Observer‐driven pseudoturnover in vegetation monitoring is context‐dependent but does not affect ecological inference

Abstract

Aims: Resurveys of vegetation plots are prone to several errors that can result in misleading conclusions. Minimizing such errors and finding alternative approaches for analyzing resurvey data are therefore important. We focused on inter-observer error and excluded other sources of variation. Our main questions were: How large is the inter-observer error (i.e. pseudoturnover) in vegetation surveys, and can it be reduced by simple data aggregation approaches? Which factors are affecting pseudoturnover and does it vary between morphological species groups or change over time? Is ecological inference robust against inter-observer differences? Location: Switzerland. Methods:Over seven years, we double-surveyed a total of 224 plots that were marked once in the field and then sampled by two observers independently on the same day. Both observers conducted full vegetation surveys, recording all vascular plant species, their cover, and additional plot information. We then calculated mean ecological indicator values and pseudoturnover.Results: Average pseudoturnover was 29% when raw species lists were compared. However, by applying simple aggregation steps to the species list, pseudoturnover was reduced to 17%. Pseudoturnover further varied among habitat types and declined over the years, indicating a training effect among observers. Most overlooked taxa, responsible for pseudoturnover, had low cover values. Mean ecological indicator values were robust against inter-observer differences. Conclusions:To minimize pseudoturnover, we suggest continuous training of observers and species-list aggregation prior to analysis. As mean ecological indicator values were robust against inter-observer differences, we conclude that they can provide a reliable estimate of temporal vegetation and ecological changes.

Landolt Indicator Values in Modern Research: A Review

“…Studies based on Landolt indicator values can be divided into three groups: studies of the flora of large areas (21% of all articles), studies of various plant communities (68%) (Table 2) and studies of individual plant species (11%). The study of flora was mainly carried out using modern methods and had a methodological character [41][42][43][44]. Over a five-year period, 19 articles were published.…”

“…Moreover, the researchers proposed a novel methodological approach that combines empirical knowledge on the determinants of soil seed bank characteristics with a modern, flexible statistical approach based on machine learning [93]. Interobserver error was analyzed in vegetation monitoring [42]. The main data on the "Database of anthropogenic vegetation of Urals and adjacent territories" registered in the Global Index of Vegetation-Plot Databases and the European Vegetation Archive are presented [24].…”

Vegetation change in acidic dry grasslands in Moravia (Czech Republic) over three decades: Slow decrease in habitat quality after grazing cessation

“…The resurvey was supervised by the original surveyor (M. Chytrý) to limit the observer-driven pseudoturnover (Boch et al, 2022). It was conducted between 4 June and 16 August.…”

Traditional haymaking is essential to sustain and restore species-rich subalpine grasslands

“…However, total species richness is not necessarily a suitable indicator to re ect environmental changes, as values mainly peak under environmental conditions with intermediate productivity and disturbance (Grime 1973;Huston 2014;Bergauer et al 2022). In addition, temporal patterns in species richness may be biased, as such measures are prone to observer differences (Archaux et al 2009;Burg et al 2015), and it has been shown that species lists of vegetation plots that were compiled by different observers usually differ from each other to a certain degree (Kapfer et al 2017;Boch et al 2022). Further, it has been shown that prior experience and training in conducting vegetation surveys can particularly affect species numbers (because of a greater ability to detect and identify species) and cover estimates over time (reviewed by Morrison 2016).…”