Introduction

Interactive Reservoir MAnagement Risk Assessment (IRMaRA) is a R Shiny interface providing probability of failure of flood and drought objective at key locations downstream of the 4 lakes regulating the Seine River based on the current storage of the reservoirs and the climatology of the basin.

This application highlights the outputs of the software ‘VGEST’ (J.-C. Bader 1992) (J.-C. Bader and Dorchies 2016) which optimises reservoir operation for one objective of flood alleviation or low flow replenishment. ‘VGEST’ computes the minimum (resp. maximum) storage in the reservoir to best meet the objective of maintaining the flow above a low-flow (resp. below a high-flow) threshold in the future.

Resulting time series of reservoir storage is statistically analysed in order to compute the risk rate of an objective to fail in the future considering the day of the year, the current storage state of the reservoir and the climatology used as input of ‘VGEST’.

Main features

Two visualisation modes are available:

Instant risk overview

This feature shows risk rates of failing objectives in the future considering the current storage state of the reservoir, the policy constraints and the date. The bar plot show the first ten objectives sorted by decreasing order of failure risk rate.

The storage state of the reservoir can be filled in 3 ways:

One objective focus

This feature shows a heat map of the failure risk rates for each possible storage state in function of the day of the year for one particular objective and a chosen policy constraints.

The chart also displays the theoretical filling curve.

The Seine River basin

4 reservoirs for 7 objectives at 9 monitoring stations

Four reservoir control the flows of the Seine River basin (See map below). There objectives are to maintain the flow between defined low-flow and high-flow thresholds at 9 monitoring stations (red dots on the map).

The Seine River basin (after Dorchies et al. (2014))
The Seine River basin (after Dorchies et al. (2014))
The thresholds at the monitoring stations in m3/s (after Dorchies et al. (2014))
Monitoring station Low-flow Vigilance Low-flow Alert Low-flow Reinforced alert Low-flow Crises High-flow Vigilance High-flow Regular High-flow Exceptional
Mery-sur-Seine 7.3 5.0 4 3.5 140 170 400
Nogent-sur-Seine 25.0 20.0 17 16.0 180 280 420
Gurgy (Yonne) 14.0 12.5 11 9.2 220 340 400
Courlon-sur-Yonne 23.0 16.0 13 11.0 550 700 900
Alfortville (Seine) 64.0 48.0 41 36.0 850 1200 1400
Châlons-sur-Marne 12.0 11.0 9 8.0 330 520 700
Noisiel (Marne) 32.0 23.0 20 17.0 350 500 650
Paris (Seine) 81.0 60.0 51 45.0 950 1600 2000
Water usage restrictions or damages associated with the flow thresholds
Code Type Name Description
l1 low Vigilance sensibilisation for water usage reduction
l2 low Alert restriction of 30% of non productive water withdrawals
l3 low Reinforced alert restriction of 50% of water withdrawals excluding drinking water supply
l4 low Crisis only the drinking water supply and respect for biological life are insured
h1 high Vigilance localized overflow flooding
h2 high Regular flood causing major overflows
h3 high Exceptional major flood, direct and widespread threat

Reservoir rule sets

The reservoir management respects a set of rules described in detail in Dehay (2012). These rules can be summed up with these few constraints:

  1. \(Q_{res}\) (minimum flow) at inlets and at Yonne outlet (inline reservoir)
  2. \(Q_{res}\) at outlets
  3. \(Q_{ref}\) (maximum flow) at inlets and outlets
  4. Priority to hydropower generation on Yonne reservoir (Max outflow of 16m3/s)
  5. QST Gradient flow daily limitation for Yonne reservoir (+/- 2m3/s per day)

With \(Q_{ref}\) the maximum flow allowed in the river at the inlets and outlets of the reservoirs and \(Q_{res}\) the minimum flow to leave in the river when the reservoir is filling up.

Usually, all these constraints should be applied but in critical situations the manager can choose to by-pass some of them in order to optimise the chance to satisfy objectives at downstream monitoring stations in the future.

IRMaRA proposes to assess the failure risk on the objectives with the following combinations of policy constraints:

  • All constraints (a+b+c+d+e)
  • 1, without \(Q_{ref}\) (a+b+d+e)
  • 1, without \(Q_{res}\) at outlets (a+c+d+e)
  • 3, without \(Q_{ref}\) (a+d+e)
  • 4, without hydropower priority for Yonne lake (a+e)
  • 5, without Yonne outflow variation limitation (a)

Flow dataset

‘VGEST’ uses naturalised flow daily time series as input. The naturalised flows come from a daily database covering the period 1900-2009 for all the gauging stations of interest (Hydratec 2011).

5000 years synthetic streamflows time series have been generated using the procedure proposed by Giuliani et al. (2018) and have been used in ‘VGEST’ in order to cover an extended variability of rare events than the original shorter time series.

About this application

This application is developed as part of the IN-WOP project (http://www.waterjpi.eu/joint-calls/joint-call-2018-waterworks-2017/booklet/in-wop) by the mixed research unit G-EAU (https://g-eau.fr)

Source codes

The source used for running IRMaRA (including ‘VGEST’ Pascal sources and R-packages) are available on the GIT repository of the project: https://gitlab.irstea.fr/in-wop

Licence

CC BY-NC-ND 4.0 The website http://irmara.g-eau.fr by David Dorchies, Jean-Claude Bader and Quan Dau is licensed under a Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0) License.

  • Irmara version number: 696cb76c
  • rvgest version number: b8bf2e52

Acknowledgement

The authors would like to thank the European Commission and the French National Research Agency (ANR) for funding in the frame of the collaborative international consortium IN-WOP financed under the 2018 Joint call of the WaterWorks2017 ERA-NET Cofund. This ERA-NET is an integral part of the activities developed by the Water JPI.

Water JPI Water Works 2017 European Commission 2018 Joint call

References

Bader, J.-C. 1992. “Consignes de Gestion Du Barrage à Vocation Multiple de Manantali: Détermination Des Cotes Limites à Respecter Dans La Retenue [Multiple Use Management of Manantali Dam: Determination of Limiting Storage Levels].” Hydrologie Continentale 7 (1): 3–12. http://horizon.documentation.ird.fr/exl-doc/pleins_textes/pleins_textes_4/hydrologie_cont/36986.pdf.
Bader, Jean-Claude, and D. Dorchies. 2016. “Calcul des limites de volumes d’eau à respecter dans des réservoirs implantés en parallèle sur un réseau hydrographique, pour permettre la meilleure satisfaction future d’un objectif commun de gestion à l’aval (soutien d’étiage ou laminage de crue) : logiciel VGEST - Application au cas du bassin de la Seine (Amélioration et extension de la méthode précédemment développée dans le cadre du programme Climaware).” Montpellier: IRD. http://www.documentation.ird.fr/hor/fdi:010070461.
Dehay, F. 2012. “Etude de l’impact du changement climatique sur la gestion des lacs-réservoirs de la Seine.” Other, Diplôme d’ingénieur de l’ENGEES ,Strasbourg. https://hal.inrae.fr/hal-02597326.
Dorchies, David, Guillaume Thirel, Maxime Jay-Allemand, Mathilde Chauveau, Florine Dehay, Pierre-Yves Bourgin, Charles Perrin, et al. 2014. “Climate Change Impacts on Multi-Objective Reservoir Management: Case Study on the Seine River Basin, France.” International Journal of River Basin Management 12 (3): 265–83. https://doi.org/10.1080/15715124.2013.865636.
Giuliani, M., J. D. Quinn, J. D. Herman, A. Castelletti, and P. M. Reed. 2018. “Scalable Multiobjective Control for Large-Scale Water Resources Systems Under Uncertainty.” IEEE Transactions on Control Systems Technology 26 (4): 1492–99. https://doi.org/10.1109/TCST.2017.2705162.
Hydratec. 2011. “Actualisation de La Base de Données Des Débits Journaliers ‘Naturalisés’ - Phase 2.” 26895 - LME/TL.