As of 2021, 675 million people globally lack access to electricity. Geospatial electrification tools can be used to identify the mix of grid-extension, mini-grids and stand-alone technologies that can supply currently unelectrified areas at the lowest cost. Several such tools have been developed, at different levels of modelling detail and complexity. In this paper, we improve the Open Source Spatial Electrification Tool (OnSSET) to develop a flexible geospatial electrification tool that can still run lighter rapid assessments for a first estimate of the technology split, but now also more detailed analysis with higher spatial and temporal resolution used for grid routing, distribution network design and optimization of hybrid mini-grid generation introduced through new algorithms. We compare the existing light and new detailed versions of the tool through a case study of the north-western parts of the Democratic Republic of the Congo. We find that the new grid routing algorithm lead to more off-grid technologies, and that the detailed design of distribution networks lead to a reduction in stand-alone technologies. The detailed optimization of hybrid mini-grids display varying effects at different demand levels. Given the increased computational effort that is observed with higher modelling detail, we discuss the implications for scenario design and selection of geospatial electrification tool for future analyses aiming to support the achievement of SDG 7.

Access to electricity is a prerequisite for development, included in both the Agenda for Sustainable Development and the African Union’s Agenda 2063. Still, universal access to electricity is elusive to large parts of the global population. In Somalia, approximately one-third of the population has access to electricity. The country is unique among non-island countries as it has no centralized grid network. This paper applies a geospatial electrification model to examine paths towards universal access to electricity in Somalia under different timelines and with regard to different levels of myopia in the modeling process. This extends the previous scientific literature on geospatial electrification modeling by studying the effect of myopia for the first time and simultaneously presenting the first geospatial electrification analysis focused on Somalia. Using the Open Source Spatial Electrification Tool (OnSSET), the least-cost electrification options towards 2030 and 2040, respectively, are compared. We find that under the shorter timeline, a deployment of mini-grids and stand-alone PV technologies alone provides the least-cost option under all but one scenario. However, under the longer timeline, the construction of a national transmission backbone would lower overall costs if there is high demand growth and/or low cost of centralized grid electricity generation. We also compare different levels of myopia in the modeling process. Here, OnSSET is first run directly until 2040, then in five-year time-steps and annual time-steps. We find that running the model directly until 2040 leads to the lowest costs overall. Running the model myopically leads to a sub-optimal, more costly technology mix, with a lock-in effect towards stand-alone systems. On the other hand, the myopic approach does provide additional insights into the development of the system over time. We find that longer-term planning favors the centralized grid network, whereas short-sighted myopic planning can lead to higher costs in the long term and a technology mix with a higher share of stand-alone PV.

The Open Source Spatial Electrifcation Tool (OnSSET) is extended to provide a long-term geospatial electrifcationanalysis of Ethiopia, focusing on the role of grid- and of-grid technologies to increase residential electricity access underdiferent scenarios. Furthermore, the model explores issues of compatibility between the electricity supply technologiesover time. Six potential scenarios towards universal access to electricity in the country are examined based on threepathways; the Ambition pathway sees high demand growth and universal access achieved by 2025, the Slow Down pathway follows a lower demand growth with a slower electrifcation rate and with a higher share of of-grid technologies,and the Big Business pathway prioritizes grid electricity frst for the industrial sector, leading to slower residential electrifcation. The results show a large focus on grid extension and stand-alone PV deployment for least-cost electrifcationin case of low grid-generation costs and uninhibited grid expansion. However, in case of a slower grid rollout rate andhigh demand growth, a more dynamic evolution of the supply system is seen, where mini-grids play an important rolein transitional electrifcation. Similarly, in the case where grid electricity generation comes at a higher cost, mini-gridsprove to be cost-competitive with the centralized grid in many areas. Finally, we also show that transitional mini-grids,which are later incorporated into the centralized grid, risk increasing the investments signifcantly during the periodswhen these are integrated and mini-grid standards are not successfully implemented. In all cases, existing barriers todecentralized technologies must be removed to ensure of-grid technologies are deployed and potentially integratedwith the centralized grid as needed.

This paper presents the first application of the scenario discovery approach in geospatial electrification modelling. 1944 electrification simulations were constructed for Burkina Faso from a combination seven input levers, including four grid-extension strategies. The scenario discovery analysis identifies a scenario described by a high grid electricity generation cost in combination with an intensification strategy for grid-extension, as most likely to lead to a high cost of electricity in Burkina Faso. Thus, to avoid such a high cost, decisions in the country could be targeted either at lowering grid electricity generation costs or to choose one of the other two gridextension strategies, or both. For each of the grid-extension strategies, a number of drivers causing a high LCOE were identified. Common drivers for all strategies were the grid electricity generation cost and discount rate. The scenario discovery approach was used to identify the key drivers of high electrification costs and their interactions, providing useful information that might not be gained from a traditional scenario-axes approach. This approach provided a structured way to analyze more parameters than found in previous electrification studies for Burkina Faso. The paper discusses on the pros compared to a traditional scenario-axes approach, such as reduced risk of perceived bias and improved ability to deal with multiple uncertain parameters, but also notes the additional computational requirements.

Ethiopia is a low-income country, with low electricity access (45%) and an inefficient power transmission network. The government aims to achieve universal access and become an electricity exporter in the region by 2025. This study provides an invaluable perspective on different aspects of Ethiopia's energy transition, focusing on achieving universal access and covering the country's electricity needs during 2015-2065. We co-developed and investigated three scenarios to examine the policy and technology levels available to the government to meet their national priorities. To conduct this analysis, we soft-linked OnSSET, a modelling tool used for geospatial analysis, with OSeMOSYS, a cost-optimization modelling tool used for medium to long-run energy planning. Our results show that the country needs to diversify its power generation system to achieve universal access and cover its future electricity needs by increasing its overall carbon dioxide emissions and fully exploit hydropower. With the aim of achieving universal access by 2025, the newly electrified population is supplied primarily by the grid (65%), followed by stand-alone (32%) technologies. Similarly, until 2065, most of the electrified people by 2025 will continue to be grid-connected (99%). The country's exports will increase to 17 TWh by 2065, up from 832 GWh in 2015, leading to a cumulative rise in electricity export revenues of 184 billion USD.

Human settlements are usually nucleated around manmade central points or distinctive natural features, forming clusters that vary in shape and size. However, population distribution in geo-sciences is often represented in the form of pixelated rasters. Rasters indicate population density at predefined spatial resolutions, but are unable to capture the actual shape or size of settlements. Here we suggest a methodology that translates high-resolution raster population data into vector-based population clusters. We use open-source data and develop an open-access algorithm tailored for low and middle-income countries with data scarcity issues. Each cluster includes unique characteristics indicating population, electrification rate and urban-rural categorization. Results are validated against national electrification rates provided by the World Bank and data from selected Demographic and Health Surveys (DHS). We find that our modeled national electrification rates are consistent with the rates reported by the World Bank, while the modeled urban/rural classification has 88% accuracy. By delineating settlements, this dataset can complement existing raster population data in studies such as energy planning, urban planning and disease response.

The introduction of geospatial data into modelling efforts carries many advantages but also introduces numerous challenges. A common challenge is the Modifiable Areal Unit Problem (MAUP), describing how results change as the spatial aggregation of data changes. Here, we have studied MAUP in geospatial least-cost electrification modelling. We do this by assessing the effects of using 26 different population bases each for Benin, Malawi and Namibia. We use the population bases to generate 2080 electrification scenarios per country and conducting a global sensitivity analysis using the Delta Moment-Independent Measure. We identify population aggregation to be highly influential to the model results with regards to method of aggregation (delta values of 0.06-0.24 depending on output studied), administrative division (0.05-0.14), buffer chosen in the clustering process (0.05-0.32) and the minimum number of neighbours within the buffer required for clustering (0.05-0.19). Based on our findings, we conclude that geospatial electrification studies are not robust concerning the choice of population data. We suggest, that modelers put larger emphasis on different population aggregation methods in their sensitivity analyses and that the methods chosen to conduct sensitivity analysis are global in nature (i.e. moving all inputs simultaneously through their possible range of values).

The authors wish to make a change in author names (adding new author—Dimitrios Mentis) to this paper [1]: Author Contributions On page 19, author contributions are updated as follows: Conceptualization, A.K., D.M. and M.H.; Methodology, A.K., A.S., B.K. and D.M.; Software, A.K., B.K., A.S. and C.A.; Validation, M.H.; Formal Analysis, A.K.; Investigation, A.K.; Resources, A.K. and B.K.; Data Curation, A.K., B.K. and A.S.; Writing—Original Draft Preparation, A.K.; Writing—Review and Editing, A.K., D.M., M.H., C.A.; Visualization, A.K. and B.K.; Supervision, M.H.; Project Administration, M.H.; Funding Acquisition, M.H., D.M. and A.K. All authors have read and agreed to the published version of the manuscript. Funding On page 19, funding sources are updated as follows: This research was funded by the World Bank under the contract number 7185716 and partially by (a) the Swedish Center for Smart Grids and Energy Storage (SweGRIDS-ABB) under grant VF-2015-0018 and (b) the ÅForsk Foundation under grant 17-604. The authors would like to apologize for any inconvenience caused to the readers and contributors by these changes. The changes do not a ect the scientific results. The manuscript will be updated, and the original will remain online on the article webpage, with a reference to this correction.

For decades, electrification planning in the developing world has often focused on extending the national grid to increase electricity access. This article draws attention to the potential complementary role of decentralized alternatives – primarily micro-grids – to address universal electricity access targets. To this aim, we propose a methodology consisting of three steps to estimate the LCOE and to size micro-grids for large-scale geo-spatial electrification modelling. In the first step, stochastic load demand profiles are generated for a wide range of settlement archetypes using the open-source RAMP model. In the second step, stochastic optimization is carried by the open-source MicroGridsPy model for combinations of settlement size, load demand profiles and other important techno-economic parameters influencing the LCOE. In the third step, surrogate models are generated to automatically evaluate the LCOE using a multivariate regression of micro-grid optimization results as a function of influencing parameters defining each scenario instance. Our developments coupled to the OnSSET electrification tool reveal an important increase in the cost-competitiveness of micro-grids compared to previous analyses.

With pressing priorities in the development agenda, policy makers in developing countries are in the difficult situation of prioritizing policy actions. Limited government and utility budgets need cost effective solutions to bring the desired development benefits of electrification, health, education and food security among others. Energy access is a prerequisite for economic activity and for human development as interacts in synergy with other development needs. As rural electrification models usually focus on the supply of electricity solely, thermal energy needs, such as cooking and water heating remain unattended and satisfied by non-renewable energy fuels. To this aim, we explore optimal electrification solutions addressing two types of energy demands, electricity and thermal energy demands for cooking. Our model builds on a 3-step electrification methodology proposed by Peña et al. including electricity as a modern source of clean energy for cooking in rural communities. The total investments needed to build and operate the microgrids, including distribution costs, is 332 million USD. This is equivalent to 1129 USD/per inhabitant. This amount does not account however the health and environment benefits that e-cooking can bring to inhabitants in Bolivian low-lands.