The abstract is read at the publisher; the summary is JClub's.
In a livestock-dense mixed land-use region in the Netherlands, land-use regression models estimated atmospheric NH 3 with cross-validated R² between 0.63 and 0.75.
- Why it matters: Atmospheric ammonia exposure is linked to adverse human health and ecosystem effects, but high-density NH 3 measurement networks are lacking, limiting prior land-use regression modeling.
- What they did: The study built LUR models using livestock- and non-livestock predictors and compared SLR, LASSO, and Random Forest across basic, emission, and hybrid predictor sets.
- The result: Random Forest did not outperform SLR or LASSO (mean CV R² 0.65 vs. 0.72 and 0.66), and pig, veal calf, laying hen, and grassland variables were positively associated with measured NH 3 concentrations.
The findingWhy it mattersWhat they didThe result
AI summary of the abstract · may contain errors · not medical advice · read the paper.