Olivier Delaigue's research while affiliated with Université Paris-Saclay and other places

... extreme flood) could affect model performance and parameter estimation. Thébault et al. (2023) reported that artificially corrupting time series has little effect on model calibration over a large dataset. Nevertheless, Perrin et al. (2007) and Ayzel and Heistermann (2021) showed that modellers can remove a large proportion of data from the calibration set, such as po-tential anomalies, without compromising the estimates of the model parameters, as long as the calibration period covers dry and wet periods. ...

... The reader is asked to refer to Slater et al. (2019) for further details about the advantages of R for all the steps of the workflow and to the R Hydrology Task View (https: //cran.r-project.org/web/views/Hydrology.html, last access: 1 August 2022, Zipper et al., 2022) for a complete list of R packages linked to hydrology. The choice of hydrological modeling R packages is particularly large (see Astagneau et al., 2021, for a recent review), providing a variety of solutions adapted to the diverse problems or case studies that can be encountered. Here again, the reader is referred to Astagneau et al. (2021) for further details about the packages and models available. ...

... Finally, the application of the chosen DA algorithms into an Open Loop (OL) hydrological model. For instance, sequential assimilation of observations in models with the Ensemble Kalman Filter (EnKF) and Particle Filter (PF) is widely employed for probabilistic hydrologic predictions and skillful operational flood forecasting systems, such as in [29][30][31][32]. ...

... We used numerical modeling to aid in decoupling the impact of these processes on streamflow. Discharge measurements were simulated for each of the 12 sub-watersheds for hydrological year (water years), or hereinafter referred to as years 2000-2022 using the R packages airGR which implements the GR5H hourly hydrological model (Astagneau et al., 2020;Coron et al., 2017Coron et al., , 2020Ficchì et al., 2019). All analyses were performed using R Statistical Software (v4.1.2; ...

... Ce projet visait à comparer et évaluer la capacité de plusieurs modèles hydrologiques à produire des prévisions d'étiages utiles pour la prise de décision en temps réel. Il a conduit à la mise au point d'une plateforme multi-modèle de prévision des bas débits pour répondre aux besoins des utilisateurs identifiés, parmi lesquels les directions régionales de l'environnement (DREAL), les directions départementales des territoires (DDT), les syndicats de bassins, etc. (Tilmant et al., 2020) Comment prévoit-on les étiages ? La prévision des débits en rivière est dépendante des caractéristiques de la réaction hydrologique du bassin versant aux conditions météorologiques futures, et de l'état de saturation du bassin à l'instant de la prévision. ...

... PREMHYCE platform https://webgr.inrae.fr/projets/projets-en-cours/onema-premhyce/; Nicolle et al., 2020). Hence the need for continuous improvement of these models. ...

... The rapid development of eDNA in recent years offers particular promise to help overcome the lack of biological data, but will require investment toward application for specific indicators [49] and implementation at large-scales [50]. For example, although eDNA analyses of aquatic diversity is improving with community approaches such as universal markers or meta-barcoding-and with comparable or greater sensitivity than traditional methods [51]-considerable effort is yet needed to build reference sequences and the genetic databases that these approaches rely on [52]. For these reasons, the potential for other mechanisms to bolster global datasets of aquatic diversity should not be ignored, including processing of museum specimens [53], fostering citizen science observation networks and platforms [54], and collating data from national biomonitoring programs [24]. ...

... This moves toward reproducible hydrology (Hall et al., 2022) and leads to the emergence of experiments of virtual laboratories (Ceola et al., 2015;Tarboton et al., 2014), opensource software (Coron et al., 2017;Slater et al., 2019), and open datasets (Addor et al., 2017;Irving et al., 2018). What about open hydrological teaching? ...

... http://software-carpentry. org) is properly embedded within hydrological science curriculums, so that future generations of hydrologists have the skills to build readable, version controlled and unit-tested software (McConnell, 2004), allowing them to engage more fully in an open scientific community by reproducing and reusing each other's research outputs." This moves toward reproducible hydrology (Hall et al., 2022) and leads to the emergence of experiments of virtual laboratories (Ceola et al., 2015;Tarboton et al., 2014), open-source software (Coron et al., 2017;Slater et al., 2019), and open datasets (Addor et al., 2017;Irving et al., 2018). What about open hydrological teaching? ...

... Two models from the 'airGR' family (Coron et al. 2017) were selected for use in the eFLaG multi-model ensemble: GR4J and GR6J, with the number identifying how many parameters are available to calibrate. Simple to understand and relatively easy to calibrate, they have previously been applied on numerous occasions to simulate the full range of river flows (e.g., Harrigan et al. 2018, Anglian Water Drought Plan, 2021. ...