Benthic eutrophication often gives origin to qualitative changes in marine and estuarine ecosystems, for example the shift in primary producers, often followed by changes in species composition and trophic structure at other levels. Through time such modifications may determine a selected new trophic structure. The development of structural dynamic models will allow to simulate such changes, using goal functions to guide ecosystem behaviour and development. The selection of other species and other food web may then be accounted by a continuous optimisation of model parameters according to an ecological goal function. Exergy has been applied in structural dynamic models of shallow lakes, and appears to be one of the most promising approaches. Theoretically, exergy is assumed to become optimised during ecosystems development, and ecosystems are supposed to self organise towards a state of an optimal configuration of this property. Exergy may then constitute not only a suitable system-oriented characteristic to express natural tendencies of ecosystems evolution, but also a good ecological indicator of ecosystems health. Biodiversity is also an important characteristic of ecosystems structure, constituting a powerful and traditional concept, which was found to be suitable to test the intrinsic ecological significance of exergy. We examined the properties of exergy (exergy and specific exergy) and biodiversity (species richness and heterogeneity) along an estuarine gradient of eutrophication, testing the hypothesis that they would follow the same trends in space and time. This hypothesis was validated in a certain extent, with exergy, specific exergy and species richness decreasing as a function of increasing eutrophication, but heterogeneity responding differently. Biodiversity measurements and their interpretation appeared subjective. Exergy and specific exergy may be a suitable alternative, that could be used as goal functions in ecological models and as holistic ecological indicators of ecosystems integrity. Nevertheless, since exergy and specific exergy showed to respond differently to ecosystems seasonal dynamics, it is advisable to use both * Corresponding author. TeL/fax: + 351 39 23603; e-mail: jcmimar@cygnus.ci.uc.pt 0304-3800/97/$17.00 © 1997 Elsevier Science B.V. All rights reserved. PII S0304-3800(97)00099-9 156 J.C. Marques et al./Ecological Modelling 102 (1997) [155][156][157][158][159][160][161][162][163][164][165][166][167] complementary. The method proposed by Jorgensen et al. (1995) to estimate exergy, which takes into account the biomass of organisms and the thermodynamic information due to genes, appeared to be operational. There is nevertheless an obvious need for the determination of more accurate (discrete) weighing factors to estimate exergy from organisms biomass. We propose to explore the assumption that the dimension of active genomes, which are primarily a function of the required genetic information to build up an organism, are proportional to the relative contents of DNA in di...