Glacio-hydrological changes along the Andes throughout the 21st Century

Glacio-hydrological changes along the Andes throughout the 21st Century | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Glacio-hydrological changes along the Andes throughout the 21st Century Alexis Caro, Thomas Condom, Antoine Rabatel, Rodrigo Aguayo, Nicolas Champollion This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4714636/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 31 Mar, 2025 Read the published version in Scientific Reports → Version 1 posted 9 You are reading this latest preprint version Abstract Assessing future glacier water contributions is crucial for resource management. However, a large gap persists regarding Andean glacier runoff research. We evaluated eight CMIP6 models (1990-2049) on a glacierized area of 27,669 km², and projected future changes in glacier runoff (on 11,282 km² related to land-terminating glaciers) using an evaluated glaciological model in 778 catchments under two extreme SSP scenarios to estimate the year of maximum glacier runoff (peak water). We evaluated the different GCMs over the historical period using in situ data and showed that some of them perform better in specific glaciological regions. For the mid-21st century, warming trends are projected across the Andes, especially in the Tropical Andes (+0.7°C), while precipitation is expected to decrease slightly in the Southern Andes (-1 to -3%). These variables significantly affect glacier dynamics and runoff estimates. Glacier runoff estimates spanning 2000-2019 and projected to 2030-2049 indicate significant declines in the Tropical Andes (-43%) and Dry Andes (-37%), and a lesser decrease in the Wet Andes (-32%). Notably, the Atuel (-62%) and Tupungato (+32%) catchments in the Dry Andes show highly contrasted changes in annual runoff across the Andes. In terms of peak water, most catchments are expected to reach it before the first half of the 21st century (between 2010 and 2028). Our study underlines the critical importance of considering seasonal variations when analyzing GCMs in hydro-glaciological simulations and emphasizes regional disparities in glacier runoff across the Andes for future water resources management. Earth and environmental sciences/Climate sciences Earth and environmental sciences/Hydrology Figures Figure 1 Figure 2 Figure 3 1. Introduction In the Andes, most of the knowledge regarding future glacier changes is derived from global simulations ( e.g. , Marzeion et al., 2012 ; Radic and Hock, 2014 ; Huss and Hock, 2015 ; Rounce et al., 2023 ). The recent study by Rounce et al. ( 2023 ) stands out as it incorporates geodetic mass balance measurements obtained at the scale of individual glaciers by Hugonnet et al. ( 2021 ) to calibrate the glacier mass balance during the historical period. It is important to note that these global-scale studies have not been specifically evaluated for the Andes, even though simulating current observations is a crucial prerequisite for predictive models (Aschwanden et al., 2013 ). The reported glacier volume changes projected throughout the 21st century by the above mentioned studies demonstrate consistent results in the Tropical Andes, with an approximate loss of glacier mass of around − 98 ± 13% by 2100 under the RCP8.5 scenario based on the Coupled Model Intercomparison Project phase 5 (CMIP5) models. However, in the southern Andes, which encompasses the largest glacierized area, there is a wider range of mass loss estimates, ranging from − 44 ± 14% to -68 ± 20% (Huss and Hock, 2015 ; Rounce et al., 2023 ). Even under the most optimistic scenarios ( i.e. RCP2.6), the reduction in glacier volume remains significant. Furthermore, a global study by Huss and Hock ( 2018 ) focused on 12 Andean catchments and estimated that glacier runoff, which includes ice and snow melt as well as rainfall on glaciers, is projected to increase in most of the catchments until 2050. However, after 2050, it is expected to decrease in all catchments except for the Santa Cruz catchment (49°S, Argentina). Local simulations of future glacier changes across the Andes have been conducted, encompassing glaciers in Colombia, Ecuador, Peru, Bolivia, and Chile ( e.g. , Frans et al., 2015 ; Réveillet et al., 2015 ; Yarleque et al., 2018 ; Vuille et al., 2018 ; Rabatel et al., 2018 ; Scheiter et al., 2021 ) as well as the Patagonian icefield ( e.g. , Schaefer et al., 2013 ; Bravo et al., 2021 ). These studies focus on different objectives ( e.g. , surface mass balance, glacier dynamic, glacier runoff) and use different CMIP5 models. In the tropical Andes, Vuille et al. ( 2018 ) estimated that the Antizana Glacier (0°S, inner tropics) is more vulnerable to warming throughout the 21st century in comparison with the Zongo Glacier (16°S, outer tropic). For Zongo Glacier, projected volume losses range from − 40 ± 7% to -89 ± 4% between 2010 and 2100 depending on the considered RCP (Réveillet et al., 2015 ) and a discharge reduction in 2100 was estimated by 25% at the annual scale and by 57% during the dry season for RCP4.5 (Frans et al., 2015 ). In the southern Andes, no study was performed in the Dry Andes, but for the Wet Andes, Scheiter et al. ( 2021 ) projected an ice volume loss between − 56 and − 97% depending on the RCP for the Mocho Choshuenco glacier in 2100. Two other studies reported future glacier changes in the Patagonian icefields. In the Northern Patagonian Icefield (NPI) a strong increase in ablation is estimated from 2050 onward with a reduction of solid precipitation from 2080 onward due to higher temperatures, with uncertainties arising from future climate and ice dynamics (Schaefer et al., 2013 ). Bravo et al. ( 2021 ) compared simulations for the period 2005–2050 with the historical period 1976–2005 and estimated a larger reduction in annual mass balance between − 1.5 to -1.9 m w.e./yr for the NPI compared to the Southern Patagonian Icefield (SPI) (-1.1 to -1.5 m w.e/yr). As glaciers continue to reduce under projected climate change scenarios, it becomes imperative to ascertain the timing of peak water (PW) - the period when glacier runoff increases before eventually declining - throughout the Andes taking into account regional differences. This knowledge holds paramount importance as it enables stakeholders to anticipate when glacier contributions to river flows will cease in the future. Because of that, the influence of uncertainties in future climate scenarios on future glacier changes has been a subject of investigation, as discussed by Marzeion et al. ( 2020 ). Their study indicates that both at the global and regional scales, the impacts of uncertainties in future climate scenarios increase over the course of the 21st century. However, in contrast, the uncertainties related to the glacier model parameterization decrease over time. Furthermore, Hausfather et al. ( 2022 ) found that more than one-quarter of the models in the Coupled Model Intercomparison Project 6 (CMIP6) (Eyring et al., 2016) have higher variability in temperature compared to the CMIP5 models. This higher variability in temperature projections could introduce additional uncertainty in the estimates of future glacier changes. Similarly, Tokarska et al. ( 2020 ) highlighted that certain CMIP6 models with high climate sensitivity ( i.e. beyond the AR5 likely range of 1.5°-4.5°C by the end of the 21st century) tend to overestimate historical warming trends. Consequently, this bias might lead to future warming projections being biased towards higher temperatures in these CMIP6 models. Conversely, CMIP6 models with climate sensitivity values within the likely range exhibit warming trends consistent with observations over the historical period, providing more reliable estimates for future climate scenarios and their impact on glaciers. This study aims to address the lack of specific estimates regarding future glacier changes and their hydrological implications in the Andean glacierized catchments. For this we use a calibrated/validated model, which incorporates corrected climate variables based on measurements during the historical periodo 2000–2019. Additionally, we have two main objectives. Firstly, we evaluate the performance of eight GCMs sourced from CMIP6 across the Andes (11°N-55°S) for both historical (1990–2019) and future periods (2020–2049). Secondly, we utilize an ensemble of evaluated GCMs (complete ensemble) and a filtered ensemble throughout the first half of the 21st century to simulate glacier runoff (including ice and snowmelt) since 2000. 2. Results 2.1. GCMs analysis for the historical and future periods 2.1.1. GCMs analysis for the historical period This section analyzes the correlations and errors between downscaled Global Climate Models (GCMs) and corrected TerraClimate data (cTC) across the Andes region, focusing on a monthly and seasonal scale during the historical period (1990–2019) for 3,213 glaciers (covering 27,669 km 2 ). At the monthly scale, GCMs and cTC temperature (r = 0.9) and precipitation (r = 0.4) exhibit statistically significant correlations on all glaciers (Figure S1 E). The mean monthly temperature error is 1.1 ± 0.1°C, and the monthly total precipitation error is 106 ± 68 mm. When considering glaciological regions, the Dry and Wet Andes show the largest errors in temperature and precipitation, followed by the Tropical Andes (Figure S2). At a seasonal scale, there are significant correlations for temperature and precipitation on a moderate proportion of glaciers (above 30%) during specific seasons (Table S1 ). For example, in the Tropical Andes, correlations are significant in JFM and AMJ for precipitation, and in OND for temperature. In the Dry Andes, a considerable number of glaciers show significant correlations in all seasons. Meanwhile, in the Wet Andes, significant correlations are observed in JFM and AMJ for temperature and during JAS and OND for precipitation. Considering that the largest glacier mass loss occurs through the transition season (OND) in the Tropical Andes and during the summer season (JFM) in the Dry and Wet Andes, we scored the eight GCMs regarding their performance for temperature and precipitation in these seasons (see Figure S1 A-D). In the Tropical Andes, the largest number of glaciers is better correlated with INM-CM5, GFDL, NorESM2 (for temperature) and FGOALS and INM-CM5 (for precipitation). For the Dry Andes, high performance was estimated from CAMS, NorESM2 (for temperature) and MPI, CESM2 (for precipitation). Regarding the Wet Andes, the models INM-CM4 (for temperature) and GFDL, INM-CM4, FGOALS, and MPI (for precipitation) exhibit the highest correlations in the majority of glaciers. Conversely, the models MPI and INM-CM4 in the Tropical Andes, INM-CM4 in the Dry Andes, and MPI, INM-CM5 and FGOALS in the Wet Andes are relevant for only a very small number of glaciers. These findings provide valuable insights into the GCMs' performance and allow considering the most relevant one to simulate the future glacier mass loss across the Andes. 2.1.2. Future climate change on glaciers We analyze the differences in mean temperature and precipitation during the periods 1990–2019 and 2020–2049 for each individual glacier considering the eight GCMs. The differences led us to define the likely ranges of climate change by glaciological region considering the percentiles 10 and 90. The objective is to identify the models that are outside the regional climate change likely ranges at the annual and seasonal scales, which are then considered as hot/cold and dry/wet models. At an annual scale, both scenarios (SSP1-2.6 and SSP5-8.5) exhibit the most significant temperature increase in the Tropical Andes (median = 0.7 and 0.9°C, respectively), followed by the Dry (median = 0.6 and 0.9°C) and Wet Andes (median = 0.4 and 0.6°C). Notably, SSP5-8.5 depicts a warmer outcome. In contrast, precipitation is projected to decrease in all glaciological regions and scenarios, except for the Tropical Andes under the SSP1-2.6 scenario (median = + 0.6%). The Dry Andes show the most considerable precipitation reduction (median = -2.8 and − 1.9%), trailed by the Wet Andes (median = -2.6 and − 0.8%) and the Tropical Andes (median = -0.8, SSP5-8.5). At a seasonal scale, the largest increase in temperature is observed in the SSP5-8.5 scenario, particularly in the Tropical Andes (JAS), followed by the Dry (OND and JFM = + 1.0°C) and Wet Andes (JFM = + 0.7°C). Regarding precipitation changes, the Tropical Andes (JFM) are estimated to experience the most significant increase, while the Dry Andes show lower negative median total precipitation during OND. Detailed percentile values can be found in Table S2 and Fig. 1. Temperature and precipitation play a crucial role in glacier mass loss during the transition (OND) and wet seasons (JFM) in the Tropical Andes. The largest glacier ablation occurs during the transition season (Autin et al., 2022 ) and any delay in precipitation during the wet season can lead to a significant increase in ablation rates (Rabatel et al., 2013 ). In the Tropical Andes, the models project a median temperature increase in 0.7°C (SSP1-2.6) and 0.9°C (SSP5-8.5) during both seasons. Precipitation is expected to decrease during the transition season by -0.9 to -0.8% and increase during the wet season by 1.4 to 1.7%. Conversely, in the Southern Andes, glacier accumulation is concentrated in autumn/winter (AMJ and JAS), although significant precipitation also occurs in spring and summer in the Wet Andes (OND and JFM). During summer, the increased precipitation contributes to reducing the strong glacier ablation rates due to an increase in albedo. In the Dry Andes, models indicate a median temperature increase in 0.6°C (SSP1-2.6) and 1°C (SSP5-8.5) during spring and summer. However, scenarios differ significantly in terms of precipitation. Under the SSP5-8.5 scenario, median precipitation is projected to decrease by 8.8% during spring and remain unchanged (0%) during summer. On the other hand, the SSP1-2.6 scenario shows an increase in median precipitation during spring (1.2%) and a reduction during summer (-2.7%). As for the Wet Andes, both scenarios of climate change exhibit a reduction in median precipitation, with the largest impact observed during summer (precipitation reduction of -2.4 to -4.5%) compared to spring (precipitation reduction of -1.9 to -4%). Additionally, the median temperature increase is more significant in summer (temperature rise of 0.5–0.7°C) than in spring (temperature rise of 0.4–0.5°C). For detailed values, please refer to Table S2. From the regional temperature and precipitation likely ranges (Fig. 1A-B for annual and Fig. 1C for seasonal), we identified hot/cold and dry/wet models. On an annual scale, for the Tropical Andes, GCMs' median values remain within likely ranges for both climate change scenarios. However, in the Dry Andes, FGOALS exhibits hot/dry characteristics and CAMS is a wet model. Moving to the Wet Andes, CESM2 and GFDL models are dry and cold, respectively, while FGOALS is a hot model. At the seasonal scale, FGOALS and CECSM2 display mean values outside the likely ranges. Specifically, in the Tropical Andes, CESM2 is a hot model during (OND). In the Dry Andes, FGOALS (OND) and CESM2 (JFM) are hot models, whereas CAMS (OND) and FGOALS (JFM) are wet models. In the Wet Andes, the FGOALS model is hot/dry (OND). In summary, considering analysis performed in sections 2.1.2 and 2.2.2 , we identified that the highest score GCMs were: CAMS, FGOALS, GFDL, INM-CM5, and NorESM2 in the Tropical Andes; GFDL, INM-CM5, MPI, and NorESM2 for the Dry Andes; CAMS, INM-CM4, and NorESM2 in the Wet Andes. In the following, we analyze changes in glacier runoff and PW considering these GCMs (filtered ensemble) and considering all the GCMs (complete ensemble). 2.2. Future glacier evolution 2.2.1. Changes in glacier runoff for the period 2030–2049 This section focuses on analyzing the mean annual changes in glacier runoff between two distinct periods: 2000–2019 and 2030–2049. To estimate these changes, we employ the filtered ensemble of GCMs. The analysis is conducted at both the glaciological region and catchment scales taking into consideration the cumulative volumes of glacier runoff by catchment. Additionally, we computed the changes in mean annual glacier runoff using the complete ensemble of GCMs. For more detailed information, refer to Figure S4 in the Supplementary Information. Figure 2 illustrates the mean annual changes in glacier runoff by comparing the historical and future periods for the 778 catchments analyzed in the Andes. At the scale of the glaciological regions, comparing the two periods 2000–2019 and 2030–2049, the simulations made using the filtered GCMs ensemble reveal a median reduction in glacier runoff of -0.1 m 3 s -1 for scenarios SSP1-2.6 y SSP5-8.5 in the Tropical Andes. In the Dry Andes, we observe differences in glacier runoff changes between the two scenarios. Simulations made with SSP1-2.6 display a higher number of catchments with reduced glacier runoff (median = -0.01 m 3 s -1 ), indicating a decrease in overall melt rates. Conversely, for SSP5-8.5, a significant number of catchments exhibit minimal changes in glacier runoff (median = -0.003 m 3 s -1 ). However, in the catchments of the highest decile (above the 90th percentile), an increase in glacier runoff (0.72 m 3 s -1 ) is observed in the SSP5-8.5 scenario. For the Wet Andes, simulations made with both scenarios indicate a reduction in glacier runoff of -0.2 m 3 s -1 , suggesting decreased glacier runoff rates. Negative values persist until the 90th percentile of catchments, indicating a consistent reduction in glacier runoff for this region. At the regional scale, changes in the glacier runoff using the filtered GCMs ensemble show similar amounts than the simulations from the complete GCMs ensemble, whereas, at the catchment scale lower changes are estimated from the filtered ensemble. Three distinct types of behavior in glacier runoff changes are observed across catchments for the SSP1-2.6 and SSP5-8.5 scenarios (Table S3). These behaviors include positive changes, negative changes, and catchments exhibiting positive changes under SSP5-8.5 and negative changes under SSP1-2.6. The largest volume changes in glacier runoff are consistently negative across all regions and time periods (2000–2019 and 2030–2049) for both scenarios. The Tropical Andes exhibit the most significant annual cumulative loss in terms of percentage regarding the historical period, with a reduction of 43% (-25.8 m 3 s -1 in SSP1-2.6). The Dry Andes follow, with a cumulative loss of 37% (-14.4 m 3 s -1 in SSP1-2.6), and the Wet Andes, with a cumulative loss of 32% (-177.2 m 3 s -1 in SSP1-2.6, cumulative loss of all glaciers in the region). However, a smaller number of catchments (n = 22) show an increase in glacier runoff, mainly in the Dry Andes, experiencing a 38% increase (+ 3.5 m 3 s -1 in SSP5-8.5). Some catchments show both increases or reductions in glacier runoff depending on the scenario. In the Dry Andes (Figure S4), these catchments show a 6% increase (+ 1.1 m 3 s -1 ) in the SSP5-8.5 scenario and a 7% reduction (-1.4 m 3 s -1 ) in the SSP1-2.6 scenario. Notably, the Olivares catchment (id = 6090889690) and Cipreses catchment (id = 6090897370) in Chile are among these catchments exhibiting contrasting behavior. Regarding the spatial distribution of catchments, increases in glacier runoff are predominant in Argentine catchments, such as the Tupungato catchment (id = 6090891240). Conversely, the negative changes, which constitute the majority of the volume changes, are distributed evenly between Chile and Argentina. Noteworthy catchments in this category include the Azufre catchment (id = 6090904960) and Atuel catchment (id = 6090900470). Table 1 displays simulations of extreme glacier runoff changes in catchments using the SSP5-8.5 scenario, categorized by glaciological regions. These results provide insights into significant variations in glacier runoff. Across all regions, the Acodado catchment in the Wet Andes experiences the most substantial reduction in glacier runoff, with a decrease of -8.4 m 3 s -1 . Meanwhile, the Atuel catchment exhibits the largest reduction in percentage terms in the Dry Andes, with a reduction of -62% (-2.4 m 3 s -1 ). These changes represent a considerable decline regarding their historical annual glacier runoff between 2000–2019. Conversely, the Huemules, Tupungato, and Olivares catchments show the most substantial increases in glacier runoff. It is noteworthy that the eastern side of the Andes (Argentina) shows the most pronounced reductions and increases in glacier runoff, indicating potential impacts on water resources and hydrological systems in the Dry Andes. Comparatively, the Tropical Andes experience smaller extreme changes in glacier runoff compared to other two glaciological regions, which can be attributed to relatively lower mean glacier runoff during the historical period. Notably, catchments such as Chawpi Urqu and Quelccaya display mean glacier runoff below 2.3 ± 1.3 m 3 s -1 for the 2000–2019 period. Table 1 Maximum reductions and increases in glacier runoff identified using scenario 8.5 between the periods 2000–2019 and 2030 − 2019 across the Andes Catchment name Mean annual Glacier runoff in the reference 2000–2019 [m 3 s -1 ] Change of glacier runoff between 2030–2049 and 2000–2019 [m 3 s -1 ] Change of glacier runoff between 2030–2049 and 2000–2019 [%] Glaciological region Country Catchment id Catchment area [km 2 ] Glacierized area (%) Latitude Negative changes Acodado 18.8 ± 2.1 -8.4 -45 Wet Andes Chile 6090024320 639 (26) 47°S Atuel 4.0 ± 0.6 -2.4 -62 Dry Andes Argentina 6090900470 348 (14) 35°S Chawpi Urqu 2.3 ± 1.3 -1.2 -45 Tropical Andes Perú-Bolivia 6090601720 198 (21) 15°S Positive changes Huemules 7.2 ± 2 1.3 18 Wet Andes Chile 6090965210 207 (43) 48°S Tupungato 2.7 ± 1.0 0.9 32 Dry Andes Argentina 6090891240 289 (28) 34°S Olivares 2.7 ± 1.1 0.5 16 Dry Andes Chile 6090889690 534 (14) 33°S Quelccaya 0.8 ± 0.4 0.3 45 Tropical Andes Perú 6090582670 182 (9) 14°S 2.2.2. Peak water throughout 21st century along the Andes The expected peak water (PW) estimation for 778 Andean catchments considering the filtered ensemble of GCMs indicates that the maximum contribution of glacier runoff to river discharge will likely occur before the first half of the 21st century. Figure 3 illustrates the distribution of PW years (percentiles 25 and 75 in both scenarios), with a concentration between 2010 and 2028 across the Andes. The closest PW years to the present occur first in the Wet Andes (2010–2024, n = 465 catchments), followed by the Tropical Andes (2014–2030, n = 183 catchments), and finally in the Dry Andes (2021–2046, n = 130 catchments). For the future period (2026–2099), PW years occur most frequently between 2026 and 2049. Tropical Andes will likely experience PW years in most catchments sooner (2026–2040, 50 to 86 catchments) than the Wet Andes (2030–2038, 40 to 30 catchments), and finally the Dry Andes (2030–2048, 82 to 110 catchments). Interestingly, in the Dry Andes, most catchments will likely show PW years later in the future (2026–2099). More details regarding the distribution of PW years by glaciological regions can be found in Figure S5. Specific locations in Fig. 3 and catchments in Table 2 provide a detailed view of the changes in PW year and the associated amounts of glacier runoff. These details allow for an examination of glacier runoff at the catchment scale within the SSP5-8.5 scenario. In the Tropical Andes, catchments in Colombian Sierra Nevada de Santa Marta ( e.g ., Pico Colon) display PW years spanning from 2020 to 2028, with a maximum glacier runoff of 0.3 m 3 s -1 . In Ecuador, PW years are projected to occur later in the second half of the 21st century, specifically from 2022 to 2084. Notably, the Altar catchment in Ecuador stands out, with a PW year estimated at 2052 ± 10 and a glacier runoff of 1.0 m 3 s -1 . In Perú, cordilleras with significant glacierized areas would experience their PW year before the first half of the 21st century. The Cordillera Blanca shows PW years ranging from 2018 to 2064, while the Cordillera Vilcanota exhibits PW years from 2024 to 2050. The Marcapata catchment, located in the Cordillera Vilcanota, holds the largest estimated glacier runoff volume in the Tropical Andes at 3.0 m 3 s -1 , with a PW year projected in 2030 ± 10. The Dry Andes shows a range of PW years from 2010 to 2062. The largest glacier runoff is simulated in the Cipreses catchment (9.3 m 3 s -1 ), followed by the Volcán catchment (4.0 m 3 s -1 ), and the Olivares catchment (3.6 m 3 s -1 ). For PW years simulated after 2049, larger glacier runoff volumes are observed in the Tupungato catchment (4.0 m 3 s -1 ) and the Yeso catchment (1.2 m 3 s -1 ). In contrast to the Tropical and Dry Andes, the Wet Andes region exhibits significant amounts of glacier runoff. Catchments in the Wet Andes, particularly in the latitudinal range of 46–48°S, considering the land-terminating glaciers of Northern Patagonian Icefield and surrounding glaciers, show a PW year range before the first half of the 21st century, from 2010 to 2048. The Acodado catchment stands out with the highest maximum glacier runoff of 17.8 3 s -1 and a PW year of 2010 ± 10, situated on the west side of the Andes. On the eastern side of the Northern Patagonian Icefield, catchments related to the Baker basin are estimated to experience PW years before 2030. Among these catchments, the Soler catchment (in the NPI) exhibits a larger PW volume of 8.5 3 s -1 compared to those found east of the NPI, like the Maitén catchment, with a runoff of 0.9 m 3 s -1 . From this analysis, we conclude that the calculation of the PW for each catchment is crucial in characterizing the local differences and spatial variability observed in the Andes. These diverse behaviors are attributed to the distinct morphometric characteristics of land-terminating glaciers and local climates in each catchment. The annual and monthly temporal series of glacier runoff for each glacier and catchment are available in the Supplementary data. For more in-depth information and specific examples, refer to Figures S6, S7, and S8 in the Supplementary information. These figures present the detailed temporal variations in glacier runoff for the respective glaciers and catchments. Table 2 Catchments highlighted by glacier runoff during the identified peak water year between 2000–2099 across the Andes Catchment name PW year PW glacier runoff [m 3 s -1 ] Location Country Catchment id Catchment area [km 2 ] Glacierized area (%) Latitude Pico Cristóbal Colón 2024 ± 10 0.3 ± 0.02 Colombia Colombia 6090000970 696 (1) 11°N Altar 2052 ± 10 1.0 ± 0.04 Ecuador Ecuador 6090249220 668 (3) 2°S Vicos 2020 ± 10 1.5 ± 0.05 Co. Blanca Perú 6090461650 279 (19) 9°S Ichiccocha 2030 ± 10 1.2 ± 0.06 Co. Blanca Perú 6090449220 246 (13) 9°S Marcapata 2030 ± 10 3.0 ± 0.09 Co.Vilcanota Perú 6090571030 730 (14) 14°S Olivares 2040 ± 10 3.6 ± 0.18 CL-AR Chile 6090889690 534 (14) 33°S Tupungato 2058 ± 10 4.0 ± 0.09 CL-AR Argentina 6090891240 289 (28) 34°S Yeso 2054 ± 10 1.2 ± 0.02 CL-AR Chile 6090892710 627 (6) 34°S Volcán 2040 ± 10 4.0 ± 0.06 CL-AR Chile 6090892940 524 (14) 34°S Cipreses 2010 ± 10 9.3 ± 0.23 CL-AR Chile 6090897370 350 (33) 35°S Soler 2026 ± 10 8.5 ± 0.24 NPI Chile 6090963530 749 (16) 47°S Maitén 2010 ± 10 0.9 ± 0.05 East of NPI Chile 6090962900 386 (7) 47°S Acodado 2010 ± 10 17.8 ± 0.39 NPI Chile 6090024320 639 (26) 47°S 3. Discussion 3.1. Climate projections analysis Presently, a comprehensive evaluation of GCMs outputs in glacierized catchments across the Andes is notably lacking, encompassing both CMIP6 and earlier datasets. Consequently, the prevailing reference source has been the IPCC AR6 report (IPCC, 2022), which examined climate variations in the Andes utilizing over 30 CMIP6 GCMs. Nevertheless, a key distinction between our simulations, focused on glaciers, and the IPCC report is the treatment of regional extension, wherein the latter considers a broader land area beyond the glacierized surfaces and situated at lower elevations. IPCC (2022) indicates an overall temperature increase for all scenarios, with the northern half of the continent experiencing more substantial warming, gradually decreasing southward. Additionally, precipitation is anticipated to decrease over the Southern Andes while increasing over the Northern Andes during the period 2041–2060, taking 1995–2014 as a reference period (under the SSP5-8.5 scenario). This pattern is consistent with the findings of Almazroui et al. ( 2021 ) who evaluated an ensemble of CMIP6 models across South America. Moreover, the GCMs ensemble tends to underestimate precipitation during the rainiest months while overestimating it during the drier months in the Southern Andes, particularly in the Southwestern region (Almazroui et al., 2021 ). In agreement with the IPCC under the SSP5-8.5 scenario, our results confirm a maximum temperature increase in the Tropical Andes (1.9°C), followed by the Dry (1.5°C), and Wet Andes (1.1°C, southward of NPI) during the period 2030–2049. For precipitation, the IPCC reports a reduction in the Dry Andes (-8%) and Wet Andes (-0.1%), alongside an increase in the Tropical Andes (4.5%). However, our findings diverge, showing a negative trend in the Tropical and Wet Andes. Notably, precipitation only exhibits an increase in the SSP1-2.6 scenario for the Tropical Andes. Regarding the Tropical Andes and the northern area of the Dry Andes, Olmo et al. ( 2022 ) conducted a study focusing on the representation of precipitation variability in the July-October season (1979–2014). By using the CESM2 and MPI-ESM1-2-HR models from CMIP6, they indicated that these models effectively capture the underlying physical mechanisms governing precipitation patterns. Our analysis revealed that these same models exhibit a high monthly correlation but a poor seasonal correlation for the OND season. However, in the Dry Andes region (18–37°S), these models exhibit a robust seasonal correlation, which holds significant importance for glacier mass balance simulations. Regarding future projections, we observed a median reduction in precipitation during the JAS and OND seasons (as projected by MPI, CESM2, and NorESM2-MM) under the SSP1-2.6 and SSP5-8.5 scenarios. These results are in line with estimations made by Agudelo et al. ( 2023 ) which indicate an increase in the occurrence of dry days (19.4%) during the austral winter (JAS) and a higher frequency of dry circulation patterns during the July-October period. Notably, the CESM2 model demonstrated the most favorable outcomes concerning precipitation variability over the southwestern region during the historical period (1901–2014) as indicated by Rivera and Arnould ( 2020 ). Given the foregoing, the use of GCMs as input data in future glacier runoff simulations should consider the following aspects. First, the importance of the seasonal temporal evaluation. Indeed, it is crucial to assess GCM performance during critical periods of glacier melt because neglecting seasonal evaluations in favor of monthly or annual assessments may overlook deficiencies in seasonal performance. Second, the GCM performance over glacierized areas must be evaluated. Indeed, using GCMs evaluated for other spatial extensions could present trends and seasonal variations that are not representative of glacierized regions but of lower elevations. 3.2. Glacier volume and runoff simulations toward 2049 Three critical aspects in the glacier dynamics considered in the simulations using OGGM need to be discussed to account for the likelihood of the presented results: The approach based on the Shallow Ice Approximation (SIA), implemented in OGGM, lacks longitudinal/transverse stress gradients and other complex mechanisms of glacier dynamics (Le Meur et al., 2004 ). As a consequence, glaciers' response to climate forcing is nearly immediate, making higher-order ice flow models (Oerlemans, 2008 ) more suitable for accurately representing ice flow. However, the use of these high-order models presents computational challenges (Jouvet, 2022), currently limiting their application in regional and global simulations. The ice thickness calibration in this study relies on data from Farinotti et al. ( 2019 ), which employs an ensemble of up to five models to estimate the ice thickness distribution. Unfortunately, this approach leads to an overestimation of approximately 20% (median) of the measured ice thickness. Moreover, our calibration parameter values for ice thickness align with values from Cuffey and Paterson ( 2010 ) and Millan et al. ( 2022 ). The simulations predict lower Glen A parameter values for glaciers with lower internal temperatures, such as those found in the glaciological zone DA1 (2.4 10 –25 s − 1 Pa − 3 ), which represents the coldest zone in the Andes. Additionally, Millan et al. ( 2022 ) estimated Glen A parameters in the Andes ranging between 5.6 10 –25 and 2.4 10 –24 s − 1 Pa − 3 , while the calibrated values used in this study range from 2.4 10 –25 and 2.4 10 –23 s − 1 Pa − 3 . Due to the limitations mentioned in (i) and (ii), our simulations of future glacier runoff changes carry uncertainties arising from the ice thickness simulation and its calibration. Moreover, the climate performance in the historical period and the various climate models and scenarios considered contribute to the overall uncertainty. Estimating melt factor values is critical, and we derive them from the calibration, using Hugonnet et al. ( 2021 ) data of geodetic mass balance. However, we must acknowledge that our historical climate dataset, which was corrected in Caro et al. ( 2024 ) and combined with Hugonnet et al. ( 2021 ) data could introduce errors in the calibrated melt factor. Consequently, this error could affect the simulated mass balance of glaciers and the related ice thickness estimation. Furthermore, the future climate data from each GCM (temperature and precipitation) was corrected relative to the historical climate dataset. The three main sources of uncertainty in future simulations of glacier runoff are: (i) the SIA, (ii) the calibration of glacier volume, and (iii) the use of GCMs and parameter values for mass balance estimation. Despite simulation biases, simulations at the Andes scale capture variations in future glacial runoff between catchments and Andean regions which sound coherent with current spatio-temporal differences. This capability arises from a consistent modeling approach, calibrated and validated across the Andes using historical data. Also, It incorporates the highest-scored climate projections specific to each region for future simulations. 3.3. PW estimation across the Andes: comparison with former studies Our 21st century PW estimates partially align with Huss and Hock ( 2018 ) findings. They identified PW in 12 Andean basins with a glacierized surface area of 9,544 km 2 , representing different proportions in the Tropical Andes (23%), Dry Andes (20%) and Wet Andes (57%). Their research revealed a PW already past in the inner tropics (2 catchments) and projected a PW occurrence between 2011–2046 in the outer tropics (4 catchments). In the Dry Andes, the estimated PW was around 2010 ± 24 (2 catchments in western side), while the Wet Andes showed a broader range from 2003 ± 11 to 2096 ± 24 (4 catchments). Interestingly, they found evidence of past PW in the northern area of the Wet Andes. In contrast, our estimations considered a larger glacierized surface area (11,282 km 2 ), filtering glaciers not accounted for in glacier national inventories and calving glaciers in the Patagonian icefields. Our results suggest that the PW will occur before the first half of the 21st century in the majority of the catchments across the Andes. According to our results, the PW has already occurred in most of the Tropical Andes catchments and will occur before 2049 in most of the Dry Andes catchments. In the Wet Andes, the majority of catchments experienced PW before the present day. Furthermore, our estimation of PW presents different ranges of years across the Andes, tied too to varying climate change scenarios. Generally, an earlier PW occurrence is associated with less warm scenarios, which aligns with observations in the Himalayan mountains and specific Andean catchments (Huss and Hock, 2018 ; Laha et al., 2021 ). 4. Conclusion We analyzed eight GCMs from CMIP6 to identify those best at reproducing historical climate (1990–2019) and projected future conditions (2020–2049) across the Andes. These models were then used to estimate changes in glacier runoff throughout the 21st century (10°N-55°S) using the calibrated and validated Open Global Glacier Model (OGGM). For the first time, we conducted a comprehensive simulation of future glacier dynamics and runoff in 778 Andean catchments. We conclude the following: The ensemble of all GCMs (complete ensemble) showed larger changes in glacier runoff between periods 2000–2019 and 2030–2049), compared to the ensemble using the highest-scoring GCMs (filtered ensemble) at the catchment scale. However, these differences were not evident at the regional scale. Glacier runoff in the Tropical and Wet Andes (filtered ensemble) showed notable reductions between 2000–2019 and 2030–2049, with median decreases of -0.1 m 3 s -1 and − 0.2 m 3 s -1 , respectively. In contrast, the Dry Andes exhibited a relatively smaller reduction (median of -0.003 m 3 s -1 ), with some catchments, like Tupungato and Olivares, experiencing at least a 16% increase. Cumulative glacier runoff in the Dry Andes increased by 38%, while the Tropical Andes experienced the most significant cumulative loss, with a reduction of 43% compared to the historical period. Projections indicate peak water from glacier runoff will occur in most Andean catchments before the mid-21st century (PW year between 2010–2028). Nevertheless, the distribution of PW years exhibits significant variations in Andean regions and catchments. The Wet Andes is expected to experience the earliest PW years (2010 to 2024), followed by the Tropical Andes (2014 to 2030), and the Dry Andes later (2021–2046). Within this region, specific attention should be given to catchments that show the highest amounts of glacier runoff, such as the Cipreses (PW in 2010 ± 10 with 9.3 m 3 s -1 ), Volcán (PW in 2040 ± 10 with 4.0 m 3 s -1 ), and the Olivares catchment (PW in 2040 ± 10 with 3.6 m 3 s -1 ). This work provides a comprehensive understanding of the variations in glacier runoff using high-scored GCMs, underlying the importance of accounting for regional and catchment-scale differences in water resource adaptation strategies. This is particularly crucial in regions most impacted by prolonged droughts in the Andes, such as the central zones of Chile and Argentina. 5. Data and Methods 5.1. Analysis of the GCMs in the historical and future periods In this study, we analyze temperature and precipitation data from two scenarios and eight GCMs simulations sourced from CMIP6 (see Table S4). Our analysis takes into account the Almazroui et al. ( 2021 ) assessment for GCMs in South America and the global evaluation of warmer models from Hausfather et al. ( 2022 ) and Tokarska et al. ( 2020 ). To establish a basis for comparison, we scale the GCMs simulations using the bias-correction method implemented in the OGGM model and compare them with the TerraClimate data (Abatzoglou et al., 2018 ), that have been corrected for temperature and precipitation (cTC, temperature, and precipitation) at the scale of the Andes on the basis of in situ data from 34 meteorological stations (Caro et al., 2024 ) during the historical period of 1990–2019 across the Andes on a glacierized area of 27,668 km 2 (considering glaciers with a surface area > 1 km 2 ). Meanwhile, we also perform comparisons between GCMs during both the historical and future periods (2020–2049). Our analysis comprises three steps: step 1: statistical downscaling of GCMs temperature and precipitation data for the historical period; step 2: calculation of annual, seasonal and monthly metrics to compare the GCMs data with the corrected TerraClimate data for the historical period; and step 3: Identification of changes in climate variables between the historical and future periods. Such simulations were deemed impossible to conduct in this context. Going into detail, these three steps consist in: Step 1. The mean monthly and annual values of temperature and precipitation from the GCMs were adjusted to the mean elevation of each glacier. This adjustment was performed using a statistical downscaling approach for two future scenarios (SSP1-2.6 and SSP5-8.5), based on the historical climate data from the cTC (1990–2019). Step 2. Correlation pattern analyses were based on monthly and seasonal (OND, JFM, AMJ, JAS) correlations between the eight GCMs and the cTC data in the period 1990–2019. We chose seasons as OND and JFM, because in these months the larger surface mass loss occurs in the Outer Tropics (transition season, Autin et al., 2022 ) and in the Dry Andes zones (austral summer). We used the Pearson coefficient of correlation and the root mean square error (RMSE) (McSweeney et al., 2015 ) as metrics to score the GCMs performance during the historical period. Step 3. We identified climate likely ranges through the GCMs ensemble considering the mean differences between the historical and future periods for temperature and precipitation at annual and seasonal (OND, JFM, AMJ, JAS) time-steps in each glacier. The climate change likely ranges are estimated by the percentiles 10 and 90 of these differences using all GCMs and glaciers (modified from DRIAS project, 2023). In addition, we estimate the median of these differences. A similar method is used by the DRIAS project of the Ministère de la Transition Écologique of the French government (see Table S2, DRIAS project, 2023, and Sørland et al., 2018 ). The annual analysis between the historical and future periods (2020–2049 and 2070–2099) is considered to test our results with previous reports of climate change (e.g., IPCC, 2022; Olmo et al., 2022 ; Agudelo et al., 2023 ). The estimation of the climate change likely ranges allows us to identify four types of models: the hot/dry; the cold/dry; the hot/wet and the cold/wet models. After these steps, we defined a complete ensemble considering eight GCMs, and a second GCM ensemble that comprise highly score GCMs, called filtered ensemble. The GCMs output data were gathered for each glacier. However, the analysis was performed at the glaciological region scale, where climate characteristics vary substantially (Caro et al., 2021 ). We did not include glaciers experiencing mass loss due to calving in our simulations conducted in the Patagonian region (Minowa et al., 2021 ). 5.2. Short description of the Open Global Glacier Model (OGGM) OGGM is a modular and open-source workflow that simulates glacier mass balance and ice dynamics using calibrated parameter values for each glacier (Maussion et al., 2019 ). The required input data are: air temperature and precipitation time series, glacier outlines and surface topography. From these inputs, annual ( e.g. , surface mass balance, glacier volume and area) and monthly (glacier melt [snow + ice] and rainfall on glaciers) outputs can be simulated. Using a glacier outline and topography, OGGM estimates flow lines using a geometrical algorithm (adapted from Kienholz et al., 2014 ). Assuming a bed shape, it estimates the ice thickness based on mass conservation and shallow ice approximation (Maussion et al., 2019 ). After these numerical steps, area and volume per glacier can be simulated. Mass balance is implemented using a precipitation phase partitioning and a temperature-index approach (Braun and Renner, 1992 ; Hock, 2003 ; Marzeion et al., 2012 ). The monthly mass balance \(\:{mb}_{i}\) at an elevation z is computed as follows: \(\:\:{mb}_{i}\left(z\right)={TC}_{p\:i}^{snow}\left(z\right){*P}_{f}-\:{M}_{f}*max\left({cTC}_{t\:i}\left(z\right)-\:{T}_{melt}\:,0\right)\) equation (1), where \(\:{TC}_{p\:i}^{snow}\) is the TerraClimate solid precipitation before being scaled by the precipitation correction factor ( \(\:{P}_{f}\) ), \(\:{M}_{f}\) is the glacier's temperature sensitivity parameter, \(\:{cTC}_{t\:i}\) is the monthly corrected TerraClimate temperature, and \(\:{T}_{melt}\) is the monthly air temperature above which ice melt is assumed to occur (from 0°C to 2.1°C). \(\:{TC}_{p\:i}^{snow}\) is computed as a fraction of the total precipitation ( \(\:{cTC}_{p}^{}\) ) where 100% is getting if \(\:{cTC}_{t\:i}\) = \(\:{T}_{i}^{rain}\) (between 2-4.1°C); and linearly interpolated in between. Here, the \(\:{M}_{f}\) was calibrated for each glacier individually using glacier volume change datasets previously described (Hugonnet et al., 2021 ). 5.2.1. Model setup in the period 2000–2050 The OGGM model was previously calibrated and evaluated during the period 2000–2019 across the Andes (10°N-55°S) in a former study by Caro et al. ( 2024 ). In this former study, we evaluated and corrected temperature and precipitation input data, evaluated the simulated glacier mass balances outputs using in situ mass balance measurements, and also the model performance at the level of three Andean catchments. The model was run for each glacier and then results were analyzed for each glaciological zone identified by Caro et al. ( 2021 ) across the Andes. During the historical (2000–2019) and future (2020–2050) periods the input data are: glacier outlines from RGI v6.0 (RGI Consortium, 2017 ) and surface topography from NASADEM (Crippen et al., 2016 ). The corrected monthly TerraClimate precipitation ( \(\:c{TC}_{p}^{}\) ) and temperature ( \(\:{cTC}_{t}^{}\) ) were used in the historical period. For the future period, monthly precipitation and temperature were selected from eight GCMs. These variables were used as input for the future simulations. The calibration procedure was applied for each individual glacier to adjust the simulated mass balance of the 2000–2019 period to the geodetic mass balance product from Hugonnet et al. ( 2021 ). The simulated glacier volume was calibrated using Farinotti et al. ( 2019 ) product at a glaciological zone scale fitting the A parameter of the Glen flow law. In addition, glacier outlines of all glaciers were associated with the year 2000. The main corrected and calibrated parameters to run OGGM across the Andes are summarized in Table S5. 5.3. Glacier runoff analysis Glacier runoff changes were estimated by adding the annual melt from each glacier by catchment, for this we considered the centroid of each glacier contour. From this annual glacier runoff time series, the difference of the mean annual glacier runoff between the periods 2000–2019 and 2030–2049 was estimated for 786 glacierized catchments. Meanwhile, PW refers to the annual glacier runoff that will initially increase and after declining in response to glacier retreat due to changes in climate conditions (Huss and Hock, 2018 ) which can be for example a long-term response of catchments to sustained warming (Hock et al., 2005). We calculate the PW inspired by Huss and Hock ( 2018 ) through the following procedure by glacier and after by catchment: i) A maximum of eight time series of the simulated glacier runoff from each glacier are compiled. Each time series comes from the different OGGM runs using different GCMs (the number of GCMs varies between the filtered and complete ensembles in each glaciological region); ii) Then, the time series were averaged, getting one time series of glacier runoff per glacier; iii) these were summed annually by catchment, allowing us to obtain the annual glacier runoff in each catchment between 2000–2099; iv) The glacier runoff was smoothed using a moving average comprising 11 years; v) On these smoothed time-series we selected the period of 20 years related to maximum glacier runoff; vi) Finally, the PW year corresponds to the median of these 20 years with a fixed uncertainty range of +/- 10 years. Declarations Competing interests The author(s) declare no competing interests. Author Contribution AC, TC, and AR were involved in the study design. AC wrote the model implementation and produced the figures, tables, and first draft of the manuscript. RA and NC contributed to the model implementation. AC performed the first level of analysis, which was improved by input from all authors. All authors contributed to the review and editing of the paper. Acknowledgement We acknowledge LabEx OSUG@2020 (Investissement d'Avenir, ANR10 LABX56). AC thanks the National Agency for Research and Development (ANID)/Scholarship Program/DOCTORADO BECAS CHILE/2019-72200174. RA was supported by the European Research Council (ERC) under the European Union's Horizon Framework research and innovation programme (grant agreement Nº101115565; ICE3 project). Data Availability All data analyzed in this article are available by glacier and catchment at https://zenodo.org/records/12714725. Other data such as the catchment outlines and their identifier (id) are available at https://doi.org/10.5281/zenodo.7890462, related to the article “Hydrological response of Andean catchments to recent glacier mass loss” https://doi.org/10.5194/tc-18-2487-2024. References Abatzoglou, J. T., Dobrowski, S. Z., Parks, S. A., & Hegewisch, K. C. TerraClimate, a High-Resolution Global Dataset of Monthly Climate and Climatic Water Balance from 1958-2015. Sci. Data. 5, 1-12.; 10.1038/sdata.2017.191 (2018). Agudelo, A. et al . Future projections of low-level atmospheric circulation patterns over South Tropical South America: Impacts on precipitation and Amazon dry season length. Journal of Geophysical Research: Atmospheres.128, e2023JD038658; 10.1029/2023JD038658 (2023). Almazroui, M. et al . Assessment of CMIP6 Performance and Projected Temperature and Precipitation Changes Over South America. Earth Syst Environ. 5, 155–183; 10.1007/s41748-021-00233-6 (2021). Aschwanden, A., Aðalgeirsdóttir, G., & Khroulev, C. Hindcasting to measure ice sheet model sensitivity to initial states. The Cryosphere. 7; 10.5194/tc-7-1083-2013 (2013). Autin, P., Sicart, J. E., Rabatel, A., Soruco, A., & Hock, R. Climate Controls on the Interseasonal and Interannual Variability of the Surface Mass and Energy Balances of a Tropical Glacier (Zongo Glacier, Bolivia, 16° S): New Insights From the Multi‐Year Application of a Distributed Energy Balance Model. Journal of Geophysical Research: Atmospheres. 127(7); 10.1029/2021JD035410 (2022). Braun, L. N. & Renner, C. B. Application of a conceptual runoff model in different physiographic regions of Switzerland. Hydrol. Sci. J. 37(3), 217-231; 10.1080/02626669209492583 (1992). Bravo, C. et al . Projected increases in surface melt and ice loss for the Northern and Southern Patagonian Icefields. Sci Rep. 11, 16847; 10.1038/s41598-021-95725-w (2021). Caro, A, Condom, T. & Rabatel, A. Climatic and Morphometric Explanatory Variables of Glacier Changes in the Andes (8–55°S): New Insights From Machine Learning Approaches. Front. Earth Sci.; 10.3389/feart.2021.713011 (2021). Caro, A. et al . Hydrological Response of Andean Catchments to Recent Glacier Mass Loss. The Cryosphere; 10.5194/tc-18-2487-2024 (2024). Crippen, R. et al . NASADEM Global Elevation Model: Methods and Progress. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences XLI-B4. 10.5194/isprs-archives-XLI-B4-125-2016 (2016). Cuffey, K. & Paterson, W. The Physics of Glaciers (Academic Press) 4th edn. (2010). DRIAS. DRIAS les futurs du climat (http://drias-climat.fr (2023). Eyring, V. et al . Overview of the Coupled Model Intercomparison Project Phase 6 (CMIP6) experimental design and organization, Geosci. Model Dev. 9, 1937-1958; 10.5194/gmd-9-1937-2016 (2016). Farinotti, D. et al . A consensus estimate for the ice thickness distribution of all glaciers on Earth. Nat. Geosci. 12, 168-173; 10.1038/s41561-019-0300-3 (2019). Frans, C. et al . Predicting glacio-hydrologic change in the headwaters of the Zongo River, Cordillera Real, Bolivia. Water Resour. Res. 51, 9029-9052; 10.1002/2014WR016728 (2015). Hausfather, Z., Marvel, K., Schmidt, G. A., Nielsen-Gammon, J. W., & Zelinka, M. Climate simulations: recognize the ‘hot model’ problem. Nature. 605, 26-29; 10.1038/d41586-022-01192-2 (2022). Hock, R. Temperature index melt modelling in mountain areas. Journal of Hydrology. 282(1-4), 104-115; 10.1016/S0022-1694(03)00257-9 (2003). Hugonnet, R. et al . Accelerated global glacier mass loss in the early twenty-first century. Nature. 592, 726-731; 10.1038/s41586-021-03436-z (2021). Huss, M., & Hock, R. A new model for global glacier change and sea-level rise. Front. Earth Sci. 3:54; 10.3389/feart.2015.00054 (2015). Huss, M., & Hock, R. Global-scale hydrological response to future glacier mass loss. Nat. Clim. Change. 8, 135-140; 10.1038/s41558-017-0049-x (2018). IPCC AR6. IPCC WGI Interactive Atlas: Regional information (Advanced). https://interactive-atlas.ipcc.ch/ (2022). Jouvet, G. et al . Deep learning speeds up ice flow modelling by several orders of magnitude. Journal of Glaciology. 68(270), 651–664; 10.1017/jog.2021.120 (2022). Kienholz, C., Rich, J. L., Arendt, A. A., & Hock, R. A new method for deriving glacier centerlines applied to glaciers in Alaska and northwest Canada. The Cryosphere. 8, 503–519; 10.5194/tc-8-503-2014 (2014). Laha, S., Banerjee, A., Singh, A., Sharma, P., & Thamban, M. The control of climate sensitivity on variability and change of summer runoff from two glacierised Himalayan catchments preprint. Hydrology and Earth System Sciences Discussions; 10.5194/hess-2021-499 (2021). Le Meur, E., Gagliardini, O., Zwinger, T. & Ruokolainen, J. Glacier flow modelling: a comparison of the shallow ice approximation and the Full-Stokes solution. Comptes Rendus Physique. 5(7), 709-722; 10.1016/j.crhy.2004.10.001 (2004). Marzeion, B. et al . Partitioning the Uncertainty of Ensemble Projections of Global Glacier Mass Change. Earth’s Future. 8; 10.1029/2019EF001470 (2020). Marzeion, B., Jarosch, A. H., & Hofer, M. Past and future sea-level change from the surface mass balance of glaciers. The Cryosphere. 6, 1295-1322; 10.5194/tc-6-1295-2012 (2012). Masiokas, M. H. et al . Current state and recent changes of the cryosphere in the Andes. Front Earth Sci. 8:99; 10.3389/feart.2020.00099 (2020). Maussion, F. et al . The open global glacier model (OGGM) v1.1. Geoscientific Model. Develop. 12, 909-931; 10.5194/gmd-12-909-2019 (2019). McSweeney, C. F., Jones, R. G., Lee, R. W., & Rowell, D. Selecting CMIP5 GCMs for downscaling over multiple regions. Climate Dynamics. 44(11-12), 3237–3260; 10.1007/s00382-014-2418-8 (2015). Millan, R. et al . Ice velocity and thickness of the world’s glaciers. Nat. Geosci. 15, 124-129; 10.1038/s41561-021-00885-z (2022). Minowa, M., Schaefer, M., Sugiyama, S., Sakakibara, D., & Skvarca, P. Frontal ablation and mass loss of the Patagonian icefields. Earth Planet. Sc. Lett. 561, 116811; 10.1016/j.epsl.2021.116811 (2021). Oerlemans, J. Minimal Glacier Models (Utrecht Publishing and Archiving Services) (2008). Olmo, M. E. et al . Circulation patterns and associated rainfall over south tropical South America: GCMs evaluation during the dry-to-wet transition season. J. Geophys. Res. Atmos. 127; 10.1029/2022JD036468 (2022). Rabatel, A. et al. Current state of glaciers in the tropical Andes: a multi-century perspective on glacier evolution and climate change. The Cryosphere , 7 (1), 81-102; 10.5194/tc-7-81-2013 (2013). Rabatel, A. et al . Toward an imminent extinction of Colombian glaciers?. Geografiska Annaler: Series A. Phys. Geogr. 100, 75–95; 10.1080/04353676.2017.1383015 (2018). Radic, V., & Hock, R. Glaciers in the Earth’s hydrological cycle. Assessments of glacier mass and runoff changes on global and regional scales. Surv. Geophys. 35, 813-837; 10.1007/s10712-013-9262-y (2014). Réveillet, M., Rabatel, A., Gillet-Chaulet, F., & Soruco, A. Simulations of changes to Glacier Zongo, Bolivia (16S), over the 21st century using a 3-D full-Stokes model and CMIP5 climate projections. Ann. Glaciol. 56, 89-97; 10.3189/2015aog70a113 (2015). RGI Consortium. Randolph Glacier Inventory. A Dataset of Global Glacier Outlines, Version 6. NSIDC: National Snow and Ice Data Center. 10.7265/4m1f-gd79 (2017). Rivera, J.A. & Arnould, G. Evaluation of the ability of CMIP6 models to simulate precipitation over Southwestern South America: climatic features and long-term trends (1901–2014). Atmos Res. 241, 104953; 10.1016/j.atmosres.2020.104953 (2020). Rounce, D. et al . Global glacier change in the 21st century: Every increase in temperature matters. Science. 379, 78-83; 10.1126/science.abo1324 (2023). Schaefer, M., MacHguth, H., Falvey, M., & Casassa, G. Modeling past and future surface mass balance of the Northern Patagonia icefield. J. Geophys. Res. Earth Surf. 118, 571-588; 10.1002/jgrf.20038 (2013). Scheiter, M., Schaefer, M., Flández, E., Bozkurt, D., & Greve, R. The 21st-century fate of the Mocho-Choshuenco ice cap in southern Chile. The Cryosphere. 15, 3637-3654; 10.5194/tc-15-3637-2021 (2021). Sørland, S. L., Schär, C., Lüthi, D. & Kjellström, E. Bias patterns and climate change signals in GCM-RCM model chains. Environ. Res. Lett. 13, 074017; 10.1088/1748-9326/aacc77 (2018). Tokarska, K. B. et al . Past warming trend constrains future warming in CMIP6 models, Science Advances. 6, eaaz9549; 10.1126/sciadv.aaz9549 (2020). Vuille, M. et al . Rapid decline of snow and ice in the tropical Andes–Impacts, uncertainties and challenges ahead. Earth Sci. Rev. 176, 195-213; 10.1016/j.earscirev.2017.09.019 (2018). Yarleque, C. et al . Projections of future disappearance of the Quelccaya, the largest tropical ice cap on Earth. Nat. Sci. Rep. 8, 15564; 10.1038/s41598-018-33698-z (2018). Additional Declarations No competing interests reported. Supplementary Files SupplementaryInformationArt.21stcenturyJuly2024.docx Cite Share Download PDF Status: Published Journal Publication published 31 Mar, 2025 Read the published version in Scientific Reports → Version 1 posted Reviews received at journal 07 Aug, 2024 Reviews received at journal 31 Jul, 2024 Reviewers agreed at journal 22 Jul, 2024 Reviewers agreed at journal 22 Jul, 2024 Reviewers invited by journal 21 Jul, 2024 Editor assigned by journal 21 Jul, 2024 Editor invited by journal 16 Jul, 2024 Submission checks completed at journal 11 Jul, 2024 First submitted to journal 09 Jul, 2024 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-4714636","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":334724578,"identity":"16335014-4965-45cf-a116-0c9e4fe2a445","order_by":0,"name":"Alexis Caro","email":"data:image/png;base64,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","orcid":"","institution":"Grenoble Alpes University","correspondingAuthor":true,"prefix":"","firstName":"Alexis","middleName":"","lastName":"Caro","suffix":""},{"id":334724579,"identity":"4ca61c04-04b2-4e98-bab6-3499baab25b3","order_by":1,"name":"Thomas Condom","email":"","orcid":"","institution":"Grenoble Alpes University","correspondingAuthor":false,"prefix":"","firstName":"Thomas","middleName":"","lastName":"Condom","suffix":""},{"id":334724580,"identity":"4a9150a5-2b64-4bc0-bbfb-23fed57c5fed","order_by":2,"name":"Antoine Rabatel","email":"","orcid":"","institution":"Grenoble Alpes University","correspondingAuthor":false,"prefix":"","firstName":"Antoine","middleName":"","lastName":"Rabatel","suffix":""},{"id":334724581,"identity":"36e28ff8-e4fb-4c2f-934e-3f0328fe2e24","order_by":3,"name":"Rodrigo Aguayo","email":"","orcid":"","institution":"Vrije Universiteit Brussel","correspondingAuthor":false,"prefix":"","firstName":"Rodrigo","middleName":"","lastName":"Aguayo","suffix":""},{"id":334724583,"identity":"ee1f9708-175f-45ed-ba80-f8a959ba2381","order_by":4,"name":"Nicolas Champollion","email":"","orcid":"","institution":"Grenoble Alpes University","correspondingAuthor":false,"prefix":"","firstName":"Nicolas","middleName":"","lastName":"Champollion","suffix":""}],"badges":[],"createdAt":"2024-07-10 00:08:14","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4714636/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4714636/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41598-025-88069-2","type":"published","date":"2025-03-31T15:57:05+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":61800396,"identity":"842073f0-3baa-4f7a-895a-453ca48e77b5","added_by":"auto","created_at":"2024-08-05 17:22:04","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":271135,"visible":true,"origin":"","legend":"\u003cp\u003eThe annual climate change likely ranges for the periods 1990-2019 and 2070-2099 can be checked in Figure S3 of the Supplementary materials.\u003c/p\u003e\n\u003cp\u003eTemperature and precipitation changes between the historical (1990-2019) and future periods (2020-2049) across the glaciological Andean regions in two climate change scenarios. Percentiles 10 and 90 (boxes, likely ranges) from mean differences in both periods are estimated considering all glaciers in the complete ensemble of GCMs. Annual climate change likely ranges are exhibited in terms of absolute (A) and relative values (B) for precipitation considering SSP1-2.6 and SSP5-8.5 scenarios. (C) show these differences by seasons considering both scenarios. Estimations are performed using 3,213 glaciers (27,668 km\u003csup\u003e2\u003c/sup\u003e) (TA = 598, DA = 370, WA = 2245). Models situated inside the boxes formed the filtered ensemble.\u003c/p\u003e","description":"","filename":"Picture1.png","url":"https://assets-eu.researchsquare.com/files/rs-4714636/v1/d7c920273906e85e728cd21c.png"},{"id":61800397,"identity":"976b0207-3330-45b2-a89a-7a3a18fdcaaf","added_by":"auto","created_at":"2024-08-05 17:22:04","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":269697,"visible":true,"origin":"","legend":"\u003cp\u003eMean annual changes in glacier runoff between the periods 2000-2019 and 2030-2049 at the catchment scale using the filtered ensemble of GCMs. For the period 2000-2019 the cTC data are used in the simulations, whereas for the period 2030-2049 an ensemble of evaluated GCMs data is used. A) and D) show the location of all catchments and the other ones considered in the zoom in, respectively. Glacier runoff differences are presented across the Andes (786 catchments) in absolute values of changes (m\u003csup\u003e3\u003c/sup\u003e s\u003csup\u003e-1\u003c/sup\u003e) for the SSP1-2.6 (B) and SSP5-8.5 scenarios (C). E) and F) represent a zoom in for glacier runoff changes between -1 and 1 m\u003csup\u003e3\u003c/sup\u003e s\u003csup\u003e-1\u003c/sup\u003e. The vertical light green and red stripes in B and C display the standard deviation of glacier runoff changes by the Tropical, Dry and Wet Andes.\u003c/p\u003e","description":"","filename":"Picture2.png","url":"https://assets-eu.researchsquare.com/files/rs-4714636/v1/8e3676f0bb7781ee143da7fa.png"},{"id":61800398,"identity":"315188dc-a6f7-4ecb-b784-9bf9c3f3eecb","added_by":"auto","created_at":"2024-08-05 17:22:05","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":241603,"visible":true,"origin":"","legend":"\u003cp\u003ePeak water year and related mean annual glacier runoff across the Andes for the SSP5-8.5 scenario throughout the 21st century. In the South American map, the peak water per catchment is exhibited for the “current” period (2000-2025 in red) and for the future (2026-2099, blue), using simulations forced by the filtered ensemble of GCMs. Detailed maps can be seen in six locations from the Sierra Nevada de Santa Marta in Colombia to the Northern Patagonian Icefield in Chile. The CL-AR location shows central Chile and Argentina.\u003c/p\u003e","description":"","filename":"Picture3.png","url":"https://assets-eu.researchsquare.com/files/rs-4714636/v1/6016f62051c270951b77aa73.png"},{"id":80082048,"identity":"507fcea3-6a66-42ad-b98c-81feeed4835d","added_by":"auto","created_at":"2025-04-07 16:06:14","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1835981,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4714636/v1/5a8bea96-0764-4e0e-bb72-a1f464bff50b.pdf"},{"id":61800399,"identity":"8ab88f44-94b0-4788-ae40-ee79ed606bb6","added_by":"auto","created_at":"2024-08-05 17:22:05","extension":"docx","order_by":5,"title":"","display":"","copyAsset":false,"role":"supplement","size":5452481,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryInformationArt.21stcenturyJuly2024.docx","url":"https://assets-eu.researchsquare.com/files/rs-4714636/v1/88b595d7a16c3f11f8fe1356.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Glacio-hydrological changes along the Andes throughout the 21st Century","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eIn the Andes, most of the knowledge regarding future glacier changes is derived from global simulations (\u003cem\u003ee.g.\u003c/em\u003e, Marzeion et al., \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2012\u003c/span\u003e; Radic and Hock, \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Huss and Hock, \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Rounce et al., \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). The recent study by Rounce et al. (\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) stands out as it incorporates geodetic mass balance measurements obtained at the scale of individual glaciers by Hugonnet et al. (\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) to calibrate the glacier mass balance during the historical period. It is important to note that these global-scale studies have not been specifically evaluated for the Andes, even though simulating current observations is a crucial prerequisite for predictive models (Aschwanden et al., \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). The reported glacier volume changes projected throughout the 21st century by the above mentioned studies demonstrate consistent results in the Tropical Andes, with an approximate loss of glacier mass of around \u0026minus;\u0026thinsp;98\u0026thinsp;\u0026plusmn;\u0026thinsp;13% by 2100 under the RCP8.5 scenario based on the Coupled Model Intercomparison Project phase 5 (CMIP5) models. However, in the southern Andes, which encompasses the largest glacierized area, there is a wider range of mass loss estimates, ranging from \u0026minus;\u0026thinsp;44\u0026thinsp;\u0026plusmn;\u0026thinsp;14% to -68\u0026thinsp;\u0026plusmn;\u0026thinsp;20% (Huss and Hock, \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Rounce et al., \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Even under the most optimistic scenarios (\u003cem\u003ei.e.\u003c/em\u003e RCP2.6), the reduction in glacier volume remains significant. Furthermore, a global study by Huss and Hock (\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2018\u003c/span\u003e) focused on 12 Andean catchments and estimated that glacier runoff, which includes ice and snow melt as well as rainfall on glaciers, is projected to increase in most of the catchments until 2050. However, after 2050, it is expected to decrease in all catchments except for the Santa Cruz catchment (49\u0026deg;S, Argentina).\u003c/p\u003e \u003cp\u003eLocal simulations of future glacier changes across the Andes have been conducted, encompassing glaciers in Colombia, Ecuador, Peru, Bolivia, and Chile (\u003cem\u003ee.g.\u003c/em\u003e, Frans et al., \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; R\u0026eacute;veillet et al., \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Yarleque et al., \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Vuille et al., \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Rabatel et al., \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Scheiter et al., \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) as well as the Patagonian icefield (\u003cem\u003ee.g.\u003c/em\u003e, Schaefer et al., \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2013\u003c/span\u003e; Bravo et al., \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). These studies focus on different objectives (\u003cem\u003ee.g.\u003c/em\u003e, surface mass balance, glacier dynamic, glacier runoff) and use different CMIP5 models. In the tropical Andes, Vuille et al. (\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2018\u003c/span\u003e) estimated that the Antizana Glacier (0\u0026deg;S, inner tropics) is more vulnerable to warming throughout the 21st century in comparison with the Zongo Glacier (16\u0026deg;S, outer tropic). For Zongo Glacier, projected volume losses range from \u0026minus;\u0026thinsp;40\u0026thinsp;\u0026plusmn;\u0026thinsp;7% to -89\u0026thinsp;\u0026plusmn;\u0026thinsp;4% between 2010 and 2100 depending on the considered RCP (R\u0026eacute;veillet et al., \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2015\u003c/span\u003e) and a discharge reduction in 2100 was estimated by 25% at the annual scale and by 57% during the dry season for RCP4.5 (Frans et al., \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2015\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eIn the southern Andes, no study was performed in the Dry Andes, but for the Wet Andes, Scheiter et al. (\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) projected an ice volume loss between \u0026minus;\u0026thinsp;56 and \u0026minus;\u0026thinsp;97% depending on the RCP for the Mocho Choshuenco glacier in 2100. Two other studies reported future glacier changes in the Patagonian icefields. In the Northern Patagonian Icefield (NPI) a strong increase in ablation is estimated from 2050 onward with a reduction of solid precipitation from 2080 onward due to higher temperatures, with uncertainties arising from future climate and ice dynamics (Schaefer et al., \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). Bravo et al. (\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) compared simulations for the period 2005\u0026ndash;2050 with the historical period 1976\u0026ndash;2005 and estimated a larger reduction in annual mass balance between \u0026minus;\u0026thinsp;1.5 to -1.9 m w.e./yr for the NPI compared to the Southern Patagonian Icefield (SPI) (-1.1 to -1.5 m w.e/yr).\u003c/p\u003e \u003cp\u003eAs glaciers continue to reduce under projected climate change scenarios, it becomes imperative to ascertain the timing of peak water (PW) - the period when glacier runoff increases before eventually declining - throughout the Andes taking into account regional differences. This knowledge holds paramount importance as it enables stakeholders to anticipate when glacier contributions to river flows will cease in the future. Because of that, the influence of uncertainties in future climate scenarios on future glacier changes has been a subject of investigation, as discussed by Marzeion et al. (\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Their study indicates that both at the global and regional scales, the impacts of uncertainties in future climate scenarios increase over the course of the 21st century. However, in contrast, the uncertainties related to the glacier model parameterization decrease over time. Furthermore, Hausfather et al. (\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) found that more than one-quarter of the models in the Coupled Model Intercomparison Project 6 (CMIP6) (Eyring et al., 2016) have higher variability in temperature compared to the CMIP5 models. This higher variability in temperature projections could introduce additional uncertainty in the estimates of future glacier changes. Similarly, Tokarska et al. (\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) highlighted that certain CMIP6 models with high climate sensitivity (\u003cem\u003ei.e.\u003c/em\u003e beyond the AR5 likely range of 1.5\u0026deg;-4.5\u0026deg;C by the end of the 21st century) tend to overestimate historical warming trends. Consequently, this bias might lead to future warming projections being biased towards higher temperatures in these CMIP6 models. Conversely, CMIP6 models with climate sensitivity values within the likely range exhibit warming trends consistent with observations over the historical period, providing more reliable estimates for future climate scenarios and their impact on glaciers.\u003c/p\u003e \u003cp\u003eThis study aims to address the lack of specific estimates regarding future glacier changes and their hydrological implications in the Andean glacierized catchments. For this we use a calibrated/validated model, which incorporates corrected climate variables based on measurements during the historical periodo 2000\u0026ndash;2019. Additionally, we have two main objectives. Firstly, we evaluate the performance of eight GCMs sourced from CMIP6 across the Andes (11\u0026deg;N-55\u0026deg;S) for both historical (1990\u0026ndash;2019) and future periods (2020\u0026ndash;2049). Secondly, we utilize an ensemble of evaluated GCMs (complete ensemble) and a filtered ensemble throughout the first half of the 21st century to simulate glacier runoff (including ice and snowmelt) since 2000.\u003c/p\u003e"},{"header":"2. Results","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\n \u003ch2\u003e2.1. GCMs analysis for the historical and future periods\u003c/h2\u003e\n \u003cdiv id=\"Sec4\" class=\"Section3\"\u003e\n \u003ch2\u003e2.1.1. GCMs analysis for the historical period\u003c/h2\u003e\n \u003cp\u003eThis section analyzes the correlations and errors between downscaled Global Climate Models (GCMs) and corrected TerraClimate data (cTC) across the Andes region, focusing on a monthly and seasonal scale during the historical period (1990\u0026ndash;2019) for 3,213 glaciers (covering 27,669 km\u003csup\u003e2\u003c/sup\u003e). At the monthly scale, GCMs and cTC temperature (r\u0026thinsp;=\u0026thinsp;0.9) and precipitation (r\u0026thinsp;=\u0026thinsp;0.4) exhibit statistically significant correlations on all glaciers (Figure \u003cspan class=\"InternalRef\"\u003eS1\u003c/span\u003eE). The mean monthly temperature error is 1.1\u0026thinsp;\u0026plusmn;\u0026thinsp;0.1\u0026deg;C, and the monthly total precipitation error is 106\u0026thinsp;\u0026plusmn;\u0026thinsp;68 mm. When considering glaciological regions, the Dry and Wet Andes show the largest errors in temperature and precipitation, followed by the Tropical Andes (Figure S2). At a seasonal scale, there are significant correlations for temperature and precipitation on a moderate proportion of glaciers (above 30%) during specific seasons (Table \u003cspan class=\"InternalRef\"\u003eS1\u003c/span\u003e). For example, in the Tropical Andes, correlations are significant in JFM and AMJ for precipitation, and in OND for temperature. In the Dry Andes, a considerable number of glaciers show significant correlations in all seasons. Meanwhile, in the Wet Andes, significant correlations are observed in JFM and AMJ for temperature and during JAS and OND for precipitation.\u003c/p\u003e\n \u003cp\u003eConsidering that the largest glacier mass loss occurs through the transition season (OND) in the Tropical Andes and during the summer season (JFM) in the Dry and Wet Andes, we scored the eight GCMs regarding their performance for temperature and precipitation in these seasons (see Figure \u003cspan class=\"InternalRef\"\u003eS1\u003c/span\u003eA-D). In the Tropical Andes, the largest number of glaciers is better correlated with INM-CM5, GFDL, NorESM2 (for temperature) and FGOALS and INM-CM5 (for precipitation). For the Dry Andes, high performance was estimated from CAMS, NorESM2 (for temperature) and MPI, CESM2 (for precipitation). Regarding the Wet Andes, the models INM-CM4 (for temperature) and GFDL, INM-CM4, FGOALS, and MPI (for precipitation) exhibit the highest correlations in the majority of glaciers. Conversely, the models MPI and INM-CM4 in the Tropical Andes, INM-CM4 in the Dry Andes, and MPI, INM-CM5 and FGOALS in the Wet Andes are relevant for only a very small number of glaciers. These findings provide valuable insights into the GCMs\u0026apos; performance and allow considering the most relevant one to simulate the future glacier mass loss across the Andes.\u003c/p\u003e\n \u003c/div\u003e\n \u003cdiv id=\"Sec5\" class=\"Section3\"\u003e\n \u003ch2\u003e2.1.2. Future climate change on glaciers\u003c/h2\u003e\n \u003cp\u003eWe analyze the differences in mean temperature and precipitation during the periods 1990\u0026ndash;2019 and 2020\u0026ndash;2049 for each individual glacier considering the eight GCMs. The differences led us to define the likely ranges of climate change by glaciological region considering the percentiles 10 and 90. The objective is to identify the models that are outside the regional climate change likely ranges at the annual and seasonal scales, which are then considered as hot/cold and dry/wet models.\u003c/p\u003e\n \u003cp\u003eAt an annual scale, both scenarios (SSP1-2.6 and SSP5-8.5) exhibit the most significant temperature increase in the Tropical Andes (median\u0026thinsp;=\u0026thinsp;0.7 and 0.9\u0026deg;C, respectively), followed by the Dry (median\u0026thinsp;=\u0026thinsp;0.6 and 0.9\u0026deg;C) and Wet Andes (median\u0026thinsp;=\u0026thinsp;0.4 and 0.6\u0026deg;C). Notably, SSP5-8.5 depicts a warmer outcome. In contrast, precipitation is projected to decrease in all glaciological regions and scenarios, except for the Tropical Andes under the SSP1-2.6 scenario (median\u0026thinsp;=\u0026thinsp;+\u0026thinsp;0.6%). The Dry Andes show the most considerable precipitation reduction (median = -2.8 and \u0026minus;\u0026thinsp;1.9%), trailed by the Wet Andes (median = -2.6 and \u0026minus;\u0026thinsp;0.8%) and the Tropical Andes (median = -0.8, SSP5-8.5). At a seasonal scale, the largest increase in temperature is observed in the SSP5-8.5 scenario, particularly in the Tropical Andes (JAS), followed by the Dry (OND and JFM\u0026thinsp;=\u0026thinsp;+\u0026thinsp;1.0\u0026deg;C) and Wet Andes (JFM\u0026thinsp;=\u0026thinsp;+\u0026thinsp;0.7\u0026deg;C). Regarding precipitation changes, the Tropical Andes (JFM) are estimated to experience the most significant increase, while the Dry Andes show lower negative median total precipitation during OND. Detailed percentile values can be found in Table S2 and Fig.\u0026nbsp;1.\u003c/p\u003e\n \u003cp\u003eTemperature and precipitation play a crucial role in glacier mass loss during the transition (OND) and wet seasons (JFM) in the Tropical Andes. The largest glacier ablation occurs during the transition season (Autin et al., \u003cspan class=\"CitationRef\"\u003e2022\u003c/span\u003e) and any delay in precipitation during the wet season can lead to a significant increase in ablation rates (Rabatel et al., \u003cspan class=\"CitationRef\"\u003e2013\u003c/span\u003e). In the Tropical Andes, the models project a median temperature increase in 0.7\u0026deg;C (SSP1-2.6) and 0.9\u0026deg;C (SSP5-8.5) during both seasons. Precipitation is expected to decrease during the transition season by -0.9 to -0.8% and increase during the wet season by 1.4 to 1.7%. Conversely, in the Southern Andes, glacier accumulation is concentrated in autumn/winter (AMJ and JAS), although significant precipitation also occurs in spring and summer in the Wet Andes (OND and JFM). During summer, the increased precipitation contributes to reducing the strong glacier ablation rates due to an increase in albedo. In the Dry Andes, models indicate a median temperature increase in 0.6\u0026deg;C (SSP1-2.6) and 1\u0026deg;C (SSP5-8.5) during spring and summer. However, scenarios differ significantly in terms of precipitation. Under the SSP5-8.5 scenario, median precipitation is projected to decrease by 8.8% during spring and remain unchanged (0%) during summer. On the other hand, the SSP1-2.6 scenario shows an increase in median precipitation during spring (1.2%) and a reduction during summer (-2.7%). As for the Wet Andes, both scenarios of climate change exhibit a reduction in median precipitation, with the largest impact observed during summer (precipitation reduction of -2.4 to -4.5%) compared to spring (precipitation reduction of -1.9 to -4%). Additionally, the median temperature increase is more significant in summer (temperature rise of 0.5\u0026ndash;0.7\u0026deg;C) than in spring (temperature rise of 0.4\u0026ndash;0.5\u0026deg;C). For detailed values, please refer to Table S2.\u003c/p\u003e\n \u003cp\u003eFrom the regional temperature and precipitation likely ranges (Fig.\u0026nbsp;1A-B for annual and Fig.\u0026nbsp;1C for seasonal), we identified hot/cold and dry/wet models. On an annual scale, for the Tropical Andes, GCMs\u0026apos; median values remain within likely ranges for both climate change scenarios. However, in the Dry Andes, FGOALS exhibits hot/dry characteristics and CAMS is a wet model. Moving to the Wet Andes, CESM2 and GFDL models are dry and cold, respectively, while FGOALS is a hot model. At the seasonal scale, FGOALS and CECSM2 display mean values outside the likely ranges. Specifically, in the Tropical Andes, CESM2 is a hot model during (OND). In the Dry Andes, FGOALS (OND) and CESM2 (JFM) are hot models, whereas CAMS (OND) and FGOALS (JFM) are wet models. In the Wet Andes, the FGOALS model is hot/dry (OND).\u003c/p\u003e\n \u003cp\u003eIn summary, considering analysis performed in sections \u003cspan class=\"InternalRef\"\u003e2.1.2\u003c/span\u003e and \u003cspan class=\"InternalRef\"\u003e2.2.2\u003c/span\u003e, we identified that the highest score GCMs were: CAMS, FGOALS, GFDL, INM-CM5, and NorESM2 in the Tropical Andes; GFDL, INM-CM5, MPI, and NorESM2 for the Dry Andes; CAMS, INM-CM4, and NorESM2 in the Wet Andes. In the following, we analyze changes in glacier runoff and PW considering these GCMs (filtered ensemble) and considering all the GCMs (complete ensemble).\u003c/p\u003e\n \u003c/div\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec6\" class=\"Section2\"\u003e\n \u003ch2\u003e2.2. Future glacier evolution\u003c/h2\u003e\n \u003cdiv id=\"Sec7\" class=\"Section3\"\u003e\n \u003ch2\u003e2.2.1. Changes in glacier runoff for the period 2030\u0026ndash;2049\u003c/h2\u003e\n \u003cp\u003eThis section focuses on analyzing the mean annual changes in glacier runoff between two distinct periods: 2000\u0026ndash;2019 and 2030\u0026ndash;2049. To estimate these changes, we employ the filtered ensemble of GCMs. The analysis is conducted at both the glaciological region and catchment scales taking into consideration the cumulative volumes of glacier runoff by catchment. Additionally, we computed the changes in mean annual glacier runoff using the complete ensemble of GCMs. For more detailed information, refer to Figure S4 in the Supplementary Information.\u003c/p\u003e\n \u003cp\u003eFigure 2 illustrates the mean annual changes in glacier runoff by comparing the historical and future periods for the 778 catchments analyzed in the Andes. At the scale of the glaciological regions, comparing the two periods 2000\u0026ndash;2019 and 2030\u0026ndash;2049, the simulations made using the filtered GCMs ensemble reveal a median reduction in glacier runoff of -0.1 m\u003csup\u003e3\u003c/sup\u003e s\u003csup\u003e-1\u003c/sup\u003e for scenarios SSP1-2.6 y SSP5-8.5 in the Tropical Andes. In the Dry Andes, we observe differences in glacier runoff changes between the two scenarios. Simulations made with SSP1-2.6 display a higher number of catchments with reduced glacier runoff (median = -0.01 m\u003csup\u003e3\u003c/sup\u003e s\u003csup\u003e-1\u003c/sup\u003e), indicating a decrease in overall melt rates. Conversely, for SSP5-8.5, a significant number of catchments exhibit minimal changes in glacier runoff (median = -0.003 m\u003csup\u003e3\u003c/sup\u003e s\u003csup\u003e-1\u003c/sup\u003e). However, in the catchments of the highest decile (above the 90th percentile), an increase in glacier runoff (0.72 m\u003csup\u003e3\u003c/sup\u003e s\u003csup\u003e-1\u003c/sup\u003e) is observed in the SSP5-8.5 scenario. For the Wet Andes, simulations made with both scenarios indicate a reduction in glacier runoff of -0.2 m\u003csup\u003e3\u003c/sup\u003e s\u003csup\u003e-1\u003c/sup\u003e, suggesting decreased glacier runoff rates. Negative values persist until the 90th percentile of catchments, indicating a consistent reduction in glacier runoff for this region.\u003c/p\u003e\n \u003cp\u003eAt the regional scale, changes in the glacier runoff using the filtered GCMs ensemble show similar amounts than the simulations from the complete GCMs ensemble, whereas, at the catchment scale lower changes are estimated from the filtered ensemble.\u003c/p\u003e\n \u003cp\u003eThree distinct types of behavior in glacier runoff changes are observed across catchments for the SSP1-2.6 and SSP5-8.5 scenarios (Table S3). These behaviors include positive changes, negative changes, and catchments exhibiting positive changes under SSP5-8.5 and negative changes under SSP1-2.6. The largest volume changes in glacier runoff are consistently negative across all regions and time periods (2000\u0026ndash;2019 and 2030\u0026ndash;2049) for both scenarios. The Tropical Andes exhibit the most significant annual cumulative loss in terms of percentage regarding the historical period, with a reduction of 43% (-25.8 m\u003csup\u003e3\u003c/sup\u003e s\u003csup\u003e-1\u003c/sup\u003e in SSP1-2.6). The Dry Andes follow, with a cumulative loss of 37% (-14.4 m\u003csup\u003e3\u003c/sup\u003e s\u003csup\u003e-1\u003c/sup\u003e in SSP1-2.6), and the Wet Andes, with a cumulative loss of 32% (-177.2 m\u003csup\u003e3\u003c/sup\u003e s\u003csup\u003e-1\u003c/sup\u003e in SSP1-2.6, cumulative loss of all glaciers in the region). However, a smaller number of catchments (n\u0026thinsp;=\u0026thinsp;22) show an increase in glacier runoff, mainly in the Dry Andes, experiencing a 38% increase (+\u0026thinsp;3.5 m\u003csup\u003e3\u003c/sup\u003e s\u003csup\u003e-1\u003c/sup\u003e in SSP5-8.5). Some catchments show both increases or reductions in glacier runoff depending on the scenario. In the Dry Andes (Figure S4), these catchments show a 6% increase (+\u0026thinsp;1.1 m\u003csup\u003e3\u003c/sup\u003e s\u003csup\u003e-1\u003c/sup\u003e) in the SSP5-8.5 scenario and a 7% reduction (-1.4 m\u003csup\u003e3\u003c/sup\u003e s\u003csup\u003e-1\u003c/sup\u003e) in the SSP1-2.6 scenario. Notably, the Olivares catchment (id\u0026thinsp;=\u0026thinsp;6090889690) and Cipreses catchment (id\u0026thinsp;=\u0026thinsp;6090897370) in Chile are among these catchments exhibiting contrasting behavior. Regarding the spatial distribution of catchments, increases in glacier runoff are predominant in Argentine catchments, such as the Tupungato catchment (id\u0026thinsp;=\u0026thinsp;6090891240). Conversely, the negative changes, which constitute the majority of the volume changes, are distributed evenly between Chile and Argentina. Noteworthy catchments in this category include the Azufre catchment (id\u0026thinsp;=\u0026thinsp;6090904960) and Atuel catchment (id\u0026thinsp;=\u0026thinsp;6090900470).\u003c/p\u003e\n \u003cp\u003eTable \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e displays simulations of extreme glacier runoff changes in catchments using the SSP5-8.5 scenario, categorized by glaciological regions. These results provide insights into significant variations in glacier runoff. Across all regions, the Acodado catchment in the Wet Andes experiences the most substantial reduction in glacier runoff, with a decrease of -8.4 m\u003csup\u003e3\u003c/sup\u003e s\u003csup\u003e-1\u003c/sup\u003e. Meanwhile, the Atuel catchment exhibits the largest reduction in percentage terms in the Dry Andes, with a reduction of -62% (-2.4 m\u003csup\u003e3\u003c/sup\u003e s\u003csup\u003e-1\u003c/sup\u003e). These changes represent a considerable decline regarding their historical annual glacier runoff between 2000\u0026ndash;2019. Conversely, the Huemules, Tupungato, and Olivares catchments show the most substantial increases in glacier runoff. It is noteworthy that the eastern side of the Andes (Argentina) shows the most pronounced reductions and increases in glacier runoff, indicating potential impacts on water resources and hydrological systems in the Dry Andes. Comparatively, the Tropical Andes experience smaller extreme changes in glacier runoff compared to other two glaciological regions, which can be attributed to relatively lower mean glacier runoff during the historical period. Notably, catchments such as Chawpi Urqu and Quelccaya display mean glacier runoff below 2.3\u0026thinsp;\u0026plusmn;\u0026thinsp;1.3 m\u003csup\u003e3\u003c/sup\u003e s\u003csup\u003e-1\u003c/sup\u003e for the 2000\u0026ndash;2019 period.\u003c/p\u003e\n \u003cdiv class=\"gridtable\"\u003e\u0026nbsp;\u0026nbsp;\u003ctable id=\"Tab1\" border=\"1\"\u003e\n \u003ccaption language=\"En\"\u003e\n \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\n \u003cdiv class=\"CaptionContent\"\u003e\n \u003cp\u003eMaximum reductions and increases in glacier runoff identified using scenario 8.5 between the periods 2000\u0026ndash;2019 and 2030\u0026thinsp;\u0026minus;\u0026thinsp;2019 across the Andes\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eCatchment name\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eMean annual Glacier runoff\u003c/p\u003e\n \u003cp\u003ein the reference 2000\u0026ndash;2019\u003c/p\u003e\n \u003cp\u003e[m\u003csup\u003e3\u003c/sup\u003e s\u003csup\u003e-1\u003c/sup\u003e]\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eChange of glacier runoff between\u003c/p\u003e\n \u003cp\u003e2030\u0026ndash;2049 and 2000\u0026ndash;2019\u003c/p\u003e\n \u003cp\u003e[m\u003csup\u003e3\u003c/sup\u003e s\u003csup\u003e-1\u003c/sup\u003e]\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eChange of glacier runoff between\u003c/p\u003e\n \u003cp\u003e2030\u0026ndash;2049 and 2000\u0026ndash;2019\u003c/p\u003e\n \u003cp\u003e[%]\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eGlaciological region\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eCountry\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eCatchment\u003c/p\u003e\n \u003cp\u003eid\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eCatchment area\u003c/p\u003e\n \u003cp\u003e[km\u003csup\u003e2\u003c/sup\u003e]\u003c/p\u003e\n \u003cp\u003eGlacierized\u003c/p\u003e\n \u003cp\u003earea\u003c/p\u003e\n \u003cp\u003e(%)\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eLatitude\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colspan=\"7\"\u003e\n \u003cp\u003eNegative changes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eAcodado\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e18.8\u0026thinsp;\u0026plusmn;\u0026thinsp;2.1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-8.4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-45\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eWet Andes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eChile\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e6090024320\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e639 (26)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e47\u0026deg;S\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eAtuel\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e4.0\u0026thinsp;\u0026plusmn;\u0026thinsp;0.6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-2.4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-62\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eDry Andes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eArgentina\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e6090900470\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e348 (14)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e35\u0026deg;S\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eChawpi Urqu\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2.3\u0026thinsp;\u0026plusmn;\u0026thinsp;1.3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-1.2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-45\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eTropical Andes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePer\u0026uacute;-Bolivia\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e6090601720\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e198 (21)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e15\u0026deg;S\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colspan=\"9\"\u003e\n \u003cp\u003ePositive changes\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eHuemules\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e7.2\u0026thinsp;\u0026plusmn;\u0026thinsp;2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e18\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eWet Andes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eChile\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e6090965210\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e207 (43)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e48\u0026deg;S\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eTupungato\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2.7\u0026thinsp;\u0026plusmn;\u0026thinsp;1.0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.9\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e32\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eDry Andes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eArgentina\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e6090891240\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e289 (28)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e34\u0026deg;S\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eOlivares\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2.7\u0026thinsp;\u0026plusmn;\u0026thinsp;1.1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e16\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eDry Andes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eChile\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e6090889690\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e534 (14)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e33\u0026deg;S\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQuelccaya\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.8\u0026thinsp;\u0026plusmn;\u0026thinsp;0.4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e45\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eTropical Andes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePer\u0026uacute;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e6090582670\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e182 (9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e14\u0026deg;S\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n \u003c/div\u003e\n \u003c/div\u003e\n \u003cdiv id=\"Sec8\" class=\"Section3\"\u003e\n \u003ch2\u003e2.2.2. Peak water throughout 21st century along the Andes\u003c/h2\u003e\n \u003cp\u003eThe expected peak water (PW) estimation for 778 Andean catchments considering the filtered ensemble of GCMs indicates that the maximum contribution of glacier runoff to river discharge will likely occur before the first half of the 21st century. Figure\u0026nbsp;3 illustrates the distribution of PW years (percentiles 25 and 75 in both scenarios), with a concentration between 2010 and 2028 across the Andes. The closest PW years to the present occur first in the Wet Andes (2010\u0026ndash;2024, n\u0026thinsp;=\u0026thinsp;465 catchments), followed by the Tropical Andes (2014\u0026ndash;2030, n\u0026thinsp;=\u0026thinsp;183 catchments), and finally in the Dry Andes (2021\u0026ndash;2046, n\u0026thinsp;=\u0026thinsp;130 catchments). For the future period (2026\u0026ndash;2099), PW years occur most frequently between 2026 and 2049. Tropical Andes will likely experience PW years in most catchments sooner (2026\u0026ndash;2040, 50 to 86 catchments) than the Wet Andes (2030\u0026ndash;2038, 40 to 30 catchments), and finally the Dry Andes (2030\u0026ndash;2048, 82 to 110 catchments). Interestingly, in the Dry Andes, most catchments will likely show PW years later in the future (2026\u0026ndash;2099). More details regarding the distribution of PW years by glaciological regions can be found in Figure S5.\u003c/p\u003e\n \u003cp\u003eSpecific locations in Fig. 3 and catchments in Table \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e provide a detailed view of the changes in PW year and the associated amounts of glacier runoff. These details allow for an examination of glacier runoff at the catchment scale within the SSP5-8.5 scenario.\u003c/p\u003e\n \u003cp\u003eIn the Tropical Andes, catchments in Colombian Sierra Nevada de Santa Marta (\u003cem\u003ee.g\u003c/em\u003e., Pico Colon) display PW years spanning from 2020 to 2028, with a maximum glacier runoff of 0.3 m\u003csup\u003e3\u003c/sup\u003e s\u003csup\u003e-1\u003c/sup\u003e. In Ecuador, PW years are projected to occur later in the second half of the 21st century, specifically from 2022 to 2084. Notably, the Altar catchment in Ecuador stands out, with a PW year estimated at 2052\u0026thinsp;\u0026plusmn;\u0026thinsp;10 and a glacier runoff of 1.0 m\u003csup\u003e3\u003c/sup\u003e s\u003csup\u003e-1\u003c/sup\u003e. In Per\u0026uacute;, cordilleras with significant glacierized areas would experience their PW year before the first half of the 21st century. The Cordillera Blanca shows PW years ranging from 2018 to 2064, while the Cordillera Vilcanota exhibits PW years from 2024 to 2050. The Marcapata catchment, located in the Cordillera Vilcanota, holds the largest estimated glacier runoff volume in the Tropical Andes at 3.0 m\u003csup\u003e3\u003c/sup\u003e s\u003csup\u003e-1\u003c/sup\u003e, with a PW year projected in 2030\u0026thinsp;\u0026plusmn;\u0026thinsp;10. The Dry Andes shows a range of PW years from 2010 to 2062. The largest glacier runoff is simulated in the Cipreses catchment (9.3 m\u003csup\u003e3\u003c/sup\u003e s\u003csup\u003e-1\u003c/sup\u003e), followed by the Volc\u0026aacute;n catchment (4.0 m\u003csup\u003e3\u003c/sup\u003e s\u003csup\u003e-1\u003c/sup\u003e), and the Olivares catchment (3.6 m\u003csup\u003e3\u003c/sup\u003e s\u003csup\u003e-1\u003c/sup\u003e). For PW years simulated after 2049, larger glacier runoff volumes are observed in the Tupungato catchment (4.0 m\u003csup\u003e3\u003c/sup\u003e s\u003csup\u003e-1\u003c/sup\u003e) and the Yeso catchment (1.2 m\u003csup\u003e3\u003c/sup\u003e s\u003csup\u003e-1\u003c/sup\u003e). In contrast to the Tropical and Dry Andes, the Wet Andes region exhibits significant amounts of glacier runoff. Catchments in the Wet Andes, particularly in the latitudinal range of 46\u0026ndash;48\u0026deg;S, considering the land-terminating glaciers of Northern Patagonian Icefield and surrounding glaciers, show a PW year range before the first half of the 21st century, from 2010 to 2048. The Acodado catchment stands out with the highest maximum glacier runoff of 17.8 \u003csup\u003e3\u003c/sup\u003e s\u003csup\u003e-1\u003c/sup\u003e and a PW year of 2010\u0026thinsp;\u0026plusmn;\u0026thinsp;10, situated on the west side of the Andes. On the eastern side of the Northern Patagonian Icefield, catchments related to the Baker basin are estimated to experience PW years before 2030. Among these catchments, the Soler catchment (in the NPI) exhibits a larger PW volume of 8.5 \u003csup\u003e3\u003c/sup\u003e s\u003csup\u003e-1\u003c/sup\u003e compared to those found east of the NPI, like the Mait\u0026eacute;n catchment, with a runoff of 0.9 m\u003csup\u003e3\u003c/sup\u003e s\u003csup\u003e-1\u003c/sup\u003e. From this analysis, we conclude that the calculation of the PW for each catchment is crucial in characterizing the local differences and spatial variability observed in the Andes. These diverse behaviors are attributed to the distinct morphometric characteristics of land-terminating glaciers and local climates in each catchment.\u003c/p\u003e\n \u003cp\u003eThe annual and monthly temporal series of glacier runoff for each glacier and catchment are available in the Supplementary data. For more in-depth information and specific examples, refer to Figures S6, S7, and S8 in the Supplementary information. These figures present the detailed temporal variations in glacier runoff for the respective glaciers and catchments.\u003c/p\u003e\n \u003cp\u003e\u003c/p\u003e\u0026nbsp;\u003ctable id=\"Tab2\" border=\"1\"\u003e\n \u003ccaption language=\"En\"\u003e\n \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e\n \u003cdiv class=\"CaptionContent\"\u003e\n \u003cp\u003eCatchments highlighted by glacier runoff during the identified peak water year between 2000\u0026ndash;2099 across the Andes\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eCatchment name\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003ePW year\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003ePW glacier runoff\u003c/p\u003e\n \u003cp\u003e[m\u003csup\u003e3\u003c/sup\u003e s\u003csup\u003e-1\u003c/sup\u003e]\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eLocation\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eCountry\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eCatchment\u003c/p\u003e\n \u003cp\u003eid\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eCatchment area\u003c/p\u003e\n \u003cp\u003e[km\u003csup\u003e2\u003c/sup\u003e]\u003c/p\u003e\n \u003cp\u003eGlacierized\u003c/p\u003e\n \u003cp\u003earea\u003c/p\u003e\n \u003cp\u003e(%)\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eLatitude\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePico Crist\u0026oacute;bal Col\u0026oacute;n\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2024\u0026thinsp;\u0026plusmn;\u0026thinsp;10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.3\u0026thinsp;\u0026plusmn;\u0026thinsp;0.02\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eColombia\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eColombia\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e6090000970\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e696 (1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e11\u0026deg;N\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eAltar\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2052\u0026thinsp;\u0026plusmn;\u0026thinsp;10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e1.0\u0026thinsp;\u0026plusmn;\u0026thinsp;0.04\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eEcuador\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eEcuador\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e6090249220\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e668 (3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2\u0026deg;S\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eVicos\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2020\u0026thinsp;\u0026plusmn;\u0026thinsp;10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e1.5\u0026thinsp;\u0026plusmn;\u0026thinsp;0.05\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCo. Blanca\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePer\u0026uacute;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e6090461650\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e279 (19)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e9\u0026deg;S\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eIchiccocha\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2030\u0026thinsp;\u0026plusmn;\u0026thinsp;10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e1.2\u0026thinsp;\u0026plusmn;\u0026thinsp;0.06\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCo. Blanca\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePer\u0026uacute;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e6090449220\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e246 (13)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e9\u0026deg;S\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eMarcapata\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2030\u0026thinsp;\u0026plusmn;\u0026thinsp;10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e3.0\u0026thinsp;\u0026plusmn;\u0026thinsp;0.09\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCo.Vilcanota\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePer\u0026uacute;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e6090571030\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e730 (14)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e14\u0026deg;S\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eOlivares\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2040\u0026thinsp;\u0026plusmn;\u0026thinsp;10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e3.6\u0026thinsp;\u0026plusmn;\u0026thinsp;0.18\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCL-AR\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eChile\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e6090889690\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e534 (14)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e33\u0026deg;S\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eTupungato\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2058\u0026thinsp;\u0026plusmn;\u0026thinsp;10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e4.0\u0026thinsp;\u0026plusmn;\u0026thinsp;0.09\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCL-AR\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eArgentina\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e6090891240\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e289 (28)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e34\u0026deg;S\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eYeso\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2054\u0026thinsp;\u0026plusmn;\u0026thinsp;10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e1.2\u0026thinsp;\u0026plusmn;\u0026thinsp;0.02\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCL-AR\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eChile\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e6090892710\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e627 (6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e34\u0026deg;S\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eVolc\u0026aacute;n\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2040\u0026thinsp;\u0026plusmn;\u0026thinsp;10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e4.0\u0026thinsp;\u0026plusmn;\u0026thinsp;0.06\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCL-AR\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eChile\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e6090892940\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e524 (14)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e34\u0026deg;S\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCipreses\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2010\u0026thinsp;\u0026plusmn;\u0026thinsp;10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e9.3\u0026thinsp;\u0026plusmn;\u0026thinsp;0.23\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCL-AR\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eChile\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e6090897370\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e350 (33)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e35\u0026deg;S\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSoler\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2026\u0026thinsp;\u0026plusmn;\u0026thinsp;10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e8.5\u0026thinsp;\u0026plusmn;\u0026thinsp;0.24\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eNPI\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eChile\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e6090963530\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e749 (16)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e47\u0026deg;S\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eMait\u0026eacute;n\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2010\u0026thinsp;\u0026plusmn;\u0026thinsp;10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.9\u0026thinsp;\u0026plusmn;\u0026thinsp;0.05\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eEast of NPI\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eChile\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e6090962900\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e386 (7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e47\u0026deg;S\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eAcodado\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2010\u0026thinsp;\u0026plusmn;\u0026thinsp;10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e17.8\u0026thinsp;\u0026plusmn;\u0026thinsp;0.39\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eNPI\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eChile\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e6090024320\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e639 (26)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e47\u0026deg;S\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n \u003cp\u003e\u003c/p\u003e\n \u003c/div\u003e\n\u003c/div\u003e"},{"header":"3. Discussion","content":"\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003e3.1. Climate projections analysis\u003c/h2\u003e \u003cp\u003ePresently, a comprehensive evaluation of GCMs outputs in glacierized catchments across the Andes is notably lacking, encompassing both CMIP6 and earlier datasets. Consequently, the prevailing reference source has been the IPCC AR6 report (IPCC, 2022), which examined climate variations in the Andes utilizing over 30 CMIP6 GCMs. Nevertheless, a key distinction between our simulations, focused on glaciers, and the IPCC report is the treatment of regional extension, wherein the latter considers a broader land area beyond the glacierized surfaces and situated at lower elevations. IPCC (2022) indicates an overall temperature increase for all scenarios, with the northern half of the continent experiencing more substantial warming, gradually decreasing southward. Additionally, precipitation is anticipated to decrease over the Southern Andes while increasing over the Northern Andes during the period 2041\u0026ndash;2060, taking 1995\u0026ndash;2014 as a reference period (under the SSP5-8.5 scenario). This pattern is consistent with the findings of Almazroui et al. (\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) who evaluated an ensemble of CMIP6 models across South America. Moreover, the GCMs ensemble tends to underestimate precipitation during the rainiest months while overestimating it during the drier months in the Southern Andes, particularly in the Southwestern region (Almazroui et al., \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). In agreement with the IPCC under the SSP5-8.5 scenario, our results confirm a maximum temperature increase in the Tropical Andes (1.9\u0026deg;C), followed by the Dry (1.5\u0026deg;C), and Wet Andes (1.1\u0026deg;C, southward of NPI) during the period 2030\u0026ndash;2049. For precipitation, the IPCC reports a reduction in the Dry Andes (-8%) and Wet Andes (-0.1%), alongside an increase in the Tropical Andes (4.5%). However, our findings diverge, showing a negative trend in the Tropical and Wet Andes. Notably, precipitation only exhibits an increase in the SSP1-2.6 scenario for the Tropical Andes.\u003c/p\u003e \u003cp\u003eRegarding the Tropical Andes and the northern area of the Dry Andes, Olmo et al. (\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) conducted a study focusing on the representation of precipitation variability in the July-October season (1979\u0026ndash;2014). By using the CESM2 and MPI-ESM1-2-HR models from CMIP6, they indicated that these models effectively capture the underlying physical mechanisms governing precipitation patterns. Our analysis revealed that these same models exhibit a high monthly correlation but a poor seasonal correlation for the OND season. However, in the Dry Andes region (18\u0026ndash;37\u0026deg;S), these models exhibit a robust seasonal correlation, which holds significant importance for glacier mass balance simulations.\u003c/p\u003e \u003cp\u003eRegarding future projections, we observed a median reduction in precipitation during the JAS and OND seasons (as projected by MPI, CESM2, and NorESM2-MM) under the SSP1-2.6 and SSP5-8.5 scenarios. These results are in line with estimations made by Agudelo et al. (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) which indicate an increase in the occurrence of dry days (19.4%) during the austral winter (JAS) and a higher frequency of dry circulation patterns during the July-October period. Notably, the CESM2 model demonstrated the most favorable outcomes concerning precipitation variability over the southwestern region during the historical period (1901\u0026ndash;2014) as indicated by Rivera and Arnould (\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eGiven the foregoing, the use of GCMs as input data in future glacier runoff simulations should consider the following aspects. First, the importance of the seasonal temporal evaluation. Indeed, it is crucial to assess GCM performance during critical periods of glacier melt because neglecting seasonal evaluations in favor of monthly or annual assessments may overlook deficiencies in seasonal performance. Second, the GCM performance over glacierized areas must be evaluated. Indeed, using GCMs evaluated for other spatial extensions could present trends and seasonal variations that are not representative of glacierized regions but of lower elevations.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003e3.2. Glacier volume and runoff simulations toward 2049\u003c/h2\u003e \u003cp\u003eThree critical aspects in the glacier dynamics considered in the simulations using OGGM need to be discussed to account for the likelihood of the presented results:\u003c/p\u003e \u003cp\u003e \u003col\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eThe approach based on the Shallow Ice Approximation (SIA), implemented in OGGM, lacks longitudinal/transverse stress gradients and other complex mechanisms of glacier dynamics (Le Meur et al., \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2004\u003c/span\u003e). As a consequence, glaciers' response to climate forcing is nearly immediate, making higher-order ice flow models (Oerlemans, \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2008\u003c/span\u003e) more suitable for accurately representing ice flow. However, the use of these high-order models presents computational challenges (Jouvet, 2022), currently limiting their application in regional and global simulations.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eThe ice thickness calibration in this study relies on data from Farinotti et al. (\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2019\u003c/span\u003e), which employs an ensemble of up to five models to estimate the ice thickness distribution. Unfortunately, this approach leads to an overestimation of approximately 20% (median) of the measured ice thickness. Moreover, our calibration parameter values for ice thickness align with values from Cuffey and Paterson (\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2010\u003c/span\u003e) and Millan et al. (\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). The simulations predict lower Glen A parameter values for glaciers with lower internal temperatures, such as those found in the glaciological zone DA1 (2.4 10\u003csup\u003e\u0026ndash;25\u003c/sup\u003e s\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e Pa\u003csup\u003e\u0026minus;\u0026thinsp;3\u003c/sup\u003e), which represents the coldest zone in the Andes. Additionally, Millan et al. (\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) estimated Glen A parameters in the Andes ranging between 5.6 10\u003csup\u003e\u0026ndash;25\u003c/sup\u003e and 2.4 10\u003csup\u003e\u0026ndash;24\u003c/sup\u003e s\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e Pa\u003csup\u003e\u0026minus;\u0026thinsp;3\u003c/sup\u003e, while the calibrated values used in this study range from 2.4 10\u003csup\u003e\u0026ndash;25\u003c/sup\u003e and 2.4 10\u003csup\u003e\u0026ndash;23\u003c/sup\u003e s\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e Pa\u003csup\u003e\u0026minus;\u0026thinsp;3\u003c/sup\u003e.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eDue to the limitations mentioned in (i) and (ii), our simulations of future glacier runoff changes carry uncertainties arising from the ice thickness simulation and its calibration. Moreover, the climate performance in the historical period and the various climate models and scenarios considered contribute to the overall uncertainty. Estimating melt factor values is critical, and we derive them from the calibration, using Hugonnet et al. (\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) data of geodetic mass balance. However, we must acknowledge that our historical climate dataset, which was corrected in Caro et al. (\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) and combined with Hugonnet et al. (\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) data could introduce errors in the calibrated melt factor. Consequently, this error could affect the simulated mass balance of glaciers and the related ice thickness estimation. Furthermore, the future climate data from each GCM (temperature and precipitation) was corrected relative to the historical climate dataset.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003c/ol\u003e \u003c/p\u003e \u003cp\u003eThe three main sources of uncertainty in future simulations of glacier runoff are: (i) the SIA, (ii) the calibration of glacier volume, and (iii) the use of GCMs and parameter values for mass balance estimation. Despite simulation biases, simulations at the Andes scale capture variations in future glacial runoff between catchments and Andean regions which sound coherent with current spatio-temporal differences. This capability arises from a consistent modeling approach, calibrated and validated across the Andes using historical data. Also, It incorporates the highest-scored climate projections specific to each region for future simulations.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003e3.3. PW estimation across the Andes: comparison with former studies\u003c/h2\u003e \u003cp\u003eOur 21st century PW estimates partially align with Huss and Hock (\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2018\u003c/span\u003e) findings. They identified PW in 12 Andean basins with a glacierized surface area of 9,544 km\u003csup\u003e2\u003c/sup\u003e, representing different proportions in the Tropical Andes (23%), Dry Andes (20%) and Wet Andes (57%). Their research revealed a PW already past in the inner tropics (2 catchments) and projected a PW occurrence between 2011\u0026ndash;2046 in the outer tropics (4 catchments). In the Dry Andes, the estimated PW was around 2010\u0026thinsp;\u0026plusmn;\u0026thinsp;24 (2 catchments in western side), while the Wet Andes showed a broader range from 2003\u0026thinsp;\u0026plusmn;\u0026thinsp;11 to 2096\u0026thinsp;\u0026plusmn;\u0026thinsp;24 (4 catchments). Interestingly, they found evidence of past PW in the northern area of the Wet Andes.\u003c/p\u003e \u003cp\u003eIn contrast, our estimations considered a larger glacierized surface area (11,282 km\u003csup\u003e2\u003c/sup\u003e), filtering glaciers not accounted for in glacier national inventories and calving glaciers in the Patagonian icefields. Our results suggest that the PW will occur before the first half of the 21st century in the majority of the catchments across the Andes. According to our results, the PW has already occurred in most of the Tropical Andes catchments and will occur before 2049 in most of the Dry Andes catchments. In the Wet Andes, the majority of catchments experienced PW before the present day. Furthermore, our estimation of PW presents different ranges of years across the Andes, tied too to varying climate change scenarios. Generally, an earlier PW occurrence is associated with less warm scenarios, which aligns with observations in the Himalayan mountains and specific Andean catchments (Huss and Hock, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Laha et al., \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e"},{"header":"4. Conclusion","content":"\u003cp\u003eWe analyzed eight GCMs from CMIP6 to identify those best at reproducing historical climate (1990\u0026ndash;2019) and projected future conditions (2020\u0026ndash;2049) across the Andes. These models were then used to estimate changes in glacier runoff throughout the 21st century (10\u0026deg;N-55\u0026deg;S) using the calibrated and validated Open Global Glacier Model (OGGM). For the first time, we conducted a comprehensive simulation of future glacier dynamics and runoff in 778 Andean catchments. We conclude the following:\u003c/p\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003eThe ensemble of all GCMs (complete ensemble) showed larger changes in glacier runoff between periods 2000\u0026ndash;2019 and 2030\u0026ndash;2049), compared to the ensemble using the highest-scoring GCMs (filtered ensemble) at the catchment scale. However, these differences were not evident at the regional scale.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eGlacier runoff in the Tropical and Wet Andes (filtered ensemble) showed notable reductions between 2000\u0026ndash;2019 and 2030\u0026ndash;2049, with median decreases of -0.1 m\u003csup\u003e3\u003c/sup\u003e s\u003csup\u003e-1\u003c/sup\u003e and \u0026minus;\u0026thinsp;0.2 m\u003csup\u003e3\u003c/sup\u003e s\u003csup\u003e-1\u003c/sup\u003e, respectively. In contrast, the Dry Andes exhibited a relatively smaller reduction (median of -0.003 m\u003csup\u003e3\u003c/sup\u003e s\u003csup\u003e-1\u003c/sup\u003e), with some catchments, like Tupungato and Olivares, experiencing at least a 16% increase. Cumulative glacier runoff in the Dry Andes increased by 38%, while the Tropical Andes experienced the most significant cumulative loss, with a reduction of 43% compared to the historical period.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eProjections indicate peak water from glacier runoff will occur in most Andean catchments before the mid-21st century (PW year between 2010\u0026ndash;2028). Nevertheless, the distribution of PW years exhibits significant variations in Andean regions and catchments. The Wet Andes is expected to experience the earliest PW years (2010 to 2024), followed by the Tropical Andes (2014 to 2030), and the Dry Andes later (2021\u0026ndash;2046). Within this region, specific attention should be given to catchments that show the highest amounts of glacier runoff, such as the Cipreses (PW in 2010\u0026thinsp;\u0026plusmn;\u0026thinsp;10 with 9.3 m\u003csup\u003e3\u003c/sup\u003e s\u003csup\u003e-1\u003c/sup\u003e), Volc\u0026aacute;n (PW in 2040\u0026thinsp;\u0026plusmn;\u0026thinsp;10 with 4.0 m\u003csup\u003e3\u003c/sup\u003e s\u003csup\u003e-1\u003c/sup\u003e), and the Olivares catchment (PW in 2040\u0026thinsp;\u0026plusmn;\u0026thinsp;10 with 3.6 m\u003csup\u003e3\u003c/sup\u003e s\u003csup\u003e-1\u003c/sup\u003e).\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e \u003cp\u003eThis work provides a comprehensive understanding of the variations in glacier runoff using high-scored GCMs, underlying the importance of accounting for regional and catchment-scale differences in water resource adaptation strategies. This is particularly crucial in regions most impacted by prolonged droughts in the Andes, such as the central zones of Chile and Argentina.\u003c/p\u003e"},{"header":"5. Data and Methods","content":"\u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003e5.1. Analysis of the GCMs in the historical and future periods\u003c/h2\u003e \u003cp\u003eIn this study, we analyze temperature and precipitation data from two scenarios and eight GCMs simulations sourced from CMIP6 (see Table S4). Our analysis takes into account the Almazroui et al. (\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) assessment for GCMs in South America and the global evaluation of warmer models from Hausfather et al. (\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) and Tokarska et al. (\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). To establish a basis for comparison, we scale the GCMs simulations using the bias-correction method implemented in the OGGM model and compare them with the TerraClimate data (Abatzoglou et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2018\u003c/span\u003e), that have been corrected for temperature and precipitation (cTC, temperature, and precipitation) at the scale of the Andes on the basis of in situ data from 34 meteorological stations (Caro et al., \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) during the historical period of 1990\u0026ndash;2019 across the Andes on a glacierized area of 27,668 km\u003csup\u003e2\u003c/sup\u003e (considering glaciers with a surface area\u0026thinsp;\u0026gt;\u0026thinsp;1 km\u003csup\u003e2\u003c/sup\u003e). Meanwhile, we also perform comparisons between GCMs during both the historical and future periods (2020\u0026ndash;2049). Our analysis comprises three steps: step 1: statistical downscaling of GCMs temperature and precipitation data for the historical period; step 2: calculation of annual, seasonal and monthly metrics to compare the GCMs data with the corrected TerraClimate data for the historical period; and step 3: Identification of changes in climate variables between the historical and future periods. Such simulations were deemed impossible to conduct in this context. Going into detail, these three steps consist in:\u003c/p\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003eStep 1. The mean monthly and annual values of temperature and precipitation from the GCMs were adjusted to the mean elevation of each glacier. This adjustment was performed using a statistical downscaling approach for two future scenarios (SSP1-2.6 and SSP5-8.5), based on the historical climate data from the cTC (1990\u0026ndash;2019).\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eStep 2. Correlation pattern analyses were based on monthly and seasonal (OND, JFM, AMJ, JAS) correlations between the eight GCMs and the cTC data in the period 1990\u0026ndash;2019. We chose seasons as OND and JFM, because in these months the larger surface mass loss occurs in the Outer Tropics (transition season, Autin et al., \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) and in the Dry Andes zones (austral summer). We used the Pearson coefficient of correlation and the root mean square error (RMSE) (McSweeney et al., \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2015\u003c/span\u003e) as metrics to score the GCMs performance during the historical period.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eStep 3. We identified climate likely ranges through the GCMs ensemble considering the mean differences between the historical and future periods for temperature and precipitation at annual and seasonal (OND, JFM, AMJ, JAS) time-steps in each glacier. The climate change likely ranges are estimated by the percentiles 10 and 90 of these differences using all GCMs and glaciers (modified from DRIAS project, 2023). In addition, we estimate the median of these differences. A similar method is used by the DRIAS project of the \u003cem\u003eMinist\u0026egrave;re de la Transition \u0026Eacute;cologique\u003c/em\u003e of the French government (see Table S2, DRIAS project, 2023, and S\u0026oslash;rland et al., \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). The annual analysis between the historical and future periods (2020\u0026ndash;2049 and 2070\u0026ndash;2099) is considered to test our results with previous reports of climate change (e.g., IPCC, 2022; Olmo et al., \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Agudelo et al., \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). The estimation of the climate change likely ranges allows us to identify four types of models: the hot/dry; the cold/dry; the hot/wet and the cold/wet models.\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e \u003cp\u003eAfter these steps, we defined a complete ensemble considering eight GCMs, and a second GCM ensemble that comprise highly score GCMs, called filtered ensemble.\u003c/p\u003e \u003cp\u003eThe GCMs output data were gathered for each glacier. However, the analysis was performed at the glaciological region scale, where climate characteristics vary substantially (Caro et al., \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). We did not include glaciers experiencing mass loss due to calving in our simulations conducted in the Patagonian region (Minowa et al., \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003e5.2. Short description of the Open Global Glacier Model (OGGM)\u003c/h2\u003e \u003cp\u003eOGGM is a modular and open-source workflow that simulates glacier mass balance and ice dynamics using calibrated parameter values for each glacier (Maussion et al., \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). The required input data are: air temperature and precipitation time series, glacier outlines and surface topography. From these inputs, annual (\u003cem\u003ee.g.\u003c/em\u003e, surface mass balance, glacier volume and area) and monthly (glacier melt [snow\u0026thinsp;+\u0026thinsp;ice] and rainfall on glaciers) outputs can be simulated.\u003c/p\u003e \u003cp\u003eUsing a glacier outline and topography, OGGM estimates flow lines using a geometrical algorithm (adapted from Kienholz et al., \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2014\u003c/span\u003e). Assuming a bed shape, it estimates the ice thickness based on mass conservation and shallow ice approximation (Maussion et al., \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). After these numerical steps, area and volume per glacier can be simulated. Mass balance is implemented using a precipitation phase partitioning and a temperature-index approach (Braun and Renner, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e1992\u003c/span\u003e; Hock, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2003\u003c/span\u003e; Marzeion et al., \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2012\u003c/span\u003e). The monthly mass balance \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{mb}_{i}\\)\u003c/span\u003e\u003c/span\u003e at an elevation z is computed as follows:\u003c/p\u003e \u003cp\u003e \u003cspan class=\"InlineEquation\"\u003e \u003cspan class=\"mathinline\"\u003e\\(\\:\\:{mb}_{i}\\left(z\\right)={TC}_{p\\:i}^{snow}\\left(z\\right){*P}_{f}-\\:{M}_{f}*max\\left({cTC}_{t\\:i}\\left(z\\right)-\\:{T}_{melt}\\:,0\\right)\\)\u003c/span\u003e \u003c/span\u003eequation (1),\u003c/p\u003e \u003cp\u003ewhere \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{TC}_{p\\:i}^{snow}\\)\u003c/span\u003e\u003c/span\u003e is the TerraClimate solid precipitation before being scaled by the precipitation correction factor (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{P}_{f}\\)\u003c/span\u003e\u003c/span\u003e), \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{M}_{f}\\)\u003c/span\u003e\u003c/span\u003e is the glacier's temperature sensitivity parameter, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{cTC}_{t\\:i}\\)\u003c/span\u003e\u003c/span\u003e is the monthly corrected TerraClimate temperature, and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{T}_{melt}\\)\u003c/span\u003e\u003c/span\u003e is the monthly air temperature above which ice melt is assumed to occur (from 0\u0026deg;C to 2.1\u0026deg;C). \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{TC}_{p\\:i}^{snow}\\)\u003c/span\u003e\u003c/span\u003e is computed as a fraction of the total precipitation (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{cTC}_{p}^{}\\)\u003c/span\u003e\u003c/span\u003e) where 100% is getting if \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{cTC}_{t\\:i}\\)\u003c/span\u003e\u003c/span\u003e \u0026lt;= \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{T}_{i}^{snow}\\)\u003c/span\u003e\u003c/span\u003e (between 0-2.1\u0026deg;C) and 0% if \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{cTC}_{t\\:i}\\)\u003c/span\u003e\u003c/span\u003e \u0026gt;= \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{T}_{i}^{rain}\\)\u003c/span\u003e\u003c/span\u003e (between 2-4.1\u0026deg;C); and linearly interpolated in between. Here, the \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{M}_{f}\\)\u003c/span\u003e\u003c/span\u003e was calibrated for each glacier individually using glacier volume change datasets previously described (Hugonnet et al., \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e \u003cdiv id=\"Sec17\" class=\"Section3\"\u003e \u003ch2\u003e5.2.1. Model setup in the period 2000\u0026ndash;2050\u003c/h2\u003e \u003cp\u003eThe OGGM model was previously calibrated and evaluated during the period 2000\u0026ndash;2019 across the Andes (10\u0026deg;N-55\u0026deg;S) in a former study by Caro et al. (\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). In this former study, we evaluated and corrected temperature and precipitation input data, evaluated the simulated glacier mass balances outputs using \u003cem\u003ein situ\u003c/em\u003e mass balance measurements, and also the model performance at the level of three Andean catchments.\u003c/p\u003e \u003cp\u003eThe model was run for each glacier and then results were analyzed for each glaciological zone identified by Caro et al. (\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) across the Andes. During the historical (2000\u0026ndash;2019) and future (2020\u0026ndash;2050) periods the input data are: glacier outlines from RGI v6.0 (RGI Consortium, \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2017\u003c/span\u003e) and surface topography from NASADEM (Crippen et al., \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). The corrected monthly TerraClimate precipitation (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:c{TC}_{p}^{}\\)\u003c/span\u003e\u003c/span\u003e) and temperature (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{cTC}_{t}^{}\\)\u003c/span\u003e\u003c/span\u003e) were used in the historical period. For the future period, monthly precipitation and temperature were selected from eight GCMs. These variables were used as input for the future simulations. The calibration procedure was applied for each individual glacier to adjust the simulated mass balance of the 2000\u0026ndash;2019 period to the geodetic mass balance product from Hugonnet et al. (\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). The simulated glacier volume was calibrated using Farinotti et al. (\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2019\u003c/span\u003e) product at a glaciological zone scale fitting the A parameter of the Glen flow law. In addition, glacier outlines of all glaciers were associated with the year 2000. The main corrected and calibrated parameters to run OGGM across the Andes are summarized in Table S5.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003e5.3. Glacier runoff analysis\u003c/h2\u003e \u003cp\u003eGlacier runoff changes were estimated by adding the annual melt from each glacier by catchment, for this we considered the centroid of each glacier contour. From this annual glacier runoff time series, the difference of the mean annual glacier runoff between the periods 2000\u0026ndash;2019 and 2030\u0026ndash;2049 was estimated for 786 glacierized catchments. Meanwhile, PW refers to the annual glacier runoff that will initially increase and after declining in response to glacier retreat due to changes in climate conditions (Huss and Hock, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2018\u003c/span\u003e) which can be for example a long-term response of catchments to sustained warming (Hock et al., 2005). We calculate the PW inspired by Huss and Hock (\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2018\u003c/span\u003e) through the following procedure by glacier and after by catchment: i) A maximum of eight time series of the simulated glacier runoff from each glacier are compiled. Each time series comes from the different OGGM runs using different GCMs (the number of GCMs varies between the filtered and complete ensembles in each glaciological region); ii) Then, the time series were averaged, getting one time series of glacier runoff per glacier; iii) these were summed annually by catchment, allowing us to obtain the annual glacier runoff in each catchment between 2000\u0026ndash;2099; iv) The glacier runoff was smoothed using a moving average comprising 11 years; v) On these smoothed time-series we selected the period of 20 years related to maximum glacier runoff; vi) Finally, the PW year corresponds to the median of these 20 years with a fixed uncertainty range of +/- 10 years.\u003c/p\u003e \u003c/div\u003e"},{"header":"Declarations","content":"\u003cp\u003e \u003ch2\u003eCompeting interests\u003c/h2\u003e \u003cp\u003eThe author(s) declare no competing interests.\u003c/p\u003e \u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eAC, TC, and AR were involved in the study design. AC wrote the model implementation and produced the figures, tables, and first draft of the manuscript. RA and NC contributed to the model implementation. AC performed the first level of analysis, which was improved by input from all authors. All authors contributed to the review and editing of the paper.\u003c/p\u003e\u003ch2\u003eAcknowledgement\u003c/h2\u003e\u003cp\u003eWe acknowledge LabEx OSUG@2020 (Investissement d'Avenir, ANR10 LABX56). AC thanks the National Agency for Research and Development (ANID)/Scholarship Program/DOCTORADO BECAS CHILE/2019-72200174. RA was supported by the European Research Council (ERC) under the European Union's Horizon Framework research and innovation programme (grant agreement N\u0026ordm;101115565; ICE3 project).\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eAll data analyzed in this article are available by glacier and catchment at https://zenodo.org/records/12714725. Other data such as the catchment outlines and their identifier (id) are available at https://doi.org/10.5281/zenodo.7890462, related to the article \u0026ldquo;Hydrological response of Andean catchments to recent glacier mass loss\u0026rdquo; https://doi.org/10.5194/tc-18-2487-2024.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eAbatzoglou, J. T., Dobrowski, S. Z., Parks, S. A., \u0026amp; Hegewisch, K. C. TerraClimate, a High-Resolution Global Dataset of Monthly Climate and Climatic Water Balance from 1958-2015. Sci. Data. 5, 1-12.; 10.1038/sdata.2017.191 (2018).\u003c/li\u003e\n\u003cli\u003eAgudelo, A. \u003cem\u003eet al\u003c/em\u003e. Future projections of low-level atmospheric circulation patterns over South Tropical South America: Impacts on precipitation and Amazon dry season length. Journal of Geophysical Research: Atmospheres.128, e2023JD038658; 10.1029/2023JD038658 (2023).\u003c/li\u003e\n\u003cli\u003eAlmazroui, M. \u003cem\u003eet al\u003c/em\u003e. Assessment of CMIP6 Performance and Projected Temperature and Precipitation Changes Over South America. Earth Syst Environ. 5, 155\u0026ndash;183; 10.1007/s41748-021-00233-6 (2021).\u003c/li\u003e\n\u003cli\u003eAschwanden, A., A\u0026eth;algeirsd\u0026oacute;ttir, G., \u0026amp; Khroulev, C. Hindcasting to measure ice sheet model sensitivity to initial states. The Cryosphere. 7; 10.5194/tc-7-1083-2013 (2013).\u003c/li\u003e\n\u003cli\u003eAutin, P., Sicart, J. E., Rabatel, A., Soruco, A., \u0026amp; Hock, R. Climate Controls on the Interseasonal and Interannual Variability of the Surface Mass and Energy Balances of a Tropical Glacier (Zongo Glacier, Bolivia, 16\u0026deg; S): New Insights From the Multi‐Year Application of a Distributed Energy Balance Model. Journal of Geophysical Research: Atmospheres. 127(7); 10.1029/2021JD035410 (2022).\u003c/li\u003e\n\u003cli\u003eBraun, L. N. \u0026amp; Renner, C. B. Application of a conceptual runoff model in different physiographic regions of Switzerland. Hydrol. Sci. J. 37(3), 217-231; 10.1080/02626669209492583 (1992).\u003c/li\u003e\n\u003cli\u003eBravo, C. \u003cem\u003eet al\u003c/em\u003e. Projected increases in surface melt and ice loss for the Northern and Southern Patagonian Icefields. Sci Rep. 11, 16847; 10.1038/s41598-021-95725-w (2021).\u003c/li\u003e\n\u003cli\u003eCaro, A, Condom, T. \u0026amp; Rabatel, A. Climatic and Morphometric Explanatory Variables of Glacier Changes in the Andes (8\u0026ndash;55\u0026deg;S): New Insights From Machine Learning Approaches. Front. Earth Sci.; 10.3389/feart.2021.713011 (2021).\u003c/li\u003e\n\u003cli\u003eCaro, A. \u003cem\u003eet al\u003c/em\u003e. Hydrological Response of Andean Catchments to Recent Glacier Mass Loss. The Cryosphere; 10.5194/tc-18-2487-2024 (2024).\u003c/li\u003e\n\u003cli\u003eCrippen, R. \u003cem\u003eet al\u003c/em\u003e. NASADEM Global Elevation Model: Methods and Progress. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences XLI-B4. 10.5194/isprs-archives-XLI-B4-125-2016 (2016).\u003c/li\u003e\n\u003cli\u003eCuffey, K. \u0026amp; Paterson, W. The Physics of Glaciers (Academic Press) 4th edn. (2010).\u003c/li\u003e\n\u003cli\u003eDRIAS. DRIAS les futurs du climat (http://drias-climat.fr (2023).\u003c/li\u003e\n\u003cli\u003eEyring, V. \u003cem\u003eet al\u003c/em\u003e. Overview of the Coupled Model Intercomparison Project Phase 6 (CMIP6) experimental design and organization, Geosci. Model Dev. 9, 1937-1958; 10.5194/gmd-9-1937-2016 (2016).\u003c/li\u003e\n\u003cli\u003eFarinotti, D. \u003cem\u003eet al\u003c/em\u003e. A consensus estimate for the ice thickness distribution of all glaciers on Earth. Nat. Geosci. 12, 168-173; 10.1038/s41561-019-0300-3 (2019).\u003c/li\u003e\n\u003cli\u003eFrans, C. \u003cem\u003eet al\u003c/em\u003e. Predicting glacio-hydrologic change in the headwaters of the Zongo River, Cordillera Real, Bolivia. Water Resour. Res. 51, 9029-9052; 10.1002/2014WR016728 (2015).\u003c/li\u003e\n\u003cli\u003eHausfather, Z., Marvel, K., Schmidt, G. A., Nielsen-Gammon, J. W., \u0026amp; Zelinka, M. Climate simulations: recognize the \u0026lsquo;hot model\u0026rsquo; problem. Nature. 605, 26-29; 10.1038/d41586-022-01192-2 (2022).\u003c/li\u003e\n\u003cli\u003eHock, R. Temperature index melt modelling in mountain areas. Journal of Hydrology. 282(1-4), 104-115; 10.1016/S0022-1694(03)00257-9 (2003).\u003c/li\u003e\n\u003cli\u003eHugonnet, R. \u003cem\u003eet al\u003c/em\u003e. Accelerated global glacier mass loss in the early twenty-first century. Nature. 592, 726-731; 10.1038/s41586-021-03436-z (2021).\u003c/li\u003e\n\u003cli\u003eHuss, M., \u0026amp; Hock, R. A new model for global glacier change and sea-level rise. Front. Earth Sci. 3:54; 10.3389/feart.2015.00054 (2015).\u003c/li\u003e\n\u003cli\u003eHuss, M., \u0026amp; Hock, R. Global-scale hydrological response to future glacier mass loss. Nat. Clim. Change. 8, 135-140; 10.1038/s41558-017-0049-x (2018).\u003c/li\u003e\n\u003cli\u003eIPCC AR6. IPCC WGI Interactive Atlas: Regional information (Advanced). https://interactive-atlas.ipcc.ch/ (2022).\u003c/li\u003e\n\u003cli\u003eJouvet, G. \u003cem\u003eet al\u003c/em\u003e. Deep learning speeds up ice flow modelling by several orders of magnitude. Journal of Glaciology. 68(270), 651\u0026ndash;664; 10.1017/jog.2021.120 (2022).\u003c/li\u003e\n\u003cli\u003eKienholz, C., Rich, J. L., Arendt, A. A., \u0026amp; Hock, R. A new method for deriving glacier centerlines applied to glaciers in Alaska and northwest Canada. The Cryosphere. 8, 503\u0026ndash;519; 10.5194/tc-8-503-2014 (2014).\u003c/li\u003e\n\u003cli\u003eLaha, S., Banerjee, A., Singh, A., Sharma, P., \u0026amp; Thamban, M. The control of climate sensitivity on variability and change of summer runoff from two glacierised Himalayan catchments preprint. Hydrology and Earth System Sciences Discussions; 10.5194/hess-2021-499 (2021).\u003c/li\u003e\n\u003cli\u003eLe Meur, E., Gagliardini, O., Zwinger, T. \u0026amp; Ruokolainen, J. Glacier flow modelling: a comparison of the shallow ice approximation and the Full-Stokes solution. Comptes Rendus Physique. 5(7), 709-722; 10.1016/j.crhy.2004.10.001 (2004).\u003c/li\u003e\n\u003cli\u003eMarzeion, B. \u003cem\u003eet al\u003c/em\u003e. Partitioning the Uncertainty of Ensemble Projections of Global Glacier Mass Change. Earth\u0026rsquo;s Future. 8; 10.1029/2019EF001470 (2020).\u003c/li\u003e\n\u003cli\u003eMarzeion, B., Jarosch, A. H., \u0026amp; Hofer, M. Past and future sea-level change from the surface mass balance of glaciers. The Cryosphere. 6, 1295-1322; 10.5194/tc-6-1295-2012 (2012).\u003c/li\u003e\n\u003cli\u003eMasiokas, M. H. \u003cem\u003eet al\u003c/em\u003e. Current state and recent changes of the cryosphere in the Andes. Front Earth Sci. 8:99; 10.3389/feart.2020.00099 (2020).\u003c/li\u003e\n\u003cli\u003eMaussion, F. \u003cem\u003eet al\u003c/em\u003e. The open global glacier model (OGGM) v1.1. Geoscientific Model. Develop. 12, 909-931; 10.5194/gmd-12-909-2019 (2019).\u003c/li\u003e\n\u003cli\u003eMcSweeney, C. F., Jones, R. G., Lee, R. W., \u0026amp; Rowell, D. Selecting CMIP5 GCMs for downscaling over multiple regions. Climate Dynamics. 44(11-12), 3237\u0026ndash;3260; 10.1007/s00382-014-2418-8 (2015).\u003c/li\u003e\n\u003cli\u003eMillan, R. \u003cem\u003eet al\u003c/em\u003e. Ice velocity and thickness of the world\u0026rsquo;s glaciers. Nat. Geosci. 15, 124-129; 10.1038/s41561-021-00885-z (2022).\u003c/li\u003e\n\u003cli\u003eMinowa, M., Schaefer, M., Sugiyama, S., Sakakibara, D., \u0026amp; Skvarca, P. Frontal ablation and mass loss of the Patagonian icefields. Earth Planet. Sc. Lett. 561, 116811; 10.1016/j.epsl.2021.116811 (2021).\u003c/li\u003e\n\u003cli\u003eOerlemans, J. Minimal Glacier Models (Utrecht Publishing and Archiving Services) (2008).\u003c/li\u003e\n\u003cli\u003eOlmo, M. E. \u003cem\u003eet al\u003c/em\u003e. Circulation patterns and associated rainfall over south tropical South America: GCMs evaluation during the dry-to-wet transition season. J. Geophys. Res. Atmos. 127; 10.1029/2022JD036468 (2022).\u003c/li\u003e\n\u003cli\u003eRabatel, A. \u003cem\u003eet al.\u003c/em\u003e Current state of glaciers in the tropical Andes: a multi-century perspective on glacier evolution and climate change. \u003cem\u003eThe Cryosphere\u003c/em\u003e, \u003cem\u003e7\u003c/em\u003e(1), 81-102; 10.5194/tc-7-81-2013 (2013).\u003c/li\u003e\n\u003cli\u003eRabatel, A. \u003cem\u003eet al\u003c/em\u003e. Toward an imminent extinction of Colombian glaciers?. Geografiska Annaler: Series A. Phys. Geogr. 100, 75\u0026ndash;95; 10.1080/04353676.2017.1383015 (2018).\u003c/li\u003e\n\u003cli\u003eRadic, V., \u0026amp; Hock, R. Glaciers in the Earth\u0026rsquo;s hydrological cycle. Assessments of glacier mass and runoff changes on global and regional scales. Surv. Geophys. 35, 813-837; 10.1007/s10712-013-9262-y (2014).\u003c/li\u003e\n\u003cli\u003eR\u0026eacute;veillet, M., Rabatel, A., Gillet-Chaulet, F., \u0026amp; Soruco, A. Simulations of changes to Glacier Zongo, Bolivia (16S), over the 21st century using a 3-D full-Stokes model and CMIP5 climate projections. Ann. Glaciol. 56, 89-97; 10.3189/2015aog70a113 (2015).\u003c/li\u003e\n\u003cli\u003eRGI Consortium. Randolph Glacier Inventory. A Dataset of Global Glacier Outlines, Version 6. NSIDC: National Snow and Ice Data Center. 10.7265/4m1f-gd79 (2017).\u003c/li\u003e\n\u003cli\u003eRivera, J.A. \u0026amp; Arnould, G. Evaluation of the ability of CMIP6 models to simulate precipitation over Southwestern South America: climatic features and long-term trends (1901\u0026ndash;2014). Atmos Res. 241, 104953; 10.1016/j.atmosres.2020.104953 (2020).\u003c/li\u003e\n\u003cli\u003eRounce, D. \u003cem\u003eet al\u003c/em\u003e. Global glacier change in the 21st century: Every increase in temperature matters. Science. 379, 78-83; 10.1126/science.abo1324 (2023).\u003c/li\u003e\n\u003cli\u003eSchaefer, M., MacHguth, H., Falvey, M., \u0026amp; Casassa, G. Modeling past and future surface mass balance of the Northern Patagonia icefield. J. Geophys. Res. Earth Surf. 118, 571-588; 10.1002/jgrf.20038 (2013).\u003c/li\u003e\n\u003cli\u003eScheiter, M., Schaefer, M., Fl\u0026aacute;ndez, E., Bozkurt, D., \u0026amp; Greve, R. The 21st-century fate of the Mocho-Choshuenco ice cap in southern Chile. The Cryosphere. 15, 3637-3654; 10.5194/tc-15-3637-2021 (2021).\u003c/li\u003e\n\u003cli\u003eS\u0026oslash;rland, S. L., Sch\u0026auml;r, C., L\u0026uuml;thi, D. \u0026amp; Kjellstr\u0026ouml;m, E. Bias patterns and climate change signals in GCM-RCM model chains. Environ. Res. Lett. 13, 074017; 10.1088/1748-9326/aacc77 (2018).\u003c/li\u003e\n\u003cli\u003eTokarska, K. B. \u003cem\u003eet al\u003c/em\u003e. Past warming trend constrains future warming in CMIP6 models, Science Advances. 6, eaaz9549; 10.1126/sciadv.aaz9549 (2020).\u003c/li\u003e\n\u003cli\u003eVuille, M. \u003cem\u003eet al\u003c/em\u003e. Rapid decline of snow and ice in the tropical Andes\u0026ndash;Impacts, uncertainties and challenges ahead. Earth Sci. Rev. 176, 195-213; 10.1016/j.earscirev.2017.09.019 (2018).\u003c/li\u003e\n\u003cli\u003eYarleque, C. \u003cem\u003eet al\u003c/em\u003e. Projections of future disappearance of the Quelccaya, the largest tropical ice cap on Earth. Nat. Sci. Rep. 8, 15564; 10.1038/s41598-018-33698-z (2018).\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-4714636/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4714636/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eAssessing future glacier water contributions is crucial for resource management. However, a large gap persists regarding Andean glacier runoff research. We evaluated eight CMIP6 models (1990-2049) on a glacierized area of 27,669 km², and projected future changes in glacier runoff (on 11,282 km² related to land-terminating glaciers) using an evaluated glaciological model in 778 catchments under two extreme SSP scenarios to estimate the year of maximum glacier runoff (peak water). We evaluated the different GCMs over the historical period using in situ data and showed that some of them perform better in specific glaciological regions. For the mid-21st century, warming trends are projected across the Andes, especially in the Tropical Andes (+0.7°C), while precipitation is expected to decrease slightly in the Southern Andes (-1 to -3%). These variables significantly affect glacier dynamics and runoff estimates. Glacier runoff estimates spanning 2000-2019 and projected to 2030-2049 indicate significant declines in the Tropical Andes (-43%) and Dry Andes (-37%), and a lesser decrease in the Wet Andes (-32%). Notably, the Atuel (-62%) and Tupungato (+32%) catchments in the Dry Andes show highly contrasted changes in annual runoff across the Andes. In terms of peak water, most catchments are expected to reach it before the first half of the 21st century (between 2010 and 2028). Our study underlines the critical importance of considering seasonal variations when analyzing GCMs in hydro-glaciological simulations and emphasizes regional disparities in glacier runoff across the Andes for future water resources management.\u003c/p\u003e","manuscriptTitle":"Glacio-hydrological changes along the Andes throughout the 21st Century","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-08-05 17:22:00","doi":"10.21203/rs.3.rs-4714636/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"editorInvitedReview","content":"","date":"2024-08-07T21:14:41+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-07-31T20:03:07+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"35235844929093867477252714841788029337","date":"2024-07-22T13:02:37+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"31392660813139967680709565770406141754","date":"2024-07-22T11:40:19+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2024-07-21T12:21:31+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2024-07-21T12:01:11+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2024-07-16T15:51:13+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2024-07-12T03:37:14+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2024-07-10T00:06:06+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"30e20f37-2538-41d0-995d-b1feffcdad49","owner":[],"postedDate":"August 5th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":35446570,"name":"Earth and environmental sciences/Climate sciences"},{"id":35446571,"name":"Earth and environmental sciences/Hydrology"}],"tags":[],"updatedAt":"2025-04-07T16:01:43+00:00","versionOfRecord":{"articleIdentity":"rs-4714636","link":"https://doi.org/10.1038/s41598-025-88069-2","journal":{"identity":"scientific-reports","isVorOnly":false,"title":"Scientific Reports"},"publishedOn":"2025-03-31 15:57:05","publishedOnDateReadable":"March 31st, 2025"},"versionCreatedAt":"2024-08-05 17:22:00","video":"","vorDoi":"10.1038/s41598-025-88069-2","vorDoiUrl":"https://doi.org/10.1038/s41598-025-88069-2","workflowStages":[]},"version":"v1","identity":"rs-4714636","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4714636","identity":"rs-4714636","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}