SecSel, a new software tool for conservation prioritization that is applicable to ordinal-scale data for multiple biodiversity features

subtracted. Marxan set conservation target as the sum of the evaluation data for each feature to be protected [5]. If the target exceeds the summed value, the difference is considered as a cost component. In Zonation, the prioritization process starts from the largest protected area cov- ering a whole region [4]. The least valuable site is initially omitted from the protected area. For a given biodiversity feature, the value of a site is the fraction of the total amount of the feature included in the protected area. The functionality of these algorithms requires that the evalua- tion data be ratio-scale for both Marxan and Zonation. Quantitative data such as area covered by vegetation to conserve, population size, popula- tion density, and biomass meet this requirement. Binary data such as presence-absence, which provide limited information, also can be handled as ratio-type data and fed to Marxan and Zonation. It is often the case that non-ratio type data that provide more information than presence- absence data are available. A typical example is a rough evaluation of population size or quality (e.g., Braun-Blanquet survey methods cover classes of plants in a phytosociological survey [7]). For example, experts are able to distinguish between stable populations with high density, pop- ulations with lower densities, and accidental observations of a few individuals. Such data can- not be added or subtracted, but they are more informative than presence-absence data. SecSel is a newly developed prioritization tool designed to handle rank-type data for priori-

tization analyses. SecSel selects areas for conservation that include local units that are highly ranked on the basis of multiple biodiversity features. Ecosystem functions and other purposes of land use may also be included in the analysis. In SecSel, a set of several local units with low ranks for a biodiversity feature is not a substitute for a single unit with a high rank. This limita- tion causes some inflexibility in site selection, but flexibility remains as long as there are ties between local units. Such ties occur in many cases when biodiversity features are roughly estimated. In many cases of conservation planning, conflicts between protection and other land use are inevitable. There are several approaches to finding compromises when such conflicts arise. For example, Chan et al. (2011) [8] have compared two approaches to incorporating ecosystem services in a Marxan analysis: handling the services as benefits to be targeted by conservation and as co-benefits or opportunity costs. With Marxan with Zones [6], which is an extended version of Marxan that can treat multiple zones for different objectives, land management uses, including biodiversity conservation, are categorized into different zones. With another widely used prioritization tool, Zonation, conflict may be simply resolved by masking out an area for land use other than biodiversity conservation. In the advanced version of Zonation, areas suitable for land uses that have negative impacts on biodiversity are given negative weight

in the ranking of priorities so that these areas are eliminated from the ones to be protected in the early stage of site selection [9]. In SecSel, ecosystem functions are also features to be considered. It is sometimes the case that use of land to exploit ecosystem functions conflicts with conservation of some of the bio- diversity features. In addition, other conflicts may occur between biodiversity features. For example, there may be pairs of biodiversity features for which conservation of one has a nega- tive impact on the other. Examples include species of woodland animals and grassland ani- mals, or aquatic species in eutrophic habitats and those in oligotrophic habitats. There are several approaches to resolving such conflicts. One approach is allocating an area for commercial land use beforehand and subsequently selecting an area for conservation from the remaining land. Such an approach is simple, but it may lead to deterioration of areas that are irreplaceable for the conservation of some features of biodiversity. Chan et al. (2011) [8] have used a widely used prioritization tool, Marxan, to try to incorporate the use of ecosystem services that may be positively or negatively correlated with conservation of biodiversity into PLOS ONE SecSel, a new software tool for conservation prioritization applicable to ordinal-scale data PLOS ONE | https://doi.org/10.1371/journal.pone.0247737 July 23, 2021 2 / 19 the National Institute for Environmental Studies, Japan. The funders had no role in study design, data collection and analysis, decision to publish, or

preparation of the manuscript. Competing interests: The authors have declared that no competing interests exist.