Currently, urban centers are responsible for consuming two-thirds of the energy in the world and for up to 70% of global greenhouse gases (GHG) emissions ^[1][31]. In this context, as urban areas lost their ability to self-supply ^[2][8], importing food from local producers became crucial. A recent publication pointed out that, in 2018, total world emissions from food transport were 511 Mt CO₂, with a global increase of 80% of these emissions since 1990 ^[3][32]. This transport implies the loss of both quantity and quality of food and is considered the main contributor to food waste by some authors ^[4][33]. Global data on food loss that occurs between harvesting and reaching the final consumer, including transport, is estimated at 14% of the total loss ^[5][34]. In this context, urban agriculture (UA) has been enthusiastically defended as actions that can help address these challenges. ^[6][35].

Covering most empty urban areas with edible green could be a new ecological achievement as such interventions could reduce urban pollution, mitigate, and adapt to climate change by increasing carbon sequestration, water infiltration and retention, and controlling the urban heat island effect ^[7][8][36,37]. Green roofs and vertical walls can also reduce heating and cooling while improving air quality and contributing to increased biodiversity and ES ^[8][37].

UA is compatible with the proposals of the circular economy movement ^[9][38], where the optimization of processes based on durable, recyclable, and renewable resources is prioritized. Thus, models that integrate food production with organic waste management have a high potential to encourage the reuse of nutrients and the development of by-products. In the United States, it is estimated that around 28% of household organic waste could be reused for food production ^[10][39]. Similarly, in Havana, organoponics was implemented using organic waste from different sources, including domestic waste ^[11][40]. In addition, given the water resource management crisis in many urban areas, UA initiatives must be jointly managed, such as the ones that capture, retain and reuse rainwater ^[12][41]. In this regard, in Munich, it is estimated that around 26% of its current rainwater supply could be harvested and then used in food production activities ^[13][42]. Considering water harvesting is especially important since droughts in the summer seasons are increasingly frequent and intense in many regions of the planet. Thus, developing models of green structures in UA sites suitable for rainwater harvesting in the rainy months can help maintain the activity during the dry months ^[14][17].

The UA benefit most reported by the studies is the food miles. Specht ^[15][43] points out that short-distance production-consumption can promote the reuse of organic waste in food production, which avoids higher GHG emissions from the biological waste process, in addition to not generating high emissions due to the low mileage in its distribution perimeter. Thus, it is argued that this can be incorporated as a climate change mitigation system. However, there is still a gap in the assessment of the impact of short-distance production-consumption concerning GHG emissions and other sustainability indicators ^[15][43].

Still on emissions, in UA they are affected mainly by the structure, land use, use of fertilizers and the distribution of their production ^[16][44]. It is argued that the most significant environmental impact in the food sector is generated by how food is produced, and not necessarily where ^[16][17][44,45], since production and harvesting are responsible for around 83% of GHG emissions from the food sector ^[18][46]. In this regard, UA tends to mitigate the environmental impact of the food chain, since in the vast majority of cases it is implemented respecting agroecological and soil regeneration practices and does not have logistic-related emissions ^[19][5]. In this sense, Hu et al. ^[16][44] in an assessment of GHG emissions from UA and conventional agriculture in Beijing noted that transport was the second most impacting factor on emissions from both modalities; however, in this regard, the UA still had lower emissions compared to conventional agriculture. Therefore, even if the mode of production may not reach an ideal in terms of emissions, UA could still result in a profitable reduction in the emissions of the total sector by dealing with smaller food distribution distances. In addition, UA also eliminates the use of intermediaries in the chain, which results in energy savings ^[20][47].

In this sense, Hu et al. ^[16][44] propose some possibilities to contain carbon emissions in the activity, such as economic measures granted to urban farmers to encourage the use of solar and wind energy and renewable fuels. However, it is noteworthy to mention that such proposals may not yet be suitable to countries of the Global South, since in the countries of the North the integration of renewable electricity is a topic best addressed in transitions ^[21][48]. Although the Global South presents a high potential for renewable energy resources, most of these countries still have limitations in terms of adequate infrastructure for such technologies. It is noteworthy that even with the use of non-renewable energy, in the Global South it is expected that by 2050 around 750 million people will still have no access to electricity ^[22][49].

Giraud ^[17][45] argues that UA can contribute to the achievement of some Sustainable Development Goals (SDG), such as increasing universal access to renewable energy (SDG#7—”Affordable and Clean Energy”), addressing air pollution in urban areas (SDG#11—Sustainable Cities and Communities) ^[23][50], contribute to Climate Action (SDG #13) and support biodiversity (SDG #15—”Life on Land”) ^[17][45]. Furthermore, Deksissa et al. ^[24][51] argue that UA integration strategies and green rainwater harvesting infrastructure could help urban areas at local and global levels in exploring adaptation mechanisms to extreme climate change events. Likewise, Marçal et al. ^[25][52] argue that in the city of Goiânia, UA plays the role of making the urban environment more resilient to the climate crisis by helping mitigate drought, floods, improve soil conservation, capturing carbon from the atmosphere and lowering local air temperature.

However, a limitation presented for the integration of Nature-based Solutions (NBS) policies is that city master plans do not usually consider food resilience in their guidelines ^[26][27][53,54]. As such, the tendency to understand grey and green spaces as separate from each other in urban plans makes the benefits of the synergistic potential that could be achieved by such measures less considered ^[28][29][55,56].

Compared to conventional agriculture, UA is understood to be more resilient due to the shorter production-consumption chain, as well as promoting systems with more diversified production ^[30][57]. In this regard, while some studies propose the assessment of sustainability in UA based on life-cycle assessment methodologies ^[31][58], other authors argue that these methodologies still underestimate the sustainable and resilient potential of the activity for urban areas ^[14][17].

Finally, it is vital to highlight that UA does not necessarily result in the preservation and regeneration of natural resources ^[32][59]. For these to happen, it is necessary to develop a transition pathway, which includes technological innovation ^[32][59], leading to new models that can maximize ES co-benefits and minimize trade-offs.

The vulnerability of ecosystems and people regarding the climate crisis significantly differ according to the region of the planet, and this is essentially due to factors such as socio-economic development, inequality, marginalization, unsustainable use of land and oceans, as well as governance ^[33][60]. The adverse impacts of climate change, development challenges and inequality mutually exacerbate each other. Mortality rates due to floods, droughts, and storms, vary drastically across regions with high and low vulnerability and thus reveal the different starting points in their movement towards climate-resilient development ^[33][60].

There is a substantial gap in understanding the political economy of urbanization and the roles of governance ^[34][61]. In South and Central America, for example, there are numerous initiatives to improve planning, but they tend to be focused on risk reduction and not on climate adaptation ^[35][18]. Thus, there is a gap in the development of planning that promotes adaptation. The solutions lie in improving governance to address global problems, encompassing participation in decision-making, ensuring greater fairness in interventions, as well as creating and improving links with different levels of government and other policies that ensure co-benefits ^[33][60].

Although there is evidence regarding the UA contribution to alleviating social and urban challenges, which include climate change, food security, biodiversity and ES, agricultural intensification, resource efficiency, urban renewal and regeneration, land management, public health, social cohesion, and economic growth ^[36][62], there is still a gap to link UA as a transformative adaptation measure ^[37][63]. It is observed that, in the vast majority of cases, the activity has been driven by community efforts ^[38][64] with shallow and unequal public support ^[26][53]. Furthermore, UA is often hindered by inadequate planning policies, which discern limited access to land, financial barriers, soil pollution and contamination, development pressure, and gentrification ^[38][64]. In addition, finding common ground in different administrative departments in different policy domains involved in UA has not proven easy, at best.

Although Clinton et al. ^[39][65] have demonstrated from global data that UA has the potential to produce tons of food, sequester tons of nitrogen, save billions of kilowatts of energy, and avoid the loss of billions of cubic meters of rainwater, this type of data is still understood in an academic and impractical way. As a result, the potential multifunctionality offered by UA is not yet engaged in land use planning in many places, which is one of the main obstacles to its adoption, especially in the Global South, where NBS are still timidly implemented ^[40][66]. Therefore, understanding UA’s multifunctionality in different geographic contexts is fundamental for its incorporation into urban planning ^[41][67].

Some authors postulate that UA presents different approaches between Global North and Global South. First, this division of the world order accepts that the first is composed of countries from North America, Europe, some Asian countries and others located in the southern hemisphere, such as Australia and New Zealand, while the Global South concentrates Latin countries, the African continent, Middle East and parts of Asia. What defines these borders is based on economic, social and political repercussions, representing the division of developed and developing nations ^[42][68]. Despite this, the model oversimplifies the characteristics of nations, since it ignores internal variations and commonalities between the Global North and the Global South.

In the context of UA, it is observed that while in the Global North countries these initiatives tend to focus on social dimensions and, in some cases, on environmental benefits, in the Global South, UA is primarily focused on food subsistence and income generation ^[43][44][69,70].

Opitz et al. ^[44][70] point out that in the Global South, production generally takes place in polluted environments ^[45][71], where health risks prevail due to poor management and environmental pollution ^[46][47][72,73]. In many cases, UA lacks legal status ^[47][73] and activities such as leisure or recreation are rarely observed ^[48][49][74,75]. In the Global North, these production spaces tend to have a temporary perspective, since they are threatened by more economically profitable land uses and largely from marketing-oriented initiatives. There is also evidence that populations on a low income, arguably the most nutritionally deprived population, are often excluded from UA ^[42][44][68,70]. Consequently, one of the main challenges reported for the Global North is to make the benefits of UA reach the populations in need, especially in the context of the existence of food deserts in many of these countries ^[42][44][68,70].

The literature understands green infrastructure initiatives related to UA as reflections of specific needs or problems. In Africa, for example, where food security is an important issue, UA is understood as an answer, while in Asia research on green spaces reflects issues of spatial planning and land distribution in high-density cities. In Latin America and the Caribbean, where there are serious issues of inequality, studies highlight vulnerability and social concerns ^[50][76].

Another aspect observed is the difference in governmental action in periods of crisis. While in developed countries it is common to observe the encouragement of activities such as UA ^[26][53], in developing countries crises can affect even more the performance of governance, as observed in the city of São Paulo during the COVID-19 pandemic ^[50][76].

Finally, Dona ^[41][67] argues that studies about the potential of UA should be focused on geographical aspects, recognizing the difficulty of establishing a universal pattern of benefits and potentials of the activity, being especially important to promote an evaluation method aimed at developing countries. Further research of UA models encompassing social and environmental aspects simultaneously is needed to understand the potential of this activity under different contexts, especially regarding the diverging Global North and Global South dynamics.

Urban food production needs to consider the different dimensions of sustainability; therefore, it must understand the management of environmental challenges, solve or alleviate social problems and promote economic returns ^[15][43].

Urban resilience and sustainability depend on the connection between different ES. The provision of ES in cities must seek equality, across socio-economic and demographic regions ^[51][77]. The integration of innovations in traditional UA activities, preserving their environmental and social role, can overcome limitations and still offer the chance to achieve the circularity of different resources present in the urban environment ^[52][53][54][78,79,80].

Accordingly, care must be taken to evaluate the suitability for UA, considering for instance the presence of air pollutants, as well as soil contamination, which expose the health of the local population to risks ^[15][43]. The implementation of UA should be directed to places with less exposure to atmospheric and soil contaminants, as well as being based on agroecological production practices to avoid in turn contamination derived from the activity itself.

An aspect not addressed in the literature is to link the implementation of the green structure of food production to the principles of preservation and regeneration of native species. Cities that consider their native vegetation incorporated into municipal land use plans can promote the activity by encouraging and delimiting cultivated species according to the native profile, for instance restricting the use of non-native species in the territory. Cities such as São Paulo encourage this type of action based on municipal plans such as the Atlantic Forest Municipal Plan ^[55][81]. Yet, there is no association of this policy with food production in the municipality. Similarly, in the city of Accra, the Land Use and Spatial Planning Act 2016 promotes the protection of different green infrastructures, such as forest reserves and green belts, but does not include UA in its guidelines ^[53][56][79,82]. Thus, the inclusion of UA in policies such as those is an opportunity in potential that needs to be further considered.

Besides, a requalification and assignment of multiple uses of urban green space would address the recommendations of international policies ^[57][83] as a fulfillment of Environmental and Urban Agendas ^[58][11], such as The New Urban Agenda with the aim to promote a new global model for sustainable urban planning ^[59][84]. In this respect, understanding the availability of green areas that can be used in synergy with UA becomes fundamental to avoid the misuse of these spaces and the loss of biodiversity.

The association of green infrastructure and UA is still lacking, even in countries that already adhere to NBS policies. In the case of flood risk management, Deksissa et al. ^[24][51] argue that even though their management measures based on green infrastructure can at the same time produce food, and well-planned UA initiatives can contribute to flooding risk management, what happens is that these are policies that are still understood separately. Therefore, the analysis of flood areas may present an opportunity to aggregate UA benefits.

Regarding the economic dimension, UA must be incorporated into commercial chains in the city, preferably in its neighborhood, to contemplate aspects of the food miles. In this regard, Sonnino ^[60][85], in an assessment of UA as a local food production system in the Rome region, argues that there is a need for a more effective connection between urban and peri-urban food-producing areas and local commerce, which includes consumers of local fairs. In this way, tracking these potential partnerships can help target locations for food production. Another potential is the connection between producers, as one can direct its production to items that the other producers do not have and vice versa, which not only promotes economic cooperation but also diversifies the availability of products, as well as act as a measure to reduce food waste ^[60][85].

The social sphere is perhaps the most fragile to be manipulated since numerous cases of UA implementation are affected by gentrification and mask greenwashing ^[61][62][86,87]. In assessing urban land use for climate adaptation in different cities in the Global South and Global North, Anguelowisk et al. ^[63][88] argue that rational and technocratic planning approaches, while defending an ideal “public good”, end up not emphasizing the asymmetrical power dynamics and the conflict over resources present in urban areas ^[64][89, ^[65]90, ^[66]91]. The study points out that in the Global South it is common to see resettlement sites distant from work opportunities, disconnected from social networks and being affected by disaster risks, which reduce the ability of communities to adapt and their long-term security. The implementation of UA must be thought out of the dispute of spaces for rich and poor, being directed in a way equivalent to the needs of each public, as well as preserving the primordial aspect of sustainability, which is to promote environmental and socio-economic development together.

Faced with such challenges, some authors propose strategies for evaluating UA initiatives. Gulya and Edmondson ^[67][19] argue that the implementation of UA should be determined according to its scale, the level of integration into the urban fabric, its social inclusion character, the quality of efficiency of food production, as well as social and environmental security promoted. In turn, Tapia et al. ^[68][92] developed a model anchored in scientific and sustainability principles, particularly related to SDG and ES narratives, to assess the benefits and negative consequences of implementing different types of UA. Similarly, Zanzi et al. ^[69][93] evaluated the sustainability in the implementation of UA by startups in Milan through a holistic framework and observed the possibility of UA to promote four facets of sustainability: economic resilience, social well-being, governance, and environmental integrity. Although relevant, both models are developed for cities belonging to the Global North, whose social and environmental characteristics are very different from those present in the Global South. Besides, the second model is based on the presence of startups, which may present limitations for other realities. Thus, there is a lack of initiatives encompassing the social and environmental particularities observed in southern countries.