Many animal species are capable of the cognitive mapping of environments depending on the changing conditions and availability of resources (Peters, 1978). Animals create these cognitive maps within their home range (hereafter HR), a relatively confined area where individuals perform their day-to-day activities (Powell, 2000). HRs are characterized according to size, shape, structure (Kenward, 2001) and particular time intervals (Harris et al., 1990; Powell, 2000; Laver and Kelly, 2008). A knowledge of animal HRs and movement patterns is fundamental for understanding the underlying mechanisms, such as survival, habitat use and dispersion (Delgado et al., 2009).

For any individual to survive and reproduce and thus increase its fitness, it must respond to changes in environmental conditions. These adaptive responses likewise affect animal spatial behaviour (Powell and Mitchell, 2012). Consequently, HR formation in the majority of vertebrates is attributed to different intrinsic and/or extrinsic factors, the most essential of which are food supply, predation pressure and conspecific density (McLoughlin and Ferguson, 2000). For example, HRs usually decrease with increasing food availability in many animal species (Desy et al., 1990; Broughton and Dickman, 1991; Akbar and Gorman, 1993; Kouba et al., 2017).

Burt (1943) argued that HRs do not apply to young adolescents because they are only wandering in search of a home region. Nonetheless, young animals also obviously use areas which are bounded to a lesser or greater extent, and in many studies, authors consider juveniles’ location of occurrence as “home ranges” (e.g., Belthoff et al., 1993; Kouba et al., 2013; Pedersen et al., 2013). However, it is important to stress that while adults choose their living space based on internal and external reasons, the HRs of offspring are, at least at the beginning of their lives, assigned to them by their parents. This fact applies especially to the fledglings of birds of prey that typically stay within the natal area because they depend on parental food provisioning after leaving the nest until they achieve independence and start natal dispersion (Newton, 1979; Kenward et al., 1993).

The phase between leaving the nest and reaching independence (the post-fledging dependence period – hereafter PFDP) is crucial due to the high mortality rate among young birds (Bustamante, 1995; Todd, 2001; Delgado et al., 2009). Given this fact, a broader knowledge regarding the spatial behaviour of birds of prey fledglings could help manage their conservation. Nevertheless, the movement patterns of juveniles have received little academic attention (see, however, Bustamante, 1995; Delgado et al., 2009; Kouba et al., 2013). Previous studies focusing on birds of prey juveniles have described only their HR sizes and/or habitat use (van Riper and van Wagtendonk, 2006; Mrykalo et al., 2007; Pedersen et al., 2013), but this research did not address the variables that determine the size of their HRs. Owl fledglings are known to extend their distance from the nest and occupy larger HRs with increasing age, most probably because of their improved mobility and progressive maturity (Belthoff et al., 1993; Penteriani et al., 2005; Kouba et al., 2013; Pedersen et al., 2013).

Our model species, the Tengmalm’s owl (Aegolius funereus), is a small, nocturnal predatory bird inhabiting the coniferous forests of the Holarctic zone (König and Weick, 2008). The diet of Tengmalm’s owls consists mainly of small rodents, especially voles of the genera Microtus and Myodes (Korpimäki and Hakkarainen, 2012), in addition to alternative prey such as shrews of the genus Sorex and small forest birds (Korpimäki, 1988; Korpimäki and Hakkarainen, 2012). Prey abundance is a crucial factor regarding Tengmalm’s owls’ foraging, spatial behaviour and reproduction. For instance, egg-laying dates are earlier and clutch sizes and overall breeding success are higher in good-food years compared to poor ones (Korpimäki and Hakkarainen, 1991; Hakkarainen et al., 2003; Korpimäki and Hakkarainen, 2012). Chicks hatch asynchronously at two-day intervals, and there are therefore age and size differences between offspring (Korpimäki, 1981; Valkama et al., 2002). Owlets usually fledge at one-day intervals in the same order as they hatched (Kouba et al., 2015).

To date, the only study that has analysed the HRs of Tengmalms’ owl fledglings during the PFDP was conducted in the Ore Mts., Czechia (Kouba et al., 2013), and the data presented in this research (from 2010 and 2011) was also used in the current study. The mean HR size of juveniles during PFDP was considerably smaller during the rich food year than in the poor food year (30.3 vs 57.7 ha); the PFDP lasted 45 and 57 days, respectively, and siblings’ nest boxes were usually located on the border of their HRs (Kouba et al., 2013). The authors further suggested that Tengmalm’s owl males probably participated in forming the HRs of juveniles because they may lead their offspring towards the best hunting area/s to decrease prey delivery distance and energy expended on flying.

Although several studies on the HR sizes of birds of prey fledglings throughout the PFDP testing single variables (e.g., habitat use) have already been published, the current study is unique because, as far as we are aware, no studies to date have tested several different variables in combination or attempted to identify which factors are decisive in determining the size of HRs in owl or diurnal raptor fledglings. Therefore, the main aims of our study were to determine (1) the average HR size of Tengmalm’s owl fledglings during nocturnal activity and diurnal roosting throughout the PFDP. In particular, we aimed to determine which (2) biotic and abiotic factors (e.g., prey abundance, weather conditions, latitude and altitude) as well as (3) individual characteristics (e.g., order of hatching, number of siblings and PFDP duration) can affect their sizes. In addition, we compared (4) our results with those of two study/geographical areas, i.e., in Central and North Europe, regions with both long and short nocturnal activity periods.

We expected to find marked differences in both the nocturnal (i) and diurnal (ii) HR sizes of fledglings between the study sites in Czechia and Finland because of entirely different habitats and environments (e.g., long vs short nights). Furthermore, we predicted that the HR sizes of juveniles will increase with (iii) decreasing prey abundance due to males encouraging fledglings to move farther from the nests in order to shorten prey delivery distance (Hakkarainen et al., 2003; Kouba et al., 2013). In addition, we expected the HRs will also increase in line with (iv) increasing duration of the PFDP because fledglings will generally move further from the nest in this period. Similarly, we predicted that (v) increasing hatching/fledging order within the sibling group will be a relevant factor because the last individuals to fledge are usually also the last to reach independence and will therefore be forced to move furthest from the nest (Kouba et al., 2013).

A total of 138 nestlings from 43 nests equipped with radio transmitters fledged successfully during the six study years: 78 individuals from 19 broods in the Czech study site and 60 individuals from 24 broods in the Finnish site ( Table 1 ; Supplementary Table S1 ). Throughout the PFDP until dispersal, 91 fledglings (58 in Czechia and 33 in Finland) were monitored successfully; the remaining 44 individuals died, two were excluded due to transmitter failure and one disappeared. As a result, HRs could only be established for these 91 juveniles. Nocturnal activity HRs were calculated for 71 fledglings, diurnal roosting HRs for 63 fledglings, and overall HRs for 43 individuals. The HRs were based on 5523 locations (2852 nocturnal and 2671 diurnal). The mean size of nocturnal activity HRs during the PFDP for Tengmalm’s owl fledglings (n = 71) calculated using 95% IID KDE was 63.7 ± 43.9 ha (± SD; median = 53.7 ha), and that of diurnal roosting HRs (n = 63) was 52.0 ± 46.1 ha (median = 32.3 ha). The mean size of overall HRs (n = 43) calculated using 95% IID KDE was 60.8 ± 44.3 ha (median = 55.9 ha; Table 1 ; Supplementary Table S1 ). The nocturnal activity HRs of siblings (n = 18 nests where two or more brood members were monitored until independence) overlapped by 62 ± 22% (mean ± SD) and the diurnal roosting HRs of siblings (n = 15) overlapped by 49 ± 23% when calculated using 95% IID KDE method. The representation of old-growth forests within 500 m radii around nest boxes of the studied broods was ca. two times higher in Finland (16 ± 7%; mean ± SD) than in Czechia (7 ± 11%) (see details in Table 2 ).

The HR sizes of fledglings with (a) nocturnal activity and those with (c) overall ones based on pooled nocturnal and diurnal data sets throughout the PFDP from Czechia and Finland did not differ from each other when calculated using either IID KDE or MCP methods (nocturnal HRs: IID KDE: t_57.7 = 0.78, p = 0.44, MCP: t_53.5 = 0.41, p = 0.68; overall HRs: IID KDE: t_26.2 = 1.22, p = 0.23, MCP: t_20.1 = 0.24, p = 0.81). However, the HR sizes of (b) diurnal roosting were approximately two times smaller in Czechia than in Finland for both methods (diurnal HRs: IID KDE: t_48.8 = 3.11, p = 0.003, MCP: t_47.6 = 2.66, p = 0.001). The fledglings’ HR sizes of (a) nocturnal activity and (b) diurnal roosting throughout the PFDP did not differ from each other using either IID KDE or MCP calculations when the data sets from both study sites were pooled (IID KDE: t_125.2 = -1.91, p = 0.06; MCP: t₁₂₇ = -2.74, p = 0.07). Therefore, we subsequently pooled data from both study areas for further analysis, but we examined the HRs recorded during nocturnal activity and diurnal roosting separately. Lastly, the fledglings’ HR sizes of (a) nocturnal activity and (b) diurnal roosting established using IID KDE differed significantly from those calculated by the MCP method (t_70 = 16.98, p < 0.001), with the HRs calculated by IID KDE being almost two times greater than those established using the MCP method.

However, the fixed effects explaining the square-root transformed HR size of both (a) nocturnal activity and (b) diurnal roosting (models a1 and a2; b1 and b2) calculated using IID KDE and MCP were virtually identical, and therefore only the results of the IID KDE calculations are presented here ( Table 3 ; for the results of the MCP calculations, see Supplementary Table S3 ). Finally, the results of models c1 and c2 regarding square-root transformed fledglings’ overall HR size established using both IID KDE and MCP were virtually identical to those of models a1 and a2 regarding the nocturnal activity. As a result, both these models (c1 and c2) are presented in Supplementary Table S3 only.

Supplementary Table S4 (models a, b, c) shows the best candidate models ranked using the AICc, AIC and HQIC information criteria. All of the information criteria rated the same five GLMM as most suitable, and therefore only the AICc results are presented here. Table 3 and Supplementary Table S3 show the five best candidate models for the three dependent variables (models a, b, c) tested, according to the AICc information criteria, including ΔAICc, Akaike weights and AICc Odds sorted by AICc (from the lowest to the highest value). The information criteria and rankings for the five best candidate models (with the most suitable model first) and Null models are shown in Supplementary Table S4 . Uniform rankings across the three information criteria make a convincing argument that the models with cross-level interactions are the five best models. The relative information loss from the comparisons between Null and best models confirms that the best and other top-ranked models are valid ( Supplementary Table S4 ). The estimate, standard error and 95% confidence interval of the fixed effects in the best and equivalent models for the three dependent variables tested are shown in Supplementary Table S5 . Finally, evaluations of all fixed effects present in the best and equivalent models according to the variable importance analysis are presented in Supplementary Table S6 .

The first major finding of our study is the comparable-sized HRs of fledglings during nocturnal activity found in both study areas, a finding which was in disagreement with prediction i. This result is important because it shows that juvenile Tengmalm’s owls occupied approximately the same HRs in their nocturnal activities throughout the PFDP regardless of the environmental conditions, i.e., geographical location and thus night length, altitude and weather (temperature and precipitation), but also habitat conditions, such as forest stand age, structure and composition. Our two study sites differ considerably in all these conditions. The Central European site is mountainous, and thus higher in altitude; it is also colder and more affluent in precipitation than the North European site. Regarding the habitat conditions, the forests on the Czech site are formed by small stands of relatively old and dense Norway spruce and extensive areas of highly loose stands of non-native young blue spruce. In comparison, the forests of the Finnish site are heavily managed, with dense middle- and old-age forest patches scattered between open clear-cut areas and dense young forests (see Kouba et al., 2013, 2023 for further details). Evidently, environmental factors are not decisive in determining the size of PFDP nocturnal activity HRs in Tengmalm’s owl fledglings. The similar nocturnal activity HR sizes in both study areas might be connected with the species’ life history and could be partly innate along with other behaviours which the fledglings exhibit during the PFDP.

In the case of diurnal roosting HRs, the results confirmed prediction ii, with fledglings in the Czech site having smaller ranges than their counterparts from Finland. Although speculative, we are convinced that this finding was due to the different habitat composition in the Czech site (Ore Mts.) with the juveniles repeatedly returning to small and dense mature Norway spruce forest patches for diurnal roosting. This explanation corresponds exactly to the roosting behaviour of Tengmalm’s owl males during breeding in the Ore Mts. and also in the Jizera Mts. in Czechia (Kouba and Tomášek, 2018; M. Kouba, unpublished data), and percentage representations of different habitat classes around nest boxes of the studied broods. These forest patches are probably ideal for diurnal roosting as they offer suitable hiding places against, for example, predators and mobbing by passerines.

Furthermore, the fact that the size of fledglings’ roosting ranges differed from the nocturnal ranges only in Czechia and not in Finland suggested that the reason for this must be specific to the Czech site. In Finland, we did not record fledglings using the same roost sites repeatedly, and this has not been observed in male Tengmalm’s owls in several other studies (Bondrup-Nielsen, 1978; Palmer, 1986; Jacobsen and Sonerud, 1987; M. Kouba and E. Korpimäki, unpublished data). The facts mentioned above further suggest the influence of the specific mosaic forest stand structure in the Ore Mts. which previously suffered severe damage from air pollution. As a result of this, the Czech study area did not offer a sufficient number of roosting sites, the remaining examples of which are now concentrated in relatively small sections of the oldest and densest forest patches. Similar results have been found for other species when the magnitude of HRs varies among populations of the same species due to variations in the physical structure of habitats (McLoughlin and Ferguson, 2000).

The next important finding of this study is that the sizes of overall fledglings’ HRs throughout the PFDP based on pooled nocturnal and diurnal locations were crucially determined by fixes recorded during nocturnal activity; thus, they were also affected by the same exact fixed effects ( Supplementary Table S5 ). The results further confirm that forest age, structure and composition are not fundamental regarding the overall and nocturnal HR sizes during the PFDP in Tengmalm’s owl fledglings, a finding which is, however, in direct contrast to results regarding juvenile mortality during this period (Kouba et al., 2023). The size of agricultural fields and old-growth forests were found to have adverse effects on the survival of Tengmalm’s owl fledglings (Kouba et al., 2023).

The calculations of overlaps between the HRs of siblings showed a very high ratio of the joint use of areas during both the nocturnal activity and diurnal roosting periods. We further found that the nest box ID explained the largest part of fledglings’ HR size variability, thus showing that siblings of a single brood had similar-sized HRs. Both of these findings clearly show that the Tengmalm’s owl sibling groups did not dissolve before the very end of the PFDP and that there was no brood division throughout the PFDP, a finding which was otherwise apparent during our daily and nightly direct field observations throughout the PFDP. No early family break-up case was observed, in contrast to those recorded by Eldegard and Sonerud, (2010, 2012), and there was no evidence of brood division as has been found in the flammulated owl (Otus flammeolus) (Linkhart and Reynolds, 1987). This finding, however, is unsurprising and probably caused by the fact that the young are cared for throughout the PFDP exclusively or almost exclusively by males (Eldegard and Sonerud, 2010, 2012).

The final main finding shows that prey abundance is not a decisive factor in determining the size of nocturnal activity and the diurnal roosting HRs of Tengmalm’s owl fledglings throughout the PFDP; thus, prediction iii has been rejected. However, a significant difference in fledglings’ HR sizes was already reported between 2010 and 2011 by Kouba et al. (2013), two years which were extremely different regarding prey abundance (18.5 times difference). When all six seasons were analysed together, the effect of prey abundance disappeared. However, HRs increased with the longer duration of PFDP (see below), and its length has a strong negative dependence on prey abundance in both the Czech and Finnish sites (Kouba et al., 2013; M. Kouba and E. Korpimäki, unpublished data). Both these explanatory variables, the PFDP duration and prey abundance, were significantly intercorrelated and were, therefore never used together in the same model. We are therefore convinced that PFDP duration was decisive in this case and that prey abundance did not directly affect the HR sizes of Tengmalm’s owl fledglings during the PFDP. This may also be associated with the fact that owlets do not hunt prey themselves but are supplied by the male parents throughout the post-fledging phase (Eldegard and Sonerud, 2010, 2012).

In line with prediction iv, the nocturnal and diurnal HRs increased with the advancing duration of PFDP. This finding reflects the fact that Tengmalm’s owl fledglings usually gradually moved away from the nest and reached the furthest distance at the end of the PFDP (Kouba et al., 2013; this study). Similar movement patterns and gradual HR increases have also been observed during the PFDP of juveniles of other owl species, for instance, in the eastern screech owl (Megascops asio) by Belthoff et al. (1993), the Eurasian eagle owl (Bubo bubo) by Penteriani et al. (2005) and Delgado et al. (2009), and the little owl (Athene noctua) by Pedersen et al. (2013). It is logical that as the fledglings get older, their flight and cognitive abilities improve and their perceptual range increases as individuals become more accustomed to their natal habitat.

However, a longer distance from the nest and a longer PFDP may not necessarily mean the existence of a larger HR. Several studies found that fledglings of birds of prey moved mainly around a focal point of their HR during PFDP, usually represented by one or more adjacent trees [e.g., tawny owls (Strix aluco) (Petty and Thirgood, 1989)] or the nest box [e.g., eastern screech owls (Belthoff and Ritchison, 1989), bald eagles (Haliaeetus leucocephalus) (Wood et al., 1998), northern goshawks (Accipiter gentilis laingi) (McClaren et al., 2005), and barn owls (Tyto alba) (Almasi et al., 2021; Cain et al., 2023)]. However, this pattern was not observed in Tengmalm’s owl fledglings in the Czech or Finnish study site. Moreover, the nest boxes of fledglings involved in our study were often located at the very edge of their HRs. Thus, as suggested earlier (Kouba et al., 2013), our results show that differences regarding juvenile behaviour and movement patterns throughout the PFDP are connected with life-history strategies in parental care and the body size of a given species. The differences recorded are most probably affected by the degree of territoriality, which is usually stronger in larger species of owls. These species also commonly have permanent territories defended by a long-term couple, and both partners care for their offspring (Mikkola, 1983; Korpimäki, 1992). In contrast, territorial behaviour is low or absent among Tengmalm’s owls (Korpimäki, 1981), at least during the PFDP, and pairs are only formed for individual breeding attempts (Korpimäki, 1989), with the male parent taking care of the young alone in the majority of cases (Hakkarainen and Korpimaki, 1994; Eldegard and Sonerud, 2009, 2012; Korpimäki and Hakkarainen, 2012).

In agreement with prediction v, nocturnal activity and diurnal roosting HRs during the PFDP increased with the increasing hatching order within a sibling group. The fledging sequence within a brood closely follows the hatching order in our study species (Kouba et al., 2015), and the oldest individuals who hatched or fledged first were usually also the first to reach independence (Kouba et al., 2013). Therefore, it is expected that the youngest siblings had the most extensive HRs in accordance with the above-described movement patterns and generally moved further from the nest than their older siblings (Kouba et al., 2013). In contrast, no such pattern has been observed in grassland passerines (Suedkamp Wells et al., 2008), in which hatching/fledging order did not affect their HR sizes, but this difference is probably caused by the occurrence of synchronous hatching in these species (Long et al., 1965; Roseberry and Klimstra, 1970) compared to the asynchronously hatched Tengmalm’s owls (Valkama et al., 2002).

We found fledglings’ nocturnal activity HRs throughout the PFDP to be 64 and 37 ha on average, calculated by IID KDE and MCP, respectively, which is up to five times less compared to the mean hunting HR of Tengmalm’s owl males during breeding (Sonerud et al., 1986; Jacobsen and Sonerud, 1987; Sorbi, 2003; Santangeli et al., 2012; Kouba et al., 2017). The difference could possibly be much greater because the fledglings’ ranges were based on locations collected daily for up to two months, and males’ radio-tracking usually only lasted for several nights. Thus, as a rule, Tengmalm’s owl fledglings spent the whole PFDP in highly restricted areas even though they gradually moved away from the nest in both the Czech and Finnish study sites. For example, we recorded that one group of four siblings spent the whole first month (2/3 of their PFDP) in an isolated old and dense spruce forest patch with an area of only 1 hectare. Several other sibling groups spent the entire PFDP in less than 10 ha HRs.

We suggest that small nightly/overall HRs with low requirements for forest age, structure and composition offer an excellent possibility for conservation management of Tengmalm’s owl fledglings, an essential element of any population viability in the long term (Begon et al., 2006). Although a more detailed analysis of habitat composition and structure of both the nocturnal activity and diurnal roosting HRs will follow on from the current study, we can already speculate about the most critical forest stands that need to be present in an area to ensure the successful PFDP survival of young Tengmalm’s owls. The presence of either young or, much more preferably, older patches of the densest forest appear to be crucial, because only these can offer the necessary protection against avian and mammalian enemies of Tengmalm’s owls (Mikkola, 1983; Hakkarainen and Korpimäki, 1996; Hayward, 1997).

Tengmalm’s owl fledglings occupied similar-sized HRs within their nocturnal activities during PFDP in both the Central and North European study areas regardless of contrasting environmental conditions, such as night length, altitude and weather, and also habitat conditions, such as forest stand age, structure and composition. Therefore, we conclude that these ecological factors are not decisive in determining the nocturnal activity HR sizes of Tengmalm’s owl fledglings. Thus, along with the findings that diurnal HRs always occurred within nocturnal HRs and were two times smaller in the Czech site formerly damaged by the air-pollution calamity, any Tengmalm’s owl conservation programs should be aimed at preserving particularly suitable roosting habitats. A management approach focused on preserving the densest and preferably oldest forest stands within areas of the study species occurrence could fundamentally benefit the population of Tengmalm’s owls and, in the context of umbrella species, also other forest-dwelling species in the Holarctic region.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The animal study was approved by Owls in Czechia were trapped, handled, blood sampled, and tagged under permits No. 530/758 R/08-Abt/UL, 35016/02-OOP/8751/02, and 173/049/ZPZ/2015/ZD-838 from the Ministry of the Environment of the Czech Republic. They were ringed under the Ringing Centre of the National Museum in Prague, permit No. 329. Fledglings in Finland were tagged and radio-tracked under the approval of the Centre for Economic Development, Transport and the Environment (Varsinais-Suomen Elinkeino-, Liikenne- ja Ympäristökeskus: permit No. VARELY/1389/2018), ringed under the ringing licence of the Finish Museum of Natural History (licence No. 524), and blood sampled under the approval of the Animal Experiment Committee of the State Provincial Office (Etelä-Suomen aluehallintovirasto ESAVI; permit No. ESAVI/3021/04.10.07/2017). All efforts were made to minimize suffering. The study was conducted in accordance with the local legislation and institutional requirements.

SSS: Investigation, Writing – original draft, Writing – review & editing. MK: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Validation, Writing – original draft, Writing – review & editing. MŠ: Formal analysis, Methodology, Visualization, Writing – original draft, Writing – review & editing. FT: Investigation, Writing – review & editing. TB: Investigation, Writing – review & editing. EK: Methodology, Supervision, Writing – review & editing.