FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4200200323> ?p ?o ?g. }

• W4200200323 abstract "<strong class=journal-contentHeaderColor>Abstract.</strong> Agricultural nitrous oxide (<span class=inline-formula>N₂O</span>) emission accounts for a non-trivial fraction of global greenhouse gas (GHG) budget. To date, estimating <span class=inline-formula>N₂O</span> fluxes from cropland remains a challenging task because the related microbial processes (e.g., nitrification and denitrification) are controlled by complex interactions among climate, soil, plant and human activities. Existing approaches such as process-based (PB) models have well-known limitations due to insufficient representations of the processes or uncertainties of model parameters, and due to leverage recent advances in machine learning (ML) a new method is needed to unlock the “black box” to overcome its limitations such as low interpretability, out-of-sample failure and massive data demand. In this study, we developed a first-of-its-kind knowledge-guided machine learning model for agroecosystems (KGML-ag) by incorporating biogeophysical and chemical domain knowledge from an advanced PB model, <i>ecosys</i>, and tested it by comparing simulating daily <span class=inline-formula>N₂O</span> fluxes with real observed data from mesocosm experiments. The gated recurrent unit (GRU) was used as the basis to build the model structure. To optimize the model performance, we have investigated a range of ideas, including (1) using initial values of intermediate variables (IMVs) instead of time series as model input to reduce data demand; (2) building hierarchical structures to explicitly estimate IMVs for further <span class=inline-formula>N₂O</span> prediction; (3) using multi-task learning to balance the simultaneous training on multiple variables; and (4) pre-training with millions of synthetic data generated from <i>ecosys</i> and fine-tuning with mesocosm observations. Six other pure ML models were developed using the same mesocosm data to serve as the benchmark for the KGML-ag model. Results show that KGML-ag did an excellent job in reproducing the mesocosm <span class=inline-formula>N₂O</span> fluxes (overall <span class=inline-formula><i>r</i>²=0.81</span>, and <span class=inline-formula>RMSE=3.6</span> <span class=inline-formula><math xmlns=http://www.w3.org/1998/Math/MathML id=M10 display=inline overflow=scroll dspmath=mathml><mrow class=unit><mi mathvariant=normal>mg</mi><mspace linebreak=nobreak width=0.125em/><mi mathvariant=normal>N</mi><mspace width=0.125em linebreak=nobreak/><msup><mi mathvariant=normal>m</mi><mrow><mo>-</mo><mn mathvariant=normal>2</mn></mrow></msup><mspace linebreak=nobreak width=0.125em/><msup><mi mathvariant=normal>d</mi><mrow><mo>-</mo><mn mathvariant=normal>1</mn></mrow></msup></mrow></math><span><svg:svg xmlns:svg=http://www.w3.org/2000/svg width=65pt height=15pt class=svg-formula dspmath=mathimg md5hash=d460c380b5fd610084b58977b186a3d4><svg:image xmlns:xlink=http://www.w3.org/1999/xlink xlink:href=gmd-15-2839-2022-ie00001.svg width=65pt height=15pt src=gmd-15-2839-2022-ie00001.png/></svg:svg></span></span> from cross validation). Importantly, KGML-ag always outperforms the PB model and ML models in predicting <span class=inline-formula>N₂O</span> fluxes, especially for complex temporal dynamics and emission peaks. Besides, KGML-ag goes beyond the pure ML models by providing more interpretable predictions as well as pinpointing desired new knowledge and data to further empower the current KGML-ag. We believe the KGML-ag development in this study will stimulate a new body of research on interpretable ML for biogeochemistry and other related geoscience processes." @default.
• W4200200323 created "2021-12-31" @default.
• W4200200323 date "2021-11-23" @default.
• W4200200323 modified "2023-10-16" @default.
• W4200200323 title "Reply on AC1" @default.
• W4200200323 doi "https://doi.org/10.5194/gmd-2021-317-cc2" @default.
• W4200200323 hasPublicationYear "2021" @default.
• W4200200323 type Work @default.
• W4200200323 citedByCount "0" @default.
• W4200200323 crossrefType "peer-review" @default.
• W4200200323 hasBestOaLocation W42002003231 @default.
• W4200200323 hasConcept C119857082 @default.
• W4200200323 hasConcept C153083717 @default.
• W4200200323 hasConcept C2781067378 @default.
• W4200200323 hasConcept C39432304 @default.
• W4200200323 hasConcept C41008148 @default.
• W4200200323 hasConceptScore W4200200323C119857082 @default.
• W4200200323 hasConceptScore W4200200323C153083717 @default.
• W4200200323 hasConceptScore W4200200323C2781067378 @default.
• W4200200323 hasConceptScore W4200200323C39432304 @default.
• W4200200323 hasConceptScore W4200200323C41008148 @default.
• W4200200323 hasLocation W42002003231 @default.
• W4200200323 hasOpenAccess W4200200323 @default.
• W4200200323 hasPrimaryLocation W42002003231 @default.
• W4200200323 hasRelatedWork W2766672297 @default.
• W4200200323 hasRelatedWork W2899084033 @default.
• W4200200323 hasRelatedWork W2972016458 @default.
• W4200200323 hasRelatedWork W2997882704 @default.
• W4200200323 hasRelatedWork W3009560883 @default.
• W4200200323 hasRelatedWork W3023163568 @default.
• W4200200323 hasRelatedWork W4224048847 @default.
• W4200200323 hasRelatedWork W4224989370 @default.
• W4200200323 hasRelatedWork W4289856193 @default.
• W4200200323 hasRelatedWork W4312914540 @default.
• W4200200323 isParatext "false" @default.
• W4200200323 isRetracted "false" @default.
• W4200200323 workType "peer-review" @default.