The contribution of volatile organic compounds (VOC) to ozone continues to receive attention and various localities have successively introduced measures to control VOC including limiting emissions from gas stations. Xi’an, as a major city of Fenwei Plain, has a large number of motor vehicles and a large demand for gasoline, and the VOC emission from gas stations can not be ignored in the city. In order to understand its impact on ozone and provide a reasonable basis for gas stations control and governance, this paper combines the SMOKE model to construct a gridded Xi’an gas stations VOC emissions, and uses WRF-CAMx model to study its effect on ozone concentration in summer. The results show that: (1) After considering the VOC emissions from gas stations, the concentrations of ozone (O3) and secondary organic aerosol (SOA) simulated by the sensitivity test in July 2018 were slightly higher than that results without the gas stations VOC emissions. (2) Similar to the baseline scenario, the daily change of O3 in July 2018 simulated by the control test considering gas stations VOC emissions shows a unimodal distribution; the diurnal change of secondary organic aerosol (SOA) is characterized by a high concentration in the morning even a peak. While the overall concentration level decreased in the afternoon, and gradually increased in the evening. (3) The results of sensitivity numerical test show that the gas stations VOC emissions in Xi’an just has a little influence on air quality, 63.5% hours its effect on ozone is within 0.01 μg/m³, and the VOC emissions from gas stations has different influence on ozone concentration in different hours. The maximum influence on hourly ozone concentration in Xi ’an is about 0.20 μg/m³.

This presentation showcases a hydrodynamic assessment of natural and nature-based features (NNBFs) for the Pascagoula River, the Escatawpa River and Grand Bay, located along the Mississippi coast of the northern Gulf of Mexico. Two separate NNBF projects are being considered to: (1) restore the historical footprint (ca. 1848) of Grand Batture Island for coastal protection purposes; and (2) reconnect the hydraulics between the Escatawpa River and Grand Bay for ecosystem services purposes. The intended coastal protection benefits of the first project include buffering agency to wave attack and attenuation of storm surge with the restored island. The intended ecosystem services benefits of the second project include replenishment of sediments to the salt marsh via increased hydroperiod (duration of tidal inundation) and availability for sediment accumulation. Astronomic tide and storm surge simulations are performed with the advanced circulation (ADCIRC) plus simulating waves nearshore (+SWAN) model to evaluate the hydrodynamic impact of the NNBF projects (Image). The simulated hydrodynamics are assessed firstly in terms of storm surge and waves for the open coast with and without the restoration of Grand Batture Island (Passeri et al., 2015), and secondly for tidal datums and inundation extent for the salt marsh with and without the hydraulic reconnection of the Escatawpa River with Grand Bay (Alizad et al., 2018). A key outcome from the analysis is the interconnectedness of the hydrodynamics within the system, where the implementation of the NNBFs results in local and non-local impacts. The numerical modeling approach with high-resolution feature definition at a system-wide scale affords such methodical evaluation of NNBFs for ecosystem restoration.

Taking machine learning models from conceptualization to production is a complex and often time-consuming practice. Solar flare prediction is a central problem in space weather forecasting and has piqued the interest of many researchers in recent years. The prediction efforts have been catalyzed by the recent advancements in machine learning and deep learning methods and the experimental results show notable performance improvements. On the other hand, operationalizing these models and building well-documented, reliable cyberinfrastructure from them remains to be a challenging issue. We will present an example training and deployment scenario for a solar flare prediction system prototype with two different modes of prediction: full-disk and active region-based. We will demonstrate the challenges we faced during the development lifecycle including the data preprocessing and integration, model training and optimization, validation, and reporting. We will also show the results from our hybrid-mode flare prediction method and factors impacting the real-life performance of our cyberinfrastructure services. The poster is available here: AGU Fall Meeting 2021.

Many of Antarctica's ice shelves experience significant surface melting each season, yet it is generally assumed that nearly all the produced meltwater re-freezes rather than being lost as runoff. Individual events of direct surface meltwater loss to the ocean have been documented, but were thought to be rare. We here present evidence of widespread through-ice-shelf drainage of supraglacial lakes, observed across various Antarctic ice shelves. We demonstrate that meltwater can accumulate for many melt seasons, in lakes that are often covered by an ice lid. Such buried lakes can drain into the sub ice shelf ocean cavity nearly instantaneously, creating an ice doline. These doline formation events are observed year-round and likely occur via a through-cutting crevasse that propagated from the lake bed by hydrofracture. The removal of a large load from the top of the floating ice shelf results in flexural uplift in the region around the drained lake basin. These surface elevation changes can be monitored using high-resolution satellite remote sensing data, such as ICESat-2 laser altimetry and digital elevation models created from WorldView stereo satellite imagery pairs. We show that doline formation has the potential significantly change the surrounding ice shelf surface hydrology, leading to different processes with opposing effects on ice shelf stability. If the surface depression formed by the former lake bed acts to capture meltwater and channels it straight down to the ocean via the newly-created drainage pathway, this process could halt supraglacial meltwater system growth and thus prevent meltwater from reaching areas more vulnerable to hydrofracture. However, if flexural uplift diverts meltwater flow around the doline and delivers it to adjacent areas on the ice shelf surface, this can lead to repeated hydrofracture in multiple locations and thus act to destabilize the ice shelf. Which of these mechanisms dominates over time on each ice shelf may ultimately be crucial for its long-term resilience in a warming climate.

Earth’s transient climate sensitivity Str is the rapid change, plateauing at ~5 yr, in global mean surface temperature GMST per change in forcing (e.g., Held et al., JGR, 2010). Str is readily evaluated from time series of total forcing F and temperature anomaly ΔT as the slope of a regression of ΔT vs F, with ΔT from model or observations and F generally modeled based on change in atmospheric composition. Prior estimates of Str have varied quite widely, mainly because of uncertainty in aerosol forcing. F is evaluated as total non-aerosol forcing, dominated by positive GHG forcing, plus negative aerosol forcing; large magnitude aerosol forcing results in small F and in turn high Str, and vice versa. Forcing time series derived from the Fifth IPCC Assessment Report (AR5, 2013) resulted in best estimate Str 0.35 K (W m-2)-1; 5% to 95% uncertainty range 0. 27 to 0.55 (Schwartz, JGR, 2018). New time series of total forcing from the (2021) Sixth IPCC Assessment report (AR6) permit similar evaluation of Str, Figure 1, as 0.46 (0.36 to 0.50) K (W m-2)-1. The increase in best-estimate Str is due to increased magnitude of best-estimate aerosol forcing in AR6 vs. AR5. Poor long-term correlation of forcing time series and observed ΔT for the 5% forcing estimate (large negative aerosol forcing added to GHG forcing, yielding low total forcing) suggests that that the corresponding bound on aerosol forcing magnitude may be an over-estimate, with the correlation substantially improved for best estimate and even more so for lowest estimate of aerosol forcing magnitude, thus more consistent with lower values of Str. A somewhat higher range of Str, 0.42 to 0.75 K (W m-2)-1, is obtained using time series of forcings obtained with individual models (Smith et al., ACP, 2020). Figure 1. Time series of total forcing F and as convolved with 5-year decaying exponential Fc (left); correlations of observed temperature anomaly ΔT (GISS) vs Fc (center); slope denotes transient sensitivity Str; and time series of ΔT (right; left axis) and Fc (right axis, scaled to ΔT by Str). Top row, lower 5% bound on forcing time series; middle row, best estimate forcing; bottom row, 95% bound. Forcing data from draft AR6 report, expected release August 9, 2021, potentially subject to change.

A statistical-equilibrium, geostrophic-turbulence regime of the stochastically forced, one-layer, reduced-gravity, quasi-geostrophic model is identified in which the numerical solutions are representative of global mean, mid-latitude, open-ocean mesoscale variability. A nominal best fit to observed SSH variance, autocorrelation, eddy, and spectral statistics is obtained for dimensional SSH stochastic-forcing variance production rate 1/4 cm^2/d, an SSH damping rate 1/62 1/wk, and a stochastic forcing autocorrelation timescale near or greater than 1 wk. This ocean mesoscale regime is nonlinear and appears to fall near the stochastic limit, at which wave-mean interaction is just strong enough to begin to reduce the local mesoscale variance production, but is still weak relative to the overall nonlinearity. Comparison of linearly-inverted wavenumber-frequency spectra shows that a strong effect of nonlinearity, the removal of energy from the resonant linear wave field, is resolved by the gridded altimeter SSH data. These inversions further suggest a possible signature in the merged altimeter SSH dataset of signal propagation characteristics from the objective analysis procedure.

Solar X-ray flares often drive coronal mass ejections and solar energetic particles (SEPs), which are key elements of space weather near the Earth and beyond. Machine learning with capability to improve space weather forecasting, prevents the adverse consequences of the space weather phenomena. We study temporal parameters of the solar X-ray flares associated with the SEP events. The temporal profile of the X-ray flare is usually divided into rising and decaying phases each approximated by a single functional representation. We observe a more complex nature of the rising and decaying phase for number of X-ray flares, which includes a pre-flare increase of intensity and a break in the decay phase. We develop a method to define, derive, and provide an optimized set of X-ray flare temporal profile parameters which takes into account the complex nature of the flare. We create a set of X-ray flare temporal properties which we next apply to the machine learning algorithms for a better forecasting of X-ray flare evolution in time and the associated SEP events.

Seismic-frequency torsional forced-oscillation measurements of shear modulus dispersion and associated strain-energy dissipation under conditions of simultaneously high temperature and pressure in our laboratory require the specimen to be enclosed within a thin-walled metal sleeve – typically of mild steel. However, the well-known transition between the austenite and ferrite phases on cooling across the interval 900-700ºC complicates its mechanical behaviour. The resulting uncertainty tends to mask the seismologically important onset, within this temperature range, of appreciably anelastic behaviour of polycrystalline olivine. In order to more closely document this important transition in mechanical behaviour, we have conducted an unpublished study in which of a specimen of polycrystalline olivine is jacketed within a copper sleeve which retains its face-centred-cubic (fcc) structure throughout the range of the measurements limited to 1050ºC by the proximity of the melting point. Here we report measurements to higher temperature (1200ºC) in which we employ austenitic (fcc) stainless steel (SS) as an alternative jacket material. Two parallel reference experiments with a SS-jacketed SS specimen and a SS-jacketed sapphire specimen were also conducted to obtain the mechanical properties of the jacket material. Such information concerning the viscoelastic behaviour of SS is used to subtract the torsional stiffness of the SS jacket from the properties of SS-jacketed olivine and thereby isolate the mechanical behaviour of olivine. The resulting dissipation spectrum for the olivine specimen consists of a monotonic background dissipation with a superimposed peak located within the 1-1000 s period range for temperatures between 900 and 1050ºC. The dissipation peak and associated shear modulus dispersion, potentially attributable to elastically accommodated grain boundary sliding, display an Arrhenius dependence upon temperature - moving systematically to shorter periods with increasing temperature. Comparison of the results obtained for sol-gel Fo90 olivine specimens enclosed within the alternative mild-steel, copper, and stainless-steel jackets will provide new insight into the nature of the seismologically important transition between the elastic and anelastic regimes.

Assessment and management of limited fresh groundwater resources on remote small islands can be complicated by heterogenous geology, natural climate cycles, and a general lack of data. The Palmyra Atoll National Wildlife Refuge is home to one of the few surviving native stands of Pisonia grandis in the central Pacific Ocean, yet these trees face pressure from groundwater salinization, with limited basic groundwater data to guide decision making. Adding to natural complexity, the geology of Palmyra was heavily altered by dredge and fill activities in the 1940s. We combined electromagnetic imaging (EMI) and hydrological field measurements from 2008-2019 with groundwater modeling to map the current distribution of fresh groundwater on the modified main island and a small, more natural islet to better understand potential physical drivers of spatiotemporal variability. Frequency-domain EMI data were collected on the main atoll islands over repeat transects in 2008 following ‘strong’ La Niña conditions (wet) and 2016 during ‘very strong’ El Niño conditions (dry). Shallow monitoring wells were installed adjacent to the geophysical transects in 2013 and screened within the fresh/saline groundwater transition zone. Temporal EMI and monitoring well data showed a strong lateral and vertical contraction of the freshwater lens in response to El Niño conditions, and transient EMI data indicated a thicker lens toward the ocean side, an opposite spatial pattern to that observed for many other Pacific islands. On an outer islet where a stand of mature Pisonia trees exist, EMI surveys revealed only a thin (<3 m from land surface) layer of brackish groundwater during El Niño. Numerical groundwater simulations were performed for a range of permeability distributions and climate conditions at Palmyra. Results revealed that the observed atypical lens asymmetry is likely due to a combination of lagoon dredging and filling with high-permeability material, allowing for more efficient submarine groundwater discharge on the lagoon side. Simulations also predict large negative changes (approximately 40% decrease) in freshwater lens volume during dry cycles and highlight threats to the Pisonia trees, yielding insight for atoll ecosystem management on Palmyra and other small Pacific islands.

Topographic maps represent a three-dimensional (3D) surface using a system of symbols in two-dimensions (2D). To facilitate students’ understanding of topographic maps, the Augmented Reality (AR) Sandbox reads the elevation of actual sand and projects the topographic map lines onto the surface of the sand. Although over 600 institutions have built AR Sandboxes to help people interpret topographic maps, classroom studies using the AR Sandbox have not found significant gains on topographic map assessments. The present study is a 2 by 2 design testing the affordances of the AR Sandbox in a laboratory setting. In the first level of the study, participants interacted with the AR sandbox (3D) or a regular computer monitor (2D) to give them feedback on five landforms they constructed in the sand. Participants initially constructed the landforms in the sand with the feedback off (i.e., sand without the overlaid projection or monitor). The feedback was then turned on and participants compared and contrasted their landform to the target topographic map. Participants were then asked to modify their landform with the feedback on (continuous), or the feedback was turned off (discrete) during modification. A mixed-ANOVA revealed significant gains on a modified version of the Topographic Map Assessment (TMA-B) from pre- to post-intervention (F(1, 74) = 80.34, p < .001). A significant interaction revealed that participants in the 2D condition had greater gains (M = 2.91, SD = 2.48) than those in the 3D (M = 1.64, SD = 2.07) condition (F(1, 74) = 6.38, p = .014), although both conditions had significant pre- to post-intervention improvement (2D: t(37) = 7.24, p < .001, d = 1.02; 3D: t(39) = 5.01, p < .001, d = 0.64). On average, the discrete feedback groups spent significantly less intervention time (M = 48.3, SD = 16.9) compared to the continuous groups (M = 58.2, SD = 18.1) (F(1, 76) = 6.20, p = .015). The findings suggest that the AR Sandbox does improve topographic map skill for individual students using our approach, and that the most efficient technique engages students in discrete episodes of feedback using the 2D computer monitor.

A new family of analytical magnetohydrostatic equilibria is presented. The methodology used to calculated these equilibria is based on work by Low (e.g. Low 1991) in the form given by Neukirch and Rastätter (1999). The equilibria have the capability of combining a non-force-free layer with a domain of linear force-free magnetic field, including a transition between these two domains. Potential applications to modelling solar magnetic fields will be discussed. Low, B.C., ApJ 370, 427 Neukirch, T. & Rastätter, L., A&A 348, 1000

There is a lot going on in introductory undergraduate science classrooms. Students bring their culture, background, and previous science experiences; instructors bring their knowledge, attitudes, and experience in science and in teaching. Students are there for a variety of reasons, and a substantial proportion will become K-12 teachers: that introductory course may be their primary science experience as an adult learner. How future teachers learn science is of critical importance to how they teach science, but few college science classes reflect the vision of the 2012 Framework for K-12 Science Education, in which “students actively engage in scientific and engineering practices in order to deepen their understanding of crosscutting concepts and disciplinary core ideas” (p. 217). The vision of the Teaching with Investigation and Design in Science (TIDeS) project is that future teachers will learn science as undergraduates the way they are expected to teach science in the K–12 classroom: engaging all students in science investigation and engineering design in a discourse-filled, context-rich, inclusive learning process. TIDeS seeks to catalyze transformation of introductory science courses by supporting faculty in the development and implementation of high-quality, rigorously tested, inclusive curricular materials that focus investigation and design. The project has two broad research questions: (1) How do the beliefs and practices of instructors change with developing and/or implementing new curricular materials? (2) What is the impact of the use of these new materials on diverse students? To address these questions, the TIDeS team developed a suite of research probes aligned with the project’s guiding principles and with each other (see figure). The probes include a semi-structured, pre-/post- faculty interview, a quantitative and qualitative classroom observation protocol, a pre-/post- student survey, a syllabus rubric, and rubrics for the curricular materials and student readings. Our preliminary data suggest that, in combination, the probes will provide a holistic picture of what teaching with investigation and design in introductory college-level science courses looks like, how it differs from an active learning classroom, and how it can support the preparation of future teachers.

Sedimentary basins strongly affect earthquake ground motions of both body and surface waves that propagate through them. Yet to characterize seismic hazards at a specific site, it is common practice to consider only the effects of near-surface geology on vertically propagating body waves despite surface waves often causing strong damage. Recently, Bowden & Tsai (2017) proposed an semi-analytical method to predict surface-wave basin amplification and noticed that certain large regional earthquake ground motions are under-predicted if surface waves are not properly accounted for. Since the theory is based on a 1-D approximation of the near-surface geologic structure and does not account for path effects, it is of interest to know how significantly such additional complexity affects the 1-D predictions. When considering deep basins, several other basin parameters play a role in the amplification of surface waves: transmission and conversion at the basin edge, basin shape, lateral resonance and focusing effects. As surface waves propagate back and forth in a highly dispersive medium, the amplification also varies strongly from the edge to the center of the basin. These effects are not always accounted for because of the cost of geophysical surveys that would accurately constrain the structure, the lack of earthquake data for empirical predictions, the poor understanding of what main factors are responsible for basin amplification, and the absence of quantitative estimates of their contribution to the overall amplification. The current study aims to provide quantitative estimates of the importance of these various path effects on surface waves amplification and also extend the current 1-D theory to more complex multi-dimensional basin structures.

Random and small-scale subsurface heterogeneities in velocity and/or density scatter the seismic wavefield when they have scale lengths on the order of the seismic wavelength. Seismic scattering is considered the origin of coda waves. Such inhomogeneities have an important effect on propagating waves, as they generate traveltime and amplitude fluctuations and may be the cause of attenuation or excitation of secondary waves. Understanding the effect of small-scale heterogeneities on the seismic wavefield is important for the characterization of the seismic source (e.g. source parameters of underground nuclear explosions) and to improve our knowledge of the Earth’s structure along the raypath. Several approaches and methods have been suggested to study the scattering of seismic waves and characterise subsurface heterogeneities. Here, we apply a combination of the analysis of the incoherent wavefield component and the coda decay with time to a dataset of over 350 teleseismic events (over 20000 traces) recorded at three seismic arrays (Warramunga, Alice Springs and Pilbara) in Australia. This combination allow us to obtain a series of parameters (correlation length, RMS velocity fluctuations of the heterogeneities and thickness of the scattering layer) that give us a measure of the spatial scale and the magnitude of the heterogeneities present in the lithosphere beneath the arrays. This is the first time such a large dataset is used for a study of these characteristics. Our new results show similar structures and scattering strength for Alice Springs and Warramunga, while revealing a different tectonic signature and stronger scattering in the case of Pilbara, possibly caused by the different thicknesses of crust and lithosphere between these regions and its different tectonic history. These stochastic models of the lithosphere are the first step in the development of a technique analogous to adaptive optics which, in this case, aims at removing the effect of the small-scale, near receiver structure from recorded wavefields, thus enabling us to improve our source characterization and to more clearly image the Earth’s interior.

The projected contribution to sea-level rise from the Greenland Ice Sheet currently has a large spread in literature, ranging from about 14 to 255 mm by the year 2100. Part of this spread is due to uncertainty in mass loss from ocean-terminating outlet glaciers in response to terminus retreat. Here, we use a diffusive-kinematic wave formulation of glacier thinning to show that steep bed features can limit thinning from diffusing inland from a glacier’s terminus. This simplified model allows us to rank 141 of Greenland’s outlet glaciers based on their potential to allow thinning to diffuse far inland and, thus, contribute to sea-level rise over the next century. We then target two glaciers: Kakivfaat Sermiat (KAK) in West Greenland and Kangerlussuaq Gletscher (KLG) in East Greenland. Both glaciers have a high potential to contribute to sea-level rise but with contrasting bed geometries; KAK has relatively low ice flux but its geometry can allow thinning to diffuse far inland while KLG has high ice flux but a geometry that will limit thinning to 30 km inland of its terminus. We simulate mass loss from each glacier, in response to prescribed terminus retreat, using a higher-order numerical model, and find very different response times of mass loss from the two glaciers over the next century. KLG reaches a new steady state by 2100, while the slow inland diffusion of thinning causes KAK to continue its response into the next century and beyond. As a result, KAK contributes nearly twice the volume of ice to sea-level rise of KLG by year 2200, suggesting that low-flux glaciers that can allow thinning to spread far into the ice sheet interior may contribute much to sea-level rise as high-flux glaciers that limit thinning to their lowest reaches. By identifying the glaciers around the ice sheet with the highest potential to contribute to sea-level rise, we hope to help focus future higher-order numerical modeling studies working toward narrowing the range in sea-level rise projections.

It is typically assumed that dinitrogen (N2) fixation and denitrification are mutually exclusive processes in riverine ecosystems because N2 fixation is favored in high light, low nitrogen (N) environments but denitrification is favored under anoxic, high N conditions. Yet recent work in marine and lake ecosystems has demonstrated that N2 fixation can happen under high N conditions and in sediments, challenging this assumption. We conducted a cross-ecoregion study to test the hypothesis that N2 fixation and denitrification would co-occur in streams and rivers across a range of reactive N concentrations. Between 2017 and 2019, we sampled 30 streams in 13 ecoregions, using chambers to quantify N2 flux using membrane inlet mass spectrometry, N2 fixation using acetylene reduction, denitrification using acetylene block, and microbial diversity using 16S gene sequencing. 25 of the study streams were part of the National Ecological Observatory Network or the StreamPULSE network, which provided data on water temperature, light, nutrients, discharge and metabolism. We found that N2 fixation rates were detectable in half of the streams surveyed, and were most frequently detected on rock, wood, and/or macrophyte substrates. Denitrification potential was detected in all streams, with rates 1-2 orders of magnitude higher than N2 fixation rates and the highest rates measured in sediments. Substrate heterogeneity, and associated variation in environmental conditions, appeared to facilitate the coexistence of N2 fixation and denitrification in the study streams. Rates of denitrification were significantly positively related to streamwater nitrate concentrations (r2 = 0.35), but N2 fixation rates were not, despite the common simplifying assumption that denitrification dominates the N2 flux in streams under high N and N2 fixation only occurs under low N conditions. Additional analyses are exploring reach to watershed characteristics, and metabolic regimes as drivers of cross-ecoregion patterns in processes.

We study the behavior of solar wind (SW) parameters over 21-24 solar cycles (SCs). We take into account that these variation can depend on three reasons: (1) type of SW, (2) phase of SC, and (3) SC number. Our catalog of large-scale SW types for 1976-2018 (see the website ftp://ftp.iki.rssi.ru/pub/omni/ [Yermolaev et al., 2009]) covers the period from the 21st to the 24th SC and allows us to analyze independently these three reasons, since we can select SW parameters for different types of SW (in particularly, slow and fast SW, Sheath, ICME, CIR) and at different phases of different SCs. Preliminary data show that with an increase in the SC number (1) the number of perturbed SW types decreases and (2) the values of plasma and magnetic field parameters decrease in each type of SW, and for each phases of SCs. The most rapid change in parameters was observed during the minimum period between the 22nd and 23rd cycles. The work was supported by the Russian Science Foundation, grant 16-12-10062. References Yermolaev, Y.I., Nikolaeva, N.S., Lodkina, I.G., Yermolaev, M. Y. (2009). Catalog of large-scale solar wind phenomena during 1976–2000. Cosmic Res 47, 81–94 https://doi.org/10.1134/S0010952509020014Yermolaev, Y.I., Nikolaeva, N.S., Lodkina, I.G., Yermolaev, M. Y. (2015). Dynamics of large-scale solar wind streams obtained by the double superposed epoch analysis, JGR, https://doi.org/10.1002/2015JA021274

Accelerating demand for energy storage has led to increasing development of brine resources in the “Lithium Triangle”, estimated to hold about 75% of the planet’s Li reserves but persistent and fundamental questions regarding the source and transit time of groundwater have confounded efforts to manage these resources effectively. The basins containing these brines lie within the massive Altiplano-Puna Plateau, home to people whose ancestors have inhabited this land for thousands of years and fragile ecosystems that exist nowhere else on Earth. This region is very dry, bordering Earth’s driest non-polar desert and as such, groundwater is the predominant and, in many areas the only source of water. Fundamental questions about the spatiotemporal dimensions of these groundwater systems have only begun to be addressed. In much of this extreme and remote region, there is a severe lack of quality baseline understanding of the regional hydrological system and connections between surface and groundwater bodies. To address these questions, we utilize an exhaustive set (~2,500 individual analyses) of environmental tracer data (δ18O, δ2H, 3H, 87Sr/86Sr), and dissolved major and minor elements in waters collected from over a dozen field campaigns in the Salar de Atacama and Altiplano of Chile and on the Puna Plateau of Argentina. Our integrated analysis pairs these data with rigorous geochemical modelling and physical hydrological measurements from the field and remote sensing products. 3H data show much of the groundwater currently discharging into these basins is non-modern (>60 yrs. old), stable isotope and geochemical data show strong connectivity but also a marked disconnect between some recharge and discharge areas. We show that “fossil” groundwater, 100-10,000 yrs. or older is widespread and fundamental to the system, sharp disconnects exist between the modern hydrological system, the water bodies it sustains, and those sustained by paleo-recharge water. By defining these connections in spatial detail and within a regional integrated framework, we greatly improve the fundamental mechanistic understanding of this and other groundwater-sustained systems. This will greatly improve the ability of communities, governments and industry to manage of these water resources in a way that is genuinely sustainable.