1. Introduction
Animal-based measures (ABMs) are considered the most reliable indicators to assess the welfare status of an animal and to identify the risk factors in the management and environmental conditions [
1]. They include a large variety of indicators such as behavior, clinical signs (e.g., skin lesions, pathologies), and physiological and productive parameters. ABM
S allow measuring how a single animal (or a group of animals) reacts to environmental and management stressors. One of the main stressors in intensive pig herds is the fluctuation of the (indoor) environmental temperature, particularly when it overcomes the thermoneutrality threshold of the animal [
2]. Further causes of environmental stress can be attributed to excessive relative humidity (rH) and harmful gas concentrations (e.g., CO
2 and NH
3) [
3,
4]. Concerning temperature control and gas removal, ventilation efficiency plays the main role, and it is mainly dependent on the ventilation system in the barn [
5]. At present, pig barns in the Mediterranean area are mainly wind-driven and, thus, naturally ventilated, while mechanical ventilation systems are less common [
6]. In naturally ventilated buildings, the external wind influences the indoor velocity magnitude and distribution [
7]. In these systems, the accurate control of the indoor conditions is not always feasible, especially on warm days, when the windows and vents of the barns are fully open [
8]. In mechanically ventilated barns, the ventilation efficiency can vary depending on the operational conditions of the ventilation system [
9]. In both natural and mechanical ventilation systems, the ventilation efficiency can vary depending on the geometry of the piggery structure; therefore, it should be carefully evaluated.
Inadequate ventilation, leading to subsequent changes in temperature, humidity, and presence of gas and dust [
10,
11,
12], has been found linked to pig behavior, health, and physiology [
9]. For some ABMs, the effects of poor ventilation on animal health are well known; for example, an insufficient air exchange can increase the occurrence of respiratory disease and thermal stress [
7,
13], while contrasting results have been observed with regard to animal behavior. In the case of aggressive behaviors and outcomes such as lesions, specific studies are lacking. Some studies reported that heat stress can lead to the development of aggressive behavior and consequent skin lesions [
14]; other works showed that high CO
2 concentration might induce higher inactivity rates of pigs and could increase the risk of overloading with subsequent occurrence of lesions in the middle area of the body and/or the prevalence of bursitis due to the prolonged contact of bone prominence with the floor [
10].
Furthermore, tail biting and relative tail lesions have been hypothesized to be influenced by the ventilation conditions (i.e., magnitude air velocity, air direction, and air exchange) and typologies (i.e., natural or mechanical). Hunter et al. [
15] observed a higher presence of lesions in the case of mechanical ventilation than in the case of natural ventilation. Therefore, it is important to consider that, according to Hunter et al. [
15], the natural ventilated building considered in the study was provided with straw litter, which is considered the “gold standard” to prevent tail biting behavior, and this aspect might have biased the results. On the contrary, the study by Scollo et al. [
16] reported higher frequencies of tail lesions in the farms adopting natural ventilation.
Lastly, no studies have been found on the effect of ventilation on the emotional state of fattening pigs. Today, it is worth noting that animals need a positive emotional state to be in a positive welfare condition [
17]; therefore, ABMs to assess pig emotional state have been proposed, such as qualitative behavioral assessment (QBA), tail posture, and tear staining [
18,
19]. It has been questioned if the impact of ventilation, dust, and air quality could be a confounding factor for some emotional state indicators [
20,
21]. It has been hypothesized that temperatures out of the acceptable range can influence tail posture, which is also an indicator of tail biting behavior in a group of pigs, and pig emotions, leading to misinterpretation of the measure itself. Similarly, high dust and gas concentrations can mislead the interpretation of tear staining as an indicator of emotion in pigs, due to inflammation of the eye and conjunctiva by poor air quality [
20].
Lastly, previous research found evidence that the age and weight of the pigs might also influence their behavioral response toward housing conditions, temperature, and air quality [
22], stressing the need to consider this variable when assessing the effect of ventilation on ABMs in a productive cycle.
Being able to identify the relationships between ABMs and ventilation conditions might be helpful to prevent the occurrence of harmful social behaviors and, therefore, increase pig welfare conditions.
This study hypothesized that there would be significant associations between housing condition management and behavior, lesions, and emotional state in undocked pigs during the fattening phase. Therefore, the first aim of the study was to quantify and qualify the main welfare issues of the pigs. The second aim was to define and compare the ventilation conditions in two different buildings of a case study farm, using computational fluid dynamics (CFD) simulations, and to investigate possible relationships between welfare indicators and housing and ventilation conditions, to evaluate how these can impact pig welfare.
4. Discussion
This study quantified and qualified the main welfare issues of pigs raised in two different buildings of the same farm. The ventilation strategy, as assessed by the CFD simulations, showed remarkable variability in the ventilation conditions of each building across the three observation days.
Overall, the QBA assessment showed that animals in the mechanically ventilated pens were in a more positive affective state, in accordance with the higher ventilation performance of the MV building, characterized by higher indoor air velocity. The QBA also evidenced a worsening in the affective states increasing with the age of the pigs. This last effect might also depend on the reduction in space allowance and the increase in temperature during summer, as well as changes in pig physiology, as previously reported [
37,
38,
39]. Therefore, the comparison between the two buildings was performed separately for each observation time (i.e., T1, T2, and T3).
At T1, pigs in the MV group showed lower tail LSI compared to NV (
Table 5) and a higher proportion of pigs with tail position up (
Table 4). Tail lesions are the outcome of tail biting behavior. Tail biting is currently considered an iceberg indicator of poor welfare, having a negative effect on the emotive state of pigs [
40]. Tail biting is an abnormal behavior, and its occurrence has been found to be strongly dependent by many managerial and environmental factors [
41]. In accordance with the result of this study, Lahrmann et al. [
27] proposed that assessing tail position would allow quickly identifying tail-bitten pigs since these pigs would keep a low tucked tail, while pigs which show few or no tail lesions would keep the tail curled and “up”.
In contrast, the behavioral analysis showed a higher frequency of negative social behavior in the group in the MV building, as well as higher stereotypy frequency. Despite that, lesion outcomes were not significantly different between the two animal groups in the different buildings. A discrepancy between these two indicators (behavior and lesion) was previously observed in other studies [
24]. A possible explanation is that the lesions are the consequence of negative social behavior that occurred in a range of time (days or weeks), while the behavioral analysis in this study was a picture of the exact moment of the assessment since they were recorded by direct observation. Moreover, a limit of the present study was that behavioral analysis was carried out using direct observation; thus, although the behaviors were recorded in the same range of time, the observations were not conducted simultaneously. Pigs in the MV building showed an indeed higher score in tear staining and dirtiness, as compared to the NV building. Tear staining is the presence of a red stain in the left eye of a pig, as a consequence of the production of a red pigment by the eye pituitary gland. In pigs, it has been proposed as an indicator of negative emotional state because of a correlation with processing negative emotions [
20,
42]. Other studies have hypothesized that tear staining might also be stimulated by excessive gas concentration, dustiness, pen soiling, or other environmental conditions [
24,
43]. On this observation day, the MV building group showed a higher proportion of dirty pigs. Pig soiling has been frequently linked to higher gases in manure [
44], and it might explain tear staining. The indoor air velocity was similar in the two buildings (
Table 7); however, the higher number of pigs/pen with lower space allowance in the MV building compared to NV at T1 might also have enhanced this mechanism.
At time T2, behavior and lesions did not show any differences, except for overall higher stereotypies in the pigs in the MV building, mainly due to the percentage of pigs showing ear biting behavior. Similar to tail biting, ear biting has been considered an indicator of poor welfare so far [
45]. Among the predisposing factors for ear biting, air quality has been reported to influence its occurrence [
46]. In MV building, the results showed a higher concentration of CO
2 as compared to NV building. CO
2 is a product of respiration, which is heavier than oxygen; therefore, it has been found to fluctuate at the pig level. It is likely to presume that, on this observation day, the inhomogeneity of the airflow and speed was not efficient to remove CO
2. Moreover, CO
2 was found to be highly related to the number of animals. The MV building had one more pig per pen and showed a lower space allowance, contributing to an increase in the CO
2 indoor concentration. This result might explain the higher presence of ear biting in this group.
Behavioral analysis evidenced also a high proportion of pigs showing dog sitting behavior on T2 and T3 observation days, in both buildings. Dog sitting has been considered a non-aggressive stereotypic behavior and an indicator of suboptimal welfare in pigs [
47,
48]. According to the study by Scollo et al. [
49], pigs reared in intensive conditions increased the frequency of sitting behavior when space allowance decreased, e.g., in the fattening phase. This has been interpreted as the lack of space to lie down [
49] or the consequence of boredom, leading to severe cognitive deprivation due to the barren environment [
50,
51]. A combination of the two factors might explain the results of the present study. Heavy pigs have a very restricted area available at the end of the cycle (because the current legislation states that pigs above 110 kg require min 1.00 m
2, and, in heavy pig production, pigs can reach up to 180 kg at the end of the rearing period). Moreover, the behavioral analysis showed that the enrichment devices available to the pigs (metal chain and a metal chain with wood) were of marginal interest since pigs spent most of their time exploring the pen and very little time on the enrichment devices. Exploring the pen (over-exploring) has been considered another sign of boredom and poor welfare in intensive pig farms since the pens are usually in barren environments that do not provide cognitive stimuli to the pigs [
52].
When considering T3, behavioral analysis evidenced a higher frequency of low tail position and negative social behavior in MV compared to NV. A low tail position has been previously associated with tail lesions; however, at this assessment, no differences were observed for tail LSI. It is important to consider that, at T3, the two buildings raised the maximum score in dirtiness, corresponding to almost all pigs in each pen having manure on >50% of the body surface; therefore, this condition might have biased the results from the lesion assessment. Pig soiling is considered the outcome of abnormal eliminative behavior in pigs [
44]. Normally, pigs on a partially slatted floor tend to release urine and/or feces on the slatted floor and rest on the full floor. When certain predisposing factors occur (see later), pigs can develop abnormal behavior, which leads to pen and pig soiling. One of the main identified factors is thermal discomfort [
53]; in fact, with high temperature, pigs raised indoor tend to rest on the slatted floor and release urine and/or feces on the full floor [
54]. In very severe heat stress conditions, pigs tend to release urine and/or feces, as well as rest, on the full floor with the purpose of heat loss [
37]. This latter condition has been considered an indicator of poor welfare since, in normal conditions, pigs prefer to avoid contact with their excreta [
44]. The optimal temperature range for heavy pigs (140–180 kg of live weight) is estimated to be 18–20 °C. Therefore, at T3, the temperatures in the two buildings were very challenging for the pigs (29–30 °C on average), and neither type of ventilation was able to significantly reduce this temperature. The indoor ventilation was consistently different at T3. The MV building showed high air velocity in the extremities, compared to the central zone. On the other hand, the NV building showed homogeneous low air velocity throughout the building length. This difference could have affected CO
2 concentration measured, which was significantly higher in the NV building as compared to MV. Accordingly, in the NV pens, higher frequencies of polydipsia were observed. Polydipsia is a stereotypy that can occur when pigs are submitted to heat stress, in an attempt to cope with hot temperatures [
55]. Moreover, behavioral analysis observed also significantly lower inactive behavior in NV pens compared to MV ones. Housing conditions also revealed that temperature, light, CO
2, and number of pigs per pen were higher in the NV case compared to MV. Those factors might have influenced the pigs’ behavior. Some studies have observed an increase in activity and aggression in the presence of high temperatures, due to heat stress and difficulty in finding a comfortable place to rest for pigs kept under intensive conditions [
14,
44]. In the present study, negative social behavior did not differ at T3, while a trend of more front LSI in NV pens was observed. Other studies, in contrast, observed an increase in lying behavior at high temperatures [
37]. The difficulty in finding a lying place could be exacerbated when the number of pigs per pen increases, as in the NV pen group. Similarly, some studies reported that increasing illumination in the pig farms can lead to an increase in activity, which does not impair pig welfare [
56]. In accord with the results, CO
2 concentration was found to be directly proportional to pig activity by Zong et al. [
57].
When the temperatures in the two building buildings were challenging, the higher air velocity in MV pens, even if not able to decrease the indoor temperature, could have contributed to a reduction in the heat perception at the pig level, as well as to a reduction in CO
2 concentration, thereby influencing pig behavior and contributing to improving their welfare [
10]. One limitation of the study was that the measurements could not be performed on the same day in both buildings, due to the farm flow chart, according to commercial agreements between farmers and buyers. However, this is the first study aimed at integrating ABM assessments and environmental measures provided by CFD simulations in heavy pigs. These preliminary results pose new questions regarding the effect of the interplay between outdoor and indoor conditions and ventilation systems on pig welfare, which will be further investigated.