Remote sensing of β-diversity: Evidence from plant communities in a semi-natural system Samuel Hoffmann, Thomas Schmitt, Alessandro Chiarucci, Severin Irl, Duccio Rocchini, Ole Vetaas, Mihai Tanase, Stéphane Mermoz, Alexandre Bouvet, Carl Beierkuhnlein To cite this version: Samuel Hoffmann, Thomas Schmitt, Alessandro Chiarucci, Severin Irl, Duccio Rocchini, et al.. Remote sensing of β-diversity: Evidence from plant communities in a semi-natural system. Applied Vegetation Science, Wiley, 2018, 22, pp.13 - 26. 10.1111/avsc.12403. hal-03272334 HAL Id: hal-03272334 https://hal.archives-ouvertes.fr/hal-03272334 Submitted on 28 Jun 2021 HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non, lished or not. The documents may come from émanant des établissements d’enseignement et de teaching and research institutions in France or recherche français ou étrangers, des laboratoires abroad, or from public or private research centers. publics ou privés. Received: 9 November 2017 | Revised: 7 August 2018 | Accepted: 6 September 2018 DOI: 10.1111/avsc.12403 RESEARCH ARTICLE Applied Vegetation Science Remote sensing of β- diversity: Evidence from plant communities in a semi- natural system Samuel Hoffmann1 | Thomas M. Schmitt1 | Alessandro Chiarucci2 | Severin D. H. Irl1 | Duccio Rocchini3,4,5 | Ole R. Vetaas6 | Mihai A. Tanase7,8 | Stéphane Mermoz7 | Alexandre Bouvet7 | Carl Beierkuhnlein1 1Department of Biogeography, BayCEER, University of Bayreuth, Bayreuth, Abstract Germany Question: Do remote sensing signals represent β- diversity? Does β- diversity agree 2 Department of Biological, Geological, with community types? and Environmental Sciences, Alma Mater Studiorum, University of Bologna, Bologna, Location: UNESCO Man and the Biosphere Reserve, La Palma, Canary Islands. Italy Methods: We recorded perennial, vascular plant species abundances in 69 plots 3Center Agriculture Food (10 m × 10 m) in three pre- defined community types along an elevational gradient of Environment, University of Trento, S. Michele all’Adige (TN), Italy 2,400 m: succulent scrubland, Pinus canariensis forest and sub alpine scrubland. The 4Centre for Integrative Biology, University of remote sensing data consists of structural variables from airborne Light Detection Trento, Povo (TN), Italy and Ranging (LiDAR) and multispectral variables from a time series of Sentinel- 2 (S2) 5Department of Biodiversity and Molecular Ecology, Fondazione Edmund images. Non- metric Multidimensional Scaling was used to assess β- diversity between Mach, Research and Innovation Centre, S. plots. K- means unsupervised clustering was applied to remote sensing variables to Michele all’Adige (TN), Italy distinguish three community types. We subsequently quantified the explanatory 6Department of Geography, University of Bergen, Bergen, Norway power of S2 and LiDAR variables representing β- diversity via the Mantel test, varia- 7Center for the Study of the Biosphere from tion partitioning and multivariate analysis of variance. We also investigated the sen- Space (CESBIO), Toulouse, France sitivity of results to grain size of remote sensing data (20, 40, 60 m). 8Department of Geology, Geography and Environment, University of Alcala, Alcalá de Results: The β- diversity between the succulent and pine community is high, whereas Henares, Spain the β- diversity between the pine and sub alpine community is low. In the wet season, Correspondence up to 85% of β- diversity is reflected by remote sensing variables. The S2 variables Samuel Hoffmann, Department of account for more explanatory power than the LiDAR variables. The explanatory Biogeography, BayCEER, University of Bayreuth, Bayreuth, Germany. power of LiDAR variables increases with grain size, whereas the explanatory power Email: [email protected] of S2 variables decreases. Funding information Conclusion: At the lower ecotone, β- diversity agrees with the pre- defined commu- This study was funded by the ECOPOTENTIAL project -­ EU Horizon 2020 nity distinction, while at the upper ecotone the community types cannot be clearly research and innovation programme, grant separated by compositional dissimilarity alone. The high β- diversity between the suc- agreement No. 641762. culent scrub and pine forest results from positive feedback switches of P. canariensis, Co-ordinating Editor: Hannes Feilhauer being a fire- adapted, key tree species. In accordance with the spectral variation hy- pothesis, remote sensing signals can adequately represent β- diversity for a large ex- tent, in a short time and at low cost. However, in- situ sampling is necessary to fully understand community composition. Nature conservation requires such interdiscipli- nary approaches. Nomenclature: Muer, Sauerbier, and Calixto (2016) Appl Veg Sci. 2018;1–14. wileyonlinelibrary.com/journal/avsc © 2018 International Association | 1 for Vegetation Science 2 HOFFMANN ET AL. | Applied Vegetation Science KEYWORDS β-diversity, conservation biogeography, elevation gradient, island biogeography, LiDAR, plant community, remote sensing, Sentinel, spectral variation hypothesis, time series, tree line, vegetation indices 1 | INTRODUCTION Chiarucci, 2005). This application rests on the spectral variation hy- pothesis (SVH) explaining the relationship between environmental The spatial and temporal change rates of species composition, i.e. heterogeneity, species diversity and spectral information (Palmer, β- diversity, have been at the heart of community ecology ever Earls, Hoagland, White, & Wohlgemuth, 2002). Environmental since Clements (1916). However, the community definition is still heterogeneity increases habitat heterogeneity and, thus, species largely debated (Chiarucci, 2007; Palmer & White, 1994; Ricklefs, diversity (i.e. habitat heterogeneity hypotheses; Simpson, 1949). 2008). The controversy revolves around the coherence and integ- Environmental heterogeneity also increases spectral heterogeneity. rity of ecological entities through different scales of space and time Therefore, spectral variation is associated with α-­ and β- diversity (Jax, 2006). In order to assess community patterns, concepts of β- (Palmer et al., 2002; Rocchini, Chiarucci, & Loiselle, 2004). However, diversity are applied that quantify the compositional dissimilarity the SVH does not apply to all ecosystems and depends on the extent between species assemblages (Anderson et al., 2011). of RS and in- situ data as well as the spatial, temporal and spectral Processes responsible for observed patterns of species co- resolution of RS data (Schmidtlein & Fassnacht, 2017). existence, usually referred to as “assembly rules”, can be determinis- This study relates to the SVH, because we investigate to what tic, stochastic, interrelated and contingent, which led Lawton (1999) degree RS signals of species assemblages can explain β- diversity, i.e. to call community ecology “a mess”. Vellend (2010) proposed the the compositional dissimilarity between species assemblages. As a following overarching processes shaping β- diversity and community case study, we sampled the semi- natural plant communities along a patterns: selection, drift, speciation and dispersal. These factors and continuous elevational gradient on La Palma, Canary Islands. First, anthropogenic activities determine β- diversity and, thus, biodiver- we test the SVH using structural RS variables from light detection sity in general (Socolar, Gilroy, Kunin, & Edwards, 2016), on which and ranging (LiDAR) and multispectral variables from a time series of human well- being depends (Cardinale et al., 2012). It is therefore Sentinel- 2 images (S2). Since RS sensors can barely account for small, important to study patterns of β- diversity as well as corresponding rare and understorey species, we expect that RS signals cannot ade- drivers. quately explain β- diversity that is derived from in- situ observations. The existence of communities implies the delineation of commu- This combination of data and techniques has not been used before to nity types. Because natural boundary sharpness varies (Auerbach & represent β- diversity with RS products. Second, we analyse to what Shmida, 1993; Wilson & Agnew, 1992), community distinction is not extent β- diversity agrees with the pre- defined community types. necessarily discrete; transition between communities can be rather continuous. This is why community limits are specifically considered 2 | METHODS as transition zones, also known as ecotones (Livingston, 1903). In early times, an ecotone was associated with a clear separation of 2.1 | Study region plant physiognomy (Clements, 1905). The recent definition of eco- tone by Lloyd, McQueen, and Lee (2000) is based on β- diversity and The island of La Palma is located at the northwest edge of the describes it as a “zone where directional change in vegetation (i.e. Canary archipelago in the Atlantic Ocean, ca. 400 km west of the qualitative and quantitative species composition) is more rapid than African coast at 28°N (Figure 1). The entire island is designated a on the other side of the zone.” Although ecotones are a standard ‘UNESCO Man and the Biosphere Reserve’. La Palma is generally entity in landscape ecology (Wiens, Crawford, Gosz, Crawford, & characterized by a subtropical-mediterranean climate. However, the Boundary, 1985),