Assessing the structural adequacy of alternative ecohydrological models using a pattern-oriented approach

The development of environmental system models is challenging because of different disciplinary philosophical approaches to uncertainty in modelling of the terrestrial hydrosphere and ecosphere. We use pattern-oriented modelling to assess model structural adequacy and to select alternative model structures within the hierarchy of a model of flood-groundwater vegetation interactions. We varied the equation structure of two key model components, flood tolerance and seasonal leaf shedding, and tested how well the model structures reproduced a set of observed patterns: (i) three species coexistence, (ii) species-specific access to groundwater, and (iii) species-specific ability to tolerate flood disturbances. We assessed (a) the role of flood frequency in biomass regulation for modelling of three coexisting species sharing the same water resources, and (b) the effect of alternative process and equation structures on the deviation of hydrological variables (transpiration, groundwater table) from average conditions. Only model structures that explicitly considered the functional relationship between flood events and biomass regulation were able to reproduce the coexistence pattern and the two secondary patterns (ii and iii). The different coexistence mechanisms had little effect on the average transpiration rates and water table depths. However, shallow and deep average groundwater tables, caused by low and high transpiration rates, were modelled more frequently with model structures that intentionally ignored species-specific phenological cycles rather than models which incorporated them. Our findings indicate that, amongst all tested model structures, the most complex one is most plausible and can explain the observed patterns in an environment controlled by the interplay between periods of water deficit and flood disturbance. It reproduced the three observed ecological patterns and enhanced the general understanding of groundwater-dependent ecosystems along ephemeral rivers. (C) 2015 Elsevier B.V. All rights reserved.