European Journal of Wildlife Research

, Volume 52, Issue 3, pp 207–212

Habitat and reproductive phenology of wild boar (Sus scrofa) in the western Iberian Peninsula

Authors

    • Departamento de Ecologia, Colégio Luís António Verney, R. Romão Ramalho 59Universidade de Évora
  • P. Fernández-Llario
    • Departamento de Biología y GeologíaInstituto de Enseñanza Secundaría “Santa Lucía del Trampal”
  • C. Fonseca
    • Departamento de BiologiaUniversidade de Aveiro
  • A. Monzón
    • Centro de Estudos em Gestão de Ecossistemas, Departamento FlorestalUniversidade de Trás-os-Montes e Alto Douro, Quinta de Prados
  • P. Bento
    • Centro de Estudos em Gestão de Ecossistemas, Departamento FlorestalUniversidade de Trás-os-Montes e Alto Douro, Quinta de Prados
  • A. M. V. M. Soares
    • Departamento de BiologiaUniversidade de Aveiro
  • P. Mateos-Quesada
    • Departamento de Biología y GeologíaInstituto de Enseñanza Secundaría “Santa Lucía del Trampal”
  • F. Petrucci-Fonseca
    • Departamento Zoologia e Antropologia, Faculdade de Ciências C2Universidade Clássica de Lisboa
Short Communication

DOI: 10.1007/s10344-005-0025-z

Cite this article as:
Santos, P., Fernández-Llario, P., Fonseca, C. et al. Eur J Wildl Res (2006) 52: 207. doi:10.1007/s10344-005-0025-z

Abstract

We examined the reproductive phenology of wild boar populations in four regions of the western Iberian Peninsula during the 1999/2000 hunting season (October–February). To estimate conception dates and birth distribution frequencies, we used foetal weights. Regions differed significantly, and we detected a relationship between region and birth distribution frequencies. Throughout the year, food availability had a major influence on the distribution of farrowing. Although a short period of high food availability leads to highly synchronous births, even in relatively harsh environmental conditions, adult females that exploit low-quality food items appear to be able to give birth at any time of the year.

Keywords

Sus scrofaBirth frequency distributionHunting seasonMediterranean Ibero Atlantic Province

Introduction

In the last 20 years, the wild boar (Sus scrofa) has become a very common species in most of the Iberian Peninsula, and the recent rapid increase in population has been the subject of several studies (Telleria and Sáez-Royuela 1985; Santos 1994; Nores et al. 1995; Fonseca 1999; Gortázar et al. 2000). Apart from their biological and ecological importance, data on the annual distribution of wild boar births contribute to a more ethical use of this game animal by helping hunters to avoid shooting nursing females.

Mauget (1980) suggested that there are two types of birth frequency distributions in wild boar populations. In the unimodal distribution, there is a single annual breeding season that occurs in spring, and in the bimodal distribution, there is a primary breeding season in late winter and a secondary period of births in summer. In the latter situation, when weaning occurs in spring or early summer, a second breeding season can occur in late summer, which can include young females farrowing for the first time and adult females giving birth for the second time in that year. Yet, a bimodal distribution of births appears to be uncommon under natural conditions, and in the absence of supplementary feeding and, typically, wild boar breed once a year (CNERA Cervides–Sanglier 1988; Boitani et al. 1995).

Mauget et al. (1984) emphasized the importance of environmental and social factors, such as photoperiod, food availability and role of the female in the group, in modulating reproduction in the wild boar. Delcroix et al. (1990) demonstrated reproductive synchronization within groups of females. Food availability can influence birth chronology and interannual birth frequency distributions (Aumaître et al. 1984; Baber and Coblentz 1987; Fruzinski and Naparty 1992; Fernández-Llario and Mateos-Quesada 1998). In autumn, females appear to delay the onset of breeding until they attain the threshold weight needed to enter oestrus and they can skip reproduction in years of severe food shortage (Mauget and Pepin 1991; Groot Bruinderink et al. 1994). Feral pigs will breed throughout the year, although there are small seasonal peaks in birth frequencies (Barret 1978; Dzieciolowski et al. 1992). Pig breeds raised in extensive farming systems, such as the Iberian breed, still retain a seasonal breading pattern showing a lower conception rate in summer (Dobao et al. 1983).

There are no published data on the reproductive phenology of wild boar populations in Portugal, and in Spain, research has focused mostly on populations in the northern, southern and south-eastern regions of the country, where unimodal birth frequency distributions occur (Sáez-Royuela 1987; Abaigar 1990; Garzón 1991; Fernández-Llario and Carranza 2000). In south-western Spain, where wild boar hunting is a strongly traditional activity of great economic importance, the reproductive phenology of the wild boar population is unstudied.

In this study, we examine the reproductive phenology of wild boar populations in four regions of the Iberian Peninsula that represent different habitats and test the null hypothesis that there is no relationship between region and birth frequency distribution in the year 1999/2000. We interpret our results from an ecological perspective and argue that decisions about the timing and duration of hunting seasons should take in consideration regional reproductive phonologies, which can vary annually.

Materials and methods

We conducted our study in the western Iberian Peninsula, where the climate is pluvioseasonal; rain is more likely in spring than in autumn, and the Azores High Pressure System promotes a drought in summer (Rivas-Martínez and Loidi 1999). Soils consist mainly of Palaeozoic siliceous materials, and a long period of human influence on the environment has led to a mosaic of plant cover types.

We gathered data from November to February during the hunting season of 1999/2000 in three regions of Portugal (Trás-os-Montes, Beiras and Alentejo) and in one region of Spain (Extremadura; Fig. 1). In those regions, the shooting technique permits the harvesting of males and females (Fernández-Llario and Mateos-Quesada 2002), and according to Fernández-Llario et al. (2003), it is appropriate for surveying wild boar populations.
https://static-content.springer.com/image/art%3A10.1007%2Fs10344-005-0025-z/MediaObjects/10344_2005_25_Fig1_HTML.gif
Fig. 1

Study areas within each country

To estimate the age of females in the field, we used size, coat colour and, when in doubt, the chronology of teeth eruption (ONC 1995; Dardaillon 1986; Fernández-Llario 1996; Fernández-Llario et al. 1996; Fernández-Llario and Carranza 2000). Females were assigned to one of three age classes: class 1 =  <1 year old, class 2 = 1–2 years old, class 3 = > 2 years old. We collected the uteri and ovaries, which were analysed in the laboratory. To determine the age of each litter, we used the Vericad (1983) formula (after Huggett and Widdas 1951),
$$T = \frac{{P^{{\frac{1} {3}}}_{s} + 2.3377}} {{0.097}}$$
where Ps is the average fresh weight (g) of the foetuses.

To estimate conception and birth dates for each litter, we assumed a 120-day gestation period. Conception and birth dates were pooled in fortnights.

To estimate the probability of two randomly drawn birth dates belonging to the same fortnight, we calculated the Simpson’s index for each region. The maximum value of pi was used as a Dominance index,
$$D = {\sum\limits_{i = 1}^N {pi^{2} } }$$
where N is the number of fortnights and pi is the proportion of the ith fortnight.

Food supply has a strong influence on the reproductive phenology of wild boar (Fernández-Llario and Mateos-Quesada 1998); therefore, in our evaluation of each region, we focused on that habitat component. To evaluate six types of plant cover [holm-oak (Quercus ilex), cork-oak (Quercus suber), chestnut (Castana sativa), oak (Quercus robur), oats (Avena spp.)/wheat (Triticum spp.) and sunflower (Helianthus spp.)/maize (Zea mays)], we used several types of data, including aerial photography, soil occupation maps and woodland inventories. The data were analysed using GIS software (ArcView).

To measure the food diversity in each region, we used the number of cover types and their respective coverage in each of the regions surveyed. Diversity was assessed using a Shannon–Weaver index, which is defined as
$$H_{\gamma } = {\sum\limits_{i = 1}^C {pi\ln {\left( {pi} \right)}} }$$
where C is the number of cover types with valuable food, and pi is the proportion of the ith cover type in the sample area.

To determine whether region and reproductive phenology (birth frequency distribution) were statistically independent, we used a chi-square analysis (Spiegel 1993). In addition, we tested the null hypothesis that the correlation coefficient between the Shannon–Weaver index and the Simpson index of the populations does not differ from zero (Spiegel 1993).

Results

We examined 191 pregnant females from Trás-os-Montes (nTM=31), Beiras (nB=21), Alentejo (nA=86) and Extremadura (nE=53). The four regions differ in their birth frequency distributions (Fig. 2 and Table 1), although each of them has similarities with populations in other regions (Table 2). The wild boar population in Alentejo had the longest birth period and no discernable peak in frequency of births. In contrast in Extremadura, more than half of the births occurred during the first fortnight of March. In Trás-os-Montes and Beiras, the birth frequency distributions were intermediate between the extreme patterns observed in the wild boar populations of Alentejo and Extremadura. We conclude that there was a significant relationship between region and birth frequency distribution in the year 1999/2000 (χ2=48.44, df=9, P<0.01).
https://static-content.springer.com/image/art%3A10.1007%2Fs10344-005-0025-z/MediaObjects/10344_2005_25_Fig2_HTML.gif
Fig. 2

Compared fortnightly relative frequency distribution of births

Table 1

Simpson and Dominance indices according to fortnightly distribution of births

Population

Simpson index

Dominance index

Alentejo

0.13

0.17

Trás-os-Montes

0.22

0.29

Beiras

0.27

0.43

Extremadura

0.28

0.49

Table 2

Birth frequency distributions in wild boar populations in several regions of Portugal, Spain and other European countries

Region, Country

Birth distribution patterns

Studies

Alentejo, Portugal

A broad main reproductive period and possibly a late summer second period of births

This study

Beiras, Portugal

Evidence of only one reproductive period

This study

Extremadura, Spain

Farrowing tends to be seasonal and synchronized

This study

Trás-os-Montes, Portugal

Evidence of only one reproductive period

This study

Andalusia, Spain

Farrowing tends to be seasonal and synchronized

Fernández-Llario and Carranza 2000

Savoie, France

Farrowing tends to be seasonal and synchronized

Baubet 1998

Piemonte, Italy

A broad main reproductive period though births may took place all year around

Durio et al. 1995

Ticino, Switzerland

A broad main reproductive period and a late autumn/early winter second period of births

Moretti 1995

Piemonte, Italy

A broad main reproductive period though a second farrowing period may occur

Durio et al. 1992

Extremadura, Spain

Evidence of only one reproductive period

Garzón 1991

Andalusia, Spain

Evidence of only one reproductive period

Abaigar 1990

Castilla y León, Spain

Evidence of only one reproductive period

Sáez-Royuela 1987

Only food items that are known to strongly influence female body condition and influence the likelihood of reproduction were included in the analysis. The Extremadura study area had the minimum possible Shannon–Weaver index value, with only Q. ilex woodland as a favourable cover type (Table 3). The other regions in our study exhibited higher food diversity. The Alentejo study area had the highest food diversity index, followed by those in Trás-os-Montes and Beiras. Despite the apparent significant regional differences in reproductive phenology, Shannon–Weaver and Simpson indices were not significantly correlated (r=−0.60, df=2, P>0.05).
Table 3

Shannon–Weaver indices relative to the number and proportion of different cover types in four regions of Portugal and Spain

 

Shannon–Weaver

Holm-oak

Cork-oak

Chestnut

Oak

Oats/wheat

Sunflower/maize

Alentejo

1.17

0.55

0.2

0.15

0.1

Trás-os-Montes

1.11

0.1

0.45

0.4

0.05

Beiras

1.05

0.45

0.35

0.2

Extremadura

0

1

In the period of study, the proportion of females in breeding condition—pregnant or lactating—in age class 1 was much higher in Alentejo than in the other regions, and the lowest value was observed in Extremadura (Fig. 3).
https://static-content.springer.com/image/art%3A10.1007%2Fs10344-005-0025-z/MediaObjects/10344_2005_25_Fig3_HTML.gif
Fig. 3

Compared proportion of wild boar breeding females, considering each class of age, during the hunting season

Discussion

The significant relationship between region and birth frequency distribution in the year of our study (1999/2000) is unlikely to be a result of regional differences in photoperiodicity, which are small. Furthermore, because some of the females we examined would have given birth in November/December, conception can occur in July and August. That said, a decrease in reproductive activity in summer, which occurs in some domestic pig breeds raised in extensive systems (Dobao et al. 1983), likely occurs in these wild boar populations. The duration of our study (from November to February) prevented us from detecting all births; namely, those that occurred in late spring and early summer.

The within-group synchronization of oestrus (Delcroix et al. 1990) cannot explain the differences found between regional birth distributions; otherwise, the shortest birth distribution period should have occurred in Alentejo, where food abundance (Santos 2002) is known to foster large female groups (Fernández-Llario and Carranza 2000).

Food availability throughout the year appears to influence birth frequency distributions as well as reproductive phenology of the studied wild boar populations.

In Alentejo, the relatively longer length of the main reproductive period might be a result of high food availability observed between late autumn and late winter. Furthermore, a second reproductive period in August/September might occur because hunting bags included lactating females in November/December and post-partum females in early winter. The availability of oats and wheat in spring and summer, sunflower and maize in summer and autumn, and mast in autumn and winter permits such a birth frequency distribution. These results are consistent with those observed in other favourable habitats (Table 2) where nutritious food is available for an extended period, the first reproductive period tends to be long and a second farrowing period can occur (Durio et al. 1992, 1995; Moretti 1995).

In Extremadura, the supply of mast oak dictates one reproduction period only, which is characterised by a sharp peak in the first fortnight of March. Single, short reproductive periods are observed in harsh environments (Table 2), such as at high elevations or during drought conditions, where farrowing tends to be seasonal and synchronized (Baubet 1998; Fernández-Llario and Carranza 2000).

In the other two regions, Beiras and Trás-os-Montes, we did not detect a second reproductive period (Table 2) which is also not found in other Iberian ecosystems (Sáez-Royuela 1987; Abaigar 1990; Garzón 1991). Given the relatively small sample sizes from Beiras and Trás-os-Montes, our results must be viewed with caution. That said, the reproductive periods observed in Beiras and Trás-os-Montes are longer than that observed in Extremadura. Although most of the regional differences in birth frequency distributions are due to seasonal differences in food availability, we did not detect a significant negative correlation between the Shannon–Weaver and Simpson indices of the regions included in our study.

In wild boar, female reproductive phenology can vary according to age. Except for Alentejo, most of the young females can be delayed in reaching breeding condition (Fig. 3). This delay in reaching breeding condition by younger females explains most of the births occurred after the reproductive peaks (Fig. 2). In Mediterranean populations, female body weight and likelihood of reproduction are correlated (Massei et al. 1996). In Extremadura, Beiras and Trás-os-Montes, young females barely attain the body weight needed to induce oestrus before April, which is much different than the situation in Alentejo, the region with the highest food diversity index, where at least 60% of the juvenile females attained the threshold weight needed to reproduce. The relatively short duration of our study did not allow us to determine whether the young females attained puberty and bred during late spring or early summer, which occurs in some wild boar populations in Italy (Monaco et al. 2003). Adult females appear to attain breeding condition using food resources other than oak, such as grass, roots and tubers, which they obtain by foraging along riverbanks. Some adult females are thought to give birth three times over 2 years, which would explain the farrowing outside the reproductive peaks. In addition, in the case of abortion, adult females might replace the aborted litter and give birth before the following spring. Aujeszky’s disease, also known as pseudorabies, is an example of a disease that has high levels of seroprevalence in some wild boar populations in Spain (Vicente et al. 2005), which can cause abortions and stillbirths in wild boar.

Although seasonal variation in precipitation is a characteristic of the Mediterranean climate, a more definitive characteristic is interannual variability in rainfall which causes Mediterranean habitats to exhibit significant interannual variability in food production, which we did not address in this study.

An understanding of the reproductive phenology of wild boar populations is useful in planning hunting seasons. If wild boar hunting is expected to sustain itself on bio-ecological and ethical grounds, the timing and duration of hunting seasons should not be the same in all regions. Rather, the hunting season should be dictated by, in part, regional reproductive phenology, which can vary annually. Given the lack of predictive models that are sufficiently accurate to determine the timing and duration of hunting seasons a priori, annual seasons should be based on the continuous monitoring of the population’s reproductive cycle. In the Mediterranean Ibero Atlantic Province, regional and annually variable hunting seasons would be an important step towards helping hunters to avoid shooting nursing females.

Acknowledgements

We are especially grateful to the estate owners that permitted us to collect data on their property. We thank Alfredo Garcia, António Silva, Artur Silvério, Carlos Pereira, Helena Almeida, Joaquim Sousa, Miguel Cortes, Ricardo Carvalho and Rui Morêda for support with the fieldwork. We gratefully acknowledge the financial aid of PAMAF and PRAXIS programmes (project PCNA/BIA/184/96 and grant XXI-BD/13394/97). Special thanks to Dr. Catarina Eira for reviewing the English of the manuscript. Our experiments complied with the laws of Portugal and Spain.

Copyright information

© Springer-Verlag 2006