Improving biodiversity surrogates for conservation assessment: A test of methods and the value of targeted biological surveys. Issue 9 (2nd May 2018)

• Record Type: Journal Article
• Authors: Ware, Chris; Williams, Kristen J.; Harding, Jo; Hawkins, Brian; Harwood, Thomas; Manion, Glenn; Perkins, Genevieve C.; Ferrier, Simon
• Editors: Di Minin, Enrico
• Is Part Of: Diversity & distributions. Volume 24:Issue 9(2018)
• Page Start: 1333
• Page End: 1346
• Publication Date: 2018-05-02
• Subjects: Australia -- biodiversity surrogate -- biological surveys -- Bush Blitz programme -- generalized dissimilarity modelling -- invertebrates; Biodiversity -- Periodicals; Biodiversity conservation -- Periodicals; 577
• DOI: 10.1111/ddi.12766
• Languages: English
• ISSNs: 1366-9516

Abstract

Abstract: Aim: Conservation assessment and planning across extensive regions rely on the use of mapped or modelled surrogates because direct field‐based inventories of biodiversity rarely provide complete spatial coverage. Surrogates are assumed to represent spatial patterns in the distribution of biodiversity, yet the validity of this assumption is rarely evaluated. Here, we use data from new biological surveys targeting poorly known taxonomic groups across sparsely surveyed landscapes to test: (1) the performance of established and novel surrogates; and (2) the value of targeted survey data in further improving surrogate effectiveness. Location: Continental Australia. Methods: Surrogates were derived from either mapped land classifications (bioregions, vegetation types), or models of spatial turnover in biodiversity composition. Models were derived by linking best‐available biological observations to high‐resolution mapped climate, terrain and soil attributes using generalized dissimilarity modelling (GDM). The performance of surrogates was evaluated using survey data for eight biological groups collected as part of the Bush Blitz programme (http://bushblitz.org ). For the GDM‐based surrogates, within‐ and cross‐taxon performance was first evaluated for models fitted to biological data available prior to Bush Blitz, and then for models enhanced through the addition of the Bush Blitz data. Results: All of the tested surrogates performed significantly better than random across all eight biological groups. GDM‐based surrogates performed over 10% better on average than the best performing combination of mapped land classifications. The addition of Bush Blitz targeted data in GDM‐based surrogates led to further improvements in surrogate performance. Main conclusions: Our results support continued investment in targeted biological survey programmes to enhance the performance of surrogates and ensure that surrogates represent a wider breadth of biodiversity. The strong performance of compositional turnover modelling, relative to mapped land classifications, suggests that this surrogate strategy deserves greater consideration in future conservation assessments and has potential for use in continental‐scale monitoring of biodiversity.