Ecological Modelling 221 (2010) 1153–1161 Contents lists available at ScienceDirect Ecological Modelling journal homepage: www.elsevier.com/locate/ecolmodel Modelling periphyton in irrigation canals O. Fovet a,∗ , G. Belaud b , X. Litrico a , S. Charpentier c , C. Bertrand d , A. Dauta e , C. Hugodot f a UMR GEAU, Cemagref, 361 rue JF. Breton, BP5095, 34196 Montpellier cedex 5, France b UMR GEAU, Montpellier SupAgro/Institut de Recherche pour le Développement, 361 rue JF. Breton, BP5095, 34196 Montpellier cedex 5, France c Société du Canal de Provence et d’´ aménagement de la région provenc ¸ ale /UMR IMEP, Le Tholonet, CS 70064, 13182 Aix-En-Provence, Cedex 5, France d UMR IMEP, Facultés des Sciences St Jérôme, Boite 441, 13397 Marseille cedex 20, France e UMR Ecolab, Centre National de Recherche Scientifique, 29 Rue Jeanne Marvig, BP 24349, 31055 Toulouse cedex 4, France f Association Syndicale Autorisée du Canal de Gignac, 1 parc du Camalcé, 34 150 Gignac, France article info Article history: Received 6 August 2009 Received in revised form 8 January 2010 Accepted 14 January 2010 Available online 8 February 2010 Keywords: Periphyton Irrigation canals Model Shear stress Hydrodynamic Experimental setup abstract Periphyton developments in water distribution canals induce major disturbances for system manage- ment, such as clogging problems when fixed algae are detached. Periphyton models can be used to simulate and improve canal management. The purpose of this paper is to review the periphyton models which integrate a hydrodynamic effect, and to discuss their relevance for application in open-channels. Afterwards, a new model of periphyton detachment that integrates the hydrodynamic factor is pro- posed. An experiment in semi-real conditions is performed to compare periphyton development under four different hydrodynamic regimes. The proposed model is compared on experimental results with two existing models. The new model reproduces well the periphyton dynamics in the four canals simul- taneously. © 2010 Elsevier B.V. All rights reserved. 1. Introduction In freshwater quality modelling, algal development is mainly considered through the phytoplanktonic compartment (Elliott et al., 2001; Reynolds et al., 2001). Yet, in running waters, periphy- ton may be the dominant form of the algal compartment, and its development may be a major disturbance for open-channel dis- tribution networks (Ferreira et al., 1999; Fovet et al., 2009). Here the detachment of algal mats requires frequent operations in order to remove them from the obstructed structures (pipes, weirs, ...). Furthermore, determining the sectors subject to algal development and the quantity of attached biomass provides useful information for the network manager. This paper aims at analyzing periphy- ton growth and detachment models applicable to such contexts. Among all environmental factors influencing periphyton dynamics, hydrodynamics has been identified as one of the most explicative (Biggs and Close, 1989; Ghosh and Gaur, 1998). Two kinds of flow impact on algae development can be involved: an effect on nutrient uptake and biomass detachment due to the physical constraint. Few periphyton models take these effects into account. Most of them have been developed empirically on natural rivers (Uehlinger et al., ∗ Corresponding author. Tel.: +33 467 046 350; fax: +33 467 166 440. E-mail address: ophelie.fovet@cemagref.fr (O. Fovet). 1996; Saravia et al., 1998) or in laboratory flumes (McIntire, 1973; Labiod et al., 2007). Most physically-based modelling approaches (Asaeda and Son, 2000; Tsujimoto and Tashiro, 2004; Toda et al., 2009) represent a biomass detachment using sediment erosion models, considering that this detachment is mainly due to bed scouring. However, many artificial networks are hard-surfaced, and therefore erosion models are not applicable. A review of existing detachment models is proposed further on. To test the validity of periphyton models, experiments have been conducted on concrete flumes built within the Gignac irrigation canal, an open-channel network supplied by the Herault River (South of France). Different hydrodynamic conditions have been applied to the flumes, while algal biomass and taxonomy have been determined at different dates during the experiment. Several existing models have been tested and a new formulation for algal detachment is proposed. This formulation is based on the mean bottom shear stress, which expresses the effort exerted by the flow on its substratum. The results provided by the different models are compared to the exper- imental data. Finally, their application to periphyton development in open-channel networks is discussed. 2. Review of periphyton models with hydrodynamic effect Many models of algae development have been designed to study eutrophication issues in freshwater systems. When dealing with 0304-3800/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.ecolmodel.2010.01.002