This is an interactive dashboard, and it needs JavaScript to show its results. If you opened it as an email attachment, save the file and open it in a web browser such as Safari or Chrome.

Wind energy could lower Alaska utility bills

Even without federal tax credits, wind energy could lower costs for GVEA members by displacing generation from expensive diesel and naphtha. Further savings can be unlocked by various operational improvements and commercial arrangements to sell surplus wind energy to Southcentral utilities. As that surplus energy flows south, it would offset scarce Cook Inlet natural gas, lowering costs for utilities across the Railbelt and, through the Power Cost Equalization program, for rural communities statewide. Wind additions would also diversify the Railbelt's energy supply, reduce greenhouse gas emissions, and improve reliability by adding much-needed capacity.

About this analysis

The purpose of this study is to advance our understanding of the economic impact of wind energy on GVEA's system, and to identify the best opportunities to save members money under a wide variety of potential scenarios.

Shovel Creek Wind is a project originated by Alaska Renewables, and now owned by Longroad Energy. Delta Wind is a project owned by Ameresco. This economic analysis was developed solely by Alaska Renewables using publicly available fuel and generation cost information filed by GVEA with the Regulatory Commission of Alaska, as well as publicly available wind resource data collected with funding from the Alaska Energy Authority (AEA). It is not a study by GVEA or the current project owner. Alaska Renewables built a custom dispatch model of GVEA's system and analyzed the operation and economics of hundreds of scenarios to compile these interactive results. The results assume no federal tax credits for wind energy.

Use the scenario controls below to explore for yourself the benefits of different energy futures.

Scenario controls

Shovel Creek Capacity iShovel Creek Wind is a proposed wind project 22 miles west of Fairbanks. The project was originated by Alaska Renewables, and is now owned by Longroad Energy, a nationally recognized investor / developer / owner / operator. 120 MW

Delta Wind Expansion iThe proposed 36 MW Delta Wind expansion project is owned by Ameresco. GVEA's Board of Directors have voted to move forward with a power purchase agreement term sheet for this project. Set Shovel Creek to 0 MW to evaluate Delta Wind on its own. 36 MW

2nd LM6000 iEvaluates building a second LM6000 turbine (2x1 combined cycle with the existing steam turbine) as part of the tested scenario. The baseline is always the current system without a 2nd LM6000. When on, results include the unit's operation and its capital cost. The 2nd LM6000 runs on ‘Quality Banked’ naphtha. Included

2nd LM6000 capital cost iSelect a capital cost assumption for the 2nd LM6000 of $80-$150M. The Methodology section explains how that capital cost becomes an annual cost to members. $100M

$80M$150M

Southcentral gas iSouthcentral utilities are expected to begin importing LNG by 2029 to meet their own needs. Those imports would allow GVEA to resume buying gas-fired power from Southcentral, on a timeframe likely sooner than a second LM6000 could be brought online. This option assumes GVEA can buy gas-fired energy from Southcentral utilities over the Railbelt intertie, up to 60 MW, and use it in place of its own oil-fired generation. The gas covers the most-used North Pole simple-cycle unit; that unit then takes over the duties of the next unit down, and so on, so the least efficient units stop running. Not included

Fuel-cost scenario iAll oil and gas fuel prices move with this selection.

• Liquid fuels are priced on GVEA's forecasted costs, from the Aug 13, 2026 Member packet.
• Gas for power from Southcentral is priced on the Railbelt Reliability Council's current forecast, low / mid / high. GVEA Mid case

Operational enablers

Implementing best-practice operational strategies to run the system more efficiently and minimize the use of expensive diesel fuel. The Methodology section lists the individual strategies.

Implementation iGVEA modeled a ladder of operational changes, each building on the last. With enablers on, the dashboard reports the lowest-cost set for the scenario you have selected, which is sometimes today's operations. The Methodology section lists every strategy in the ladder. Included

Commercial enabler

Surplus-wind sale price iGVEA could sell excess wind that would otherwise be wasted to other utilities on the Railbelt grid. Select a price assumption for that energy. 12 ¢/kWh

0¢7.5¢15¢

Levelized savings to GVEA members

-/yr

levelized annual savings (2025$)

present value of savings over the 30-year project life

Levelized rate impact

vs. the baseline

Residential member savings

at kWh/mo

Oil displaced

diesel + naphtha, per year

CO₂ emissions reduced

per year

GVEA's dispatched energy mix

Share of energy actually served to GVEA load by source (excludes surplus wind that is spilled or sold to other utilities). Even at the largest build, 120 MW Shovel Creek plus 36 MW Delta Wind, wind supplies only about 22% of the energy GVEA serves. The rest stays open to other sources, including low-cost natural gas if a pipeline is ever built.

Rate savings of selected generation across GVEA's fuel-price scenarios

Levelized rate savings (¢/kWh) for the current configuration under GVEA's low, mid and high fuel-price projections. Your selected scenario is highlighted.

Where the wind energy goes

2nd LM6000: utilization and cost of energy

Annual capacity factor

hours online per year

generated per year

Key Results

GVEA has several options to meaningfully reduce costs compared with its current generation system. New wind, a second LM6000 and resumed purchases of gas-fired power from Southcentral each lower costs on their own, and the right combination lowers them further.

Each option saves members money on its own

Levelized annual savings from each option by itself, compared with GVEA's current generation system. Choose a fuel-price case:

★ Lowest-cost option

-levelized rate reduction

-annual member savings

-present value over 30-yr life

Wind with resumed Southcentral gas purchases delivers the greatest savings

More wind saves members more money

Shovel Creek only With 36 MW Delta Wind Levelized savings at Mid fuel, $M per year. Hover a point for its value.

More wind, more savings

Still true with a 2nd LM6000

Still true with Southcentral gas

Wind is a hedge against high fuel prices

Cumulative savings to GVEA, discounted

Operational results: how GVEA dispatch changes

How each generator's annual dispatch changes from the baseline (no Shovel Creek, no Delta, no enablers) to the selected scenario. First: change in fuel burned (gallons/yr of diesel or naphtha, as each unit uses). Then: change in the number of unit starts. Green is an improvement (less fuel, fewer starts).

Adding wind sharply cuts the oil burned in GVEA's turbines and the hours those units run.

Change in fuel burned (gallons/yr)

Some units do see more starts:

Change in unit starts (per year)

If start counts are an operational constraint for a given unit, other strategies and adjustments, especially in how the battery energy storage system is dispatched, could be implemented to bring them down. Note the dispatch model was not optimized to reduce starts.

Operations unlock the value

GVEA could also procure flexible regulation services from efficient Southcentral gas plants instead of cycling its own oil units, a lower-cost path to integrating wind that further eases strain on GVEA's oil-based generators (see Benefits to Southcentral Alaska below).

Benefits to Southcentral Alaska

Interior wind can offset scarce Cook Inlet natural gas

Surplus wind generated in the Interior, energy GVEA can't use in the moment, can flow south over the Railbelt intertie and displace natural gas burned in Southcentral power plants. As Cook Inlet gas runs short and utilities turn to costly imported LNG, every avoided unit of gas matters for the whole grid.

Surplus wind exported

- GWh/yr

excess Interior wind sent south

- over 30 yr

Natural gas avoided

- BCF/yr

at a ~50%-efficient LM6000 combined cycle

- over 30 yr

Southcentral gas-cost avoided

$- M/yr

gross value of imported LNG not burned; fuel only, before the cost of supplying the wind

- over 30 yr

Imported LNG price $15 /MCF

$10$15$20

Net exchange with Southcentral

Why this matters across the Railbelt, and beyond

For decades, GVEA and the Interior imported low-cost, gas-fired power from Southcentral utilities across the Alaska Intertie. The decline of Cook Inlet natural gas has brought that practice to a near halt: Southcentral utilities face a supply shortfall as early as 2027 and are preparing to import liquefied natural gas (LNG) at a delivered cost significantly higher than historical Cook Inlet gas prices.

Wind built in the Interior can flip that flow. Surplus generation can move south to offset gas demand in Southcentral, stretching remaining Cook Inlet reserves, reducing the volume of expensive LNG that must be imported, and helping lower costs for every Railbelt utility and its ratepayers.

The Railbelt utilities' own study points the same way. The Alaska Railbelt Wind Integration Study, prepared by Energy and Environmental Economics (E3) for GVEA, Chugach Electric, Matanuska Electric and Homer Electric and released in 2025, found that the Railbelt can reliably integrate 300 MW of new wind, enough to serve over a quarter of its annual electricity needs. A full year of simulated 5-minute operations showed no loss-of-load events, and only about 1% of wind output was curtailed. E3 estimated that this wind would cut Railbelt-wide operating costs by $97 to $126 million a year by 2030, mostly by burning less natural gas, and lower Railbelt CO₂ emissions by about 25%.

Those results assume the Railbelt is operated as one coordinated system. E3 modeled the four utilities as a single balancing area and recommended moving toward integrated, Railbelt-wide unit commitment and dispatch, transmission scheduling without wheeling charges, and more flexible gas scheduling. Its sensitivities put more flexible gas scheduling at about $14 million a year of additional savings and stronger transmission ties between regions at about $13 million a year. E3's figures exclude the cost of the wind itself, so the study does not estimate the effect on rates; this dashboard adds that step for GVEA's members.

The intertie can also carry a two-way partnership. Rather than cycling its own oil-fired units to balance wind, GVEA could procure flexible regulation services from efficient Southcentral gas plants, a lower-cost way to integrate wind into the Railbelt that eases wear on GVEA's oil generators and saves members money. For Southcentral utilities, it is a chance to better recover costs from the flexible, efficient units they already own, turning idle capacity into revenue.

Those benefits reach beyond the road system. Alaska's Power Cost Equalization (PCE) program keeps electricity affordable for roughly 200 rural communities and 82,000 residents, and was created so rural Alaskans share in the same energy investments that lower costs on the Railbelt. A stronger, lower-cost Railbelt supports the broader statewide energy economy, and the communities PCE serves.

Methodology, assumptions & data sources

What's shown: the levelized impact on GVEA's rates for the selected scenario versus one common baseline: GVEA's existing generation fleet serving its projected load, with no new wind, no second LM6000 and no operational enablers. Loads in every scenario, including the baseline, follow GVEA's own load projections in its August 13, 2026 member book. Every scenario's annual cost is rebuilt directly from the dispatch model's raw outputs (non-fuel variable cost + regulation cost − negative-power revenue + wind PPA cost + each fuel's consumption × its levelized price). The saving is the baseline's annual cost minus the scenario's, plus surplus-wind sales. The rate impact is that saving spread over the load the baseline serves, which puts every scenario on one denominator, so the ¢/kWh figure and the present value always agree in sign. Negative = lower rates (savings).
Fuel-cost scenarios: the Low / Mid / High selector uses GVEA's own low, mid and high fuel-price projections presented at the August 13, 2026 Special Board Meeting. Levelized over 2030-2059 in constant 2025 $/MMBtu, the diesel blend is 23.0 / 28.7 / 37.7, the naphtha blend 15.4 / 19.3 / 25.3, and Southcentral natural gas 10.4 / 12.5 / 14.8 (coal held at $4.30). The gas projections are the Railbelt Reliability Council's latest, converted to real 2025 dollars. GVEA's forecast runs through 2045; for 2046 to 2059 this analysis holds each fuel's real price flat at its 2045 level rather than extending the trend, which would compound above inflation for another 14 years. Diesel is modeled as GVEA's historical ~90% high-sulfur / 10% ULSD blend. The second LM6000 would run on higher-priced QB naphtha. Based on GVEA's filings, the historical relationship between QB naphtha and diesel blend pricing is a ratio of 0.77, resulting in QB naphtha pricing of 17.7 / 22.2 / 29.1 (levelized over 2030-2059, 2025 $/MMBtu).
Levelized rate impact: levelized in constant 2025 dollars over the 30-year analysis window, 2030-2059, using a nominal discount rate of 5.0% and 2.3% inflation (real rate ≈ 2.64%). The model workbook levelizes fuel prices over 2030-2045, so the rate impacts shown here do not match the workbook's Impact on Rates tables. The annual dollar figure is that levelized rate impact times GVEA's annual load. When the second LM6000 is on, its capital charge, fixed O&M and capital maintenance are levelized over the same 30 years at the same rate (see 2nd LM6000 option).
Lifetime totals are present values (30 years, 2030-2059). The headline present value of savings over the 30-year project life, and every other lifetime dollar total on this page, covers the 30 years from 2030 through 2059, is expressed in constant 2025 dollars, and is discounted to 2030 at GVEA's 2.64% real rate. The window matches the 30-year life of the wind assets and the 30-year LM6000 loan, so no costs or benefits are cut off at the end. Each year is rebuilt separately. The scenario burns the same annual fuel volumes every year (MMBtu by fuel type, from the dispatch model), and those volumes are priced at that year's real $/MMBtu from GVEA's annual fuel-price forecast in constant 2025 dollars (diesel blend and naphtha blend by scenario; QB naphtha at a 0.77 ratio to the diesel blend). Each year's saving is the baseline cost minus the scenario cost, plus surplus-wind sales. The 30 annual savings are then discounted to 2030 and added up, using GVEA's own rate: 5% nominal and 2.3% inflation, which is a 2.64% real discount rate. The levelized rate impacts use the same 30 years, rate and 2030 base year, so both headline numbers are on one consistent basis. A saving in 2059 carries a weight of 0.47 relative to a saving in 2030. The year-by-year chart shows the annual savings before discounting; its caption gives both the undiscounted sum and the present value.
How each cost is treated over time: fuel follows GVEA's annual real price forecast. O&M and regulation are held constant in real 2025 dollars every year. The wind PPA is constant in real terms for its 25-year term and flat in nominal dollars after that, as described under Wind cost is included. No battery cost is charged, because the PACE-funded battery is assumed to be available (see Battery). The second LM6000's capital charge declines each year, its fixed O&M is held constant in real terms, and its capital maintenance is charged every fifth year, as described under 2nd LM6000 option. Surplus-wind sales are credited at the selected price, held constant in real terms.
Battery (PACE-funded BESS): operational stages 2-4 need battery capacity for regulation (10 MW at stage 2, 25 MW at stages 3-4). This analysis assumes GVEA's proposed 46 MW PACE-funded battery energy storage system is built and available for these regulation purposes. GVEA expects to build it regardless of Shovel Creek, under its USDA PACE award, so no battery cost is charged to any scenario.
Wind cost is included. Each scenario's cost embeds the wind PPA price (the project's price schedule converted to 2025 dollars) and regulation costs. Regulation had to be provided for all wind on the system, including Eva Creek and the existing Delta Wind. The Delta Wind expansion's price is an estimate, because its PPA terms are not public. Each wind PPA runs 25 years, through 2054, with its price constant in real terms. For 2055 to 2059 the price is held flat in nominal dollars, so its real value falls by about 2.3% a year. The PACE-funded battery is assumed to already be available on the system for wind regulation, so no BESS capital cost is charged. Costs of procuring firm capacity elsewhere are not included. Surplus-wind sales are credited on top at the price you select.
Southcentral gas: an option, not part of the baseline, and modeled by GVEA rather than estimated here. Southcentral utilities are expected to begin importing LNG by 2029 to meet their own needs. Those imports would allow GVEA to resume buying gas-fired power from Southcentral, on a timeframe likely sooner than a second LM6000 could be brought online. In these runs the Southcentral purchase is dispatched after the North Pole combined-cycle units and ahead of the oil-fired units, is allowed to provide regulation, and is limited to 60 MW with a 10% outage rate. It is treated as added load on the Southcentral pool's Eklutna units at an average heat rate of 8.7 MMBtu/MWh, so no minimum generation is assumed. GVEA pays for the gas burned, at the RRC's projected price for the selected fuel case, plus $100/MWh of non-fuel charges: CEA charged GVEA $45/MWh for energy and $98/MWh of other charges in 2024, and those non-fuel charges are assumed to continue, escalating with inflation. Because these runs redispatch the whole system, the oil displaced, the unit starts and the surplus wind all come from the model itself. Emissions net out both sides, subtracting the CO₂ from the gas burned in Southcentral at EPA's 53.06 kg CO₂/MMBtu. GVEA reran only two operating stages with Southcentral power, current operations and Backup NPSC regulation, so the stage control snaps to one of those when the option is on.
Surplus-wind sales: "surplus wind" is wind available but not usable on GVEA's system in the hour, so it would otherwise be spilled (curtailed). Instead it is exported to Southcentral. The price follows the same path as the wind PPA: constant in real terms for the 25-year contract term, then flat in nominal dollars, so it fades in real terms after 2054. Credit = each scenario's surplus wind (MWh) × the sale price (slider, 0-15 ¢/kWh; GVEA modeled $75/MWh = 7.5 ¢/kWh) ÷ annual load; 0 ¢/kWh means no sales. The operational enablers reduce how much wind is spilled, so sales are worth more with fewer enablers running.
Operational enablers (regulation upgrades) form a nested chain: Zehnder offline regulation (0 MW BESS) → +backup Zehnder (10 MW BESS) → +backup NPSC (25 MW BESS) → +real-time regulation sizing. Each was only modeled on top of the previous one. They change the dispatch (how much wind is used vs. spilled, and how much oil is displaced), which flows through the levelized cost. Each stage is described below.
Stage 1, Zehnder offline regulation: a unit is scheduled offline in the day-ahead commitment but committed to come online reliably within the 25-minute start time required for slow regulation. Scheduling certain units to provide offline regulation lets the system hold less spinning thermal reserve, so more low-cost wind can be accepted. Standard practice: non-spinning, quick-start reserves are a routine reliability product in organized markets.
Stage 2, Backup Zehnder regulation: a unit's full capacity is scheduled to provide up-regulation in the day-ahead commitment, and it is brought online in real time only if online and offline reserves drop below one hour of stored energy. A battery provides that one-hour buffer, so backup units run far less. Modeled with a 10 MW, 2-hour battery, on top of Zehnder offline regulation. Why it is reasonable: it is equivalent to the standby and supplemental reserve tiers used widely, and the battery simply guarantees ride-through while the unit starts.
Stage 3, Backup NPSC regulation: the North Pole simple-cycle units also provide backup regulation, in addition to the Zehnders. With the one-hour battery buffer, NPSC can fill this role, and its better heat rate lowers fuel burn when reserves are actually needed. Modeled with a 25 MW battery. A logical extension of the backup-reserve concept to the most efficient available units.
Stage 4, Real-time regulation sizing: instead of holding the reserve committed a day ahead, the system re-sizes reserve requirements in real time using the much more accurate 1-2 hour-ahead forecast. This cuts unnecessary backup-unit runtime and lets more wind be accepted. Best practice: look-ahead, forecast-driven reserve sizing is a core feature of modern real-time markets and advanced energy management system dispatch.
Emissions reduced (tons CO₂e/yr): the drop in annual CO₂-equivalent emissions versus the baseline. Each unit's modeled fuel use is combined with standard EPA CO₂ factors (distillate 73.96, naphtha ≈68, coal ≈97.2 kg CO₂/MMBtu). Beyond the environmental benefit, verified emissions reductions help GVEA unlock a large share of the federal funding under its already-awarded USDA New ERA grant.
Dispatched energy mix: energy actually served to GVEA load by source, excluding surplus wind that is spilled or sold to other utilities.
2nd LM6000 option: the baseline never includes a second LM6000. Turning the option on adds a second LM6000, making the North Pole plant a 2x1 combined cycle, to the tested scenario. Only the new unit runs on QB naphtha; the existing LM6000 continues on its current naphtha blend. The scenario uses the model's dispatch with the unit in service and charges the unit's capital cost. Set the capital cost with the $80M-$150M slider. How the capital charge is built: GVEA's August 13 member book finances the unit at 5% interest over 30 years and recovers it through rates as depreciation + interest + a TIER margin. TIER (Times Interest Earned Ratio) is the interest coverage that co-op lenders require; the member book collects it as interest × 0.79. GVEA's first-year utility-rate cost is $10.3M at $80M and $18.6M at $150M, interpolated linearly in between. That first-year figure is not the cost in every year. Interest and the TIER margin are charged on the remaining loan balance, which falls as 1/30 of the capital is depreciated each year, and the charges are in nominal dollars that inflation erodes at 2.3% a year. The dashboard therefore rebuilds the charge year by year: depreciation, plus 1.79 × 5% × the remaining balance, plus the small remainder of GVEA's first-year figure held flat in nominal dollars, all converted to 2025 dollars. At $80M the real charge falls from $10.3M in 2030 to about $1.7M in 2059, when the loan is nearly repaid. The lifetime present value uses this declining series year by year. The levelized rate impact uses its levelized equivalent over 2030-2059, about $5.9M/yr at $80M and $10.5M/yr at $150M, divided by annual generation. Fixed O&M and capital maintenance: this analysis adds fixed O&M of $17/kW-yr on the unit's 43 MW nominal gas turbine rating from its air permit, about $0.7M a year held constant in real terms. The roughly 60 MW capacity addition also counts extra output from the existing steam turbine, which is not included in this charge. It also adds the member book's required capital maintenance of about $12.5M every five years, in 2025 dollars. It falls due every fifth year, and is charged here as a level annual amount with the same present value, about $2.4M a year, so the year-by-year savings curve stays smooth. With fixed O&M the two together add about $3.1M a year. The model workbook also carries a flat $10M a year capital cost in every LM6000 scenario; this analysis removes it so the unit's capital is counted only once. Operationally the unit lowers system cost (cheaper naphtha displaces diesel), so its net effect depends on the capital assumption. This option does not include any costs for refinery upgrades or fuel storage needed to supply the additional unit with sufficient quality-bank (QB) naphtha. For reference, GVEA projects the second LM6000 alone would lower rates by roughly $13-18/MWh, while Shovel Creek without it can lower rates by up to ~$30/MWh (Shovel + Delta up to ~$33/MWh) in the 30-year High fuel case.
How this is built: a single script reads GVEA's scenario workbook and writes the three data files the dashboard uses: each scenario's annual non-fuel cost and fuel volumes, the dispatch of every unit, and the annual fuel-price series. Every assumption applied outside the workbook is written alongside them, so any number here can be traced back to a workbook cell or a stated assumption.
Sources: fuel volumes and generation come from a custom hourly dispatch model of GVEA's system; fuel-price projections and the wind/LM6000 rate-impact comparison are from GVEA's August 13, 2026 Special Board Meeting member book and GVEA COPA / 2025 RCA filings.

Fuel prices used in this analysis

GVEA's low, mid and high projections, shown in nominal dollars per MMBtu. The model itself works in constant 2025 dollars. Solid lines are the diesel blend, dashed lines the naphtha blend. GVEA's forecast ends in 2045; every later year holds the 2045 price flat in real terms.