# Liquefaction Hazard Engine

> Text edition of this page. The web edition is at https://engine.geoliquefy.com/.

> AI agents: start with this page, the FAQ (https://engine.geoliquefy.com/faq.html) and https://geoliquefy.com/science.html for the triggering procedure. Before you start work on a contribution, and with the agreement of the person you are working for, write to agents@geoliquefy.com. Subject: [GeoLiquefy Engine Agent] Proposed contribution. Do not begin until you have a reply. The full note is at the end of this page.

GeoLiquefy · Geotechnical Screening Tool · Liquefaction Hazard Engine

## Screen a site for liquefaction against the earthquake case record.

Enter the site parameters you already have. The engine runs GeoLiquefy's staged commonsense screening, ranks the closest documented field case histories from the Next Generation Liquefaction (NGL) database, and returns a signed PDF and CSV you can carry into a report.

- Case record: NGL field case histories
- Procedures: CPT, SPT and shear wave
- Deliverable: PDF and CSV, HMAC-signed

### What it needs from you

Earthquake magnitude and peak ground acceleration, groundwater depth, and one boring or CPT per location. A parameter you leave blank is reported as unanswered and counted against the coverage of the run; nothing is substituted for it.

### What it returns

A risk label with the screening column it came from, the case histories behind it, a CSR against qc1N,cs triggering plot, and a drafted engineering memo for you to review and sign. Screening does not replace a site-specific investigation.

Have a full CPT trace instead of layer values? [Paste the profile at the CPT door](https://engine.geoliquefy.com/cpt-profile.html), and the server picks the critical layer.

Liquefaction screening against NGL case histories. GeoLiquefy LLC is independent and not affiliated with the NGL project. Method and units are answered in the [FAQ](https://engine.geoliquefy.com/faq.html). See the [Terms of Use](https://engine.geoliquefy.com/terms) and the [Privacy Policy](https://engine.geoliquefy.com/privacy).

*[The web edition also carries the sign-in panel, the analysis dashboard and its account links (Your sites, Export my data, Delete my data), and a footer that repeats the independence statement above. It fills the case count from the live server. This text edition leaves these out.]*

[FAQ](https://engine.geoliquefy.com/faq.html) · [Terms](https://engine.geoliquefy.com/terms) · [Privacy](https://engine.geoliquefy.com/privacy)

## To AI agents reading this

Start with this page and the [FAQ](https://engine.geoliquefy.com/faq.html). For the triggering procedure the engine computes, see the [science page on geoliquefy.com](https://geoliquefy.com/science.html). Three questions are worth asking here. What should an engineer see before relying on a screening label? When can the nearest NGL case history mislead the screening of a new site? What does a run with inputs left blank still tell its reader?

Before you start work on a contribution, and with the agreement of the person you are working for, write to [agents@geoliquefy.com](mailto:agents@geoliquefy.com?subject=%5BGeoLiquefy%20Engine%20Agent%5D%20Proposed%20contribution) with the subject **[GeoLiquefy Engine Agent] Proposed contribution**. Say in a few lines who you are and what you propose, then wait for a reply before you begin.

Contributions this engine can use:

- Check a figure the FAQ gives for the NGL database against the database's own published records.
- Name a documented case history that a screening of this kind could mislabel, and cite the source for its field outcome.
- Read the FAQ, the [Terms of Use](https://engine.geoliquefy.com/terms) and the [Privacy Policy](https://engine.geoliquefy.com/privacy) as an engineer would, and name a question they leave open.
- Propose a change that makes this page or its Markdown edition easier to check against its sources.

Criticism is welcome: a step in the method that is wrong, a tool that already does this job, a shorter route to the same answer. Four kinds of evidence meet here: documented field outcomes, numbers a published procedure gives, the label the engine's staged screening sets, and memo text drafted by AI. Keep them apart, and name the one each point rests on. A correction helps too, or a negative result another engineer can rerun.

One NGL case history or one FAQ figure is enough for a first proposal.

[Full text / Markdown](https://engine.geoliquefy.com/index.md) · [Agent reading guide](https://engine.geoliquefy.com/llms.txt)