No New Tunnel Downtown? Sound Transit Explores Ballard Link Alternatives

Earlier this year, Sound Transit revealed updated cost estimates for its major ST3 projectstotalling $14 to $20 billion (in 2025 dollars) morethan its current long-range financial plan can afford. Over half of that excess cost is due to a massive jump in costs to build the planned Ballard Link Extension, which was expected to cost about $11.9 billionbut is now estimated at $20.1 to $22.6 billion. Much of the cost of the project is related to construction of a second tunnel through Downtown Seattle. While advocates havelong called for the agency to study alternative routesfor the Ballard Link Extension, Claudia Balducci was the first Sound Transit Board member to openly call for study of how the Ballard Link Extension could be built without a second tunnel through downtown. Last week, agency staff presented the results their investigation into the feasibility of two ways to build the Ballard Link extension without a second downtown tunnel: interlining with the current tunnel, or stub-ending underneath Westlake

Two Alternative Approaches to the Ballard Link ExtensionConcept Slide comparing ST3, Interlining, and Stub-end configurations for Ballard Link

The current ST3 plan involves building a new dedicated rail tunnel from Seattle Center via SLU to Chinatown. This new Ballard Line would continue south towards SeaTac and eventually Tacoma.

• 1 Line: Tacoma-Ballard in new tunnel
• 2 Line: Redmond-Mariner in existing tunnel
• 3 Line: West Seattle-Everett in existing tunnel

All lines were planned to run every 6 minutes peak, 10 minutes off-peak. However, it is the most expensive option and will likely at a minimum require truncating the line at Smith Cove due to insufficient funds.

Interline and Stub-end concepts map in detail

Both the Interline concept (Alternative 1B) and Stub-end concept (Alternative 2) would build the same path and stations as the existing ST3 plan for Seattle Center, South Lake Union and Denny. They would start to differ south of Denny.

The Interline concept would instead join the existing downtown tunnel at Pine Street and use Symphony Station as the first shared station.

• 1 Line: Tacoma-Ballard in existing tunnel
• 2 Line: Redmond-Mariner in existing tunnel
• 3 Line: West Seattle-Everett in existing tunnel

The Stub-end concept would follow virtually the same path as ST3 until downtown, where it would instead terminate at the 2nd Westlake station beneath the existing Westlake station.

• 1 Line: Tacoma-Northgate* in existing tunnel
• 2 Line: Redmond-Mariner in existing tunnel
• 3 Line: West Seattle-Everett in existing tunnel
• (5) Line: Westlake-Ballard in stub tunnel

Interlining in Depth

The Interline concept would merge into the existing downtown tunnel at Pine Street and 3rd Avenue. Symphony Station would be the first shared station. This is where transfers from the Ballard line will take place.

Alternative 1A Diagram

There were a couple different alternatives for the Interlining concept. The first Alternative 1A had some operational constraints with northbound trains to Ballard blocking southbound trains from Northgate.

Alternative 1B Diagram

Sound Transit instead focused on the main Alternative 1B (shown above) which would instead completely separate the motions between trains to Ballard and trains from Northgate. This would have much fewer operational concerns. The segment from Symphony station to Chinatown/ID station will have to share all three train lines.

(There was a third Alternative 1C for the interlining concept which might be discussed further in a separate article. It would construct a new cut and cover (C&C) connection to the DSTT following the alignment of the DSTT Stub Tunnel, the abandoned bus tunnel access to the former Convention Place bus station. Sound Transit suggested a sharp 100 degree turn to avoid the Seattle Convention Center which was labeled as a fatal flaw.)

This alternative will require property acquisition both east and west of 3rd Avenue. A second major drawback of the interlining concept is that construction of the T junction will require closure of 1 and 2 line between Symphony and Westlake stations likely for a couple years. 3rd Avenue will likely need to be closed to bus routes for at least 10 months as well.

No extra maintenance facilities will need to be built because this Ballard segment will still be connected to the original tunnel and the existing OMFs.

Service Pattern and Frequency Bottleneck by Tunnel

• 1 Line: Tacoma-Ballard in existing tunnel
• 2 Line: Redmond-Mariner in existing tunnel
• 3 Line: West Seattle-Everett in existing tunnel

Sound Transit states that the 1, 2, and 3 Lines using the interlining concept will maintain a similar service pattern to what was planned under ST3. However, utilizing only a single tunnel will create a bottleneck, limiting train frequency during peak periods. This will require upgrading the downtown tunnel.

	ST3 plan	Interlining headway (no upgrade)	Interlining trains per hour
1 Line	5 min (new tunnel)	7.5 min (existing tunnel)	8 trains per hour
2 Line	8 min (existing tunnel)	10 min (existing tunnel)	6 trains per hour
3 Line	8 min (existing tunnel)	10 min (existing tunnel)	6 trains per hour
max frequency	3 min	3 min	20 trains per hour

For example without upgrading the downtown tunnel this might constrain the 1 Line to ~7 min frequency and the 2 and 3 line to only 10 minute frequency if all three are running in the tunnel simultaneously. This is assuming a max of 3 minute frequency since Sound Transit has previously expected both 2 Line and 3 Line to run with 6 minute frequency with the existing tunnel.

	Interlining (no upgrade)	Interlining (upgraded)	Interlining (upgraded)
1 Line	7.5 min (existing tunnel)	5 min	12 trains per hour
2 Line	10 min (existing tunnel)	6.66 min	9 trains per hour
3 Line	10 min (existing tunnel)	6.66 min	9 trains per hour
max frequency	3 min	2 min	30 trains per hour

With some tunnel upgrades this could potentially be upgraded to 2 minute frequency. Specifically Sound Transit notes that implementing CBTC (Communications-Based Train Control) would allow shorter train headways, higher service frequency, and
faster recovery from disruptions. This would allow the 1 line to run around 5 min frequency and the 2 and 3 line at ~7 min frequency.

Interlining Easy Transfers

One major advantage of the Interline concept is that many transfers are much easier. Especially travelers heading in the same north/south direction could just wait on the same platform. For example people heading from SeaTac to Northgate would either need to climb up to the mezzanine and down to the platform at SODO or transferring at the 2nd Westlake station climb 4~5 flights of floors to reach the original Westlake station. Instead with the interline concept one could just wait at the same northbound platform.

Stub-end in Depth

The Stub-end concept will have the Ballard line end at a new terminal station north of 5th Avenue just next to the existing Westlake station. Note this is pretty similar to the existing ST3 plan following the same route from Denny to Westlake and almost the same new secondary Westlake station at 5th Ave.

One notable difference between this Stub-end concept and the previous 2014 Stub concept by SDOT is that the 2nd Westlake station is built underneath the existing tunnel. This greatly raises construction costs and limits savings, but future-proofs it for a future tunnel extension.

Service Pattern and Transfers

• 1 Line: Tacoma-Northgate* in existing tunnel
• 2 Line: Redmond-Mariner in existing tunnel
• 3 Line: West Seattle-Everett in existing tunnel
• 5? Line: Westlake-Ballard in stub tunnel

Riders to and from Ballard would instead need to transfer at Westlake station. A silver lining is that travelers from Northgate/Lynnwood to SeaTac/Federal Way would keep their existing one-seat ride.

Frequency Bottleneck by Number of Trains

Implementing the Stub-end concept is complicated by two factors: the need for tunnel upgrades similar to the Interline concept, and a serious train shortage made worse by the route change.

The new concept plan requires Tacoma trains to run further-to Northgate/Lynnwood/Everett-instead of to the original Ballard destination. Running service all the way to Everett is not cost-free and risks inadequate train frequency due to the existing lack of rolling stock. To limit the number of new trains required, Sound Transit proposes a compromise: terminate the Tacoma line at Northgate.

For this alternative, the Northgate Station pocket track would be utilized as an interim terminus for trains from Tacoma. If Communications-Based Train Control (CBTC) is implemented and trains run under semiautomatic train operation, 1 Line trains could be turned at Northgate without delaying 2 Line or 3 Line trains.

Missing OMF

The Ballard stub is isolated because it lacks a connection to either SODO or the existing tunnel. The lack of a connection means that trains from the existing OMF's cannot reach the stub. Therefore, the stub-end concept requires the construction of an extra maintenance facility.

Sound Transit investigated the Interbay Armory Site as a location for a new Operations and Maintenance Facility (OMF). They determined that running the isolated Ballard stub would require a fleet of at least 40 trains, which in turn necessitates a facility of at least 7.3 acres (ideally 10 acres). Since the Interbay Armory offers approximately 13 acres, the site has sufficient space to meet this requirement.

ST3 as planned versus Stub-end concept with linesFuture Extensions

Both the Interlining and Stub-end concepts allow for a future second tunnel to be built. For both concepts TBMs would be launched from the south at the Massachusetts Street portal. In order to take the TBM out of the ground short extensions will be built south of their destination.

For the interline concept, a new cut-and-cover stub tunnel could be dug branching south of Denny Station. The TBM would be retrieved once it reaches the stub tunnel.

For the stub-end concept a short Sequential Excavation Method (SEM) tunnel section would be dug south of the 2nd Westlake station. The TBM would also be retrieved once it reaches that short tunnel.

Ballard Link Concept comparison

	ST3	Interlining	Stub-end
Cost	most expensive truncate to Smith Cove	cheaper* easier to reach ballard	cheaper* easier to reach ballard
OMF	Can use existing OMF	Can use existing OMF	Must build new OMF in Interbay
Existing Operational Impact	Operational impact low.	Operational impact high. Need to close Lines 1 / 2 for multiple years	Operational impact low
Constructability	High impacts Requires building 2nd Westlake, Pioneer Square, and Chinatown Stations	High impacts Requires building T junction at 3rd/Pine	Low impacts Requires building 2nd Westlake station
Transfers (Prioritizing Northgate to SeaTac as the most traveled corridor)	Worst transfers Multi-story climb to transfer at Westlake	Best transfers Same platform transfers for same direction	Okay transfers Maintains Northgate to Seatac trains but all Ballard travelers must transfer
Service Pattern	Highest frequency 1 Line 5 min 2 Line 6 min 3 Line6 min	Lower frequency (without upgrades) 1 Line ~7 min 2 Line ~10 min 3 Line~10 min	Lower frequency (without upgrades) 1 Line ~7 min 2 Line ~10 min 3 Line~10 min

Cost Savings

The cost savings for both of these concepts are quite large.

Interlining concept (Alternative 1B)Stub-end concept (Alternative 2)

The Stub-end concept would defer the SODO to Westlake segment saving $7.2 billion dollars. The Interlining concept plus the 2nd Westlake station would save $8.4 billion dollars. However these savings would be offset by additional property takings and construction cost for building the new T junction or the new OMF for around $1.5~$2.0 billion. The interline concept would also require a bus bridge for a couple years. These costs seems reasonable.

	Interlining	Stub-end
Savings	-$8.4B (skip westlake station)	-$7.2B (includes 2nd westlake station)
ROW acquisition	$0.3B to $1.3B	$0.3B to $0.4B
T junction/ OMF	$0.7B to $1.0B	$0.8B to $1.2B
Bus Bridge	$0.5B to $0.75B
Tunnel upgrades	$1 to $3B	$1 to $3B
Project Delay (ignored)	$1.2B to $2.4B	$1.2B to $2.4B
Total savings	$2.6B to $5.9B	$2.6B to $5.1B

Next Sound Transit lists the cost of tunnel upgrades and project delay. On closer inspection, these incredibly high estimates seem severely inflated in order to claim that neither the interline concept nor the stub-end concept would garner any savings at all. It's a bit too much of a coincidence that sum of all high end estimates at $21.4 billion magically match with the same cost as the original ST3 Ballard Link Extension at $21.4 billion.

First there is the cost of upgrading the existing tunnel with ventilation, egress exits, and CBTC as previously explained. Sound Transit has estimated the upgrade at a wide range of $1 billion to $3 billion. This high estimate seems a bit incredulous that implementing these changes would be more expensive than building the Ballard bridge segment.

Secondly there is a supposed 24-48 month project delay that will cost an additional $1.2 to 2.4 billion dollars. However, this a bit deceptive to include project planning delays considering Sound Transit does not even have enough money for current estimates of West Seattle Link at $12 billion dollars, and are already planning on truncating West Seattle Link to Delridge Station due to lack of funds. With the current Ballard Link estimates at $22 billion it doesn't really matter if planning takes longer since Sound Transit already lacks sufficient funds to build the full Ballard Link

Counterarguments

Sound Transit staff and some board members continue to argue strongly for retaining the original ST3 plan, asserting that the second downtown tunnel is a vital investment for the long term system. They caution that relying solely on the existing tunnel creates significant operational risks because the lack of crossovers and the current block-based signaling system severely limit resilience and the ability to recover from disruptions. The second tunnel also provides operational flexibility and establishes more manageable line lengths as well as fewer maintenance facilities (unlike the Stub-end concept).

Conclusion

The Ballard Link Extension, as planned under ST3, is not financially viable under current cost estimates and will at minimum truncate at Smith Cove. (Similar to the likely West Seattle Link Extension truncation to Delridge Station ) These Ballard Link alternative concepts of interlining and stub-end can be difficult to build, but with such large cost savings they are necessary to have a remote chance of reaching Ballard. Sound Transit and Seattleites now face difficult choices between:

Option A (current plan): Accept the cost overrun, skip reaching Ballard, and truncate Ballard Link at Smith Cove
Option B (Interlining): Choose significant cost savings at the expense of multi-year closures of the existing Line 1 and Line 2 downtown for construction
Option C (Stub-end): Choose significant cost savings with the tradeoff of mandatory transfers for Ballard riders and large land acquisition for a new maintenance facility.

References

• Second Downtown Tunnel Feasibility Assessment 12/11/25
  • Main report. Warning 110 page pdf
• Enterprise Initiative: System Performance & ST3 Policy Review Executive Committee 8/7/2025
  • (All lines planned to run every 6 minutes peak, 10 minutes off-peak.)
• UPDATE: Long-term light rail fleet & service challenges Community Oversight Panel 4/10/24
  • (1 Line: 5-minutes peak, 2 & 3 Lines: 8-minutes)
• C-07: Transit Tunnel (International District to Northgate) improvements enabling increases in system frequency
• Transit Expansion Study Ballard to Downtown Seattle 05/16/2014 Final Report