The choice of season and duration of the MRR protocol applied to two saproxylic beetles proved appropriate for the study area, as the two species were rare or absent at the beginning and end of the sampling period, with most specimens recorded in the second third of the study.
For both species, the number of marked individuals decreased sharply from August to September, although the number of samples was lower in September (Table 1), as we only sampled until 9 September as weather prevented further fieldwork. Nevertheless, most of the season was undoubtedly successfully covered, as a rapid decline in abundance can be attributed to heavy rainfall (at some sites in the Julian Alps > 200 mm of precipitation on 29 and 30 August) and the drop in temperature at the end of August 2020 (Cegnar 2020). We therefore assume that not many animals were alive after our last sampling date due to the low temperatures.
Furthermore, a simple method of inspecting dead and damaged trees also proved to be efficient (see Vrezec et al. 2011). However, the proposed use of additional pitfall traps placed near the inspected trees due to the nocturnal activity of those beetles (Vrezec et al. 2011) was unsuccessful, despite reportedly increased nocturnal activity (M. asper: Hardersen et al. 2017, Vrezec et al. 2008; see also Brelih et al. 2006).
In the area, M. a. funereus seems to be more abundant than R. alpina. While the assumptions for a closed population were fulfilled for M. a. funereus and a ranking of differently probable models was possible, we did not attempt to apply such closed models to the population of R. alpina. This species is capable of moving far larger distances (Drag et al. 2011), their recapture rate was low and many dead animals were found at the end of the study period, which all indicate an open population. However, abundance estimation with Jolly-Seber open population models requires a robust dataset with plenty of recaptures (e.g. Jugovic et al. 2018, but see Schwarz 2001), which prevented the use of such model for our data.
The best model (Mth) for M. a. funereus indicated the existence of two groups of animals with different capture probabilities, which in our case are most likely represented by a small group of males (about 4% of the total population) that occupy a suitable tree or stump for a long time and defend it from other beetles (Polak 2012), and a second group with the rest of the animals that do not show such territorial behaviour. Territorial males are also much more likely to be caught (as they stay in the same place longer) than the rest of the males and females in the population. Males of both species have been found to have longer antennae than females (Pogorelec 2021). When resting in exposed, sunny places (Russo et al. 2011), territorial males of M. a. funereus indicate occupation of a tree to other conspecifics (Campanaro et al. 2017, Polak 2012), but possibly also to smaller, less competitive R. alpina, with widely spread antennae. R. alpina probably avoids M. a. funereus, as both species were found on the same beech tree in less than one fifth of the cases studied (18.5%).
The observed overall sex ratio was close to 1 : 1 in M. a. funereus, while it was obviously more male-biased in R. alpina. Besides an overall shorter lifespan of females than males in R. alpina (up to 15 and 35 days for females and males, respectively; see Campanaro et al. 2017), the low proportion of females found can also be explained by their immediate copulation and subsequent oviposition shortly after emerging from the wood, which is then followed by their rapid death (Campanaro et al. 2017, Drag et al. 2011). Stronger territorial behaviour of males (M. a. funereus: Polak 2012; R. alpina: Duelli and Wermelinger 2005) in combination with a higher dispersal ability in R. alpina than in M. a. funereus could also contribute to the fact that males of R. alpina are much more detectable, so that the male-biased ratio in this species could also be an overestimation to some extent.
As some males of M. a. funereus display territorial behaviour and stay on the same suitable tree for a long time, this is also reflected in their respective migrations, which can be numerous but short in males, while they can be much longer, although less numerous, in females. In this species, females frequently move between trees looking for mating opportunities (Hardersen et al. 2017, Polak 2012), as they may mate with several males during the season (Polak 2012), and to find a suitable tree for oviposition (Hardersen et al. 2017, Polak 2012).
According to Zimmermann et al. (2011), connectivity of 3% of the population could be sufficient to ensure effective genetic flow between neighbouring local populations. However, this estimate based on our data is extremely low for territorial M. a. funereus males (observed distances ≤ 2 m), but this extremely short distances might be related to the study design, which included only small habitat patches with sufficiently suitable woody material. The estimated dispersal abilities of non-territorial males (3% of the population should be able to disperse to a distance of 420 m; IPF fits the data quite well: R2 = 0.72) and females (up to 3 km) were much further, although it should be stressed that the IPF for females did not fit the data very well (R2 = 0.42). In general, however, M. a. funereus is expected to have low movement ability due to its inability to fly (Drovenik and Pirnat 2003, Monné 2017). Therefore, in this flightless species, fewer suitable trees are available to males, so competition for suitable trees and the chance of reproduction leading to mutual fighting might be more frequent than in R. alpina. Rossi de Gasperis et al. (2018) observed more frequent reproduction of large M. asper males, probably due to multiple victories in fights against smaller males. As both species use strong mandibles in these fights, which can lead to injuries of the opponent, it is not unexpected that we observed broken antennae in one third (33.3%) of M. a. funereus males (n = 30), while no such injuries were recorded in R. alpina males (n = 18) (personal observation). The latter could be the result of the less pronounced territoriality of R. alpina (but see Duelli and Wermelinger 2005), which is also indicated by the much lower recapture rate than in M. a. funereus and is related to the different flight ability of the two species (which is absent in M. asper). According to Drag et al. (2011), the dispersal ability of R. alpina is quite variable and depends on the spatial arrangement of the habitat. It may remain predominantly within the patch and move between suitable trees, but is sometimes forced to travel considerable distances between habitat patches, possibly also due to (pre)occupied trees by larger competitors, as our data show. Nevertheless, the vast majority of individuals seem to stay within a few hundred metres of their natal site (Drag et al. 2011).
Here we have demonstrated the importance of old beech forests with an abundance of dead and partially dead trees, damaged trees and stumps for both beetle species. This tree species has already been described as very important for many saproxylic beetles, including M. asper (Leonarduzzi et al. 2017). In detail, there is a difference in the colonisation of different tree types between the two species. While more and fewer specimens of M. a. funereus were found on dying and already dead trees than would have been expected on the basis of their availability, respectively, the opposite is true for R. alpina. Dying but still standing trees have a sparser canopy, thicker bark, but not much undergrowth yet, all characteristics that have a positive effect on the presence of R. alpina (Russo et al. 2011). Since both species prefer standing trees (cf. Campanaro et al. 2017, Hardersen et al. 2017), but these are much more occupied by M. a. funereus than by R. alpina in the study area, this again suggests an inferiority of R. alpina in mutual interactions with larger M. a. funereus. It should also be emphasised that different types of trees may be preferred for different life functions of the two species: In the case of M. a. funereus, we have often observed the presence of several beetles on larger, dying but still standing trees, where females can find suitable sites for oviposition (see Kariyanna et al. 2017), as such trunks may provide sufficient food for their long development and greater insulation from the moist decay conditions typical of forests with many dead trees (Campanaro et al. 2017). Slightly fewer animals were found on dead lying, damaged trees and stumps. These trees were mainly occupied by males using them as mating sites (Hardersen et al. 2017). Both categories (but in reverse order of rank) were also best occupied by R. alpina. For both species, dead tree stumps can also serve as mating sites, but are used less frequently than the previous two categories. As Hardersen et al. (2017) have shown, tree stumps were clearly the least used by M. asper compared to logs and log piles (other categories were not considered). Only partially damaged (still living) trees are occupied least often by either species.
As larger (compared to smaller) trees provide a more permanent habitat in terms of food and moisture (Jurc 2004), a general preference for larger trees is also evident in our study area, although a wide range of trees of different thickness could be occupied, as reported for R. alpina (Russo et al. 2011). That is, more animals were found on trees with larger diameters, and the vast majority of all animals were observed on tree trunks with a diameter of more than 50 cm. We found that for larger diameters, the difference in the mean number of individuals per trunk between the two species was more pronounced in favour of the largest of the two species (mean difference: diameter < 80 cm − 0.8 specimens/trunk; diameter > 80 cm − 7.8 specimens per trunk; see Fig. 5). Finally, slightly more specimens of R. alpina than of M. a. funereus were found only on trees with the smallest diameter (< 60 cm, Fig. 5), which seem to be less favoured by both species, but more available for the smallest of the two.
Implications for conservation
Both species studied require an abundance of thick dead and partially dead beech trees. At sites where they occur together, the less competitive R. alpina can partially avoid larger M. a. funereus with movement on to only slightly damaged trees or their stumps if these are available. Forest reserves left to natural succession are extremely important for many organisms, including saproxylic insects, which contribute significantly to forest biodiversity by accelerating wood decomposition and nutrient cycling. The most important reason for protecting saproxylic insects and maintaining suitable deadwood habitats is that the loss of biodiversity would have a negative impact on the sustainability of ecosystem functions. Therefore, it would also be necessary to raise awareness of the importance of the saproxyles' role in an ecosystem and to protect these species by law. While R. alpina is largely protected by law, this is currently not the case for M. a. funereus. This is only due to a taxonomic revision of M. funereus, which has been synonymised with M. asper as a subspecies. This taxonomic change would then inevitably also require an urgent reassessment for the legal protection of this subspecies.