SEARCH · USGS Science

Results for “International Journal of Coal Geology”

Search indexed USGS publications on groundwater, aquifers, geologic maps, mineral resources and earthquakes. Explore source records by subject and place.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

1,660 records · Page 27Linked to original sources

USGS2025

Hydrogeologic framework and conceptual model of the Red River alluvial aquifer east of Lake Texoma, southeastern Oklahoma, 1980–2022

The 1973 Oklahoma Groundwater Law (Oklahoma Statutes §82-1020.5) requires that the Oklahoma Water Resources Board conduct hydrologic investigations of the State’s groundwater basins to support a determination of the maximum annual yield for each groundwater basin. At present (2025), the Oklahoma Water Resources Board has not established a maximum annual yield for the Red River alluvial aquifer east of Lake Texoma. To support the evaluation and determination of a maximum annual yield, a hydrogeologic framework and conceptual groundwater-flow model were developed to assess groundwater availability in the Red River alluvial aquifer east of Lake Texoma. The scope of this hydrologic investigation is the alluvium and terrace containing the Red River alluvial aquifer in Oklahoma between Lake Texoma, the Texas State line, and the Arkansas State line, an extent referred to in this report as “the eastern part of the Red River alluvial aquifer.” Parts of the alluvium and terrace extent in Arkansas and Texas are included in some analyses to address hydrologic influences from outside the aquifer’s boundaries in Oklahoma. The eastern part of the Red River alluvial aquifer in southeastern Oklahoma consists of approximately 401,280 acres of Quaternary alluvium and terrace deposits associated with the Red River and its major tributaries. Mean annual recharge to the aquifer for the 1980–2022 study period was estimated to be 8.62 inches per year, or 17.98 percent of the mean annual precipitation over the same period (47.94 inches). This mean annual recharge rate is equivalent to an inflow of approximately 288,250 acre-feet per year for the eastern part of the Red River alluvial aquifer. Recharge estimated using the Soil-Water-Balance code accounts for 98.7 percent of the conceptual-model inflows to the eastern part of the Red River alluvial aquifer. Saturated-zone evapotranspiration accounts for 11.9 percent and net streambed seepage accounts for 87.4 percent of the outflows in the conceptual model.

Arkansas, Oklahoma, Texas↗

USGS2025

Dendroseismological investigation of redwood trees along the North Coast section of the San Andreas Fault

Sequoia sempervirens (coast redwood) tree rings have the potential to annually resolve late-Holocene earthquakes on the northern San Andreas Fault based on direct (e.g., physical damage) and indirect (e.g., co-seismic environmental change) impacts, but scarcity of suitable samples and challenges crossdating this long-lived species have limited progress. More precise dating of the pre-1906 (penultimate) earthquake can improve hazard assessment and understanding of rupture segmentation. We target old trees (maximum >815 yr) along the North Coast section of the fault (increment cores via rope-climbing, 11 living trees; plunge cuts, 23 stumps) and employ complementary disturbance detection methods including radial-growth averaging (tree- and series-level), cataloging anatomical indicators (e.g., traumatic resin ducts, TRD), and dating structural components (e.g., reiterated trunks, leans). Multi-centennial ring-width chronologies at Fort Ross (1569−2023) and Gualala (1397−2023) promote continued study with incomplete crossdating limiting utilization of some series. Growth pulses (reductions, releases) and TRD dispersed across the record reflect dynamic environments that obfuscate detection of earthquake signals. The 1906 earthquake did not leave strong signatures on most trees, and when it did, within-tree response varied from normal presentation to discoloration, TRD, and missing rings. Synchrony of indicators at both locations identified 1678−1680 (6 of 15 trees) and 1698−1700 (8 of 16 trees) as the strongest disturbances among dated rings in the time range of the penultimate earthquake, peaking at 1698 (15.7 % of possible growth and anatomical indicators), but the triggering mechanisms for these events are unknown.

California↗

USGS2026

Prioritizing resource protection and understanding potential susceptibility of springs to surficial changes in a low-temperature geothermal system

Geothermal systems are vulnerable to changes in water budget and composition, requiring science-based management. This study uses a dataset of spring water temperatures, time series of groundwater residence time tracers (tritium and carbon-14), and stable isotopes of water to understand geothermal flow in a low-temperature geothermal system in north west Colorado, United States (Steamboat Springs). The geothermal system is bisected by the Yampa River, necessitating a stream mass balance approach to quantify total discharge. Time series analysis of water temperature data provides a ranked list of features more susceptible to surficial changes, which is corroborated using time series of tritium which indicate spatially distinct patterns of mixing between modern and pre-modern groundwater. All springs contain a portion of pre-modern groundwater that is thousands to tens of thousands of years old, a period coinciding with melting of extensive Pleistocene glaciers that was likely one of the recharge sources to the geothermal system. Stream mass balance indicates that greater than 80% of the total geothermal discharge is derived from diffuse or small springs, highlighting the extensive nature of the geothermal outflow zone and the association with local geologic structures. This study provides baseline data to support management of the Steamboat Springs geothermal system and indicates the utility of these approaches in developing science-based geothermal management.

Colorado↗

USGS2024

Seismic response comparison of a historical masonry church subject to real and simulated ground motions

In recent years, advanced numerical models and high-performance computing have facilitated the utilization of ground motion time series in the assessment of the non-linear dynamic behavior of historic masonry structures. Since recorded accelerograms can be sparse for specific analysis conditions, stochastic ground motion simulations have become a viable alternative to overcome this limitation. This study simulates the recorded acceleration time series of the Central Italy 2016 earthquake event at the closest station to the town of Macerata using a site-based stochastic approach. The simulated motions are seismologically evaluated using a goodness-of-fit method in terms of various intensity measures. The simulated records, in conjunction with real records, are used to study the non-linear dynamic behavior of San Filippo Neri church located in Macerata. The church of San Filippo represents an important example of Baroque religious architecture in central Italy, which was damaged and closed off to the public after the 2016 earthquake events. The construction was investigated with a vast diagnostic campaign which included on-site testing and dynamic identification tests. The collected data is used to calibrate the dynamic response of a three-dimensional finite element model of the church. The model is finally used to compare the non-linear seismic responses under real and simulated ground motions with the site recorded damage. The results of structural responses demonstrate a strong agreement between the real and simulated records, providing evidence to support the validation of the site-based stochastic simulation.

Macerata↗

USGS2025

Comparison of hydrologic data and water budgets between 2003–08 and 2018–23 for the eastern part of the Arbuckle-Simpson aquifer, south-central Oklahoma

The Arbuckle-Simpson aquifer is divided spatially into three parts (eastern, central, and western). The largest groundwater withdrawals are from the eastern part of the Arbuckle-Simpson aquifer, which provides water to approximately 39,000 people in Ada and Sulphur, Oklahoma, and surrounding areas. The Arbuckle-Simpson aquifer, including the eastern part, is designated a sole source aquifer for its service area. Based primarily on data collected between 2003 and 2008, a series of comprehensive hydrologic studies of the Arbuckle-Simpson aquifer was published to provide the information necessary to perform groundwater-flow model simulations so that the Oklahoma Water Resources Board could determine how much water could be withdrawn from the aquifer while maintaining flow to springs and streams. As part of the Phase 1 studies, an aquifer water budget was developed from a numerical model for the period 2003–08. For this report, Phase 1 refers to the 2003–08 data collection period, although for some of the analyses, data collected prior to 2003 were used to inform model development work. Allocation of water from this aquifer was then established by the Oklahoma Water Resources Board in 2013. Additional well-spacing rules were also established by the Oklahoma Water Resources Board for sensitive sole source groundwater basins. To determine how the water budget for the eastern part of the Arbuckle-Simpson aquifer has changed over time, recently collected hydrologic data (2018–23) were compared to data collected during 2003–08. The analysis of changes in the aquifer water budget from 2003–08 to 2018–23 could help resource managers better understand changes in the overall balance of water in storage and the potential effects on streamflow, changes in groundwater levels, and the effects of different water uses in the aquifer area on available water in the eastern part of the Arbuckle-Simpson aquifer and streams overlying the eastern part of the Arbuckle-Simpson aquifer.

Oklahoma↗

USGS2026

Use of precariously balanced rocks to constrain postglacial earthquake magnitudes in New England, United States

We use the age and fragility of 13 precariously balanced rocks (PBRs) in Massachusetts, New Hampshire, and southwestern Maine to evaluate the estimated probabilistic seismic hazard (PSH) from the U.S. Geological Survey’s National Seismic Hazard Model (NSHM), and also develop a map of maximum postglacial earthquake magnitudes in the region. The PBRs are erratic boulders left behind by the retreat of the Laurentide ice sheet between 13 and 18 ka, and published ice retreat maps are used to estimate their fragility ages. The fragilities and ages of the rocks appear to be consistent with the NSHM, in that the estimated earthquake magnitudes and resulting ground motions that would topple them are only expected on timescales longer than their fragility ages. This result, on the North American passive margin, contrasts with PBR studies at active plate boundaries, where the PBR constraints imply large reductions to estimated PSH. It may be that the PBRs are not old enough in the eastern United States to reveal the same PSH modeling issues as those revealed in the plate boundary settings. The study provides constraints on maximum magnitudes of postglacial earthquakes across much of Massachusetts, New Hampshire, and southwestern Maine, with the constraints limited to the ages of the PBRs.

Maine. Massachusetts, New Hampshire↗

USGS1995

Water quality of the Boca Raton canal system and effects of the Hillsboro Canal inflow, southeastern Florida, 1990-91

The City of Boca Raton in southeastern Palm Beach County, Florida, is an urban residential area that has sustained a constant population growth with subsequent increase in water use. The Boca Raton network of canals is controlled to provide for drainage of excess water, to maintain proper coastal ground-water levels to prevent saltwater intrusion, and to recharge the surficial aquifer system from which the city withdraws potable water. Most of the water supplied to the Boca Raton canal system and the surficial aquifer system, other than rainfall and runoff, is pumped from the Hillsboro Canal. The Biscayne aquifer, principal hydrogeologic unit of the surficial aquifer system, is highly permeable and there is a close relation between water levels in the canals and the aquifer. The amount of water supplied by seepage from the conservation areas is unknown. Because the Hillsboro Canal flows from Lake Okeechobee and Water Conservation Areas 1 and 2, which are places of more highly mineralized ground water and surface water, the canal is a possible source of contamination. Water samples were collected at 10 canal sites during wet and dry seasons and analyzed for major inorganic ions and related characteristics, nutrients, and trace elements. All concentrations were generally within or less than the drinking-water standards established by the Florida Department of Environmental Protection. The high concentrations of sodium and chloride that were detected in samples from the Boca Raton canal system are probably from the more mineralized water of the Hillsboro Canal. Other water-quality data, gathered from various sources from 1982 through 1991, did not indicate any significant changes nor trends. The data include pesticide and metal analyses of water samples and bottom sediments collected at four canal sites in the Boca Raton study area by the U.S. Geological Survey during 1982-84. The effects of the Hillsboro Canal on the water quality of the Boca Raton canal system are indicated by increased concentrations of sodium, chloride, dissolved solids, and total organic carbon. Concentrations of the constituents in the canal water generally decrease with distance from the Hillsboro Canal pumping station and are the result of dilution by receiving canal waters.

Florida↗

USGS2024

Earthquake effects surveyed during the nineteenth century as ecological features of Chinookan tidelands

Lasting effects of a Cascadia earthquake in 1700 were documented during surveys of Chinookan tidelands near the mouth of the Columbia River between 1805 and 1868. The effects resemble estuarine consequences, near Anchorage, of the 1964 Alaska earthquake: fatal drowning of subsided meadows and forests by post-earthquake tides, rebirth of marshes and forests through post-earthquake sedimentation, and uplift. Chinookan remains of killed forests were recorded by James Graham Cooper, John J. Lowell, and Cleveland Rockwell. Cooper, attached to a railroad survey and the Smithsonian Institution, wrote of redcedar stumps and trunks standing dead in tidal marshes of Shoalwater (now Willapa) Bay. Two such snags served as bearing trees for Lowell as he platted a Shoalwater Bay township under contract with the General Land Office. Rockwell, of the US Coast Survey, flecked landward edges of tidal flats west of Astoria with symbols that evoke remains of a bygone spruce forest. The Lewis and Clark Expedition, while in that area in 1805–1806, mapped and puzzled over tideland vegetation that post-1700 succession helps explain.

Alaska, British Columbia, California, Oregon, Wash↗

USGS2025

Using visualization science to inform the design of environmental decision-support tools—A case study of the U.S. Geological Survey Waterwatch

Environmental decision-support tools are increasingly being used to serve both expert and non-expert audiences. Many existing tools are primarily expert-focused, and redesigning them can be challenging because experts and non-experts interact with tools differently, existing users may be resistant to changes, and there is little guidance on how to prioritize redesign efforts and demonstrate their efficacy. In this report, we present a case study of a user-centered redesign of an established environmental decision-support tool—the U.S. Geological Survey WaterWatch. WaterWatch supports flood, drought, and other water resource management decisions through the display of water levels at gages across the United States. Using a participatory process, we identified a functional change (replacing the existing rainbow colormap), created an alternative design, and tested the alternative’s usability through two general public surveys. The results showed that replacing the rainbow colormap with a more intuitive diverging colormap improves usability, regardless of the audience’s subjective preference for the rainbow color scheme. In addition, we demonstrated the importance of using legends to improve the audience’s understanding of the map symbols. This study demonstrates how user-centered design approaches can be used to inform the design of high-profile products and tools.

Scientific Investigations Report↗

USGS2026

Assessing the state of hydrologic science in the Upper Klamath Basin—A comprehensive review of data, tools, and models

Water demand in the Upper Klamath Basin (UKB) from various stakeholders and ecological needs often outstrips available supply, leading to persistent management challenges. This study reviews the state of hydrologic science within the UKB as of 2025—specifically, the tools, data, and models available for assessing five key components of the water system: (1) surface water; (2) precipitation; (3) evapotranspiration; (4) groundwater; and (5) water use. The UKB water supply is critical for Native American communities, regional agriculture, and federally listed fishes and faces challenges from competing needs, climate variability, and operational/regulatory requirements. We assess existing datasets, regional and national models, and historical studies to understand the available resources and identify gaps that may hinder integrated water assessments and management. Our findings indicate areas where improvements in data collection and model precision could improve the accuracy of water-availability forecasts and support water-management practices. This review can inform near-term forecasting, assist in optimizing water-resource data collection and management strategies, and support regional water-availability assessments of the basin.

Oregon↗

USGS2026

Storage timescales and the crystal growth history recorded in the layered reservoir of the Unit 3 Kulanaokuaiki Tephra Member of the Uēkahuna Ash from Kīlauea volcano

The Kulanaokuaiki Tephra Unit 3 (900 C.E.) was a subplinian basaltic scoria eruption, the largest known to have occurred at Kīlauea’s summit. The initiation mechanism and cause(s) of this highly hazardous eruption remain poorly understood, particularly as this style of activity has not been observed in the modern period. We use olivine diffusion timescales from Fe-Mg and lithium zoning to assess magma storage timescales and evaluate evidence for mixing prior to eruption. Of the 200 crystals assessed, 55% are normally zoned for forsterite and 45% are non-zoned. Zoned crystals (> 1 mm) ( n = 96) have average compositions of Fo 89 cores and Fo 81 rims, while small crystals 0.5–1 mm ( n = 14) have Fo 85 cores and Fo 82 rims. Modeled Fo diffusion timescales range from < 1 to 12.6 years (median of 3.0 years). Lithium maps in 32 of the forsterite-zoned crystals reveal a variety of lithium zoning types, most of which we attribute to crystal growth. Many lithium profiles are coupled to phosphorus within the outer ~ 100 μm of the crystal ( n = 17), reflecting growth processes and precluding their use for diffusion studies. Overall, our Unit 3 olivine dataset displays no evidence of magmatic priming shortly prior to eruption, unlike prior observations for the basaltic subplinian eruption of the Keanakāko‘i Tephra Unit E (1650 C.E.). Thus, explosive eruption magmas may be stored for considerably longer than previously thought, and magmatic priming is not a necessary precursor for Kīlauea’s explosive eruptions. We hypothesize that gas-driven filter pressing from crystallization of microlites and macrocrysts in the reservoir slurry contributed to the explosivity of the Unit 3 eruption.

Hawaii↗

USGS2025

The role of alkali bicarbonate-sulfate brines in the genesis of carbonatite REE resources at the Bear Lodge Alkaline Complex, Wyoming

Rare-earth element (REE) resources in the Bear Lodge Alkaline Complex, Wyoming, are hosted in a variably leached carbonatite dike swarm spatially related to bodies of diatreme breccia. This study examines fluid inclusions in carbonatite dikes, peripheral fluorite breccias, and smoky quartz veins to reconstruct the physiochemical conditions of REE mineralization. Results reveal a multi-stage evolution of carbonatitic fluids: 1) high-temperature (330–432°C), CO 2 -rich magmatic fluids, 2) REE-enriched alkali bicarbonate-sulfate brines, and 3) low-temperature meteoric (117–182°C) or diluted magmatic fluids. Multiphase inclusions (stage-2) contain burbankite, nahcolite, and alkali sulfate daughter crystals, linking the brines to early burbankite [(Na,Ca) 3 (Sr,Ba,REE) 3 (CO 3 ) 5 ] mineralization in carbonatite. REE mobility has been linked to the stability of aqueous complexes with ligands such as Cl-, SO 4 2- , and HCO 3 - . However, light REEs were not initially transported significant distances, rather they crystallized as burbankite from brines or brine-melts associated with the carbonatite magma. Crystallization of burbankite removed Na from the system, leaving residual fluids with high K/Na ratios and promoting potassic fenitization throughout the complex. Fractionation of light and heavy REEs and involvement of low temperature (<300°C), high K/Na bicarbonate brines may have facilitated transport and formation of peripheral occurrences that are more enriched in heavy REEs at Bear Lodge.

Wyoming↗

USGS2025

Estimated hydrogeologic, spatial, and temporal distribution of self-supplied domestic groundwater withdrawals for aquifers of the Virginia Coastal Plain

Water use from private-domestic wells accounts for nearly 40 percent of total groundwater withdrawals in the Virginia Coastal Plain Physiographic Province (henceforth called the Virginia Coastal Plain). However, because self-supplied domestic water use generally falls below the Virginia Department of Environmental Quality (VDEQ) reporting and management threshold of 300,000 gallons per month, quantifying these withdrawals is challenging. This report builds upon the foundation of previous U.S. Geological Survey investigations by providing revised techniques to improve estimates of the aquifer source, spatial distribution, and monthly magnitude of these groundwater withdrawals. The aquifer sources of private-domestic wells in the Virginia Coastal Plain were estimated by cross-referencing 8,264 well records from the VDEQ and the Virginia Department of Health to a digital model of the Virginia Coastal Plain hydrogeologic framework. This analysis highlights the regional importance of the Yorktown-Eastover, Potomac, and surficial aquifers. Collectively, these three aquifers account for 80 percent of self-supplied domestic groundwater withdrawals. The population using self-supplied domestic water was estimated using census blocks, well-use ratios, building footprints, and land-use and land-cover data to produce a high-resolution, disaggregated, raster-based dataset. This approach improves upon previous models at the census-block or road-network scale by reducing the low-density spread of the self-supplied domestic population across undeveloped areas and concentrating the population and its corresponding water use in the areas where it is most likely to occur. Results show that an estimated 475,332 people comprise the 2020 self-supplied domestic population of the Virginia Coastal Plain, an increase of 5.7 percent since 2010, and the greatest concentrations of self-supplied domestic population surround large cities. Estimates could be further refined with the addition of current and complete spatial data on public water-system service areas. The quantity of water used by the self-supplied domestic population was estimated by modifying published state per-capita water-use coefficients with the corresponding monthly variability assessed from Virginia Coastal Plain public water-system withdrawal data. This analysis estimates an average increase of 12 percent from June through August and an average decrease of 8 percent from December through March from the baseline annual average of 80 gallons per day per capita, which generally matches similar studies in the eastern United States. The application of these revised methodologies for the estimation of private-domestic wells and the self-supplied domestic population improves understanding of domestic groundwater use in the Virginia Coastal Plain across hydrogeologic, spatial, and temporal scales. These revisions help better inform water-resource managers and decision makers and support higher resolution groundwater modeling. Furthermore, these methods are transferrable to other areas where self-supplied domestic water withdrawals are important to the overall water budget.

Virginia↗

USGS2024

Tungsten skarn quantitative mineral resource assessment and gold, rare earth elements, graphite, and uranium qualitative assessments of the Kuldjuktau and Auminzatau Ranges, in the central Kyzylkum region, Uzbekistan

A new quantitative mineral resource assessment for tungsten skarn was conducted for the Auminzatau and Kuldjuktau mountain ranges in Central Uzbekistan, along with qualitative assessments of orogenic gold, rare earth elements (REEs), amorphous graphite, and uranium. By integrating a variety of geological, geochemical, geophysical, and remote sensing data sets, estimates of undiscovered tungsten skarn deposits in permissive tracts are combined with grade and tonnage distributions of known deposits to generate probabilistic estimates of undiscovered resources. Undiscovered deposits in Auminzatau are estimated to contain median resources of 98 thousand metric tons (kt) of WO 3 with a 70 percent (%) probability of at least 28 kt and a 10% probability of at least 468 kt, of which 16 kt to 293 kt may be economic to extract. In Kuldjuktau, the undiscovered deposits are estimated to contain median resources of 27 kt of WO 3 with a 60% probability of at least 12 kt and a 10% probability of at least 208 kt, of which 5 kt to 132 kt may be economic to extract. Our results suggest that the Auminzatau–Kuldjuktau Mountains area is highly prospective for additional discovery of significant Au and U resources and has low prospectivity for discovery of significant REE and graphite resources.

Minerals↗

USGS2025

A review of post-wildfire adaptations of surface-water-quality models: Synthesis, gaps, and opportunities

As wildfires increasingly affect water-supply watersheds, the demand for models to predict water-quality responses is increasing. This work reviews and synthesizes existing post-wildfire applications of water-quality models in the context of geographic and ecohydrological distribution, hydrologic and water-quality response process representation, model parameterization, model and input data scales, model calibration data availability, as well as calibration and performance evaluation approaches. Emphasis is placed on models that simulate water-quality output, rather than sediment and erosional response as the primary focus. Here, identified gaps and opportunities to advance the post-wildfire application of water-quality models include: 1. applying models in under-represented geographic and ecohydrologic regions, 2. simulating multiple streamflow generation mechanisms, including groundwater, with an emphasis on shifting dominant flow pathways as the landscape recovers following wildfire, 3. adding studies that include the simulation of metals, 4. incorporating more biogeochemical and in-stream processes to model applications, 5. applying finer spatial and temporal resolution of precipitation data input as well as finer spatial resolution hydrologic response units, 6. implementing fully distributed grid or element models or finer resolution response units to capture burn severity heterogeneity, 7. collecting enhanced water-quality data for model calibration and validation, 8. conducting model-intercomparison studies, and 9. developing model parameter value guidance in post-wildfire applications. These identified gaps and opportunities may assist users in deciding on key processes and approaches to consider in modeling post-wildfire water-quality conditions.

Science of the Total Environment↗

USGS1976

Hydrogeology of stratified drift in Farmington, Connecticut: Available data and future needs

The purpose of this report is twofold. The first is to update hydrogeologic information on the major stratified-drift deposits that underlie much of Farmington. The second is to outline data requirements for future ground-water evaluation and management. The scope of the report is limited to the stratified drift, as it is the only aquifer capable of sustaining large withdrawals for public or industrial supply. This aquifer is composed of interbedded layers of gravel, sand, silt and clay. Most of this material was deposited by glacial meltwaters, but locally the aquifer contains some unconsolidated deposits of nonglacial origin. The most extensive stratified-drift deposits are in the valleys of the Farmington and Pequabuck Rivers.(See plate B.) A few small areas of stratified drift near East Farmington Heights and Oakland Gardens are not discussed in this report as their potential for large-scale development is slight.

Connecticut↗

USGS2024

Hydroacoustic observations reveal drivers of mixing and salinization of a karst subterranean estuary during intense precipitation

Karst subterranean estuaries within globally ubiquitous carbonate aquifers are coastal groundwater ecosystems that provide an essential water resource for human populations. To understand the drivers of salinization within a coastal aquifer in the Yucatan Peninsula (Mexico), we employed hydroacoustics in flooded caves to observe how oceanic and atmospheric events facilitate mixing between the meteoric lens (fresh-brackish groundwater) and the saline groundwater on tidal and episodic timescales. Precipitation during Tropical Storm Carlotta increased the flow and salinity of the meteoric lens without evidence for vertical mixing across the halocline. We postulate that vertical migration of haloclines in the conduit relative to those within the rock matrix during precipitation creates lateral density gradients that drive mixing, and ultimately creates a brackish layer within the meteoric lens. These results provide a mechanistic explanation for vertical and lateral exchange in a coastal carbonate aquifer, which has implications for groundwater response to future climatic change.

Ox Bel Ha cave system, Yucatan peninsula↗

USGS2025

Case study of deep learning image segmentation for the purposes of rapid 2D petrographic analysis in volcanic rocks

Automation using deep learning methods is a useful alternative to manual methods of petrographic segmentation, but often requires user familiarity with coding and/or algorithms. We examine the Dragonfly TM program's deep learning tools for application by users with a variety of skill levels as a method for petrographic image segmentation. An image processing methodology, bimodal image stacking, was created for low-input-data, high-efficacy training of models which can then be applied to varied samples. Using backscatter electron images we show that the resulting model segmentations agree with manual segmentation total and modal crystallinity values within 5%, and calculated plagioclase crystal size distribution (CSD) values within 2σ, despite limitations in discriminating mafic phases. Model creation and training takes <24 hours, 1–3 hours of which are supervised, and the resultant model can then be applied to new uncharacterized samples in <15 minutes per image. This allows for non-experts to create and utilize deep learning models to segment images of variable brightness and texture, at low user-time cost and resulting in size and shape data which are within uncertainty of manual segmentation. While some limitations are noted (for example, sieve-textured phases may need manual correction, and different minerals with similar BSE intensity may not be resolved as separate phases), this methodology can be utilized for general application of models to wide ranges of volcanic crystalline and bubble textures, and to create a library of models for rapid petrological analysis during volcanic eruptions.

Alaska↗