ABSTRACT Glaciers are thinning and retreating as climate warms, thus eroding less of the Earth’s surface. However, other hydrological factors in glacierized catchments are likely producing a

transient increase in sediment delivery, resulting in ‘peak sediment’. Estimating the trajectory of the peak sediment is ecologically and socially important but scientifically challenging

because of the delayed and non-linear response of glacier sediment export to climate forcing. This study used time series of suspended sediment concentration starting in the 1960s from 11

Andean rivers at subtropical latitudes to analyse past changes in sediment export and infer its future behaviour. The recent decade has experienced anomalously high sediment concentration in

of the glacierized catchments, the peak sediment generated by the anthropogenic deglaciation started two centuries ago has already passed its maximum. SIMILAR CONTENT BEING VIEWED BY OTHERS

SEDIMENT DISCHARGE FROM GREENLAND’S MARINE-TERMINATING GLACIERS IS LINKED WITH SURFACE MELT Article Open access 13 February 2024 RECENT INTENSIFIED EROSION AND MASSIVE SEDIMENT DEPOSITION IN

TIBETAN PLATEAU RIVERS Article Open access 24 January 2024 CHANGES IN GLOBAL FLUVIAL SEDIMENT CONCENTRATIONS AND FLUXES BETWEEN 1985 AND 2020 Article 02 January 2025 INTRODUCTION Mountain

glaciers have a great erosive capacity, contributing to topographic relief of the Earth1. Once glaciers produce sediment and solutes, these are displaced by the ice itself or by wind and

water to lowlands, where they play a fundamental role in landform evolution, ecosystems and human activities such as water management and hydropower generation2,3. For example, in the

subtropical Andes, decadal-scale increments in sediment export have had negative consequences on reservoir capacity and drinking water supply4,5,6 and a potentially positive impact on

coastal ecology due to greater availability of nutrients7. Climate warming is expected to decrease the erosive capacity of glaciers as they thin and retreat due to the reduced gravitational

driving stress and sliding speeds, thus reducing sediment production8,9,10,11. However, glaciers’ sediment export may increase after deglaciation begins due to the upstream extension of the

subglacial hydrological system that connects long-term stored sediment4,10,12 (Fig. 1). Similar to the concept of a transient increase in glacier runoff, known as ‘peak water’13, this

transient increase in sediment export was termed as ‘peak sediment’14,15. Peak sediment depends mainly on climate variability, glacier thermal regime, bed topography, and the mass and

distribution of stored sediment. For a typical valley glacier, it can last for more than a century and extend beyond the peak water10,16. In line with the theory of peak sediment14, the

global, present-day deglaciation, which started approximately in the 19th century due to anthropogenic warming17, is driving an increase in sediment export from several glacierized regions,

such as the Andes, European Alps, Artic, and Himalayas18,19,20. Similarly, climate warming is also expected to generate a transient rise in sediment production in the periglacial environment

and a transition in the sediment transport regime of mountain rivers from thermal to pluvial3,21. Despite knowledge of glacier and hydrology’s evolution in response to climate warming,

evaluating the timing of peak sediment remains difficult because glaciers’ thermal regime, stored subglacial sediment, and hydro-sedimentological connectivity are usually unknown.

Notwithstanding these difficulties, peak sediment is projected to end between 2100 and 2200 globally15. In contrast, another study suggests that two alpine catchments are currently

experiencing their maximum and will end between 2070 and 210022. Here, six-decade-long sediment discharge series of eleven subtropical Andean rivers between 27 and 35 °S were used to shed

light on future glacier sediment export (Fig. 2a and Table S1). In particular, the evolving relationship between water and sediment released by glaciers was evaluated from the 1960s onwards.

This relationship yields insights into the temporal evolution of glacial sediment availability in response to changing glaciological, climatic and hydrology forcing. The rivers’ catchments

contain variable ice volume, but, without exception, gauges are between a few tens and hundred kilometres away from glaciers’ terminus. Gauges measure suspended sediment concentration and

mass in the stream. We focus on the suspended sediment concentration (SSC, mg l−1) because the sediment mass transported by glacial meltwater in the region is comparable in magnitude to that

generated by snowmelt outside glaciers, but usually has a higher solid:liquid ratio as it circulates over the sediment-rich glacial bed23. In addition, snowmelt events produce more water

and stream power than ice melt events24, so they have more sediment from the bed and banks of the mainstem. SUBTROPICAL ANDEAN GLACIERS This high terrain has glaciers with both cold and

polythermal regimes, including valley glaciers of up to 24 km2 and surface velocities of up to ~300 m yr−1 in the 2010s25,26,27. In the glacierized Maipo River basin, located in the

central-southern sector of the study area (Fig. 2a), there was a sharp increase in the mean and extreme turbidity during the warm and dry seasons of the last decade4. The positive

interannual correlation of turbidity with ice melt, insignificant with rainfall, and negative with snowmelt outside or over glaciers allowed to link this increase to changes in the

connectivity of the subglacial and proglacial areas, as these are the sources of sediment whose transport would be favoured by an increase in glacial streams28. This interpretation was also

based on (a) the low land use change4, (b) the absence of GLOFs in the catchment since 195429, and (c) the low frequency of slope instability associated with permafrost degradation and

glacial retreat30. RESULTS REGIME DETECTION By applying a sequential “regime shift” detection method31 to the annual series of the three glacierized catchments with the most complete

records, we identified three regimes or periods with specific SSC values (Methods). A high SSC regime is observed from the late 1960s until the mid-1970s (I), followed by a low SCC regime

from the late 1970s to the 2000s (II), and then a new high SSC regime in the 2010s (III) (Fig. 3a–c). Interestingly, SSC values during regime III are about half of those of the first period

in the 1970s. When including the remaining catchments in the analysis, it is observed that the higher magnitude of SSC during the first period occurs only in glacierized catchments (Fig. 

2b). The change between regimes II and III corresponds to the sharp increase in turbidity in the Maipo River. Although regime I was not previously analysed, judging from its below-average

precipitation, like regime III32 (Fig. 3d), and the pattern shown in the scatterplot of Fig. 2b, it is likely to have a cryospheric origin, similar to regime III. However, despite its

possible common origin with the current high SSC regime, regime I has a higher magnitude which could be due to three potential temporal changes: a) exhaustion of sediment stored, b)

diminished meltwater accessibility to sediment along the glaciers’ bed because of lesser development of glacio-hydrological morphologies like moulins and supra- and sub- glacial channels,

and c) decreased glacier water discharge and associated sediment transport capacity16 because of diminished glacier area or climatic forcing (i.e., lower temperature and higher snowfall

inhibiting subglacial runoff). Identifying which of the three processes, or their combination, determines the current state of glacial sediment transport can provide clues about future

sediment flux from these catchments. For example, if the lower SSC magnitude of the current regime compared to regime I was caused by glacial sediment exhaustion or a reduction in meltwater

discharge due to the loss of glacier area, sediment export would be expected to decrease in the future. On the contrary, if the lower SSC magnitude of the current regime was because climatic

forcing or meltwater accessibility were lower compared to regime I, sediment export could increase if these variables also do so. ANALYSIS OF THE REGIMES To understand SSC changes, we first

examine the magnitude and causes of variations in glacier runoff using annual estimates of ice melt and glacier area in the Maipo River basin. This is the only basin for which these outputs

are available and it has similar topographic and climatic characteristics to the other basins in the south of the study area (Fig. 3c; Table S1). Both data come from glacio-hydrological

simulations using the TOPKAPI-ETH model, which was run using geodetic glacier mass balance, remotely sensed snow cover area, local hydro-meteorological measurements, and an independent daily

gridded snow water equivalent reconstruction24. This model was validated in the region against streamflow and snow cover area measurements, with satisfactory results, especially for annual

integrations of the outputs such as those used here (24; Methods). For each SSC regime, annual averages of total ice melt and ice melt normalized per glacierized area were calculated. The

total ice melt indicates the magnitude of glacial streamflow, whereas the specific ice melt expresses the climatic forcing strength and, indirectly, the connection between the atmosphere and

subglacial system because it increases the development of glacio-hydrological morphologies33. When comparing the three regimes, the two high SSC periods experienced greater total and

specific ice melt compared to regime II (Fig. 4). Yet, regime I had greater total melt and lower melt per unit area than regime III (Fig. 4). Although these differences are not significant

at a 95% level, they are consistent with the larger glacierized area and the slightly colder and wetter climate during regime I in the 1970s24,34 (Fig. 3d). We assessed the role of sediment

availability in the SSC changes by examining temporal variations in the interannual relationship between ice melt and the number of extreme turbidity events (ETE). The ETE were defined as

days with SSC values above the 85th quantile during the warm and dry season (Methods). We used ice melt as a proxy for proglacial streamflow, and the number of ETE was used as a proxy for

its SSC because glacial sediment evacuation is better detected in the right tail of the SSC probability distribution at gauges far from glaciers4,35. We only examined the frequency of the

ETE, i.e., the number per year, to avoid bias caused by rainstorms, which produce SSC magnitudes much higher than purely thermal melting4,36. The ETE were calculated for the Aconcagua River

basin, the glacierized catchment with the fewest measurement gaps, ensuring that results are independent of missing data (Fig. 3a; Table S1). This basin is adjacent to the Maipo River basin,

where the ice melt data were calculated, and both basins have a high correlation in ice melt and sediment delivery (the number ETE calculated for the Maipo River basin were also tested;

Methods). The relationship between ice melt and ETE frequency was evaluated through the coefficient of correlation (_r_) and the intercept and slope of their linear regression considering

ice melt as a covariate. The _r_ quantifies the relationship’s strength, while the slope expresses the land erodibility, the shape and gradient of the channel, and the extent to which new

sediment sources become available as streamflow rises37,38,39. The intercept indirectly indicates sediment availability because it is the minimum SSC value that streamflow can generate in

the given context and is independent of its magnitude and increase39. The correlation _r_ between ice melt and ETE during the two high SSC regimes (I and III) was positive, similar in

magnitude and significantly different from the negative correlation during regime II (Fig. 5). The negative correlation signals that the ETE of that period were triggered by precipitation

and snowmelt, and hence during years when a thick seasonal snowpack protected ice from summer temperatures and hindered the percolation of meltwater into glacial bed through moulins and

crevasses40,41. The two high SSC regimes also have similar positive slopes in the relationship between ETE and ice melt, which are significantly steeper than those negative of the low SSC

regime (Fig. 5). The higher slope indicates that ETE varied more strongly with ice melt over the high SSC regimes I and III compared to the low SSC regime II. Conversely, the intercept -a

proxy of sediment availability when _r_ is positive- was significantly higher during regime I and then it gradually decreased until the current regime when it slightly recovered (Fig. 5).

The ongoing deglaciation and the higher specific ice melt of the current regime raise doubts that a lower accessibility of meltwater along glacial bed has caused this decrease. In turn, the

likely lower magnitude of the glacial streamflow should not have modified the intercept of the relationship (Fig. 4). In this sense, the lower current sediment availability compared to

regime I could be caused by exhausted glacial sediment. Combining this result with the fact that regional glacial water flux is decreasing24, it is inferred that the peak sediment has

largely passed its maximum and, therefore, glacier sediment flux will slowly decrease in the future. Although, as mentioned above, sediment mass flux was not analysed in detail, its highest

values during the 1970s and 1980s support this outcome (Fig. S3). DISCUSSION A basin-scale relationship was detected between the long-term decrease of SSC and glacier volume (Fig. 2b).

However, the northern Pulido River basin experienced an increase despite being noticeably glacierized (Fig. 2b). In comparison to the others that had decreased, this basin contains glaciers

at a higher elevation and consequently colder and possibly with less meltwater in their bed (Table S1). Local studies corroborate that high-elevation northern glaciers have mostly cold

thermal regimes26,27. In this sense, the glaciers of this basin may have been insensitive to the favourable conditions for increased sediment discharge of the 1970s. Therefore, in this

basin, as in the others of its type, the peak sediment may not have passed its maximum yet (Fig. 2a). It must also be noted that, for basins with a preponderance of cold-based glaciers, peak

sediment could be lower due to the possible less stored sediment associated with reduced sliding velocities as a result of decreased meltwater volumes42. The consistent findings across the

catchments may suggest a regional pattern in sediment export. Considering unmonitored basins with a glacier volume greater than 2 × 10−4 km3 km−2 (Fig. 2b) and an average glacial elevation

less than that of the Pulido River basin, i.e., 5 km ASL, eight more basins of the size of those analysed may have had decreases in glacial sediment flux (Fig. 2a). If basins with higher

glacial elevations are added to the analysis, it turns out that possibly most of the subtropical Andean rivers are experiencing or will experience during this century the maximum of the peak

sediment (Fig. 2a). The ice melt-ETE relationship shows a pronounced multidecadal variability, which would be due to the effects of regional climate and the memory of the

glacio-hydrological system on the hydraulic connectivity of glacial sediment (Fig. 5). On the one hand, the natural decadal variability of the Pacific Ocean’s climate, known as the Pacific

Decadal Oscillation, results in periods of below (above) average precipitation34,43 (Fig. S4) and, therefore, glaciers with less (more) snow accumulation and a subglacial realm more (less)

connected to the atmosphere40. Nevertheless, years of high and low precipitation also occur during decades in which the opposite climatic situation predominates, suggesting that the

variability of the ice melt-ETE relationship could also be caused by the multiannual memory of the glacio-hydrological system. For example, creating a moulin in the glacier’s upper reaches

would entail several years of negative mass balance in which each ablation season would begin from a point more advanced than the previous one44,45,46. In this context, a positive mass

balance year might not close its roof with snow bridges that limit the access of meltwater to the glacier bed47. In the same way, during a period of positive and neutral mass balance years,

a negative one could be insufficient for a new moulin to reach the glacier bed. We propose that the observed changes in sediment export were mainly generated in the subglacial and proglacial

environments. However, secondary sediment sources may include: a) the degrading permafrost48, b) the occasional paraglacial landslides in the valley walls favoured by glacial unloading30,

and c) the GLOF of the early 1980s in the Cachapoal River basin (Table S1; 29), the only one in the study area since the 1960s. CONCLUDING REMARKS Series of fluvial sediment concentration,

unique in the world for their length and glacial information, were analysed to advance knowledge of a recently discovered cryospheric process called ‘peak sediment’. This process is caused

by global warming of recent centuries and has an important impact on hydropower generation, drinking water supply and river and coastal ecosystems. The results suggest that the peak sediment

has passed its maximum for polythermal glaciers in the subtropical Andes but has not yet done so for colder glaciers at higher elevations. It was also observed that in addition to long-term

trends, the sediment concentration of rivers has a substantial multidecadal variability controlled by regional climate and the evolution of glacio-hydrologic morphologies. Therefore, the

shape and timing of the peak sediment strongly depend on topography, climate and hydro-sedimentological connectivity, a fact that should be considered when inferring the state of other

regions on the Earth. METHODS RIVER SEDIMENT DATA Daily measurements of suspended sediment concentration (SSC; mg l−1) derived from water samples at the Chilean gauging stations were

obtained from the database of the Chilean Water Directorate (DGA; http://www.dga.cl/). SSC was sampled at the river surface and at approximately the same point using a bottle. The routine

sampling procedure was the same for all the operators of the gauging stations49. The samples were filtered, ignited to remove organic matter, and weighed at the DGA laboratory49. To

calculate the suspended sediment yield (SSY) of Fig. S3, the SSC measurements were multiplied by the concurrent streamflow measurements made by the DGA. Once the data were downloaded, the

gauges with 50% or more daily data in the period 1964–2017 were selected. The period was defined based on the hydrological annual cycle, which goes from April to March in the region. The

_Río Illapel en las Burras_ gauge was discarded despite having sufficient data due to its abnormal values possibly related to an error in field measurement, laboratory processing or

measurement transcription (Sup. Material of Vergara et al.23). For gaps of one day, data was filled with the average between the previous and following days. Following, monthly averages were

calculated for those months with more than ten days of measurements. If the number of measurements was less, the average between the values of the same month of the previous and subsequent

years was used, and failing that, the average between the last and following months of the same year. For year-round averages, years with less than ten months with more than ten daily

measurements and at least one missing month that could not be filled by months of nearby hydrological years or months of the same year were discarded. For averages of the warm season, i.e.,

from October to March, those seasons with less than four months with more than ten measurements and at least one missing month that could not be filled with the other two methods were

discarded. Emphasis was placed on this season because it is when the subglacial system is activated. In addition, sediment mass flux during the remaining months is much lower (Fig. 1c of

Vergara et al.4). To compare the averages of the periods 1964–1981 and 2000–2017 (Fig. 2b & S1), only gauges with at least six years with data in each period were used. This resulted in

9 gauges being used for the whole year and 11 for the warm season. All gauges are located between 27 and 35 °S (Table S1). ICE MELT DATA To infer streamflow at the glacier snout, daily ice

melt modelled for all glaciers in the Maipo River basin for the period 1955–2020 was used24 (Table S1). This estimate was produced by the TOPKAPI-ETH model, a spatially distributed,

physically oriented glacio-hydrological model suitable for calculating glacier mass balance and snow cover evolution in mountainous catchments. The model includes parameterizations of the

most important hydrological processes at high altitudes, such as snow accumulation, albedo decay, snow gravitational redistribution, ice and snow melt, and ice melt under debris. Ayala et

al.24 applied numerous instances of the model to the study catchment, with spatial resolutions ranging from 100 m for glaciers (focusing on ice melt contribution and glacier mass balance) to

1 km for the whole catchment (focusing on seasonal snow cover). At each glacierized grid cell, the snow remaining at the end of each hydrological year is converted to ice. The main outputs

of the model are the annual time series of glacier volume and area and the daily time series of rainfall, snowmelt and ice melt, which are further split into the fraction coming from

glaciers and the one that comes from outside. For more information about the model applied, consult Ayala et al.24. CALCULATION OF THE REGIME SHIFTS For each series of Fig. 3, S2, S3 and

S4a, temporal changes were assessed using the Sequential Regime Shift Detection method31. This technique is based on the calculation of a regime shift index combined with a sequential

application of Student’s _t_-tests to determine the significance and timing of regime shifts. Three parameters must be set to detect the regimes: the target significance level, the cut-off

length and the outlier weighting factor (_h_). The significance level of the difference between mean values of neighbouring regimes is based on Student’s two-tailed _t_-tests with unequal

variance31. Regimes longer than the cut-off length will all be identified. Shorter regimes can also pass the test if the differences are large enough, but their detection depends on the

selected Cut-off length and Significance level. The use of _h_ allows for the detection of regimes unaffected by the influence of a single outlier with a disproportionate effect. Values

outside ± _h_ standard deviations will be weighted inversely proportional to their distance from the average of their corresponding regime. We used a target significance level of 0.1, a

cut-off length of 15 years considering the series lengths, and an _h_ of 4, considering the series’ high natural variability. Trials with different combinations of parameter values gave very

similar results. ETE CALCULATION AND THEIR RELATIONSHIP TO ICE MELT The number of ETE were calculated for the warm season and the Aconcagua River basin (Fig. S4a), as the other highly

glacierized basins have relatively few measurements (Table S1). Furthermore, its SSC data had the highest correlation with the reference turbidity series of the region, which has hourly

resolution, has no gaps, and measures a fine grain size that is little influenced by stream power and more sensitive to hillslope and cryospheric processes (Fig. S6; Section 2.1 of Vergara

et al.4). All this suggests that Aconcagua’s temporal series best reflects how glacial sediment flux fluctuated and gives confidence about the results. Note that turbidity series could not

be used to study long-term changes because it began operating in 1990 (Fig. 3a). A trial was performed calculating the ETE for the Maipo River basin, which gave similar results, especially

for the intercept (Fig. S7). To construct the ETE series, only warm seasons with at least 70% of daily data were used. In order to verify that rain played a minor role, a series of

melt-driven ETE was generated, excluding days with precipitation, which gave similar results (Fig. S8 & Table S2). These account for 91% of the total. Although the scatterplot between

ice melt and ETE suggests that the assumptions for a linear regression are not significantly violated (Fig. S9), the relationship was also analysed using a power law, which gave analogous

results (Fig. S10). Although the ETE and ice melt series correspond to different basins, it is important to note that the basins are adjacent and have similar hypsometries and glacier

properties (Table S1). In fact, annual ice melt values of the Maipo basin glaciers and of the largest, most erosive glacier in the Aconcagua basin -the Juncal Norte- are highly correlated

(R2 = 0.86; Fig. S5; Sup. Note 1). Furthermore, as noted above, the SSC and turbidity of the basins also have a high interannual correlation amongst them (Fig. S6a, b). DATA AVAILABILITY All

databases used are publicly accessible. The daily SSC and streamflow records were obtained from the DGA website (http://www.dga.cl/servicioshidrometeorologicos/Paginas/default.aspx).

ERA5-Land data were downloaded from Copernicus website (https://cds.climate.copernicus.eu/datasets/reanalysis-era5-land-monthly-means?tab=overview). The glacier volume data were downloaded

from https://www.research-collection.ethz.ch/handle/20.500.11850/315707. The ice melt data are available in Vergara et al.4. The Regime Shift Detection Software was downloaded from

https://sites.google.com/view/regime-shift-test/home?authuser=0. REFERENCES * Herman, F., De Doncker, F., Delaney, I., Prasicek, G. & Koppes, M. The impact of glaciers on mountain

erosion. _Nat. Rev. Earth Environ._ 2, 422–435 (2021). Article  Google Scholar  * Li, D. et al. High Mountain Asia hydropower systems threatened by climate-driven landscape instability.

_Nature Geoscience_. https://doi.org/10.1038/s41561-022-00953-y (2022). * Zhang, T. et al. Shifted sediment-transport regimes by climate change and amplified hydrological variability in

cryosphere-fed rivers. _Sci. Adv_. https://doi.org/10.1126/sciadv.adi5019 (2023). * Vergara I., Garreaud R. & Ayala Á. Sharp increase of extreme turbidity events due to deglaciation in

the subtropical Andes. _Journal of Geophysical Research: Earth Surface_. https://doi.org/10.1029/2021JF006584 (2022). * Rivera, J. A., Marianetti, G. & Scaglione, M. Assessment of

precipitation events affecting the distribution of drinkable water in the Gran Mendoza area, Argentina. _Cuad. Geogr.áficos_ 61, 204–222 (2022). Article  Google Scholar  * Los Andes

Newspaper 2022. https://www.losandes.com.ar/sociedad/dique-potrerillos-por-que-se-limpia-el-descargador-de-fondo-y-que-ocurriria-si-no-se-hiciera/. * Masotti, I. et al. The influence of

river discharge on nutrient export and phytoplankton biomass off the Central Chile Coast (33°–37°S): Seasonal cycle and interannual variability. _Front. Mar. Sci._ 5, 1–12 (2018). *

Adhikari, S. & Marshall, S. J. Influence of high-order mechanics on simulation of glacier response to climate change: Insights from Haig Glacier, Canadian Rocky Mountains. _Cryosphere_

7, 1527–1541 (2013). Article  Google Scholar  * Dehecq, A. et al. Twenty-first century glacier slowdown driven by mass loss in High Mountain Asia. _Nat. Geosci._ 12, 22–27 (2019). Article 

CAS  Google Scholar  * Delaney, I. & Adhikari, S. Increased Subglacial Sediment Discharge in a Warming Climate: Consideration of Ice Dynamics, Glacial Erosion, and Fluvial Sediment

Transport. _Geophys. Res. Lett._ 47, 1–11 (2020). Article  Google Scholar  * Jó, V., van Wyk de Vries, M., Ignéczi, Á., Mari, L., & Nagy, B. Glacier slowdown and rapid ice loss in the

Tinguiririca and Cachapoal Basin, Central Andes of Chile. _Global and Planetary Change_, 231. https://doi.org/10.1016/j.gloplacha.2023.104287 (2023). * Schiefer, E., Hassan, M. A., Menounos,

B., Pelpola, C. P. & Slaymaker, O. Interdecadal patterns of total sediment yield from a montane catchment, southern Coast Mountains, British Columbia, Canada. _Geomorphology_ 118,

207–212 (2010). Article  Google Scholar  * Huss, M. & Hock, R. Global-scale hydrological response to future glacier mass loss. _Nat. Clim. Change_ 8, 135–140 (2018). Article  Google

Scholar  * Antoniazza, G. & Lane, S. N. Sediment yield over glacial cycles: A conceptual model. _Prog. Phys. Geogr._ 45, 842–865 (2021). Article  Google Scholar  * Zhang, T. et al.

Warming-driven erosion and sediment transport in cold regions. _Nat. Rev. Earth Environ._ https://doi.org/10.1038/s43017-022-00362-0 (2022). * Delaney, I., Anderson, L. & Herman, F.

Modeling the spatially distributed nature of subglacial sediment transport and erosion. _Earth Surf. Dyn._ 11, 663–680 (2023). Article  Google Scholar  * Masiokas, M. H. et al. Glacier

fluctuations in extratropical South America during the past 1000 years. _Palaeogeogr., Palaeoclimatol., Palaeoecol._ 281, 242–268 (2009). Article  Google Scholar  * Bendixen, M. et al. Delta

progradation in Greenland driven by increasing glacial mass loss. _Nature_ 550, 101–104 (2017). Article  CAS  Google Scholar  * Li, D. et al. Exceptional increases in fluvial sediment

fluxes in a warmer and wetter High Mountain Asia. _Science_ 374, 599–603 (2021). Article  CAS  Google Scholar  * Schmidt, L. K., Francke, T., Grosse, P. M., Mayer, C. & Bronstert, A.

Reconstructing five decades of sediment export from two glacierized high-alpine catchments in Tyrol, Austria, using nonparametric regression. _Hydrol. Earth Syst. Sci._ 27, 1841–1863 (2023).

Article  Google Scholar  * East, A. E. & Sankey, J. B. Geomorphic and sedimentary effects of modern climate change: Current and anticipated future conditions in the western United

States. _Rev. Geophysci._ 58. https://doi.org/10.1029/2019RG000692 (2020). * Schmidt, L. K., Francke, T., Grosse, P. M. & Bronstert, A. Projecting sediment export from two highly

glacierized alpine catchments under climate change: exploring non-parametric regression as an analysis tool. _Hydrol. Earth Syst. Sci._ 28, 139–161 (2024). Article  CAS  Google Scholar  *

Vergara, I., Garreaud, R., Araneo, D. & Leyton, F. Environmental control of the present-day sediment export along the extratropical Andes. _Geomorphology_ 441, 1–9 (2023). Article 

Google Scholar  * Ayala, Á. et al. Glacier runoff variations since 1955 in the Maipo River basin, semiarid Andes of central Chile. _The Cryosphere_, 1–39. https://doi.org/10.5194/tc-2019-233

(2020). * Millan, R., Mouginot, J., Rabatel, A. & Morlighem, M. Ice velocity and thickness of the world’s glaciers. _Nat. Geosci._ 15, 124–129 (2022). Article  CAS  Google Scholar  *

Gacitúa, G. et al. 50 MHz helicopter-borne radar data for determination of glacier thermal regime in the central Chilean Andes. _Ann. Glaciol._ 56, 193–201 (2015). Article  Google Scholar  *

Masiokas, M. H. et al. A Review of the Current State and Recent Changes of the Andean Cryosphere. _Front. Earth Sci._ 8. https://doi.org/10.3389/feart.2020.00099 (2020). * Lane, S. N.,

Bakker, M., Gabbud, C., Micheletti, N. & Saugy, J.-N. Sediment export, transient landscape response and catchment-scale connectivity following rapid climate warming and alpine glacier

recession. _Geomorphology_ 277, 210–227 (2017). Article  Google Scholar  * Iribarren Anacona, P., Mackintosh, A. & Norton, K. P. Hazardous processes and events from glacier and

permafrost areas: Lessons from the Chilean and Argentinean Andes. _Earth Surf. Process. Landf._ 40, 2–21 (2015). Article  Google Scholar  * Sepúlveda, S. A., Tobar, C., Rosales, V.,

Ochoa-Cornejo, F. & Lara, M. Megalandslides and deglaciation: modelling of two case studies in the Central Andes. _Nat. Hazards_ 118, 1561–1572 (2023). Article  Google Scholar  *

Rodionov, S. N. A sequential algorithm for testing climate regime shifts. _Geophys. Res. Lett._ 31, L09204 (2004). * Masiokas, M. H., Villalba, R., Luckman, B. H. & Mauget, S. Intra- to

Multidecadal Variations of Snowpack and Streamflow Records in the Andes of Chile and Argentina between 30° and 37°S. _J. Hydrometeorol._ 11, 822–831 (2010). Article  Google Scholar  *

Rawlins, L. D., Rippin, D. M., Sole, A. J., Livingstone, S. J. & Yang, K. Seasonal evolution of the supraglacial drainage network at Humboldt Glacier, northern Greenland, between 2016

and 2020. _Cryosphere_ 17, 4729–4750 (2023). Article  Google Scholar  * Garreaud, R. D. et al. The Central Chile mega drought (2010–2018): A climate dynamics perspective. _Int. J. Climatol._

40, 421–439 (2020). Article  Google Scholar  * Skålevåg, A., Korup, O. & Bronstert, A. Inferring sediment-discharge event types in an alpine catchment from sub-daily time series,

_Hydrol. Earth Syst. Sci_. [preprint], https://doi.org/10.5194/hess-2023-300 (2024). * Vergara, I., Garreaud, R., Moreiras, S., Araneo, D. & Beigt, D. Exploring the association between

landslides and fluvial suspended sediment in a semi-arid basin in central Chile. _Geomorphology_ 402, 108129 (2022). Article  Google Scholar  * Asselman, N. E. M. Fitting and interpretation

of sediment rating curves. _J. Hydrol._ 234, 228–248 (2000). Article  Google Scholar  * Yang, G. et al. Sediment rating parameters and their implications: Yangtze River, China.

_Geomorphology_ 85, 166–175 (2007). Article  Google Scholar  * Swift, D. A. et al. The hydrology of glacier-bed overdeepenings: Sediment transport mechanics, drainage system morphology, and

geomorphological implications. _Earth Surf. Process. Landf._ 46, 2264–2278 (2021). Article  Google Scholar  * Fountain, A. Effect of snow and firn hydrology on the physical and chemical

characteristics of glacial runoff. _Hydrological Process._ 10, 509–521 (1996). Article  Google Scholar  * Lane, S. N. & Nienow, P. W. Decadal‐scale climate forcing of Alpine glacial

hydrological systems. _Water Resour. Res._ 55, 2478–2492 (2019). Article  Google Scholar  * Cuffey, K. M. & Paterson, W. S. B. Basal Slip. In K. M. Cuffey, & W. S. B. Paterson

(Eds.), _The physics of glaciers_ (4th ed.). Burlington, MA: Butterworth-Heinemann (2010). * Jacques-Coper, M. & Garreaud, R. D. Characterization of the 1970s climate shift in South

America. _Int. J. Climatol._ 35, 2164–2179 (2015). Article  Google Scholar  * Fountain, A. G. & Walder, J. S. Water flow through temperate glaciers. _Rev. Geophys._ 36, 299–328 (1998).

Article  Google Scholar  * Catania, G. A. & Neumann, T. A. 2010. Persistent englacial drainage features in the Greenland Ice Sheet. _Geophys. Res. Lett._ 37.

https://doi.org/10.1029/2009GL041108 (1998). * Irvine-Fynn, T. D. L., Hodson, A. J., Moorman, B. J., Vatne, G. & Hubbard, A. L. Polythermal glacier hydrology: A review. _Rev. Geophys._

49, 1–37 (2011). * Gulley, J. D., Benn, D. I., Müller, D. & Luckman, A. A cut-and-closure origin for englacial conduits in uncrevassed regions of polythermal glaciers. _J. Glaciol._ 55,

66–80 (2009). Article  Google Scholar  * Tapia Baldis, C. & Trombotto Liaudat, D. Rockslides and rock avalanches in the Central Andes of Argentina and their possible association with

permafrost degradation. _Permafr. Periglac. Process._ 30, 330–347 (2019). Article  Google Scholar  * Tolorza, V. et al. Suspended sediments in Chilean rivers reveal low postseismic erosion

after the Maule Earthquake (Mw 8.8) during a severe drought. _J. Geophys. Res. Earth_ 124, 1378–1397 (2019). Article  Google Scholar  * Herman, F. et al. Erosion by an Alpine glacier.

_Science_ 350, 193–195 (2015). Article  CAS  Google Scholar  * Mernild, S. H., Lipscomb, W. H., Bahr, D. B., Radić, V. & Zemp, M. Global glacier changes: A revised assessment of

committed mass losses and sampling uncertainties. _Cryosphere_ 7, 1565–1577 (2013). Article  Google Scholar  * Farinotti, D. et al. A consensus estimate for the ice thickness distribution of

all glaciers on Earth. _Nat. Geosci._ 12, 168–173 (2019). Article  CAS  Google Scholar  Download references ACKNOWLEDGEMENTS I.V. and R.G. are partially supported by CR2 - FONDAP/ANID

1522A0001. I.D. is funded by SNF Project No. PZ00P2_202024. AUTHOR INFORMATION AUTHORS AND AFFILIATIONS * Center for Climate and Resilience Research, Santiago, Chile Iván Vergara & René

Garreaud * IPATEC, CONICET-UNCo, Bariloche, Argentina Iván Vergara * Department of Geophysics, University of Chile, Santiago, Chile René Garreaud * Institut des dynamiques de la surface

terrestres (IDyST), Université de Lausanne, Lausanne, Switzerland Ian Delaney * Centro de Estudios Avanzados en Zonas Áridas (CEAZA), La Serena, Chile Álvaro Ayala Authors * Iván Vergara

View author publications You can also search for this author inPubMed Google Scholar * René Garreaud View author publications You can also search for this author inPubMed Google Scholar *

Ian Delaney View author publications You can also search for this author inPubMed Google Scholar * Álvaro Ayala View author publications You can also search for this author inPubMed Google

Scholar CONTRIBUTIONS I.V. and R.G. designed the study. I.V. conducted the analyses. A.A. calculated the ice melt of the Juncal Norte glacier. I.V., R.G., I.D. and A.A. interpreted the

analyses and contributed to the paper writing. CORRESPONDING AUTHOR Correspondence to Iván Vergara. ETHICS DECLARATIONS COMPETING INTERESTS The authors declare no competing interests. PEER

REVIEW PEER REVIEW INFORMATION _Communications Earth & Environment_ thanks Ting Zhang, Caio Breda and the other, anonymous, reviewer for their contribution to the peer review of this

work. Primary Handling Editor: Carolina Ortiz Guerrero. A peer review file is available. ADDITIONAL INFORMATION PUBLISHER’S NOTE Springer Nature remains neutral with regard to jurisdictional

claims in published maps and institutional affiliations. SUPPLEMENTARY INFORMATION TRANSPARENT PEER REVIEW FILE SUPPLEMENTARY INFORMATION RIGHTS AND PERMISSIONS OPEN ACCESS This article is

licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any

medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed

material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are

included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and

your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this

licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/. Reprints and permissions ABOUT THIS ARTICLE CITE THIS ARTICLE Vergara, I., Garreaud, R., Delaney, I. _et al._ Deglaciation

in the subtropical Andes has led to a peak in sediment delivery. _Commun Earth Environ_ 5, 630 (2024). https://doi.org/10.1038/s43247-024-01815-8 Download citation * Received: 20 June 2024 *

Accepted: 18 October 2024 * Published: 26 October 2024 * DOI: https://doi.org/10.1038/s43247-024-01815-8 SHARE THIS ARTICLE Anyone you share the following link with will be able to read

this content: Get shareable link Sorry, a shareable link is not currently available for this article. Copy to clipboard Provided by the Springer Nature SharedIt content-sharing initiative