Abstract
Interactions between co-occurring species, including competition and predation, comprise critical processes regulating local community structure, habitat use, and diversity. We monitored nesting habitats, breeding chronology, and reproductive success rates to describe the patterns of spatiotemporal organization of three co-habiting species: the “native” turtle dove, the “invasive” collared dove, and the “predatory” Maghreb magpie. We defined nesting site parameters, breeding chronology dates, and success rates to explain how these species are dispersed in space and time. Similarly, predation attacks were evaluated. Patterns of habitat use were best explained by fear of predation and competition. Both doves selected nesting sites far away from the predatory Magpie to protect their nests. Equally, sympatric Columbidae turtle dove and collared dove were segregated horizontally and vertically only in space to reduce competition inside olive orchards. On the other side, Maghreb magpie started the breeding activity first, most probably to benefit from food abundance targeted in doves’ nests (eggs and nestlings). Further, breeding success was higher in both doves, despite predation pressure exercised by the Magpie and other reptiles. Magpie nests were colonized by the great spotted cuckoo (Clamator glandarius). Finally, this study provides the first and only detailed data on nest-niche of the turtle dove co-occurring with competitor and predator species, in the entire Northwest Africa range. Additionally, our data provide an opportunity of large-scale comparative studies of the nesting niche and breeding performances of the turtle dove, collared dove, and Maghreb magpie complex.
1. Introduction
Turtle dove Streptopelia turtur is the only long-distance migratory Columbidae in the Western Palearctic that has suffered a speedy and severe deterioration principally in the Northern slope of the Mediterranean. The European subspecies S.. turtur has decreased severely in the Great Britain (−78%) [1, 2], Spain and Portugal (−70%) [3, 4], and Austria (−54%) [4] between 1980 and 2020. Despite the ubiquity of dove distribution, the Asian population have also declined [5]. In Europe, deterioration causes remain unclear [6]. However, numerous factors have been suggested counting intensive hunting [3], degradation breeding and foraging resources [7], and infection with parasites principally Trichomonas gallinae [8]. However, because impacting factors affect different stages in the life cycle of the turtle dove, more studies are needed to estimate the impact of each factor and involve other pressures counting predators and invasive species [6] and these are suggested to increase the effectiveness of conservation measures [9].
In North Africa, mainly in western zones, the turtle dove is a summer migrant and breeder [10–12]. Morocco hosts an important breeding population of doves, mainly Streptopelia turtur arenicola [13]. It is mostly observed in agricultural fields that represent 1.5% of the total land area of Morocco [14–16] and woodland environments [17, 18] that represent 12.7% of the national surface [19]. In farmlands, breeding doves are mostly observed in irrigated perimeters dominated by oranges, olives, and apples [15, 20–22]. Despite its favorable status when compared with European subspecies, S. t. arenicola is currently suffering from human stressors counting farming practices [21, 23] and natural enemies counting predators [24] and sympatric competitors [16, 25].
In both farmlands and woodlands, S. t. arenicola is mostly impacted by reptiles and raptors that attack breeding pairs and their clutches [21, 22]. These natural enemies cause an annually loss of 15% to 30% of clutches in apples, olives, and orange orchards [15, 16, 24]. Equally, competitors, principally sympatric Columbidae, were currently revealed to impact negatively the breeding doves in Moroccan farmlands [16, 24]. S. t. arenicola was dominated by invasive collared dove (Streptopelia decaocto) [16], by sedentary laughing dove Streptopelia senegalensis [26], and by wood pigeon (Columba palumbus) [27] in agricultural and forest ecosystems during breeding season. However, these studies did not clarify how the vulnerable migrant doves manage to avoid the impacts of both predators and competitors on different stages of the life cycle in breeding habitats, which is suggested to offer more data for conservationists and policy makers [14]. Equally, the investigation of cohabitation between vulnerable bird and its natural enemies is suggested to serve as great model in ecology for animal species.
In this study, we used field investigations to document the interactions among the native “vulnerable” turtle dove (Streptopelia turtur arenicola), the “invasive-competitor” Eurasian collared dove (Streptopelia decaocto), and the “predatory” Maghreb magpie (Pica mauritanica) in Northwest Africa. We analyzed the patterns of (i) nesting site selection of these birds to examine whether they are segregated on some nest site characteristics, (ii) breeding chronology to investigate if there is any temporal segregation, and (iii) reproductive success to evaluate the impact of predation attacks and competition on reproductive rates of the migrant dove. More specifically, we searched if turtle doves manage spatially and temporally to avoid completion and predation in its breeding habitats.
2. Materials and Methods
2.1. Study Area
The study was conducted in the irrigated zone (357 000 ha) of the Saiss farming plain, placed in the neighborhood of the Fez empirical city (Central Morocco, Figure 1). The study sites are located between the Middle Atlas and Rif Mountainous chains, at an altitude of 600 m. The monitored sites are bounded by different aquatic ecosystems counting rivers (El Jawahir, Sebou, and Boufekrane) and dams (El Mahrez, Allal Fassi, and El Gaada dam) that offer required water sources (one of the most foraging elements conditioning the presence or absence of avian species).
Saïss plain is dominated by a semiarid climate with a strong continental impact. Annual precipitations are around 800 mm. Temperature varies widely between summer (July-August) dominated by hot temperatures (40–45°C) and cold winter (10–15°C). On the other hand, cereal crops (60%) dominate the Saiss plain (main foraging seeds required by doves) and are used for both industrial purposes and food benefits. Cereals are dominated by wheat (Triticum turgidum and T. aestivum), maize (Zea mays), and fodder crops of alfalfa (Medicago sativa). Orchards cover only 40% and are dominated by olives (Olea europaea) that occupy 14.4%. Olive orchards constituting potential nesting supports are organized in form of patches (isolated farms or groups of neighboring farms) throughout the plain.
2.2. Data Collection
Based on the previous observations of both doves in North African orchards [16] and the current breeding case of Magpie in olives [28], birds were particularly monitored in olive orchards. Investigations of the reproductive season were realized between the end of March and mid-September based on the phenological status of both doves in Morocco [26]. Three potential olive orchards (Figure 1) were designated and monitored for two seasons (2017–2019). The selected olives were bordered by water streams and cereal plots (potential forage for both doves), in addition to a dump of Ain Lbida where magpies were observed regularly from 2015 (unpublished data). Each farm was divided into two areas (based on roads in the marginal zones of orchard and farming personnel abundant in the periphery) (Figure 2): central sector (olive trees inside the orchard close to the epicenter) and marginal sector (first three olive-tree-lines situated in the orchard boundaries). Monitored farms were assigned with specific identifiers. Moreover, all monitored variables including tree density, tree-lines, and farming practices were similar between marginal and central sectors. The separation of two sectors is a practical way to search for segregation between two or more species nesting in the same plot [16].
Each week, when weather conditions were favorable (absence of rain and moderate temperatures), surveys started from 06h00 to 18h00. We managed to identify nests and their status (new, incubation, rearing, or failure). We found that nests were localized via phone GPS (Geotracker) in all monitored sectors/orchards and olive trees. A nest was counted active when incubating adults, at least one egg, at least one nestling, or droppings were detected. The position of each nesting tree was referenced and then reported in an Open Source GIS (Quantum GIS 3.14 (2020)).
2.3. Nesting Strategies
To describe nesting strategies, we focused on nesting site variables: (i) two variables for delimitation of macrohabitat, including nesting orchards and nest location inside or in the marginal zone of the farm (to differentiate potential cohabitating species); (ii) Seven variables for microhabitat breeding locations (cm) (very useful for coexisting species inside the same orchard, zone, and/or olive-tree), including nesting-olive-tree height (NTH), elevation of the nest above the ground (NHG), distance to lower canopy (NDLC), and the distance separating nest and tree central trunk (NDCT); and (iii) nest morphology variables (cm), including nest big axis (NBD), nest small axis (NSD), and nest cup depth (NDP) (Figure 3). Nest dimensions were measured because we expected that nest morphology will be variable depending on the presence of the competitor doves on the same tree or nesting zone. When there were no birds in a nest when located, the three variables for nest position and three variables for nest dimensions were immediately measured using a clinometer. If a bird was present, these measurements were taken early in the morning, when birds often leave the nest to look for food.
To explicate any potential segregation of breeding sites, distances separating nest location to the adjacent regional road (DR), nearest infrastructures (buildings, households, and farming storages) (DH), and nearby cereal farms (DC) were estimated for each species with QGIS (distances were estimated based on displayed maps). Equally, we noted the supporting tree (ST), and we quantified the number of farmers (daily encountered workers) inside each orchard (NP) for each visit, as well as herbaceous cover under nesting olives (VUCN). These variables were measured only for turtle and collared doves because of the similarity in their ecological requirements, while the Maghreb magpie is very different in terms of food requirements (feed in the dump and from other bird’ nests, which make foraging sites unclear to estimate distances) and human impact (North African populations nest always far from the highly populated area), and these make difficult to compare between Magpie and doves.
2.4. Breeding Chronology
To describe any segregation in the breeding chronology of bird species; (i) first to last nests per season, (ii) first eggs to last laying date per season, and (iii) first chicks to last hatching date per season were monitored for each bird. On the other hand, for nests detected after laying periods or rearing stages, laying time could not be confirmed with exactitude. We therefore based on the feature of both eggs and chicks (i.e., feathers of nestlings) and we relied on descriptive accounts of known-aged clutches in the surrounding population and breeding site to conclude nearly nest laying time and then nesting date.
2.5. Reproductive Success
Breeding success rates were assessed by the calculation of succeeded nests (active nests/built nests), hatching eggs (hatched eggs/laid eggs), and survived chicks (chicks leaving their nests/hatched chicks). In parallel, failure factors, including predation, desertion, and others were recorded. Further, the nature of predators (to estimate the attacks of Magpie on both doves) was searched via shells, meat fragments, feathers, fresh animals, and human traces inside or in the neighborhood of nests. For example, rats typically leave eggshell fragments, while snakes feed on nests without leaving a trace [29].
2.6. Statistics
We checked for normality and homogeneity of variance for all variables via the Kolmogorov–Smirnov test. The one-way ANOVA test was used to assess differences in nest placement and dimension among studied species. Similarly, breeding success rates, including nest occupation (active nests/built nests), laying (hatched eggs/laid eggs), and fledging (chicks leaving their nests/hatched chicks) were analyzed with ANOVA. For all these parameters, we considered three nesting orchards and two breeding seasons. Statistics were done in STATGRAPHICS Centurion software, version XVII, and results were given as sample size and mean ± SD. On the other hand, nest-niche parameters were compared only between doves due to their cohabitation inside the same orchards and the possibility to compare between their nesting parameters, while Maghreb magpie nests were isolated and have different foraging sites.
To assess the main factors of nest-niche separation between doves, distances separating nests of each bird to the adjacent cereal farms (DC), adjacent vehicle road (DR), adjacent infrastructure (homes or other constructions) (DH), active employees inside groves (NP) (documented for each visit), herbaceous cover under nesting-olives (VUCN) (assessed by superficial covered by plants in the circle of three m around nesting tree), type of supporting tree (ST), and the elevation of nesting-olives (NTH) were considered as illustrative factors (principal factors), while the nest position in the central sector (OC) or in the peripheral sector (OP) of the farms for each bird was counted as response variables (1: nest located in the periphery or interior, 0: nest not located in the periphery or interior), and were examined with PCA (only factors with eigenvalues >1.0 were considered).
Similarly, to examine the relevance of threatening factors to influence the productivity of each bird during the breeding season, desertion, predation attacks, demolition, and infestation (only for magpie infested by the parasite) were counted as predictors of fledging likelihood in Columbidae and magpie (responses: with 1 (at least one chick has emancipated) and 0 (no chicks produced)), and examined via binomial error structure (generalized linear model) with a log it link function.
3. Results
3.1. Nest-Niche Selection
Columbidae, turtle, and collared dove selected the same breeding habitats inside olive orchards N1 and N2, while the Maghreb magpie was isolated alone inside orchard N3 (Figure 1). However, the migrant dove selected mostly nesting trees in the central sector of the olive farms (87%; 96 nests), while the invasive dove selected principally olive nesting trees in the marginal area of orchards (79%; 40 nests) (Figure 4). Further, turtle dove nests were placed far from human impact, counting infrastructures, roads, and farmers (Figure 5). On the opposite, the collared dove nested close to infrastructures and in the areas dominated by farmers.
Breeding species selected different nest positions on olive-supporting trees (except distance from the nest to the central trunk of the tree NDCT) (Table 1). Both doves selected taller olive trees but nested at a lower height in comparison with magpie. Similarly, nest dimensions were variables among species. Magpie nests were characterized by grater dimensions, while doves’ nests were comparable and categorized by medium-sized platforms.
3.2. Breeding Dates
The breeding timeline of studied species, counting nest construction, laying, and hatching dates at Fez are summarized in (Figure 6). Magpie started nesting activities first during the fourth week of March. Turtle dove and collared dove nesting dates were on the fourth week of April. Similarly, laying activities were earlier in Magpie (third week of April), while T. dove and C. dove were late (second week of May). On the other hand, hatching dates were different among studied birds. First chicks were recorded for the magpie (first week of May), followed by collared dove (fourth week of May) and later turtle dove (first week of June).
3.3. Reproductive Rates and Failure Factors
Productivity was variable among monitored species (Table 2). Migrant and resident doves have the highest breeding success rates during nesting (N = 7, DF = 2, F = 140.21, ), laying (N = 7, DF = 2, F = 82.34, ), and fledging phases (N = 7, DF = 2, F = 12.79, ) in comparison with Maghreb magpie. However, the invasive dove showed the highest breeding success during nesting and laying phases in comparison to the native dove.
Predation, mainly magpie attacks were the most failure causing loss of both doves’ clutches. Predators attacked 29.25% of turtle dove and 25.31% of collared dove clutches. Human disturbance, desertion, and destruction have caused loss of 18.98% and 5.83% of clutches in turtle doves, as well as 13.45% and 2% of collared dove clutches successively. On the other hand, Maghreb magpie clutches were failed mainly due to desertion (38.46% of nests), destruction (15.38% of nests), and parasitism (3.84% of nests, 20.83% of eggs, and 25% of nestlings) (Table 3).
4. Discussion
To our knowledge, this is the deep study of the nest-niche ecology of the vulnerable turtle dove in the presence of its sympatric competitor and predator [16, 25]. Our main goal was to provide comprehensive data on the segregation of nesting sites, breeding dates, and reproductive success of turtle dove co-occurring with collared dove and Maghreb magpie. These findings are the first and only provided results related to nest-niche ecology of turtle dove, collared dove, and magpie in Morocco and the entire Northwest African zone, which is of great interest for future comparative studies and the employment of a possible conservation plan of the endangered Moroccan Turtle dove (globally threatened) and endemic North African Maghreb magpie populations.
This study revealed the significant segregation of nest-niche and breeding chronology between turtle doves and co-occurring collared dove and Maghreb magpie. Both doves nested jointly in orchards N1 and N2, while the magpie was isolated in the orchard N3. Similarly, nest placement was different among monitored birds. Despite the use of the same orchards, T. doves have nested mainly on olive trees situated in the central sector of the orchards, while C. dove nests were abundant in the peripheral areas of the olive orchards. The spatial segregation between both doves and Maghreb magpie is suggested to separate between nests of both Columbidae and predatory Maghreb magpie considered as a direct predator for passerine birds [30], and to protect clutches from predation risk [31]. Moreover, the spatial segregation recorded between nests of both Columbidae is suggested to reduce competition between the invasive collared dove and the native turtle dove, as mentioned currently in Moroccan farmlands [16, 25], Algeria [32], and previously in the west of the Iberian Peninsula [33], where the expansions of the invasive dove in urban areas have forced T. dove to select breeding sites far away in forest ecosystems, to avoid competition for available food and nesting resources [1, 7]. In our case, the competition avoidance is also achieved via vertical segregation recorded in nest placement; the collared dove selected nest placement at a greater height and taller trees compared to turtle doves as revealed currently by [16] in the same area. On the other hand, Maghreb magpie breeding chronology was earlier in comparison with both doves. Magpie started nesting activities in March, while both doves initiated the construction of their nests on the last week of April. Similarly, first chicks were occurred firstly for the magpie during May, followed by collared dove and later the turtle dove. The early breeding chronology recorded in the Maghreb magpie is most likely related to food abundance targeted in both doves’ nests; the magpie started breeding activities first to benefit from eggs and nestlings of both doves (particularly late nests) to ensure sufficient food for its own nestlings (this is confirmed in mentioned failure factors). These earliest breeding dates have been mentioned in other predators; the boreal owl species start their breeding early to benefit from the abundance and availability of prey species’ nests (they attack eggs and nestlings) [34]. On the other hand, the similarity of breeding chronology between competitive doves is expected since these Columbidae are so close in their biological and ecological features include breeding [1, 7].
Breeding success was significantly higher in doves compared to the magpie during all breeding stages. Clutches of both doves were highly threatened by predation attacks, mainly from the magpie and reptiles, as well as from human disturbance, including olive-tree cutting and pulverization of pesticides in coincidence with breeding activities. These results confirm the effect of predation on both doves [15, 21, 24] and highlight the attacks conducted by the North African magpie (despite the absence of precise statistical evidence, field observations were sufficient to confirm predation attacks). On the other hand, Maghreb magpie clutches were lost mainly due to human disturbance and parasitism. Magpie nests were attacked and colonized by the great spotted cuckoo (Clamator glandarius) (Figure 7), and this is widely reported in Europe and recently in Africa, where the great spotted cuckoo (Clamator glandarius) is considered as an obligate clutch parasite and the Eurasian magpie (Pica pica) population as the regular host [28, 30, 35, 36] due to the lack of antiparasitic defenses.
5. Conclusion
This is the first study providing deep insights into the breeding habitat use, time partitioning, breeding success of the native turtle dove co-occurring with the invasive collared dove and the predatory Maghreb magpie in Morocco and entire North Africa. We have shown that both doves select nesting sites far away from the magpie to protect their clutches. Equally, competitive doves partitioned nesting sites horizontally (T. dove in the center and C. dove in the periphery of the orchards) and vertically (higher nest placement in C. dove) to reduce competition. We found that magpie starts breeding activities early to benefit from the nests of both doves and other neighboring species. On the other hand, breeding success was higher in doves despite predation attacks conducted by the Maghreb magpie and human disturbance. Correspondingly, the intelligent strategy of early breeding adopted by the magpie is broken by parasitic colonization of its nests by the great spotted cuckoo. Finally, predation fear and competition avoidance were more important in explaining spatial partitioning of breeding habitats among studied species, while food availability was more important in explaining temporal segregation. However, all these aspects, including the relationship between magpie attacks and breeding chronology of doves and potential segregation of foraging resources in co-occurring doves, need deep and close monitoring to get the real picture of these complex ecological issues.
Data Availability
All necessary data are included within the article with clear careful statement. The full data are available from corresponding author upon reasonable request for any future studies.
Conflicts of Interest
On behalf of all authors, the corresponding author states that there are no conflicts of interest.
Authors’ Contributions
WS and IM conducted the field work and collected all data. IM, HA, and MH statistically analyzed data. FF, DO, and MD contributed to results’ interpretation. WS, IM, MD, and FF were major contributors in writing the manuscript. All authors read and approved the final manuscript.
Acknowledgments
The authors thank the owners of the farms for kindly allowing them to work on their properties. This study was supported by the Regional Department of High Commission for Water, Forests and Desertification Control, in Fez, Morocco.