Η παρούσα διπλωματική εργασία έχει ως στόχο την καταγραφή και την ερμηνεία των επιπτώσεων των ανθρώπινων επιδράσεων στο αβιοτικό περιβάλλον και τη βλάστηση του κάτω μέρους του Εύηνου ποταμού και την εξαγωγή συμπερασμάτων για τη διατάραξη της φυσιολογικής του ισορροπίας, με σκοπό την βέλτιστη διαχείρισή του στο μέλλον. Τέλος, στοχεύει στη δημιουργία ενός οικολογικού μοντέλου λειτουργίας του ποταμού. / The present thesis aims at as the recording and the interpretation of repercussions of human effects in the abiotic environment and the vegetation of down part of Evinos river and the export of conclusions on the perturbation of his physiologic balance, aiming at his most optimal management in the future. Finally, it aims in the creation of ecological model of operation of river.

5 Absctract Environment and demographic characteristics of host population are important factors affecting incidence and severity of plant diseases. The aim of this study was to reveal possible effects of microclimatic and edaphic characteristics and host population demographic factors on the biotrophic pathogen Puccinia cribrata, microcyclic rust parasitising clonal evergreen plant Vinca minor. I studied the ecology of this pathosystem in natural populations of V. minor in the Soča valley in the Julian Alps in Slovenia. Rugged terrain of the Soča valley with different bedrocks allowed to study pathogen distribution at topographically diverse forest and scrub sites, greatly differing in soil depth and canopy structure. P. cribrata was significantly more likely to occur in the themophilous forest with shallow calcareous soils and open canopy than in the mesophilous forest types. Its prevalence (proportion of diseased stems in population) was also highest in this type of vegetation. Most important host population factors were population size and population density positively affecting incidence and prevalence respectively. Additionally, I tested various vegetation and topographic indexes derived from remote sensing for usability in prediction models. It appears that vegetation indexes connected with canopy...