Journal for Nature Conservation 61 (2021) 126004 Contents lists available at ScienceDirect Journal for Nature Conservation journal homepage: www.elsevier.com/locate/jnc Genetic conservation strategies of endemic plants from edaphic habitat islands: The case of Jacobaea auricula (Asteraceae) Javier Bobo-Pinilla a, b, *, 1, Esteban Salmerón-Sánchez c, **, 1, Juan Francisco Mota c, Julio Peñas d a Department of Botany, University of Salamanca, 37007, Salamanca, Spain Biobanco de ADN Vegetal, Edificio Multiusos I+D+i, 37007, Salamanca, Spain c Department of Biology and Geology, CEI⋅MAR and CECOUAL, University of Almería, Almería, Spain d Plant Conservation Unit, Department of Botany, University of Granada, 18071, Granada, Spain b A R T I C L E I N F O A B S T R A C T Keywords: In situ conservation Ex situ conservation Gypsohalophytic flora Relevant Genetic Units for Conservation Threatened species Conservation genetics is a well-established and essential scientific field in the toolkit of conservation planning, management, and decision-making. Within its framework, phylogeography allows the definition of conservation strategies, especially in threatened endemic plants. Gypsum and salt-rich outcrops constitute a model example of an edaphic island-like habitat and contain rare and endemic species, many of them threatened. This is the case of Jacobaea auricula, an Iberian gypsohalophytic species with biological, ecological, and conservation interest. Genetic-based criteria were used to preserve the highest possible percentage of the species’ genetic pool as well as to dispose of a set of genotypes for translocation and/or reinforcement planning of degraded populations. Relevant Genetics Units for Conservation (RGUCs) were selected as in situ conservation planning. As a complementary ex situ measure, the optimal contribution for the populations to maximize the genetic pool within each genetic cluster was calculated. To preserve the maximum genetic diversity and the highest percentage of rare AFLP bands possible, eight RGUCs were selected; the ex situ conservation design included twenty-one populations, gathering all haplotypes and ribotypes. Our genetic conservation proposal of J. auricula would improve the implementation of future genetic conservation measures, as a species model of endemic plants from edaphic habitat islands. 1. Introduction It is necessary to develop and apply strategies and methods for the conservation of biodiversity due to historical losses of biodiversity (Margules & Pressey, 2000; Pärtel et al., 2005). Conservation biology aims to preserve current genetic diversity and the diversification processes that are taking place at species-level (Forest et al., 2007). Genetic diversity must be preserved as it holds the survival ability of the species (Hoban et al., 2020; Pérez-Collazos et al., 2008); to this effect, population genetics data are essential for both conceptual and applied biodiversity conservation programs. Moreover, conservation genetics is a well-established scientific field that will be essential (among other methods) in the toolkit of conservation planning, management, and decision-making (Frankham et al., 2004; Holderegger et al., 2019). Regarding diversification processes, phylogeny and phylogeography can enlighten how interactions between evolutionary and ecological processes influence diversity at multiple scales (Webb et al., 2002). For this reason, these disciplines could improve the proactive conservation planning (Avise, 2009; Médail & Baumel, 2018). Genetic patterns, species potential habitat and intraspecific phylogenetic relationships are essential to appropriately address species conservation (Commander et al., 2018). Therefore, the use of genetic diversity structure is necessary for defining conservation strategies, especially for threatened endemic plants within biodiversity hotspots. Unfortunately, most of the phylogeographic studies have not placed much emphasis on establishing management and conservation proposals neither in situ units nor ex situ, i.e. maintaining genetic diversity in ex situ collections. According to Médail & Baumel (2018), who performed a review of the studies dealing with the genetic diversity structure of narrow endemic plants in the Mediterranean Basin hotspot, only 27 % of * Corresponding author at: Department of Botany, University of Salamanca, 37007, Salamanca, Spain. ** Corresponding author at: Department of Biology and Geology. University of Almería, 04120. Almería, Spain. E-mail addresses: javicastronuevo@usal.es (J. Bobo-Pinilla), esanchez@ual.es (E. Salmerón-Sánchez). 1 both authors have contributed equally to this manuscript. https://doi.org/10.1016/j.jnc.2021.126004 Received 25 November 2020; Received in revised form 11 April 2021; Accepted 13 April 2021 Available online 24 April 2021 1617-1381/© 2021 The Author(s). Published by Elsevier GmbH. This is (http://creativecommons.org/licenses/by/4.0/). an open access article under the CC BY license J. Bobo-Pinilla et al. Journal for Nature Conservation 61 (2021) 126004 these studies used the information generated to establish priorities for the conservation of the species, and around 18 % inferred conservation units. Both in situ and ex situ conservation genetic strategies are essential for protecting rare and threatened plant species (Volis & Blecher, 2010). The success of conservation depends on whether the species are able to survive in the habitat and how they could be regenerated if it were necessary in the future. Conservation genetics leads us to use genetic patterns in the conservation decision-making process (DeSalle & Amato, 2004), where an important objective is to search how many and which populations deserve conservation priority. In order to preserve the highest amount of genetic diversity possible in the least number of populations and/or areas, several estimators have been proposed over time: Evolutionary Significant Units (Ryder, 1986), Management Units (Moritz, 1994), Operational Conservation Units (Doadrio et al., 1996), Fundamental Geographic and Evolutionary Units (Riddle & Hafner, 1999), Functional Conservation Units (Maes et al., 2004), and Relevant Genetic Units for Conservation (RGUCs; Pérez-Collazos et al., 2008). The latter approach combines two methods that use genetic data (considering both common and rare alleles) to estimate the minimum number of conservation units that should be targeted for an adequate representation of the total genetic variability of a species. This method is based on the idea that rare alleles are essential in conservation because they represent unique evolutionary products that could provide the species with the ability to adapt to environmental changes (Bengtsson et al., 1995; Lopez et al., 2009; Pérez-Collazos et al., 2008; Shaw & Etterson, 2012). Moreover, this method allows the selection of those populations that hold the highest values of diversity and/or rarity within the geographical areas. The selection of RGUCs has been used to propose sampling strategies for species such as Boleum asperum Desv. (Pérez-Collazos et al., 2008), Borderea pyrenaica Miégev. (Segarra-Moragues & Catalán, 2010) and Astragalus edulis Bunge (Peñas et al., 2016). The conservation proposals are often focused on passive protection which results inadequate for reducing accelerated losses of natural species and habitats (Fenu et al., 2019; Mace & Purvis, 2008). In situ conservation of all of the populations of threatened species is often not feasible at large scales due to the costs, but it is feasible to apply ex situ conservation to the most threatened species which require greater effort (Fay & Krauss, 2003). When creating a germplasm bank, the gathering of all the genetic diversity of the species is essential as this will allow the proposal of viable translocation measures in the future (Caujapé-Castells & Pedrola-Monfort, 2004; Pearse & Crandall, 2004) and represents the basis of the ex situ conservation strategy. Ex situ collections may contribute effectively to plant species conservation if their use is supported by a thorough understanding of the limiting factors, such as scarcity of source material, low viability, low genetic variation, and socioeconomic factors, among others (Abeli et al., 2020; Hyvärinen, 2020). The seed banks should contain an optimal number of haplotypes and allele copies (and the type of allele targeted) and thus, must contain populations for the maximization of genetic diversity. The habitats with gypsum outcrops frequently associated with saltrich deposits are interesting biodiversity hotspots (Gutiérrez et al., 2008). These soils present particular physical and chemical characteristics which are inhabited by numerous plant species that have significant adaptations to survive in them, such as the gypsohalophytic flora (Denaeyer–De Smet, 1970). Gypsum and salt-rich outcrops occupy disjunct areas in territories with arid or semiarid climate conditions. These habitats comprise a model example of an edaphic island-like habitat and are interesting for the study of plant distributions, gene flow, genetic diversity, and diversification (Escudero et al., 2015; Moore et al., 2014; Mota et al., 2011). The gypsohalophytic flora is rich in rare and endemic species, many of them threatened (Pérez–García et al., 2011), and characterize the Iberian gypsum steppes habitat, which is included within the UE “Priority habitat 1520” (Gypsophiletalia order) (Evans, 2006; Mota et al., 2011). To preserve this priority habitat in the Iberian Peninsula, the inclusion of fifty-one localities has been proposed (Mota et al., 2011). Other proposals have been presented to preserve specifically gypsohalophytic species, such as the establishment of micro-reserves (Eugenio et al., 2013; Salazar et al., 2011), or the inclusion within nature protection areas in Natura 2000 network (Salazar et al., 2011). Unfortunately, these in situ protection proposals were applied at local level, and do not take into account the levels of genetic diversity of the population nor how such gene diversity is distributed throughout the whole area of distribution of the species. Jacobaea auricula (Bourg. Ex Coss) Pelter. (Asteraceae) is a characteristic species of gypsum and salt-rich habitats (Salmerón-Sánchez et al., 2017), with biological, ecological, and conservation interest. This is an herbaceous perennial species from the eastern part of the Iberian Peninsula that has a discontinuous distribution in scattered and small populations. This disjunct distribution is edaphically restricted to gypsiferous or marl soils, salt marshes and saltland pastures bordering lagoons or seasonal water courses (Ascaso & Pedrol, 1991; Pérez–García et al., 2011; Salazar & Peñas, 2011; Salazar et al., 2011). Several intraspecific morphological discontinuities associated with geographical areas have been described, which has traditionally led to the recognition of three different subspecies (Ascaso & Pedrol, 1991; De La Torre et al., 1997). All three subspecies are included in different Spanish regional lists of threatened species (Anthos, 2020; Mota et al., 2011), as well as in the Spanish Red List of Vascular Flora (VV.AA., 2000) where they are considered to be in the threat category VU (Vulnerable). Salmerón-Sánchez et al. (2017) studied the phylogeographical and evolutionary history of the species, and whether the classical taxonomic differentiation in subspecies is genetically supported. In this research, the authors concluded that it would be premature to recognize infraspecific taxa in J. auricula and that these molecular taxonomic results should be considered for conservation purposes of the species. To assist the preservation and management plans of flora associated with gypsum and salt-rich outcrops, J. auricula is studied as a focal species for developing genetic conservation strategies. To achieve this, the specific objectives are: a) to select RGUCs, on the basis of the possession of both common and rare alleles developing an in situ conservation planning, and b) to select populations from which to collect seeds as an ex situ proposal, aiming to store the greatest genetic variability that would contribute to the future creation of new populations or to reinforce existing ones. 2. Materials and methods 2.1. Selection of relevant Genetic Units for Conservation (RGUCs) Amplified fragment length polymorphisms (AFLP) dataset of Jacobaea auricula obtained by Salmerón-Sánchez et al. (2017) were used as a source of genetic data to select the Relevant Genetic Units for Conservation (RGUCs). Taking into account the availability of data for many of the endangered species and the methodologies regarding the conservation proposals, the use of existing AFLP datasets allows to carry out successful conservation approaches. The used dataset includes a total of 285 samples from 32 populations distributed along the full range of the species. Four selected AFLP primer combinations produced 1625 reproducible fragments which allowed for the determination of the population genetic structure and genetic diversity of this species. Among other genetic parameters obtained in this phylogeographical analysis, gene diversity indices, frequency and distribution of rare bands present in each population as well as the inference of distinct genetic clusters (Table1 and Fig. 1; data obtained from Salmerón-Sánchez et al., 2017) are useful to design conservation priorities (Médail & Baumel, 2018). In this study, our proposal is the use of this information to set conservation units, specifically, RGUCs (Pérez-Collazos et al., 2008). RGUCs selection relies on two premises based on the population structure and on the probabilities of the loss of rare alleles (those with an overall frequency lower than 10 %, and present in less than 20 % of the populations; Table S1). In the method, the calculated values of 2 J. Bobo-Pinilla et al. Journal for Nature Conservation 61 (2021) 126004 Table 1 Geographic and genetic diversity and rarity features of the populations of J. auricula. Assignment to the genetic clusters detected by Salmerón-Sánchez et al. (2017); Nei’s GD, Nei’s gene diversity index; DW, frequency down-weighted marker values; the last column refers to whether the populations are in a protected area; higher values of genetic diversity and rarity per cluster are indicated in bold. Nº Pop. Locality Genetic cluster Longitude/ latitude Nei’s GD DW Haplotypes Ribotypes 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 Lo, Ribafrecha Na, Sesma Na, Peralta, Barranco de Vallacuera Na, Fitero Z, between Tudela and Ejea Z, Barranco Val de Vares, monte de la Mediana Z, Bujaraloz, Laguna del Pez L, between La Sentiu de Sio and Balaguer L, La Noguera between Camarasa and Cubells L, Biosca-Sanahuja, Les Gesses T, La Albarca, Pla de la Devesa, Barranc de la Bova So, Monteagudo de las Vicarias So, Monteagudo de las Vicarias (b), Los Chorlitos Te, Las Cuerlas, Laguna de Gallocanta Te, Cubla-Villastar, Los Centenares M, Villaconejos M, Aranjuez, El Salobral To, Villacañas, lagunas de Peña Hueca Cu, El Pedernoso, llanos de Montilla Ab, Casas de Ves-Balsa de Ves, Corral del Caracol A, San Vicente de Raspeig A, Saladar de Agua Amarga A, Laguna de la Mata Mu, Jumilla, Sierra Santa Ana, La Buitrera Mu, El Rincón (Lorca) Al, Huercal Overa, Rambla de Santa Bárbara Gr, Galera, Barranco del Agua Gr, Cúllar, Rambla Amarguilla Gr, Baza, salar de Baza Al, Rambla de Tabernas Al, Rambla El Cautivo Al, La Sartenilla B B B B B B B B A B B B B B B B B B B B D D D D C C C C C C C C −2.37 /42.35 −2.08◦ /42.49◦ −1.85◦ /42.37◦ −1.88◦ /42.04◦ −1.38◦ /42.12◦ −0.75◦ /41.52◦ −0.26◦ /41.38◦ 0.87◦ /41.80◦ 0.93◦ /41.85◦ 1.31◦ /41.84◦ 0.90◦ /41.30◦ −2.14◦ /41.39◦ −2.18◦ /41.39◦ −1.53◦ /40.97◦ −1.13◦ /40.25◦ −3.51◦ /40.09◦ −3.63◦ /39.99◦ −3.35◦ /39.51◦ −2.77◦ /39.49◦ −1.19◦ /39.29◦ −0.57◦ /38.38◦ −0.53◦ /38.28◦ −0.68◦ /38.02◦ −1.33◦ /38.42◦ −1.88◦ /37.87◦ −1.96◦ /37.37◦ −2.57◦ /37.73◦ −2.62◦ /37.56◦ −2.74◦ /37.55◦ −2.45◦ /37.01◦ −2.44◦ /37.01◦ −2.41◦ /37.02◦ 0,123 0,121 0,105 0,098 0,109 0,108 0,119 0,119 0,100 0,112 0,112 0,114 0,120 0,129 0,120 0,115 0,104 0,114 0,111 0,113 0,092 0,097 0,091 0,097 0,096 0,088 0,090 0,087 0,074 0,071 0,079 0,083 5,858 5,940 5,278 4,699 5,599 5,437 5,730 5,236 8,672 6,728 5,983 5,935 5,949 7,637 4,882 5,579 5,405 4,154 4,682 6,315 5,989 5,566 7,146 6,090 4,374 4,376 7,254 9,717 4,227 3,902 3,945 4,103 III(2) VII(2) VII(2) III(2) III(2) VII(2) V(2) VI(3) IX(3) IV(2) I(1) & III(1) I(2) I(2) III(2) III(2) I(1) & III(1) III(2) IX(2) IX(3) I(3) II:2 I:2 II(3) IX(3) II(1) & VIII(2) VIII(3) II(2) I(2) I(2) II(2) II(2) II(2) I (2) I (2) I (2) I (2) I (2) I (2) I (2) I (2) I (2) I (2) I (2) I (2) I (2) I (2) I(1) & III, IV(1) I(1) & III, IV(1) III,IV(1) & I,V(1) IV(2) I(1) & III, IV(1) I(1) & I, IV(1) I(1) & III, IV(1) I(1) & I,III(1) I(1) & III, IV(1) ◦ ◦ I(1) & II,III(1) II(2) II(2) II(2) II(2) II(2) II(2) II(2) Fig. 1. (a) Location of the populations of J. auricula studied. Populations were assigned to the different genetic clusters following the results of STRUCTURE analysis over the AFLP dataset (Salmerón-Sánchez et al., 2017) (cluster A = blue; cluster B = red; cluster C = yellow; cluster D = green). (b) Neighbor-joining tree based on distance matrix of FST between every pair of populations. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article). probability of rare-allele loss are compared with the degree of interpopulation subdivision (Caujapé-Castells & Pedrola-Monfort, 2004; Pérez-Collazos et al., 2008). Before carrying out the analysis to determine the selection of populations to be preserved, it is necessary to establish the consistency of a priori potential subdivisions with genetic parameters. Plant genetic 3 J. Bobo-Pinilla et al. Journal for Nature Conservation 61 (2021) 126004 diversity is spatially structured at different scales as a result of environmental influences, life-history traits, and the demographic past history of the species (Engelhardt et al., 2014). This could lead to the need for treatment of several subgroups. In the case of J. auricula, it is possible to adopt two different criteria; the application of Bayesian methods over the AFLP dataset (Salmerón-Sánchez et al., 2017) has allowed determining the number of genetic units on the basis of the detected polymorphism. STRUCTURE v. 2.3.4 (Pritchard et al., 2003) software showed the existence of four genetic clusters (A, B, C, and D; Fig. 1 & Table 1). When populations showed genetic admixture, they were assigned to the predominant cluster. The second criterion is focused on the use of plastid and ribosomal sequences for the establishment of phylogeographic patterns in the species (Salmerón-Sánchez et al., 2017). In both criteria, the groups are inferred exclusively from genetic data, although we finally selected the clustering generated in STRUCTURE, as they yielded a better split among groups of populations. These genetic clusters were considered as geographic units or sampling areas. To support the geographic areas proposed, genetic relationships among populations were analyzed using a neighbor-joining tree (NJ; Saitou & Nei, 1987) based on distance matrix of FST between every pair of populations. For that, 1000 resampled FST distance matrices among populations were constructed by bootstrapping (Felsenstein, 1985) in AFLPSURV (Vekemans, 2002). FST values were calculated following Lynch and Milligan method (1994) after the estimation of allelic frequencies by means the method developed by Zhivotovsky (1999). Software package PHYLIP V3.6 (Felsenstein, 2005) was used to estimate the length of tree branches (FITCH, Fitch & Margoliash, 1967). Following Ceska et al. (1997), the total number of populations that should be preserved (n) to represent a given proportion of the genetic diversity (P) was estimated with the modified equation P = 1 - FnST (Segarra-Moragues & Catalán, 2010). FST value was calculated using ARLEQUIN 3.5.1.2 (Excoffier & Lischer, 2010). A proportion of 99 % and 99.9 % of the total genetic diversity was set for the populations of J. auricula. To calculate the probabilities of loss of the rare alleles (in our case, rare bands as we work with dominant markers), the expression L = (1 p)2N (Bengtsson et al., 1995) was used; here p represents the band frequency and N the number of populations in which a rare band is present. This expression is equally applicable for both codominant and dominant markers (Pérez-Collazos et al., 2008). For each rare AFLP band, the observed and expected probabilities of loss (Lo and Le, respectively), and the representative value (R) were calculated, following the method described by (Pérez-Collazos et al., 2008). R-value indicates the proportion of rare alleles (bands in our case) captured by sampling only one population (Bengtsson et al., 1995; Caujapé-Castells & Pedrola-Monfort, 2004; Pérez-Collazos et al., 2008; Segarra-Moragues & Catalán, 2010). R-values were also calculated in each genetic cluster to obtain the proportion of rare bands captured by sampling one population within them. The Preferred Sampling Area (PSA) for each rare AFLP band was chosen regarding the higher frequency within the areas (Table S1). Regarding the PSA percentages and the R-values of the genetic clusters, the optimal proportion of the populations to be sampled in each cluster was calculated. For each PSA, the populations were chosen by considering the higher value of Nei’s gene diversity index (Salmerón-Sánchez et al., 2017) to manage the maximum amount of diversity. The number of rare bands found in each population (Table 1) was also considered to test which percentage of them would be recovered after selecting the minimum number of populations of J. auricula following our managing scheme. Moreover, in order to verify the success of the proposal, the number of AFLP bands that would be captured if the population selection process had been random was calculated. Average values were calculated over 100 repetitions. 2.2. Contribution of the populations to global genetic diversity for ex situ conservation In order to create a seed bank that maximizes genetic diversity, the software Metapop2 v2.2.1 (López-Cortegano et al., 2019) was used on the AFLP dataset. This software calculates the expected proportional contribution (Cx; Table 2) of each population (within the genetic clusters) to a theoretical synthetic pool with maximum global gene diversity ∑ (Dmax). The software maximizes the function Dmax = 1 − nij=1 f ij ci cj where fij is the average coancestry between populations i and j, and ci and cj is the contribution of subpopulation i and j to the pool (Toro & Caballero, 2005). Randomization process was applied over 1000 repetitions in order to check the success of the selection. Metapop2 v2.2.1 also calculates the proportional contribution of each population to Nei’s gene diversity [ΔHnei, (Nei, 1978)] and the proportional contribution of the average Nei’s minimum genetic distance (ΔHdist) between populations. These contributions (amount of genetic diversity and distance gained or lost) are calculated by disregarding each population one by one from the analysis in each genetic cluster; as a practical approach values under 2% were not considered. Moreover, as an estimation of the distribution of the genetic diversity, the software calculates the proportion of gene diversity explained within and among populations in each area (Petit et al., 1998). 2.3. Mapping genetic diversity and rarity patterns To create a genetic diversity (Nei, 1978) and rarity (DW; Schönswetter & Tribsch, 2005) gradient map the Multilevel b-spline tool Table 2 Metapop2 v2.2.1 results in each of the genetic groups considered (B, C and D): ΔHnei, proportional increment/decrement of the within-population gene diversity when the population data is removed in the analysis; ΔHdist, proportional increment/decrement of Nei’s average genetic distance between populations when the population data is removed in the analysis; ΔHt, total variation; Cx, expected proportion of seeds from the populations in order to obtain the maximum diversity values in a synthetic population within each genetic cluster, values under 2% were not considered. Cluster B Cluster C Cluster D 4 Population ΔHnei ΔHdist ΔHt Cx (%) 1 2 3 4 5 6 7 8 10 11 12 13 14 15 16 17 18 19 20 25 26 27 28 29 30 31 32 21 22 23 24 −0.396 −0.237 0.377 0.535 0.140 0.155 −0.188 −0.167 0.072 0.061 0.045 −0.056 −0.404 −0.122 −0.088 0.318 −0.081 0.049 −0.024 −2.128 −0.730 −1.703 0.352 1.395 1.546 0.613 0.196 0.510 −0.529 0.588 −0.551 0.308 0.102 −0.248 −0.168 0.104 −0.311 −0.157 0.021 −0.870 −0.503 0.072 0.067 −0.143 0.202 0.228 −0.305 0.334 −0.161 −0.373 −2.968 −1.798 0.132 −0.809 −0.442 −0.327 0.998 −0.010 −0.475 1.061 −7.669 −7.993 −0.088 −0.135 0.129 0.367 0.244 −0.156 −0.345 −0.145 −0.799 −0.443 0.117 0.011 −0.546 0.080 0.140 0.013 0.253 −0.112 −0.397 −5.096 −2.528 −1.571 −0.457 0.953 1.220 1.611 0.186 0.035 0.532 −7.081 −8.544 7.30 – – – – 6.30 11.80 2.80 16.50 11.10 – 0.70 19.20 – 0.20 5.80 – 8.80 9.50 34.80 18.30 18.30 11.90 – 2.00 – 14.70 15.40 13.50 34.10 37.00 J. Bobo-Pinilla et al. Journal for Nature Conservation 61 (2021) 126004 (Conrad et al., 2015) implemented in QGIS (QGIS-Development-Team, 2017) software was used. This tool interpolates the specific values of the populations drawing the genetic diversity and rarity patterns. captured by choosing only one population of the entire range of the species (i.e. R-value) was 16.6 %. Considering the different genetic clusters independently, R-values of 22.73 % (cluster B), 36.33 % (cluster C), and 69.20 % (cluster D) were obtained (Table 3 and Fig. 2). Regarding the genetic cluster A, the R-value was not calculated as just one population belongs to this cluster. Based on the PSA distribution of these rare bands and R-values, the optimal proportion of the populations to be sampled within each genetic cluster was 0.31 (cluster B) : 0.26 (cluster C) : 0.43 (cluster D). With respect to the total number of populations to be sampled, FST value for the total dataset was 0.323. But considering that cluster A has only one population, FST value was recalculated excluding this population, an FST value of 0.312 was obtained. As a result, only four populations are needed (n = 3.95) to gather 99 % of the AFLP bands, whereas six populations are needed (n = 5.95) targeting to 99.9 % the proportion of AFLP bands to preserve. Giving the optimal proportion of the three clusters, 1.19–1.79 populations from cluster B, 1.07–1.61 populations from cluster C and 1.70–2.56 populations from cluster D should be targeted. The populations with the higher values of genetic diversity within each genetic cluster were chosen: populations 14 and 1 (Nei’s GD values of 0.129 and 0.123 respectively) from cluster B, populations 25 and 26 (Nei’s GD values of 0.096 and 0.090 respectively) from cluster C, and populations 22, 24, and 21 (Nei’s GD values of 0.097, 0.097 and 0.092 respectively) from the cluster D were selected. The selection proposed gathers 77.54 % of the AFLP bands while the random selection showed a value of 73.33 %. Regarding the number of rare bands found in each genetic cluster, the 100 %, 29.9 %, 27,9%, and 67.4 % of bands were recovered in clusters A, B, C, and D, respectively. Considering all the selected populations, 50.3 % of the rare bands were recovered. 2.4. Plastid and ribosomal DNA patterns Given the importance of including plastid DNA in conservation proposals as it represents the evolutionary history of plant species (Carvalho et al., 2019), the sequences of the three regions of the plastid DNA obtained for J. auricula in Salmerón-Sánchez et al. (2017) were downloaded from GenBank (psbA-3′ trnKmatK (Shaw et al., 2005), rpl16 (Small et al., 1998) and trnQ-5′ rps16 (Shaw et al., 2007); Table S2). The 75 sequences were assembled and edited using Geneious v 5.5.7 (Drummond et al., 2012) and aligned with Clustal W2 2.0.11 (Larkin et al., 2007). Further adjustments were made by visual inspection. The resulting sequences were concatenated; given the relative high number of haplotypes found by Salmerón-Sánchez et al. (2017), the program Gblocks (Castresana, 2000) was used to trim gapped regions and to remove non informative mutations. Finally, an unrooted haplotype network was constructed using TCS 1.21 (Clement et al., 2000). On the other hand, the use of ribosomal sequences has allowed to carry out a robust ascription to different clades in the elaboration of molecular phylogenies (Silva et al., 2015), setting a priori subdivisions of the populations (Bacchetta et al., 2008). Furthermore, nrDNA sequences are useful in the detection of hybridization events (Widmer & Baltisberger, 1999), as nucleotide additive patterns could be result of recent hybridization events (Aguilar et al., 1999; Plume et al., 2013). It is important to consider the possible existence of intraspecific hybrids and their importance in establishing species management plans (Chan et al., 2019). Ribosomal sequences and ribotypes obtained by Salmerón-Sánchez et al. (2017) were considered in our analyses due to the importance of this type of molecular marker. 3.2. Contribution of the populations to global genetic diversity for ex situ conservation 3. Results The intra- and inter-population contributions to the total genetic diversity in each area were 78.83 % and 21.17 %, for cluster B, 80 % and 20 % for cluster C, and 76.27 % and 23.73 % for cluster D. The variation of the genetic diversity and distance when removing populations within the cluster B were not significant, being the higher gain of genetic diversity 0.53 % when removing population 3 and the higher loss of genetic diversity of 0.40 % when removing population 14. With respect to the genetic distance, the decrease was 0.87 % when removing population 10 while the increase values were not significant. The optimal contribution calculated for the cluster B included 10 populations (Table 2) being the higher proportional values detected for populations 14, 10, 7, and 11 (with 19.20 %, 16.50 %, 11.80 %, and 11.10 % respectively). Regarding cluster C, the greatest decrease in genetic diversity (2.12 %) and distance (2.97 %) were found when removing population 25 while the greatest increase in genetic diversity and distance was detected for populations 30 and 31 (1.55 % and 1.00 %, respectively). The optimal contribution calculated for the cluster C included 5 populations (Table 2) being the higher proportional values detected for populations 25, 26, 27, and 32 (with 34.8 %, 18.3 %, 18.3 %, and 14.7 % respectively). The values calculated for the cluster D showed higher decrease values for genetic distance when removing populations 23 and 24 (4.43 % and 7.99 %, respectively). No significant values regarding genetic diversity were found. The optimal contribution calculated for the cluster D included all the populations (Table 2) being the proportional values detected 37 %, 34.1 %, 15.4 %, and 13.5 % for populations 24, 23, 21, and 22 respectively. The Nei’s genetic diversity values calculated for the synthetic populations in each genetic cluster were 0.149, 0.112, and 0.130 (for clusters B, C, and D respectively) while the random selection of seeds 3.1. Selection of RGUCs From a total of 1625 AFLP bands, 815 met the rarity requirements (Table 3). Of them, 16 were exclusive to cluster A, 311 were exclusive to cluster B, 110 were exclusive to cluster C, and 62 were exclusive to cluster D. After choosing the PSA for each of the rare bands (Table S1), a total of 115 bands were assigned to cluster A, 336 to cluster B, 173 to cluster C, and 183 to cluster D (1% of the rare AFLP bands were not assigned to any PSA; Table 3). The proportion of rare AFLP bands Table 3 Distribution of rare AFLP bands (those with an overall frequency lower than 10 %, and present in less than 20 % of the populations) and RGUCs calculation values in the different genetic clusters (A, B, C and D) considered and throughout the full range of J. auricula. PSA (Preferred Sampling Area); R-value (percentage of rare AFLP bands captured by sampling one population within the genetic clusters); n (calculated number of populations to be sampled to include a fixed diversity value; i.e., 99 % and 99.9 %); n values were corrected (see Material & Methods) to adjust the method giving that cluster A has only one population. Full range Total nº AFLP bands Nº rare AFLP bands Exclusive rare AFLP bands Nº rare AFLP bands (by PSA) % of rare AFLP bands (by PSA) R-value (%) Optimal proportion n n (99 % - corrected) n (99 % - integer) n (99.9 % - corrected) n (99.9 % - integer) 1625 815 – – – 16.9 – 4.075 3.955 – 5.95 – A B C D – 115 16 115 14.11 – – – – 1 – 1 – 591 311 336 41.23 22.73 0.32 – 1.24 2 1.87 2 – 267 110 173 21.23 36.33 0.26 – 1.05 2 1.58 2 – 227 62 183 22.45 69.20 0.42 – 1.67 2 2.51 3 5 J. Bobo-Pinilla et al. Journal for Nature Conservation 61 (2021) 126004 Fig. 2. Regression lines of the average rare bands frequency (x-axis) with the negative logarithms of the observed and expected probabilities of loss [−log(Lo) (grey diamonds) and − log(Le) (black circles)] over the full set of rare AFLP bands and over the genetic clusters (B, C and D) of J. auricula. The quotient between the slopes of the observed and the expected regression lines indicates the percentage of rare AFLP bands represented when sampling a single population within the clusters (R-value). Fig. 3. (a) Nei’s gene diversity and (b) rarity patterns (red = low; yellow = medium; green = high). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article). 6 J. Bobo-Pinilla et al. Journal for Nature Conservation 61 (2021) 126004 from the populations within the clusters would result in values of 0.144, 0.103, and 0.119; this evidences an increase of 3.3 % (cluster B), 7.6 % (cluster C), and 8.1 % (cluster D) of the Nei’s genetic diversity values of the proposal with respect to the random selection. along the distributional range, two haplotypes (II and VIII), which differ by only one step from haplotype I, were exclusive to the southern populations. Haplotype IX also differ one step from haplotype VIII and was distributed to the central-south populations but was also found in population 9 to the north. Haplotype III is also one step away from the central haplotype and was found as the main haplotype at the northern populations. Haplotypes IV, V, and VI and are one step away from haplotype III and were exclusive to the northern populations 8, 7, and 10 respectively. Finally, haplotype VII differs by only one step from haplotype V and was also exclusive to three northern populations (i.e., 2, 3, and 6). Regarding ribosomal sequences, in the phylogeographic study of Jacobaea auricula (Salmerón-Sánchez et al., 2017), five different ribotypes were found. This provided evidence of intraspecific hybridization in eight populations (15, 16, 17, 19, 21, 22, 23, and 25). The distribution of ribotypes is shown in Table 1. 3.3. Map of genetic diversity and rarity patterns The Nei’s gene diversity pattern showed a clear signal of low diversity to the south of the distribution range while the values at the north of the range were higher (Fig. 3a); the most impoverished populations to the south were populations 29 and 30 with a diversity value of 0.74 and 0.71 respectively, while in the north populations 3, 4, and 17 had relative low values of genetic diversity (0.105, 0.098, and 104 respectively). In contrast, populations 1 and 14 held the higher values of genetic diversity to the north (0.123 and 0.129) while populations 22, 24, and 25 held the higher values to the south (0.097, 0.097, and 0.096 respectively). The rarity pattern (Fig. 3b) also showed the lower values in the southern distributional range (i.e., 3.902, 3.945, and 4.103 in populations 30, 31, and 32 respectively) but also the higher rarity values (7.254 and 9.717 in populations 27 and 28) together with some populations from the north of the Iberian Peninsula (8.672 and 7.637 in populations 9 and 14). 4. Discussion 4.1. In situ conservation: selection of Relevant Genetic Units for Conservation (RGUCs) As Falk & Holsinger (1991) suggest, the highest priority for in situ conservation systems is to capture the core of variability present in the species. The method of choosing RGUCs (Pérez-Collazos et al., 2008) allows the selection of the minimum number of populations of Jacobaea auricula that should be preserved to mitigate the possible loss of genetic diversity of the species. Our conservation managing proposal would account for a moderate percentage of rare AFLP bands (50 %) when considering the whole distribution of the species. We obtained different results for each genetic group; clusters A and C of our selection accounted for 100 % and 67 % of AFLP rare bands respectively while the results of clusters B and C were lower, capturing 30 % and 28 % of the total number of rare AFLP bands respectively. This is a considerable 3.4. Plastid and ribosomal DNA pattern The alignment of the concatenated DNA sequences after the removal of the gaps and uninformative mutations presented 2322 bp which included 33 mutations – eleven of which were considered informative (Table S2). Considering the different regions amplified, six substitutions were found for the psbA-3′ trnK-matK region, four substitutions were found for the rpl16 region, whereas the trnQ-rps16 region only contained two substitutions. These mutations defined a total of 9 haplotypes (Fig. 4). Whereas the central haplotype of the network was distributed Fig. 4. (a) Geographical distribution of the haplotypes found in J. auricula; (b) Haplotype network (size of the circles represents the number of samples; black dots represent haplotypes not found). 7 J. Bobo-Pinilla et al. Journal for Nature Conservation 61 (2021) 126004 Fig. 5. Geographical distribution of the selected populations for both in situ (red) and ex situ (green) genetic conservation. Populations selected in the reinforcement plan are colored in yellow. Black dots are non-selected populations. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article). proportion bearing in mind that, in our proposal, we selected eight populations as RGUCs (see locations in Fig. 5). This is only 25 % of the total number of the populations analyzed. Other authors who have considered the same methodology to create a management proposal required a higher proportion of populations to conserve similar levels of genetic diversity as in J. auricula (Caujapé-Castells & Pedrola-Monfort, 2004; Ciofi & Bruford, 1999; Mota et al., 2019; Peñas et al., 2016; Pérez-Collazos et al., 2008; Segarra-Moragues & Catalán, 2010); although it is true that the number of populations to be conserved will depend on the total FST value for the whole distribution of the species and on the different genetic subdivisions considered. In contrast, the percentage of rare alleles or fragments which would be recovered by selecting the PSA was higher in these studies. This depends on the type of molecular marker used in the proposal of management and the number of rare loci for each species. In general, when a greater number of populations were selected, the number of rare alleles recovered was greater. For example, in Andro& cymbium gramineum (Cav.) McBride (Caujapé-Castells Pedrola-Monfort, 2004), eight out of thirteen populations were selected, gathering 97 % of rare alleles. In the case of Boleum asperum, four out of eight populations were selected (Pérez-Collazos et al., 2008) which included 85.10 % of the rare bands. Also, in Borderea pyrenaica Miégeville, five out of the eleven populations studied were selected (Segarra-Moragues & Catalán, 2010), containing 97.5 % of the rare alleles. In other studies, the results were similar to those obtained in J. auricula. Thus, in Convolvulus boissieri Steud., where 6 out of the 15 populations studied were selected, around 61 % of the rare fragments were preserved (Mota et al., 2019). Also, these different outputs based on the recovery rate of rare fragments or alleles can be a consequence of the number of molecular markers studied, or of the molecular marker type. Thus, a higher number of rare bands analyzed (815 in J. auricula, 273 in A. edulis, 102 in C. boissieri and 47 in B. asperum) were found in the studies based on dominant AFLP markers with respect to those that used codominant markers, such as isoenzymes (38 rare alleles in the A. gramineum) or microsatellite markers (24 in B. asperum). Selection of such populations also would allow the capture of most of haplotypes (six out nine), being distributed as follows: haplotype I in population 22, haplotype II in population 21; haplotype III in populations 1 y 14; haplotype VI in population 19; haplotype VIII in populations 25 y 26 and haplotype IX in population 24. This output was similar to the obtained in C. boissieri (seven out fifteen; Mota et al., 2019) or in A. edulis (five out seven; Peñas et al., 2016) which is considerable as our approach does not take into account haplotypes in the selection of the PSA. Among the non-captured haplotypes, two were exclusive to one single population (haplotype IV and V, located in populations 10 and 7 respectively), and the remaining was present in three populations (haplotype VII, in populations 2,3 and 5). With respect to ribosomal sequences, all ribotypes except V (exclusive to population 17) were captured. Moreover, we found evidence of intraspecific hybridization in eight of the populations considered (15, 16, 17, 19, 21, 22, 23, and 25; see Table 1). Of them, populations 21, 22, and 25 would be included. Given the controversy regarding the role of intraspecific hybridization in natural populations (Chan et al., 2019), we should be careful in the inclusion of these populations in our management proposal. However, this would be an opportunity for a deeper study of the effect of natural intraspecific hybridization in J. auricula that allows us to assess the suitability of hybridization as a conservation tool (as in the case of Pinus torreyana Parry ex Carrière; Hamilton et al. (2017)). Twelve out the 32 populations of J. auricula studied are located within different protected areas (such as micro-reserves, wetlands or natural parks, Sites of Community Importance or Special Areas of Conservation of Natura 2000 network; see Table 1) either partially or in its complete distribution; of them, our RGUC selection included two populations (9 and 22). The fact of already having two populations that are inside protected areas facilitates our work when proposing this in situ conservation strategy; in any case, populations within protected areas also need genetic conservation as complementary strategy. At the present, protection policies applied to J. auricula are based on the morphological subdivision of the species (Ascaso & Pedrol, 1991; De 8 J. Bobo-Pinilla et al. Journal for Nature Conservation 61 (2021) 126004 La Torre et al., 1997). As a consequence, only subspecies are included in the different regional and Spanish red lists, under different threats categories (Anthos, 2020). Other protection proposals, such as reserve selection of gypsophile flora (Pérez–García et al., 2011), coincide partially in some localities that we have selected for genetic conservation of J. auricula (Table 1 and Fig. 5), but is evident the existence of discrepancies as reserve selection is based on the analysis of the whole gypsophile flora. Our approach, instead, considers the detected genetic groups present in the whole species. From our point of view, present protection could be better targeted and would have a more effective scientific basis if the RGUC concept is followed. Thus, the RGUCs selection would complement present areas of protection of the species throughout its range of distribution. values when compared with the rest of the populations of the respective clusters (Fig. 5). Our proposal includes the reinforcement of these populations with the optimal proportions calculated here for each cluster. Two considerations must be made regarding the reinforcement of the populations: 1) the risk of inbreeding depression (Barrett & Kohn, 1991; Keller & Waller, 2002), and 2) outbreeding depression (Hufford & Mazer, 2003; Tallmon et al., 2004). Regarding inbreeding depression, the creation of an efficient seed bank that consider the genetic diversity and rarity makes population reinforcements reliable and helps to increase the success of the proposal (Fenu et al., 2019; Lienert, 2004). Outbreeding depression must also be considered given the wide range of J. auricula. The ecologic differences of the areas that the plant inhabits makes it probable that the introduction of individuals from different conditions decrease the survival and reproductive ability as local adaptations could have been developed within the areas (Fenster & Galloway, 2000; Lema & Nevitt, 2006). The genetic pattern of four clusters found by Salmerón-Sánchez et al. (2017) reduces the risk of outbreeding depression as the reintroductions are made from populations from the same cluster (Kaulfuß & Reisch, 2017; Shemesh et al., 2018). 4.2. Ex situ conservation: selection for seed bank and reinforcement planning Although it appears proven the value of the RGUCs in terms of in situ genetic conservation (Peñas et al., 2016; Pérez-Collazos et al., 2008), the conservation proposals focused on passive protection may not be stopping the diversity loss (Fenu et al., 2019). The mere creation of protected areas seems to underestimate the adaptive potential that some populations may contain (Jump et al., 2009; Volis, 2019). Moreover, considering the percentage of rare bands recovered after the selection of eight RGUCs, and due to the restricted distribution of the non-captured ribotype and haplotypes, it is also necessary to establish ex situ conservation measures. Ex situ and in situ techniques should be used in a combined way to maximize the success of the proposal increasing in situ protection ability (Engelmann et al., 2007; Hawkes et al., 2000; Li & Pritchard, 2009; Volis & Blecher, 2010). In the creation of a seed bank, collecting most of the diversity is essential to ensure functionality (Caujapé-Castells & Pedrola-Monfort, 2004; Pearse & Crandall, 2004). Traditionally plant material has been collected from several populations from different habitats assuming that diversity was distributed along populations, particularly in wide distribution species (Hamrick & Godt, 1990; Hamrick et al., 1991). Moreover, the capture of alleles present at a very low frequency is unlikely in samples of realistic size when no genetic information is provided (Lawrence et al., 1995). The establishment of systematic strategies to maximize the collection of the genetic diversity and rarity is thus imperative (Farnsworth et al., 2006). The optimal proportion of seeds calculated in our study provides a direct estimation in order to maximize the genetic diversity in the seed bank within the four different clusters. Also, this method allows the optimization of the sampling effort establishing a lower limit under which we could dismiss the importance of the given population to the whole seed bank (e.g., 2%). At least 21 of the 32 populations (ten from cluster B, six from cluster C and all the populations from clusters D and A) should be selected to create a seed bank for the species. This selection not only gathers all the haplotypes and ribotypes of the species but also those present within the four genetic clusters (see locations in Fig. 5). Further studies are needed to correct the optimal proportions calculated when taking into consideration the differential gemination rates of the different areas and populations (Bacchetta et al., 2008). Moreover, in order to ensure the success of the ex situ proposal and giving the self-compatible characteristics of the species (Kunin, 1997), the seed production in self and cross pollination should be studied. The reinforcement of populations of rare and threatened species has become essential for biodiversity conservation (Armstrong & Seddon, 2008). These proposals aim to increase the survival of a given species (Commander et al., 2018; Volis & Blecher, 2010). The genetic diversity pattern of J. auricula shows low values in all the populations in the southern distributional range; a similar pattern has been found for other edaphic endemic species as Gypsophila struthium Loefl. (Martínez-Nieto et al., 2013). The most impoverished populations are populations 29–32 (with values under 0.085 Nei’s GD). Furthermore, populations 4 and 17 from cluster B and population 23 from cluster D also hold low genetic 5. Conclusions In situ-based conservation relying on global floristic criteria does not guarantee the conservation of genetic diversity of the different populations of the species. This is even the case where these species appear linked to a very specific habitat, such as outcrops of gypsum and saltsoils. Our proposal would improve the implementation of future genetic conservation measures for J. auricula – a model of endemic plants from edaphic habitat islands – allowing us to preserve the highest proportion of the gene pool possible of the species by combining both in situ and ex situ approaches. This ensures that future translocation and reinforcement planning of the most degraded populations will be possible. Funding This work was partially supported by the Spanish Ministerio de Economía y Competitividad through the projects CGL2010-16357 and CGL2012-32574. E. Salmerón-Sánchez was supported by the University of Almería, through the projects ‘Assessment, Monitoring and Applied Scientific Research for Ecological Restoration of Gypsum Mining Concessions (Majadas Viejas and Marylen) and Spreading of Results (ECORESGYP)’ sponsored by the company EXPLOTACIONES RÍO DE AGUAS S.L. (TORRALBA GROUP); ‘Provision of services, monitoring and evaluation of the environmental restoration of the mining concessions Los Yesares, María Morales and El Cigarrón’ sponsored by the company Saint Gobain Placo Iberica S. We would like to thank M. Montserrat Martínez-Ortega helped with field work and initial analyses. Declaration of Competing Interest The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results. Acknowledgements The authors would like to thank M. Montserrat Martínez-Ortega, Luz M. Muñoz-Centeno, Fabián Martínez-Hernández, Sara Barrios and Teresa Malvar for their participation in DNA extractions, molecular analyses and in general for the help provided. We also thank Sara Barrios, María Santos, Santiago Andrés, Blas Benito and Antonio Abad for the help provided in the collection of plant material. Finally, we are thankful to Francisco J. Pérez-García for his valuable comments concerning halogypsophyte species. 9 J. Bobo-Pinilla et al. Journal for Nature Conservation 61 (2021) 126004 Appendix A. Supplementary data Iberian gypsum steppes. In M. B. Morales, & J. Traba (Eds.), Steppe ecosystems: Biological diversity, management and restoration: Environmental research advances (pp. 197–210). Nova Science Publishers Inc. Evans, D. (2006). The habitats of the European Union habitats directive. Biology and Environment, 106(3), 167–173. https://doi.org/10.3318/BIOE.2006.106.3.167. Excoffier, L., & Lischer, H. E. L. (2010). Arlequin suite ver 3.5: A new series of programs to perform population genetics analyses under Linux and Windows. Molecular Ecology Resources, 10(3), 564–567. https://doi.org/10.1111/j.17550998.2010.02847.x. Falk, D. A., & Holsinger, K. E. (1991). Genetics and conservation of rare plants (p. 283). Farnsworth, E. J., Klionsky, S., Brumback, W. E., & Havens, K. (2006). A set of simple decision matrices for prioritizing collection of rare plant species for ex situ conservation. Biological Conservation, 128(1), 1–12. https://doi.org/10.1016/j. biocon.2005.09.010. Fay, M. F., & Krauss, S. L. (2003). Orchid conservation genetic in the molecular age. In K. W. Dixon, S. P. Kell, R. L. Barrett, & P. J. Cribb (Eds.), Orchid conservation (pp. 91–112). Natural History Publications. Issue 6. Felsenstein, J. (1985). Phylogenies and the comparative method. The American Naturalist, 125(1), 1–15. Felsenstein, J. (2005). PHYLIP (phylogeny inference package) version 3.6. Seattle: Department of Genome Sciences, University of Washington. http://evolution.geneti cs.washington.edu/phylip.html. Fenster, C. B., & Galloway, L. F. (2000). Inbreeding and outbreeding depression in natural populations of Chamaecrista fasciculata (Fabaceae). Conservation Biology, 14 (5), 1406–1412. https://doi.org/10.1046/j.1523-1739.2000.99234.x. Fenu, G., Bacchetta, G., Charalambos, S. C., Fournaraki, C., Giusso del Galdo, G. P., Gotsiou, P., Kyratzis, A., Piazza, C., Vicens, M., Pinna, M. S., & de Montmollin, B. (2019). An early evaluation of translocation actions for endangered plant species on Mediterranean islands. Plant Diversity, 41(2), 94–104. https://doi.org/10.1016/j. pld.2019.03.001. Fitch, W. M., & Margoliash, E. (1967). Construction of phylogenetic trees. Science, 155 (3760), 279–284. https://doi.org/10.1126/science.155.3760.279. Forest, F., Grenyer, R., Rouget, M., Davies, T. J., Cowling, R. M., Faith, D. P., Balmford, A., Manning, J. C., Procheş, Ş., van der Bank, M., Reeves, G., Hedderson, T. A. J., & Savolainen, V. (2007). Preserving the evolutionary potential of floras in biodiversity hotspots. Nature, 445(7129), 757–760. https://doi.org/ 10.1038/nature05587. Frankham, R., Ballou, J. D., Briscoe, D. A., & McInnes, K. H. (2004). A primer of conservation genetics. Cambridge University Press. https://doi.org/10.1017/ CBO9780511817359. Gutiérrez, F., Guerrero, J., & Lucha, P. (2008). A genetic classification of sinkholes illustrated from evaporite paleokarst exposures in Spain. Environmental Geology, 53 (5), 993–1006. https://doi.org/10.1007/s00254-007-0727-5. Hamilton, J. A., Royauté, R., Wright, J. W., Hodgskiss, P., & Ledig, F. T. (2017). Genetic conservation and management of the California endemic, Torrey pine (Pinus torreyana Parry): Implications of genetic rescue in a genetically depauperate species. Ecology and Evolution, 7(18), 7370–7381. https://doi.org/10.1002/ece3.3306. Hamrick, J. L., & Godt, M. J. W. (1990). Allozyme diversity in plant species. Plant population genetics, breeding, and genetic resources (pp. 43–63). Sinauer Associates Inc.. https://doi.org/10.1016/0022-4596(78)90152-4. Hamrick, J. L., Godt, M. J. W., Murawski, D. A., & Loveless, M. D. (1991). Correlation between species traits and allozyme diversity: Implications for conservation biology. In D. A. Falk, & K. E. Holsinger (Eds.), Genetics and conservation of rare plants (pp. 75–86). Oxford University Press. Hawkes, J. G., Maxted, N., Ford-Lloyd, B. V., Hawkes, J. G., Maxted, N., & FordLloyd, B. V. (2000). The genetic resources of plants and their value. The ex situ conservation of plant genetic resources (pp. 1–18). Netherlands: Springer. https://doi. org/10.1007/978-94-011-4136-9_1. Hoban, S., Bruford, M., D’Urban Jackson, J., Lopes-Fernandes, M., Heuertz, M., Hohenlohe, P. A., Paz-Vinas, I., Sjögren-Gulve, P., Segelbacher, G., Vernesi, C., Aitken, S., Bertola, L. D., Bloomer, P., Breed, M., Rodríguez-Correa, H., Funk, W. C., Grueber, C. E., Hunter, M. E., Jaffe, R., … Laikre, L. (2020). Genetic diversity targets and indicators in the CBD post-2020 Global Biodiversity Framework must be improved. Biological Conservation, 248, Article 108654. https://doi.org/10.1016/j. biocon.2020.108654. Holderegger, R., Balkenhol, N., Bolliger, J., Engler, J. O., Gugerli, F., Hochkirch, A., Nowak, C., Segelbacher, G., Widmer, A., & Zachos, F. E. (2019). Conservation genetics: Linking science with practice. Molecular Ecology, 28(17), 3848–3856. https://doi.org/10.1111/mec.15202. Hufford, K. M., & Mazer, S. J. (2003). Plant ecotypes: Genetic differentiation in the age of ecological restoration. Trends in Ecology & Evolution, 18(3), 147–155. https://doi. org/10.1016/S0169-5347(03)00002-8. Hyvärinen, M.-T. (2020). Rubus humulifolius rescued by narrowest possible margin, conserved ex situ, and reintroduced in the wild. Journal for Nature Conservation, 55, Article 125819. https://doi.org/10.1016/j.jnc.2020.125819. Jump, A. S., Marchant, R., & Peñuelas, J. (2009). Environmental change and the option value of genetic diversity. Trends in Plant Science, 14(1), 51–58. https://doi.org/ 10.1016/j.tplants.2008.10.002. Kaulfuß, F., & Reisch, C. (2017). Reintroduction of the endangered and endemic plant species Cochlearia bavarica—Implications from conservation genetics. Ecology and Evolution, 7(24), 11100–11112. https://doi.org/10.1002/ece3.3596. Keller, L. F., & Waller, D. M. (2002). Inbreeding effects in wild populations. Trends in Ecology & Evolution, 17(5), 230–241. https://doi.org/10.1016/S0169-5347(02) 02489-8. Kunin, W. E. (1997). Population biology and rarity: On the complexity of density dependence in insect—Plant interactions. In W. E. Kunin, & K. Gaston (Eds.), The Supplementary material related to this article can be found, in the online version, at doi:https://doi.org/10.1016/j.jnc.2021.126004. References Abeli, T., Dalrymple, S., Godefroid, S., Mondoni, A., Müller, J. V., Rossi, G., & Orsenigo, S. (2020). Ex situ collections and their potential for the restoration of extinct plants. Conservation Biology, 34(2), 303–313. https://doi.org/10.1111/ cobi.13391. Aguilar, J. F., Rosselló, J. A., & Feliner, G. N. (1999). Molecular evidence for the compilospecies model of reticulate evolution in Armeria (Plumbaginaceae). Systematic Biology, 48(4), 735–754. https://doi.org/10.1080/106351599259997. CSIC - Real Jardín Botánico (RJB), & Fundación Biodiversidad. (2020). Anthos. Sistema de información de las plantas de España. www.anthos.es. Armstrong, D. P., & Seddon, P. J. (2008). Directions in reintroduction biology. Trends in Ecology & Evolution, 23(1), 20–25. https://doi.org/10.1016/j.tree.2007.10.003. Ascaso, J., & Pedrol, J. (1991). De plantis vas-cularibus praesertim ibericis. Fontqueria, 31, 137–138. Avise, J. C. (2009). Phylogeography: Retrospect and prospect. Journal of Biogeography, 36 (1), 3–15. https://doi.org/10.1111/j.1365-2699.2008.02032.x. Bacchetta, G., Bueno Sánchez, A., Fenu, G., Jiménez-Alfaro, B., Mattana, E., Piotto, B., & Virevaire, M. (Eds.). (2008). Conservación ex situ de plantas silvestres. Principado de Asturias / La Caixa. Barrett, S. C. H., & Kohn, J. R. (1991). Genetic and evolutionary consequences of small population sizes in plants: Implications for conservation. In D. A. Falk, & K. A. Holsinger (Eds.), Genetics and conservation of rare plants (pp. 3–30). Oxford University Press. Bengtsson, B. O., Weibull, P., & Ghatnekar, L. (1995). The loss of alleles by sampling: A study of the common outbreeding grass Festuca ovina over three geographic scales. Hereditas, 122(3), 221–238. https://doi.org/10.1111/j.1601-5223.1995.00221.x. Carvalho, Y. G. S., Vitorino, L. C., de Souza, U. J. B., & Bessa, L. A. (2019). Recent trends in research on the genetic diversity of plants: Implications for conservation. Diversity, 11(4), 62. https://doi.org/10.3390/d11040062. Castresana, J. (2000). Selection of conserved blocks from multiple alignments for their use in phylogenetic analysis. Molecular Biology and Evolution, 17(4), 540–552. https://doi.org/10.1093/oxfordjournals.molbev.a026334. Caujapé-Castells, J., & Pedrola-Monfort, J. (2004). Designing ex-situ conservation strategies through the assessment of neutral genetic markers: Application to the endangered Androcymbium gramineum. Conservation Genetics, 5(2), 131–144. https:// doi.org/10.1023/B:COGE.0000029997.59502.88. Ceska, J. F., Affolter, J. M., & Hamrick, J. L. (1997). Developing a sampling strategy for Baptisia arachnifera based on allozyme diversity. Conservation Biology, 11(5), 1133–1139. https://doi.org/10.1046/j.1523-1739.1997.95527.x. Chan, W. Y., Hoffmann, A. A., & Oppen, M. J. H. (2019). Hybridization as a conservation management tool. Conservation Letters, 12(5), e12652. https://doi.org/10.1111/ conl.12652. Ciofi, C., & Bruford, M. W. (1999). Genetic structure and gene flow among Komodo dragon populations inferred by microsatellite loci analysis. Molecular Ecology, 8(12 Suppl. 1), S17–30. http://www.ncbi.nlm.nih.gov/pubmed/10703549. Clement, M., Posada, D., & Crandall, K. A. (2000). TCS: A computer program to estimate gene genealogies. Molecular Ecology, 9(10), 1657–1659. https://doi.org/10.1046/ j.1365-294X.2000.01020.x. Commander, L., Coates, D., Broadhurst, L., Offord, C. A., Makinson, R. O., & Matthes, M. (Eds.). (2018). Guidelines for the translocation of threatened plants in Australia. Australian Network for Plant Conservation. De La Torre, A., Alonso, M. A., & Vicedo, M. (1997). Senecio auricula s.l. en la península ibérica: problemas taxonómicos y posición fitosociológica. Anales de Biología, 22(22), 103–116. Denaeyer–De Smet, S. (1970). Note on the chemical composition of salts secreted by various gypsohalophytic species of Spain. Bull. Soc. Roy. Bot. Belg., 103, 273–278. DeSalle, R., & Amato, G. (2004). The expansion of conservation genetics. Nature Reviews Genetics, 5(9), 702–712. https://doi.org/10.1038/nrg1425. Doadrio, I., Perdices, A., & Machordom, A. (1996). Allozymic variation of the endangered killifish Aphanius iberus and its application to conservation. Environmental Biology of Fishes, 45(3), 259–271. https://doi.org/10.1007/ BF00003094. Drummond, A., Ashton, B., Buxton, S., Cheung, M., Cooper, A., Duran, C., Field, M., Heled, J., Kearse, M., Markowitz, S., Moir, R., Stones-Havas, S., Sturrock, S., Thierer, T., & Wilson, A. (2012). Geneious v 5.5.7. Biomatters Ltd.. http://www.gen eious.com Engelhardt, K. A. M., Lloyd, M. W., & Neel, M. C. (2014). Effects of genetic diversity on conservation and restoration potential at individual, population, and regional scales. Biological Conservation, 179, 6–16. https://doi.org/10.1016/j.biocon.2014.08.011. Engelmann, F., Dulloo, M. E., Astorga, C., Dussert, S., & Anthony, F. (Eds.). (2007). Complementary strategies for ex situ conservation of coffee (Coffea arabica L.) genetic ressources. A case study in CATIE, Costa Rica. Topical reviews in Agricultural biodiverstity. Bioversity International. Escudero, A., Palacio, S., Maestre, F. T., & Luzuriaga, A. L. (2015). Plant life on gypsum: A review of its multiple facets. Biological Reviews, 90(1), 1–18. https://doi.org/ 10.1111/brv.12092. Eugenio, M., Molina, C., & Montamarta, G. (2013). The conservation of high interest plant species offers the chance to preserve unique and vulnerable representatives of 10 J. Bobo-Pinilla et al. Journal for Nature Conservation 61 (2021) 126004 biology of rarity (pp. 150–173). Netherlands: Springer. https://doi.org/10.1007/97894-011-5874-9_9. Larkin, M. A., Blackshields, G., Brown, N. P., Chenna, R., Mcgettigan, P. A., McWilliam, H., Valentin, F., Wallace, I. M., Wilm, A., Lopez, R., Thompson, J. D., Gibson, T. J., & Higgins, D. G. (2007). Clustal W and clustal X version 2.0. Bioinformatics, 23(21), 2947–2948. https://doi.org/10.1093/bioinformatics/ btm404. Lawrence, M. J., Marshall, D. F., & Davies, P. (1995). Genetics of genetic conservation. I. Sample size when collecting germplasm. Euphytica, 84(2), 89–99. https://doi.org/ 10.1007/BF01677945. Lema, S. C., & Nevitt, G. A. (2006). Testing an ecophysiological mechanism of morphological plasticity in pupfish and its relevance to conservation efforts for endangered Devils Hole pupfish. Journal of Experimental Biology, 209(18), 3499–3509. https://doi.org/10.1242/jeb.02417. Li, D.-Z., & Pritchard, H. W. (2009). The science and economics of ex situ plant conservation. Trends in Plant Science, 14(11), 614–621. https://doi.org/10.1016/j. tplants.2009.09.005. Lienert, J. (2004). Habitat fragmentation effects on fitness of plant populations – A review. Journal for Nature Conservation, 12(1), 53–72. https://doi.org/10.1016/j. jnc.2003.07.002. Lopez, S., Rousset, F.Ç., Shaw, F. H., Shaw, R. G., & Ronce, O. (2009). Joint effects of inbreeding and local adaptation on the evolution of genetic load after fragmentation. Conservation Biology, 23(6), 1618–1627. https://doi.org/10.1111/j.15231739.2009.01326.x. López-Cortegano, E., Pérez-Figueroa, A., & Caballero, A. (2019). metapop2: Reimplementation of software for the analysis and management of subdivided populations using gene and allelic diversity. Molecular Ecology Resources, 19(4), 1095–1100. https://doi.org/10.1111/1755-0998.13015. Lynch, M., & Milligan, B. G. (1994). Analysis of population genetic structure with RAPD markers. Molecular Ecology, 3(2), 91–99. https://doi.org/10.1111/j.1365294x.1994.tb00109.x. Mace, G. M., & Purvis, A. (2008). Evolutionary biology and practical conservation: Bridging a widening gap. Molecular Ecology, 17(1), 9–19. https://doi.org/10.1111/ j.1365-294X.2007.03455.x. Maes, D., Vanreusel, W., Talloen, W., & Van Dyck, H. (2004). Functional conservation units for the endangered Alcon Blue butterfly Maculinea alcon in Belgium (Lepidoptera: Lycaenidae). Biological Conservation, 120(2), 229–241. https://doi. org/10.1016/j.biocon.2004.02.018. Margules, C. R., & Pressey, R. L. (2000). Systematic conservation planning. Nature, 405 (6783), 243–253. https://doi.org/10.1038/35012251. Martínez-Nieto, M. I., Segarra-Moragues, J. G., Merlo, E., Martínez-Hernández, F., & Mota, J. F. (2013). Genetic diversity, genetic structure and phylogeography of the Iberian endemic Gypsophila struthium (Caryophyllaceae) as revealed by AFLP and plastid DNA sequences: connecting habitat fragmentation and diversification. Botanical Journal of the Linnean Society, 173(4), 654–675. https://doi.org/10.1111/ boj.12105. Médail, F., & Baumel, A. (2018). Using phylogeography to define conservation priorities: The case of narrow endemic plants in the Mediterranean Basin hotspot. Biological Conservation, 224, 258–266. https://doi.org/10.1016/j.biocon.2018.05.028. Moore, M. J., Mota, J. F., Douglas, N. A., Olvera, H. F., & Ochoterena, H. (2014). The ecology, assembly and evolution of gypsophile floras. Plant ecology and evolution in harsh environments (pp. 97–128). https://digitalcommons.oberlin.edu/faculty_schol. Moritz, C. (1994). Defining “evolutionarily significant units”. Tree Genetics & Genomes, 9, 373–375. Mota, J. F., Salmerón-Sánchez, E., Pérez-García, F., Martínez-Hernández, F., MendozaFernández, A. J., Medina-Cazorla, J. M., & Merlo, M. E. (2019). Catálogo Delphi de la flora edafoendémica de los blanquizales dolomíticos béticos: bases para su conocimiento y conservación. In J. Peñas de Giles, & J. Lorite Moreno (Eds.), Biología de la conservación de plantas en Sierra Nevada: Principios y retos para su preservación (pp. 193–210). Editorial Universidad de Granada. Mota, J. F., Sánchez-Gómez, P., & Guirado, J. S. (Eds.). (2011). Diversidad vegetal de las yeseras ibéricas. El reto de los archipiélagos edáicos para la biología de la conservación. ADIF-Mediterráneo Asesores Consultores. Nei, M. (1978). Estimation of average heterozygosity and genetic distance from a small number of individuals. Genetics, 89(3), 583–590. Pärtel, M., Kalamees, R., Reier, Ü., Tuvi, E. L., Roosaluste, E., Vellak, A., & Zobel, M. (2005). Grouping and prioritization of vascular plant species for conservation: Combining natural rarity and management need. Biological Conservation, 123(3), 271–278. https://doi.org/10.1016/j.biocon.2004.11.014. Pearse, D. E., & Crandall, K. A. (2004). Beyond FST: Analysis of population genetic data for conservation. Conservation Genetics, 5(5), 585–602. https://doi.org/10.1007/ s10592-003-1863-4. Peñas, J., Barrios, S., Bobo-Pinilla, J., Lorite, J., & Martínez-Ortega, M. M. (2016). Designing conservation strategies to preserve the genetic diversity of Astragalus edulis Bunge, an endangered species from western Mediterranean region. PeerJ, 4(1), e1474. https://doi.org/10.7717/peerj.1474. Pérez-Collazos, E., Segarra-Moragues, J. G., & Catalán, P. (2008). Two approaches for the selection of Relevant Genetic Units for Conservation in the narrow European endemic steppe plant Boleum asperum (Brassicaceae). Biological Journal of the Linnean Society, 94(2), 341–354. https://doi.org/10.1111/j.1095-8312.2008.00961.x. Pérez–García, F., Martínez–Hernández, F., Garrido-Becerra, J. A., MendozaFernández, A. J., Medina-Cazorla, J. M., Martínez-Nieto, M. I., Salmerón-Sánchez, E., Guirado, J. S., Merlo, M. E., & Mota, J. F. (2011). Biogeografía de la conservación en los aljezares ibéricos: Patrones corológicos y selección de reservas. In J. F. Mota, P. Sánchez-Gómez, & J. Guirado (Eds.), Diversidad vegetal de las yeseras ibéricas. El reto de los archipiélagos edáficos para la biología de la conservación (pp. 303–304). ADIF-Mediterráneo Asesores Consultores. Petit, R. J., El Mousadik, A., & Pons, O. (1998). Identifying populations for conservation on the basis of genetic markers. Conservation Biology, 12(4), 844–855. https://doi. org/10.1111/j.1523-1739.1998.96489.x. Plume, O., Straub, S. C. K., Tel-Zur, N., Cisneros, A., Schneider, B., & Doyle, J. J. (2013). Testing a hypothesis of intergeneric allopolyploidy in vine cacti (Cactaceae: Hylocereeae). Systematic Botany, 38(3), 737–751. https://doi.org/10.1600/ 036364413X670421. Pritchard, J. K., Wen, X., & Falush, D. (2003). Documentation for STRUCTURE software, version 2.3. Chicago, IL: University of Chicago. http://pritch.bsd.uchicago.edu/struc ture.html. QGIS-Development-Team. (2017). QGIS geographic information system. Open Source Geospatial Foundation. http://qgis.org. Riddle, B. R., & Hafner, D. J. (1999). Species as units of analysis in ecology and biogeography: Time to take the blinders off. Global Ecology and Biogeography, 8(6), 433–441. https://doi.org/10.1046/j.1365-2699.1999.00170.x. Ryder, O. A. (1986). Species conservation and systematics: The dilemma of subspecies. Trends in Ecology & Evolution, 1(1), 9–10. https://doi.org/10.1016/0169-5347(86) 90059-5. Saitou, N., & Nei, M. (1987). The neighbor-joining method: A new method for reconstructing phylogenetic trees. Molecular Biology and Evolution, 4, 406–425. Salazar, C., & Peñas, J. (2011). Senecio auricula subsp. ca stellanu s Ascaso & Pedrol. In J. Mota, P. Sánchez-Gómez, & J. Guirado (Eds.), Diversidad vegetal de las yeseras ibéricas. El reto de los archipiélagos edáficos para la biología de la conservación (pp. 301–302). ADIF-Mediterráneo Asesores Consultores. Salazar, C., Peñas, J., & Sánchez-Gómez, P. (2011). Senecio auricula Bougeau ex Coss. subsp. auricula. In J. F. Mota, P. Sánchez-Gómez, & J. Guirado (Eds.), Diversidad vegetal de las yeseras ibéricas. El reto de los archipiélagos edáficos para la biología de la conservación (pp. 297–299). ADIF-Mediterráneo Asesores Consultores. Salmerón-Sánchez, E., Martínez-Ortega, M. M., Mota, J. F., & Peñas, J. (2017). A complex history of edaphic habitat islands in the Iberian Peninsula: phylogeography of the halo-gypsophyte Jacobaea auricula (Asteraceae). Botanical Journal of the Linnean Society, 185(3), 376–392. https://doi.org/10.1093/botlinnean/box058. Schönswetter, P., & Tribsch, A. (2005). Vicariance and dispersal in the alpine perennial Bupleurum stellatum L. (Apiaceae). Taxon, 54(3), 725–732. https://doi.org/10.2307/ 25065429. Segarra-Moragues, J. G., & Catalán, P. (2010). The fewer and the better: Prioritization of populations for conservation under limited resources, a genetic study with Borderea pyrenaica (Dioscoreaceae) in the Pyrenean National Park. Genetica, 138(3), 363–376. https://doi.org/10.1007/s10709-009-9427-2. Shaw, R. G., & Etterson, J. R. (2012). Rapid climate change and the rate of adaptation: Insight from experimental quantitative genetics. New Phytologist, 195(4), 752–765. https://doi.org/10.1111/j.1469-8137.2012.04230.x. Shaw, J., Lickey, E. B., Schilling, E. E., & Small, R. L. (2007). Comparision of whole chloroplast genome sequences to choose noncoding regions for phylogenetic studies in angiosperm: The tortoise and the Hair III. American Journal of Botany, 94(3), 275–288. https://doi.org/10.3732/ajb.94.3.275. Shaw, J., Lickey, E. B., Beck, J. T., Farmer, S. B., Liu, W., Miller, J., Siripun, K. C., Winder, C. T., Schilling, E. E., & Small, R. L. (2005). The tortoise and the hare II: Relative utility of 21 noncoding chloroplast DNA sequences for phylogenetic analysis. American Journal of Botany, 92(1), 142–166. https://doi.org/10.3732/ ajb.92.1.142. Shemesh, H., Shani, G., Carmel, Y., Kent, R., & Sapir, Y. (2018). To mix or not to mix the sources of relocated plants? The case of the endangered Iris lortetii. Journal for Nature Conservation, 45, 41–47. https://doi.org/10.1016/j.jnc.2018.08.002. Silva, J. L., Lim, S.-Y., Kim, S.-C., & Mejias, J. A. (2015). Phylogeography of cliff-dwelling relicts with a highly narrow and disjunct distribution in the western Mediterranean. American Journal of Botany, 102(9), 1538–1551. https://doi.org/10.3732/ ajb.1500152. Small, R. L., Ryburn, J. A., Cronn, R. C., Seelanan, T., & Wendel, J. F. (1998). The tortoise and the hare: Choosing between noncoding plastome and nuclear Adh sequences for phylogeny reconstruction in a recently diverged plant group. American Journal of Botany, 85(9), 1301–1315. https://doi.org/10.2307/2446640. Tallmon, D., Luikart, G., & Waples, R. (2004). The alluring simplicity and complex reality of genetic rescue. Trends in Ecology & Evolution, 19(9), 489–496. https://doi.org/ 10.1016/j.tree.2004.07.003. Toro, M. A., & Caballero, A. (2005). Characterization and conservation of genetic diversity in subdivided populations. Philosophical Transactions Biological Sciences, 360 (1459), 1367–1378. https://doi.org/10.1098/rstb.2005.1680. Vekemans, X. (2002). AFLP-SURV version 1.0. Belgium: Laboratoire de Génétique et Ecologie Végétale, Université Libre de Bruxelles. Volis, S. (2019). Conservation-oriented restoration – A two for one method to restore both threatened species and their habitats. Plant Diversity, 41(2), 50–58. https://doi. org/10.1016/j.pld.2019.01.002. Volis, S., & Blecher, M. (2010). Quasi in situ: A bridge between ex situ and in situ conservation of plants. Biodiversity and Conservation, 19(9), 2441–2454. https://doi. org/10.1007/s10531-010-9849-2. VV.AA. (2000). Lista Roja de Flora Vascular Española (valoración según categorías UICN). Conservación vegetal, 6, 11–38. extra. 11 J. Bobo-Pinilla et al. Journal for Nature Conservation 61 (2021) 126004 Webb, C. O., Ackerly, D. D., McPeek, M. A., & Donoghue, M. J. (2002). Phylogenies and community ecology. Annual Review of Ecology and Systematics, 33(1), 475–505. https://doi.org/10.1146/annurev.ecolsys.33.010802.150448. Widmer, A., & Baltisberger, M. (1999). Molecular evidence for allopolyploid speciation and a single origin of the narrow endemic Draba ladina (Brassicaceae). American Journal of Botany, 86(9), 1282–1289. https://doi.org/10.2307/2656776. Zhivotovsky, L. A. (1999). Estimating population structure in diploids with multilocus dominant DNA markers. Molecular Ecology, 8(6), 907–913. https://doi.org/10.1046/ j.1365-294x.1999.00620.x. 12