Abstract

Hunting statistics affirm that the number of wild boar (Sus scrofa) has increased consistently throughout Europe over recent decades. In this model analysis, we analyzed data on hunting bags in the Czech Republic to elucidate the effects of hunter effort on wild boar population dynamics. We presumed that hunting pressure destroys the regular social system whereby reproduction of younger, subordinate females is suppressed, consequently increasing overall reproductive capacity. Data on annual hunting bags for wild boar came from the official state statistics and covered 77 districts from 27 seasons between 1993 and 2019. “Change in hunting pressure”, i.e. the change in the number of boar harvested in any given year as a percentage of the numbers taken in the following year, was analyzed with a multivariate Generalized Linear Mixed Model (GLMM) for repeated measures. We applied the information-theoretic approach for estimating the effects of different factors on this Change in hunting pressure, using expanded information criteria. The GLMM of the best model revealed that the change in hunting pressure was growing with increasing total hunting bags for wild boar. In contrast, with an increase in the proportion of hunted sows within the overall bag, the Change in hunting pressure decreased. Furthermore, the interaction between the Percentage of hunted sows and Hunting bags for all wild boars confirmed these opposite trends. Finally, with the increasing coldness of the district, winter precipitation, and summer temperature, the Change in hunting pressure decreased.

Keywords:Sus scrofa; Social suppression; Population control; Hunting bags; Model analysis

Introduction

Hunting statistics assembled in 18 European countries confirm that wild boar (Sus scrofa) increased consistently throughout Europe between 1982 and 2012 [1]. More recent data from the Czech Republic suggest that this trend has continued [2]. The most important factor contributing to this population increase is the tremendous increase in the environment’s carrying capacity resulting from intensive crop production across Europe. Other factors, such as milder winters, reforestation, supplementary feeding and compensatory population responses of wild boar to hunting pressure, might also explain increased population growth [3-6]. With populations of large predators not yet well-established in many areas of Europe, shooting is the leading cause of mortality in wild boars, with only very few animals dying of natural causes, e.g., diseases [7,8]. With this in mind, it may be of the utmost importance to consider how wild boar hunting is organized and performed and the impact of hunting practices on population growth [7-10].

Hunting bag data, as illustrated above, are valuable for estimating long-term trends. However, this data type is often less detailed than the information obtained from comprehensive field studies. Despite this limitation, hunting bag data can provide insights beyond identifying long-term trends. Hence, we performed a model analysis on this data to reveal more nuanced relationships, which could inspire future detailed field studies. In the model analysis, we consider the effect of sustained hunting pressure on the species’ social dynamics and what effect that may have on population productivity. This work does not aim to challenge or replace detailed field studies. Instead, by exploring trends in population from over 70 districts of the Czech Republic sharing general strategies of wildlife management over many years, we seek to find general trends that could help manage wild boar more effectively in the future within the country. Several studies have underlined the central role of females in the wild boar social system and described patrilineality in this species, where social groups may consist of females of up to three generations [11-13].

Competitive strategies aiming at inhibiting the reproduction of others are commonly observed among female mammals living in relatively stable social groups [14,15]. We may expect the same in the wild boar females. Older popular scientific German literature argues that the stable social system of the wild boar inhibits the reproduction of young gilts e.g., [16]. Current scientific publications, however, need to be reminded of this idea, and more formal scientific evidence is still absent. If young females, free from the suppressive influence of a socially stable system, could start reproducing earlier and more productively, the more intensive the level of unselective hunting, the more likely it is that reproductive rates within the population as a whole will increase.

Furthermore, wild boars exhibit high synchrony of reproduction within one social group [17] and produce large litters [18]. Hence, cooperative rearing of offspring may play a significant role in shaping the observed social structure. Such multigenerational and female-dominated social units offer advantages for females in enabling the foraging and rearing of young people when multiple litters are present simultaneously in a group. The benefits of living in socially stable groups include increased resources that can increase the number and/or quality of offspring produced by the suppressing female and help with rearing offspring (including care and, in some cases, milk) [19].

Second, we considered the concepts of r- and K-selection proposed by MacArthur and Wilson [20], refined by Pianka [21,22], and others. According to Rindos [23], r and K-selection “are not mutually exclusive alternatives; rather they are but abstract extremes, the vast majority of organisms inhabiting environments that are neither totally r-selecting nor totally K-selecting”. This selection strategy may change intra specifically along ecological gradients [24]. On a microscale, localized shifts occur along the K-/r-selection continuum, even within a species [25]. Thus, it would fit theoretically into non-hunted wild populations or those with low hunting pressure. Therefore, as a surplus to lack of the social suppression mentioned above after an increased hunting pressure, favorable r-selection conditions will give the organism a competitive edge in unstable environmental conditions by directing high energy input into reproduction (i.e., many offspring) with a correspondingly low input into each of its progeny. It seems it could work for the wild boar population under intensive hunting pressure. Such a presumption may be realistic. In France and Italy, generation turnover times were 3.6 years in the lightly hunted population and only 2.3 years in the heavily hunted population [26]. In Châteauvillain-Arc-en-Barrois, north-eastern France, long-term monitoring (22 years) of a wild boar population subject to markedly increasing hunting pressure, birth dates have advanced by up to 12 days throughout the study period [27]. In Germany, Gethöffer et al. [28] have documented that exogenous factors have a strong impact on both reproductive seasonality and the percentage of reproducing individuals in an age group. However, they presented no information about the effect of hunting pressure per se. The potential positive influence of collaborative social relationships could be further supported by the fact that in Northern Germany, population density did not affect litter size [29]. Thus, social cooperation could compensate for the adverse effects of the population pressure.

In addition to releasing reproductive suppression of young females, intensive hunting of wild boar can theoretically affect the reproduction of wild boar females in at least two directions. Selective hunting of breeding sows can first decrease the reproduction of the most productive individuals. Orphaned wild boar females dispersed from the natal group [30-33] did not form groups based strictly on relatedness but merged into larger groups. Bieber et al. [34] emphasized that wild boar females are aggressive against unknown conspecifics. Therefore, such disruption might suppress their reproduction temporarily. At the same time, it may also be stressful for orphaned young females to stay within the natal groups for the lack of social support. It would mean a lack of social buffering documented across species [35,36] etc.

Numerous studies have addressed this effect in domestic pigs. Social deprivation is a severe stressor affecting many behavioral and physiological functions of gregarious species, including pigs [37]. The need to re-establish social hierarchy induces intense social stress [38,39]. There is growing evidence that positive social interactions can attenuate the effects of stressful life experiences [40,41]. Thus, lack of social support from familiar conspecifics may affect the pigs’ behaviour, autonomic nervous and neuroendocrine systems, immune system, etc. So far as we know, there is no such data for wild boars. However, based on the studies on domestic pigs, we may presume that the significant destructive effect of removing adult females on the prosperity of the wild boar population is quite substantial, most likely on younger females in particular. In contrast to females, wild boar males tend to form associations with unrelated males, which conforms to male-biased dispersal in this species [42], and their effect on the productivity rate of the wild boar population is marginal.

In our model analysis, we set up two hypotheses (Table 1): (i) If an increased hunting pressure disordered the social system, it would release the young females’ reproductive suppression and maximize their reproductive capacity, shifting K- strategy to rstrategy effects.; (ii) Accumulating the proportion of harvested sows will either increase (if it would contribute to further destroying the social stability of the population) or decrease reproduction (if removing the adult sows would worsen the stress and well-being of young females). The published literature does not allow us to assert with any confidence which of these is likely to have the greater effect, and thus we cannot offer confident prediction as to the direction of any effect, however, it is clear that it should not be neutral.

Material and Methods

Since it is generally impossible to get accurate data about the size of free-ranging populations of wild boar, annual reports on hunting bags are commonly used as a proxy for actual numbers of individuals [43-45]. Unless specified below, the data come from the Forest Management Institute, Brandýs nad Labem, Section of Hunting Management at Frýdek-Místek, where all information on hunting bags and other hunting-related records have been stored and processed for official state statistics. For the analysis, hunting bags (categorized to all wild boar and then separately to sows, boars, yearlings and piglets) from all districts of the Czech Republic (n=77) over 29 seasons (1990 and 1991, 1994 to 2019) were available. For each district, we also had the information on the district size (in hectares), the size of agricultural, forested, water, and other areas within the district, the number of registered hunters and hunters’ trained dogs. All data were analyzed using the SAS System (SAS, version 9.4). District size varied considerably (Table 2). Therefore, we first adjusted the variables of total hunting bag for all wild boar, number of registered hunters, and hunters’ trained dogs per 1000 ha of district size. Next, to eliminate a longlasting average rise of hunting bags and the disparity of the size of districts, we calculated the change in the number of wild boars harvested by comparing harvest in a given year with that taken in the following season (current ‘year’s bag as a percentage of the following ‘year’s bag). This variable, denoted “Change in hunting pressure”, has a value over 100 %, indicating an increase between successive seasons, while a value lower than 100% indicates a decrease in harvesting.)

All other variables were expressed as a proportion within the district. i.e., the percentage of the total hunting bag comprised of sows, boars, yearlings and piglets per district and year. In the same way, we calculated the percentage of agricultural land, forested land, water, and other land types for each district. We then made multicollinearity diagnostics after Schreiber-Gregory and Jackson [46]. Our first step was to explore the correlation matrix of hunting bag, cull composition and variables of proportional land cover. The Percentage of hunted sows correlated with the Percentage of hunted boars (r=0.67), and the Percentage of hunted yearlings correlated with the Percentage of hunted piglets (r=- 0.83). As expected, the percentage of agricultural areas correlated negatively with the percentage of forested areas (r=-0.95), and the number of Registered hunters correlated with the number of Hunters’ trained dogs (r=0.70). Since there was no relationship between the hunting bag and area data, we analyzed the data separately in the next step through the Variance Inflation Factor and Tolerance (using the VIF, TOL, and COLLIN options in PROC REG).

We had to remove the variable reflecting the total hunting bag from the set because it showed high collinearity due to a minimal Eigenvalue (1E-12) combined with a high Condition index (1669907). Then, we removed the percentage of hunted boars, significantly correlated with the percentage of hunted sows. However, the problem of collinearity was not solved. Even after removal from the data set of the percentage of yearlings [tolerance=0.05, eigenvalue = 0.0007, Condition index = 71.78], there was still an indication of the collinearity with the percentage of hunted piglets (Eigenvalue = 0.01, Condition index = 14.42). Thus, no combination of the hunting bag data was free from collinearity, and all variables had to be involved in subsequent GLMMs individually. A similar situation was apparent with the area data. First, we removed the percentage of forested areas because of the correlation with the percentage of agricultural areas. Even after its removal, there remained an indication of collinearity with the percentage area of water (eigenvalue = 0.01, Condition index = 13.66) with the percentage of other areas (eigenvalue = 0.02, Condition index = 12.10).

Each district was assigned to individual climate units after Quitt [47]. These units are based on long-term measured climate characteristics and represent a given district’s average longterm climate-specific measurements. The features are expressed as a percentage covering the district according to the following climate areas; very cold, cold, moderately warm, warm and very warm. Except for the characteristic of very cold, for all other aspects, there are three levels of humidity (poor, average, and rich precipitations), giving 13 more or less inter-correlated characteristics. Although no variables had an exceptionally high correlation (about 0.8 or higher), we examined multicollinearity as above. We found a high rate of multicollinearity among the characteristics. To cope with the multicollinearity of the climatic classification data, we applied the principal component analysis (PCA, PROC PCA) after Schreiber-Gregory and Jackson. We decided on five factors to keep for the final model from the eigenvalues of the correlation matrix (with five eigenvalues greater than 1) and a general review of the scree plot. Then, by utilizing a parallel analysis criterion that gave five factors, we verified the estimate and generated five components (Quitt_PCA1, Quitt_PCA2, Quitt_ PCA3, Quitt_PCA4, Quitt_PCA5). For further analysis, we selected two components, best representing double-sided limit values (Quitt_PCA2 and Quitt_PCA5); the Quitt_PCA2 characterized a cold climate, while Quitt_PCA5 was a warm environment.

Monthly data for temperature and precipitation were downloaded from the Czech Hydrometeorological Institute web pages, and means were generated for extended periods (averages across several months, starting with successive months up to half a year). After inspection, we used mean values from November of the previous year to April (winter) and May to October (summer) for precipitation and temperature. Then, as before, we calculated correlations and tested for multicollinearity. Precipitation and temperature data did not correlate. They did not correlate with the Quitt_PCAs either. On the other hand, for both characteristics, we discovered multicollinearity between winter and summer data according to the Collinearity Diagnostics, specifically in the relationship of the eigenvalue column to the condition index column. The multicollinearity was indicated by small eigenvalues (0.02 and 0.0086 for precipitation, and 0.05 and 0.0009 for temperature) with large corresponding condition numbers (11.12 and 18.56 for precipitation, and 7.7 and 56.05 for temperature).

We used a multivariate Generalized Linear Mixed Model (GLMM, PROC MIXED). Because we analyzed data on the same districts over all hunting seasons/years, analyses were designed for repeated measures (using PROC MIXED with REPEATED = Year and SUBJECT=District). For the dependent variable Change in hunting pressure, we constructed a set of a priori hypotheses (Table 3) covering all three tested presumptions (Table 1) with fixed factors listed in (Table 2). Several count variables were logtransformed (natural logarithm transformation) to improve the normality of residuals and to reduce skewness (marked with “t” at the end of the variable name in (Table 3). Where appropriate, we included interaction terms.

We used the information-theoretic approach to estimate the factors’ effects on the dependent variable [48,49]. Since the introduction of Akaike’s Information Criterion AIC, [50] more information criteria have been developed with differing mathematical properties and philosophies of model selection in mind. After Christensen [51], we used expanded information criteria AIC, AICC [52], BIC [53], CAIC [54], and HQIC [55] to select a true model of the seasonal or social. Then we compared the candidate models by ranking them based on the information criteria (PROC RANK). The model with the lowest value (i.e., closest to zero) is considered to be the “best” model. The differences (Δi) between the Fit statistic values (the smallest values indicating the best fitting model) were sorted according to AIC for the five best models. We used Δ = 2 as a criterion, a distance widely interpreted as indicating a “substantial level of empirical support” [49: p. 170]. Akaike weight wi can be interpreted as the probability that Mi is the best model (in the AIC sense, that minimizes the Kullback– Leibler discrepancy), given the data and the set of candidate models e. g., We calculated Δ AIC, Akaike weights wi for all models, and for estimating the strength of evidence in favour of one model over the other we calculated the AIC Odds by dividing their Akaike weights wmin/wj. As recommended by various authors e. g., [56,57], we obtained Δ BIC, BIC weights wi, and BIC Odds using the same formulas, just replacing AIC with BIC values. Compared with AIC, BIC, the advantage of this is that it penalizes models with more parameters. Thus, the BIC weights wi may be appreciably different from AIC weights wi.

To see if the best model has merit, we adapted an approach from Burnham and Anderson (2002) and used again the two most common information criteria AIC and BIC. We compared our best model to the null model for the dependent variable using Δ AIC (AIC null – AIC best model) and a relative information loss [exp ((AIC_null − AICi_best)/2)], and Δ BIC (BIC null – BIC best model) and a relative information loss [exp ((BIC_null − BICi_best)/2)]. We calculated coefficient estimates, standard errors, and 95% confidence intervals for each fixed effect in the best model. Associations between the dependent variable and countable fixed effects are presented by fitting a random coefficient model using GLMM described by Tao et al. (2002). We calculated the predicted values of the dependent variable and plotted them against the fixed effects with predicted regression lines.

Results

In our model analysis, statistics on hunting bags (total of 2933041 harvested wild boar) revealed highly unbalanced hunting of wild boar in Czechia with a predominant proportion of piglets (mean ± STD, min, max; 57.19% ± 9.93, 0.00, 88.41), and yearlings (34.02% ± 13.31, 0.00, 100.00), and with a much lower proportion of sows (4.33% ± 3.97, 0.00, 33.00) and 5% of adult males (4.45% ± 5.48, 0.00, 37.30). Total hunting bags for wild boar tended to increase over time because the total number of yearlings and piglets harvested grew, while by contrast the number of adults shot tends to have remained constant. Information criteria and rankings for the five best candidate models (“best” model first) and Null Model are shown in (Table 4). Uniform ranking across the four information criteria makes a convincing argument that the model with the cross-level interaction is the “best” model. As an addition to (Table 4), five best candidate models and Null model with Δ AIC, Akaike weights, AIC Odds and Δ BIC, BIC weights, BIC Odds, sorted by AIC (from the lowest to the highest value) are presented in (Table 5). We can interpret Akaike and BIC weights as the probability that Model 1 is the correct model amongst the candidate set is 75% [58]. According to the Odds, Model 1 is 6 times more likely to be the best model in terms of Kullback– Leibler discrepancy than the next-best Model 2. Moreover, the factors Percentage of hunted sows (Percent_sowst), Total hunting bag (Tot_hunt_bagt), and their interaction are apparently the most influential factors because they are involved in 7 best models (models 6 and 7 not shown). Relative information loss from comparing the Null and the best models confirms that the best model has merit (Akaike and BIC weights). The estimate, standard error and 95% confidence interval of the fixed factors in the best model are shown in (Table 6). The 95% confidence intervals of all fixed effects did not include zero. Therefore, they would also be considered significant if Null Hypothesis Testing were used. With increasing total hunting bags (all wild boar regardless of age and sex), the Change in hunting pressure was also growing. In contrast, with the increasing percentage of hunted sows, the change in hunting pressure was decreased. These opposite trends were further confirmed in the interaction between the percentage of hunted sows and total hunting bags (Figure 4). Finally, with increasing Quitt_PCA2 (Figure 5), winter precipitation and summer temperature (Figure 6), the change in hunting was decreasing.

Discussion

Presuming that annual reports on hunting bags correlate with and are an adequate proxy for actual numbers of free-ranging wild boar, our model analysis suggests several important conclusions. As predicted, the social structure of the hunted population seems to be the crucial factor in modifying the effect of hunting on subsequent population development. With increasing hunting pressure, the population increased (Figure 2). We suggest that the mechanism behind that is that the influence of a socially stable system in the suppression of reproduction of subdominant sows is probably diminishing. In addition, intensive hunting thus develops unpredictable living conditions for the animals. This change should release mechanisms shifting intraspecifically the range from K- towards r-selection. Therefore, theoretically, the age of onset of female reproduction decreases and probably simultaneously increases the size of their litters. So intense hunting is counter-productive in an attempt to control the wild boar population growth. This conclusion corresponds well with the data from other European countries [59].

On the other hand, increasing the proportion of sows within the hunting bag does not seem in itself to contribute to the observed increase in productivity associated with increased culling pressure overall. This lack of increase in population productivity with increased proportion within the hunting bag of mature females might suggest instead that hunting sows may increase social deprivation within the group and severely stress some of the other reproductive females, having in consequence, a negative effect on their behavioral and physiological functions. Consequently, removing the adult sows is likely acting against the increasing population (Figure 3). We could speculate that increasing the percentage of hunted sows might reach a point where the release of suppression of reproduction could match the stressful effects associated with the absence of a leading sow. However, perhaps the proportion of female guards would have to be greater than the range in our data, as the data inspection shows no sign of countervailing trends. Therefore, the proportion of sows hunted should be substantially increased in Czechia. Our model analysis did not support generally recommended higher hunting rates of piglets e.g., for controlling population growth. The GLMM models containing the percentage of harvested piglets were not featured among the models selected as the best fit. By removing predominantly piglets, their mothers may reduce output in milk and any other energy associated with caring for their offspring. Saving energy output benefits the sows’ physical condition, particularly for younger sows [60]. Better physiological condition of sows results in larger litters [61]. Surviving piglets have better living conditions, their body weight increases and mortality decreases, and young females thus quickly reach a threshold body mass necessary for breeding for the first time [62].

Our analysis also revealed an effect of climatological factors, although there are differences in this aspect between various studies [63,64]. In the present study, districts classified as climatically colder tended to be less suitable for the wild boar. High winter precipitation and warm summer temperatures also harmed population growth. Thus, the effects of the weather may be attributed to the nutritional condition of females, litter size production, the success of hunting [65], etc. Our results would be comparable with Frauendorf et al. regarding summer temperature but not winter precipitation. Another adverse effect of the cold environment and increased winter precipitation may be associated with reduced wild boar movement in such conditions [66,67], perhaps leading to reduced quality of nutrition and, therefore, survival of the animals. Thus, high winter precipitation may generally make it difficult for the survival of young boar. On the other hand, our analysis could not consider local habits, such as improvement of nutrition, supplementary feeding and feeding at baiting sites [68]. Local nutritional conditions are generally crucial for pig reproduction and survival [69,70].

Conclusion

As concluded previously, new approaches are urgently required to mitigate human–wild boar conflicts, otherwise destined to grow further [71]. Since the current model analysis results depend on the hunting bags, they should be handled with care. Nonetheless, we believe the findings could inspire wild boar management. Our results suggested that increasing hunting pressure destroying the social system is not the solution. Therefore, hunters should replace their approach with alternatives to reduce wild boar numbers while maintaining their social structure, catching entire social units in traps. Various trap types for trapping wild pigs are available from a conventional trap design, drop nets, and recently developed suspended traps [72-75]. Experience mainly from the US in eradicating invasive wild pigs has shown high efficiency. For example, suspended traps removed 88.1% of the estimated population of wild pigs, whereas drop nets removed 85.7%, and corral traps removed 48.5% [76].

References

1. Massei G, Kindberg J, Licoppe A, Gacic D, Sprem N, et al. (2014) Wild boar populations up, numbers of hunters down? A review of trends and implications for Europe. Pest Management Science 71: 492-500.
2. Tkadlec E (2021) Meze růstu populace prasete divokého na dohled? Svět Myslivosti 22: 8-11.
3. Morelle K, Lejeune P (2015) Seasonal variations of wild boar Sus scrofa distribution in agricultural landscapes: a species distribution modelling approach. European Journal of Wildlife Research 61(1): 45-56.
4. Gabor TM, Hellgren EC, Van den Bussche RA, Silvy NJ (1999) Demography, sociospatial behaviour and genetics of feral pigs (Sus scrofa) in a semi-arid environment. Journal of Zoology 247(3): 311-322.
5. Canu A, Scandura M, Merli E, Chirichella R, Bottero E, et al. (2015) Reproductive phenology and conception synchrony in a natural wild boar population. Hystrix-Italian Journal of Mammalogy 26(2): 77-84.
6. Gethöffer F, Keuling O, Maistrelli C, Ludwig T, Siebert U (2023) Heavy youngsters-habitat and climate factors lead to a significant increase in body weight of wild boar females. Animals 13(5): 898.
7. Keuling O, Baubet E, Duscher A, Ebert C, Fischer C, et al. (2013) Mortality rates of wild boar Sus scrofa L. in central Europe. European Journal of Wildlife Research 59(6): 805-814.
8. Toïgo C, Servanty S, Gaillard JM, Brandt S, Baubet E (2008) Disentangling natural from hunting mortality in an intensively hunted wild boar population. Journal of Wildlife Management 72(7): 1532-1539.
9. Keuling O, Podgórski T, Monaco A, Melletti M, Merta D, et al. (2018) Eurasian wild boar Sus scrofa (Linnaeus, 1758). In: M Melletti, E Meijaard, (Eds.). Ecology, Conservation and Management of Wild Pigs and Peccaries 202-233.
10. Johann F, Handschuh M, Linderoth P, Dormann CF, Arnold J (2020) Adaptation of wild boar (Sus scrofa) activity in a human-dominated landscape. BMC Ecology 20(1): 4.
11. Kaminski G, Brandt S, Baubet E, Baudoin C (2005) Life-history patterns in female wild boars (Sus scrofa): mother–daughter postweaning associations. Canadian Journal of Zoology 83(3): 474-480.
12. Poteaux C, Baubet E, Kaminski G, Brandt S, Dobson FS, et al. (2009) Socio-genetic structure and mating system of a wild boar population. Journal of Zoology 278(2): 116-125.
13. Podgórski T, Lusseau D, Scandura M, Sonnichsen L, Jedrzejewska B (2014) Long-lasting, kin-directed female interactions in a spatially structured wild boar social network. Plos One 9(6): e99875.
14. Stockley P, Bro-Jorgensen J (2011) Female competition and its evolutionary consequences in mammals. Biological Reviews 86: 341-366.
15. Wasser SK, Barash DP (1983) Reproductive suppression among female mammals: implications for biomedicine and sexual selection theory. Quarterly Review of Biology 58: 513-538.
16. Meynhardt H (1978) Schwarzwild-Report. Vier Jahre unter Wildschweinen. Leipzig Radebeul: Neumann Verlag.
17. Delcroix I, Mauget R, Signoret JP (1990) Existence of synchronisation of reproduction at the level of the social group of the European wild boar (Sus-scrofa). Journal of Reproduction & Fertility 89(2): 613-617.
18. Servanty S, Gaillard J-M, Allainé D, Brandt S, Baubet E (2007) Litter size and fetal sex ratio adjustment in a highly polytocous species: the wild boar. Behavioral Ecology 18(2): 427-432.
19. Mauget R (1982) Seasonality of reproduction in the wild boar. In: DJA Cole, GR Foxcroft, (Eds.). Control of pig reproduction. (Butterworth: London), pp 509-526.
20. MacArthur RH, Wilson E (1967) The theory of island biogeography. Princeton, NJ: Princeton University Press. 202 p.
21. Pianka ER (1970) On r-selection and K-selection. American Naturalist 104: 592-597.
22. Pianka ER (1972) r and K selection or b and d selection. American Naturalist 106(951): 581-588.
23. Rindos D (2013) The origins of agriculture: an evolutionary perspective. San Diego, New York: Academic Press.
24. Tchernov E, Horwitz LK (1991) Body size diminution under domestication - unconscious selection in primeval domesticates. Journal of Anthropological Archaeology 10(1): 54-75.
25. Safriel UN, Ritte U (1983) Universal correlates of colonising ability. In: IR Swingland, PJ Greenwood, (Eds.). Ecology of animal movement. (Oxford University Press: Oxford), pp 215-243.
26. Servanty S, Gaillard JM, Ronchi F, Focardi S, Baubet E, et al. (2011) Influence of harvesting pressure on demographic tactics: implications for wildlife management. Journal of Applied Ecology 48(4): 835-843.
27. Gamelon M, Besnard A, Gaillard JM, Servanty S, Baubet E, et al. (2011) High hunting pressure selects for earlier birth date: Wild boar as a case study. Evolution 65(11): 3100-3112.
28. Gethöffer F, Sodeikat G, Pohlmeyer K (2007) Reproductive parameters of wild boar (Sus scrofa) in three different parts of Germany. European Journal of Wildlife Research 53(4): 287-297.
29. Frauendorf M, Gethöffer F, Siebert U, Keuling O (2016) The influence of environmental and physiological factors on the litter size of wild boar (Sus scrofa) in an agriculture dominated area in Germany. Science of the Total Environment 541: 877-882.
30. Casas-Díaz E, Closa-Sebastià F, Peris A, Miño A, Torrentó J, et al. (2013) Dispersal record of Wild boar (Sus scrofa) in northeast Spain: Implications for implementing disease-monitoring programs. Wildlife Biology in Practice 9: 19-26.
31. Jerina K, Pokorny B, Stergar M (2014) First evidence of long-distance dispersal of adult female wild boar (Sus scrofa) with piglets. European Journal of Wildlife Research 60(2): 367-370.
32. Keuling O, Lauterbach K, Stier N, Roth M (2010) Hunter feedback of individually marked wild boar Sus scrofa L.: dispersal and efficiency of hunting in northeastern Germany. European Journal of Wildlife Research 56: 159-167.
33. Keuling O, Stier N, Roth M (2009) Commuting, shifting or remaining? Different spatial utilisation patterns of wild boar Sus scrofa L. in forest and field crops during summer. Mammalian Biology 74: 145-152.
34. Bieber C, Rauchenschwandtner E, Michel V, Suchentrunk F, Smith S, et al. (2019) Forming a group in the absence of adult females? Social Networks in yearling wild boars. Applied Animal Behaviour Science 217: 21-27.
35. Cohen S, Wills TA (1985) Stress, social support, and the buffering hypothesis. Psychological Bulletin 98(2): 310-357.
36. Hennessy MB, Kaiser S, Sachser N (2009) Social buffering of the stress response: Diversity, mechanisms, and functions. Frontiers in Neuroendocrinology 30: 470-482.
37. Kanitz E, Hameister T, Tuchscherer M, Tuchscherer A, Puppe B (2014) Social support attenuates the adverse consequences of social deprivation stress in domestic piglets. Hormones and Behavior 65(3): 203-210.
38. Arey DS, Edwards SA (1998) Factors influencing aggression between sows after mixing and the consequences for welfare and production. Livestock Production Science 56: 61-70.
39. Rault JL (2012) Friends with benefits: Social support and its relevance for farm animal welfare. Applied Animal Behaviour Science 136: 1-14.
40. Tuchscherer M, Kanitz E, Puppe B, Hameister T, Tuchscherer A (2014) Social support modulates splenocyte glucocorticoid sensitivity in piglets exposed to social deprivation stress. Physiology & Behavior 131: 25-32.
41. Tuchscherer M, Kanitz E, Tuchscherer A, Puppe B (2016) Effects of social support on glucocorticoid sensitivity of lymphocytes in socially deprived piglets. Stress-the International Journal on the Biology of Stress 19(3): 325-332.
42. Truvé J, Lemel J (2003) Timing and distance of natal dispersal for wild boar Sus scrofa in Sweden. Wildlife Biology 9: 51-57.
43. Fanelli A, Perrone A, Ferroglio E (2021) Spatial and temporal dynamics of wild boars Sus scrofa hunted in Alpine environment. European Journal of Wildlife Research 67(3): 10.
44. Imperio S, Focardi S, Santini G, Provenzale A (2012) Population dynamics in a guild of four Mediterranean ungulates: density-dependence, environmental effects and inter-specific interactions. Oikos 121: 1613-1626.
45. Segura A, Acevedo P, Rodriguez O, Naves J, Obeso JR (2014) Biotic and abiotic factors modulating wild boar relative abundance in Atlantic Spain. European Journal of Wildlife Research 60: 469-476.
46. Schreiber-Gregory DN, Jackson HM (2017) Multicollinearity: What is it, why should we care, and how can it be controlled. Proceedings of the SAS® Global Forum 2017 Conference. (SAS Institute Inc.: Cary, NC), pp Conference Paper 1404.
47. Quitt E (1971) Klimatické oblasti Č Climatic regions of Czechoslovakia. Praha: Academia.
48. Richards SA, Whittingham MJ, Stephens PA (2011) Model selection and model averaging in behavioural ecology: the utility of the IT-AIC framework. Behavioral Ecology and Sociobiology 65: 77-89.
49. Burnham KP, Anderson DR (2002) Model selection and multimodel inference: A practical information-theoretic approach. Second edition. New York, Berlin, Heidelberg: Springer-Verlag.
50. Akaike H (1973) Information theory and an extension of the maximum likelihood pronciple. In: BN Petrov, F Csaki, (Eds.). Proc the 2nd international symposium on information theory. (Akademia Kiado: Budapest), pp 267-281.
51. Christensen W (2018) Model selection using information criteria (Made easy in SAS®). SAS Conference Proceedings: Western Users of SAS Software 2018, September 5-7, 2018, Sacramento, California: Paper 2587-2018.
52. Hurvich CM, Tsai C-L (1989) Regression and time series model selection in small samples. Biometrika 76(2): 297-307.
53. Schwarz G (1978) Estimating the dimension of a model. Annals of Statistics 6(2): 461-464.
54. Bozdogan H (1987) Model selection and Akaike information criterion (AIC) - the general-theory and its analytical extensions. Psychometrika 52: 345-370.
55. Hannan EJ, Quinn BG (1979) Determination of the order of an autoregression. Journal of the Royal Statistical Society Series B-Methodological 41(2): 190-195.
56. Buckland ST, Burnham KP, Augustin NH (1997) Model selection: An integral part of inference. Biometrics 53(2): 603-618.
57. Wagenmakers EJ, Farrell S (2004) AIC model selection using Akaike weights. Psychonomic Bulletin & Review 11(1): 192-196.
58. Canchola JA, Neilands TB (2006) Mixed Model Selection via the Information-Theoretic Approach using SAS. SAS Conference Proceedings: Western Users of SAS Software 2006 September 27-29.
59. Keuling O, Strauss E, Siebert U (2016) Regulating wild boar populations is "somebody else's problem"! - Human dimension in wild boar management. Science of the Total Environment 554: 311-319.
60. Andersen IL, Naevdal E, Boe KE (2011) Maternal investment, sibling competition, and offspring survival with increasing litter size and parity in pigs (Sus scrofa). Behavioral Ecology and Sociobiology 65(6): 1159-1167.
61. Malmsten A, Jansson G, Lundeheim N, Dalin AM (2017) The reproductive pattern and potential of free ranging female wild boars (Sus scrofa) in Sweden. Acta Veterinaria Scandinavica 59(1): 52.
62. Servanty S, Gaillard JM, Toigo C, Brandt S, Baubet E (2009) Pulsed resources and climate-induced variation in the reproductive traits of wild boar under high hunting pressure. Journal of Animal Ecology 78(6): 1278-1290.
63. Bergqvist G, Paulson S, Elmhagen B (2018) Effects of female body mass and climate on reproduction in northern wild boar. Wildlife Biology 2018(1):1-6.
64. Oja R, Kaasik A, Valdmann H (2014) Winter severity or supplementary feeding-which matters more for wild boar? Acta Theriologica 59: 553-559.
65. Rösslová M, Vacek S, Vacek Z, Prokůpková A (2020) Impact of climatic factors on the success of hunting various game species in Czech Republic. Applied Ecology and Environmental Research 18(2): 2989-3014.
66. Thurfjell H, Spong G, Ericsson G (2014) Effects of weather, season, and daylight on female wild boar movement. Acta Theriologica 59: 467-472.
67. Johann F, Handschuh M, Linderoth P, Heurich M, Dormann CF, et al. (2020) Variability of daily space use in wild boar Sus scrofa. Wildlife Biology 12.
68. Zeman J, Hrbek J, Drimaj J, Kudláček T, Heroldová M (2018) Habitat and management influence on a seasonal diet composition of wild boar. Biologia 73: 259-265.
69. Drimaj J, Kamler J, Hošek M, Plhal R, Mikulka O, et al. (2020) Reproductive potential of free-living wild boar in Central Europe. European Journal of Wildlife Research 66: 11.
70. Rosell C, Navàs F, Romero S (2012) Reproduction of wild boar in a cropland and coastal wetland area: implications for management. Animal Biodiversity and Conservation 35(2): 209-217.
71. Massei G, Cowan DP, Coats J, Bellamy F, Quy R, et al. (2012) Long-term effects of immunocontraception on wild boar fertility, physiology and behaviour. Wildlife Research 39(5): 378-385.
72. Alexandrov T, Kamenov P, Stefanov D, Depner K (2011) Trapping as an alternative method of eradicating classical swine fever in a wild boar population in Bulgaria. Revue Scientifique Et Technique-Office International Des Epizooties 30(3): 911-916.
73. Metcalf EM, Parker ID, Lopez RR, Higginbotham B, Davis DS, et al. (2014) Impact of gate width of corral traps in potential wild pig trapping success. Wildlife Society Bulletin 38(4): 892-895.
74. Fahlman A, Lindsjo J, Norling TA, Kjellander P, Agren EO, et al. (2020) Wild boar behaviour during live-trap capture in a corral-style trap: implications for animal welfare. Acta Veterinaria Scandinavica 62(1): 59.
75. Gaskamp JA, Gee KL, Campbell TA, Silvy NJ, Webb SL (2021) Effectiveness and efficiency of corral traps, drop nets and suspended traps for capturing wild pigs (Sus scrofa). Animals 11: 1565.
76. Colomer J, Rosell C, Rodriguez-Teijeiro JD, Massei G (2021) 'Reserve effect': An opportunity to mitigate human-wild boar conflicts. Science of the Total Environment 795: 11.