Revealing Paleo-Groundwater and Interbasin Flow as Fundamental to Resource Sustainability on the Arid Altiplano-Puna Plateau Brendan J. Moran1, David F. Boutt1, LeeAnn Munk2, Joshua Fisher3, Felicity Arengo4, Patricia Marconi5, Diego Frau6 1Univeristy of Massachusetts, Amherst; [email protected]. 2University of Alaska, Anchorage. 3Advanced Consortium on Cooperation, Conflict, and Complexity, Earth Institute, Columbia University, 4Center for Biodiversity and Conservation, American Museum of Natural History, 5Fundacion Yuchan, Salta, Argentina, 6Instituto Nacional de Limnologia/CONICET, Universidad del Litoral, Santa Fe, Argentina Motivation -& Objective Results Discussion

Water resources on the arid high-Andean plateau are critical to sustaining both indigenous 25%~75% Range within 1.5IQR 25%~75% Range within 1.5IQR 18 5000 Median Line 87 86 Median Line Mean Right: δ O data vs 0.72200 0.72200 Mean Left: Sr/ Sr data from communities and fragile Ramsar World Heritage ecosystems yet accelerating demand for Outliers Tres Quebradas Inflow Puna Basement Rocks Outliers elevation of sampling. 4500 mineral resources and the effects of climate change have led to concerns about the 0.72000 0.72000 this work. Potential Lapse rate estimated for 4000 sustainability of these resources. Persistent and fundamental questions regarding the source 0.71800 0.71800 end-member signatures this region from 3500 and movement of groundwaters, which sustain most surface waters here make managing 0.71600 0.71600 indicted by dashed lines. Sr Sr Rohrmann et al., 2014. 3000 86 86 0.71400 0.71400 Right: O and H isotope Sr/ these resources particularly difficult. Sr/ 87 87 0.71200 0.71200 data from each basin, each Ellipses indicate waters 2500 Estimated Lapse rate We seek to address the following questions: Puna Ignimbrites & Ashes at similar elevation with (Rohrmann et al., 2014) 0.71000 0.71000 with different color/shape. 2000 H H -(‰) VSMOW Sample Elevation Elevation (mamsl) Sample 2

1. What is the nature of hydrogeologic connectivity within the plateau; between topographically δ 0.70800 0.70800 Local Evaporation Lines distinct signatures. Blue closed basins and between modern infiltration (<60 yrs.) and the paleo-groundwater system? SdA Brines 1500 0.70600 0.70600 indicate predicted arrows are potential 1000 Snow Thermal Spring Vega Stream Lagoon Brine 3000-3500 3500-4000 4000-4600 2. How connected are surface water bodies (wetlands, lakes, salt lakes and salars) on the Puna to the Spring -14.0 -13.0 -12.0 -11.0 -10.0 -9.0 -8.0 -7.0 -6.0 -5.0 -4.0 Elevation (mamsl) signature of source waters. connections with waters δ18O - VSMOW (‰) groundwater (aquifers) and what is distribution and magnitude of these connections? 69°0'0"W 68°0'0"W 67°0'0"W down-gradient. Rain Snow Hombre Muerto Caro Antofalla Grande/Diamante 3. What are the dynamic response times of surface and groundwaters to perturbations from climate δ18O - VSMOW (‰) Carachipampa N Carchipampa NE 3Q Los Aparejos Rio Chaschuil Rio Vilavil Hombre Muerto Caro Antofalla Grande/Diamante Carachipampa N Carachipampa NE 3Q Los Aparejos Rio Chaschuil Rio Vilavil change and groundwater extractions? 25°0'0"S [1] Water Types 80 # Streams 87 86 Right: Piper diagram of ) Ca 80 Evaporation Mineral Precipitation - ! # Springs Left: Sr/ Sr v. 1/Sr Salar de ++ 60 # Lagoons + Mg Puna data collected in this study 60 # Groundwater + Cl Background Hombre Muerto -- 0.72200 Basement from water across the 1 Snow 440 Ca-SO4 69°0'0"W 68°0'0"W 67°0'0"W 66°0'0"W ! ++ High Elevation Basins Laguna Grande Spring ! 40 and available published SO 1 ! Low Elevation Basins study area. Signature Left: Regional map !!! Tres Quebradas 0.72000 Salar de Atacama 20 0"!!!! Laguna Grande Laguna ! 0!!!! brines Major Basins Hombre Muerto major streams #" 0! 20 # data. Data shapes identify NE Carachipampa * !!! Hombre Muerto of Altiplano Puna - ) " Carachipampa from major rock types ) # "! ! 0.71800 Caro " # ' #* Carachipampa ! Tres Quebradas El Penon Springs " ! # basin, colors are water brines Ca-HCO3 " ! (Meixner et al, 2019) in plateau, basins #) Na-Cl # 0.71600 Carachipampa Northern Carachipampa "" , #1# Carachipampa high Spring ! + Sr 24°0'0"S # type. Important water # # ,#, discussed in this ,+ elevation basins 86 this region and general + 0.71400 , -- 20

3 Sr/ group and evolutionary 20 Na 80 Na-HCO3 80 87 work, sample Salar de + ) Puna end-member mixing lines + K 40 SO 0.71200 - + CO + Rio ++ Ignimbrites Rio ) 40 3 pathways are labeled and 60 + 60 -- locations in this Antofalla 4 & Ashes noted. Major distinctions Mg 60 1 Aguas Los Patos HCO " 0.71000 ! 60 colored by basin. 40 ) 0 * # 40 work and location of - # in water-rock interactions Calientes ) ) 80 ' !# # " " ! " 0.70800 Salar de Cerro Galan * ' 80 # " !" 25°0'0"S " # 20 - ## #"#! ) " ) 20 profiles presented Depletion "##00 ! ## #,+ , #! Thermal waters along flow paths are Hombre Muerto #!!"!" ## " #) +,, ,0!0 #"#" !!!!!!,,,!"* ##), "! 0!!*! ! 1!!1!!!!!!!!!!!!!!!!$!!+!#,#)!,+ # 1 ")!!!!!! 0.70600 Laguna Caro 80 60 40 20 $ herein. 0 2 4 6 8 10 12 identifiable here. 20 40 - 60 80 Salar de ++ Cl 1/Sr (ppm)

26°0'0"S Antofagasta Ca Antofalla Below: Average precipitation region-wide 1998-2009 de la Sierra [3] Stream Spring Vega Lagoon from the TRMM satellite, reanalyzed by Bookhagen at % meq/kg Wet Period Wet Period 25%~75% 18 Range within 1.5IQR 26°0'0"S al., (in review). Data points are lc-excess filtered δ O -5 18 Median Line Antofagasta 0.6 Mean Laguna Grande Laguna Left: Time series of δ O 0.6 de la Sierra values from groundwater and spring waters. Outliers 69°0'0"W 68°0'0"W 67°0'0"W 66°0'0"W [4] Grande CarachiPampa El Penon in spring and stream Salar δ18O ‰ 0.5 0.5 Springs & Gw -6 (lc-excess filtered) Carachipampa waters. Color of line -13.9 - -10.9 23°0'0"S -10.89 - -9.4 El Penon Salar corresponds to basin (same 0.4 0.4 -9.39 - -8.1 27°0'0"S -8.09 - -6.7 SALLJ 18 -7 Tres Quebradas -6.69 - -4.8 mod as plot above) and arrows mod 0.3 Salar Limit of Interally -4.79 - -0.6 0.3

TRMM2b31 H, R H, R Groundwater 3 Drained Area Precipitation ? to map indicate location of 3 Major Fault Spring Systems mm/year Stream Major Drainges Lagoon -8 0.2 0.2 Salars 24°0'0"S Enrichment notable stream sites. Brine 500 250 0 Thermal water Lagoons Precipitation Watersheds Laguna Los Aparejos 012.5 25 50 75 100 Rio

28°0'0"S Kilometers 0.1 0.1 Vilavil -9 June 2011 June 2012 February 2018 February 2019 Tres Quebradas 0.0 0.0 Tothian interbasin flow regime 27°0'0"S 3Q to Cortaderas Date W E Spring Vega Stream Lagoon Brine 3000-3500 3500-4000 4000-4600 25°0'0"S Salar Left: 100000 [3] Elevation (mamsl) Left: Site map of study area. 10000 1000 (mg/l)

Conceptualization - 100 R statistics from this Limit of Interally Topographic watersheds are Cl Above: mod of regional flow 10 Chaco Drained Area outlined in red. Important study by water type grouping and -4

from Toth (1963). 26°0'0"S -5 -6 elevation of sampling. basins, rivers and communities -7 -8 These flow O (‰) -9 18 18 δ O R in waters calculated from δ Left: are labeled; colors of basin -10 δ18O Low Elevation mod -11 High Elevation Pampa moisture regimes commonly -12 Pacific moisture 3 Rio labels correspond to data across -13 H. Statistics are from three regions 5,000 5,000 develop long, Watershed Streams Chaschuil boundary 27°0'0"S [2] 18 2 this work. Transects presented Springs of comparable size. The Great Basin δ O, δ H Samples 4,750 4,750 interbasin flow Salars Laguna Chaco ? Major Dissolved Ions Samples Laguna Lakes and Central Plains data are below are numbered. 3H percent Los Aparejos 4,500 4,500 Negra Vega paths in arid 87 86 3 3Q Salar groundwaters. Data extracted from Sr/ Sr H, Rmod modern content of water shown 4,250 4,250 mountainous areas. 015 30 60 90 120 Kilometers 0.70786 - 0.70866 50.1 - 59% 87 86 ? Avg. = 0.23 NWIS dataset and this study as relative circle size; Sr/ Sr 4,000 4,000 Laguna Los Avg. = 0.21 0.70866 - 0.71076 40.1 - 50% Avg. = 0.14 (including Salar de Atacama region) Aparejos data shown on color scale. Blue 3,750 3,750 28°0'0"S 0.71076 - 0.71373 30.1 - 40% from Moran et al., 2019. R in arrows show major modern flow 3,500 3,500 mod

Conceptual Framework Elevation (mamsl) 0.71373 - 0.71749 20.1 - 30% Rio 3,250 Chaschuil 3,250 precipitation value used is from (a) N paths, green hollow arrows are 0.71749 - 0.72057 10.1 - 20% Left: Conceptual NE major interbasin or 3,000 3,000 Great Basin, USA Central Plains, USA Altiplano-Puna Plateau Lindsey et al., 2014 model of the SE 0 - 10% S 2,750 paleo-groundwater flow path. 2,750 Plateau 0 10 20 30 40 50 60 22°30'0"S 0.60 δ18O hydrogeologic flow Major Fault 010 20 40 60 80 N Major Drainges Salars Lagoons Watersheds Distance (km) Systems 3 Kilometers H, Rmod: 0% 10% 20% 30% 40% 50% Snowmelt regime of the Salar de 0.50 Carchi Pampa to Cerro Galan Atacama (SdA) basin Carachi Pampa to Hombre Muerto N S NE Northern Plateau NW Salar de Atacama to Laguna Blanca SE S 0.40 Rain 100000

23°0'0"S Hyper-arid to [2] Rio Los Patos from Moran et al., [1] δ18O 100000 the west. NE -4 10000 [4] mod Negligible 10000 Transition 0.30 recharge 2019; in profile view -5 1000 1000 (mg/l) 18 H, R Mixing

Zone - and inflow to δ O 3 Mixing

-6 (mg/l) 100 100 Cl the system - 18 Altiplano/Puna below. Watershed is 10 0.20 δ O -7 Cl ~4,200 m a.s.l. O (‰) 10 Nucleus 1 Mixing 18 Rainout -6.2

δ -8 23°30'0"S 2,300 m a.s.l. solid black outline, 0.1 -6.4 Peine 18 -6.6 0.10 δ O (rainout) - -6.4 Block -9 -6.8 -6.6 SE sub-watersheds are - -7.0 10 -7.2

-6.8 O (‰) Southern Plateau δ18O 0.00 11 - 18 18 -7.4

-7.0 δ O δ colored. Rainout -7.6 -16.00 -14.00 -12.00 -10.00 -8.00 -6.00 -4.00 -2.00 S O (‰) -7.2 -7.8 18 12 18 δ O 18 -7.4 -8.0 H2O-δ O (‰) 24°0'0"S 5,500 δ 6,000 6,000 Above right: Major -7.6 Cerro Galan Streams Watershed Springs Vegas Streams 5,750 boundary 5,750 hydrochemical facies 5,000 5,000 5,000 Springs Rio Streams Watershed Salars 18 Lg. Vega Rio 5,500 5,500 Punilla 4,750 boundary Rio 4,750 Lakes Above: Predominately fossil waters (values below ~10% R ) are related to their δ O within SdA catchment. Vega Grande Springs Trapiche mod Below: Percent modern water in 86 sam- 4,500 Punilla 5,250 5,250 ? 4,500 Salars 4,500 18 Lakes Lakes Headwaters Laguna 24°30'0"S 3 Salar de signatures. Waters with high percentage of modern water and relatively low δ O values Colors correspond to 5,000 Diamante 5,000 ples determined by H decay. The salar Hombre 012.5 25 50 75 100 4,000 4,250 4,250 18 km Salar de Antofagasta Muerto 4,750 4,750 indicate a high elevation source (likely snow-melt) while rainwater has higher δ O sub-watersheds. 4,000 4,000 Laguna Laguna 68°30'0"W 68°0'0"W 67°30'0"W 67°0'0"W nucleus and sub-watersheds are shaded. Antofalla Laguna de la Sierra Grande 3,500 4,500 Cabi 4,500 (b) P Blanca 3,750 3,750 values. Potential mixing between these waters can be identified. Divide PPlateau ? 18 Laguna δ O 18 δ O Rmodern Content 4,250 4,250

Plocal Elevation (mamsl) Alumbrera Precipitation 3,500 3,500 18 Below: Same data as 3,000 δ O Cerro Toco N Elevation (mamsl) Carachi Pampa 0.0 - 0.06 Salar 4,000 4,000 Rio 3,250 Salar 3,250 Elevation (mamsl) Vega 0.07 - 0.13 de Atacama Vilavil GWR GWR at right, presented 2,500 3,750 3,750 LGM water table 0.14 - 0.20 3,000 3,000 References Peine Block El Penon

Modern water table (falling) 23° S 0.20 - 0.26 3,500 3,500 Contributions from along transect through 2,750 2,750 0.27 - 0.59 Pena Blanca (n=2) 2,000 Bookhagen, B. (in review): High resolution spatiotemporal distribution of rainfall seasonality and extreme events hydrogeologic watershed 0 20 120 3,250 Carachipampa 3,250 Topographic Watershed Grosjean et al. 1995 40 60 80 100 0 50 100 150 200 250 300 350 400 3 Salar Vertically exaggerated, water table depths not to scale SE (blue) zone. Gw Sw H, R : 0% 10% 20% based on a 12-year TRMM time series, in review. Pena Blanca Spring 3 mod 30% 40% 50% Distance (km) H, Rmod: 0% 10% 20% 30% 40% 50% Distance (km) 3,000 3,000 1.8 25%~75% Tóth, J. (1963). A theoretical analysis of groundwater flow in small drainage basins. Journal of Geophysical Re- (a) (b) Range within 1.5IQR 5 (n=2) 2,750 2,750 1.6 High Elevation Lakes & Lagoons W Median Line E 0 10 20 30 40 50 60 70 80 search, 68(16), 4795–4812. https://doi.org/10.1029/jz068i016p04795 Mean NE Above & Right: Cross-sections through study region, labeled 1-4 and on map. Red circle size relative to Rmod content in waters. 1.4 Outliers 0.37 [SC 19] Distance (km) 4,200 18 - 3 Moran, B. J., Boutt, D., & Munk, L. A. (2019). Stable and Radioisotope Systematics Reveal Fossil Water as Funda- H, Rmod: 0% 10% 20% 30% 40% 50% 1.2 Talabre Spring Blue arrows indicate modern water transport, green arrows indicate interbasin flow. δ O and Cl along transect are plotted above. mental Characteristic of Arid Orogenic-Scale Groundwater Systems. Water Resources Research,. https://- 4,000 1.0 High Elevation Inflow 76 102 0.06 [SC:1] 7 (n=2) doi.org/10.1029/2019WR026386. H (TU) 0.8 78 (n=2) 3,800 3 Toro Blanco Spring 81 0.6 Rohrmann, A., Strecker, M. R., Bookhagen, B., Mulch, A., Sachse, D., Pingel, H., … Montero, C. (2014). Can stable 82 Lejía gw (n=3) Conclusions 3,600 0.4 83 & 3 186 139 Laguna Lejía isotopes ride out the storms? The role of convection for water isotopes in models, records, and paleoaltimetry studies in 0.2 18 87 86 84 Lejía laguna (n=2) the central Andes. Earth and Planetary Science Letters, 407, 187–195. https://doi.org/10.1016/j.epsl.2014.09.021. 3,400 Nucleus 1. An exhaustive set (~350) of environmental tracer data (δ O, δ2H, 3H, Sr/ Sr), and dissolved major ions in waters across this integrated system 0.0 Avg. Nucleus MP-12 Avg. Marginal 13(n=2) Nucleus Marginal Trans. Lagoons Trans. South East High Elev.High Elev. Brine (n=11) 187 Godfrey, L. V., Jordan, T. E., Lowenstein, T. K., & Alonso, R. L. (2003). Stable isotope constraints on the transport of Brine Brine Pools Zone Inflow Inflow Inflow Lakes Brine (n=10) 3,200 n=12 n=10 n=10 n=4 n=8 n=9 n=25 n=2 n=2 2 188 & 190 SDA216 reveals substantial spatial heterogeneity in both interbasin and modern, shallow flow regimes; controlled by geologic structure and topographic features. water to the Andes between 22° and 26°S during the last glacial cycle. In Palaeogeography, Palaeoclimatology, Palaeo- Elevation(mamsl) Marginal Brine Avg. Trans. Pools MP-08 9 (n=2) Nuclues (n=10) PP-01 12 3,000 0.23 [SC:239] Avg. Lagoons 195 MP-07 Laguna Miscanti ecology (Vol. 194, pp. 299–317). Elsevier B.V. https://doi.org/10.1016/S0031-0182(03)00283-9. Brine (n=4) PP-03 Avg. Trans. Zone 70 11 2. Pre-modern ‘fossil’ groundwater is fundamental in this system, most of the water discharging to large basin floors is composed of fossil water. The 0.09 [SC:230] Transitional Pools (n=8) 197 8 (n=3) Lindsey, B.D., Jurgens, B.C., and Belitz, K., 2019, Tritium as an indicator of modern, mixed, and premodern ground- 2,800 69 185 Laguna Miñiques Lagoons Large MRTs MP-09 (n=2) 0.30 [SC:237] Low Elevation Inflow 180 (n=2) SE modern and fossil flow systems have very distinct transit time distributions and therefore sharp disconnects over short distances exists between them. water age: U.S. Geological Survey Scientific Investigations Report 2019–5090, 18 p., https://doi.org/10.3133/ 2,600 Transition Non-modern Plateau 0.18 [SC:60] Zone 0.03 [SC:8] sir20195090. Water 2,400 0.04 [SC:29] South 3. Our conceptual model of this integrated hydrologic system characterizes spatiotemporal connections. Using this understanding, potential impacts on Meixner, A., Sarchi, C., Lucassen, F., Becchio, R., Caffe, P. J., Lindsay, J., … Kasemann, S. A. (2019). Lithium con-

24° S centrations and isotope signatures of Palaeozoic basement rocks and Cenozoic volcanic rocks from the Central Andean 2,200 122 05 10 20 30 40 critical and threatened wetland ecosystems and water resources from development or climate change scenarios can be greatly improved 5 0 5 10 15 20 25 30 35 40 45 50 km arc and back-arc. Mineralium Deposita. https://doi.org/10.1007/s00126-019-00915-2 Distance from Transitional Pools (km) 68° W