How to Estimate Sun Hours for a Garden Bed
Compare daylight with modeled direct sun using a reproducible Sydney example, explicit screen dimensions and a record you can repeat through a growing season.
Last reviewed September 18, 2026A garden can contain full-sun and shade conditions only a few metres apart. Estimate each important bed at its own point and compare the dates that cover the real growing season.
Open a worked example
Each link restores the stated place, moment and modeled assumptions in the studio.
Inputs for a useful comparison
Use the illustrative point at 33.8642° S, 151.2166° E on 15 October 2026, Australia/Sydney. The example begins with no obstacles and a flat horizon. The alternate scene adds a hypothetical 30 m high, 12 m wide screen 10 m north of the point, with normal 0°. This is a constructed geometry lesson, not a survey of Sydney’s Royal Botanic Garden.
Pin the bed, not the address
Place the receiving point inside the planting area. Street-address coordinates or the center of a large yard can miss the shadow edge that matters for a bed beside a wall or tree.
Model the seasonal skyline
Add measured buildings and walls as vertical screens. Trees use that same opaque geometry: vary height and width to test uncertainty, but do not interpret the result as dappled-light transmission. A manual horizon profile can represent a skyline if its angles are measured; it does not import terrain automatically.
Match dates to the crop calendar
Test the beginning, middle, and end of the growing period. A midsummer total can overstate exposure for spring or autumn crops. Use the same sample dates when comparing candidate beds.
Interpret hours as a screening metric
Plant labels often group exposure into broad bands, but species, climate, heat, reflected light, and irrigation still matter. Use direct-sun hours to compare locations and identify field observations, not as a guarantee of plant performance.
Keep a repeatable observation record
For each bed, record the scene name, coordinates, date, time zone, screen dimensions and their sources. Add three columns to your notes: modeled direct sun, observed shade times, and unmodeled changes such as leaves or temporary structures.
Use the same dates for all beds. At a shadow boundary, make separate point studies for each end and the centre; a single total does not describe every square metre of a long bed.
Method and a worked example
Two-minute midpoint integration counts actual elapsed time with the solar center above the geometric horizon and no modeled obstruction. An outdoor point has no facing-direction condition. Standard sunrise/set use the refracted upper limb; this is why standard daylight and zero-obstacle geometric direct sun differ.
The clear example gives 12 h 54 min standard open-horizon daylight (rounded from 773.89 min) and 12 h 46 min modeled geometric direct sun (766 min). The hypothetical north screen reduces modeled direct sun to 10 h 52 min. These are engine results for declared assumptions, not measured sunshine or a plant recommendation.
How to interpret the comparison
The difference between the two direct-sun scenarios is caused by the added screen. The smaller difference between standard daylight and geometric direct sun comes from the horizon conventions. Clouds can reduce actual bright sunshine further. The annual Year heatmap stays open-horizon even when your point has modeled obstacles.
Limits of the estimate
The example omits terrain, canopy gaps, reflections and weather. A tree screen is opaque, not a leaf-transmission model. Two-minute sampling can miss a brief gap and cannot certify a transition to the nearest minute.
What to do next
Open the two scenes and inspect the north screen near midday. Save a named scene for your own bed; check its centre and ends on the same three growing-season dates. Export CSV for solar angles and JSON for the assumptions. Record actual shade separately with date, local time and observation notes.
Questions about this method
Does dappled tree light count as direct sun?
The screen model does not resolve dappled light or partial leaf transmission. Record that uncertainty and observe the actual bed; changing an opacity label cannot provide a canopy measurement.
Should I use one point for a long garden bed?
No when a shadow boundary crosses it. Test the ends and center separately to see whether the bed contains materially different exposure zones.