The potential leaching of atrazine in the area, around Trasimeno Lake is evaluated using the PELMO model and the uncertainty associated with the simulation process is assessed. Simulation has been performed for all combinations obtained from weather (15 years), soil profile (115 different soil profiles), and pesticide properties. The amount of atrazine leached below I m depth is used as an indicator of the potential leaching. Two approaches are compared: stochastic and "megaplot". The stochastic approach is based on 400 simulations performed through a Monte Carlo generator simultaneously modifying key input data such as soil texture, organic carbon and pesticide properties. Megaplot is based on simulation of the 44 unique combinations of weather, soil and crop characteristics identified in the area. The uncertainty of the stochastic approach is about 62% but it is difficult to upscale to a large scale. The uncertainty of the megaplot approach ranges between 55 and 88% and upscaling to a large scale is easier.
Pesticide leaching potential in the Trasimeno Lake area. Assessment of uncertainty associated with the simulation process
Esposito A
2001
Abstract
The potential leaching of atrazine in the area, around Trasimeno Lake is evaluated using the PELMO model and the uncertainty associated with the simulation process is assessed. Simulation has been performed for all combinations obtained from weather (15 years), soil profile (115 different soil profiles), and pesticide properties. The amount of atrazine leached below I m depth is used as an indicator of the potential leaching. Two approaches are compared: stochastic and "megaplot". The stochastic approach is based on 400 simulations performed through a Monte Carlo generator simultaneously modifying key input data such as soil texture, organic carbon and pesticide properties. Megaplot is based on simulation of the 44 unique combinations of weather, soil and crop characteristics identified in the area. The uncertainty of the stochastic approach is about 62% but it is difficult to upscale to a large scale. The uncertainty of the megaplot approach ranges between 55 and 88% and upscaling to a large scale is easier.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.