Energy–water and GHG nexus: A South African water industry case

In South Africa (SA), water and electricity both play a vital role in the country's path to meeting its development agenda and correcting injustices of the past. The importance of these resources thus cannot be overemphasized. Water is vital for the agricultural sector, manufacturing/industries sector, households and power generating utilities. The energy that is required for water distribution from its sources to treatment plants and/to the intended consumers is mostly in the form of electricity (Schoeman et al. 2016).

In 2008 and 2014, SA experienced electricity shortages, which led to domestic and industrial power cuts, resulting in the economy being negatively affected. This makes this review of the SA water industry's energy efficiency essential, to ensure sustainability in the water industry.

The objective of this study is to review the SA energy–water–GHG nexus and compare it to international best practice. The use of international best practises assisted in the construction of a decision support model and identification of potential water–energy and GHG saving initiatives in the SA water industry, and scenario-based cased studies are developed to test this initiative.

Electricity is a critical input for delivering municipal water and wastewater services. In SA, coal power stations use wet cooling systems in the production of electricity and as a result more energy is used in the country by water utilities for pumping. Therefore as the demand for energy increases, this has a direct negative effect on the quality of SA water resources. This interdependence between water and energy is called the energy–water nexus and it is important for human existence, being the basic pillar for economic, technological and agricultural development (Balestieri 2013). According to Xu et al. (2014), water supply accounts for about 2–3% of worldwide electricity consumption and of this value, 70% is consumed by water distribution networks. The basic function of water distribution systems (WDS) is to supply users with the required amount of water, delivered at an adequate pressure for domestic and industrial consumption (Mays 2000; Balestieri 2013). Most WDS are characterised by three major components, namely: distribution piping, distribution storage/reservoirs and pumping stations (Mays 2000). These components can further be classified into the subcomponents: pipes, valves, pumps, drivers, power transmission units, controls and storage tanks. Distribution pipes are the most common component in the WDS and contains fittings such as pipes, valves and flow measurements (Mays 2000). The pipe element of the distribution system always leads to the storage system, which is closely associated with tanks. The water tanks are used to supply water during high system demands or when pumps cannot meet the pressure requirements, ensuring that system demands are met at all times (Pabi 2013).

Water distribution networks require an energy boost to overcome heights and friction in the pipes, for example in pump stations, which are generally associated with electrical consumption. In water distribution networks, the largest part of the operating cost is energy used for pumping (service pumps and booster pumps), which provides the required discharge and pressure at varying demand points (Berry 2007). Part of the energy (electrical/hydraulic) is lost due to the low efficiency of equipment (pumps, valves, and pipes), bad operational practices, inadequate designs and the presence of redundant structures (leakages in pipes and cracks in tanks/reservoirs) (Copeland 2017).

Research has shown that significant energy savings can be achieved through water efficiency initiatives (Balestieri 2013). These water efficiency initiatives also contribute to the reduction of carbon footprints, which otherwise would have been generated by the processes of treating and distributing water (D'Ercole et al. 2016). Table 1 below sets out best practise and energy efficiency measures, that can be applied in water networks.

In pump operation, the desirable operating condition is for pumps to be operating close to the maximum efficiency point, called the Best Efficiency Point (BEP) (Pabi 2013). BEP is the intersection point of the water system's resistance curve and the pump's head–flowrate (H–Q) curve. However in a WDS, the demand changes daily and seasonally, resulting in changes in the BEP. In order to handle these varying conditions in WDS, variable speed drives (VSDs) are the most suitable energy efficiency initiative based on their operation principle.

Balestieri (2013) states that VSDs are an energy-efficient technology for controlling pump flows, which are dependent on the interaction of the pump characteristic curve and process system curve for effective application. VSDs efficiently control pump flows by varying the pumps' rotating speed. This is achieved by using an electronic controller that adjusts the frequency of power supplied to a motor in order to change the motor's rotational speed and meet operational load. Pumps in a water network are designed to meet the head and flow at peak demand hours, which means that in off-peak hours the same pump will apply excess pressure to the network, resulting in a higher energy consumption than required (Babaei 2015). VSDs are able to align the pump output flow/pressure directly to the process requirements, which improves the process performance. Thus, the use of VSDs on motor pumps, which rotate at a lower speed at off-peak hours, are recommended as one method for water networks to save energy (Babaei 2015).

Leakage is an essential element in the supply-demand balance for water utilities. The primary components of leakage management are (CIWEM, 2015):

This involves the reduction of excess pressure and volume of water lost through leaks. Stabilizing the pressure in water networks also helps to reduce the frequency of pipe bursts (CIWEM, 2015).

This is the process of proactively looking for unreported leaks and burst pipes. According to the European Commission (2015) ALC consists of two stages, namely leak monitoring & localisation, and leak location & pin pointing.

The simplest and most cost effective way of repairing leakages is by repairing the leaks that have been found through ALC and those that have been reported to the company promptly & effectively (CIWEM, 2015).

This is the process of investing in facilities and asset renewal in order to reduce the occurance of new leaks.

Table 2 sets out examples of leak management initiatives in Europe and the energy savings that they realised.

Leakage rates for all municipalities in SA are given in Table 3, where it is illustrated that rural areas are the most challenging in terms of water leakages.

Schoeman et al. (2016) states that the Department of Water Affairs and Sanitation (DWAS) is currently faced with old infrastructure breakdown or failures, as well as a lack of proper maintenance of infrastructure. Research has shown that only 53% of municipalities in SA have adequate maintenance capacity (Johannesburg Water SOC (Ltd) 2015).

The South African Institute of Civil Engineering (SAICE 2011) estimates that the DWAS requires about R139 billion for infrastructure replacement. Although the necessary technologies and methodologies to solve water loss problems are known and available, the cost of implementation inhibits problem mitigation (Mays 2000). With the rising cost of electricity and supply shortage in the country, energy saving has become more important (Breytenbatch et al. 2016).

In addition to the energy efficiency initiatives in water distribution networks, there is a need to view water as an energy carrier. Municipal wastewater is a potential source of chemical energy, which has organic carbon that can be recovered as biogas in sludge digestion. Wastewater also contains organic energy from organic pollutants discharged in sewer systems. Part of this organic energy is recovered in the form of biogas from sludge digestion in wastewater treatment plants (WWTPs) (Frijns, 2013). Table 4 sets out examples of WWTPs around the world that generate useful energy.

This section presents the energy efficiency initiatives/measures identified for municipal water supply systems (Gauteng Province (GP), SA). Three energy efficiency measures are presented, namely: use of VSDs, leak management and energy recovery from wastewater, in the City of Johannesburg (CoJ) municipality. Furthermore, a decision support system (WDS model supported by the MULTI-MOORA technique) is proposed, assisting the WDS industry to identify the best performance indicators as per specific requirements. Results, discussions and conclusions are also presented.

In SA, GP, KwaZulu-Natal (KZN) and Western Cape (WC) are the top three water consumers, per industry sector. GP, as the economic hub of SA, consumes the most water out of the above-mentioned provinces, mainly because it accommodates most of the country's water consuming activities (manufacturing, construction, financial services, real estate services & government services) and population, as illustrated in Table 5 below (Statistics South Africa 2014). In 2016 Gauteng contributed 33.8% to the South African GDP, therefore this province can be identified as the highest risk area in SA in terms of the potential lack of reliable water supply.

GP province has the highest population density and growth rate in SA, with an estimated population of 12.27 million. It is made up of three metropolitan municipalities (CoJ, Ekurhuleni & City of Tshwane) and three district municipalities (Westrand, Metsweding & Sedibeng). These municipalities accommodate major manufacturing/processing factories, financial & retail services, who consume great amounts of water and generate huge amounts of wastewater every year (Wegelin 2010).

Operationally the CoJ's water system is divided into seven subsystems (Johannesburg Metropolitan Municipality 2017): (A) Midrand/Ivory-Park/Diepsloot/KyaSands, (B) Randburg/Greenside/Northcliff, (C) Roodepoort/Constantia Kloof/Northgate, (D) Soweto/Arnadale, (E) Alexandra/Wynberg/Sandton, (F) Lenasia/Ennerdale/Orange-Farm, and (G) Lenasia/Eldorado Park/Orange-Farm.

Within the CoJ water network there are 89 water reservoirs, 10 depots, 28 water towers, 6 wastewater works, 31 water & 35 sewage pumping stations, 92,164 valves & hydrants and 12,581 km of water pipelines, supplying approximately 1,574 ML/day (Johannesburg Water SOC (Ltd), 2015).

CoJ draws most of its water from the Vaal River, with Olifants River and Crocodile River also used as water sources (Johannesburg Water SOC (Ltd) 2015).

The current pumping systems for all seven regions within CoJ are analysed and compared with an alternative pumping system which uses VSDs and leak management. The difference in energy usage, greenhouse emissions (CO₂, SO_X) and NO_X emissions in the systems is noted. The potential energy generation from WWTPs is also analysed within CoJ.

Detailed potential energy & CO₂, SO_X and NO_X emission savings that can be achieved in the WDS are presented. To conduct these calculations, Microsoft Excel is used, in which graphs are plotted highlighting the theoretical current system versus proposed system that are energy efficient. This analysis is based on literature on SA water systems and assumptions which are detailed in the case study section.

The limitations of this study are: the lack of operational information/data that relates to the CoJ water network; pump ratings and configurations, flowrates, pressure and head of the system and operating emissions from all 7 regions within CoJ. Therefore, for the purpose of highlighting important principles and energy efficient initiatives and decision support modelling, which have brought tremendous success in European countries and the USA, a scenario-based model is established.

The approach is to present models of the results as scenarios based on the individual and then collective impacts of the best practice options. The results of this study are separated into functional specific and scenario options, illustrating the potential benefits of implementing the options presented via the literature study above.

The first energy efficient initiative to be analysed is the use of VSDs in the CoJ water network. All 7 regions within the CoJ are considered in order to highlight the potential energy savings that can be achieved when using VSDs as compared to using a constant speed pump with varying water demand. The following assumptions are made for the purpose of the calculations:

• Currently all the regions in CoJ use constant speed pumps;

• The pumps operate 24 hours a day and 365 days a year (8,760 hours).

Figure 1 below represents the relationship between water demand in SA municipalities and time, over a period of 24 hours. The water demand profile in Figure 1 is assumed to be the same for all regions in the CoJ. From Figure 1, the peak hours in SA municipalities is between 07:00–10:00 hrs and 18:00–20:00 hrs (which is the period that requires maximum pumping capacity i.e. 100% and 90% respectively, water flow volume when using VSDs). For the rest of the time, the demand is moderately distributed requiring low to medium high pumping capacity (50–80% water flow volume when using VSDs).

A VSD motor pump is ideal for this type of system where the demand for water varies depending on the time of the day. VSDs efficiently control flow by varying the pumps' rotational speed, matching it to the operational load. For the CoJ the pumps' rotational speed ranges from 50% to 100%. According to the Carbon Trust 2013, VDSs cannot reduce the flow to 0% flow. This is attributed to the reduction of cooling capacity.

Table 6 below illustrates the different pump rotational speed (% of full system flow) and the amount of time at each pump rotational speed setting as per Figure 1. From Table 6 below, it is shown that the pump rotational speed operates between 90–100% flow at peak hours i.e. 5–10am and 17:00–20:00 hrs.

Figure 2 below illustrates a typical loading control configuration for different pump controls (% of full system flow vs % system flow power). For this study, the curve of interest is the VSD curve. From Figure 2, the % of power utilized by a VSD of the original full load is obtained for each % flow setting, i.e. when the VSD motor pump is operating at 50%, the pump uses only 20% of the full load power. Therefore, if a constant speed pump is used between 0:00 hrs and 05:00am, 80% of the energy used would be going to waste as only 20% is needed to pump the same volume of water based on the water demand.

m_SOX is the potential mass of SO_X (kg) that can be saved per region, EF_SOX is the SO_X factor (kg/KWh) given as 0.00869, m_NOX is the potential mass of NO_X (kg) that can be saved per region, EF_NOX is the NO_X factor (kg/KWh) given as 0.004, m_CO2 is the potential mass of CO₂ (kg) that can be saved per year and EF_CO2 is the CO₂ factor (kg/KWh) given as 1.008.

The energy savings in Table 8 above is represented graphically by Figure 3 below. It illustrates the current system of 15 KW and the proposed VSD pumping system. The potential energy saving per 24 hour cycle is the area between the two graphs.

From Figure 3 above the greatest energy saving from all regions is when the pumps' rotational speed is between 50–80% of the maximum flow. This corresponds to low water demand periods i.e. from 20:00 hrs–05:00am and 12:00pm–17:00 hrs. The potential energy savings are considerable at higher flow rates i.e. 90–100% flow, when using a VSD.

Tables 9 and 10 below illustrates the potential energy, CO₂, NO_X and SO_X savings for replacing the current constant pumping system with 30 KW and 45 KW pumping systems with VSDs for all the regions (A, B, C, D, E, F & G). From Tables 9 and 10 below, KWh is the potential energy per annum (8,760 hours of operation), CO₂ is the potential carbon dioxide savings in kg per annum, SO_X is potential SO_X savings in kg per annum and NO_X is the potential NO_X savings in kg per annum.

From Tables 7–9, there is high potential of energy savings when using VSDs, as compared to a system with constant speed, with respect to varying water demand. It can be observed that the higher the pump ratings the more potential there is for energy, CO₂, NO_X and SO_X savings i.e. 45 KW pump has the greatest energy potential savings. These savings are as a result of the pumps running at a speed depending on the reservoir tank levels and water demand. That means that the pumps in all regions run slower for much of the day as seen from Figure 1 (as peak demand is only 8 hours per day). And this could significantly contribute to the city's efforts towards adhering to the Paris Agreement of 2015 regarding greenhouse gas emissions.

Based on the above stated assumptions, the CoJ water utility has the potential to save a total of: 2,527,334 KWh of energy, 2,603,154 kg of CO₂ emissions, 10,109 kg of NO_X emissions, and 21,963 kg of SO_X emissions per annum.

From literature, the biggest contributor to non-revenue water in SA is leakages, with the GP losing about 27% of the water that it receives from Randwater per year (CSIR/CIDB 2006). This provides an area to explore in water distribution networks, as countries like the United States of America and the United Kingdom have an average leakage rate of 20% (Walker 2014). The following section discusses the energy associated with leakages in WDS in the CoJ. Furthermore, this section highlights the associated energy savings that can be achieved in the CoJ WDS. The following assumption are made:

• a.

  The total daily flowrate of 1,574 ML/d (City of Johannesburg Metropolitan Municipality, 2017)

• b.

  The leakage rate is 27% in all regions in the CoJ (EPA 2012)

• c.

  The flow distribution is the same for all regions

From Table 11, 97,667ML/y of water can be saved within the CoJ water network.

Using Equations (5)–(7), the amount of potential gas savings is: 21,164 of CO₂, 179 kg of SO_X and 86 kg of NO_X.

The third energy efficient initiative to be analysed is the recovery of energy from WWTPs in the CoJ. The CoJ has six WWTPs, ranging from medium size to macro size, which will be considered in this section, as illustrated below in Table 13. The following assumptions are made:

• The wastewater arriving at the WWTPs includes both domestic (black water & grey water) and industrial wastewater (City of Johannesburg Metropolitan Municipality 2017);

• Biogas has the following composition: CH4 (65%), CO2 (30%) and H2S (5%) (Chua 2013);

• Only 0.68% of the flowrate of the wastewater enters the digester as sludge (Chua 2013);

• Biogas production of 0.0153 m³/L sludge;

• 365 days operation;

• 1 KWh = 3,600,000 J (3.6 MJ);

• 1 m³ of CH₄ = 36 MJ;

• 36 MJ = 10 KWh; and

• 1 m³ of CH₄ = 10 KWh.

The total energy that can be produced from the six WWTPs in the CoJ is 300,259 KWh, with an associated CO₂ emission savings of 309,267 kg, NO_X emission savings of 1,201 kg and SO_X emission savings of 2,609 kg, under the assumption stated above. The energy potential is heavily dependent on the size of the water treatment plant, as seen in Table 14 and 15 below. Table 14 has four large WWTPs at a flowrate of 132 ML/d and 2 medium WWTPs at a flowrate of 60 ML/d, the potential energy savings is 438,216 KWh.

While Table 15 below has 3 small WWTPs with a flowrate of 30 ML/d and three medium WWTPs at a flowrate of 60 ML/d, has an energy potential of 182,590 KWh.

From Tables 13–15, it is illustrated that the there is more energy potential in a network that has a large WWTP/high wastewater flowrate (Q).

This section proposes a decision support system for evaluating the performance indicators based on impact ratios. Using this, a best chain of metrics corresponding to any performance indicator can be chosen by a water utility for attaining nexus goals.

Fuzzy logic deals with the information by allowing a partial set beside a crisp set (Sahu et al. 2015, 2016). Fuzzy logic usually tackles the true value in a range between entirely true and false. Fuzzy logic finds its applications in assisting decision making.

If are the set of professionals in the group decision making procedure, are the set of choices/performance indicators at level L₁, at level L₂ and at level L₃ are the set of metrics (Sahu et al. 2015).

Suppose that is the fuzzy value of metrics, assigned by k_K, where is a Triangular Fuzzy Number set (TFNs) against (set of choices/performance indicators) corresponding to infinite metrics of

In above equation, denoted the submission of priority rating of metrics (depended input) at level L₃, under an individual measures (inputs) c_j at level L₂.

The Multi-Objective Optimization by Ratio Analysis (MOORA) technique is proposed by Brauers & Ginevicius 2009. The technique is extended using reference points and full multiplicative form to design a robust approach, called MULTI-MOORA.

Here ⁠, indicates the product of measures c_j at level L₂ against choices/performance indicator at level L₁ to be maximized with ⁠, being the number of indicators to be maximized.

indicates the product of measures c_j at level L₂ against choices/performance indicator at level L₁ to be minimized with ⁠, being the number of indicators to be minimized.

This is an empirical case study of a South African water distribution company (A), located in Johannesburg, supplying water to Gauteng's manufacturing, construction, cement, power generation, etc. sectors. This company executed the proposed multi-layered structural WDS model resulting from the literature survey. For the purpose of the model the L₁ indicators (outputs) together with the L₂ indicators (inputs) are all detailed in Appendix A, Table S1. The proposed model was established by inserting 29 metrics at L₃ level, which align with internal and external measures at L₂ level, linked with three performance indicators at level L_0. This model aides the South African water distribution company to identify the most influential or significant performance indicators and their chains (metrics) to maximise savings of energy–water and reduce emissions of GHG. (The nexus concept states that saving water means saving energy and minimizing emission of GHG). The company analyzed the results by computing the impact of three core performance indicators using fuzzy priority ratings against individual L₃ metrics assigned by company experts. After evaluating the results, the company could potentially focus on performance indicators and t-linked metrics to achieve nexus goals. The study steps are depicted below:

Firstly, the authors have undertaken a literature survey in the field of WDSs metrics and measures and identified some metrics and their internal and external measures, which aligned with the performance indicators to create the multi-layered structural WDS model shown in Table 18. The metrics definitions are shown in Appendix A, Table S2.

In order to undertake the study, a team from South African water distribution company (A) was created, including staff members with expertise in water treatment, pump drivers and sanitation, maintenance, WDS networking, security, etc. from different departments. Finally, a team of five professionals based on their experience and high qualification was finalized.

Next, the team, assisted by linguistic variables shown in Appendix A, Table S3, to validate the vagueness associated with L₃ metrics. The team was requested to assign the linguistic variables (priority rating) against L₃ metrics, shown in Appendix A, Table S4. Later, all are transformed into TMFs and aggregated by using the average rule Equation (15). Subsequently, all are defuzzified by using Equation (16).

After defuzzification, Equation (17) was applied to move from level L₃ to L₂, and then MULTI-MOORA was applied to aid decision-making. The steps are given below:

Equation (18) was used to evaluate the overall impact of nexus performance indicators; the results are detailed in Appendix A, Table S5, where the C₂ performance indicator was found to be the most influential (C₂=34.21%,) than both C₁ (32.11%) and C₃ (33.67%) out of 100%.

Equation (19) was used to compute the overall impact of nexus performance indicators; the consequences being depicted in Appendix A, Table S5, where the C₂ performance indicator is shown to be more influential (C₂=35.10%) than both C₁ (31.01%) and C₃ (33.89%) out of 100%.

Equation (20) was used to calculate the overall impact of nexus performance indicators, with the results listed in Appendix A, Table S5, where again the C₂ performance indicator performed better (C₂=38.61%) than both C₁ (30.62%) and C₃ (33.67%) out of 100%.

After undertaking the comparative analysis, the South African utility was advised to implement the chain of metrics corresponding to the C₂ PI in order to achieve their nexus goals.

Energy usage in the CoJ water supply system can be optimized by reducing the energy usage in pumps, reducing water leakages in distribution systems and co-generation of energy from WWTPs. The present amount (scenario 0) of water that is lost to leakage is 456 ML/d, which is associated with 2,164 KWh of energy (for treatment and distribution) and 21,429 kg of gas emissions (CO₂, SO_X, NO_X). In the present case, three scenarios are identified:

• 1.

  Use of VSD and leak management

• 2.

  Leak management and co-generation of electricity

• 3.

  Use of VSD, leak management and co-generation of electricity

• 4.

  Use of a decision support model for attaining nexus goals

The scenarios are compared in terms of performance indicators aimed at energy consumption reduction and C0₂ emission reduction.

In scenario 1, VSDs are installed in all seven regions of the CoJ: Region A (25 KW), Region B (25 KW), Region C (25 KW), Region D (200 KW), Region E (100 KW), Region F (100 KW) and Region G (15 KW). The CoJ leakage rate is reduced from 29% to 17%.

The introduction of VSDs and leakage management measures reduces the total energy consumption to 54% and the amount of CO₂ emissions to 54%, and 97,667 ML/y of water is saved per annum. This scenario is characterised by a sharp decrease in both the energy consumption and CO₂ production (Figures 4 and 5).

In scenario 2, leak management and co-generation of electricity in the CoJ water network system were evaluated. These two initiatives have the potential to reduce the energy consumption by 15% and CO₂ emission by 16.5%. The small impact of energy change in scenario 2 is because the leakage rate in scenario 0 is 29% and scenario 2 is only proposing a 12% decrease in leakage rate to fit international best practice of 17%. Scenario 2 performs well in reduction of CO₂, NO_X and SO_X emissions.

In scenario 3 the introduction of VSDs, co-generation of electricity and the reduction of leakage rate from 29% to 17%. This scenario has a significant affect on energy and CO₂ emission reduction, i.e. 60% and 60.5% respectively. Therefore scenario 3 is considered to be the most promising scenario in terms of offering energy savings and CO₂ emission reduction, as illustrated in Figure 6 below.

Eventually, in scenario 4, a decision support system was developed, consisting of a conceptualized WDS model simulated by the MULTI-MOORA technique. This scenario is important for saving water–energy and CO₂ emission reduction. CoJ's water utility was advised to implement the suite of metrics corresponding to the C₂ PI in order to achieve their nexus goals. In real life, the WDS plant can use its own professional team to identify the metrics and linked PIs to achieve their nexus goals, using the simulation technique to support their decision-making. The model is flexible in nature, so utilities can also change and manipulate the metrics and PIs as per industrial practices.

A summary of scenarios 1, 2 & 3 is given in Table 16 below.

The 3 scenarios deal with three nexus strategies, i.e. leak management, use of VSDs, and energy production from WWTPs, all aiming to make WDS operation more energy efficient. The WDS operators can attain green standards by minimizing water loss and energy use, as well as CO₂, NO_X and SO_X emissions (nexus goals), using any of the three nexus strategies.

Scenario 4 proposed decision support system is relevant for WDS operators in identifying the most effective metrics/practices based on impact ratios. The model also serves to improve the overall water supply performance index.

The current study reviewes the energy–water nexus in a SA WDS and a comprehensive technical analysisis is presented to analyse possible energy consumption and CO₂ emissions reduction for the CoJ water network. The CoJ has a population of 4.94 million people and is the economic hub of SA.

Four scenarios are considered, where scenarios 1, 2 & 3 involve energy consumption reduction, co-generation of energy from WWTPs and water loss (leakage) reduction. Scenario 4 dealt with a flexible decision support system. The present situation (scenario 0) is characterised by being an inefficient system in which leakage is above the internationally recognised best practice rate of 17%.

The major contribution of scenarios 1-3 is that a combination of energy co-generation, leakage reduction and use of VSDs is proposed in order to reduce energy consumption by 60% and CO₂ by 60.5%. In general the method of analysis used above is shown to be an important tool for understanding the different interactions between energy and water to achieve balance optimization. However the use of such a single objective aimed at reduction of energy usage/reduction of CO₂ emissions is very limited as it does not give the complete picture of the complex interdependences between a WDS, energy consumption and water leakage.

Furthermore, in scenario 4, the WDS operator of CoJ is advised to implement a suite of metrics corresponding to the C₂ PI in order to achieve the nexus goal, with results shown by bar chart, in Figure 6.