Matskási István (szerk.): A Magyar Természettudományi Múzeum évkönyve 86. (Budapest 1994)

Kemenes, I. ; Demeter, A.: Uni- and multivariate analyses of the effects of environmental factors on the occurrence of otters (Lutra lutra) in Hungary

ANNALES HISTORICO-NATURALES MUSEI NATIONALIS HUNGARICI Tomus 86. Budapest, 1994 p. 133-138 Uni- and multivariate analyses of the effects of environmental factors on the occurrence of otters (Lutra lutra) in Hungary by I. KEMENES & A. DEMETER, Budapest KEMENES, I. & DEMETER, A. (1994): Uni- and multivariate analyses of the effects of environmental factors on the occurrence of otters (Lutra lutra) in Hungary. - Annls hist.-nat. Mus. natn. hung. 86: 133-138. Abstract - Univariate analyses of otter distribution records in Hungary (369 sites) revealed that the occurrence of this species showed moderate positive association with the density of the bankside vegetation and the depth of water, and loose association with steepness of the bank of the watercourse. A variety of other, non-scalable environmental factors correlated with one another also seemed to have an effect on the occurrence of otters. Multiple logistic regression analysis of the presence (1) or absence (0) of signs of otters as dependent variable, and water depth, steepness of the bank, density of the vegetation and the presence or absence of various disturbance factors (agricultural use of the water bank, obvious signs of pollution of the water etc.) as independent variables, was carried out. It confirmed that the density of the bank vegetation and depth of the water are indeed the two main factors that increase the probability of finding otters at a particular site. However, the effects of other environmental variables proved to be much more complex than inferred from qualitative assessments. With one figure and 3 tables. INTRODUCTION A survey of 369 sites in aquatic habitats in Hungary revealed that more than 50% of them were still inhabited by otters (Lutra lutra L.) (KEMENES 1991). As the otter is extinct over much of Europe (MASON & MACDONALD 1986), it seemed important to examine the role of a variety of environmental factors which allow the occurrence of otters in Hungary. Likewise, it was important to elucidate environmental factors which were likely to drive away otters from particular areas. In earlier studies aimed at detecting the major factors influencing the occurrence of otters, a variety of habitat characteristics were recorded and their importance investigated. Pollution of the water proved to be the most limiting factor: the presence of heavy metal ions or PCBs and other pollutants was always correlated with the absence or decline of otter populations in the study areas (MASON 1989). In non-polluted habitats a variety of other factors, such as the type of the vegetation on the bank (MACDONALD & MASON 1985) or the depth of the water (GLIMMERVEEN & OUWERKERK 1984) have influence on the distribution of otters. Therefore, the first aim of the present work was to investigate in Hungary the correlations of single environmental factors with the presence or absence of otters. On the other hand, a variety of other, non-scalable environmental factors also seemed to have effect on the occurrence of otters. These observations led us to employ a multivariate method which also took into account the relationships between a number of scalable and non-scalable environmental parameters which might influence the occurrence of otters. This method, multiple logistic regression analysis, is a technique that is becoming increasingly used in ecological applications (TREXLER & TRAVIS 1993). METHODS 1. Establishing the presence of otters in aquatic habitats During a nationwide survey between 1988 and 1990, 369 sites at ponds, lakes and rivers were visited in Hungary. A standard field survey technique, described by CHAPMAN & CHAPMAN (1982) and MACDONALD (1983) was used to establish the presence of otters at each study site. The basic unit area used during the survey was the 100