Diet of Wood Pigeon (Columba palumbus) in Forest Areas of Souk Ahras Region (North- Eastern Algeria): Management Implications

1LEAT Laboratory, Department of Biology, Faculty of Nature and Life Sciences, University of Souk Ahras, P.O. 1553, Annaba Road, Souk Ahras, Algeria. 2Department of Nature and Life Sciences, Faculty of Sciences, University of Oum El-Bouaghi, P.O. 321, Constantine Road, Oum El Bouaghi, Algeria. Article Information Received 08 July 2019 Revised 13 May 2020 Accepted 08 August 2020 Available online 06 November 2020


INTRODUCTION
D etermining a bird's diet is crucial to many aspects of avian ecology (Sam et al., 2017). The studies carried on the wild bird's diet suggested that the food resources play a major role in determining how much to invest in reproduction, and it ultimately affects the population size (Ruffino et al., 2014). Another way in which food can influence the bird population is via its impact on breeding timing (Newton, 1980;Poulin et al., 1994). Seasonality has a fundamental impact on food availability, which in turn affects abundance, bird distribution, breeding success and the ultimate survival (Mengesha and Bekele, 2008).
Columbidae is one of the most threatened bird families in the world however it has received relatively little conservation attention (Walker, 2007). The wood pigeon belongs to the family Columbidae. It is a Palearctic species that breeds across the European continent, from Russia and O n l i n e

F i r s t A r t i c l e
Algeria, Merabet et al. (2014) reported that Oxalis Oxalis pes-caprae and olives are important food items consumed by wood pigeon. The seasonal variation of the pigeon diet is another story. Many studies highlighted the temporal variation in the consumption of food items by wood pigeon. Seeds and fruits of trees dominate winter and spring diets (Murton, 1965;ÓhUallachain and Dunne, 2013;Merabet et al., 2014). Gutiérrez-Galán et al. (2016) found that fruits of trees also occurred during autumn in a worthwhile proportion. During summer, the wood pigeon feeds almost on cereals, and many studies agree with the great consumption of this food category (Colquhoun, 1951;Jimenez et al., 1994;Gutiérrez-Galán et al., 2016).
In North Africa, only a few studies document the wood pigeon that were limited to annotated checklists and catalogues of bird species in Algeria (Isenmann and Moali, 2000), Morocco (Thévenot et al., 2003) and Tunisia (Isenmann et al., 2005).
Wood pigeon diet in Algeria has been insufficiently examined, and most of the studies are only descriptive (Guezzoul et al., 2004;Merabet et al., 2014). These works showed that the wood pigeon consumes mostly plants with a preference for seeds of Oxalis, barley Hordeum vulgare and wheat Triticum durum in the plain of Mitidja (Merabet et al., 2014;Guezzoul et al., 2004). More research on the diet of this species is sorely needed and especially in our study area, where the wood pigeon is still found in forest areas and it rarely spreads over agricultural and urban areas. Besides wood pigeon is a game species that needs better management interventions. In order to redress this lack of information, we have studied the diet composition of this species and its seasonal variation in three forests within the Souk Ahras region, in North-eastern Algeria.
The primary aim is to understand the ecology of the wood pigeon in the region of Souk Ahras and increase the knowledge of the diet of this species. The research objectives were to evaluate the ecological requirements of this game bird and to develop conservation management strategies for this species in forest areas.

Study area
The study was carried out from October 2015 to March 2017 at three forests within the region of Souk Ahras, in North-eastern Algeria (Fig. 1, Table I), where wood pigeon are reported to be hunted.

Data collection
In total, 86 individual wood pigeon (54 males and 32 females) were collected and examined between October 2015 to March 2017. The birds were hunted by local hunters. They were dissected for crop content analysis. All the crops were removed and emptied into Petri dishes and the contents were weighed on a digital scale with an accuracy of 0.0001g. The food items were examined and separated under a binocular microscope LEICA with a magnification of 10 X 40 mm. The sorted items were conserved in ethanol at 70°.

O n l i n e F i r s t A r t i c l e
Food items were classed into three categories: cereals: (seeds of barley), fruits: (acorns and other fruits consumed by wood pigeon), grass clover (leaves of clover Trifolium sp.).
Outside the legal hunting season, we observed the feeding sites and recorded the species on which the wood pigeons were feeding. The observations on the feeding sites were not detailed; our results are based on general simple observations and the information given by the local residents.
The body weight of entire birds was measured with an electronic balance (to the nearest 1.0g), distinguishing between sexes and season.
Seasonal changes in diet were evaluated by comparing wood pigeons haunted in autumn (October to November), winter (December to February) and high spring (March to April). The determination of sex was by examination of gonads following dissection.

Statistical analyses
Data on richness and body weight were analyzed for both the overall total and for each season and sex. To analyze differences in diet composition between seasons as well as between sexes, a Permutational Multivariate Analysis of Variance procedure (PERMANOVA) was used with the ADONIS function (Anderson, 2001) included in the "vegan" package for R 3.1.0 (R Core Team, 2016) software (Oksanen et al., 2007). This procedure acts as a matrix based non-parametric analysis of variance. PERMANOVA analyzes and segment sums of squares using semi-metric and metric distance matrices using permutation methods (Anderson, 2005).
To analyze seasonal differences in the relative consumption of food categories, overall crop weight and dietary richness, a non-parametric Kruskal-Wallis one-way analysis of variance (ANOVA) test, using the R function, kruskal.test, was used (Kruskal and Wallis, 1952). The test used a single factor with three levels: One for each of the seasons (autumn, winter and high spring). Then Student 't' test analysis was used (Zimmerman and Zumbo, 1993) to test for differences in relative consumption of food categories, overall crop weight and dietary richness between sexes.

RESULTS
Overall, we analyzed 86 crops of wood pigeon; all were hunted in Souk Ahras region between October 2015 and March 2017.
During the study period, 16 species of plant foods were identified. Thirty species were found in the crops and five of them were observed during the fieldwork.
Considering all the consumed items in the analyzed crops, the wood pigeon was exclusively herbivorous; no animal species were recorded during this study. Table II summarizes all the plant species consumed by pigeons or recorded during the fieldwork.
The diet composition was examined to see if there was a difference between seasons and sexes. Permanova analysis revealed a clear difference between seasons (Adonis: F 2.83 =7.9391, p<0.001), but not between sexes (Adonis: F 2.83 =1.0719, p˃0.05).
The total weight of consumed items by all studied wood pigeons (86 individuals) was 2276.1g and the average weight of crop contents was 28.45 ± 10.86g. In comparison between males and females, the average weight (± sd) of the crop content in males (29.42±10.9g) was almost similar to that of females (26.89±10.79g). Statistically, there was no significant difference between males and females (t test: t 1.84 = -0.7622, p ˃ 0.05) ( Fig.  2A). The variability of the average weight through seasons was significant (Kruskal-Wallis test: χ 2 2.83 = 14.8451, p< 0.001) (Fig. 2B). The different food categories consumed by the wood pigeon varied through seasons (Fig. 3). The tree fruits were the best represented food category in the wood pigeon diet. This category accounted (93.25± 17.43%) for the total weight of consumed items. It was present in the diet of 98.83% of the analyzed crops of birds.The main constituents of this food category account for fruits of mastic tree (34.52%) and acorns of cork oak (30.58%). There was a significant seasonal difference in the weight proportion of tree fruits in the diet (Kruskal-Wallis: χ 2 2.83 = 15.55, p< 0.01) (Fig. 3A).
The cereals made up 6.00± 17.29% of the bird diet and occurred in 12.79 % of crops. Cereals were exclusively represented by barley. In addition, wood pigeon were observed feeding on wheat and oats Avena sativa during O n l i n e

F i r s t A r t i c l e
Diet of Wood Pigeon (Columba palumbus) in Forest Areas 5 the fieldwork. The items in this category were consumed only in summer and autumn. There was a significant seasonal difference in the weight proportion of consumed cereal grains (Kruskal-Wallis test: χ 2 2.86 = 18.9492, p< 0.01) (Fig. 3B).
The clover showed a highly significant difference in term of relative proportion between seasons (Kruskal-Wallis test: χ 2 86. 2 = 25.8857, p< 0.01) (Fig. 3C). This food category represented a minor proportion of the diet of wood pigeon (0.76 ±3.74 %). In autumn, the wood pigeon diet was dominated by tree fruits constituting (84.83 ±25.03%) and occurred in 97.05% of crops. Considering the weight proportion, the most important items in this category were mastic tree fruits (48.59 %), followed by Mediterranean hackberry Celtis australis (19.76%), and zeen oak Quercus canariensis (11.26%). The autumn diet included other species recorded in small amounts (Mount Atlas mastic tree Pistacia atlantica, Mock privet Phyllerea angustifolia, Aleppo pine and zeen oak).
Cereals also constituted a large proportion of the autumn diet (17± 25.03%) and were found in 32.35% of the analyzed samples.
The consumption of tree fruits was the most recorded in winter, the diet was exclusively made by tree fruits representing 100% of crop contents mass and being present in all the analyzed crops (100%). Two species were found in major proportions viz., cork oak (46.32%) and mastic tree (33.99%). Mock privet, zeen oak and Sensitive plant Mimosa pudica together represented less than 20% of the consumed materials.
In spring also, the diet analysis revealed that the tree fruits which represented (94.57± 8.97%) of the crop contents, occurred in all the examined crops. The most eaten fruits were cork oak (36.53%), Aleppo pine (30.55%) and olives (30.11%). Berries of Ivy accounted for 2.78% of the spring diet. The clover was found with a low weight proportion (5.43± 8.97%) but was reported in 33.33% of studied crops.

DISCUSSION
Few quantitative data exist on the diet of the wood pigeon in Algeria and North Africa. This study provides a detailed description of the species diet in a Mediterranean forest in North-eastern Algeria.
In our study area, the number of plant species consumed by wood pigeon was smaller than reported in similar studies in Algeria (Guezzoul et al., 2004;Merabet et al., 2014) and in Europe (Murton et al., 1963;Schnock and Seutin, 1973;ÓhUallachain and Dunne, 2013;Gutiérrez-Galán et al., 2016). This lower food diversity may be due to the higher availability of a few preferred food items over long periods of the year (Whelan and Brown, 2005;Molokwu et al., 2011;. It has already been seen in other studies in the case of acorns (Purroy et al., 1984;Díaz and Martín, 1998) and olives (Merabet et al., 2014), a fact also reported in our results. Wood pigeon extensively foraged on the most abundant and available species i.e. acorns of cork oak and fruits of mastic tree.
Animal materials were absent from the wood pigeon diet in our study area. This result is consistent with the findings of other researchers (Folk, 1984;Purroy et al., 1984;Jiménez et al., 1994;Gutiérrez-Galán et al., 2016). In this study, the analyzed birds were not hunted during the breeding season when many herbivorous birds take in as many insects as possible (Ehrlich et al., 1988). Columbiformes consume additional calcium-rich material during their egg-laying period (Murton et al., 1964;Ó hUallachain and Dunne, 2013).
The basis of the wood pigeon diet is made up of tree fruits. This result is similar to those of Murton (1965), Purroy et al. (1984), ÓhUallachain and Dunne (2013) and Gutiérrez-Galán et al. (2016). Among the most prominent consumed items in this food category, fruits of mastic tree and acorns of cork oak were the most taken species.
Traditionally, the diet varies throughout the day, the year, and between years, sites, sexes, and even between individuals (Sutherland et al., 2004). According to our results, variation in diet composition of the wood pigeon has been significant through the seasons. This variation can be related to the fact that the consumption of food items depends on their availability in the study area (Fuentes, 1994;O'Donnell and Dilks, 1994;Gutiérrez-Galán et al., 2016).
During the autumn, wood pigeon fed predominantly on mastic tree fruits. This important use of mastic tree seems to be related to the peak in fruit abundance, which reaches high levels between October and January (Herrera, 1984). It is also coincident with the fruiting season of mastic tree. In addition, it could be related to the digestive physiology of birds and the pulp composition of fruits of the Pistacia genus (Fuentes, 1994).
Contrary to our finding, in the plains of Mitidja (Algeria), mastic tree fruits represented a minor item in the wood pigeon diet. However, olives were more preferred (Merabet et al., 2014). This difference may be due to the fact that the fruits of the mastic tree have less fat value (15.9%) than olives (41.9%) (Jordan and Herrera, 1981). It may also be linked to difference in the habitat structure between Souk Ahras forests largely dominated by fruit trees and the Mitidja Plain which is mainly an agricultural region.
The wood pigeon completed its autumn diet with other items including acorns of cork oak. Significant

O n l i n e F i r s t A r t i c l e
A. Kaouachi et al. consumption of acorns during autumn has been highlighted by many others authors (Purroy et al., 1984;Díaz and Martín, 1998;Verdu et al., 2010;Bendjoudi et al., 2015;Gutiérrez-Galán et al., 2016). During the winter, tree fruits dominated the diet; mastic tree and cork oak were the most exploited by wood pigeon. Both species are widely available in the study area. It has been proposed that frugivory can be advantageous because of the low energetic cost involved in fruit-feeding (O'Donnell and Dilks, 1994). Many authors reported a preferred consumption of acorns in winter (Verdu et al., 2010;Bendjoudi et al., 2015;Gutiérrez-Galán et al., 2016). The acorns are an important energy source as they contain high levels of fatty acids (Heitmeyer and Fredrickson, 1981;Reinecke, 2017). Additionally, efficient consumption of oak acorns may decrease the time required to fill their stomachs (Reinecke, 2017).
In spring, the food spectrum of wood pigeon became more diverse as other food items became available, related to the growth of ground vegetation (Petty, 1999). The other items included Aleppo pine, olives and clover. It is important to note that the occurrence of Aleppo pine was due to the fact that the wood pigeons were hunted in pine forests.
Although clover constituted a minor percentage of the spring diet, it occurred in 30% of crops. It has been suggested that pigeons feed on foliage to complement the scarcity of fruits (O'Donnell and Dilks, 1994;Oliveira et al., 2002;Ando et al., 2017). The green foods are an important source of many essential elements for wood pigeon (Cramp, 1972).
On the other hand, olives were represented in large proportions during high spring. As in other studies (Cramp, 1972;Huysentruyt et al., 2013;Perea and Gutiérrez-Galán, 2016;Gutiérrez-Galán et al., 2016), the consumption of olives by wood pigeon reaches considerable values from March on wards when pigeons start to reproduce.
Another study in Algeria (Merabet et al., 2014) found that oxalis substituted acorns and olives as the most prominent food item in the spring diet of wood pigeon. The absence of this species in the crops in our study can be explained by the fact that oxalis is not widely found in the study area.
Our observations on the field work during summer showed that the wood pigeon food depended on cereals.
Flocks of wood pigeon were observed in cultivated fields on the edge of forests, feeding mostly on wheat, barley and oats. Similarly, consumption of cereals has been highlighted by several authors (Purroy et al., 1984;Gutiérrez-Galán et al., 2016). Other studies found that cereals dominated the autumn/ late summer diet. This difference could stem from the changes in agricultural practices from winter-sown to spring-sown crops (ÓhUallachain and Dunne, 2013). Some birds were also seen ingesting Mediterranean hackberry during late summer. Merabet et al. (2014) reported a similar result in Algeria.
The pattern of seasonal variation of the diet was similar to most studies on wood pigeon diet (ÓhUallachain and Dunne, 2013;Merabet et al., 2014;. The consumption level of of some food items does not necessarily reflect strong selection or preference but likely an opportunistic diet (Marrero et al., 2004). The Absolute preferences can only be inferred outside of the context of fruit abundance (Parrish, 1997).

CONCLUSION
We identified the food spectrum of the wood pigeon in Souk Ahras region. Overall, the diet of the wood pigeon fitted into the known pattern. The temporal variation in the diet was significant. As in other studies, our preliminary data showed that the tree fruits dominated the diet during all the seasons excepting the summer in which cereals were the most consumed. Our data collected from three forest areas suggest that the wood pigeon is a generalist/ opportunistic feeder, feeding on various food items and switching to alternative species when preferred food was not available. Furthermore, we suggest that the relationship between the composition of the diet and the availability of foods still requires clarification. Finally, this study has important implications for the conservation and the management of the wood pigeon in the forests of Souk Ahras and possibly other similar woodland areas.