Rural Electrification Pathways: An Implementation of LEAP and GIS Tools in Mali

Remote rural populations do not often have the luxury of viable multisource electricity generation systems. Considering fossil fuels for remote populated areas is not often a viable option due to the fuel transportation costs and the population’s socioeconomic status. Extending the grid is often economically prohibitive. This paper proposes possible ways in which Mali could increase the rate of population with access to electricity by 2050 using Low Emission Analysis Platform (LEAP) and geographical information tools. The current energy situation is assessed, and multiple demand and supply scenarios are created to find the most viable option in environmental and economic dimensions. A minimum of 50% reduction of biomass consumption in the residential sector and a maximum of 71% was achieved through the combination of grid extension and decentralized solar PV. Solar PV becomes the preferable option when enough time for the effects of electricity on income is given. When these effects are not present, solar PV is still a better option, as the amount of biomass replaced with electricity is reduced.


Introduction
As of 2019, 44% of the global population still lives in rural areas [1]. Even though globally, 82% of the rural population has access to electricity [2] in Sub-Saharan Africa, the access rate barely reaches 31% [3].
The impact electricity access can have on rural populations can be gleaned through the indicators used on multi-criteria decision-making studies [4,5], as most include among their dimensions economic, social, and environmental aspects. Among the indicators themselves, the share of the population with school education, electricity used for income-generating activities, indoor/outdoor pollution, access to entertainment and information, etc., speak about betterment in lifestyle for communities with access to electricity. Other aspects that electricity access has on rural populations also include the betterment of gender inequality and migration patterns [6].
Remote rural populations do not often have the luxury of viable multisource electricity generation systems, even though they are sometimes the most viable option for specific case scenarios [7]. In Sub-Saharan Africa, particularly in Mali, resource assessment points to either the use of renewables such as solar energy, hydropower where available [8], or fossil fuels. Hydropower is limited by the seasonality of Sub-Saharan climate, commonly characterized by rainy and dry seasons, and by location, as with no countrywide electricity grid, the generation remains local. Reaching remote populated areas with modern energy is not a simple task [9]. Recent studies highlight the need for increased research on institutional barriers, frameworks, and incentives to rural electrification [10]. The advantages of using decentralized renewable energy plants versus fossil-fueled generators have been to a report published by the United Nations Development Programme (UNDP) and the World Bank [39], was USD 4400 per km without accounting for labor and transportation; in 2014, the cost of grid extension including labor, materials, and other costs was of USD 19,070 per km [40]. In 2019, the cost of extending the grid ranged from USD 30,798 to 35,282 per km [41]. The increase in the cost of grid extension has yet to show saturation and might continue to grow, which would be expected when considering progressive growth in labor costs.
Energies 2021, 14, x FOR PEER REVIEW 3 of 20 poles, transportation of material, and fuel less cost-efficient. The cost of extending the grid in Mali has been increasing through the years. In 2000, the cost of grid extension, according to a report published by the United Nations Development Programme (UNDP) and the World Bank [39], was USD 4400 per km without accounting for labor and transportation; in 2014, the cost of grid extension including labor, materials, and other costs was of USD 19,070 per km [40]. In 2019, the cost of extending the grid ranged from USD 30,798 to 35,282 per km [41]. The increase in the cost of grid extension has yet to show saturation and might continue to grow, which would be expected when considering progressive growth in labor costs.

Figure 1.
Population distribution and electricity grid of Mali. Using data from [42,43].
Anaerobic digestion has been described as an excellent alternative to convert waste to energy to meet environmental and energetic targets [44]. It can provide energy when weather-dependent sources would not be able to [45], balancing a grid of renewable energy systems [46]. When enough energy from other renewables is available, storage, cogeneration, and other productive uses become possible such as co-generating electricity and cold storage to reduce agricultural losses [47]. Integration of biogas plants to renewable grids such as biomass, hydropower, solar photovoltaic (PV), etc., in the right amounts, can lead to vastly more viable costs than grid extension [48]. Locations such as Sub-Saharan Africa, and specifically Mali [49], in which solar radiation is abundant [50,51] could make use of solar PV supported by biomass [48]. In its spatial component, the present study takes only the solar portion and compares it to grid extension. This limitation is a result of the lack of availability of reliable countrywide data to include it within the calculations.
To argue that an increase in electricity consumption leads to an increase in income generation, the electricity available has to include productive uses. The multi-tiered framework proposed by [5], which is built considering access to the productive application of energy, is used as a base for scenario building throughout this paper.
Therefore, the objective that the present work tries to accomplish is to identify whether complete rural electrification is achievable in Mali and what are some of the requirements for this to occur. Two main steps are taken to achieve this. First, the current Anaerobic digestion has been described as an excellent alternative to convert waste to energy to meet environmental and energetic targets [44]. It can provide energy when weather-dependent sources would not be able to [45], balancing a grid of renewable energy systems [46]. When enough energy from other renewables is available, storage, co-generation, and other productive uses become possible such as co-generating electricity and cold storage to reduce agricultural losses [47]. Integration of biogas plants to renewable grids such as biomass, hydropower, solar photovoltaic (PV), etc., in the right amounts, can lead to vastly more viable costs than grid extension [48]. Locations such as Sub-Saharan Africa, and specifically Mali [49], in which solar radiation is abundant [50,51] could make use of solar PV supported by biomass [48]. In its spatial component, the present study takes only the solar portion and compares it to grid extension. This limitation is a result of the lack of availability of reliable countrywide data to include it within the calculations.
To argue that an increase in electricity consumption leads to an increase in income generation, the electricity available has to include productive uses. The multi-tiered framework proposed by [5], which is built considering access to the productive application of energy, is used as a base for scenario building throughout this paper.
Therefore, the objective that the present work tries to accomplish is to identify whether complete rural electrification is achievable in Mali and what are some of the requirements for this to occur. Two main steps are taken to achieve this. First, the current energy situation of Mali is modeled using LEAP to identify the present and provable future electricity needs. Then, tiered electrification scenarios are studied using both LEAP and GIS to consider the location and characteristics of the different populated areas of Mali, taking into account the effects of new or increased electricity access on income generation. The present paper is divided into material and methods, results, discussion, and conclusions.

Materials and Methods
This study utilizes two software packages to produce the results needed to accomplish the set objectives. LEAP, or Low Emission Analysis Platform, version 2020.1.0.20 Dictionary Version: 463 [52] and the open-source Geographic Information System QGIS 3.16 (64 bit) [53].
LEAP is a software widely used for energy modeling that matches supply and demand and can compare the interactions between different policies and scenarios [54]. Scenario analyses from 2016 to 2050 were performed using this software to ascertain what Mali will need in terms of energy and propose possible ways to supply it.
The key assumptions used to model the energy scenario of Mali were a population of 17,994,837 [36] with a population growth of 3%, an average household size of 6.3 [36], and an urban share of 40.78% [36] (The data used to model Mali's energy consumption and current supply was compiled from different sources including EDM-SA (Energie Du Mali SA) [55,56]; National Institute of Statistics INSAT (Institut National de la Statistique) [37]; System of energetic information SIE (Système d'Information Energetique) [38,55] [57].). The number of households was defined as the total population divided by the average household size. Urban and rural households were assumed to be represented by the number of total households multiplied by the urban and rural shares, respectively (e.g., the total population divided by average household size multiplied by the urban share equals the number of households in urban areas).
Certain assumptions were necessary to model the energy distribution between urban and rural sectors and between each of the electrified and non-electrified sub-sectors.
For biomass, the share of the total biomass consumed by urban and rural areas was found to be 50.36% consumed in rural areas and 49.63% consumed in urban areas [58]. These shares were multiplied by the total biomass consumption in 2016 (data provided by USTT-B). The share of electrified and non-electrified homes in each area was multiplied by the respective urban and rural biomass consumption to find the amounts used by electrified and non-electrified sub-sectors: TSS Bci is the total consumption of biomass per sub-sector, SS Bci is the sub-sector biomass consumption ratio (%), TRBc is the total residential biomass consumption, and SS% is the sub-sector ratio.
The electricity access rates (26% rural electricity access and 86% urban electricity access) were normalized into 23.21% of the total electricity consumed in the residential sector consumed by rural households and 76.79% by urban households. This value was then multiplied by the total electricity consumed in the residential sector: where NEa i is the normalized electricity access rate of each sub-sector (%) and SSec is the sub-sector electricity access rate. For petroleum, the usage of lamp oil was taken as representative of the petroleum products consumed in the residential sector. The region of Kayes was found to use 98% of the total lamp oil consumption in Mali and therefore was used as representative of the whole country. In Kayes, 73% of the population live in rural areas and 27% in urban areas [59]. The consumption was multiplied by the normalized urban and rural electrifica- tion rates of Mali (77% of the electricity consumed in the residential sector is consumed by the urban electrified homes and 23% by the rural electrified homes) to find the shares of lamp oil consumption in the electrified and non-electrified sectors of both rural and urban areas. These shares were then multiplied by the total amount of petroleum products consumed in the residential sector. Table 1 presents the values assumed representative of the consumption in Mali: SS PPc = NRes PPi × Residential petroleum product consumption (5) NRes PPi is the normalized residential petroleum product consumption rate per subsector (%), SS PPi is the petroleum products consumption of each sub-sector, and SS PPc is the total consumption of petroleum products in each sub-sector.
According to The World Bank [36], the industrial sector's growth (represented by the world development indicator: Industry, value added as % of GDP) has had a trend of 2.5% growth per year from 1967 to 2007, decreasing over the years between 2008 and 2016. The activity level of the industrial sector (percentage of the population working in the industrial sector) in Mali is set to grow in proportion to the total population [57]. The industrial sector grew from 0.24% of the total population in 2016 to 0.53% by 2020 and is further expected to grow to 1.02% by 2030. Both the commercial and the transportation sectors are assumed to follow the population growth of 3% per year.
Transmission and distribution losses are taken as 22% from Système D'information Energétique SIE 2018 [38]; this value remains unchanged throughout the calculation years.
The availability (maximum percentage of hours that the process is available in each dispatch period) of Hydropower was taken as 44.59%, of Thermal stations as 14.39%, and renewables (modeled as solar PV) as 25% [60]. The Process energy efficiency for the thermal stations was given the value of 37.71% [60]. Value assumed valid for Mali.
According to the KfW Development Bank report, the conversion efficiency of LPG stoves is about 60% [61], while the conversion efficiency of traditional firewood for cooking is about 15% [61]. This difference in efficiency was taken into account for biomass replacement scenarios. Electricity and LPG stoves are assumed to have the same conversion efficiency.
According to Fondation Energía sin Fronteras [62], the biomass production in Mali is mainly firewood (see Figure 2). Therefore, the biomass consumption of Mali has been modeled as solely firewood.
As Mali produces its firewood, the resource was considered non-finite for the calculation years.
LEAP facilitates the study of the interaction between different scenarios. This makes the effects of single measures and combinations of measures able to be studied. This paper takes advantage of this functionality and studies ten scenarios built by combining two base scenarios with different measures and targets (see Table 2).
NATa and NATb are the base scenarios presented in Table 3. COM represents complete rural electrification, and T1 to T5 represent different tiers of electrification, see Table 4. NxC_Tn, with x = (1, 2) and n ranging from 1 to 5, are the ten scenarios studied. Each of these ten indicators is built by combining the effects of the sub-scenarios a-c.
One of the current targets of Mali, as well as other sub-Saharan countries, is to reduce the consumption of biomass of the residential sector to help combat the encroaching desertification. In this sense, the scenarios NATa and NATb look to reduce the biomass consumption by replacing it with LPG in scenario NATa and with electricity in scenario NATb.
As Mali produces its firewood, the resource was considered non-finite for the calculation years.
LEAP facilitates the study of the interaction between different scenarios. This makes the effects of single measures and combinations of measures able to be studied. This paper takes advantage of this functionality and studies ten scenarios built by combining two base scenarios with different measures and targets (see Table 1). NATa and NATb are the base scenarios presented in Error! Reference source not found.. COM represents complete rural electrification, and T1 to T5 represent different tiers of electrification, see Error! Reference source not found.. NxC_Tn, with x = (1, 2) and n ranging from 1 to 5, are the ten scenarios studied. Each of these ten indicators is built by combining the effects of the sub-scenarios a-c.    The scenarios T1 to T5 are scenarios where different electrification tiers are used. These electrification tiers were defined by [5,63], where the minimum daily capacity per household for different tiers of electrification was defined (see Table 4). As the tiers of electrification consider the eventual use of electric appliances for cooking and water heating purposes, biomass consumption has been assumed reduce according to the tier of electrification.  [5]. See [5,63] for more in-depth information.
Currently, Mali does not have a countrywide electricity network. See Figure 1. The LEAP model used does not consider the distinction between locations where a grid or network is available to transmit and distribute the electricity. Further analysis was made using QGIS to account for grid extension [53].
QGIS enables the geographical component to be included in modeling and can account for the difference between locations. This enables the component of "presence of electricity grid" to be included in the study.
As previously mentioned, extending the grid is not always the best option; as such, the methodology depicted in Figure 3 was developed and followed. , specific photovoltaic output [66], wind power density [67], and main and secondary rivers [68]) specific to the area delimited by it. The final product expected is a map of Mali with   [65], specific photovoltaic output [66], wind power density [67], and main and secondary rivers [68]) specific to the area delimited by it. The final product expected is a map of Mali with the best possible electrification option cost-wise under the assumptions considered.
Besides those used on the previous step (LEAP), these assumptions include an expected decrease of the Levelized cost of Electricity (LCOE) for solar PV-battery systems and an increase in disposable income generated due to the increase in electricity access/consumption.
For each cell or unit, the tiered net present cost (NPC) for the solar PV system required to supply the demand of the relevant unit is calculated following Equation (6): where T i NPC is the net present cost per tier, LCOE is the levelized cost of electricity for solar PV, H is the number of households in the cell, Tc i is the minimum daily electricity consumption per tier of electrification (see Table 4), L is the lifetime of the project, DR is the discount rate, and PR is the performance ratio of the plant.
As the study presents scenarios up to 2050, three values of LCOE were used: 0.25 [17,69], 0.12 [16,70], and 0.069 [14]. These are meant to represent the gradual decrease in price for solar PV electricity generation throughout the study timeframe and are assumed to represent the LCOE for 2030, 2040, and 2050, respectively. The discount rate was assumed to be 8% [71], and the performance ratio to be 0.3 [72].
As mentioned before, this paper assumes that electricity consumption directly influences the increase of disposable income. The rate at which this increase is expected to occur follows the formula defined by EoI_Index 2. This formula was modified to fit Mali's maximum average monthly salary [73] under the assumption that poor populations behave similarly on low electricity consumption levels.
where I H is the income per household (assuming one income maker per household), MAI is the maximum average monthly income taken as USD 2646 [73], ß 0 = 0.01412 and ß 1 = 0.98665, and x is the electricity consumption in kWh. Affordability and viability were calculated to select which tier of solar electrification, if any, is best suited for each unit. In this paper, affordability is defined as the ability of the population to pay for the electricity services without compromising their ability to supply for their other needs after the effects of increased income generation due to increased electricity consumption are estimated (affordability is present when Equation (8) is true). Viability refers to whether the technology suggested is more or less expensive than the alternative.
where S sI is the sustainable portion of the daily income used for electricity services, Cte pop is the population growth in the year of interest, I H is the increase of income due to increased electricity consumption, Num H is the number of households present in the geographical unit, T D is the minimum daily demand by tier of electrification (see Table 4), and LCOE Y is the assumed LCOE for the year of interest.
To calculate the sustainable portion of daily income to be used for electricity S sI , the poverty levels per region [74] were used. The dataset contains the population living with less than USD 5.5 per day, less than USD 3.2 per day, and USD 1.9 per day. The calculation assumes that the number of people under each category has at least the amount of the category. Two percent of the daily income is taken as a sustainable quantity to be spent on electricity in Equation (9).
Pop <1.9 is the amount of population assumed to have at least USD 1.9 of daily income, Pop <3.2 is the amount of population assumed to have at least USD 3.2 of daily income, and Pop T is the total population living within the area delimited by the geographical unit.
This process was performed for each geographical unit and the years 2030, 2040, and 2050 taking into account population growth and decrease of PV LCOE.

Results
Scenario NATa considered the possibility of transitioning from biomass consumption in the rural residential sector to an LPG-dominated demand as a first step, to then transition to an electricity-dominated demand (scenarios N1C_T). Although this study does not provide a specific timeframe for the transition between tiers of electrification, the process can be inferred from the differences between the tiered electrification scenarios (see Figure 4). As the electrification tier increases, the demand for biomass and oil products decreases. As scenario NATa and following tiered scenarios (N1C) consider the biomass demand of the urban areas to transition to LPG as soon as 2030, by 2050, the demand for oil products will be great.  In scenario NATb, the urban residential sector does not transition to LPG usage by 2030 but rather to electricity by 2050; this results in an increased demand for electricity and lower demand for oil products. As this scenario does not consider a transition in the rural sector from biomass to LPG to electricity, by 2050, no variation in the oil product demands between electrification tiers is apparent (see Error! Reference source not found.  In scenario NATb, the urban residential sector does not transition to LPG usage by 2030 but rather to electricity by 2050; this results in an increased demand for electricity and lower demand for oil products. As this scenario does not consider a transition in the rural sector from biomass to LPG to electricity, by 2050, no variation in the oil product demands between electrification tiers is apparent (see Figure 5).
For Mali to supply the demand by 2050 of any of the scenarios mentioned above, see Table 2, considerable increases in electricity generation capacity will be needed, be it from renewable or non-renewable sources. This is highlighted by the extra electricity needed, labeled "Electricity" in Figure 6. Although, with the transition from biomass for cooking to oil products in the urban sector, this extra electricity is overtaken by the oil products requirement that will ensue, see Figure 6.
In scenario NATb, the urban residential sector does not transition to LPG usage by 2030 but rather to electricity by 2050; this results in an increased demand for electricity and lower demand for oil products. As this scenario does not consider a transition in the rural sector from biomass to LPG to electricity, by 2050, no variation in the oil product demands between electrification tiers is apparent (see Error! Reference source not found.).

Figure 5. Energy demand by 2050 for scenarios based on NATb.
For Mali to supply the demand by 2050 of any of the scenarios mentioned above, see Table 2, considerable increases in electricity generation capacity will be needed, be it from renewable or non-renewable sources. This is highlighted by the extra electricity needed, labeled "Electricity" in Error! Reference source not found.. Although, with the transition from biomass for cooking to oil products in the urban sector, this extra electricity is overtaken by the oil products requirement that will ensue, see Error! Reference source not found.. Similarly, when the urban residential sector transitions from biomass to electricity, the amount in electricity imports needed increases from less than the 500 ktoe needed in the previous scenarios to 1775 ktoe in N2C T3 and 1311 ktoe in N2C T5. Mali's target for electricity expansion by 2030 considers a limit in electricity imports of up to 680 ktoe [57]. However, the requirements for a transition to electricity will require at least double this amount by 2050 even when T5 electrification/generation is available in the rural areas; see Error! Reference source not found..   Similarly, when the urban residential sector transitions from biomass to electricity, the amount in electricity imports needed increases from less than the 500 ktoe needed in the previous scenarios to 1775 ktoe in N2C T3 and 1311 ktoe in N2C T5. Mali's target for electricity expansion by 2030 considers a limit in electricity imports of up to 680 ktoe [57]. However, the requirements for a transition to electricity will require at least double this amount by 2050 even when T5 electrification/generation is available in the rural areas; see Figure 7.
While the results obtained with LEAP do not represent a timeline of events to reach rural electrification by 2050, it presents the picture of a possible future.
For the QGIS results, three timestamps were used: 2030, 2040, and 2050. These timestamps can be seen in Figure 8 for 2030, Figure 9 for 2040, and Figure 10 for 2050. As the methodology describes, each colored spot on the map is a hexagonal geographical unit of 5 km of internal diameter. Each colored unit represents which tier of electrification is both viable and affordable for the geographical unit.
Similarly, when the urban residential sector transitions from biomass to electricity, the amount in electricity imports needed increases from less than the 500 ktoe needed in the previous scenarios to 1775 ktoe in N2C T3 and 1311 ktoe in N2C T5. Mali's target for electricity expansion by 2030 considers a limit in electricity imports of up to 680 ktoe [57]. However, the requirements for a transition to electricity will require at least double this amount by 2050 even when T5 electrification/generation is available in the rural areas; see Error! Reference source not found.. While the results obtained with LEAP do not represent a timeline of events to reach rural electrification by 2050, it presents the picture of a possible future.
For the QGIS results, three timestamps were used: 2030, 2040, and 2050. These timestamps can be seen in Error! Reference source not found. for 2030, Error! Reference source not found. for 2040, and Error! Reference source not found. for 2050. As the methodology describes, each colored spot on the map is a hexagonal geographical unit of 5 km of internal diameter. Each colored unit represents which tier of electrification is both viable and affordable for the geographical unit. When all populated areas are electrified according to the tiers presented in the previous figures, the potential reduction of the biomass consumption reaches a minimum of 50% reduction from the normal (no tiers of electrification) biomass consumption on 2030 with no income effect; see Table 5. When the effects of increased income due to the increase of electricity consumption are accounted for, the possible reduction of biomass consumption reaches its maximum in 2050: 71%.

Discussion
Given the current electrification rates and generation capacity, substantial investment will be necessary to supply modern energy to the population of Mali. This study compared three approaches in which this could be done. The first focused on replacing traditional biomass consumption with LPG and using decentralized solar PV to electrify rural populations. The second considered progressive investments in decentralized solar PV to supply the rural population of Mali with electricity according to five stacking electrification tiers. The last approach considered grid extension and centralized electricity generation.
Both decentralized approaches showcase a need for an increased centralized generation to supply the urban population, the size of which varies depending on whether the biomass consumption of the urban residential sector is replaced with LPG or with electricity. In all approaches, the consumption of biomass was at least halved.
The provenance of the electricity generated for centralized distribution was not considered in this study, as the main focus was electrifying the rural population. Therefore, an environmental comparison between the transition to LPG or electricity in the residential sector of Mali was not performed. Consequently, a conclusion about which one is better is impossible without further study. As for cooking purposes alone, the increase in efficiency between switching from biomass to LPG or electricity provides no significant differences to ascertain which one is the best approach. Although it is a given that if the electricity is generated through green means, read renewable energy, transitioning to electricity would incur lower emissions than LPG.
Electrifying rural populations in countries where the national grid does not reach the vast majority of the populated areas, such as Mali, is no easy task. This study used five tiers of electrification and compared the economics of extending the grid versus installing solar PV plants on the locations where the electricity is needed using two scenarios. The first considers that electricity consumption has no direct effect on income generation. The second considers a direct influence. When this influence is considered, numerous populated regions can afford higher electrification tiers. Although, even when it is not considered (the influence of electricity consumption on income), installing decentralized solar PV instead of extending the grid incurs fewer costs when the tier of electrification, and therefore the amount of electricity consumed, is low. Decentralized systems become a better solution as the distance between the existing grid and the location of interest increases. For the vast majority of remote locations, installing solar PV even for the highest tier of electrification is more economically viable than extending the grid when the influence of electricity on income is considered.
As the prices for solar PV are expected to decrease in the future, the areas where the economic viability of solar PV over grid extension are set to increase. This reduction in prices opens the possibility of progressive grid extension and stacking solar PV installations for Mali to reach high electrification rates without the need for either expensive grid extension or fossil-based electricity generation.
When the benefits of new or increased electricity consumption are already present most of the populated areas of Mali could afford acceptable levels of electrification. It would be a matter of waiting for these benefits to ensue before they can start paying for either the service or the equipment. Therefore, any business or support model intended to electrify rural populations should consider providing a gap between the time when access to the electricity service starts and when the population begins to pay for it. Because this study did not include a demonstration site on its methodology, the time it would take for the discussed increase in disposable income to appear is uncertain.

Conclusions
In this study, the current electricity access and generation situation of Mali was used to generate multiple electrification scenarios. The scenarios were focused on providing electricity to the residential sector of Mali and mainly to the currently not electrified remote rural areas. In Section 1 of this study, the projected energy demand was met with different tiers of solar electrification. This shed light on the possible reduction of biomass consumption and fossil fuels.
In Section 2 of the study, the effects of new or increased electrification were taken into account. A spatial methodology was used to find which solar electrification tier was viable and affordable for different areas in Mali. This step was done for cases where the effects of electricity demand on income generation were present and were not present. This produced a graphical comparison of how, when enough time is given for the effects to ensue, the ability of the population to afford higher tiers of electrification increases. These results obtained through the methodology presented propose that the inclusion of the effects of electricity demand on income generation is an essential factor to consider for rural electrification modeling.
Most remote rural areas in developing countries lack the economic resources to pay a monthly fee for electricity; this is why the private sector is so reticent about investing in it. Pay-as-you-go financing with the proper data gathering tools could serve as both a viable way for the newly electrified populations to pay for the services and as an indicator on when the demand has increased and when a new expansion is required. With this, two objectives could be met. First, investment is only made when the need is proven/expected. Second, research-wise, this dataset could present a timescale for when precisely the benefits of increased or new electrification emerge, making a refinement of rural energy modeling considering the effects of electrification on income generation possible.
The inclusion of biogas production from rural waste is a possible improvement in the present work. Substituting the need for higher tiers of solar electrification with biogas sourced cooking and heating; could present faster and more affordable electrification for remote rural areas. In its current form, this study concludes that solar PV is the best option to electrify remote rural areas even when the effects of electrification on income are yet to emerge. When they are present, the potential of the population to afford higher tiers of electrification modify the results enough to consider it an essential factor to be considered in future studies. Data Availability Statement: Publicly available datasets were analyzed in this study. This data can be found on the referenced sources.