Samples collection and preservation
In Summer 2018, we collected 280 soil samples from five Alpine forelands
(Fig. 1): Amola (coordinates of the center of the foreland: N 46.215° E
10.697°), Morteratsch (N 46.438° E 9.936°), Rutor (N 46.669° E 6.992°),
Sforzellina (N 46.351° E 10.510°) and Grande di Verra (N 45.895° E
7.749°). For each foreland, we selected three to eight sites for which
the date of glacier retreat is known on the basis of the literature,
dated images and field surveys, focusing on the period between the end
of the Little Ice Age (~1850) and recent years (Marta et
al., 2021). Soil samples were representative of different stages of soil
development depending on the time elapsed between the retreat of
glaciers and sampling activities (hereafter referred to as “time since
glacier retreat”; ranging from 12 to 168 years).
At each site, we established five regularly spaced plots at distances of
about 20 m. At each plot, we collected five soil cores within one-meter
distance and we kept the 0-5 cm and 7.5-20 cm portions to be
representative of two different soil depths, hereafter called
“surface” and “deep” soils, respectively (Fig. 1). For each of the
five cores, we pooled portions of the same depth together to form one
composite sample of ~200 g and we homogenized it. We
took 15 g of soil from each composite sample and desiccated it
immediately in sterile boxes with 40 g of silica gel. Previous analyses
showed that this approach enables a cost-effective and long-term
preservation of soil eDNA (Guerrieri et al., 2021). Soil eDNA collection
was performed wearing gloves and the sampling tools were decontaminated
with a portable blow torch (>1000°C) before the collection
of each sample. We did not include soil litter and avoided roots, leaves
and other large plant organs.
Soils were sampled “by depth” rather than “by horizons”, as is
common practice in eDNA-based studies (Dickie et al., 2018) and in soil
monitoring networks involving multiple glacier forelands (e.g., Khedim
et al., 2021; Orgiazzi, Ballabio, Panagos, Jones, & Fernández-Ugalde,
2018; Rime et al., 2015; Schweizer, Hoeschen, Schlüter, Kögel-Knabner,
& Mueller, 2018) because soil horizons are not yet differentiated in
early stages of soil development, and because this approach allows
obtaining a standardized pattern that can be applied across soils from
multiple areas at very different development stages (Dickie et al.,
2018; Khedim et al., 2021; Rime et al., 2015). Thus, the two categories
“surface” (0-5 cm) and “deep” (7.5-20 cm) are used to define soil
samples collected at two different soil depths, regardless of the
horizons.