Kaszab Zoltán (szerk.): A Magyar Természettudományi Múzeum évkönyve 75. (Budapest 1983)
Rajczy, M. ; Padisák, J.: DIVDROP analysis - a new method for the interpretation of species importance in diversity changes
groups according to their DIVDROP effect. There are two species among them which had an equal, overall effect through the 30 days. Further two species had a nearly continuous decreasing importance, and nine species had an increase effect on diversity right after the storm. Representatives for these three cases are given in Figs IB, C, D. Only two species: Aphanizomenon flos-aquae RALFS ex BORN, et FLAH. f. klebahnii ELENK. and Coelosphaerium kuetzingicnum NÄG. have a frequency measure of 5. Their removal from the data-set usually caused a great change (mainly increase) in diversity (Fig. IE, F). Shannon diversity (Fig. 2) shows a roughly constant decrease with a maximum around 9th day of the study period. This trend cannot be due to any decrease in the number of species coexisting day to day, for the number of species slightly but continuously increased during the 30 consecutive days (Fig. 3). Most of the species above e s = 0.05 had positive effect on diversity so they also could not be repsonsible for such a trend. There were only three species (Figs 1A, E, F) which might be suspected to have caused this trend. Nitsehia sp. 1. does not seem likely to have been responsible for the trend because we could observe only two days when it had negative effect on diversity. In Fig 4A, B the DIVDROP effects of Aphanizomenon flos-aquae f. klebahnii and Coelosphaerium kuetzingicnum are plotted against Shannon diversity. As it can fairly well be seen, the rise in diversity (the local maximum around 3 August) might have been caused by a decreasing negative effect of Coelosphaerium kuetzingicnum. After this slight peak a continuous decrease in diversity can be observed. First, on 6—8 August Aphanizomenon flos-aquae f. klebahnii was the species which was responsible for that, then, between 9 and 14 August Coelosphaerium kuetzingianum. So the trend in diversity can satisfactory be explained through the patterns of these two species till 14 August. The local minimum of diversity around 16 August fits well to that of in number of species (cf. Fig. 3). The following decrease in diversity could not be explained by any single effect, it might have been an additive result of many small effects, e. g. the pattern in the number of species or the decreasing positive effect of less important species (see the Cosmarium bioculatum like pattern, Fig. 1). So the storm which broke out on 9 August had a profound effect on the phytoplankton community. Before the storm Aphanizomenon flos-aque f. klebahnii began to be overwhelmingly predominant, then a crash occured correlated with the storm and the community became predominated by another species of blue-green algae. The eifect of the storm is also obvious considering the importance patterns of the less dominant species. The terrestrial example 20 consecutive relevés of Salvio (nutanti-nemorosae )-Festucetum rupicolae were recorded along a forest-road in N-Hungary (SZUJKÓ-LACZA & RAJCZY 1983). The trend of changes in diversity along the relevé series can be seen in Fig. 5. Relevés 1 an 2 had the highest diversity values. They were very even because had no predominant grass species — the ground was barren on many spots —, and as they belonged to an other community they were neglected from the following enumeration. Thereafter two local maxima were found to exist at relevés 4—5 and 15—16. Between these maxima diversity as a trend shows a moderate decrease. After relevé 16 this decrease becomes definite. The DIVDROP run (with the automatic selector described previously) resulted 10 plots of "reduced" diversities of the 10 selected species (total number of species was 82). First we grouped the species into two: Group A — Species having a general effect on the trend of diversity. Two species belonged to this group, Brachypodium pinnatum (L.) P. B. (Fig. 6) and Festuca pseudodalmatica KRAJ. ex DOM. (Fig. 7). Group B •— Species having no general effect on the trend. This group is divisible into three subgroups:
