Hydrol. Earth Syst. Sci., 19, 1107–1123, 2015
www.hydrol-earth-syst-sci.net/19/1107/2015/
doi:10.5194/hess-19-1107-2015
© Author(s) 2015. CC Attribution 3.0 License.
A groundwater recharge perspective on locating tree plantations
within low-rainfall catchments to limit water resource losses
J. F. Dean1,2,*, J. A. Webb1,2, G. E. Jacobsen3, R. Chisari3, and P. E. Dresel4
1Agricultural Sciences Department, La Trobe University, Bundoora, Victoria, Australia
2National Centre for Groundwater Research and Training, Adelaide, Australia
3Institute for Environmental Research, ANSTO, Sydney, Australia
4Department of Environment and Primary Industries, Bendigo, Victoria, Australia
*now at: Biological and Environmental Sciences, University of Stirling, Scotland, UK
Correspondence to: J. F. Dean (joshua.dean@sitr.ac.uk)
Received: 8 July 2014 – Published in Hydrol. Earth Syst. Sci. Discuss.: 28 August 2014
Revised: 22 January 2015 – Accepted: 7 February 2015 – Published: 26 February 2015
Abstract. Despite the many studies that consider the im- ing should be avoided in the dominant zone of recharge, i.e.
pacts of plantation forestry on groundwater recharge, and the topographically low areas and along the drainage lines,
others that explore the spatial heterogeneity of recharge in and should be concentrated on the upper slopes, although this
low-rainfall regions, there is little marriage of the two sub- may negatively impact the economic viability of the planta-
jects in forestry management guidelines and legislation. Here tion.
we carry out an in-depth analysis of the impact of reforesta-
tion on groundwater recharge in a low-rainfall (< 700 mm an-
nually), high-evapotranspiration paired catchment character-
ized by ephemeral streams. Water table fluctuation (WTF) 1 Introduction
estimates of modern recharge indicate that little groundwa-
ter recharge occurs along the topographic highs of the catch- Tree plantations are known to have the potential to reduce
ments (average 18 mm yr−1); instead the steeper slopes in groundwater recharge and surface water flows (e.g. Bell et
these areas direct runoff downslope to the lowland areas, al., 1990; Benyon, 2002; Bosch and Hewlett, 1982; Jobbagy
where most recharge occurs (average 78 mm yr−1). Recharge and Jackson, 2004; Scanlon et al., 2007; van Dijk et al.,
estimates using the chloride mass balance (CMB) method 2007), particularly in low-rainfall, high-evapotranspiration
were corrected by replacing the rainfall input Cl− value with regions where the high transpiration demands of the trees
that for streamflow, because most recharge occurs from infil- make them a significant user in the water balance (e.g.
tration of runoff through the streambed and adjacent low gra- Benyon et al., 2006; Fekeima et al., 2010; Jackson et al.,
dient slopes. The calculated CMB recharge values (average 2005; Schofield, 1992). This is often regarded as a negative
10 mm yr−1) are lower than the WTF recharge values (aver- aspect of tree plantations, but may be a positive outcome if
age 47 mm yr−1), because they are representative of ground- the aim of a particular forestry project is to reduce ground-
water that was mostly recharged prior to European land clear- water levels, e.g. to decrease groundwater salinization (dis-
ance (> BP 200 years). The tree plantation has caused a pro- cussed further below). Groundwater recharge in low-rainfall
gressive drawdown in groundwater levels due to tree water regions is also affected by a variety of other factors that cause
use; the decline is less in the upland areas. substantial spatial variability – in particular topography, soil
The results of this study show that spatial variations in characteristics and geology (e.g. Delin et al., 2000; Scan-
recharge are important considerations for locating tree plan- lon et al., 2002; Schilling, 2009; Webb et al., 2008; Win-
tations. To conserve water resources for downstream users ter, 2001). However, the important conclusions made in the
in low-rainfall, high-evapotranspiration regions, tree plant- recharge studies have not been brought together with the re-
sults of tree plantation studies and directly applied to water
Published by Copernicus Publications on behalf of the European Geosciences Union.
1108 J. F. Dean et al.: Locating tree plantations from a groundwater recharge perspective
resource management problems accompanying the establish- tainable action plans for surface water and groundwater (Dal-
ment of tree plantations (Farley et al., 2005). haus et al., 2008; Jackson et al., 2005; Nicholson et al., 2006).
Since the earliest work on defining groundwater systems, A whole-catchment approach is key to managing ground-
recharge has been shown to be controlled predominantly by water recharge in the context of land use change (Cartwright
topography: the majority of groundwater recharge occurs at et al., 2007). However, despite the evidence that recharge is
topographic highs, and discharge is mostly in topographic often concentrated in topographic lows, groundwater man-
lows where the upward hydraulic gradient prevents recharge agement strategies in southeastern Australia typically operate
from occurring (Domenico and Schwartz, 1998; Schilling, on the assumption that recharge occurs primarily in the upper
2009). However, in arid and semi-arid regions, recharge fol- parts of catchments, particularly along the ridgelines. Current
lowing rainfall events often occurs predominantly in local regulations for tree plantations in Australia focus on the per-
depressions and along ephemeral streams (diverging from centage of a given catchment that can be forested, rather than
early conceptual models), due to the focusing of overland what areas should be planted to maintain or intercept ground-
flow in these areas. Water tables under ephemeral streams water recharge, depending on the management application.
are generally below the streambed (except during extended Here we present the findings from a paired catchment
rainfall events), and therefore upwards groundwater gradi- study in southwestern Victoria, Australia, where one catch-
ents do not occur most of the time. Infiltration beneath these ment is planted with a tree plantation, and the adjacent catch-
areas may also be encouraged by the presence of preferential ment is covered with pasture. This approach largely removes
pathways, along which infiltrating water may more readily the variables of climate, topography, soil and geology, with
reach the water table (Delin et al., 2000; Scanlon et al, 2002; the only major difference between the two catchments be-
Schilling, 2009; Winter, 2001). In southeastern Australia in ing vegetation cover. Previous paired catchment studies on
particular, it has been observed that recharge can vary signif- the impact of tree plantations tended to focus on surface wa-
icantly within catchments due to multiple modes of recharge ter responses to afforestation, while groundwater has been
(Cartwright et al., 2007). somewhat neglected (Brown et al., 2005). In this study con-
Vegetation can significantly impact groundwater recharge ceptual models of groundwater flow (based on 14C and tri-
due to transpiration and by intercepting rainfall and over- tium groundwater dating) and groundwater recharge esti-
land flow (Scanlon et al., 2002; Winter, 2001); changing mates (based on the water table fluctuation and chloride mass
land use can therefore affect recharge patterns. For exam- balance methods) are used to assess the impact of a Eucalyp-
ple, land salinization has occurred in large parts of south- tus globulus plantation on the hydrologic and hydrogeologic
eastern Australia due to the replacement of native forest by regime. This contextualization is then used to discuss the best
pasture and crops that use less water; this has led to increased areas to site tree plantations within low-rainfall catchments.
recharge which raised water tables, causing saline groundwa-
ter to come to the land surface and discharge into surface wa-
ter features (Allison et al., 1990; Bennetts et al., 2006, 2007). 2 Background
In contrast, afforestation of cleared farmland is likely to de-
This study is part of a multi-site, paired-catchment investiga-
crease recharge, due to the high rate of transpiration by the
tion into the impacts of land use and climate change on the
actively growing, closely planted trees, as well as the inter-
quality and quantity of groundwater and surface water re-
ception of overland flow and evaporation from the canopy
sources in western Victoria, Australia (Adelana et al., 2014;
(Benyon et al., 2006). In particular, the evergreen eucalyptus
Camporese et al., 2013, 2014; Dean et al., 2014; Dresel et al.,
tree plantations commonly planted in southeastern Australia
2012).
take up and transpire significantly more water than pasture,
their canopy intercepts more rainfall and allows it to evapo-
2.1 Site description
rate, and their roots reach greater depths than grasses, mean-
ing they can extract water over a larger volume of the soil The study area consists of a pair of small, adjacent catch-
column (Bosch and Hewlett, 1982; Feikema et al., 2010; Hi- ments at Mirranatwa in southwestern Victoria – one (referred
bbert, 1967). This recharge reduction is the reason why some to as the eucalypt catchment) is covered predominantly in a
studies have suggested using targeted tree plantations to re- recently planted (July 2008) E. globulus (Blue Gum) plan-
duce recharge in areas where there are high rates of saline tation (0.8 km2), the other (referred to as the pasture catch-
groundwater discharge (e.g. Bennetts et al., 2007). Tree plan- ment) is mostly pasture for grazing sheep (0.4 km2; Fig. 1).
tations also sequester carbon dioxide, prompting ongoing de-
bate over the trade-off between increased water use by trees 2.1.1 Geology
versus their increased carbon sequestration potential (Farley
et al., 2005). As such, efforts over the past few decades in Both catchments are underlain by the same weath-
southeastern Australia to reforest land that was cleared in the ered/fractured aquifer, the Devonian Dwyer Granite (390–
late 1800s by European settlers (Schofield, 1992) are now 395 Ma; Hergt et al., 2007; Van den Berg, 2009). The up-
causing difficulties for land managers trying to define sus- per ∼ 20 m of the granite is well-weathered, porous and
Hydrol. Earth Syst. Sci., 19, 1107–1123, 2015 www.hydrol-earth-syst-sci.net/19/1107/2015/
J. F. Dean et al.: Locating tree plantations from a groundwater recharge perspective 1109
Figure 1. Left: location of the study site in southwestern Victoria, Australia; right: location of the streams, weirs and bores and their reference
numbers. “L” denotes the presence of a water level logger in a bore.
permeable saprolite; below this is relatively fresh, fractured ter months of May to September (Dean et al., 2014). Runoff
bedrock. The fractured granite aquifer extends no deeper ratios for the pasture and eucalypt catchments are 3.0 and
than 150 m, as below this depth the fracture conductivity is 3.3 % respectively (based on the stream hydrograph records
negligible due to the high lithostatic pressure (Boutt et al., from February 2011 to February 2014), and both streams are
2010; Cook, 2003; Dept. Sustainability and Environment, ephemeral.
2012). The granite saprolite is generally thicker beneath the Vegetation of the area prior to European settlement was
lower parts of the catchment than along the ridges, and is mostly open eucalypt woodland (Department of Sustain-
overlain by up to 7 m of alluvial/colluvial material along and ability and Environment, Victoria). Following European set-
adjacent to drainage lines. This alluvium/colluvium is clay- tlement there was extensive land clearance, and the catch-
rich and impermeable in places, causing temporally variable ments were entirely converted to pasture by 1869 (White
artesian behaviour in some of the bores along the drainage et al., 2003). 76 % of the northern catchment was subse-
lines in both catchments. The topography of the site (hills in quently converted to an E. globulus plantation in July 2008
the middle of a broad valley, Fig. 1) means that both catch- (Fig. 1). Prior to the planting of the eucalypts, the eucalypt
ments are local groundwater systems, and there are no re- plantation catchment (Euc – Table 1) was used for graz-
gional groundwater inputs. There is 50 m of relief in the eu- ing, and was virtually identical to the pasture grazing catch-
calypt catchment, and 30 m in the pasture catchment; both ment (Pas – Table 1) immediately to the south. During the
catchments comprise reasonably steep hills separated by a planting of the trees the eucalypt catchment was ripped to
marked break in slope from the more or less flat topography an average depth of 800 mm and mounded to an average
along the drainage lines (Fig. 1). height of 300 mm. The tree density is 1010 stems per ha
(2.2 m between trees along a row, and 4.5 m between rows),
2.1.2 Climate and land use and fertilizer was applied following ripping and mounding
at 60 kg ha−1 (McEwens Contracting, personal communica-
The climate is Mediterranean, maritime/temperate (Cfb in tion, 2011). The tree rows run east–west across the slope in
the Köppen classification); the average annual rainfall since the main northeastern part of the catchment, and north–south
records began in 1901 for the area is 672 mm (±125σ ), (∼ down the slope) to the west of H Addinsalls Road (Fig. 2).
while pan evaporation is around 1350 mm annually, exceed-
ing rainfall for the majority of the year, excepting the win-
www.hydrol-earth-syst-sci.net/19/1107/2015/ Hydrol. Earth Syst. Sci., 19, 1107–1123, 2015
1110 J. F. Dean et al.: Locating tree plantations from a groundwater recharge perspective
280m 290m 300m 280m gers in the older bores, groundwater levels were generally
measured manually bi-monthly.
310m 290m There are two small dams in each catchment, ranging in
size from 10 to 50 m2; they are not large enough to signifi-
270m cantly impact the hydrology of the site (Fig. 1). The roads at
R the site are single lane and unsealed, and although they areow orientation less permeable than the normal ground surface and therefore
290mpromote runoff, their very small area means that they have
negligible impact on the site hydrology.
280m
260m
3 Methods
Groundwater levels, surface water flow and rainfall data were
collected from August 2009 to February 2013 for this study,
with some older long-term groundwater level data from
manual measurements going back as far as 1986 available
Legend:
from the Victorian Department of Environment and Primary
Bore
290m Industries archives. Groundwater and surface chemistry isWeir 0m 250m 500m available from sampling campaigns from August 2010 to Au-
gust 2011 (Dean et al., 2014).
<6 m to water table, 30% of catchment
<8 m to water table, 38% of catchment 3.1 Rainfall and streamflow
Daily rainfall measurements were available from a Bureau
Figure 2. Orientation of the tree rows in the eucalypt plantation and
of Meteorology station (089019) approximately 2 km south
the area where tree roots may be able to reach groundwater up to
depths of 6 and 8 m below the surface. of the study site; rainfall was also measured in the study
catchments and showed an excellent correlation with the
Bureau of Meteorology station. Due to significant gaps in
the on-site data, the Bureau of Meteorology station data
2.1.3 Catchment instrumentation was used for consistency throughout the study period. To
determine rainfall patterns, cumulative deviation from the
The pasture catchment has 13 bores drilled to different monthly mean (CDM) values were calculated alongside daily
depths, and the eucalypt catchment has 10 bores (the bores values (Sect. 4.1.1), whereby the difference between a given
may be considered to be piezometers – they are screened monthly rainfall total and the average for that month (calcu-
towards the bottom of the casing over a discrete 2 m inter- lated from the entire station’s data record of 1901 to 2012),
val; Table 1). Seven bores in the eucalypt catchment and two was cumulatively summed from one month to the next (mod-
bores in the pasture catchment were drilled for this project in ified from Craddock, 1979). The CDM values represent the
late 2009; the other bores were installed in the late 1980s longer-term rainfall patterns, with a sustained negative trend
in the pasture catchment, and the mid-1990s in the euca- for drought periods and positive values indicating wetter than
lypt catchment. A groundwater logger was installed in ev- usual periods, and match well with the longer-term hydro-
ery bore in the eucalypt catchment in August 2009, measur- graphs (Sect. 4.1).
ing at a minimum 4 h time interval, and eight bores in the Streamflow in both catchments is ephemeral, and was
pasture catchment have loggers measuring at the same fre- measured at 30 min intervals at V-notch weirs at both catch-
quency. There is a v-notch weir at the end of each catch- ment outlets and summed to annual totals, and a total for the
ment on both streams, with one bore immediately adjacent complete study period, 2009–2013. To allow comparison be-
to the eucalypt catchment weir and two next to the pasture tween catchments, volumes were converted to depth equiv-
catchment weir (Fig. 1). The bores adjacent to the weirs have alents (mm) by dividing by the respective catchment area.
Campbell CS450-L pressure transducers (accuracy±0.01 m) Streamflow is derived predominantly from direct runoff, as
measuring water level and electrical conductivity (EC) at the proportion of groundwater input into the stream is small
30 min intervals, while the other bores have Schlumberger (discussed further below).
Mini Diver loggers (accuracy±0.025 m) measuring only wa-
ter level. At the weirs the surface water level was measured 3.2 Grain size analysis
using a standard V-notch construction, and electrical con-
ductivity (EC) was recorded using a logger in the weir pool The grain size of the saprolite was used to estimate the aver-
(Dresel et al., 2012). Prior to installation of groundwater log- age specific yield value for this aquifer over the whole study
Hydrol. Earth Syst. Sci., 19, 1107–1123, 2015 www.hydrol-earth-syst-sci.net/19/1107/2015/
Row orientation
J. F. Dean et al.: Locating tree plantations from a groundwater recharge perspective 1111
Table 1. Groundwater characteristics and bore construction.
Bore ID Earliest data Screen depth Surface elevation Radiocarbon age 1σ – error Activity of 1σ – error Logger Groundwater
from bore (m below surface) (m AHD) (BP yr) 3H (TU) Cl− (mg L−1)
Pasture Catchment
Pas72 – Low 31 Aug 1986 9.4–11.6 259.55 1665 ±30 BD N 3292
Pas73 – Low 31 Aug 1986 4–6.1 259.54 2055 ±30 N 3110
Pas75 – Low 31 Aug 1986 12–13.6 263.93 935 ±35 Y 2231
Pas76 – Low 31 Aug 1986 2.2–4.2 263.98 575 ±30 BD Y 1595
Pas95 – Low 26 Aug 2009 22.8–24.8 254.13 3540 ±30 BD Y (weir) 2732
Pas96 – Low 26 Aug 2009 5–7.55 254.18 345 ±25 1.12 ±0.09 Y (weir) 2553
Pas74 – Up 31 Aug 1986 6.2–8.5 268.62 790 ±30 0.44 ±0.04 Y 306
Pas77 – Up 31 Aug 1986 17.7–19.7a 271.11 Modern 2.84 ±0.13 N 28
Pas78 – Up 31 Aug 1986 17.3–19.4 277.45 650 ±90 BD Y 1185
Pas79 – Up 31 Aug 1986 23.65–25.65a 283.23 Modern 2.55 ±0.12 N 38
Pas80 – Up 31 Aug 1986 23.3–24.4 288.23 115 ±30 1.24 ±0.08 Y 2290
Pas81 – Up 31 Aug 1986 7.1–8.9 272.12 690 ±100 0.79 ±0.08 N 668
Pas82 – Up 31 Aug 1986 23.2–24.8 283.54 430 ±30 0.60 ±0.05 Y 329
Eucalypt catchment
Euc84 – Low 12 Nov 1996 5.6–7.5 268.67 785 ±30 Y 3909
Euc85 – Low 12 Nov 1996 7.9–10 268.66 b BD Y 3537
Euc89 – Low 30 Oct 2009 26–28 261.80 7330 ±50 Y 2833
Euc90 – Low 30 Oct 2009 13–15 261.93 6980 ±45 Y 2788
Euc92 – Low 30 Oct 2009 26.2–29.2 255.43 20770 ±90 BD Y (weir) 1490
Euc93 – Low 2 Mar 2010 11–14 263.31 725 ±30 0.73 ±0.06 Y 1357
Euc83 – Up 12 Nov 1996 14.8–16.7 274.21 685 ±30 BD Y 2064
Euc91 – Up 30 Oct 2009 33.9–35.9 280.02 415 ±30 0.39 ±0.04 Y 1114
Euc94 – Up 30 Oct 2009 28–30 286.05 2060 ±30 BD Y 2891
Euc97 – Up 30 Oct 2009 43.1–45.1; 291.74 5655 ±35 0.30 ±0.04 Y 3494
57.6–59.6
BD: below detectable; a assumed screen depths; b CO2 concentration too low for analysis.
site, as the geology of the two catchments is very similar the radiocarbon ages have been compromised by “dead” car-
(see Sect. 3.6.1). During drilling of five bores on the euca- bon in the regolith; standard error of groundwater ages is 25–
lypt catchment, samples of the regolith were taken at 1 m in- 100 years (Dean et al., 2014). In addition, seven bores in the
tervals to a depth of 10 m, or until bedrock was encountered. eucalypt catchment and 11 bores in the pasture catchment
Samples were sieved using a 2 mm sieve and the material that (including the shallowest and deepest bores and a range in
passed through was then analysed using a Malvern Master- between), were analysed for tritium (standard error in these
sizer 2000. measurements was 0.04–0.13 tritium units (TU); Dean et al.,
2014; Table 1). The methodologies for both are described in
3.3 Groundwater composition more detail in Dean et al. (2014).
All 23 groundwater bores across the entire site were sampled 3.5 Radon (222Rn)
once each over a period of a year, from August 2010 to Au-
gust 2011. Seasonal variability in groundwater composition Radon surveys were carried out on groundwater and surface
is considered negligible due to the age of the groundwater at water samples in both catchments to ascertain whether there
the study site (mostly > 200 years; Table 1), and repeat sam- is a significant contribution of groundwater to surface water
222
pling produced virtually identical field parameters (Dean et flow. The Rn content of surface water and groundwater
al., 2014). Subsamples for Cl− were filtered with 0.45 µm fil- was measured using the gas-extraction for H2O accessory of
ter paper and analysed using Ion Chromatography. Ground- the Durridge RAD-7 radon detector. The RAD-7 is an alpha
water sampling, Cl− analyses and calculations of volume- particle detector that measures the decay of the radon daugh-
− ters, 214Po and 218weighted, average rainfall Cl concentrations are described Po. Samples from weirs, bores and dams
in more detail in Dean et al. (2014). (disconnected surface water bodies; Fig. 1) were collected in
250 mL vials and aerated for 5 min to degas the radon into
3.4 14C analysis and tritium analysis the air circulation within the instrument, which takes four
measurements (5 min each), and then gives the mean 222Rn
Dating of the groundwater was carried out to determine the concentration in Bq L−1; the average standard error for mea-
time period over which recharge has occurred. Groundwater surements using this instrument is 10 %.
samples from all the bores at the study site were 14C dated
and no corrections were applied, as there is no indication that
www.hydrol-earth-syst-sci.net/19/1107/2015/ Hydrol. Earth Syst. Sci., 19, 1107–1123, 2015
1112 J. F. Dean et al.: Locating tree plantations from a groundwater recharge perspective
3.6 Groundwater recharge (a) 
259.8 1030
To ensure robust estimates of groundwater recharge, two Euc90 - Low 15 day moving averageBarometric pressure rainfall
different, well-established methods were used, namely the
water table fluctuation method and chloride mass balance 259.7
method. While both methods are in widespread use, they
have known deficiencies that are discussed below.
259.6 1000
3.6.1 Water table fluctuations
The water table fluctuation (WTF) method for measuring 259.5
groundwater recharge was first applied in the 1920s (Healy
and Cook, 2002; Meinzer, 1923) and has since been refined
(e.g. Jie et al., 2011; Scanlon et al., 2005; Sophocleus, 1991). 259.4 970
The principle of this method is that rises in the groundwater 13/03/2011 28/03/2011 12/04/2011 27/04/2011 12/05/2011
(b) 
hydrograph of an unconfined aquifer provide an estimate of 261 180
Euc90 - Low Raw logger data
recharge to the water table, calculated from 260.5 15 day average 160
Rainfall
260 140
1h 120
R = Sy , (1)
259.5
1t 100259
80
258.5
where recharge (R) is the product of the specific yield of the 60
258
aquifer (Sy) and the change in hydrograph height (1h) over a
40
257.5 20
given time interval (1t). This method assumes that recharge
257 0
occurs vertically from piston flow and that water discharges 04/09/2009 04/09/2010 04/09/2011 03/09/2012
continuously from the aquifer, causing a drop in the water ta-
Figure 3. (a) Barometric pressure (in equivalent cm of H O),
ble when recharge is not occurring. Therefore the change in 2
groundwater logger data, rainfall and the 15 day moving average
hydrograph height from which recharge is calculated is the
used for the water table fluctuation method estimates of groundwa-
sum of the rise in the hydrograph, together with the decline
ter recharge. The black dots represent the average groundwater level
in the hydrograph that would have occurred in the absence of for the preceding 15 days. (b) Full record for the bore used in (a)
recharge over the same time period (Healy and Cook, 2002; – Euc90 – showing the complete removal of the large amount of
Jie et al., 2011). Several techniques have been developed to barometric noise, but keeping the overall trend of the 15 day period.
estimate the hydrograph decline: the graphical approach – in
which the exponential decay curve of the hydrograph is man-
ually extended to coincide with the peak of the next recharge exponential (as observed in the hydrographs). Because the
event (Delin et al., 2007); the master recession curve ap- assumption of an exponential recession curve is implicit in
proach – in which regression functions are assigned to simu- the graphical and master recession curve WTF methods, the
late the potential hydrograph decline for each data time-step RISE approach was adopted, i.e. the decay curve of the hy-
(Heppner et al., 2007); and the RISE approach – in which drograph was ignored. Applying the RISE approach means
the assumption is made that in the absence of recharge, no that the values calculated in this study potentially underesti-
decline in the water table occurs (Jie et al., 2011; Rutledge, mate actual recharge, but when compared with the graphical
1998). approach carried out for sections of the hydrographs where
It proved difficult to apply the graphical and master reces- exponential recession curves were evident, gave very similar
sion curve methods in the present study because these meth- values.
ods focus on the section of the hydrograph recession limb Raw bore hydrograph data collected using data loggers at
which decays exponentially, whereas the recession limbs in the site contain small fluctuations due to the impact of baro-
the Mirranatwa hydrographs often had significant sections metric pressure on the water column in the bore (Fig. 3a;
which were steep and straight (Fig. 3); this can lead to the Rasmussen and Crawford, 1997). The fluctuations in the
underestimation of actual groundwater recharge, as has been water level and the barometric pressure are normally in-
highlighted elsewhere (Cuthbert, 2014). In addition, because versely correlated (Butler et al., 2011), and can be readily
the streams in both study catchments are ephemeral, ground- corrected (Rasmussen and Crawford, 1997; Toll and Ras-
water discharge as baseflow occurs only occasionally; the mussen, 2007). At the study site these fluctuations are clearly
majority of groundwater discharge occurs at the bottom of positively correlated with barometric fluctuations (Fig. 3a),
the catchments and downstream of the catchment boundaries. and as a result normal barometric compensation techniques
This intermittent baseflow means that the recession curve could not be applied. Two types of groundwater level sen-
in the hydrographs following a recharge event may not be sors were used: Schlumberger Mini Diver loggers (accuracy
Hydrol. Earth Syst. Sci., 19, 1107–1123, 2015 www.hydrol-earth-syst-sci.net/19/1107/2015/
Water level  (mAHD) Groundwater level (m AHD) 
Rainfall (mm) Barometric pressure (cm H2O);  rainfall (mm + 970) 
J. F. Dean et al.: Locating tree plantations from a groundwater recharge perspective 1113
±0.025 m) and Campbell CS450-L pressure transducers (ac- Table 2. Median grain size compositions for sampled profiles used
curacy±0.01 m); the Campbell sensors are vented and there- to estimate a range of values for Sy in Eq. (1).
fore technically do not need compensating for barometric
pressure changes, while the Schlumberger sensors require Bore ID Clay Silt Fine sand Coarse sand
barometric compensation and barometric loggers were in- (%) (%) (%) (%)
stalled in the middle of both catchments to collect baromet- Euc89 – Low 3 39 38 19
ric data for this purpose. The barometric effect shown in Euc91 – Low 3 39 40 18
Fig. 3a is consistent across all the Schlumberger sensors in Euc93 – Low 3 36 43 18
both catchments, regardless of landscape position. Figure 3a Euc94 – Up 3 35 44 18
is based on Fig. 1 from Butler et al. (2011), and the data from Euc97 – Up 3 34 43 20
this study was prepared in the same manner, so the positive
correlation is not an artefact of data processing error. Baro-
metric forcing was evident in the Campbell sensor data also, tion within the groundwater system, that direct recharge (R,
despite their being vented, so these data were treated in the in mm) occurs via piston flow, and that runoff is negligible:
same way as the Schlumberger data (see below).
A 15-day moving average was used to remove the baro- Cp
R = P , (2)
metric fluctuations but retain the overall response to rainfall Cgw
(Fig. 3b). The 15-day time step is a narrow enough time pe-
riod to incorporate recharge events and reflect the general where P is the amount of rainfall (mm), Cp is the con-
−
trend of the hydrograph, but removes the small barometri- centration of Cl in P , and Cgw is the concentration of
−
cally forced fluctuations that bear no relationship to rain- Cl in groundwater (Allison and Hughes, 1978; Scanlon et
fall (Fig. 3). Recharge was then calculated using Eq. (1), al., 2002). R was calculated at all bores using the ground-
−
where 1h was taken as the sum of the increases in ground- water Cl content (Table 1), and rainfall Cl
− content was
water level over the time step, and then summed for the en- the median value from three different sampling periods at
tire length of the record. When there was a drop in ground- nearby sites (Fig. 1): 1954–1955 at Cavendish (Hutton and
water level from one time step to the next, this was taken Leslie, 1958), 2003–2004 at Hamilton (Bormann, 2004),
as zero recharge. The measurement uncertainty of the log- and 2007–2010 at Horsham (Nation, 2009); all Cl
− values
gers (±0.025 m) was used as the threshold for recognition were volume weighted based on rainfall during the sampling
of recharge for each 15-day time step. The RISE method periods in these studies. These three sampling periods in-
was also used to calculate recharge for the longer-term hy- clude a wet period (1954–1955) and two dry periods (2003–
drographs (generally bi-monthly measurements taken prior 2004 and 2007–2009). The median rainfall Cl
− from all of
to logger installation). these studies is 4.3± 0.9 mg L
−1, and the annual rainfall is
A specific yield value of 0.095± 0.014 was calculated for 672± 125 mm (1σ ); the uncertainties associated with each
the unconfined saprolite aquifer from the average grain size value were used to estimate the overall uncertainty in the
(clay to coarse sand; Table 2) of all the bore samples anal- recharge values calculated. R is strongly governed by Cp in
ysed (see Sect. 3.2), using the general relationship between this equation, so it is important to take into account the vari-
specific yield and grain size in Healy and Cook (2002, Ta- ability in Cp.
bles 1 and 2). The estimation of specific yield is a potential
source of considerable error in recharge calculations as it can
4 Results and discussion
vary spatially, although it can be assumed to be independent
of time (Healy and Cook, 2002). The specific yield value
4.1 Groundwater recharge estimates
calculated here is comparable to other values from weath-
ered granites in the region (0.043 – Hekmeijer and Hocking, Recharge estimates calculated using both the WTF and CMB
2001; 0.075 – Edwards, 2006). When calculating recharge methods range from 0.8±0.3 to 161 24 mm yr−1± (Table 3),
for the study site, this specific yield was applied to bores that a very wide range that matches recharge calculations from
are screened within the saprolite, and is assumed to be repre- similar climatic areas in Australia (5–250 mm yr−1; Alli-
sentative for the whole site because of the relatively uniform son and Hughes, 1978; Cook et al., 1989), and elsewhere
nature of the soils (Table 2). from low-rainfall regions around the world (0.2–35 mm yr−1;
Scanlon et al., 2006).
3.6.2 Chloride mass balance
4.1.1 Water table fluctuation method
The chloride (Cl−) mass balance (CMB) method for calcu-
lating recharge is based on the relationship between Cl− in The groundwater hydrographs vary significantly across the
groundwater and in precipitation, assuming that all Cl− in study site (Fig. 4), indicating substantial variation in ground-
the groundwater is derived from rainfall and remains in solu- water recharge. Because hydrographs from the upper parts
www.hydrol-earth-syst-sci.net/19/1107/2015/ Hydrol. Earth Syst. Sci., 19, 1107–1123, 2015
1114 J. F. Dean et al.: Locating tree plantations from a groundwater recharge perspective
Table 3. Recharge (R) values using different methods for all the bores across both catchments.
R (mm yr−1)
Bore ID Groundwater Groundwater Cl− with Water table Long-term hydrograph water
Cl− stream input correction fluctuation method table fluctuation method
Pasture catchment – lowland landscape position
Pas72 – Low∗ 0.9± 0.3 6.8± 4.6 L D
Pas73 – Low∗ 0.9± 0.3 7.2± 4.8 L D
Pas75 – Low 1.3± 0.5 3.9± 2.6 58± 9 38± 6
Pas76 – Low 1.8± 0.7 5.5± 3.7 77± 11 D
Pas95 – Low∗ 1.1± 0.4 24± 16 C D
Pas96 – Low 1.1± 0.4 26± 17 161± 24 D
Pasture catchment – upland landscape position
Pas78 – Up 2.5± 0.9 C 36± 5 D
Pas80 – Up 1.0± 0.4 C 12± 2 30± 5
Pas82 – Up 8.8± 3.3 C 26± 4 28± 4
Pasture catchment – possible fracture flow
Pas74 – Up 9.4± 3.5 C 65± 10 56± 8
Pas77 – Up 102± 38 C L D
Pas79 – Up 76± 29 C L D
Pas81 – Up 4.3± 1.6 C L D
Eucalypt catchment – lowland landscape position
Euc84 – Low∗ 0.7± 0.3 1.7± 1.3 C C
Euc85 – Low∗ 0.8± 0.3 1.9± 1.4 C C
Euc89 – Low 1.0± 0.4 5.7± 4.3 59± 9 D
Euc90 – Low 1.0± 0.4 5.8± 4.4 74± 11 D
Euc93 – Low 2.1± 0.8 8.0± 6.1 40± 6 D
Eucalypt catchment – upland landscape position
Euc83 – Up 1.4± 0.5 C 10± 2 19± 3
Euc91 – Up 2.6± 1.0 C 17± 3 D
Euc94 – Up 1.0± 0.4 C 1.7± 0.2 D
Euc97 – Up 0.8± 0.3 C 26± 4 D
Eucalypt catchment – possible fracture flow
Euc92 – Low∗ 1.9± 0.7 C C D
∗ Confined bores; L: no logger present; D: no data; C: this calculation was not done for that bore as it did not meet the required conditions (see
Sects. 3.6.1 and 3.6.2).
of the catchment show a limited response to rainfall pat- ments (average 78 mm yr−1; 12 % of rainfall; Fig. 4; Table 3).
terns, both in the detailed groundwater logger data (Fig. 4) These recharge trends have been consistent over the past 20–
and the longer-term monitoring data for the older bores 30 years (Fig. 5).
(Fig. 5), recharge values calculated using the WTF method The greater recharge in the lower-lying areas is predom-
are relatively low for these areas in both catchments (average inantly because the steeper slopes in the upland areas di-
18 mm yr−1; 3% of rainfall). rect runoff downslope to the lowland areas, which are conse-
In contrast, bores on or close to drainage lines show a quently saturated for longer with a greater volume of runoff.
much greater sensitivity to sustained rainfall and stream- In addition, runoff velocities across the lower areas decrease
flow events (e.g. for bore Pas96, rises in the hydrograph due to the reduction in slope, allowing more infiltration
directly correspond to flow in the ephemeral stream chan- into the soil. Runoff from the upland areas is aided by the
nel; Fig. 6). As a result, recharge values calculated from low-permeability, silty soils (Table 2), and infiltration in the
logger data and longer-term hydrographs using the WTF lower-lying areas, particularly through the streambed, is in-
method are relatively high for low-lying areas in both catch- creased by the greater depth of weathering (9 m depth to
Hydrol. Earth Syst. Sci., 19, 1107–1123, 2015 www.hydrol-earth-syst-sci.net/19/1107/2015/
J. F. Dean et al.: Locating tree plantations from a groundwater recharge perspective 1115
265 200 271 200
Pas74 - Up Euc83 - Up 
264 270
263 100 269 100
WTF: 65 ± 10; CMB: 9.4 ± 3.5 WTF: 10 ± 2; CMB: 1.4 ± 0.5 
262 268
261 0 267 0
267 200 269 200
Pas78 - Up Euc91 – Up 
266 268
WTF: 36 ± 5; CMB: 2.5 ± 0.9 
265 100 267 100
WTF: 17 ± 3; CMB: 2.6 ± 1.0 
264 266
263 0 265 0
268 200 266 200
Pas80 - Up Euc94 – Up 
267 265
266 100 264 100
WTF: 12 ± 2; CMB: 1.0 ± 0.4 WTF: 1.7 ± 0.2; CMB: 1.0 ± 0.4 
265 263
264 0 262 0
267 200 282 200
Pas82 - Up WTF: 26 ± 4; CMB: 0.8 ± 0.3 Euc97 - Up 
266 281
265 100 280 100
WTF: 26 ± 4; CMB: 8.8 ± 3.3 
264 279
263 0 278 0
265 200 270 200
Pas75/6 - Low Euc84/85 - Low 
264 269
263 100 268 100
WTF: 58 ± 9; CMB: 1.3 ± 0.5 WTF: -- ; CMB: 0.7 ± 0.3 
262 267
261 0 266 0
253 200 264 200
Pas95 - Low WTF: -- ; CMB: 1.1 ± 0.4 Euc93 - Low 
252 263
251 100 262 100
WTF: 40 ± 6; CMB: 2.1 ± 0.8 
250 261
249 0 260 0
255 200 261 200
Pas96 - Low Euc89 - Low 
254 260
253 100 259 100
WTF: 161 ± 24; CMB: 1.1 ± 0.4 WTF: 59 ± 9; CMB: 1.0 ± 0.4 
252 258
251 0 257 0
Aug 2009 Aug 2010 Aug 2011 Aug 2013 261 200
Euc90 - Low 
260
259 100
WTF: 74 ± 11; CMB: 1.0 ± 0.4 
258
257 0
Aug 2009 Aug 2010 Aug 2011 Aug 2013 
Figure 4. Bore hydrographs, rainfall and recharge estimates (in mm yr−1 from Table 3), for the water table fluctuation and chloride mass
balance methods. Hydrographs are sorted by landscape position – lowland or upland.
www.hydrol-earth-syst-sci.net/19/1107/2015/ Hydrol. Earth Syst. Sci., 19, 1107–1123, 2015
Water depth (mAHD) 
Rainfall (mm) 
1116 J. F. Dean et al.: Locating tree plantations from a groundwater recharge perspective
Manual measurements Logger data Cumulative deviation from mean monthly rainfall
268 600
Pas75 - Low 400
266 200
0
264 -200
-400
262 -600
31-Aug-86 31-Aug-91 30-Aug-96 30-Aug-01 31-Aug-06 31-Aug-11
267 600
Pas74 - Up 
265
0
263
261 -600
31-Aug-86 31-Aug-91 30-Aug-96 30-Aug-01 31-Aug-06 31-Aug-11
271 600
Pas80 - Up 
269
0
267
265 -600
31-Aug-86 31-Aug-91 30-Aug-96 30-Aug-01 31-Aug-06 31-Aug-11
270 600
Pas82 - Up 
268
0
266
264 -600
31-Aug-86 31-Aug-91 30-Aug-96 30-Aug-01 31-Aug-06 31-Aug-11
272 600
Euc84 - Low 
270
0
268
266 -600
31-Aug-86 31-Aug-91 30-Aug-96 30-Aug-01 31-Aug-06 31-Aug-11
272 600
Euc85 - Low 
270
0
268
266 -600
31-Aug-86 31-Aug-91 30-Aug-96 30-Aug-01 31-Aug-06 31-Aug-11
273 600
Euc83 - Up 
271
0
269
267 -600
31-Aug-86 31-Aug-91 30-Aug-96 30-Aug-01 31-Aug-06 31-Aug-11
Figure 5. Long-term hydrographs for bores with available data and cumulative deviation from mean monthly rainfall to show the relationship
between groundwater levels and long-term rainfall patterns.
Hydrol. Earth Syst. Sci., 19, 1107–1123, 2015 www.hydrol-earth-syst-sci.net/19/1107/2015/
Groundwater level (mAHD) 
Rainfall (mm) 
J. F. Dean et al.: Locating tree plantations from a groundwater recharge perspective 1117
253.2 0.45
Pas96 – Low (downstream) where “RO” (mm) is the estimated amount of runoff that0.4
253.0
0.35 would reach a given bore, and C
−
ro is the estimated Cl con-
252.8 0.3 centration of the runoff (volume weighted).
0.25
252.6
0.2 The volume of runoff at a particular bore (RO) is calcu-
252.4 0.15 lated using streamflow as a proxy for runoff, by dividing the
0.1
252.2
0.05 average streamflow per year by the amount of the catchment
252.0 0 that could theoretically provide runoff to the bore location
22/03/11 03/04/11 15/04/11 27/04/11 09/05/11 21/05/11 02/06/11 14/06/11
(i.e. a bore in the middle of the catchment is only going to
Groundwater level (raw data) Streamflow
264.5 0.45
Pas75 – Low (upstream) receive approximately half the runoff that could potentially0.4
264.3
0.35 recharge a bore at the bottom of the catchment). The Cl
−
264.1 0.3 concentration of the runoff (Cro) is calculated from the aver-
0.25
263.9
0.2 age EC measured at each weir (May 2010 to February 2013),
263.7 0.15 converted to Cl− using the EC −:Cl ratio for the study site
0.1
263.5 data set (0.39 and 0.37 for the pasture and eucalypt catch-
0.05
263.3 0 ments respectively). Equation (3) was only applied to bores
22/03/11 03/04/11 15/04/11 27/04/11 09/05/11 21/05/11 02/06/11 14/06/11
in the lowland parts of the landscape where runoff is likely to
Figure 6. Pasture stream hydrographs (Dwyer’s Creek) and bores recharge the groundwater. Because of the highly variable na-
hydrographs from the bottom of the catchment (Pas96) and midway ture of the streamflow Cl−, the potential variation in recharge
up the catchment (Pas75). values calculated from Eq. (3) is large, and this is seen in the
error values (1σ – Table 3).
The recalculated recharge values generated from Eq. (3)
bedrock in the pasture catchment and 30 m in the eucalypt are much closer to the WTF recharge values, but are still
catchment, except at the very bottom of this catchment). generally a factor of 5 to 15 lower. This may reflect the
Two of the lowland bores (Euc84 and Euc85) show very fact that the groundwater across the study site is mostly
similar recharge patterns to upland bores (e.g. Euc83), i.e. > 200 years old, indicating that the CMB values are gener-
little recharge, due to the presence of a localized confining ally representative of recharge rates under native vegetation
layer (both bores frequently go artesian; Fig. 4). prior to land clearance during European settlement in the late
Two of the upper-slope bores show high recharge (Pas74 1800s, whereas the WTF values represent recent recharge
and Pas 78), due to preferential recharge down fractures in (August 2009 to February 2013). The older, pre-European
the granite (Sect. 4.2; Fig. 5). settlement vegetation caused lower recharge, as these trees
transpire much more water from groundwater and the soil
4.1.2 Chloride mass balance method
zone than modern pasture. This disparity between modern
and pre-European recharge rates has been observed else-
Recharge values calculated from the CMB method (Eq. 2)
where in southeastern Australia (e.g. Allison et al., 1990;
are much lower than the WTF method values, often by an or-
Bennetts et al., 2006, 2007; Cartwright et al., 2007).
der of magnitude or more (Table 3): e.g. Pas96 has recharge
−1 −1 The CMB method estimates of recharge do not vary signif-values of 1.1± 0.4 mm yr (CMB) and 161± 24 mm yr
−1 icantly between the two catchments, showing that both catch-(WTF), and Pas82 has a CMB value of 8.8± 3.3 mm yr
−1 ments behaved in a similar fashion before measurements be-and a WTF value of 26± 4 mm yr . Furthermore, the bore
gan, prior to the establishment of the plantation. This corrects
hydrographs used to calculate the WTF recharge values indi-
for the lack of a calibration period prior to the change in land
cate that there is much more recharge occurring in the low-
use, a potential source of considerable error (Brown et al.,
land areas than is indicated by the CMB values.
2005).
The most likely explanation for the mismatch between
the CMB and WTF results is that the input Cl− value used
in the CMB method should be for runoff/streamflow rather 4.2 Topographic controls on recharge
than rainfall, because most recharge occurs from infiltration
Recharge estimates using the WTF method (Table 3) show
of surface flow through the streambed and across the low-
that within the local groundwater systems of the study catch-
gradient slopes adjacent to the streams, as previously dis-
ments, variations in recharge predominantly reflect differ-
cussed.
ences in topography. Dominant areas of recharge are not
To account for this difference, the CMB values were re-
− along the topographic highs of the catchments, as in the tra-calculated using the volume and Cl content of streamflow
ditional conceptual model of recharge, but are instead analo-
(assumed to be the same as runoff here) in place of rainfall
gous with more arid regions, where most recharge occurs in
in Eq. (2):
topographic depressions (Scanlon et al., 2002).
Cro Recharge rates increase as surface elevation decreases
R = RO , (3)
Cgw (Fig. 7). The steeper slopes of the upland areas promote
www.hydrol-earth-syst-sci.net/19/1107/2015/ Hydrol. Earth Syst. Sci., 19, 1107–1123, 2015
Groundwater level (mAHD) 
Streamflow (ML/day) 
1118 J. F. Dean et al.: Locating tree plantations from a groundwater recharge perspective
B
Euc94 Pas80
B
Euc91 B’
280m Pas78 B’
Pasture 
Eucalypt catchment
270m catchment Pas74
Euc90 Euc93 Pas76
260m 1.0 ± 0.4
12 ± 2 9.4 ± 3.5
1.8 ± 0.7 65 ± 10
1.0 ± 0.4 2.5 ± 0.9 77 ± 11
1.7 ± 0.2 36 ± 5
250m
2.1 ± 0.8
1.0 ± 0.4 40 ± 6 Aug 2013 water table
74 ± 11 Aug 2009 water table
2.6 ± 1.0
17 ± 3
2.6 ± 1.0 Recharge (mm/yr) from groundwater chloride mass 240m balance method
17 ± 3 Recharge (mm/yr) from water table fluctuation 
method
230m
0m 250m 500m 750m 1000m 1250m 1500m
Figure 7. Cross-section from bore Euc91 across both catchments to bore Pas74 showing recharge rates based on both methods used in this
study, and the water table change over the course of the study period (see Fig. 1 for bore locations).
runoff rather than infiltration, aided by low-permeability, tralia where there was disparity between the residence times
silty soils (Table 2). Overland flow is focused into topo- of groundwater samples (Cartwright et al., 2007). Neverthe-
graphic lows and along drainage lines. Here the granite is less, the dominant recharge control across both catchments
most weathered, as indicated by the greater depth to bedrock is topography rather than fracture heterogeneity, as shown by
here (9 m in the pasture catchment, and 30 m in the eu- the relatively flat hydrographs for most of the upland bores,
calypt catchment except at the very bottom of this catch- and strongly oscillating hydrographs in the lowland bores
ment), encouraging recharge to occur, particularly through (Fig. 4).
the streambed.
4.4 The interplay between ephemeral streamflow and
4.3 Influence of fractures on groundwater recharge groundwater recharge and discharge
The 14C data (Table 1) show that most of the groundwater at The streams at the study site are ephemeral, flowing on aver-
the study site is older than the tree plantation, but the ground- age only 40 % of the time at the catchment outlets. When
water in some bores (Pas74, Pas 80, Pas81, Pas 82, Pas 96, they are dry, recharge can occur readily along and near
Euc91, Euc93 and Euc97) also contains measurable tritium, the streambeds as upwards groundwater gradients are not
indicating a component of younger groundwater (< 50 years present, because the water table is below the base of the
old). Recharge in fractured rock aquifers like granite is con- stream. As a result, bores in the lower parts of the catchments
trolled to some extent by the fracture network (Cook, 2003), (e.g. Pas96 near the outlet of the pasture catchment; Fig. 6)
which forms multiple recharge pathways. In the study area show a clear, sometimes instantaneous link between recharge
this has allowed mixing of young groundwater (containing and runoff.
tritium) with much older groundwater (as shown by the 14C Following extended periods of wet weather, the ephemeral
dates; Table 1). The hydrograph for the upslope bore Pas74 stream at the bottom of the eucalypt catchment is fed by
(Fig. 4) shows high recharge following rainfall events (in groundwater discharge, as shown by the significant levels of
contrast to most of the other upslope bores), most likely be- 222Rn measured at the weir (11 Bq L−1; Fig. 8); however, the
cause it is located on a fracture in the granite that allows rapid elevated 222Rn measured in the eucalypt stream could just
recharge, as shown by the dilute groundwater with low Cl− be due to the close proximity of the granite bedrock to the
concentrations (Dean et al., 2014) and the presence of signif- surface at the bottom of this catchment. This is suggested
icant amounts of tritium (Table 1). by the high 222Rn values in Pas95 and Euc92, both screened
This dual porosity (matrix and fracture flow) influence on in granite bedrock, compared to the lower 222Rn values in
recharge has been observed elsewhere in southeastern Aus- Euc90 and Pas96, which are screened in the weathered gran-
Hydrol. Earth Syst. Sci., 19, 1107–1123, 2015 www.hydrol-earth-syst-sci.net/19/1107/2015/
J. F. Dean et al.: Locating tree plantations from a groundwater recharge perspective 1119
Eucalypt Pasture  A Euc97 A’
250 catchment  catchment
290m
280m
Euc83
200 270m Euc85
Euc93 (3) Euc89 Euc92
260m
(2)
150 (1)
250m
Water table
100 240m Projected water table (1)Projected water table (2)(3)
Projected granite bedrock
230m Granite bedrock
50 220m
0m 250m 500m 750m 1000m
Eucalypt Pasture 
Pcalatcnhtamtieonnt Fcatrcmhment
0 10 (3)
1 (2)
0.1
A (1)
0.01
A’
0.001
0.0001
0.00001
0 10 20 30 40 50 60
Figure 8. 222Rn concentrations in the streams, measured at the Percentage of time that indicated flow was equaled or exceeded
weirs of both sites, and nearby bores. Surface water from further
Figure 9. Long section from bores Euc97 to Euc92 showing the ef-
up the catchments is represented by water from dams located ups-
fect of the shallow granite on the water table under different flow
lope in both catchments. The relatively high levels in the ground-
conditions shown in the flow duration curve below: (1) where low
water are a result of the decay of uranium present in the allanite and
flows in the eucalypt catchment stream are sustained for longer due
zircon of the granite.
to some groundwater discharge compared to virtually no ground-
water discharge in the pasture catchment, (2) where the water ta-
ble is at the surface and runoff is transported more quickly out of
ite saprolite (Fig. 8). Regardless, the shallow granite bedrock
the eucalypt than in the pasture catchment, and (3) where there are
at the outlet of the eucalypt catchment (less than 2 m below some rare, very high flows, much higher than observed in the pas-
the surface; Fig. 9) forces groundwater towards the surface ture catchment.
here. In contrast, the bedrock at the bottom of the pasture
catchment is 9 m deep, so the water table lies more consis-
tently below the base of the stream and there is less ground- The groundwater hydrographs indicate that during the
water discharge; as a result, the pasture catchment has fewer study period, recharge occurred readily in the lowland areas
low flows than the eucalypt catchment (Fig. 9) and lower of both catchments, particularly when there was enough rain-
222Rn levels (1 Bq L−1 at the weir; Fig. 8). fall to generate consistent flow in the streams, while much
In both catchments, during periods of little or no rainfall, less recharge is evident on the upper slopes. There is rel-
the water table lies below the surface, so recharge can occur atively little groundwater discharge along the streams, as
along the length of the channel. When it begins to rain and the shown by the 222Rn data (Fig. 8), and groundwater within
system wets up, the water table rises at the downstream end the catchments is lost predominantly through evapotranspi-
of the catchment and groundwater begins to discharge here ration, particularly when the water table is within 2 m of the
(this occurs more frequently and to a greater extent in the eu- ground surface (as commonly occurs in southeastern Aus-
calypt catchment). Continued rain raises the water table so it tralia, e.g. Bennetts et al., 2006, 2007); a small amount flows
connects to the stream further upstream, increasing the length out at the bottom of the catchment.
of the stream that receives groundwater discharge (Fig. 9;
Adelana et al., 2014). When rainfall ceases, the water table 4.5 Vegetation controls on recharge
drops and progressively disconnects from the stream, start-
ing upstream, until it is completely disconnected throughout The bore hydrographs in the eucalypt catchment show a clear
the catchment. This means that during smaller rainfall events, overall declining trend of up to 3 m during the study period,
when the water table remains below the land surface and does evident even in artesian bores (Euc84 and Euc85), and re-
not connect to the stream, recharge occurs along the length gardless of landscape position (Fig. 4). This decline is not
of the stream. During larger rainfall events, as the water table evident in hydrographs from the pasture catchment (Fig. 4),
comes to the surface along the stream channel, the area of where the water table has increased by 0.5–1 m during the
potential recharge decreases. whole study period as a result of consecutive wet summers of
2010/2011 and 2011/2012 (Fig. 7). The tree plantation was
a little over 1 year old when the main measurements of this
www.hydrol-earth-syst-sci.net/19/1107/2015/ Hydrol. Earth Syst. Sci., 19, 1107–1123, 2015
Radon (Bq/L) 
upstream 
downstream 
upstream 
downstream 
Flow (mm/day)
1120 J. F. Dean et al.: Locating tree plantations from a groundwater recharge perspective
study began, and as the age of the plantation increased, a availability resulting from reforestation will be offset by the
steeper decline in groundwater depth was observed (Fig. 4). beneficial gains of the carbon sequestration within the new
The water level decrease in the eucalypt catchment, with trees (Schrobback et al., 2011).
no corresponding drop in the pasture catchment, is attributed A reduction of groundwater recharge by plantations, as
to greater water use by the trees, as has been demonstrated documented in this study, lowers the water table and can re-
elsewhere (e.g. Adelana et al., 2014; Bosch and Hewlett, duce stream flow. If this is the object of the reforestation, for
1982). The water table decline is less in the upland areas example to reduce saline groundwater discharge, then this
(Fig. 9), probably because recharge rates here are lower, so land use change may well serve its purpose (Bennetts et al.,
that the decrease in recharge due to tree water use has had rel- 2007). However, the recent drought in southeastern Australia
atively little impact. Furthermore, in the upland areas the wa- (1997–2010) has exacerbated concerns that trees may be a
ter table is too deep for the vegetation to access the ground- significant user of local and regional water resources, reduc-
water directly; Benyon et al. (2006), in a study in the same ing groundwater recharge, discharge and surface water avail-
region of southeastern Australia, found that deep-rooted eu- ability (Jackson et al., 2005).
calypts can only access groundwater up to a depth of 6–8 m. In order to reduce the impact on water availability, current
In the lowland areas the trees are able to reach the ground- regulation of tree plantations in southeastern Australia fo-
water (Fig. 2), and this, combined with the interception of cuses on the percentage of a catchment that may be planted.
potential recharge in the soil zone by the growing plantation, However, the present study shows that the location of the
causes the observed decline in groundwater level (Fig. 4). plantation within the catchment is significant also, with a
Although tree roots can provide preferential pathways for in- smaller water table decline seen in the upland areas of the
filtration of rainfall to the water table (Burgess et al., 2001), eucalypt catchment. Therefore to reduce the impact of plan-
any effect of this is masked by the overall impact of euca- tations on groundwater recharge, tree planting should be
lypt water use. The increasing rate of decline in groundwater avoided in the dominant zone of recharge, i.e. the topograph-
depth over time can be attributed to the greater water usage ically low areas and along the drainage lines, and should be
by the trees as they grow (Fig. 4). concentrated on the upper slopes, where the water tables are
The narrow areas immediately adjacent to the drainage deeper and the trees are less likely to access the groundwater
lines in the eucalypt catchment are covered in grass and transpire it directly. At present, tree plantations in Victo-
and therefore there is less direct interception of potential ria cannot be planted within 20 m of drainage lines, to avoid
recharge, but in fact these areas show the biggest decline in erosion of creek banks when the trees are removed (Dept. of
groundwater level (Fig. 7). The highest rates of recharge oc- Environment and Primary Industries, Victoria); we suggest
cur along the drainage lines and the adjacent trees will there- that this currently restricted area along the drainage lines be
fore have a substantial impact there, in particular because expanded to include as much of the low topography parts of
they are directly accessing the groundwater. the site as practicable.
With groundwater levels in the eucalypt catchment still in The expansion of the drainage line exclusion zone in
decline at the end of the study period, five years since the tree plantations will have an added benefit in many regions
establishment of the eucalypt plantation, there is no sign that of southeastern Australia where the groundwater is saline.
the system is reaching equilibrium under the new land use. This is because the parts of the catchments where the saline
Brown et al. (2005) indicate that equilibrium would not be groundwater is within a few metres of the land surface (gen-
expected until more than 5 years after the land use change erally the lowland areas) can have a negative effect on tree
occurred. health; at the study site, the trees closer to the drainage lines
are shorter and thinner than those upslope.
4.6 Management of tree plantations and recharge However, excluding tree planting from low-elevation ar-
eas reduces the number of trees that can be planted within a
Afforestation of farmland was widespread in southeastern catchment, and also means that trees are not planted in areas
South Australia and southwestern Victoria (known as the where (good quality) groundwater is shallowest and can be
Green Triangle) from the 1980s through to the 2000s, with most readily accessed for tree growth. As the primary pur-
the plantation area expanding by 5–14 % to 30 000 ha in Vic- pose of many tree plantations is the production of wood and
toria alone (Adelana et al., 2014; Benyon et al., 2006; Ierodi- pulp products for economic gain, this restriction will slow
aconou et al., 2005). However, the subsequent development economic returns. To overcome this, consideration could be
of tree plantations in the region has been hindered by a poor given to planting lower-water-use trees that can better cope
timber market (HVP Plantations, personal communication, with the upslope areas where water supplies for tree growth
2013) and concerns that plantations use more groundwater may be limited.
and surface water than other land uses like farming. As a re- This management strategy of balancing economic and hy-
sult, tree plantations in the state of South Australia must now drologic perspectives when locating tree plantations within
be licensed as groundwater users (Govt. of South Australia, catchments will be applicable to other low-rainfall, high-
2009), while it is hoped that the potential reduction in water
Hydrol. Earth Syst. Sci., 19, 1107–1123, 2015 www.hydrol-earth-syst-sci.net/19/1107/2015/
J. F. Dean et al.: Locating tree plantations from a groundwater recharge perspective 1121
evaporation regions, and should be considered for tree plan- Ian Cartwright, Guillaume Bertrand, a third anonymous reviewer
tations in similar climatic areas worldwide. and the editor, Przemyslaw Wachniew, for their constructive feed-
back and helpful comments throughout the review process.
Neither the NCGRT, AINSE nor DEPI Victoria (as funding
5 Conclusions sources) were involved in the design of this specific study, nor
were they involved in the collection or analysis of the resulting data.
While the importance of topography and ephemeral streams
to focused recharge in low-rainfall regions around the world Edited by: P. Wachniew
has been known for some time, the implications of this as-
pect of the groundwater resource literature have not been in-
corporated into plantation management guidelines and leg-
islation. In this study, it is shown that the majority of mod-
ern recharge at the study site, calculated from the water table References
fluctuation method, occurs in the lower parts of both study
catchments (12 % of rainfall versus 3 % in the upland areas). Adelana, M., Dresel, E., Hekmeijer, P., Zydor, H., Webb, J.,
Overland flow is focused into topographic lows and along Reynolds, M., and Ryan, M.: A comparison on streamflow
drainage lines where greater infiltration can occur. Recharge and water balances in adjacent farmland and forest catch-
calculations using a corrected chloride mass balance method ments in south-western Victoria, Australia, Hydrol. Process.,
gave lower values than modern recharge estimates because doi:10.1002/hyp.10281, online first, 2014.
the groundwater across the study site is mostly > 200 years Allison, G. B. and Hughes, M. W.: The use of environmental
old, representing recharge under native eucalypt forest prior chloride and tritium to estimate total recharge to an unconfined
aquifer, Aust. J. Soil Res., 16, 181–195, 1978.
to European land clearance. Relatively little groundwater
Allison, G. B., Cook, P. G., Barnett, S. R., Walker, G. R., Jolly, I. D.,
discharges into the streams or flows out at the bottom of
and Hughes, M. W.: Land clearance and river salinisation in the
the catchment; groundwater within the catchments is lost
western Murray Basin, Australia, J. Hydrol., 119, 1–20, 1990.
predominantly through evapotranspiration. Overall the tree Bell, R. W., Schofield, N. J., Loh, I. C., and Bari, M. A.: Ground-
plantation in this study caused a drawdown in groundwater water response to reforestation in the Darling Range of Western
levels, increasing over time as the trees aged, compared to a Australia, J. Hydrol., 115, 297–317, 1990.
slight rise in groundwater levels in the pasture catchment. Bennetts, D. A., Webb, J. A., Stone, D. J. M., and Hill, D. M.: Un-
The results of this study lead to the conclusion that both derstanding the salinisation process for groundwater in an area
the hydrogeological and economic frameworks for commer- of south-eastern Australia, using hydrochemical and isotopic ev-
cial forestry need to be considered. If conserving ground- idence, J. Hydrol., 323, 178–192, 2006.
water recharge is a primary objective, tree planting should Bennetts, D. A., Webb, J. A., Stone, D. J. M., and Hill, D. M.:
Dryland salinity processes within the discharge zone of a local
be avoided in the dominant zone of recharge, and concen-
groundwater system, southeastern Australia, Hydrogeol. J., 15,
trated on the upper slopes, where recharge is low enough
1197–1210, 2007.
that any further reduction will have minimal impact. We sug-
Benyon, R. G.: Water use by tree plantations in the green triangle,
gest expanding present regulations for tree plantations which A review of current knowledge, The Glenelg Hopkins Catchment
specify that trees cannot be planted within a certain distance Management Authority, Hamilton, Australia, 2002.
of drainage lines, including as much of the low topography Benyon, R. G., Theiveyanathan, S., and Doody, T. M.: Impacts of
parts of the site as practicable. Consideration should be given tree plantations on groundwater in south-eastern Australia, Aust.
to the potential negative impact on the financial viability of J. Bot., 54, 181–192, 2006.
a tree plantation. This management strategy is applicable to Bormann, M. E.: Temporal and spatial trends in rainwater chemistry
low-rainfall, high-evaporation regions worldwide. across Central and Western Victoria, Honours Thesis, La Trobe
University, Melbourne, Australia, 2004.
Bosch, J. M. and Hewlett, J. D.: A review of catchment experiments
Acknowledgements. This work would not have been possible with- to determine the effect of vegetation changes on water yield and
out the assistance and support of the land owners, Iven, Iris and evapotranspiration, J. Hydrol., 55, 3–23, 1982.
Marcia Field, Macquarie Bank Foundation, and the plantation man- Boutt, D. F., Diggins, P., and Mabee, S.: A field study (Mas-
agement, McEwen’s Contracting Pty Ltd. We would like to ac- sachusetts, USA) of the factors controlling the depth of ground-
knowledge Phil Cook and Peter Hekmeijer at the Department of water flow systems in crystalline fractured-rock terrain, Hydro-
Primary Industries Victoria, our collaborators in the National Cen- geol. J., 18, 1839–1854, 2010.
tre for Groundwater Research and Training (of which this re- Brown, A. E., Zhang, L., McMahon, T. A., Western, A. W., and
search is part of programme 4 – http://www.groundwater.com.au) Vertessy, R. A.: A review of paired catchment studies for deter-
for funding and support, the Australian Institute of Nuclear Science mining changes in water yield resulting from alterations in vege-
and Engineering (AINSE – grant number AINSTU0710) for fund- tation, J. Hydrol., 310, 28–61, 2005.
ing and support, and funding from the State of Victoria through Burgess, S. S. O., Adams, M. A., Turner, N. C., White, D. A., and
the Department of Environment and Primary Industries, Future Ong, C. K.: Tree roots: conduits for deep recharge of soil water,
Farming Systems Research Division. We would also like to thank Oecologia, 126, 158–165, 2001.
www.hydrol-earth-syst-sci.net/19/1107/2015/ Hydrol. Earth Syst. Sci., 19, 1107–1123, 2015
1122 J. F. Dean et al.: Locating tree plantations from a groundwater recharge perspective
Butler, J. J., Jin, W., Mohammed, G. A., and Reboulet, E. C.: New Farley, K. A., Jobbágy, E. G., and Jackson, R. B.: Effects of af-
insights from well responses to fluctuations in barometric pres- forestation on water yield: a global synthesis with implication
sure, Ground Water, 49, 525–533, 2011. for policy, Glob. Change Biol., 11, 1565–1576, 2005.
Camporese, M., Dean, J. F., Dresel, P. E., Webb, J. A., and Daly, Feikema, P. M., Morris, J. D., and Connell, L. D.: The water balance
E.: Hydrological modelling of paired catchments with compet- and water sources of a Eucalyptus plantation over shallow saline
ing land uses, Proceedings of the 20th International Congress on groundwater, Plant Soil, 332, 429–449, 2010.
Modelling and Simulation, Adelaide, Australia, 1–6 December Govt. of South Australia: Managing the water resource impacts of
2013, 1819–1825, 2013. plantation forests, Dept. of Water, Land and Biodiversity Conser-
Camporese, M., Daly, E., Dresel, P. E., and Webb, J. A.: Simpli- vation, Adelaide, Australia, 2009.
fied modelling of catchment-scale evapotranspiration via bound- Healy, R. W. and Cook, P. G.: Using groundwater levels to estimate
ary condition switching, Adv. Water Resour., 69, 95–105, 2014. recharge, Hydrogeol. J., 10, 91–109, 2002.
Cartwright, I., Weaver, T. R., Stone, D., and Reid, M.: Constrain- Hekmeijer, P. and Hocking, M.: Feasibility of groundwater pump-
ing modern and historical recharge from bore hydrographs, 3H, ing from granite slopes, Centre for Land Protection Research,
14C, and chloride concentrations: Applications to dual-porosity Bendigo, Victoria, Australia, Technical report no. 71, 2001.
aquifers in dryland salinity areas, Murray Basin, Australia, J. Hy- Heppner, C. S., Nimmo, J. R., Folmar, G. J., Gburek, W. J., and
drol., 332, 69–92, 2007. Risser, D. W.: Multiple-methods investigation of recharge at a
Cook, P. G.: A Guide to Regional Groundwater Flow in Fractured humid-region, fractured rock site, Pennsylvania, USA, Hydro-
Rock Aquifers, CSIRO Land and Water, Adelaide, Australia, geol. J., 15, 915–927, 2007.
2003. Hergt, J., Woodhead, J., and Schofield, A.: A-type magmatism in
Cook, P. G., Walker, G. R., and Jolly, I. D.: Spatial variability of the Western Lachlan Fold Belt? A study of granites and rhyolites
groundwater recharge in a semiarid region, J. Hydrol., 111, 195– from the Grampians region, Western Victoria, Lithos, 97, 122–
212, 1989. 139, 2007.
Craddock, J. M.: Methods of comparing annual rainfall records for Hibbert, A. R.: Forest treatment effects on water yield, in: Interna-
climatic purposes, Weather, 34, 332–346, 1979. tional Symposium on Forest Hydrology, edited by: Sopper, W. E.
Cuthbert, M. O.: Straight thinking about groundwa- and Lull, H. W., Pergamon, Oxford, UK, 30 August–10 Septem-
ter recession, Water Resour. Res., 50, 2407–2424, ber 1965, 527–543, 1967.
doi:10.1002/2013WR014060, 2014. Hutton, J. T. and Leslie, T. I.: Accession of non-nitrogenous ions
Dalhaus, P. G., Cox, J. W., Simmons, C. T., and Smitt, C. M.: Be- dissolved in rainwater to soils in Victoria, Aust. J. Agr. Res., 9,
yond hydrogeologic evidence: challenging the current assump- 59–84, 1958.
tions about salinity processes in the Corangamite region, Aus- Ierodiaconou, D., Laurenson, L., Leblanc, M., Stagnitti, F., Duff, G.,
tralia, Hydrogol. J., 16, 1283–1298, 2008. Salzman, S., and Versace, V.: Consequences of land use change
Dean, J. F., Webb, J. A., Jacobsen, G. E., Chisari, R., and Dresel, on nutrient exports: a regional scale assessment in south-west
P. E.: Biomass uptake and fire as controls on groundwater Victoria, Australia, J. Environ. Management, 74, 305–316, 2005.
solute evolution on a southeast Australian granite: aboriginal Jackson, R. B., Jobbagy, E. G., Avissar, R., Baidya Roy, S., Barrett,
land management hypothesis, Biogeosciences, 11, 4099–4114, D. J., Cook, C. W., Farley, K. A., le Maitre, D. C., McCarl, B.
doi:10.5194/bg-11-4099-2014, 2014. A., and Murray, B. C.: Trading water for carbon with biological
Delin, G. G., Healy, R. W., Landon, M. K., and Bhlke, J. K.: Effects carbon sequestration, Science, 310, 1944–1947, 2005.
of topography and soil properties on recharge at two sites in an Jie, Z., van Heyden, J., Bendel, D., and Barthel, R.: Combination of
agricultural field, J. Am. Water Resour. As., 36, 1401–1416, 200. soil-water balance models and water-table fluctuation methods
Delin, G. N., Healy, R. W., Lorenz, D. L., and Nimmo, J. R.: Com- for evaluation and improvement of groundwater recharge calcu-
parison of local- to regional-scale ground-water recharge in Min- lations, Hydrogeol. J., 19, 1487–1502, 2011.
nesota, USA, J. Hydrol., 334, 231–249, 2007. Jobbagy, E. G. and Jackson, R. B.: Groundwater use and saliniza-
Dept. Sustainability and Environment: Groundwater SAFE: Secure tion with grassland afforestation, Glob. Change Biol., 10, 1299–
allocations, future entitlements, Victorian Gov. Dept. of Sustain- 1312, 2004.
ability and Environment, Melbourne, Australia, 2012. Meinzer, O. E.: The occurrence of groundwater in the United States
Domenico, P. A. and Schwartz, F. W.: Physical and Chemical Hy- with a discussion of principles, US Geololgical Survey, Wash-
drogeology, John Wiley and Sons Inc., New York, USA, 528 pp., ington, D.C., USA, Water-Supply Paper 489, 1923.
1998. Nation, E.: Groundwater Recharge for Agriculture: Rainfall Chem-
Dresel, P. E., Hekmeijer, P., Dean, J. F., Harvey, W., Webb, J. A., istry Report, Bureau of Rural Sciences, Canberra, Australia, 62
and Cook, P.: Use of laser-scan technology to analyse topography pp., 2009.
and flow in a weir pool, Hydrol. Earth Syst. Sci., 16, 2703–2708, Nicholson, C., Dalhaus, P., Anderson, G., Kelliher, C., and
doi:10.5194/hess-16-2703-2012, 2012. Stephens, M.: Corangamite Salinity Action Plan 2005–2008,
Edwards, M.: A hydrological, hydrogeological and hydrogeochem- Corangamite Catchment Management Authority, Colac, Victo-
ical study of processes leading to land and water salinisation ria, Australia, 133 pp., 2006.
in the Mount William Creek catchment, southeastern Australia, Rasmussen, T. C. and Crawford, L. A.: Identifying and Remov-
PhD thesis, La Trobe University, Bundoora, Melbourne, Aus- ing Barometric Pressure Effects in Confined and Unconfined
tralia, 2006. Aquifers, Ground Water, 35, 502–511, 1997.
Rutledge, A. T.: Computer programs for describing the recession of
ground-water discharge and for estimating mean ground-water
Hydrol. Earth Syst. Sci., 19, 1107–1123, 2015 www.hydrol-earth-syst-sci.net/19/1107/2015/
J. F. Dean et al.: Locating tree plantations from a groundwater recharge perspective 1123
recharge and discharge from streamflow records–update, US Ge- Sophocleus, M. A.: Combining the soilwater balanceand water-
ological Survey, Reston, Virginia, USA, Water-Resources Inves- level fluctuation methods to estimate natural groundwater
tigations report 98-4148, 43 pp., 1998. recharge: practical aspects, J. Hydrol., 124, 229–241, 1991.
Scanlon, B. R., Healy, R. W., and Cook, P. G.: Choosing appropriate Toll, N. J. and Rasmussen, T. C.: Removal of barometric pressure
techniques for quantifying groundwater recharge, Hydrogeol. J., effects and earth tides from observed water levels, Ground Water,
10, 18–39, 2002. 45, 101–105, 2007.
Scanlon, B. R., Reedy, R. C., Stonestrom, D. A., Prudic, D. E., and van Dijk, A. I. J. M., Hairsine, P. B., Arancibia, J. P., and Dowl-
Dennehy, K. F.: Impact of land use and land cover change on ing, T. I.: Reforestation, water availability and stream salinity:
groundwater recharge and quality in the southwestern US, Glob. A multi-scale analysis in the Murray-Darling Basin, Australia,
Change Biol., 11, 1577–1593, 2005. Forest Ecol. Manag., 251, 94–109, 2007.
Scanlon, B. R., Keese, K. E., Flint, A. L., Flint, L. E., Gaye, C. B., VandenBerg, A. H. M.. Rock unit names in western Victoria, Seam-
Edmunds, M., and Simmers, I.: Global synthesis of groundwa- less Geology Project, Geological Survey of Victoria Report, Geo-
ter recharge in semiarid and arid regions, Hydrol. Process., 20, science, Victoria, Australia, 130, 2009.
3335–3370, 2006. Webb, J. A., Williams, B. G., Bailue, K., Walker, J., and Anderson,
Scanlon, B. R., Jolly, I., Sophocleous, M., and Zhang, L.: Global J. W.: Short-term groundwater dynamics at a paddock scale, Pro-
impacts of conversions from natural to agricultural ecosystems ceedings of Water Down Under Adelaide, Australia, 15–17 April
on water resources: Quantity versus quality, Water Resour. Res., 2008, 1493–1500, 2008.
43, W03437, doi:10.1029/2006WR005486, 2007. White, M., Oates, A., Barlow, T., Pelikan, M., Brown, J., Rosen-
Schilling, K. E.: Investigating local variation in groundwater gren, N., Cheal, D., Sinclair, S., and Stutter, G.: The Vegetation
recharge along a topographic gradient, Walnut Creek, Iowa, of North-West Victoria, Department of Sustainability and Envi-
USA, Hydrogeol. J., 17, 397–407, 2009. ronment, Melbourne, Australia, 2003.
Schofield, N. J.: Tree planting for dryland salinity control in Aus- Winter, T. C.: The concept of hydrologic landscapes, J. Am. Water
tralia, Agroforest. Syst., 20, 1–23, 1992. Resour. As., 37, 335–349, 2001.
Schrobback, P., Adamson, D., and Quiggin, J.: Turning water into
carbon: carbon sequestration and water flow in the Murray-
Darling Basin, Environ. Resour. Econ, 49, 23–45, 2011.
www.hydrol-earth-syst-sci.net/19/1107/2015/ Hydrol. Earth Syst. Sci., 19, 1107–1123, 2015