Perspectives in Plant Ecology, Evolution and Systematics 30 (2018) 89–102 Contents lists available at ScienceDirect Perspectives in Plant Ecology, Evolution and Systematics journal homepage: www.elsevier.com/locate/ppees Research article Time lag between glacial retreat and upward migration alters tropical alpine ☆ T communities Anaïs Zimmera,b,c, Rosa I. Menesesb, Antoine Rabateld, Alvaro Sorucoe, Olivier Danglesf,g,h, ⁎ Fabien Anthelmea,b,c, a AMAP, IRD, CNRS, INRA, Université de Montpellier, Montpellier, France b Museo Nacional de Historia Natural, Herbario Nacional de Bolivia (LPB), Cota Cota, Casilla 10077 Correo Central, La Paz, Bolivia c Instituto de Ecología, Universidad Mayor San Andrés, Calle 27, Cota Cota, Campus Universitario, La Paz, Bolivia d Université Grenoble Alpes, CNRS, IRD, Institut des Géosciences de l’Environnement (IGE, UMR 5001), F-38000 Grenoble, France e Instituto de Geología y del Medio Ambiente, Universidad Mayor San Andrés, Calle 27, Cota Cota, Campus Universitario, La Paz, Bolivia f Institut de Recherche pour le Développement (IRD), EGCE, 91198 Gif-sur-Yvette Cedex, France g Université Paris-Sud 11, 91405 Orsay Cedex, France h Pontificia Universidad Católica del Ecuador, Facultad de Ciencias Exactas y Naturales, Quito, Ecuador ARTICLE INFO ABSTRACT Keywords: Species range shifts and possible species extinctions in alpine regions are hypothesized being influenced by the Biological soil crust increasing time lag between the velocity of global warming and the slowness of primary succession. We tested this Climatic debt hypothesis in tropical alpine environments above 4700 m a.s.l. (Central Andes) and we explored the underlying Chronosequence mechanisms at work by using four sites gradually deglaciated since the acceleration of warming in the late 1970’s. Nurse plant These post-glacial chronosequences, made available by a multidisciplinary approach combining glaciology and Species range shifts ecology, are extremely rare and provide a pertinent space-for-time substitution for the study of climate change Tropical andes effects. We found consistent patterns in plant succession (abundance, species richness and functional strategies) along the four chronosequences. Dispersal limitation was a prominent constraint for succession, even at the end of the chronosequences, leading to an overrepresentation of anemochorous species in comparison with adjacent ecosystems. Nurse plants were infrequent and their low maturity seemed to make them poorly efficient as facil- itators, contrarily to the expectations made by the stress-gradient hypothesis in alpine regions. This suggests that, despite the accelerating rate of warming, the dynamics of primary succession remains slow, generating a climatic debt and hampering the adaptation to climate change in alpine plant communities. 1. Introduction warming (Gottfried et al., 2012; Lenoir and Svenning, 2015; Harsch and HilleRisLambers, 2016). Located at the upper limit of life, alpine species Under the pervasive effects of climate warming on ecosystems, are particularly sensitive to these range shifts because new, upward characterizing the biodiversity-climate change relationship has become habitats are often devoid of life and soil, or they are simply absent. The a major scientific challenge (Lavergne et al., 2010; Chen et al., 2011; migration lag experienced by alpine plants, i.e. the time spent between Urban, 2015). Especially, a detailed description of patterns and me- a climatic fluctuation and the moment when plants effectively reach the chanisms of vegetation succession is required to identify the ecological new site, has been shown to cause a significant reduction in their spatial processes that will shape future biodiversity (Walker and Wardle, distribution in a study centred on the Holocene scale (Dullinger et al., 2014). In mountain systems, despite the local availability of biotic and 2012). Given the high magnitude and velocity of current and predicted abiotic refuges in heterogeneous areas (Scherrer and Körner, 2011; climate change, it is expected that this migration lag might affect ne- Anthelme et al., 2014a), the majority of organisms need to perform gatively the altitudinal distribution of alpine species (Svenning and upward range shifts to adapt to the direct and indirect effects of Sandel, 2013). A key issue is to know whether plant migration lag can Abbreviations: CSR, competitor, stress-tolerant, ruderal; BSC, biological soil crust; LIA, little ice age ☆ This article is part of a special issue entitled Alpine and arctic plant communities: a worldwide perspective published at the journal Perspectives in Plant Ecology, Evolution and Systematics 30C. ⁎ Corresponding author at: UMR AMAP/DIADE, Boulevard de la Lironde, TA A-51/PS2, F-34398 Montpellier Cedex 5, France. E-mail addresses: [email protected] (A. Zimmer), [email protected] (R.I. Meneses), [email protected] (A. Rabatel), [email protected] (A. Soruco), [email protected] (O. Dangles), [email protected] (F. Anthelme). http://dx.doi.org/10.1016/j.ppees.2017.05.003 Received 11 October 2016; Received in revised form 24 May 2017; Accepted 30 May 2017 Available online 13 June 2017 1433-8319/ © 2017 Elsevier GmbH. All rights reserved. A. Zimmer et al. Perspectives in Plant Ecology, Evolution and Systematics 30 (2018) 89–102 result in a climatic debt for natural communities that would be no nurse plants, thus indirectly reducing plant diversity through cascade longer in equilibrium with climate (Devictor et al., 2012; Svenning and effects (sensu Malatesta et al., 2016)? And (2) would the short time Sandel, 2013). available for the development of alpine nurse plants in recently de- Elevation has been shown positively correlated with the intensity of glaciated sites impact negatively plant communities because of limited warming in mountainous areas, with tropical alpine ecosystems being ontogenic variations between the nurse and the beneficiary (Anthelme among the most heavily affected by global warming because of their and Dangles, 2012)? high elevation, generally above 4000 m a.s.l. (Bradley et al., 2006; MRI, More generally, how the various pioneer organisms interact early 2015). This is observable with the tropical Andean glaciers, which have after glacial retreat is poorly known and requires further investigation experienced a more pronounced shrinking than other glaciers world- (Matthews and Vater, 2015; Erschbamer and Caccianiga, 2016). Among wide since the late 1970’s(Rabatel et al., 2013). The impacts of these these organisms, biological soil crust (BSC), an association of bacteria, changes on biodiversity seem to be rapid and severe, as demonstrated lichens, algae, mosses and fungi, has been proposed to be another driver with aquatic and terrestrial invertebrates (Jacobsen et al., 2012; Cauvy- of primary succession after glacial retreat, through facilitative interac- Fraunié et al., 2016; Moret et al., 2016). They are in line with the recent tions with other plants (Türk and Gärtner, 2001; Breen and Lévesque, evidence that the biodiversity of South America −the continent that 2008; Matthews and Vater, 2015). Can they ensure successful primary shelters more than 90% of tropical alpine regions worldwide- is the succession under a rapidly changing climate irrespective of nurse plant most affected by changing climate with 23% of its species at risk vs. effects? 7.9% worldwide, under an optimistic warming scenario (Urban, 2015). By characterizing the abiotic environment, the species diversity, Tropical alpine environments are also impacted by climate warming at plant–plant and plant-rock spatial associations along primary succes- ecosystem level, through glacier shrinking, which has been shown to sion in four sites in the tropical Andes gradually deglaciated since the modify water input and increase ecosystem fragmentation (Dangles late 1970’s (post-glacial chronosequences) we aimed at testing these et al., 2017) and by increasing plant growth and organic carbon pro- specific hypotheses and their related research questions. Our results duction (Cooper et al., 2015). All in all, this makes tropical Andean were discussed into the recent conceptual framework on species range ecosystems flagship descriptors for the characterization of the biodi- shifts consecutive to global changes (Lenoir and Svenning, 2015), ex- versity-climate change interactions in a changing world. However, al- ploring future scenarios for the tropical alpine biodiversity. though secondary succession in the alpine tropics has received con- siderable attention (e.g. Sarmiento et al., 2003; Bueno and Llambí, 2. Material and methods 2015), primary succession has been overlooked, so far, especially its characterization in a changing world (but see Suárez et al., 2015). 2.1. Study area and study sites Our first specific hypothesis is that the unrivalled velocity of warming in tropical alpine regions may exacerbate the dispersal filter At elevations above 500 m a.s.l., the tropical Andes extend over 1.5 with new species assemblages being even more dominated by ane- millions km2. They are one of the most important biodiversity hotspots mochorous species than what has been observed along longer post- worldwide; by, with a high proportion of endemic plants (Anthelme glacial chronosequences, so far (Stöcklin and Bäumler, 1996; et al., 2014b). Tropical alpine regions in the Andes range between Erschbamer and Caccianiga, 2016; Marta et al., 2016). Indeed, wind is 3200 m a.s.l. and more than 5000 m a.s.l., even