Summary

  • Climate scientist Daniel Swain and U.S. Geological Survey research geologist Jonathan Warrick identify the alignment of an incoming Kelvin wave, a super El Niño, and 8 to 12 inches of climate-driven sea-level rise as the binding variable for the 2026–2027 winter coastal damage forecast.
  • Substrate evidence places compounding of multiple drivers above either the Kelvin wave alone or El Niño storm patterns alone in causal ranking; two inversion tests support the ranking.
  • The California coastal system is fragile under the projected conditions: a threshold-dependent damage function, concentrated transportation and drainage exposures within six feet of sea level in the San Francisco Bay Area, and a bounded emergency-preparation balancing loop operating against three reinforcing loops.
  • The season’s discriminating signals will be the count of storm events arriving inside two distinct high-tide windows — November through December per the California Coastal Commission and January through March per Warrick — and whether atmospheric river landfalls coincide with peak astronomical tides.

California coastal communities are preparing for a winter in which three independent oceanographic and atmospheric drivers converge along the state’s coastline. Climate scientist Daniel Swain has stated that “coastal flooding is at this point practically certain,” while USGS research geologist Jonathan Warrick has stated that the systems “will pack a far bigger punch” together than separately. The substrate places the binding variable not in any one driver but in their alignment, with Marin County chief resilience officer Nick Brubaker reporting that the new information “certainly changes our baseline assumptions on how we’re going to be planning for impact.” Two of the three drivers — the Kelvin wave and the long-running sea-level rise — are documented in the substrate as insufficient by themselves to produce the projected damage, while the El Niño storm season operates against a baseline that has already crossed local damage thresholds in places. The structural accounting that follows ranks the drivers, audits the fragility profile, and maps the loops that govern what the next six months will deliver.

Why alignment, not any single driver, is the binding variable

The substrate supports a compound-causation reading over either of the single-driver readings. Mike Jacox, a research oceanographer with the National Marine Fisheries Service, has stated that on their own, Kelvin waves “aren’t always hazardous.” Jacox has separately stated that “when there are very high tides or storm surges, their effects can be amplified” — language that ties the wave’s damage to coincident events rather than to its presence. Swain has stated that in places where freshwater and saltwater combine, “the damage can be devastating,” and the article reports the Kelvin wave’s foot of elevation “would compound” the underlying 8 to 12 inches of rise. Warrick’s statement that the systems “will pack a far bigger punch” supplies the direct substrate support for the compound reading.

Two inversion tests discriminate the ranking. Removing the Kelvin wave while holding sea-level rise and El Niño fixed leaves the January high-tide flooding Marin County saw “without any added effects from a Kelvin wave” — a documented baseline state. Removing the El Niño while holding the Kelvin wave and sea-level rise fixed leaves manageable wave amplification that, per Jacox’s framing, remains below the damage threshold in most conditions. Brubaker’s statement that the wave “changes our baseline assumptions on how we’re going to be planning for impact” is consistent with a marginal-contribution reading in which the Kelvin wave is the variable that pushes ordinary tidal and storm events past local capacity thresholds.

Two baseline observations reject the null hypothesis that the season produces no significant departure from recent winters. Swain has reported that “just the first few inches of elevation in southern California have already caused enough erosion to close the Amtrak line along the coast, and this is just the beginning.” Brubaker has reported high-tide flooding in Marin County in January without any added effects from a Kelvin wave, noting that “we could see some significant flooding” as creeks and streams become less able to drain. Each observation establishes that a stressor is already producing measurable damage before either the Kelvin wave or the El Niño arrives.

What each driver does and does not do in isolation

The Kelvin wave is a band of warm ocean water that swells tides from beneath the surface, moving as a slosh of warm water across the Pacific before colliding with the South American coastline and surging north toward Alaska. Its behavior differs from a surface wave or tsunami; it does not climb and crash. Per Jacox, Kelvin waves “can reduce the growth of phytoplankton that form the base of the marine food web.” Their amplifying effects are conditional on coincident tides or surges.

El Niño is characterized by high sea surface temperatures in the central and eastern tropical Pacific. The Pacific Ocean surface temperatures now reported exceed those recorded before the 1997–1998 winter, raising the likelihood that elevated sea levels will coincide with severe storms during the year’s highest tides in November and December. The El Niño has already fueled destructive storms in Hawaii and unleashed Pacific hurricanes that have hammered beach towns.

Climate-driven sea-level rise has added 8 to 12 inches to California’s coast over the past century. The California Coastal Commission memo cited in the substrate identifies that the highest tides of the year in parts of California arrive in November and December, when the state tends to be hit by wet weather. The November–December highest-tide window per the California Coastal Commission and the January–March storm-and-tide window per Warrick are two distinct high-tide windows that bracket the storm season and strengthen the alignment claim rather than dilute it.

The fragile profile of the coastal system

Taleb and Douady (2012) characterize fragility as concave sensitivity to stressors — a system losing disproportionately more from rare large stresses than its typical performance would predict. The California coastal system, as described in the substrate, exhibits this concave response profile to the stress combination now arriving.

Concave exposures — those that lose from volatility — are concentrated in three primary locations plus several hidden exposures. First, transportation and population infrastructure within six feet of sea level in the San Francisco Bay Area, identified by Swain as a concentrated exposure where “there’s a huge amount of people and infrastructure within 6ft of sea level there.” Second, the Amtrak coastal line in Southern California, already closed by erosion from climate-driven rise alone, with Swain stating that “just the first few inches of elevation in southern California have already caused enough erosion to close the Amtrak line along the coast.” Third, the drainage network, where Brubaker reports that “engorged creeks and streams will be less able to drain as sea levels rise.”

Hidden concavities include the freshwater–saltwater mingling Swain describes — “both of these in some places can combine where you have freshwater outflows and saltwater inflows mingling” — where the damage mode extends beyond the flood event itself. Insurance and reinsurance repricing has accelerated in coastal California in recent years, transferring rather than eliminating the cost of future events. Federal disaster declaration and FEMA reimbursement function as a downstream backstop rather than a real-time buffer. The Amtrak coastal line closure illustrates concentration risk in transportation corridors — single-line dependencies that amplify the effect of any given closure. Brubaker’s drainage observation applies to both coastal and estuarine environments, and is consistent with a concavity the existing drainage network cannot fully resolve.

Antifragile elements — components that gain from volatility — are limited in the substrate. Emergency-management preparedness qualifies as a short-term adaptive gain: sandbag distribution, drain clearing, and infrastructure evaluation across Bay Area jurisdictions. Distributed drain and culvert clearing, public emergency-alert signup drives, and the practice of distributing sandbags and flood barriers ahead of known events all incrementally reduce fragility with each cycle. The broader coastal system is not gaining from the disorder; it is amortizing exposure into federal and state disaster recovery, which transfers rather than eliminates the cost.

The damage function appears threshold-dependent rather than linear: small increments in sea-level elevation produce outsized damage once a particular elevation is crossed, as the Amtrak closure and the Marin January flooding together suggest.

Reinforcing loops versus bounded preparedness

Three reinforcing loops and one balancing loop operate across overlapping timescales.

Reinforcing loop R1 — sea-level rise and exposure. Climate-driven sea-level rise produces a higher baseline water level, which produces larger tidal and storm-surge amplitude reaching infrastructure, which produces more frequent inundation events, which produces increased pressure to harden or retreat from coastal exposure, which produces uneven adaptation that closes only part of the gap, which produces continued exposure to the next storm event. This pattern matches a “Fixes That Fail” archetype widely attributed to Senge (1990): each adaptation defers a problem that the underlying driver continues to produce.

Reinforcing loop R2 — erosion and reduced wave buffering. Storm wave action erodes beaches and bluffs, producing narrower beaches that absorb less wave energy, so that the next storm reaches further inland, producing more erosion. A related expression runs from higher baseline to more frequent coastal erosion to narrower beaches and reduced natural wave buffering to higher effective baseline at the infrastructure interface. A separate R2-alt runs from storm damage coincident with high tides to infrastructure failure to reduced drainage and protective capacity to greater damage from the next coincident event.

Reinforcing loop R3 — insurance, capital, and tax base. Rising coastal risk produces higher premiums or non-renewal, producing property value pressure, producing reduced local tax base, producing reduced infrastructure investment, producing greater residual vulnerability.

Balancing loop B1 — emergency preparation and response. A heightened risk signal produces emergency-management preparedness — drain clearing, sandbag distribution, infrastructure checks, public alert signups — which produces reduced damage from any given event. Brubaker’s statement that the new information “certainly changes our baseline assumptions on how we’re going to be planning for impact” represents a within-season adjustment of this balancing loop. The loop is bounded: emergency preparation reduces damage from any given event, but the underlying stressor trend raises the operating point faster than preparation lowers it. The balancing loop operates with a long delay relative to the storm events it is intended to mitigate, and its leverage is bounded by preparedness measures that do not alter the underlying stock variables.

Timescales and structural observation

In the immediate October-through-March window, the emergency-preparation balancing loop dominates, with the storm-and-tide reinforcing loop waiting to dominate if alignment occurs. In the medium-term one-to-five-year window following the season, the recovery-and-repricing loop dominates as insurance markets, FEMA allocations, and local-government balance sheets process the season’s losses. In the long-term five-year-plus window, the climate-driven sea-level rise loop and the insurance-and-tax-base loop govern the trajectory. The compound risk visible now is the short-term manifestation of those two long-timescale loops operating on multi-decade horizons, with the 8 to 12 inches of long-running rise as the substrate-level evidence that the loops have already produced measurable state change before this winter begins.

The system has three reinforcing loops in tension with one balancing loop, with a long delay on the balancing channel. The substrate identifies two distinct high-tide windows in different months — November–December per the California Coastal Commission and January–March per Warrick — that bracket the storm season. The balancing loop’s leverage is bounded by preparedness; the reinforcing loops compound across multiple storm events within a single winter.

Discriminating signals and candidate trajectories

Four candidate trajectories can be distinguished from the available evidence. The near-miss scenario assumes emergency preparations succeed, storm tracks keep the most intense systems offshore, and California’s coastline absorbs the season with localized damage. The 1997–1998 analog scenario assumes the layered stressors align with the largest tides and storms, reproducing the compound damage previously associated with levee failures north of Sacramento, harbor inundation across Santa Cruz and the San Francisco Bay Area, and federal disaster declarations across 21 of California’s 58 counties. The freshwater-dominant scenario treats coastal effects as secondary to atmospheric river impacts inland, where heavy rainfall in the Sacramento and San Joaquin watersheds drives flood damage through riverine and estuarine pathways; Swain’s note that freshwater outflows and saltwater inflows can “combine” in places where the two meet, and Brubaker’s observation that “engorged creeks and streams will be less able to drain as sea levels rise,” apply directly to the freshwater pathway. The cascading infrastructure failure scenario emphasizes that disruption to transportation and utility networks can produce damage substantially exceeding direct flood losses, particularly in the Bay Area.

Scenarios not represented in the available reporting — post-fire debris-flow risk following a wet winter after drought, regional drought-to-deluge transitions, and seismically triggered compound events — are not evaluated because the substrate does not supply evidence for them; the omitted connections continue to operate whether or not they appear in the structural accounting.

Discriminating signals most likely to separate the candidate trajectories: timing of atmospheric river landfalls relative to high tides; duration of inland rainfall; and whether storm surge coincides with peak astronomical tides at vulnerable Bay Area and Southern California locations.

What the substrate weakens

Historical storm-pattern planning, levee-and-seawall strategies designed against single-driver benchmarks, and the use of 1997–1998 as a baseline analog are each weakened by the substrate. The article reports current Pacific surface temperatures warmer than those recorded before the 1997–1998 winter, which means any planning baseline drawn from that season understates the present state. Warrick’s statement that the systems “will pack a far bigger punch” and Brubaker’s report that the wave “changes our baseline assumptions on how we’re going to be planning for impact” point toward planning frameworks in which alignment, rather than peak impact, is the binding variable.

Closing diagnostic posture

Warrick has stated that “this coming Kelvin wave is just the beginning,” that “the thing to look out for is what will happen in January, February and March – that’s when we get the biggest storms, we get our largest waves and we have some very large tides,” and that “when it comes to these storms and the tides, we really are kind of rolling the dice.” Brubaker has stated that “we’re using all of the data that we can, but we don’t have a perfect measure,” and that his focus has been outreach “to make sure that people are aware that this may be one of the most significant El Nino years that we’ve seen.” The dice are weighted by decades of accumulated exposure, and the probability distribution they implement is governed by loops operating far outside the immediate preparation horizon. Action is being taken within a bounded preparation loop while the larger reinforcing loops continue on timescales the preparation cannot reach.

Analytical techniques used in this piece

This analysis applies the methods below. Each links to a short, plain-English explainer you can read and reuse.

Differential Diagnosis
Lists the candidate explanations for a symptom and rules them out one by one.
Fragility / Antifragility Audit
Asks whether a system gains or loses from volatility, shocks, and disorder (Taleb).
Systems Dynamics (Structural)
Maps a system’s structure — stocks, flows, and the architecture that shapes its behavior.
Superforecasting (Tetlock)
The habits — calibration, updating, track records — that make some forecasters reliably better.