Can artificial intelligence improve seasonal forecasts and inform reservoir operations?
Increasingly variable hydrologic regimes combined with more frequent and intense extreme events are challenging water systems management worldwide. These trends emphasize the need of accurate medium- to long-term predictions to timely prompt anticipatory operations. In this work, we introduce the Cl...
Main Authors: | , , , , |
---|---|
Other Authors: | , , , , |
Format: | Conference Object |
Language: | English |
Published: |
2020
|
Subjects: | |
Online Access: | http://hdl.handle.net/11311/1209034 https://www.sisclima.it/wp-content/uploads/2021/04/ClimRisk2020_Book-of-abstracts.pdf |
id |
ftpolimilanoiris:oai:re.public.polimi.it:11311/1209034 |
---|---|
record_format |
openpolar |
spelling |
ftpolimilanoiris:oai:re.public.polimi.it:11311/1209034 2023-12-31T10:20:49+01:00 Can artificial intelligence improve seasonal forecasts and inform reservoir operations? M. Giuliani M. Zaniolo G. Davoli P. Block A. Castelletti Giuliani, M. Zaniolo, M. Davoli, G. Block, P. Castelletti, A. 2020 http://hdl.handle.net/11311/1209034 https://www.sisclima.it/wp-content/uploads/2021/04/ClimRisk2020_Book-of-abstracts.pdf eng eng ispartofbook:ClimRisk2020 - Book of Abstract ClimRisk2020: Time for Action! Raising the ambition of climate action in the age of global emergencies. http://hdl.handle.net/11311/1209034 https://www.sisclima.it/wp-content/uploads/2021/04/ClimRisk2020_Book-of-abstracts.pdf info:eu-repo/semantics/conferenceObject 2020 ftpolimilanoiris 2023-12-06T18:08:42Z Increasingly variable hydrologic regimes combined with more frequent and intense extreme events are challenging water systems management worldwide. These trends emphasize the need of accurate medium- to long-term predictions to timely prompt anticipatory operations. In this work, we introduce the Climate State Intelligence framework that relies in Artificial Intelligence tools to capture the concurrent state of multiple global climate signals, such as El Nino Southern Oscillation (ENSO) and the North Atlantic Oscillation (NAO), and improve seasonal forecasts. The use of multiple climate signals ensures the portability of this framework to different geographic locations, including regions where traditional teleconnections are weak. Adopting a data-driven modeling approach, we use these teleconnections and other observed preseason sea surface temperature anomalies to forecast local meteorological variables on a seasonal time scale. The resulting forecasts are subsequently transformed using a dynamic hydrologic model into streamflow predictions, which are used as additional inputs for informing water systems operations. Finally, we quantify the operational value of the hydrologic forecasts as the corresponding gain in system performance with respect to a baseline solution that does not use any forecast information. We apply the framework to the Lake Como basin, a regulated lake in northern Italy which is mainly operated for flood control and irrigation supply. Numerical results show the existence of notable teleconnection patterns dependent on both ENSO and NAO over the Alpine region, which contribute in generating skillful seasonal precipitation and hydrologic forecasts. The use of this information for conditioning the lake operations produces an average 44% improvement in system performance with respect to a baseline solution not informed by any forecast, with this gain that further increases during extreme drought episodes. Conference Object North Atlantic North Atlantic oscillation RE.PUBLIC@POLIMI - Research Publications at Politecnico di Milano |
institution |
Open Polar |
collection |
RE.PUBLIC@POLIMI - Research Publications at Politecnico di Milano |
op_collection_id |
ftpolimilanoiris |
language |
English |
description |
Increasingly variable hydrologic regimes combined with more frequent and intense extreme events are challenging water systems management worldwide. These trends emphasize the need of accurate medium- to long-term predictions to timely prompt anticipatory operations. In this work, we introduce the Climate State Intelligence framework that relies in Artificial Intelligence tools to capture the concurrent state of multiple global climate signals, such as El Nino Southern Oscillation (ENSO) and the North Atlantic Oscillation (NAO), and improve seasonal forecasts. The use of multiple climate signals ensures the portability of this framework to different geographic locations, including regions where traditional teleconnections are weak. Adopting a data-driven modeling approach, we use these teleconnections and other observed preseason sea surface temperature anomalies to forecast local meteorological variables on a seasonal time scale. The resulting forecasts are subsequently transformed using a dynamic hydrologic model into streamflow predictions, which are used as additional inputs for informing water systems operations. Finally, we quantify the operational value of the hydrologic forecasts as the corresponding gain in system performance with respect to a baseline solution that does not use any forecast information. We apply the framework to the Lake Como basin, a regulated lake in northern Italy which is mainly operated for flood control and irrigation supply. Numerical results show the existence of notable teleconnection patterns dependent on both ENSO and NAO over the Alpine region, which contribute in generating skillful seasonal precipitation and hydrologic forecasts. The use of this information for conditioning the lake operations produces an average 44% improvement in system performance with respect to a baseline solution not informed by any forecast, with this gain that further increases during extreme drought episodes. |
author2 |
Giuliani, M. Zaniolo, M. Davoli, G. Block, P. Castelletti, A. |
format |
Conference Object |
author |
M. Giuliani M. Zaniolo G. Davoli P. Block A. Castelletti |
spellingShingle |
M. Giuliani M. Zaniolo G. Davoli P. Block A. Castelletti Can artificial intelligence improve seasonal forecasts and inform reservoir operations? |
author_facet |
M. Giuliani M. Zaniolo G. Davoli P. Block A. Castelletti |
author_sort |
M. Giuliani |
title |
Can artificial intelligence improve seasonal forecasts and inform reservoir operations? |
title_short |
Can artificial intelligence improve seasonal forecasts and inform reservoir operations? |
title_full |
Can artificial intelligence improve seasonal forecasts and inform reservoir operations? |
title_fullStr |
Can artificial intelligence improve seasonal forecasts and inform reservoir operations? |
title_full_unstemmed |
Can artificial intelligence improve seasonal forecasts and inform reservoir operations? |
title_sort |
can artificial intelligence improve seasonal forecasts and inform reservoir operations? |
publishDate |
2020 |
url |
http://hdl.handle.net/11311/1209034 https://www.sisclima.it/wp-content/uploads/2021/04/ClimRisk2020_Book-of-abstracts.pdf |
genre |
North Atlantic North Atlantic oscillation |
genre_facet |
North Atlantic North Atlantic oscillation |
op_relation |
ispartofbook:ClimRisk2020 - Book of Abstract ClimRisk2020: Time for Action! Raising the ambition of climate action in the age of global emergencies. http://hdl.handle.net/11311/1209034 https://www.sisclima.it/wp-content/uploads/2021/04/ClimRisk2020_Book-of-abstracts.pdf |
_version_ |
1786831370393223168 |