2 Application of a Machine Learning Technique for Developing Short-Term Flood …
33
For the city of Cuenca, the current flood early warning system consists simply on
a real-time monitoring of control points located upstream zones of interest (urban
areas). When the flow at a control point exceeds a certain threshold, authorities
activate warning alarms for prevention purposes. The main disadvantage is that the
anticipation of an extreme event is limited by the transit time between the control
station and the zone of interest. The methodology proposed in this study is therefore
a further step for dealing with flood and hydrological drought events. We proposed
forecasting extreme flows with lead times up to 24 h.
Although further exploration of the RF technique is still required for improving
model performances, the models and the methodology followed by this study can
be immediately used and the results interpreted by decision-makers and politicians.
The next logic step is building up a platform for intelligently communicating results
with the people, bearing in mind that they are not necessarily familiar to computer
and engineering sciences.
References
Biau G, Scornet E (2016) A random forest guided tour. Test 25:197–227
Bontempi G., Taieb SB, Le Borgne Y-A (2012) Machine learning strategies for time series
forecasting. EBISS 62–77
Brath A, Montanari A, Toth E (2004) Analysis of the effects of different scenarios of historical data
availability on the calibration of a spatially-distributed hydrological model. J Hydrol 291:232–
253. https://doi.org/10.1016/j.jhydrol.2003.12.044
Braud I, Ayral P-A, Bouvier C, Branger F, Delrieu G, Dramais G, Le J, Leblois E, Nord G, Vandervaere J.P (2016) Advances in flash floods understanding and modelling derived from the FloodScale project in South-East France. FLOODrisk 2016—3rd Eur. Conf. Flood Risk Manag. https://
doi.org/10.1051/e3sconf/20160704005
Breiman L (2017). Classification and regression trees. Routledge.
Breiman L (2001) Random forests. Mach Learn 45:5–32. https://doi.org/10.1023/A:101093340
4324
Brouwer R, Van Ek R (2004) Integrated ecological, economic and social impact assessment of
alternative flood control policies in the Netherlands. Ecol Econ 50:1–21. https://doi.org/10.1016/
j.ecolecon.2004.01.020
Buytaert W, Célleri R, De Bièvre B, Cisneros F, Wyseure G, Deckers J, Hofstede R (2006) Human
impact on the hydrology of the Andean páramos. Earth-Science Rev 79:53–72. https://doi.org/
10.1016/j.earscirev.2006.06.002
Buytaert W, Cuesta-Camacho F, Tobón C (2011) Potential impacts of climate change on the environmental services of humid tropical alpine regions. Glob Ecol Biogeogr 20:19–33. https://doi.
org/10.1111/j.1466-8238.2010.00585.x
Chang FJ, Hwang YY (1999) A self-organization algorithm for real-time flood forecast.
Hydrol Process 13:123–138. https://doi.org/10.1002/(SICI)1099-1085(19990215)13:2%3c123::
AID-HYP701%3e3.0.CO;2-2
Cortez P (2010). Sensitivity analysis for time lag selection to forecast seasonal time series using
neural networks and support vector machines. Int. Jt. Conf. Neural Netw. (IJCNN) 2010: 1–8.
https://doi.org/10.1109/IJCNN.2010.5596890
Dawson CW, Wilby RL (2001) Hydrological modelling using artificial neural networks. Prog Phys
Geogr 25:80–108
33
For the city of Cuenca, the current flood early warning system consists simply on
a real-time monitoring of control points located upstream zones of interest (urban
areas). When the flow at a control point exceeds a certain threshold, authorities
activate warning alarms for prevention purposes. The main disadvantage is that the
anticipation of an extreme event is limited by the transit time between the control
station and the zone of interest. The methodology proposed in this study is therefore
a further step for dealing with flood and hydrological drought events. We proposed
forecasting extreme flows with lead times up to 24 h.
Although further exploration of the RF technique is still required for improving
model performances, the models and the methodology followed by this study can
be immediately used and the results interpreted by decision-makers and politicians.
The next logic step is building up a platform for intelligently communicating results
with the people, bearing in mind that they are not necessarily familiar to computer
and engineering sciences.
References
Biau G, Scornet E (2016) A random forest guided tour. Test 25:197–227
Bontempi G., Taieb SB, Le Borgne Y-A (2012) Machine learning strategies for time series
forecasting. EBISS 62–77
Brath A, Montanari A, Toth E (2004) Analysis of the effects of different scenarios of historical data
availability on the calibration of a spatially-distributed hydrological model. J Hydrol 291:232–
253. https://doi.org/10.1016/j.jhydrol.2003.12.044
Braud I, Ayral P-A, Bouvier C, Branger F, Delrieu G, Dramais G, Le J, Leblois E, Nord G, Vandervaere J.P (2016) Advances in flash floods understanding and modelling derived from the FloodScale project in South-East France. FLOODrisk 2016—3rd Eur. Conf. Flood Risk Manag. https://
doi.org/10.1051/e3sconf/20160704005
Breiman L (2017). Classification and regression trees. Routledge.
Breiman L (2001) Random forests. Mach Learn 45:5–32. https://doi.org/10.1023/A:101093340
4324
Brouwer R, Van Ek R (2004) Integrated ecological, economic and social impact assessment of
alternative flood control policies in the Netherlands. Ecol Econ 50:1–21. https://doi.org/10.1016/
j.ecolecon.2004.01.020
Buytaert W, Célleri R, De Bièvre B, Cisneros F, Wyseure G, Deckers J, Hofstede R (2006) Human
impact on the hydrology of the Andean páramos. Earth-Science Rev 79:53–72. https://doi.org/
10.1016/j.earscirev.2006.06.002
Buytaert W, Cuesta-Camacho F, Tobón C (2011) Potential impacts of climate change on the environmental services of humid tropical alpine regions. Glob Ecol Biogeogr 20:19–33. https://doi.
org/10.1111/j.1466-8238.2010.00585.x
Chang FJ, Hwang YY (1999) A self-organization algorithm for real-time flood forecast.
Hydrol Process 13:123–138. https://doi.org/10.1002/(SICI)1099-1085(19990215)13:2%3c123::
AID-HYP701%3e3.0.CO;2-2
Cortez P (2010). Sensitivity analysis for time lag selection to forecast seasonal time series using
neural networks and support vector machines. Int. Jt. Conf. Neural Netw. (IJCNN) 2010: 1–8.
https://doi.org/10.1109/IJCNN.2010.5596890
Dawson CW, Wilby RL (2001) Hydrological modelling using artificial neural networks. Prog Phys
Geogr 25:80–108