Accounting for tree line shift, glacier retreat and primary succession in mountain plant distribution models. Issue 12 (11th September 2014)

• Record Type: Journal Article
• Authors: Carlson, Bradley Z.; Georges, Damien; Rabatel, Antoine; Randin, Christophe F.; Renaud, Julien; Delestrade, Anne; Zimmermann, Niklaus E.; Choler, Philippe; Thuiller, Wilfried; Rouget, Mathieu
• Is Part Of: Diversity & distributions. Volume 20:Issue 12(2014:Dec.)
• Page Start: 1379
• Page End: 1391
• Publication Date: 2014-09-11
• Subjects: Biodiversity -- Periodicals; Biodiversity conservation -- Periodicals; 577
• DOI: 10.1111/ddi.12238
• Languages: English
• ISSNs: 1366-9516

Abstract

<abstract abstract-type="main" id="ddi12238-abs-0001"> <title>Abstract</title> <sec id="ddi12238-sec-0001" sec-type="section"> <title>Aim</title> <p>To incorporate changes in alpine land cover (tree line shift, glacier retreat and primary succession) into species distribution model (SDM) predictions for a selection of 31 high‐elevation plants.</p> </sec> <sec id="ddi12238-sec-0002" sec-type="section"> <title>Location</title> <p>Chamonix Valley, French Alps.</p> </sec> <sec id="ddi12238-sec-0003" sec-type="section"> <title>Methods</title> <p>We fit linear mixed effects (LME) models to historical changes in forest and glacier cover and projected these trends forward to align with 21st century IPCC climate scenarios. We used a logistic function to model the probability of plant establishment in glacial forelands zones expected to become ice free between 2008 and 2051–2080. Habitat filtering consisted of intersecting land cover maps with climate‐driven SDMs to refine habitat suitability predictions. SDM outputs for tree, heath and alpine species were compared based on whether habitat filtering during the prediction period was carried out using present‐day (static) land cover, future (dynamic) land cover filters or no land cover filter (unfiltered). Species range change (SRC) was used to measure differences in habitat suitability predictions across methods.</p> </sec> <sec id="ddi12238-sec-0004" sec-type="section"> <title>Results</title> <p>LME predictions for 2021–2080 showed continued glacier retreat, tree line rise and primary succession in glacier forelands. SRC was highest in the unfiltered scenario (−10%), intermediate in the dynamic scenario (−15%) and lowest in the static scenario (−31%). Tree species were the only group predicted to gain overall range by 2051–2080. Although alpine plants lost range in all three land cover scenarios, new habitat made available by glacier retreat in the dynamic land cover scenario buffered alpine plant range loss due to climate change.</p> </sec> <sec id="ddi12238-sec-0005" sec-type="section"> <title>Main conclusions</title> <p>We provide a framework for combining trajectories of land cover change with SDM predictions. Our pilot study shows that incorporating shifts in land cover improves habitat suitability predictions and leads to contrasting outcomes of future mountain plant distribution. Alpine plants in particular may lose less suitable habitat than standard SDMs predict due to 21st century glacier retreat.</p> </sec> </abstract>