Introduction

With the current rate of population growth particularly in the developing countries of Africa, Asia, and Latin America (Cilliers, Cilliers, Lubbe, & Siebert, 2013), the world population is estimated to reach 9.6 billion inhabitants by 2050, with around 70% expected in urban areas (United Nations, 2014). According to United Nations Habitat (2014, p. 268), between 2010 and 2015, Africa has been urbanized at a higher annual rate (1.1%) than the world average rate of urbanization (0.9%), which is also reflected by the number of cities with more than 1 million inhabitants that has drastically risen from one city in 1950 to 52 in 2014 and is expected to reach 68 in 2025. Such demographic explosion in urban areas causes a loss of green space cover due to fragmentation (Kong & Nakagoshi, 2005; McDonald et al., 2009; Millennium Ecosystem Assessment, 2005), as exemplified by the case of Abuja city (Nigeria) where a 7.1% annual loss of green space cover has been noted between 2000 and 2006 (Fanan, Dlama, & Oluseyi, 2011). Similarly, a loss of 380 indigenous species has been recorded in the city of Plzen (Czech Republic) over a period of 120 years (Chocholouskova & Pyšek, 2003).

In some African cities, the loss of indigenous species has been so high that the exotic species currently represent up to 80% of the flora (Bigirimana, Bogaert, De Cannière, Lejoly, & Parmentier, 2011; Rija, Said, Mwamende, Hassan, & Madoffe, 2014). Some of these species even become invasive, which can have consequences for biodiversity and ecosystem function, as well as for the economy of the countries involved (Doren, Volin, & Richards, 2009). Clearly, such urbanization-driven ecosystem simplification and biotic homogenization affects the ecological quality of green spaces, reducing their role as potential habitat for plants and animals, as well as recreational area for urban dwellers (Millennium Ecosystem Assessment, 2005).

In fact, the presence of green spaces regulates the microclimate, reduces noise, prevents water and air pollution, and contributes to the decontamination of municipal waste, thus, playing a major role in biodiversity preservation (Bolund & Hunhammar, 1999; Kong, Yin, James, Hutyra, & He, 2014). Therefore of it is prime importance to understand the spatial pattern and the quality of green spaces in urban areas, in order to design appropriate strategies for their sustainable management. While studies on the impact of urbanization on spatial pattern and ecological quality of green spaces are well documented in Europe, America and Asia (Gong, Chen, & Yua, 2013; Kong & Nakagoshi, 2005; La Rosa & Privitera, 2013), little is known for sub-Saharan African cities (Barau, Ludin, & Said, 2013; Bernholt, Kehlenbeck, Gebauer, & Buerkert, 2009; Bigirimana et al., 2011; Cilliers et al., 2013; Fanan et al., 2011; McConnachie, Shackleton, & McGregor, 2008; Osseni, Mouhamadou, Tohozin, & Sinsin, 2015), despite their favorable geographical location for green space establishment (Mensah, 2014); and Lubumbashi city in the south-eastern part of the Democratic Republic of the Congo forms no exception to this hypothesis.

As the pattern and ecological quality of green spaces has been only scantily investigated in Lubumbashi city (Leblanc & Malaisse, 1978; Malaisse & Mutihac, 1976; Maréchal, Useni, Bogaert, Munyemba, & Mahy, 2018), a rapid and unplanned urbanization is generally observed. The continuous destruction of green spaces is evident in the city center where they are currently replaced by built-up, as well as in the new allotments in city periphery, where the discontinuous built-up establishment is accompanied by a systematic destruction of preexisting vegetation (Bruneau & Pain, 1990; Nkuku & Rémon, 2006; Useni, Cabala, et al., 2017), leading to a type of zone referred to as the peri-urban zone of the city (André, Mahy, Lejeune, & Bogaert, 2014).

This study was designed to fill these gaps by characterizing the green spaces along an urbanization gradient in Lubumbashi. Considering the fact that the traditional urban–rural opposition was completed taking into account the peri-urban zone (Bogaert, Bioloso, Vranken, & André, 2015; Trefon & Kabuyaya, 2015), it was crucial in the present to study green spaces in urbanized areas by considering separately urban and peri-urban green spaces. Therefore, the established theory of the urban–rural gradient (Kong & Nakagoshi, 2005; Li, Li, Zhu, Song, & Wu, 2013; McDonnell & Hahs, 2008; Niemelä et al., 2002; Qureshi, Breuste, & Lindley, 2010; Rija et al., 2014) was used in this study.

This study tests the following hypotheses: (a) green spaces in the urban zone are more fragmented and characterized by smaller areas than the peri-urban ones, (b) the proportions and types of green spaces are different between the urban zone and the peri-urban zone, and (c) compared to the peri-urban zone, green spaces in the urban zone are more equipped and characterized by higher proportions of anthropogenic vegetation while showing less vegetation strata and a higher proportion of exotic species.

Methods

Study Area

This study was carried out in Lubumbashi (Figure 1), the capital of the Upper Katanga Province, in the eastern part of the Democratic Republic of the Congo. Located between 11°27′ and 11°47′ S and between 27°19′ and 27°40′ E, the city is subdivided in seven municipalities, covering an area of about 747 km² (Nkuku & Rémon, 2006), mostly dominated by red and yellow lateritic soils (Leblanc & Malaisse, 1978). The climate is humid subtropical (Cw type of the Köppen classification system; Bultot, 1950), with one rainy season (November–March) and one dry season (May–September), separated by October and April as transitional months. From a phenological point of view, Malaisse (1974, p. 270) has distinguished five seasons: early rainy season (October–November), main rainy season (December–February), late rainy season (March–April), cold dry season (May–July), and warm dry season (August–September). The average annual rainfall is close to 1,200–1,300 mm, with an annual average temperature around 20℃ and an annual mean relative humidity of 66% (Leblanc & Malaisse, 1978). At the beginning of the past century, the vegetation was dominated by the miombo woodland, which, due to diverse anthropogenic activities, is progressively replaced by a wooded savannah and then to a shrubby and ultimately to grassy savannah (Useni, Malaisse, et al., 2017).

Figure 1.

Map of the study area (Lubumbashi city, Upper Katanga province in DR Congo), where sampled green spaces are represented by black dots stuck onto the 2014 SPOT image.

Results

Spatial Pattern of Green Spaces

Table 1 and Figure 2 show that the urban zone presented significantly lower total area, mean area, maximum area, median area, and minimum area of the green spaces than the peri-urban zone; at converse, however, the number of green spaces was higher compared to the peri-urban zone. This observation was confirmed by the index of the largest patch (peri-urban zone: 14.9% vs. urban zone 8.1%). This is an indication of predominance of smaller green spaces in the urban area (except for the zoological garden and the golf field in the Western part of the city), which might in turn suggest either a tendency to green spaces fragmentation or a scarcity of spaces limiting the possibility to establish green spaces of bigger size. In general, green spaces in the urban area present acreages virtually threefold lesser than peri-urban green spaces.

Table 1.

Landscape Metrics Calculated for the Green Spaces of Lubumbashi City in 2015.

Figure 2.

Spatial pattern of green spaces sampled along the urbanization gradient in Lubumbashi.

Ecological Functionalities of Green Spaces Along Urbanization Gradient

Data in Table 2 show that more attached and roadside green spaces were observed in the urban zone compared to the peri-urban zone (p < .05). It was apparent that wastelands and informal green spaces, green buffers, and fields are more frequent in the peri-urban zone (p < .05). However, no significant differences could be found for the proportion of parks or squares, gardens, nurseries, playgrounds, cemeteries, pastures, and wetlands between both zones (p > .05; Table 2), which might be suggestive that the aforementioned types of green spaces were unaffected by the degree of urbanization.

Table 2.

Green Space Types of Lubumbashi City in 2015 Based on Existing Typologies (Kong & Nakagoshi, 2005; La Rosa & Privitera, 2013; Mensah, 2014).

Results showed a higher proportion of green spaces colonized by invasive species in the peri-urban zone, which was paradoxically accompanied by a higher proportion of natural vegetation (p < .05; Table 3). In line with a number of reports in the literature (Bromilov, 2010; Chukwuka et al., 2007; Ipou et al., 2011; Lowe et al., 2000; Muoghalu & Chuba, 2005; Tiébré et al., 2012), Tithonia diversifolia, Imperata cylndrica, Leucaena leucocephala, Lantana camara, and Ricinus comunis were also identified as alien invasive species in the study area. Moreover, following Fisher’s exact test, significant differences regarding the equipment of the green spaces and the ownership status were noted between the two zones, with the urban zone presenting more equipped green spaces, most probably due to the lesser proportion of private green spaces in this zone (p < .05; Table 3). The degree of urbanization has no significant effect on the presence of different strata, the presence of bare soil, the presence of anthropogenic and seminatural vegetation, and on the accessibility of green spaces (p > .05).

Table 3.

Vegetation Characteristics and Management Parameters of Green Spaces Located in the Urban and Peri-Urban Zones of Lubumbashi City in 2015.

A total of 108 species, including 71 exotic species (65.7%) belonging to 34 families were recorded within the study area, with 83 species found in the urban zone and 87 in the peri-urban. Among identified families, Poaceae and Fabaceae constitute the main group, although the most occurring species remained annual plants. Detailed information in relation botanical family of different species, their life forms, and their position in the urban rural gradient are provided in the Appendix. The urban zone was characterized by 61 exotic species (73.5%) against 52 (59.8%) for the peri-urban zone. The ecological analysis showed that the flora of Lubumbashi was dominated by phanerophyta (46 spp., 42.6% of total), while a lesser contribution was found for the therophyta (31 sp., 28.7% of total). Other life forms were hardly present (Table 4). Nonetheless, no significant differences in life forms were noted between the two zones of the urban–rural gradient (chi-square, Table 4).

Table 4.

Life Forms of the Species in Urban and Peri-Urban Green Spaces of Lubumbashi City.

Discussion

The accelerated spatial dynamics of Lubumbashi city as other African cities has negative effects on green spaces; the city is indeed experiencing an environmental crisis linked particularly to its uncontrolled urbanization. Hence in this study, urban and peri-urban green spaces were characterized separately, so as to clearly distinguish the associated effects of the urbanization degree on them. In this context, a decision tree (André et al., 2014) was utilized in view of its straightforward and time-effective approach, in addition to being based on morphological characteristics (André et al., 2014). These zones highlight new management methods of green spaces but above all typological choices of specific green spaces. We used typologies applied elsewhere (Kong & Nakagoshi, 2005; La Rosa & Privitera, 2013; Mensah, 2014) regrouping green spaces with respect to land cover, land use, and ownership, which are similar to categories of green spaces previously identified in Lubumbashi (Leblanc & Malaisse, 1978), although types such as nurseries, pastures, cemeteries, wetlands (Figure 3), and green buffers constitute additional features. Since they have been previously identified as an ideal compromise to characterize spatial pattern (Bogaert & Mahamane, 2005), metrics were utilized in this study to assess the spatial impact of urban growth on green spaces in Lubumbashi. However, considering that the shape of urbanized patches is entirely determined by the anthropogenic context (Cornelis & Hermy, 2004; Forman & Godron, 1986), the shape index of green spaces was not calculated in this study.

Figure 3.

A wetland-like green space in the urban zone of Lubumbashi, with herbaceous species Imperata cylindrica and Cyperus spp in the foreground of the picture.

As the general trend in developing countries, unbuilt lands in urban and peri-urban zones are matter of less rigorous governance and management, mostly driven by various sociopolitical reasons (S. Shackleton, Chinyimba, Hebinck, Shackleton, & Kaoma, 2015). Results of our study demonstrated an increasing number of small green spaces with the degree of urbanization in Lubumbashi, which is exacerbated by the lack of a master plan for the management of the city, thus resulting in further destruction of preexisting green spaces, as they are systematically replaced by diverse types of built infrastructures. These findings are in line with observations in Abuja (Nigeria; Fanan et al., 2011), Greater Dhaka (Bangladesh; Byomkesh, Nakagoshi, & Dewan, 2012), and many European and U.S. cities (Fuller & Gaston, 2009; McDonald, Forman, & Kareiva, 2010) where a close relationship was evidenced between the pattern of urbanization and the intensity of green space removal and degradation. It is apparent from current results that the green spaces located in the urban zone of Lubumbashi tend to become more disconnected. And despite their lower number, it was shown that only a third of peri-urban green spaces are public, which can be explained not only by the sociopolitical context in the city characterized by land speculation (Groupe Huit, 2009) but also by the higher number of green spaces owned by private institutions like schools or churches (Nkuku & Rémon, 2006).

It is well known that the proportions of particular green space types are related to the degree of urbanization. Hence, like in Lubumbashi, the presence of roadside and attached green spaces in the urban zone is influenced by historical context and current political planning processes of the city (Nkuku & Rémon, 2006) and in older postcolonial municipalities, green spaces alongside public buildings, and streets lined with trees are frequently observed. As to wastelands and informal green spaces, their higher proportion in the peri-urban zone is explained by the expansion of discontinuous built-up (André et al., 2014) on ancient vegetated land. In Africa, the peri-urban zone with its higher proportion of fields is generally home to many economic activities sustaining the daily life of many households (André et al., 2014; Trefon & Kabuyaya, 2015). With this in mind, it is logical to assume that the rapid population growth that amplifies the competition for land could be a threat to those agricultural lands (Figure 4), and in general to all green space types, therefore causing the aforementioned environmental crisis.

Figure 4.

Brassica chinensis on a field-like green space in the peri-urban zone of Lubumbashi city.

In reaction to such an unfortunate perspective, local authorities and, in some cases, private institutions have devoted efforts to create new green spaces (Figure 5), frequently equipped, on the rare open spaces still existing in the city. For this reason, a close relationship between the presence of anthropogenic vegetation and the equipment of the green spaces was observed in the urban zone of Lubumbashi. Maréchal et al. (2018) confirmed this observation in Lubumbashi. The peri-urban zone showed an elevated proportion of green spaces colonized by invasive species. According to the intermediate disturbance hypothesis, the urban–rural interface shows a higher invasibility due to both urban edge effects and increased pressure of invasive species from the surrounding rural matrix (Catford et al., 2012; Doren et al., 2009; Radeloff et al., 2005). The peri-urban zone also showed a higher proportion of green spaces characterized by natural vegetation, normally typical of rural zones (Maréchal et al., 2018). Since peri-urban zones are assimilated to exogenous edge zones of urban patches (Bogaert et al., 2015), they present hybrid ecological conditions of both undisturbed (rural) and disturbed (urban) habitats (André et al., 2014; Niemelä et al., 2002; Radeloff et al., 2005; Rija et al., 2014).

Figure 5.

A garden-like green space, installed by the initiative of local authorities in the peri-urban zone of Lubumbashi city, dominated by populations of Acacia auriculiformis. This green space, adjacent to cassava field, is used for prayer and reading activities.

The process of urbanization has highly transformed the landscape of Lubumbashi and has created heterogeneous urban vegetated lands with new environmental conditions, leading to an installation of exotic species at an appreciable degree (Figure 6). Similar situations have also been observed in other African cities, though in varied proportions (Bernholt et al., 2009; Bigirimana et al., 2011; Rija et al., 2014). The urbanization of a city has considerable impacts on the composition of urban flora (Godefroid, 2001; McKinney, 2008; Ricotta, Godefroid, & Rocchini, 2010; Tait, Daniels, & Hill, 2005). For the case of Lubumbashi, as revealed in the current study, it is worth mentioning that the species richness observed is approximately 4 times lower compared to values previously advanced by Leblanc and Malaisse (1978, pp. 70–90). This situation could be in relation with the removal of the original vegetation cover and its replacement by built-up, since 68% of species found earlier were located on wastelands and informal green spaces (Leblanc & Malaisse, 1978), currently occupied by built-up. This situation is particularly alarming in the peri-urban zones as these are dynamic zones that are characterized by rapid change as a result of an extension of the city and associated infrastructures (Chirisa, 2010; S. Shackleton et al., 2015).

Figure 6.

A square in the urban zone of Lubumbashi city. In the foreground of the picture is the ligneous species Delonix regia, while Cynodon dactylon, an exotic species, constitutes the prevalent herbaceous species.

The life form of plants is an adaptive response to environment, in addition to providing an ecological classification that may be indicative of habitat conditions (Shaltout & El-Sheikh, 2002). Surprisingly, the higher presence of phanerophyta appears to be linked to the presence of roadside green spaces in the urban zone and remnant patches of miombo forest in the peri-urban zone. These two factors may explain the lower presence of phanerophyta in the city of Bujumbura (Bigirimana et al., 2011), although it is also surrounded by the miombo woodland and savannah vegetation. The dominance of phanerophyta observed in Lubumbashi, which also corroborates results previously obtained for the city of Kisangani in the Congo basin (Mutabana, 1982), appears to be critical in contributing to more resilient social–ecological systems, hence it should be taken into consideration in any urban planning initiative and in designing new housing developments (S. Shackleton et al., 2015). The presence of therophyta observed in almost all the studied zones might be due to disturbances in habitat conditions, favored by built-up expansion. Indeed, it is evident from the literature that the capacity to produce large quantities of seeds in therophyta, despite the reduced competitive capacity, would explain the maintenance of their populations in disturbed environments such as cultivated fields or fallows in urbanized areas (Wang et al., 2014).

Current findings demonstrated that both zones of the urban–rural gradient present a similar level of accessibility to the green spaces, which might be an indication of a free of charge access to the majority of green spaces. Likewise, it is thought that Lubumbashi green spaces are visited irrespective of the type of their vegetation or their management characteristics (Qureshi, Breuste, & Jim, 2013), which appears to be the opposite to the situation in cities like Santa Cruz (Bolivia), where a preference is noted for larger green spaces, with more amenities (Wendel, Zarger, & Mihelcic, 2012).

Implications for Conservation

A range of issues have been evidenced regarding land management in the city of Lubumbashi. The most flagrant cases concern the existence of less equipped and poorly structured quarters in the peri-urban zone, an uncontrolled extension of the city through creation of housing development endangering the existence of green spaces and associated infrastructures, and a lack of coordinated actions concerning the urban management in the city (Groupe Huit, 2009). As an evident demographic explosion is expected in the coming decades in the city, and considering the huge density of population observed in ancient colonial quarters, it appears logical to assume that such demographic explosion will most likely occur in the peri-urban zone where lands are being allotted at a worrying pace. Currently, in this peri-urban zone, the majority of lands are owned by privates, who are managing those lands and corresponding vegetal covers. In this context, an exacerbation of conflict between green space preservation and housing is expected in the future in the peri-urban zone, particularly fields. Conservation of fields could be critical to food security because their area diminishes every year to the benefits of infrastructure which has the priority over agricultural activities in the city, probably due to the ignorance of municipal offices.

Conversely, gardens around private businesses, factories, schools, and public office buildings could be important for species conservation and a range of ecosystem services in urban zone. Unfortunately, there is a strong trend to densification of the existing properties through the neighborhoods with the increase in population and the decrease in new allocations of parcels for building new houses and offices in urban zone. But in addition, when these attributions of new properties are done, they often public green spaces that previously supported a diversity of cultivated plants. Indeed, the species diversity of urban green spaces could also be forced to decrease.

Our study demonstrated that the urbanization in Lubumbashi city has brought about a tendency toward smaller and private green spaces, which could, in the long term, spoil green spaces-related benefits such as recreational opportunities, psychological well-being of citizens (Maas, Verheij, Groenewegen, de Vries, & Spreeuwenberg, 2006), and economic and ecological advantages (Bolund & Hunhammar, 1999; C. M. Shackleton & Blair, 2013). Such diverse benefits contribute to more resilient social–ecological systems and need to be a part of the future urban planning initiatives as well as in the design of new housing developments in Lubumbashi city. Indeed, it is now generally accepted that sustainability principles cannot be disregarded in urban planning in the perspective to create the conditions for comfortable and dignified living (C. M. Shackleton, Hebinck, Kaoma, Chishaleshale, & Chinyimba, 2014). Furthermore, an operational program of urbanism should be implemented so as to establish allotments capable of guarantying a minimum of green space to the annually occurring surplus of population in Lubumbashi city. Therefore, it is necessary for local populations to be aware of the importance of green spaces and their role in counteracting adverse effects associated with urbanization.

This study provides a better knowledge of the vegetation inside Lubumbashi city and its spatial organization, which is a first step toward monitoring the evolution of the vegetation in the current context of anarchic expansion of the city. Globally, obtained results highlighted the importance to preserve green spaces in the peri-urban zone of Lubumbashi city, where relatively high proportions of indigenous species and natural vegetation were recorded.

This study highlighted the compositional differences in green spaces between the planned (urban zone) and unplanned (peri-urban zone) neighborhoods in Lubumbashi city, which is probably associated with historical and current political planning processes in the city. Bearing this in mind, for future urbanization plan, a call is particularly made to local authorities and planners for the implementation of effective measures targeting to increase plant diversity within the city, for example, by restoring public urban green spaces using indigenous species identified in this study, with the aim to counteract exotic founded in high proportion in the studied area.

The urbanization process of Lubumbashi city resulted in the progressive elimination of natural vegetation like miombo woodland with its indigenous flora and its replacement by green space types dominated by exotic species. This is the well-known process of biological homogenization (McKinney, 2008). However, these recently created green spaces, and with high level of anthropization, cannot fulfill the ecological functions that were provided by the natural ones due to the difference in structure and species composition (Li et al., 2013). It is important to conserve and restore natural or seminatural green space in Lubumbashi city using indigenous species founded in our study.

Integrating indigenous species in roadside green spaces in the peri-urban zone could also increase the spatial connectivity of the green infrastructure, which is worth consideration in any plan of new housing developments in the peri-urban zone. The current study also revealed a notable occurrence of invasive species in the study area, especially in the peri-urban zone. But in the current context where the city of Lubumbashi does not have a master plan and monitoring program for invasive species, the invasion debt by new species and the extinction debt for all species are both high in the medium and long term. In this context, a regular inventory of the local flora is warranted in order to adopt efficient approaches to counteract the propagation of the large spectrum of exotic invasive species observed in the studied area. Likewise, flora monitoring could help to identify the species of socioeconomical interests. The current study listed species like the endemic Bulbostylis pseudoperennis (Goeth.) that exhibit potential in the restoration of degraded metal-rich habitats and the revegetation of metal bare soils in the katangan copperbelt area (Ilunga wa Ilunga et al., 2015). The same goes for Leucaena leucocephala, Imperata cylindrica, Panicum maximum, Tithonia diversifolia, and Hyparrhenia spp., which are essential for agropastoral activities; Aloe vera known for medicinal properties; while Abelmoschus esculentus, Ipomoea batatas, and Citrus spp. are of nutritional interest. In another respect, despite their potential ecological threats, biomass of Tithonia diversifolia, Mimosa spp., and Eicchornia crassipes (also listed in this study) could be oriented toward replenishment of soil fertility for agricultural production. It is hence evident with respect to how they are managed, be it invasive, exotic species can still prove useful for life in a given ecosystem.