What was built A raster map classifying San Francisco's solar energy potential into five irradiance zones using Global Horizontal Irradiance data — from fog-prone, low-output areas to high-irradiance zones optimal for solar investment.
Who it was for Urban energy planners and infrastructure investors evaluating solar installation returns at the municipal scale.
Why it mattered Solar potential in San Francisco varies substantially by neighborhood — fog, topography, and building density all affect output. The analysis makes that variation spatially explicit and decision-ready at the city scale.

The Challenge

Solar potential isn't uniform across a city. Topography, building density, fog patterns, and orientation all affect how much sunlight reaches a given location — and in San Francisco specifically, the difference between the sunniest and cloudiest neighborhoods is substantial. This map makes that difference visible.

Global Horizontal Irradiance data was classified into five zones, rendered as a gradient from low-irradiance areas (cool tones) to high-irradiance zones (red and orange). The output gives urban energy planners and infrastructure investors a spatial answer to a spatial question: where in the city is solar installation most likely to produce meaningful returns?

The value is in the decision support — not just showing where the sun hits hardest, but producing output at a resolution and classification that can actually drive infrastructure investment decisions at the municipal scale.

This analysis was used to evaluate priority zones for solar energy development based on spatial patterns in solar exposure and land suitability.

GHI Classification — 5 Zone Output
Zone 5 — Highest irradiance (optimal solar investment)
Zone 4 — High irradiance
Zone 3 — Moderate irradiance
Zone 2 — Low irradiance
Zone 1 — Lowest irradiance (fog-prone, shaded)

"The question wasn't 'does San Francisco get sun.' It was 'where does it get enough sun to make solar investment worthwhile.' That's a raster analysis problem."

Process

  1. 1
    Research Identified GHI dataset appropriate for municipal-scale solar analysis in San Francisco.
  2. 2
    Data Prep Processed raster data; established classification breaks for 5-class output.
  3. 3
    GIS Work Applied raster classification; designed five-class color gradient; verified spatial accuracy against known SF geography.
  4. 4
    Output Published raster map showing optimal solar installation zones across San Francisco.
Principle 02
The map is for a person making a decision

A raster classification is only as useful as the decision it enables. The five-zone output wasn't chosen for aesthetic reasons — it was chosen because infrastructure investors and urban planners need discrete, actionable zones, not a continuous gradient that leaves the interpretation open.