The study sites were characterised at the biological level, and biodiversity was estimated at different levels of integration.
i) A description of habitat diversity was provided, and the composition of the local communities determined in different seasons. The Diptera fauna was characterised at each study site. Bird counting were conducted in different seasons in the Smir lagoon and dam lake, at the mouth of Zoura Oued (river) and in the Kneiss area. In the Smir lagoon several stations were regularly monitored every two months throughout the year at different distances from the lagoon mouth, and the bottom macrofauna was characterised with respect to water and vegetation features. In the beach-dune system species richness was evaluated in spring and autumn using standard trapping techniques with pitfall traps along transects perpendicular to the shoreline. Samples were sorted to order level and only for isopods and coleopterans to family level and then morphotyped into recognisable taxonomic units (RTUs). Species were studied both quantitatively and qualitatively using ecological coefficients (relative abundance) and Fisher's diversity index. To evaluate the evenness of the community Shannon-Weaver index was calculated and compared to Brillouin index. For both indices the maximum values were computed and were used to obtain the evenness of the community through Pielou. Also Simpson's dominance index was considered. Percentages of similarities between zones and on the total of the localities was analysed in the two seasons through Renkonen index. The overall analysis permits comparisons between geographically distant sites and gives indications on habitat quality and on their relevance for conservation.
ii) Key species were then chosen and their potential to adapt to environmental changes estimated by comparing the variation of genetic structure and behavioural adaptation of populations from beaches subject to different impacts. The issue of biodiversity can be approached at various levels of organisation, the most basic being genetic diversity, which encompasses the variation in the genetic make-up among individuals both within a population and between populations. The intraspecific genetic variation of common talitrids dwelling in Mediterranean sand beaches has been analysed using allozyme electrophoresis. The results for the genetic structure of talitrid species have led to the estimation of the levels of genetic differentiation of North-African populations. The genetic peculiarity of this geographic group has been shown. Estimates of genetic heterozygosity and polymorphism have been provided and external factors have been suggested, which could affect the genetic variability and evolutionary trend of the North-African talitrids. The species Talitrus saltator was also studied at the DNA level, which can be more informative for inter-individual variation. A dune plant, Cakile maritima, was also chosen as key species and the genetic variation within population at the individual level was analysed through DNA techniques. These were developed for both the plant and the animal species within the project. We can conclude that we have identified techniques for the early identification of reduction in biodiversity, likely as a consequence of increased environmental impact in both key species. For the plant, functional markers have been isolated that may be used to estimate heterozigosity and polymorphism in specific genes. Moreover, for both organisms, the ISSR technique has been shown to be a good method for fast, easy to apply and inexpensive analysis of intra-population variability. The results of the analysis performed on the plant and on the animal suggest that the characterisation of single species genetic variability and the eventual reduction could be a useful biological indicator of environmental impact.
iii) Changes in animal behaviour are usually the earliest biological indicators of environmental change, and a comparison of the behaviour patterns of key species from selected beaches with those already on record provides a basis for bioassays of beach stability. These assume that population samples from stable beaches are likely to show less diverse behavioural patterns than those from beaches under threat. The rationale for such assumption is that adaptation to particular environmental features requires time (lifetime for individual adaptation, several generations for genetic adaptation). Only individuals coming from coasts relatively stable in time can adapt to the specific features of the sites of origin. Estimates of specific adaptation can thus serve as a tool for coastal stability. The tests are based on the activity rhythms and orientation of a key species, the sand-beach amphipod, Talitrus saltator. The tests require reasonable numbers of live animals and are relatively easy to perform. Moreover, the animals are unharmed, and can be returned to their environment. Activity rhythms have been analysed at the individual level using recording chambers constructed ad hoc and electronic data logger. The results are simple to analyse using software packages developed within the project. Recording equipment have been available to partners, and could be made available on request to interested students or managers. For the study of orientation on the beach, transportable chambers were constructed of easy use, and made available to partners for future monitoring at sites. New statistic methods for a multivariate analysis of angular data have been developed within the project. The application of multiple regression analysis permits to estimate the effect of several influencing factors on the orientation of a key species. Then the test can be used as a bio-assay of environmental impact.