Keyword

1 Introduction

For (Wolsink, 2018) the energy available from renewable sources is abundant, but the real scarcity derives from the space needed for the infrastructure. Natural landscape and principles of co-existence with existent energy and non-energy infrastructure limit these resources’ technical realization. The scientific publications on cities’ renewable energy potential mainly rely on geographic information system (GIS) data such as solar and wind maps. However, planning systems that realize the potential of renewable energy resources require going beyond GIS tools. In this study, we consider two additional dimensions to GIS maps in the assessment of cities’ renewable energy potential: competing space and competing resources.

1.1 Study Sample City

The sample city of Dakar overlaps the boundaries of the administrative region, meaning it has a territory of 547 km2 with four administrative subdivisions: Dakar Peninsula (District 1), Guediawaye (District 2), Pikine (District 3), and Rufisque (District 4).

1.2 Renewable Energy Economic and Technical Potential

The concept of competing space includes three layers of technical constraints. The first layer of constraints is the landscape formatted by natural processes over time. The presence of mountains (or valleys), irregular fields (slope), and natural screens such as canopies are all parameters that impose limits to renewable energy resources. At this point, it is worth clarifying the usual confusion between the terms “sustainable” and “renewable” energy resources, with the criteria of landscape co-presence to argue the difference. For instance, it is indisputable to call solar energy a renewable resource, but if the capture of solar irradiation requires to clear the canopy, we consider co-presence to be outraged; thus, this energy cannot be named sustainable. The same goes for scenarios where valleys are levelled to instal hydropower technology.

The second layer of constraints is co-presence in a finite area of the energy technology with human and other animals’ settlements and with competing infrastructure. Thus, the assessment of the technical potential excludes areas such as animal parks, birds’ reserves, cemeteries, etc.

The third layer of constraints to energy potential realization is the requirement to minimize the distance between the energy transformation system and feed-in-point to the grid in order to increase the energy production potential (EPP) through lowering losses (Masurowski et al., 2016).

The competing resources in the city are conventional energy generation resources that feature the value of having technology infrastructure readily available in the supply mix, which can be numerically accounted as economies of scale. For a technically realizable site to be economically viable, the costs associated with its realization should be lower than that of existing alternatives, which were in majority diesel power generation in the study reference year (2016). The core objective of the chapter is to identify the technical and economic constraints that place boundaries on the infinite theoretical potential of bioenergy, solar photovoltaic and wind energy in our sample city.

2 Methodological Approach

The assessment method is the comparison of unit costs of electricity generation from solar photovoltaic, waste-to-energy and wind energy with the average cost of electricity generation in the study reference year (2016). The cost calculated is the long-run generation cost of electricity. The excel-based computation model uses materials from publications listed below, which are accessible online.

2.1 Assessment of Solar Photovoltaic Potential

The co-presence layer constraints integration of solar photovoltaic systems in the city. The assessment of the city's solar photovoltaic potential relies on second-hand data downloaded from the Global Atlas Map accessible online (Solargis, 2020). From this theoretical potential, we add layers of excluding parameters (e.g. green and water areas), in order to isolate sites where the theoretical potential can be realized with technology, making it the technical potential.

2.2 Assessment of Waste-to-Energy Potential

The distinction between a renewable and a sustainable energy resource excludes from the technical potential of bioenergy, forests and biomass from all areas in the city classified as green or wet. The assessment of the city waste-to-energy potential relies on second-hand data from the city waste collection agency (Unite de Coordination et de Gestion des Dechets, 2016) and from the results of a study on Dakar’s waste characterization completed in 2017 as part of a PhD thesis (Fall, 2017). The theoretical energy potential is the potential of energy recovery from quantities of waste collected in the baseline year. From the theoretical potential, we derive the technical potential based on the assumption of a 100% waste collection rate for the city.

2.3 Assessment of Wind Energy Potential

The co-presence with competing infrastructure layer constraints the installation of wind energy technology in the city. The assessment of the city wind energy potential relies on second-hand data downloaded from the Global Wind Atlas accessible online (DTU Wind Energy & ESMAP, 2020). From this theoretical potential, we add layers of excluding parameters (e.g. residences), in order to isolate sites where the theoretical potential can realize with technology, making it the technical potential. Due to our third layer of technical constraint that minimizes the distance between the generation site and grid feed-in-point, we did not consider the wind-offshore potential of Dakar.

The assessment of each of the three technologies economic potential relies on excel-based computation of the long-run generation cost of the technology, which formula is:

Equation 5.1 Long-run generation cost

$$ \mathrm{LRGC}=\frac{\left(\mathrm{CRF}\ \left.\mathrm{CAPEX}\right)\right.}{\mathrm{H}}+\mathrm{SRGC} $$

Equation: Long-Run Generation Cost Formula

Where,

$$ \mathrm{SRGC}=\frac{{\mathrm{F}}_{\cos t}}{\mathrm{EF}}+\mathrm{Other}\ \mathrm{O}\&\mathrm{M}+\frac{\left({\mathrm{CO}}_{2\ \mathrm{price}}\ast {\mathrm{CO}}_{2\ \mathrm{fsef}}\right)}{\mathrm{EF}} $$

  • CRF is the capital recovery factor.

  • CAPEX is the technology capital cost expressed in EUR per kW installed.

  • H is full load hours, which are defined as equivalent hours of production at full (rated) capacity. It is the capacity factor expressed in hours.

  • Fcost is the fuel cost.

  • EF is the efficiency factor of the energy conversion process.

  • Other O&M stands for other operation and maintenance costs.

  • CO2 price is the price of 1 tonne CO2-eq emitted by the conversion process.

  • CO2fsef is the conversion process-specific emission factor.

3 Results

3.1 Estimation of Solar Photovoltaic Technical Potential

Figure 5.1 downloaded from Solargis shows the city theoretical potential of solar photovoltaic.

Fig. 5.1
figure 1

Dakar Theoretical solar photovoltaic potential. Source: Global Solar Atlas Website

Full size image

Legend: Report generated on 9 January 2020 from https://globalsolaratlas.info/. The energy potential considering the direct normal irradiation (DNI) is 1523 kWh/m2 per year. The surface area of the city is 547 km2. Therefore, the theoretical potential of Dakar on solar photovoltaic is 833,081 GWh per year. Table 1 displays the percent of the land occupied by various uses in the city (Direction de l'Urbanisme et de l'Architecture, 2016).

From Table 5.1, we derive the surface that can host solar photovoltaic systems equivalent to 141.7 km2. Therefore, the technical potential of solar photovoltaic energy is 215,768 GWh. Considering a capacity factor of 17% equivalent to 1489.2 full load hours, this technical potential is 689.9 GW.

Table 5.1 Land uses in Dakar (extracted from Urbanism Masterplan towards 2035)
Full size table

3.2 Estimation of Waste-to-Energy Technical Potential

We assess the technical potential of waste-to-energy considering the calorific value of the city waste fractions recoverable as energy. These fractions are organic, plastics, paper, textile and wood waste components. The net calorific value is the approximate net calorific value of common municipal solid waste fractions provided by the World Energy Council (2016), quoting the International Solid Waste Association (ISWA). However, the previous document does not provide calorific value information on the wood component of municipal waste. Therefore, we assume the most common wood species in Dakar, the beefwood tree (Casuarina esquisetifolia), as a source of the wood waste. Information on the calorific value of the beefwood tree (20.9 MJ/kg) is from the World Agroforestry Centre database accessible online. Table 5.2 shows the technical potential of the city waste fractions recoverable as energy.

Table 5.2 Waste-to-energy potential of Dakar
Full size table

The technical waste-to-energy potential of Dakar in the baseline year is 39.6 MW power that can be converted with anaerobic digestion (organics) and Dendro Liquid Energy (others) technologies.

3.3 Estimation of Wind Energy Technical Potential

Figure 5.2 downloaded from Global Wind Atlas displays the theoretical potential of wind energy for Dakar.

Fig. 5.2
figure 2

Dakar Theoretical wind energy potential. Source: Global Wind Atlas

Full size image

Legend: Report generated on 10 January 2020 from https://globalwindatlas.info/. The energy potential considering the Mean Power Density of the 10% windiest areas of the city is 320 W per m2. The surface area of the city is 547 km2. Therefore, the theoretical potential of wind energy is 175,040 MW.

As opposed to solar photovoltaic, wind technologies cannot be integrated into the built environment of the city. Wind technologies are ideally installed in peripheral areas of a city to minimize interactions with competing infrastructure such as communication infrastructure. Table 5.1 on soil occupations from the National Directorate in charge of Land Planning (DPU) indicates that about 3.4% of the city’s land is made of these nude areas that do not have any geographical or natural affectation (Direction de l'Urbanisme et de l'Architecture, 2016). From this, we can derive a surface area equivalent to 18.6 km2 that can host wind systems with a technical potential for wind energy equivalent to 5951.4 MW.

3.4 Estimation of the Economic Energy Potential

Table 3 lists the input parameters of the Excel model to calculate each technology long-run generation cost in the baseline year (2016) (Table 5.3).

Table 5.3 LRGC input parameters
Full size table

CAPEX is from IRENA (2019) for anaerobic digestion, solar photovoltaic and wind energy. For Dendro liquid energy, CAPEX is from Ghougassian (2012).

Full load hours are derived for each technology from the capacity factor provided by IRENA (2019).

Capital recovery factor (CRF) is computed with a discount rate at 9% (Bah, 2015) and an assumption of 15 years lifetime. CRF is 12.4% for all technologies.

Fuel cost at 0 for all renewable energy technologies is an assumption of the study.

The efficiency factor (EF) is the efficiency of the conversion process. Figures are from Fraunhofer ISE (2019) for solar photovoltaic, and Fraunhofer IWES (Fraunhofer IWES, 2014) for wind energy. For DLE technology, EF is from Ghougassian (2012). For anaerobic digestion, EF is computed with data provided by Wendy et al. (2013), namely:

  • Biogas production: 303.6 m3 per ton of municipal solid waste.

  • Conversion factor biogas to electricity (in MWh per m3) is 0.0024.

  • Energy content of the organic waste is 1.11 kWh per kg of waste (4 MJ per kg) (World Energy Council, 2016).

Therefore, the efficiency of the anaerobic digestion energy recovery is 65.6%. This rate meets the R1 formula established by the European Commission Waste Directive, which provides that only thermal waste-to-energy conversion with efficiency equal to or higher than 0.65 can be regarded as an energy recovery operation.

Other O&M are from IRENA (2019) for anaerobic digestion, solar photovoltaic and wind energy. For DLE, the other O&M costs are from Ghougassian (2012).

Carbon prices (CO2 price) available in the literature varies widely per country, and inside the country per sector/project, and per partner buyer. Theretofore, we considered the mean value of the recommendations of the High-Level Commission on Carbon Pricing (World Bank Group, 2019) cited by the United Nations Framework Convention on Climate Change, which is $40–$80 per metric tonne by 2020. Therefore, the carbon price in the study is EUR 54 (60 USD).

CO2 specific emission factor of the conversion process is estimated at 0 for solar photovoltaic and wind energy. For municipal solid waste-to-energy, the Intergovernmental Panel on Climate Change (2006) estimates that 1 GJ energy conversion of municipal solid waste produces on average:

  • 100 kg CO2 emissions, which are equivalent to 0.36 kg per kWh

  • 30 kg CH4 emissions, which are equivalent to 0.107 kg per kWh. Considering a global warming potential (GWP) of 25, the CO2 equivalent emissions (CO2-eq) are 2.7 kg per kWh

  • 4 kg N2O emissions, which are equivalent to 0.014 kg per kWh. Considering a global warming potential of 298, the CO2 equivalent emissions are 4.29 kg per kWh

Therefore, carbon emissions per kWh generation are estimated at 7.35 kg CO2-eq.

Conversion rate USD to EUR is 0.9 that was the average rate in 2016 (www.statista.com consulted 7 January 2020).

4 Discussion of Results

4.1 Solar Photovoltaic Potential

The long-run generation cost of solar photovoltaic is EUR cents 16.1 per kWh. This cost is lower than the electricity tariff applied to households (EUR 18.3 cents per kWh), but is higher than the average electricity generation cost (EUR 7 cents per kWh) in the baseline year. Therefore, we confirm an economic potential for decentralized systems installed by citizens who would bear less cost than when using electricity from grid connection. However, at utility scale solar PV is economically not viable in the baseline conditions.

The sensitivity analysis considers the impact on the long-run generation cost of the capital cost (CAPEX) and the discount rate. The unit capital cost of solar photovoltaic can vary with capacity installed due to economies of scale. Large-scale (e.g. utility scale) power plants usually have lower costs than distributed systems. We analyze the sensitivity of LRGC to capital cost in two scenarios and the sensitivity of LRGC to discount rate in two other scenarios.

  1. (a)

    Scenario 1.1: Capital cost is at the level to reach grid parity, all other parameters remaining unchanged. This scenario can also be considered one of utility scale power plants currently in operation, which had a generation cost of around EUR 7 cents in 2016 (SENELEC, 2017).

  2. (b)

    Scenario 1.2: Capital cost is at the level to reach grid parity. Discount rate is set at 7.5% (OECD average), and all other parameters remain unchanged.

  3. (c)

    Scenario 2.1: Capital cost is at the level to reach consumer neutrality (i.e. cost = end-user tariff), all other parameters remaining unchanged.

  4. (d)

    Scenario 2.2: Capital cost is at the level to reach consumer neutrality. Discount rate is set at 7.5% (OECD average), and all other parameters remain unchanged.

Figure 5.3 shows solar photovoltaic capital cost related to scenarios 1 and 2 in comparison to the baseline scenario.

Fig. 5.3
figure 3

Solar photovoltaic scenarios comparison

Full size image

In Scenario 1.1, the capital cost should decrease by 68% to reach grid parity, which is equivalent to a capital cost of EUR 465 per kW installed. In Scenario 1.2, the capital cost should decrease by 65% to reach grid parity with a discount rate of 7.5%, which is equivalent to a capital cost of EUR 509 per kW installed.

In Scenario 2.1, the cost should increase by 13% to reach consumer neutrality over decentralized solar photovoltaic and grid generation, which is equivalent to EUR 1642 per kW installed. In Scenario 2.2, the unit cost should increase by 24% to reach consumer neutrality over decentralized solar photovoltaic and grid generation, which is equivalent to EUR 1798 per kW installed.

4.2 Waste-to-Energy Potential

The cost of organic waste conversion to electricity with anaerobic digestion technology (EUR 3.6 cents per kWh) is lower than Dakar’s average generation cost in the study reference year. The cost of other calorific waste materials (plastics, paper, textiles and wood clogs) conversion to electricity with Dendro Liquid Energy technology (EUR 1.9 cents per kWh) is lower than Dakar’s average generation cost in the reference year. Therefore, the installation of a 39.6 MW waste to electricity conversion system is economically viable in Dakar. Fig. 5.4 shows the bioenergy generation costs, in comparison to the kWh electricity generation cost, with the grid generation cost in the baseline year as a benchmark.

Fig. 5.4
figure 4

Waste-to-energy costs comparison (EUR cents per kWh)

Full size image

Compared to the city of Vienna, where similar estimates were completed, discount rate and the quantities of waste collected are parameters that are notably different.

In 2016, Dakar was home to a population of 3.4 million habitants (ANSD, 2020). Fall (2017) estimates the waste generated in Dakar at 171.82 kg per capita per year, which is equivalent to 589,234 tonnes per year. Fig. 5.5a displays a comparison of the energy demand of Dakar and the potential of energy produced by waste at three waste collection quantities, which are:

  1. (a)

    Scenario 1.1: quantities equivalent to the waste collected in 2016.

  2. (b)

    Scenario 1.2: quantities equivalent to the total of waste generated (based on Fall estimates), with equivalent proportions of organics and other waste materials.

  3. (c)

    Scenario 1.3: waste quantities equivalent to that of the city of Vienna (Austrian Federal Ministry of Sustainability and Tourism, 2017), keeping the same materials proportions.

Fig. 5.5
figure 5

Sensitivity of energy potential (GWh) to waste quantities (tonnes) and capital cost (EUR). (a) Simulation of quantities (b) Simulation of costs

Full size image

Figure 5b displays a comparison of the energy cost in the baseline year (discount rate = 9%) and the DLE cost at two levels of discount rate, which are:

  1. (a)

    Scenario 2.1: discount rate is equivalent to 7.5%, which is an average for OECD cities that include Vienna (IRENA, 2019).

  2. (b)

    Scenario 2.2: discount rate is equivalent to 9% as in the baseline.

  3. (c)

    Scenario 2.3: Discount rate is equivalent to 14% (fictive).

Scenarios 1.1 and 1.2 (quantities) provide Dakar with the potential to partially cover its energy demand in the baseline year, through recycling its waste, with proportions equivalent to 14% and 68%, respectively. Scenario 1.3 provides Dakar with the potential to cover the overall city electricity demand in 2016. The city’s total electricity demand was 1673.6 GWh (ANSD, 2019).

In scenarios 2.1 to 2.3 (capital cost discount rate), the generation cost remains below the generation cost and the tariff in the baseline year. The generation cost (EUR 7 cents per kWh) is relatively high, compared to the costs at different discount rates. The discount rate should be 85% for DLE to reach grid parity, which is unlikely.

Another parameter that is of interest in the waste-to-energy scenarios is CO2 price. CO2 pricing assumes a voluntary commitment of Dakar to reduce its carbon emissions as a component of its nationally determined contributions (NDC); Senegal being a non-Annex 1 party to the United Nations Framework Convention on Climate Change (UNFCCC, 2020). We assume the fourth scenario with voluntary carbon pricing that returns a long-run generation cost of:

  • EUR 64.1 cents for anaerobic digestion technology

  • EUR 51.5 cents per kWh for Dendro liquid energy technology

Therefore, prices for carbon emissions make the waste-to-energy technology cost seven to nine times higher than the average cost of electricity generation, which makes waste-to-energy technology too expensive and no longer a competitive option for Dakar:

4.3 Wind Energy Potential

The long-run generation cost of wind energy is EUR 8.3 cents per kWh. This cost is slightly higher than that of Dakar’s generation cost for the baseline year. In addition, we assumed that wind energy technology could not be integrated into the built environment, which excludes comparison with end-user tariffs for decentralized generation. Therefore, in the baseline conditions, wind technology is not economically viable. In addition, when we consider the cost as exclusion criteria to compare renewable energy technology with competing energy technologies, the result favours existent generation options (mainly diesel units), as they feature economies of scale accounted in terms of experience on technology use.

The study simulated one sensitivity scenario in wind energy, which sets the discount rate at 7.5%, equivalent to Vienna’s figure. The analysis returns that at this rate, the wind cost is EUR 7.7 cents per kWh, which is still higher than the grid generation cost at the baseline. The discount rate should be 5.7% to reach grid parity.

5 Conclusion

This chapter assessed the technical and economic potential of three renewable energy resources available in Dakar, going beyond the geographic information system (GIS) data. The main lesson derived from this assessment is that at local level context-specific parameters like space, technology costs and capital costs matter in making a renewable energy resource economically viable or not. When considering Dakar’s space restrictions and the presence of competing (conventional) energy resources, we found that waste-to-energy is the most economical alternative to grid generation considering the reference energy scenario. The technology is readily available, and the city has experience with recycling its waste-to-energy with anaerobic digestion. One example is the Thecogas waste-to-energy plant operational in the main slaughterhouse of Dakar. Dendro Liquid energy is another option for recycling the other calorific waste materials that the city generates in significant quantities (paper, plastics, textiles and wood clogs). Solar photovoltaic and wind energy technologies are also supplying options for the city, as the resources are available and technically usable. However, technology cost and cost of capital (discount rate) should significantly decrease in order for these technologies to be competitive compared to supply from the grid. Still, solar PV can be an alternative to grid, but only at the level of the end-user where tariffs are higher. Wind onshore faces both limitations of space restriction and lack of economic competitiveness, compared to competing resources.