Methods
Site description
Stem, soil and groundwater samples were collected at Elsey National
Park, a site located in northern Australia, about 400km southeast of
Darwin (15.00°S, 133.19°E). This region experiences a wet-dry tropical
climate (Aw according to the Köppen-Geiger classification), with
an average annual rainfall of 1017 mm for the period 2004–2022 (Bureau
of Meteorology’s Cave Creek Station #14650), >95% of
which falls between November and April. Elsey National Park is an area
of regional groundwater discharge from an extensive limestone aquifer
(Jolly et al., 2004; Karp, 2008; Lamontagne et al., 2021). The park
supports a diverse array of vegetation types along a groundwater depth
gradient, ranging from vine forests in areas with shallow groundwater
(i.e. \($\approx$\)1 m below ground level) to woodland savannas in
areas where the water table is deeper (i.e. >10 m below
ground level).
We conducted two sampling campaigns in October 2021 and October 2022,
coinciding with the later part of the dry season. During that period,
water availability is at a minimum and we expected some trees to rely
entirely on groundwater, whereas others without access to groundwater
would likely experience water stress as soil moisture stores become
depleted (Canham et al., 2021; Duvert et al., 2022). Because our aim was
to capture a range of tree species and water stress conditions, we
sampled trees across several sites covering varying habitats and depths
to water table. We sampled common tree species from seasonally dry
woodlands (E. tetrodonta, E. chlorostachys), to riparian
species (M. argentea, M. dealbata), as well as species
tolerant of both dry conditions and short-term seasonal inundation
(T. arostrata, H. arborescens). For the purposes of this
study, we sampled a total of seven tree species and 17 individual trees
(Table 1).
Stem sampling and leaf water potential
measurements
Originally, our plan was to sample a total of 40 stems from nine tree
species, including two replicate trees per species and two replicate
stems per individual tree – except for Hakea arborescens, for
which we aimed to increase the number of replicate trees to eight. This
specific focus on H. arborescens was motivated by preliminary
leaf water potential data, which showed a considerable variability among
trees of this species.
We encountered various challenges associated with the Cavitron
extraction procedure, including difficulties in sampling stem segments
that were straight enough to be secured in the rotor, and an inability
for some stems to extract sufficient volumes of xylem water for isotope
analysis. As a result, we were only able to analyse a total of 18 stems
from seven tree species. This still permitted the sampling of one to two
replicate trees per species and one to two replicate stem segments per
individual tree – except for H. arborescens, for which we
sampled eight stems from seven individual trees (Table 1).