Science review Cornell University, Ithaca, New York (research origin); continental U.S. simulation focus; northern Rockies / Glacier National Park named in pop cover
Cornell’s 2015 Zombie Math: Where Science Said to Hide (And What It Really Meant) (2026)
Dated look at Cornell’s 2015 “statistical mechanics of zombies” — Rockies hideouts, disease-model teaching tool, not a real outbreak forecast.

Sign in to save stories to your shelf.
For years, preppers argued bug-out spots like it was a contact sport.
In March 2015, this desk breathlessly reported that Cornell researchers had finally put numbers on the nightmare: if the dead walk, get sparse, get high, think Rockies — Glacier National Park vibes included in the pop retelling.
That page is a historical predecessor. This 2026 rewrite keeps the fun and stamps the calendar: the work was a teaching-friendly epidemiology / statistical-mechanics toy, presented around the 2015 American Physical Society March Meeting and later published in Physical Review E (2 November 2015): “You can run, you can hide: The epidemiology and statistical mechanics of zombies” by Alexander A. Alemi, Matthew Bierbaum, Christopher R. Myers, and James P. Sethna.
Not a CDC bulletin. Not a prophecy. A model with teeth jokes.
What the 2015 paper actually did
According to the authors’ abstract and APS/ScienceDaily framing at the time, zombies were a spicy cousin of classic SIR-style disease models — useful for walking students through continuous dynamics, lattice simulations, phase-transition talk, and finally a continental United States stochastic outbreak simulation.
Headline takeaway the internet kept: for “realistic” fictional parameters, cities cook fast; remote, low-density, elevated northern Rockies terrain looks comparatively survivable inside the simulation.
That is model geography, not a park ranger endorsement.
The numbers people still quote (with caveats)
Our 2015 predecessor loved the chain-reaction arithmetic — zombie zero bites, population doubles in nightmare fashion, urban centers collapse on compressed timescales while rural spread waits on a shuffling vector.
Treat those figures as illustrative simulation talk from 2015 coverage, not empirical undead telemetry. Change the bite rate, the mobility rule, or the lattice assumptions and the apocalypse calendar moves. That is the point of a model: sensitivity, not scripture.
Alemi’s contemporaneous quotes emphasized the pedagogical angle — zombies drag you through techniques used on real epidemics, “albeit in a fun context.” Believe the teaching pitch. Side-eye anyone selling Glacier as your divinely appointed bunker because a physicist ran a lattice.
Glacier, Nebraska, and the “now everyone knows” problem
Pop articles (including our old one) name-checked Glacier National Park and remote Midwest fantasies (Walking Dead Nebraska energy). Funny consequence the predecessor noted: once a study goes viral, the “ideal” hideout is meme-contaminated.
In 2026, that gag still works — and so does the adult reminder: national parks are for recreation under real rules, not LARPing an outbreak. Winter-freeze zombie theories, “zombie-proof cabins,” and panic-room mansions remain entertainment folklore. Product ads from the mid-2010s (log-cabin marketers with tongue-in-cheek guarantees) were marketing bits, not peer review.
Later status
As of this desk’s September 2026 check, the Cornell zombie paper remains what it was: a completed 2015 academic exercise with a stable journal citation (Phys. Rev. E 92, 052801). It did not become an operational forecasting tool for a real pathogen. Later Glacier research you may stumble on in searches concerns wildlife and recreation ecology, not undead SIR models — do not mash those topics together.
No pathogen playbook belongs here. Zombies stay fictional. Real disease communication belongs to public-health authorities, not cryptid blogs.
How science-fiction models help (and mislead)
They help when they teach contact networks, spatial heterogeneity, and why density matters. They mislead when headlines imply “science has saved us” from a monster that does not exist. Our 2015 title leaned into salvation theater. 2026 prefers: science lent us a classroom metaphor; your bug-out argument still needs water, not just altitude.
Conspiracy-curious angle, grain of salt attached: every viral model becomes a Rorschach test. Preppers see validation. Skeptics see nerd cosplay. Both can be partly right if they keep the date stamp visible.
Practical reading rules for “zombie science” headlines
- Ask for the publication year and venue.
- Label fictional pathogens as fictional.
- Separate pedagogy from policy.
- Do not treat park names in a paper as survival GPS.
- Pair with entertainment lists; leave real outbreak guidance to official channels.
What we refused to do
We will not present the 2015 Cornell work as breaking news in 2026. We will not invent updated “official” hideout coordinates. We will not provide pathogen enhancement, synthesis, or weaponization content of any kind. We will not hard-sell bunkers.
Salt freely. If the dead rise, the paper will not save you. If they do not — and they have not — you still got a memorable intro to statistical mechanics.
What “doomed” meant in the U.S. simulation
The continental simulation — as authors and APS-week explainers framed it — was not a single cinematic wave washing evenly from sea to sea. It was a geography of heterogeneous risk: dense metros flash over; sparse high country buys time. “Doomed” in the playful abstract sense meant that under their chosen fictional parameters, long-term continental coexistence with aggressive undead lookalikes looked ugly for Homo sapiens as a whole — while individual hideout choice still changed your personal survival clock.
That distinction matters. Prepper Twitter heard “go to Glacier.” Statisticians heard “spatial structure dominates mean-field SIR.” Both readings touched the elephant. Only one belongs in a national-park Leave No Trace brochure.
Sister models and the genre
By 2016, our own desk was already connecting zombie-model fashion to vampire arithmetic papers (see the companion refresh on Hartl/Mehlmann and Efthimiou/Gandhi). The pattern is durable into 2026: academics borrow monsters to teach; media borrows academics to sell end-times SEO. Your job as a reader is to keep the arrows of citation pointed the right way.
Also: Tiger Log Cabins’ tongue-in-cheek “zombie-proof” marketing from that era was advertising humor, not an ISO standard. Fancy button-down fortresses in design blogs likewise remain architecture porn. Neither is Cornell-approved PPE.
Sources
- Claim element: Historical predecessor (March 4, 2015) — Attribution: Monstrum Athenaeum original — Link: Science Has Once Again Saved Us…
- Claim element: Journal publication (Nov 2, 2015) — Attribution: Alemi et al., Phys. Rev. E — Link: You can run, you can hide…
- Claim element: APS meeting / ScienceDaily framing (Feb 2015) — Attribution: ScienceDaily summary of Cornell zombie modeling — Link: Zombie outbreak? Statistical mechanics…
- Claim element: Author PDF (if available) — Attribution: Alex Alemi publications mirror — Link: zombies.pdf
About this article
Entertainment and historical science-communication commentary only — not emergency preparedness instruction, not public-health guidance, and not a forecast of any real outbreak. Zombies are fictional. No pathogen synthesis, enhancement, weaponization, or lab advice is provided or implied. National parks and private property are subject to real laws; this article is not a trespass or illegal-camping invitation. This 2026 article is a brand-new legacy refresh; the 2015 URL is a historical predecessor, not a prior publication of this text. MonstrumAthenaeum.org is for mature audiences who can tell a lattice model from a news alert.
Original publication credit
Original title: Science Has Once Again Saved Us From The Zombie Apocalypse (Well..Some Of Us). The preserved copy has no named writer credit. Originally published March 4, 2015.
