PG&E Operational Flow Orders

Operational flow orders and emergency flow orders PG&E declared on its gas system: gas day, order type, reason (high or low inventory), stage, tolerance band in percent, noncompliance charge and customers affected.

PG&EOutages & Grid Events data CAISO market

Open in Data Ask the agent

One row per: gas day x order type x reason; Event (per gas day) cadence
Evidence: Observed market data; Published observations from the named source
Time basis: Date-labelled records; UTC · event_time

Overview

A gas day with no row had no order. Published by PG&E.

Query contract: Preserve one canonical row per (gas_day, order_type, reason) before aggregating.
Analysis rules: Count orders by gas day; a gas day with no row had no order.
Interpretation: An OFO restricts shippers' imbalance tolerance; it is not by itself a curtailment of gas to power plants.Published by PG&E; attribute it to PG&E whenever it is reported.

Access

Download: Query it in SQL, download it, or read it with an API key. Download data
MCP: Connect an MCP client and load the contract with describe_data before calling query_data.CAISO MCP API reference
HTTP endpoint: POST /api/v1/datasets/pge_cgt_ofo_events/query
Full history: Delivered to Snowflake, Databricks or your own storage on a schedule by Data services.
More coverage

Schema

11 columns

Column	Type
`event_time`	`DateTime64(6, 'UTC')`
`gas_day`	`Date32`
`order_type`	`LowCardinality(String)`
`order_description`	`Nullable(String)`
`reason`	`LowCardinality(String)`
`stage`	`LowCardinality(Nullable(String))`
`tolerance_pct`	`Nullable(Float64)`
`noncompliance_charge`	`Nullable(Float64)`
`noncompliance_charge_text`	`Nullable(String)`
`customers_affected`	`Nullable(Int64)`
`ingested_at`	`DateTime64(6, 'UTC')`

Sample data

event_time	gas_day	order_type	order_description	reason	stage	tolerance_pct	noncompliance_charge	noncompliance_charge_text	customers_affected	ingested_at
2026-10-23 00:00:00.000000	2026-10-23	System Wide	System-Wide OFO	High Inventory	2	5	1	1	∅	2026-10-31 00:51:52.806007
2026-10-23 00:00:00.000000	2026-10-23	System Wide	System-Wide OFO	High Inventory	2	5	1	1	∅	2026-10-31 03:51:54.928901
2026-10-23 00:00:00.000000	2026-10-23	System Wide	System-Wide OFO	High Inventory	2	5	1	1	∅	2026-10-31 06:51:52.834344
2026-10-06 00:00:00.000000	2026-10-06	System Wide	System-Wide OFO	Low Inventory	2	9	1	1	∅	2026-10-31 00:51:52.806007
2026-10-06 00:00:00.000000	2026-10-06	System Wide	System-Wide OFO	Low Inventory	2	9	1	1	∅	2026-10-31 03:51:54.928901
2026-10-06 00:00:00.000000	2026-10-06	System Wide	System-Wide OFO	Low Inventory	2	9	1	1	∅	2026-10-31 06:51:52.834344

Coverage

Complete: 17.5%
Gap days: 8,605
Days with data: 1,820

Show coverage history

Coverage: 17.5%Days with data: 1,820Gaps: 8,605

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

2000

1999

1998

gaplessmore

Related data

Related outages & grid events data
Dataset	Description
PG&E Maintenance CapacityPG&E · Per posting	PG&E's posted maintenance on its backbone paths: path, segment, start and end dates, and the remaining capacity in MMcf/d during the work. Read the newest snapshot_date. Published by PG&E.
PG&E Outage AreasPG&E · 10-15 minute snapshots	The polygons PG&E draws for larger outages on its outage map, per build, as GeoJSON, with the same customers, cause and restoration fields as `pge_outages.outages`, joined on `outage_id` and `event_time`. Read with FINAL.
PG&E Outage MapPG&E · 10-15 minute snapshots	Every outage on PG&E's public outage map, one row per outage per map build (every 10-15 minutes): estimated customers out, cause, crew status, outage start, the estimated time of restoration (`current_etor`), circuit, city, county and a point. `kind` is `outage` or `psps` (a Public Safety Power Shutoff). An outage's episode is its rows across builds grouped by `outage_id`; the newest build is `event_time = (SELECT max(event_time) FROM pge_outages.outages)`. Read with FINAL.
PG&E PSPS FootprintsPG&E · 10-15 minute snapshots	The footprint of each Public Safety Power Shutoff on PG&E's outage map, per build: the event, its stage, customers and Medical Baseline customers affected, the de-energization window and estimated restoration, as GeoJSON. Empty between shutoffs. Read with FINAL.