A hydrogeological study assessed the evolution of groundwater levels in a site in Milano, Italy, where a road is under construction. In the area the depth to groundwater had been historically very low – less than two meters – until the industrial development and population increase, started in the 50’, determined a high increase in groundwater withdrawal and a subsequent decrease of groundwater table. Level started to increase gradually during the 70’, reaching a height able now to interfere with the road being built. This is consistent with the observed series of both public and private groundwater withdrawal rate from 1979 to 2005 in the town of Milano, which show a reduction with linear trend. Other collected data are irrigation, a part of which reaches the groundwater table, and rainfall. This has been previously analyzed to determine the variability and distribution during the period 1917-2010. The average rainfall has decreased by 200 mm since 1985, and its distribution is compatible with a Gaussian. Therefore, a groundwater flow was built with Modflow2000 and calibrated in steady-state condition with 2011 data. Since annual precipitation is responsible for low-period groundwater increase, a first simulation under a Monte Carlo approach was set by sampling 100 times the rainfall described as a Gaussian distribution with minimum and maximum equal to the minimum and maximum rainfall recorded during the observation period. One Modflow simulation was run for each sampling, observing the resulting groundwater head in some points along the road. As a result, probability curves of exceeding groundwater levels were delivered to the road designer. Also a deterministic simulation has been run to determine a precautionary groundwater level in the next 50 years. The rainfall was set equal to the average and well withdrawal was reduced by 35% following the observed evolution. This result is highly uncertain due to the impossibility to forecast the future industrial and civil groundwater demand.
Monte Carlo forecast of groundwater level rise in Milano, Italy
FRANCANI, VINCENZO;COLOMBO, LORIS
2014-01-01
Abstract
A hydrogeological study assessed the evolution of groundwater levels in a site in Milano, Italy, where a road is under construction. In the area the depth to groundwater had been historically very low – less than two meters – until the industrial development and population increase, started in the 50’, determined a high increase in groundwater withdrawal and a subsequent decrease of groundwater table. Level started to increase gradually during the 70’, reaching a height able now to interfere with the road being built. This is consistent with the observed series of both public and private groundwater withdrawal rate from 1979 to 2005 in the town of Milano, which show a reduction with linear trend. Other collected data are irrigation, a part of which reaches the groundwater table, and rainfall. This has been previously analyzed to determine the variability and distribution during the period 1917-2010. The average rainfall has decreased by 200 mm since 1985, and its distribution is compatible with a Gaussian. Therefore, a groundwater flow was built with Modflow2000 and calibrated in steady-state condition with 2011 data. Since annual precipitation is responsible for low-period groundwater increase, a first simulation under a Monte Carlo approach was set by sampling 100 times the rainfall described as a Gaussian distribution with minimum and maximum equal to the minimum and maximum rainfall recorded during the observation period. One Modflow simulation was run for each sampling, observing the resulting groundwater head in some points along the road. As a result, probability curves of exceeding groundwater levels were delivered to the road designer. Also a deterministic simulation has been run to determine a precautionary groundwater level in the next 50 years. The rainfall was set equal to the average and well withdrawal was reduced by 35% following the observed evolution. This result is highly uncertain due to the impossibility to forecast the future industrial and civil groundwater demand.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
