**5. Conclusions**

The above case studies involving Singapore and Kigali offer two examples of cities in which urban agriculture differs markedly from the point of view of methods of production used and the role played in the spatial and functional structure. This reflects both the natural, socioeconomic and political context and the institutional and legal structures that are in place. A matter of significance for the latter is the scale of the presence of agriculture in the cities selected on which the nature of managemen<sup>t</sup> is dependent. In the case of Singapore, this is currently, in fact, a marginal activity that does not represent a policy priority for the authorities, even in the context of the inhabitants' almost total dependence on imports for their food security. In turn, in Kigali, agriculture is an integral element of the city ecosystem, and one whose presence and role is sufficient to ensure consideration being given to it as the development policy in the Rwandan capital is pursued.

For many decades, urban agriculture was regarded by planners and governmen<sup>t</sup> representatives in many countries as a form of resistance to urban development priorities and was, therefore, prohibited from the urban space [31]. Despite its role in poverty alleviation and food security improvement, there appeared to be little political will to support urban agriculture [89]. Nevertheless, progress is being made, and the attitude toward the activity is gradually changing. In many cities of the world, the legislative restrictions on urban agriculture have been removed and systems designed to support urban agriculture are even put in place [31]. In the case of the two cities analyzed in this paper, urban agriculture is indeed included in the land-use policy. Thus, while it takes into account the planning documents of both Singapore and Kigali (i.e., the Master Plan 2019 in the case of Singapore and the Kigali Master Plan 2013), the vision for the future situation looks quite dissimilar in the two metropolises. Moreover, in the two cases presented, the approaches toward urban agriculture to a different extent comply with the principles of smart, resilient and soft urban development. Certain solutions promoted by the cities' decisionmakers might be considered as being in line with the concepts, while others are quite the opposite. In the case of Singapore, the emphasis is on modern agriculture based around high-technology and not occupying the valuable and limited resource that space represents for the city. Its place, therefore, falls within a technocratic vision of the smart city, with inhabitants being encouraged to use modern technology in urban agriculture, while support is not forthcoming for initiatives based around the 'more conventional' community gardens or farms in general cultivating land in a traditional way. Bottom-up initiatives and participation of residents are limited, which is inconsistent with the assumptions of at least two pillars of the smart city concept, which are smart people and smart governance. Neglecting the grassroots and community activities of inhabitants as well as the social function of urban gardens is also inconsistent with the assumptions of the soft city, which is based on simple solutions developed by the citizens themselves. Innovative methods of agricultural production such as hydroponic modules often placed on rooftops or indoors, have several socio-economic benefits, as their presence increases the multifunctionality of the urban space and provides for the shortening of value chains, which falls within the smart, resilient and soft city concepts. At the same time, however, such high-technology production systems have fewer benefits in environmental and ecological terms, as they do not help preserve natural resources or increase the share of green areas. Moreover, the reduction in the agricultural land outlined in the current Singapore Master Plan will lead to a further decrease in the share of green areas within the city limits.

In Kigali, the authorities seek optimal means of using land whose geophysical features preclude it from being built on, ye<sup>t</sup> urban agriculture is to be preserved almost exclusively in peri-urban areas. At the same time, authorities offer inhabitants a certain freedom to make use of unused land in cultivation, in line with their needs. This conferment of a right of initiative among citizens is in line with the smart city concept and its smart governance aspect, as well as the soft city concept, as it gives the inhabitants the possibility to create and shape their own urban space, according to their needs. Moreover, the creation of kitchen gardens and the transformation of small-scale wastelands into cultivated plots

leads to the improvement of the residents' food security and enhances their ability to cope with stresses, such as a drastic increase in food prices, which is in line with the urban resilience concept. In Singapore, the authorities are more restrictive in these matters, with a large proportion of the farms studied, especially those performing social or community functions, not having their land leases prolonged.

While the two cities do clearly differ significantly where urban agriculture is concerned, it is possible to note certain similarities as well. In the first place, it is worth recalling the system of land ownership, wherein the authorities in both states are the main decisionmakers when it comes to land management. While in Rwanda (unlike Singapore), they are not actually the main owners, Article 13 of the Organic Law confers very broad powers with regard to control over economic activity in general within the city. In neither case is it possible to regard urban agriculture as a priority activity. Though the issue appears in planning documents, with reference made to the significance from the point of view of inhabitants' food security and wellbeing, the activity is not important enough for the space of economic significance to be allocated. So, while the scale of urban agriculture in Kigali is incomparably greater than in Singapore, even here, the activity only takes in those areas incapable of being used in construction. Singapore simply has a smaller share of this kind of area unsuited to being built up. Furthermore, the fact that Rwanda is developing dynamically at present suggests that upcoming decades will see demands for land resources increase, just as it did in Singapore, which is, after all, a model for the Rwandan authorities. This should ensure urban agriculture's expulsion from whichever areas are capable of supporting more profitable types of economic activity. It remains in question whether the Kigali's authorities will further follow Singapore's footsteps and start supporting high-tech initiatives, enabling the efficient use of scarce land resources. As innovative methods of agricultural production are indeed gaining in importance in fast developing cities of the Global South, they may also prove effective and beneficial in the capital of Rwanda. Nevertheless, it is important to emphasize that, in contrast to Singapore, whose experience is quite consolidated, in Kigali, the policy agenda as well as the approach to managing urban agriculture might still evolve in a different direction. As Kigali's urban tissue is currently under dynamic transformation, solutions integrating urban agriculture into urban space planning can still be introduced. By combining innovative solutions with the grassroots initiatives of the inhabitants, Kigali could create sustainable urban food systems, while achieving a high level of urban resilience.

**Author Contributions:** Conceptualization, A.G. and K.G.; methodology, A.G.; validation, A.G., K.G.; formal analysis, A.G.; investigation, A.G. and K.G.; resources, A.G.; writing—original draft preparation, A.G.; writing—review and editing, A.G. and K.G.; visualization, A.G. and K.G; project administration, A.G.; funding acquisition, A.G. Both authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the Faculty of Geography and Regional Studies of the University of Warsaw (SOP-87/19; SOP-130/19; SOP-38/2020; SWIB 36/2021) and Rada Konsultacyjna ds. SRN of the University of Warsaw (20/II/2019).

**Conflicts of Interest:** The authors declare no conflict of interest.
