Answers are grounded in this paper and cite their sources.
Accurate quantification of GHG emissions from agricultural soils based on key environmental and management factors, is crucial for developing effective mitigation strategies. This study applies a machine learning approach using data from multiple montane grasslands in central Europe to model the spatial and temporal dynamics of N 2 O emissions, using meteorological, soil, and management data. The primary aim is to advance predictive modelling of N 2 O emissions from grassland soils for spatial upscaling and scenario analysis. Specifically, we assess whether a regionally trained model can accurately estimate N 2 O emissions from an independent site excluded from training. We collected data from five fertilized grasslands across southern Germany, northern Switzerland and western Austria (122 soil-site-treatment-year combinations), and trained a long short-term memory (LSTM) algorithm to model the influence of drivers within 5 subsequent days on N 2 O emissions. The dataset includes daily N 2 O emission measurements along with key emission drivers such as soil moisture and temperature in 10 cm soil depth, daily precipitation, occurrence of fertilization events (zero or one flag), as well as static soil characteristics such as pH and bulk density. The trained LSTM model showed varying predictive performance when evaluated on two test datasets, with higher performance at some site-years and lower at others. One test set included temporally independent data (hold-out years) (RMSE = 28 µg N 2 O-N m −2 h −1 ), while the other included cross-site data from a soil type withheld from training (RMSE = 11 µg N 2 O-N m −2 h −1 ). The model captured N 2 O dynamics, including the magnitude and timing of emission peaks driven by slurry application and environmental factors, though predictive performance varied across individual site-years. These results demonstrate that LSTM-based models can predict N 2 O emissions for most site-years across test sites with similar environmental and soil characteristics, representing an alternative to process-based models for predicting soil N 2 O emissions.
No takes yet. Share an insight, caveat, or question.
Payrosangari et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: