Joining the incompatible: Exploiting purposive lists for the sample-based estimation of species richness

Abstract

The lists of species obtained by purposive sampling by field ecologists can be used to improve the sample-based estimation of species richness. A new estimator is here proposed as a modification of the difference estimator in which the species inclusion probabilities are estimated by means of the species frequencies from incidence data. If the species list used to support the estimation is complete the estimator guesses the true richness without error. In the case of incomplete lists, the estimator provides values invariably greater than the number of species detected by the combination of sample-based and purposive surveys. An asymptotically conservative estimator of the mean squared error is also provided. A simulation study based on two artificial communities is carried out in order to check the obvious increase in accuracy and precision with respect to the widely applied estimators based on the sole sample information. Finally, the proposed estimator is adopted to estimate species richness in the Maremma Regional Park, Italy.

Molecular Heterogeneity in Large-Scale Biological Data: Techniques and Applications

“…For a single sample, extrapolation or inference of diversity indices can only be extended to the sample population-the repertoire (or molecules) that is contained in the sample and could be observed if sequencing were exhaustive. Extensions to the individual level require multiple samples per individual, resulting in incidence data (91,112). Modeling this data is more complex, as it requires modeling both the missing species within each sample and the missing species between the samples.…”

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

“…Floras can be defined as species inventories for a given territory in a given time, aimed to link taxonomic, geographic, and temporal information (Palmer et al, 1995;Chiarucci and Bonini, 2005;Bedini et al, 2016). From a sampling perspective, floristic inventories are usually compiled by subjectively searching for and collecting plants, trying to obtain an exhaustive list of species over a large area, even though this task is often very challenging (Chiarucci and Palmer, 2005;Vondrák et al, 2016;Chiarucci et al, 2018). With this approach, generally referred to as 'preferential' or as 'opportunistic', surveyors select specific sites or habitats based on their subjective expectations (Chiarucci et al, 2018).…”

“…From a sampling perspective, floristic inventories are usually compiled by subjectively searching for and collecting plants, trying to obtain an exhaustive list of species over a large area, even though this task is often very challenging (Chiarucci and Palmer, 2005;Vondrák et al, 2016;Chiarucci et al, 2018). With this approach, generally referred to as 'preferential' or as 'opportunistic', surveyors select specific sites or habitats based on their subjective expectations (Chiarucci et al, 2018). Palmer et al (2002) defined the highly subjective combination of ability, experience, expertise and intuition that guide the botanist in the field as the "botanic internal algorithm".…”

“…On the other hand, in a probabilistic sampling approach species are identified and listed only if occurring in a priori selected sampling units, generally corresponding to plots (Chiarucci et al, 2018). In this approach, independent sampling units can be selected, thus satisfying the condition that each site has the same probability of being sampled.…”

“…Accordingly, the same authors inferred that using just over one-tenth of the workforce and resources required by a probabilistic method, a preferential strategy detected twice the number of species. On the other hand, the use of probabilistic approaches to draft a floristic inventory is encouraged (Palmer et al, 2002;Chiarucci and Bonini, 2005;Wiser et al, 2011), in order to allow rigorous statistical analyses (Chiarucci et al, 2018). Indeed, only probabilistic approaches meet the conditions of randomness, known probability and independence of statistical sampling (Chiarucci, 2007), thus (i) allowing comparisons among different regions and times or among unequal sampling efforts (Colwell and Coddington, 1994;Gotelli and Colwell, 2001;Koellner et al, 2004;Kalkhan et al, 2007), (ii) avoiding the onset of artefacts on the analyses (Palmer et al, 2008), and (iii) facilitating the reproducibility of methods.…”