prilagođeno pretraživanje po punom tekstu

ŠUMARSKI LIST 3-4/2018 str. 49     <-- 49 -->        PDF

1994; Buesching et al., 2008). In addition, many studies highlighted that the cover of dead wood was an important environmental factor for the bank vole’s microhabitat selection, spatial distribution and abundance (Miklós & Žiak, 2002; Lešo et al., 2016; Zwolak et al., 2016). On the other hand, several studies focused on the survey of the difference between demographical parameters to investigate the response of small mammals to forestry practices (e.g. Savola et al., 2013; Gasperini et al., 2016).
In this study, to evaluate the response of bank vole on the protected (unmanaged), managed (by thinning interventions) and reforested habitat, we estimated the population biological traits of this species. Based on open population design, we have more specifically analysed the spatial and seasonal pattern of survival and capture probability as well as habitat dependence of population size.
Material and Methods
Materijali i metode
Study area and sampling method – Područje istraživanja i metode uzorkovanja
As part of the DRAVA-INTERECO project in 2007, trapping was performed in two sample areas in Lankoci forest, Hungary (46°18´N, 17°02´E) and one sample plot in the Repaš forest in Croatia (46°10´N, 17°05´E). One of the designated Hungarian sampling sites here was an old (> 100 yr) strictly protected alder gallery forest (Paridi quadrifoliae-Alnetum) in Lankóci Forest Nature Reserve, distinguished thereinafter as ‘protected forest habitat’ (PFH-HU). This association type occurs mostly on relatively lower terrain of higher floodplain areas, mostly on alluvial forest soil. Before river regulations, areas of this forest type used to be inundated only at times of higher floods. Such forest stands today are found almost exclusively along watercourses and oxbows in flood-prevented areas, thus they have developed during the course of gallery forest or bog forest succession. The second sample area is located besides a strictly protected forest, where clear-cutting was performed in the year 2000 on a plot of more than 1 hectare. This ‘reforested habitat’ (RH-HU) has been gradually becoming covered in forest re-growth, the development of vegetation having accelerated during recent years (2004-2006) with higher precipitation. Due to the higher ground level, this forest stand was characterized by the vegetation of oak-elm-ash gallery forest (Fraxino pannonicae-Ulmetum). The third area was selected in the Repaš forest in Croatia, in a more arid, mature (< 90 yr) oak-hornbeam forest (Carpino betuli - Quercetum roboris “typicum”) stand, which differed from the other two, besides microclimatic features, in its vegetation structure (dominance of pedunculate oak and common hornbeam) and fluvisol forest soil (Vrbek et al., 2008). Due to the fact that it was subject to intensive forestry management, it was thereinafter distinguished as ‘habitat under forest management’ (HUFM-CRO).
The capture-mark-recapture (CMR) method was applied for population- and community-level monitoring (Horváth et al., 2008a; b). Based on the established monitoring protocol, a standard trapping grid of 11×11 traps was applied in each of the three habitats, with the same box-type live-traps (75×95×180 mm) positioned 10 m apart, thus the estimated demographical values were projected onto an area of 1 hectare (Horváth & Kovačič, 2007). Small mammal synchronous monitoring in the forest habitats was pursued for four sampling periods in 2007 (July, August, September and October) and a standard 5-night trapping session was carried out in every month. Based on the number of trap stations being operated in particular grids, and on the number of sampling nights, we used the entire 2007 pool of 7260 trap night data for the population dynamical evaluation of the character species. Just like the traps themselves, the trapping technique was also alike in all cases: bacon and cereals mixed with aniseed extract and vegetable oil were used as bait. The traps were checked two times a day starting at 7 am in the morning and 7 pm in the evening, with the traps being left triggered during the day. The captured individuals were tattoo-marked on their toes, this method ensuring individual identification for the entire capture history of every small mammal individual. Upon capture, we recorded the sex (also gravidity or lactation in females), age, body mass, trap number and individual code for each animal. Age was determined from body mass and external body features.
Statistical analysis – Statistička analiza
Three capture parameters were specified based on the CMR method: total number of animals (ni) (Table 1), number of recaptures (ri) and number of known individuals (mi). The difference in the success of the recapture we tested based on ratio of recapture (rr% = ri / ni × 100). Based on daily data, these parameters were investigated by variance analysis, comparing three habitats and four months considering each area. Firstly, we examined variables for normality using