Prioritisation of pesticides and target organ systems for dietary cumulative risk assessment based on the 2019-2021 monitoring cycle.

Aiming at accelerating the implementation of cumulative risk assessment to pesticide residues, this report describes a two-step prioritisation analysis, on individual pesticides and on target organ systems, that allows to identify (i) low-priority substances expected to have a marginal contribution to cumulative risk, and (ii) high priority organ systems to be addressed in future cumulative risk assessments. The analysis encompassed 350 substances and 36 raw primary commodities of plant origin surveyed in the monitoring cycle 2019-2021, carried out in 30 population groups, covering 3 age classes, and 17 EU countries. Probabilistic exposure calculations, for chronic and acute effects, were executed on the occurrence and consumption data by a two-dimensional procedure, modelling variability and uncertainty.

Air pollutant emissions from global food systems are responsible for environmental impacts, crop losses and mortality.

Food systems are important contributors to global emissions of air pollutants. Here, building on the EDGAR-FOOD database of greenhouse gas emissions, we estimate major air pollutant compounds emitted by different stages of the food system, at country level, during the past 50 years, resulting from food production, processing, packaging, transport, retail, consumption and disposal. Air pollutant estimates from food systems include total nitrogen and its components (NO, NH and NO), SO, CO, non-methane volatile organic compounds (NMVOC) and particulate matter (PM, PM, black carbon and organic carbon).

Near-real-time global gridded daily CO2 emissions 2021.

We present a near-real-time global gridded daily CO emissions dataset (GRACED) throughout 2021. GRACED provides gridded CO emissions at a 0.1° × 0.

Prediction of severe thunderstorm events with ensemble deep learning and radar data.

The problem of nowcasting extreme weather events can be addressed by applying either numerical methods for the solution of dynamic model equations or data-driven artificial intelligence algorithms. Within this latter framework, the most used techniques rely on video prediction deep learning methods which take in input time series of radar reflectivity images to predict the next future sequence of reflectivity images, from which the predicted rainfall quantities are extrapolated. Differently from the previous works, the present paper proposes a deep learning method, exploiting videos of radar reflectivity frames as input and lightning data to realize a warning machine able to sound timely alarms of possible severe thunderstorm events.