Skip to content
← All retrospectives

CAISO · California

Three projects filed the same day. Two were built, one withdrew.

Colorado River Substation 230 kV

Sixteen interconnection requests since 2011 at one substation: three reached commercial operation, eleven withdrew, two remain active.

Built
3
Withdrawn
11
Still in queue
2

The queue record

Every interconnection request the dataset records at this point of interconnection, in filing order.

Interconnection requests recorded at this point of interconnection, with request date, capacity and outcome.
RequestedProjectTechnologyCapacityOutcome
2011-03-31COLORADO RIVER 48 MWSolar48 MWWithdrawn2012-04-26
2011-03-31MESA SOLARSolar500 MWWithdrawn2012-04-18
2011-03-31SP BLYTHE 1Solar221 MWWithdrawn2016-11-09
2011-11-15MESA VERDE SOLARSolar80 MWWithdrawn2013-05-06
2011-11-15BLYTHE MESA SOLAR IISolar500 MWWithdrawn2012-02-24
2012-04-02BLYTHE MESA SOLAR II - C5Solar500 MWWithdrawn2012-07-26
2012-04-02MULE MOUNTAIN - C5Solar150 MWWithdrawn2013-02-28
2015-04-30CRIMSONBattery350 MWBuilt2022-10-28 · 7.5 yrs
2015-04-30IRISHGas600 MWWithdrawn2017-09-05
2015-04-30ARLINGTONSolar + battery409.9 MWBuilt2023-10-20 · 8.5 yrs
2017-05-01QUARTZITE 10Solar154.59 MWWithdrawn2017-07-14
2017-05-01MESAVILLE SOLARSolar + battery476.33 MWWithdrawn2020-05-11
2018-04-16QUARTZITE SOLAR 11Solar + battery150 MWBuilt2024-12-17 · 6.7 yrs
2019-04-15CALYPSO SOLARSolar + battery400 MWStill in queue
2020-04-15AMETHYST SOLARSolar + battery200 MWWithdrawn2020-07-06
2020-04-15GRACE ENERGY CENTERSolar + battery508.56 MWStill in queue

What the record shows

  • CRIMSON, IRISH and ARLINGTON were all filed on 2015-04-30. Two reached commercial operation; the third withdrew in 2017.
  • The largest request in that cohort — IRISH, at 600 MW — is the one that withdrew. Capacity did not track survival at this node.
  • The two projects that were built from that cohort took 7.5 and 8.5 years from request to commercial operation.
  • Every one of the seven requests filed between 2011 and 2012 withdrew; five of them within about a year of filing.
  • MESAVILLE SOLAR, at 476.33 MW, stayed in the queue for three years before withdrawing in 2020.

What a decision record would have to make explicit

  • Same-day filings at one node diverged completely, so queue position and filing date alone did not separate them.
  • A decision record covering this node in 2015 would have had to state, as unknowns, the network conditions and study exposure that distinguished the three requests — none of which are in the public queue record.
  • The dataset does not record why any of these projects withdrew, so no cause can be attributed from this record alone.

What this does and does not show

This is a reading of a public dataset, not a customer case study: no engagement, customer or client is involved. The dataset records outcomes — what was built, what withdrew, and when — but not why any project withdrew, so no cause is attributed here. Nothing on this page claims GridVision predicted, influenced or would have changed any outcome, and a pattern at one point of interconnection does not generalise to another.

Source: Lawrence Berkeley National Laboratory, Queued Up: 2026 Edition — project-level interconnection queue dataset. Interconnection requests through year-end 2025. Used under CC BY 4.0; attribution to Lawrence Berkeley National Laboratory and GridTracker.