With the majority of the human population concentrated in urban areas [10], cities have a vital stake in determining the sustainability of our Planet. The provided literature review is organized into topics with a focus on leading state-of-the-art research work on urban energy, water, and environment systems. The topics include energy supply, energy usage, transport, circular economy, water and waste issues, air quality, and environment as well as planning and governance at the urban level. The coverage emphasizes those for cities that are signatories or members of major city networks whenever possible, namely Covenant of Mayors (CoM) [11] and the Cities Climate Leadership Group (C40) [12].

Since June 2016, the CoM initiative is enlarged as the Global Covenant of Mayors for Climate and Energy [13]. In addition, new targets for signatory cities are adopted towards achieving at least a 40% reduction of CO₂ emissions from baseline years by the year 2030 [14]. The renewed targets underline the importance of persistent efforts towards reaching net-zero emissions by mid-century to have a chance of limiting global warming to at most 1.5 °C by 2100 [15]. There are over 7,650 signatories with close to 6,000 actions plans [16]. Analyses based on monitoring reports indicate that emissions from building and transport sectors will reduce by 49% and 23% over baseline years by 2020, respectively [17]. Local energy generation would be able to account for about 20% of the emission reductions [17].

The C40 initiative started as a climate mitigation network of some of the largest 40 cities that has since increased to about 100 cities, including observers [12]. Similarly, the C40 initiative has put forth trajectories for its cities towards reaching net-zero emissions by the year 2050 considering the total carbon budget and different city typologies [5]. While all cities are to reach net-zero emissions by the year 2050, the year in which a peaking of emissions is reached can vary by about a 10-year timeframe [5]. All cities, however, are called to contribute to “bending the curve” of global CO₂ emissions by the year 2020 [5]. In both contexts, research work in support of more sustainable urban systems is a top priority.

Given the above literature review on urban systems, there remains a need to support a thorough benchmarking process at the local level towards comparing the sustainable development of energy, water and environment systems in cities. In addition to studies that focus on individual cities or a group of cities, benchmarking studies that involve multiple cities across multiple aspects can provide scientific contributions in quantifying the level of proximity or distance in the overall performance of a city relative to others.

The present paper builds upon previous stages of the research work [53]-[56] in a way that is particularly important for supporting a “science for cities” through benchmarking based on a composite indicator. In addition to a new city sample, the results are used in a scenario analysis that is undertaken for a particular city for the first time. The manuscript is organized such that the method of the research work is first put forth with an emphasis on the selection of the present city sample. The subsequent two sections then underline key aspects of the process of data collection and the results for the present city sample. A scenario that is developed for the home venue of the Rio Earth Summit, namely Rio de Janeiro, is also put forth and discussed in the results. An outlook for the use of the results in support of more sustainable urban systems is presented in the concluding remarks.

The sample of the present research work is a new world sample that also represents a greater share of cities from outside the European signatories of the CoM initiative, including cities in the C40 initiative. Table 2 summarizes the 26 cities in the new sample along with the main reference(s) [59]-[116] for the original CoM Sustainable Energy and/or Climate Action Plan, namely SEAP or SECAP, updated monitoring report, and present status in the context of the C40 initiative. Existing research work that analysed aspects of energy, water, and environment related issues for the cities in the sample is further marked based on references [18]-[52] as a means of linking the cities in the sample with the literature review. Such a new sample brings the total number of cities in the first five samples to 102 cities.

The selection of the cities in the present sample involved a multi-criteria approach to improve the geographical diversity of the sample. First, the availability of energy and CO₂ emissions related data is a prerequisite in all samples. In addition, the total number of cities that were included from a given country in previous samples was compared to the total number of CoM signatories from all countries to identify areas of priority in the sample. At the same time, the SDEWES Index is systematically shared through the scientific platform of the SDEWES Conference series [58]. A selection of cities that can represent the cities of participating authors can increase the potential impact of the research results given that researchers can be seen as “change agents” [117], [118] in their cities. For this reason, the number of SDEWES authors from all countries that participated in the 11^th Conference on SDEWES [119] was evaluated as an additional criterion. The venue of this conference as Lisbon, Portugal was analysed in a previous sample [55] and was excluded for inclusion.

According to the method of this research work, the SDEWES Index is applied for the first time to the cities that take place in Table 2 to determine the benchmark leader of the present sample. In this process, data is collected for each indicator for each city in the sample and normalized based on the Min-Max method [120]. The direction of the indicators as summarized in [56] determines the descending or ascending order of the normalized values. Since the Min-Max method is relative to a given sample, the performances of these cities are benchmarked to all other cities to which the SDEWES Index has been applied to date for consistency based on the use of common minimum and maximum values [121]. Any outliers are identified based on higher order moments, including skewness greater than 2 and kurtosis greater than 3.5 [122]. Any outliers are addressed according to winsorization that can be used to substitute such values with the subsequent value in the data set given that these cases remain below threshold shares [122]. The steps of data collection and processing enable the next steps for normalization and aggregation to obtain index values for each city.

Eq. (1) provides the means of aggregating the normalized values of the five indicators in each dimension per any city in the sample C_j from j = 1 to j = 26. Here, I_x.y(C_j) is the normalized value of a given indicator in a given dimension numbered x with indicator number y for a given city C_j. The normalization is conducted with the range 0 to 10 so that the maximum possible score in a given dimension and the overall index score is 50. Here, α_x is weights with a summation of unity across all dimensions in consistency with previous practices [53]-[56]. The weights are further subjugated to 10,000 Monte Carlo simulations to consider any possible changes based on the mean simulated values using random weights scaled to unity. The results are compared according to quartiles of city performance. Figure 1 summarizes the dimensions of the SDEWES Index based on seven dimensions D₁ to D₇.

SDEWES C j = ∑ x=1 7 ∑ y=1 5 α x I x.y C j Where ∑ x=1 7 α x = 1 (1)

Figure 1.

Summary of the dimensions of the SDEWES Index D1 to D7 [57]

One of the cities that satisfied the criteria for inclusion in the present sample was later announced as the venue of the 1^st Latin American SDEWES Conference [123], namely Rio de Janeiro. In this respect, Rio de Janeiro is further subjugated to scenarios that consider the targets in its strategic outlook for the 500^th anniversary of the city based on the plan Visão Rio 500 (Vision Rio 500) [124]. In this vision, the city aims to become a “global benchmark” for sustainable development [124]. Moreover, Rio de Janeiro holds importance as the first city in a developing country to adopt a carbon neutrality target [125] and has joined other cities during the 23^rd Session of the Conference of Parties (COP23) to the UNFCCC to undertake a carbon neutrality target by 2050 [126]. Hence, the application of a scenario to the existing performance of Rio de Janeiro becomes a related objective of the research work. Local aims are also compared with opportunities within the scope of the SDEWES Index.

The scenario is developed to be a “normative scenario” that is defined as a scenario in which a specific future that may be achievable through certain measures or targets is prescribed [127]. In turn, normative scenarios can be used to encourage the realization of those measures. The scenario is developed for a timeframe from the present (t₀) up to the year 2050. Target years in Visão Rio 500 are used to determine annual rates of progress in certain indicators or the years when the best performance of the sample may be reached. When such targets are not explicit, those that are adopted in a scenario for an “Average City” in a subsequent integrated sample of 120 cities are applied [121]. Here, the aim was to reduce the gap between a city performing at an average level and the best performing city.

Eqs. (2-3) govern the annual increases in performance that are taken in the scenario for Rio de Janeiro, which is city C₂₁ based on the alphabetical order of cities as presented in Table 2. Here, i_x.y(C₂₁) _t0 is the data value of a given indicator based on existing levels of performance. In contrast, i_x.y(C₂₁)' _t(z) represents the data values in the normative scenario in a future time t(z). Both values are prior to the normalized values in eq. (1). The aspect of implementing eqs. 2 or 3 integrates a new time dimension into the index to enable dynamic evaluations for a scenario. Rio de Janeiro (C₂₁) thus becomes the first city within the phases of the SDEWES Index to which a scenario is applied based on local targets:

i x.y C 21 t z ′ = i x.y C 21 t 0 +z i x.y C T ts − i x.y C 21 t0 / t s − t 0 (2)

i x.y C 21 t z ′ = i x.y C 21 t 0 −z i x.y C 21 t0 − i x.y C T ts / t s − t 0 (3)

In eq. (2), average annual increases that would be needed to reach the targeted (or best) value by the target year are added to existing levels of performance for indicators in which higher values are more desirable. Eq. (3) represents other instances for indicators in which lower values are desirable so that annual reductions need to take place towards bringing existing levels of performance to the targeted (or best) value. In both eqs. (2) and (3), i_x.y(C_T) _ts represents the target value in the future that should be reached by the target year (t_s). The number of calendar years that has elapsed between t₀ and any future time t_(z) is accounted by a multiplier z. The context of eqs. (2) and (3) thus represents an application of a normative scenario based on the attainment of specific target values by target years. Based on quantitative projections for such a scenario, the SDEWES Index in eq. (1) for Rio de Janeiro is re-calculated using the normalized values of all data inputs, I_x.y (C₂₁)' _t(z).

Based on data for the 26 cities in Table 3 for D₁, the energy usage of residential, tertiary and municipal buildings in an average city is 21.6 TWh. Such a value exceeds those of the transport sector for private and public vehicles as well as the municipal vehicle fleet at 11.2 TWh. On an annual basis, the average city requires 16.89 MWh of energy per capita to provide energy services in buildings, transport, any industry (non-ETS), and public lighting. Recent monitoring reports of the cities in the sample further support a trend of progress towards reductions in urban energy usage within CoM signatories, particularly the building sector [128]. Since the cities in the sample are located in various geographical settings with differing climates, the impact of the climatic context on urban energy demand is represented by the total degree days weighted by an average Coefficient of Performance (COP) for the heating and cooling seasons. For example, Bangalore has an absence of Heating Degree Days (HDD) while Helsinki has limited Cooling Degree Days (CDD). In an average city, the total degree days weighted by average seasonal COP values are 1,173. Overall, the efficiency with which primary energy spending reaches the end-use sectors is found to be 73.5%.

The presence of a strategic framework for climate mitigation and the characteristics of the urban energy supply structure within which energy and CO₂ saving measures take place are evaluated in the scope of D₂ as summarized in Table 4. In the present sample, 21 cities implement SEAP, SECAP or equivalent plans while 5 cities monitor related data without a similar action plan with detailed measures. In addition to the urban energy supply structure, the status of improving the energy performance of buildings, increasing density of public transport, and upgrading the public lighting infrastructure is subjugated to sub-indicators.

One of the major shortcomings of urban energy supply structures is the allocation of energy resources that have a high useful work potential, i.e. high exergy, to demands that require low exergy content [131]-[132]. The most prevalent example is the use of natural gas to meet the low exergy demands of space heating [132]. Cities with Heat Only Boilers (HOB), heat only district energy networks, or any similar reliance are differentiated from those cities in which these may not be the dominant features of the heating and/or cooling sector.

More efficient options include those for District Heating or Cooling (DH/C) networks based on Combined Heat and Power (CHP), the integration of residual sources of heat, including the residual heat of urban wastewater, and developments towards Fourth Generation District Heating (4GDH) networks that operate with lower supply temperatures [133]. These energy supply options are scored categorically considering penetration and the use of renewable energy. The average city receives a score of 1.2 out of a possible score of 3.0 that represents a city with a DH/C network based on CHP with more than a 75% share in the urban thermal energy load and developments toward 4GDH networks. In Copenhagen, district heating based on biomass and waste covers 98% of the city with progress towards 4GDH [79]. Suburbs at Albertslund and Høje Taastrup are renovated for low-temperature district heating in which the heat loss has been reduced from 40% to 13% with 555 MWh annual energy savings [134]. Helsinki utilizes Thermal Energy Storage (TES) in support of more efficient and flexible energy supply structures. TES units 35 meters below the ground support the network with 14,500 district heating and 250 district cooling connections [135].

Table A1 details the characteristics of the urban energy supply structure for all cities. For example, the district energy system of Gothenburg uses large-scale heat pumps and includes the integration of waste heat from industrial processes [85]. By the year 2030, the remaining share of fossil fuels in the DH system will be phased-out with the use of renewable and residual energy [84]. By the same year, Berlin is pursuing an energy concept to phase-out coal usage based on solutions for gas or biogas based CHP, renewable energy, and power-to-gas solutions [72]. Within this concept, Berlin is investigating the use of low-exergy transmission pipes with different temperature levels [136]. Antwerp plans to use residual heat from the port area to heat buildings in a Nieuw Zuid district [137]. Vilnius is increasing the share of cogenerated energy to maximize primary energy savings [138].

Other cities plan to integrate large-scale solar heat and geothermal energy (Amsterdam [59]) or lakewater (Bregenz [76]) in the future as part of urban energy efforts. In Braga, renewable generation based on PV, biomass CHP, and solar thermal is planned to reduce fossil fuel use by about 12,321 MWh per year [75]. In contrast, the opportunity to progress towards more advanced urban energy systems is not being utilized in other cities. Presently in Bangalore, district cooling that is more efficient than individual air conditioning units is not planned while related initiatives are starting in pilot Indian cities [139]. Individual air conditioning units are prevalent in Rio de Janeiro that has similar options for district cooling while only building-scale trigeneration units were analysed for the Olympic Games [140].

In addition to the energy supply structure, renovation measures to improve the energy performance of residential, commercial, and municipal buildings are expected in local plans. The average energy use of residential (173 kWh per m²) and non-residential buildings (251 kWh per m²) per year in the European building stock under normal climate conditions [141] represents significant room for improvement. In contrast, cities that have implementations of nearly Net-Zero Energy Buildings (nZEB), energy-plus buildings, carbon neutral buildings and/or districts can obtain advantages towards transforming the local building stock. Buildings that meet the nZEB target have high energy performance for reduced energy loads that are covered by a significant share of renewable energy sources [142]. In most cases, primary energy use intensity should not be higher than 50 kWh per m² per year [142]. The details of nZEB related developments are given in Table A2. Cities that have successful undertakings of such targets include those at the building and some at the district levels. In London, a zero-carbon target for residential development has come into effect [98] while the zero energy development site BedZED [143] relies on renewable energy sources. In Amsterdam, local regulations encouraged only climate-neutral buildings to be built from 2015 onwards [59]. The building of the China Academy of Building Research in Beijing operates as a nearly nZEB [144] while the Nieuw Zuid area of Antwerp will be a passive house district with 2,000 dwellings [145]. An average city receives a score of 1.4 out of a possible score of 2.0 for energy savings in buildings based on refurbishment measures and pilot buildings and/or districts that reached or neared net-zero energy targets.

A well-integrated public transport network with high levels of passenger usership is one of the most effective measures against traffic congestion with multiple benefits beyond energy and CO₂ savings, including those for air quality and the quality of life. Table A3 provides an evaluation of the density of public transport, including urban rail that has an advantage in operational energy usage per passenger kilometer [146]. The average urban rail density is 0.15 km/km² with a daily ridership of 9,020 passengers per km. Some cities further support the bus, trolleybus, and/or urban rail network with decentralized options based on pilot or mature bicycle sharing programs. Based on Table 4, a total of 6 cities, including Copenhagen and Helsinki, receive the highest possible score in the indicator for the density of public transport, some of which also correspond to cities with less than 25% extra traveling time due to daily congestion [147]. In contrast, traffic congestion reaches over 60% and 80% in peak morning and evening hours in Beijing and Rio de Janeiro [147].

In addition to energy and CO₂ emission saving measures for the sectors of energy generation, buildings, and transport, another major domain of action and reporting in cities is public lighting. Public lighting based on solid-state lighting using Direct Current (DC) from renewable power is one of the most efficient options that are available. Amsterdam has such an implementation in the port area with both energy and material savings due to the elimination of converters and alternating current cables with plans to diffuse this measure to the rest of the city [148]. Other best practices include London that also receives a top score while the average city receives a score of 1.3 out of 2.0 in this indicator. Other cities have disadvantages, including higher electricity use per km of lit roads in Rio de Janeiro [149].

The widespread availability of renewable energy sources provides a vibrant opportunity to transform the energy base of cities. The integration of multiple renewable energy sources into urban systems can improve self-sufficiency and optimize the overall energy structure. In addition, a responsive demand side that has sufficient flexibility to follow and/or store the energy supply from variable renewables in quantity and quality across the time dimension is essential. Table 5 provides an evaluation of the renewable energy potential of cities and the current status of utilizing this potential based on the indicators in D₃. The existing gap in utilizing the renewable energy potential of cities is most evident in the cases of Batna and Lviv with shares of 1.00% and 6.54% of renewable energy in the electricity supply as well as negligible shares of green energy in the transport sector. In contrast, these and other cities have some of the highest potentials for renewable energy, including Batna that has an annual mean solar energy potential of 5,750 Wh/m²/day and above average geothermal energy potential.

Based on Table 5, an average city has an annual mean of 4,466 Wh/m²/day of solar energy potential based on solar insolation on an optimally inclined plane, 4.8 m/s of wind energy at a height of 50 m above ground level, and 60 mW/m² of geothermal energy based on the heat flow intensity. The average share of renewable energy in electricity production is 38.46% of which 13 cities are above this average value. The city that is closest to the average is Helsinki with a value of 39.00%. In the transport sector, the average city in the sample has a share of 5.44% of green energy, including biofuels and electricity with at least a 45% renewable energy share. Among the cities in the sample, Bregenz, Copenhagen, Gothenburg, Helsinki, Rio de Janeiro, and São Paulo take place as the cities that are above the average in both indicators for renewable energy utilization. The remaining cities have mixed performances across the indicators for the share of renewable energy in the electricity and transport sectors.

As put forth in local strategies, policy efforts are ongoing to increase the penetration of renewable energy in urban energy systems, which would further increase the performance of a given city in D₃. For example, in Florence, the penetration of solar energy installations in public buildings is 9.6 kW per 100 inhabitants that is low in comparison to other Italian cities [150]. Under on-going measures in its SEAP, Florence seeks to change this situation with a target of applying PV and/or solar thermal units to at least 10% of building structures [82]. The utilization of renewable energy sources to obtain energy outputs with a high useful work potential (exergy) is prioritized in D₃ while the cascaded use of renewable energy to lower exergy demands will produce additional benefits for CO₂ emission savings as captured in D₅. In addition, the use of wind energy and solar PV technologies has the potential of displacing about 2,500 litres of water consumption per MWh that would have been used in coal or nuclear based power plants [151]. Renewable energy thereby has a central role in addressing multiple domains in the Sustainable Development Goals [152].

Levels of water usage in cities can either prevent or necessitate undertakings to expand the water supply network and tap into the use of additional limited water resources. For this reason, the ability to manage water consumption by increasing the efficiency of water usage and recycling water resources becomes crucial. Already, some cities are targeting to have at most zero increases in potable water usage as a model for other cities to follow [158]. In the present sample, the performance of an average city indicates a water usage of 7.9 m³ per capita when measured based on the blue water footprint of domestic consumption, which excludes green and grey water footprints as well as any blue water that may be used for products consumed elsewhere [159]. In Grand Lyon, for example, the blue water footprint of domestic consumption is 11.3 m³ per capita [160] that is higher than other localities in France and the majority of other values in the sample. In contrast, the value is still more favourable than the per capita values of 14.0 m³ in Florence and 12.2 m³ in Gothenburg.

Aspects of water usage are interlinked with aspects of water quality in such a way that the excessive extraction of resources from water reservoirs can reduce the dilution capacity of water bodies and deteriorate water quality while posing critical issues of water stress. In contrast, the recycling of stormwater and wastewater can improve waterway health and in so doing, also benefit other cities that may share the same water basins [161]. Based on measurements of nitrogen, potassium, pH, conductivity, and dissolved oxygen, the value of the water quality index [162] in an average city is 81.8 out of a possible score of 100. Overall, 14 cities are at or above while 12 cities are below this average with the highest and lowest values being put forth by Helsinki with a value of 99.1 and Batna with a value of 37.7, respectively. An important source of water for Helsinki comes from Lake Päijänne that is treated and then distributed to homes. In the past decades, an extensive effort to limit the flow of chemicals from agriculture and pulp industries to water bodies was put forth to obtain the present levels of water quality [163]. In contrast, the operation of the wastewater treatment plant of Batna is currently suboptimal among other issues, including impact from the contaminant load of the industry [164] that continues to impact overall water quality.

Based on readings from urban monitoring stations for the two most recent data series [165]-[166], Braga, Bilbao, and Rio de Janeiro recorded the most improvement in annual mean PM₁₀ values. In these cities, the annual mean PM₁₀ values improved by more than 25% from the values in the previous data series. In Rio de Janeiro, the previous reading was 67.0 µg/m³ [165] while the most recent reading from the same urban monitoring stations is given to be 48.8 µg/m³ [166]. Other cities marked less of an improvement, including Beijing that was able to improve an extreme value of 121 µg/m³ [165] to only 108.0 µg/m³ [166]. Coal burning in the urban areas of Beijing was officially banned while increasing rates of car ownership in the city and industrialization in the neighbouring areas continue to affect air quality [167].

All cities, however, did not record improvements in values among which at least 8 cities had degradation of annual mean PM₁₀ concentration values. Those of Tianjin degraded from 96 µg/m³ [165] to more than 125 µg/m³ [166] and similarly, those of Bangalore deteriorated from 103.3 µg/m³ [165] to 118.1 µg/m³ [166]. Based on Table 6, annual mean PM₁₀ concentration across all 26 cities in the sample is 39.7 µg/m³ or 26.3 µg/m³ with the winsorized values of 4 outlier values. The ability to obtain further improvements in annual mean PM₁₀ values of which 70% may be attributed to PM_2.5 concentrations is crucial given that each 5 µg/m³ increase in PM_2.5 can lead to a 20% higher risk of developing deposits in coronary arteries [168] that trigger coronary diseases as a major public health concern.

Stress on environmental quality is further imposed based on the demand for land area to support the current levels of food and material consumption in society. The ecological footprint method quantifies the demand for land across six categories [169], including those for built-up land and various kinds of agricultural produce. The average level of ecological footprint in the sample is 4.36 global hectares (gha) per capita, which indicate the area that is needed to support the annual consumption pattern of one individual. The per capita values for the two Chinese cities of Beijing at 3.92 gha and Tianjin at 2.88 gha [170] compare differently with the national value at 3.49 gha per capita while both values are below the sample average. In comparison to biocapacity as the capacity of natural systems to regenerate, the average ecological deficit in the sample is 1.15 gha per capita.

Such an ecological deficit further underlines the need for a rapid transformation to more sustainable production and consumption patterns, including those through such paradigm shifts as the shared economy. In the metropolitan area of Helsinki, for example, only 4% of private vehicles are found to be necessary to sustain current levels of vehicle transport considering shared mobility schemes [171]. In addition, as shared mobility is able to replace all private car journeys, CO₂ emissions could be curbed by about 34% and the land for public parking could be utilized for productive purposes [171]. Such analyses exemplify the co-benefits of shared mobility alongside those for reduced congestion.

The decarbonisation of urban systems based on the use of renewable energy in the urban context in buildings, transport, and the industry remains to be a challenge around the world although progress is being made. For example, among the cities in the present sample, the urban and inter-city rail network of Amsterdam as well as other Dutch cities is now supplied with 100% renewable energy [177] while the built area surfaces of urban rail stations in Antwerp [178] are being used for building-integrated photovoltaics. In contrast, the energy system of Amsterdam is largely based on the use of natural gas for individual space heating of buildings. This represents a thermodynamic mismatch with ramifications on additional primary energy spending as well as CO₂ emissions [179]. By the year 2020, however, the city has plans in place to realize an efficient expansion in the district heating network with close to a doubling of buildings that are connected to a district energy scheme [63]. The integration of renewable energy in urban waterworks also presents opportunities for cities.

Table 7 provides the data compilation for the indicators in D₅ and represents the most recent status of CO₂ emissions for the cities in the sample. Based on monitoring reports submitted under the CoM [180] or local statistical yearbooks, an average city in the sample emitted 9.21 million tonnes of CO₂ emissions from buildings and 2.98 million tonnes of CO₂ emissions from transport. Including all other sectors except industries that may be included in EU ETS, which is not reported in the framework of the CoM, an average city emitted 0.28 tonnes of CO₂ emissions per MWh. One of the cities that appear to be closest to reaching a complete decarbonisation of the energy sector based on a relatively lower CO₂ intensity is Gothenburg that emits 0.13 tonnes of CO₂ per MWh [86]. Another example may be given from Copenhagen that currently emits about 0.17 tonnes of CO₂ per MWh [81].

The presence of industry in the urban context may be a disadvantage for present levels of energy spending and CO₂ emissions. Based on Table A4, 15 cities had 4 or more energy and CO₂ intense industries and/or clusters located within the urban context. The utilization of residual heat from industry that is in proximity to urban areas with greater than 50 TJ/km² of heat demand density, however, can provide a promising opportunity to substitute primary energy spending in the energy supply structure of cities [181]. For example, the residual heat of two thermal power plants with the main activity of electricity generation that is less than 10 km from central Amsterdam can provide 24.91 TJ of heat on an annual basis [182]. Among other industries, the non-ferrous metals industry that is located in close proximity to Antwerp can provide 0.77 TJ of heat on an annual basis. According to such results from the Heat Roadmap Europe Pan-European Thermal Atlas (Peta) 4.2 [182], a significant share of current primary energy spending can be substituted based on residual heat from industry.

In addition, it is important that airports servicing the cities are included in the scope of urban energy systems, especially since the energy and CO₂ emissions impact of airports can often exceed those of particular districts of the city. Airports that have undertaken measures for on-site renewable energy generation and smart grid applications as well as innovative energy harvesting measures such as those in [183] obtain an advantage over other airports in which such implementations may not have been undertaken. The pilot power-to-gas facility in one of the airports of Berlin, namely Brandenburg Airport, started to use wind and solar energy to power a 500 kW alkaline electrolyser to produce hydrogen gas. The gas is then used for the energy needs of transport on the groundside as well as cogeneration in the terminals [184]. In total, the 3 airports that are certified with climate neutrality under Airport Carbon Accreditation (ACA) have best practices in the uses of renewable energy. For example, Bangalore Kempegowda International Airport is increasing on-site electricity generation based on solar energy towards a 50% renewable energy share [185]. Similarly, Helsinki Airport now has the largest solar power plant among Nordic airports [186].

The planning of urban systems towards cleaner and more liveable cities is an important component of optimizing the interactions between people, technology and the environment in the context of cities as complex systems. More liveable cities are dependent on multiple aspects including urban infrastructure, urban form, green spaces, and social welfare [189]. In this respect, infrastructure for urban waste and wastewater management enables an essential urban service that can also support the transition to a more resource-efficient, closed-loop society. As amended under the Circular Economy Package [190], it is expected that at least 65% of municipal waste is recycled by the year 2035 while at most 10% will be landfilled towards zero waste. The Urban Waste Water Treatment Directive [191] stipulates limits for compliance in collecting, treating, and discharging subjected wastewater loads.

Based on the data compilation for the sub-indicators in Table A5, an average city in the sample has a recycling and composting share of 30%. Recycling and composting can take place among the most environmentally friendly options in waste management [192]. The leading cities are Berlin (68%), Bregenz (67%) and Antwerp (63%) while the lagging cities are Batna, Beijing, Lviv, Rio de Janeiro, São Paulo, and Tianjin with a share of 5% or less. In contrast, separate waste collection is a critical basis for increasing the share of recycling and composting in the waste hierarchy towards zero waste municipalities [193]. The cities with relatively higher recycling and composting shares, however, are not always the cities with the lowest waste generation per capita, both of which are important according to the principles of waste hierarchy. In the present sample, an average city produces 398 kg of waste per capita. Berlin, Copenhagen, and Helsinki take place as the cities that have below average waste production per capita as well as recycling and composting shares above 40%.

The level of compliance with the Urban Waste Water Treatment Directive is improving with deficiencies in about 12% of cases for more stringent treatment [191]. Among cities in the sample, the evaluation of the compliance summaries of wastewater treatment plants indicates that those of 17 cities are directly compliant based on levels of Biochemical (BOD) and Chemical (COD) Oxygen Demand as well as Total Dissolved Solids (TSS) in treated wastewater [194]. Moreover, important amounts of nutrients can be recovered from municipal wastewater streams based on nutrient recycling programs. Some cities, including Amsterdam, Helsinki and Grand Lyon, treat 100% of wastewater with nutrient recovery techniques to recover nitrogen, potassium, and/or phosphorus as chemical commodity [195]. In other cities, selected wastewater treatment plants are found to have non-compliances in certain parameters as well as partial collection of the wastewater load as given in Table A6. For example, Vila Nova de Gaia generates 28,000 population equivalents wastewater load while primary and secondary treatment is not compliant with criteria, including BOD [191].

Important parameters to distinguish aspects of compact urban form include the number of non-continuous urban centers, the share of the population living in core urban areas, and the sprawl index [196]. Cities that have polycentricism, a relatively high concentration of inhabitants in core urban areas, and sprawl index far less than zero can be more efficient and pose less of a threat on land-use changes through compact urban form [197] in contrast to cities in which these conditions are the opposite. Urban intensification that prioritizes the development of central nodes has been pursued as a strategy to increase the quality of life through more compact urban form in such cities as Gothenburg [198]. In fact, the city has managed to achieve a negative sprawl index at −1.97% with 56.15% of the population living in core urban areas [196]. In contrast, the highest urban sprawl is seen to take place in Madrid at 4.57%, followed by Porto that has a similarly high sprawl index at 4.30% [196].

The allotment of green space and the protection of green corridors are other priorities that must be undertaken simultaneously while intensifying urban development [199]. In the present sample, 5 cities have green area as a share of total urban area at or above 40%. Green areas also limit the share of impermeable surfaces in the city as an important climate adaptation measure to reduce the risks of inundation and heat waves. For example, the share of green area is 53.71% while impermeable surfaces are 28.85% [200] in Gothenburg that seeks to be a Water-Wise city [201]. The area of protected green corridors within a 150 km radius of the city, including national parks, nature reserves, and Ramsar Convention sites is evaluated for all cities. Amsterdam has the greatest diversity of such sites with over 3,184 km² of protected land area. Based on Table A7, cities with both favourable compact urban form and green areas include Amsterdam and Gothenburg while an average city scores 1.8.

In relatively more liveable cities, quality of life is further shaped by the availability of economic and educational opportunities, including income and perceptions of well-being [202]. Based on the values in Table 8, an average city in the sample had a Gross Domestic Product (GDP) of 30,969 PPP USD per capita at 2010 constant prices. In addition, an average citizen scored satisfaction with daily experience at 7.2 out of 10 when adjusted based on inequality. On average, 32.5% of the population had attained the level of tertiary education, which can increase employment opportunities [203]. Cities that receive a higher standing in such parameters of social welfare in Table 8 are Gothenburg, London, and Copenhagen.

The cross-cutting field of R&D, innovation and sustainability policy can support the performance of a given city in the above indicators given that the knowledge base of local stakeholders are integrated with the measures in related strategies, including Sustainable Urban Mobility Plans. First, for such a synergetic context, priorities in energy, transport, the environment, as well as smart and sustainable cities must be upheld in the strategies of the R&D and innovation systems in which the local stakeholders interact. Table A8 provides an evaluation of the R&D and innovation policy orientation within a multi-level governance framework [219] based on related priorities and the resources that are available for their implementation. Policy tools include the City of the Future initiative for German cities [220], including Berlin, and the support of the Ministry of Cities in Brazil for low-carbon transport [221], including Rio de Janeiro and São Paulo. Furthermore, at least 8 cities are demonstration sites of European smart city projects [222]. The highest gross domestic expenditure on R&D (GERD) as a percentage of GDP is 3.27% while the average is 1.69%.

Clean technology patents that are classified with the codes Y02 and Y04 [223] are one of the outputs of the functions of R&D and innovation systems. These patents include those on building technologies (Y02B), energy generation (Y02E), transport (Y02T), smart grids (Y04S), as well as carbon capture and storage (Y02C) [223]. Based on Table A9, the largest quantity of clean technology patents belong to Germany while the highest share of Y02 and Y04 coded patents in total national patents belongs to Morocco. In addition, Copenhagen and Paris act as hubs of innovation in their countries given that more than 35% of international patent applications have applicants located in these cities [196]. Copenhagen contributes to the international patent applications of Denmark with a share of 55% [196].

The presence of local universities and research institutes are further assets for the local knowledge base. Table A10 identifies the strength of cities based on the presence of institutes of higher education and research in the city. The quality of education and research are further integrated into the evaluations based on a weighted scoring for institutions from the city that take place in the top 1,000 institution rankings of SCImago [224]. The SCImago institution rankings are based on excellent performance in research, innovation outputs, and societal impact [225]. The cities with the greatest advantage in the local knowledge base include the cities of London, Beijing and Tianjin. In addition, the sustainable city initiatives of Amsterdam are supported by the Amsterdam Institute for Metropolitan Solutions that aims to bring the city closer to the principles of a circular economy [226]. In contrast, extended collaboration networks and knowledge transfer becomes increasingly important for cities that may not have similar assets in local knowledge production. Cities that may be seen to include weaknesses in both local institutes and the h-index include Batna and Lviv.

Based on Table 9, an average city in D₇ receives a score of 2.2 out of 3.0 in R&D and innovation policy orientation and 2.1 out of 3.0 in clean technology patents. In addition, an average city has about 6 universities of which 4 takes place in the SCImago rankings. The strength of knowledge production as further represented by the h-index is an average value of 562. Overall, the assets that are available to transform cities into hubs of sustainable innovation must be orchestrated to support the CO₂ mitigation targets of cities for the year 2020 and beyond. In an average city, the target is to reduce CO₂ emissions by 25% from the baseline by the year 2020, including the annualized value of targets that are set beyond 2020. The highest target is the target of Copenhagen in which the city will be climate neutral by the year 2025 [79] that requires an annualized reduction of at least 67% by the year 2020.

Figure 2 provides the normalized values of the indicators in D₁ for the 26 cities in the current sample. These normalized values represent the total unweighted scores for D₁ prior to the index aggregation in eq. (1). These values are further organized into quartiles in which the top performing cities that take place in the top 25% of these values are marked within the range for Q₄. Accordingly, the top performing cities in D₁ based on aspects of energy usage and climate in descending order are Salé (40.571), Karlovac (40.278), Bregenz (40.042), and Vilnius (38.971). Among these cities, Karlovac gained an advantage in D₁ during the most recent monitoring phase that indicated over a 40% reduction in the energy usage of buildings and 9.7% reduction in the energy usage of transport from the baseline year [95]. On a per capita basis, energy usage decreased from 18.08 MWh to 10.21 MWh [95]. The cities of Batna and Bilbao also take place in Q₄.

Considering all cities in Figure 2, the sample average C_AV for D₁ is marked in the last stacked bar at a value of 31.052 and extended with a dotted line as a reference value for the other cities. In comparison to this reference value, the ratios of the values C_j over C_AV range between 1.31 and 0.44 as given in the triangular markings. In the latter case, the city of Tianjin performs below the average in most indicators, including energy usage per capita based on a value of 39.73 MWh per capita and a final to primary energy ratio of 68.1%. Cities that minimize energy usage in buildings and transport relative to both population and climate with a favourable energy system efficiency perform better in D₁.

Figure 3 indicates that Copenhagen obtains a perfect score in D₂ based on a score of 50.000 for the sum of the normalized values of the indicators that evaluate the penetration of energy and CO₂ saving measures. Such a performance represents the achievement of the top values in each indicator in this dimension. Other cities that take place in the top 25% at Q₄ include Berlin, London, and Grand Lyon that have normalized values between 46.667 and 44.667 while Helsinki, Amsterdam, and Gothenburg follow the performances of these cities based on the next best values above the top quartile threshold of 41.500. These cities in Q₄ adopt a more integrated approach towards optimizing the urban energy system from both the supply and demand sides. In contrast, the sample average C_AV for D₂ is 30.327 with cities performing at ratios as low as 0.57 of this average value due to limitations in measures that encompass the urban energy supply structure and the demand side. From Salé (20.667) to Batna (18.667), the only strength is the monitoring of SEAP while the only relative strength of Bangalore (18.500) is the public transport network.

The results in Figure 4 indicate that Rio de Janeiro (34.494), São Paulo (32.799), and Gothenburg (29.294) attain the highest performances in D₃ among the 26 cities in the sample. The performances of these top performing cities have also increased from the previous reporting phases. For example, the renewable energy share in the electricity grid for Rio de Janeiro and São Paulo have increased by 9.9% in one year and reached 81.19% [156]. After these cities, the relatively high performance of Gothenburg is also supported by the second highest green energy share in transport including both biofuels and electricity at 14.34%, which represents progress towards decarbonizing the transport sector. Based on aspects of both renewable energy potential and utilization, the average city receives a score of 22.403 in D₃ of which cities perform at ratios of 1.54 and 0.53 of this average value.

Figure 5 provides the normalized values of the indicators in D₄ in which the city of Helsinki (39.802) is found to be the top city in aspects of water usage and environmental quality. Helsinki is able to combine a relatively low footprint for blue water consumption with excellent water and air quality. In addition, the biocapacity per capita is much greater than the ecological footprint per capita, which places the city in a more favourable position considering the balance between the demand for land across various categories and the natural regenerative capacity. The cities of São Paulo (39.594) and Rio de Janeiro (37.420) surpass the performance of Helsinki in only one indicator, namely ecological footprint per capita, while the performance of Gothenburg (37.323) surpasses that of Helsinki in the indicator for the annual mean PM₁₀ concentration. Other cities that take place in Q₄ are Vilnius (35.968), Braga (33.153), and Karlovac (32.624). In contrast, the average city in the sample receives a normalized value of 30.261 in D₄. Cities have performances at a ratio of at most 1.32 and at least 0.59 of this average value based on the triangular marks for each city.

The latter case of a 0.59 ratio with the average value represents Batna (17.890) that receives the lowest value in D₄ based on shortcomings in water quality, annual mean PM₁₀ concentration, and biocapacity per capita. The cities of Tianjin (23.521) and Beijing (22.140) perform better than Batna only in the indicator for water quality with major shortcomings in annual mean PM₁₀ concentration as well as biocapacity per capita. Here, only the ecological footprint per capita in Tianjin is lower than the value for Beijing [170].

Figure 6 indicates that Bregenz (37.709) has the best performance in the indicators of D₅ for CO₂ emissions and industrial profile. Copenhagen (37.620), Braga (37.346), and Vila Nova de Gaia (36.921) are the next best performing cities in D₅ with performances in Q₄. Three additional cities, namely, Karlovac (36.907), Helsinki (36.486), and Gothenburg (35.612) further take place in the top 25% of cities in D₅. In comparison to performances in previous dimensions, Bregenz and Karlovac had also taken in the top quartile in D₁ while Copenhagen, Helsinki, and Gothenburg had attained top performances in D₂. In addition, Bregenz, Copenhagen, Vila Nova de Gaia, and Gothenburg had obtained top places in D₃ for renewable energy and utilization. In contrast, the cities that are closest to the sample average for D₅, namely Bydgoszcz (27.548) and Lviv (27.207), have certain weaknesses, including those in CO₂ emissions intensity of the energy mix. The cities that take place in the lowest 25% at Q₁ have values well below the bottom quartile threshold of 25.289.

Based on the results in Figure 7, Gothenburg (39.107), Amsterdam (37.272), and Copenhagen (36.985) are the best performing cities in D₆ for urban planning and social welfare. These cities have advantages in multiple indicators under D₆, including those for waste and wastewater management, compact and green urban form, and inequality adjusted well-being. Best practices from Gothenburg include a recycling and composting share of 48.0%, which is above the European average at 43.6% [204], and a negative sprawl index at −1.97% that indicates a retracting urban area while 56.15% of the population is living in core urban areas [196]. In addition, Gothenburg has favourable performances in indicators for social welfare, including a tertiary education rate at 51.0% for the population aged 30-34. Other cities that take place in the top 25% of D₆ are London (35.874), Helsinki (35.265), Berlin (35.121), and Bregenz (34.289). The average city receives a score of 26.071 in D₆ with cities performing at a ratio of 1.50 above or 0.33 below this average value.

Figure 8 puts forth the results for the quartile performances of the cities in D₇ in which London (37.871), Berlin (37.754), and Copenhagen (31.460) are the top cities. Especially the top two cities obtain relatively strong performances in all of the five indicators. Among these cities, London has the strongest base of local public/private universities and research institutions, including those taking place in the SCImago top 1,000 institutions [224]. Such an outcome is rivalled only by the performances of Beijing and Tianjin that also take place in the top quartile for D₇. In addition to R&D policy orientation, strengths of Copenhagen include its progress towards its climate neutrality target, particularly in the heating and cooling sector [231], with knowledge sharing with other cities. Following these cities, Amsterdam (30.281) and Helsinki (27.680) also take place in the top quartile.

The normalized values that are provided in Figures 2-8 for each dimension of the SDEWES Index enable the aggregation of the values according to eq. (1) for each city in the sample. According to the method, the weights are also subjugated to 10,000 Monte Carlo experiments, the results for which are provided in Figure 9 based on an ordering with the original aggregated index value. In comparison to the mean simulated values of the Monte Carlo experiments, there is only one position in which an interchange of rank takes place between two adjacent cities in the sample. Antwerp advances while Madrid recedes by one rank each where higher ranks have decreasing rank number. The order of all other cities and the inclusion of cities in quartiles of performance are stable.

Figure 9.

Results of the Monte Carlo simulations and performance quartiles

The colouring in Figure 9 is further representative of quartiles of city performance. The sample is normalized according to the overall integrated city sample [121] so that the number of cities that are allocated to each quartile may not be equal for the present city sample. Cities that take place in the lower 25% of the integrated sample (Q₁) obtain an aggregated index score below 26.891, including the cities of Lviv and Berlin alongside 8 other cities. The cities in Q₁ have weaknesses in multiple dimensions and are termed as the “Challenged Cities” [56] of a given sample. The remaining cities that perform below or at the median value take place in the second quartile of Q₂, which includes Florence and Bilbao. These cities have a more limited set of weaknesses in any dimension and have been termed as the “Solution-Seeking Cities” [56]. Above the median value, the third quartile of Q₃ includes the cities of Porto and Karlovac for a total of 5 cities. These cities represent the “Transitioning Cities” [56] that are able to attain performances that are relatively better well-rounded across multiple dimensions. In contrast, the highest level of performance is reserved for cities in the top quartile Q₄ that are termed as the “Pioneering Cities” [56]. The 6 cities that take place with this status starting from the top performer are Copenhagen, Helsinki, Gothenburg, Bregenz, Amsterdam, and Vilnius.

Table 10 summarizes the SDEWES Index scores for the 26 cities based on eq. (1). The percentage difference in the performance of each city with the median value of the integrated sample is also given. Copenhagen that ranks in first place with an index score of 36.038 performs 24.1% above the median value while Bilbao scores the closest to the median value with a difference of −0.1% based on a value of 29.021. Other cities perform well below the median value, including Lviv, Batna, Bangalore, Beijing and Tianjin.

Some of the best practices for the top three cities in the present sample are overviewed in Table 11. There is not a single best practice that can directly take a city to top ranks. Rather, one of the most important guiding principles to improve performances in the SDEWES Index is the implementation of an integrated approach across all dimensions simultaneously. The best practices that are selected for inclusion in Table 11 span urban energy, water and environment systems among other possible examples. The replication potential of the best practices can depend on multiple factors, such as local energy characteristics. However, the guiding approach that is put into action in the best practices, such as ensuring the more rational use of energy resources, protecting water resources, and minimizing waste and CO₂ emissions, is directly applicable to all cities.

Based on the results in Figure 9 and Table 10, Rio de Janeiro takes place as one of the cities in the lower 25% quartile (Q₁) among the “Challenged Cities.” As one of the cities in this quartile, the city has weaknesses in certain dimensions. Based on data in Figures 2-8, the SDEWES Index results for Rio de Janeiro are compared in Figure 10 across all dimensions collectively. In comparison, the city performs close to the sample average in D₁, below the average in D₂, above the average in D₃ and D₄, below the average in D₅ and D₆, and above the average in D₇ to obtain a score of 25.981 as the overall Index value.

Figure 10.

SDEWES Index results for the present performance of Rio de Janeiro

Any future developments in the city could enable increases in the performance of Rio de Janeiro. Table 12 summarizes local aspirations in the context of targets to realize Visão Rio 500 under the third theme of being “a green, sustainable, and resilient city” [124]. The first subtheme is sustainable development towards zero carbon [124]. The target years of related aspirations for this subtheme are provided along with connections to the scope of the SDEWES Index. Exemplary opportunities that are available to realize these aspirations and improve dimension and overall index scores of the city also take place within Table 12. Some of the principles are marked to have potential spill-overs to more than one dimension.

Already, the City Hall has issued the Qualiverde label that provides financial incentives to buildings with reduced environmental impact [235]. The use of solar panels, green roofs, and efficient lighting is indicated to be increasing [235]. In the aspect of urban transport, new Bus Rapid Transit lines and a light rail transport in Porto Maravilha are transforming mobility options with benefits for energy, emissions, time, and social cohesion. In addition, 400 km of bicycle paths represent the greatest extent in Latin America [124]. Under waste management, however, 3% of total waste generation is recycled whereas the recyclable fraction is about 40% [124] that requires greater separate collection. In the meantime, a new waste treatment center (Seropédica) has been opened that involves bioenergy production while the old landfill at Gramacho is installed with a landfill gas recovery system [124].

The exemplary opportunities in Table 12 further extend to the use of wastewater sludge to recover nutrients and materials, including those that could be recovered to produce concrete fillers and aggregate binders as building construction material [233]-[234]. The replacement of individual cooling units in buildings with district cooling is another solution. Best practices of better performing cities can be adapted to the context of Rio de Janeiro.

Under Visão Rio 500, the second subtheme towards “a green, sustainable, and resilient city” involves resilience to climate change for minimizing damage and saving lives [124]. The ambitions of this subtheme are summarized in Table 13 along with the principles and target years. The rivers of the city can supply up to 9 thousand tons of water per day for treatment as tap water in regular seasons [124]. In contrast, the rivers can also pose treats to urban inundation. Currently, flood control measures are implemented throughout the city, including in Grande Tijuca and Praça da Bandeira while the Jacarepaguá Basin is reinforced with a macro-drainage system [124]. The limitation of urban sprawl to avoid development into areas that may be prone to risks of landslides and floods is also another valid option based on the indicator on compact and green urban form in D₆ of the SDEWES Index.

The third subtheme through which Rio de Janeiro seeks to be a global benchmark as “a green, sustainable, and resilient city” is based on protecting its water resources in the form of beaches, bays, ponds and rivers [124]. As an extended interpretation of the name of the city, Rio de Janeiro aims to become a “City of Waters” while upholding the highest standards for water quality and water sanitation. Table 14 provides the ambitions of the city under this subtheme together with the principles and target years. The ambitions as given in Table 14 are important given that some of the water bodies of the city, including the Bay of Guanabara and the lagoon of Barra, are poorly used due to frequent pollution issues [124].

Rio de Janeiro is recognized as a city with insufficient infrastructure for water treatment based on low coverage of wastewater collection and missing connections [124]. The share of wastewater treated to total water consumed had been about 50% [236] while a new wastewater treatment plant is operational [237]. An interim target is set towards ensuring 80% coverage of wastewater treatment to overcome current insufficiencies in providing for basic sanitary needs [124]. Water stress is another area of concern such that periods of drought can impact the limited water supplies from the reserve of the Guandu and Paraíba do Sul River [124]. As a remedy to these concerns, alternative sources of water supplies are expected to be tapped. At the same time, access to water supplies that may be further away from the city may increase energy needs for the transport of water. The use of gravity distribution networks may be maximized and electricity for water pumping can be met with the co-location of renewable energy in water facilities among possible options (Table 14).

The ambitions, principles and target years as summarized in Tables 12-14 are used to consider a Visão Rio 500 scenario for Rio de Janeiro under the SDEWES Index. Only those indicators that are directly impacted by the targets as defined in Tables 12-14 are changed according to Visão Rio 500. Those that are used in the Average City to catch the best performing cities by target years as given in [121] are used for the other indicators. Targets that required an update based on Visão Rio 500 include 60% lower emissions from 2012 baseline emissions by the year 2050 (see Table 12). The realization of this target requires an annual average reduction of about 1.58% each year based on eq. (3), which is lower than those for an Average City scenario [121]. In practical terms, reductions in CO₂ emissions involve the effect of both efficiency and renewable energy measures. For this reason, CO₂ emission reductions based on the Visão Rio 500 target need to be decomposed. Only expected contributions from any efficiency measures are represented in the scenario values of indicators on urban energy usage that is represented in dimension D₁. The possible share of CO₂ emission reductions based on efficiency measures in total CO₂ emission reductions is adapted from International Energy Agency projections [4] (see Table 15).

In comparison to the recent analysis provided by the C40 network in [5], however, a target year of 2065 for carbon neutrality would not be compatible with the remaining CO₂ budget that can provide a fair chance of limiting global warming to at most 1.5 °C by the end of century. Since the adoption of Visão Rio 500, the city of Rio de Janeiro was announced as one of the C40 cities to adopt a carbon neutrality target by the year 2050 [125], which can modify the original target for reaching carbon neutrality given a stability of political will.

In aspects of urban water and waste issues, the target of enabling a reduction in per capita water usage by 50% by the year 2050 (see Table 14) was integrated into scenario values based on reductions at about 0.097 m³ per year based on eq. (3). The target of guaranteeing the water quality of all surface water sources by the year 2050 (see Table 14) required an extension of the related target so that Rio de Janeiro attained the level of the best performing city by the same target year based on eq. (2). The generation of 50% less waste by the year 2050 (see Table 12) was taken to involve reductions of about 1.42% each year from existing levels based on eq. (3). Two of the Visão Rio 500 targets, namely the provision of basic sanitation infrastructure for the entire city and ensuring that no residence will be at risk of landslides and floods (see Tables 13-14) have the same target year at 2035. This implied an extended target of Rio de Janeiro reaching the values of the best performing city for the related indicators in D₆ involving wastewater management and urban form.

In contrast to the overall performance of Rio de Janeiro as given above in Figure 10, the performance in Figure 11 compares the results of the city under a normative scenario for the year 2050 based on targets in the local plan of Visão Rio 500 and the C40 carbon neutrality target. As indicated previously, the target values of the best performing city in the integrated sample are taken for all other indicators [121]. Based on the normative scenario, the index score of Rio de Janeiro at 44.762 in the year 2050 would surpass the score of the currently top performing city under dynamic conditions for the scenario and static conditions for the other cities. In reality, however, dynamic conditions apply to all cities equally and relative performance will continue to change the values of the best performing city in the sample across time. The evaluation of the present performance and future opportunities are discussed in Table 16.

Figure 11.

SDEWES Index results for a normative scenario for Rio de Janeiro