Recibido: 06/08/2024
Aceptado: 24/02/2025
Disponible en línea: 03/06/2025
Publicado: 31/07/2025

REVISTA DE TELEDETECCIÓN
Asociación Española de Teledetección
(2025) 66, 22235
ISSN 1133-0953
EISSN 1988-8740
https://doi.org/10.4995/raet.2025.22235
Francisco Castillo-Vergara
1,2*, Edwin Loarte
1,2, Katy Medina
1,2, Sofia Rodriguez-Venturo
1,3, Eladio Tuya
2
1 Centro de Investigación en Ciencias de la Tierra, Ambiente y Tecnología (ESAT), Universidad Nacional Santiago Antúnez de Mayolo, Av. Centenario 200, Huaraz, Peru.
2 Facultad de Ciencias del Ambiente (FCAM), Universidad Nacional Santiago Antúnez de Mayolo, Av. Universitaria 115, Huaraz, 02002, Peru.
3 Facultad de Ciencias Biológicas, Universidad Nacional Mayor de San Marcos, Av. Venezuela s/n cuadra 34, Lima, 15081, Peru.
Abstract: In high mountain ecosystems, glaciers, lakes and wetlands play an important role in local water resources, where a variation in glacier dynamics can affect the rest of the ecosystem components due to their hydrological connections. The objective of the study was to analyze the relationship of glacier dynamics with lakes and wetlands in Llullan, Quillcay and Yanayacu catchments of Cordillera Blanca (Peru) in the period from 1989 to 2019. Using Geographic Information Systems (GIS) and remote sensing, a multitemporal analysis was carried out using Landsat satellite images, and subsequently linear correlation was calculated and a Principal Component Analysis (PCA) was applied. The results of glacier dynamics parameters indicated a loss of between 22 and 56% of glacier area; from 24 and 63% of glacier volume; glacier front retreat varied between 10 and 18 m yr-1; and Equilibrium Line Altitude (ELA) rose from 137 and 227 m. The lake dynamics evidenced growth between 5 and 82% of their area and between 2 and 151% of their volume; and permanent wetlands showed a tendency to increase over time (from 44 and 289%). PCA indicated that PC1 and PC2 explained between 71 and 91% of the total variance of the original 7 variables, and glacier dynamics were found to have a good-very good inverse correlation with lake dynamics (between 0.62 and 0.99) and good with wetland dynamics (from 0.64 and 0.69). The study reflected the interconnectivity of high mountain ecosystem elements, where an acceleration of glacier retreat would trigger lakes and wetlands to reach their maximum capacity in shorter periods and begin to retreat earlier than expected.
Key words: glacier dynamics, lakes, wetlands, high mountain, multitemporal analysis, Cordillera Blanca.
Variación espaciotemporal de la dinámica glaciar y su relación con los cambios en ecosistemas de alta montaña en la cordillera Blanca, Perú
Resumen: En los ecosistemas de alta montaña, los glaciares, lagunas y humedales pueden compensar la escasez hídrica desempeñando un papel importante en los recursos hídricos locales, donde, una variación de la dinámica glaciar puede afectar al resto de los componentes de los ecosistemas debido a sus conexiones hidrológicas. El objetivo del estudio fue analizar la relación de la dinámica glaciar con las lagunas y humedales de alta montaña en la cordillera Blanca (Perú), dentro de las unidades hidrográficas Llullán, Quillcay y Yanayacu (periodo 1989-2019). Mediante el uso de los Sistema de Información Geográfica (SIG) y teledetección se realizó un análisis multitemporal utilizando imágenes satelitales Landsat, y posteriormente se calculó la correlación lineal y se aplicó un Análisis de Componentes Principales (ACP). Los resultados de los parámetros de la dinámica glaciar indicaron una pérdida entre 22-56% de área glaciar; un 24-63% de volumen glaciar; el retroceso del frente glaciar varió entre 10-18 m/año; y la Línea de Equilibrio Glaciar (ELA) se elevó entre 137-227 m. La dinámica de las lagunas evidenció un crecimiento entre 5-82% de su área y entre 2-151% de su volumen; y los humedales permanentes presentaron una tendencia a incrementarse con el tiempo (del 44% y 289%). El ACP indicó que los CP1 y CP2 explicaban entre el 71-91% de la varianza total de las 7 variables originales, y se encontró que la dinámica glaciar presentaba una correlación inversa elevada-muy elevada con la dinámica de las lagunas (entre 0,62-0,99) e intensa con la dinámica de humedales (entre 0,64-0,69). El estudio reflejó la interconectividad de los elementos de los ecosistemas de alta montaña, donde, un aceleramiento del retroceso glaciar desencadenaría que las lagunas y humedales alcancen su máxima capacidad en periodos más cortos y empiecen a retroceder antes de lo previsto.
Palabras clave: dinámica glaciar, lagunas, humedales, alta montaña, análisis multitemporal, cordillera Blanca.
In recent decades, the cryosphere has been greatly reduced in extent and volume (WGMS, 2017), primarily in response to climate change that has led to a the global increase in temperatures and associated changes in local precipitation patterns (Marzeion et al., 2014). In these changing environmental conditions, questions arise regarding the sustainability and interrelationship of glacier runoff with downstream rivers, lakes and wetlands, which are the most important sources of fresh surface water on Earth (Vörösmarty, 2009), especially in high mountain regions that are used by the population and their economic activities (Silverio and Jaquet, 2005; Armstrong et al., 2018; Denzinger et al., 2021). Their study, conservation and sustainable use are priority issues worldwide. If these reservoirs were to disappear, the different human activities would lose their main supply of freshwater, especially in times of drought (Francou and Wagnon, 1998; Vuille et al., 2008).
In the case of tropical glaciers, such as those in the Cordillera Blanca, they are considered essential components of Andean hydrological systems (Baraer et al., 2012). According to the results published by INAIGEM (2018), between 1962 and 2016 Peru lost more than 50% of its glacier coverage (1280.9 km2) and, within this, the Cordillera Blanca lost ~38% (277.4 km2) of its area with a retreat rate of ~4.97 km2 yr-1. The glacier retreat of the Cordillera Blanca is generating changes (increase) in the annual runoff in the Río Santa basin and in the dynamics of lakes and wetlands. The glacier contribution of the Río Santa basin is from 10 and 20% of the total annual discharge and can reach up to ~40% in the dry season (Mark and Seltzer, 2003). However, this trend will only be observed until the glacier cover disappears (Mark and Mckenzie, 2007).
Much research has focused on determining the impact of glacier retreat in the Cordillera Blanca on surface runoff in the Río Santa basin (e.g. Mark and Seltzer, 2003; Mark et al., 2005; Baraer et al., 2012; Carey et al., 2017). However, few studies have focused on the relationship between glacier retreat and the dynamics of other important components of hydrogeological systems (Zeballos et al., 2014), such as the lakes and wetlands. They are being hydrologically interconnected with glacier runoff and susceptible to changes in time and space of glacier retreat (Polk et al., 2017).
Due to climate change, the sustainability of glaciers, lakes and high mountain wetlands is at risk. This raises the question of how these ecosystems, and the important aquifers they that contain, will respond in the event of the disappearance of glaciers. The question becomes even more relevant when glaciers are predicted to disappear in many basins before the end of this century (Zeballos et al., 2014). Therefore, lakes and wetlands could become the main regulating factors of high mountain hydrological systems (Francou, 2013).
During the last few years, social and scientific interest in the study of glaciers in mountainous regions has increased substantially (Adler et al., 2019). Hence, this research aimed to analyze the relationship between changes in glacier dynamics on high mountain lakes and wetlands in the Cordillera Blanca (having as study areas the Llullan, Quillcay and Yanayacu catchments) by a multitemporal analysis of Landsat coverages, in the period 1989-2019 with a three-year interval.
The study area included the glaciers, lakes and high mountain wetlands of the Llullan (north), Quillcay (center) and Yanayacu (south) catchments from the Cordillera Blanca, in the department of Ancash, Peru (Figure 1) located between 8º57’04’’ to 9º50’39’’ S and 77º13’26’’ to 77º49’39’’ W.

Figure 1. Study area: Llullan, Quillcay and Yanayacu catchments of Cordillera Blanca.
The study of glacier, lake and wetland dynamics was carried out through the multitemporal analysis of Landsat 5 and 8 satellite images. We chose, preferably, those scenes captured during the dry season in Peru (June-September), and which also presented a lower number of clouds, shadow and snow, as recommended in the Methodological Manual of the Peruvian National Glacier Inventory (INAIGEM, 2017).
A total of 22 satellite images were downloaded (from: https://glovis.usgs.gov/ and https://earthexplorer.usgs.gov/) as the main scenes (two for each year of study, Table 1), and 36 satellite images were downloaded as support scenes (including 4 Sentinel-2 scenes obtained from: https://browser.dataspace.copernicus.eu/), which were used to correct the initial glacier coverage (1989) and to complete some areas with abundant cloud cover, snow or other uncontrolled factors. All satellite images were corrected radiometrically (conversion of values to floating-point reflectance) and atmospherically (DOS1 method; BOA reflectance to TOA reflectance) using the Semi-Automatic Classification (SCP) plugin in QGIS. Then, they were reprojected to the UTM 18S zone and cropped to the study area.
Table 1. Landsat Collection 2 satellite images used as main scenes in multitemporal analysis.
Satellite |
Sensor |
Year |
Month |
Day |
Code |
Landsat 5 |
TM |
1989 |
August |
24 |
LT05_L1TP_008066_19890824_20200916_02_T1 |
LT05_L1TP_008067_19890824_20200916_02_T1 |
|||||
Landsat 5 |
TM |
1992 |
June |
29 |
LT05_L1TP_008066_19920629_20200914_02_T1 |
LT05_L1TP_008067_19920629_20200914_02_T1 |
|||||
Landsat 5 |
TM |
1995 |
August |
25 |
LT05_L1TP_008066_19950825_20200912_02_T1 |
LT05_L1TP_008067_19950825_20200912_02_T1 |
|||||
Landsat 5 |
TM |
1998 |
September |
2 |
LT05_L1TP_008066_19980902_20200908_02_T1 |
LT05_L1TP_008067_19980902_20200908_02_T1 |
|||||
Landsat 5 |
TM |
2001 |
August |
9 |
LT05_L1TP_008066_20010809_20200906_02_T1 |
LT05_L1TP_008067_20010809_20200906_02_T1 |
|||||
Landsat 5 |
TM |
2004 |
May |
13 |
LT05_L1TP_008066_20040513_20200903_02_T1 |
LT05_L1TP_008067_20040513_20200903_02_T1 |
|||||
Landsat 5 |
TM |
2007 |
July |
25 |
LT05_L1TP_008066_20070725_20200830_02_T1 |
LT05_L1TP_008067_20070725_20200830_02_T1 |
|||||
Landsat 5 |
TM |
2010 |
August |
18 |
LT05_L1TP_008066_20100818_20200823_02_T1 |
LT05_L1TP_008067_20100818_20200823_02_T1 |
|||||
Landsat 8 |
OLI/TIRS |
2013 |
September |
11 |
LC08_L1TP_008066_20130911_20200912_02_T1 |
LC08_L1TP_008067_20130911_20200913_02_T1 |
|||||
Landsat 8 |
OLI/TIRS |
2016 |
November |
22 |
LC08_L1TP_008066_20161122_20200905_02_T1 |
LC08_L1TP_008067_20161122_20200905_02_T1 |
|||||
Landsat 8 |
OLI/TIRS |
2019 |
August |
27 |
LC08_L1TP_008066_20190827_20200826_02_T1 |
LC08_L1TP_008067_20190827_20200826_02_T1 |
In the case of wetland cover mapping, most of the scenes in Table 1 were used. The changed scenes were for 1992, 2004 and 2016 satellite images (Table 2). These changes were necessary to standardize the study month and to be able to correctly map for permanent wetlands (those that last through the dry season). The satellite images from the rainy season were discarded from the analysis, because they showed a disproportionate increase in area of wetlands.
Table 2. Replacement Landsat Collection 2 satellite images for multitemporal analysis of wetland cover.
Satellite |
Sensor |
Year |
Month |
Day |
Code |
Landsat 5 |
TM |
1992 |
August |
16 |
LT05_L1TP_008067_19920816_20200914_02_T1 |
Landsat 5 |
TM |
2004 |
August |
1 |
LT05_L1TP_008067_20040801_20200903_02_T1 |
Landsat 8 |
OLI/TIRS |
2016 |
August |
18 |
LC08_L1TP_008067_20160818_20200906_02_T1 |
For the Digital Elevation Model (DEM), three scenes were downloaded from the ALOS-PALSAR sensor (orthometric heights with EGM96 vertical datum, obtained from: https://search.asf.alaska.edu/#/) on January 20, 2011 (AP_07790_FBD_F6980_RT1, AP_07790_FBD_F6990_RT1 and AP_07790_FBD_F7000_RT1). This DEM was used because of its spatial resolution (12.5 m) and, in addition, because it is recommended for use in high mountain areas such as the Cordilleras of Peru (INAIGEM, 2017).
The delimitation of the glacier area was carried out using the Normalized Difference Snow Index (NDSI) (Hall et al., 1995), where the radiometric threshold varied between 0.25 and 0.45 depending on the intrinsic characteristics of the satellite image of each year of study. Turpo et al. (2022) recommended that thresholds be adjusted by visual inspection in each classification region and year, and with INAIGEM (2023) determined that, in tropical Andean glaciers, the NDSI threshold used as an indicator of the presence of glacier cover is ≥ 0.1. In the case of debris-covered glaciers, their automated delimitation was more difficult, therefore, these coverages were digitized freehand through the visual interpretation of satellite images. For this purpose, high-resolution Google Earth Pro images from the years 2003, 2005, 2011, 2012 and 2013 (according to availability in each study area) were used. To perform the final manual correction of glacier cover, especially debris-covered glaciers, we used the 2016 data published in the National Glacier Inventory of Peru (INAIGEM, 2018, 2021), as well as high-resolution images available in Google Earth Pro and SAS Planet software.
The individualization of the glaciers was carried out using the Strahler cut-off limits provided by INAIGEM and following the details of the Methodological Manual of the National Glacier Inventory of Peru (INAIGEM, 2017). These cut-off polygons were corrected and completed in the case of extinct glaciers found in each catchment of study.
To obtain the glacier volume, the methodology established by Linsbauer et al. (2009) and Paul and Linsbauer (2012) called Glacier Bed Topography (GlabTop) was used (systematized in a Python script). This method calculates the glacier volume from the estimation of the average thickness of the central ice flow that depends on the slope of the glacier surface. Within the Python script modelling, the following formulas were used, which directly relates the ice thickness to the basal shear stress (Van Wyk de Vries et al., 2022):
where;
where: H is the ice-thickness, τb is the basal shear stress measured in kPa, f is a shape factor accounting for lateral drag along the glacier margins, ρi is the ice density, g is gravitational acceleration, α is the ice-surface slope angle and Δzi is the glacier-surface elevation range in km. For this model, the input data used were the ALOS-PALSAR DEM, individualised glacier areas and glacier flow lines. The estimated error of this method is related to an uncertainty of 30%, and for some individual glaciers, up to 50% (Linsbauer et al., 2012; Pieczonka et al., 2018), and, for the research, the average data error was taken as 40%.
For the mapping of the glacier fronts, the properties of the combination of bands (SWIR colour, brightness and contrast enhancements), contour lines and high-resolution satellite images were used. For this, the highest and lowest point of each glacier was established, then, following the direction of the central flow, the line of the glacier front was digitized. The retreat of the glacier front was obtained by the difference of the central flow line in a comparison period (every three years). The position error of the glacier front to the multitemporal measurements obtained from satellite images was calculated with the following formula (Hall et al., 2003; Silverio and Jaquet, 2005):
where: Efr is the error associated with the measurement of the glacier front, α indicates the spatial resolution of the first image used, β the spatial resolution of the second image used, and ε the registration error or Root Mean Square (RMS). According to USGS (2021), Landsat Tier 1 satellite scenes are consistent in their georegistration and exhibit an average image-to-image RMS ≤ 12 m.
To obtain the glacier ELA, the equivalent value obtained from the digitization of the Snow Line Altitude (SLA) was used. The SLA value can be considered as a good indicator of ELA in tropical glacier areas (Rabatel et al., 2012; Loarte et al., 2015). For this purpose, three strategically distributed glaciers (per catchment) were chosen, in which the SLA was digitized 400 m above the central flow line using the combination of bands and contrast stretches of the satellite images. Then, altitude values were extracted (using the ALOS-PALSAR DEM) every 12 m along the digitized line, and the values obtained for each glacier and catchment were averaged. In this case, the error estimate was referential to the standard deviation value found from the values extracted from the digitized line of the SLA (Loarte, 2011).
The delimitation of the lakes area was performed using the Normalized Difference Water Index (NDWI) (McFeeters, 1996), where the radiometric threshold varied between 0.01 and 0.20 depending on each satellite scene of study. Ji et al. (2009) recommend that the spectral index threshold should be adjusted according to the current study situations because this value is not a constant. Wang et al. (2020) and INAIGEM (2023) determined that the threshold of NDWI used as an indicator of the presence of water bodies is ≥ –0.05. In this case, the areas obtained with the index were corrected and smoothed using high resolution satellite images available in each study area.
To obtain the volume of the lakes, the empirical formulas proposed by Huggel et al. (2002), Loriaux and Casassa (2013) and Kapitsa et al. (2017), were used; these formulas gave relatively similar results, so the values obtained were averaged. To estimate the error in the volume for each lake, we took the value of the standard deviation of the data obtained with the three methodologies mentioned above.
In this case, the only parameter that could be monitored with satellite images was the wetland area. For this, the Modified Soil-Adjusted Vegetation Index 2 (MSAVI2) proposed by Qi et al. (1994), was used, where a threshold of 0.26 was determined for its delimitation, being this value calibrated according to the wetland area reported by Rios (2020) in Quillcay catchment for 2019. Within the automatic delimitation, the coverage was corrected with the criteria of altitude > 3700 m a.s.l. (Santiago et al., 2021; Rios, 2024) and slope between 0-15° (Rios, 2020; Rios and Fuentealba, 2020). Finally, manual correction was performed to refine those areas incorrectly mapped due to their spectral similarity (e.g., native forests and grasslands), using high-resolution satellite images from Google and ESRI (basemaps).
Within area calculations, it can be assumed that the error associated with their measurements possesses a normal or Gaussian distribution, i.e., the error in the delimitation of areas can be estimated using statistical approximations. Therefore, the error of a unit area based on a standard deviation (1σ) can be expressed as follows (Hanshaw and Bookhagen, 2014; Wang et al., 2020):
where: P is the perimeter (m), G the spatial resolution of the satellite image (m), A the area (m), and E the individual relative error. The coefficient 0.6872 indicates that ~69% of the pixels are prone to error.
Likewise, the cumulative error for an entire study region (or subregions) can be calculated with the following formula (Wang et al., 2020):
where: ET is the total area error of a region or subregion (%), i is the element in the n° i in the region or subregion under study, and a is the individual relative error of a unit area.
Principal Components Analysis (PCA) was applied to detect relationships between variables and identify data structure. To do this, data adequacy was first assessed using the Kaiser-Meyer-Olkin (KMO) test and the communality values. Principal components (PC) were then extracted using the eigenvalue criterion (>1), with those with the greatest amount of total variance explained. Finally, a component matrix was obtained to analyze the weight of each retained PC relative to each original variable. The final matrix was rotated using Varimax rotation to obtain loadings that facilitated interpretation and calculation of the final values for each PC.
Subsequently, the Shapiro-Wilk normality test was applied to determine the type of distribution of the data and, finally, Pearson’s linear correlation (for normal data) and Spearman’s Rho (for non-normal data) tests were performed to determine the relationship between the parameters, with a significance level (α) of 5%.
The analysis of the spatiotemporal dynamics in the Llullan, Quillcay and Yanayacu catchments of the Cordillera Blanca showed the changes over time in the physical parameters of the glaciers, lakes and wetlands, where the evolution of the landscape is dominated by changes in the area of the ecosystem elements.
The annual and average errors obtained for the parameters of the glaciers, lakes and wetlands of each catchment are summarized in Table 3. In the case of glacier area and volume, the Quillcay catchment presented the highest average errors, while in the glacier ELA it was the Llullan catchment that presented the highest average error. For the glacier fronts, the annual and average errors were equal because the same satellite was used for all years and study areas. In the case of lakes, the Yanayacu catchment presented the greatest error of the three catchments, and in the case of wetlands, it was the Quillcay catchment. It has been detected that a greater presence of area and/or quantity of elements is associated with a greater estimated error within each study parameter.
Table 3. Annual and average error of the parameters of the dynamics of glaciers, lakes and wetlands.
Parameter |
Annual error |
Average error |
||||||||||
1989 |
1992 |
1995 |
1998 |
2001 |
2004 |
2007 |
2010 |
2013 |
2016 |
2019 |
||
Llullan Catchment |
||||||||||||
Glacier area (km2) |
0.38 |
0.41 |
0.42 |
0.41 |
0.41 |
0.42 |
0.42 |
0.42 |
0.42 |
0.45 |
0.45 |
0.42 |
Glacier volume (km3) |
0.40 |
0.39 |
0.38 |
0.36 |
0.36 |
0.35 |
0.35 |
0.34 |
0.34 |
0.31 |
0.31 |
0.35 |
Glacier front (m) |
54.43 |
54.43 |
54.43 |
54.43 |
54.43 |
54.43 |
54.43 |
54.43 |
54.43 |
54.43 |
54.43 |
54.43 |
Glacier ELA (m) |
7 |
16 |
21 |
20 |
4 |
17 |
28 |
29 |
31 |
34 |
25 |
21 |
Lake area (km2) |
0.06 |
0.12 |
0.09 |
0.12 |
0.09 |
0.13 |
0.14 |
0.15 |
0.12 |
0.18 |
0.16 |
0.12 |
Lake volume (km3) |
0.002 |
0.005 |
0.005 |
0.005 |
0.004 |
0.005 |
0.006 |
0.007 |
0.003 |
0.007 |
0.007 |
0.005 |
Quillcay Catchment |
||||||||||||
Glacier area (km2) |
2.99 |
2.91 |
2.90 |
3.78 |
3.73 |
2.67 |
2.66 |
2.39 |
2.45 |
0.97 |
1.12 |
2.60 |
Glacier volume (km3) |
0.72 |
0.69 |
0.67 |
0.57 |
0.55 |
0.51 |
0.49 |
0.48 |
0.48 |
0.42 |
0.42 |
0.55 |
Glacier front (m) |
54.43 |
54.43 |
54.43 |
54.43 |
54.43 |
54.43 |
54.43 |
54.43 |
54.43 |
54.43 |
54.43 |
54.43 |
Glacier ELA (m) |
14 |
17 |
11 |
14 |
20 |
16 |
17 |
23 |
13 |
11 |
20 |
16 |
Lake area (km2) |
0.11 |
0.11 |
0.11 |
0.12 |
0.14 |
0.17 |
0.17 |
0.18 |
0.20 |
0.20 |
0.20 |
0.16 |
Lake volume (km3) |
0.002 |
0.002 |
0.002 |
0.002 |
0.003 |
0.003 |
0.003 |
0.003 |
0.003 |
0.003 |
0.003 |
0.003 |
Wetland area (km2) |
0.012 |
0.007 |
0.012 |
0.009 |
0.013 |
0.015 |
0.015 |
0.011 |
0.016 |
0.017 |
0.019 |
0.013 |
Yanayacu Catchment |
||||||||||||
Glacier area (km2) |
1.28 |
1.14 |
1.15 |
0.71 |
0.69 |
0.65 |
0.62 |
0.66 |
0.63 |
0.16 |
0.16 |
0.71 |
Glacier volume (km3) |
0.11 |
0.09 |
0.09 |
0.07 |
0.07 |
0.07 |
0.06 |
0.06 |
0.05 |
0.05 |
0.04 |
0.07 |
Glacier front (m) |
54.43 |
54.43 |
54.43 |
54.43 |
54.43 |
54.43 |
54.43 |
54.43 |
54.43 |
54.43 |
54.43 |
54.43 |
Glacier ELA (m) |
10 |
12 |
13 |
18 |
23 |
21 |
13 |
14 |
12 |
22 |
21 |
16 |
Lake area (km2) |
0.34 |
0.36 |
0.36 |
0.37 |
0.34 |
0.39 |
0.36 |
0.36 |
0.38 |
0.41 |
0.41 |
0.37 |
Lake volume (km3) |
0.005 |
0.005 |
0.005 |
0.005 |
0.005 |
0.005 |
0.005 |
0.005 |
0.005 |
0.005 |
0.005 |
0.005 |
Wetland area (km2) |
0.004 |
0.001 |
0.005 |
0.003 |
0.003 |
0.005 |
0.006 |
0.002 |
0.015 |
0.017 |
0.015 |
0.007 |
The analyzed parameters of glacier dynamics were glacier area, volume, front and ELA (Figure 2). These analyses were developed for each glacier; however, the results are presented for each catchment to make a comparison from a broader landscape point of view.

Figure 2. Glacier dynamics in the Llullan, Quillcay and Yanayacu catchment between 1989 and 2019. A) Area variation; B) volume variation; C) front retreat, and D) ELA variation.
The first analysis was the variation in glacier area (Figures 2A and 3). The results show how glacier retreat evolved in the three catchments from 1989 to 2019. In the Llullan catchment, glaciers lost a total of 6.77 km2 in the last 30 years (retreat rate of 0.23 km2 yr-1), representing a loss of ~22% with respect to its initially estimated area. Llullan lost the least area between 1998 and 2001 (0.12 km2) and the largest, between 2013 and 2016 (2.10 km2). In the Quillcay catchment, the glaciers lost almost twice the area retreated than in the Llullan (12.83 km2), with a retreat rate of 0.43 km2 yr-1 (loss of ~30% of its initial area). Quillcay lost the least area between 2004 and 2007 (0.47 km2) and the largest, like the Llullan, between 2013 and 2016 (2.65 km2). In the case of the Yanayacu catchment, the lost was 7.47 km2 (retreat rate of 0.25 km2 yr-1), which, represents a loss of ~56% of its initial area. Yanayacu lost the least glacier area between 1998 and 2001 (0.14 km2, like Llullan) and the greatest, between 1989 and 1992 (1.88 km2).

Figure 3. Glacier dynamics: A) Llullan catchment and Artesonraju glacier, B) Quillcay catchment and Shallap glacier, and C) Yanayacu catchment and Gueshgue glacier; between 1989 to 2019.
The second analysis was the variation in glacier volume. The results obtained show a similar evolution to the glacier retreat area data (Figure 2B). The glaciers of the Llullan catchment lost 0.24 km3 of glacier volume during the 30 years of study (retreat rate of 0.008 km3 yr-1), representing a loss of ~24% with respect to their initial volume. Lullan lost the least volume between 1998 and 2001 (0.004 km3) and the greatest, between 2013 and 2016 (0.067 km3). In the Quillcay catchment, glaciers lost 0.76 km3 of volume (retreat rate of 0.025 km3 yr-1), representing a loss of ~42% of its initial total. Quillcay lost the least volume between 2010 and 2013 (0.004 km3) and the greatest, between 1995 and 1998 (0.252 km3). For its part, the Yanayacu catchment lost 0.17 km3 of volume (~63% of its initial volume) with a rate of retreat of 0.006 km3 yr-1, between 1998 and 2001 Yanayacu lost the least volume (0.005 km3).
The third analysis focused on estimating the retreat of glacier fronts (Figure 2C). In the Llullan catchment, the results show an average retreat rate of 10.30 m yr-1, followed by the Yanayacu catchment with 15.11 m yr-1, and the Quillcay catchment with 18.20 m yr-1. The periods with the least retreat of the fronts were between 2004 to 2007 and 2010 to 2013 (15.11 to 37.64 m yr-1), and the greatest retreats between 1989 to 1992 and 2013 to 2016 (46.22 to 95.29 m yr-1).
The last parameter analyzed was the variation of glacier ELA (Figure 2D). The results show that the glaciers of the Llullan catchment increased their ELA by 137 m between 1989 and 2019 (variation of 5 m yr-1), while between 1998 and 2001 there was a drastic decrease of 256 m, and between 2001 and 2004 they showed the highest ELA elevation (of 208 m). In the Quillcay catchment, the glaciers increased their ELA by 138 m (variation of 5 m yr-1). Between 1998 and 2001, Llullan catchment presented the greatest reduction in its ELA (146 m) and between 2001 and 2004 it rose by 110 m. Yanayacu showed the highest elevation of its ELA, increasing its value by 227 m during the 30 years of study (variation of 8 m yr-1). Between 1998 and 2001, its ELA decreased by 273 m, and between 1989 and 1992, it increased by 170 m.
The parameters analyzed for lake dynamics were area and volume (Figure 4). These analyses were carried out for each lake; however, the results are presented for each catchment.

Figure 4. Lake dynamics in Llullan, Quillcay and Yanayacu catchments between 1989 and 2019. A) Area variation and B) volume variation.
The first analysis was performed on area variation (Figure 4A and Figure 5). The lakes of the Llullan catchment increased their area by 0.87 km2 at the end of the 30 years of study, representing an increase of ~82% with respect to their initial area (growth rate of 0.03 km2 yr-1). This catchment had a great spatial dynamic compared to the other study catchments. Between 2010 and 2013 this catchment presented a reduction in its area of 0.54 km2 and, aversely, presented a greater increase between 2013 and 2016 (0.59 km2). In the Quillcay catchment, the lakes increased their area by 0.72 km2 (growth rate of 0.02 km2 yr-1) estimating an increase of ~60% with respect to their initial area. It presented an area reduction of 0.02 km2 between 2016 and 2019 and had a higher growth (of 0.26 km2) between 2001 and 2004. Yanayacu catchment presented a lower evolution compared to the other catchments because its increasing was only 0.15 km2 in the last 30 years (increase of ~5%). The period where it showed a decrease was between 1998 and 2001 (0.04 km2), and like the Quillcay catchment, it presented a greater increase during 2001 and 2004 with (0.05 km2).

Figure 5. Lake dynamics: A) Llullan catchment and Paron lake, B) Quillcay catchment and Palcacocha lake, and C) Yanayacu catchment and lake in formation at the foot of Gueshgue glacier; between 1989 and 2019.
The second analysis was on volume variation, reflecting the results a behavior like the area variation (Figure 4B). The lakes of Llullan, Quillcay and Yanayacu catchments increased in volume by 0.041, 0.018 and 0.001 km3 respectively during the 30 years of study. In the case of Llullan catchment, the increase was explained by the dynamics of Paron lake (historically regulated and generator of multiple socio-environmental conflicts over water use), which represented a potential danger due to its large size (and volume) during the rainy season. The Llullan catchment presented an increase of ~151% in its initially estimated volume, Quillcay an increase of ~111% and Yanayacu only ~5%. Regarding the periods of lesser or greater increase, it was varied and without dominant periods, however, a similarity proportional to the variation of the area of lakes and to the retreat of glaciers was detected.
The only possible parameter to evaluate the dynamics of permanent wetlands using remote sensing images was the area (Figures 6 and 7). The results showed the reduction and growth of permanent wetlands areas between 1989 and 2019 in Quillcay and Yanayacu catchments. In Llullan catchment, it was verified with image processing and fieldwork that there were not permanent wetlands.

Figure 6. Wetland dynamic in Quillcay and Yanayacu catchments from 1989 and 2019.

Figure 7. Wetland dynamics in the: A) Quillcay catchment and upper part of Quillcayhuanca valley and B) Yanayacu catchment and its lower-middle part.
The wetlands of Quillcay catchment increased a total of 0.76 km2 in area (growth rate of 0.025 km2 yr-1), representing an increase of ~44% with respect to its initial area. The period where it presented a greater decrease was between 1989 and 1992 with 0.99 km2 and, on the contrary, it obtained a greater growth between 1992 to 1995 and 2010 to 2013 with 0.96 km2. In the Yanayacu catchment, wetlands increased a total of 0.91 km2 in area (growth rate of 0.030 km2 yr-1), estimating an increase of ~289%. In the period 2007 and 2010 there was a decrease of 0.31 km2 and from 2010 and 2013 there was the greatest growth in its extent (1.04 km2). The permanent wetlands of both catchments have a great spatial dynamic compared to glaciers and high mountain lakes.
The PCA showed that the value of the KMO test were 0.70, 0.72 and 0.59 for Llullan, Quillcay and Yanayacu catchments, respectively. All values of the extracted communalities were found to be above 0.50. These results indicated that factor analysis could be used with the sample data obtained.
The PCA was performed with the 7 variables studied, and the following were selected: 1 PC for Llullan, and 2 PCs for Quillcay and Yanayacu, based on the eigenvalue criterion (> 1) and accounting for between 71 and 91% of the total variance explained. The loadings (normal and rotated) of the selected parameters in the PCs of each catchment are shown in Table 4. In the Llullan catchment, PC1 had higher positive loadings in the area and volume of the lakes, and higher negative loadings in the area and volume of the glaciers, and it was not possible to obtain the rotated loadings since only one PC was extracted. In the Quillcay catchment, PC1 had higher positive and negative loads in the same variables as Llullan; whereas, in PC2 the highest load was found in the glacier front. In the Yanayacu catchment, PC1 had higher positive loads in the area of the lakes and wetlands, and higher negative loads in the area and volume of the glaciers (as in the other catchments); whereas, in PC2 there were only positive loads, being the highest in the glacier front. In a general analysis, in the three catchments it was determined that within the first CP the highest absolute loads were obtained for the parameters of glacier area, glacier volume and lake area, whereas within the second CP the highest absolute load was always with the glacier front. The PCA (for both normal and rotated values) determined the importance of the area and volume parameters within the study of glacier, lake and wetland dynamics.
Table 4. Component loadings of the PCA.
Component loading |
|||||||||
Catchment |
Llullan |
Quillcay |
Yanayacu |
||||||
Component |
PC1 |
PC1 |
PC2 |
PC1 |
PC2 |
||||
Eigenvalue |
4.31 |
5.10 |
1.21 |
4.40 |
1.53 |
||||
Variance (%) |
71.84 |
72.92 |
17.33 |
62.79 |
21.80 |
||||
Cumulative variance (%) |
71.84 |
72.92 |
90.25 |
62.79 |
84.60 |
||||
Parameter/Variable |
Component matrix |
||||||||
Normal |
Normal |
Rotated |
Normal |
Rotated |
|||||
Component |
PC1 |
PC1 |
PC2 |
PC1 |
PC2 |
PC1 |
PC2 |
PC1 |
PC2 |
Glacier area |
–0.90 |
–0.97 |
–0.86 |
–0.45 |
–0.91 |
0.35 |
–0.97 |
||
Glacier volume |
–0.89 |
–0.98 |
–0.90 |
–0.40 |
–0.91 |
0.34 |
–0.97 |
||
Glacier front |
0.71 |
0.48 |
0.83 |
0.95 |
0.93 |
0.89 |
|||
Glacier ELA |
0.71 |
0.80 |
0.46 |
0.49 |
0.78 |
0.72 |
0.49 |
0.50 |
0.71 |
Lake area |
0.93 |
0.97 |
0.93 |
0.98 |
0.89 |
0.40 |
|||
Lake volume |
0.92 |
0.95 |
0.93 |
0.71 |
0.40 |
0.53 |
0.63 |
||
Wetland area |
- |
0.71 |
–0.51 |
0.87 |
0.87 |
0.87 |
|||
In the three catchments, the lakes and mountain wetlands were found to be statistically significant to the changes in glacier dynamics (Tables 5, 6 and 7).
Table 5. Normality and linear correlation test for the Llullan catchment.
Statistical test |
Parameter |
||||||
Glacier area |
Glacier volume |
Glacier front |
Glacier ELA |
Lake area |
Lake volume |
Wetland area |
|
Normality |
|||||||
Shapiro-Wilk |
0.74 |
0.75 |
0.27 |
0.06 |
0.34 |
0.29 |
- |
Correlation test |
|||||||
Glacier area with… |
|||||||
Pearson |
1.00 |
1.00 |
–0.48 |
–0.53 |
–0.77 |
–0.74 |
- |
Significance (bilateral) |
- |
0.00 |
0.14 |
0.10 |
0.01 |
0.01 |
- |
Glacier volume with… |
|||||||
Pearson |
1.00 |
1.00 |
–0.45 |
–0.53 |
–0.74 |
–0.72 |
- |
Significance (bilateral) |
0.00 |
- |
0.16 |
0.09 |
0.01 |
0.01 |
- |
Table 6. Normality and linear correlation test for Quillcay catchment.
Statistical test |
Parameter |
||||||
Glacier area |
Glacier volume |
Glacier front |
Glacier ELA |
Lake area |
Lake volume |
Wetland area |
|
Normality |
|||||||
Shapiro-Wilk |
0.80 |
0.21 |
0.12 |
0.86 |
0.01 |
0.00 |
0.35 |
Correlation test |
|||||||
Glacier area with… |
|||||||
Pearson |
1.00 |
0.98 |
–0.45 |
–0.73 |
- |
- |
–0.64 |
Spearman’s Rho |
- |
- |
- |
- |
–0.99 |
–0.95 |
- |
Significance (bilateral) |
- |
0.00 |
0.17 |
0.01 |
0.00 |
0.00 |
0.03 |
Glacier volume with… |
|||||||
Pearson |
0.98 |
1.00 |
–0.41 |
–0.71 |
- |
- |
–0.67 |
Spearman’s Rho |
- |
- |
- |
- |
–0.99 |
–0.95 |
–0.71 |
Significance (bilateral) |
0.00 |
- |
0.21 |
0.02 |
0.00 |
0.00 |
0.02 |
Table 7. Normality and linear correlation test for Yanayacu catchment.
Statistical test |
Parameter |
||||||
Glacier area |
Glacier volume |
Glacier front |
Glacier ELA |
Lake area |
Lake volume |
Wetland area |
|
Normality |
|||||||
Shapiro-Wilk |
0.79 |
0.77 |
0.39 |
0.49 |
0.40 |
0.00 |
0.01 |
Correlation test |
|||||||
Glacier area with… |
|||||||
Pearson |
1.00 |
1.00 |
0.22 |
–0.69 |
–0.87 |
- |
- |
Spearman’s Rho |
- |
- |
- |
- |
- |
-0.61 |
-0.69 |
Significance (bilateral) |
- |
0.00 |
0.55 |
0.02 |
0.00 |
0.05 |
0.02 |
Glacier volume with… |
|||||||
Pearson |
1.00 |
1.00 |
0.23 |
–0.67 |
–0.87 |
- |
- |
Spearman’s Rho |
- |
- |
- |
- |
- |
–0.64 |
–0.69 |
Significance (bilateral) |
0.00 |
- |
0.57 |
0.02 |
0.00 |
0.03 |
0.02 |
In the Llullan catchment, the results of the Shapiro-Wilk test showed that all variables had a normal distribution. In the correlation tests, it was obtained that the glacier area and volume possessed between them a perfect direct correlation (~1.00, p ≤ 0.05), whereas, they possessed a moderate inverse correlation with glacier front retreat and changes in the glacier ELA (between 0.45 and 0.53, not significant). Regarding the influence of glacier dynamics parameters on lake dynamics, a good inverse correlation was found (between 0.72 and 0.77, p ≤ 0.05), i.e., the smaller the glacier area or volume, the larger of lake area and volume and vice versa. In the case with wetland dynamics, it was not possible to perform the analysis because no permanent wetlands were detected within this catchment (Table 5).
In the Quillcay catchment, the results of Shapiro-Wilk test show that the variables of area and volume of lakes do not have a normal distribution. In the linear correlation, it was observed that glacier area and volume possessed between them a very good-perfect direct correlation (between 0.98 and 1.00, p ≤ 0.05) and, whereas, they had a moderate inverse correlation with glacier front retreat (between 0.41 and 0.45, not significant), and good with glacier ELA (between 0.71 and 0.73, p ≤ 0.05). On the influence of glacier dynamics parameters on lake dynamics, a very good inverse correlation was found (between 0.95 and 0.99, p ≤ 0.05), and in the case of wetland dynamics, a good inverse correlation was discovered (between 0.64 and 0.67, p ≤ 0.05) (Table 6).
Finally, in the Yanayacu catchment, the Shapiro-Wilk test showed that only the lake volume and wetland area variables did not have a normal distribution. The linear correlation shows that glacier area and volume possessed between them a perfect direct correlation (~1.00, p ≤ 0.05); also, they had a low inverse correlation with glacier front retreat (between 0.22 and 0.23, not significant), and good with glacier ELA (between 0.67 and 0.69, p ≤ 0.05). In the case of influence of glacier dynamics parameters on lake dynamics, we found a very good inverse correlation (~0.87, p ≤ 0.05) with lake area and good with volume (between 0.61 and 0.64, p ≤ 0.05). Finally, with respect to wetland dynamics, a good inverse correlation was found (~0.69, p ≤ 0.05) (Table 7).
The estimated glacier retreat in the three catchments (between 22 and 56%) from 1989 and 2019, is within the range reported by other authors for cordillera Blanca (e.g., Racoviteanu et al., 2008; Silverio and Jaquet, 2017; INAIGEM, 2018). However, a greater glacier retreat was evident in the southern zone (Yanayacu catchment) compared to the central and northern zones (Llullan and Quillcay catchments) of Cordillera Blanca.
The retreat rates of the three catchments were lower than the 4.97 km2 yr-1 of retreat estimated by INAIGEM (2018) for the Cordillera Blanca. The retreat rates of Llullan and Yanayacu catchments were very similar (0.23 and 0.25 km2 yr-1, respectively), while the Quillcay catchment was almost doubled those values (0.43 km2 yr-1). Despite similar retreat rates between Llullan and Yanayacu, the latter lost ~56% of its glacier area and Llullan only ~22%. This difference may be due to the number of glaciers present in each catchment, as well as the type of glacier cover. It is important to note that the Llullan catchment has a considerable number of debris covered glaciers (e.g., Huancotepampa, Piramide 1 and Jatunraju), which have shown very little retreat compared to the clean glaciers (Benn et al., 2012).
Another relevant factor to consider in glacier retreat is the number and size of glaciers. The more glaciers there are, the longer they will last over time (as in the Quillcay catchment); likewise, the size of the glaciers influences their conservation because smaller glaciers are usually not very resistant to the environmental changes they experience and, therefore, retreat faster (Garg et al., 2019; Zhao et al., 2020). Fragmentation of ice bodies to smaller, isolated and lower altitude blocks was a problem in the Quillcay and Yanayacu catchments, demonstrating the marked retreat of glaciers (Racoviteanu et al., 2008; Schauwecker et al., 2017).
Regarding the loss of glacier volume, the most worrisome aspect is the reduction of the estimated water reserve potential, which is even greater (in percentage terms) than the glacier surface area lost. This problem was highlighted in the Quillcay catchment, where ~42% of the initial volume was lost, evidenced by the retreat of the extensive glacier tongues of Palcaraju, Tullapraju, Cayesh, Shallap and the 30 extinct glaciers found in the study period. Currently, there are no studies on the estimation of glacier volume at the catchment or cordillera level applying models such as GlabTop, but there are studies on point glaciers and in short periods of comparison (e.g. Hidrandina SA, 1989; Cerna, 2012; Rojas and Tordocillo, 2018). Hidrandina SA (1989) estimated the volume through an empirical formula for the glacier cordilleras of Peru, however, it is not possible to compare the results with this study due to the nature of the methodologies used. To date, there are still high uncertainties in all the methods used for the estimation of glacier thickness and volume; however, the results obtained with GlabTop in different studies justify its application (Linsbauer et al., 2012; INAIGEM, 2018; Van Wyk de Vries et al., 2022; Zapana et al., 2024). The data from this study can serve as a reference and starting point for a more accurate quantification of the volume and potential of water reserves historically and in the future.
The analysis of glacier fronts is usually more complex because the most evident retreat is in valley glaciers and less in mountain glaciers. In fact, the data obtained established reference points for each study area. An inverse and non-significant correlation to the retreat of the area and/or volume was evidenced, i.e., glacier retreat cannot determine the retreat of the front, therefore, this dynamic must be studied according to other elements, such as, the type of glacier (valley or mountain), climatic (temperature and precipitation) and topographic factors (slope and aspect). In general, the smaller glaciers were those with the least retreat of the front compared to the larger ones, which are generally located within flatter surfaces, where the glacier front tends to extend more and therefore presents a greater retreat. In most of the glaciers studied, the retreat of the fronts was predominant over the periods of stabilization (Bernex and Tejada, 2010; Cuadros, 2014).
The ELA in Quillcay and Yanayacu catchments showed a significant inverse correlation with glacier area and volume (Osmaston, 2005), which also increased over time. In Llullan catchment, it presented a non-significant correlation, which could be due to the presence of debris-covered glaciers. Yanayacu catchment presented the highest elevation of the ELA and the greatest loss of glacier area. In many cases, the ELA is used to evaluate the evolution of retreat because it determines the division between the accumulation and ablation zone of a glacier (Ubeda, 2011; Rabatel et al., 2012). The ELA method used for the study was very useful for estimating this parameter, considering that fieldwork is usually very expensive and difficult to perform (Loarte et al., 2015). Seltzer (1993) emphasizes that when past ELA with present ELA was compared, a measure of climate change can be obtained (Loarte, 2011; Rabatel et al., 2012; Schauwecker et al., 2017).
This research reaffirmed that glacier loss has increased in recent decades (Seehaus et al., 2019) leading to an impact on water resources in high mountains, and to negative impacts on ecosystems and related activities (Francou and Wagnon, 1998; Silverio and Jaquet, 2005; IDEAM, 2012; Francou, 2013; Engdahl and Maxwell, 2015; Polk et al., 2017). Despite this, it is important to mention the complexity of the issue, on the one hand, the decrease of the ice reserve would increase the immediate availability of the water resource, whereas, the reserve potential will decrease when it is reduced too much to maintain them (Zappa et al., 2003; IDEAM, 2012; Francou, 2013; Seehaus et al., 2019).
Lake dynamics, like glacier dynamics, are often complex and can be influenced by factors other than natural ones. The three catchments showed a tendency to increase their area at the end of the study period. Yanayacu catchment showed a greater number and area of lakes, but with little growth, representing a greater degree of stability. Quillcay showed continuous growth, stabilizing in the last 10 years, and Llullan showed greater dynamics with periods of growth and decrease, mainly due to anthropic intervention in the regulation of the Paron lake. This lake has been regulated since the 1990s by Duke Energy Company and, since 2008, by the rural community Cruz de Mayo (Carey et al., 2012), although it has growth significantly since 1989.
The Llullan catchment showed a similar growth rate to Quillcay. While Yanayacu had a minimal variation (increase of ~5%) due to the presence of permanent lakes (maximum storage capacity) and the scarce formation of new lakes. During the study period, in the three catchments new lakes originated in the areas of glacier retreat that presented with favorable conditions for water storage (INAIGEM, 2019), the growth of a lake will be conditioned by the topographic factors.
As for the volume, its behavior was a direct reflection of the variation of its area, empirically based on it as well (Huggel et al., 2002; Loriaux and Casassa, 2013; Kapitsa et al., 2017). The calculation of volumes represented a first multitemporal approach to quantify the water reserve potential in the Cordillera Blanca and may be comparable to the recent study of Muñoz et al. (2020), which tried to find a more significant correlation for the volume of a lake located in tropical glacier zones.
Regarding wetland dynamics, this component presented a greater complexity in its variation, showing periods of growth and decrease according to the years evaluated (Lorini, 2014). In terms of the index used, the results of MSAVI2 were more satisfactory for wetland mapping than NDVI (Aponte-Saravia and Ospina-Noreña, 2019), despite being an unconventional index. This may be due to the improved sensitivity of the index on regions with high percentage of bare soil and areas with snow cover (Finessi and Cogliati, 2019). In the Llullan catchment, evidence of permanent wetlands was not found due to the rugged topography and the small amount of valley bottoms with slopes less than 15°, with a greater presence of shrub communities (shrublands and forests).
In the correlation tests, the results show a significant inverse relationship (between 0.61 to 0.99) of glaciers with lakes in the three catchments. The connection itself is evident, glacier retreat has historically favored the growth and appearance of new lakes in the areas of glacier retreat (Polk et al., 2017; INAIGEM, 2019; Emmer et al., 2020), the loss of glacier area and volume is directly associated with the increase in the surface and volume of the lakes (Emmer et al., 2014). However, the growth of these ponds will be conditioned by ideal topographic factors in which glacier water can be stored (INAIGEM, 2018), especially on flat surfaces with gentle slopes (Emmer et al., 2020). However, with the passage of time and because glaciers have retreated to the point of becoming mountain glaciers, the flat surfaces have been (and will be in the future) limited.
In the case of wetlands, a significant inverse correlation with glaciers was also found (between 0.64 and 0.69) in the Quillcay and Yanayacu catchments. For several authors (e.g., Cochi et al., 2014; Garcia and Otto, 2015; Polk et al., 2017; INAIGEM, 2019) glacier retreat acts as a spatial driving force that increases the area of permanent wetlands, considering that glaciers are the main source of surface water in the dry season (Vuille et al., 2008). Due to glacier retreat, increased surface area and volume of lakes (Emmer et al., 2014), wetlands could be varying as a statistical function of these changes (Polk et al., 2017). As wetlands are influenced by other factors, the study obtained a not so high level of correlation. The most discussed factor is the influence of precipitation, in periods with low precipitation it has been found that permanent wetlands present a progressive decrease in their area (Aponte, 2018) and in the rainy season, temporary wetlands tend to increase (Garcia and Otto, 2015).
Our results for wetlands differ from Zeballos et al. (2014) and Aponte (2018) that mentioned a direct correlation with the reduction of glacier area. Likewise, Buitrón and Fernández (2012) mentioned that wetlands show a tendency to decrease with time, therefore, wetland dynamics is very complex. Several studies of similar high mountain ecosystems show that the area of wetlands can increase, decrease or remain stable (Lorini, 2014; Polk et al., 2017).
Within the study of dynamics, the level of correlation cannot consider other important elements that may affect the variation of these elements; however, it is an indicator of the interconnectivity of ecosystems. This research gives us an indication of what could come in the close future, to the point where some of the elements cease to exist. The influence between these variables may decrease as glacier coverage decreases (Polk et al., 2017), therefore, it is highlighted that glacier dynamics is related to the dynamics of the rest of the high mountain ecosystems as an interacting circulatory system (IDEAM, 2012; Sun et al., 2019).
The dynamics of glaciers, lakes and wetlands are very complex issues and are associated with multiple factors; using remote sensing, GIS and multitemporal analysis historical data can be obtained for a better understanding, being, Landsat satellite images an important data source for the quality of results obtained and its historical record worldwide.
The research demonstrated the existence of a significant inverse correlation of glacier dynamics (area and volume) on the spatial variation of lakes (area and volume, between 0.61 and 0.99, p ≤ 0.05) and wetlands (area, between 0.64 and 0.69, p ≤ 0.05) of high mountain, in the Llullan, Quillcay and Yanayacu catchments of Cordillera Blanca from 1989 and 2019. The results reflected that the lower the glacier area and volume, the greater the area and volume of the lakes, and the area of the wetlands.
In the lakes, there has been evidence of growth and their appearance in the areas of glacier retreat, however, this is conditioned by the topographic factors. In the case of wetlands, the presence of palustrine vegetation cover has become more permanent over time, generally in areas adjacent to pre-existing vegetation cover. Despite the level of correlation found, it is important to mention that there is no definitive certainty that glacier retreat is the primary and determining factor in wetland dynamics and will always be influenced by other factors (e.g., precipitation patterns). While changes in glacier dynamics inversely influenced lakes and wetlands from 1989 and 2019, their future projection may be constrained and altered by other morphological, topographical, and climatic variables. This research highlights the importance of the interconnectivity between these elements within the high mountain hydrological cycle as sources of the water reserve potential of a catchment with glaciers.
In general terms, it is concluded that, in the short term, glacier retreat generates a favorable response on the eco-hydrological environment, where water resources such as lake areas and wetlands increase. In the long term, glaciers will shrink, so water resources will decrease (increasing annual variability), and ecological degradation will increase, i.e., the change of the cryosphere will have a negative impact on the high mountain eco-hydrological environment.
This research was funded by CONCYTEC through the PROCIENCIA program within the framework of the “Basic Research Projects 2024-03” competition, under contract [PE501088551-2024-PROCIENCIA], project “Analysis of chronosequences of glacier retreat areas and their relationship with the emergence of terrestrial and aquatic ecosystems in Cordillera Blanca, Peru (CronoEco-Glaciar)”.
Adler, C., Huggel, C., Orlove, B., Nolin, A. (2019). Climate change in the mountain cryosphere: impacts and responses. Regional Environmental Change, 19(5), 1225–1228. https://doi.org/10.1007/s10113-019-01507-6
Aponte, J. (2018). Dinamica de Cambios Superficiales en Humedales Cordillera La Viuda-Perú, mediante indices espectrales y modelo digital de elevacion, periodo 1985-2016. Universidad Nacional de Colombia.
Aponte-Saravia, J., Ospina-Noreña, J. E. (2019). Evaluando el desempeño de índices espectrales para identificar humedales alto andinos. Revista de Teledetección, 53, 59–72. https://doi.org/10.4995/raet.2019.10580
Armstrong, R. L., Rittger, K., Brodzik, M. J., Racoviteanu, A., Barrett, A. P., Khalsa, S. J. S., Raup, B., Hill, A. F., Khan, A. L., Wilson, A. M., Kayastha, R. B., Fetterer, F., Armstrong, B. (2018). Runoff from glacier ice and seasonal snow in High Asia: separating melt water sources in river flow. Regional Environmental Change, 19(5), 1249–1261. https://doi.org/10.1007/s10113-018-1429-0
Baraer, M., Mark, B. G., Mckenzie, J. M., Condom, T., Bury, J., Huh, K. I., Portocarrero, C., Gómez, J., Rathay, S. (2012). Glacier recession and water resources in Peru’s Cordillera Blanca. Journal of Glaciology, 58(207), 134–150. https://doi.org/10.3189/2012JoG11J186
Benn, D. I., Bolch, T., Hands, K., Gulley, J., Luckman, A., Nicholson, L. I., Quincey, D., Thompson, S., Toumi, R., Wiseman, S. (2012). Response of debris-covered glaciers in the Mount Everest region to recent warming, and implications for outburst flood hazards. Earth-Science Reviews, 114(1–2), 156–174. https://doi.org/10.1016/j.earscirev.2012.03.008
Bernex, N., Tejada, M. (2010). Cambio climático, retroceso glaciar y gestión integrada de los recursos hídricos.
Buitrón, C., Fernández, J. (2012). Estudio espacial multitemporal de variaciones en superficie observadas a través de imágenes satelitales Landsat en una región del parque nacional Sajama Bolivia.
Carey, M., French, A., O’Brien, E. (2012). Unintended effects of technology on climate change adaptation: An historical analysis of water conflicts below Andean Glaciers. Journal of Historical Geography, 38(2), 181–191. https://doi.org/10.1016/j.jhg.2011.12.002
Carey, M., Molden, O. C., Rasmussen, M. B., Jackson, M., Nolin, A. W., Mark, B. G. (2017). Impacts of Glacier Recession and Declining Meltwater on Mountain Societies. Annals of the American Association of Geographers, 107(2), 1–10. https://doi.org/10.1080/24694452.2016.1243039
Cerna, M. (2012). Cuantificacion del cambio de volumen de los glaciares de la Cordillera Blanca utilizando los modelos de elevacion digital topograficos y de imagenes ASTER: Nevados Champara y Huascaran. Universidad Nacional Mayor de San Marcos.
Cochi, N., Prieto, G., Dangles, O., Rojas, A., Ayala, C., Condori, B., Casazola, J. L. (2014). Metodología para evaluar el potencial productivo y la dinámica socioecológica de la ganadería en bofedales altoandinos. Ecología En Bolivia, 49(3), 120–131.
Cuadros, B. (2014). Evolución de la cobertura glaciar del nevado Ananea entre 1985-2010. Revista Científica Investigación Andina, 14(2), 57–67.
Denzinger, F., Machguth, H., Barandun, M., Berthier, E., Girod, L., Kronenberg, M., Usubaliev, R., Hoelzle, M. (2021). Geodetic mass balance of Abramov Glacier from 1975 to 2015. Journal of Glaciology, 67(262), 331–342. https://doi.org/10.1017/jog.2020.108
Emmer, A., Harrison, S., Mergili, M., Allen, S., Frey, H., Huggel, C. (2020). 70 years of lake evolution and glacial lake outburst floods in the Cordillera Blanca (Peru) and implications for the future. Geomorphology, 365(107178), 1–11. https://doi.org/10.1016/j.geomorph.2020.107178
Emmer, A., Vilímek, V., Klimeš, J., Cochachin, A. (2014). Glacier Retreat, Lakes Development and Associated Natural Hazards in Cordilera Blanca, Peru. Environmental Science and Engineering, 231–252. https://doi.org/10.1007/978-3-319-00867-7
Engdahl, N. B., Maxwell, R. M. (2015). Quantifying changes in age distributions and the hydrologic balance of a high-mountain watershed from climate induced variations in recharge. Journal of Hydrology, 522, 152–162. https://doi.org/10.1016/j.jhydrol.2014.12.032
Finessi, F., Cogliati, M. (2019). Estudio de las áreas de vegetación en el norte de la provincia de Neuquén con teledetección. 14° Encuentro Del Centro Internacional de Ciencias de La Tierra.
Francou, B. (2013). El rápido retroceso de los glaciares en los Andes tropicales: Un desafío para el estudio de la dinámica de los ecosistemas de alta montaña. Ecología En Bolivia, 48(2), 69–71.
Francou, B., Wagnon, P. (1998). Cordilléres andines, sur les hauts sommets de Bolivie, du Pérou et d’Equateur. Glénat.
Garcia, E., Otto, M. (2015). Caracterización Ecohidrológica de humedales alto andinos usando imágenes de satélite multitemporales en la cabecera de cuenca del Río Santa, Áncash, Perú. Ecología Aplicada, 14(2), 115–125. https://doi.org/10.21704/rea.v14i1-2.88
Garg, P. K., Shukla, A., Jasrotia, A. S. (2019). On the strongly imbalanced state of glaciers in the Sikkim, eastern Himalaya, India. Science of the Total Environment, 691, 16–35. https://doi.org/10.1016/j.scitotenv.2019.07.086
Hall, D., Bayr, K., Schöner, W., Bindschadler, R., Chien, J. (2003). Consideration of the errors inherent in mapping historical glacier positions in Austria from the ground and space (1893-2001). Remote Sensing of Environment, 86(4), 566–577. https://doi.org/10.1016/S0034-4257(03)00134-2
Hall, D., Riggs, G., Salomonson, V. (1995). Development of Methods for Mapping Global Snow Cover Using Moderate Resolution Imaging Spectroradiometer Data. Remote Sensing of Environment, 54, 127–140.
Hanshaw, M. N., Bookhagen, B. (2014). Glacial areas, lake areas, and snow lines from 1975 to 2012: Status of the cordillera vilcanota, including the Quelccaya Ice Cap, northern central Andes, Peru. The Cryosphere, 8(2), 359–376. https://doi.org/10.5194/tc-8-359-2014
Hidrandina SA. (1989). Inventario de glaciares del Perú. Unidad de Glaciología e Hidrología.
Huggel, C., Kääb, A., Haeberli, W., Teysseire, P., Paul, F. (2002). Remote sensing based assessment of hazards from glacier lake outbursts: A case study in the Swiss Alps. Canadian Geotechnical Journal, 39(2), 316–330. https://doi.org/10.1139/t01-099
IDEAM. (2012). Glaciares de Colombia más que montañas con hielo. Instituto de Hidrología, Meteorología y Estudios Ambientales.
INAIGEM. (2017). Manual metodológico de inventario nacional de glaciares. Instituto Nacional de Investigación en Glaciares y Ecosistemas de Montaña.
INAIGEM. (2018). Inventario nacional de glaciares: las Cordilleras Glaciares del Perú. Instituto Nacional de Investigación en Glaciares y Ecosistemas de Montaña.
INAIGEM. (2019). Informe de la Situación de los Glaciares y Ecosistemas de Montaña 2018.
INAIGEM. (2021). Inventario de lagunas y glaciares del Perú 2016. Geoportal INAIGEM. https://www.arcgis.com/apps/webappviewer/index.html?id=5379ac94516a4cb0a7f1cd0fa7bfc94b
INAIGEM. (2023). Memoria descriptiva del Inventario Nacional de Glaciares y Lagunas de Origen Glaciar del Perú. Instituto Nacional de Investigación en Glaciares y Ecosistemas de Montaña (INAIGEM).
Ji, L., Zhang, L., Wylie, B. (2009). Analysis of dynamic thresholds for the Normalized Difference Water Index. Photogrammetric Engineering & Remote Sensing, 75(11), 1307–1317. https://doi.org/10.14358/PERS.75.11.1307
Kapitsa, V., Shahgedanova, M., Machguth, H., Severskiy, I., Medeu, A. (2017). Assessment of Evolution of Mountain Lakes and Risks of Glacier Lake Outbursts in the Djungarskiy (Jetysu) Alatau, Central Asia, using Landsat Imagery and Glacier Bed Topography Modelling. Natural Hazards and Earth System Sciences Discussions, May, 1–54. https://doi.org/10.5194/nhess-2017-134
Linsbauer, A., Paul, F., Haeberli, W. (2012). Modeling glacier thickness distribution and bed topography over entire mountain ranges with glabtop: Application of a fast and robust approach. Journal of Geophysical Research: Earth Surface, 117(3), 1–17. https://doi.org/10.1029/2011JF002313
Linsbauer, A., Paul, F., Hoelzle, M., Frey, H., Haeberli, W. (2009). The Swiss Alps without glaciers - a GIS-based modelling approach for reconstruction of glacier beds. Proceedings of Geomorphometry, 243–247. https://doi.org/10.5167/uzh-27834
Loarte, E. (2011). Determinación de la variación espacio-temporal de la Línea de Equilibrio (ELA) a través la Línea de Nieve (SLA) de los glaciares de la Cordillera Blanca, durante el periodo 2001-2010. Universidad Nacional Santiago Antúnez de Mayolo.
Loarte, E., Rabatel, A., Gomez, J. (2015). Determinación de la variación espacio-temporal de la altura de la línea de equilibrio de los glaciares de la Cordillera Blanca, Perú. Revista Peruana Geo-Atmosférica RPGA, 4, 19–30.
Loriaux, T., Casassa, G. (2013). Evolution of glacial lakes from the Northern Patagonia Icefield and terrestrial water storage in a sea-level rise context. Global and Planetary Change, 102, 33–40. https://doi.org/10.1016/j.gloplacha.2012.12.012
Lorini, H. (2014). Estrategia de adaptación al cambio climático para humedales altoandinos.
Mark, B. G., Mckenzie, J. M. (2007). Tracing increasing tropical Andean glacier melt with stable isotopes in water. Environmental Science and Technology, 41(20), 6955–6960. https://doi.org/10.1021/es071099d
Mark, B. G., McKenzie, J. M., Gómez, J. (2005). Hydrochemical evaluation of changing glacier meltwater contribution to stream discharge: Callejon de Huaylas, Peru. Hydrological Sciences Journal, 50(6), 975–988. https://doi.org/10.1623/hysj.2005.50.6.975
Mark, B. G., Seltzer, G. O. (2003). Tropical glacier meltwater contribution to stream discharge: A case study in the Cordillera Blanca, Peru. Journal of Glaciology, 49(165), 271–281. https://doi.org/10.3189/172756503781830746
Marzeion, B., Cogley, J. G., Richter, K., Parkes, D. (2014). Attribution of global glacier mass loss to anthropogenic and natural causes. Science, 345, 919–921. https://doi.org/10.1126/science.1254702
McFeeters, S. K. (1996). The use of the Normalized Difference Water Index (NDWI) in the delineation of open water features. International Journal of Remote Sensing, 17(7), 1425–1432. https://doi.org/10.1080/01431169608948714
Muñoz, R., Huggel, C., Frey, H., Cochachin, A., Haeberli, W. (2020). Glacial lake depth and volume estimation based on a large bathymetric dataset from the Cordillera Blanca, Peru. Earth Surface Processes and Landforms, 45(7), 1510–1527. https://doi.org/10.1002/esp.4826
Osmaston, H. (2005). Estimates of glacier equilibrium line altitudes by the Area × Altitude, the Area × Altitude Balance Ratio and the Area × Altitude Balance Index methods and their validation. Quaternary International, 138–139, 22–31. https://doi.org/10.1016/j.quaint.2005.02.004
Paul, F., Linsbauer, A. (2012). Modeling of glacier bed topography from glacier outlines, central branch lines, and a DEM. International Journal of Geographical Information Science, 26(7), 1173–1190. https://doi.org/10.1080/13658816.2011.627859
Pieczonka, T., Bolch, T., Kröhnert, M., Peters, J., Liu, S. (2018). Glacier branch lines and glacier ice thickness estimation for debris-covered glaciers in the Central Tien Shan. Journal of Glaciology, 64(247), 835–849. https://doi.org/10.1017/jog.2018.75
Polk, M. H., Young, K. R., Baraer, M., Mark, B. G., McKenzie, J. M., Bury, J., Carey, M. (2017). Exploring hydrologic connections between tropical mountain wetlands and glacier recession in Peru’s Cordillera Blanca. Applied Geography, 78, 94–103. https://doi.org/10.1016/j.apgeog.2016.11.004
Qi, J., Chehbouni, A., Huete, A. R., Kerr, Y. H., Sorooshian, S. (1994). A Modified Soil Adjusted Vegetation Index. Remote Sensing of Environment, 48(2), 119–126. https://doi.org/10.1016/0034-4257(94)90134-1
Rabatel, A., Bermejo, A., Loarte, E., Soruco, A., Gomez, J., Leonardini, G., Vincent, C., Sicart, J. E. (2012). Can the snowline be used as an indicator of the equilibrium line and mass balance for glaciers in the outer tropics? Journal of Glaciology, 58(212), 1027–1036. https://doi.org/10.3189/2012JoG12J027
Racoviteanu, A. E., Arnaud, Y., Williams, M., Ordoñez, J. (2008). Decadal changes in glacier parameters in the Cordillera Blanca, Peru, derived from remote sensing. Journal of Glaciology, 54(186), 499–510. https://doi.org/10.3189/002214308785836922
Rios, R. (2020). Delimitación de Ecosistemas de Montaña usando Árbol de decisiones con datos de sensores remotos. https://www.inaigem.gob.pe/wp-content/uploads/2020/02/ViernesCientificoRRR_blanco.pdf
Rios, R. (2024). Metodología para el mapeamiento detallado de ecosistemas de montaña en el Parque Nacional Huascarán [Universidad Nacional Agraria La Molina]. In Universidad Nacional Agraria La Molina. https://repositorio.concytec.gob.pe/bitstream/20.500.12390/187/3/2017_Puicon_Evaluacion-resistencia-natural.pdf
Rios, R., Fuentealba, B. (2020). Árbol de decisiones para delimitación de ecosistemas de montaña en la U.H. Pariac - Rajucolta, aplicando teledetección.
Rojas, J., Tordocillo, J. (2018). Estimación del Área de la Superficie y el Cambio de Volumen del Glaciar del Nevado Champará (Cordillera Blanca, Perú) a partir de las Imágenes y los Modelos de Elevación Digital del Sensor ASTER / Terra (2000-2010). Revista de Glaciares y Ecosistemas de Montaña, 4, 23–42. https://doi.org/10.36580/rgem.i4.23-42
Santiago, F., Mallqui, H., Rios, R. (2021). Mapeo de la cobertura vegetal en la subcuenca Quillcay (Ancash, Perú) con el clasificador de Árbol de decisiones. Aporte Santiaguino, 14(1), 78–91. https://doi.org/10.32911/as.2021.v14.n1.761
Schauwecker, S., Rohrer, M., Huggel, C., Endries, J., Montoya, N., Neukom, R., Perry, B., Salzmann, N., Schwarb, M., Suarez, W. (2017). The freezing level in the tropical Andes, Peru: An indicator for present and future glacier extents. Journal of Geophysical Research, 122(10), 5172–5189. https://doi.org/10.1002/2016JD025943
Seehaus, T., Malz, P., Sommer, C., Lippl, S., Cochachin, A., Braun, M. (2019). Changes of the tropical glaciers throughout Peru between 2000 and 2016 - Mass balance and area fluctuations. The Cryosphere, 13(10), 2537–2556. https://doi.org/10.5194/tc-13-2537-2019
Seltzer, G. O. (1993). Late-Quaternary Glaciation Andes for Climate Central Change. Mountain Research and Development, 13(2), 129–138.
Silverio, W., Jaquet, J. M. (2005). Glacial cover mapping (1987-1996) of the Cordillera Blanca (Peru) using satellite imagery. Remote Sensing of Environment, 95(3), 342–350. https://doi.org/10.1016/j.rse.2004.12.012
Silverio, W., Jaquet, J. M. (2017). Evaluating glacier fluctuations in Cordillera Blanca (Peru) by remote sensing between 1987 & 2016 in the context of ENSO. Archives Des Sciences, 69(2), 145–161.
Sun, X., Zhang, R., Huang, W., Sun, A., Lin, L., Xu, H., Jiang, D. (2019). The response between glacier evolution and eco-geological environment on the Qinghai-Tibet Plateau. China Geology, 2(1), 1–7. https://doi.org/10.31035/cg2018078
Turpo, E. Y., Borja, M. O., Espinoza-Villar, R., Moreno, N., Camargo, R., Almeida, C., Hopfgartner, K., Yarleque, C., Souza, C. M. (2022). Mapping Three Decades of Changes in the Tropical Andean Glaciers Using Landsat Data Processed in the Earth Engine. Remote Sensing, 14(9). https://doi.org/10.3390/rs14091974
Ubeda, J. (2011). El impacto del cambio climático en los glaciares del complejo volcánico Nevado Coropuna, (Cordillera Occidental de los Andes Centrales). Universidad Complutense de Madrid.
USGS. (2021). Landsat collection 1. https://www.usgs.gov/landsat-missions/landsat-collection-1?qt-science_support_page_related_con=1#qt-science_support_page_related_con
Van Wyk de Vries, M., Carchipulla-Morales, D., Wickert, A. D., Minaya, V. G. (2022). Glacier thickness and ice volume of the Northern Andes. Scientific Data, 9(1), 1–16. https://doi.org/10.1038/s41597-022-01446-8
Vörösmarty, C. J. (2009). The Earth’s natural water cycles. In World Water Development Report 3. https://doi.org/10.1016/B978-0-12-409548-9.11203-5
Vuille, M., Francou, B., Wagnon, P., Juen, I., Kaser, G., Mark, B. G., Bradley, R. S. (2008). Climate change and tropical Andean glaciers: Past, present and future. Earth-Science Reviews, 89(3–4), 79–96. https://doi.org/10.1016/j.earscirev.2008.04.002
Wang, X., Guo, X., Yang, C., Liu, Q., Wei, J., Zhang, Y., Liu, S., Zhang, Y., Jiang, Z., Tang, Z. (2020). Glacial lake inventory of High Mountain Asia (1990–2018) derived from Landsat images. Earth System Science Data Discussions, 12, 2169–2182. https://doi.org/10.5194/essd-12-2169-2020
WGMS. (2017). Global glacier change bulletin No. 2 (2014–2015). https://doi.org/10.5904/wgms-fog-2017-10
Zapana, M., Peña, R., Pachac, Y. (2024). Glacier Volume Estimation Using the Glaptop Model in the Cordillera Blanca of Peru. ESPOCH Congresses: The Ecuadorian Journal of S.T.E.A.M., 3(4), 62–71. https://doi.org/10.18502/espoch.v3i4.17165
Zappa, M., Pos, F., Strasser, U., Warmerdam, P., Gurtz, J. (2003). Seasonal water balance of an Alpine Catchment as Evaluated by different methods for spatially distributed snowmelt modelling. Nordic Hydrology, 34(3), 179–202. https://doi.org/10.2166/nh.2003.0003
Zeballos, G., Soruco, Á., Cusicanqui, D., Joffré, R., Rabatel, A. (2014). Uso de imágenes satelitales, modelos digitales de elevación y sistemas de información geográfica para caracterizar la dinámica espacial de glaciares y humedales de alta montaña en Bolivia. Ecología En Bolivia, 49(3), 14–26.
Zhao, X., Wang, X., Wei, J., Jiang, Z., Zhang, Y., Liu, S. (2020). Spatiotemporal variability of glacier changes and their controlling factors in the Kanchenjunga region, Himalaya based on multi-source remote sensing data from 1975 to 2015. Science of the Total Environment, 745, 140995. https://doi.org/10.1016/j.scitotenv.2020.140995
To cite this article: Castillo-Vergara, F., Loarte, E., Medina, K., Rodriguez-Venturo, S., Tuya, E. 2025. Spatiotemporal variation of glacier dynamics and its relationship with changes in high mountain ecosystems in the Cordillera Blanca, Peru. Revista de Teledetección, 66, e22235. https://doi.org/10.4995/raet.2025.22235
* Corresponding author: fcastillov@unasam.edu.pe