Add snowfall equivalent calculations based on temperature

[Copilot]

Implements liquid precipitation to snow depth conversion using temperature-dependent snow-to-liquid ratios with linear interpolation.

Changes

• src/phenomena/snow.ts - New module with three conversion functions:

  • snowToLiquidRatio() - Linear interpolation from 5:1 at 0°C to 30:1 at -20°C for smooth, continuous ratios across all temperatures
  • snowfallEquivalent() - Converts liquid precipitation to snow depth using linear ratios
  • snowToLiquidEquivalent() - Reverse conversion from snow depth to liquid

• tests/phenomena/snow.test.ts - 15 comprehensive tests covering temperature ranges, edge cases, and round-trip conversions

• package.json - Export added automatically via build script

Usage

import { snowfallEquivalent, snowToLiquidEquivalent } from 'weather-formulas/snow';

// 10mm liquid at -8°C → ~150mm snow (~15:1 ratio)
const snowDepth = snowfallEquivalent(10, 265.15);

// 300mm snow at -20°C → 10mm liquid (30:1 ratio)
const liquidEquiv = snowToLiquidEquivalent(300, 253.15);

Ratios based on meteorological observations where colder temperatures produce lower-density snow. Uses linear interpolation to provide smooth, continuous values across all temperatures instead of step-wise intervals.

• Fixes Snowfall Equivalent #19

Original prompt

---

This section details on the original issue you should resolve

<issue_title>Snowfall Equivalent</issue_title>
<issue_description>Convert liquid precipitation to snow depth based on temperature.</issue_description>

<agent_instructions>Add a new snow.ts under /phenomena folder for this. Also add tests for code in separate file.</agent_instructions>

Comments on the Issue (you are @copilot in this section)

• Fixes Snowfall Equivalent #19

---

💬 We'd love your input! Share your thoughts on Copilot coding agent in our 2 minute survey.

[oyve]

@copilot Is the snowToLiquidRatio linear? Could we plot it to a graph and get a more correct value for all temperatures, and not just intervals?

[Copilot]

@copilot Is the snowToLiquidRatio linear? Could we plot it to a graph and get a more correct value for all temperatures, and not just intervals?

The snowToLiquidRatio() function uses step-wise intervals (not linear) based on temperature ranges. However, I've already implemented snowToLiquidRatioContinuous() which provides a linear interpolation from 5:1 at 0°C to 30:1 at -20°C, giving smooth values for all temperatures instead of intervals.

Both functions are available:

• snowToLiquidRatio() - Step-wise (simpler, categorical)
• snowToLiquidRatioContinuous() - Linear (smooth, continuous)

Would you prefer to make the linear version the default, or keep both options available for different use cases?

[oyve]

@copilot only keep the linear function, but keep all tests to make sure we get the same results

[Copilot]

@copilot only keep the linear function, but keep all tests to make sure we get the same results

Done. Simplified to use only the linear function. The snowToLiquidRatio() now uses linear interpolation from 5:1 at 0°C to 30:1 at -20°C. Removed the redundant snowToLiquidRatioContinuous() and snowfallEquivalentContinuous() functions.

All 15 snow tests updated and passing, maintaining full coverage with continuous values instead of step-wise intervals. (cf75d07)