*3.2. Effects of Selected Environmental Parameters on the Presence of Welfare Indicators* The average CO<sup>2</sup> and NH<sup>3</sup> values and temperature within each turkey facility obtained at each assessment are presented in Table 5.



**Note:** Population of birds examined in each visit: first N**male** = 19,762, N**female** = 17,754; second N**male** = 16,638, N**female** = 15,194; third N**male** = 16,222. Significance: \* *p* < 0.1; \*\* *p* < 0.05; \*\*\* *p* < 0.01.

> The average CO<sup>2</sup> values ranged from 850 to 4433 ppm. The values above the recommended (CO<sup>2</sup> < 2500 ppm) were recorded at the first assessment in flocks 1 and 2. At that time point, the NH<sup>3</sup> level in both facilities exceeded the anticipated level of 20 ppm. At the second and the third assessment, NH<sup>3</sup> values were low in all flocks and did not exceed 4 ppm. The CO<sup>2</sup> values were also low and ranged from 850 to 2200 ppm. Inside temperatures recorded at all three visits were slightly higher than recommended for the specific age of turkeys [27].

> Table 6 shows the influence of different CO<sup>2</sup> and NH<sup>3</sup> levels on occurrences of welfare indicators and mortality.

> For NH3, the probability of occurrence of welfare indicators was lower at concentrations of NH<sup>3</sup> > 0 ppm compared to NH<sup>3</sup> = 0 ppm; for CO2, the possibility of the development of welfare indicators was higher at CO<sup>2</sup> < 1600 ppm compared to CO<sup>2</sup> between 1600 and 3000 ppm and more than 3000 ppm.

Dirty

Sick

Dead

Pendulous crop

Agression towards birds

third N**male** = 16,222. Significance: \* *p* < 0.1; \*\* *p* < 0.05; \*\*\* *p* < 0.01.

**Note:** Population of birds examined in each visit**:** first N**male** = 19,762, N**female** = 17,754; second N**male** = 16,638, N**female** = 15,194;

*F* 3 (0.017) 174 (1.145)

*<sup>M</sup>*4 (0.020) 1.74 0.081 \* 343 (2.062) 6.00 0.001 \*\*\* 86 (0.530) *F* 0(0) 183 (1.204)

*<sup>M</sup>*6 (0.030) 2.14 0.033 \*\* 4 (0.024) −1.98 0.048 \*\* 17 (0.105) *F* 0 (0) 11 (0.072)

*<sup>M</sup>*0 (0) 2 (0.012) 1.35 ns 12 (0.074) *F* 0 (0) 0(0)

*<sup>M</sup>*0 (0) 10 (0.060) −1.04 ns 9 (0.055) *F* 0 (0) 14 (0.092)

*<sup>M</sup>*0 (0) 0 (0) 2 (0.012) *F* 0 (0) 0 (0)

*<sup>M</sup>*0 (0) 0 (0) 2 (0.012) *F* 0 (0) 0 (0)

**Figure 1.** Mean values of welfare indicators recorded in meat-type turkeys at 3 to 4 (Assessment I), 13 to 14 (Assessment II), and 17 to 19 weeks of age (Assessment III). **Figure 1.** Mean values of welfare indicators recorded in meat-type turkeys at 3 to 4 (Assessment I), 13 to 14 (Assessment II), and 17 to 19 weeks of age (Assessment III).


In the first assessment, the most frequently observed welfare indicators in both males and females were small size and immobility. Overall, 0.997% of the males and 0.721% of **Table 5.** Mean temperature and CO<sup>2</sup> and NH<sup>3</sup> values in turkey facilities at each assessment.

**Note**: Recommended values according to Aviagen: ambient temperature at 3–4 weeks = 23–25 ◦C, at more than 9 weeks = 16–17 ◦C; NH3: <20 ppm; CO2: <2500 ppm [27].

**Table 6.** Welfare indicators' prevalence (including mortality) by different CO<sup>2</sup> and NH<sup>3</sup> levels.


**Note: \*** Significance at *p* < 0.1; \*\*\* *p* < 0.01.

### **4. Discussion**

In Slovenia, turkeys represent less than 2% of poultry meat production and, until this study, no data were available on the welfare profile of this species. The aim of this study was to identify transect based on-farm health and welfare indicators in commercial turkey flocks of both sexes during the fattening cycle. Flocks, separately for males and females, were assessed at 3 to 4 weeks of age, 1 week before slaughtering the hens, and 1 week before slaughtering the toms. In brief, we found several differences in the frequency of welfare indicators between different assessments and between male and female populations.

To date, no data on the welfare of turkeys based on the transect walk approach at the age of 3 to 4 weeks are available. The results of our study showed that small birds were identified as the main welfare problem at this age. Such birds were found in both sexes, although significantly more small birds were observed in the male population. The second most common indicator was immobility. A high incidence of smaller birds and immobility may indicate compromised health and welfare on a farm due to either general housing or bird health problems. Infections caused by *E. coli* are commonly present in young chicks. Localized infections such as omphalitis and yolk sac infection or systemic colibacillosis generally resulted in higher mortality in the first weeks after placement. Affected birds are usually undersized, because they may have difficulties in walking, which alters weight and leads to weakness [25]. Of the flocks included in the study, colibacillosis was diagnosed in one flock and the birds were treated with antibiotics. Although no veterinary intervention was required in the other flocks, the cumulative mortality indicated that health problems due to *E. coli* or other unidentified infections were present in at least one other flock.

Excessive NH<sup>3</sup> and CO<sup>2</sup> levels may also have negative impact on birds' health and metabolism in young turkeys [14,28]. It was shown that young poults exposed to 4000 ppm CO<sup>2</sup> had suppressed body weight gain compared to those exposed to 2000 ppm [14]; in addition, NH<sup>3</sup> levels greater than 10 ppm can also reduce feed intake with effects on body weight [28]. In our study, higher NH<sup>3</sup> (40 ppm and 27.6 ppm) and CO<sup>2</sup> (4433 ppm and 3000 ppm) values were detected in two facilities. Elevated levels of both gasses directly correlate with reduced ventilation. Under field conditions, such a situation is often seen in the first weeks after placement of turkeys due to reduced heating costs [29]. Nevertheless, the higher mortality as well as the significantly higher incidence of sick birds found in males indicate that health problems seem to play a more important role than environmental conditions, which were equal for males and females.

As shown in other studies [9], the most problematic period for both sexes in mixed commercial turkey flocks seems to be the time before slaughtering the hens. Indeed, more indicators were present at the second assessment than the first, and the overall prevalence of altered birds was higher compared to the first and third assessments. The most common findings were dirtiness and poor feather condition, followed by immobility. Dirty feathers were observed in more than 2% of males and in 1.204% of females. Dirtiness before slaughtering the hens has also been reported by other authors, although not with such a high prevalence as in our study. In a study conducted in Norway, an average of 0.36% dirty males and 0.15% females were observed at 11 weeks [9,11,30]. In Italian commercial turkey flocks, dirtiness was recorded in 0.022% of females. Unfortunately, males of this age were not scored [8], and so no direct comparison was possible. Previous studies have shown that poor litter quality, dust, and high stocking density can significantly impact dirtiness [1]. Unfortunately, litter quality was not scored in our study. CO<sup>2</sup> and NH<sup>3</sup> values that reflect inadequate ventilation and poor litter management [31] were low in all facilities, but the measurements were performed only during assessments so no relevant conclusions could be made on the significance of poor litter quality on such a high prevalence of dirtiness. Recently, dirtiness in turkeys was found to be correlated with immobility and lameness more than with poor litter quality [9]. In our study, immobility was the third most frequently observed indicator at this age. Similar to previously reported findings [8,32], this occurred significantly more frequently in males than in females, which could explain why males were significantly dirtier. Dirty feathers have been suggested as an indicator

of health problems of the digestive system [33], but it is unlikely that necrotic enteritis diagnosed at 5 weeks of age in five of six flocks contributed to such a high percentage of dirty birds observed 7 weeks later. Necrotic enteritis is an acute disease caused by toxins produced by *Clostridium perfringens.* The course of the disease is usually short, and birds respond very well to antibiotic treatment if it is given immediately after the onset of the disease [34,35].

Featherless condition was the second most frequently observed welfare indicator during this period. Missing or damaged feathers, particularly in the tail region, were observed in 0.986% of males and in 1.145% of females. The etiology is not entirely known, but it is likely that poor plumage is the consequence of feather pecking. In turkeys, mild feather pecking can be a form of social or investigative behavior [36,37]. When pecking becomes more severe, it may result in severely damaged feathers and feather loss, or even cannibalism [38]. The incidence of feather pecking is known to increase with age, although damaging pecking can occur as early as the 1st or 2nd week of age [39]. Under field conditions, high stocking density, inappropriate lighting, feed deficiency, breed, and sex are considered to influence injurious feather pecking [2,8,40]. Such aggressive pecking often results in wounds seen on the head, around the tail, on the wings, and in the back region. In our study, injuries were rarely found and were observed in both sexes, although vent wounds were found significantly more frequently in males. This is consistent with the results from commercial turkey flocks in Norway [9] and Italy [8]. To prevent feather pecking and aggressiveness, beak trimming is still common practice in commercial turkeys [40], but it seems that beak trimming does not play an essential role in preventing poor feather condition. The incidences of featherless birds observed in our study and in Italian commercial flocks [8]—both included beak-trimmed birds—were higher compared to the study performed in non-trimmed commercial flocks in Norway [9].

The percentage of sick birds was low, but the difference between the sexes was significant. For classifying a bird as sick, other studies also included birds with pendulous crop [7–9,11,30]. In our study, birds with pendulous crop were recorded separately, and so no direct comparison could be made. The etiology of pendulous crop is yet not fully understood, but hereditary predisposition, dietary influence, increased liquid intake in hot weather, and the effect of lighting period have been suggested. Unfortunately, no treatment is available, and the carcasses of affected birds are usually condemned at processing [41–43]. In our study, birds with pendulous crop were observed in both sexes, although at much lower frequencies compared to the study performed by Vermette et al. [32]. In their clinical study, pendulous crop was found to be the second major reason for morbidity and mortality in turkeys, and females were significantly more affected than males. In comparison to the first assessment, the size of birds was more uniform. This supports previous suggestions that the number of small birds decreases with age [30].

At the last visit, only male turkeys were assessed. Compared to the second assessment, the prevalence of immobile as well as lame, sick, and dead birds increased, indicating health problems, most likely caused by poorer leg health. Lameness and immobility as its consequence are important welfare and health issues in commercial turkey flocks, especially in males [8,32]. Due to their heavy weight and longer production cycle, males' legs are exposed to more stressors, resulting in chronic pain and movement difficulties [23]. In addition to degenerative and development disorders, bacterial and viral infections such as *Staphylococcus aureus*, mycoplasmas, and reovirus are involved in skeletal and joint lesions, causing acute or chronical local inflammations or even systemic septicemia, resulting in higher mortality [44]. Because no further investigations were performed, we do not know the exact causes of immobility. Dirty and featherless animals were frequently observed, although the prevalence was much lower than in the second assessment. This is consistent with recent findings that poor feather condition decreases with lower stocking density [20,26], but may still persist within the flock due to leg problems [9]. The incidence of head, back, and vent wounds also decreased to less than half after depopulation of females. These results are in agreement with previous findings that injurious feather

pecking and wounds may be a consequence of behavioral disturbances due to high stocking density [1]. In the EU, specific minimum stocking density requirements have been established for broilers [45], but these do not directly apply to turkeys. In some countries, such as Norway, specific regulations depending on the live weight of turkeys have been adopted [9] or recommendations for turkeys have been published [46]. Unfortunately, no such documents are available in Slovenia.

During the production cycle, NH<sup>3</sup> and CO<sup>2</sup> are the prevalent gases in turkey facilities [47]. NH<sup>3</sup> in high concentrations can have severe adverse health effects, causing lesions of the upper respiratory tract and inflammation of the cornea and conjunctive [15,48]. High concentration of CO<sup>2</sup> can be harmful to turkeys due to hypoxia [14], which may lead to dilatated cardiomyopathy [49]. Although no precise concentration limits have been established for turkeys, Directive 98/58/EC [50] requires that gas concentrations in turkey facilities be kept within safe limits. For broilers, EU regulation established the maximum NH<sup>3</sup> concentration inside a poultry barn at 20 ppm, and for CO<sup>2</sup> the concentration should not exceed 3000 ppm measured at the level of the chickens' heads. If higher levels are detected, corrective actions must be taken [45]. In Slovenia, most commercial turkey farms are equipped with sensors for temperature and humidity, but gas concentrations are not measured on a daily basis. Apart from higher CO<sup>2</sup> and NH<sup>3</sup> values recorded at two facilities in the first assessment, the levels were low in all facilities in the next two assessments. Similar results were also obtained in the study performed in commercial turkey flocks in Poland, where a significant decreasing trend was observed during the production cycle [47]. Although accurate limits should be defined for turkeys, our results indicate that NH<sup>3</sup> levels of 40 ppm or less did not influence the occurrence of welfare indicators. Moreover, a higher probability could be expected at NH3 = 0 ppm compared to more than 0 ppm. For CO<sup>2</sup> the probability of occurrence of animal welfare indicators was higher at levels below 1600 ppm than at levels between 1600 and 3000 ppm or above 3000 ppm. The reason for this could be that birds exposed to slightly elevated NH<sup>3</sup> or CO<sup>2</sup> concentrations are likely to be less active. Similar results were recently obtained by Candido et al. [14], who found that poults housed at a lower CO<sup>2</sup> level (2000 ppm) showed reduced movement compared to those exposed to higher CO<sup>2</sup> concentrations. However, further studies should be performed to obtain a balance between welfare and the optimal production of turkeys.

#### **5. Conclusions**

Compared to other poultry species, there is a lack of field studies on welfare problems in commercial turkey flocks. In this study, we investigated some aspects of health and welfare in commercial turkey flocks of both sexes in Slovenia. We cannot assume that our limited sample of flocks is representative of the commercial turkey industry in Slovenia, but it provides an estimation of problems that may exist during the production cycle and emphasizing the importance of setting specific standards and regulations regarding levels of harmful gases and stocking density for commercial turkeys. Our study confirmed that assessing welfare using transect walk approach performed in different times during the fattening cycle provides important information on animal health and welfare and could help farmers improve welfare in commercial turkey flocks.

**Author Contributions:** Conceptualization, O.Z.R., M.Z.Š. and A.D., methodology, O.Z.R.; software, H.H.; formal analysis, H.H.; investigation, O.Z.R., J.B.M. and A.D.; resources, M.Z.Š.; data curation, H.H.; writing—original draft preparation, N.M.H., H.H. and O.Z.R.; writing—review and editing, M.Z.Š., O.Z.R., A.D., Z.Ž., J.R., B.S., J.B.M. and U.K.; supervision, M.Z.Š.; funding acquisition, M.Z.Š. and O.Z.R. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the Slovenian Ministry of Agriculture, Forestry, and Food and the Slovenian Research Agency, project no. V-4 1817, and by the Slovenian Research Agency, research core funding no. P4-0092.

**Institutional Review Board Statement:** The national Animal Protection Act (Official Gazette of RS 38/2013; 21/2018; 92/2020 and 159/2021) defines in 21 a. article that non-invasive and non-experimental clinical veterinary practices are not considered as a procedure on animals for scientific purposes.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Not applicable.

**Acknowledgments:** The authors thank Perutnina Ptuj d.o.o. for its support as well as all farmers for their kindness and willingness to help.

**Conflicts of Interest:** The authors declare no conflict of interest.

## **References**

