New Data on Migration Time, Breeding Phenology, and Breeding Success of European Turtle Doves in Their Highest Breeding Habitats in North Africa

Laboratory of Functional Ecology and Environment, Faculty of Sciences and Technology, Sidi Mohamed Ben Abdellah University, P.O. Box 2202, Imouzzer Street, Fez, Morocco Laboratory of Pharmacology and Environmental Health, Faculty of Science Dhar El Mahraz, USMBA Fez, B.P. 1796, Fez-Atlas 30003, Morocco Laboratory of Biotechnology and Valorisation of Phytogenetic Ressources, Faculty of Sciences and Technics, Sultan Moulay Sliman University, Beni Mellal, Morocco Laboratoire de Géo-biodiversité et Patrimoine Naturel, Scientific Institute, Mohamed V University, Av. Ibn Battota, 10 BP 703, Rabat, Morocco


Introduction
e current advances in the study of avian migration have underscored the important linkages between wintering, stopover, and breeding zones [1][2][3][4] and the need to consider this interconnectivity in conservation measures [5]. It is becoming increasingly clear that the timing of spring migration, breeding chronology, and feeding quality can greatly influence reproductive success and population trends [6][7][8], especially for long-distance migratory species [9].
Birds are hypothesized to adopt their migration dates and breeding chronology in agreement with environmental factors dominating in wintering and breeding grounds [10].
is means that migration dates and breeding chronology were selected early and/or lately in order to avoid the harsh conditions, counting temperatures, and rainfall of such habitat [11]. In fact, many Anatidae and Palearctic migrant species have coincided their migration dates with suitable spring climate conditions and food availability during summer in breeding quarters, in order to maximize their breeding success [12]. Since climate change, mainly high temperatures can result in missed foraging opportunities, loss of body condition, and impaired parental care [13]. Also, avian responses to increasing temperatures may include breeding earlier in the season [14] or selecting cooler nest microsites [15].
With the development of technologies, especially satellite telemetry, the examination of individuals and their migration and/or breeding strategies facing different climatic, as well as geographical conditions, has greatly advanced with more abilities from wintering to breeding grounds [16,17]. However, the field monitoring of the strategies adopted by these populations needs an exhaustive presence in the field [18,19]. Additionally, many ecological and biological aspects must be monitored for a long period before conclusions. is is particularly important in the case of rapidly declining species, where threatening factors are still unclear in both breeding and wintering zones [20,21].
European turtle doves (Streptopelia turtur) are the only migrant Columbidae that winters in Sub-Saharan Africa and breeds in North Africa and Europe [22,23]. Based on breeding and wintering areas, this game has undergone long-term declines, especially in Western Europe [24,25]. Over the past decades, the intensification of agriculture and habitat loss are classified as principal causes leading to the declining of turtle doves [22,26,27]. ese factors cause a failure of breeding success and reduction of migration capacities. Although the breeding ecology and migration dates of turtle doves have been examined in detail [28,29], there has been no comparable examination of the detailed breeding chronology and reproductive success in high-altitude habitats [22]. In all previous studies, research concerning breeding ecology was limited to an altitude ranging from 100 to 600 m in both Europe and North Africa [22,23,[30][31][32][33][34][35][36][37]. erefore, understanding the migration and breeding phenology at high-altitude habitat will open a new window in conservation measures, in particular with recent evidence indicating that turtle doves breed and feed in mountainous zones [22,23]. In parallel, [38] have reported a significant shifting in migration dates and breeding chronology between high and low altitude breeding sites. erefore, the clarification of migration dates and breeding success, as well as the impact of the principal environmental factors dominating, will situate the high-altitude zones in terms of suitability, mainly rate of breeding success, and in terms of habitat that deserves more attention in conservation concern for declining doves.
In this paper, we present a four-year analysis of breeding phenology and migration dates of turtle doves in their highest breeding habitats. We examine the evidence that elevations in altitudinal parameters have coincided with late migration dates and breeding chronology in doves as reported in many other avian species. In the Western Atlantic Red Knot (Calidris canutus rufa) and Savannah Sparrows (Passerculus sandwichensis), the migration and breeding chronology were late in high-altitude mountains and influenced rate of success [39,40]. erefore, if the change in breeding chronology occurs, then how it will influence breeding success in the 'vulnerable' turtle dove.

Study Area.
e present study was carried out in Midelt province, Ait Ayach valley (32°41′N, 4°44′W) Morocco. e monitored sites were located at the Northern foot of Jbel Ayachi Mountains, in the Molouya high plateau, 1400 to 1600 m altitude above sea level ( Figure 1). Midelt was dominated by forests that cover 17095 ha, including Xerophytes with 4 772 ha, Holm oak 3375 ha, and Red juniper 2280 ha, while irrigated perimeter cover more than 1000 ha [33]. e orchards were dominated by fruit trees, including apples, plums, and peach trees.
In terms of climate, the temperature in Midelt ranges from 8 to 21.2°C with an annual average of 16.9°C, and precipitations vary largely between 65 mm and 481 mm from one year to another (data were extracted from 1957 to 2010 from Midelt Weather Station). All these characteristics, including vegetation cover and climatic parameters, would play a major role in the bioecology of the turtle dove.

Data Collection.
Twenty apple orchards (7.47 ha in total) at Ait Ayach (Midelt) were selected and monitored during four years, from 2015 to 2018. For each season, the survey of dove's arrival dates (first observed birds during spring) started from early March and continued to the building of the first nest. Equally, departure dates (date of last observed birds in the breeding site) were recorded at the leaving of the last birds from the whole breeding area. After the installation of doves in the designated sites (after the arrival of breeding pairs), the nest initiation (first nest recorded after the arrival date), laying date (first egg after the nest initiation), hatching (first fledged chick), and chicks' nest leaving (first chick to leave its nest) were surveyed with three visits per month (each visit with three days in field), based on turtle dove breeding biology reported in North Africa [30,31,41,42]: (i) the first visit was conducted at the beginning of every month, (ii) the second visit from 10 th to 20 th days, (iii) the third visit from 20 th to 30 th days. Common Bird Census (CBC) methodology [36], instigated to provide sound and observational information on farmland bird populations in agricultural lands, was used and research took place from 06.00 to 18.00 hours. In our case, this methodology was adopted inside apple orchards and nests or breeding pairs were researched line by line based on the tree lines of apple trees. In every single visit, recorded nests were monitored from the construction to the nest leaving or nest failure. Furthermore, reproductive success rates were evaluated by the calculation of hatching eggs (hatched eggs/laid eggs), fledging, and flying chicks (chicks leaving their nests/fledged chicks).
Moreover, failure factors were cited (basing on carcasses and other aspects, such as the status of the nest and feathers in or outside the nest) along with breeding stages, in order to compare between breeding phases.

Data Analysis.
e normality of data was checked for all variables with the Kolmogorov-Smirnov test. Arrival and departure dates were defined and means were calculated (all arrival or departure dates during four years). Results were given as mean ± SD. Climatic conditions (temperature and rainfall) were accessed from the Midelt Weather Station from 2015 to 2018. We analysed separately the relationship between breeding chronology, including nesting, laying, hatching, nest leaving and habitat characteristics, counting altitude, temperature, and rainfall by simple regression (linear model), and all variables with P walue greater than 0.5 were eliminated. In parallel, the analysis of variance with the lack-of-fit test was conducted to evaluate the best-fit for models by considering chronology (nesting, laying, hatching, and nest leaving dates) as dependent variables and habitat parameters (altitude, temperature, and rainfall) as factors. Rates of breeding success were compared between breeding phases, including nesting, laying, and rearing, via one-way ANOVA test, considering twenty apple orchards.
Statistical analysis was performed using the STAT-GRAPHICS Centurion software (version XVI.I), and graphs were created by GraphPad Prism Mac 6.0 h software.

Migration and Breeding Chronology.
During four years, the first arrival dates in spring were on 28.25 ± 2.05 March, while the last doves were observed on 28.00 ± 1.47 September at Midelt. On the other hand, turtle doves arrived solitary, particularly during the first weeks of April. After these periods, doves were observed in groups, especially in pairs. e breeding chronology of turtle doves, including nest initiation, laying, fledging, and chick's flying dates at Midelt is summarized in Figure 2. Nesting (initiation of nest construction) began during the last week of April at Midelt (26.5 ± 0.64 April), and the nesting period dated from 26 April to 6 August. Similarly, laying dates were earlier, and the first eggs were found on 28.00 ± 0.7 April, only two days after the nest construction date. e last laying date was on 12 August. On the other hand, the first fledged chicks were on 17.50 ± 2.72 May, and chicks started flying on 3.50 ± 2.33 June.

Relationships between Breeding Parameters and Climatic
Conditions. e relationships between the breeding chronology of turtle doves (nest initiation, laying dates, and hatching dates) and climatic parameters at Midelt are summarized in Figure 3 and Table 2. e breeding chronology of turtle doves, mainly nesting dates were earlier with increasing temperatures. On the contrary, elevated altitudes and increasing temperatures shifted lately nesting and fledging dates of turtle doves at Midelt.
Our findings clearly indicate that the European turtle doves breed in Midelt as a mountainous zone with an altitude ranging between 1400 and 1600 m. is species arrived on 28.25 ± 2.05 March and started its breeding season        [50], and in Moroccan North Atlantic (100 m above sea level and 450 km to Midelt) 24 ± 0.16 March [51]. In all these preceding studies, particularly those cited in low breeding sites (between 100 m and 514 m above sea level), nesting and laying dates ranged between 19 to 23 March and 10 to 20 April, respectively. erefore, despite the limitation of this study to one highaltitude site, the cited literature confirms that the arrival and breeding chronology dates are later in high-altitude breeding lands. At this point, the latitudinal influence on arrival and breeding dates was excluded, because Tadla where Dove's chronology was earlier being in the same latitudinal level with Midelt, while the northern Atlantic is in the North of Midelt.
e suggested reason behind this late breeding chronology at Midelt is the environmental factors accompanying the elevation of the altitude (Figure 3). In fact, high temperatures and low rainfall promote an early nesting and check nest leaving (Table 2). Such results were mentioned in many other migratory birds [52][53][54]. In the barn swallow (Hirundo rustica), the arrival dates vary from 12 April at 110 altitude to 24 April at 760 m [55]. Furthermore, the temperature was warm at lowlands and predicted the arrival dates of swallows. On the other hand, departure dates were on 28.00 ± 1.47 September, before the installation of cold temperatures in dove's breeding site.
Across three breeding seasons, breeding success was different from nesting to chicks rearing (Table 1). In total, 71.16% of eggs were hatched, and 79.56% of chicks have flown. Moreover, 57% of chicks (flown chicks/laid eggs) have flown at Midelt, which is comparable to previous results cited in other North African breeding habitats [35,38,45]. In Algeria, the nesting success varies between 56% and 60.32% [36,37,46], and in Morocco success rate varies between 54,1 % and 55.5% [39,48,56]. Failure factors were different. e predation (Figure 4), mainly reptiles, caused the loss of a significant clutch part (20.98 of nests were predated, counting 18.89% of eggs and 10.54% of chicks), as well as the desertion of nests (5.14% of nests were deserted, counting 28 eggs (4.41%) and 20 chicks (4.39%)). Furthermore, 32 eggs were unhatched and 25 dead chicks were lost (chicks were found died every cold morning). Similar results were cited in Orange orchards. In fact, [29,39,48] have reported that the human disturbance in orange orchards at Tadla (150 km to Midelt) counting fruit harvesting, tree pruning, and the overuse of pesticides in coincidence with turtle dove breeding periods, are the main factors causing nest desertion.

Conclusion
In summary, this study provides a deep analysis of turtle doves' arrival dates, breeding chronology, and reproductive success in its North African highest breeding zone. Results reveal the late arrival and breeding seasons in high-altitude habitats compared to lowlands. However, breeding success was similar between low-and high-altitude zones. All these strategies, including late arrival and breeding chronology, prevent turtle doves against low temperatures in high habitats (protection of eggs and chicks) and human disturbance in managed agricultural farms.

Data Availability
e data used to support the findings of this study are included within the article.