Aliso Canyon Workshop: Energy Division Pushes Inventory Cut as SoCalGas Warns of Supply Shortfalls

The CPUC’s first Aliso Canyon Biennial Assessment Workshop exposed a direct conflict over whether Southern California gas reliability depends on storage or on optimistic assumptions about pipeline performance.

Energy Division recommends reducing Aliso Canyon's maximum inventory by 10 Bcf (from 68.6 to 58.6 Bcf), the maximum step permitted under a 2024 decision (D.24-12-076). SoCalGas argues the current 68.6 Bcf maximum should be raised, not reduced. ALJ Jamie Ormond was present, steering the record toward safety concerns.

Energy Division's Assessment

Energy Division's presentation was led by Khaled Abdelaziz (natural gas modeling lead) with Eileen Hlavka covering the economic analysis and Jean Spencer making clarifying interventions during the Q&A session. The Energy Division assessment comprises four analyses required by D.24-12-076.

• Analysis 1 (Demand Reduction): This is a binary test comparing 1-in-10 peak day demand forecasts against a 4,121 MMcfd closure threshold established by a contractor several years ago. Forecasted peak demand exceeds the threshold in both study periods (by 441 MMcfd for winter 2025-26 and 76 MMcfd for winter 2030-31), which means that Aliso Canyon cannot be closed entirely. This analysis does not determine a minimum inventory level.
• Analysis 2 (Gas Balance Reliability): This analysis uses an in-house stochastic daily mass balance model. At pipeline supply capacity of 3,200 MMcfd (reflecting resolution of the L4000/L4002 Northern Zone outage), only about 1% of Aliso Canyon's inventory is needed (but some inventory is still required). Abdelaziz stressed this is "a necessary but not sufficient test" because it doesn't conserve energy or account for cost-mitigation withdrawals.
• Analysis 3 (Hydraulic Modeling): This analysis uses Synergi Gas 4.9.5 by DNV to simulate the full pipeline network hour by hour. For winter 2025-26, the simulation succeeded but required 550 MMcfd of continuous withdrawal from Aliso Canyon; without it, linepack collapses and curtailments become imminent. For winter 2030-31, Aliso Canyon is not needed, but only if three major system upgrades are completed, no unplanned outages occur, and demand reductions materialize as forecast. Aliso Canyon is not needed for summer high-demand days in either period.
• Analysis 4 (Economic Analysis): This is a threshold comparison designed as an off-ramp: if SoCal Citygate forward prices exceed 50% above Henry Hub and historical levels, the CPUC should consider maintaining inventory regardless of reliability findings. The threshold was not triggered for winter 2025-26. Hlavka noted that since the fall 2025 report, forward prices for winter 2026-27 have actually dropped due to mild weather and high storage inventories, meaning market conditions are now more supportive of reducing inventory.

Energy Division's bottom line: the CPUC should reduce Aliso Canyon's levels by 10 Bcf, though it hedges that a smaller reduction "may be appropriate" given LNG export uncertainty.

SoCalGas's Rebuttal

SoCalGas's presentation was split between Andrew Sawin (senior engineer, Gas Transmission Planning) on reliability and Michelle Dandridge (senior manager, Strategic Planning, Transmission & Storage) on economics. Their central thesis: the assessment systematically understates storage value through unrealistic supply assumptions.

• The main dispute is receipt point utilization (the percentage of available pipeline capacity that shippers actually schedule and deliver). Energy Division follows the 85/100% Receipt Point Utilization framework from Attachment A of D.24-12-076 (85% on Northern and Southern zones, 100% on Wheeler Ridge). SoCalGas calls this "unrealistic and imprudent," citing actual Receipt Point Utilization during recent storms: 47% in 2021, 38% in 2024, and 25% during Winter Storm Fern in January 2026. SoCalGas also argues that pipeline outages (L4000/L4002, L5000, L225) were not properly modeled. Their corrected supply assumptions show approximately 800 MMcfd less available supply than Energy Division assumed.
• Winter Storm Fern is SoCalGas's centerpiece: over 8 Bcf was withdrawn from storage in six days, with Northern Zone supplies at 9% of capacity, and Wheeler Ridge collapsing from 765 to 115 MMcfd due to the Castaic L225 landslide. SoCalGas's own mass balance and peak day analyses at corrected supply assumptions show deficits at both 58.6 and 68.6 Bcf; only 86.2 Bcf (the previous pre-leak maximum) produces a surplus.
• On economics, Dandridge argued the assessment's economic analysis is too narrow: it doesn't model storage's price-mitigation role, doesn't capture Unbundled Storage Program impacts, and uses forward prices that assume current storage levels, creating a circularity problem. A 10 Bcf reduction would cut the fully-subscribed Unbundled Storage Program from 25 to 15 Bcf, directly increasing customer price volatility and reducing ratepayer revenues.
• Dandridge also cited what SoCalGas characterized as an inconsistency in the biennial assessment's forward-look pricing for winter 2026-27, claiming that when corrected, prices actually breach the 50% economic threshold.

In sum, SoCalGas's position is that no reduction in storage levels should occur.

Q&A Session

The Q&A sessions exposed vulnerabilities on both sides.

• Energy Division's weakness is the economic analysis. Spencer acknowledged a circularity: "part of the reason prices are low is because storage is really high." She explicitly invited parties to address the limitations in testimony, suggesting that Energy Division wants help building a better framework.
• SoCalGas's weaknesses are in their assumption choices. Abdelaziz elicited that the L4000/L4002 pressure reduction (which anchored much of SoCalGas's reliability modeling) actually ended in summer 2025 and did not persist into the winter. SoCalGas used it anyway for "consistency."
• On the L225 Wheeler Ridge assumption for 2030-31, Sawin acknowledged that SoCalGas did not compute the joint probability of that outage coinciding with a 1-in-10 peak day. Abdelaziz noted prior work showed such joint probabilities are "very, very low" and questioned whether worst-case assumptions are consistent with a 1-in-10 regulatory standard.
• On Receipt Point Utilization, Abdelaziz provided context: during the original investigation, no stakeholder besides SoCalGas advocated below 85%. SoCalGas's corrected supply of 2,400 MMcfd against 3,600 nominal works out to approximately 67% ("the lowest capacity I've seen over the past six years").
• Dr. Issam Najm (Porter Ranch Neighborhood Council) also got Sawin to confirm that low Receipt Point Utilization figures reflect what customers chose to schedule, not whether gas was physically unavailable (complicating SoCalGas's framing of Receipt Point Utilization as purely a reliability metric).

Closing Orders

Ormond directed SoCalGas and Energy Division to file their presentation slides, with party comments due April 28 (10 pages max) and reply comments due May 5 (5 pages). She explicitly invited suggestions for improving the analytical methodology. More significantly, Ormond ordered SoCalGas to produce a narrative history of outages over the past decade that caused serious reliability concern i.e., events "that make people at SoCalGas's hair fall out," with a May 15 deadline.

INSTANT ANALYSIS

Three disputes will shape the outcome of this proceeding.

• First, Receipt Point Utilization, the single largest driver of the supply-calculation gap. Energy Division has the procedural high ground (Attachment A of D.24-12-076), but SoCalGas has Winter Storm Fern.
• Second, the economic analysis: both sides acknowledge it is rudimentary, and the party that offers the CPUC a credible framework (rather than just criticizing the existing one) may have outsized influence.
• Third, whether Aliso Canyon retirement by 2030-31 is realistic given Energía Costa Azul supply diversions, L225 vulnerability, and system upgrades that haven't been publicly identified.