Impact of Feeding Habits on Population Ecology and Breeding Biology of Grey Francolin (Francolinus pondicerianus) in District Bhakkar, Punjab, Pakistan

Aftab Raza Khan1, Azhar Abbas Khan2*, Javed Iqbal3, Arif Muhammad Khan4, Zeshan Hassan2, Hafiz Muhammad Aatif2 and Umbreen Shahzad2 1Department of Biological Sciences, Government Post Graduate College, Bhakkar, Pakistan. 2College of Agriculture, Bahauddin Zakariyya University, Bahadur Campus Layyah, Pakistan. 3Department of Biological Sciences, Government Degree College, Mankera, Pakistan. 4Department of Biotechnology, University of Sargodha Main-Campus, Sargodha, Pakistan. Article Information Received 01 December 2019 Revised 22 May 2020 Accepted 07 July 2020 Available online 14 January 2021


INTRODUCTION
G rey francolin, Francolinus pondicerianus (Gmelin, 1789) is widely associated with drier parts of the Indus plains and penetrates Thar desert in Sindh, and Dera Ghazi Khan, Cholistan and Thal deserts, Kala Chitta Range, Salt Range and Potwar plateau in Punjab, the bird also occurs in suitable habitats occurring in Balochistan, Khyber Pakhtunkhwa and Azad Kashmir. Grey francolin, a medium O n l i n e

F i r s t A r t i c l e
population density of grey francolins as 1.59 ± 0.39 per hectare (159 per km 2 ) in irrigated tract and in forest tracts it was 0.87 ± 0.14 per hectare (87 per km 2 ) birds. The grey francolin indicated a declining trend in the past, and was an indicator species for farmland ecosystems suggesting wildlife populations not always getting benefit from different agricultural schemes (Kleijn et al., 2001;Kleijn and Sutherland, 2003).
In agro-ecosystem of Pothwar Plateau, the average breeding performance of grey francolin was observed as 7.0±0.36 (clutch size), 6.00±0.36 days (egg laying period), 20.6±0.50 days (incubation period) and 5.33±1.22 (hatching success) (Hussain et al., 2012). The analysis of 140 crops showed that the species had a higher dependence on arthropods during summer and on plants during winter. They observed that 10 crop contents had arthropods which constituted 28% of food of grey francolin which preferred a wider base of food contents, advantageous for survival under all the odds without lowering their populations seriously (Khan et al., 2015). The frequency distribution of food types within crops of grey-winged francolins (Scleroptila afra) in Losotho indicated that arthropods constituted 7.3% of food items by volume (Little et al., 1992). Keeping in view the declining patterns of this phasianid species, the present study was conducted in different areas of District Bhakkar during June 2018 to May 2019, wherefrom so far, no data were available. The present study provides information on population ecology and breeding biology of grey francolin, a valuable asset of the Thal desert ecosystem of District Bhakkar, with reference to feeding habits.

Study area and sampling design
District Bhakkar (31.6 o N 71.0 o E, 149m (522 feet) above mean sea level) is part of Thal desert, Punjab, Pakistan. Thal sand dunes and xeric vegetation are the main features of the desert landscape. Bhakkar is divided into four tehsils: Kalurkot, Darya Khan, Bhakkar and Mankera. Average annual temperature is 24.6 o C while mean annual precipitation is 213 mm.
A detailed survey was carried out during June 2018 to May 2019 in different areas of District Bhakkar. Population of grey francolin was estimated by using line transect method as explained by Burnham et al. (1980). In order to find out the potential population of grey francolin, a few preliminary surveys were conducted which provided a base to divide the study area into 16 stands (1. Taragran wala 2. Ghulamaan 3. Rakh Ghulamaan 4. Chak 19 ML 5. Rakh Mahota 6. Rakh Dalaan 7. Joyia 8. Jhamat Shumali 9. Dajal 10. 51 TDA 11. 57 TDA 12. 61 TDA 13. Rakh Karlu Wala 14. Rakh Mankera 15. Rakh Kapahi and 16. Rakh Mahni), with possible dense population of grey francolin. Although habitat conditions in these 16 stands might be different from one another, they were selected based on similar cropping pattern in the study area. One stand is situated along the Indus belt (stand 9) while others are in the plains region.
A permanent fixed transect line (1.5 km) was established in each of 16 stands, trying to maintain a straight line passing diagonally through stand area covering all possible microhabitat variations. The grey francolin was searched by direct (sighting) and indirect (calls, faecal pellets) methods. Each transect line was walked by two workers at a moderate speed of 2-3 km/hour in the morning (0600 to 0930 h) and in the evening (1600 to 1830 h) from June 2018 to May 2019. The data were collected fortnightly from each stand. Four stands (transects) were walked in one day in each tehsil and walked by different workers, completing the exercise in four days every fortnight. Each line transect covered an average band of about 40 m on each side as record suggested an approximate distance of sighted birds from the transect line, hence a uniform transect width of 80 m (0.08 km) was fixed for analysis of line transect data. Data on age (young and adult) and sex (male and female based on spurs on legs of males) of each sighted bird were recorded.

Collection and analysis of food contents
Thirty crops were obtained from grey francolins collected randomly from local hunters after fulfilling formalities of the respective Department from different stands in the study area during different seasons (spring from February-April, n= 9; summer from May-July, n= 8; autumn from August-October, n= 6; winter from November-January, n= 7). Sixty crops were obtained randomly from adults during different seasons in spring (n= 15), summer (n= 15), autumn (n= 15) and winter (n= 15). The weight of each crop was measured and the crop packed separately in plastic bags. Each crop was preserved in 7% formalin solution and was dissected in the laboratory for removal of food contents. The crops were weighed for food contents. The identifiable food contents were separated and categorized to the lowest possible taxonomic ranks following the work of Faruqi et al. (1960). Some crop contents could not be identified and were designated as unidentified plants or animals. Arthropods were collected randomly from different stands by quadrat sampling method. A quadrat of 0.25 m 2 size was used fortnightly each month at a rate of one minute for arthropod collection. All samples were measured in volume (ml) by water displacement method.

O n l i n e F i r s t A r t i c l e
Threat survey and data analysis Besides direct field observations, questionnaires were used to determine threats to grey francolin population. Random sampling methodology was applied during the current study. Detailed structured open and close-ended questionnaires were used to get information on various threat factors in different stands of the study area. The questionnaires included information on the use of pesticides, hunting, habitat destruction and other factors. Questionnaires (160) were filled from local hunters for the study from different stands (10 from each). On the basis of this information stands were designated as affected areas (stand 12) and control areas (stand 16) in order to evaluate impact of threat factors on population ecology, breeding biology (clutch size, incubation period, hatching success, fledging success, egg weight and egg width) and feeding habits. The stand 12 was considered an affected area due to low population of grey francolin and habitat destruction while stand 16 was taken as control area due to high grey francolin population and habitat intact. Prior to data collection, an informed consent was taken from the participants for their willingness to share the information.
Significance of difference between male: female sex ratio for calendar months, and adult crop contents in different months and from different stands was tested at 0.05 level by using Chi square test (χ2). The formula, number of sighted birds divided by transect area (transect length × average transect width, 1.5 km × 0.08 km= 0.12 km 2 ) was used to calculate the density of bird population (per km 2 ) in each stand. Independent samples t-test and One-way-ANOVA at significance level of 0.05 were analyzed by using SPSS 16 software package. Pearson coefficient of linear correlation was calculated to determine association of densities of different calendar months with different variables. Pearson coefficient of linear correlation and Regression model was used to determine association of arthropod abundance in different calendar months. Dispersion index, which is the scattering of grey francolin from the transect line, was calculated by dividing group variance by group mean for each category (Odum, 1971). The values of index <1 suggested a uniform, =1 random and >1 clumped population.

RESULTS AND DISCUSSION
Human-mediated wildlife reshuffling has raised concerns over the conservation of biodiversity. Globally the population of grey francolin is categorized as "Least Concern" in the Red List of threatened species (IUCN 2019-20). In investigated desert areas, the grey francolin presented a high density of 1.97±0.65/km 2 (Fig. 1) and 1.76±0.62/km 2 in stand 16 (Table I). Pearson coefficient of linear correlation and Regression model suggested a direct link of seasonal factors with average densities of grey francolin and arthropod abundance (individuals/m 2 ) (p ˂ 0.05) (not shown in tables). t-test and One-Way-ANOAV results indicated that there was no significant difference between morning and evening samples of grey francolin (data not shown). Chi square test (χ2) indicated that there was no significant difference between different samples of males and females of grey francolin during different calendar months. Stand data suggested that grey francolin had an average of 0.14 young/adult and 0.33 young/female ratios (data not shown). Grey francolin maintained a fairly clumped dispersion during different calendar months and showed an overall dispersion index of 2.1±0.43 (Fig. 2). Pearson coefficient of linear correlation between dispersion index and precipitation indicated values of p > 0.05, t = 0.65 (Fig. 2). Dispersion indices for morning and evening samples for different calendar months were not much different (p > 0.05, df = 21, t = 0.71; p > 0.05, df = 02, F = 0.90). The morning and evening samples in different stands revealed a clumped dispersion (mornings =15.61±3.5, evenings = 14.40±2.78). t-test (p > 0.05, df = 30, t = 0.25) and One-Way-ANOVA (p > 0.05, df = 1, F = 0.25) indicated that there was no significant difference between morning and evening samples.  The analysis of seasonal samples revealed significant increase in body mass, crop mass and crop volume of adult males while young females elucidated a positive increase than males with slight but non-significant variation, respectively (Table II). The rigorous consumption of leaves, seeds and arthropods was observed in all seasons and stands for adult grey francolin (Figs. 3 and 4) while arthropods (0.11±0.01) were highly preferable food for young ones (Fig. 5, p ˂ 0.05). t-test results elucidated a positive relationship between seasonal fluctuations and feeding habits of grey francolin (p ˂ 0.05) (data not shown). Chi square test (χ2) explained linear relationship between different food items of leaves, fruits, grit, other invertebrates, arthropods and other animals in adults during different months and stands (Table V). A statistically positive trend was observed between feeding habits and density and dispersion index of adult and juvenile grey francolins. t-test and One-Way-ANOVA results clearly reflected a highly significant association of crop mass and crop volume with body mass, density and dispersion index (p ˂ 0.05) (Table III). p ˂ 0.05 (t-test) elucidated a direct

O n l i n e F i r s t A r t i c l e
A.R. Khan et al. link of density and dispersion index with different threat factors (Table IV) and arthropod abundance with volume of arthropods in diet of adults in different seasons with slight variations in juveniles (data not shown). A significant variation in clutch size, hatched eggs and fledging success was observed in stand 12 (p ˂ 0.05, t-test). Population density of grey francolin (13.01±6.39/ km 2 ) was found to be higher in the morning (mean 29.25) due to foraging than in the evening (mean 20.67). These results matched with the findings of Rotella and Ratti (1988). Maximum population density of grey francolins (1.97 birds/ km 2 ) was in April. Population density of grey francolin in Lal Suhanra National Park, Bahawalpur, Pakistan was 8.40 birds/ km 2 (Khan, 2010). The lowest density 0.30±0.23 was reported in December 2018. It might be due to low temperature, draught, hunting, use of pesticides and local migrations for new food sources (Odum, 1971;Khan et al., 2015). Population recruitments expressed seasonal changes in age structure when chicks were more than two months old after September. It was difficult to differentiate young ones from adults under field conditions. Male: Female sex ratio was skewed towards males in grey francolins (1.41: 1). It was due to severe shooting and temperature variations (Khan et al., 2015). The present findings of dispersion index of grey francolin (2.1±0.43) showed a clumped dispersion. Jansen et al. (2001) studied different levels of clumping of redwing francolin (F. levaillantii) in South Africa. Odum (1971) suggested that dispersion index had survival value for bird species. Uniform dispersion indicated a decreased intraspecific competition and a group protection from environmental hazards.
Leaves contributed 0.094±0.006/km 2 of average volume of crop contents while arthropods expressed 0.161±0.022/km 2 for adult grey francolin during different calendar months. Little (1992) indicated high proportion of arthropods by volume in crops of adult grey francolin in all seasons. Leaves, seeds and arthropods were highly consumed food diversity by Phasianid species. Tsachalidis et al. (2007) and Paralikidis et al. (2010) confirmed such findings. Phasianid species number was affected by human agricultural practices which modified the diet of these species (Chamberlain et al., 2000).
The data regarding breeding biology exhibited mean hatching success and fledging success of 2.33±0.33, 1.67±0.33 per clutch respectively in the study area (stand 12), and 5.4±0.36, 4.2±0.33 per clutch in control area (stand 16) respectively. In agro ecosystem of Pothwar Plateau, the average breeding performance of grey francolin was observed as 5.1±0.65 (hatching success) and 4.3±0.66 (fledging success) in forest habitat; 5.5±2.10 (hatching success) and 5.0±1.95 (fledging success) in cultivated habitat (Hussain et al., 2012). Since we lack previous published work in Bhakkar region, the above cited study might be appropriate for comparison.

CONCLUSION
Seasonal variations affected the feeding habits of grey francolin, which resulted in fluctuations in density and dispersion index of this species. Illegal hunting, habitat destruction and use of pesticides are the main threats to this valuable asset of Thal desert ecosystem. In grey francolin, there was a positive trend of crop mass and crop volume with body mass. Future studies are also warranted to evaluate their predatory performance against a wide range of arthropods and especially grasshoppers, crickets and locusts. It is recommended to conserve foraging areas and furthermore promote the judicious use of pesticides and control illegal hunting to save these game and cage birds that help the agriculturist by consuming crop pests.

Statement of conflict of interest
The authors have declared no conflict of interest.

O n l i n e F i r s t A r t i c l e
Grey Francolin Feeding Habits in Deserts