Earth Science Applications Research Articles

Precipitation isotope time series predictions from machine learning applied in Europe

Hydrogen and oxygen isotope values of precipitation are critically important quantities for applications in Earth, environmental, and biological sciences. However, direct measurements are not available at every location and time, and existing precipitation isotope models are often not sufficiently accurate for examining features such as long-term trends or interannual variability. This can limit applications that seek to use these values to identify the source history of water or to understand the hydrological or meteorological processes that determine these values. We developed a framework using machine learning to calculate isotope time series at monthly resolution using available climate and location data in order to improve precipitation isotope model predictions. Predictions from this model are currently available for any location in Europe for the past 70 y (1950-2019), which is the period for which all climate data used as predictor variables are available. This approach facilitates simple, user-friendly predictions of precipitation isotope time series that can be generated on demand and are accurate enough to be used for exploration of interannual and long-term variability in both hydrogen and oxygen isotopic systems. These predictions provide important isotope input variables for ecological and hydrological applications, as well as powerful targets for paleoclimate proxy calibration, and they can serve as resources for probing historic patterns in the isotopic composition of precipitation with a high level of meteorological accuracy. Predictions from our modeling framework, Piso.AI, are available at https://isotope.bot.unibas.ch/PisoAI/.

Outlook for Exploiting Artificial Intelligence in the Earth and Environmental Sciences

AbstractPromising new opportunities to apply artificial intelligence (AI) to the Earth and environmental sciences are identified, informed by an overview of current efforts in the community. Community input was collected at the first National Oceanic and Atmospheric Administration (NOAA) workshop on “Leveraging AI in the Exploitation of Satellite Earth Observations and Numerical Weather Prediction” held in April 2019. This workshop brought together over 400 scientists, program managers, and leaders from the public, academic, and private sectors in order to enable experts involved in the development and adaptation of AI tools and applications to meet and exchange experiences with NOAA experts. Paths are described to actualize the potential of AI to better exploit the massive volumes of environmental data from satellite and in situ sources that are critical for numerical weather prediction (NWP) and other Earth and environmental science applications. The main lessons communicated from community input via active workshop discussions and polling are reported. Finally, recommendations are presented for both scientists and decision-makers to address some of the challenges facing the adoption of AI across all Earth science.

Application of Remote Sensing in Earth Sciences – A Review

The application of remote sensing sciences in the field of geology is very diverse and wide. One of its most important applications in earth sciences is geological mapping. Mineral exploration using remote sensing techniques is done in different ways, one of them is the mapping alteration zones related to mineral resources. Given the importance of remote sensing and geosciences in today's industry and given that deposit-related alteration areas are one of the most important exploratory keys. in this review study the mapping methods and alteration zones detection using remote sensing techniques and other applications of remote sensing in earth sciences and its generalities are explained.

A novel geomagnetic assimilation method of vector geomagnetic data

Geomagnetic field has important applications in Earth Science, earthquake prediction, aerospace, national defense construction and other areas. Therefore, accurate modeling of regional geomagnetic field is very important. Geomagnetic assimilation method can effectively improve the accuracy of regional vector geomagnetic data, and then promote the establishment of high-precision geomagnetic regional model. The geomagnetic stations at different latitudes and longitudes have different influences on the geomagnetic assimilation of vector geomagnetic data. However, traditional geomagnetic assimilation methods do not take into account the differences in latitude and longitude, which result in low accuracy of the geomagnetic assimilation of vector geomagnetic data. To solve the problem, this paper proposes a new geomagnetic assimilation method based on the three weighting factors of direction, distance and a parameter k. In this method, the longitude differences and latitude differences of different stations are taken as the weight elements to participate in the weight determination, and parameter k is set to further modify the weight, so as to achieve the high precision geomagnetic assimilation to regional vector geomagnetic data. Based on the real observation data of five stations in the European region of station network in the world data center (WDC), this paper conducted geomagnetic assimilation of the vector geomagnetic data of different stations. The experimental results show that at five stations which include some central stations and some marginal stations, for the F, H, D and I geomagnetic elements, the geomagnetic assimilation accuracy of the proposed method is better than that of the method based on the distance weighting method both on magnetically quiet day and on magnetically disturbed day. In addition, the results also show that the geomagnetic assimilation accuracy of FUR station, which is a marginal station, is significantly improved by the proposed method. On magnetically disturbed day, the error of F component is reduced by 0.6 nT, and the error of H component is reduced by 1 nT. On magnetically quiet day, the error of F component is reduced by 0.4 nT, and the error of H component is reduced by 0.1 nT. The above results prove the effectiveness and feasibility of the proposed method in the geomagnetic assimilation of vector geomagnetic data, especially for the vector geomagnetic data of marginal area.

Investigation on the accuracy of ground penetrating radar in the tunnel based on improved analectic hierarchy process

With the rapid development of tunnel engineering in China, geological conditions encountered in tunnel construction are becoming more and more complex. Construction safety has become the top priority of tunnel construction. To avoid geological hazards such as landslides, water inrush, and mud gushing caused by blind construction, ground penetrating radar (GPR) is widely used in the medium distance ahead of a tunnel face. The advanced prediction of GPR is often interfered by factors such as environment, operation, data processing, image interpretation, and other factors, which cause final results to have errors. Therefore, it is highly necessary to track and test prediction effects, and also, to choose scientific and effective evaluation methods for accuracy of the prediction result. Based on the geological radar forecast data of the Yujiashan tunnel, this paper sums up the interference factors of the ground radar forecast and discusses the advantages and disadvantages of different evaluation methods. On this basis, the analytic hierarchy process (AHP) which has many applications in earth sciences, physics, nanotechnology and etc., is improved by integrating the principle of maximum membership of fuzzy comprehensive evaluation. Then, a four-level evaluation model is established, which judges the accuracy of GPR geological prediction. Meanwhile, two geological predictions in the Yujiashan tunnel are selected to be applied in the model. Finally, the recognition level of the accuracy evaluation model is more accurate. The comparison between the actual excavation exposure and the radar image interpretation results shows that the geological radar advance prediction results are more accurate. This verifies the rationality and applicability of using the improved AHP to evaluate the accuracy of geological radar advance prediction.

The Organic Isotopologue Frontier

Organic molecules are key components of the Earth-life system. Their stable isotope composition provides information on various problems such as past climate, energy resources, or the synthesis of prebiotically relevant molecules on Earth and elsewhere. Organic molecules are made of isotopologues, molecules differing in the number and/or position of isotope substitution. Recent years have witnessed a boom in technological development dedicated to isotopologue measurement, leading to an unprecedented degree of information regarding organic (bio)synthesis. While applications in Earth and planetary sciences has been limited so far to simple hydrocarbons, typically methane, isotopologue proxies are expected to rapidly emerge in biogeochemistry, providing new types of environmental and biological tracers. This review describes principles and measurement techniques, as well as present and potential biogeochemical applications. ▪ Stable isotopes of organic molecules are widely used in biogeo-chemistry. ▪ Isotope analysis at the intramolecular level is expected to provide new information on the origin of molecules. ▪ Recent technological developments unlocked the potential of intramolecular isotope analysis, providing new proxies in biogeochemistry and new opportunities to clarify questions related to Earth and planetary sciences.

Remote sensing cloud computing platform development and Earth science application

Remote sensing cloud computing platform development and Earth science application

A Multi-Cooperative Deep Convolutional Neural Network for Spatiotemporal Satellite Image Fusion

Remote sensing satellite images with high temporal and high spatial resolution play a critical role in earth science applications. However, it is difficult for a single satellite to obtain such images due to technical and cost constraints. Therefore, spatiotemporal image fusion based on deep learning has received extensive attention in recent years. This article proposes a multicooperative deep convolutional neural network (MCDNet) for spatiotemporal satellite image fusion. This method is a new multinetwork model in which multiple networks work together to reconstruct the predicted image. The multinetwork model consists of a super-resolution network, a difference reconstruction network, and a collaborative training network. First, the super-resolution network uses the combination of a novel multiscale mechanism and dilated convolutions to make full use of the spectral information of the coarse image and upgrade it to a transitional image that matches the fine image. The difference reconstruction network uses structural relevance to complete the reconstruction of the fine difference image. The collaborative training network extracts the hidden information from the fine image and uses the time relevance to restrict the training of the difference reconstruction network. Finally, the fine difference image and the known fine image are combined to complete the image fusion. The new compound loss function can help multinetwork models better complete cooperative training. Through experiments on two datasets and comparison with existing fusion algorithms, the subjective and objective results prove that MCDNet can effectively reconstruct higher-quality prediction images.

Very High Spatial Resolution Downscaled SMAP Radiometer Soil Moisture in the CONUS Using VIIRS/MODIS Data

Satellite remote sensing has been providing passive microwave soil moisture (SM) retrievals of a global spatial coverage and a high revisit frequency for research and applications in earth and environmental sciences, specifically after the soil moisture active and passive (SMAP) was launched in 2015. But, the spatial resolution of SM data is restricted to tens of kilometers, which is insufficient for regional or watershed scale studies. In this article, an SM downscaling algorithm was developed based on the vegetation modulated apparent thermal inertia relationship between SM and changes in land surface temperature (LST). The algorithm used data sets from the North America Land Data Assimilation System Noah model outputs and the advanced very high resolution radiometer data of the long term data record from 1981 to 2018. Here, the downscaling model was applied to visible/infred LST data from the visible infrared imaging radiometer suite at 400-m and the moderate resolution imaging spectroradiometer at 1 km to downscale the L2 radiometer half-orbit 9 km SMAP SM from 2018 to 2019 for the contiguous United States. The 400-m/1-km downscaled SM products were validated using 125 in situ SM ground measurements acquired from the International Soil Moisture Network. The validation results summarized by SM network show that the overall unbiased RMSE for 400-m of the improved/original downscaling algorithms and 1-km SM outperform 9-km SM by 0.01, 0.007, and 0.012 m3/m3 volumetric soil moisture, respectively, which indicates a fairly good performance of the downscaling algorithm. It is also found that precipitation has an impact on the 9-km SMAP SM.

Which is the best 9Be carrier for 10Be/9Be accelerator mass spectrometry?

Commercial 9Be solutions used for chemical preparation of samples for accelerator mass spectrometry contain the cosmogenic long-lived radionuclide 10Be at elevated but different 10Be/9Be levels. Within a systematic study of recently produced solutions, comparison to published data and new data on customised solutions from minerals, we recommend - if no customised solution is available - the 9Be solutions from Australian Chemical Reagents (ACR) or from LGC. They contain 10Be/9Be at the 3.4 × 10−15 level, which is still suitable for the majority of Earth science applications, compared to customised solutions at the 10−16 level for lowest-level studies. Commercial solutions from Scharlab having different lot numbers, i.e. an identification number assigned to a particular lot of material from a single manufacturer, vary in 10Be/9Be by up to a factor of nine. Hence, it seems an advisable strategy to buy a bigger quantity of a single production batch (such as 10 × 100 ml bottles of 9Be at 1 g l−1) and have them tested once at any AMS facility before first use.•The best 9Be carrier for low-level 10Be/9Be applications is a customised one from minerals like phenakite.•The best 9Be carriers for medium- and high-level 10Be/9Be applications are currently from Australian Chemical Reagents (ACR) or from LGC.•As 9Be carriers from Scharlab of different batches (LOT) contain 10Be/9Be at different levels, it is advisable to buy a bigger number of bottles of the same LOT of commercial carriers after being identified to have reasonably low isotope ratios.

Pragmatic solvers for 3D Stokes and elasticity problems with heterogeneous coefficients: evaluating modern incomplete LDL<sup><i>T</i></sup> preconditioners

Abstract. The need to solve large saddle point systems within computational Earth sciences is ubiquitous. Physical processes giving rise to these systems include porous flow (the Darcy equations), poroelasticity, elastostatics, and highly viscous flows (the Stokes equations). The numerical solution of saddle point systems is non-trivial since the operators are indefinite. Primary tools to solve such systems are direct solution methods (exact triangular factorization) or approximate block factorization (ABF) preconditioners. While ABF solvers have emerged as the state-of-the-art scalable option, they are invasive solvers requiring splitting of pressure and velocity degrees of freedom, a multigrid hierarchy with tuned transfer operators and smoothers, machinery to construct complex Schur complement preconditioners, and the expertise to select appropriate parameters for a given coefficient regime – they are far from being “black box” solvers. Modern direct solvers, which robustly produce solutions to almost any system, do so at the cost of rapidly growing time and memory requirements for large problems, especially in 3D. Incomplete LDLT (ILDL) factorizations, with symmetric maximum weighted-matching preprocessing, used as preconditioners for Krylov (iterative) methods, have emerged as an efficient means to solve indefinite systems. These methods have been developed within the numerical linear algebra community but have yet to become widely used in applications, despite their practical potential; they can be used whenever a direct solver can, only requiring an assembled operator, yet can offer comparable or superior performance, with the added benefit of having a much lower memory footprint. In comparison to ABF solvers, they only require the specification of a drop tolerance and thus provide an easy-to-use addition to the solver toolkit for practitioners. Here, we present solver experiments employing incomplete LDLT factorization with symmetric maximum weighted-matching preprocessing to precondition operators and compare these to direct solvers and ABF-preconditioned iterative solves. To ensure the comparison study is meaningful for Earth scientists, we utilize matrices arising from two prototypical problems, namely Stokes flow and quasi-static (linear) elasticity, discretized using standard mixed finite-element spaces. Our test suite targets problems with large coefficient discontinuities across non-grid-aligned interfaces, which represent a common challenging-for-solvers scenario in Earth science applications. Our results show that (i) as the coefficient structure is made increasingly challenging, by introducing high contrast and complex topology with a multiple-inclusion benchmark, the ABF solver can break down, becoming less efficient than the ILDL solver before breaking down entirely; (ii) ILDL is robust, with a time to solution that is largely independent of the coefficient topology and mildly dependent on the coefficient contrast; (iii) the time to solution obtained using ILDL is typically faster than that obtained from a direct solve, beyond 105 unknowns; and (iv) ILDL always uses less memory than a direct solve.

Advances in Random Fiber Lasers and Their Sensing Application.

Compared with conventional laser, random laser (RL) has no resonant cavity, reducing the requirement of cavity design. In recent years, the random fiber laser (RFL), a novel kind of RL, has made great progress in theories and experiments. The RFL has a simpler structure, a more flexible design, and higher reliability. It has valuable applications for earth sciences, biological life sciences, and national defense security, due to these unique properties. This paper reviews the development of RFLs in the last decade, including their configurations based on various optical fibers and their output properties, especially the method of control. Moreover, we also introduce their applications in the optical fiber sensing system, which is a very important and practical orientation to study. Finally, this paper presents the prospects of RFLs.

Noise Analysis of a Passive Resonant Laser Gyroscope.

Large-scale laser gyroscopes have found important applications in Earth sciences due to their self-sufficient property of measurement of the Earth’s rotation without any external references. In order to extend the relative rotation measurement accuracy to a better level so that it can be used for the determination of the Earth orientation parameters (EOP), we investigate the limitations in a passive resonant laser gyroscope (PRG) developed at Huazhong University of Science and Technology (HUST) to pave the way for future development. We identify the noise sources from the derived noise transfer function of the PRG. In the frequency range below , the contribution of free-spectral-range (FSR) variation is the dominant limitation, which comes from the drift of the ring cavity length. In the to frequency range, the limitation is due to the noises of the frequency discrimination system, which mainly comes from the residual amplitude modulation (RAM) in the frequency range below 2 Hz. In addition, the noise contributed by the Mach–Zehnder-type beam combiner is also noticeable in the 0.01 to 2 Hz frequency range. Finally, possible schemes for future improvement are also discussed.

High-resolution mapping of floodplain topography from space: A case study in the Amazon

Terrain elevation is essential for land management, navigation, and earth science applications. Remote sensing advancements have led to an increase in the availability of a range of digital elevation models with global to quasi-global land coverage. However, the generation of these models in water bodies requires specialized approaches, such as the delimitation of the shorelines (isobaths) of lakes over time. Therefore, the processing costs are high in complex areas with many lakes. Currently, there is no systematic topographic mapping of lakes and channels in large and complex floodplains using remote sensing data. We present here the first high-resolution topographic mapping (30 m) of the non-forested portion of the middle-lower Amazon floodplain using a new method based on in-situ Amazon river water levels and a flood-frequency map derived from the Landsat Global Surface Water Dataset. Validation using locally derived bathymetry showed a root mean square error (RMSE) of 0.89 m for floodplain elevation and a good representation of spatial patterns with Pearson's correlation coefficient of 0.77. Our approach for improving topographic representation in open water areas is an alternative to SRTM3 DEM or MERIT DEM, which represents these areas as a flat surface. We also generated the Amazon River bathymetry using nautical charts from the Brazilian Navy (average RMSE of 7.5 m and bias of 5 m), and floodplain depths maps corresponding to the high- and low-water periods of the river flood wave. The results show that the storage volume in the open-water floodplain varies 104.3 km3 on average each year (from 11.9 km3 in low-water to 116.2 km3 in high-water). The method can be applied to any temporarily flooded area to provide the often missing underwater digital topographic data required for hydrological, ecological, and geomorphological studies. The data set developed in this study can be found at https://doi.org/10.17632/vn599y9szb.1.

A review and evaluation of 39 thermal conductivity models for frozen soils

Frozen soil thermal conductivity (λeff) is a critical thermo-physical property that is required for environmental, earth science, geotechnical and geo-environmental applications and associated numerical modeling. Measurement of λeff in frozen soils is difficult and prone to errors, especially near the freezing/thawing point of soil water (e.g., −4 to 0 °C). Available steady-state or transient methods to measure λeff based on the soil temperature response to applied heat often result in melting of soil ice, violating the conduction-only assumptions of these methods and result in biased λeff measurements. Therefore, the choice of λeff models is often influence by their ease of implementation. A great number of such models have been developed during the last few decades since the latest comprehensive review of frozen soil thermal conductivity models in the early 1980s. There is a need to revisit this topic by comparing the models and evaluating their performance to provide information to the novice and expert alike, in order to guide them on their advantages, limitations and applications. A total of 39 models were categorized as: 1) linear and non-linear regression models (8 models); 2) physical models (6 models); 3) mixing models (6 models); 4) normalized models (17 models); and 5) models of other types based on their characteristics (2 models). These models were assessed with a large compiled dataset consisting of 331 λeff measurements taken at temperatures <−4 °C on 27 soils from seven studies, using steady-state or transient methods. Three performance indices including root mean square error (RMSE), average deviations (AD) and Nash-Sutcliff Efficiency (NSE) were used to evaluate performance of these models. The results showed that the models of Becker et al. (1992) (BB1992), Tian et al. (2016) (TZ2016), Zhang et al. (2018) (ZM2018) and Wang et al. (2017) (WL2017) were the best performing models in their affiliated category. However, none of the models performed satisfactorily, with NSE = 0.51, RMSE = 0.46 W m−1 °C−1 and AD = −0.04 W m−1 °C−1 for the best performing model among all the models investigated. Future studies that focus on conceptualization and development of new frozen soil thermal conductivity models for accurate and wide application is recommended.

Pressure Generation above 35GPa in a Walker-Type Large-Volume Press**Supported by the National Natural Science Foundation of China (Grant Nos. 51320105007, 11634004 and 41902034), the Fundamental Research Funds for the Central Universities of China (Grant No. 45119031C037), and the Program for Changjiang Scholars and Innovative Research Team in University (Grant No. IRT1132).

Pressure generation to a higher pressure range in a large-volume press (LVP) denotes our ability to explore more functional materials and deeper Earth’s interior. Pressure generated by normal tungsten carbide (WC) anvils in a commercial way is mostly limited to 25GPa in LVPs due to the limitation of their hardness and design of cell assemblies. We adopt three newly developed WC anvils for ultrahigh pressure generation in a Walker-type LVP with a maximum press load of 1000 ton. The hardest ZK01F WC anvils exhibit the highest efficiency of pressure generation than ZK10F and ZK20F WC anvils, which is related to their performances of plastic deformations. Pressure up to 35GPa at room temperature is achieved at a relatively low press load of 4.5MN by adopting the hardest ZK01F WC anvils with three tapering surfaces in conjunction with an optimized cell assembly, while pressure above 35GPa at 1700K is achieved at a higher press load of 7.5MN. Temperature above 2000K can be generated by our cell assemblies at pressure below 30GPa. We adopt such high-pressure and high-temperature techniques to fabricate several high-quality and well-sintered polycrystalline minerals for practical use. The present development of high-pressure techniques expands the pressure and temperature ranges in Walker-type LVPs and has wide applications in physics, materials, chemistry, and Earth science.

Set Pair Analysis in Earth and Environmental Sciences: Development, Challenges, and Future Prospects

Set pair analysis (SPA) is a method to describe and deal with uncertainty in systems by using the connection number. It has been widely used in earth and environmental sciences and has become increasingly important in decision-making for environmental protection, risk evaluation, and disaster prediction. This paper reviewed the development of SPA with focus on its application in earth and environmental sciences. In addition, the challenges faced by SPA development in earth and environmental sciences were discussed. The development of SPA enriches and improves the ways of treating uncertainties in decision-making processes. However, there are still several theoretical problems to be solved. This review deepens our understanding into the application of SPA in earth and environmental sciences, promoting its wider development and application in these research fields.

Optimal X-ray micro-CT image based methods for porosity and permeability quantification in heterogeneous sandstones

Summary 3-D X-ray micro-CT (XCT) is a non-destructive 3-D imaging method, increasingly used for a wide range of applications in Earth Science. An optimal XCT image-processing workflow is derived here for accurate quantification of porosity and absolute permeability of heterogeneous sandstone samples using an assessment of key image acquisition and processing parameters: image resolution, segmentation method, representative elementary volume (REV) size and fluid-simulation method. XCT image-based calculations obtained for heterogeneous sandstones are compared to two homogeneous standards (Berea sandstone and a sphere pack), as well as to the results from physical laboratory measurements. An optimal XCT methodology obtains porosity and permeability results within ±2 per cent and vary by one order of magnitude around the direct physical measurements, respectively, achieved by incorporating the clay fraction and cement matrix (porous, impermeable components) to the pore-phase for porosity calculations and into the solid-phase for permeability calculations. Two stokes-flow finite element modelling (FEM) simulation methods, using a voxelized grid (Avizo) and tetrahedral mesh (Comsol) produce comparable results, and similarly show that a lower resolution scan (∼5 µm) is unable to resolve the smallest intergranular pores, causing an underestimation of porosity by ∼3.5 per cent. Downsampling the image-resolution post-segmentation (numerical coarsening) and pore network modelling both allow achieving of a REV size, whilst significantly reducing fluid simulation memory requirements. For the heterogeneous sandstones, REV size for permeability (≥1 mm3) is larger than for porosity (≥0.5 mm3) due to tortuosity of the fluid paths. This highlights that porosity should not be used as a reference REV for permeability calculations. The findings suggest that distinct image processing workflows for porosity and permeability would significantly enhance the accurate quantification of the two properties from XCT.

Assessing the efficiency of a complex of geophysical methods of rock mass research under the conditions of an underground mine

Introduction. At present, geophysical methods of rock mass research have found wide application in mining and Earth sciences. They are used to perform prospecting and evaluation work in the massif, namely, to identify structural irregularities, voids and various contact zones of the media with sufficient accuracy. The accuracy of the results obtained depends on the method used in particular conditions, the nature and parameters of the target, the external conditions of the natural experiment, as well as the tasks set in the work. Research methodology. The work included a set of geophysical studies, namely, methods of spectral seismic profiling and GPR in an underground mine. Measurements were made to find different objects at different angles to the object in question in order to determine the distance to it. Research results. As a result of investigating with a complex of geophysical methods, a metal shelf was found under the bulk rock mass and two shelf tiers were subsequently separated when carrying out vertically directed measurements. In the process of searching for the underlying excavation in the massif during vertically directed geophysical measurements, which were carried out in the horizontal excavation 20 meters from the shaft, changes in spectrum density at a depth of about 55 m were detected, which coincides with the actual location of the sought excavation. The results of cavity search in rock mass when making measurements at an angle to the object sought showed that there are no sufficiently clear boundaries and it is impossible to estimate the distance to the cavity in the filled stopes.

Structural stabilities of iron silicides at high pressures and temperatures

Iron silicides provide extensive applications in functional device and earth science. Very little is known about these alloys at high pressures and temperatures. Here, combining swarm-intelligence-based structure prediction methodology and quasi-harmonic approximation (QHA), we investigate Fe-Si compounds at pressures of up to 500 GPa. Evolutionary simulations correctly find the known B20-FeSi structure and some other experimentally observed compounds at atmospheric pressure and finite temperature. It is shown that among the possible iron silicides, only B2-FeSi is thermodynamically stable at high pressures, and the increasing temperature tends to stabilize B2-FeSi compound rather than other iron silicides.