Spatial and seasonal use of habitat by birds in northern Nigeria

ABSTRACT

Capsule: Although woodland habitat supports higher avian species diversity and abundance, shrubland may be more important for migratory species.

Aims: To investigate the use of habitat, site and season by bird species according to feeding guild, residence status and population trend in Dutse, Northern Nigeria.

Methods: A point transect was used to count and identify birds between 0630 and 1100 hours from August 2015 to February 2016 across habitats, sites and seasons in Dutse.

Results: Frugivore abundance was significantly higher in woodland habitat. The woodland and shrubland habitats had a higher proportion of resident and migratory species, respectively. Birds with stable population trends were more abundant in woodland, whereas the abundance of birds with decreasing trends were generally similar across most sites. Most of the birds with decreasing trends were migratory species. The dry season had a higher abundance and diversity of birds across different feeding guilds, residence status and population trend status than the wet season.

Conclusion: Both woodland and shrubland in Dutse were important bird habitats, but migrant species preferentially used shrubland. The shrubland habitat and the dry season should be prioritized for conservation consideration, especially for migrant birds in Dutse.

Bird distribution may vary spatially and temporally due to their habitat preferences and environmental conditions (Anderson & Cribble 1998, Ballance et al. 2006, Amorim et al. 2009). As one of the best-studied taxa, birds are known for their various ecological roles and significance in many ecosystem processes (Şekercioğlu et al. 2004, Millennium Ecosystem Assessment 2005, Şekercioğlu 2006). For example, about 70% of bird species are important seed pollinators and dispersers (Clout & Hay 1989, Anderson et al. 2003, Holly et al. 2007); approximately 4% of the world's flora are pollinated by nectarivorous bird species (Rebelo 1987, Pauw & Louw 2012). The Long-billed Malachite Sunbird Nectarinia famosa plays the vital role as an exclusive pollinator of plant species in Cape Town, South Africa (Pauw & Louw 2012). Frugivores help with seed dispersal, as exemplified by the White-thighed Hornbill Ceratogymna cylindricus and the Black-casqued Hornbill Ceratogymna atrata (Holbrook et al. 2002). Understanding the spatiotemporal distribution of birds in relation to their ecological functions is of great importance, especially in light of rapid anthropogenic-driven changes to many African landscapes (Fernandez-Juricic et al. 2004, Tøttrup et al. 2012, Milne & Bennett 2013).

The Senegal and Niger rivers are two important river basins that support most of the Sahelian habitats in West Africa and are the reason why West African habitats, including those in Nigeria, serve as wintering grounds for many migrant bird species (Morel & Morel 1992). As a result, more than two billion Palearctic migrant birds spend their wintering period in these areas (Rabøl 1987, Hahn et al. 2009, Zwarts et al. 2009, Mallord et al. 2016). These huge numbers of visitors may influence the ecology of resident bird species through increased competition for habitat and spatial use in different seasons (Rabøl 1987). Migrants are less neophobic, eurytopic and generalist (Leisler 1992, Ivande & Cresswell 2016) on their wintering grounds, making it easier for them to adapt to various habitats and resources (Leisler 1992). Some studies have shown that, in some instances, migrants may outnumber their resident congeners in their habitats (Lack 1985, Kelsey 1992). This may result in increased competition for space and habitat use amongst resident and migrant birds.

Habitat loss and environmental degradation have been associated with population decline in migrant birds on their West African wintering grounds (Kelsey 1992, Wilson & Cresswell 2006, Cresswell et al. 2007, Adams et al. 2014, Vickery et al. 2014). In comparison to residents, migrant birds may be more susceptible to loss or degradation of habitat (Vickery et al. 2014). Some studies in Nigeria have shown that human activities such as wood removal, grazing, forest conversion to farmland, hunting and other disturbances have led to habitat loss and a decline in numbers of both resident (e.g. the vulnerable Beaudouin's Snake Eagle Circaetus beaudouini, BirdLife International 2017a) and migrant birds (e.g. the near threatened Pallid Harrier Circus macrourus, Kelsey 1992, BirdLife International 2017b). However, the causes of declines of many birds in West Africa, including Nigeria, are still poorly understood due to lack of information on habitat use and foraging ecology of both resident and migrant birds (Jones 1985, Kelsey 1992, Leisler 1992, Vickery et al. 2014, Mallord et al. 2016).

Bird population trends are good environmental and biodiversity indicators (Gregory et al. 2005, Bachler & Liechti 2007). Available information from the International Union for Conservation of Nature (IUCN) Red List and Birdlife International showed that many birds in Nigeria are exhibiting decreasing population trends (e.g. Crested Lark Galerida cristata, Green-backed Heron Butorides striata, Blue-naped Mousebird Urocolius macrourus, Bruce's Green Pigeon Treron waalia) (Birdlife International 2016, IUCN 2016). A number of these bird species are known to occur in Dutse, a West African Sahelian habitat in northwestern Nigeria, where the impact of human activities and environmental change is on the increase.

Muhammad et al. (2018) found that season and landscape variables (vegetation type and site) were significant drivers in structuring Dutse bird assemblages. This paper will extend that work by investigating how spatial use differs seasonally based on feeding guild, residence status and status of current population trends of birds in Dutse. In addition, we also provide the first comprehensive checklist of birds in Dutse which will serve as an important local baseline for avian information. Knowledge of spatial use by birds in different seasons in Dutse will aid in identifying bird species of conservation concern as well as in determining avian conservation priorities in light of limited resources.

Methods

Study area and design

Dutse is the capital of the Jigawa state in northwestern Nigeria (11°42′04″N 9°20′31″E), and has a hot, dry semi-arid climate. Like most northern states in Nigeria, Dutse experiences a short wet season and a long dry season. The dry season usually starts from October to May while the wet season usually lasts for four months from June to September. Total annual rainfall ranges from 600 mm in the north to 1000 mm in the south. Mean daily temperatures can be as low as 20°C and up to 42°C from March to September. The cooler temperatures occur between October and February. Relative humidity is low from January to March when it is moderated by the Harmattan (cold-dry and dusty trade wind blowing over the West African subregion) but highest in August during the peak of the wet season. The vegetation of Dutse is primarily Sudan savannah and is characterized by few scattered trees and extensive open woodland and grasslands. Most trees are from the families Ceselpinaceae and Mimosaceae. Dutse is located close to an Important Bird Area (IBA), the Hadejia – Nguru Wetlands; these two sites share similar climatic conditions and have some bird species in common (Muhammad et al. 2018).

Dutse is mainly agrarian, with about 80% of the total land area being used for agricultural purposes. Grazing activities are prominent as many of the people engage in livestock rearing. Most of the farmers in Dutse cannot afford mechanized farming, therefore, cattle or rudimentary tools were used to plough land and there was a reliance on rain for irrigation. These practices have resulted in small-scale farming, with the majority of the smallholder farmers practicing a polyculture farming system on farms averaging 2.5 ha. In most cases, groundnut, sorghum and beans are planted simultaneously within the same area and land is left fallow during the dry season. Harvesting of firewood is common as many of the households and bakeries in Dutse still depend on it for fuel.

Four sites in Dutse, namely Wangara, Warwade, Malamawa and Model were selected for data collection (Figure 1). The sites were categorized into two different habitat types based on the major vegetation type; Wangara was dominated by shrub vegetation, whereas the other three sites were mainly woodland dominated by trees greater than 3 m in height. Tree types and structure differed among the sites. The dominant tree species in Model was Date Palm Phoenix dactylifera while the other three sites were dominated by Neem Azadirachta indica, Baobab Adansonia digitata, Doum Palm Hyphaene thebaica, Acacia species (such as Balanites aegyptiaca, Acacia seyal, Acacia senegalensis) and Ficus species. The model was closest to the township of Dutse and therefore had more human settlement and farms compared to the other sites.

Figure 1. Map of (a) the Jigawa state, Nigeria, showing (b) Dutse and the study sites.

Six transects, each 2 km in length, were established at each site. Eleven points were laid on each transect, such that each successive point and transect were 200 m apart, which allowed for sampling independence and habitat variable recording (Greenberg et al. 1997, Manu & Cresswell 2007, Sorace & Visentin 2007, Mitchell et al. 2008, Chang et al. 2013). In total, 264 points on 24 transects of 48 km in cumulative length were used to survey birds across sites, habitats and seasons (Table 1).

Table 1. Summary of sampling effort, habitat variables, bird species richness, abundance and diversity estimates in Dutse, Nigeria.

Site	Model	Malamawa	Wangara	Warwade
Dominant habitat	Woodland	Woodland	Shrubland	Woodland
No of transects in cold season	24	24	24	24
No of transects in wet season	18	18	18	18
Total no of transects surveyed	42	42	42	42
Total no of individuals recorded	4148	3489	2580	3429
Observed species richness	86	74	77	87
Chao (richness)	66	66	65	66
Chao (Shannon)	4.06	3.88	4.00	3.98
Chao (Simpson)	0.981	0.974	0.979	0.977
Average tree density per transect	9.1	9.2	8.9	7.7
Average no of shrubs per transect	22.7	21.7	22.9	27.4
No. of vulnerable species	1			1
No. of threatened species				1

Bird counts

The survey was carried out from August 2015 to February 2016, spanning both the wet and the cool, dry seasons. All bird recordings were conducted in calm weather conditions from dawn at 06:30 to 11:00 (Jones et al. 1996, Cresswell et al, 2007, Manu & Cresswell 2007, Manu et al. 2007, Sorace & Visentin 2007, Manu et al. 2010). Bird count would start immediately upon arrival at a point (Stevens et al. 2010). Birds were recorded for five minutes within a 50 m radius at each point; the short survey period reduced over-estimation of densities and allowed for more points to be surveyed in a given time duration (Lee & Marsden 2008). From one point to another on each transect, the observer walked at a slow and steady pace of approximately 40 m per minute. Birds recorded by sound were assumed to be single individuals unless calls were heard simultaneously or from different directions. For consistency, all bird surveys and habitat recordings were carried out by the first author throughout the study using a pair of Olivon binoculars (8 × 40). A standard field guide was used to confirm the identification of sighted birds (Borrow & Demey 2014).

Data analyses

Point abundance was calculated by summing up the total number of recorded birds by sight or sound per point, transect and site. All birds were categorized by feeding guild (granivores, insectivores, frugivores, omnivores, nectarivores, carnivores or piscivores) and by population trend (stable, decreasing, increasing and unknown), with the aid of the Handbook of the Birds of the World (del Hoyo et al. 2016) and the IUCN Red List of Threatened Species (IUCN 2016). Using Borrow and Demey (2014), residence status of bird species was classified as breeding visitor (BV), non-breeding visitor within main range (NBVMR), non-breeding visitor with sparse occurrence (NBVSO), resident (R), and resident but partially migratory (RPM).

Chao richness and Shannon and Simpson diversity estimates were calculated by the site. To determine if the survey effort was generally sufficient and to compare diversity estimates across sites, sample-size-based rarefaction and extrapolation (R/E) curves and sample completeness curves were generated for each site to show species richness (Hill number order (q) = 0) and Simpson diversity (q = 2) (Chao et al. 2014). Similarity Percentage (SIMPER) analysis using the Bray–Curtis dissimilarity was performed to ascertain which bird species contributed most to the variation in bird assemblages across habitats, sites and seasons.

A one-way multivariate analysis of variance (manova) was performed to test for differences in the abundance of birds by feeding guild, residence status and population trend across the categorical variables of habitats (woodland and shrub), sites (Wangara, Warwade, Malamawa and Model) and seasons (wet and dry). The Pillai trace was used as the test statistic. For the feeding guild, only insectivores, granivores and frugivores were included in the manova analysis due to the low numbers recorded for other feeding guilds. For the purposes of the manova analysis, birds with the statuses of R and RPM were considered as residents whereas BV, NBVMR, and NBVSO were considered as migrants; we also excluded birds with unknown population trend due to low sample size. We used Gignac & Szodorai (2016) to define partial eta-squared value in reporting effect size. SIMPER was performed using the Paleontological Statistics Software Package (PAST) version 3.18 (Hammer et al. 2001). All other analyses were conducted using R (R Development Core Team 2016), with additional functions used from the packages vegan (Oksanen et al. 2017), for diversity estimate calculation, iNEXT (Hsieh et al. 2014) for sampling curves, and heplots (Fox et al. 2017) for determination of partial eta-squared values.

For the purpose of providing a comprehensive avian checklist for Dutse, some bird species that were recorded outside the point count surveys were added to the checklist, although these were not included in the statistical analyses.

Results

A total of 13,656 individual birds of 128 species from 46 families were encountered during the survey, both within and outside the count areas (online Appendix S1). Sample completeness curves suggested that sampling effort to characterize avian species diversity across the sites was sufficient (Figure 2). Even though sample coverage and estimated species richness were similar for all four sites, the Simpson diversity was highest in Model, followed by Wangara, Warwade and Malamawa (Table 1, Figure 2). Of the 14 vulnerable and near threatened bird species in Nigeria, the Beaudouin's Snake Eagle (categorized as vulnerable under the IUCN Red List of Threatened Species) and Pallid Harrier (categorized as near threatened) were both recorded in Dutse (Appendix S1, Table 1). Among the bird families, Accipitridae had the highest number of bird species (30%), followed by Sylviidae (22%) and Estriididae (15%).

Figure 2. Sample-size-based rarefaction (left column) and extrapolation (right column) curves, showing species richness (q = 0) and Simpson diversity (q = 2) by sites in Dutse, Nigeria.

Results based on the SIMPER analysis showed the most important bird species responsible for the dissimilarities in species composition both across and between habitats, sites and seasons (Figure 3); most of these bird species are gregarious and colonial in nature. The average dissimilarities between habitats, sites and season were 52.24%, 51.34% and 87.97%, respectively.

Figure 3. Bird species that cumulatively contributed >1% dissimilarity of the species composition across seasons (a), sites (b) and habitats (c) in Dutse, Nigeria.

The majority of the recorded bird species in this study were insectivores (48%), followed by frugivores (16%), granivores (16%), carnivores (15%), nectarivores (3%) and piscivores (2%) (Appendix S1). However, in terms of numerical importance, frugivores accounted for the highest abundance (40%), followed by granivores (33%), insectivores (24%), nectarivores (2%), carnivores (1%) and piscivores (<1%). Within the same feeding guild, abundance of birds was significantly different across habitats (F_{1, 259} = 14.84, P < 0.001, Pillai's Trace = 0.15), sites (F_{3, 774} = 6.50, P < 0.001, Pillai's Trace = 0.21), and seasons (F_{1, 518} = 18.10, P < 0.001, Pillai's Trace = 0.10) with moderate (partial eta-squared, η² = 0.15), low (η² = 0.07) and moderate (η² = 0.10) effect sizes, respectively. Frugivore abundance was significantly higher in woodland habitat than in shrubland (Figure 4, Table 2). Likewise, frugivore abundance was significantly different across sites, being highest in the Model site, followed by Warwade, Malamawa and Wangara (Figure 4, Tables 2 & 3). The abundance of insectivorous and granivorous birds was not significantly different across habitats or sites. The dry season harboured higher abundances of insectivores, granivores and frugivores than the wet season (Figure 4, Table 2).

Figure 4. Bird abundance by feeding guild, residence status and population trend between habitats (shrubland and woodland), sites (Model, Wangara, Malamawa and Warwade) and seasons (wet and dry) in Dutse, Nigeria.

Table 2. manova results of mean abundances of bird species by feeding guild, residence status and population trend across habitats, sites and seasons.

	Variable	DF	F-value	Pr(>F)
Frugivore	Habitat	1	31.38	P < 0.001
	Site	3	14.63	P < 0.001
	Season	1	29.42	P < 0.001
Insectivore	Season	1	40.99	P < 0.001
Granivore	Season	1	7.21	0.008
Resident	Habitat	1	11.86	P < 0.001
	Site	3	4.95	0.002
	Season	1	32.16	P < 0.001
Migrant	Habitat	1	19.90	P < 0.001
	Site	3	10.74	P < 0.001
	Season	1	64.66	P < 0.001
Stable	Habitat	1	11.17	P < 0.001
	Site	3	4.78	0.003
	Season	1	28.32	P < 0.001
Decreasing	Habitat	1	0.05	0.820
	Site	3	3.51	0.016
	Season	1	8.27	0.004
Increasing	Habitat	1	0.05	0.827
	Site	3	1.55	0.203
	Season	1	23.05	P < 0.001

Table 3. Pair wise comparison of bird abundance by feeding guild, residence status, and population trend across sites.

	Pair wise	DF	Pillai	F	Pr(>F)
Guild	Wangara – Model	1	0.33	20.67	P < 0.001
	Wangara – Malamawa	1	0.14	6.82	P < 0.001
	Wangara – Warwade	1	0.26	14.35	P < 0.001
Status	Wangara – Model	1	0.19	15.20	P < 0.001
	Wangara – Malamawa	1	0.22	17.54	P < 0.001
	Wangara – Warwade	1	0.19	14.49	P < 0.001
Trend	Malamawa – Model	1	0.07	3.00	0.033
	Malamawa – Wangara	1	0.13	6.43	P < 0.001
	Malamawa – Warwade	1	0.10	4.78	0.003

Of the recorded bird species in Dutse, 72% were resident birds, 10% non-BVs, 8% partially migratory residents, 6% non-BVs, and 4% BVs (Appendix S1). There were statistically significant differences in bird abundance by residence status across habitats (F_{1, 260} = 22.6, P < 0.001, Pillai's Trace = 0.15), sites (F_{3, 518} = 9.16, P < 0.001, Pillai's Trace = 0.19) and seasons (F_{1, 519} = 41.51, P < 0.001, Pillai's Trace = 0.12). Partial eta-squared values indicated moderate effect size across habitats (η² = 0.15), sites (η² = 0.10) and season (η² = 0.14). The abundances of resident and migratory birds differed significantly between woodland and shrubland habitats (Table 2); resident birds were more abundant in woodland than in shrubland, whereas the abundance of migrants was higher in shrubland (Figure 4). The abundance of resident and migrant birds differed significantly between sites (Tables 2 & 3), with higher abundance of residents recorded in Model and Malamawa sites followed by Warwade and Wangara (Figure 4). On the other hand, the highest abundance of migrant birds was recorded in Wangara followed by Model (Figure 4). Dry season had higher abundance of both residents and migrant birds than wet season (Figure 4, Table 2).

Population trend assessments by the IUCN Red List of Threatened Species (IUCN, 2016) suggested that 57% of the recorded bird species in the study area were stable, 20% decreasing, 18% increasing and 5% unknown (Appendix S1). The majority of the bird species with decreasing population trends were migrants, with a few resident species. These resident species include the Crested Lark Galarida cristata, Hoopoe Upupa epops, Yellow-billed Oxpecker Buphagus africanus and Northern Carmine Bee-eater Merops nubicus. There were significant differences in the abundance of birds by population trend across habitats (F_{1, 259} = 4.45, P = 0.004, Pillai's Trace = 0.05), sites (F_{3, 774} = 3.26, P < 0.001, Pillai's Trace = 0.11) and seasons (F_{1, 518} = 14.53, P < 0.001, Pillai's Trace = 0.08) although the effect sizes were small (η²< 0.1 for all). There was no significant difference in bird abundance by population trend between habitats except for the stable population trend (Table 2). Woodland habitat had higher abundance of birds with a stable population trend than shrubland (Figure 4). The abundance of birds with stable and decreasing population trends was significantly different between sites but not for those with the increasing trend (Table 2). Comparison between sites showed that the abundance of bird species with decreasing trend was similar between Model, Wangara and Warwade which were higher than that in Malamawa (Figure 4), whereas Model had higher abundance of birds with stable population trend. The dry season had a significantly higher abundance of birds with stable, decreasing and increasing trends than the wet season (Table 2, Figure 4).

Discussion

This study provides the first comprehensive checklist of birds in Dutse, Nigeria and characterizes spatiotemporal utilization patterns of birds by feeding guild, residence status and population trend. This work extends the initial published study by Muhammad et al. (2018) on seasonality and landscape effects on Dutse bird assemblages. According to Birdlife International (2016), Nigeria has a total of 863 bird species, while 894 had been recorded in Fishpool & Evans (2001). Most of these bird species (94.7%) have been categorized as least concern. Our study recorded a total of 128 bird species which accounted for 15% of the known Nigerian bird species richness. Among the bird families, Accipitridae, which includes the near threatened Pallid Harrier, had the highest number of species but recorded abundance was low compared to other bird families.

Dutse appears to serve as a foraging ground for a large number of bird species and families (Appendix S1). The dominant and highly gregarious White-rumped Seedeater Serinus leucopygius, Red-billed Quelea Quelea quelea, Village Weaver Ploceus cucullatus, Rose-ringed Parakeet Psittacula krameri and Chestnut-bellied Starling Lamprotornis pulcher were always recorded in large colonies. Their relatively high abundance and dominance across habitat, site and season could be explained by their affinity with trees and farmlands that provide food and shelter. As an example, the White-rumped Seedeater can be found using trees to get supplementary food from farmlands. These species also tend to be gregarious and display a colonial lifestyle (Borrow & Demey 2014).

This study found frugivory to be the dominant feeding guild in Dutse with the highest abundance in the woodland habitat and in the Model site dominated by woodland. This suggests that habitats and sites with a higher number of trees may attract more birds due to fruit availability, cover and nesting areas. Several studies have associated the abundance and diversity of fruits in the tropical region (like Nigeria) with frugivore abundance and diversity (Orians 1969, Bleher et al. 2003). In Dutse, most of the tree species in the woodland areas produce fruits and nectar that attract frugivores and nectarivores (Lack 1987, Gentry 1990, Borghesio & Laiolo 2004, Seoane et al. 2013); typically fruiting occurs at the end of the wet season and through some months during the dry season.

The higher abundance of birds by feeding guilds recorded in the dry season could be explained by the presence of migrant birds wintering in Dutse such as insectivores (e.g. warblers), granivores (e.g. Sudan Golden Sparrow Passer luteus and Four-banded Sandgrouse Pterocles quadricinctus) and frugivores (e.g. African Golden Oriole Oriolus auratus). The higher number of insectivorous species recorded in this study could be associated with the abundance of farmland (secondary habitat) which provides additional food sources such as insects especially during cultivation and harvest.

This study coincided with the wintering period for the majority of migrants between September and October (Aidley & Wilkinson 1987, Hulme & Cresswell 2012), which allowed them to be recorded. Even though a greater percentage of the recorded bird species were residents, a good number of breeding and non-breeding migratory bird species were recorded as well as many which were Palearctic migrants. Out of the 251 migratory bird species in Nigeria (Birdlife International 2016), 35 species (14%) were recorded in this study, which accounted for 20% in abundance, suggesting that Dutse may serve as an important wintering ground. A study in similar habitat in Northern Nigeria recorded 15 migrants and 76 Afrotropical bird species (Jones et al. 1996).

Higher abundance of resident birds was mostly associated with woodland habitat, especially in the Model area, while migrants were more abundant in shrubland habitat. The majority of the migrants recorded were shrubland species such as Common Whitethroat Sylvia communis, Woodchat Shrike Lanius senator, Northern Wheatear Oenanthe oenanthe and Montagu's Harrier C. pygargus, farmland related species, such as Yellow Wagtail Motacilla flava, or birds that preferred medium-sized trees, such as Common Redstart Phoenicurus phoenicurus.

This study found that 57%, 20% and 18% of the recorded bird species to be displaying stable, decreasing and increasing population trends, respectively, with migrant birds having a higher number of species with decreasing trends, for example Montagu's Harrier, Woodchat Shrike and Yellow Wagtail. Other studies have reported decreasing population trends of migrant birds wintering in central and northern Nigeria, which highlights the conservation concerns particularly for the wintering grounds (Sanderson et al. 2006, Hulme & Cresswell 2012, IUCN 2016).

Common bird species that are highly gregarious and colonial, such as Chestnut-bellied Starling, Brown Babbler Turdoides plebejus, Greater Blue-eared Starling Lamprotornis chalybaeus and White-billed Buffalo Weaver Bubalornis albirostris were usually recorded in large numbers and had stable population trends. These birds were typically found in woodland habitats, and were abundant at the Model site and during the dry season. While these bird species do not appear to be of any particular conservation concern presently, ongoing monitoring of their population trends is warranted in light of anticipated future spatiotemporal changes in Dutse.

The name ‘Dutse’ means hilly rocks. These are common landscape components that are found abundantly within and around the Dutse town. However, due to the nature of the terrain and difficulty in accessing rocky areas, this habitat was not included in the sampling areas. Species associated with rocky habitats were, therefore, not recorded at the point counts, but species such as Cinnamon-breasted Bunting Emberiza tahapisi and Stone Partridge Ptilopachus petrosus, were known to occur in abundance in Dutse. Similarly, bird species like Northern Carmine Bee-eater, White Stork Ciconia ciconia, and some highly seasonal species such as the Abdmin Stork Ciconia abdimii were often recorded outside the point count stations. In addition, this study does not adequately cover aquatic and nocturnal species. A few water-related bird species such as Green-backed Heron B. striata, Black-headed Heron Ardea melanocephala and nocturnal species such as Pearl-spotted Owlet Glaucidium perlatum and Africa Scops Owl Otus senegalensis were occasionally observed in Dutse, but not recorded during the point count surveys. Future surveys in Dutse should address these gaps.

In conclusion, both woodland and shrubland in Dutse are important bird habitats. Resident and migrant birds both use the woodland area, however, migrant birds appeared to be associated more closely with shrubland. The high number of both resident and migrant birds wintering in this area highlights the importance of Dutse as an important wintering ground in the region. Due to records of two species of birds of concern, Wangara is a potential candidate area for avian conservation priority. Farming as well as other anthropogenic activities are expected to increase in Dutse due to growing population size and demand for food (Zangina 2015). This will likely put even more pressure on areas like Model due to its central location. Intensive farming practices, deforestation and overgrazing of livestock have affected soil stability and fertility (Mansur 2014) and will likely cause further degradation of habitats for resident and migrant birds. Comprehensive land development plans coupled with improved agricultural practices that incorporate biodiversity concerns will be important for avian conservation and management in Dutse. Shrubland areas, the Model and Wangara sites and the dry season should be prioritized when planning for conservation efforts, especially for the migrant bird species.

Acknowledgment

We would like to thank the Nigerian government for providing a scholarship to the first author to undertake this study as part of his PhD research. We wish to thank our field assistant, Mr Habibu Kawaye for his time and tireless effort during the data collection process.

ORCID

Sulaiman I. Muhammad http://orcid.org/0000-0002-8711-0653

Amy Y. Then http://orcid.org/0000-0001-6558-8237

Rosli Ramli http://orcid.org/0000-0003-0410-9604