Modelling the known and unknown plant biodiversity of the Amazon Basin

Abstract

Aims The overall aim of this study is to provide the data needed for Amazonian conservation and the sustainable management of the region. To this end I model the hypothetical distribution of plant species richness across the Amazon Basin, the distribution of the proportion of this species richness that can be accounted for by described species, and hence the distribution of the biodiversity which remains unknown.Location Amazonia, Neotropics.Methods Species richness across the Amazon Basin is estimated by comparing the occurrences of 1584 species of Magnoliophyta whose taxonomy and geographical distributions are relatively well known. These data are used to collate checklists for squares of 1°latitude by 1°longitude. Comparison of the checklists allows estimation of the relative expected diversity in the vicinity of each degree square. Summing the distributions of the hypothetical real ranges gives the proportion of the biodiversity that can be accounted for by described species. Subtraction of the second distribution from the first gives a distribution of the contribution to the overall biodiversity that the model predicts, potentially, results from as yet undescribed species. ResultsCollections documented in recent botanical monographs show an extremely biased distribution with the best knowledge being found in a very few relatively well-collected areas. At the degree square level, this model predicts that gamma biodiversity in the Amazon Basin is uniformly high across most of the basin. The model predicts that four large areas of the basin are particularly poorly known, and that they should contain large numbers of uncollected species. Main conclusionsThe model presented here highlights the difficulties of quantifying Amazonian plant diversity and its distribution. The low density of collections, and especially their extremely clumped distribution, undermines confidence in theories that seek to explain the apparent distribution of biodiversity. The model's prediction is substantially different from published predictions of the distribution of alpha diversity. Testing of this model in the areas identified as lacunae would require collecting programmes designed to collect fertile material of rare species. If the model's predictions are approximately accurate, the plant biodiversity of the Amazon Basin is considerably underestimated.

Tree communities of white-sand and terra-firme forests of the upper Rio Negro

“…These findings highlight the lacuna we have in our botanical knowledge of the forests of the upper Rio Negro. Deepening our botanical knowledge can be achieved by longterm projects aiming at systematically collecting botanical material (Hopkins 2007). We suggest that establishing longterm monitoring plots designed to capture variation in tree species diversity and composition at multiple spatial scales combined with flora projects planned to collect fertile material of rare species, would make an important contribution to our understanding of tree species diversity and composition in the upper Rio Negro.…”

Distribution of pteridophyte communities along environmental gradients in Central Amazonia, Brazil

“…Low and clumped density of botanical collections is a constraint for biodiversity mapping in Amazonia (Nelson et al 1990;Hopkins 2007) and consequently, biodiversity surrogates such as vegetation physiognomies are used as predictors of species distribution. Nevertheless, studies comparing floristic differences between Amazonian physiognomies like savannah, white-sand, and terra firme forests will not surprisingly find higher beta-diversity than studies focusing on the floristic turnover within a single forest physiognomy.…”

The Citizen Science Landscape: From Volunteers to Citizen Sensors and Beyond

“…It is, however, important to recognise that these datasets can be compromised by potential lack of precision, inherent biases, and uncertainties which are often present within these extensive studies [11,22,84]. For example, you may have more reports of species in areas that are highly populated by humans than in those that are sparsely populated, or more reports of species that are less cryptic than others.…”

“…For example, you may have more reports of species in areas that are highly populated by humans than in those that are sparsely populated, or more reports of species that are less cryptic than others. It is therefore a challenge to determine whether the data is correct or the reports are biased; this is the reason why many citizen science programs are so rigidly composed and use standardised protocols which are replicated across many stratified surveyed plots (see Table 5 and [11,22,84]). It is therefore important to ensure, in hypothesis driven studies, that sampling design does not introduce bias, and that counts are shaped by the data and not the ability of the observer International Journal of Zoology to detect or record data [85].…”