Hindcasting modelling for restoration and conservation planning: application to stream fish assemblages

Although ecosystems are increasingly threatened worldwide, the resources available to set up management actions, such as conservation or restoration, remain severely limited. Methods designed to spatially allocate conservation and restoration actions while maximizing their ecological benefits are urgently needed. Whereas conservation planning methods have been extensively developed in past decades, improvements in restoration planning are still needed to build efficient management tools. This methodological gap may be explained by the difficulties encountered when estimating non‐disturbed ecological conditions (i.e. reference conditions) which are usually required to compare past and present ecosystem states. Here, the species composition of stream fish assemblages in 607 river stretches of the Pas‐de‐Calais department in the north of France was predicted using species distribution models. Present occurrence was predicted according to four environmental variables, including two related to human‐induced disturbances (i.e. proportion of river stretch length affected by hydraulic works and local geomorphological alterations). The fish assemblages potentially present in the absence of such disturbances were then predicted using hindcasting modelling which involves artificially setting to zero the values of the disturbance variables in the models. A framework based on a multi‐faceted approach of diversity was applied to the present‐day fish assemblages and those predicted under the two non‐disturbed scenarios (i.e. restoration scenarios) to assess their suitability for management. While the theoretical restoration of the natural flow regimes was likely to result in few changes in fish assemblage composition and consequently in their multi‐faceted diversity, the restoration of natural geomorphological characteristics was predicted to reduce the taxonomic diversity but increase the functional diversity, the natural heritage importance and the socio‐economic value of the fish assemblages. This study provides environmental decision‐makers with a tool to identify precisely and simultaneously the conservation and restoration actions that have to be undertaken as a priority, by comparing the present and non‐disturbed multi‐faceted diversity indices. Copyright © 2015 John Wiley & Sons, Ltd.