We present the chemical and isotopic compositions of gases and fluxes of CO2 from the hydrothermal features of Newberry Volcano, a large composite volcano located in Oregon's Cascade Range with a summit caldera that hosts two lakes, Paulina and East Lakes. Gas samples were collected from 1982 to 2021 from Paulina Hot Springs (PHS) on the shore of Paulina Lake, East Lake Hot Springs (ELHS) on the shore of East Lake, and Obsidian Flow Gas Seep (OFGS), an area of diffuse gas emissions. Surveys of CO2 flux were conducted in 2020 at OFGS (1400 m2) and East Lake (4.1 km2). Gases from all three sites were CO2-rich (>= 79 mol% in dry gas) but showed considerable compositional variability over time due to interaction with ground and surface water. An increase in H2S concentrations and decline in CO2/H2S ratios in ELHS gases coincided with a drop in East Lake water level from 1999 to 2021. ELHS and OFGS gases were high in CH4 relative to PHS and the delta 13C of CH4 values for ELHS gases (-72.2 and -63.6 %o) reflected a predominantly biogenic origin. The dominant source of N2 and Ar in PHS, ELHS, and OFGS samples was likely groundwater. Helium isotopic ratios (6.47 to 8.02 Rc/Ra) support a persistent source of magmatic He beneath Newberry caldera and consistently high values measured at OFGS and PHS relative to ELHS suggest distinct fluid flow paths from depth to the surface features. The delta 13C of CO2 and CO2/3He values (-8.9 to-5.35 %o and 1.3 x 109 to 4.6 x 1010, respectively) measured in gases reflect con-tributions of CO2 from both mantle and crustal sources. Measured CO2 fluxes at OFGS and East Lake ranged from 1 to 8808 and < 1 to 364 g m- 2 d-1, respectively. A CO2 emission rate of 0.5 t d-1 was calculated for OFGS. The CO2 emission rate estimated for East Lake was 30 t d-1 and when compared to prior estimates, reflects steady-state lake degassing. An enhanced geochemical monitoring plan should be implemented, including annual sampling of gases at ELHS, OFGS, and PHS for geochemical analysis, installation of a continuous lake-level monitoring station at East Lake, and annual CO2 flux surveys at OFGS. Geochemical data should be integrated with data from the real-time seismic and GPS network to better resolve and interpret potential changes in the Newberry magmatic-hydrothermal system.

It is now well established that earthquakes cause various hydrogeological responses at distances thousands of kilometers from the epicenter. What remains unexplained is the large amplitude and intensity of some responses. Following the 2004 Mw 9.1 Sumatra earthquake, groundwater 3,200 km from the epicenter erupted violently from a well and formed a water fountain reaching a height exceeding 60 m. We model the relevant processes by combining tidal analysis of groundwater level with numerical simulations using a two-dimensional finite-element model. We suggest that the eruption resulted from a combination of factors, including a rapid increase of crustal permeability and runaway CO2 exsolution and bubble nucleation induced by the passage of seismic waves. Our results may have implications for some engineering applications such as oil production and CO2 sequestration, and the eruption of hydrothermal features such as geysers.

About 14.5 months after the 2018 eruption and summit collapse of Kilauea Volcano, HawaiModified Letter Turned Commai, liquid water started accumulating in the deepened summit crater, forming a lake that attained 51 m depth before rapidly boiling off on December 20, 2020, when an eruption from the crater wall poured lava into the lake. Modeling the growth of the crater lake at Kilauea summit is important for assessing the potential for explosive volcanism. Our current understanding of the past 2500 years of eruptive activity at Kilauea suggests a slight dominance of explosive behavior over effusive. The deepened summit crater and presence of the crater lake in 2019 raised renewed concerns about explosive activity. Groundwater models using hydraulic-property data from a nearby drillhole successfully forecast the timing and rate of lake filling. Here we compare the groundwater-model predictions with observational data through the demise of the crater lake, examine the implications for local water-table configuration, consider the potential role of evaporation and recharge (neglected in previous models), and briefly discuss the energetics of the rapid boil-off. This post audit of groundwater-flow models of Kilauea summit shows that simple models can sometimes be used effectively to simulate complex settings such as volcanoes.

Electrical power production from geothermal energy has a solid record of success for permeable reservoirs such as The Geysers in northern California and geothermal systems in Iceland and New Zealand, among other places. Such permeable reservoirs, however, represent only a small fraction of the available heat energy in the earth’s shallow crust. Most of the available energy resides in rocks with insufficient permeability and storage to produce commercial volumes of heated fluids

First posted January 7, 2021 For additional information, contact: Contact HVOHawaiian Volcano ObservatoryU.S. Geological Survey1266 Kamehameha AvenueSuite A-8Hilo, HI 96720 Kīlauea Volcano, on the Island of Hawaiʻi, is surrounded and permeated by active groundwater systems that interact dynamically with the volcanic system. A generalized conceptual model of Hawaiian hydrogeology includes high-level dike-impounded groundwater, very permeable perched and basal aquifers, and a transition (mixing) zone between freshwater and saltwater. Most high-level groundwater is associated with the low-permeability intrusive complexes that underlie volcanic rift zones and calderas and also act to compartmentalize the groundwater system. Hydrogeologic studies of Kīlauea in recent decades, accompanied by deep research drilling, have shown that high-level groundwater is more widespread than once understood, that permeability decreases dramatically at depth, particularly in rift zones, and that freshwater can occur at depths of as much as several kilometers below the local water table. Copious groundwater recharge causes near-surface conductive heat flow to be near zero over much of Kīlauea. Approximately 95 percent of groundwater discharge occurs offshore, accompanied by approximately 99 percent of the approximately 6,000 megawatts of heat supplied by magmatic intrusion. Here, we summarize current understanding of the groundwater system of Kīlauea Volcano and describe transient changes during the decade or more preceding the 2018 eruption sequence. The changes in groundwater chemistry and thermal structure beneath Kīlauea summit hold implications for volcanic-volatile transport and the potential for explosive volcanism. Between 2008 and 2018, the magma conduit beneath the lava lake likely created an adjacent zone of very hot rock that significantly delayed liquid groundwater inflow to the draining magma conduit. Sulfate concentrations in groundwater beneath Kīlauea summit, sampled at the National Science Foundation-funded drill hole 1.5 kilometers south-southwest of the lava lake, declined substantially between 2010 and present. This decline likely reflects, at least in part, the decreased effectiveness of volatile condensation and solution into groundwater (scrubbing). The vent opening in 2008 presumably focused volatile flux into the vicinity of the vent, and progressive drying of the surroundings further restricted interaction with the groundwater system. The decrease in sulfate concentrations in the drill hole between 2010 and 2018 likely reflects decreased effectiveness of scrubbing.

During the 2018 eruption of Kilauea Volcano, Hawai'i, scientists relied heavily on a conceptual model of explosive eruptions triggered when lava-lake levels drop below the water table. Numerical modeling of multiphase groundwater flow and heat transport revealed that, contrary to expectations, liquid water inflow to the drained magma conduit would likely be delayed by months to years, owing to the inability of liquid water to transit a zone of very hot rock. The summit of Kilauea subsequently experienced an similar to 2-month period of consistent repeated collapses, and the crater now extends below the equilibrium position of the water table. Liquid water first emerged into the deepened crater in late July 2019. The timing of first appearance of liquid water (about 14 months postcollapse) and the rate of crater lake filling (currently similar to 27 kg/s) were well-predicted by the numerical modeling done in late spring 2018, which forecast liquid inflow after 3 to 24 months at rates of 10 to 100 kg/s. A second-generation groundwater model, reflecting the current crater geometry, forecasts lake filling over the next several years. The successful 2018 to present forecasts with both models are based on unadjusted in situ permeability estimates (1 to 6 x 10(-14) m(2)) and water-table elevations (600 to 800 m) from a nearby research drillhole and geophysical surveys. Important unknowns that affect the reliability of longer-term forecasts include the equilibrium water-table geometry, the rate of evaporation from the hot and growing crater lake (currently similar to 29,000 m(2)at 70-80 degrees C), and heterogenous permeability changes caused by the 2018 collapse.

The hydrothermal systems associated with the restless high-threat volcanoes at Lassen and Long Valley, California, each release large amounts of arsenic (As) to surface waters - similar to 6 and similar to 8 metric tons/yr, respectively. The hydrothermal As output can increase during volcanic unrest, as illustrated by a two-fold increase during unrest at Lassen in 2014-15. During that period of unrest, increased As concentrations and fluxes were measured up to 75 km downstream from the Lassen source, in Mill Creek near the confluence with the Sacramento River. In eastern California, the Long Valley hydrothermal system feeds into the Los Angeles Aqueduct (LAA), and the Los Angeles Department of Water and Power (LADWP) actively manages the LAA system to remove hydrothermal As. In 1980, during a series of similar to M(w)6 earthquakes, the discharge of a particular group of hydrothermal vents in Long Valley increased approximately 7-fold, though the total increase in As flux to the LAA system at that time is unknown. The likely mechanism for increased hydrothermal discharge in each case is permeability enhancement due to strong ground motion. A review of the global literature on earthquake hydrology suggests a worst-case scenario of a roughly 10-fold increase in permeability, with commensurate increase in the hydrothermal As flux persisting for days to months. Here we evaluate the potential impact of such increases in hydrothermal As flux on the California water-supply system.

The many successes in volcano forecasting over the past several decades owe mainly to pattern recognition, both in monitoring data and the geologic record. During the early stages of the 2018 Klauea eruption, the conceptual model of Stearns (1925, ), based on the explosive 1924 Klauea eruption, was highly influential. This model postulates that explosions are triggered by liquid water inflow into a recently vacated magma conduit. Modern quantitative modeling approaches, supplemented by hydrogeologic data unavailable in 1925, yield a more nuanced view. Results demonstrate that liquid water inflow would likely be delayed by months to years, owing to the inability of liquid water to transit a zone of very hot rock surrounding the conduit. The exercise demonstrates the use of physically based modeling to supplement traditional volcano forecasting approaches during an ongoing event.

Hydrologic responses to earthquakes such as streamflow increases, water‐level changes, and changes in geyser eruption frequency often reflect changes in permeability caused by seismic waves. The dynamic nature of permeability, as revealed by coseismic hydrologic phenomena, holds implications for groundwater systems, geothermal resources, mineral resources, and geologic hazards. Analysis of water‐level responses to solid Earth tides and changes in atmospheric pressure provides a passive way to continuously monitor changes in permeability and storage properties in tectonically active regions.

From 2009 to 2015 the U.S. Geological Survey (USGS) systematically monitored hydrothermal behavior at selected Cascade Range volcanoes in order to define baseline hydrothermal and geochemical conditions. Gas and water data were collected regularly at 25 sites on 10 of the highest-risk volcanoes in the Cascade Range. These sites include near-summit fumarole groups and springs/streams that show clear evidence of magmatic influence (high 3He/4He ratios and/or large fluxes of magmatic CO2 or heat). Site records consist mainly of hourly temperature and hydrothermal-flux data. Having established baseline conditions during a multiyear quiescent period, the USGS reduced monitoring frequency from 2015 to present. The archived monitoring data are housed at (doi:10.5066/F72N5088). These data (1) are suitable for retrospective comparison with other continuous geophysical monitoring data and (2) will provide context during future episodes of volcanic unrest, such that unrest-related variations at these thoroughly characterized sites will be more clearly recognizable. Relatively high-frequency year-round data are essential to achieve these objectives, because many of the time series reveal significant diurnal, seasonal, and inter-annual variability that would tend to mask unrest signals in the absence of baseline data. Here we characterize normal variability for each site, suggest strategies to detect future volcanic unrest, and explore deviations from background associated with recent unrest.

Springs at La Soufrière of Guadeloupe have been monitored for nearly four decades since the phreatic eruption and associated seismic activity in 1976. We conceptualize degassing vapor/gas mixtures as square‐wave sources of chloride and heat and apply a new semianalytic solution to demonstrate that chloride and heat pulses with the same timing and duration result in good matches between measured and simulated spring temperatures and concentrations. While the concentration of chloride pulses is variable, the local boiling temperature of 96°C was assigned to all thermal pulses. Because chloride is a conservative tracer, chloride breakthrough is only affected by one‐dimensional advection and dispersion. The thermal tracer is damped and lagged relative to chloride due to conductive heat exchange with the overlying and underlying strata. Joint analysis of temperature and chloride allows estimation of the onset and duration of degassing pulses, refining the chronology of recent magmatic intrusion.

Permeability is the dominant parameter in most hydrogeologic studies. There is abundant evidence for dynamic variations in permeability in time as well as space, and throughout the crust. Whether this dynamic behavior should be included in quantitative models depends on the problem at hand.

Distal volcano-tectonic (dVT) seismicity typically precedes eruption at long-dormant volcanoes by days to years. Precursory dVT seismicity may reflect magma-induced fluid-pressure pulses that intersect critically stressed faults. We explored this hypothesis using an open-source magmatic-hydrothermal code that simulates multiphase fluid and heat transport over the temperature range 0 to 1200°C. We calculated fluid-pressure changes caused by a small (0.04km3) intrusion and explored the effects of flow geometry (channelized vs. radial flow), magma devolatilization rates (0–15kg/s), and intrusion depths (5 and 7.5km, above and below the brittle-ductile transition). Magma and host-rock permeabilities were key controlling parameters and we tested a wide range of permeability (k) and permeability anisotropies (kh/kv), including k constant, k(z), k(T), and k(z, T, P) distributions, examining a total of ~1600 realizations to explore the relevant parameter space. Propagation of potentially causal pressure changes (ΔP≥0.1 bars) to the mean dVT location (6km lateral distance, 6km depth) was favored by channelized fluid flow, high devolatilization rates, and permeabilities similar to those found in geothermal reservoirs (k~10−16 to 10−13m2). For channelized flow, magma-induced thermal pressurization alone can generate cases of ∆P≥0.1 bars for all permeabilities in the range 10−16 to 10−13m2, whereas in radial flow regimes thermal pressurization causes ∆P<0.1 bars for all permeabilities. Changes in distal fluid pressure occurred before proximal pressure changes given modest anisotropies (kh/kv~10–100). Invoking k(z,T,P) and high, sustained devolatilization rates caused large dynamic fluctuations in k and P in the near-magma environment but had little effect on pressure changes at the distal dVT location. Intrusion below the brittle-ductile transition damps but does not prevent pressure transmission to the dVT site.