Vol. I, No. 43 · San Francisco — Monday, September 28, 2026 · Price: half a foot above the tables

BY MAXINE DOLAN, CITY DESK

At 11:12 on the night of July 13, the water at the federal tide gauge in the Presidio rose past the line NOAA uses to count a day of flooding. It cleared the line by about a tenth of an inch. On June 14 it had done the same at 11:18 p.m. No night in June or July had ever done it before. In NOAA’s count at the station, which runs back to 1920, every earlier flood day fell between November and February.

San Francisco Bay has now topped that line on six days in 2026, in three bursts: four straight days at New Year’s, then two single summer nights. That is the most in any year since 1983. From 1920 through 1968 it never topped the line once. The Bay’s average level sits about nine inches higher than in the 1920s, so an ordinary big tide now does most of the work a winter storm used to do. Tides are predicted years ahead. The next test already has a date: the morning of December 24.

The margin is thin. July 13 cleared the line by less than the three-quarters of an inch NOAA gives as the accuracy of a single reading. Without that night, 2026 only ties 2005’s five. And 2005 started nothing: one flood day followed in 2006, then none from 2007 through 2013.

The line is NOAA’s threshold for minor flooding at this gauge, 1.87 feet above the average daily high tide. The agency calls water over it high-tide flooding, “sometimes referred to as sunny day or nuisance flooding.” Every flood day this year was minor. The highest, on January 3, still came within 3.4 inches of the gauge’s record high water, set on January 27, 1983.

It is a reading at one pier, and on its own it says nothing about which streets, if any, got wet. It is also not the only line. The National Weather Service keeps its own minor-flood mark at the same gauge, eight inches lower. The Bay crosses that one every year, summers included: on 541 days from 1980 through 2025, 117 of them in June or July. NOAA’s higher line marks the rarer water, and it is the one NOAA’s flood-day count uses. Every figure here is measured against it.

The gauge has measured the Bay since June 29, 1854. For the first 49 years of NOAA’s flood count there is nothing to count. Then come 32 flood days between 1969 and 2024, bunched into a few winters: eight in 1983, three in 1998, five in 2005.

What changed this summer is how little help the tide needed. NOAA’s tide tables predict each high tide from the pull of the moon and sun. Anything the Bay does beyond the tables comes from wind, air pressure, river runoff and the ocean itself. The gauge’s recorded high tides topped the line on 29 days from 1980 through 2024, two more than NOAA’s official count, which tallies days a little differently. Two of the 29 fell in March 1983, a month where NOAA’s count has none. On those 29 days the Bay ran a median 14 inches above the tables. A flood here usually took a big tide plus more than a foot of help.

The New Year’s flood days this January still had that kind of help, 11 to 15 inches. The summer ones had much less. On June 14 the Bay ran 8 inches above the tables, and on July 13 it ran 6.

Six inches is enough because of where the tables top out. They are built on the Bay’s average level from 1983 to 2001, and the highest tide they can produce falls 5.7 inches short of the flood line. The Bay has moved up since. Through August it ran an average of 4.7 inches above the tables at each day’s highest tide. It ran at least 5.7 inches above them on 70 of 235 days. In 2001 and 2003 that average sat within half an inch of zero, and in 2025 it was 2.7 inches.

Two things lifted the Bay, and the gauge cannot fully separate them. One is the long rise. NOAA puts the station’s trend at 0.78 inches a decade since 1897, and the Bay’s average for 2016 through 2025 sits 8.8 inches above its 1920s average. The other is this year’s ocean. Against the tables, the Bay has run about two inches higher in 2026 than in 2025. Swings that size come and go: the annual average fell 2.1 inches from 2023 to 2025.

This year’s swing came with a warming Pacific. NOAA’s El Niño index stood 0.4 degrees Celsius below normal for December through February and 1.8 degrees above it for June through August, the months of the two summer flood days. In an El Niño, NOAA says, sea levels on the West Coast are generally higher, from warmer, expanded water and shifted weather. That is no guarantee here. Nine of the 32 earlier flood days came in the El Niño winters of 1983 and 1998. The strongest El Niño in the index, peaking in the winter of 2015–16, brought none.

The tides are strong this year, but not rare. Through September the tables held eight predicted high tides within 2.6 inches of their ceiling. Eight earlier years since 1969 held as many or more across a full year, led by 1986 with 13 and 2022 with 11. Together those eight years produced two flood days. Last year the tables held five such tides, and the Bay, running two inches lower against them, produced none.

The Port of San Francisco says the city faces coastal flood risk today. It is planning shoreline heights for 3 to 7 feet of sea level rise by 2100. The gauge’s last century delivered about nine inches. Add a strong year for tides and a high year for the ocean, and nine inches was enough to put the Bay over NOAA’s flood line on a July night.

The next candidate is already printed. At 10:37 a.m. on December 24, the tables call for a high tide less than a tenth of an inch below their ceiling. If the Bay runs as far above the tables that morning as it has on three days in ten this year, it crosses the line again.

Confidence is high that NOAA counted six days, on its verified data through August 31. Preliminary readings for September show the Bay stayed below the line. Confidence is lower on July 13 alone, which cleared the line by less than one reading’s accuracy. It is moderate on what 2026 means for the years ahead, since a strong tide year and a high ocean can fade together, as 2005’s did. On July 13 the tables had called for a tide 6 inches short of the line. The Bay made up the difference on its own.

Four donors, two of them a married couple, gave $5 million in 13 days of April to one of them. San Francisco caps gifts to candidates at $500, but committees formed for ballot measures have no limit. In the same stretch of her last cycle, London Breed’s committees drew 5 percent of measure money.

BY CONSTANCE FEENEY, CITY HALL

Between April 1 and April 13, four people wrote checks of $1 million or more to a committee then called Clean Up City Hall, Mayor Lurie’s Ballot Measure Committee. Chris Larsen, executive chairman of Ripple, and his wife, Lyna Lam, each gave $1 million on April 1. Michael Moritz, the former Sequoia Capital partner, gave $2 million on April 7, and John Pritzker $1 million on April 13.

Since July 1, 2025, the two ballot-measure committees that carry Mayor Daniel Lurie’s name have reported $14.8 million in contributions. That is 49.5 percent of the $29.9 million given to every committee San Francisco lists as formed for a ballot measure, in a stretch that takes in this year’s June and November ballots. Labor and business political action committees that also spend on measures are outside that count, as are gifts of goods and services, in every cycle compared here. In the same months of 2023 and 2024, London Breed’s committees drew 5.1 percent. Half the money raised since last July to pass or defeat the laws San Franciscans vote on directly went to two committees carrying one officeholder’s name.

The size is possible because of how the rules are built. A candidate for city office may not accept more than $500 from any one contributor. A committee formed to pass or defeat a ballot measure has no such cap, even when an officeholder controls it. Berkeley once capped gifts to measure committees at $250. The Supreme Court struck that cap down on December 14, 1981, as a violation of the First Amendment.

The older committee filed from May 2025 as the Daniel Lurie Ballot Measure Committee, Safe, Affordable MUNI for a Thriving San Francisco. Since September 22 it has filed as Mayor Lurie’s Yes on H Committee, and it has taken in $4.3 million. The newer one took its first check on April 1 and has filed since September 21 as Mayor Lurie’s Yes on D, E & F Committee. It has taken in $10.5 million.

All four letters are on the November 3 ballot. D would make it harder to put measures on the ballot by petition. E changes how departments adopt technology and lets larger contracts skip the supervisors’ approval. F would let the mayor reorganize most departments and fire their heads without cause. H would authorize new revenue for transit. The June ballot carried a different Proposition D, a tax on outsized executive pay, and the fight over it drew the two biggest measure committees of the cycle after the mayor’s.

The money came in large pieces. Of 75 contributions from 67 donors, 21 were $250,000 or more, and together they make up $11.7 million, or 79 percent. Six were under $1,000; together they come to $360. Moritz, who funds the news site The San Francisco Standard, is the largest single giver at $2.49 million in two checks. He also gave $105,596 in goods and services, mostly a campaign mailer. Larsen gave $1.49 million, so he and Lam together match Moritz’s checks to the dollar, and Ripple Labs gave the Muni committee another $500,000. OpenAI, Anthropic and a committee sponsored by Airbnb gave $500,000 each, and Visa gave $250,000.

Mayors have run ballot committees before, at another scale. Ed Lee’s, first funded in 2012 and last used for Propositions J and K in 2016, raised $949,700 over its life. Breed’s two, for Proposition E in March 2024 and Proposition O that November, raised $1,181,572 between them. Take the mayor’s committees out of this cycle and everyone else raised $15.1 million, down from $16.5 million in 2023 and 2024. The growth in ballot money this cycle is the mayor’s. The mayor’s share follows the name on the filing, not who steers the money. Committees allied with a mayor that do not carry the name count as everyone else, in every cycle.

The city’s own regulator has been working on the gap. In April, Ethics Commission staff proposed applying the $500 candidate limit to other committees a candidate controls, whenever that candidate is actively running for city office. On August 14 the commission approved it, 4 to 0. It now needs a supermajority of the Board of Supervisors, and it would take effect no earlier than January 1, 2027, two months after the November vote. Whether it would ever reach committees like the mayor’s turns on when he counts as actively running. The draft gives an official who already controls such a committee 30 days after starting to run to shed money above the limit.

Confidence is high on the totals. They are the committees’ own reports of cash gifts, in their latest amended versions, covering contributions through September 19. Gifts reported since on the separate late-contribution form are not in them, and the Ethics file refreshes nightly. Breed’s committees raised $882,941 in the 14 and a half months before September 19, 2024. Clean Up City Hall took in more than six times that in its first 13 days.

ON THE CALENDAR

Ignite Talks @ The AI Conference 2026. A night of rapid-fire talks tied to The AI Conference: each speaker gets five minutes and 20 slides that advance on their own every 15 seconds. 607 people had registered as of early Sunday, twice the 303 on the page when the desk’s calendar first logged it on August 27. With 393 spots still open, the room is about 60 percent spoken for. It is a single night; the desk’s tracker lists no series behind it.

Wednesday, September 30, 7:00 PM · Pier 48, Mission Bay · not free · lu.ma/tzsufkl6

From the desk’s calendar: public Luma pages, counts as of 1:48 a.m., September 27. Tell the desk about the gathering it missed.

SOURCES & RECEIPTS

1. NOAA Center for Operational Oceanographic Products and Services (CO-OPS), station 9414290, San Francisco (Presidio), established June 29, 1854 (station metadata). High-tide flood days from NOAA’s derived htf_annual and htf_monthly products at the NOS minor threshold, 13.69 ft station datum (moderate 14.62 ft). 1920–1968: zero every year. 38 flood days 1920 through July 2026; the 32 before 2026 fell in January (19), February (5), November (1) and December (7); 2026: four in January, one in June, one in July. 1983: six minor, two moderate. 2006: one; 2007–2013: none. No data 1971; 1970 and 2024 partial years (17.5% and 43.7% complete). 2. NOAA CO-OPS verified high/low water and tide predictions (datagetter, station datum, local time), 1980–2026; the station’s high/low series returns no data before 1980. 2026 flood-day highs: January 1, 13.90 ft; January 2, 14.06; January 3, 14.36; January 4, 13.91; June 14, 13.76 (11:18 p.m.); July 13, 13.70 (11:12 p.m.). Predicted highs those days: 13.02, 13.14, 13.11, 12.90, 13.08, 13.18. Days whose recorded high reached 13.69 ft, 1980–2024: 29 (NOAA’s htf count for the same years: 27), including March 1 and 2, 1983. Observed minus predicted on those 29: median 1.19 ft, range 0.52–2.21. Average of each day’s higher high minus prediction: 2001, −0.03 ft; 2003, +0.03; 2025, +0.23; 2026 (January–August, 235 days), +0.39, with 70 days at or above +0.48. Tidal datums, 1983–2001 epoch: mean higher high water 11.82 ft; highest astronomical tide 13.21 ft (December 23, 2003); record high water 14.64 ft (January 27, 1983, 9:30 a.m.). Predicted highs of 13.0 ft or more, full years 1969–2025: eight or more in 1969 (8), 1972 (9), 1982 (10), 1986 (13), 1987 (8), 1990 (8), 2004 (10) and 2022 (11), with two NOAA flood days among them (1969, 1982); 2025, five (no flood days; average above the tables +0.23 ft); 2026, eight through September 27. December 24, 2026, 10:37 a.m.: predicted 13.21 ft. September 2026 preliminary maximum 13.05 ft (September 8). National Weather Service minor flood threshold at this gauge, 13.02 ft (NOAA floodlevels, beside the NOS minor 13.69): reached on 541 days 1980–2025, at least twice every year, 117 of them in June or July. NOAA CO-OPS sensor specifications (updated December 2025): accuracy of an individual water level measurement, ±0.02 m (0.79 in) relative to datum. 3. NOAA sea level trends, station 9414290: 0.78 ± 0.03 inches per decade, September 1897–December 2025. Monthly mean sea level: the 2016–2025 average (2024 omitted as incomplete) is 0.73 ft (8.8 in) above the 1920–1929 average; annual averages 2023, +0.38 ft, and 2025, +0.20 ft, relative to 1983–2001 mean sea level. 4. NOAA Ocean Service, “What is high tide flooding?” (oceanservice.noaa.gov/facts/high-tide-flooding.html). Port of San Francisco, Waterfront Resilience Program (sfport.com/wrp). NOAA Ocean Service, “Coastal Flooding in California: What You Need to Know” (December 2015): “when there is an El Niño, sea levels on the West Coast are generally higher due to warmer, expanded ocean waters and changing weather patterns.” NOAA Climate Prediction Center, Oceanic Niño Index (1950–): December–February 2026, −0.39°C; June–August 2026, +1.80; peaks +2.14 (December–February 1983), +2.37 (November–January 1997–98), +2.59 (November–January 2015–16, the index’s highest). NOAA flood days in January and February 1983, six; February 1998, three; 2015 and 2016, none. All pulled September 28, 2026; queries in receipts.json.

2. San Francisco Ethics Commission, Campaign Finance – Transactions (DataSF pitq-e56w; most recent version of each filing), Form 460 Schedule A monetary contributions, pulled September 28, 2026 (file loaded September 27). Committees 216006060 and 214099226: 75 contributions from 67 contributors, $14,806,860, July 23, 2025–September 19, 2026 ($10,502,500 and $4,304,360). Checks of $1 million or more: Larsen and Lam, April 1; Moritz, April 7; Pritzker, April 13. 21 contributions of $250,000 or more total $11,725,000; six under $1,000 total $360. Moritz: $2,000,000 (April 7) and $487,500 (July 29); Schedule C in kind, $100,000 slate mailer (September 3) and $5,595.78 food and drinks (May 18). Larsen: $1,000,000 and $487,500 (August 19). Occupations and employers as filed; Larsen files as Christian Larsen. Lam as Larsen’s wife: San Francisco State University, “A Transformative $25 Million Gift From Chris Larsen, Lyna Lam, and Rippleworks” (develop.sfsu.edu/chris-larsen). Moritz as former Sequoia Capital partner and funder of The San Francisco Standard: Wikipedia, “Michael Moritz.” 6. Same file, committees filed as “Primarily Formed Measure,” Schedule A: July 1, 2025–September 19, 2026, $29,903,544 (1,751 contributions); July 1, 2023–September 19, 2024, $17,382,755, of which Breed’s committees (208751473, 211703442) $882,941; July 1, 2011–September 19, 2012, $6,024,909, of which Lee’s (130721317) $50,000. Lifetime: Lee $949,700 (May 2012–November 2016; $399,500 of it in 2012, $305,000 in 2016; filed as “Mayor Ed Lee for San Francisco Committee, Yes on J and K,” the November 2016 sales-tax pair per SPUR’s November 2016 voter guide, spur.org/voter-guide/2016-11); Breed $826,350 plus $355,222. June 2026 Proposition D committees: “Yes on C, No on D to Protect San Francisco’s Small Businesses…” $5,218,450 (contributions February 5–May 27, 2026) and “Yes on D – Stand Up for SF, sponsored by labor organizations” $3,095,000 (December 15, 2025–June 1, 2026), the largest measure committees after the mayor’s two. Clean Up City Hall, April 1–13, 2026: $5,360,000 in seven checks. Committee names and renamings as dated by filing. 7. San Francisco Ethics Commission, “Contribution Limits and Restrictions” ($500 per contributor, per election cycle, to candidates for city office). Citizens Against Rent Control v. City of Berkeley, 454 U.S. 290 (decided December 14, 1981). 8. San Francisco Ethics Commission, “Campaign Finance Project to Reinforce Contribution Limits” (April 2026) and August 14, 2026 agenda item 5 staff memo: extend the candidate limit to other candidate-controlled committees while the candidate is actively running; supermajority of the commission and the Board of Supervisors required; operative no earlier than January 1, 2027; proposed 1.114(a)(4) gives an official controlling a committee before actively running 30 days to dispose of funds above the limit. Commission vote, 4–0, August 14, 2026: Mission Local, “S.F. ethics commission votes to end politicians’ campaign finance loophole.” 9. San Francisco Public Press, “November 2026 San Francisco Election Guide” (Propositions D, E, F and H). Local News Matters, “Prop D vs. Prop C: San Francisco’s high-stakes fight over business taxes” (May 10, 2026): June 2 ballot; Proposition D, the “Overpaid CEO Salary Tax.” Queries in receipts.json.

The Frisconian · Published from forty-nine square miles surrounded by reality · thefrisconian.com · @deaconsanti