Characterizing surrogacy performance in the systematic conservation planning of riverine networks

Abstract

Priority‐area selection is a core phase of systematic conservation planning, often carried out using a single (surrogate) taxon. Efficient surrogates are expected to yield taxonomically representative priority areas that embrace the populations not only of the surrogate but also the surrogated taxa. Compared with the terrestrial realm, surrogacy performance of riverine taxa has received much less attention. This study compared the surrogacy performance of fishes (FI), macrophytes (MP), and benthic macroinvertebrates (MI) in terms of total area, connectedness, spatial congruence, and taxonomic representativeness of priority areas in the Middle Danube basin (Hungary). Setting three target values for each surrogate group, nine area prioritization designs were run by using a purpose‐written connectivity‐centric algorithm to emphasize the importance of longitudinal connectivity. FI provided the smallest, MP the intermediate, and MI the largest priority areas or solutions. Connectedness was greatest for FI, being one order of magnitude higher than for the other two groups. Pairwise spatial congruence was highest between FI and MP, lowest between MP and MI, and intermediate for FI and MI. MI yielded the most representative solutions, although the number of occurrences of the surrogated taxa in the solution, as a criterion of representation, modified the ratio of the taxa represented. Areas compiled from the overlapping parts of the surrogate‐specific priority sets proved to be smaller than, and similarly representative of, single‐taxon solutions. Taxon‐rich groups such as MI can serve as efficient surrogates, but that can result in larger solutions than for less taxon‐rich surrogates. Apart from the size, the compactness of the solutions seems to be determined by the identity of the surrogate taxa, and FI can be alternative surrogates in connectivity‐centric prioritization. At the same time, multi‐group approaches can enhance the robustness of area prioritization in terms of representativeness compared with single‐taxon procedures.