Abstract

There is a broad scientific consensus that ammonia (NH3) emissions have a primary impact on air quality. In particular, the intensive farming of livestock, primarily bovine and swine, is recognised as being responsible for approximately 97% of NH3 emissions in Lombardy, Italy. A relevant issue is related to the impact on airborne particulate matter (PM2.5) of NH3, whose high levels are responsible for thousands of premature deaths in the same region. The AGRIMONIA dataset (https://agrimonia.net/) provides airborne pollutant concentrations, weather, agricultural emissions, livestock, and land and soil use in the years 2016–2021, harmonised at the daily scale for Lombardy. Thanks to a heteroskedastic Hidden Dynamic Geostatistical Model, the AGRIMONIA datasets is used to show that a 50% reduction in NH3 emissions in the wintertime would imply a reduction of PM2.5 by 7.0% overall (rural and urban plain) and by 12.8% in rural provinces where farming is more intensive. The GRINS project (https://grins.it/) aims to develop a national platform called “AMELIA” where hundreds of variables will be harmonised to the same resolution i.e. municipality and daily. For municipal air quality indicators data from in-situ networks, models, and satellites will be fused using advanced statistical models for large spatiotemporal data.