in situ network would be driven by international, regional and national policy on conservation and utilization of plant genetic resources for food and agriculture (PGRFA) (Figure 4 in red) and it is implemented at national level (Figure 4 in dark green). Since the purpose of the integrated strategy is to preserve CWR genetic resources for use in crop improvement and to maintain cultivar development options, a fundamental element is making conserved CWR germplasm available to the user community (Figure 4 in purple) and to achieve this, the interface between in situ, ex situ and use of conserved diversity needs to be strengthened. As indicated by the cyclical flow of the related strategies in Figure 4, planning and implementing global in situ conservation of CWR will be an iterative process requiring periodic review and updating as CWR conservation and utilization policy, science and practice develops. Promoting awareness of the value of CWR to food and economic security as well as raising additional funding, will be critical to support this process and ensure long-term in situ CWR conservation.

Finally, the establishment of the network at each geopolitical level requires some form of integrated governance structure to manage and sustain the network. Governance may be defined as “the activity of governing a country or controlling a company or an organization; the way in which a country is governed or a company or institution is controlled…” (Oxford Learner's Dictionaries, 2022). One element of governance are the rules or minimum inclusion criteria for a CWR site/population to join the network. Such rules are likely to vary depending on the geopolitical level of the network, but the sort of criteria that might apply were originally proposed by Iriondo et al. (2012), amended by Maxted (2016), and further amended here.

Instinctively, it would appear simplest and most beneficial for the protected area manager or farmer to each grant access to the genetic resources they conserve and for users to contact them directly. However, in practice, this is unlikely to ever work smoothly, because a priori we do not know which of the literally millions of in situ or on-farm conserved populations the user is likely to request samples from. Therefore, ensuring all protected area managers or farmers that manage sites containing CWR populations are sufficiently aware of their rights and obligations under the Nagoya Protocol and the ITPGRFA to supply germplasm and ensure their own ABS rights are secured is impractical. A more practical option is that the protected area manager or farmer supply population backup samples to a local/national named gene bank or rather GRC and the center supplies subsamples of these to users, securing the suppliers rights and ensuring the users' needs are met (see Figure 6). Figure 6 provides four alternative options for how the user might be linked to the in situ conserved resource via the GRC.

Considering the four options, Option 1 would place significant additional workload and resource expenditure on the GRC as they are treated the same way as ex situ collections. Option 2 does not facilitate access to the in situ conserved resources, so is undesirable. Option 3 is demand and supply based and because of seasonality of population seed supply, could significantly delay provision of the conserved resources to the user but would be the cheapest for GRC to implement. Option 4 is most preferable as it would be relatively cheap to implement and would mean in situ population samples would be made accessible alongside the ex situ conserved material. The regular resampling of the host in situ populations would mean the sample would better reflect their current genetic diversity content which is continually evolving. Presence of the in situ sample in the GRC would mean it could be characterized and evaluated alongside the ex situ samples. The provision of accessibility to in situ population samples via the gene bank/GRC would acknowledge the fact that they have the appropriate expertise in user seed supply and might see this as a natural extension of their existing role.