Grappling with systemic discrimination and bias in geosciences can be overwhelming to the point that one may feel powerless to fix them. Despite the sweeping nature of this challenge, faculty, principal investigators, and other scientists with leadership roles have unparalleled power to mitigate harm in environments they oversee. Here, we identify ways that scientists in these roles can immediately address bias in three common spaces --- the lab, field, and classroom. We highlight key actions that can be taken to improve the quality of life of marginalized students and other trainees quickly, while important but comparatively slow institutional changes proceed.

Reconstructing water availability in terrestrial ecosystems is key to understanding past climate and landscapes, but there are few proxies for aridity that are available for use at terrestrial sites across the Cenozoic. The isotopic composition of tooth enamel is widely used as paleoenvironmental indicator and recent work suggests the potential for using the triple oxygen isotopic composition of mammalian tooth enamel (∆’17Oenamel) as an indicator of aridity. However, the extent to which ∆’17Oenamel values vary across environments is unknown and there is no framework for evaluating past aridity using ∆’17Oenamel data. Here we present ∆’17Oenamel and δ18Oenamel values from 50 extant mammalian herbivores that vary in physiology, behavior, diet, and water-use strategy. Teeth are from sites in Africa, Europe, and North America and represent a range of environments (humid to arid) and latitudes (34S to 69N), where mean annual δ18O values of meteoric water range from -26.0‰ to 2.2‰ (VSMOW). ∆’17Oenamel values from these sites span 146 per meg (-283 to -137 per meg, where 1 per meg = 0.001‰). The observed variation in ∆’17Oenamel values increases with aridity, forming a wedged-shape pattern in a plot of aridity index vs. ∆’17Oenamel that persists regardless of region. In contrast, the plot of aridity index vs. δ18Oenamel for these same samples does not yield a distinct pattern. We use these new ∆’17Oenamel data from extant teeth to provide guidelines for using ∆’17Oenamel data from fossil teeth to assess and classify the aridity of past environments. ∆’17Oenamel values from the fossil record have the potential to be a widely used proxy for aridity without the limitations inherent to approaches that use δ18Oenamel values alone. In addition, the data presented here have implications for how ∆’17Oenamel values of large mammalian herbivores can be used in evaluations of diagenesis and past pCO2.

Distribution: A. Approved for public release – distribution is unlimited QuakeCast is a novel system for short-term earthquake prediction using global ionosphere Total Electron Content (TEC) data. In the last 20 years, earthquakes have caused over 800,000 deaths and $650 billion in economic damage. While current seismic early warning systems provide up to a minute’s notice by detecting an earthquake’s non-damaging P-waves prior to the damaging S-waves, earlier warnings could allow more significant emergency preparations. Electromagnetic ionospheric phenomena have recently been observed many days before major earthquakes, notably in 2015 in Nepal. QuakeCast explores whether such signals could be used to predict earthquakes in a global dataset of ionosphere and earthquake data. The ionosphere data consists of global TEC data from NASA’s Crustal Dynamics Data Information System (CDDIS), for the years 2005-2015 at 15 minute intervals. The earthquake data consists of over 10,000 events of at least M5 from the International Seismological Centre (ISC-GEM) Global Instrumental Earthquake Catalogue over the same time period. We used the dataset to explore the following classification problem: given a 24-hour sequence of ionosphere TEC data in a 30° latitude by 30° longitude window, is the sequence preseismic (an earthquake occurs in the next time step) or nominal (no earthquake)? We built two models to address this. The first is a classical logistic regression model trained on radially binned data. Analysis of the classifier weights indicate a localized effect in the ionosphere: TEC data spatially and temporally closer to the event contributes more significantly to the prediction. The second model is a deep learning ConvLSTM autoencoder trained to reproduce nominal TEC data sequences. Since autoencoders reproduce data which resemble their training data better than data which does not, we used the reconstruction error to classify sequences as anomalous (preseismic) or nominal. Both methods were found to perform significantly better than a random null classifier, indicating that the ionosphere contains information useful for predicting earthquakes at least 15 minutes in advance. While further research is needed to incorporate additional data features and address noise, we believe that these results are a promising development toward forecasting earthquakes using geoelectric signals.

A document by Richard W Eastes. Click on the document to view its contents.

Nonlinear wave-particle interactions contribute to the acceleration and precipitation of electrons in the outer radiation belt. Recent simulations and spacecraft observations suggest that oblique whistler-mode chorus can cause loss cone overfilling through nonlinear Landau resonance and thus break the strong diffusion limit of quasilinear theories. Here we show with test-particle simulations that a single element of parallel-propagating chorus can also break the diffusion limit through nonlinear cyclotron resonance, as long as its amplitude remains high. This is due to the strong scattering at low pitch angles caused by individual chorus subpackets. We further demonstrate that the subpacket modulations create a discernible pattern in the precipitating electron fluxes, with peaks correlated with the largest subpackets. Such flux patterns may be connected to weak micropulsations within diffuse auroras.

Snowfall has important significance in water resources management and disaster prevention worldwide. Accurate prediction of both mean and extreme snowfall is challenging because of multiple controlling mechanisms at different spatial and temporal scales. By using a 65 years long in-situ snowfall observation, we evaluated seven different machine learning algorithms for predicting monthly snowfall in the Lower Peninsula of Michigan (LPM). The Bayesian Additive Regression Trees (BART) demonstrates the best fitting (R2 = 0.88) and out-of-sample prediction skills (R2 = 0.58) for the monthly mean snowfall followed by the Random Forest model. The BART also demonstrate strong predictive skills for seasonal and the extreme monthly snowfall. Both machine learning models also demonstrate signals of key physical processes controlling the snowfall including topography, local/regional environmental factors, and teleconnections. Particularly, models with the non-parametric framework can incorporate signals from multiple scales and nonlinear responses from the snowfall to environmental factors and that substantially improved the model prediction skills. The multiscale machine learning approach provides a reliable and computationally efficient alternative approach to predict/forecast weather and climate and has potential to be applied to other extreme weather prediction scenarios.

The burgeoning field of socio-hydrology aims to understand the dynamics of coupled human-water systems in order to inform sustainable water management. However, socio-hydrological methods have traditionally relied on the quantification of qualitative social data, which runs into significant epistemological challenges between the positivist paradigm of hydrology and the interpretivist paradigm of much of social science. This article builds on recent literature that advocates for a pluralistic approach, retaining the methodological and epistemological differences inherent to social sciences and hydrology, and thereby revealing a more complete understanding of situated human-water relationships. In this vein, we propose rapid ethnography as a tool to complement hydrological modeling. We demonstrate the utility of this technique with a case study on hydrological entanglements of rural livelihoods in Andes, Antioquia, Colombia–a region dominated by agriculture, but with the presence of artisanal and small-scale gold mining (ASGM). Our ethnographic study builds on previous hydrological models and quantitative social studies of the region by exploring a myriad of local risk perceptions of hydrological systems, in particular, 1) climate change, 2) water contamination, and 3) hydrological erosion and landslides. We explore how these risk perceptions informed livelihood decisions that paradoxically threatened the very hydrological resources that the livelihoods depended on; yet, this cycle was difficult to interrupt given prevailing economic and political constraints. Ultimately, we seek to advocate for pluralistic approaches in socio-hydrology to help understand complex human-water relationships and ultimately identify critical points of intervention for sustainable water resource management, model-based or otherwise.

Geophysical monitoring of geologic carbon sequestration is critical for risk assessment during and after carbon dioxide (CO2) injection. Integration of multiple geophysical measurements is a promising approach to achieve high-resolution reservoir monitoring. However, joint inversion of large geophysical data is challenging due to high computational costs and difficulties in effectively incorporating measurements from different sources and with different resolutions. This study develops a differentiable physics model for large-scale joint inverse problems with reparameterization of model variables by deep neural networks and implementation of a differentiable programming approach of the forward model. The main novelty is the use of automatic differentiation and parallel computing for efficient multiphysics data assimilation. The application to the Sleipner benchmark model demonstrates that the proposed method is effective in estimation of reservoir properties from seismic and resistivity data and shows promising results for CO2 storage monitoring.

Previous studies have found that Northern Hemisphere aerosol-like cooling induces a La Nina-like quasi-equilibrium response in the tropical Indo-Pacific. Here, we explore a coupled atmosphere-ocean feedback pathway by which this response is communicated. We override ocean surface wind stress in a comprehensive climate model to decompose the total ocean-atmosphere response to forced extratropical cooling into the response of surface buoyancy forcing alone and surface momentum forcing alone. In the subtropics, the buoyancy-forced response dominates: the positive low cloud feedback amplifies sea surface temperature (SST) anomalies which are then communicated to the tropics via wind-driven evaporative cooling. In the deep tropics, the momentum-driven Bjerknes feedback creates zonally asymmetric SST patterns in the Indian and Pacific basins. Although subtropical cloud feedbacks are model-dependent, our results suggest this feedback pathway is robust across a suite of models such that models with a stronger subtropical low cloud response exhibit a stronger La Nina response.

We characterize the vortex and dust devil activity at Jezero from pressure and winds obtained with the MEDA instrument on Mars 2020 over 415 sols (Ls=6-213º). Vortices are abundant (4.9 vortices per sol with pressure drops >0.5 Pa when correcting from gaps in coverage) and peak at noon. At least one in every 5 vortices carries dust from RDS-MEDA data, and intense vortices are more likely to carry dust. Seasonal variability was small but dust devils were abundant during a dust storm (Ls=152-156º). Vortices are more frequent and intense over terrains with lower thermal inertia favoring a higher daytime surface-to-air temperature gradient. We fit measurements of wind and pressure during dust devil encounters to models of vortices, and investigate their physical characteristics. Diameters range from 5 to 135 m with a mean of 20 m. Three 100-m size events passed within 30 m of the rover. From the close encounters we estimate a dust devil activity of 2.0-3.0 dust devils km$^{-2}$ sol$^{-1}$. A comparison of MEDA observations with a Large Eddy Simulation of Jezero at Ls=45º produces a similar result. We estimate that large dust devils with diameters $>100$ m have a density of 0.1 dust devils km-2sol-1, implying that dust lifting is dominated by the largest vortices in the region. At least one vortex had a central pressure drop of 9.0 Pa and internal winds of 25 ms-1. The MEDA wind sensors were partially damaged during two dust devil encounters, and we detail these events.

Effective permeability is a key physical property of porous media that defines its ability to transport fluid. Digital rock physics combines modern tomographic imaging techniques with advanced numerical simulations to estimate effective rock properties. Digital rock physics is used to complement or replace expensive and time-consuming or impractical laboratory measurements. However, with increase in sample size to capture multimodal and multiscale microstructures, conventional approaches based on direct numerical simulation (DNS) are becoming very computationally intensive or even infeasible. To address this computational challenge, we propose a hierarchical homogenization method (HHM) with a data-driven surrogate model based on 3-D convolutional neural network (CNN) and transfer learning to estimate effective permeability of digital rocks with large sample sizes up to billions of voxels. This workflow (HHM-CNN) divides the large digital rock into small sub-volumes and predicts the sub-volume permeabilities through a CNN surrogate model of Stokes flow at the pore scale. The effective permeability of the full digital rock is then predicted by solving the Darcy equations efficiently on the upscaled model in which the permeability of each cell is assigned by the surrogate model. The proposed method is verified on micro-CT data of both sandstones and carbonates as well as the reconstructed high-resolution digital rock obtained by multiscale data fusion. The computed permeabilities of our proposed hierarchical approach are consistent with the results of the DNS on the full digital rock. Compared with conventional DNS algorithms, the proposed hierarchical approach can largely reduce the computational time and memory demand.

Constraining the properties of continental margin where dynamic interplay between the lithosphere and convective upper mantle actively occurs can provide implications to understand the evolution and modification process of cratonic lithosphere. Cratonic lithosphere in the eastern Eurasian plate has experienced multiple episodes of tectonic events at the continental margin during the Phanerozoic, which are characterized by collision and subduction of continental and oceanic plates with extensive magmatic intrusion, extension, and rifting. To better understand the complex tectonic processes and their influences on the continental lithosphere, we imaged upper mantle structure beneath the Archean-Proterozoic basement in the Korean Peninsula (KP) by seismic tomography. Our dataset is entirely based on relative arrival time of teleseismic body wave recorded by dense seismic array across the southern KP in order to extract velocity variations from the local area and reduce the effects from deeper upper mantle structures. The images show a sharp and large lateral velocity variation within a short distance (< 200 km) in KP, which indicates highly variable lithospheric structure. An anomalously thick high-velocity structure beneath the southwestern part suggests a fragment of a long-lasting cratonic lithosphere (~150 km) at the continental margin. The absence of deep lithosphere and mostly occupied by hot, buoyant, low-velocity upper mantle beneath the Gyeonggi massif, Gyeongsang continental arc-backarc system, and eastern continental margin of KP shows highly modified regions. In addition, there are clear spatial correlations of low-velocity upper mantle with partial melting zones and the localities of Cenozoic basalts, high surface heat flux, and high topography with recent uplift along the eastern mountain ranges. A strong variation in the upper mantle velocity structure without clear distinctions in surface geology across different massifs suggests heterogeneous modification of continental lithosphere by recent and transient processes, such as the opening of the East Sea (Sea of Japan) or a subduction of Philippine Sea Plate. Dynamic interaction of the prominent lithospheric structure and convective upper mantle has controlled reactivation and destabilization at the cratonic margin of KP.

In the context of climate change, a clear understanding of the processes and factors driving global warming is a major concern. During past geological times, Earth suffered several intervals of global warmth but the priming factors remain equivocal. Yet a careful appraisal of all processes being implied during those past events is essential to evaluate how they can inform future climates, in order to provide decision makers a clear understanding of the processes at play in a warmer world. In this context, the global warmth of the Cretaceous era, specifically during the Cenomanian-Turonian, is of particular interest. Here we use the IPSL-CM5A2 Earth System model to unravel the forcing parameters of the Cenomanian-Turonian greenhouse climate. We perform six simulations, from the preindustrial to the Cretaceous by implementing one additional boundary condition change at a time, i.e. (1) polar ice cap retreat, (2) pCO2 increase to 1120 ppm, (3) vegetation and soil parameters, (4) solar constant reduction (~ -1%) and (5) paleogeography (90Ma). Between the first preindustrial simulation and the last Cretaceous simulation, a global warming of more than 11°C is simulated. Most of this warming is driven by the increase in pCO2 to 1120 ppm. Paleogeographic changes represent the second major contributor to the warming while the solar constant reduction counteracts most of this geographically-driven warming. Finally, changes in vegetation and soil parameters as well as the retreat of polar ice caps have a minor impact at the global scale. A full assessment of the processes driving warming or cooling under each boundary condition change will be presented. Ultimately, our work supports the overarching influence of atmospheric carbon dioxide in driving the Earth’s global climate and global warming.

Sulfate is a major atmospheric pollutant and radiative forcing (RF) factor that influences air quality, cloud microphysics and climate. Therefore, a better evaluation of sulfate concentrations and RF patterns is essential for policy-making and the management of air pollution and climate change. This study comprehensively estimates the global distribution of sulfate concentrations and RFs and analyzes the sources of uncertainty in the Community Earth System Model version 2 (CESM2) and the Parallel Offline Radiative Transfer (PORT) model. Compared with the observations, the incorporation of detailed in-cloud aqueous-phase chemistry and the enhanced wet deposition flux of sulfate significantly improved the simulations of sulfur species both near the ground and at high altitudes, which is beneficial for a more accurate estimation of the global sulfate RF. The improved simulated RF of sulfate from 1850 to 2015 is -0.382 Wˑm-2. This study finds that wet deposition is the key process governing both the horizontal and vertical distributions of sulfate concentrations. The overestimation of surface sulfate and the underestimation of high-altitude sulfate made by the model are essential uncertainty factors of the sulfate RF estimation. This study emphasizes the importance of improving the simulation of global sulfate distribution as well as its RF, which may strongly pressure the near-future warming potential when witnessing a rapid transition to a carbon neutral world that is phasing out fossil fuel. A more accurate assessment of sulfate levels and radiation effects will play a remarkable guiding role in the formulation of global emission reduction-related policies in the future.

For over fifty years, cores recovered from ocean basins have generated extensive fossil, lithologic, and chemical archives that have revolutionized the fields of plate tectonics and oceanography, and significantly improved our understanding of climate change. Although scientific ocean drilling (SOD) data are openly available after each expedition, formats for these data are heterogeneous. Furthermore, lithological, chronological, and paleobiological data are typically separated into different repositories, limiting researchers’ abilities to discover and analyze integrated SOD data sets. Emphasis within Earth Sciences on adhering to FAIR Data Principles and the establishment of community-lead databases provide a pathway to unite SOD data and further harness the scientific potential of the investments made in offshore drilling. Here, we describe a workflow for compiling, cleaning, and standardizing key SOD records, and importing them into the Paleobiology Database (PBDB) and Macrostrat, systems with versatile, open data distribution mechanisms. These efforts are being carried out by the extending Ocean Drilling Pursuits (eODP) project. eODP has processed all of the lithological, chronological, and paleobiological data from one SOD repository, along with numerous other datasets that were never deposited in a database; these were manually transcribed from original reports. This compiled dataset contains over 78,000 lithological units from 1,048 drilling holes from 390 sites. Over 26,000 fossil-bearing samples, with 5,280 taxonomic entries from 13 biological groups, are placed within this lithologic spatiotemporal framework. Information is available via the PBDB and Macrostrat application programming interfaces, which render data retrievable by a variety of parameters, including age, taxon, site, and lithology.

For the analysis of water distribution systems, hydraulic simulation is of great importance; however, there are no suitable and holistic simulation tools for Intermittent Water Supply (IWS), so researchers are forced to utilise EPA-SWMM software, which needs some modifications to make it suitable for IWS analysis. This study focuses on providing a generic IWS analysis method in EPA-SWMM. Firstly, IWSs were classified according to coping strategies (e.g., household tank installation) customer satisfaction, supply schedule and duration to provide a unified terminology to the body of knowledge and assist in predicting the systems’ behaviour, which should be reflected in the modelling process. Then EPA-SWMM based modelling approaches were compared paying attention to the case studies applied and two improved hydraulic modelling approaches have been proposed depending on the customer satisfaction of the network (namely i) unsatisfied IWS, ii) satisfied and partially satisfied IWS). Ultimately, customer satisfaction is the key, as the IWS systems are driven by user water consumption behaviours that are based on the demand-supply relationship. The application of the proposed methods to two different case studies and comparison with the state-of-the-art methods demonstrated their capability, compatibility with EPA-SWMM operating principles, and the ability to simulate the IWS system behaviour in a more realistic manner. In addition, the proposed methods are generic, unlike the available ones, which, in most cases, are only applicable to the specific case study.

The properties of the inner core boundary (ICB) carry important information about the growth of the solid inner core and solidification of the liquid outer core. In this presentation, we study the ICB by analyzing the spectrum difference of PKiKP (a compressional wave reflected from the ICB) and PcP (a compressional wave reflected from the core-mantle boundary) phase pair. Our dataset includes the seismic data recorded by the Hi-net in Japan for two earthquakes occurring in Myanmar on 4th September 2009 and 4th February 2011, with their PKiKP waves sampling the ICB beneath mid-eastern China. Little waveform and spectrum differences are observed between PKiKP and PcP waves in the seismic data. By comparing spectrum shapes of the seismic data and the synthetics of models with various topographic fluctuations and transitional thicknesses of ICB, we conclude that the studied ICB region is flat and sharp. Within the resolvability of the seismic data, the ICB topographic fluctuation of the region is less than 200 m in height and the transition of the boundary is thinner than 100 m in thickness.

The timescale over which magma rises to the surface during a volcanic eruption is a critical parameter that controls eruption style but is challenging to determine. Melt embayments, which are pockets of magma trapped within crystals but are still open to the host magma, tend to develop volatile (e.g., H2O, CO2) concentration gradients during ascent through the process of diffusion. These gradients are increasingly used as output constraints for diffusion models to calculate total ascent time. However, there are two main assumptions associated with the method that have yet to be properly evaluated. First, the diffusion models tend to be carried out in 1D, which may be less accurate than 3D models if the embayment narrows near the opening. This geometry has more complicated volatile flux pathways that 1D models cannot account for. Second, the models typically assume that volatile contents follow equilibrium saturation trends defined by depth below the surface, which may not be accurate for high silica magmas. For these compositions, a significant supersaturation of volatiles may be needed for volatile exsolution to start, depending on the availability of suitable crystal phases for bubble nucleation. This nucleation lag delays the start of diffusion as well and could influence modeled timescales. We developed 3D diffusion models to evaluate the degree of inaccuracy in calculated timescales introduced by these assumptions for high silica magmas. We built synthetic embayments with variable 3D geometries and imposed both equilibrium and disequilibrium ascent conditions in our models. Our results indicate that the inaccuracy introduced by embayment geometry in 1D diffusion models increases as the diameter of the embayment opening decreases. This degree of inaccuracy is also tied to diffusion duration – as it increases, so does the relative error in timescales retrieved by 1D models fitting 3D-generated profiles. Once the diffusion front extends beyond the narrow, “necked” region of the embayment, relative error increases with increasing diffusion time. However, volatile disequilibrium behavior does not appear to significantly impact modeled timescales. These results indicate that embayment geometry is a critical variable that must be accounted for whereas delayed nucleation is of secondary influence.