USGS ScienceSearch

SEARCH · USGS Science

Results for “Journal of Hydrology, Regional Studies”

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

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

1,662 records · Page 84Linked to original sources

Did they feel it? Legacy maroseismic data illuminates an engimatic 20th century earthquake

The challenges and the importance of preserving legacy instrumental records of earthquakes are now well-recognized (e.g., Richards & Hellweg, 2020, https://doi.org/10.1785/0220200053 ). Seismologists may not be aware of parallel challenges and opportunities with legacy macroseismic data for earthquakes in the United States. For much of the 20th century, macroseismic data were collected by a series of U.S. government agencies using a standard questionnaire distributed on postcards. Published summaries of postcards provide macroseismic data akin to modern Did You Feel It? questionnaire responses. In this paper we focus on the M 6.5 Fickle Hill, California earthquake, on 21 December 1954 (Hellweg et al., 2025) as a proof-of-concept, illustrating the potential of what we dub Did They Feel It? (DTFI) data to improve our understanding of significant 20th century U.S. earthquakes for which instrumental data are sparse. Legacy macroseismic data interpreted following modern conventions can potentially constrain traditional ShakeMaps at a level of detail and accuracy that in some respects rival maps for modern earthquakes. The updated ShakeMap for the 1954 Fickle Hill earthquake, also drawing from recently published media and first-person accounts, supports the location, depth, and stress drop value estimated from available instrumental data (Hellweg et al., 2025).

California

Haunted Summerville: Ghostly lights or earthquake lights?

Among the colorful local lore in the Charleston, South Carolina, area, are a number of ghost stories, shared not only over campfires but also in published books. Among the most well-known of the stories is the tale of the Summerville Light. Local lore holds that a strange light sometimes seen in a remote area is a lantern carried by the ghost of a local woman who once waited hours for her husband, who turned out to have been decapitated earlier that day in a train accident (DePoppe, 2023). Extant sources suggest the ghost stories began to circulate in the 1950s to 1960s. So pervasive was the lore that (Old) Sheep Island Road became known among local residents as Light Road, with a local stretch of road known today as Old Light Road. Reviewing the location where the lights appear as well as the nature of accounts, I suggest that many if not all of the anecdotal observations can be most readily attributed to natural phenomena, including earthquake lights from earthquakes that were too small to be felt. Accounts of lights near Summerville cluster in proximity to the generally accepted epicenter of the 1886 Charleston, South Carolina, earthquake, where foreshocks to the 1886 mainshock were apparently concentrated, and within a few kilometers of three M3.5 – 4.4 earthquakes in 1959 and 1960.

South Carolina

Evaluation of stream capture related to groundwater pumping, middle Humboldt River Basin, Nevada

Historical, future, and potential stream capture from groundwater pumping in the middle Humboldt River Basin (MHRB), Nevada, is estimated using a calibrated numerical groundwater flow model. The model was developed to estimate (1) stream capture, which is the change in flux between the groundwater system and the Humboldt River and tributaries, and (2) change in streamflow, which is the change in streamflow estimated for the Imlay gage on the Humboldt River (U.S. Geological Survey streamgage 10333000). Historical stream capture for water years (WYs) 1961–2015 is estimated using recorded and estimated groundwater pumping during that period. Future (predictive) stream capture was based on historical stresses (WYs 1961–2015) using a scenario that simulated non-mine pumping from WY 2015 at a uniform rate for 100 years into the future. Potential stream capture throughout the middle Humboldt River Basin from groundwater pumping during varying durations of time are presented in a series of capture maps. Maps also are presented that show the potential to capture from groundwater evapotranspiration, as well as the storage changes for pumping duration of 100 years. Estimates of historical stream capture from the mainstem Humboldt River during the early 1960s are less than 400 acre-feet per year (acre-ft/yr) when groundwater withdrawals and pumping rates were relatively small compared to more recent times. In the late 1980s and early 1990s, groundwater withdrawals increased and estimated historical stream capture also increased from about 4,000 acre-ft/yr in the late 1980s and early 1990s to as much as 18,800 acre-feet (acre-ft) in WY 1998. In WY 2015, estimated historical stream capture declined to about 13,000 acre-ft because of decreasing groundwater withdrawals and lower streamflow during the drought of WYs 2012–15, resulting in less stream water available for capture. Stream capture was estimated for 100 years into the future based on WY 2015 non-mine pumping rates and mine-dewatering activity through WY 2015. Stream capture is forecast to increase to about 23,000 acre-ft/yr, and streamflow in the Humboldt River could decrease by as much as 19,000 acre-ft/yr. Pumping for mine-dewatering and the associated discharge of that water affects streamflow in the Humboldt River at Imlay, Nevada (U.S. Geological Survey streamgage 10333000). Historically, from WYs 1991 to 2015, streamflow was greater at Imlay gage during active mine-dewatering from mine-water discharge operations and increased by as much as 105,000 acre-ft in WY 1998. The increase was attributed mostly to the discharge of groundwater from mine-related dewatering operations directly into the mainstem Humboldt River or its tributaries, with some of this increase associated with return flows from discharge to rapid infiltration basins. Results indicate that streamflow at Imlay gage is expected to decrease by as much as 1,600 acre-ft/yr 30 years after mine-related pumping and discharge are discontinued. The streamflow reductions at the Imlay gage are expected to then decrease to around 500 acre-ft/yr, 100 years after mine-related pumping and discharge are discontinued. Potential capture maps were produced for pumping durations of 10, 25, 50, and 100 years. Capture map results indicate that areas of greater potential stream capture occur adjacent to the Humboldt River and for upstream tributaries areas north of the Humboldt River.

Nevada

REDPy: A Python tool for automated repeating earthquake detection and visualization

Detecting and cataloging seismic events are among the most fundamental tasks in seismology. Many standardized tools for these tasks exist, including the open‐source package repeating earthquake detector in Python (REDPy). REDPy generates an organized catalog of seismic events from continuous waveform data, in which events are automatically separated into groups (“families”) by their waveform similarity through cross‐correlation. REDPy also automatically generates various outputs that allow a user to visualize important trends in the catalog, which may be used in real time or in retrospective analyses to allow rapid identification of interesting features. The code was designed for near‐real‐time volcano monitoring but is applicable across a broad range of use cases in seismology and seismoacoustics. In this article, the utility and performance of REDPy are demonstrated on two highly seismogenic volcanic eruption sequences: the onset of the dome‐building eruption of Mount St. Helens, Washington, from 2004 to 2005, and the entirety of the summit caldera collapse sequence of Kīlauea, Hawai‘i, in 2018. This article is meant to be a companion to the documentation of the code; in addition to detailing the basic required inputs, script functionality, and resulting outputs, the reasonings behind several important design decisions are also discussed.

Seismological Research Letters

Oblique contraction along the fastest ocean-continent transform plate boundary focuses rock uplift west of the Fairweather fault, southeast Alaska

Contraction along the Yakutat–North America plate boundary drives 4.6–9.0 mm/year Holocene rock uplift rates along Earth's fastest slipping (≥49 mm/year) ocean–continent transform fault, the Fairweather Fault. Between Icy Point and Lituya Bay, the near-vertical Fairweather fault focuses rock uplift and rapid right-lateral slip by accommodating both vertical and fault-parallel strain during oblique-slip and separate, predominantly strike-slip ruptures. Unusually high uplift rates, indicated by radiocarbon and luminescence dating, result from a 10-km-wide, asymmetric, positive flower structure along a 20°, ∼30-km-long restraining double bend in the Fairweather fault. The principal reverse fault in the flower structure, the offshore, blind Icy Point–Lituya Bay fault, ruptures no more than every 460–1040 years evidenced by uplifted Holocene shorelines. Maximum 3–5 m coseismic uplifts imply 3.1–10 m dip slip per event and earthquake magnitudes of M w 7.0–7.5. The Yakutat block collides obliquely into North America, and our model entails oblique slip on the Fairweather fault with and without corupture on the reverse fault. Oblique slip is evident by vertically offset (>25 m) fluvial and marine terraces and by the primary Fairweather fault strand that strikes >20° to the west of plate-boundary motion.

Alaska

Deformation mechanisms in quartz veins and shear zones elucidate the origin of gold mineralization at Pogo, Alaska

Pogo is a quartz vein hosted, ca. 8 Moz gold deposit. Although it has similarities to orogenic and magmatic-hydrothermal deposits, its origin remains enigmatic. Observations from surface exposures, underground workings, and drill core provide new constraints on quartz vein origins with implications for mineralization. Abundant, largely barren metamorphic segregation quartz veins are found throughout interior Alaska and share characteristics with mineralized quartz veins at Pogo. Pogo quartz veins show crystal plastic deformation fabrics consistent with greenschist to amphibolite facies metamorphism and a lack of internal shear. In contrast, gold in Pogo quartz veins occurs in brittle microfaults and fractures with sulphides such as arsenopyrite and pyrrhotite. Major and minor cataclastic shear zones also exist at Pogo. Cataclastic shear zones commonly cut the mineralized quartz veins and porphyroclasts are dominated by quartz with the same plastic deformation fabrics found in the major Pogo quartz veins. The porphyroclasts, and the carbonaceous clay-rich matrix they sit in, also contain gold indicating that the shear zones postdate quartz and earliest gold deposition. These observations suggest that competency contrasts between the quartz veins and their phyllosilicate-rich host rocks controlled transient permeability formed by late, preferential brittle deformation events localizing sulphide ± gold deposition.

Alaska

Cursed? Why one does not simply add new data sets to supervised geothermal machine learning models

Recent advances in machine learning (ML) identifying areas favorable to hydrothermal systems indicate that the resolution of feature data remains a subject of necessary improvement before ML can reliably produce better models. Herein, we consider the value of adding new features or replacing other, low-value features with new input features in existing ML pipelines. Our previous work identified stress and seismicity as having less value than the other feature types (i.e., heat flow, distance to faults, and distance to magmatic activity) for the 2008 USGS hydrothermal energy assessment; hence, a fundamental question regards if the addition of new but partially correlated features will improve resulting models for hydrothermal favorability. Therefore, we add new maps for shear strain rate and dilation strain rate to fit logistic regression and XGBoost models, resulting in new 7-feature models that are compared to the old 5-feature models. Because these new features share a degree of correlation with the original relatively uninformative stress and seismicity features, we also consider replacement of the two lower-value features with the two new features, creating new 5-feature models. Adding the new features improves the predictive skill of the new 7-feature model over that of the old 5-feature model; albeit, that improvement is not statistically significant because the new features are correlated with the old features and, consequently, the new features do not present considerable new information. However, the new 5-feature XGBoost model has a statistically significant increase in predictive skill for known positives over the old 5-feature model at p = 0.06. This improved performance is due to the lower-dimensional feature space of the former than that of the latter. In higher-dimensional feature space, relationships between features and the presence or absence of hydrothermal systems are harder to discern (i.e., the 7-feature model likely suffers from the “curse of dimensionality”).

Geothermal Resources Council Transactions

Recycling radiogenic osmium by crustal foundering in subduction zones: Evidence from pyroxenite xenoliths in the north Andean arc

Removal of mafic-ultramafic lower crust (e.g., via delamination) is fundamental to making andesitic continental crust, yet direct evidence of this process remains elusive. A unique suite of garnet clinopyroxenite and hornblendite (arclogite) xenoliths from the northern volcanic zone of the Andes, erupted to the surface in the mid-Pleistocene, have bulk-rock geochemistry and osmium (Os) isotopic compositions that demonstrate their crustal affinity, yet have equilibration pressures and temperatures below the arc Mohorovičić discontinuity (~53 km) and some as deep as within the sub-arc mantle wedge (~105 km). Garnet websterites from the same xenolith suite, sourced from depths approaching the Wadati-Benioff zone (~140 km), have elemental and isotopic compositions indicative of a mantle origin, likely formed as products of peridotite-liquid reactions above the subducting slab. Variability in bulk-rock 143 Nd/ 144 Nd values and garnet oxygen isotope ratios ( 𝛿 18 O) for these samples is attributed to minor assimilation of subducted components and/or older crustal material, but assimilation alone cannot account for the highly radiogenic Os isotopic compositions. These results provide direct petrologic evidence for modern arclogite formation and foundering in the Andean orogen, the archetypal active continental subduction system. Elevated 187 Os/ 188 Os values of Mercaderes arclogites imply that recycling of gravitationally unstable arc cumulates during continental crust formation introduces highly radiogenic Os into the convective mantle, which has implications for mass fluxes across the crust-mantle boundary as well as the sources and evolution of mantle heterogeneity as seen in orogenic peridotite massifs and in the sources of oceanic basalts

Earth and Planetary Science Letters

Rare earth elements on the Moon

Rare earth elements (REEs) are a scarce but vital resource for our modern economies and lifestyles. Since the late 1990s, China has supplied the vast majority of the world’s refined REEs. Increasing global demand has broadened the search for REE deposits to unconventional places, including the Moon. Although most lunar rocks have very low REE concentrations, Apollo samples showed that one type of lunar rock containing potassium (K), REEs, and phosphorus (P)—known by the acronym KREEP—has high concentrations of REEs. Data from orbiting satellites have identified locations where substantial deposits of KREEP are likely. The viability of mining these deposits depends on the evolution of REE economics, the development of the Earth-Moon infrastructure, and the findings from future lunar mineral exploration missions.

Fact Sheet

Digitizer Suite: The Albuquerque Seismological Laboratory Digitizer Testing Suite

Laboratory testing of digitizers and seismometers helps ensure that prior to deployment the instrumentation can produce high quality data and is operating within specifications. In this work we detail the software package called: the Albuquerque Seismological Laboratory (ASL) Digitizer Test Suite. This Java software package provides several algorithms to verify various performance parameters of digitizers commonly used for recording analog seismic instruments. The goal of these tests is not to be exhaustive, but to identify common failures that could compromise the integrity of seismic data being recorded on the digitizer. For example, Sandia National Laboratories (e.g., Slad and Merchant, 2018) routinely do comprehensive testing of digitizers for various monitoring missions. While these tests reports are valuable for comprehensively characterizing a recording system, it would be resource intensive to conduct such tests on every seismic recorder used in a network. We focus on tests that include ways to estimate the sensitivity, timing, self-noise, and clip-level of the digitizer, as well as the fidelity of the signal being recorded. The software is publicly available and provides a way for the community to verify the integrity of a digitizer using a minimum amount of outside equipment.

Seismological Research Letters

Mineralization processes at Escanaba Trough

Escanaba Trough off northern California is the southernmost and only sediment covered segment of the Gorda Ridge. Hydrothermal mineralization was first discovered in Escanaba Trough in 1985, and subsequent investigations included the delineation of the NESCA (northern Escanaba) and SESCA (southern Escanaba) sections of Escanaba Trough, as well as drilling by Ocean Drilling Program (ODP) Leg 169 in 1996. Preliminary research conducted in the 1980s distinguished two classes of sulfide mineral assemblages within the Escanaba Trough: one dominated by pyrrhotite (Fe 1-x S) and one containing ‘mixed sulfides’ including chalcopyrite (CuFeS 2 ) and sphalerite (ZnS). In 2022, we conducted a research expedition to the Escanaba Trough to more thoroughly characterize mineralization processes and consider preservation of precipitated minerals throughout hydrothermally active and inactive sections of Escanaba Trough, with the specific objective of characterizing the host minerals, and processes of enrichment for specific elements which are currently listed as critical elements and can also be toxic under certain conditions (Sb, As, Ba, Co, Ga, Ge, Te). The following will include initial results from that work.

California

AAPG Energy and Minerals Division Tight Oil and Gas Committee Activities and Commodity Report for 2021-2022: Tuscaloosa Marine Shale, Gulf Coast basin, Louisiana and Mississippi

The Upper Cretaceous Tuscaloosa marine shale (TMS) potential production area encompasses 20.4 million acres across central Louisiana (LA), southern Mississippi (MS), southwestern Alabama (AL), and a small southwestern section of the Florida panhandle (Hackley et al., 2018). It remains a minor and largely undeveloped unconventional shale oil play with production from the TMS confined along the east-west LA-MS State boundary (Fig. 1). The mean undiscovered, technically recoverable resources in the TMS are estimated at 1.5 billion barrels of oil and 4.6 trillion cubic feet of gas (Hackley et al., 2018). Geochemical analyses of source rock solvent extracts and oil samples indicate that, in the play area, the TMS is the primary source of shale oil produced from the TMS as well as of accumulated oils in the underlying conventional reservoirs of the lower Tuscaloosa (Hackley et al., 2020). The API gravity of TMS oils ranges from approximately 34 to 46 degrees (Hackley et al., 2020; Croke et al., 2020).

AAPG Wiki

Groundwater flowpath characteristics drive variability in per- and polyfluoroalkyl substances (PFAS) loading across a stream-wetland system

Groundwater-dependent ecosystems in areas with industrial land use are at risk of exposure to a PFAS chemicals. We investigated one such system with several known PFAS source areas, where high and low permeability sediments (glacial) coupled with groundwater-lake and groundwater/surface-water interactions created complex ‘source to seep’ dynamics. Using heat-tracing and chemical methods, numerous preferential groundwater discharge zones were identified and sampled across the upper Quashnet River stream-wetland system in Mashpee, MA, USA, downgradient of Joint Base Cape Cod (JBCC). Surface-water and groundwater samples were analyzed for 40 PFAS compounds between March and October 2022. Samples were collected from groundwater seeps identified as preferential discharge points (PDPs), wells upgradient of the stream-wetland system, contributing flow-through kettle lakes, and along Quashnet River surface-waters. PFAS from sampled waters contained perfluorinated carboxylates (PFPea, PFHxA, PFNA), perfluorinated sulfonates (PFBS, PFPeS, PFHxS, PFOS), fluorotelomer sulfonates (6:2, 8:2 FtS), and perfluoroalkyl sulfonamides (PFOSA). Samples from PDPs and wells had measured PFAS concentrations ranging from non-detect to 4677 ng/L ng/L (mean = 418 ng/L, std. = 709 ng/L), and a range of deuterium excess values (3.2 to 15.9 per mil) indicative of varying degrees of groundwater-lake interaction prior to groundwater flowpath emergence at PDPs. Correlations ( p < 0.01) between deuterium excess, %PFAS precursors, and terminal PFAS compounds highlighted potential precursor transformations associated with lake-groundwater exchange along flowpaths sourcing PDPs. However, some seepages had higher total PFAS concentrations (>1000 ng/L) than upgradient kettle lakes despite showing lake (evaporative) isotopic signatures, indicating the potential for groundwater flowpath convergence at wetland discharge zones and the influence of lakebed PFAS precursor reactions. Results from these synoptic surveys address gaps in the existing PFAS literature by demonstrating the importance of subsurface fate and transport on PFAS compound concentrations and mass loading in preferential groundwater discharge zones.

Massachusetts

Enhanced geothermal systems electric-resource assessment for the Great Basin, southwestern United States

The U.S. Geological Survey recently (2025) completed a provisional assessment of the geothermal-electric resources associated with high-temperature, low-permeability rock formations of the Great Basin, Southwestern United States. If sufficient technological advances to commercialize enhanced geothermal systems occur, then a current best provisional estimate for electric-power generation capacity of 135 gigawatts electric are available from the upper 6 kilometers of the Earth’s crust. This estimate is a potential substantial increase of the installed geothermal electricity-generating capacity from <1 to 10 percent of current total U.S. power production capacity.

California, Idaho, Nevada, Oregon, Utah

Mid-Cretaceous porphyritic magmatism in the Beartooth Mountains of the northern Laramide foreland and its connection to Frontier Formation sediment dispersal in Bighorn basin, Montana-Wyoming (USA), and Cordilleran tectonism

Porphyritic intrusions in the southeast Beartooth Mountains and porphyritic cobbles from the Torchlight conglomerate of the Cenomanian Frontier Formation in Bighorn basin (Montana-Wyoming, USA) bear striking resemblance. We utilize geologic mapping, petrography, geochemistry, and geochronology to characterize rocks from both locations and test whether the Beartooth Mountains area was the provenance for Bighorn basin cobbles. Mapping on Line Creek Plateau in the Beartooth Mountains shows three units with porphyritic phenocrysts consisting dominantly of plagioclase, plagioclase + potassium feldspar, and plagioclase + quartz. Petrography shows embayed quartz, opacitic rims, and zoned plagioclase. Geochemistry characterizes the intrusions as trachydacite to trachyte and trace elements exhibit enrichment in light rare earth elements and slight depletion in heavy rare earth elements relative to primitive mantle. Zircon U-Pb dating shows substantial Archean inheritance and ca. 104−92 Ma dates; some samples display multiple mid-Cretaceous date populations. Bighorn basin porphyritic cobbles exhibit similar modal mineralogy, near-identical geochemistry, and Archean inheritance and mid-Cretaceous geochronology. We suggest the porphyritic cobbles were sourced from basement-hosted porphyritic intrusions now exposed in the Beartooth Mountains, given that geochemistry does not agree with coeval volcanic sources along the Frontier Formation depositional fairway (Idaho batholith) or sedimentary-hosted Beartooth Mountains porphyritic sills intruded into Cambrian shale. Stratigraphic evidence does not support kilometer-scale mid-Cretaceous Beartooth Mountains exhumation to expose basement-hosted porphyritic intrusions, and we therefore present a volcanic hypothesis to bring rocks to the surface. This mid-Cretaceous multimillion-year magmatism is unexplained by tectonic models for the north-central Laramide foreland, and thus we further explore the emplacement processes, spatiotemporal significance, and tectonic implications of these rocks.

Montana, Wyoming

World minerals outlook—Cobalt, gallium, helium, lithium, magnesium, palladium, platinum, and titanium through 2029

Given the rapid expansion in the demand for mineral commodities that underpin worldwide economic growth and technological advancement, information regarding expected country-level mine production and production capacity is becoming increasingly important to industry stakeholders, end users, and policymakers. Production capacity can limit future supply, depending on how rapidly that capacity is able to expand. Current capacity can be evaluated on the basis of past production. Decreases to future capacity can be taken into account from announcements of planned shutdowns of mines or processing facilities, which are frequently publicized well in advance of such closures. Likewise, capacity expansions, which usually involve multiple stages—such as permitting, financing, and construction (all of which take time)—can also be estimated. As such, it is possible to evaluate midterm future capacity based on estimates of today’s capacities along with consideration of future investment plans. This World Minerals Outlook provides estimated capacities for cobalt, gallium, helium, lithium, magnesium, palladium, platinum, and titanium for 2025 through 2029. The results of this analysis indicate that two mineral commodities important to the manufacture of lithium-ion batteries—cobalt and lithium—are expected to have large capacity growth in the next few years, likely owing to expectations for increased demand for these batteries. For gallium, helium, palladium, and platinum, capacity is expected to remain stable or exhibit moderate growth. Still, these expected capacity levels are higher than current production, allowing for future production growth. The production capacity outlook is opaque for magnesium and titanium metal, which have a significant fraction of current production in nonmarket economies, such as China and Russia. Ultimately, though, where free market conditions prevail, full utilization of capacity potential for those commodities is likely to depend on supply deficits and prices that are above production costs.

Scientific Investigations Report

Characterization and validation of tidally calibrated strains from the Alto Tiberina Near Fault Observatory Strainmeter Array (TABOO-NFO-STAR)

Six horizontal borehole tensor strainmeters (TSM1-6) installed from Fall 2021 to Spring 2022 comprise the Alto Tiberina Near Fault Observatory Strainmeter Array (STAR), providing an unprecedented opportunity to investigate seismic and aseismic deformation from hazardous high- and low-angle normal faults in Italy. Prior to use in tectonic applications, they require in-situ calibration and correction for non-tectonic signals. We tidally calibrate the instruments, characterize the calibration uncertainty, and test the results against environmental and earthquake signals originating from local to teleseismic distances. The STAR sites demonstrably deviate from assumptions common to the standard manufacturer's calibrations, including negative areal coupling at TSM3-6. While the tidally calibrated strains have ~3-56% uncertainty, the calibrated dynamic strains show interstation precision and accuracy to nanostrain levels, and static coseismic offsets in the array footprint are within uncertainty. TSM3 records a complex series of strains that may arise from dynamically triggered near-borehole fracture slip and fluid flow that does not appear to affect its sensitivity to lower strain rate deformation. Future calibration improvement may be afforded with longer stable timeseries, particularly for TSM4. Overall, our analyses demonstrate expanded geodetic capability for detecting deformation in the Alto Tiberina Near Fault Observatory.

Alto Tiberina Near Fault Observatory

Extracting data from maps: Lessons learned from the artificial intelligence for critical mineral assessment competition

The U.S. Geological Survey (USGS), Defense Advanced Projects Research Agency (DARPA), NASA Jet Propulsion Laboratory (JPL), and MITRE ran a 12-week machine learning competition aimed at accelerating development of AI tools for critical mineral assessments. The Artificial Intelligence for Critical Mineral Assessment Competition solicited innovative solutions for two challenges: 1) automated georeferencing of historical maps, and 2) automated feature extraction from historical maps. Competitors used a new dataset of historical map images to train, validate, and evaluate their models. Automated georeferencing pipelines attained a median root-mean square error of 1.1 km. Prompt-based extraction (i.e., with user input) of polygons, polylines, and points from geologic maps yielded median F1-scores of 0.77, 0.56, 0.35, respectively. Geologic maps pose numerous challenges for AI workflows because they vary significantly. However, despite its short duration, the competition yielded promising results that have since spurred further innovation in this area and led to the development of new AI tools to semi-automate key, time-consuming parts of the assessment workflow.

Applied Computing and Geosciences