Publication Abstracts

Agredazywczuk et al. 2026

Agredazywczuk, P., T. Ouma, M. Barthel, A. Otinga, R. Njoroge, K. Butterbach-Bahl, R. Daelman, J.E. Hickman, W. lbrahim, M. Laub, S. Leitner, A. Shumba, K.L. Tully, S. Wachiye, J. Zheng, M. Bauters, R. Kiese, R. Cardinael, G. Obozinski, J. Six, and E. Harris, 2026: Nitrous oxide emissions across Sub-Saharan Africa: Meta-analysis and data-driven modelling. Glob. Biogeochem. Cycles, 40, no. 7, e2026GB009141, doi:10.1029/2026GB009141.

Food security and avoiding large-scale land use change in Sub-Saharan Africa (SSA) requires increasing agricultural productivity, necessitating greater fertiliser use. This may increase soil nitrous oxide (N2O) emissions, a potent greenhouse gas. This study used Machine learning (ML) models to predict N2O emissions under future climatic and fertiliser scenarios across SSA. Three models were trained (Random Forest (RF), XGBoost (XGB), and feedforward neural networks (FNN)) on existing forest, grassland, and cropland N2O measurements. The analysis identified the main drivers influencing N2O emissions: temperature, soil moisture, rainfall, and fertiliser (cropland). SSA N2O emissions totalled 252-538 Gg N yr-1 (1 Gg = 109g), with forests contributing 119-342 Gg N yr-1, grasslands 70-132 Gg N yr-1, and croplands 63-64 Gg N yr-1. Ranges reflect the full uncertainty across all models. Unexpectedly, severe climate change (SSP5-8.5 scenario) may decrease total N2O emissions by 4%-37% across SSA, possibly due to drier soils. However, when climate change was combined with tripled fertiliser use (0-63 kg N yr-1 to 0-189 kg N yr-1), the models predicted a wide range of cropland emissions increases of 6%-23% (XGB-RF), iwith FNN predicting 139% from baseline projections. These findings highlight that while climate change may reduce overall N2O emissions from forests and grasslands, agricultural intensification will likely become an increasingly significant emission source. To improve prediction accuracy, more comprehensive N2O monitoring across SSA is needed. This work underscores the need for international investment in monitoring infrastructure and an accessible regional data repository to guide sustainable agricultural development and SSA climate policy.

Export citation: [ BibTeX ] [ RIS ]

BibTeX Citation

@article{ag06000g,
  author={Agredazywczuk, P. and Ouma, T. and Barthel, M. and Otinga, A. and Njoroge, R. and Butterbach-Bahl, K. and Daelman, R. and Hickman, J. E. and lbrahim, W. and Laub, M. and Leitner, S. and Shumba, A. and Tully, K. L. and Wachiye, S. and Zheng, J. and Bauters, M. and Kiese, R. and Cardinael, R. and Obozinski, G. and Six, J. and Harris, E.},
  title={Nitrous oxide emissions across Sub-Saharan Africa: Meta-analysis and data-driven modelling},
  year={2026},
  journal={Global Biogeochemical Cycles},
  volume={40},
  number={7},
  pages={e2026GB009141},
  doi={10.1029/2026GB009141},
}

[ Close ]

RIS Citation

TY  - JOUR
ID  - ag06000g
AU  - Agredazywczuk, P.
AU  - Ouma, T.
AU  - Barthel, M.
AU  - Otinga, A.
AU  - Njoroge, R.
AU  - Butterbach-Bahl, K.
AU  - Daelman, R.
AU  - Hickman, J. E.
AU  - lbrahim, W.
AU  - Laub, M.
AU  - Leitner, S.
AU  - Shumba, A.
AU  - Tully, K. L.
AU  - Wachiye, S.
AU  - Zheng, J.
AU  - Bauters, M.
AU  - Kiese, R.
AU  - Cardinael, R.
AU  - Obozinski, G.
AU  - Six, J.
AU  - Harris, E.
PY  - 2026
TI  - Nitrous oxide emissions across Sub-Saharan Africa: Meta-analysis and data-driven modelling
JA  - Glob. Biogeochem. Cycles
JO  - Global Biogeochemical Cycles
VL  - 40
IS  - 7
SP  - e2026GB009141
DO  - 10.1029/2026GB009141
ER  -

[ Close ]

• Return to 2026 Publications

• Return to Publications Homepage