Biases in conservation: a regional analysis of Spanish vertebrates

Apart from being the main cause of biodiversity loss, humans are also essential for sustainability and biological conservation. Regulations for the conservation of threatened species are key to respond to this challenge. Here we evaluate the efficacy of mandatory regulations in Spain, a decentralized country, for the effective protection of threatened terrestrial vertebrates. We studied the association between the number of both “vulnerable” and “endangered” species on regional lists of threatened species and approved management plans for such species (mandatory) with ecological (e.g. percentage of forest and agricultural areas), social (e.g. environmental awareness and GPD per capita) and geographical (e.g. total and protected areas) variables on the regional scale. Our results showed that an approved management plan is available for only 20% of threatened terrestrial vertebrates and there were important taxonomical and regional biases. Higher protection levels appeared in regions with higher percentages of protected area, more citizens’ environmental awareness, lower GPD per capita and shorter regional lists. Herpetofauna has fewer approved management plans than mammals or birds. Our results highlight the importance of integrating the perspective, knowledge and practices of all stakeholders (Academia, governments and society) to effectively apply environmental regulations.


INTRODUCTION
One of the major challenges that today's society faces is mitigating the biodiversity loss curve (Ceballos et al., 2015;Ceballos et al., 2017;Mace et al., 2018;Ripple et al., 2019;WWF, 2020) as it threatens the essential ecosystem services provided for humanity (Díaz et al., 2006), harming both human well-being and ecosystem functioning (Mace et al., 2005;Tilman et al., 2014). Global changes, such as habitat loss and fragmentation, soil erosion, water scarcity, development disparities or global warming (Schellnhuber et al., 1997), are the main threats for biodiversity conservation (Bellard et al., 2014). Despite humans being the main cause of biodiversity loss, they are also key to find strategies for sustainable development and effective conservation policies. For example, Hoffmann et al. (2015) calculated that without conservation measures, at least 148 ungulates species would increase the endangered status worldwide, and Bolam et al. (2020) estimated that bird and mammals extinction rates would have been 2.9-4.2-fold higher without conservation policies. However, conservation policies are generally driven by taxonomical and geographical biases (Lawler et al., 2006;Díaz et al., 2018). Albeit it helps to protect some species, it can undermine the overall biodiversity conservation goal (Di Marco et al., 2017;Christie et al., 2020a) Conservation actions focus on a small charismatic group of animals, mainly motivated by as the number of studied species is not proportional to the number of species found in nature.
For example, although there are more amphibian species than mammals, scientific research encompasses 10-fold more mammal species than amphibians (Clark & May, 2002). Rosenthal et al. (2017) demonstrated that endothermic vertebrates are extremely overrepresented in scientific publications, and Dos Santos et al. (2020) showed that scientific capacity of countries within species' range, followed by body mass and years since taxonomical description, explain conservation research efforts. In addition to these predictors, the presence of species in manmade habitats also can explain the conservation-related research effort (Silva et al., 2020).
Besides taxonomical biases, species conservation and research efforts are also subject to strong geographical biases. For example, some ecosystems from the northern hemisphere like temperate forests dominate the scientific literature (Lawler et al., 2006). Di Marco et al. (2017) found that around 40% of biological conservation studies are carried out in the USA, Australia and the UK, but only 10% in Africa and 6% in Southeast Asia. Due to taxonomic biases in scientific research, conservation policies and societal preferences are closely related issues, as we explained in the previous examples, governments decisions and social interest can largely define which animal groups are studied the most.
Here we used threatened status and conservation policies of terrestrial vertebrates in Spain to analyze taxonomical and geographical conservation biases. In this study, we quantitatively analyzed the national and regional lists of threatened species, and the management plans for terrestrial vertebrates, in Spain to evaluate the factors associated with fulfilling conservation policies, and to find potential geographical and taxonomical biases. Although conservation actions in Spain have substantially increased in the last few decades (Morillo & Gómez-Campo, 2000), a complete review of the conservation status of its threatened vertebrate species, and its adjustment to existing regulations, is lacking. We hypothesize that (1) the compliance with conservation regulations in Spain varies between regions, due to their ecological, social, and geographic differences. In this sense, higher percentage of threatened species under a management plans are found in regions with a higher Gross Domestic Product (GDP) per capita, because it is presumed that they have more technical and administrative capacity to develop the conservation policies; higher percentage of protected area and forest area, where conservation policies have traditionally been focused on; and more environmental awareness, being citizen pressure one of the major drivers that lead to political action. On the contrary, lower percentage of threatened species under a management plan are found in regions with high number of threatened species in their lists and higher percentage of agricultural area; (2) endotherm species, like mammals and birds, have higher rates of approved management plans than herpetofauna, due to most of reptiles and amphibians are generally less popular species

MATERIALS AND METHODS
This research was conducted in Spain, which includes 17 Regions known as "Autonomous Communities" (15 insular regions and 2 island regions; map in Appendix D). The wildlife conservation regulation of Spain, as a decentralized country, includes three levels: European directives, national laws and regional normative (Table 1, Appendix A). Each region must approve its own regional list of threatened species with at least two categories: "Endangered", which includes taxa or populations whose survival is unlikely if the causal factors of the current situation continue to act; "Vulnerable", taxa or populations at risk of moving to the previous category in the immediate future if the adverse factors that act on them are not corrected.
They must include the species already listed on the National List that are present in their territory. They can also include more species or increase the degree of certain species' threat according to scientific reports. Furthermore, each region has to develop a Conservation Plan for the threatened taxa or populations included in the "Vulnerable" category within a maximum 5-year period, and a Recovery Plan for the taxa or populations included in the "Endangered" category within a maximum 3-year period (Appendix A). The conservation measures contemplated in both these Conservation and Recovery management plans aim to recover populations of target species by reducing threats, and protecting or improving both current and potential habitats. In addition, these management plans seek to encourage research, dissemination, awareness and the involvement of society in conservation actions.
We used data on state and regional regulations of terrestrial vertebrates (mammals, birds, reptiles, amphibians). National law correspond to Law 42/2007, of December 13, on Natural Heritage and Biodiversity, and Royal Decree-Law 139/2011, of February 4, about the Spanish List of Threatened Species, as well as all its modifications. We also included data from regional normative on the management of biodiversity, and those regulations with a modified regional list of threatened species (Appendix A) and the management plans of every species (Appendix B). We considered any regulations approved before 2020.
For all 17 Regions, we obtained the number of threatened species categorized as "Vulnerable" and "Endangered", the year that each species was included on the list and the * + (Equation 1).
We used binomial generalized lineal models (GLM) to determine the association between the level of protection of threatened terrestrial vertebrates in each Region (i.e., percentage of species with a management plan) with the ecological, social and geographical variables of the Regions under study (Table C3) In order to check whether there were significant differences in the number of protected species among the different considered taxonomical groups (mammals, birds, reptiles, amphibians), we performed a Chi-square test for all pairs of groups. All the statistical analyses were carried out with R, version 3.6.3 (R Core Team, 2020). Model averaging was performed using MuMIn R package.

Geographical analysis of the protection level
The Spanish List of Threatened Species includes 110 species of terrestrial vertebrates ( Table   2). The range of the number of species included on the list per Region varies from 6 to 82 (Table 3). On average, each Region has 36.1±19.7 species on its list: 25.4±16.7 species classified as "Vulnerable" (VU) within a range from 4 to 70 species, and 10.73±5.2 cataloged as "Endangered" (EN) from 2 to 18 species (Table 3). Two Regions (Castilla y León and Catalonia) do not have an approved list of threatened species (they were excluded from the analysis).
Of the total listed species, 20% have an approved management plan (17% of the "Vulnerable" species and 48% of the "Endangered species). Significant differences were found in the number of approved management plans between conservation status (α = 0.05, P = 0.005). The mean of the Index of Adjustment to the list was 0.38±0.26, within a range from 0.06 to 1 ( Table 3).
The general average for the approval of the management plans in the different Regions is 9.5±6.5 years. No significant differences were found in the time taken to approve management plans (t = 0.45, df = 24, P = 0.66) between the "Endangered" (9.13 ± 6.87 years) and "Vulnerable" species (10.29±6.05 years). Only four regions indicated the average values of periods of time according to legal requirements (Table 6 and Table C1).
A negative association was observed between compliance with regulations and the number of threatened species plus GPD per capita, as was a positive association with the percentage of protected area and environmental awareness ( Figure 1). No significant association was noted for the other tested variables with compliance to regulations. Model averaging included four models, which encompassed 96% of the weight: number of protected species and GPD per capita (47%); number of protected species and percentage of protected areas (27%); number of protected species (18%) and number of protected species; and environmental awareness (4%) ( Table 4).

Taxonomical analysis of the protection level
Birds were the taxonomical group with the highest percentage of threatened species covered by a management plan (22%), followed by mammals (16%), amphibians (14%) and lastly reptiles (13%) ( Table 5). There were significant differences among all pairs of groups (P < 0.01).

DISCUSSION
Here we used data from Spanish conservation regulations to analyze potential geographical and taxonomical biases in the conservation of threatened terrestrial vertebrates in Spain. We found that the mandatory regulations for the conservation of wildlife species in Spain are complied with only by 20% of the species, which highlights generalized underprotection of terrestrial vertebrates. Moreover, delays in the applications for management plans were found, which represented were exceeded the application period described in laws by more than 2-fold for "Vulnerable" species and 3-fold for "Endangered" species. We also found regional differences due to ecological, geographical and social factors, as well as taxonomical biases.
Before discussing differences between taxa and regions, it is important to understand that the approval date of a management plan (the data we used in our work) is only a first step of the project cycle, which means that a management plan could have been approved but do not have an effective operational implementation. The management plans could fail after their approval if they are drafted by experts who do not know the correct project management approach (Battisti, 2018), they do not involve local agencies and citizens (Battisti et al., 2020), or periodical inspections are not carried out.

Regional biases
Not all Spanish regions protected their fauna equally. There was a wide variability between different regions, with some protecting every threatened species and some that almost did not develop management plans for any of them. Furthermore, two Regions (Catalonia and Castilla y León) have no approved regional lists. The Regions with extensively threatened species lists tend to cover a lower percentage of them using a management plan, probably due to budgetary, technical and administrative limitations. The application of joint management plans for a group of similar species, which are affected by the same threats, or plans to protect specific habitats could be useful in the event of resource scarcity for regions with longer lists, like steppe birds management plans (Junta de Andalucía, 2011).
Although we hypothesized that regions with higher economic capacity (higher GPD per capita) would provide larger budgets for conservation policies, we found that the number of threatened species with management plan tended to decrease GDP per capita rose. Therefore, the wealthiest regions are also the most urban and industrialized ones (Cámara de Comercio de España, 2018) and attach less importance to nature (Esteban Curiel, 2000;Mayer et al., 2009;Battisti & Zocchi, 2018), which indirectly implies that less social pressure is placed on species protection. However, the opposite phenomenon can emerge in some regions (e.g. increasing ecotourism from developed countries).
The percentages of protected areas in Regions were the other main variable that can explain compliance with regulations, followed by environmental awareness. Both showed a positive relation to the number of species with a management plan, which highlights the importance of protection measures and environmental awareness. More economical resources and research efforts are allocated to protected areas than to non-protected territories (West et al., 2006). Hence there is more knowledge about the species inhabiting them and more elements to draw up management plans. Moreover, many protected areas are of outstanding cultural, tourist and historic interest (Wray et al., 2010) with, thus, more interest in protecting their species can be expected of many stakeholders, such as local inhabitants, NGOs, regional politicians and researchers (Ament et al., 2016;Garibaldi & Turner, 2004). Environmental awareness varies widely among different Spanish Regions (Esteban Curiel, 2002;Instituto Nacional de Estadística, 2008), and the positive relation between environmental awareness and the proportion of species with a management plan makes it difficult to identify the causality of this relation. On the one hand, high rates of environmental awareness in Regions can pressurize governments to develop conservation measures or, on the contrary, conservation plans encourage dissemination actions to raise society's awareness, and these actions may increase environmental awareness. Moreover, the Regions with more protected areas lead to better environmental education and promote wildlife tourism, which also increases environmental awareness. In this sense, conservation education, the first step to get a greater environmental awareness (Jacobson et al., 2015), could also explain the regional biases in the management plans.
Birds, followed by mammals, obtain higher percentages of approved management plans than herpetofauna (Restani & Mazluff, 2002;Titley et al., 2017). These differences could be explained by managements plans being drawn up for less popular species, like most reptiles and amphibians, being prioritized less. On the one hand, mammals and birds, phylogenetically close to humans, are naturally more attractive to us (Martin-Forés et al., 2013;Titley et al., 2017). Additionally, there are many unfounded beliefs in and myths about herpetofauna in many regions, which have traditionally been considered harmful to local populations (Ceríaco, 2012;Pollard et al., 2015;Rueda Núñez, 2014;Uyeda, 2016). Due to this social rejection to reptiles and amphibians, they could face serious difficulties to be officially protected, even when they could be more threatened by extinction than popular species (Davies et al., 2018).
Moreover, a taxonomical bias in species research can influence conservation policies because conservation problems are identified earlier in the most studied taxonomical groups, and more knowledge and tools are available for their conservation (Burgman, 2004;Hutchings, 2004;Martín-López et al., 2009). Donaldson et al. (2018, using data for over 10,000 species, found that certain taxa were the focus of the most of the published papers, whilst other threatened species barely received research attention. They concluded that "while greater research on threatened species alone cannot ensure their protection, understanding taxonomic bias may be helpful to address knowledge gaps in order to identify research directions and inform policy."

Long delays in approving management plans
Our results show a major delays in drawing up and enforcing conservation measures and huge differences among Spanish Regions. Spanish National Law specifies that a Conservation Plan must be applied for "Vulnerable" species within a maximum 5-year period, and a Recovery Plan for "Endangered" species within a maximum 3-year period. However, the reality lies far from these requirements as management plans take around 10 years to be approved (when considering only the species with an approved plan, less than 30% of the total threatened species). The delay in performing conservation measures and actions could affect species preservation, especially that of endangered species that are close to extinction (Essl et al., 2015). In the current global change context, timing conservation actions is crucial, particularly for those species that are vulnerable to climate change (Naujokaitis- Lewis et al., 2018). Furthermore, early approval of management plans is more economically efficient and improves biodiversity recovery (Fuller et al., 2007). Given the administrative, social and economic peculiarities of the different countries, worldwide comparative studies related to delays in conservation policies are complex to be performed, but we consider not just occasional and local studies are necessary, but official and systematic informs, driven by countries governments and supranational organizations, guaranteeing compliance with the regulations and the established times.

Reducing biases in species conservation
Some highlights and general recommendations to reduce conservation biases can be deduced from our results. On the one hand, as several social, ecological and geographical variables are related to complying with environmental regulations, integration among Academia, governments and society is essential for their application (Armitage et al., 2007;Felt et al. 2007;Future Earth, 2014;Carmen et al., 2015). Academia provides systematic literature and measurable information to set up environmental policies (Heink et al., 2015;Dicks et al., 2016). Governments must apply and create the necessary regulations to protect threatened species. Society can pressurize governments so that regulations are drafted and subsequently applied. Finally, the local working groups and environmental volunteers could improve the actual implementation of management plans. Therefore, the ultimate reason of the wildlife underprotection in Spain and in so many other countries could be lack of political commitment and barely any social claim to protect biodiversity (Mehring et al., 2016).
Moreover, given the biodiversity crisis and the taxonomical and regional biases of threatened species in Spain, we consider that greater national coordination is necessary, along with enough budgetary allocations for researchers and technicians to cover all the taxonomical groups, and to make regular evaluations of both lists of threatened species and management plans. Effective biodiversity protection would ultimately result in a direct and indirect increase in ecosystem services, which could be considered mid-and long-term investments. What works in conservation? Using expert assessment of summarised evidence to identify practices that enhance natural pest control in agriculture. Biodiversity and    a) The inclusion of a taxon or population in the "Endangered" category will entail, within a maximum 3-year period, adopting a recovery plan that includes the most appropriate measures for fulfilling the objectives and, wherever appropriate, the designation of critical areas.
In critical areas, and in areas of the potential reintroduction or expansion of these taxa or populations defined as such in the recovery plans, conservation measures and management instruments, specific to these areas or integrated into other plans, will be established to prevent the negative effects on the species that have motivated the designation of these areas.
b) The inclusion of a taxon or population in the "vulnerable" category will entail, within a maximum 5-year period, adopting a conservation plan that includes the most appropriate measures for fulfilling the pursued objectives.
c) For those taxa or populations that share the same conservation problems or similar geographical areas, plans to simultaneously cover several taxa or populations may be developed.
d) For the species or populations that live exclusively, or a high proportion of them, in protected natural areas, the Natura 2000 Network or areas protected by international instruments, plans may be integrated into the corresponding planning and management figures for those spaces.
2. The Autonomous Communities will prepare and approve conservation and recovery plans for threatened terrestrial species.
Regional Laws.  Table C1. Average years taken to approve the management plans for threatened species in each Region since they appeared on the regional list, broken down into threatened categories.