With generation capacity of electricity systems in Canada projected to grow up to 116% by 2050 to accommodate electrification and population growth, power system planners face a growing challenge in implementing new generation capacity which retains system reliability while addressing public concerns of affordability, greenhouse gas emissions, and other socio-environmental impacts.

While optimizing for reliable and affordable electricity is the core focus of power system planning, planners are increasingly leveraging policies such as emission caps or moratoriums to address political and social pressure to limit social and environmental impacts of new electrical infrastructure. To establish policies which robustly support objectives despite limited knowledge of the future, contemporary methods use established uncertainty analysis tools such as stochastic optimization or Monte Carlo simulations but remain limited in modeling uncertainties with no underlying probability distribution and often have limited consideration for subjective or intangible socio-environmental impacts. This study addresses this gap by introducing a chronological myopic modelling approach to Monte Carlo uncertainty analysis where uncertainties are modeled as random walks, and socio-environmental outcomes are emphasized. This method is of particular interest for systems with high exposure to uncertainty and is thus applied to a case study in Yukon, Canada for its small and electrically isolated grid, but is broadly applicable to centrally planned grids.

For this study, we modeled the Yukon electricity system and measured the effects of introducing policy instruments to constrain power system planning decisions between the years 2022 and 2050. Policies evaluated include emissions limits, restrictions on imported fuels, and moratoriums on highly polluting or impactful generators. We let uncertain parameters take many random walks through time to form many hypothetical futures. For each future, we sequentially and myopically solve for cost-optimal capacity expansion with a multi-year lookahead given expected or forecasted values for each parameter. Uncertainties are resolved as the model progresses chronologically, solving for dispatch decisions of generation capacity built from prior model periods. The result is a distribution of system performance with measured outputs including costs, greenhouse gas emissions, aggregated Likert scoring of community and ecological impacts, and proxy measurements for system reliability and stability.

Early results indicate that regardless of policy, Yukon’s electricity system is most vulnerable to changing hydrological conditions, demand increases, and a peakier load profile, with all measured outputs exhibiting sensitivity to these parameters. In a pair-wise comparison of futures under each policy, systems planned without new fossil generators had on average 8.5% less generation costs and 99% less emissions than in the absence of policy but exhibited 42% higher ecological impacts owing to higher proportions of hydroelectric development. Additionally, we present a decomposition of uncertain parameters and correlate positive outcomes in the presence or absence of individual generators, revealing robust generation assets.

As the climate crisis intensifies, switching to renewable energy remains a critical piece of the solution to ensure rapid decarbonization. However, renewable energy generation is highly reliant on the ambient environmental conditions, making it difficult to estimate the long-term generation—a task that is likely to get more difficult under climate change. Accounting for the impact of climate change is particularly difficult, as there remains uncertainty related to the magnitude of climate change within the mid- and long-term in addition to the relatively unknown impacts of climate change on generation of renewable energy technologies. In this work, we aim to fill this gap by leveraging machine learning to investigate the impact of climate change on state-level renewable energy generation across the US. Using data from the Energy Information Administration (EIA), we project the solar, wind, and hydropower generation across multiple US states under two key climate change scenarios. Our goal is to answer two key questions: (1) How will climate change impact renewable energy generation; and (2) Do these impacts differ across states? To answer these questions, we leveraged several machine learning techniques, as well as an ensemble of models, to first model the observed relationship between renewable energy generation and the surrounding weather and climate. Then, we used those same models to project the changes to the system, given the most recent IPCC climate change scenarios. Here, we will present the results from the projection analysis across multiple US states, including the states of California, New York, Florida, and Georgia, which contain some of the largest electric utilities in the country. The results indicate significant changes across different states and seasons, which could impact grid management and planning. Ultimately, the results will provide critical insights into the sustainability of renewable energy technologies over the long-term, given the reality of climate change.

Utility-scale solar photovoltaic (PV) is pivotal to decarbonizing the power sector, driven by declining costs, supportive policies, and its scalability. The Midwest, with its vast agricultural lands, is rapidly emerging as a key region for solar expansion. However, existing studies on the economic impact of solar have predominantly focused on local benefits, overlooking the opportunity costs of converting agricultural lands into solar installations. In addition, traditional power planning research has yet to integrate economic impact assessments directly into capacity models for guiding optimal utility-scale siting decisions.

This study bridges these gaps by endogenously integrating local economic metrics into a power system planning model to assess the effects of economic impacts on utility-scale solar siting. We analyze all counties within the Great Lakes region—comprising Illinois, Indiana, Michigan, Minnesota, Ohio, and Wisconsin—to develop localized supply and marginal benefit curves. The supply curves capture the costs of capacity additions while accounting for zoning regulations, land exclusions, and parcel-level constraints on agricultural lands. Marginal benefit curves quantify the economic contribution per county from project lifecycle expenditures that support interconnected local industries, as well as accounting for the opportunity cost of agricultural land conversion. These curves feed into a multi-objective power system planning model aimed at minimizing system costs and maximizing local economic benefits for solar investments through 2040.

Results reveal that counties with larger economies and below-average crop productivity deliver the highest value-added per megawatt (MW) of installed capacity. For example, large counties (population above 50,000) in Minnesota generate up to $38,400 per MW, whereas this value declines by 31% in smaller counties (population below 15,000). Moreover, the conversion of agricultural land in large Minnesota counties results in annual reductions of 12% ($4,900 per MW) for low crop productivity counties versus 16% ($6,000 per MW) in counties with high crop productivity. A scenario that reallocates solar investments towards counties offering higher economic benefits indicates potential shifts of up to 53% ($860 million) in Ohio, with a corresponding decrease of 47% ($500 million) in Wisconsin, all without increasing overall system costs.

Our findings underscore the necessity of integrating economic considerations into utility-scale solar planning to balance decarbonization with rural economic development. By discussing the trade-offs and synergies between cost minimization and community benefits, this research provides actionable insights for policymakers and energy planners to promote equitable and sustainable solar deployment.

Energy and water systems are deeply interconnected, leading to complex interdependencies that change in magnitude with changing climate, socioeconomic, and policy landscapes. Energy systems rely on water at every stage of transformation--directly, for activities like cooling power plants or as a “feedstock” for hydropower and electrolysis, and indirectly, for mining primary fuels or cultivating biomass. Similarly, water systems require energy for a range of applications, such as groundwater extraction, reservoir operations, and water conveyance and treatment. Concurrent management of these interdependent and often competing energy and water flows is crucial for sustaining key societal functions, which require understanding their relative magnitudes and interdependencies across a range of futures to enable informed planning and management. Yet, such forward-looking, internally-consistent evaluations are missing, which hampers evidence-based planning of integrated energy-water systems. Consistent accounting of future energy, water, and combined energy-water flows, while incorporating multisector feedbacks among energy, water, and land systems under climate, socioeconomic, and technological change is a daunting task. To address this, we have used the GCAM-USA version of the Global Change Analysis Model, which includes the necessary sectoral, spatial, and technological detail to enable quantification of future energy-water flows under global change drivers through an internally consistent framework. We have evaluated a suite of scenarios that span a range of plausible climate and socioeconomic futures for the US. The modeling outputs have been leveraged to populate an open-source tool developed at PNNL to visualize complex sectoral and spatial dynamics through dynamic Sankey diagrams. These diagrams are an effective tool to illustrate the flow of key resources, fuels, and commodities from source to end use, capturing the impacts of scenario constraints on technological transformations and resource allocation decisions. Energy-water flows exhibit significant change across climate and socioeconomic futures over the 21st century. In futures with lower radiative forcing, the amount of electricity needed grows over time to simultaneously meet the dual challenges of supporting a growing population and addressing climate change . The role of renewable technologies becomes more pronounced in the future, with transformation pathways showing a switch from dominantly used natural gas to wind and solar resources for electricity production. Similarly, tradeoffs between competing water uses become prominent as large magnitudes of water needed for thermoelectric cooling of power plants get amplified by irrigation requirements to grow biomass to meet electrification targets. Such relationships and relative dependence between energy and water use, especially for emerging technologies such as hyperscaler data centers, suggest that strategies for sustainable transitions would require flexibility in the production, transformation, and consumption of resources, technologies, and commodities. To ensure sustainable transitions, special attention will be needed to manage and minimize the grave and simultaneous pulls on both energy and water resources in the future.

India is the third-largest producer of electricity globally which is undergoing a significant transformation towards a renewable energy (RE)-dominated power system. At present, thermal power constitutes 61% of the nation’s installed capacity and 75% of its electricity generation, positioning India as the second-largest coal consumer and the third-largest emitter of CO2 worldwide. Aligning with global commitments to mitigate greenhouse gas emissions, India has set ambitious targets of achieving 450 GW RE installed capacity by 2030, with aspirations to transform into a RE-dominated system by 2050. This transition marks a fundamental shift from a conventional, firm power system to one characterized by intermittent, variable, and uncertainty associated with renewable energy system. Such a transformation poses significant challenges, including the need for balancing variable supply with variable demand, managing the economic and social impacts of stranded thermal power assets, addressing resource constraints driven by weather variability, and adapt to the rigidity of existing conventional power infrastructure. Addressing these challenges requires a comprehensive and integrated approach. One promising solution lies in the integration of supply-side management (SSM) and demand-side management (DSM) interventions in transition planning. SSM focuses on optimizing the generation and supply of electricity to match the demand, while DSM emphasizes strategies to reshape and manage consumer energy consumption patterns.

This study, using mathematical modelling approach, explores the effectiveness of integration of SSM and DSM strategies in mitigating the variabilities introduced by renewable energy sources. A linear programming-based mathematical model is developed to optimally match supply and demand by adopting economically attractive solutions. On the supply side, we have minimised the total cost of the electricity system by optimising the power generation mix of the system. On the demand side, interventions such as load shifting are employed to reshape load curves, enabling better alignment with variable RE generation. DSM options, including load curtailment and shifting, are analyzed under both incentive-based and penalty-based pricing strategies to encourage consumer participation. The Karnataka electricity system, a high renewable energy-rich state in India, is used as a case study for model implementation and validation. The results demonstrate that the integrated model effectively moderates demand variability, reduces high storage costs, and enhances the utilization of renewable energy capacities. Additional benefits include deferred capacity additions, improved utilization of existing infrastructure, minimized reliance on thermal power plants, and a reduction in overall demand variability. This study underscores the potential of integrated SSM and DSM interventions in facilitating a cost-efficient and sustainable transition to a RE-dominated electricity system.

Behind the meter rooftop solar plus storage (PVESS) has the potential to benefit the hosting customers by providing affordability, environmental, and reliability and resilience value. Whereas the bill reduction and environmental benefits of PVESS are well studied, its monetary resilience benefits are less understood. The increasing trend of power interruptions driven by extreme weather events heightens the need to understand these benefits. This study leverages various publicly available datasets to perform a cost benefit analysis of adding to determine the resilience value of PVESS for a typical single family home in each county in the continental U.S. We find that PVESS is very effective to technically mitigate interruptions across the country. However, the monetary benefits in the base case only cover about 14% of battery costs, with no county exceeding 60%. This is somewhat expected, given that PVESS provide other monetary benefits that are not part of this analysis. Through sensitivities, we find that higher frequency of extreme weather events roughly triples the resilience value of PVESS and that higher values of lost load double the same metric. Our sensitivity analysis shows that the benefit cost ratio of PVESS for customers living in areas with higher-than-average frequency of long duration interruptions and value of lost load is already above one even without considering other value streams. We conclude with recommendations that regulators and utilities could implement to enable customers to calculate and capture the resilience value of PVESS more efficiently.