Introduction
Urbanization in developing countries has for long been identified as a major process with multiple consequences at the global scale (Cohen, 2006; Henderson & Turner, 2020). This process comprises both the emergence of megacities and the rapid growth of small and medium cities (Cohen, 2006; United Nations, Department of Economic and Social Affairs, Population Division 2018). As a result, urban areas are projected to house 60 per cent of people globally by 2030, with almost 90 per cent of this growth taking place in Asia and Africa (un.org). The corresponding rise of the “indoor biome” raises new eco-evolutionary questions regarding species, and species communities, associated with this expanding environment (Martin et al., 2015; Hulme-Beaman et al., 2016). Long composed mainly of native species, these communities (like others from more natural biomes) have been progressively colonized by introduced species benefitting from human-caused global changes (Vitousek et al., 1997).
Among other groups of living organisms, rodents comprise species that are especially prone to take advantage of the modification of habitats by human activities, being core species in an urbanization context (Capizzi et al., 2014; Rothenburger et al., 2017). Their impacts are diverse and multidimensional, including notably biodiversity loss (Doherty et al. 2016), threats to food security (Singleton et al., 2021), disease transmission (Han et al., 2015), economic burdens (Dossou et al., 2020; Diagne et al., 2023) and societal decay (Colombe et al., 2019). Some of these rodents are well-known as major invasive alien species (hereafter ‘invasive rodents’) worldwide (Lowe et al., 2000; Capizzi et al., 2014). This is the case of Rattus rattus , the black rat, and Mus musculus , the house mouse, which are both listed amongst “100 of the world’s worst invasive species” of the planet (Lowe et al., 2000). The ongoing expansion of those invasive rodents in several parts of the world (see for instance Dalecky et al., 2015; Zeng et al., 2018; Yu et al., 2022) leads to multispecific assemblages of small mammals (mainly rodents) that combine invasive and native species in a variety of ecological and evolutionary contexts. One of these contexts is represented by commensalism in anthropogenic environments, where the species concerned literally “live within houses”, in close proximity with humans (Hulme-Beaman et al., 2016). There, despite ever-increasing studies on the distribution (including invasion history of invasive small mammals), impacts and dynamics of individual species over space and time, the multispecific assemblages of these rodent-dominated communities has rarely been studied from a community ecology perspective.
Species sampling in anthropogenic habitats is often complicated because it involves going into people’s homes or industrial or commercial buildings. Moreover, even if rodent communities in such habitats are often depauperate, diversity is generally not taken into account because one focus on a (or a pair of) target species in relation with specific questions raised by it/them. This is the case in the review by Feng & Himsworth (2014) on R. norvegicus and R. rattus , in the studies focusing on the impact of urban characteristics on the genetic structure of rodent populations in different cities (R. norvegicus in American cities: Combs et al. 2018; M. musculus in Dakar (Senegal): Stragier et al. 2020), or in experiments on species cohabitation and interspecific competition involving M. musculusin SW Argentina (Castillo et al., 2003; Gomez et al., 2008). However, some studies have already considered more complete communities. For instance, Panti-May et al. (2012) measured data on abundance, population and habitat use parameters of M. musculus and R. rattusamong their native counterparts in households of a rural area of Mexico as part of a study on zoonotic disease transmission. Masi et al. (2010) evaluated the respective importance of socioeconomic and environmental risk factors for urban rodent (including R. rattus , R. norvegicus and M. musculus ) infestation in Sao Paulo, Brazil. Cavia et al. (2009) analyzed the relation between rodent community composition and diversity and the landscape structure in the city of Buenos Aires, showing a clear trend of habitat partitioning between invasive R. rattus, R. norvegicus and/or M. musculus(dominant in parklands, shantytowns or industrial–residential neighborhoods) and native species (dominant in a natural reserve but also present in parklands). In Africa, Olaseha et al. (1994) presented general considerations on the importance of housing and sanitation on the presence of rats (R. rattus and R. norvegicus ) and mice (M. musculus ) based on questionnaires completed by interviews in towns and villages of a rural area in southwestern Nigeria. Demby et al. (2001), followed by Fichet et al. (2005, 2009) provided information on small mammal distribution in urban as well as in rural areas of Guinea, in relation to Lassa virus distribution and prevalence. Taylor et al. (2008) gave a few elements of urban distribution of rodents in Durban (South Africa) in a small mammal community largely dominated by R. norvegicus . Monadjem et al. (2011) compared movement patterns and possible interactions ofMastomys natalensis (a native rodent species) and R. rattus in distant sites of Tanzania, Malawi and Namibia using telemetry and Rhodamine B marker. In the capital city of Niger, Niamey, Garba et al. (2014) analysed the distribution of native and invasive rodents in a series of sites corresponding to habitation districts, cultivated gardens and industrial zones. They showed the dominance of the nativeM. natalensis over the invasive R. rattus and M. musculus and spatial segregation between them, that they interpreted as the result of an ongoing native-to-invasive species turn over. Hima et al. (2019) assembled an important dataset on commensal small mammal distribution in a series of localities along the Benin-Niger “corridor”, between Cotonou and Niamey. They showed the dominance of either invasive R. rattus in Cotonou (see also Houemenou et al., 2014) or native M. natalensis in Niamey, with segregation patterns between Rattus spp. and M. natalensis , and a very regular and important presence of Crocidura spp. (incl. C. olivieri ), especially at lower latitudes. None of these studies has nonetheless addressed in detail the co-distribution and coexistence of a set of species (both native and invasive) belonging to a whole small mammal community in human-made environments, especially at a fine spatial scale. At best, they considered co-occurrence patterns at the scale of a country, a region or a whole city, but never at the level of the housing units or the buildings, where inter-individual (be they intra- or interspecific) interactions actually occur. Yet it is precisely at this fine scale that the ecological interactions take place which probably determine the trajectory of the communities in terms of their distribution in space and time.
Niche / ressource partitioning represents a way to manage coexistence among competing species within habitats (Pianka, 1973; Chesson, 2000). Indeed, the complex and interactive effects of species niche overlap, niche breadth and environmental heterogeneity on species co-occurrence patterns have been highlighted repeatedly (see a synthesis in Bar Massada, 2015). In the particular case of commensal small-mammal communities, information on and analyses of species co-distribution and co-existence, habitat partitioning (if any), and interspecific interactions between species (including invasive ones) are lacking, being however of paramount importance to better understand: i) the way invasive species spread at the microhabitat scale; ii) the consequences of this spread on the distribution of native species at the microhabitat scale ; iii) the actual associations between species likely to represent zoonotic disease reservoirs at the very contact with humans.
In Africa, both R. rattus and M. musculus colonized most countries via boats of European or Arab settlers, often centuries ago (Happold, 2013). Long confined to coastal areas and larger cities, they have been spreading continuously over inland areas thanks to the development of infrastructures and associated human exchanges (e.g. movements of goods and people) that accompanies the ongoing urbanization of rural areas. This is the case in Senegal (West Africa) where bothR. rattus and M. musculus have experienced recent range expansion eastward from the western Atlantic coastal areas (Duplantier et al., 1991; Dalecky et al., 2015; Konečný et al., 2013). Being exclusively commensal in this country, they encounter native species that inhabit human settlements, leading to inevitable interactions which ultimately determine the patterns of cohabitation between them. To describe this coexistence and try understand the underlying interactions, we sampled communities of commensal small mammals from localities of various size within the southern half of Senegal (corresponding to the distribution range of R. rattus in the country) within a 3-year time period. The sites sampled were widely invaded by the black rat and/or the domestic mouse, which most of the time cohabited with a wide spectrum of native rodent and shrew species. We aim at providing novel insights from the following questions: i) which invasive and native species compose the small mammal community across the different localities targeted? ii) what are the preferred habitat types and ecological niches of each of these species? iii) do these species show particular interspecific associations (segregation or aggregation) globally and/or locally? To answer these questions, we investigated here the composition, geographic distribution, micro-habitat use, ecological niche breadth and overlap, and species co-occurrence within the target small mammal community at various spatial scales.