Degrees of compositional shift in tree communities vary along a gradient of temperature change rates over one decade: Application of an individual-based temporal beta-diversity concept

Abstract

Temporal patterns in communities have gained much attention recently. However, the persistence of identical individuals and turnover among other individuals within the same species over time have not been considered. To deal with this issue, I propose new index concepts for beta-diversity and the relative speed of compositional equilibrium in relation to individual turnover based on individual identity information. I applied this new approach to individually tracked tree monitoring data in deciduous and evergreen broad-leaved forests across the Japanese archipelago. A new index explored the relative contributions of mortality and recruitment processes. Clear patterns emerged showing that an increase in the temperature change rate facilitated the relative contribution of mortality components. The relative speed of compositional equilibrium decreased with increasing temperature change rates in deciduous forests and increased with increasing warming rates in evergreen forests. The new concepts provide a way to identify novel and high-resolution temporal patterns in communities. Data accessibility statement All the data used in the present study were obtained from open sources. The forest monitoring data were extracted from the website of the Biodiversity Center of Japan (http://www.biodic.go.jp/moni1000/index.html, in Japanese: accessed on November 11, 2019) published in part by Ishihara et al. (2011). Climate data are from the Japanese Automated Meteorological Data Acquisition System (AMeDAS) database (http://www.jma.go.jp/jma/indexe.html, Japan Meteorological Agency accessed December 12, 2019). Taxonomic information was obtained from the Ylist (Yonekura & Kajita 2003: http://ylist.info/ylist_simple_sea rch.html, in Japanese: accessed December 8, 2019).

Individual-based multiple-unit dissimilarity: novel indices and null model for assessing temporal variability in community composition

“…Two types of dissimilarity indices have been proposed in previous studies: incidence-based and abundance-based dissimilarity indices (Baselga 2013b; Nakadai, 2020). Specifically, the widely used Bray–Curtis dissimilarity index is an abundance-based extension of the Sørensen index (Legendre and Legendre 2012; Baselga 2013b).…”

“…Both z o1 and z o2 take a value only of 1 or 0 (presence and absence, respectively). Therefore, P is the total number of individuals present at both T1 and T2, whereas M and R are the numbers of individuals unique to T1 and T2, respectively (Nakadai 2020).…”

“…In some cases, there is turnover of different individuals belonging to the same species even when the community composition is stable over time, which contributes to a dynamic compositional equilibrium across time periods. Therefore, the components that change over time ( M and R ) are re-organized into two further components: compositional equilibrium ( E ) and shift ( B and C ), respectively (Nakadai 2020; Fig. 1).…”

“…To address this issue, Nakadai (2020) proposed the “individual-based beta-diversity” concept, and provided novel indices to evaluate individual turnover and compositional shift by comparing individual turnover between two periods at a given site. Specifically, I developed an individual-based temporal beta-diversity approach for the Bray–Curtis dissimilarity index (i.e., an abundance-based temporal beta-diversity index) (Nakadai 2020). Over time, some individuals replace others, and therefore ecological communities are dynamic and vary to some degree both spatially and temporally (Mori et al 2018; Tatsumi et al 2019).…”

“…The persistence and turnover of individuals over time is a unique feature of temporal beta-diversity. Nakadai (2020) introduced the “individual-based beta-diversity” concept, and provided novel indices to evaluate individual turnover and compositional shift by comparing individual turnover between two periods at a given site. However, the proposed individual-based indices are applicable only to pairwise dissimilarity, not to multiple-temporal (or more generally, multiple-unit) dissimilarity.…”