BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//talks.staging.osgeo.org//flowpath-2025//speaker//TGSBLC
BEGIN:VTIMEZONE
TZID:CET
BEGIN:STANDARD
DTSTART:20001029T040000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10
TZNAME:CET
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20000326T030000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3
TZNAME:CEST
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:pretalx-flowpath-2025-UDDGHD@talks.staging.osgeo.org
DTSTART;TZID=CET:20250611T184000
DTEND;TZID=CET:20250611T185000
DESCRIPTION:Deep learning paradigms are widespread approaches for manifold 
 environmental applications. Their strength consists in the possibility of 
 processing massive quantities of data as well as performing complex tasks\
 , with unsupervised procedures. The use of deep learners\, based on neural
  computing\, proved its effectiveness for data classification\, data gener
 ation\, time series prediction\, etc.. In particular\, recurrent neural ne
 tworks are suitable for time series prediction. Here\, recurrent neural ne
 tworks are used to model groundwater levels of the large shallow porous aq
 uifer of Brindisi. This is an aquifer with a catchment of approximately 10
 00 km2 located in the upper east Salento peninsula\, in southeast Italy. I
 t is hosted by the shallow Quaternary deposits consisting of weakly cement
 ed sands\, which diffusively outcrops in its catchment. This aquifer is re
 charged by local rainfalls\, which do not significantly change across the 
 catchment\, in terms of volumes and time distribution. This shallow aquife
 r was monitored by the former National Hydrographic Bureau\, whereas piezo
 metric wells were used to acquire groundwater level measurements with a fr
 equency of one measure every three days. However\, these time series\, ava
 ilable for 6 wells between 1952 and 2002\, are not complete\, since severa
 l gaps\, i.e. missing data exist. These time series\, together with the ti
 me series of rainfall data are used to train a recurrent neural network\, 
 which is able to predict average monthly groundwater levels as well as rec
 onstructing missing data\, whereas time series are incomplete. Differently
  from machine learning approaches\, multiple well groundwater time series\
 , as well as multiple rain gauge time series will be used together\, in or
 der to train the deep learner.
DTSTAMP:20260427T221726Z
LOCATION:Room R3
SUMMARY:Groundwater level prediction based on deep learning - Lorenzo Di Ta
 ranto
URL:https://talks.staging.osgeo.org/flowpath-2025/talk/UDDGHD/
END:VEVENT
END:VCALENDAR