When energy doesn’t add up: use of an energyshed framework in assessing progress towards renewable energy transitions

Global progress in energy transitions to support climate mitigation goals has been slower than anticipated; this has prompted shifts away from traditional paradigms of regulated energy ownership towards a model of energy democratization by local communities and individuals. For example, in the United States, local communities in over 250 cities, counties, and states have made pledges to reach 100% renewable electrification by target dates ranging from 2020 to 2050. However, the availability of infrastructure and the competition for renewable energy resources, as well as lack of awareness of these limitations, present significant barriers to overcome. In this study, we explored a subset of 31 of these cities to assess their current electricity generation and how much further they have to go to meet their goals. Through an energyshed framework, we estimated powerplant electricity allocation to each city assuming competition for power from various renewable and non-renewable resource types, as well as look at the ‘best case scenario’ assuming 100% allocation of renewable-sourced electricity for a handful of cities in order to understand the existing and planned energy mixes for 2021 and the following 20 years. It is likely most cities will meet 10% of their energy demand with renewable energy, with best cases scenarios reaching between 35% and 65% renewable penetration, within the next 20–30 years. This highlights the need for infrastructural development in the energy sector, as well as intentional planning efforts in order to make these energy goals a reality.


Introduction
The energy landscape across the world is changing [1].The need for renewable energy transitions in support of climate action is well established-yet the pathways to achieving this integration remain uncertain.Concerns of infrastructure stability, electricity grid congestion, energy storage, resource exhaustion, land-use changes, and biodiversity loss can muddy the waters on achievable benchmarks and steady progress forward [2][3][4][5].These concerns present valuable and necessary caution in the conversations of energy transitions.However, this can often lead to 'analysis paralysis' for implementing the solutions already presented by the scientific community.This begs the question 'how far do we have to go?' in meeting national, regional, and city level goals for renewable integration.
In recommitting to the Paris Agreement, the United States hopes to contribute towards goals of limiting global warming-namely through greenhouse gas (GHG) emission reduction [6].To achieve this, the United States (and many other countries) are leaning into the transition from fossil fuels to more carbon neutral energy supplies.According to the Energy Information Administration (EIA), the United States is expected to increase its renewable energy mix from 21% in 2021 to 44% by 2050 [7].This target would show progress toward meeting Nationally Determined Contributions within the Paris Agreement and the United Nations Sustainable Development Goals.However, the recent pathways of integrating renewable energy to meet sustainability goals are often at odds with other aspects of sustainability such as environmental preservation Figure 1.Pledged city population distribution and status-cities that have made a pledge to source 100% renewable electricity/energy are shown by their population and status as of 2022 [9].The selected cities are the 31 cities include in this study, and are all also considered 'Committed' as they are not yet fully powered by renewable energy.'Powered' cites are those that had made the 100% commitment, and are reported by the Sierra Club as being 100% powered by renewable energy [9].and conservation.The rate and magnitude of this transition may strain environmental resources to a point that leads to irreversible and significant harm to ecosystems, biodiversity [3] and water availability [8].Yet, the goals for limiting global temperature increases calls for drastic and expedited efforts in GHG emission reduction-and the projected efforts of the United States may not meet these targets.Additionally, the current condition of distribution and storage infrastructure in the electricity system appears to be ill-equipped to handle a large-scale energy transition.
Amidst these national concerns and accompanying gridlock, there is effort at local scales to contribute to renewable energy transitions.This mobilization of resources and communities demonstrates a desire to move forward with energy goals.Over 250 communities have committed to achieving 100% renewable energy by 2050, and so far, 20% have reported to have met their goals [9] (figure 1).These cities have set a benchmark for other wishing to following in their steps-however additional context helps to highlight why the remaining communities may have more work yet ahead of them.Many of the cities that have yet to meet renewable energy goals are of populations significantly larger than those already 'Powered' .The 'Powered Cities' , in fact, only account for 3% of the total population of all cities that have made a commitment.'Powered Cities' are categorized as such based on reportedly receiving '100% clean, renewable electricity' [9].Although this is a notable accomplishment, there is some doubt regarding the true nature of what qualifies as receiving 100% renewable energy.
Recent research has shown that regional transitions towards renewable electricity generation could have higher ecological impacts, if pursued in a haphazard fashion [3].Urban areas already have well established links to the rural areas outside their formal incorporation [2,10], including the energy sector and associated networks.The concept of 'Energysheds' seeks to better define this space of connecting urban energy sinks to their generation sources, powerplant allocations, and overall energy 'footprints' that ought to be considered within the anticipated energy transitions [10][11][12].The term energyshed is still relatively new but has gained national attention for the benefit it offers to understanding the urban energy transitions that are already well underway [13,14].An energyshed is defined in several various ways, but the roots of energysheds as an operational mechanism are defined as 'a network of powerplants and transmission infrastructure required to supply electricity to a given point or zone of consumption…on the electricity grid' [10,11].This idea is largely driven by the dynamics of supply (from powerplants) and demand (from cities) while also considering the overlapping and interconnected nature of energy distribution systems.This concept also plays an important role in mapping the allocation of powerplant generation to determine relative energy mixes for cities.The energyshed framework offers a unique approach to contextualizing the inherent friction in the energy system dynamics described above, as well as revealing potential insights and priorities to future approaches for energy transitions [14].
In our study, a subset of cities that have committed to 100% renewable energy by 2050 were assessed through an energyshed framework, with the goal of better understanding the current and projected efforts of their transitions (figures 1 and 2).Given the complexities of the electric grid, we present our results through two scenarios: (1) a realistic assessment of modeled powerplant allocation and the implied sharing of electricity from power-producing resources in the 31 overlapping energysheds, and (2) a mutually exclusive assessment for a handful for cities that assumes new powerplant construction within their energysheds have a 100% allocation to towards this city.The results presented herein explore how far these cities have to go in achieving their goals, as well as suggest what challenges and opportunities may exist in their transitions.

Overview
The approach to this study was three-fold-(1) generate energysheds for a subset of cities with a pledge to achieve 100% renewable energy, (2) predict and summarize energy mixes using modeled allocation data for powerplants constructed after 2010 and in planning post 2021, and (3) summarize best case scenarios for a subset of cities assuming mutually exclusive powerplant allocation of new construction within their energysheds.The selection of the 31 cities used within the study was based on an overlap of available powerplant/city allocation data from the DeRolph et al [11] dataset and the cities committed to 100% renewable energy by 2050 from the Sierra Club Ready for 100 dataset.The four cities selected for the mutually exclusive best-case scenario were based on representing a range of geography, energy mixes, and progress.

Energyshed development
Using the dataset of city powerplant allocation from DeRolph et al [11], each city within the study was attributed a list of powerplants identified by EIA ID and their energy allocation to that specific city (in MWh).Each cities' list of contributing powerplants was then mapped in ArcGIS using the EIA Powerplant dataset from 2021 [15].From the 2021 EIA-860 Schedule 3 Generator dataset, each powerplant could be attributed its individual generators by technology type, capacity (in MW), capacity factor, status (operational, proposed, retired/cancelled), and status year [15].The base energysheds for each city were thus created to only include the generator units that were operational in 2010 at the powerplants identified from DeRolph et al (which accounts for 85% of city energy demand in 2010).
Around each city, three concentric buffers were developed to define the energyshed spatially for that city.Distance from each powerplant to the city's centroid were sorted from most proximate to distally located.As each powerplant had an energy allocation amount, the three buffers reflect the incremental distances from the city where 75%, 90%, and 99% of total demand was met by powerplant allocation.In order to remain conservative in estimates of expanding energyshed assets, the 75% buffer was used to add in generator units that were within it for the 2021 energyshed composition, as well as the 'planned' energyshed composition.For the energyshed composition to reflect the current powerplant generation, generator units that had been retired between 2010 and 2021 were excluded from the 2021 energyshed composition, and those that are slated for retirement after 2021 were excluded from the 'planned' energyshed composition.Additionally, any new generator units to have come online, or that are planned to come online within these timeframes, were added into their respective datasets.This process allows for the creation of three separate energyshed compositions for each city-each reflecting the generator assets expected to meet majority of energy demand from 2010, 2021, and existing future plans.

Allocation based energy mixes 2.3.1. Predicting powerplant allocation post-2010
Given that only the powerplants that exist within the 2010 dataset have allocation data, powerplants and generators that have been constructed since 2010 require their energy allocation to be determined.To do so, we used generalized linear models (GLMs) with the gamma link function to predict our response variable of powerplant demand fraction.The relationship between demand fraction (Dmd) and powerplant allocation (in MWh) is illustrated in equation ( 1), For each city, five GLMs were created, based on the combinations of predictor variables provided below in supplemental figure 1 (SF 1).The GLM with the lowest AIC value was selected, and the model was used on the original dataset from 2010 to predict the demand fraction for that powerplant, in order to validate the model.The residuals of this model were then plotted to check for any instances of heteroskedasticity (SF 1).The selected GLM was then used to predict the demand fraction for powerplants in the 2021 and planned datasets that did not have previously recorded allocation.

Developing energy portfolios
Following the determination of individual powerplant demand fraction, each generator unit at a powerplant was assigned the same demand fraction.Using this, the individual annual generator energy allocation could be calculated, as shown in equation ( 2).The annual generator energy (Gen) in MWh was determined using equation ( 3), where P is the generator power capacity (MW) and cf is the capacity factor [15].As powerplants often consist of multiple generator technology types, the energy portfolios needed to be summarized at the generator level rather than the powerplant level, The summation of each city's annual energy generation by resource type was determined from these results, which was then transformed into a percent composition for each of the three time frames.The percent change of resource types (figure 5) was based on the change between the 2021 energy composition and the planned composition.For the purposes of energy technology categorization, the extensive list of technology types from the EIA Powerplant and Generator datasets [15] was simplified to a general Resource type which were then categorized into either 'Nonrenewable Energy' or 'Renewable Energy' (supplemental table 3).

Mutually exclusive energy portfolios
Using the spatial framework of each of the four city energysheds, the 'best case scenario' was determined assuming that all new construction since the 2010 energyshed allocation work that is within a cities energyshed was 100% allocated to that city.Given that energysheds overlap, each of these city's results are mutually exclusive-two cities that have overlapping energysheds cannot be considered to both be operating under this 'best case scenario' .This is to represent the case where new construction proximate to a city with a pledge towards 100% renewable energy has invested in energy asset development under an agreement that allocated 100% of that powerplants generation to the city.An implicit assumption of this scenario is that electricity transmission distribution systems will require substantial updates to directly wire power producing assets to city end user nodes.For each of the cities, the energysheds developed include a list of all powerplants and generator units with allocation data from 2010, as well as the plants and generator units that are operational as of 2021 and are planned within the next 20 years.Generation for these units was determine using equation (3).Only new construction since 2010 was assumed to have 100% allocation to its corresponding city-powerplants operational in 2010 within the cities energyshed retained their allocation results from the DeRolph et al dataset.

Results
Across the 31 cities present in this study, none are expected to meet their goal of 100% renewable based on existing or planned infrastructure development.On average, these cities are projected to reach only a 10% renewable energy mix within the next 20 years.Even in the four best base scenarios developed, cities appear to cap off renewable energy penetration between 35% and 65% in the next two decades.

Energyshed dynamics
As expected, the energysheds developed for the 31 cities have significant overlap, as shown in figure 2. This overlap is driven by a share of existing powerplant generation, as well as overlapping demand from planned powerplants in the coming years.Despite 250 communities across the United States pledging to reach 100% renewable energy in the next 30 years, the 2021 EIA-860 Generator dataset includes only 1842 proposed generator units (providing an additional 170 870 MW of grid capacity) [15].Across the 31 energysheds assessed within this study, over 1000 of these units are accounted for-despite large regions of the United States not overlapping with the study areas (figure 2).Of the proposed generator units, 1265 can be classified as renewable energy sources, totaling an anticipated 113 000 MW of capacity.While majority of proposed projects are renewable, 90% of these generator units of are from solar and wind additions-implying a great need for energy storage [15].

Trends in energy transitions
There are several trends in the energy transitions that occur across cities within the study.Within the nonrenewable energy dynamics, there is largely a shift from coal power to natural gas and natural gas combined cycle powerplants (figure 3).Across the 31 cities, coal declined from 39% to 30% from 2010 to the planned 2040 energy mix, with most of this transition having occurred by 2021 (figures 3(a) and (b)).Across the United States, coal has declines from 45% to 17% in the same period.Natural gas appears to have replaced much of coal power, as its initial 34% has risen to 51% of the energy mix across these 31 cities.
Regarding renewable energy-wind and solar host the majority of new development, with the hydropower and geothermal composition remaining fairly constant in the total energy mix (figures 3(a) and (c)).In looking at only the renewable energy sources, the proportion of hydropower and biomass is cut in half from 2010 to the anticipated portfolio in 2040.This is largely due to limited new development of these resources, and a significant increase in wind and solar construction, consistent with the US EIA reports [7].Wind energy is expected to increase from 5% of the renewable energy mix to 25% based on existing plans in these cities.Solar is anticipated to move from less than 1% of the total renewable mix to 26% (figure 3(c)).The initial energy composition for 2010 and the anticipated modeled energy totals for 2020 and 2040 are available in SI tables 1 and 2 (nonrenewable and renewable energy sources).

Two development scenarios
In order to compare the powerplant allocation approach to the mutually exclusive development approach, figure 4 illustrates that these methods may be seen as a 'business as usual' and 'best case' scenarios.For example, in Columbia SC, the allocation scenario predicts only a slight shift in energy composition from 2010 to the planned build outs.This is based on assuming that new powerplants built within Columbia's energyshed are following the same allocation patterns and distributions as before.While this may be the case for some development, the mutually exclusive scenario shows what the energy mix may look like if all new construction within Columbia's energyshed was dedicated to this city alone.The future energy mix moves from a potential of 8% to a best case of 65% renewable energy integration (figure 4(a)).A similar trend is shown in Akron, OH, Las Vegas, NV, and Kansas City, MO.Las Vegas (figure 4(d)) shows the smallest shift in energy composition, even with 100% allocation within its energyshed.This suggests that there is a significant lack of new development proximate to Las Vegas to facilitate its energy transition.
Because these results necessitate that all new construction within a city's energyshed are completely dedicated to that city, these scenarios are mutually exclusive, and not representative of a possible reality.It does, however, provide an absolute cap for the range of uncertainty in energy mixes compared to the business-as-usual approach of powerplant allocation.Even in instances of 100%, mutually exclusive powerplant allocation, cities are not expected to meet 100% renewable energy integration based on current development plans.Providing these two scenarios allows for a measure of uncertainty in our results to be contextualized.Although it is unlikely that all new development within a city is energyshed will be allocated to it-there is a possibly that some development could be intended for direct wiring or decentralized generation.The mutually exclusive scenario provides an upper limit for our estimates of what is the best case for renewable energy integration.Alternatively, the 'business as usual' allocation approach suggests a more realistic, and conservative estimate, of city renewable energy integration based on current powerplant development.Additional uncertainty remains based on the accuracy of our models (SI figure 1) as well as the assumption that major transmission infrastructure has not shifted significantly.

Progress towards the goal
All of the cities demonstrate movement towards an increased share of renewable energy (figure 5).Although the average across the 31 cities using the allocation approach is around 10%, there is one city-Portland, Development scenarios-energy mixes for each of the four cities selected has an 'allocation scenario' and a 'mutually exclusive scenario' depicted.Allocation scenarios reflect that of all 31 cities where powerplant allocation was predicted.Mutually exclusive scenarios for the four shown cities show energy mixes assuming all new powerplant development within that city energyshed was allocated 100% to that city.Oregon-which stands apart.Unlike many of the other cities in this study, Portland's starting renewable energy mix was already significantly higher than most (69%).While the city does have a limited increase in their renewable energy mix (expected increase to 71%), their overall composition from the reference year places them well above the rest.In addition to Portland, Las Vegas appears to be making progress to their goal-surpassing 30% of their energy mix being from renewable energy sources given planned powerplant construction and facility retirement (figure 5).Rochester, New York, and Salt Lake City, Utah are on a similar path of progress with Salt Lake City having the largest percent increase in renewable energy mix-from 5% to over 25%.

Discussion
Of the over 250 communities setting their sights on 100% renewable energy integration, only 52 of these have met their electrification goals [9].Notably, the cities that have met their goals have an average population of 55 700 (ranging from 400 to 390 000 residents), whereas the cities that are committed but have yet to meet their goals have an average population of 540 000 (ranging from 100 to 1800 000 residents) (figure 1).The cities selected for this study all have relatively higher populations among cities committed to 100% renewable energy, and these represent larger urban areas that have drastically different pathways to success than the smaller cities working towards the same goals.While our study focused on cities as the focal point for an energyshed, the energyshed model can be scaled down to block groups or up to metropolitan areas as developed in the initial DeRolph energyshed assessment as a means to assess aggregate energy demand from rural and suburban areas, as well as city boundaries DeRolph et al [11].Those identified as 'Powered Cities' typically are receiving '100% clean, renewable electricity' [9] from either a municipal utility provider citing 100% wind or solar, or from a regional power alliance serving several of these neighboring communities [9].However, given the current limitations of storage and grid infrastructure, our data causes us to question the authenticity of a city as large as 390 000 (Oxnard CA) or even 90 000 residents (Santa Monica, CA) being fully powered by renewable energy 100% of the time.
Although there is movement in the right direction for the remaining committed cities, most of these communities began their pledges towards 100% renewable energy in the early 2000s and thus progress has been slow.With many having benchmarks near 2020, and end goals in 2050, none of the 31 cities within this study are even halfway to meeting their goals, despite nearing the halfway mark in many of their timeframes.Although this subset of 31 cities reflects only 12% of the cities with a commitment, they represent 84% of the total population of these 250 cities.The EIA projects that by 2050 the country wide renewable energy share will be nearly 45%, with the current renewable generation sitting at nearly 13% of total production [7].This implies, assuming energy production remains constant, that renewable energy generation will need to triple in the next 30 years to meet the 45% share.The EIA expects carbon emissions to dip slightly through 2035, but then continue their climb as energy production continues to outpace the rate of expected renewable energy transition [7].With this as the case, and energy demand only increasing, the communities that have made a pledge to be 100% renewable energy by 2050 have significant ground to cover given projected trends.

Moving forward
Our results are well aligned with the trends illustrated in figure 6, as many of the cities exhibit similar trends in their energy mix progression.As shown in figures 2 and 6, there is currently a limited amount of planned powerplant construction that would significantly bolster the efforts towards 100% renewable energy integration.However, with the new funding and support programs taking place under the Infrastructure Investment and Jobs Act [16], it is still possible for progress to be made in the next few years.Improvements to electricity grid resilience, reliability, and flexibility are all essential to the integration of renewable energy [17].Tools that focus on the improvement to resource and operational flexibility of the energy sector are critical [4].Given the focus on improving the energy sectors' infrastructure, resilience, storage, and clean energy technology, the Department of Energy has employed the use of the energyshed framework to help policy makers, engineers, researchers, and communities more aware of what opportunities, potential pathways, and barriers are present for local scale initiatives in light of these large-scale transitions [13].Residential solar, battery storage, and other distributed energy resources are additional assets that may continue to shape the energy transitions at more local scales.One such example is the continued investment and development of rooftop solar at the residential scale.The analysis herein does not account for rooftop solar, as it is not included in the national scale powerplant datasets.Although it is not included in the assessment, rooftop solar provides an opportunity for homeowners and small businesses to reduce their electricity bill through energy sales back to the grid or offset their demand from the grid with rooftop generation/storage.The limitation here, however, it that large portions of city energy demand comes from commercial and industrial sectors, where the scale of solar needed to impact operations would necessitate a powerplant development that would likely be included in our available datasets.Other efforts explore and support that there is not only an infrastructural limitation in energy transitions, but social, economic, and political barriers [4,[18][19][20][21][22][23][24].

The need for storage
In general, because the hydropower capacity is expected to remain largely stagnant-while wind and solar increase-the overall mix of hydropower in the renewable portfolio often decreases.In each of the cities and across the Unites States, shown in figure 3, the total energy portfolio shows hydropower remaining 'constant' as an overall percentage of the energy supply.However, in the renewable energy nested chart (figure 3(c)), it is clear that over time, hydropower will encompass a smaller percentage of renewable generation.Yet, as it stands, hydropower is our current best solution to grid scale energy storage for renewable energy, as various modes of operation can allow for delayed release to meet peak demands or as baseload dispatch [26].Although hydropower, particularly large hydropower, is not likely a community level solution, it does contribute to balancing out the variability of wind and solar power due to grid scale storage for these technologies.Hydropower has a complex past and future, but small scale non-powered dam conversions or minimum flow turbine operations may contribute to more localized renewable energy efforts [27][28][29][30].Where the increase in renewable energy is still lagging behind the decommissioning of coal powerplants, natural gas energy generation is often seen to increase (figures 3(b) and (e))-both in the United States and globally [31,32].
Natural gas was intended to be a 'transition fuel' to help the energy sector cut back on emissions by replacing coal powerplants [31,32].However, there may be indirect effects of using natural gas as a bridge-in that it has prolonged reliance on fossil fuels and limited the transitions of the grid and storage infrastructure [31].While having successfully offered a significant replacement to coal emissions (figure 6), the growth of natural gas and its label as a 'transition fuel' has done little to reframe the mindset of operational, grid scale renewable energy integration and storage [27,33].As an example, San Francisco and Boston show a decrease in renewable energy composition from 2010 to 2021-primarily from the decommissioning of hydropower facilities and an increase in natural gas (figure 5).Without redefining the role of natural gas in the energy sector, large scale efforts for grid level energy storage will remain one of the most significant roadblocks to successful renewable energy transitions.

Competition or cooperation?
Given current planned construction, it appears that wind and solar energy are the main paths towards renewable energy integration [7] (figure 7).As such, some communities may begin to compete for land and powerplant allocation.Without the proper planning and infrastructure development, these transitions may be met with grid congestion and instability in distribution.One of the additional insights provided by an energyshed approach encourages developers to consider what geographic regions have overlapping interests and generation needs [10,11,13].In regard to the mutually exclusive development scenarios-it is also worth noting the significant infrastructure changes that would be required to hardwire in powerplants to a single city while passing through the surrounding regions.Based on the current grid infrastructure and operations, hardwiring powerplants can be done-but is unlikely at the scale demonstrated within this study.This suggests that communities must cooperate and communicate their energy transition and development plans with one another-as opportunities for optimized infrastructure development and expansion may be made clear.

Conclusion
Evident in the magnitude of energyshed overlap from just this subset of 31 US cities, there are many regions across the country that could foster cooperation for funding, land acquisition, infrastructure development, distribution networks, and storage efforts for renewable energy integration.The needs of successful energy transitions require reframing the issue as not one that is purely technical, economic, social, or political but that is interdisciplinary and requires collaboration and communication across multiple sectors.As local government officials and managers host the conversations for achieving 100% renewable energy within their community, there is a question of who is engaged as a stakeholder in developing these plans.Much of the energy sector is driven by supply and demand, operating on its own unique market structure, which requires the presence of energy economists [34].Socio-political support for renewable energy transitions must come from both urban and rural stakeholders where the various needs and contributions of these groups are explicit and mutually beneficial [2,10,19].
There are certainly many contributing and compounding factors to explain the lag in transition success despite growing support.Chief among these seem to remain as (1) economics, (2) leadership, and (3) energy literacy.(1) Although the levelized cost of energy (LCOE) for renewables has decreased, it has only done so in more recent years-however their intermittency still presents an issue with meeting peak energy demands due to lack of economic grid storage options.Within the energy markets themselves, facilities that can meet peak energy demands are more profitable.LCOE does not fully account for energy market prices during different periods of the day.It is only now being recognized as a need to reevaluate the economic assessments of energy technology and operation to accommodate for renewable intermittency and storage needs [35,36].As for (2) leadership, it is commendable that community leaders are taking up matters and demonstrating motivation towards energy transitions.However, it is still perceived by many that very often there is gridlock, mismanaged goal setting, and sociopolitical roadblocks at various scales of government that continue to hinder accountable and actionable progress [24,37,38].One of the other major factors that underpins majority of the discourse surrounding energy transitions is the lack of (3) energy literacy.Our world today is highly 'energy unconscious' and energy illiterate despite how intimately energy and energy use are intertwined with our daily lives and work [37][38][39].Most of the general population would not know where the nearest powerplant to them is-or how power is delivered to their home.
As research suggests, there are many alternate routes for achieving high renewable energy penetration pathways-each with their own set of trade-offs and implied infrastructural changes as the co-evolution of communities and energy systems takes place [5,10].Many cities will sponsor or endorse renewable energy projects outside of their municipality or city limits.As these projects develop and are distributed to surrounding areas, there is a traceable measurable connection to a community aiming towards 100% renewable energy goals.In contrast, purchasing renewable energy credits outside of a cities energyshed and labeling it as 'powered by 100% renewable energy' can be misleading to the community, and potential detrimental to the understanding of the infrastructural needs and challenges we face for a successful energy transition.The energyshed framework may help to better highlight possible futures for meeting renewable energy transition timelines, as well as ensuring that additional constraints and considerations are fully contextualized to the vast energy landscape.

Figure 2 .
Figure 2. Energysheds for selected cities-the 31 selected cities and their energysheds are shown.Additionally, all operational and currently planned powerplants within these energysheds are also included.Energysheds are cut off at the US borders, as only US powerplants were used in the assessment.In reality, the Western and Eastern Interconnections branch in Canada, which would expand the energysheds of the affected cities.

Figure 3 .
Figure 3. Energy mixes-panel (a) shows the energy mixes for the 31 cities in the study through 2010, 2021, and modeled scenarios.Panel (b) shows the relative nonrenewable energy mix and (c) shows the relative renewable energy mix.

Figure 4 .
Figure 4.Development scenarios-energy mixes for each of the four cities selected has an 'allocation scenario' and a 'mutually exclusive scenario' depicted.Allocation scenarios reflect that of all 31 cities where powerplant

Figure 5 .
Figure 5. Renewable mix percentage over time-each of the cities in the study are shown with their respective renewable energy mix percentage.This percentage is shown from the original 2010 dataset based on powerplant allocation from DeRolph et al.The 2021 and Planned quantities are based on results from the energyshed development and the GLMs developed for each city.

Figure 6 .
Figure 6.Energy transitions across the United States-the energy mixes across the United States are shown, separated into renewable and nonrenewable resources.These totals come from the US Energy Information Administration [15, 25].

Figure 7 .
Figure 7. Anticipated resource shifts-the transition away from and towards particular resource types for each city are shown.A positive percent change implies an increase in that resources' generation.A negative percent change implies a decreased reliance on that particular resource type.The reference period for change is from 2021 energyshed composition to the planned compositions.