Introduction
Ecosystems all over the world are affected by climate change (Parmesan,
2006; Pecl et al., 2017). High latitude and altitude ecosystems are
especially exposed because of arctic amplification and
elevation-dependent warming (Pepin et al., 2015; Serreze & Barry,
2011). In temperate and boreal regions, one of the most striking
ecological transitions in high elevation areas is the change from boreal
lowland forests, to alpine highlands without forests (Fig. 1a). This
marked ecotone, termed the forest line (here) or treeline, varies in
altitude and species composition across the globe, and has in many
regions moved towards higher elevations during the last decades (Harsch,
Hulme, McGlone, & Duncan, 2009). Although many mountain regions are
influenced by land use, the uppermost alpine forest lines are mainly
temperature driven (Körner, 2012). Hence, as a response to climate
warming, changes in vegetation are predicted, including an upward shift
of the forest line into mountain ecosystems (Beckage et al., 2008; Chen,
Hill, Ohlemuller, Roy, & Thomas, 2011). Recent studies have documented
positive climate feedbacks from expanding alpine forest lines (de Wit et
al., 2014) and an increase in the elevation of alpine forest lines will
probably increase the local temperature and therefore accelerate the
ongoing expansion further (Rydsaa, Stordal, Bryn, & Tallaksen, 2017).
Large areas in Fennoscandia are covered by mountain birch (Betula
pubescens ssp. czerepanovii ) forest, that also forms the
alpine forest lines in most areas in this region (Bryn & Potthoff,
2018). At the forest-alpine ecotone, the vegetation is normally shrub
dominated in a transition zone, before shifting to vegetation dominated
by ericaceous plant species, commonly referred to as low-alpine
vegetation (Fig. 1b). In Norway, an upward shift in forest lines has
been observed, both due to climate warming and land use change (Bryn &
Potthoff, 2018). The expansion of subalpine trees and shrubs leads to
increased primary production, and thus aboveground carbon (C) fixation.
However, studies from alpine and arctic regions have shown that soil
organic C content is significantly higher under alpine heaths than under
shrub and forest vegetation (Parker, Subke, & Wookey, 2015; Sorensen et
al., 2018). Parker et al. (Parker et al., 2015) showed that soil
C turnover was faster and belowground C pools smaller in the forest and
shrub vegetation compared to arctic heath. Similarly, Sørensen et
al. (Sorensen et al., 2018) showed that C stocks were lowest beneath
shrub vegetation and significantly higher beneath ericaceous plants in
the alpine heath. Furthermore, it is documented that litter decomposes
faster in the shrub and mountain birch forest vegetation than in the
alpine heath, independent of the litters resistance to decomposition
(Parker et al., 2015). These differences in soil C pools may largely be
regulated by belowground microorganisms, including fungi, similar to in
lowland systems (Clemmensen et al., 2013; Frey, 2019). Improved
knowledge of these communities and their functional roles are important
for a deeper understanding of the C pool dynamics across the forest line
ecotone.
Belowground fungi contribute to soil C processes in various ways, and
can roughly be grouped into parasites, mutualistic symbionts and
saprotrophs, although there is a blurry transition between nutritional
strategies (Selosse, Schneider-Maunoury, & Martos, 2018). Saprotrophic
fungi decompose and recycle dead organic matter, thus releasing C to the
atmosphere. Mycorrhizal fungi form mutualistic symbiosis with plant
roots, where they receive freshly fixed C from their plant host in
exchange for water and nutrients. Saprotrophic and mycorrhizal fungi are
thus crucial components of the global C cycle, as well as the nitrogen
(N) and phosphorus (P) cycles (Lindahl & Tunlid, 2015; Smith & Read,
2008; Talbot, Allison, & Treseder, 2008). Mycorrhizal fungi can be
divided into functional groups depending on their plant hosts, structure
and function (Smith & Read, 2008). The three main groups are arbuscular
mycorrhizal (AM), ectomycorrhizal (EcM), and ericoid mycorrhizal (ErM)
fungi, which establish symbiosis with mainly herbs and graminoids (AM),
trees and shrubs (EcM), and ericoid plants (ErM), respectively (Smith &
Read, 2008). EcM fungi, which dominate in northern forest ecosystems,
can promote turnover of organic matter in some systems (Bodeker et al.,
2014; Frey, 2019; Lindahl & Tunlid, 2015; Talbot et al., 2008) or C
sequestration in others (Averill, Turner, & Finzi, 2014; Koide,
Fernandez, & Malcolm, 2014; Orwin, Kirschbaum, St John, & Dickie,
2011). Root-associated dark septate endophytes (DSE) are also common in
alpine and arctic vegetation and often found in the same environments as
ErM fungi (Newsham, Upson, & Read, 2009; Olsrud, Michelsen, &
Wallander, 2007). It has recently been shown that DSE can promote
nutrient uptake in plants (Hill et al., 2019). Taxonomically, it can be
hard to separate between fungi forming DSE and ErM since both groups are
dominated by ascomycetes, especially Leotiomycetes. Both DSEs and ErM
fungi have melanized hyphae resistant to decomposition, which has been
proposed to play a central role in soil C sequestration (Clemmensen et
al., 2015; Fernandez & Koide, 2013). Above the forest line, ericoid
dwarf shrubs dominate the alpine heath. Thus, as the forest migrates
upwards, a shift in belowground dominating functional groups is
expected, potentially shifting current soil C dynamics. Other
micro-eukaryotes, including invertebrates and protists, are also
essential members of the soil biota and the soil food web (Geisen et
al., 2018; Phillips et al., 2019), although their contribution to
belowground C processes is not well studied.
To improve our understanding of the belowground biota across the
mountain birch forest line ectone, we conducted a survey designed to
measure belowground compositional changes across the ecotone in nine
sites spread across southern Norway. Using DNA metabarcoding, we
targeted the ITS2 (fungi only) and 18S (all eukaryotes) regions of
ribosomal DNA from soil, and we measured ergosterol (a proxy for fungal
biomass) to obtain both qualitative and quantitative (fungi only)
information about the belowground biota. We aimed to couple these
results with both abiotic (soil edaphic factors, climate and bedrock)
and biotic factors (aboveground vegetation), to investigate the
relationships between above- and belowground changes across the ecotone.
More specifically, we wanted to investigate whether: (i) there is a
strong compositional change in the soil biota, both in terms of
taxonomic and functional groups, across the forest line ecotone; (ii)
the changes in taxonomic and functional groups of soil fungi are
strongly correlated with changes in aboveground vegetation, and (iii)
more fungal biomass is found above the forest line, which may be due to
the presence of more ErM/DSE fungi with recalcitrant mycelium. To assess
the generality of our observations, we analysed soil samples obtained
along ecotones in nine replicated sites spread across southern Norway.