**4. Behavioral assessment during intensive rearing**

To our knowledge, this is the first comprehensive work providing results of all AP behavioral assays in gallinaceous birds to natural predators in non-contrived settings. For assessing AP behavior, I used the flight initiation test, predator test, novel object test, emergence test, and restraint test [10, 24, 31, 33, 34]. The tests were modified as per the requirements and nature of the Chukars. No birds were exposed to predators during breeding and farming as usually done in intensive farming. In this way, I aimed to assess the significant AP behavior in coveys exposed to the predators in the release pens. Different sample sizes reduced the chances of bias in behavioral monitoring.

#### **4.1 Flight initiation test**

The FI test was carried out to find the flight initiation distance (FID) for human intrusion in captivity. The flight initiation behavior was assessed individually by an observer walking toward the flying pen. Firstly, two observers (a volunteer and the author) started walking toward the pen while observing the birds. The recording of observations was started from a distance of 20 meters. Secondly, the observer walked along all sides of the pen. The minimum distance between the enclosure and the observer was 1.5 feet. Meanwhile, the different behaviors of Chukars during the intrusion and stimulation were also assessed and recorded. The FI response was characterized in three cohorts, namely, (1) Weak response: birds slightly changed their position or just walked; (2) Moderate response: Chukars took a short flight from the observer and disturbances; (3) High response: birds flew from one side of the flying pen to the other to escape the disturbance.

A total of 20 observations were carried out on 230 subjects to obtain results on the FI of the Chukar partridges (see Appendix 1). During data collection for the FI test, the sample size was kept at a minimum of n = 5 to a maximum of n = 21. The mean sample size was 11.5 1.18. These samples were collected from the birds found very near to the mesh wire. I categorized flight initiation responses into weak, medium, and high responses. The maximum number of birds showed high responses (n = 89), followed by medium (n = 71) and weak (n = 70) responses. The results showed that weak and high FI responses are associated with higher variance in the response. The Kruskal-Wallis test showed no significant differences (p = 0.1817) between the medians of all three behaviors. High responses were normally distributed as opposed to the weak and medium responses, while medium responses were more skewed (i.e., more abnormally distributed).

#### **4.2 Flight initiation distance FID**

Flight Initiation Distance (FID) data collection was carried out independently of the FI test. During the observations for FID, the intrusion distance for Chukars was noted. It is the distance between the human and bird right before the initiation of flight. The bird perceives an intruder (human) as a danger. The bird takes flight when the minimum distance is approached. An observer approached the birds of the flying pens from 10 m outside. When the bird took flight, the distance between the observer and the bird was measured using a measuring tape. A total of 278 observations were recorded. The response categories were divided into 5 cohorts.


#### **Table 1.**

*Showing the distance cohorts and frequencies (numbers and percentages) of FID in those distance cohorts.*

The observation was discarded if the bird being intruded did not take flight but just changed its position.

For finding the flight initiation distance of Chukar partridge, a total of 278 observations were recorded. The frequencies of these observations were divided into 5 cohorts (**Table 1**). I found a mean distance of 5.34 0.4 and a variance of 4.55. The minimum and maximum FID recorded were 1.6 m and 9.93 m, respectively. Most of the birds showed FID between 2 m to 4 m (38.85%) followed by 4 m to 6 m (23.38%). The maximum FID was shown by 16 birds (5.76%). Results also indicate that experimental birds have a high FI response in the case of humans. This is beneficial as the trained birds cannot be approached, trapped, or captured after release to the wild. Zaccaroni [21] used the human stimuli for eliciting fights in Rock partridge. That study found more distant escape responses as compared to the non-human intervened birds. Fernández-Juricic et al. [35] described longer FIDs in the species with larger groups or living in flocks, which makes sense based on the more conscious and sensitive birds at the edges or near the disturbances. Hingee and Magrath [36] further elaborated the phenomenon, finding that flight in some birds can cause other less-evoked birds to fly away.

#### **4.3 Predator test (PT)**

For Predator Test (PT), the ex-post-facto design in a semi-contrived manner was followed. The natural prevalence and exposure of both avian and mammalian predators were the stimulus for the PT. The AP behavior against raptors was studied during the daytime. Mammalian predators were active at night. The effect of their presence on the birds was assessed. The same types of variables or parameters were used for both types of predators. The behaviors of the birds were also filmed for strengthening the field observations. I noted the behavior of birds until the resuming of the normal activities (feeding or resting) of half of the birds being observed after the predator's appearance. I observed and recorded typical AP behaviors for the birds *viz,* vigilance, crouching, TI and escape. These parameters were characterized as (1) Vigilance: if the birds showed any sign of activeness i.e., outstretched neck with support on front toes, slow staggering gait and actively searching intimate environment; (2) Crouching: this freezing behavior was demarcated by the reduction in the movement or just squatting frequently; (3) Tonic immobility (TI): the persistent hunker position for >30s; and (4) Escape: walking and running, also including flying attempts.

*Intensive Farming and Welfare Regarding Anti-Predator Behavior of Chukar… DOI: http://dx.doi.org/10.5772/intechopen.106371*

#### *4.3.1 For avian predators*

For assessing AP behavior against avian predators, a total of 29 observations were recorded in 515 birds. I noted observations for no more than n = 30 birds. On the other hand, the least sample size was n = 5 birds. Descriptive statistics (see Appendix 2) shows that the mean sample size is 17.75 1.4. The maximum number of birds were found vigilant (n = 18) followed by crouching (n = 15), TI (n = 8), and Escape (n = 6). It can be depicted that crouching and vigilant responses were abundant among the birds. The Mann–Whitney pairwise test shows that there are significant differences in the means of all the behaviors except crouching and vigilance; TI and escape responses. The data also show that 6.65 0.8, 6.75 0.7, 2.44 0.4, and 1.89 0.3 are the means for vigilance, crouching, TI, and escape behaviors for the avian predators, respectively. The highest responses are observed the crouching and vigilance (**Figure 2**). This study evidenced that aerial predators can cause more crouching than vigilance, TI, and escape responses. There were slight differences between crouching and vigilance behaviors. Previous research supported the same results in free-living [37] and captive-reared game birds [38]. Both studies presented those partridges mainly crouched in response to aerial predators. The AP responses are usually accompanied by species-specific alarm calls [39].

#### *4.3.2 For mammalian predators*

To assess the behavior of the Chukar partridge upon the encounter with terrestrial predators, I carried out a total of 14 observations on 168 birds. The exposure was noncontrived; in other words, I did not have any impact on their encounter with predators. The results show that escape response was the most frequently recorded (n = 118) in the terrestrial predator case, followed by vigilance (n = 26) and crouching and TI (n = 8 each) (Appendix 3). These results are represented graphically (**Figure 3**). The means for all of the categories of behavior are significantly different (p < 0.05). Moreover, Mann–Whitney pairwise test showed significant differences in the means of escape and vigilance, TI and escape. Moreover, there are no significant differences in the means of TI and crouching; TI and vigilance.

Undoubtedly, this study also cleared that escape reaction was the most prevalent behavior in the case of terrestrial predators (see Appendix 3). The least prevalent responses were vigilance, crouching, and TI. By contrast, [38] argued that vigilance was the most common reaction for the exposure to terrestrial predators. It indicated

#### **Figure 2.**

*Graph showing bar chart with standard error (SD) and deviations in the plots at 95% confidence interval for PT against the avian/aerial predators.*

#### **Figure 3.**

*Graph showing a bar chart with SD and deviations in the plots at 95% confidence interval for PT against the mammalian/terrestrial predators.*

the good response to the terrestrial predator by the present birds. Partridges respond differently to different predators [39]. For example, common behavior against terrestrial predators is vigilance, for easy detection and to avoid attack on them. Alternatively, against avian predators the combination of camouflage, TI and crouching can be effective. Interestingly, it was also observed that birds resumed their normal maintenance activities promptly after the exposure to the predators.

The present study had strengths over previous ones for exposure of living predators in de facto design and vast space use. The limitations of the previous results can be attributed to their small-sized experimental cages as compared to those used in our study. Earlier studies by [33, 38, 39], were conducted in cages or pens of very small sizes as compared to the present study. As stated by [40], the experimental cages can disrupt normal physiology and illicit the stress response during the initial transference in the cages. The conducting of experiments in smaller enclosures might have affected the results of the previous studies. Furthermore, the use of dummy predators can be beneficial due to: (1) The experiments and exposure can be presented at one's disposal, (2) Birds cannot be attacked by the dummy predators, and (3) The AP responses of Chukar are stronger for the live predators than dummy predators [41]. Even in captivity, the living predators may cause the deaths of birds in captivity. Meanwhile, the exposure can also stimulate short-term AP behavior in conspecifics in the same pen and the long-run survival in the wild. This behavior can be AP calls, alertness, or flying rigorously. In previous studies, the proper wild behavior would not have been shown by the experimental birds.

#### **4.4 Novel object test (NOT)**

For the novel object (NO), test birds were not isolated from the flock. A camera was placed both in and alongside the mesh wire to test the neophobia i.e., the fear of new things. The camera was kept on for recording video. When placing the camera, the disturbance to the birds was kept minimal. If more than three birds were disturbed during the placement of the camera, the observation was discarded for analysis. An observer also carried out a behavioral assay from 40 feet away. A pair of Russian Sehfeld Military binoculars was also used to see birds clearly. The test was continued for 5 minutes or until the birds showed the sign of acclimatization to the object. The NO test was dissolved in the following parameters: (1) Alert and Gazing: birds only gazed at the novel object (camera) and stood alert with no motion, (2) Approach: they *Intensive Farming and Welfare Regarding Anti-Predator Behavior of Chukar… DOI: http://dx.doi.org/10.5772/intechopen.106371*

#### **Figure 4.**

*Experimental settings of the NOT with a covey of 7 Chukars (one behind the pillar) standing vigilant against the novel object (camera).*

approached the camera, and (3) Escape: they were fearful of the object and flew away. All three categories of the responses were kept mutually exhaustive.

For NOT, a camera was introduced in the pen and different responses to that novel object were observed (**Figure 4**). The NOT was done on a total of 291 birds. The NOT showed that alertness and gazing was the most frequently observed behavior (n = 166). The approach to the introduced novel object and escape responses were on second (n = 90) and third (n = 35) numbers in frequency. The mean value for the alert and the gazing response was 6.64 0.58, for approach 3.6 0.49. The median value for escape was 1. The ANOVA and Mann–Whitney pairwise test showed that there are statistically significant differences (p < 0.05) present in the means of all the behaviors associated with NOT. The descriptive results have been tabulated in Appendix 4.

The briskness in the NOT shows that birds were aware of the danger associated with the introduced object. Very low number of birds perceived the NO as a real danger and escaped, while others approached the object to explore it. Such exploration in the wild would result in the death or trapping of the birds. Furthermore, the birds were not isolated from the covey, so definitely proper behavior against NO is crucial. When birds were in a group, they approached the NO massively (Appendix 4). This finding is in contrast with a study by [42], who found that ravens show neophobia (phobia of new objects) when they are in groups. In other words, they approached the NO more quickly when alone as compared with when they are with conspecifics. Here, partridges preferred to escape when they were alone or in a small covey.

#### **4.5 Escape/emergence test (ET)**

ET was done during the transportation of birds into the flying or releasing pens. Transporting modules (80 cm length x 60 cm width x 40 cm height) made of wood were used. A single module has two compartments, each 20 cm in height. All other sides of the module had meshed wire, except the roof and floor sides. Birds could see outside from the sides of the modules. Birds were captured from the training pen

using a hand net and kept in the poultry module for 5 minutes for acclimatization. Then birds were carried to the flying pen for liberation. The doors of the modules were opened without disturbance and observers stood as far as possible. The emergence observations were carried out on all birds for 3 minutes. After 3 minutes the test was terminated, and the remaining birds were deemed to show no escape reaction. The escape behavior was divided into three categories: (1) Immobile: after opening the door of transporting module, birds remain inactive or not tried to escape within 3 minutes of the test; (2) Walking: it was an escape response of relatively lower intensity, birds just walked out of the module briskly; and (3) Flying: birds emerged under 3 minutes and took a flight to escape of the module.

ET was performed during releasing of birds to the releasing pen. The escaping pattern of birds from the transporting modules was noted. Results showed that the maximum number of birds (n = 360) did not bother to escape out of the module and remained immobile (24.0 2.9) m (see Appendix 5). The second most frequent response (n = 278) was the flying response (18.53 2.4). It is followed by the least frequent (n = 112) walking response (7.46 1.0). A minimum of 5 birds were found immobile in contrast to 0 and 2 for walking and flying in a single batch. Tukey's pairwise test confirmed the significant comparison between walking and immobile and flying and walking categories. I found no statistically significant differences between the means of immobile and flying categories (p = 0.2123). It is also clear that most of the birds (n = 270) did not try to escape from the boxes at all. The escape from the boxes simulates the potential danger in the intimate environment or attack on the conspecifics. Rather, the majority of the birds preferred to stay motionless in the boxes, while a few others preferred to take flight in a possible perilous situation. The birds flew with a short sound of "keer" [39]. The calls by the escaped birds might elicit stimuli in the birds in boxes for the potential perilous situation outside. Despite this, a work by [43] also found that Chukars did not struggle to escape during the experiment since they showed habituation to the study apparatus.

#### **4.6 Flight angles (FA) test**

The birds which showed the flight escape reaction were also subjected to the flight angles test. Flight angles are a proxy of the flight intensity during the escape. The intensity of the flight taken was observed and noted by two independent observers. For this purpose, flight reaction was categorized into three cohorts: (1) Strong flight: the bird took a flight of nearly ≤90° angle, (2) Medium flight: an escape flight of ≤45° angle, (3) Weak flight: the flight of ≤30°. An angle meter was used to measure the angles of the flights. During the flying escape of n = 278 birds, I noted the flight angles. I found that medium flights were the most common mode of escape from the boxes. The mean number of birds for strong flights, medium and weak were 6.26 1.0, 9.2 1.8, and 3.06 0.8, respectively. Out of every batch of maximum strong flights, medium, and weak flights were 15, 24, and 9. Strong flights indicate good escape behavior. The maximum number of birds that took a flight was recorded as n = 38. On the other hand, two were the minimum birds that took flight from any batch. Tukey's pairwise test showed that there was a significant difference between weak and medium flights (p = 0.005). Nevertheless, Mann–Whitney pairwise test showed no statistically significant differences between strong and weak flights, or between medium and strong flights. The descriptive collected data is shown in Appendix 6. Without training, most flights of the birds were in the category of medium ≤45° flight, which is enough to escape a terrestrial predator, but not for avoiding an aerial predator.

*Intensive Farming and Welfare Regarding Anti-Predator Behavior of Chukar… DOI: http://dx.doi.org/10.5772/intechopen.106371*

Nevertheless, in this study all the birds were naïve to the AP behavior; no AP trainings were provided [38] showed that if red-legged partridge were trained for AP behaviors, they showed a stronger and prolonged response as compared to the inept birds. Present results are not in compliance with this statement, as no AP training was provided before the collection of data. The observed AP behavior cannot be expected to closely overlap with that of the wild conspecifics [33]. If birds were sourced from the wild, there would be different results for AP response. On the other hand, the present study clearly showed that even if early-life trainings were not provided to the game birds, they can still show more than good AP behavior. The differences in the behaviors of birds by simulating the predators (previous studies) and using real natural predators (present study) have significant differences in the intensities and types of shown behaviors.
