More species, less effort: Designing and comparing sampling strategies to draft optimised floristic inventories

Abstract Floristic inventories are an essential part of basic and applied research in botany. Despite their long history, floristic research is still carried out following non-objective (preferential) sampling approaches. Accordingly, final outputs (i) are extremely variable in the quality and quantity of collected data and hardly repeatable, (ii) rely on the researcher ability, and (iii) miss the basic assumptions to allow inferential statistical analyses. The aim of this work is to explore the drafting of a floristic inventory by means of geostatistical approaches to locate sampling units (plots) in the study area. We planned, carried out and then compared two different sampling strategies: (i) ‘basic strategy’, a stratified random sampling design based solely on a spatial optimization criterion (no prior information is available), and (ii) ‘advanced strategy’, a sampling design based on the maximisation of the spectral heterogeneity among sampling units, quantified in terms of Normalized Difference Vegetation Index values (NDVI). The strategy that maximises collected floristic information was assessed based on a combination of descriptive and quantitative statistics, such as (i) the completeness of the floristic inventory, (ii) the steepness of the rarefaction curves, (iii) the sampling time effort, and (iv) the plot contribution to the total β diversity. The 'advanced strategy' detected more species than the 'basic strategy' in all the sampling sites. The 'advanced strategy' accumulates species more quickly than the 'basic strategy'. The 'advanced strategy' selects sampling units more homogeneously contributing to total β diversity; in addition, they are better spatially arranged across the study area to capture environmental peculiarities of sampling sites. The 'advanced strategy' needs a little more effort in the design of the sampling strategy, but it is more effective than the 'basic strategy' in drafting a species inventory. We provide here the R routine to perform the 'advanced strategy', which can be profitably and freely used in any other geographic location and vegetation context.