Next Article in Journal
Assessment of CH4 and CO2 Emissions from a Municipal Waste Landfill: Trends, Dispersion, and Environmental Implications
Previous Article in Journal
Emission Rates for Light-Duty Truck Towing Operations in Real-World Conditions
Previous Article in Special Issue
Investigating the Trend Changes in Temperature Extreme Indices in Iran
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Atmosphere
Volume 16
Issue 7
10.3390/atmos16070751
atmosphere-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Exploring the Causes of Multicentury Hydroclimate Anomalies in the South American Altiplano with an Idealized Climate Modeling Experiment

by
Ignacio Alonso Jara
1,*,
Orlando Astudillo
2,3,
Pablo Salinas
2,
Limbert Torrez-Rodríguez
2,4,
Nicolás Lampe-Huenul
5 and
Antonio Maldonado
2,3
1
Departamento de Ciencias Históricas y Geográficas, Universidad de Tarapacá, Arica 1000000, Chile
2
Centro de Estudios Avanzados en Zonas Áridas (CEAZA), Colina del Pino, La Serena 1700000, Chile
3
Departamento de Biología Marina, Universidad Católica del Norte, Coquimbo 1710115, Chile
4
Departamento de Ingeniería Mecánica, Universidad de La Serena, La Serena 1700000, Chile
5
Departamento de Ingeniería en Computación e Informática, Facultad de Ingeniería, Universidad de Tarapacá, Arica 1000000, Chile
*
Author to whom correspondence should be addressed.
Atmosphere 2025, 16(7), 751; https://doi.org/10.3390/atmos16070751
Submission received: 3 February 2025 / Revised: 4 March 2025 / Accepted: 20 March 2025 / Published: 20 June 2025
(This article belongs to the Special Issue Extreme Climate in Arid and Semi-arid Regions)
Browse Figures
Versions Notes

Abstract

:
Paleoclimate records have long documented the existence of multicentury hydroclimate anomalies in the Altiplano of South America. However, the causes and mechanisms of these extended events are still unknown. Here, we present a climate modeling experiment that explores the oceanic drivers and atmospheric mechanisms conducive to long-term precipitation variability in the southern Altiplano (18–25° S; 70–65 W; >3500 masl). We performed a series of 100-year-long idealized simulations using the Weather Research and Forecasting (WRF) model, configured to repeat annually the oceanic and atmospheric forcing leading to the exceptionally humid austral summers of 1983/1984 and 2011/2012. The aim of these cyclical experiments was to evaluate if these specific conditions can sustain a century-long pluvial event in the Altiplano. Unlike the annual forcing, long-term negative precipitation trends are observed in the simulations, suggesting that the drivers of 1983/1984 and 2011/2012 wet summers are unable to generate a century-scale pluvial event. Our results show that an intensification of the anticyclonic circulation along with cold surface air anomalies in the southwestern Atlantic progressively reinforce the lower and upper troposphere features that prevent moisture transport towards the Altiplano. Prolonged drying is also observed under persistent La Niña conditions, which contradicts the well-known relationship between precipitation and ENSO at interannual timescales. Contrasting the hydroclimate responses between the Altiplano and the tropical Andes result from a sustained northward migration of the Atlantic trade winds, providing a useful analog for explaining the divergences in the Holocene records. This experiment suggests that the drivers of century-scale hydroclimate events in the Altiplano were more diverse than previously thought and shows how climate modeling can be used to test paleoclimate hypotheses, emphasizing the necessity of combining proxy data and numerical models to improve our understanding of past climates.

1. Introduction

Annual precipitation in the South American Altiplano is scarce (100–400 mm) and largely limited to the austral summer months (December, January, and February; hereafter DJF) (Figure 1a), representing the main hydrological resource for the inhabitants of the high Andes of Bolivia, Perú, Chile, and Argentina [1,2]. Moreover, DJF precipitation is critical for the sustainability of mountain glaciers, lakes, salt flats, and wetland systems [3,4,5] and is the main driver of regional groundwater recharge [6]. Hence, understanding the drivers and variability of hydroclimate change in the Altiplano is of primary socio-environmental interest. Interannual rainfall variability in the Altiplano is pronounced and results from a complex interplay between upper and lower troposphere circulation, which are ultimately modulated by sea surface temperatures (SSTs) in the equatorial Pacific (i.e., El Niño Southern Oscillation; ENSO) and the tropical Atlantic [2,7]. At longer timescales, tree-ring chronologies indicate the existence of significant decadal and multidecadal hydroclimate changes linked to equatorial Pacific SSTs [8,9]. However, much less is known about the drivers of precipitation on centennial timescales, which hampers a long-term perspective of past variations and the projection of future responses.
Holocene paleoclimate records have long documented the existence of widespread multicentury droughts and pluvials in the Altiplano and adjacent eastern and western cordilleras, e.g., [10,11,12,13]. However, the causes and mechanisms of these long-term anomalies remain little explored. For instance, published hydroclimate reconstructions suggest the existence of a ~300 yr pluvial event that took place about 2000 years ago in the drier southern portion of the Altiplano (18–25° S; 70–65 W; >3500 masl; masl = meters above sea level). Figure 1b shows the location of those reconstructions, and Figure 1c summarizes the proxy evidence that supports the existence of this multicentury pluvial, which included elevated abundances of the high-Andean pollen taxa Apiaceae [14,15], depleted oxygen-isotope values in peatland cores [16], and reduced dust fluxes [17]. Detailed information on the resolution, dating error, and the period of the hydroclimate anomalies for each record is provided in the Supplementary Information (Table S1). This humid event may have also been recorded as rising levels in Lake Titicaca [10,18] and as a decline in oxygen-isotope values in the central Andes at 10° S [19], although no evidence is found in the records from the tropical Andes of Peru, Ecuador, and Colombia further north [20,21,22]. These spatial differences prompted the suggestion of an extra-tropical driver, such as the advection of South Pacific or South Atlantic moisture over the Altiplano, as a potential cause [15,17]. Yet, the spatial distribution of past climate events does not provide direct evidence of their underlying atmospheric mechanisms. Climate model simulations offer an innovative approach to identifying paleoclimate drivers and atmospheric processes [23,24].
Transient climate simulations forced by global circulation models are now commonly employed to reconstruct past trends and assess atmospheric processes. However, they are usually implemented with a coarse spatial resolution (>200 km), and therefore, they fail to resolve the physical processes that control precipitation in regions with complex topography [25,26]. This issue has been recently raised [12,27], showing that a globally forced transient simulation could not capture the full range of the centennial-scale variability observed in the Holocene proxy records from South America and the central Andes region (14–24° S), respectively. The absence of long-term droughts and pluvials in global climate models impedes the ability to attribute specific forcings or atmospheric mechanisms, highlighting the necessity to employ novel approaches, including regional models, that better represent local rainfall patterns resulting from the Andean relief [24].
For the present study, we employ a regional climate model to perform a series of idealized modeling experiments that explore the long-term precipitation responses in the southern Altiplano and their relationship with oceanic variability, tropospheric circulation, and moisture transport. Our experiments aim to test the role of the Pacific and Atlantic SST anomalies in forcing century-long pluvials in the southern Altiplano. To do so, two different yearly ocean-atmosphere configurations that led to positive DJF precipitation anomalies during historical times were identified and used as lateral boundary conditions for a series of independent 100-year-long simulations with a perpetual forcing mode. We validated the simulation outputs, analyzed the regional hydroclimate trends, identified the underlying ocean/atmospheric mechanisms, and finally discussed the implications for regional paleoclimate reconstructions. Our idealized modeling setup allowed us to address the following questions: (1) Can persistent SST anomalies in the Pacific and/or Atlantic basins drive centennial-scale precipitation changes in the southern Altiplano? And (2) what are the atmospheric mechanisms responsible for the existence of extended (≥100 yr) hydroclimate events in this region?
Atmosphere 16 00751 g001
Figure 1. (a) Percentage of DJF precipitation in South America. Precipitation data are taken from the satellite and rain gauge precipitation CHIRPS dataset [28]. (b) Map of the southern Altiplano, including the location of the paleoclimate records mentioned in the text: Laguna Ceusis [14], Lago Chungará [15], Santa Victoria mire [17], and Cerro Tuzgle [16]. (c) Proxy evidence for a Holocene hydroclimate event in the southern Altiplano about 2000 years before the present (BP; gray shading). Present equals to 1950 CE. The map in (b) was produced with the QGIS geographic information system. QGIS Association, http://www.qgis.org, accessed on 23 November 2023.
Figure 1. (a) Percentage of DJF precipitation in South America. Precipitation data are taken from the satellite and rain gauge precipitation CHIRPS dataset [28]. (b) Map of the southern Altiplano, including the location of the paleoclimate records mentioned in the text: Laguna Ceusis [14], Lago Chungará [15], Santa Victoria mire [17], and Cerro Tuzgle [16]. (c) Proxy evidence for a Holocene hydroclimate event in the southern Altiplano about 2000 years before the present (BP; gray shading). Present equals to 1950 CE. The map in (b) was produced with the QGIS geographic information system. QGIS Association, http://www.qgis.org, accessed on 23 November 2023.
Atmosphere 16 00751 g001

2. Study Area

The Altiplano basin, spanning 200,000 km2 across the border of Bolivia, Peru, and Chile, is situated within the subtropical Andes Cordillera of South America at an elevation exceeding 3500 m above sea level. It is an intermontane tectonic trench filled by sedimentary and volcanic basements, bounded by an eastern cordillera and the Amazon and Chaco basins to the east and by a western cordillera and the hyper-arid Pacific coast to the west. Due to its relative aridity, the southern portion of the Altiplano (south of 18° S) is marked by a large number of evaporite basins, including the Uyuni and Coipasa salt flats, two of the world’s largest saline basins. The southern Altiplano is a mountainous region where surface conditions vary significantly across small spatial scales and features an extremely seasonal hydroclimate regime, with a DJF precipitation peak that represents about 70–80% of the annual budget (Figure 1a). DJF precipitation—rain and snow—results from the advection of the continental moisture associated with the South American Monsoon System (SASM) [29]. The monsoon season is marked by the formation of a lower troposphere (>800 hPa) tropical Atlantic anticyclone, which transports oceanic moisture into the Amazon basin. On the western Amazon, the Atlantic moisture flux is redirected southwards by the Andes Cordillera, forming an intense northerly flow of moisture referred to in the literature as the South American Low-Level Jet (SALLJ) [30,31]. In addition, a zone of maximum cloudiness and precipitation is formed from the core of the Amazon basin to the Atlantic coast of southeastern Brazil, which is referred to as the South Atlantic Convergence Zone (SACZ) [32,33]. The SACZ is one of the main features of the SASM, representing a pathway of moisture back to the Atlantic. Summertime advection of moisture to the Altiplano is promoted by the formation of the Bolivian High (BH), an upper troposphere (<500 hPa) anticyclonic circulation centered aloft the Altiplano (Figure S1 in the Supplementary Materials). The BH formation is linked to the strong monsoonal heating over the Amazon, the central Andes, and the SACZ regions [34] and generates an upper-level easterly flow that transports continental moisture up to the Altiplano [35]. Yet, further away from the tropics, the southern Altiplano is at the periphery of the SASM domain, and therefore, it receives considerably less moisture than the northern and eastern portions of the Andean plateau [36]. By the end of the austral summer, the SASM and BH degenerate and an upper-level westerly flow prevails over the Altiplano for the rest of the year.
Wet DJF seasons are usually associated with upper-level easterly wind anomalies, which are accentuated when the BH is displaced to the south [37]. Expectedly, there is a strong correlation between DJF precipitation and the upper troposphere zonal flow over the historical period (1951–2021; Figure S2a in the Supplementary Materials). In addition, anomalous wet summers result from elevated moisture content in western Amazonia and/or from enhanced orographic lifting of the SALLJ [38]. ENSO variability is an important driver of interannual DJF variability in the southern Altiplano, largely by modulating upper-level zonal wind anomalies [39,40,41,42]. The linkage between ENSO and the hydroclimate of the Altiplano is clearly revealed by the characteristic ENSO tongue that emerges from the spatial correlation between interannual DJF precipitation and SSTs (Figure S2b). Wet summers are generally observed during cold SSTs in the tropical Pacific (La Niña) and vice versa [42,43].
Atlantic SSTs, on the other hand, influence the strength of lower-level trade winds and moisture influx towards the Andes [36,44]. In general, warmer SSTs in the northern tropical Atlantic (0–30° N) promote moisture transport into the Amazon basin and further south [39]. SSTs over the southern tropical Atlantic (0–30° S) also modulate the precipitation variability in the Altiplano through the intensification/weakening of the SACZ. Cold SST anomalies over this region strengthen the SACZ and the BH, promoting intense precipitation over the Altiplano [45,46]. However, cold waters over the southern tropical Atlantic could also promote the strengthening of the South Atlantic high-pressure circulation, which could displace the SACZ northwards, reducing continental moisture transport [47]. For this reason, the correlation between SSTs in the southern Atlantic and the precipitation in the southern Altiplano over the historical period is not as strong as the one observed in the tropical Pacific (Figure S2b). DJF precipitation also exhibits decadal variability; however, the amplitude of these changes is smaller compared with the interannual oscillations [43]. Decadal trends are associated with long-term variations in tropical Pacific SSTs, with the upper-level zonal wind anomalies resulting from changes in the position of the BH and with the strengthening/weakening of the SALLJ [48,49]. Ref. [50] showed that dry (wet) summers in the southern Altiplano are associated with the positive (negative) phase of the Pacific Decadal Oscillation (PDO). There is no clear evidence for a strong influence of Atlantic SSTs over decadal rainfall variability in the Altiplano [51].

3. Database and Methods

Experimental Setup

Two annual cycles, characterized by anomalously wet austral summers, were identified through an Empirical Orthogonal Function (EOF) analysis of the ERA5 reanalysis dataset [52], which computed the leading mode of DJF rainfall variability over the 1951–2021 period (Figure S3a in the Supplementary Materials). These periods, the 1983/1984 and 2011/2012 DJF seasons, exhibited strong positive summer precipitation anomalies and were used as boundary conditions for two independent regional modeling experiments (Figure 2 and Table 1). The WRF simulations were designed to evaluate whether the atmospheric and SST conditions responsible for the humid 1983/1984 and 2011/2012 DJF seasons could sustain a century-long pluvial event in the southern Altiplano. The aim was to investigate the physical mechanisms contributing to prolonged hydroclimate anomalies without assuming that the climate states of these short-lived seasons directly mirrored those of multicentury events. Although these idealized simulations do not necessarily reflect realistic past boundary conditions, they provide a straightforward approach to assess paleoclimate hypotheses by evaluating the plausibility of a proposed forcing in terms of the model’s ability to reproduce the spatial or temporal features observed in proxy records [53], such as extended hydroclimate events.
Two 100-year climate simulations were conducted using repeated 3-hourly ERA5 atmospheric lateral boundary conditions (LBCs) (Figure S4 in the Supplementary Materials) and SST surface forcing (Figure 2), derived from the annual periods encompassing the humid 1983/1984 and 2011/2012 DJF seasons. These cyclical LBCs and surface conditions establish a forcing mode suppressing interannual variability, generating idealized modeling runs with their own internal dynamics. In addition, two independent control runs were conducted to assess (1) whether the results correspond directly to the imposed atmospheric LBCs and SST annual forcing associated with the extreme DJF seasons and (2) whether the observed trends result from artifacts of the experimental design (e.g., domain setup, parametrizations, and/or cyclical forcing) rather than the specific LBCs. For (1), a 100-year control experiment was conducted using an annual period encompassing the 2003/2004 DJF season—an average hydroclimate season—as boundary conditions (Figure 2c). For (2), an annual period encompassing the 1997/1998 DJF season was employed as LBCs for a second control simulation, which featured strong El Niño conditions and was one of the driest historical austral summers in the southern Altiplano (Figure 2c). This latter control simulation was forced with extreme oceanic and atmospheric conditions markedly different from those used in the other three simulations despite the application of the same parametrizations and cyclical forcing schemes.
All the modeling runs used boundary conditions that were updated at a 3-hourly temporal resolution, including atmospheric LBCs and SST forcing, derived from the ERA5 dataset. A detailed list of the start and end times for all four modeling runs is provided in Table 2. The regional Weather Research and Forecasting (WRF) model version 3.0 [54] was employed, which is a cutting-edge system for weather prediction that supports a wide range of meteorological applications across scales, with detailed features described in reference [54]. For the experiments, a continental-scale domain encompassing tropical and subtropical South America and the eastern Pacific and western Atlantic margins was employed (90–25° W; 15° N–40° S; Figure 2). This domain covers the region with the strongest precipitation–SST relationship in the tropical Pacific (Figure S2b in the Supplementary Materials) and diverse Atlantic oceanic patterns, enabling the modeling runs to be sensitive to these distinct oceanic boundary conditions. The simulations were limited to 100 years due to the substantial computational resources required for extended periods in such a large domain.
The WRF runs were conducted on a 55 km horizontal grid spacing and 43 vertical levels, ranging from the surface up to 50 hPa, with hourly instantaneous results. A timestep of 90 s was used for the model integration (Table 2). ERA5 reanalysis (0.25° × 0.25° or about 30 km horizontal spacing) was used for the initial and lateral boundary conditions. This climate dataset has proven to accurately depict temperature and precipitation in the Altiplano [55]. Notably, the resolution of the WRF simulation is coarser than the driving ERA5 dataset, representing an upscaling of the lateral grid. The coarse resolution of the modeling grid resulted from the substantial computational resources required for extraordinarily long simulations at a continental scale, a limitation that also prevented the use of a second embedded domain with finer resolution over the Altiplano. Nonetheless, the grid size employed is sufficient to analyze long-term precipitation trends and explore large-scale atmospheric mechanisms in mountainous regions [56]. In terms of parameterizations, the scheme adopted by [57] was followed. The WRF Single Moment 6-class (WSM6) scheme [58] was used for microphysics, with the radiation schemes following the Rapid Radiative Transfer Model [59] for longwave and the Dudhia scheme [60] for shortwave. The MM5 similarity surface layer scheme [61] and the thermal diffusion scheme were employed for the land surface based on the MM5 5-layer soil temperature model [62]. The Yonsei University parameterization of the planetary boundary layer [58] and the Kain–Fritsch scheme for atmospheric convection [63] were also utilized. These parameterizations, identical to those in [56], demonstrated superior performance compared to other schemes in the highly complex topography of the Andes. Total DJF precipitation trends in the 1983/1984 and 2011/2012 simulations, along with the control runs, were reported to analyze the simulated lower and upper troposphere circulation and moisture transport at local and continental scales.

4. Results

We first validated the spatial coherence of the WRF simulations after upscaling from the ERA5 boundary domain. Figure 3 shows a comparison between the total precipitation during the first summer of the 1983/1984 and 2011/2012 WRF simulations (left column) and their corresponding DJF values in the ERA5 dataset (middle column). This comparison offers a straightforward way to evaluate how the WRF model is simulating precipitation patterns at different spatial scales because the first summer of each simulation is not affected by the perpetual forcing mode. Figure 3 reveals a broadly similar spatial pattern of precipitation at both continental and regional scales, although significantly higher values are observed in the WRF compared with the ERA5 dataset. At continental scales, peak precipitation is observed over the central Amazon basin, the Ecuadorian/Colombian Pacific coast, and along the Peruvian and Bolivian Andes in both datasets. In the southern Altiplano, both WRF simulations generate higher precipitation amounts with maxima over the eastern Andes flanks compared to their corresponding ERA5 periods (Figure 3b,d). Pearson’s correlation indicates that the WRF is significantly positively correlated to ERA5 in both the 1983/1984 (r = 0.32; p < 0.01) and 2011/2012 (0.46; p < 0.01) simulations, although the correlation is not strong. Overall, the spatial distribution of precipitation at continental and regional scales was not dramatically changed by upscaling from the finer ERA5 grid. To further validate the coherence of the simulations, we compared the first summer seasons of the 1983/1984 and 2011/2012 WRF simulations with their corresponding DJF periods in an independent dataset: the satellite and rain gauge precipitation estimates of the CHIRPS product [28]; right column in Figure 3. This comparison exhibits an overall good agreement between the spatial pattern of the WRF simulations and the CHIRPS datasets, with precipitation maxima over the Amazon basin, the eastern Andean flanks, and the SACZ region. This latter band of convective precipitation is particularly well represented in both the WRF and CHIRPS domains during the 2011/2012 season (Figure 3c). However, once again, the WRF model produces significantly higher precipitation over these regions in the first summer of both simulations. These differences could be explained by the fact that the simulations with regional models at scales coarser than 50 km tend to overestimate rainfall, which is partially associated with the misrepresentation of the gradients imposed by the topography that are not explicitly resolved in models at this scale [64,65]. Over the southern Altiplano, the CHIRPS dataset shows lower precipitation compared with the WRF simulations, especially for the eastern margins of the plateau. These differences are largely explained because the WRF simulations were forced with ERA5 conditions, which produce significantly higher precipitation than the CHIRPS dataset for the southern Altiplano. Notwithstanding these differences, significant correlations between the WRF and CHIRPS are observed over the southern Altiplano for the 1983/1984 (r = 0.47; p < 0.01) and the 2011/2012 (r = 0.44; p < 0.01) simulations. Overall, the WRF model has the ability to properly represent the main spatial precipitation patterns captured by gridded datasets at continental and regional scales.
Figure 4 shows the total DJF precipitation trends for the southern Altiplano in all four 100 yr modeling runs (the 1983/1984, 2011/2012 simulations, plus the 2003/2004 and 1997/1998 control runs). All simulations are able to generate similar summer precipitation values during the initial years to those observed in the first year of each experiment, consistent with the year-to-year imposition of the same annual atmospheric lateral and oceanic surface forcing. Long-term negative trends are visible for the 1983/1984 and 2011/2012 simulations, as well as over the 2003/2004 control run. The 1983/1984 run shows the strongest downturn, with a reduction of 75% by the end of the simulation period, whereas this reduction is slightly less pronounced (71%) for the 2011/2012 simulation. These downward trends could reflect the fact that the model has been forced by extreme pluvial conditions, which may require a continuous supply of moisture that is impossible to recharge without any interannual variability. This scenario is supported by the analysis of the integrated vapor transport, which shows substantial reductions in the moisture reaching the Altiplano for the 1983/1984 and 2011/2012 simulations (Figure S5 in the Supplementary Material). Yet, the 2003/2004 control run, forced by mild SST conditions and a neutral hydroclimate DJF season, exhibits a likewise downward trend with 66% less DJF precipitation at the end of the simulation (Figure 4c), indicating that the observed responses in the 1983/1984 and 2011/2012 simulations are not controlled by the extreme SST field imposed. Similar downward trends in these three modeling runs may suggest that the simulated changes are not responding to the lateral boundary conditions but, instead, that they might be a modeling artifact related to the experimental setup. However, the negative drifts in our simulations are limited to the austral summer season (Figure S6 in the Supplementary Materials) and that the average annual precipitation for the entire South American continent shows not common trajectories but distinct responses for each simulation, with positive trends observed in certain parts of the simulation domain, as it will be discussed in the next section. One potential explanation for the common trajectories is the large size of the model domain employed in our study. In continental-scale domains, the lateral boundary forcing becomes less important than the internal dynamics and feedback [66]. This could explain the downward drifts in the simulations even though they are forced by different boundary conditions. Additionally, the downward trends could have resulted from the cyclical forcing employed in the experimental setup. However, unlike the three other simulations, the 1997/1998 cyclical control shows no reduction over the 100 yr period (Figure 4b). This result indicates that the downward trends observed in the 1983/1984 and 2011/2012 experiments are unrelated to the cyclical forcing or any specific parametrization used as an experimental design.
The ultimate cause for the pronounced interannual drying in three of the four experiments is linked to an alteration in the upper and lower troposphere circulation, driven by an intensified anticyclonic circulation in the southwestern Atlantic (Figure S7). Figure S7 further reveals a progressive cooling of surface air temperature (2 m), centered beneath the high-pressure anomalies (20–35° S). The simulated cooling is then interpreted as a consequence of enhanced westerly winds in the southern branch of the anticyclone, which reduce evaporation and sensible heat flux from the underlying SSTs. Notably, the magnitude of these changes correlates with the magnitude of the downward precipitation trends over the Altiplano, with the stronger cooling observed in the 1983/1984 simulation. In contrast, the 1997/1998 experiment shows no anticyclonic intensification and no significant cooling. These simulated dynamics are in line with [45], which demonstrated that persistent warm SST anomalies north of 40° S enhance the subtropical anticyclone, suppressing vertical motion and deep convection over the SACZ region. Despite generating humid DJF seasons, the imposed warm SST fields over the southwestern Atlantic observed in the 1983/1984 and 2011/2012 summers (Figure 2) promote this drying mechanism in the long term. Over time, this dynamic progressively cools surface air and disrupts the southward advection of Atlantic moisture, thereby curtailing the humidity influx into the Altiplano and causing the observed downward precipitation trends. This dynamic is also consistent with the historical variability in the ERA5 reanalysis, where dry (wet) DJF seasons are characterized by a strengthening (weakening) of the high-pressure cell over the southern Atlantic (Figure S8). This observation suggests that the simulated drifts result from the natural ocean and atmosphere dynamics documented in historical times rather than from any artificial processes resulting from the experimental setup. In summary, all this evidence supports that the selected boundary conditions are generating realistic climate dynamics across the simulation domain, largely resulting from the internal model dynamics and not from the extreme boundary conditions or other experimental setups. Hence, we posit that an evaluation of the causes and processes leading to the observed hydroclimate changes could yield valuable insights into the causes of long-term changes, as evidenced in paleoclimate records.

5. Discussion

While our results can be used to gain new insights into the long-term hydroclimate variability in the Altiplano, it is worth noting that the WRF runs presented here excluded past external forcings such as orbital configurations, greenhouse gases, volcanic, and solar activity. By employing a perpetual climatology that explicitly suppressed any form of interannual variability, our WRF simulations represent an idealized scenario rather than a realistic representation of past climate states, and therefore, the paleoclimate inferences that emerge from it should be taken with caution. Despite that our experiments generate information about the causes of long-term climate variability that are not restricted to the distant past, the simulated results can be helpful in the interpretation of proxy-based reconstructions, especially for records covering the mid or late-Holocene, where orbital and greenhouse parameters approached pre-industrial levels. In this section, the causes of century-long hydroclimate anomalies are further elaborated based on the atmospheric dynamics observed in our 1983/1984 and 2011/2012 simulations.
The mechanisms behind the negative precipitation trends observed in the 1983/1984 and 2011/2012 simulations are explored in Figure 5, which compares the climatological difference between the initial (years 1 to 10) and final (years 91 to 100) decades in both simulations. The long-term precipitation decline in the 1983/1984 run is marked by a prevailing upper-level westerly flow, a strong reduction in upward motion (Figure 5a) over the Altiplano, as well as a significant decline in specific humidity over the entire central Andes and the Pacific coast (Figure 5b). The high westerly wind anomaly is linked to a northwestward migration of the BH, promoted by the intensification of the South Atlantic anticyclone (Figure 5b; vectors). Despite the intensified anticyclonic circulation in the southern Atlantic, moderate increases in upward motion and specific humidity are observed over the SACZ region (Figure 5a and Figure 5b, red and purple areas, respectively). Active convection over the SACZ region might contribute to the negative precipitation trend in the southern Altiplano by diverting moisture back to the southern tropical Atlantic and, thus, away from the Andes Cordillera. Furthermore, any remaining moisture reaching the Andes is further expelled from the Altiplano by the prevailing upper-level westerly wind anomaly.
Comparing the latitude/pressure plots for the initial and final decades of the 1983/1984 simulation reveals a change from an easterly to a westerly wind regime aloft the Altiplano, as well as a transition from positive to negative anomalies in specific humidity at the mid and upper levels (Figure 6a,b). This latter change is clearly reflected in Figure S9 (Supplementary Materials) as a strong reduction in the northerly moisture transport from 500 to 300 hPa (left column). Altogether, our results suggest that the SST anomalies responsible for the strong pluvial of December of 1983 and January–February of 1984—that is, warmer SSTs over the southern tropical Atlantic and the Atlantic coast of southern Brazil, Uruguay, and Argentina—are unable to generate extended pluvial conditions in the region. We do not find any evidence of extra-tropical moisture transport to the Altiplano by the end of the simulation based on the absence of any extra-tropical area showing increases in specific humidity or a clear southerly low-level flow. In this regard, our study does not support the inference that an extra-tropical moisture source might have caused an extended pluvial event in the southern Altiplano [15,17]. Our 1983/1984 simulation is consistent with the observation that a warmer southern tropical Atlantic can sustain a vigorous SACZ over extended periods of time [67]. The hydroclimate of the Altiplano has been linked to the location/intensity of the SACZ at different timescales [46]; our results show that an active SACZ would not necessarily generate a century-scale pluvial in the southern Altiplano if westerly wind anomalies prevail in the upper troposphere. The combination of an intensified SACZ and a southward movement of the BH has been suggested as a driver for millennial-scale pluvial events over the Altiplano during the Last Glacial Termination (18,000–11,700 years BP) [68]. This scenario differs from our long-term experiment, and it may suggest that the interplay between the BH and the SACZ and the resulting hydroclimate impacts on the Altiplano may not have been stationary in time. However, these inferences should be taken carefully as boundary conditions during the Last Termination were much different from the ones used in this experiment.
When contrasting the initial and final decades of the 2011/2012 simulation (Figure 5c,d), the establishment of a strong westerly flow at the upper troposphere and strong reductions in upward motion and specific humidity are clearly observable. Yet, unlike the 1983/1984 run, the final decade of the 2011/2012 simulation shows a decline in upward motion around the SACZ area, which indicates a weakening of this climate feature (Figure 5c). The primary cause of the precipitation decay in the 2011/2012 simulation seems to be related to a strong and general reduction in moisture availability over the southwestern Amazon and the lowlands of Bolivia and northern Argentina, as evidenced by the 500 hPa specific humidity differences (Figure 5d). This reduction leads to sustained drying over the southern Altiplano despite the persistence of a low-level northerly flow, as revealed in Figure S9. The latitude–pressure plots of Figure 6c,d expose a clear transition from an easterly to a westerly zonal wind anomaly above the Altiplano, which further prevents any marginal tropical moisture from penetrating the high-Andean plateau. Despite all the limitations of our experiment, these results suggest that the SST anomalies causing the historical 2011/2012 DJF pluvial in the southern Altiplano—that is, the moderate La Niña conditions over the tropical Pacific combined with cold temperatures in the southern tropical Atlantic—are incapable of forcing a century-long pluvial event.
Regardless of any change in moisture availability or low-level circulation, strong upper-level westerly wind anomalies are observed by the end of the 1983/1984 and 2011/2012 WRF simulations, which emphasizes the relevance of a high-tropospheric flow in controlling century-scale hydroclimate changes in the southern Altiplano. Notably, the Atlantic SST anomalies, observed during the 2011/2012 historical summer, resemble the negative phase of the SAOD, with colder SST anomalies in the subtropics and warmer anomalies in the southern Atlantic (Figure 2d). In historical times, this configuration tends to weaken the lower troposphere Atlantic gyre and the inland moisture transport, generating dry conditions in the central Andes and northern Brazil [67,69]. Our simulation indicates that this mechanism might act cumulatively in our cyclical simulations, reducing the inland moisture transport and SACZ activity year after year, irrespective of the conditions prevailing in the tropical Pacific. Global circulation models partially support these results, reproducing extended periods of reduced DJF precipitation in the Altiplano during the last millennium in association with a lower-level westerly wind anomaly over the continental Equator and a weakening of the SACZ [27]. It is further noted that, by the end of both WRF simulations, the Atlantic branch of the Inter-Tropical Convergence Zone (ITCZ) has shifted northward to the 5–15° N band, revealed by the vertical motion and specific humidity differences in Figure 5. Such a northerly position is unusual in the austral summer, and it is likely caused by the cold anomalies recorded in the subtropical Pacific and Atlantic basins (Figure S7). The northward shift of the ITCZ is ultimately responsible for the declining humidity levels and convective activity over the Amazon and the central Andes while resulting in strong positive precipitation anomalies over the tropical Andes north of the Equator (Figure 7). Interestingly, a long-term (70–80 years) increase (~50–60%) in the total March, April, and May (MAM) precipitation is observed in both simulations (Figure S6). This is the only season experiencing significant precipitation surges, and by the end of the 100 yr period, total DJF and MAM precipitation are roughly equivalent in both the 1983/1984 and 2011/2012 runs. This seasonal shift might explain the differences between annual versus summer precipitation reconstructions in the Altiplano and/or the potential incursion of extra-tropical precipitation. However, these MAM surges are insufficient to drive significant increases in the total annual precipitation over the Altiplano. In sum, our WRF simulations suggest that the oceanic conditions conducive to historical DJF pluvials in the southern Altiplano fail to generate the extended periods of increased precipitation documented in the Holocene paleoclimate records, such as the one recorded ~2000 years ago (Figure 1c). The cumulative effect of historical SSTs and atmospheric forcings in the WRF runs results in lower and upper troposphere alterations that, in contrast, lead to significant negative precipitation anomalies.
Which SST anomalies drove the multicentury pluvials that are observed in the Holocene proxy records, then? Extended intervals with strong La Niña conditions might be a potential cause, as only weak (1983/1984) and moderate La Niña (2011/2012) summers were tested (Table 1 and Figure 2). We note that one of the alkenone-derived SST reconstructions presented in [70] exhibits a cooling event in the Pacific coast of southern Perú (15° S) between 2300 and 1900 years BP, closely aligned with the southern Altiplano pluvial event described in the introduction. This may suggest that a cooling of the southwestern Pacific coast could be a potential SST driver; however, both the 1983/1983 and 2011/2012 simulations were forced by warm SSTs over the southern Peruvian coasts (Figure 2). Future modeling experiments with a broader set of oceanic forcing configurations will help to evaluate this possibility.
Although the resulting hydroclimate trajectories for the southern Altiplano are opposite to the one documented 2000 years ago, our simulation domain was large enough to include tropical and subtropical South America, providing information about long-term hydroclimate anomalies that could be relevant for the interpretation of proxy records. At the continental scale, both the 1983/1984 and 2011/2012 WRF runs exhibit strong precipitation surges, which were observed over the western fringes of the Amazon, the tropical Andes, and northern South America (Figure 7). Opposite trends between hydroclimate reconstructions from the Altiplano and the tropical Andes have been described in the literature, e.g., [14,71]. For instance, during the Little Ice Age (500–100 years BP; LIA), humid conditions were recorded in the tropical Andes of Peru, Ecuador, and Colombia [19,72,73], while drier climates have been suggested for the southern Altiplano and the Atacama Desert [12,13,74]. The northward migration of the BH and the trade winds observed in the simulations offer a potential mechanism to explain such differences, although these findings should be approached with care because the proposed LIA scenario may differ from the historical boundary conditions used in this study. We also note that the continental hydroclimate trends observed in our simulations, particularly the 1983/1984 run, resemble spatial variability of the second mode of the past SASM variations identified by [75] based on the statistical decomposition of oxygen-isotope paleoclimate records. This mode of variability tracks the SASM contractions/expansions that result in precipitation anomalies of opposite signs along its northern and southern margins. This monsoon dynamic could result in drier conditions over the southern Altiplano and northern Brazil, along with wet climates over the tropical Andes and northern South America. The 1983/1984 and 2011/2012 simulations provide a potential mechanism for this type of monsoon dynamic, resulting from a progressive cooling of the subtropical Pacific and Atlantic basins, expansion of the South Atlantic high-pressure cell, strong upper-level westerly wind anomalies, and a northward migration of the ITCZ. The authors of [75] have dated their second mode of monsoon variability between 460–300 years BP.

6. Conclusions

To our understanding, this study presents the first regional climate modeling experiment of long-term hydroclimate responses in the Altiplano. Despite the idealized nature of our simulations, novel information with regard to the potential mechanisms governing precipitation variability on centennial timescales has been provided. A detailed analysis of the control experiments led us to conclude that the selected SST forcings are generating distinctive atmospheric dynamics. Importantly, these dynamics are unrelated to the imposition of extreme hydrological seasons or the cyclical nature of the experiment.
Despite being forced with boundary conditions leading to the humid summers of 1983/1984 and 2011/2012, the simulations show long-term negative trends. Progressive drying in both simulations is the result of the strong upper-level westerly wind anomalies that developed in response to a sustained cooling of the southwestern Atlantic and a strengthening of the anticyclonic circulation. These results indicate that the SST anomalies responsible for such a humid DJF season, which included warmer SSTs over the southwestern Atlantic and moderate La Niña conditions, are incapable of generating extended pluvial conditions in the southern Altiplano. In addition, we did not find evidence for an extra-tropical source as a cause of long-term hydroclimate change. At continental scales, the simulation exhibits notable precipitation surges over the western Amazon basin, the tropical Andes, and northern South America. Hence, the mechanisms leading to persistent drying over the southern Altiplano can be used to explain the divergent trends observed in the paleoclimate records from tropical South America.
We were unable to provide direct evidence for the causes of the Holocene multicentury pluvials recorded in the proxy records, albeit an outline of the potential drivers and processes explaining the long-term hydroclimate trajectories have been presented and discussed in detail. Since summertime precipitation is a critical socio-environmental resource for the inhabitants of the high Andes of South America, our results may help to outline future hydroclimate trajectories in a region where climate projections show considerable disagreement [76,77]. Our WRF modeling experiment represents an innovative methodology for testing paleoclimate hypotheses derived from the growing number of proxy sequences available in South America.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/atmos16070751/s1. Reference [78] is cited in the Supplementary Materials.

Author Contributions

I.A.J., O.A. and A.M. conceptualized the study. The WRF modeling runs were performed by P.S. and O.A. The analysis and visualization of the data were conducted by L.T.-R., N.L.-H. and I.A.J. I.A.J. led the writing of the manuscript with contributions from all authors. All authors have read and agreed to the published version of the manuscript.

Funding

This work was conducted with the financial support of the ANID postdoctoral grant #319018, ANID-MILENIO NCS2022_009, and Proyecto UTA Mayor N° 5818-23. Limbert Torrez-Rodriguez has received funding support from FONDECYT through grant #1201742, whereas Orlando Astudillo was supported by FONDECYT project #1231174 and from ANID (Concurso de Fortalecimiento al Desarrollo Científico de Centros Regionales 2020-R20F0008-CEAZA). Antonio Maldonado was supported by FONDECYT 1221106 grant.

Data Availability Statement

The ERA5 reanalysis data were retrieved from https://cds.climate.copernicus.eu (accessed on 15 September 2023). The CHIRPS precipitation data were obtained from https://www.chc.ucsb.edu/data/chirps (accessed on 15 September 2023). The Niño3.4 index was obtained from https://climatedataguide.ucar.edu (accessed on 15 September 2023), while the SAOD index (SAODI) was obtained from http://lijianping.cn/dct/page/65592 (accessed on 15 September 2023). Our three WRF runs (monthly climatological means) and all Python programming codes used for the climate analysis and visualizations will be freely available from the GitHub repository at https://github.com/AlexelProgramador (accessed on 15 September 2023).

Acknowledgments

The modeling work was supported by the supercomputing infrastructure of the NLHPC (ECM-02). The authors express their gratitude to the Center for Advanced Studies in Arid Zones (CEAZA) and the National Laboratory of High-Performance Computing Chile (NLHPC) for the availability to run the WRF simulations.

Conflicts of Interest

The authors declare that they have no relevant financial or non-financial interests to disclose.

Abbreviations

The following abbreviations are used in this manuscript:
WRFWeather Research and Forecasting (WRF)
DJFDecember–January–February
MAMMarch–April–May
ENSOEl Nino Southern Oscillation
PDOPacific Decadal Oscillation
BHBolivian High
LBCsLateral boundary conditions
SALLJSouth American Low-Level Jet
SAODSouth Atlantic Ocean Dipole
SACZSouth Atlantic Convergence Zone
SASMSouth American summer monsoon
ITCZIntertropical Convergence Zone
SSTSea surface temperature

References

  1. García, M.; Raes, D.; Jacobsen, S.-E.; Michel, T. Agroclimatic constraints for rainfed agriculture in the Bolivian Altiplano. J. Arid Environ. 2007, 71, 109–121. [Google Scholar] [CrossRef]
  2. Canedo-Rosso, C.; Uvo, C.B.; Berndtsson, R. Precipitation variability and its relation to climate anomalies in the Bolivian Altiplano. Int. J. Climatol. 2019, 39, 2096–2107. [Google Scholar] [CrossRef]
  3. Vuille, M.; Carey, M.; Huggel, C.; Buytaert, W.; Rabatel, A.; Jacobsen, D.; Soruco, A.; Villacis, M.; Yarleque, C.; Timm, O.E. Rapid decline of snow and ice in the tropical Andes–Impacts, uncertainties and challenges ahead. Earth-Sci. Rev. 2018, 176, 195–213. [Google Scholar]
  4. Satgé, F.; Hussain, Y.; Xavier, A.; Zolá, R.P.; Salles, L.; Timouk, F.; Seyler, F.; Garnier, J.; Frappart, F.; Bonnet, M.-P. Unraveling the impacts of droughts and agricultural intensification on the Altiplano water resources. Agric. For. Meteorol. 2019, 279, 107710. [Google Scholar]
  5. Anderson, T.G.; Christie, D.A.; Chávez, R.O.; Olea, M.; Anchukaitis, K.J. Spatiotemporal peatland productivity and climate relationships across the Western south American altiplano. J. Geophys. Res. Biogeosci. 2021, 126, e2020JG005994. [Google Scholar] [CrossRef]
  6. Blin, N.; Hausner, M.; Leray, S.; Lowry, C.; Suárez, F. Potential impacts of climate change on an aquifer in the arid Altiplano, northern Chile: The case of the protected wetlands of the Salar del Huasco basin. J. Hydrol. Reg. Stud. 2022, 39, 100996. [Google Scholar]
  7. Bennett, M.; New, M.; Marino, J.; Sillero-Zubiri, C. Climate complexity in the Central Andes: A study case on empirically-based local variations in the dry Puna. J. Arid Environ. 2016, 128, 40–49. [Google Scholar]
  8. Crispín-DelaCruz, D.B.; Morales, M.S.; Andreu-Hayles, L.; Christie, D.A.; Guerra, A.; Requena-Rojas, E. High ENSO sensitivity in tree rings from a northern population of Polylepis tarapacana in the Peruvian Andes. Dendrochronologia 2022, 71, 125902. [Google Scholar]
  9. Morales, M.; Christie, D.; Villalba, R.; Argollo, J.; Pacajes, J.; Silva, J.; Alvarez, C.; Llancabure, J.; Gamboa, C.S. Precipitation changes in the South American Altiplano since 1300 AD reconstructed by tree-rings. Clim. Past 2012, 8, 653. [Google Scholar]
  10. Abbott, M.B.; Binford, M.W.; Brenner, M.; Kelts, K.R. A 350014C yr high-resolution record of water-level changes in Lake Titicaca, Bolivia/Peru. Quat. Res. 1997, 47, 169–180. [Google Scholar] [CrossRef]
  11. Baker, P.A.; Fritz, S.C.; Garland, J.; Ekdahl, E. Holocene hydrologic variation at Lake Titicaca, Bolivia/Peru, and its relationship to North Atlantic climate variation. J. Quat. Sci. Publ. Quat. Res. Assoc. 2005, 20, 655–662. [Google Scholar] [CrossRef]
  12. González-Pinilla, F.J.; Latorre, C.; Rojas, M.; Houston, J.; Rocuant, M.I.; Maldonado, A.; Santoro, C.M.; Quade, J.; Betancourt, J.L. High-and low-latitude forcings drive Atacama Desert rainfall variations over the past 16,000 years. Sci. Adv. 2021, 7, eabg1333. [Google Scholar]
  13. Valero-Garcés, B.L.; Grosjean, M.; Schwalb, A.; Geyh, M.; Messerli, B.; Kelts, K. Limnogeology of Laguna Miscanti: Evidence for mid to late Holocene moisture changes in the Atacama Altiplano (Northern Chile). J. Paleolimnol. 1996, 16, 1–21. [Google Scholar]
  14. Jara, I.A.; Maldonado, A.; de Porras, M.E. Late Holocene dynamics of the south American summer monsoon: New insights from the Andes of northern Chile (21° S). Quat. Sci. Rev. 2020, 246, 106533. [Google Scholar]
  15. Jara, I.A.; Maldonado, A.; González, L.; Hernández, A.; Sáez, A.; Giralt, S.; Bao, R.; Valero Garcés, B.L. Centennial-scale precipitation anomalies in the southern Altiplano (18° S) suggest an extratropical driver for the South American summer monsoon during the late Holocene. Clim. Past 2019, 15, 1845–1859. [Google Scholar]
  16. Kock, S.T.; Schittek, K.; Wissel, H.; Vos, H.; Ohlendorf, C.; Schäbitz, F.; Lupo, L.C.; Kulemeyer, J.J.; Lücke, A. Stable oxygen isotope records (δ18O) of a high-Andean cushion peatland in NW Argentina (24° S) imply South American Summer Monsoon related moisture changes during the Late Holocene. Front. Earth Sci. 2019, 7, 45. [Google Scholar]
  17. Hooper, J.; Marx, S.K.; May, J.-H.; Lupo, L.C.; Kulemeyer, J.J.; Pereira, E.; Seki, O.; Heijnis, H.; Child, D.; Gadd, P. Dust deposition tracks late-Holocene shifts in monsoon activity and the increasing role of human disturbance in the Puna-Altiplano, northwest Argentina. Holocene 2020, 30, 519–536. [Google Scholar] [CrossRef]
  18. Guédron, S.; Delaere, C.; Fritz, S.C.; Tolu, J.; Sabatier, P.; Devel, A.-L.; Heredia, C.; Vérin, C.; Alves, E.Q.; Baker, P.A. Holocene variations in Lake Titicaca water level and their implications for sociopolitical developments in the central Andes. Proc. Natl. Acad. Sci. USA 2023, 120, e2215882120. [Google Scholar]
  19. Kanner, L.C.; Burns, S.J.; Cheng, H.; Edwards, R.L.; Vuille, M. High-resolution variability of the South American summer monsoon over the last seven millennia: Insights from a speleothem record from the central Peruvian Andes. Quat. Sci. Rev. 2013, 75, 1–10. [Google Scholar]
  20. Bird, B.W.; Abbott, M.B.; Rodbell, D.T.; Vuille, M. Holocene tropical South American hydroclimate revealed from a decadally resolved lake sediment δ18O record. Earth Planet. Sci. Lett. 2011, 310, 192–202. [Google Scholar] [CrossRef]
  21. Jaramillo, D.; Vélez, M.I.; Escobar, J.; Pardo-Trujillo, A.; Vallejo, F.; Villegas, J.C.; Acevedo, A.L.; Curtis, J.; Rincon, H.; Trejos-Tamayo, R. Mid to late holocene dry events in Colombia’s super humid Western Cordillera reveal changes in regional atmospheric circulation. Quat. Sci. Rev. 2021, 261, 106937. [Google Scholar]
  22. Haug, G.H.; Hughen, K.A.; Sigman, D.M.; Peterson, L.C.; Rohl, U. Southward migration of the intertropical convergence zone through the Holocene. Science 2001, 293, 1304–1308. [Google Scholar] [PubMed]
  23. Zhu, C.; Zhang, J.; Liu, Z.; Otto-Bliesner, B.L.; He, C.; Brady, E.C.; Tomas, R.; Wen, Q.; Li, Q.; Zhu, C. Antarctic warming during Heinrich Stadial 1 in a transient isotope-enabled deglacial simulation. J. Clim. 2022, 35, 7353–7365. [Google Scholar]
  24. Orrison, R.; Vuille, M.; Smerdon, J.E.; Apaéstegui, J.; Campos, J.L.P.; Cruz, F.W.; Della Libera, M.E. South American Monsoon variability over the last millennium in paleoclimate records and isotope-enabled climate models. Clim. Past Discuss. 2022, 18, 2045–2062. [Google Scholar]
  25. Ludwig, P.; Gómez-Navarro, J.J.; Pinto, J.G.; Raible, C.C.; Wagner, S.; Zorita, E. Perspectives of regional paleoclimate modeling. Ann. N. Y. Acad. Sci. 2019, 1436, 54–69. [Google Scholar] [PubMed]
  26. Cook, K.; Vizy, E. South American climate during the Last Glacial Maximum: Delayed onset of the South American monsoon. J. Geophys. Res. Atmos. 2006, 111, D02110. [Google Scholar]
  27. Rojas, M.; Arias, P.A.; Flores-Aqueveque, V.; Seth, A.; Vuille, M. The South American monsoon variability over the last millennium in climate models. Clim. Past 2016, 12, 1681–1691. [Google Scholar]
  28. Funk, C.; Peterson, P.; Landsfeld, M.; Pedreros, D.; Verdin, J.; Shukla, S.; Husak, G.; Rowland, J.; Harrison, L.; Hoell, A.; et al. The climate hazards infrared precipitation with stations—A new environmental record for monitoring extremes. Sci. Data 2015, 2, 150066. [Google Scholar]
  29. Vera, C.; Higgins, W.; Amador, J.; Ambrizzi, T.; Garreaud, R.; Gochis, D.; Gutzler, D.; Lettenmaier, D.; Marengo, J.; Mechoso, C.R. Toward a unified view of the American monsoon systems. J. Clim. 2006, 19, 4977–5000. [Google Scholar] [CrossRef]
  30. Romatschke, U.; Houze, R.A., Jr. Extreme summer convection in South America. J. Clim. 2010, 23, 3761–3791. [Google Scholar] [CrossRef]
  31. Yabra, M.S.; Nicolini, M.; Borque, P.; Skabar, Y.G.; Salio, P. Observational study of the South American low-level jet during the SALLJEX. Int. J. Clim. 2022, 42, 9676–9696. [Google Scholar] [CrossRef]
  32. Kodama, Y. Large-scale common features of subtropical precipitation zones (the Baiu frontal zone, the SPCZ, and the SACZ) Part I: Characteristics of subtropical frontal zones. J. Meteorol. Soc. Jpn. Ser. II 1992, 70, 813–836. [Google Scholar] [CrossRef]
  33. Nielsen, D.M.; Belém, A.L.; Marton, E.; Cataldi, M. Dynamics-based regression models for the South Atlantic Convergence Zone. Clim. Dyn. 2019, 52, 5527–5553. [Google Scholar] [CrossRef]
  34. Lenters, J.; Cook, K.H. On the origin of the Bolivian high and related circulation features of the South American climate. J. Atmospheric Sci. 1997, 54, 656–678. [Google Scholar] [CrossRef]
  35. Garreaud, R.; Vuille, M.; Clement, A.C. The climate of the Altiplano: Observed current conditions and mechanisms of past changes. Palaeogeogr. Palaeoclimatol. Palaeoecol. 2003, 194, 5–22. [Google Scholar] [CrossRef]
  36. Vuille, M.; Keimig, F. Interannual variability of summertime convective cloudiness and precipitation in the central Andes derived from ISCCP-B3 data. J. Clim. 2004, 17, 3334–3348. [Google Scholar] [CrossRef]
  37. Aceituno, P.; Montecinos, A. Circulation anomalies associated with dry and wet periods in the South American Altiplano. In Proceedings of the Fourth International Conference on Southern Hemisphere Meteorology and Oceanography, Hobart, Australia, 29 March–2 April 1993; pp. 330–331. [Google Scholar]
  38. Segura, H.; Espinoza, J.C.; Junquas, C.; Lebel, T.; Vuille, M.; Garreaud, R. Recent changes in the precipitation-driving processes over the southern tropical Andes/western Amazon. Clim. Dyn. 2020, 54, 2613–2631. [Google Scholar] [CrossRef]
  39. Sulca, J.; Vuille, M.; Silva, Y.; Takahashi, K. Teleconnections between the Peruvian Central Andes and Northeast Brazil during extreme rainfall events in Austral summer. J. Hydrometeorol. 2016, 17, 499–515. [Google Scholar] [CrossRef]
  40. Rojas-Murillo, K.; Lupo, A.R.; Garcia, M.; Gilles, J.; Korner, A.; Rivera, M.A. ENSO and PDO related interannual variability in the north and east-central part of the Bolivian Altiplano in South America. Int. J. Clim. 2022, 42, 2413–2439. [Google Scholar] [CrossRef]
  41. Lavado-Casimiro, W.; Espinoza, J.C. Impactos de El Niño y La Niña en las lluvias del Perú (1965–2007). Rev. Bras. Meteorol. 2014, 29, 171–182. [Google Scholar] [CrossRef]
  42. Vuille, M.; Bradley, R.S.; Keimig, F. Interannual climate variability in the Central Andes and its relation to tropical Pacific and Atlantic forcing. J. Geophys. Res. Atmos. 2000, 105, 12447–12460. [Google Scholar]
  43. Garreaud, R. The Andes climate and weather. Adv. Geosci. 2009, 22, 3–11. [Google Scholar]
  44. Espinoza, J.C.; Marengo, J.A.; Ronchail, J.; Carpio, J.M.; Flores, L.N.; Guyot, J.L. The extreme 2014 flood in south-western Amazon basin: The role of tropical-subtropical South Atlantic SST gradient. Environ. Res. Lett. 2014, 9, 124007. [Google Scholar] [CrossRef]
  45. Robertson, A.W.; Mechoso, C.R.; Kim, Y.-J. The influence of Atlantic sea surface temperature anomalies on the North Atlantic Oscillation. J. Clim. 2000, 13, 122–138. [Google Scholar]
  46. Lenters, J.; Cook, K.H. Summertime precipitation variability over South America: Role of the large-scale circulation. Mon. Weather Rev. 1999, 127, 409–431. [Google Scholar] [CrossRef]
  47. Laureanti, N.C.; Chou, S.C.; Nobre, P.; Curchitser, E. On the relationship between the South Atlantic Convergence Zone and sea surface temperature during Central-East Brazil extreme precipitation events. Dyn. Atmos. Ocean. 2024, 105, 101422. [Google Scholar]
  48. Segura, H.; Espinoza, J.C.; Junquas, C.; Takahashi, K. Evidencing decadal and interdecadal hydroclimatic variability over the Central Andes. Environ. Res. Lett. 2016, 11, 094016. [Google Scholar]
  49. Sulca, J.; Vuille, M.; Dong, B. Interdecadal variability of the austral summer precipitation over the Central Andes. Front. Earth Sci. 2022, 10, 954954. [Google Scholar] [CrossRef]
  50. Seiler, C.; Hutjes, R.W.; Kabat, P. Climate variability and trends in Bolivia. J. Appl. Meteorol. Climatol. 2013, 52, 130–146. [Google Scholar] [CrossRef]
  51. Grimm, A.M.; Saboia, J.P.J. Interdecadal variability of the South American precipitation in the monsoon season. J. Clim. 2015, 28, 755–775. [Google Scholar]
  52. Hersbach, H.; Bell, B.; Berrisford, P.; Hirahara, S.; Horányi, A.; Muñoz-Sabater, J.; Nicolas, J.; Peubey, C.; Radu, R.; Schepers, D. The ERA5 global reanalysis. Q. J. R. Meteorol. Soc. 2020, 146, 1999–2049. [Google Scholar] [CrossRef]
  53. Harrison, S.P.; Bartlein, P.J.; Prentice, I.C. What have we learnt from palaeoclimate simulations? J. Quat. Sci. 2016, 31, 363–385. [Google Scholar] [CrossRef]
  54. Skamarock, W.C.; Klemp, J.B. A time-split nonhydrostatic atmospheric model for weather research and forecasting applications. J. Comput. Phys. 2008, 227, 3465–3485. [Google Scholar] [CrossRef]
  55. Birkel, S.; Mayewski, P.; Perry, L.; Seimon, A.; Andrade-Flores, M. Evaluation of reanalysis temperature and precipitationfor the Andean Altiplano and adjacent cordilleras. Earth Space Sci. 2022, 9, e2021EA001934. [Google Scholar] [CrossRef]
  56. Babaei, M.; Alizadeh, O.; Irannejad, P. Impacts of orography on large-scale atmospheric circulation: Application of a regional climate model. Clim. Dyn. 2021, 57, 1973–1992. [Google Scholar] [CrossRef]
  57. Schumacher, V.; Fernández, A.; Justino, F.; Comin, A. WRF high resolution dynamical downscaling of precipitation for the Central Andes of Chile and Argentina. Front. Earth Sci. 2020, 8, 328. [Google Scholar] [CrossRef]
  58. Hong, S.-Y.; Lim, K.-S.S.; Lee, Y.-H.; Ha, J.-C.; Kim, H.-W.; Ham, S.-J.; Dudhia, J. Evaluation of the WRF double-moment 6-class microphysics scheme for precipitating convection. Adv. Meteorol. 2010, 2010, 707253. [Google Scholar] [CrossRef]
  59. Mlawer, E.J.; Taubman, S.J.; Brown, P.D.; Iacono, M.J.; Clough, S.A. Radiative transfer for inhomogeneous atmospheres: RRTM, a validated correlated-k model for the longwave. J. Geophys. Res. Atmos. 1997, 102, 16663–16682. [Google Scholar] [CrossRef]
  60. Dudhia, J. Numerical study of convection observed during the winter monsoon experiment using a mesoscale two-dimensional model. J. Atmos. Sci. 1989, 46, 3077–3107. [Google Scholar] [CrossRef]
  61. Berg, L.K.; Zhong, S. Climatology. Sensitivity of MM5-simulated boundary layer characteristics to turbulence parameterizations. J. Appl. Meteorol. 2005, 44, 1467–1483. [Google Scholar] [CrossRef]
  62. Huang, M.; Huang, B.; Huang, A.H.-L. Implementation of the 5-layer thermal diffusion scheme in weather research and Forecasting model with Intel many integrated core (MIC) architecture. In High-Performance Computing in Remote Sensing IV; SPIE: Bellingham, DC, USA, 2014; pp. 67–75. [Google Scholar]
  63. Kain, J.S. The Kain–Fritsch convective parameterization: An update. J. Appl. Meteorol. 2004, 43, 170–181. [Google Scholar] [CrossRef]
  64. Rauscher, S.A.; Coppola, E.; Piani, C.; Giorgi, F. Resolution effects on regional climate model simulations of seasonal precipitation over Europe. Clim. Dyn. 2010, 35, 685–711. [Google Scholar] [CrossRef]
  65. Torrez-Rodriguez, L.; Goubanova, K.; Muñoz, C.; Montecinos, A. Evaluation of temperature and precipitation from CORDEX-CORE South America and Eta-RCM regional climate simulations over the complex terrain of Subtropical Chile. Clim. Dyn. 2023, 61, 3195–3221. [Google Scholar] [CrossRef]
  66. Caldwell, P.; Chin, H.-N.S.; Bader, D.C.; Bala, G.J. Evaluation of a WRF dynamical downscaling simulation over California. Clim. Change 2009, 95, 499–521. [Google Scholar] [CrossRef]
  67. Chaves, R.R.; Nobre, P.J. Interactions between sea surface temperature over the South Atlantic Ocean and the South Atlantic Convergence Zone. Geophys. Res. Lett. 2004, 31, L03204. [Google Scholar] [CrossRef]
  68. Martin, L.C.; Blard, P.-H.; Lavé, J.; Condom, T.; Prémaillon, M.; Jomelli, V.; Brunstein, D.; Lupker, M.; Charreau, J.; Mariotti, V.; et al. Lake Tauca highstand (Heinrich Stadial 1a) driven by a southward shift of the Bolivian High. Sci. Adv. 2018, 4, eaar2514. [Google Scholar] [CrossRef]
  69. de Mendonça Silva, I.; Medeiros, D.M.; Mendes, D.; Mesquita, M.d.S. Intraseasonal and interannual mechanisms of summer rainfall over Northeast Brazil using a hidden Markov model. Clim. Dyn. 2023, 61, 5805–5826. [Google Scholar] [CrossRef]
  70. Salvatteci, R.; Schneider, R.R.; Blanz, T.; Mollier-Vogel, E.J. Deglacial to Holocene ocean temperatures in the Humboldt Current System as indicated by alkenone paleothermometry. Geophys. Res. Lett. 2019, 46, 281–292. [Google Scholar] [CrossRef]
  71. Apaéstegui, J.; Cruz, F.W.; Vuille, M.; Fohlmeister, J.; Espinoza, J.C.; Sifeddine, A.; Strikis, N.; Guyot, J.L.; Ventura, R.; Cheng, H. Precipitation changes over the eastern Bolivian Andes inferred from speleothem (δ 18 O) records for the last 1400 years. Earth Planet. Sci. Lett. 2018, 494, 124–134. [Google Scholar] [CrossRef]
  72. Bird, B.W.; Abbott, M.B.; Vuille, M.; Rodbell, D.T.; Stansell, N.D.; Rosenmeier, M.F. A 2300-year-long annually resolved record of the South American summer monsoon from the Peruvian Andes. Proc. Natl. Acad. Sci. USA 2011, 108, 8583–8588. [Google Scholar] [CrossRef]
  73. Vuille, M.; Burns, S.; Taylor, B.; Cruz, F.; Bird, B.; Abbott, M.; Kanner, L.; Cheng, H.; Novello, V. A review of the South American monsoon history as recorded in stable isotopic proxies over the past two millennia. Clim. Past 2012, 8, 1309–1321. [Google Scholar]
  74. Mujica, M.I.; Latorre, C.; Maldonado, A.; González-Silvestre, L.; Pinto, R.; de Pol-Holz, R.; Santoro, C.M. Late Quaternary climate change, relict populations and present-day refugia in the northern Atacama Desert: A case study from Quebrada La Higuera (18° S). J. Biogeogr. 2015, 42, 76–88. [Google Scholar]
  75. Campos, J.; Cruz, F.; Ambrizzi, T.; Deininger, M.; Vuille, M.; Novello, V.F.; Strikis, N. Coherent South American Monsoon variability during the last millennium revealed through high-resolution proxy records. Geophys. Res. Lett. 2019, 46, 8261–8270. [Google Scholar] [CrossRef]
  76. Minvielle, M.; Garreaud, R.D. Projecting rainfall changes over the South American Altiplano. J. Clim. 2011, 24, 4577–4583. [Google Scholar] [CrossRef]
  77. Vera, C.S.; Díaz, L.B.; Saurral, R.I. Influence of anthropogenically-forced global warming and natural climate variability in the rainfall changes observed over the South American Altiplano. Front. Environ. Sci. 2019, 7, 87. [Google Scholar]
  78. Skamarock, W.C.; Klemp, J.B.; Dudhia, J.; Gill, D.O.; Barker, D.M.; Duda, M.G.; Huang, X.-Y.; Wang, W.; Powers, J.G. A description of the advanced research WRF version 3. NCAR Tech. Note 2008, 475, 113. [Google Scholar]
Atmosphere 16 00751 g002
Figure 2. Hydroclimate and oceanic anomalies (z-scores) during the DJF periods selected for the simulation experiments: (a) precipitation and SST anomalies for the 1983/1984 summer; (b) same as (a) for the 1997/1998 summer (control run); (c) same as (a) for the 2003/2004 summer (control run); (d) same as (a) for the 2011/2012 summer. All anomalies are calculated as standard deviation units from the 1951–2021 mean (z-scores). The area marked by the dashed black lines denotes the simulation domain used in the modeling experiment. All data correspond to the ERA5 reanalysis dataset.
Figure 2. Hydroclimate and oceanic anomalies (z-scores) during the DJF periods selected for the simulation experiments: (a) precipitation and SST anomalies for the 1983/1984 summer; (b) same as (a) for the 1997/1998 summer (control run); (c) same as (a) for the 2003/2004 summer (control run); (d) same as (a) for the 2011/2012 summer. All anomalies are calculated as standard deviation units from the 1951–2021 mean (z-scores). The area marked by the dashed black lines denotes the simulation domain used in the modeling experiment. All data correspond to the ERA5 reanalysis dataset.
Atmosphere 16 00751 g002
Atmosphere 16 00751 g003
Figure 3. Upscaling and model validation. Comparison between the 1983/1984 and 2011/2012 summers in the WRF simulation, the ERA5 reanalysis, and the CHIRPS precipitation datasets. (a) Left panel: total precipitation for the initial DJF period in the WRF 1983/1984 simulation. Mid panel: total 1983/1984 DJF precipitation in the ERA5 reanalysis. Right panel: total 1983/1984 DJF precipitation in the CHIRPS dataset. (b) Same as (a) for the southern Altiplano; (c) same as (a) but for the 2011/2011 DJF simulation; (d) same as (c) but for the southern Altiplano.
Figure 3. Upscaling and model validation. Comparison between the 1983/1984 and 2011/2012 summers in the WRF simulation, the ERA5 reanalysis, and the CHIRPS precipitation datasets. (a) Left panel: total precipitation for the initial DJF period in the WRF 1983/1984 simulation. Mid panel: total 1983/1984 DJF precipitation in the ERA5 reanalysis. Right panel: total 1983/1984 DJF precipitation in the CHIRPS dataset. (b) Same as (a) for the southern Altiplano; (c) same as (a) but for the 2011/2011 DJF simulation; (d) same as (c) but for the southern Altiplano.
Atmosphere 16 00751 g003
Atmosphere 16 00751 g004
Figure 4. Total summertime precipitation averages (mm) for the southern Altiplano region simulated in our four 100 yr WRF runs. (a) 1983/1984 simulation. (b) 1997/1998 simulation. (c) 2003/2004 simulation. (d) 2011/2012 simulation. The gray shading encompasses one standard deviation from the regional averages. The red rectangles denote the initial and final decades of the modeling runs used in Figure 5 and Figure 6.
Figure 4. Total summertime precipitation averages (mm) for the southern Altiplano region simulated in our four 100 yr WRF runs. (a) 1983/1984 simulation. (b) 1997/1998 simulation. (c) 2003/2004 simulation. (d) 2011/2012 simulation. The gray shading encompasses one standard deviation from the regional averages. The red rectangles denote the initial and final decades of the modeling runs used in Figure 5 and Figure 6.
Atmosphere 16 00751 g004
Atmosphere 16 00751 g005
Figure 5. Comparative panel depicting differences between the average atmospheric and moisture circulation during the initial and final decades of the 1983/1984 and 2011/2012 WRF simulations. (a) Difference between the final and initial decade of the 1983/1984 simulation for vertical motion at 500 hPa (colored) along with composite upper-level (200 hPa) zonal and meridional wind differences (streamlines). Negative (positive) differences indicate increased upward (downward) motion in the final decade relative to the initial decade, colored in blue (red). (b) Difference for specific humidity at 500 hPa (colored) along with composite lower-level (800 hPa) zonal and meridional wind differences (streamlines). The red rectangle denotes the region where the longitudinal mean anomalies are calculated for Figure 6. (c) Same as (a) for the 2011/2012 simulation; (d) same as (b) for the 2011/2012 simulation.
Figure 5. Comparative panel depicting differences between the average atmospheric and moisture circulation during the initial and final decades of the 1983/1984 and 2011/2012 WRF simulations. (a) Difference between the final and initial decade of the 1983/1984 simulation for vertical motion at 500 hPa (colored) along with composite upper-level (200 hPa) zonal and meridional wind differences (streamlines). Negative (positive) differences indicate increased upward (downward) motion in the final decade relative to the initial decade, colored in blue (red). (b) Difference for specific humidity at 500 hPa (colored) along with composite lower-level (800 hPa) zonal and meridional wind differences (streamlines). The red rectangle denotes the region where the longitudinal mean anomalies are calculated for Figure 6. (c) Same as (a) for the 2011/2012 simulation; (d) same as (b) for the 2011/2012 simulation.
Atmosphere 16 00751 g005
Atmosphere 16 00751 g006
Figure 6. Pressure (hPa)–latitude cross-section with anomalies (z-scores) for specific humidity (colored), integrated vertical and meridional circulation (vectors), and zonal winds (contours) for the initial (years 1 to 10) and final (years 91 to 100) decades for the 1983/1984 and 2011/2012 simulations. The longitudinal means for all variables are calculated over the region, delimited by the red rectangles in Figure 5c,d. Anomalies in the pressure–latitude sections were calculated as standard deviations from the 100 yr means. Continuous (dashed) contours indicate westerly (easterly) zonal wind anomalies. The red line represents the profile of the maximum elevation of the Andes Cordillera, and the position of the Altiplano is marked in yellow. (a,b) First and final decade of the 1983/1984 simulation. (c,d) First and last decades of the 2011/2012 simulation.
Figure 6. Pressure (hPa)–latitude cross-section with anomalies (z-scores) for specific humidity (colored), integrated vertical and meridional circulation (vectors), and zonal winds (contours) for the initial (years 1 to 10) and final (years 91 to 100) decades for the 1983/1984 and 2011/2012 simulations. The longitudinal means for all variables are calculated over the region, delimited by the red rectangles in Figure 5c,d. Anomalies in the pressure–latitude sections were calculated as standard deviations from the 100 yr means. Continuous (dashed) contours indicate westerly (easterly) zonal wind anomalies. The red line represents the profile of the maximum elevation of the Andes Cordillera, and the position of the Altiplano is marked in yellow. (a,b) First and final decade of the 1983/1984 simulation. (c,d) First and last decades of the 2011/2012 simulation.
Atmosphere 16 00751 g006
Atmosphere 16 00751 g007
Figure 7. Total DJF precipitation trends (mm per decade) in South America for (a) the 1983/1984 and (b) 2011/2012 simulations. Trends are calculated using a linear regression for the entire 100 yr simulation period.
Figure 7. Total DJF precipitation trends (mm per decade) in South America for (a) the 1983/1984 and (b) 2011/2012 simulations. Trends are calculated using a linear regression for the entire 100 yr simulation period.
Atmosphere 16 00751 g007
Table 1. Summary of precipitation and SST anomalies during DJF seasons selected as boundary conditions for the regional simulations.
Table 1. Summary of precipitation and SST anomalies during DJF seasons selected as boundary conditions for the regional simulations.
DJF Precipitation Anomaly (z-Score)DJF Nino 3.4 Index (z-Score)DJF Southeastern Pacific SST Anomaly (Figure 2)DJF South Atlantic Ocean
Dipole Index (SAODI)
1983/19832.4−0.5Warm0.5
1997/1998−2.42.3Cold0.92
2003/20040.230.4Cold0.63
2011/20121.7−0.7Warm−0.68
Table 2. Setup details for the four WRF simulation runs presented in the article.
Table 2. Setup details for the four WRF simulation runs presented in the article.
Simulation
Run		Initial
Date/Time		Final
Date/Time		Spin-Up
Time		Timesteps
(seconds)		Number
of Cycles
1983/198410 November 1983,
09:00		10 November 1984,
06:00		10 days90100
1997/1998
(control)		21 October 1997,
18:00		21 October 1998,
15:00		10 days90100
2003/2004
(control)		2 May 2003,
15:00		2 May 2004,
12:00		10 days90100
2011/201221 June 2011,
09:00		21 June 2012,
06:00		10 days90100
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Jara, I.A.; Astudillo, O.; Salinas, P.; Torrez-Rodríguez, L.; Lampe-Huenul, N.; Maldonado, A. Exploring the Causes of Multicentury Hydroclimate Anomalies in the South American Altiplano with an Idealized Climate Modeling Experiment. Atmosphere 2025, 16, 751. https://doi.org/10.3390/atmos16070751

AMA Style

Jara IA, Astudillo O, Salinas P, Torrez-Rodríguez L, Lampe-Huenul N, Maldonado A. Exploring the Causes of Multicentury Hydroclimate Anomalies in the South American Altiplano with an Idealized Climate Modeling Experiment. Atmosphere. 2025; 16(7):751. https://doi.org/10.3390/atmos16070751

Chicago/Turabian Style

Jara, Ignacio Alonso, Orlando Astudillo, Pablo Salinas, Limbert Torrez-Rodríguez, Nicolás Lampe-Huenul, and Antonio Maldonado. 2025. "Exploring the Causes of Multicentury Hydroclimate Anomalies in the South American Altiplano with an Idealized Climate Modeling Experiment" Atmosphere 16, no. 7: 751. https://doi.org/10.3390/atmos16070751

APA Style

Jara, I. A., Astudillo, O., Salinas, P., Torrez-Rodríguez, L., Lampe-Huenul, N., & Maldonado, A. (2025). Exploring the Causes of Multicentury Hydroclimate Anomalies in the South American Altiplano with an Idealized Climate Modeling Experiment. Atmosphere, 16(7), 751. https://doi.org/10.3390/atmos16070751

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Article metric data becomes available approximately 24 hours after publication online.
Back to TopTop