JZJohnny ZhangZhou

Johnny ZhangZhou

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Johnny ZhangZhou has a total of 111 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    A Machine Learning‐Based Geothermal Gradient Framework for Constraining Lithospheric Biomass

    by Wenyu Zhao, Harrison B. Smith, J. ZhangZhou

    Earth's deep subsurface hosts a large microbial biosphere, but the magnitude of lithospheric biomass remains poorly constrained due to sparse observations and uncertain thermal limits. Here, we present a physically constrained global assessment of the continental and oceanic lithospheric biomass shallower than the 122°C isotherm. Using 8,452 geothermal gradient observations aggregated to a 1° × 1° grid, we trained separate XGBoost models for continental and oceanic domains to reconstruct a continuous global geothermal gradient field. Temperature‐bound habitable depths derived from this framework yield comparable habitable volumes of 6.8 × 10 8 km 3 for the continental lithosphere and 7.3 × 10 8 km 3 for the oceanic lithosphere. Recalculated continental biomass agrees with previous estimates, yielding 2–6 × 10 29 cells (4.2–12.6 Gt C). In contrast, oceanic biomass is highly sensitive to the treatment of shallow seawater‐influenced samples across alternative extrapolation schemes. Excluding these samples yields estimates of 0.5–1.6 × 10 28 cells (0.1–0.3 Gt C), substantially lower than previous estimates, whereas including them inflates biomass estimates by up to 4 orders of magnitude. Combined continental and oceanic biomass estimates total 4.3–12.9 Gt C. These results indicate that lithospheric biomass is more limited than previously inferred and demonstrate that physically constrained integration provides a robust framework for quantifying the deep biosphere.

  • Open Access

    Coupling dynamical accretion and chemical differentiation: A unified framework for the diversity of Earth and Mars

    by Zhihui Kong, Beibei Liu, J. ZhangZhou, Haolan Tang, Zhengbin Deng, Qun-Ke Xia, Simon L. Grimm, Man Hoi Lee, Yi Huang, Liping Qin, et al.

    The physical and geochemical differences between Earth and Mars provide fundamental constraints on terrestrial planet formation, yet a self-consistent framework linking dynamical and chemical aspects remains elusive. Here we present an integrated modeling framework that couples high-resolution N-body simulations with impact-driven metal-silicate equilibration to track the dynamical accretion history and chemical differentiation for Earth and Mars. Using a narrow ring planetesimal accretion scenario, we show that Earth and Mars analogs naturally sample systematically different solid reservoirs within the protoplanetary disk. Earth analogs preferentially accrete reduced material around the planetesimal ring center, whereas Mars analogs acquire a larger fraction of oxidized material exterior to the ring. This leads to diverse bulk redox states, with composition further modified by impact-dependent pressure-temperature equilibration conditions during core formation. As a result, Earth analogs experience deeper equilibration and more efficient transfer of iron into the core, producing mantles with low iron oxide contents and larger core mass fractions. In contrast, Mars analogs equilibrate at shallower conditions, retain more iron in their mantles, and develop smaller cores. Our results demonstrate that the dynamical and geochemical differences between Earth and Mars emerge from the coupled effects of accretion pathways, the disk's radial redox structure, and impact-controlled differentiation rather than from any single process. Our unified framework physically explains the geochemical diversity of terrestrial planets and offers a potential pathway to interpret compositions of rocky planets in exoplanetary systems.


  • Open Access

    Calibrating Mantle Redox Conditions Using Ferric Iron in Clinopyroxene Xenoliths: A Machine Learning Approach

    by Chenyang Ye, Chuntao Liu, Zhou Zhang

    The oxygen fugacity (fO2) of the mantle governs the behaviors of multivalent elements (e.g., Fe, V) and the speciation of C–O–H fluids, influencing mantle melting, magmatic evolution and volatile distribution across tectonic settings. However, the estimation of mantle fO2 is limited by challenges in measuring Fe3+ in minerals like clinopyroxene (Cpx) due to analytical constraints and inconsistencies between oxybarometer methods. Here, we applied machine learning (ML) to predict Cpx Fe3+ content and equilibrium pressure-temperature and fO2 conditions. We employed a nested cross-validation approach to minimize coincidental perfomance biases. Our models outperformed previous ferric iron and thermobarometer models on both metrics and generalization. The ML-based oxybarometer shows adequate generalization with R2 peaking at 0.74, average MAE is 0.95, and average RMSE is 1.45. We compiled an application dataset comprising of 9,832 global mantle xenolith samples. For the subset with Fe3+ measurements (n≈600), ML-predicted fO2 closely matches thermodynamic estimates, supporting the robustness and global applicability of our approach. Applying the model to the rest samples lacking Fe3+ analyses expands geographic coverage to data-sparse provinces (e.g., South America, India, and Eastern Europe), and reveals coherent global redox gradients. Xenoliths from cratonic mantle domains show no temporal fO2 trends since Mesoproterozoic. Comparative analysis across cratonic mantle xenoliths, abyssal peridotites, and oceanic intraplate xenoliths indicates that mantle residues are initially oxidized by short-term metasomatism, but eventually equilibrate to stable redox conditions through interactions with neutral or reducing agents.

  • Open Access

    The oxidation of mantle wedge in Earth’s middle age

    by Chun-Tao Liu, Chen-Yang Ye, Qun-Ke Xia, Zhou Zhang

    The redox state of Earth’s mantle exerts a fundamental influence on volcanic degassing and the composition of the atmosphere, yet its long-term evolution remains uncertain. Here we use a machine-learning classifier to identify primitive arc basalts and reconstruct mantle wedge’s oxygen fugacity over time. Our results show that the redox state of mantle wedge raised in Earth’s middle-age. This Mesoproterozoic oxidation was asynchronous with surface oxygenation, suggesting that deep Earth processes, such as the enhanced fluxes of serpentinite-derived fluids, drove the oxidation of mantle wedge. The establishment of an oxidized mantle wedge may have reduced volcanic oxygen sinks and facilitated atmospheric oxygen accumulation in Mesoproterozoic, with implications for the rise of early eukaryotic life.

  • Open Access

    Geochemistry π: Machine Learning for Geochemists Who Don’t Want to Code

    by J.Zhou ZhangZhou

    Geochemistry π is an open-source automated machine learning Python framework. Geochemists need only provide tabulated data (e.g. excel spreadsheet) and select the desired options to clean data and run machine learning algorithms. The process operates in a question-and-answering format, and thus does not require that users have coding experience. Version 0.7.0 includes machine learning algorithms for regression, classification, clustering, dimension reduction and anomaly detection. After either automatic or manual parameter tuning, the automated Python framework provides users with performance and prediction results for the trained machine learning model. Based on the scikit-learn library, Geochemistry π has established a customized automated process for implementing machine learning. The Python framework enables extensibility and portability by constructing a hierarchical pipeline architecture that separates data transmission from algorithm application. The AutoML module is constructed using the Cost-Frugal Optimization and Blended Search Strategy hyperparameter search methods from the A Fast and Lightweight AutoML Library, and the model parameter optimization process is accelerated by the Ray distributed computing framework. The MLflow library is integrated into machine learning lifecycle management, which allows users to compare multiple trained models at different scales and manage the data and diagrams generated. In addition, the front-end and back-end frameworks are separated to build the web portal, which demonstrates the machine learning model and data science workflow through a user-friendly web interface. In summary, Geochemistry π provides a Python framework for users and developers to accelerate their data mining efficiency with both online and offline operation options. All source code is available on GitHub (https://github.com/ZJUEarthData/geochemistrypi), with a detailed operational manual catering to both users and developers (https://geochemistrypi.readthedocs.io/en/latest/).

  • Open Access

    The sub-arc mantle has remained oxidized since the Neoproterozoic oxygenation event

    by Chun-Tao Liu, Chen-Yang Ye, Qun-Ke Xia, J. ZhangZhou

    The redox state of the sub-arc mantle impacts magma degassing at arc volcanoes and arc-related ore deposits. However, the evolution of the oxygen fugacity (fO2) of the sub-arc mantle during the Neoproterozoic Oxygenation Event (NOE) remains poorly understood. Here, we applied machine learning to identify arc basalts, then calculated V/Sc ratios of primitive arc basalts since ~0.8 billion years ago (Ga). Our results show that average V/Sc remained nearly constant at ~7.16, suggesting an oxidized sub-arc mantle since ~0.8 Ga. This implies that increased surficial oxygen levels during the NOE may not directly oxidize the sub-arc mantle. Instead, its oxidation likely stems from the influx of oxidized fluids and melts via subducted serpentinized oceanic crust. The persistently oxidized sub-arc mantle may have controlled the long-term oxidation of arc magmas and thus influenced the oxidation of Earth’s surficial environments since ~0.8 Ga, though it was not the NOE’s primary driver.


  • Open Access

    Orange-Volcanoes: A new open and collaborative platform to perform data-driven investigations and machine learning analyses in petrology and volcanology

    by Alessandro Musu, Valerio Parodi, Marko Toplak, Alessandro Carfì, Mónica Ágreda-López, Fulvio Mastrogiovanni, J. ZhangZhou, Diego Perugini, Donato Belmonte, Penny E. Wieser, et al.


  • Mapping Crustal Vp/Vs in North America With a Machine Learning Approach

    by Xingxing Gao, Yunfeng Chen, Wenyu Zhao, J. ZhangZhou

    Vp/Vs (Poisson's ratio) provides critical information for constraining the bulk crustal composition, stress state, and tectonic evolution of the Earth. The receiver function technique has been extensively utilized to constrain the crustal Vp/Vs, yet the reliability of measurements can be affected by complex structures and uneven distribution of seismic stations. Consequently, the interpolated Vp/Vs maps can often be biased by unreliable observations, especially in data‐sparse regions. We tackle these issues by proposing a machine learning model that integrates multiple geophysical data sets to estimate Vp/Vs, leveraging the physical and structural properties of the crust. We train the model by compiling an extensive data set of global Vp/Vs measurements at 13,314 seismic stations and employ XGBoost to map Vp/Vs with other key crustal properties. Experiments using data from the (a) United States and (b) United States and Canada demonstrate superior prediction accuracy, achieving an overall value of 0.84 in both cases. Feature importance analysis indicates that crustal tectonic type, geographic coordinates, mid‐crust shear‐wave velocity, and crustal thickness primarily capture Vp/Vs variations, together explaining over 70% of reduction in the normalized root‐mean‐square error. The inclusion of other features further refines small‐scale Vp/Vs variation. Compared to cubic and Kriging interpolations, the predicted Vp/Vs map from machine learning exhibits less local extremes and a better alignment with the first‐order crustal structure across the continent. This study highlights the capability of machine learning to uncover complex geophysical relationships for reliable Vp/Vs estimates and its potential to constrain crustal composition at a continental scale.


  • Open Access

    Modelling redox state via V-Sc-Ti-Yb partitioning in mantle derived melts

    by C.-T. Liu, C.-Y. Ye, J. ZhangZhou

    The ratios of multivalent V to homovalent elements (e.g., V/Sc, V/Ti, V/Yb) are pivotal proxies for assessing mantle oxygen fugacity (fO2), yet their geochemical behaviours during partial melting remain debatable. Here, we conducted petrological modelling to explore the responses of these ratios to fO2 variations in sub-arc and mid-ocean ridge basalt (MORB) mantle sources, considering diverse mineral assemblages and chemical compositions. Our results suggest that V/Sc is relatively unaffected by changes in mantle source characteristics and melting extent, positioning it as a more dependable indicator of mantle fO2 and as a tool for tracing the evolution of mantle fO2 in terrestrial planets. In contrast, V/Ti and V/Yb are both more sensitive to the degree of melting and V/Ti is notably affected by mantle chemical composition. These insights refine our understanding of redox conditions across various tectonic environments on Earth. For instance, applying our model to data sets of primitive basalts reveals that the contemporary sub-arc mantle is generally more oxidised than the MORB mantle, consistent with impacts of slab-derived fluids and sediments in arc regions.


  • Open Access

    Estimating Biomass in the Lithosphere: A Machine Learning Approach

    by Wenyu Zhao, Johnny Zhangzhou

    Research on sampling and analysis of fluids and rocks within both continental and oceanic crusts has revealed a significant portion of Earth's prokaryotic biomass residing in the rock-hosted biosphere, extending to depths of several kilometers. This deep subsurface life within the lithosphere may host a substantial amount of Earth's biomass. However, more constrained estimates of this biomass are still lacking. In this study, we first determine the habitable volume of the lithospheric biosphere and then estimate its potential biomass. Temperature, a critical factor influencing the limits of life, is considered the primary focus in this context. We employ a machine learning approach to estimate the habitable volume of the lithosphere above the 122°C isotherm, which is considered the upper limit for prokaryotic life. Our methodology involves selecting a range of geophysical and geological features that are thought to impact geothermal gradients. These include Moho depth, lithosphere-asthenosphere boundary (LAB) depth, topography, susceptibility, tectonic units, gravity mean curvature, vertical magnetic field, and distances to ridges, trenches, transform faults, young rifts, and volcanoes, as well as P-wave and S-wave velocities in the crust. We used a Gradient Boosted Regression Tree algorithm to develop two models correlating geothermal gradients in continental and oceanic settings with these geophysical and geological features. The models were applied to a binned grid with a 0.5° x 0.5° resolution, enabling the estimation of the depth to the 122°C isotherm for each grid based on the machine learning model's predictions of geothermal gradients. Our results reveal substantial habitable volumes within the lithosphere. The habitable volume in the continental crust may account for about 7% of the total crustal volume, while in the oceanic crust, it could be around 5%. Given the cell density range of 102 to 106 cells per gram in oceanic rocks and 102 to 104 cells per gram in continental rocks, along with an estimated carbon mass of 2x10-14 grams per cell, the lithospheric biomass is estimated to be between 0.008 and 33.2 Gt Carbon. For context, the biomass found in plants and marine microbes is estimated to be around 450 Gt Carbon and 2.9 Gt Carbon, respectively.