Site Sunlight Analysis: A Reproducible Shadow Study
Run a preliminary site solar analysis with a worked wall example. Compare summer and winter direct-sun hours, document inputs and understand model limits.
Last reviewed October 22, 2026A useful sunlight analysis connects a decision to a receiving point, measured geometry and a date. This worked study asks a narrow question: how does a wall due south change direct-sun duration at one point in London? It is a synthetic experiment, not a survey of a real property.
Open a worked example
Each link restores the stated place, moment and modeled assumptions in the studio.
Separate solar position from site shading
Start with the sun path map for the place, date and local time, then use the shadow calculator for the stated obstacle geometry. Keep the receiving point, time zone and obstacle assumptions together when comparing scenarios.
Astronomy Engine documents the astronomical calculation library used here. It does not supply the dimensions or transparency of a building or tree. NOAA’s calculation notes explain refraction and atmospheric limits; NOAA also states that its calculator is no longer actively maintained. These references describe methods, not an inspection of the modeled property.
Inputs for a useful comparison
Observer: 51.5080° N, 0.1281° W, Europe/London, receiver height 0 m. Model one opaque vertical screen 12 m due south, 9 m tall and 20 m wide, with its normal at 180°. No terrain, trees, other buildings or window-facing filter. Compare 21 June and 21 December 2026.
Choose the result before choosing the tool
Sun-path analysis answers where the sun is. Shadow analysis asks whether a modeled object intercepts the ray. Direct-sun duration counts unblocked time at a point. Irradiance describes incident energy per area; photovoltaic yield adds module, orientation, temperature and system behavior. Those outputs are related but not interchangeable.
- For a garden, compare receiving points and morning versus afternoon exposure.
- For a window, also specify the outward-facing bearing and receiving height.
- For a possible solar-panel position, use this study to screen obstructions, then evaluate irradiation and system performance separately.
Keep the experiment reproducible
Hold the coordinates and obstacle geometry fixed while changing the date. Use the scene links below the introduction to restore this example. The times are local noon: June uses BST, December GMT. The exposure totals integrate the entire local day, not just the selected noon instant.
To reproduce the open-point baseline, remove the wall from the restored scene. To reproduce the wall case, reopen its scene link. Save each variation with a clear name and export its field sheet before editing.
Use a site worksheet before making a decision
Record the point, time zone, date, obstacle dimensions, receiving height and where every measurement came from. Mark unknown heights as unknown. A missing imported building is not evidence of a clear sightline.
Check the predicted transitions against observations on site. If the purpose is a formal daylight report, planning submission or an energy investment, this preliminary model is not a substitute for the appropriate measured inputs and analysis.
Method and a worked example
SunCarta samples the geometric solar center every two minutes across the actual local civil day. A sample counts as direct sun only when the center is above 0° and its ray is not intercepted by the finite screen. This duration is not the conventional sunrise-to-sunset daylight duration, which uses a different solar-limb/refraction definition.
On 21 June 2026, the open point receives 984 modeled minutes (16 h 24 m); adding this wall leaves 984 minutes, so the modeled loss is zero. On 21 December, the open point receives 456 minutes (7 h 36 m), while the wall case receives 104 minutes (1 h 44 m): 52 minutes before noon and 52 after noon. The modeled loss is 352 minutes (5 h 52 m). These are sampled model outputs, not measured sunshine records.
How to interpret the comparison
The same obstacle can have almost no effect on summer direct-sun duration and a large winter effect. The wall subtends about 36.9° vertically on its centerline (arctan(9/12)); in winter the low sun is intercepted across much of its path. In summer it passes above the wall when its bearing crosses the screen. The screen is finite, so winter light can still reach the point around its edges.
Limits of the estimate
A real building has depth, a roof and multiple facades; this example uses one zero-depth vertical screen. Trees change with leaf cover. Uneven ground and the actual skyline matter. This preliminary geometry study does not calculate irradiance, photovoltaic yield, room illuminance or compliance with daylight or planning standards.
What to do next
Open both scenes, then switch to the same location and your measured geometry. Read the direct-sun windows, explore monthly sample days and export a field sheet with your assumptions. For electricity estimates, use an irradiance and PV-performance model rather than multiplying these hours by panel capacity.
Six decisions from a reproducible synthetic study
These examples calculate geometry at public reference coordinates. The windows, wall and garden are hypothetical. They are not observations or inspections of the places.
1. Will December sun reach this window?
South-facing window at 51.5080, −0.1281, Europe/London. Receiver height 1 m; window 1.8 m wide × 1.5 m high with 1 m sill. A 3 m high, 12 m wide opaque wall stands 6 m due south. Bearings are clockwise from true north. No terrain or other obstacles are modeled.
| Date | Modeled direct sun | Morning / afternoon |
|---|---|---|
| 2026-06-21 | 9 h 20 m | 3 h 38 m / 5 h 42 m |
| 2026-12-21 | 0 h 54 m | 0 h 28 m / 0 h 26 m |
Two-minute midpoint samples count sun in front of the vertical plane, above the horizon and outside the modeled wall. They do not estimate indoor brightness, glazing transmission or weather.
Open this configured window in the existing planner →2. Compare two hypothetical apartments before a visit
London and Madrid public reference coordinates, both with a hypothetical 180° window, the selected date and their local time zones. No obstacles are modeled in this comparison: the real skyline is unknown. Equal civil dates do not mean equal simultaneous UTC instants.
| Reference point | Zone | Modeled direct sun |
|---|---|---|
| 51.508, -0.1281 | Europe/London | 7 h 36 m |
| 40.4154, -3.7074 | Europe/Madrid | 9 h 6 m |
During a visit, confirm the actual window bearing, receiving height, neighboring roofline, trees and overhangs. This table is a controlled example, not a ranking of real apartments.
Continue with the apartment checklist →3. The shadow of a three-metre wall
For a vertical height above level ground, shadow length = 3 / tan(solar elevation). This separate calculation uses a ground receiver, not the 1 m window receiver. At or below 0° elevation, a daytime shadow length is not assigned.
| Local time | Elevation | Length on flat ground |
|---|---|---|
| 06:00 | -17.91° | Sun below geometric horizon |
| 09:00 | 5.45° | 31.42 m |
| 12:00 | 15.05° | 11.16 m |
| 15:00 | 5.17° | 33.19 m |
| 18:00 | -18.34° | Sun below geometric horizon |
A sloped surface, roof depth or tree canopy changes the geometry. Keep height in metres and angle in degrees before converting to radians for the tangent.
Open the same wall and date for a shadow study →4. Choose three points for a hypothetical garden
Three ground points lie 5 m west, at the reference point and 5 m east. They use the same fixed wall and date; the window-facing filter is removed.
- -5 m east: 3 h 8 m modeled direct sun at 51.508000, -0.128172.
- 0 m east: 0 h 54 m modeled direct sun at 51.508000, -0.128100.
- 5 m east: 3 h 8 m modeled direct sun at 51.508000, -0.128028.
Measure each receiving point and obstacle; do not reuse a horizon measured elsewhere. Two-minute sampling is a numerical resolution, not survey accuracy. Trees and seasonal leaf cover are absent from this synthetic scene. These hours are not a universal crop recommendation.
Compare garden receiving points →5. Record uncertainty instead of guessing
Example field sheet: bearing 180° true north (confirm compass setting); distance 6 m; wall height 3 m with a hypothetical ±0.5 m uncertainty; width 12 m; receiver 1 m; units metres; measured date and method still needed for a real site.
- Wall height 2.5 m: 3 h 14 m modeled window sun.
- Wall height 3 m: 0 h 54 m modeled window sun.
- Wall height 3.5 m: 0 h 54 m modeled window sun.
Repeat the model at plausible extremes. A height range does not include uncertainty in distance, bearing, roof geometry or the chosen point.
6. Save a study another person can read
The scene JSON opens in the planner’s existing scene-file input. The complete study JSON adds both dates, hourly totals, shadow arithmetic, garden points, sensitivity and assumptions. CSV contains hourly solar position samples in UTC and local time, not observed weather or window exposure. The printable field sheet records the selected scene and the engine.
No example action requests your location or saves your coordinates. The exported examples contain only these public reference coordinates and synthetic geometry.
Calculated with Astronomy Engine 2.1.19 / SunCarta methodology 2.0. Methodology and limits · Astronomy Engine · NOAA bearing/elevation definitions. NOAA’s linked older calculator uses a west-positive longitude convention; SunCarta uses east-positive signed longitude.
Questions about this method
Is sunlight analysis the same as solar-panel analysis?
No. Direct-sun hours describe geometry at a point. Solar-panel output also depends on irradiation, orientation, temperature and system losses. SunCarta’s shading study does not estimate kWh.
Why is direct-sun duration shorter than sunrise-to-sunset daylight?
The direct-sun integrator requires the geometric solar center to be above 0°. Standard sunrise and sunset use a conventional limb/refraction definition; the two durations need not match.
Can I reproduce the numbers?
Yes. Open the June and December scenes, read the direct-sun totals and remove the wall to compare the open-point baseline. Keep the coordinates, receiving height and dates unchanged.