1 Sara Knox Characterizing the geospatial features and hydrology of four pro-glacial valleys in the Cordillera Blanca, Peru McGill University EPSC 480: Honors Research Project Jeffrey McKenzie April 2009
1
Sara Knox
Characterizing the geospatial features and hydrology of four pro-glacial valleys in the Cordillera Blanca, Peru
McGill University EPSC 480: Honors Research Project
Jeffrey McKenzie April 2009
2
INTRODUCTION
Many large rivers throughout the world originate in mountainous regions. High
mountains and their glaciers, such as the Himalayas, the Andes, and the Alps, supply
water to hundreds of millions of people and assure water throughout the year (Coudrain
et al. 2005). In the outer-tropics, which experience highly seasonal precipitation patterns,
glaciers are significant water reservoirs that buffer river runoff by providing melt-water
throughout the year (Mark and Seltzer 2005). Many rivers draining glaciated basins,
especially in the Himalaya-Hindu Kush (HKH) region, the South-American Andes, New
Guinea, and East Africa, are sustained by glacier runoff during the dry season (IPCC
2007, Bradley 2006, Barnett et al. 2005).
Studies worldwide indicate an extensive retreat of mountain glaciers in the non-
polar regions of the world throughout the 20th century, with a marked increase in recent
decades (Coudrain et al. 2005, Kaser 1999). The retreat of tropical and subtropical
glaciers is of particular concern since this is where the majority of the world’s population
is located and where the greatest biodiversity is found (Mark and Seltzer 2005). As
glaciers retreat, there is an initial increase in discharge in the short term, but as they
continue to decrease in size they become less significant seasonal reservoirs (Mark and
McKenzie 2007, Coudrain et al. 2005, Mark et al. 2005). In the tropics, glacier recession
will likely result in more variable seasonal stream flow, with a loss of glacial buffering
during the dry season (Mark and McKenzie 2007, Mark et al. 2005). Droughts in the dry
season are therefore predicted for regions which rely heavily on glacial melt water as
their primary dry-season water supply (IPCC 2007).
3
Peru is an example of a country which will suffer significantly from glacier
recession. Most of the country’s population lives along the Pacific coastal plains and
western slopes of the Andes, where the land is arid and heavily dependant on runoff from
the mountains (Vergara et al. 2007). Furthermore, Peru’s economically profitable
agricultural sector is highly water intensive, with large demands for irrigation water from
the desert coastal region which is already water stressed (Spang 2006). The majority of
Peru’s most glacierized mountain range, the Cordillera Blanca (Figure 1), drains to the
Pacific via tributary streams to the Rio Santa (Mark and McKenzie 2007). The Rio Santa
and its tributaries provide an important source of water for hydroelectric power
generation, irrigation, mining, industrial, and domestic purposes (Spang 2006). Modeling
tropical pro-glacial hydrology is therefore greatly important in predicting the future
impact of glacial retreat on water resources.
Several hydrological models have been developed to simulate pro-glacial
discharge across the Cordillera Blanca (Juen et al. 2007, Pouyaud et al. 2005). Although
these models have been generally successful in reproducing discharge in catchments
where melt water is the dominant input, they tend to highly simplify groundwater
contribution mechanisms. However, studies have shown that during the dry season,
groundwater often represents the largest hydrological input to stream discharge in some
glacierized basins (Baraer et al. 2009). Baraer et al. (2009) found that although
groundwater contributes just slightly over 10% to the total water budget of the
Querococha valley (Figure 1), during the dry season, groundwater is the largest
contributor to basin outflow; in 1999 and 2007, the relative contribution of groundwater
to dry season discharge was 68% and 79% respectively. Therefore, in modeling pro-
4
glacial systems it is important to accurately incorporate groundwater contributions and
pathways in addition to glacial melt and runoff.
A potential location for the temporal storage of groundwater in the Cordillera
Blanca is pampas; treeless plains that are formed from paludified moraine-dammed lakes
(Mark and McKenzie 2007). The pampas are characterized by low permeability and
organic rich unconsolidated material (Mark and McKenzie 2007). Although very little
literature exists on such geomorphic features, similar formations are observed in other
regions of the South American Andes. “Bofedale” is the local name given to valley
bottom formations in the Cordillera Real, Bolivia which are composed of basal moraines
and fluvio-glacial deposits overlain by peat (Caballero et al. 2002). The precise
hydrologic role of the pampas is not entirely understood and merits further study.
To address the pressing concern posed by glacier recession, Baraer et al. (2009)
are developing hydrologic models that can include climate change related impacts on
water resources at the scale of the Cordillera Blanca. A major constraint in the model
development is the poor understanding of tropical pro-glacial hydrology in large (area
greater than 5000 km2), multi-valley, glacial fed catchments. To quantify and gain
insight into pro-glacier hydrology in the tropics, Baraer et al. (2009) used a three
component mixing model, called the hydrochemical basin characterization method
(HBCM), to quantify the contribution of glacier melt water, groundwater and surface
runoff to streams for four valleys and sub-basins within the Cordillera Blanca, Peru
(Figure 1). To interpret the results of the HBCM within the geographical context and
develop a better understanding of the factors that control the hydrological processes
5
within and across catchments, the geospatial characteristics of the watersheds and their
sub-basins must be studied.
The objectives of this study are to 1) Map the total area of the pampas in the pro-
glacial valleys of the Cordillera Blanca due to the potential hydrological importance of
these features 2) Characterize the basins and sub-catchments selected for the HBCM
analysis by compiling geospatial data including watershed size, glacierized area, glacier
hypsometry, and pampa area 3) Correlate the geospatial data with the HBCM results for
the 2008 dry season in an attempt to explain the relative groundwater and melt water
contributions observed in each valley and nested watershed, and investigate basin
characteristics that could potentially be used to predict the hydrology of pro-glacial
valleys in the Cordillera Blanca.
STUDY AREA
The most glacierized tropical mountain range on Earth, the Cordillera Blanca,
spans over 130km between 8º-10º S latitude in Peru (Vuille et al. 2008). As described
above, the majority of the glacierized area within the Cordillera Blanca discharges to the
Pacific Ocean via tributary streams to the Rio Santa, which also receives runoff from the
nonglacierized Cordillera Negra mountain range. The Rio Santa originates at Lake
Conococha and flows northwest over 300km, draining a total watershed of 12 200km2.
The upper Rio Santa watershed, the Callejon de Huaylas, is 4900km2 and is delimited by
the hydroelectric power plant at Huallanca (1800 m.a.s.l.) (Figure 1). It is estimated that
10-20% of the total annual discharge, and 40% of the dry season discharge, in the
Callejon the Huaylas is comprised of glacial melt (Mark et al. 2005).
6
The glacierized Cordillera Blanca has undergone a significant reduction in glacier
volume throughout the past century. The total glacierized area of the Cordillera Blanca is
presently estimated to be 570 km2 based on 2003 SPOT images, an overall loss of
glacierized area of 22.4% since 1970 (Racoviteanu et al. 2008a). In addition, since 1970
there has been an average rise in glacier terminus elevations of 113 m and an average rise
in the median elevation of glaciers of 66m. Finally, the number of glaciers has increased,
which is indicative of disintegration of ice bodies (Racoviteanu et al. 2008a).
Historical discharge records are available for about 40 years in some tributaries
that flow to the Rio Santa. Monthly average discharge from these gauged streams
discharging into the Callejon de Huaylas is higher during the months of October-April,
closely reflecting the seasonality of precipitation typical of the outer tropics (Mark and
Seltzer 2003). Over 80% of precipitation falls between October and May and the austral
winter months of June to September have nearly no precipitation (Mark and McKenzie
2007). Conversely, monthly averaged air temperature remains nearly constant
throughout the year, and the diurnal variation exceeds the annual variation as is
characteristic of the tropics (Kaser and Georges 1999).
The Cordillera Blanca is approximately 10Ma and sits on the magmatic Andean
arc caused by the subduction of the Nazca plate under the South American plate. The
western side of the Cordillera Blanca is underlain by bedrock exhumed by a NW-SE
trending normal fault, and consists of 80% to 90% batholith, the rest being isolated
regions of tonalite and diorite (Baraer et al. 2009). The lithology of the batholith is
characterized by high silicate contents. The eastern side of the Cordillera Blanca is
comprised of the Jurassic sedimentary Chicama formation composed of weathered rock
7
structures including shale, argillite, and sandstone, and some areas are characterized by
extensive pyrite (Baraer et al. 2009).
To investigate the similarities and differences in the hydrology of pro-glacial
valleys in the Cordillera Blanca, the catchments of Llanganuco, Quilcayhuanca,
Querococha and Pumapampa (Figure 1) were selected for the HBCM analysis. They are
located across the geographical extent of the mountain range; Llanganuco is located at
the northern end of the Cordillera Blanca where the valleys are deeply incised and the
valley walls are very steep, and Pumapampa is situated at the southern end where the
valley walls have a gentler topography and there is a smaller elevation gradient between
the valley bottom and the watershed divide.
METHODS
Data Sources
To derive the total pampa area for the Cordillera Blanca and geospatial
parameters of the four catchments and selected sub-basins, a digital elevation model
(DEM), satellite imagery, georeferenced geology maps, and glacier shapefiles of the
Cordillera Blanca were processed within a geographic information system (GIS: ArcGIS
9.3). The DEM is an ASTER DEM product with 30m cell size. The projection for the
DEM and satellite imagery is Universal Transverse Mercator (UTM) Zone 18S. The
satellite imagery is also from the ASTER sensor which was launched in 1999 and
acquires high spatial resolution data in 14 bands, from the visible to the thermal infrared
wavelengths (NASA-JPL 2009). Nine scenes, acquired between 2000 and 2005, were
needed to cover the Cordillera Blanca. Bands 1 (520 – 600 nm) and 2 (630 – 690 nm) are
in the visible portion of the electromagnetic spectrum (VIS) and band 3 (760 – 860 nm) is
8
in the near-infrared (NIR) (NASA-JPL 2009). The spatial resolution of the sensor for
these bands is 15m. Glacier shapefiles, derived from 2003 SPOT5 imagery, covering
90% of the Cordillera Blanca were obtained from the GLIMS Glacier Database
(Racoviteanu 2008b). The Pumapampa basin is located outside the SPOT5 imagery thus
glacier coverage for this watershed was digitized from a 2002 ASTER scene. Two
1:1000000 scale geology maps published by the Peruvian Insituto de Geologia y Mineria
were georeferenced and projected to UTM Zone 18S and used as the basis for the
geology analysis.
Pampa delineation
As noted above, pampas are treeless plains located in valley bottoms and thus are
characterized as regions with a low gradient. To determine regions of low grade, a slope
layer was derived from the DEM using grid-based modeling. A semi-automated method
was then used to delineate the pampas: Pampas represent regions with a slope less than or
equal to 10º, an elevation greater than 3500m, and which are found within glacierized
valleys surrounded by steep slopes. Determining the location of the pampas was further
refined by examining a false color composite of the ASTER scenes and excluding regions
where no vegetation was present.
Geospatial analysis
The hydrological tools of ArcGIS 9.3 were used to delineate the valleys and sub-
basins corresponding to those of the HBCM. The pour-points for catchments and sub-
basins were based on GPS mapped mixing point locations used for the HBCM analysis
(Figures 2-5). The glacierized area in each catchment and sub-basin was determined
from the GLIMS glacier shapefiles (and digitized 2002 ASTER scene for the Pumapampa
9
basin) using the Analysis tools of ArcGIS 9.3. For glacierized area found in each
catchment, grid-based modeling and zonal functions were used to calculate its minimum,
maximum, mean, and median elevation, average slope and average aspect. The
hypsometric profiles of the glacierized areas were determined using the Spatial Analyst
toolbar. The Analysis tools were also used to determine the area of pampa in each basin.
The average length to width ratio of the pampas and the total river length flowing through
the pampas were determined using the Measure tool. The geologic characterization of
each catchment was digitized from the georeferenced geology maps.
The hydrological data on the contribution of groundwater and melt water to 2008
dry season discharge (June and July) for sub-basins in the Llanganuco, Quilcayhuanca,
Querococha and Pumapampa valleys was obtained from the HBCM analysis (Baraer et
al. 2009). The geospatial parameters for catchments and nested watersheds were then
correlated with the results from the HBCM analysis.
RESULTS AND DISCUSSION
Geospatial data
The geospatial data for the pro-glacial catchments of interest are summarized in
Table 1. The valleys differ considerably with regards to glacierized area and percentage
surface area covered by glaciers (Figure 6). The degree of glacierization is greatest in the
most northern catchment (35.6%) and decreases in the southern valleys, although
Querococha has the lowest percent watershed area glacierized (2.8%). Larger glacierized
areas have a higher maximum elevation and a lower minimum elevation (ie. greater
altitudinal range), as well as a higher median elevation (Table 1). The glacierized area in
the Querococha basin is found entirely below the snowline elevation of the Cordillera
10
Blanca (~5200 m) (Klein et al. 1995; Mark and Seltzer 2005) (Figure 7). Conversely, the
majority of the glacierized area in the Llanganuco is located slightly above 5200 m
(Figure 7). The hypsometric curves for the Quilcayhuanca and Pumapampa basins show
that the majority of the glaciated area in these valleys is located around the same
elevation (just below 5200 m) but that a larger mass of ice is present at higher elevations
in the Quilcayhuanca valley. The glacierized areas within the basins also have varying
average slopes. Llanganuco has glaciers with the highest average slopes (31º),
Quilcayhuanca and Querococha contain glaciers with similar average slopes (~28º), and
the glacierized area in the Pumapampa basin has the lowest average slope (26º). These
results are similar to the average slope of all glaciers in the Cordillera Blanca (32º)
(Racoviteanu et al. 2008a). The average aspect for the glacierized area in most
watersheds is southwest for except for Llanganuco where the average aspect of the
glaciers is south (Table 1). This is consistent with Racoviteanu et al. (2008a) who found
that the average orientation of all glaciers in the Cordillera Blanca is 193º (southwest).
The total area of pampas, as defined above, in the glacierized valleys of the
Cordillera Blanca is 63.6 km2. The pro-glacial valleys located at the southern end of the
Cordillera Blanca have a greater area covered by pampas than those located at the
northern end (Figure 1). For instance, 11.6% of the Pumapampa basin is covered by
pampas versus 0.8% for the Llanganuco watershed (Figure 8). Basins with a higher
percentage of pampas also have a greater total river length flowing through the pampas
(Table 1). The geometry of the pampas differs between the basins; Quilcayhuanca has
the longest and most narrow pampas, followed by Llanganuco, whereas the length to
width ratio of the pampas in Querococha is the smallest (Table 1).
11
Approximately 60% of the Llanganuco watershed is composed of the Chicama
formation and plutonic intrusion comprises the remainder of the basin (Figure 9). In the
Quilcayhuanca watershed, the Chicama formation underlies a much greater proportion of
the basin (86%) and the plutonic intrusion, mainly found in the southwest end of the
catchment, is present in the rest of the basin (Figure 9). In the Querococha watershed, the
Chicama formation makes up 69% of the basin, plutonic intrusion is present around the
southern boundary, comprising 22% of the basin, and 9% of the basin is underlain by
quaternary fluvial glacial sediments which are found along the southwest border of the
basin (Figure 9). The Pumapampa basin has a significantly different geological setting
than the three other basins; the Chicama formation underlies only 15% of the catchment,
the majority of the basin (66%) is composed of volcanic ignimbrite calipuy and the rest
of the watershed is comprised of cretaceous sedimentary rocks (Figure 9).
The valleys described above were sub-divided into nested watersheds
corresponding to the mixing points used in the HBCM (Figures 2-5). The geospatial data
for the sub-basins are presented in Tables 2 and 3.
HBCM results for the 2008 dry season
The HBCM was used for most of the mixing points within the partially glaciated
Llanganuco, Quilcayhuanca, Querococha and Pumapampa basins to determine the
absolute contribution of groundwater and melt water to 2008 dry season discharge. Due
to restricted water sampling throughout the Pumapampa and Llanganuco valleys and
limitations of the HBCM, the end-member analysis in these basins is restricted to a
limited number of sub-basins representing 30% and 38% of the total basin area
respectively. The absolute average groundwater and melt water contribution to total
12
discharge at the outlet of each basin is summarized in Figure 10. Melt water contribution
is greatest in the Llanganuco valley (0.90 m3/s), and is considerably lower in the
Quilcayhuanca (0.41 m3/s) and Querococha (0.06 m3/s) basins. The HBCM analysis was
unable to quantify the melt water contribution to surface water in the Pumapampa basin.
Conversely, groundwater contribution is lowest in the Llanganuco basin (0.045 m3/s) but
greater in the other watersheds, contributing between 0.15 m3/s and 0.19 m3/s to the
discharge at the Pumapampa and Querococha outlets. The absolute groundwater and
melt water contribution to discharge at the sub-basin level is summarized in Tables 4 and
5.
Correlation between geospatial parameters and HBCM results
1. Melt water
At the sub-basin level, there are trends which can be observed between geospatial
parameters and melt water contribution to dry season discharge. Firstly, larger sub-basins
have a greater melt water contribution (Figure 11). Melt water discharge also increases
with larger amounts of glacier coverage (Figure 12). Additionally, there is a strong
positive correlation between the size of the ablation zone (roughly the glaciated area
below 5200 m) and melt water contribution (Figure 13). The elevation data for the
glaciated area in the nested watersheds is also correlated with melt water contribution;
basins whose glaciers have a higher maximum and median elevation, a greater elevation
range and a lower minimum elevation also have greater melt water discharge. However
these parameters are strongly related to glaciated area and thus the latter feature is likely
the predominant factor contributing to higher melt water production. The average slope
and aspect of the glaciated area is not correlated with melt water production. Based on a
13
stepwise multiple regression analysis for various combinations of geospatial parameters,
only the glaciated area in each sub-basin significantly predicts melt water discharge (P <
0.01), explaining 84% of the variability in the minimum melt water production and 85%
of the variation in minimum melt water contribution.
At the catchment scale, the relationships between geospatial parameters and melt
water contribution are constrained by sample size, and trends which emerge at the sub-
basin level are not evident at the larger scale. For example, there is no apparent
relationship between basin area, glaciated area or ablation area and absolute melt water
contribution. This is likely explained by the fact that melt water discharge in the
Llanganuco basin could only be quantified for 38% of the entire catchment. However, at
the valley scale, melt water contribution increases with larger amounts of glacier
coverage (Figure 14). This can’t be assessed at the sub-basin level since melt water
originates from catchments that are assumed to be 100% glaciated. Again average slope
and aspect of the glaciers doesn’t appear to influence melt water production.
Kaser and Osmaston (2002) found that melting occurs predominantly in the
ablation zone below the equilibrium line altitude (ELA), thus the glaciated area below the
snow line, controlled by hypsometry, should be the predominant factor influencing melt
water contribution. However, total ice covered area in the nested watersheds is the only
parameter which was found to significantly predict melt water contribution. This
suggests that melting could potentially occur above the snow line. Conversely, the fact
that the glacierized area below the ELA isn’t the predominant factor controlling melt
could be due to the limited number of sub-basins and the resolution of the DEM. More
extensive sampling in the valleys would allow this analysis to be conducted for a larger
14
number of sub-basins. Local topographic factors such as slope and aspect also influence
melting rates (Racoviteanu et al. 2008a). Slope and aspect were averaged for the entire
ice mass within the catchments and not for individual glaciers, thus there is considerable
variability within these parameters which likely explains why they are not significant in
predicting melt water discharge. It is important to acknowledge that the estimates of
glacier elevation, slope, aspect and hypsometry are highly sensitive to the quality of the
DEM used for this analysis.
Although physical parameters such as total ice covered area and glacier
hypsometry directly influence melt water contribution (Mark and Seltzer 2003), other
factors such as precipitation, humidity, cloud cover, and albedo greatly influence ablation
(Vuille et al. 2008, Mark and Seltzer 2005). This is further complicated by climate
variability at decadal time scales, such as El nino events which tend to result in enhanced
melt (Coudrain et al. 2005). Therefore both physical and climate parameters should be
considered when quantifying the contribution of melt water to total discharge.
2. Groundwater
The only geospatial parameter which is related to groundwater contribution at the
sub-catchment level is basin area (Figure 15a). When examining this trend within each
valley, sub-basin area appears to influence groundwater production to a differing degree
(Figure 15b). However this observation is limited by sample size and thus requires
further investigation. At the valley scale, groundwater contribution does not appear to
increase with increasing catchment area (Figure 16). This is likely explained by the fact
that groundwater contribution could only be quantified for a limited number of sub-basins
in the Llanganuco and Pumapampa valleys and thus groundwater discharge isn’t
15
representative of the entire catchment area. A multiple regression analysis revealed no
geospatial parameter which significantly predicts groundwater contribution.
The absence of correlations between pampa parameters and groundwater
contribution to surface water suggests that pampas - defined as paludified moraine
dammed lakes - may not be temporal storage sites for groundwater. This is also
supported by their unique major ion chemistry which is inconsistent with surface water
and melt water hydrochemistry (Baraer et al. 2009), and their low hydraulic conductivity.
Potential sources of groundwater might be slope deposits, including lateral moraines and
talus slopes, and sand and gravel aquifers in valley bottoms. Rock faces are typically
made of massive batholiths and therefore are not a potential storage site.
CONCLUSION
The valleys of Llanganuco, Quilcayhuanca, Querococha and Pumapampa have
considerably different geospatial and hydrological characteristics. Notably, the more
northern catchments have a higher degree of glacierization, but a smaller pampa area than
the southern basins. Melt water production decreases from the most northern to the most
southern valley, whereas groundwater production is similar in the basins of
Quilcayhuanca, Querococha, Pumapampa, but considerably lower in the Llanganuco
catchment. However, melt water and groundwater contribution in the Llanganuco and
Pumapampa basins could only be quantified for 38% and 30% of the entire catchment
area respectively (Baraer et al. 2009). The geospatial parameters were analyzed in
conjunction with the HBCM results to investigate physical features which could explain
the differences in hydrology. Only the glaciated area in each sub-basin was found to
significantly predict melt water discharge. Ablation area, slope and aspect are known to
16
influence melt water production, however these features were not found to significantly
predict melt water contribution presumably due to the limited number of sub-basins and
coarseness of the slope and aspect data. Future work should involve repeating this
analysis for a larger number of sub-basins in addition to determining slope and aspect
parameters at the glacier level rather than for the entire glaciated area in the sub-basins.
The hydrogeology of glaciated terrains is very complex and a key feature of pro-
glacial systems is the variability in hydraulic conductivity (Fetter 2001, p.285). Deposits
formed in association with glaciers range from till with mud-textured matrix to well-
sorted glacio-fluvial sands and gravels which are of particular hydrogeologic importance
(Robinson et al. 2008). Understanding the hydrogeology of valleys in the Cordillera
Blanca is further complicated by the numerous glacial advances and retreats since the last
glacial maximum and the heterogeneous geology of the Cordillera Blanca. Thus, since
no geospatial parameters significantly predict absolute groundwater contribution, further
hydrogeological investigations in these pro-glacial catchments are warranted to provide a
better understanding of the groundwater processes in these complex environments.
17
Tables
Table 1. Summary of geospatial parameters for four valleys in the Cordillera Blanca including basin area (km2), degree of glacierization (km2 and %), elevation (m), slope (º), and aspect (º) of glaciated area, pampa area (km2), length to width ratio of pampas and total length of river flowing through the pampas (km).
Valley
Basin Area (km2)
Glaciated Area (km2)
% glaciated
area
Minimum elevation
of glaciated area (m)
Maximum elevation
of glaciated area (m)
Median elevation
of glaciated area (m)
Maximum-minimum elevation
of glaciated area (m)
Average slope (º)
of glaciated
area
Average aspect (º) of
glaciated area
Area of Pampa
in basin (km2)
% Pampa
area
Average length
to width ratio of pampas
Total river
length through pampa (km)
Llanganuco 91.05 32.43 35.6 4377 6577 5258 2200 30.7 182.0 0.74 0.8 4.65 3.67 Quilcayhuanca 87.66 17.98 20.5 4353 6180 5139 1827 27.9 217.2 3.55 4.1 13.17 15.42 Querococha 62.92 1.75 2.8 4647 5181 4946 534 28.1 208.1 2.54 4.0 1.64 9.64 Pumapampa 69.16 9.34 13.5 4704 5462 5128 758 25.6 222.2 8.00 11.6 4.20 19.97
Table 2. Summary of the glacier parameters in selected sub-basins of the valleys of interest. Glacier parameters include degree of glacierization (km2 and %), elevation (m), slope (º), and aspect (º) of glaciated area.
Valley Sub-basin
Basin Area (km2)
Glaciated Area (km2)
% glaciated area
Glaciated Area below
5200 m (km2)
Minimum elevation
of glaciated area (m)
Maximum elevation of
glaciated area (m)
Median elevation
of glaciated area (m)
Maximum-minimum
elevation of glaciated area (m)
Average slope (º)
of glaciated
area
Average aspect (º)
of glaciated
area Llanganuco Kinzl Up 17.35 12.00 69.2 4.76 4377 6577 5319 2200 32.8 214.4 Quilcayhuanca Laguna Cuchilla 4.16 2.46 59.1 1.03 4735 6063 5259 1328 33.5 202.1 Quilcayhuanca Laguna Tulipacocha 15.18 9.12 60.1 5.60 4353 6180 5138 1827 25.8 221.3 Querococha Yan out 1.50 0.76 50.6 0.76 4647 5128 4896 481 26.6 197.0 Querococha YOG up 1.81 0.91 50.0 0.91 4816 5181 4973 365 26.1 222.9
18
Table 3. Summary of the pampa parameters in selected sub-basins of the valleys of interest. Pampa parameters include pampa area (km2), percentage of sub-basin covered by pampas, length to width ratio of pampas, total length of river flowing through the pampas (km) and area pampa in the above (upstream) sub-basin.
Valley Sub-basin Basin
Area (km2)
Area of Pampa in
basin (km2)
%Pampa area
Average length to
width ratio of pampas
Total river length
through pampa (km)
Area of pampa
(km2) in sub-basin
above
%Pampa area in
sub-basin above
Llanganuco Kinzl 6.29 0.00 0.00 Quilcayhuanca Casa de agua 7.56 0.38 3.37 1.69 1.35 0.49 16.38 Quilcayhuanca Cayesh Low 6.10 0.45 7.33 7.93 2.60 0.16 1.27 Quilcayhuanca Quilcay above Choco 2.98 0.49 16.38 3.48 1.61 1.11 7.85 Quilcayhuanca Quilcay Out 20.82 1.33 11.86 14.28 5.37 0.38 3.37 Quilcayhuanca Valley2 8.03 0.66 2.07 3.27 3.36 Querococha Q2-2 6.42 0.48 7.45 3.17 1.65 Querococha Q2-3 7.13 0.62 8.75 0.84 1.49 Querococha Q2-1 10.06 0.19 1.91 2.17 6.25 1.10 8.13 Pumapampa Above Huayllo 3.91 0.32 8.27 1.81 0.76 1.21 10.26 Pumapampa Puma Q stn 15.49 2.99 19.31 4.70 9.46 0.32 8.27 Pumapampa Pumapampa below Ruri 1.02 0.30 3.42 5.62 0.99
19
Table 4. HBCM results for the 2008 dry season maximum and minimum melt water contribution (m3/s).
Table 5. HBCM results for the 2008 dry season maximum and minimum groundwater contribution (m3/s).
Valley Sub-basin
minimum groundwater
contribution (m3/s)
maximum groundwater
contribution (m3/s)
Llanganuco Kinzl 0.04 0.04
Quilcayhuanca Casa de agua 0.01 0.04
Quilcayhuanca Cayesh Low 0.02 0.03
Quilcayhuanca Quilcay above Choco 0.00 0.00
Quilcayhuanca Quilcay Out 0.06 0.11
Quilcayhuanca Valley2 0.03 0.03
Querococha Q2-1 0.10 0.11
Querococha Q2-2 0.04 0.04
Querococha Q2-3 0.05 0.05 Pumapampa Above Huayllo 0.02 0.09 Pumapampa Puma Q stn 0.06 0.07 Pumapampa Pumapampa below Ruri 0.03 0.03
Valley Basin minimum melt water contribution (m3/s)
Maximum melt water contribution (m3/s)
Llanganuco Kinzl Up 0.90 0.90
Quilcayhuanca Laguna Cuchilla 0.07 0.07
Quilcayhuanca Laguna Tulipacocha 0.32 0.35
Querococha Q2-2 0.02 0.03
Querococha Q2-3 0.03 0.03
20
FIGURES
Figure 1. Map of Callejon de Huaylas with catchments and pampas.
21
Figure 2. Map of sub-basins and mixing points in the Figure 3. Maps of sub-basins and mixing points in the
Llanganuco catchment. Quilcayhuanca catchment.
22
Figure 4. Map of sub-basins and mixing points in the Figure 5. Maps of sub-basins and mixing points in the
Querococha catchment. Pumapampa catchment.
23
0.00
5.00
10.00
15.00
20.00
25.00
30.00
35.00
40.00
Llanganuco Quilcayhuanca Querococha Pumapampa
Glaciated Area (km2)
% glaciated area
Figure 6. Glaciated Area (km2) and percentage glacierized area based on 2003 SPOT5 imagery (2002 ASTER imagery for the Pumapampa valley) for four pro-glacial valleys in the Cordillera Blanca, Peru.
4000
4400
4800
5200
5600
6000
6400
6800
0 10000 20000 30000 40000 50000 60000 70000
m
Llanganuco
Quilcayhuanca
Pumapampa
Querococha
Figure 7. Hypsometric curves for the glaciated area in four pro-glacial valleys in the Cordillera Blanca, Peru.
Area (m2)
24
0.00
2.00
4.00
6.00
8.00
10.00
12.00
14.00
Llanga
nuco
Quilc
ayhuan
ca
Quer
ococh
a
Pumap
ampa
Pampa Area (km2)
Pampa Area/Basin
Area (%)
Figure 8. Area of pampas (km2) and percentage of the basin covered by pampas for four pro-glacial valleys in the Cordillera Blanca, Peru.
Figure 9. Geological characterization of the A) Llanganuco B) Quilcayhuanca
D
15%
66%
19%
Chicama
Formation
Volcanic
Ignimbrite
Calipuy
Cretaceous
Sedimentary
Rocks
B
14%
86%
Plutonic
Intrution
Chicama
Formation
C
25%
40%
35%
Plutonic
Intrution
Chicama
Formation
Quaternary
Fluvial Glacial
Sediments
A
60%
40%
Plutonic
Intrusion
Chicama
Formation
25
C) Querococha D) Pumapampa valleys.
26
Figure 10. Absolute contribution (m3/s) of average melt water and groundwater to total 2008 dry season discharge at the outlet of the Llanganuco, Quilcayhuanca, Querococha and Pumapampa catchments.
y = 0.0378x
R2 = 0.7601
0
0.2
0.4
0.6
0.8
1
1.2
0 5 10 15 20 25 30
Sub-basin Area (km2)
Min
imu
m m
elt
wa
ter
con
trib
uti
on
(m
3/
s)
Figure 11. Relationship between sub-basin area (km2) and minimum melt water contribution (m3/s).
27
y = 0.0597x
R2 = 0.8371
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 2 4 6 8 10 12 14
Glaciated Area (km2) within sub-basin
Min
imu
m m
elt
wa
ter
con
trib
uti
on
(m
3/
s)
Figure 12. Relationship between glaciated area in sub-basin (km2) and minimum melt water contribution (m3/s).
Figure 13. Relationship between area (km2) below the ELA (~below 5200m) and minimum melt water contribution (m3/s).
28
Figure 14. Relationship between percent of basin glaciated and average melt water contribution (m3/s) and.
y = 0.0028x + 0.0167
R2 = 0.3336
0
0.02
0.04
0.06
0.08
0.1
0.12
0 5 10 15 20 25 30
Sub-basin area (km2)
Min
imu
m g
rou
dn
wa
ter
con
trib
uti
on
(m
3/
s)
Figure 15a. Relationship between sub-basin area (km2) and minimum groundwater contribution (m3/s).
29
Figure 15b. Relationship between sub-basin area (km2) within each valley and minimum groundwater contribution (m3/s).
0
0.05
0.1
0.15
0.2
0.25
0.00 10.00 20.00 30.00 40.00 50.00 60.00 70.00
Basin Area (km2)
Gro
un
dw
ate
r co
ntr
ibu
tio
n (
m3
/s)
Querococha
Quilcayhuanca
Llanganuco
Pumapampa
Figure 16. Relationship between total basin area (km2) and average groundwater contribution (m3/s).
30
REFERENCES ASTER. 10 Jan. 2009. NASA Jet Propulsion Laboratory. 25 March 2009. < http://asterweb.jpl.nasa.gov/index.asp> Baraer M., McKenzie J., Mark B.G., Knox S. 2009. Characterizing groundwater input to a tropical proglacial hydrologic system in the Cordillera Blanca, Peru. In preparation. Barnett, T. P., J. C. Adam, and D. P. Lettenmaier. 2005. Potential impacts of a warming climate on water availability in snow-dominated regions. Nature 438 (7066):303- 309. Bradley, Raymond S., Mathias Vuille, Henry F. Diaz, and Walter Vergara. 2006. CLIMATE CHANGE: Threats to Water Supplies in the Tropical Andes. Science 312 (5781):1755-1756. Caballero, Y., V. Jomelli, P. Chevallier, and P. Ribstein. 2002. Hydrological characteristics of slope deposits in high tropical mountains (Cordillera Real, Bolivia). CATENA 47 (2):101-116. Coudrain, A., Francou, B., and Kundzewicz, Z. W. 2005. Glacier shrinkage in the Andes and consequences for water resources—Editorial. Hydrological Sciences Journal 50 (6): 925–932. Fetter C.W. 2001. Applied Hydrogeology. Prentice Hall, NJ, 598 p. IPCC. 2007. Climate Change 2007: Impacts, Adaptation and Vulnerability. Contributions of Working group 2 to the fourth assessment Report of the Intergovernmental Panel on Climate Change. (eds: Becker A. and Bruce J.) Intergovernmental Panel on Climate Change, Cambridge and New York. Juen, Irmgard, Georg Kaser, and Christian Georges. 2007. Modelling observed and future runoff from a glacierized tropical catchment (Cordillera Blanca, Peru). Global and Planetary Change 59 (1-4):37-48. Kaser, G., Ostmaston, H., 2002. Tropical glaciers. Cambridge: Cambridge University Press, 2007 p. Kaser, G. 1999. A review of the modern fluctuations of tropical glaciers. Global and Planetary Change 22 (1–4): 93–103. Kaser, Georg, and Christian Georges. 1999. On the Mass Balance of Low Latitude Glaciers with Particular Consideration of the Peruvian Cordillera Blanca. Geografiska Annaler, Series A: Physical Geography 81 (4):643-651.
31
Klein, A.G., Isacks, B.L. and Bloom, A.L., 1995. Modern and Last Glacial Maximum Snowline in Peru and Bolivia: implications for regional climatic change. Bulletin de l'Institut Français d'Études Andines 24:607–617. Mark, B. G., and J. M. McKenzie. 2007. Tracing Increasing Tropical Andean Glacier Melt with Stable Isotopes in Water. Environmental Science and Technology 41 (20):6955-6960. Mark B.G., and G.O. Seltzer. 2005. Deglaciation in the Peruvian Andes: climatic forcing, hydrologic impact and comparative rates over time. In: U. Huber, H.K.M. Bugmann and M.A. Reasoner, Editors, Global Change and Mountain Regions: An Overview of Current Knowledge vol. 23, Springer, Dordrecht: 205–214. Mark, B. G., McKenzie, J. M., Gomez, J. 2005. Hydrochemical evaluation of changing glacier meltwater contribution to stream discharge: Callejon de Huaylas, Peru. Hydrological Sciences Journal 50: 975-987. Mark, Bryan G., and Geoffrey O. Seltzer. 2003. Tropical glacier meltwater contribution to stream discharge: a case study in the Cordillera Blanca, Peru. Journal of Glaciology 49:271-281. Pouyaud, B., M. Zapata, J. Yerreb, J. Gomez, G. Rosas, W. Suarez, P. Ribstein (2005), On the future of the water resources from glacier melting in the Cordillera Blanca, Peru. Hydrological Sciences Journal-Journal Des Sciences Hydrologiques. 50(6): 999-1022. Racoviteanu, A., Arnaud, Y., Williams, M. Decadal changes in glacier parameters in Cordillera Blanca, Peru derived from remote sensing. Journal of Glaciology. 2008. 54:499-510. Racoviteanu. 2008b. GLIMS Glacier Database, edited, National Snow and Ice Data Center/World Data Center for Glaciology. Robinson, Zoe P., Ian J. Fairchild, and Andrew J. Russell. 2008. Hydrogeological implications of glacial landscape evolution at Skeioarársandur, SE Iceland. Geomorphology 97(1-2):218-236. Spang, Edward. 2006. Alpine Lakes and Glaciers in Peru: Managing Sources of Water and Destruction. Research sponsored by Tufts Institute of the Environment research Grant. Vergara, W., Deeb, A., Valencia, A., Bradley, R., Francou, B., Zarzar, A., Grünwaldt, A., Haeussling, S., 2007. Economic impacts of rapid glacier retreat in the Andes. Eos, Transactions American Geophysical Union 88 (25), 261–263. Vuille, Mathias, Bernard Francou, Patrick Wagnon, Irmgard Juen, Georg Kaser, Bryan G. Mark, and Raymond S. Bradley. 2008. Climate change and tropical Andean glaciers: Past, present and future. Earth-Science Reviews 89 (3-4):79-96.