Meteorologist, oceanographer, Python solvem probler, experimenter, Raspberry Pi tinkerer, geek dad, manager, space enthusiast, ham radio lover, science columnist, ADHDer, synthetic intellect curious, linux admin cosplayer.

Draft Corrections — Post 27: The Forecast on a VAX (Cane-Zebiak 1986)

Post: _posts/2026-04-28-The-forecast-on-a-VAX.md Verified against: research/people/Mark_Cane.md, research/people/Stephen_Zebiak.md, research/computers/Cane_Zebiak_model.md, NOAA CPC ONI v5 table, Lamont and IRI press releases.

Overall verdict

The post is overall in very good shape. Strong agreement with primary and secondary sources on the bibliography, the model architecture, the forecast, and the Vetlesen Prize. A handful of small factual issues and a couple of phrasing tightenings would improve it.

Required corrections (factual errors)

1. Cane’s MIT entry year — should be 1972, NOT 1968

Post line 38 reads: “In 1968 he entered the doctoral program at the Massachusetts Institute of Technology in atmospheric and ocean dynamics.”

This is incorrect. Per Cane’s own CV (Mark_long_CV_2-15-15.pdf):

  • 1966-70: Senior Programmer Analyst, GISS (he met Charney here)
  • 1970-72: Assistant Professor of Mathematics, New England College
  • 1972-75: MIT PhD program (entered 1972, defended 1975)

Suggested fix: Change “In 1968 he entered” to “In 1972 he entered” — and ideally add a phrase about the 1966-72 GISS / New England College interlude, which is biographically important: it’s the period he met Charney in person (Charney was brought in as a consultant on a project Cane was working on at GISS, per Cane’s own Deep Convection Ep. 6 narrative).

Drafted replacement: “After four years as a senior programmer analyst at the NASA Goddard Institute for Space Studies in New York, where he met Charney for the first time – the older man had been brought in as a consultant on the project Cane was working on – and two years as an assistant professor of mathematics at New England College in rural New Hampshire, Cane entered the doctoral program at the Massachusetts Institute of Technology in 1972.”

1b. Lamont distance from Manhattan — should be ~18 miles, NOT 30

Post line 14: “on a wooded campus on the New York side of the Hudson River about thirty miles north of Manhattan”

Verified sources put Lamont at 18 miles north of Manhattan (Palisades, NY, just across the Hudson from Yonkers). Columbia Magazine’s Ewing feature gives the often-quoted “fifteen miles north of Columbia”; Lamont’s own visitor page and Hudson Valley sources give “18 miles north of Manhattan.” Thirty miles is too far.

Suggested fix: Change “about thirty miles north” to “about eighteen miles north” or “about fifteen miles north of the Columbia University campus” or “about twenty miles north” (which is close enough to be defensible).

1c. Excerpt’s “forecast was issued in autumn 1986”

Post excerpt (line 11): “The forecast was issued in autumn 1986. The El Nino arrived on schedule.”

Per Barnett et al. 1988 Science 241, 192-196 (verified by direct PDF text extraction): “Between late winter and spring of 1986, three different objective schemes predicted a moderate warming of the equatorial Pacific SSTs, an El Niño, to begin in late summer to early fall of 1986.”

The forecast was issued in late winter / spring 1986 and published in the Nature paper of 26 June 1986. The El Niño event arrived in autumn 1986. The excerpt has the timing inverted — it’s the El Niño that arrived in autumn, not the forecast.

Suggested fix: Replace “The forecast was issued in autumn 1986” with one of:

  • “The forecast was published in mid-1986.”
  • “The forecast was issued in spring 1986, with the El Niño predicted to arrive later that year.”
  • Simply: “The forecast was issued in 1986. The El Niño arrived on schedule.”

Suggested clarifications (not strictly errors)

2. The VAX 11/780 specifically is not documented in any primary source

The post asserts in two places (lede and image caption) that the machine was a “DEC VAX 11/780”. The research notes flag this as unverified by primary source. The 1986 Nature and 1987 MWR papers do not name the hardware. Secondary sources just say “VAX” or “minicomputer at Lamont”.

The VAX 11/780 was indeed the most-deployed VAX model at universities in the early-to-mid 1980s, but Lamont could plausibly have had a 11/750, 11/780, 11/785, or MicroVAX. We do not know.

Recommendation: Either soften to “DEC VAX” in the lede, or keep “VAX 11/780” but add a footnote acknowledging that the specific model is not documented in the primary papers and rests on the standard university-VAX assumption of the era.

Drafted footnote addition to footnote 1: “The 1986 *Nature and 1987 Monthly Weather Review papers do not name the specific VAX model. The VAX 11/780 is the standard departmental machine of the era and the one Lamont is most likely to have had, but the exact model is not documented in the primary papers.”*

3. “Elder scientists scoffed” — attribution to Vetlesen citation vs press release

Post body (line 26) says: “a citation that explicitly noted that ‘elder scientists scoffed’“. Footnote 3 attributes the phrase to the Lamont Vetlesen announcement AND the Krajick IRI feature.

Strictly: the phrase appears in the Krajick/Lamont press release announcing the 2017 prize (“When they predicted another El Niño in 1986, elder scientists scoffed, but it arrived later that year”), not in the formal Vetlesen Prize citation, which uses more anodyne language (“their seminal contributions to understanding the cyclic phenomenon known as the El Nino-Southern Oscillation, or ENSO”).

Recommendation: Change “a citation that explicitly noted” to “an accompanying announcement that explicitly noted”. Or rephrase as “with the accompanying press materials noting”.

Drafted replacement for the relevant clause: “…with a $250 000 cheque and an announcement that explicitly noted that ‘elder scientists scoffed’ at his original forecast.”

3b. “Borough Park” neighborhood — unsourced

Post line 34: “He grew up in the Borough Park section of Brooklyn in the late 1940s and 1950s…”

The Borough Park neighborhood specifically is not in any source I located. The Forward 2017 interview gives “a neighborhood that was Jewish, Irish and Italian” and “very Jewish, Yiddish culture” but does not name Borough Park. Cane graduated from Midwood High School, which is in the Midwood neighborhood; Borough Park is a different Brooklyn neighborhood ~4 miles to the west. Jewish/Irish/Italian Brooklyn neighborhoods of that era could include Midwood, Flatbush, Borough Park, Bensonhurst, or Bay Parkway, among others.

Recommendation: Either replace “Borough Park” with “Brooklyn” (general), or “the Jewish-Irish-Italian Brooklyn of the late 1940s and 1950s”, or simply omit the specific neighborhood name. Saying Borough Park specifically may be an inference rather than a sourced fact.

3c. “Public-school system that put Black and white children in the same classrooms”

Post line 34 (in the same long sentence about Brooklyn): “the public-school system that put Black and white children in the same classrooms”

The Forward 2017 interview’s racial-integration material is about the Brooklyn Dodgers (“It brought us together. It was racially integrated, so as a child I had African American heroes”), not about classroom integration. Brooklyn public schools in the 1950s were de facto segregated by neighborhood despite no de jure segregation; a Jewish neighborhood like the one Cane describes would have had very few Black students.

Recommendation: Soften to something like “the Brooklyn Dodgers as a racially integrated source of civic pride, the public-school system, the conviction that mathematics and physics were honourable lifetime occupations…” — drop the classroom-integration phrase or replace with something more accurate to the source. Or: “the public schools and the racially integrated baseball team as symbols of civic possibility…”

4. “Politically inconvenient at a Cold War institution about funding priorities”

Post line 42: “He had, in his own retrospective view, also been politically inconvenient at a Cold War institution about funding priorities for atmospheric and ocean science.”

This specific framing — “politically inconvenient at a Cold War institution about funding priorities for atmospheric and ocean science” — is not in any source I’ve located. The Forward 2017 interview and the Krajick 2017 IRI feature both establish that Cane was politically active (civil rights work, Vietnam War protest) but neither attributes his MIT tenure denial to politically inconvenient views about science funding priorities.

The Deep Convection Episode 7 host gloss (“this is probably a decision MIT has come to regret”) is not Cane’s words and is non-specific.

Recommendation: Either remove the sentence or soften it. The honest answer is that Cane has been deliberately reticent about the reasons.

Drafted replacement: “He had, in his own retrospective accounts, also been politically active in ways that were perhaps unusual for the MIT atmospheric science faculty of the period.”

Or, more cautiously: simply remove the sentence and let the next sentence (“He left MIT in 1984”) stand.

5. “Stayed at MIT as a postdoctoral researcher”

Post line 40: “He stayed at MIT as a postdoctoral researcher and then as a member of the faculty.”

Per Cane’s CV, after his 1975 MIT PhD he did NOT stay at MIT:

  • 1975-76: NAS/NRC Research Associate, GISS
  • 1976-78: Senior Analyst, Goddard SFC (Sigma Data Services)
  • 1976-77: Adjunct Asst Prof of Geology, Columbia
  • 1978-79: Staff Scientist, Goddard Lab for Atmospheric Sciences
  • 1977-79: Visiting Scientist, MIT (Meteorology Dept) — overlapping with the Goddard staff position
  • 1979-84: MIT faculty (Assistant and Associate Professor)

He was at GISS/Goddard for his postdoc and only came back to MIT as a Visiting Scientist in 1977 (alongside his Goddard position) before joining the MIT faculty in 1979.

Suggested fix: Replace with something like “He spent the next four years as a postdoctoral researcher at NASA’s Goddard Institute for Space Studies and as a visiting scientist at MIT, then joined the MIT faculty in 1979 as a tenure-track assistant professor.”

6. “His then-postdoctoral advisor Edward Sarachik”

Post line 42: “with his then-postdoctoral advisor Edward Sarachik”

Sarachik was a senior MIT colleague who collaborated with Cane on the 1976-1979 J. Mar. Res. “Forced baroclinic ocean motions” series. Cane’s actual postdoc was at NASA Goddard / GISS 1975-1979. The CV does not name a single postdoctoral advisor; the collaboration with Sarachik came through MIT after Cane’s PhD, while Cane was at Goddard/GISS as a postdoc and visiting MIT.

Suggested fix: Soften to “his MIT colleague Edward Sarachik” or “his senior collaborator Edward Sarachik” rather than “postdoctoral advisor.”

7. “Doherty Senior Scientist that same year” (1984)

Post line 44: “Lamont-Doherty Earth Observatory … hired him as a Doherty Senior Scientist that same year.”

Per CV:

  • 1984-87: Senior Research Scientist, Lamont-Doherty Geological Observatory
  • 1987-97: Doherty Senior Research Scientist, Lamont-Doherty Geological Observatory

Cane was hired in 1984 as a Senior Research Scientist (without the “Doherty” prefix). The Doherty title came in 1987.

Also, in 1984 the institution was still Lamont-Doherty Geological Observatory (1969-1993). The post’s mention “Lamont-Doherty Earth Observatory” for 1984/1986 in lines 14 and 44 is a minor anachronism — the “Earth Observatory” name only came in 1993. The post already correctly explains the rename chronology in line 52. The post may simply have chosen to use the modern name throughout for clarity, which is defensible.

Suggested fix: Change “Doherty Senior Scientist” to “Senior Research Scientist” in line 44. The Lamont-Doherty Earth/Geological naming can be left as a stylistic choice if the post is OK with mild anachronism.

Cross-reference checks

  • Cross-reference to Post on Charney (“The Man Who Tamed the Equations”) — correctly states Charney was Cane’s PhD advisor. Charney died in 1981; Cane finished in 1975, so the chronology works.
  • Cross-reference to Post on Manabe at GFDL — the GFDL comparison (“a thousand times faster, a thousand times less expensive”) is in the right order-of-magnitude ballpark; defensible.
  • Cross-reference to Post on Bjerknes (“He Made Walker Right”) — Bjerknes feedback as the “single non-linearity” is consistent with the literature.
  • Cross-reference to Post on Lorenz (“The Butterfly That Broke the Forecast”) — the predictability point is correctly made: Lorenz’s two-week limit applies to atmospheric weather, not to the slow coupled-mode dynamics of ENSO.

Priority of corrections (suggested order)

Hard factual errors (should fix):

  1. Cane entered MIT in 1972, not 1968. (Per CV.)
  2. Lamont is ~18 miles north of Manhattan, not 30. (Multiple sources.)
  3. Excerpt’s “forecast was issued in autumn 1986” inverts the timing — the forecast was issued in late winter/spring 1986; the El Niño event arrived in autumn 1986. (Per Barnett 1988 Science.)
  4. Cane was NOT an MIT postdoc — his postdoc 1975-79 was at NASA GISS/Goddard. He was a Visiting Scientist at MIT 1977-79 before becoming faculty in 1979.
  5. Cane was hired at Lamont in 1984 as Senior Research Scientist, not “Doherty Senior Scientist” — the Doherty title came in 1987.

Phrasing tightenings (recommended):

  1. “Borough Park” — not in any source; either remove or replace with “Brooklyn”.
  2. “Politically inconvenient at a Cold War institution about funding priorities” — not in any source; soften or remove.
  3. “His then-postdoctoral advisor Edward Sarachik” — soften to “his MIT collaborator” or “his senior MIT colleague”.
  4. “Public-school system that put Black and white children in the same classrooms” — the Forward source describes Brooklyn Dodgers as the racially integrated institution, not the classrooms.
  5. “Citation that explicitly noted that ‘elder scientists scoffed’” — the citation itself is anodyne; the phrase appears in the accompanying Lamont press release. Soften to “announcement that explicitly noted”.

Soft flags (note but not change):

  1. “VAX 11/780” specifically not documented in primary sources, but plausible. Consider adding a footnote acknowledging the specific model is the standard university-VAX assumption of the era.
  2. “About one million instructions per second”, “few megabytes of memory”, “about as much power as a domestic clothes dryer” — order-of-magnitude correct for a 1985-86 VAX 11/780 but not directly sourced for Lamont’s specific machine.
  3. “A simulated year on the VAX took something like an hour of CPU time” — order-of-magnitude estimate; not directly sourced.
  4. Use of “Lamont-Doherty Earth Observatory” for 1984/1986 contexts is a mild anachronism (institution was “Lamont-Doherty Geological Observatory” 1969-1993; “Earth Observatory” only from 1993). The post already explains the rename in line 52.

No-change verifications (post is correct)

  • Nature 321, 827-832, June 1986. CORRECT.
  • Mark Cane born 20 October 1944, Brooklyn NY, parents Jacob and Frances. CORRECT (per IRI/Krajick + Wikidata).
  • Vetlesen Prize 2017, $250,000, jointly with Philander. CORRECT.
  • IRI founded 1996 at Lamont. CORRECT.
  • Zebiak as Cane’s first MIT PhD student. CORRECT.
  • Zebiak Director-General of IRI 2003-2012. CORRECT.
  • 1986-87 El Niño peak ~1.2°C Nino 3.4. CORRECT per NOAA CPC ONI.
  • Adrian Gill 1980 QJRMS paper as atmosphere component. CORRECT.
  • Cane-Patton 1984 JPO numerical core. CORRECT.
  • Battisti 1988 / Schopf-Suarez 1988 / Battisti-Hirst 1989 / Jin 1997 — all citations CORRECT.
  • Barnett, Graham, Cane, Zebiak, Dolan, O’Brien, Legler 1988 Science 241, 192-196. CORRECT.
  • Sean Dolan disappears from the literature after 1988. CORRECT (post explicitly flags this).
  • Henry IV Part I Glendower/Hotspur reference — quoted by Seager 2015 in Oceanography 28(1). CORRECT.
  • “If you build it, they will come” Cane quote in Forward 2017. CORRECT.