Articles | Volume 18, issue 9
https://doi.org/10.5194/gmd-18-2521-2025
© Author(s) 2025. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
https://doi.org/10.5194/gmd-18-2521-2025
© Author(s) 2025. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Development and technical paper
| 
08 May 2025
Development and technical paper |
| 08 May 2025
| 08 May 2025
NN-TOC v1: global prediction of total organic carbon in marine sediments using deep neural networks
Naveenkumar Parameswaran
CORRESPONDING AUTHOR
GEOMAR Helmholtz Centre for Ocean Research Kiel, Kiel, Germany
Department of Mathematics, Kiel University, Kiel, Germany
Everardo González
GEOMAR Helmholtz Centre for Ocean Research Kiel, Kiel, Germany
Ewa Burwicz-Galerne
MARUM – Center for Marine Environmental Sciences, University of Bremen, Bremen, Germany
Malte Braack
Department of Mathematics, Kiel University, Kiel, Germany
Klaus Wallmann
GEOMAR Helmholtz Centre for Ocean Research Kiel, Kiel, Germany
Related authors
No articles found.
Alexandra N. Loginova, Andrew W. Dale, Frédéric A. C. Le Moigne, Sören Thomsen, Stefan Sommer, David Clemens, Klaus Wallmann, and Anja Engel
Biogeosciences, 17, 4663–4679, https://doi.org/10.5194/bg-17-4663-2020, https://doi.org/10.5194/bg-17-4663-2020, 2020
Short summary
Short summary
We measured dissolved organic carbon (DOC), nitrogen (DON) and matter (DOM) optical properties in pore waters and near-bottom waters of the eastern tropical South Pacific off Peru. The difference between diffusion-driven and net fluxes of DOC and DON and qualitative changes in DOM optical properties suggested active microbial utilisation of the released DOM at the sediment–water interface. Our results suggest that the sediment release of DOM contributes to microbial processes in the area.
Sebastian Beil, Wolfgang Kuhnt, Ann Holbourn, Florian Scholz, Julian Oxmann, Klaus Wallmann, Janne Lorenzen, Mohamed Aquit, and El Hassane Chellai
Clim. Past, 16, 757–782, https://doi.org/10.5194/cp-16-757-2020, https://doi.org/10.5194/cp-16-757-2020, 2020
Short summary
Short summary
Comparison of Cretaceous OAE1a and OAE2 in two drill cores with unusually high sedimentation rates shows that long-lasting negative δ13C excursions precede the positive δ13C excursions and that the evolution of the marine δ13C positive excursions is similar during both OAEs, although the durations of individual phases differ substantially. Phosphorus speciation data across OAE2 and the Mid-Cenomanian Event suggest a positive feedback loop, enhancing marine productivity during OAEs.
Sonja Geilert, Patricia Grasse, Kristin Doering, Klaus Wallmann, Claudia Ehlert, Florian Scholz, Martin Frank, Mark Schmidt, and Christian Hensen
Biogeosciences, 17, 1745–1763, https://doi.org/10.5194/bg-17-1745-2020, https://doi.org/10.5194/bg-17-1745-2020, 2020
Short summary
Short summary
Marine silicate weathering is a key process of the marine silica cycle; however, its controlling processes are not well understood. In the Guaymas Basin, silicate weathering has been studied under markedly differing ambient conditions. Environmental settings like redox conditions or terrigenous input of reactive silicates appear to be major factors controlling marine silicate weathering. These factors need to be taken into account in future oceanic mass balances of Si and in modeling studies.
Tronje P. Kemena, Angela Landolfi, Andreas Oschlies, Klaus Wallmann, and Andrew W. Dale
Earth Syst. Dynam., 10, 539–553, https://doi.org/10.5194/esd-10-539-2019, https://doi.org/10.5194/esd-10-539-2019, 2019
Short summary
Short summary
Oceanic deoxygenation is driven by climate change in several areas of the global ocean. Measurements indicate that ocean volumes with very low oxygen levels expand, with consequences for marine organisms and fishery. We found climate-change-driven phosphorus (P) input in the ocean is hereby an important driver for deoxygenation on longer timescales with effects in the next millennia.
Konstantin Stolpovsky, Andrew W. Dale, and Klaus Wallmann
Biogeosciences, 15, 3391–3407, https://doi.org/10.5194/bg-15-3391-2018, https://doi.org/10.5194/bg-15-3391-2018, 2018
Short summary
Short summary
The paper describes a new way to parameterize G-type models in marine sediments using data about reactivity of organic carbon sinking to the seafloor.
Ulrike Lomnitz, Stefan Sommer, Andrew W. Dale, Carolin R. Löscher, Anna Noffke, Klaus Wallmann, and Christian Hensen
Biogeosciences, 13, 1367–1386, https://doi.org/10.5194/bg-13-1367-2016, https://doi.org/10.5194/bg-13-1367-2016, 2016
Short summary
Short summary
The study presents a P budget including the P input from the water column, the P burial in the sediments, as well as the P release from the sediments. We found that the P input could not maintain the P release rates. Consideration of other P sources, e.g., terrigenous P and P released from the dissolution of Fe oxyhydroxides, showed that none of these can account for the missing P. Thus, it is likely that abundant sulfide-oxidizing bacteria release the missing P during our measurement period.
K. Wallmann, B. Schneider, and M. Sarnthein
Clim. Past, 12, 339–375, https://doi.org/10.5194/cp-12-339-2016, https://doi.org/10.5194/cp-12-339-2016, 2016
Short summary
Short summary
An Earth system model was set up and applied to evaluate the effects of sea-level change, ocean dynamics, and nutrient utilization on seawater composition and atmospheric pCO2 over the last glacial cycle. The model results strongly suggest that global sea-level change contributed significantly to the slow glacial decline in atmospheric pCO2 and the gradual pCO2 increase over the Holocene whereas the rapid deglacial pCO2 rise was induced by fast changes in ocean dynamics and nutrient utilization.
A. W. Dale, S. Sommer, U. Lomnitz, I. Montes, T. Treude, V. Liebetrau, J. Gier, C. Hensen, M. Dengler, K. Stolpovsky, L. D. Bryant, and K. Wallmann
Biogeosciences, 12, 1537–1559, https://doi.org/10.5194/bg-12-1537-2015, https://doi.org/10.5194/bg-12-1537-2015, 2015
A. W. Dale, V. J. Bertics, T. Treude, S. Sommer, and K. Wallmann
Biogeosciences, 10, 629–651, https://doi.org/10.5194/bg-10-629-2013, https://doi.org/10.5194/bg-10-629-2013, 2013
Related subject area
Biogeosciences
China Wildfire Emission Dataset (ChinaWED v1) for the period 2012–2022
Process-based modeling of solar-induced chlorophyll fluorescence with VISIT-SIF version 1.0
Including the phosphorus cycle into the LPJ-GUESS dynamic global vegetation model (v4.1, r10994) – global patterns and temporal trends of N and P primary production limitation
A comprehensive land-surface vegetation model for multi-stream data assimilation, D&B v1.0
Sources of uncertainty in the SPITFIRE global fire model: development of LPJmL-SPITFIRE1.9 and directions for future improvements
The unicellular NUM v.0.91: a trait-based plankton model evaluated in two contrasting biogeographic provinces
FESOM2.1-REcoM3-MEDUSA2: an ocean–sea ice–biogeochemistry model coupled to a sediment model
Satellite-based modeling of wetland methane emissions on a global scale (SatWetCH4 1.0)
Emulating grid-based forest carbon dynamics using machine learning: an LPJ-GUESS v4.1.1 application
Systematic underestimation of type-specific ecosystem process variability in the Community Land Model v5 over Europe
Soil nitrous oxide emissions from global land ecosystems and their drivers within the LPJ-GUESS model (v4.1)
Lambda-PFLOTRAN 1.0: a workflow for incorporating organic matter chemistry informed by ultra high resolution mass spectrometry into biogeochemical modeling
An improved model for air–sea exchange of elemental mercury in MITgcm-ECCOv4-Hg: the role of surfactants and waves
BOATSv2: new ecological and economic features improve simulations of high seas catch and effort
A dynamical process-based model for quantifying global agricultural ammonia emissions – AMmonia–CLIMate v1.0 (AMCLIM v1.0) – Part 1: Land module for simulating emissions from synthetic fertilizer use
Simulating Ips typographus L. outbreak dynamics and their influence on carbon balance estimates with ORCHIDEE r8627
Biological nitrogen fixation of natural and agricultural vegetation simulated with LPJmL 5.7.9
Learning from conceptual models – a study of the emergence of cooperation towards resource protection in a social–ecological system
Development and assessment of the physical-biogeochemical ocean regional model in the Northwest Pacific: NPRT v1.0 (ROMS v3.9–TOPAZ v2.0)
The biogeochemical model Biome-BGCMuSo v6.2 provides plausible and accurate simulations of the carbon cycle in central European beech forests
Parameterisation toolbox for physical-biogeochemical model compatible with FABM. Case study: the coupled 1D GOTM-ECOSMO E2E for the Sylt-Romo Bight, North Sea
TROLL 4.0: representing water and carbon fluxes, leaf phenology, and intraspecific trait variation in a mixed-species individual-based forest dynamics model – Part 2: Model evaluation for two Amazonian sites
Estimation of above- and below-ground ecosystem parameters for the DVM-DOS-TEM v0.7.0 model using MADS v1.7.3: a synthetic case study
H2MV (v1.0): Global Physically-Constrained Deep Learning Water Cycle Model with Vegetation
DeepPhenoMem V1.0: deep learning modelling of canopy greenness dynamics accounting for multi-variate meteorological memory effects on vegetation phenology
Impacts of land-use change on biospheric carbon: an oriented benchmark using the ORCHIDEE land surface model
Implementing the iCORAL (version 1.0) coral reef CaCO3 production module in the iLOVECLIM climate model
Assimilation of carbonyl sulfide (COS) fluxes within the adjoint-based data assimilation system – Nanjing University Carbon Assimilation System (NUCAS v1.0)
ML4Fire-XGBv1.0: Improving North American wildfire prediction by integrating a machine-learning fire model in a land surface model
Alquimia v1.0: A generic interface to biogeochemical codes – A tool for interoperable development, prototyping and benchmarking for multiphysics simulators
Quantifying the role of ozone-caused damage to vegetation in the Earth system: a new parameterization scheme for photosynthetic and stomatal responses
Radiocarbon analysis reveals underestimation of soil organic carbon persistence in new-generation soil models
Exploring the potential of history matching for land surface model calibration
EAT v1.0.0: a 1D test bed for physical–biogeochemical data assimilation in natural waters
Using deep learning to integrate paleoclimate and global biogeochemistry over the Phanerozoic Eon
Modelling boreal forest's mineral soil and peat C dynamics with the Yasso07 model coupled with the Ricker moisture modifier
Dynamic ecosystem assembly and escaping the “fire trap” in the tropics: insights from FATES_15.0.0
In silico calculation of soil pH by SCEPTER v1.0
Simple process-led algorithms for simulating habitats (SPLASH v.2.0): robust calculations of water and energy fluxes
A global behavioural model of human fire use and management: WHAM! v1.0
Terrestrial Ecosystem Model in R (TEMIR) version 1.0: simulating ecophysiological responses of vegetation to atmospheric chemical and meteorological changes
biospheremetrics v1.0.2: an R package to calculate two complementary terrestrial biosphere integrity indicators – human colonization of the biosphere (BioCol) and risk of ecosystem destabilization (EcoRisk)
Modeling boreal forest soil dynamics with the microbially explicit soil model MIMICS+ (v1.0)
Optimal enzyme allocation leads to the constrained enzyme hypothesis: the Soil Enzyme Steady Allocation Model (SESAM; v3.1)
Implementing a dynamic representation of fire and harvest including subgrid-scale heterogeneity in the tile-based land surface model CLASSIC v1.45
Inferring the tree regeneration niche from inventory data using a dynamic forest model
Optimising CH4 simulations from the LPJ-GUESS model v4.1 using an adaptive Markov chain Monte Carlo algorithm
The XSO framework (v0.1) and Phydra library (v0.1) for a flexible, reproducible, and integrated plankton community modeling environment in Python
AgriCarbon-EO v1.0.1: large-scale and high-resolution simulation of carbon fluxes by assimilation of Sentinel-2 and Landsat-8 reflectances using a Bayesian approach
SAMM version 1.0: a numerical model for microbial- mediated soil aggregate formation
Zhengyang Lin, Ling Huang, Hanqin Tian, Anping Chen, and Xuhui Wang
Geosci. Model Dev., 18, 2509–2520, https://doi.org/10.5194/gmd-18-2509-2025, https://doi.org/10.5194/gmd-18-2509-2025, 2025
Short summary
Short summary
The China Wildfire Emission Dataset (ChinaWED v1) estimated wildfire emissions in China during 2012–2022 as 78.13 Tg CO2, 279.47 Gg CH4, and 6.26 Gg N2O annually. Agricultural fires dominated emissions, while forest and grassland emissions decreased. Seasonal peaks occurred in late spring, with hotspots in northeast, southwest, and east China. The model emphasizes the importance of using localized emission factors and high-resolution fire estimates for accurate assessments.
Tatsuya Miyauchi, Makoto Saito, Hibiki M. Noda, Akihiko Ito, Tomomichi Kato, and Tsuneo Matsunaga
Geosci. Model Dev., 18, 2329–2347, https://doi.org/10.5194/gmd-18-2329-2025, https://doi.org/10.5194/gmd-18-2329-2025, 2025
Short summary
Short summary
Solar-induced chlorophyll fluorescence (SIF) is an effective indicator for monitoring photosynthetic activity. This paper introduces VISIT-SIF, a biogeochemical model developed based on the Vegetation Integrative Simulator for Trace gases (VISIT) to represent satellite-observed SIF. Our simulations reproduced the global distribution and seasonal variations in observed SIF. VISIT-SIF helps to improve photosynthetic processes through a combination of biogeochemical modeling and observed SIF.
Mateus Dantas de Paula, Matthew Forrest, David Warlind, João Paulo Darela Filho, Katrin Fleischer, Anja Rammig, and Thomas Hickler
Geosci. Model Dev., 18, 2249–2274, https://doi.org/10.5194/gmd-18-2249-2025, https://doi.org/10.5194/gmd-18-2249-2025, 2025
Short summary
Short summary
Our study maps global nitrogen (N) and phosphorus (P) availability and how they changed from 1901 to 2018. We find that tropical regions are mostly P-limited, while temperate and boreal areas face N limitations. Over time, P limitation increased, especially in the tropics, while N limitation decreased. These shifts are key to understanding global plant growth and carbon storage, highlighting the importance of including P dynamics in ecosystem models.
Wolfgang Knorr, Matthew Williams, Tea Thum, Thomas Kaminski, Michael Voßbeck, Marko Scholze, Tristan Quaife, T. Luke Smallman, Susan C. Steele-Dunne, Mariette Vreugdenhil, Tim Green, Sönke Zaehle, Mika Aurela, Alexandre Bouvet, Emanuel Bueechi, Wouter Dorigo, Tarek S. El-Madany, Mirco Migliavacca, Marika Honkanen, Yann H. Kerr, Anna Kontu, Juha Lemmetyinen, Hannakaisa Lindqvist, Arnaud Mialon, Tuuli Miinalainen, Gaétan Pique, Amanda Ojasalo, Shaun Quegan, Peter J. Rayner, Pablo Reyes-Muñoz, Nemesio Rodríguez-Fernández, Mike Schwank, Jochem Verrelst, Songyan Zhu, Dirk Schüttemeyer, and Matthias Drusch
Geosci. Model Dev., 18, 2137–2159, https://doi.org/10.5194/gmd-18-2137-2025, https://doi.org/10.5194/gmd-18-2137-2025, 2025
Short summary
Short summary
When it comes to climate change, the land surface is where the vast majority of impacts happen. The task of monitoring those impacts across the globe is formidable and must necessarily rely on satellites – at a significant cost: the measurements are only indirect and require comprehensive physical understanding. We have created a comprehensive modelling system that we offer to the research community to explore how satellite data can be better exploited to help us capture the changes that happen on our lands.
Luke Oberhagemann, Maik Billing, Werner von Bloh, Markus Drüke, Matthew Forrest, Simon P. K. Bowring, Jessica Hetzer, Jaime Ribalaygua Batalla, and Kirsten Thonicke
Geosci. Model Dev., 18, 2021–2050, https://doi.org/10.5194/gmd-18-2021-2025, https://doi.org/10.5194/gmd-18-2021-2025, 2025
Short summary
Short summary
Under climate change, the conditions necessary for wildfires to form are occurring more frequently in many parts of the world. To help predict how wildfires will change in future, global fire models are being developed. We analyze and further develop one such model, SPITFIRE. Our work identifies and corrects sources of substantial bias in the model that are important to the global fire modelling field. With this analysis and these developments, we help to provide a basis for future improvements.
Trine Frisbæk Hansen, Donald Eugene Canfield, Ken Haste Andersen, and Christian Jannik Bjerrum
Geosci. Model Dev., 18, 1895–1916, https://doi.org/10.5194/gmd-18-1895-2025, https://doi.org/10.5194/gmd-18-1895-2025, 2025
Short summary
Short summary
We describe and test the size-based Nutrient-Unicellular-Multicellular model, which defines unicellular plankton using a single set of parameters, on a eutrophic and oligotrophic ecosystem. The results demonstrate that both sites can be modeled with similar parameters and robust performance over a wide range of parameters. The study shows that the model is useful for non-experts and applicable for modeling ecosystems with limited data. It holds promise for evolutionary and deep-time climate models.
Ying Ye, Guy Munhoven, Peter Köhler, Martin Butzin, Judith Hauck, Özgür Gürses, and Christoph Völker
Geosci. Model Dev., 18, 977–1000, https://doi.org/10.5194/gmd-18-977-2025, https://doi.org/10.5194/gmd-18-977-2025, 2025
Short summary
Short summary
Many biogeochemistry models assume all material reaching the seafloor is remineralized and returned to solution, which is sufficient for studies on short-term climate change. Under long-term climate change, the carbon storage in sediments slows down carbon cycling and influences feedbacks in the atmosphere–ocean–sediment system. This paper describes the coupling of a sediment model to an ocean biogeochemistry model and presents results under the pre-industrial climate and under CO2 perturbation.
Juliette Bernard, Elodie Salmon, Marielle Saunois, Shushi Peng, Penélope Serrano-Ortiz, Antoine Berchet, Palingamoorthy Gnanamoorthy, Joachim Jansen, and Philippe Ciais
Geosci. Model Dev., 18, 863–883, https://doi.org/10.5194/gmd-18-863-2025, https://doi.org/10.5194/gmd-18-863-2025, 2025
Short summary
Short summary
Despite their importance, uncertainties remain in the evaluation of the drivers of temporal variability of methane emissions from wetlands on a global scale. Here, a simplified global model is developed, taking advantage of advances in remote-sensing data and in situ observations. The model reproduces the large spatial and temporal patterns of emissions, albeit with limitations in the tropics due to data scarcity. This model, while simple, can provide valuable insights into sensitivity analyses.
Carolina Natel, David Martin Belda, Peter Anthoni, Neele Haß, Sam Rabin, and Almut Arneth
EGUsphere, https://doi.org/10.5194/egusphere-2024-4064, https://doi.org/10.5194/egusphere-2024-4064, 2025
Short summary
Short summary
Complex models predict forest carbon responses to future climate change but are slow and computationally intensive, limiting large-scale analyses. We used machine learning to accelerate predictions from the LPJ-GUESS vegetation model. Our emulators, based on random forests and neural networks, achieved 97 % faster simulations. This approach enables rapid exploration of climate mitigation strategies and supports informed policy decisions.
Christian Poppe Terán, Bibi S. Naz, Harry Vereecken, Roland Baatz, Rosie A. Fisher, and Harrie-Jan Hendricks Franssen
Geosci. Model Dev., 18, 287–317, https://doi.org/10.5194/gmd-18-287-2025, https://doi.org/10.5194/gmd-18-287-2025, 2025
Short summary
Short summary
Carbon and water exchanges between the atmosphere and the land surface contribute to water resource availability and climate change mitigation. Land surface models, like the Community Land Model version 5 (CLM5), simulate these. This study finds that CLM5 and other data sets underestimate the magnitudes of and variability in carbon and water exchanges for the most abundant plant functional types compared to observations. It provides essential insights for further research into these processes.
Jianyong Ma, Almut Arneth, Benjamin Smith, Peter Anthoni, Xu-Ri, Peter Eliasson, David Wårlind, Martin Wittenbrink, and Stefan Olin
Geosci. Model Dev. Discuss., https://doi.org/10.5194/gmd-2024-223, https://doi.org/10.5194/gmd-2024-223, 2024
Revised manuscript accepted for GMD
Short summary
Short summary
Nitrous oxide (N2O) is a powerful greenhouse gas mainly released from natural and agricultural soils. This study examines how global soil N2O emissions have changed from 1961 to 2020 and identifies key factors driving these changes using an ecological model. The findings highlight croplands as the largest source, with factors like fertilizer use and climate change enhancing emissions. Rising CO2 levels, however, can partially mitigate N2O emissions through increased plant nitrogen uptake.
Katherine A. Muller, Peishi Jiang, Glenn Hammond, Tasneem Ahmadullah, Hyun-Seob Song, Ravi Kukkadapu, Nicholas Ward, Madison Bowe, Rosalie K. Chu, Qian Zhao, Vanessa A. Garayburu-Caruso, Alan Roebuck, and Xingyuan Chen
Geosci. Model Dev., 17, 8955–8968, https://doi.org/10.5194/gmd-17-8955-2024, https://doi.org/10.5194/gmd-17-8955-2024, 2024
Short summary
Short summary
The new Lambda-PFLOTRAN workflow incorporates organic matter chemistry into reaction networks to simulate aerobic respiration and biogeochemistry. Lambda-PFLOTRAN is a Python-based workflow in a Jupyter notebook interface that digests raw organic matter chemistry data via Fourier transform ion cyclotron resonance mass spectrometry, develops a representative reaction network, and completes a biogeochemical simulation with the open-source, parallel-reactive-flow, and transport code PFLOTRAN.
Ling Li, Peipei Wu, Peng Zhang, Shaojian Huang, and Yanxu Zhang
Geosci. Model Dev., 17, 8683–8695, https://doi.org/10.5194/gmd-17-8683-2024, https://doi.org/10.5194/gmd-17-8683-2024, 2024
Short summary
Short summary
In this study, we incorporate sea surfactants and wave-breaking processes into MITgcm-ECCOv4-Hg. The updated model shows increased fluxes in high-wind-speed and high-wave regions and vice versa, enhancing spatial heterogeneity. It shows that elemental mercury (Hg0) transfer velocity is more sensitive to wind speed. These findings may elucidate the discrepancies in previous estimations and offer insights into global Hg cycling.
Jerome Guiet, Daniele Bianchi, Kim J. N. Scherrer, Ryan F. Heneghan, and Eric D. Galbraith
Geosci. Model Dev., 17, 8421–8454, https://doi.org/10.5194/gmd-17-8421-2024, https://doi.org/10.5194/gmd-17-8421-2024, 2024
Short summary
Short summary
The BiOeconomic mArine Trophic Size-spectrum (BOATSv2) model dynamically simulates global commercial fish populations and their coupling with fishing activity, as emerging from environmental and economic drivers. New features, including separate pelagic and demersal populations, iron limitation, and spatial variation of fishing costs and management, improve the accuracy of high seas fisheries. The updated model code is available to simulate both historical and future scenarios.
Jize Jiang, David S. Stevenson, and Mark A. Sutton
Geosci. Model Dev., 17, 8181–8222, https://doi.org/10.5194/gmd-17-8181-2024, https://doi.org/10.5194/gmd-17-8181-2024, 2024
Short summary
Short summary
A special model called AMmonia–CLIMate (AMCLIM) has been developed to understand and calculate NH3 emissions from fertilizer use and also taking into account how the environment influences these NH3 emissions. It is estimated that about 17 % of applied N in fertilizers was lost due to NH3 emissions. Hot and dry conditions and regions with high-pH soils can expect higher NH3 emissions.
Guillaume Marie, Jina Jeong, Hervé Jactel, Gunnar Petter, Maxime Cailleret, Matthew J. McGrath, Vladislav Bastrikov, Josefine Ghattas, Bertrand Guenet, Anne Sofie Lansø, Kim Naudts, Aude Valade, Chao Yue, and Sebastiaan Luyssaert
Geosci. Model Dev., 17, 8023–8047, https://doi.org/10.5194/gmd-17-8023-2024, https://doi.org/10.5194/gmd-17-8023-2024, 2024
Short summary
Short summary
This research looks at how climate change influences forests, and particularly how altered wind and insect activities could make forests emit instead of absorb carbon. We have updated a land surface model called ORCHIDEE to better examine the effect of bark beetles on forest health. Our findings suggest that sudden events, such as insect outbreaks, can dramatically affect carbon storage, offering crucial insights into tackling climate change.
Stephen Björn Wirth, Johanna Braun, Jens Heinke, Sebastian Ostberg, Susanne Rolinski, Sibyll Schaphoff, Fabian Stenzel, Werner von Bloh, Friedhelm Taube, and Christoph Müller
Geosci. Model Dev., 17, 7889–7914, https://doi.org/10.5194/gmd-17-7889-2024, https://doi.org/10.5194/gmd-17-7889-2024, 2024
Short summary
Short summary
We present a new approach to modelling biological nitrogen fixation (BNF) in the Lund–Potsdam–Jena managed Land dynamic global vegetation model. While in the original approach BNF depended on actual evapotranspiration, the new approach considers soil water content and temperature, vertical root distribution, the nitrogen (N) deficit and carbon (C) costs. The new approach improved simulated BNF compared to the scientific literature and the model ability to project future C and N cycle dynamics.
Saeed Harati-Asl, Liliana Perez, and Roberto Molowny-Horas
Geosci. Model Dev., 17, 7423–7443, https://doi.org/10.5194/gmd-17-7423-2024, https://doi.org/10.5194/gmd-17-7423-2024, 2024
Short summary
Short summary
Social–ecological systems are the subject of many sustainability problems. Because of the complexity of these systems, we must be careful when intervening in them; otherwise we may cause irreversible damage. Using computer models, we can gain insight about these complex systems without harming them. In this paper we describe how we connected an ecological model of forest insect infestation with a social model of cooperation and simulated an intervention measure to save a forest from infestation.
Daehyuk Kim, Hyun-Chae Jung, Jae-Hong Moon, and Na-Hyeon Lee
EGUsphere, https://doi.org/10.5194/egusphere-2024-1509, https://doi.org/10.5194/egusphere-2024-1509, 2024
Short summary
Short summary
Physical–biogeochemical ocean global models is difficult to analyze oceanic environmental systems. To accurately understand the physical–biogeochemical processes at the regional scale, physical and biogeochemical models were coupled at a high resolution. The results successfully simulated the seasonal variations of chlorophyll and nutrients, particularly in the marginal seas, which were not captured by global models. The model is an important tool for studying physical–biogeochemical processes.
Katarína Merganičová, Ján Merganič, Laura Dobor, Roland Hollós, Zoltán Barcza, Dóra Hidy, Zuzana Sitková, Pavel Pavlenda, Hrvoje Marjanovic, Daniel Kurjak, Michal Bošel'a, Doroteja Bitunjac, Maša Zorana Ostrogović Sever, Jiří Novák, Peter Fleischer, and Tomáš Hlásny
Geosci. Model Dev., 17, 7317–7346, https://doi.org/10.5194/gmd-17-7317-2024, https://doi.org/10.5194/gmd-17-7317-2024, 2024
Short summary
Short summary
We developed a multi-objective calibration approach leading to robust parameter values aiming to strike a balance between their local precision and broad applicability. Using the Biome-BGCMuSo model, we tested the calibrated parameter sets for simulating European beech forest dynamics across large environmental gradients. Leveraging data from 87 plots and five European countries, the results demonstrated reasonable local accuracy and plausible large-scale productivity responses.
Hoa T. T. Nguyen, Ute Daewel, Neil Banas, and Corinna Schrum
EGUsphere, https://doi.org/10.5194/egusphere-2024-2710, https://doi.org/10.5194/egusphere-2024-2710, 2024
Short summary
Short summary
Parameterisation is key in modeling to reproduce observations well but is often done manually. This study presents a Particle Swarm Optimizer-based toolbox for marine ecosystem models, compatible with the Framework for Aquatic Biogeochemical Models, thus enhancing its reusability. Applied to the Sylt ecosystem, the toolbox effectively (1) identified multiple parameter sets that matched observations well, thus providing different insights into ecosystem dynamics, (2) optimized model complexity.
Sylvain Schmitt, Fabian Fischer, James Ball, Nicolas Barbier, Marion Boisseaux, Damien Bonal, Benoit Burban, Xiuzhi Chen, Géraldine Derroire, Jeremy Lichstein, Daniela Nemetschek, Natalia Restrepo-Coupe, Scott Saleska, Giacomo Sellan, Philippe Verley, Grégoire Vincent, Camille Ziegler, Jérôme Chave, and Isabelle Maréchaux
EGUsphere, https://doi.org/10.5194/egusphere-2024-3106, https://doi.org/10.5194/egusphere-2024-3106, 2024
Short summary
Short summary
We evaluate the capability of TROLL 4.0, a simulator of forest dynamics, to represent tropical forest structure, diversity and functioning in two Amazonian forests. Evaluation data include forest inventories, carbon and water fluxes between the forest and the atmosphere, and leaf area and canopy height from remote-sensing products. The model realistically predicts the structure and composition, and the seasonality of carbon and water fluxes at both sites.
Elchin E. Jafarov, Helene Genet, Velimir V. Vesselinov, Valeria Briones, Aiza Kabeer, Andrew L. Mullen, Benjamin Maglio, Tobey Carman, Ruth Rutter, Joy Clein, Chu-Chun Chang, Dogukan Teber, Trevor Smith, Joshua M. Rady, Christina Schädel, Jennifer D. Watts, Brendan M. Rogers, and Susan M. Natali
Geosci. Model Dev. Discuss., https://doi.org/10.5194/gmd-2024-158, https://doi.org/10.5194/gmd-2024-158, 2024
Revised manuscript accepted for GMD
Short summary
Short summary
Thawing permafrost could greatly impact global climate. Our study improves modeling of carbon cycling in Arctic ecosystems. We developed an automated method to fine-tune a model that simulates carbon and nitrogen flows, using computer-generated data. Using computer-generated data, we tested our method and found it enhances accuracy and reduces the time needed for calibration. This work helps make climate predictions more reliable in sensitive permafrost regions.
Zavud Baghirov, Martin Jung, Markus Reichstein, Marco Körner, and Basil Kraft
EGUsphere, https://doi.org/10.5194/egusphere-2024-2044, https://doi.org/10.5194/egusphere-2024-2044, 2024
Short summary
Short summary
We use an innovative approach to study the Earth's water cycle by blending advanced computer learning techniques with a traditional water cycle model. We developed a model that learns from meteorological data, with a special focus on understanding how vegetation influences water movement. Our model closely aligns with real-world observations, yet there are areas that need improvement. This study opens up new possibilities to better understand the water cycle and its interactions with vegetation.
Guohua Liu, Mirco Migliavacca, Christian Reimers, Basil Kraft, Markus Reichstein, Andrew D. Richardson, Lisa Wingate, Nicolas Delpierre, Hui Yang, and Alexander J. Winkler
Geosci. Model Dev., 17, 6683–6701, https://doi.org/10.5194/gmd-17-6683-2024, https://doi.org/10.5194/gmd-17-6683-2024, 2024
Short summary
Short summary
Our study employs long short-term memory (LSTM) networks to model canopy greenness and phenology, integrating meteorological memory effects. The LSTM model outperforms traditional methods, enhancing accuracy in predicting greenness dynamics and phenological transitions across plant functional types. Highlighting the importance of multi-variate meteorological memory effects, our research pioneers unlock the secrets of vegetation phenology responses to climate change with deep learning techniques.
Thi Lan Anh Dinh, Daniel Goll, Philippe Ciais, and Ronny Lauerwald
Geosci. Model Dev., 17, 6725–6744, https://doi.org/10.5194/gmd-17-6725-2024, https://doi.org/10.5194/gmd-17-6725-2024, 2024
Short summary
Short summary
The study assesses the performance of the dynamic global vegetation model (DGVM) ORCHIDEE in capturing the impact of land-use change on carbon stocks across Europe. Comparisons with observations reveal that the model accurately represents carbon fluxes and stocks. Despite the underestimations in certain land-use conversions, the model describes general trends in soil carbon response to land-use change, aligning with the site observations.
Nathaelle Bouttes, Lester Kwiatkowski, Manon Berger, Victor Brovkin, and Guy Munhoven
Geosci. Model Dev., 17, 6513–6528, https://doi.org/10.5194/gmd-17-6513-2024, https://doi.org/10.5194/gmd-17-6513-2024, 2024
Short summary
Short summary
Coral reefs are crucial for biodiversity, but they also play a role in the carbon cycle on long time scales of a few thousand years. To better simulate the future and past evolution of coral reefs and their effect on the global carbon cycle, hence on atmospheric CO2 concentration, it is necessary to include coral reefs within a climate model. Here we describe the inclusion of coral reef carbonate production in a carbon–climate model and its validation in comparison to existing modern data.
Huajie Zhu, Mousong Wu, Fei Jiang, Michael Vossbeck, Thomas Kaminski, Xiuli Xing, Jun Wang, Weimin Ju, and Jing M. Chen
Geosci. Model Dev., 17, 6337–6363, https://doi.org/10.5194/gmd-17-6337-2024, https://doi.org/10.5194/gmd-17-6337-2024, 2024
Short summary
Short summary
In this work, we developed the Nanjing University Carbon Assimilation System (NUCAS v1.0). Data assimilation experiments were conducted to demonstrate the robustness and investigate the feasibility and applicability of NUCAS. The assimilation of ecosystem carbonyl sulfide (COS) fluxes improved the model performance in gross primary productivity, evapotranspiration, and sensible heat, showing that COS provides constraints on parameters relevant to carbon-, water-, and energy-related processes.
Ye Liu, Huilin Huang, Sing-Chun Wang, Tao Zhang, Donghui Xu, and Yang Chen
Geosci. Model Dev. Discuss., https://doi.org/10.5194/gmd-2024-151, https://doi.org/10.5194/gmd-2024-151, 2024
Revised manuscript accepted for GMD
Short summary
Short summary
This study integrates machine learning with a land surface model to improve wildfire predictions in North America. Traditional models struggle with accurately simulating burned areas due to simplified processes. By combining the predictive power of machine learning with a land model, our hybrid framework better captures fire dynamics. This approach enhances our understanding of wildfire behavior and aids in developing more effective climate and fire management strategies.
Sergi Molins, Benjamin Andre, Jeffrey Johnson, Glenn Hammond, Benjamin Sulman, Konstantin Lipnikov, Marcus Day, James Beisman, Daniil Svyatsky, Hang Deng, Peter Lichtner, Carl Steefel, and David Moulton
Geosci. Model Dev. Discuss., https://doi.org/10.5194/gmd-2024-108, https://doi.org/10.5194/gmd-2024-108, 2024
Revised manuscript accepted for GMD
Short summary
Short summary
Developing scientific software and making sure it functions properly requires a significant effort. As we advance our understanding of natural systems, however, there is the need to develop yet more complex models and codes. In this work, we present a piece of software that facilitates this work, specifically with regard to reactive processes. Existing tried-and-true codes are made available via this new interface, freeing up resources to focus on the new aspects of the problems at hand.
Fang Li, Zhimin Zhou, Samuel Levis, Stephen Sitch, Felicity Hayes, Zhaozhong Feng, Peter B. Reich, Zhiyi Zhao, and Yanqing Zhou
Geosci. Model Dev., 17, 6173–6193, https://doi.org/10.5194/gmd-17-6173-2024, https://doi.org/10.5194/gmd-17-6173-2024, 2024
Short summary
Short summary
A new scheme is developed to model the surface ozone damage to vegetation in regional and global process-based models. Based on 4210 data points from ozone experiments, it accurately reproduces statistically significant linear or nonlinear photosynthetic and stomatal responses to ozone in observations for all vegetation types. It also enables models to implicitly capture the variability in plant ozone tolerance and the shift among species within a vegetation type.
Alexander S. Brunmayr, Frank Hagedorn, Margaux Moreno Duborgel, Luisa I. Minich, and Heather D. Graven
Geosci. Model Dev., 17, 5961–5985, https://doi.org/10.5194/gmd-17-5961-2024, https://doi.org/10.5194/gmd-17-5961-2024, 2024
Short summary
Short summary
A new generation of soil models promises to more accurately predict the carbon cycle in soils under climate change. However, measurements of 14C (the radioactive carbon isotope) in soils reveal that the new soil models face similar problems to the traditional models: they underestimate the residence time of carbon in soils and may therefore overestimate the net uptake of CO2 by the land ecosystem. Proposed solutions include restructuring the models and calibrating model parameters with 14C data.
Nina Raoult, Simon Beylat, James M. Salter, Frédéric Hourdin, Vladislav Bastrikov, Catherine Ottlé, and Philippe Peylin
Geosci. Model Dev., 17, 5779–5801, https://doi.org/10.5194/gmd-17-5779-2024, https://doi.org/10.5194/gmd-17-5779-2024, 2024
Short summary
Short summary
We use computer models to predict how the land surface will respond to climate change. However, these complex models do not always simulate what we observe in real life, limiting their effectiveness. To improve their accuracy, we use sophisticated statistical and computational techniques. We test a technique called history matching against more common approaches. This method adapts well to these models, helping us better understand how they work and therefore how to make them more realistic.
Jorn Bruggeman, Karsten Bolding, Lars Nerger, Anna Teruzzi, Simone Spada, Jozef Skákala, and Stefano Ciavatta
Geosci. Model Dev., 17, 5619–5639, https://doi.org/10.5194/gmd-17-5619-2024, https://doi.org/10.5194/gmd-17-5619-2024, 2024
Short summary
Short summary
To understand and predict the ocean’s capacity for carbon sequestration, its ability to supply food, and its response to climate change, we need the best possible estimate of its physical and biogeochemical properties. This is obtained through data assimilation which blends numerical models and observations. We present the Ensemble and Assimilation Tool (EAT), a flexible and efficient test bed that allows any scientist to explore and further develop the state of the art in data assimilation.
Dongyu Zheng, Andrew S. Merdith, Yves Goddéris, Yannick Donnadieu, Khushboo Gurung, and Benjamin J. W. Mills
Geosci. Model Dev., 17, 5413–5429, https://doi.org/10.5194/gmd-17-5413-2024, https://doi.org/10.5194/gmd-17-5413-2024, 2024
Short summary
Short summary
This study uses a deep learning method to upscale the time resolution of paleoclimate simulations to 1 million years. This improved resolution allows a climate-biogeochemical model to more accurately predict climate shifts. The method may be critical in developing new fully continuous methods that are able to be applied over a moving continental surface in deep time with high resolution at reasonable computational expense.
Boris Ťupek, Aleksi Lehtonen, Alla Yurova, Rose Abramoff, Bertrand Guenet, Elisa Bruni, Samuli Launiainen, Mikko Peltoniemi, Shoji Hashimoto, Xianglin Tian, Juha Heikkinen, Kari Minkkinen, and Raisa Mäkipää
Geosci. Model Dev., 17, 5349–5367, https://doi.org/10.5194/gmd-17-5349-2024, https://doi.org/10.5194/gmd-17-5349-2024, 2024
Short summary
Short summary
Updating the Yasso07 soil C model's dependency on decomposition with a hump-shaped Ricker moisture function improved modelled soil organic C (SOC) stocks in a catena of mineral and organic soils in boreal forest. The Ricker function, set to peak at a rate of 1 and calibrated against SOC and CO2 data using a Bayesian approach, showed a maximum in well-drained soils. Using SOC and CO2 data together with the moisture only from the topsoil humus was crucial for accurate model estimates.
Jacquelyn K. Shuman, Rosie A. Fisher, Charles Koven, Ryan Knox, Lara Kueppers, and Chonggang Xu
Geosci. Model Dev., 17, 4643–4671, https://doi.org/10.5194/gmd-17-4643-2024, https://doi.org/10.5194/gmd-17-4643-2024, 2024
Short summary
Short summary
We adapt a fire behavior and effects module for use in a size-structured vegetation demographic model to test how climate, fire regime, and fire-tolerance plant traits interact to determine the distribution of tropical forests and grasslands. Our model captures the connection between fire disturbance and plant fire-tolerance strategies in determining plant distribution and provides a useful tool for understanding the vulnerability of these areas under changing conditions across the tropics.
Yoshiki Kanzaki, Isabella Chiaravalloti, Shuang Zhang, Noah J. Planavsky, and Christopher T. Reinhard
Geosci. Model Dev., 17, 4515–4532, https://doi.org/10.5194/gmd-17-4515-2024, https://doi.org/10.5194/gmd-17-4515-2024, 2024
Short summary
Short summary
Soil pH is one of the most commonly measured agronomical and biogeochemical indices, mostly reflecting exchangeable acidity. Explicit simulation of both porewater and bulk soil pH is thus crucial to the accurate evaluation of alkalinity required to counteract soil acidification and the resulting capture of anthropogenic carbon dioxide through the enhanced weathering technique. This has been enabled by the updated reactive–transport SCEPTER code and newly developed framework to simulate soil pH.
David Sandoval, Iain Colin Prentice, and Rodolfo L. B. Nóbrega
Geosci. Model Dev., 17, 4229–4309, https://doi.org/10.5194/gmd-17-4229-2024, https://doi.org/10.5194/gmd-17-4229-2024, 2024
Short summary
Short summary
Numerous estimates of water and energy balances depend on empirical equations requiring site-specific calibration, posing risks of "the right answers for the wrong reasons". We introduce novel first-principles formulations to calculate key quantities without requiring local calibration, matching predictions from complex land surface models.
Oliver Perkins, Matthew Kasoar, Apostolos Voulgarakis, Cathy Smith, Jay Mistry, and James D. A. Millington
Geosci. Model Dev., 17, 3993–4016, https://doi.org/10.5194/gmd-17-3993-2024, https://doi.org/10.5194/gmd-17-3993-2024, 2024
Short summary
Short summary
Wildfire is often presented in the media as a danger to human life. Yet globally, millions of people’s livelihoods depend on using fire as a tool. So, patterns of fire emerge from interactions between humans, land use, and climate. This complexity means scientists cannot yet reliably say how fire will be impacted by climate change. So, we developed a new model that represents globally how people use and manage fire. The model reveals the extent and diversity of how humans live with and use fire.
Amos P. K. Tai, David H. Y. Yung, and Timothy Lam
Geosci. Model Dev., 17, 3733–3764, https://doi.org/10.5194/gmd-17-3733-2024, https://doi.org/10.5194/gmd-17-3733-2024, 2024
Short summary
Short summary
We have developed the Terrestrial Ecosystem Model in R (TEMIR), which simulates plant carbon and pollutant uptake and predicts their response to varying atmospheric conditions. This model is designed to couple with an atmospheric chemistry model so that questions related to plant–atmosphere interactions, such as the effects of climate change, rising CO2, and ozone pollution on forest carbon uptake, can be addressed. The model has been well validated with both ground and satellite observations.
Fabian Stenzel, Johanna Braun, Jannes Breier, Karlheinz Erb, Dieter Gerten, Jens Heinke, Sarah Matej, Sebastian Ostberg, Sibyll Schaphoff, and Wolfgang Lucht
Geosci. Model Dev., 17, 3235–3258, https://doi.org/10.5194/gmd-17-3235-2024, https://doi.org/10.5194/gmd-17-3235-2024, 2024
Short summary
Short summary
We provide an R package to compute two biosphere integrity metrics that can be applied to simulations of vegetation growth from the dynamic global vegetation model LPJmL. The pressure metric BioCol indicates that we humans modify and extract > 20 % of the potential preindustrial natural biomass production. The ecosystems state metric EcoRisk shows a high risk of ecosystem destabilization in many regions as a result of climate change and land, water, and fertilizer use.
Elin Ristorp Aas, Heleen A. de Wit, and Terje K. Berntsen
Geosci. Model Dev., 17, 2929–2959, https://doi.org/10.5194/gmd-17-2929-2024, https://doi.org/10.5194/gmd-17-2929-2024, 2024
Short summary
Short summary
By including microbial processes in soil models, we learn how the soil system interacts with its environment and responds to climate change. We present a soil process model, MIMICS+, which is able to reproduce carbon stocks found in boreal forest soils better than a conventional land model. With the model we also find that when adding nitrogen, the relationship between soil microbes changes notably. Coupling the model to a vegetation model will allow for further study of these mechanisms.
Thomas Wutzler, Christian Reimers, Bernhard Ahrens, and Marion Schrumpf
Geosci. Model Dev., 17, 2705–2725, https://doi.org/10.5194/gmd-17-2705-2024, https://doi.org/10.5194/gmd-17-2705-2024, 2024
Short summary
Short summary
Soil microbes provide a strong link for elemental fluxes in the earth system. The SESAM model applies an optimality assumption to model those linkages and their adaptation. We found that a previous heuristic description was a special case of a newly developed more rigorous description. The finding of new behaviour at low microbial biomass led us to formulate the constrained enzyme hypothesis. We now can better describe how microbially mediated linkages of elemental fluxes adapt across decades.
Salvatore R. Curasi, Joe R. Melton, Elyn R. Humphreys, Txomin Hermosilla, and Michael A. Wulder
Geosci. Model Dev., 17, 2683–2704, https://doi.org/10.5194/gmd-17-2683-2024, https://doi.org/10.5194/gmd-17-2683-2024, 2024
Short summary
Short summary
Canadian forests are responding to fire, harvest, and climate change. Models need to quantify these processes and their carbon and energy cycling impacts. We develop a scheme that, based on satellite records, represents fire, harvest, and the sparsely vegetated areas that these processes generate. We evaluate model performance and demonstrate the impacts of disturbance on carbon and energy cycling. This work has implications for land surface modeling and assessing Canada’s terrestrial C cycle.
Yannek Käber, Florian Hartig, and Harald Bugmann
Geosci. Model Dev., 17, 2727–2753, https://doi.org/10.5194/gmd-17-2727-2024, https://doi.org/10.5194/gmd-17-2727-2024, 2024
Short summary
Short summary
Many forest models include detailed mechanisms of forest growth and mortality, but regeneration is often simplified. Testing and improving forest regeneration models is challenging. We address this issue by exploring how forest inventories from unmanaged European forests can be used to improve such models. We find that competition for light among trees is captured by the model, unknown model components can be informed by forest inventory data, and climatic effects are challenging to capture.
Jalisha T. Kallingal, Johan Lindström, Paul A. Miller, Janne Rinne, Maarit Raivonen, and Marko Scholze
Geosci. Model Dev., 17, 2299–2324, https://doi.org/10.5194/gmd-17-2299-2024, https://doi.org/10.5194/gmd-17-2299-2024, 2024
Short summary
Short summary
By unlocking the mysteries of CH4 emissions from wetlands, our work improved the accuracy of the LPJ-GUESS vegetation model using Bayesian statistics. Via assimilation of long-term real data from a wetland, we significantly enhanced CH4 emission predictions. This advancement helps us better understand wetland contributions to atmospheric CH4, which are crucial for addressing climate change. Our method offers a promising tool for refining global climate models and guiding conservation efforts
Benjamin Post, Esteban Acevedo-Trejos, Andrew D. Barton, and Agostino Merico
Geosci. Model Dev., 17, 1175–1195, https://doi.org/10.5194/gmd-17-1175-2024, https://doi.org/10.5194/gmd-17-1175-2024, 2024
Short summary
Short summary
Creating computational models of how phytoplankton grows in the ocean is a technical challenge. We developed a new tool set (Xarray-simlab-ODE) for building such models using the programming language Python. We demonstrate the tool set in a library of plankton models (Phydra). Our goal was to allow scientists to develop models quickly, while also allowing the model structures to be changed easily. This allows us to test many different structures of our models to find the most appropriate one.
Taeken Wijmer, Ahmad Al Bitar, Ludovic Arnaud, Remy Fieuzal, and Eric Ceschia
Geosci. Model Dev., 17, 997–1021, https://doi.org/10.5194/gmd-17-997-2024, https://doi.org/10.5194/gmd-17-997-2024, 2024
Short summary
Short summary
Quantification of carbon fluxes of crops is an essential building block for the construction of a monitoring, reporting, and verification approach. We developed an end-to-end platform (AgriCarbon-EO) that assimilates, through a Bayesian approach, high-resolution (10 m) optical remote sensing data into radiative transfer and crop modelling at regional scale (100 x 100 km). Large-scale estimates of carbon flux are validated against in situ flux towers and yield maps and analysed at regional scale.
Moritz Laub, Sergey Blagodatsky, Marijn Van de Broek, Samuel Schlichenmaier, Benjapon Kunlanit, Johan Six, Patma Vityakon, and Georg Cadisch
Geosci. Model Dev., 17, 931–956, https://doi.org/10.5194/gmd-17-931-2024, https://doi.org/10.5194/gmd-17-931-2024, 2024
Short summary
Short summary
To manage soil organic matter (SOM) sustainably, we need a better understanding of the role that soil microbes play in aggregate protection. Here, we propose the SAMM model, which connects soil aggregate formation to microbial growth. We tested it against data from a tropical long-term experiment and show that SAMM effectively represents the microbial growth, SOM, and aggregate dynamics and that it can be used to explore the importance of aggregate formation in SOM stabilization.
Cited articles
Arndt, S., Jørgensen, B., LaRowe, D., Middelburg, J., Pancost, R., and Regnier, P.: Quantifying the degradation of organic matter in marine sediments: A review and synthesis, Earth-Sci. Rev., 123, 53–86, https://doi.org/10.1016/j.earscirev.2013.02.008, 2013. a
Baturin, G. N.: Issue of the relationship between primary productivity of organic carbon in ocean and phosphate accumulation (Holocene-Late Jurassic), Lith. Miner. Resour., 42, 318–348, https://doi.org/10.1134/S0024490207040025, 2007. a
Becker, J. J., Wood, W. T., and Martin, K. M.: Global Crustal Heat Flow Using Random Decision Forest Prediction, in: AGU Fall Meeting Abstracts, vol. 2014, NG31A–3788, https://ui.adsabs.harvard.edu/abs/2014AGUFMNG31A3788B/abstract (last access: 3 February 2024), 2014. a
Berner, R. A.: Processes of the Long-Term Carbon Cycle: Organic Matter and Carbonate Burial and Weathering, in: The Phanerozoic Carbon Cycle: CO2 and O2, Oxford University Press, ISBN 9780195173338, https://doi.org/10.1093/oso/9780195173338.003.0005, 2004. a
Boudreau, B. P.: Diagenetic models and their implementation, vol. 505, Springer Berlin, ISBN 978-3-642-64399-6, https://doi.org/10.1007/978-3-642-60421-8, 1997. a
Bradley, J. A., Hülse, D., LaRowe, D. E., and Arndt, S.: Transfer efficiency of organic carbon in marine sediments, Nat. Commun., 13, 7297, https://doi.org/10.1038/s41467-022-35112-9, 2022. a, b, c
Burdige, D. J.: Burial of terrestrial organic matter in marine sediments: A re-assessment, Global Biogeochem. Cy., 19, GB4011, https://doi.org/10.1029/2004GB002368, 2005. a, b
Burdige, D. J.: Preservation of Organic Matter in Marine Sediments – Controls, Mechanisms, and an Imbalance in Sediment Organic Carbon Budgets?, Chem. Rev., 107, 467–485, https://doi.org/10.1021/cr050347q, 2007. a
Crameri, F.: Scientific colour maps, Zenodo [data set], https://doi.org/10.5281/zenodo.8409685, 2023. a
De Veaux, R. D. and Ungar, L. H.: Multicollinearity: A tale of two nonparametric regressions, in: Selecting Models from Data, edited by: Cheeseman, P. and Oldford, R. W., Springer New York, New York, NY, 393–402, ISBN 978-1-4612-2660-4, 1994. a
Diesing, M., Thorsnes, T., and Bjarnadóttir, L. R.: Organic carbon densities and accumulation rates in surface sediments of the North Sea and Skagerrak, Biogeosciences, 18, 2139–2160, https://doi.org/10.5194/bg-18-2139-2021, 2021. a, b
Emerson, S. and Hedges, J. I.: Processes controlling the organic carbon content of open ocean sediments, Paleoceanography, 3, 621–634, https://doi.org/10.1029/PA003i005p00621, 1988. a, b, c, d
Flanders Marine Institute: Global Oceans and Seas, version 1, https://doi.org/10.14284/542, https://www.marineregions.org/ (last access: 25 August 2023), 2021. a
Gal, Y. and Ghahramani, Z.: Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learning, in: Proceedings of The 33rd International Conference on Machine Learning, New York, New York, USA, edited by: Balcan, M. F. and Weinberger, K. Q., Proceedings of Machine Learning Research, vol. 48, PMLR, 19 June 2016, 1050–1059, https://proceedings.mlr.press/v48/gal16.html (last access: 24 February 2022), 2016. a
Goyet, C., Healy, R., and Ryan, J.: Global Distribution of Total Inorganic Carbon and Total Alkalinity Below the Deepest Winter Mixed Layer Depths, ORNIJCDIAC-127 NDP-076, https://doi.org/10.3334/zCDIAC/otg.ndp076, 2000. a
Hall, S. J.: The continental shelf benthic ecosystem: current status, agents for change and future prospects, Environ. Conserv., 29, 350–374, http://www.jstor.org/stable/44520615 (last access: 26 June 2022), 2002. a
He, K., Zhang, X., Ren, S., and Sun, J.: Delving deep into rectifiers: Surpassing human-level performance on imagenet classification, in: Proceedings of the IEEE International Conference on Computer Vision, Santiago, Chile, 1026–1034, 7–13 December 2015. a
Hedges, J. I. and Keil, R. G.: Sedimentary organic matter preservation: an assessment and speculative synthesis, Mar. Chem., 49, 81–115, https://doi.org/10.1016/0304-4203(95)00008-F, 1995. a, b, c, d
Joint Panel on Oceanographic Tables and Standards: Processing of Oceanographic Station Data, UNESCO, Paris, ISBN 978-92-3-102756-7, 1991. a
Jørgensen, B. B., Wenzhöfer, F., Egger, M., and Glud, R. N.: Sediment oxygen consumption: Role in the global marine carbon cycle, Earth-Sci. Rev., 228, 103987, https://doi.org/10.1016/j.earscirev.2022.103987, 2022. a, b, c, d
Kim, S. S. and Wessel, P.: New global seamount census from the altimetry-derived gravity data, Geophys. J. Int., 186, 615–631, https://doi.org/10.1111/j.1365-246X.2011.05076.x, 2011. a
Kullback, S. and Leibler, R. A.: On Information and Sufficiency, Ann. Math. Stat., 22, 79–86, http://www.jstor.org/stable/2236703 (last access: 19 April 2023), 1951. a
LaRowe, D., Arndt, S., Bradley, J., Estes, E., Hoarfrost, A., Lang, S., Lloyd, K., Mahmoudi, N., Orsi, W., Shah Walter, S., Steen, A., and Zhao, R.: The fate of organic carbon in marine sediments – New insights from recent data and analysis, Earth-Sci. Rev., 204, 103146, https://doi.org/10.1016/j.earscirev.2020.103146, 2020a. a, b
LaRowe, D. E., Arndt, S., Bradley, J. A., Burwicz, E., Dale, A. W., and Amend, J. P.: Organic carbon and microbial activity in marine sediments on a global scale throughout the Quaternary, Geochim. Cosmochim. Ac., 286, 227–247, https://doi.org/10.1016/j.gca.2020.07.017, 2020b. a
LeCun, Y., Bengio, Y., and Hinton, G.: Deep learning, Nature, 521, 436–444, https://doi.org/10.1038/nature14539, 2015. a
Lee, T. R., Wood, W. T., Skarke, A., Phrampus, B. J., and Obelcz, J.: Data files associated with the k-nearest neighbor global prediction of isopachs for present to middle Miocene, Zenodo [data set], https://doi.org/10.5281/zenodo.3675364, 2020. a, b
Legge, O., Johnson, M., Hicks, N., Jickells, T., Diesing, M., Aldridge, J., Andrews, J., Artioli, Y., Bakker, D. C. E., Burrows, M. T., Carr, N., Cripps, G., Felgate, S. L., Fernand, L., Greenwood, N., Hartman, S., Kröger, S., Lessin, G., Mahaffey, C., Mayor, D. J., Parker, R., Queirós, A. M., Shutler, J. D., Silva, T., Stahl, H., Tinker, J., Underwood, G. J. C., Van Der Molen, J., Wakelin, S., Weston, K., and Williamson, P.: Carbon on the Northwest European Shelf: Contemporary Budget and Future Influences, Frontiers in Marine Science, 7, 143, https://doi.org/10.3389/fmars.2020.00143, 2020. a
Lucas, M. and Giles, H.: Quantifying Uncertainty in Random Forests via Confidence Intervals and Hypothesis Tests, J. Mach. Learn. Res., 17, 1–41, http://jmlr.org/papers/v17/14-168.html (last access: 4 August 2023), 2016. a
National Geophysical Data Center: The NGDC Seafloor Sediment Grain Size Database, first Version, NOAA National Centers for Environmental Information, https://doi.org/10.7289/V5G44N6W, 1976. a
Pape, T., Bünz, S., Hong, W.-L., Torres, M. E., Riedel, M., Panieri, G., Lepland, A., Hsu, C.-W., Wintersteller, P., Wallmann, K., Schmidt, C., Yao, H., and Bohrmann, G.: Origin and Transformation of Light Hydrocarbons Ascending at an Active Pockmark on Vestnesa Ridge, Arctic Ocean, J. Geophys. Res.-Sol. Ea., 125, e2018JB016679, https://doi.org/10.1029/2018JB016679, 2020. a, b
Paradis, S., Nakajima, K., Van der Voort, T. S., Gies, H., Wildberger, A., Blattmann, T. M., Bröder, L., and Eglinton, T. I.: The Modern Ocean Sediment Archive and Inventory of Carbon (MOSAIC): version 2.0, Earth Syst. Sci. Data, 15, 4105–4125, https://doi.org/10.5194/essd-15-4105-2023, 2023. a, b
Parameswaran, N., González, E., Burwicz-Galerne, E., Braack, M., and Wallmann, K.: Code for Global Prediction Of Total Organic Carbon In Marine Sediments Using Deep Neural Networks (nn-toc), GitHub [code], https://github.com/paramnav/nn-toc (last access: 13 November 2024), 2024a. a
Parameswaran, N., González, E., Burwicz-Galerne, E., Braack, M., and Wallmann, K.: Dataset for the Global Prediction Of Total Organic Carbon In Marine Sediments Using Deep Neural Networks (nn-toc), Zenodo [data set], https://doi.org/10.5281/zenodo.11186224, 2024b. a, b, c, d
Pavlis, N. K., Holmes, S. A., Kenyon, S. C., and Factor, J. K.: The EGM2008 Global Gravitational Model, abstract 2008AGUFM.G22A..01P, 2008 General Assembly of the European Geosciences Union, Vienna, Austria, https://ui.adsabs.harvard.edu/abs/2008AGUFM.G22A..01P (last access: 7 October 2014), 2008. a, b
Phrampus, B. J., Lee, T. R., and Wood, W. T.: Predictor Grids for “A Global Probabilistic Prediction of Cold Seeps and Associated Seafloor Fluid Expulsion Anomalies (SEAFLEAs)”, Zenodo [data set], https://doi.org/10.5281/zenodo.3459805, 2019. a
Rényi, A.: On measures of entropy and information, in: Proceedings of the fourth Berkeley symposium on mathematical statistics and probability, volume 1: contributions to the theory of statistics, Fourth Berkley Symposum on Mathematical Statistics and Probablity June 20-July 30, 1960, Statistical Laboratory of the University of California vol. 4, University of California Press, 547–562, 1961. a
Restreppo, G. A., Wood, W. T., and Phrampus, B. J.: Oceanic sediment accumulation rates predicted via machine learning algorithm: towards sediment characterization on a global scale, Geo-Mar. Lett., 40, 755–763, https://doi.org/10.1007/s00367-020-00669-1, 2020. a
Restreppo, G. A., Wood, W. T., and Phrampus, B. J.: A machine-learning derived model of seafloor sediment accumulation, Mar. Geol., 440, 106577, https://doi.org/10.1016/j.margeo.2021.106577, 2021. a, b
Romankevich, E., Vetrov, A., and Peresypkin, V.: Organic matter of the World Ocean, Russ. Geol. Geophys., 50, 299–307, https://doi.org/10.1016/j.rgg.2009.03.013, 2009. a, b
Sala, E., Mayorga, J., Bradley, D., Cabral, R. B., Atwood, T. B., Auber, A., Cheung, W., Costello, C., Ferretti, F., Friedl, er, A. M., Gaines, S. D., Garilao, C., Goodell, W., Halpern, B. S., Hinson, A., Kaschner, K., Kesner-Reyes, K., Leprieur, F., McGowan, J., Morgan, L. E., Mouillot, D., Palacios-Abrantes, J., Possingham, H. P., Rechberger, K. D., Worm, B., and Lubchenco, J.: Protecting the global ocean for biodiversity, food and climate, Nature, 592, 397–402, https://doi.org/10.1038/s41586-021-03371-z, 2021. a
Shannon, C. E.: A mathematical theory of communication, The Bell System Technical Journal, 27, 379–423, https://doi.org/10.1002/j.1538-7305.1948.tb01338.x, 1948. a, b
Song, S., Santos, I. R., Yu, H., Wang, F., Burnett, W. C., Bianchi, T. S., Dong, J., Lian, E., Zhao, B., Mayer, L., Yao, Q., Yu, Z., and Xu, B.: A global assessment of the mixed layer in coastal sediments and implications for carbon storage, Nat. Commun., 13, 4903, https://doi.org/10.1038/s41467-022-32650-0, 2022. a, b
Song, T., Pang, C., Hou, B., Xu, G., Xue, J., Sun, H., and Meng, F.: A review of artificial intelligence in marine science, Front. Earth Sci., 11, https://doi.org/10.3389/feart.2023.1090185, 2023. a
Thyng, K. M., Greene, C. A., Hetland, R. D., Zimmerle, H. M., and DiMarco, S. F.: True Colors of Oceanography, Oceanography, 29, 9–13, https://doi.org/10.5670/oceanog.2016.66, 2016. a
van der Voort, T. S., Blattmann, T. M., Usman, M., Montluçon, D., Loeffler, T., Tavagna, M. L., Gruber, N., and Eglinton, T. I.: MOSAIC (Modern Ocean Sediment Archive and Inventory of Carbon): a (radio)carbon-centric database for seafloor surficial sediments, Earth Syst. Sci. Data, 13, 2135–2146, https://doi.org/10.5194/essd-13-2135-2021, 2021. a, b
Wei, C.-L., Rowe, G. T., Escobar-Briones, E., Boetius, A., Soltwedel, T., and Caley, M. J.: Global patterns and predictions of seafloor biomass using random forests, PLoS ONE, 5, e15323, https://doi.org/10.1371/journal.pone.0015323, 2010. a, b
Whittaker, J., Goncharov, A., Williams, S., Müller, R. D., and Leitchenkov, G.: Global sediment thickness dataset updated for the Australian-Antarctic Southern Ocean, Geochem. Geophy. Geosy., 14, 3297–3305, https://doi.org/10.1002/ggge.20181, 2013. a
Short summary
Our research uses deep learning to predict organic carbon stocks in ocean sediments, which is crucial for understanding their role in the global carbon cycle. By analysing over 22 000 samples and various seafloor characteristics, our model gives more accurate results than traditional methods. We estimate that the top 10 cm of ocean sediments hold about 156 Pg of carbon. This work enhances carbon stock estimates and helps plan future sampling strategies to better understand oceanic carbon burial.
Our research uses deep learning to predict organic carbon stocks in ocean sediments, which is...
