# The Science of Powder (Part 3): From SLR to a dryness score — fact and inference

Furano City publishes a 0–100 dryness index and one anchor point. How we wired it into our model, and why the line between must be labelled inference.

- URL: https://www.ykhokkaido.com/en/blog/powder-science-03-dryness
- Published: 2026-09-19
- Section: Powder
- Series: The Science of Powder (part 3)
- Tags: Snow-liquid ratio, Dryness, Powder model

> This article documents where the parameters in SnowWind's own model come from. Anything marked as our inference has not been confirmed by the original research team and does not represent the position of Furano City or any research institution.

---

This part is about a small detail that decides whether the rest of the series deserves your trust.

## The facts first

Furano City, together with Kitami Institute of Technology and Hokkaido University's Institute of Low Temperature Science, publishes an index called **fuwa-sara**: a 0–100 score whose core variable is forecast snow-liquid ratio. The work appeared in the Cold Region Technology Conference proceedings, Vol.39 (2023).

The published material contains one crucial anchor:

> **50 points ⇔ snow-liquid ratio 2.0 cm/mm**

That number is not arbitrary. 2.0 cm/mm is precisely the Japan Meteorological Agency's dry-snow threshold, so the index places "just barely dry snow" exactly at the midpoint.

All of the above is **sourced fact**.

## Now what we added

We wanted a dryness axis on the same scale inside our own powder model, which meant we needed a conversion from snow-liquid ratio to 0–100.

The problem: the published material gives us **one point**.

One point does not determine a line. To get from (2.0, 50) to a full mapping you must first assume the mapping is linear. Once you assume that, the anchor pins the slope:

```
dryness = 25 × SLR (capped at 100)
```

That formula genuinely runs in our code. It is **not the published mapping** — no mapping formula was published. Linearity is our assumption, and we have no way to verify it.

So on our pages and in our code comments, this line is always tagged **SNOWWIND INFERENCE**, never presented as fact.

## Why say so out loud

In this line of work, dressing an inference up as a fact carries almost no short-term cost. Nobody checks, and it reads better.

It carries two long-term costs. First, if the original team later publishes the real mapping and it turns out not to be linear, everything we have ever written loses credibility at once. Second, readers take our numbers into the mountains and make decisions with them, and those decisions have consequences.

So we apply one convention across the whole site: **every claim carries a source tier.**

- **SOURCE FACT** — has an external source we can name
- **SNOWWIND INFERENCE** — our reasoning, and it may be wrong
- **MODEL PARAMETER** — a value we chose; not right or wrong, only well or badly tuned

## A worked example

Take the sentence "the snow is dry today" and split it:

| Claim | Tier |
|---|---|
| The JMA dry-snow line is 2.0 cm/mm | SOURCE FACT |
| Furano's index anchors 50 points at 2.0 cm/mm | SOURCE FACT |
| dryness = 25 × SLR | SNOWWIND INFERENCE |
| SLR estimated from temperature via a Kuchera-style heuristic | SNOWWIND INFERENCE |
| The multiplier range that dryness maps onto | MODEL PARAMETER |

Only the first two rows are citable facts. The rest is our model, and a different team would make different choices.

## Three things to take away

1. Furano's dryness index and the 50 ⇔ 2.0 anchor are published, citable facts.
2. Our 25 × SLR conversion is an inference, because one anchor cannot define a line.
3. Credibility comes from labelling, not from tone. An article that will not separate fact from inference has earned trust in neither.

The series continues with how temperature is turned into an SLR estimate, how wind destroys a surface after landing, and why even the driest snow will not hold up over a thin base. For Furano's terrain data, see [Furano's runs & lifts page](/resorts/furano/guide).

## FAQ

### What is Furano's "fuwa-sara" index?

A 0–100 powder index published by Furano City with Kitami Institute of Technology and Hokkaido University's Institute of Low Temperature Science. Its core variable is forecast snow-liquid ratio, and the work appeared in Cold Region Technology Conference Vol.39 (2023).

### Where does the 25 × SLR mapping come from?

It is SnowWind's inference. The published material gives one anchor: 50 points corresponds to a snow-liquid ratio of 2.0 cm/mm. Getting a whole line from one point requires assuming linearity, and that assumption is ours — no mapping formula was published.

### Why label inferences at all?

Because a reader cannot tell them apart otherwise. Writing an inference as though it were a fact reads as more authoritative in the short run and accumulates risk for everyone in the long run. Every claim we make carries a source tier.

## Sources

- Shirakawa, Matoba & Osuga, "Development of a fuwa-sara index for powder snow using snow-liquid ratio", Cold Region Technology Conference Vol.39 (2023)
- Furano City, "Furano Bonchi Powder Project", bonchipowder.com
- Japan Meteorological Agency — dry/wet snow classification by snow-liquid ratio