Glacier melt is an important source of water for high Andean rivers in central Chile, especially in dry years, when it can be an important contributor to flows during late summer and autumn. However, few studies have quantified glacier melt contribution to streamflow in this region. To address this shortcoming, we present an analysis of meteorological conditions and ablation for Universidad Glacier, one of the largest valley glaciers in the central Andes of Chile at the head of the Tinguiririca River, for the 2009–2010 ablation season. We used meteorological measurements from two automatic weather stations installed on the glacier to drive a distributed temperature-index and runoff routing model. The temperature-index model was calibrated at the lower weather station site and showed good agreement with melt estimates from an ablation stake and sonic ranger, and with a physically based energy balance model. Total modelled glacier melt is compared with river flow measurements at three sites located between 0.5 and 50 km downstream. Universidad Glacier shows extremely high melt rates over the ablation season which may exceed 10 m water equivalent in the lower ablation area, representing between 10 and 13 % of the mean monthly streamflow at the outlet of the Tinguiririca River Basin between December 2009 and March 2010. This contribution rises to a monthly maximum of almost 20 % in March 2010, demonstrating the importance of glacier runoff to streamflow, particularly in dry years such as 2009–2010. The temperature-index approach benefits from the availability of on-glacier meteorological data, enabling the calculation of the local hourly variable lapse rate, and is suited to high melt regimes, but would not be easily applicable to glaciers further north in Chile where sublimation is more significant.
The central region of Chile (30–37
In this region, winter precipitation is driven by the interactions between the westerlies circulation and the Andean natural barrier; and summer runoff is strongly influenced by the storage and release from glaciers and snow covers (Garreaud, 2013). Accurate knowledge of the processes involved in the runoff generation from mountainous areas is vital to understand and predict the availability of water resources and contribution to sea level rise (Mernild et al., 2016), especially considering the ongoing and projected future decrease in glacier volume under climate warming scenarios (Pellicciotti et al., 2014; Ragettli et al., 2016).
At these latitudes, the Andes present several peaks over 6000 m above sea
level (a.s.l.) and have a mean elevation of
There have been only a few physically based distributed glacio-hydrological
modelling investigations in the Andes of Chile (Pellicciotti et al., 2014;
Ayala et al., 2016), which is an important limitation for the understanding
of future glacier contribution to river flows, considering the current trends
of glacier shrinkage (e.g. Bown
et al., 2008; Le Quesne et al., 2009; Malmros et al., 2016) and negative mass
balance (Mernild et al., 2015) in the region. One of the most studied
glaciers in the region is Juncal Norte Glacier. Pellicciotti et al. (2008)
investigated the point-scale energy balance and melt regime using an
automatic weather station (AWS) located in the glacier ablation zone, showing
that the ablation process is dominated by incoming shortwave radiation. Using
a physically based distributed glacier-hydrological model, Ragettli and
Pellicciotti (2012) estimated that melted glacier ice from Juncal Norte
Glacier contributed 14 % of the basin (241 km
We present an analysis of meteorological conditions and ablation for Universidad Glacier, a large valley glacier in central Chile, located in a climatic transition zone with a Mediterranean climate type, between the humid temperate south and arid north of the country. The main aims are (1) to identify the principal meteorological drivers of ablation and their patterns and trends during a full ablation season; (2) to compare methods of ablation estimation using degree-day and energy balance models; and (3) to estimate the contribution of glacier melt to downstream river flows and its water resource implications. The aims are addressed using point energy balance and distributed temperature-index models forced with data from two AWS located on the glacier ablation and accumulation zones, and stream gauging records both proximal to the glacier snout and 50 km downstream at mid-altitude on the Tinguiririca River.
Location of Universidad Glacier in central Chile.
Panel
Universidad Glacier (34
Scientific investigations at Universidad Glacier were initiated by
Lliboutry (1958), who described some morphological characteristics of the
glacier surface, including ogives, blue bands, penitents and moraines, noting
the absence of penitents above 3800 m a.s.l., in contrast to glaciers
further north in Chile. According to his observations, the lower part of the
glacier had a sudden advance around 1943. After this event, a spectacular
recession (
This study focuses on the ablation season (1 October to 31 March) of the 2009/2010 hydrological year, when the discharge, meteorological and glaciological conditions were monitored. The 2009/2010 hydrological year is of significance as it marks the beginning of a period of extreme aridity (2010–2015) in central and southern Chile (Bosier et al., 2016) which extended into 2017 according to data from the Dirección General de Aguas de Chile (Chilean Water Cadastre, DGA) and Dirección Meteorológica de Chile (National Weather Service, DMC).
Data collected include meteorological observations at two AWS, surface lowering monitoring from ablation stakes and a sonic ranger (Fig. 1), satellite-derived snow cover distribution and discharge measurements in the proglacial stream. Following the analysis of energy fluxes at the location of the lower AWS, a temperature-index model was calibrated and applied at the glacier scale. The resulting melt amounts were used to estimate total glacier discharge, which is compared with downstream discharge records.
Stake ablation measurements.
Two AWS were installed on the surface of the glacier (Fig. 1): one on the
ablation zone (AWS1, 34
Hourly time series of observed meteorological variables.
Three stakes installed on the ablation zone of the glacier between 30 September and 3 October 2009 were read on 21 November while the surface was still snow covered at each stake (Fig. 1, Table 1). Stake 1 was located close to AWS1 and was used to assess point melt estimations from the different models. Snow density was measured using the standard Mount Rose procedure (U.S. Department of Agriculture, 1959) on the days of installation and re-measurement of stakes. We calculated the mean snow density (Table 1) and water equivalent (w.e.) surface ablation for each stake.
A Campbell Scientific SR-50 sonic ranging sensor was installed next to AWS1. The sensor recorded surface lowering continuously every 15 min during a 73-day period. SR-50 data were filtered using a Hampel filter (Pearson, 2002) and then hourly means were calculated. Lowering measurements were converted to w.e. ablation values using snow density measured at stakes (Table 1).
To derive snowline elevation, we used the MODIS/Terra L3 global daily snow cover product (MOD10A1, Hall et al., 2002) with a spatial resolution of 500 m, which retrieves subpixel fractional snow cover area. MOD10A1 was developed using a regression with the Landsat Thematic Mapper (30 m spatial resolution) Normalized Difference Snow Index (NDSI), offering a much more accurate approach for detecting snow covered area than previous satellite snow cover products (Cortés et al., 2014). In order to map the snow line throughout the monitored period, we obtained the hypsometric curve of the Tinguiririca Basin from an ASTER GDEM V2 with a resolution of 30 m (Tachikawa et al., 2011) and then calculated the snowline altitude for the austral summer of 2009–2010 in the upper Tinguiririca Basin. The MODIS snow cover product was used only if the cloud fraction for each satellite image was less than 30 %. The snowline elevation on days of high cloud cover was estimated using a linear interpolation between the last day before and the first day after the data gap. The time series of snowline elevation is used as a model input to define snow or ice surface areas on the glacier. We used the MOD10A1 product since it provides a reliable identification of the ice surface of Universidad Glacier, which is partially covered by debris and aerosols. The MOD10A1 product gives the fractional snow cover for each pixel in the range 0 to 100. To ensure a correct snowline altitude, we assumed the presence of snow in the pixel only when the fractional value was 100. However, we acknowledge some uncertainty in the snowline altitude.
We applied a standard degree-day model (DDM) (e.g. Hock, 2003, 2005) at an hourly time step, in order to estimate glacier surface melt during the 2009/2010 ablation season. The model was forced with hourly temperature data from AWS1.
Melt is calculated by multiplying the hourly positive temperature
Melt,
Boxplot showing the statistical distribution of hourly lapse rates calculated between AWS1 and AWS2 in the common period. Upper and lower box limits are the 75 and 25 % quartiles, the red horizontal line is the median, the filled circle is the mean, and crosses are outlying values.
To extend the model to a distributed scale (distributed DHM, DDHM hereafter),
we calculated the temperature lapse rate (LR) using both AWS in the common
period (Fig. 2). Following the recommendation of Petersen and
Pellicciotti (2011), we estimated a daily LR cycle (Fig. 3). The mean hourly
LR on an average day oscillates between
Using the hourly LR, we distribute air temperatures over the entire glacier surface on a 30 m grid at an hourly time step, using the ASTER GDEM V2 and the glacier outline which was digitized from an ASTER image of 27 March 2010. For October and November we assumed the same hourly lapse rate observed in the common period (December to March). Calculated melt values were not adjusted for reduction under debris cover on a medial moraine in the ablation zone.
A point-scale energy balance model (EBM hereafter) was applied using weather station data collected at the AWS1, between 1 October 2009 and 29 January 2010. We restricted use of data only up until this date because a sharp change in net radiation and incoming shortwave radiation occurred after 29 January; therefore data from late January onwards are of questionable accuracy.
Energy available for ablation,
Stability corrections were applied to turbulent fluxes using the bulk
Richardson number (
The estimation of the river discharge was based on the determination of the
cross-sectional geometry and the monitoring of water level in the proglacial
stream. Water level in the stream was monitored using a submersible pressure
transducer (KPSI Series 500), installed 500 m downstream of the glacier
terminus (2428 m a.s.l.), which registered hourly water levels from
24 November 2009 until 14 April 2010. The proglacial stream receives the
waters draining from a catchment with a total area of 86 km
Wind roses showing the hourly wind direction and the wind speed frequency at AWS1 (local time).
In order to convert automatic water level measurements into discharge, we
applied the widely used Manning equation (Phillips and Tadayon, 2006; Fang et
al., 2010; Gascoin et al., 2011; Finger et al., 2011) which combines
environmental parameters such as stream slope, bed roughness and river
section shape and area, for uniform open channel flow. It defines the
discharge
The geometry of the channel cross section was measured in the field at the
location of the pressure transducer. The hydraulic radius is a measure of
channel flow efficiency and is defined as the ratio of the cross-sectional
area to its wetted perimeter. We used the ASTER GDEM of 30 m resolution to
estimate a slope of 0.03
We also make use of two other streamflow gauge measurements (see Fig. 1). The
first is operated by a private company, Pacific HydroChile, located 1700 m
from the glacier snout and recording data every hour. The second one is
operated by DGA, and is located on the Tinguiririca River at 560 m a.s.l.,
50 km downstream from Universidad Glacier. The contributing watershed to
this lower gauge has an area of 1436 km
At each grid cell and time step, glacier melt obtained with the DDHM was
transformed into discharge using a linear reservoir model (Baker et al.,
1982; Hock and Noetzli, 1997). For hourly time intervals, the proglacial
discharge
Snowline elevation estimated using the MODIS snow cover product. The grey area corresponds to the altitude range of Universidad Glacier, the dashed line shows the equilibrium line altitude range estimated using an ASTER image of 27 March 2010 and black points show the AWS elevations.
Daily mean net radiation, incoming shortwave radiation, turbulent latent and sensible heat fluxes and the calculated energy available for melt at AWS1 (2650 m a.s.l.). On 21 and 22 November there are no data due to maintenance of AWS1.
Time series of meteorological variables are shown in Fig. 2. During the
December–March period, air temperature is almost constantly above
0
Mean monthly energy fluxes at AWS1.
Comparison of cumulative melt estimated by the point-scale degree-hour model (grey area between green lines), point-scale energy balance model, sonic ranger and stake 1 located near AWS1 (2650 m a.s.l.).
The snowline altitude derived from MODIS data is shown in Fig. 5. At the beginning of the ablation season, the entire glacier surface was covered by snow. The snowline altitude increased gradually until mid-January and thereafter stabilized between 3800 and 4000 m a.s.l. There is some variability in the snowline position, probably due to varying proportions of cloud cover on different days. This snowline altitude range, derived from the MODIS MOD10A1 snow cover product (Sect. 2.5), is slightly higher than the altitude of the ELA estimated with the ASTER image from the end of March of 2010 (3500 to 3700 m a.s.l., Fig. 1), possibly due to differences in spatial resolution in the two types of imagery. In the first half of the ablation season a high percentage of cloud cover (greater than 30 %) affected snowline detection.
Figure 6 shows the daily mean of observed energy fluxes (net radiation and
incoming shortwave radiation), turbulent fluxes calculated by the EBM (latent
and sensible heat) and the resulting energy available for melt at AWS1,
calculated by the model. Daily mean melt energy closely matches daily mean
net radiation through much of the ablation season due to compensation between
generally positive
Sonic ranger measurements and stake observations (Fig. 1) were compared to
melt estimated with the EBM and DHM at the location of AWS1 (Fig. 7).
Sublimation represents a small percentage (2.8 %) of the total ablation
calculated with the EBM, reflecting the predominantly positive air
temperatures and, hence, that ablation is dominated by melt. Snow disappeared
at this location (
Spatial distribution of cumulative glacier melt from Universidad
Glacier using two different
Melt simulations from the DHM and EBM agreed well with the stake and sonic
ranger ablation measurements. The DHM tended to lag behind the EBM and sonic
ranger until 21 November, after which the EBM and sonic ranger estimates fall
within the DHM range for
Total cumulative melt of Universidad Glacier using the degree-hour model. The red and blue lines and areas represent the cumulative melt at the locations of stakes 2 and 3, respectively. Points indicate the stake measurements. The area in grey enclosed by dashed black lines represents the lowest altitude of the glacier.
Time series of hourly discharge in the proglacial stream from the water level pressure sensor and the HydroChile gauging station.
Figure 8 shows the accumulated melt for each pixel of Universidad Glacier
estimated by the DDHM during the period 1 October 2009 to 31 March 2010 using
ice
During the study period we estimated an average streamflow of
12 m
The hourly mean hydrographs have strong daily amplitude cycles during the high discharge months (Fig. 10) and exhibit a characteristic shape for a glaciated catchment, with a steep rise and gradual decline (Nolin et al., 2010; Willis, 2011). Discharge peaked typically at 16:00, from a minimum at 10:00 which, considering the large size of the glacier, indicates an efficiently subglacial channelized drainage system flow typically of periods of dominant glacier ice melting (Willis, 2011).
At the hourly scale, water discharge estimated at the HydroChile station
showed high correlation with the values derived from the water pressure
sensor installed near the glacier front (
Comparison of cumulative melt calculated with the distributed degree-hour model (grey area) and streamflow measurements from the water level sensor data and the HydroChile station.
Total glacier melt calculated with the DDHM is compared with the discharge
records estimated from the pressure sensor and the gauging records from the
HydroChile station, at 500 and 1700 m from the glacier snout, respectively,
between 24 November 2009 and 31 March 2010 (Figs. 1 and 11). At an hourly
time step, glacier melt and proglacial discharge estimations have
correlations of 0.72 (pressure sensor station) and 0.75 (HydroChile station).
Melt estimated from the glacier represents between 42 and 58 % of the
streamflow estimated from the pressure sensor, depending on the ice
Daily mean melt from the distributed degree-hour model, and
discharge measurements from the water level sensor, HydroChile station and
DGA station. Mean daily air temperature at AWS1 is plotted on the right
Monthly melt from Universidad Glacier represents between 10 and 13 %
(depending on ice
The daily variability of all stream gauging series was similar between December and January. The DGA station measurements mainly show the additional influence of the air temperature variations on snowmelt across the catchment, since the rainfall in the period of Fig. 12 was 0 mm. In February and March, the DDHM calculated melt and the DGA station streamflow display similar temporal variations, with 1 to 2 days of lag.
Monthly discharge from Universidad Glacier as a percentage of the
total discharge in the Tinguiririca River, measured at the DGA station.
Ranges in the percentages are for
Our results suggest that a simple empirical melt model (DDHM) is suitable for estimating glacier melt contribution to streamflow from glaciers in the central region of Chile. This interpretation is based on the close correlation between melt estimates from the DHM and melt estimates from an energy balance model, ablation stake and sonic ranging sensor at a point scale, and agreement between estimates of total glacier runoff and discharge estimations in the proglacial stream. This good agreement results from, first, on-glacier measurements of meteorological data at two locations, enabling the use of a local hourly calibrated lapse rate to extrapolate air temperature inputs to the distributed melt model; second, locally calibrated degree-hour factors; and third, knowledge of the spatial distribution of snow and ice cover from satellite data. Forcing distributed temperature-index melt models with off-glacier data can be problematic due to the difficulty in estimating the temperature distribution across the glacier (Shaw, 2017; Shaw et al., 2017). At a point scale, a locally calibrated temperature-index model forced with off-glacier air temperature data can lead to improvement over use of on-glacier temperature data, due to damping of temperature within the glacier boundary layer (Guðmundsson et al., 2009). However, recent glacier studies have revealed high variability in the local air temperature lapse rate, due to variations in the strength and thickness of the katabatic boundary layer and changes associated with cloud cover and the synoptic-scale wind field (Petersen and Pellicciotti, 2011; Petersen et al., 2013; Ayala et al., 2015), which are difficult to account for in off-glacier data. Hence, the availability of temperature measurements for two on-glacier locations at different elevations provided suitable data for driving the DDHM.
Although we consider the model outputs to be robust, it is important to bear in mind that empirical temperature-index models do not attempt to simulate the real physical processes of glacier ablation, and the DDHM ignores other influences on rates and spatial patterns of ablation, such as topographic shading, blowing snow, debris cover and subsurface fluxes. Hence, the DDHM may not be suitable for longer-term mass balance studies where climatic and surface factors may undergo change.
The key sources of uncertainty in the results are the
following. The degree-hour factor of ice. A lack of stake measurements and only
a short period of sonic ranger data on ice mean there is some uncertainty in
a representative ice The snowline altitude derived from the MOD10A1
product. Although glacier surface characteristics on the tongue allow
differentiation between ice and snow, the resolution of the snow product is
similar to the width of the glacier tongue. A lag of 10 to 12 days was found
between the MOD10A1 product and field observations of the transition from
snow to ice at the AWS1 site (Fig. 7). Furthermore, in the highest zone of
the glacier, fewer debris and aerosols cover the ice surface, making it
harder to distinguish between ice and snow, which could have led to errors in
identifying surface type. DDHM melt estimates were not adjusted
for the effects of moraine and patchy distributed debris in the ablation zone
(Fig. 1). The moraines are of substantial thickness in lower areas of the
tongue and likely to reduce ablation below the highest values shown in Fig. 8
in the terminus zone. However, other areas of the ablation zone are affected
by a thin and patchy layer of debris or aerosol, which is likely to increase
ablation through local albedo reduction (Fyffe et al., 2014). Although
quantification of the effects of debris on melt is beyond the scope of this
study, it would be expected that impacts of thick morainic debris and thin
patchy debris elsewhere will tend to compensate in overall melt estimations
for the glacier. Snow density, which is required to convert stake
and ultrasonic sensor measurements of snow into w.e. melt for model
validation and calculation of degree-hour factors, was measured only two
times in the early ablation period. Single, fixed Sublimation was ignored in the DDHM. However, Universidad
Glacier has an ablation regime dominated by melt, more typical of temperate
glaciers further south in Chile (Brock et al., 2007); therefore, this
omission is likely to have led to only a small overestimate of glacier
runoff. Groundwater flows and evaporative losses from glacier melt water are
unknown but considered negligible. The date of the ASTER GDEM is
not known, which could have produced small errors in temperature distribution
due to elevation changes in the glacier surface between the dates of GDEM
acquisition and model analyses.
During periods of low positive temperature and high insolation, the DHM tends to underestimate melt, and vice versa during periods of high temperature, due to the high temperature sensitivity of simple temperature-index models (Pellicciotti et al., 2008). This implies spatial and temporal errors will occur, i.e. overestimation of melt during warm weather and on the lower glacier, and melt underestimation during cold weather and on the upper glacier. Such error will tend to compensate over time and in summation of total glacier melt, but will lead to short-term inaccuracies.
The finding that Universidad Glacier, while accounting for just 2 % of
the total basin area, contributed a monthly mean of between 10 and 13 %
of total streamflow from the entire upper Tinguiririca Basin over the
December 2009 and March 2010 period, demonstrates the importance of glaciers
for river flows in central Chile during the summer months. The overall
glacier melt contribution to the Tinguiririca River would be much larger
considering that the total glacier area of the basin is 81 km
The recent and ongoing retreat of Universidad Glacier is a direct consequence
of atmospheric warming (Le Quesne et al., 2009; Wilson et al., 2016) and the
relevance of glacier melt contribution highlighted in this work implies that
serious negative impacts on river discharge are expected over the next
decades. Considering that the estimated upward migration (200 m) of the zero
degree isotherm between 1975 and 2001 in central Chile (Carrasco et al.,
2005) far exceeds the elevational retreat (
From our analyses, it is impossible to assess whether the Tinguiririca River's discharge has already reached the “peak water” expected for glacierized basins as a consequence of deglaciation (Casassa et al., 2009). The observed recent positive trend in the discharge of the Tinguiririca River (Masiokas et al., 2006) suggests that peak water is yet to occur. In contrast, recent modelling work has shown that peak water has already passed further north in the Juncal Norte Basin and that future runoff is likely to sharply decrease (Ragettli et al., 2016). Estimations of the future runoff trend and melt contribution from Universidad Glacier are beyond the scope of this work. However, the possibility of increased persistence and recurrence of droughts in central Chile (Bosier et al., 2016) would increase the hydrological importance of Universidad Glacier in the future and therefore more research is needed in order to address these issues.
It has been shown that on high-altitude glaciers in northern Chile and during
the dry season of the outer tropics of Peru and Bolivia, melt rates are
reduced as more ablation occurs through sublimation (Winkler et al., 2009;
Sagredo and Lowell, 2012; MacDonell et al., 2013). On the other hand, to the
south of
Although melt rates cannot be compared directly between different glaciers in
different years, two other studies in Chile provide a reference for the DDHM
results for Universidad Glacier in the 2009–2010 season. Pellicciotti et
al. (2014) estimated the total melt in the lower ablation zone of Juncal
Norte Glacier (33
Recently Ayala et al. (2016) showed that the glacier melt contribution at the
river outlet of glaciers Bello, Yeso (debris-free glaciers) and Piramide
(debris-covered glacier) in the central Andes (
At a basin scale, glacier contribution to downstream discharge in the
Tinguiririca River is of a similar magnitude to previous results for the
central Andes. For example, Ragettli and Pellicciotti (2012) estimated that
14 % of the total streamflow of the Juncal River Basin (241 km
In this study, we have investigated the meteorological conditions, ablation and melt water contribution to downstream river flow of Universidad Glacier, located in central Chile, during the 2009–2010 summer ablation season. We used a point-scale energy balance and a distributed degree-hour melt model, driven by data from two on-glacier weather stations. The main outcomes of this work are the following.
The distributed degree-hour model provides a robust simulation of surface melt, especially on the glacier tongue where good agreement was found between melt estimated from the point-scale degree-hour model, energy balance model, ablation stake measurements and sonic ranger records. Almost continuously positive air temperatures in the ablation zone between November and March are appropriate for the application of a simple temperature index method to calculate glacier melt; however, some melt overestimation was identified for the accumulation zones due to more frequent negative air temperatures at higher elevations.
Meteorological conditions result in very high ablation season melt totals, which reach 10 m w.e. on the lower tongue. This finding is attributed to the high insolation due to a low percentage of cloud cover, combined with a predominantly positive air temperature.
By comparing total glacier melt with discharge measurements at 50 km
downstream on the Tinguiririca River, we estimate that the monthly mean
contribution of Universidad Glacier is between 10 and 13 % of the
streamflow observed in the upper Tinguiririca Basin for the period
December 2009 to March 2010. This estimated contribution reaches a maximum of
15 to 20 % in March. The total contribution of all glaciers to streamflow
in the upper Tinguiririca Basin will be considerably larger considering that
Universidad Glacier only represents 36 % of the total glacier area of the
basin (
The successful application of a simple temperature-index melt model to estimate total seasonal melt at Universidad Glacier is partly a consequence of the predominant high melt regime of this glacier, which favours the application of the degree-hour model. In this sense, estimation of streamflow contributions from glaciers in northern Chile is more challenging as an increasing proportion of ablation energy is consumed by sublimation (MacDonell et al., 2013) which cannot be estimated from simple temperature-index methods.
Climatic warming, leading to a rapid rise in the zero-degree isotherm (Carrasco et al., 2005) and upward expansion of glacier melt contributing area into the accumulation zone, means Universidad Glacier will continue to make a crucial, and perhaps an increasing contribution to downstream flows in the next few decades, particularly as smaller glaciers in the basin disappear. In the long term, glacier shrinkage will lead to a depletion of glacier melt and in downstream streamflow in the Tinguiririca River, particularly in late summer. This will have severe implications for human activities in the river valley such as mining, domestic consumption, industry, tourism, forestry and agriculture (Aitken et al., 2016) and hydropower generation (Valdés-Pineda et al., 2014). Hydropower generation on the Tinguiririca River at La Higuera and La Confluencia (Pelto, 2011) will be affected by interannual variability in water supply and future streamflow trends in the medium to long term. Finally, more long-term high-elevation stations in the Andes are necessary to establish the interannual variability of glacier contribution to river discharge in order to help manage future water availability, considering climate change and the increasing demand for water in the region (Meza et al., 2012).
The data used in this study are presented in the figures. Data and metadata from Automatic Weather Stations (Figs. 2, 4 and 6) are available upon request from the corresponding author.
The authors declare that they have no conflict of interest.
This work was supported by CECs, which is funded by the Chilean Government through the Centers of Excellence Base Financing Program of Comisión Nacional de Investigación Científica y Tecnológica de Chile (CONICYT). Pablo Zenteno and Camilo Rada assisted with data collection. The Dirección General de Aguas de Chile (DGA) also provided data and support to this paper. We would like to thank Christophe Kinnard for sharing his data. We would like to thank Fabian Drenkhan and one anonymous referee for their constructive and useful comments and recommendations. Edited by: Jan Seibert Reviewed by: Fabian Drenkhan and one anonymous referee