I have 40 panels on my roof feeding two Tesla inverters, and if that setup has taught me one thing, it’s that solar production is unforgiving about geometry. Panels don’t average their way through problems. A shadow in the wrong place at the wrong hour doesn’t shave a little off the top — it can knock out a whole section of the array while the shadow sits there. Trees are the most common cause, and they’re also the trickiest, because they grow, they leaf out seasonally, and they’re often not even on your property.
This article covers how shade actually hurts production (it’s worse than intuition suggests), how to measure the problem before spending money, and which mitigation options genuinely work.
Why a Little Shade Causes a Lot of Loss
Solar panels are built from cells wired in series, and panels themselves are often wired in series into “strings.” Series wiring means current is limited by the weakest link. Shade one cell hard enough and it chokes the current for everything wired through it — the electrical equivalent of a kink in a garden hose.
Modern panels have bypass diodes that route around shaded sections, which helps, but the loss is still disproportionate: a shadow covering 5% of a string’s surface area can cost well more than 5% of that string’s output while the shadow is present. On a plain string-inverter system without panel-level electronics, one badly placed branch shadow at 10 a.m. every day adds up over a year.
The flip side: shade at the right time matters far less. A shadow that crosses your array at 7 a.m., when the sun is low and production is minimal anyway, costs almost nothing. A shadow parked over the panels from 10 a.m. to 2 p.m. — the peak production window — is the expensive kind. The question is never “is there shade?” but “when, where, and for how long?”
How to Measure Shading Before You Spend Money
Don’t guess, and don’t rely on eyeballing the yard at noon in January. The sun’s path shifts dramatically across the year, and a tree that’s harmless in June can shadow your roof all afternoon in December (or vice versa).
The best free starting point is PVWatts, NREL’s production calculator (pvwatts.nrel.gov). It won’t model your specific trees, but it gives you a realistic unshaded baseline for your roof’s location, tilt, and azimuth — the number every shade loss gets measured against. Run it first; everything else is a discount off that figure.
For the shade itself:
- Professional shade measurement. Installers use instruments like the Solar Pathfinder or Solmetric SunEye to capture the horizon and obstructions from your actual roof, producing a month-by-month “solar access” percentage. Any reputable installer should do this as part of a quote — if one won’t, that tells you something.
- Satellite-based estimates. Tools that use aerial imagery and lidar can flag roof sections with poor sun exposure. Useful for a first pass, not a substitute for on-roof measurement.
- Your own observation, done systematically. Photograph your roof from the same spot at 9 a.m., noon, and 3 p.m., near the solstices and equinoxes if you’re patient. Crude, free, and surprisingly informative.
If you already have panels, your monitoring app is the measurement tool. I check mine regularly, and a recurring dip at the same time of day, on sunny days, on the same part of the system, is shade signing its name.
What Shade Scenarios Do to Output: A Worked Example
The table below is a clearly hypothetical illustration — an example 8 kW south-facing array that would produce about 11,000 kWh per year with no shade at all. The percentages are the kind of ranges a shade study might return; your roof will differ, which is exactly why you measure.
| Shade scenario (example array) | Approx. annual loss | Example production | What it feels like |
|---|---|---|---|
| No meaningful obstructions | 0–2% | ~10,800–11,000 kWh | Baseline |
| Low winter-only shade (sun low in sky) | 3–6% | ~10,300–10,700 kWh | Barely visible on the annual bill |
| One tree, morning shade before 9 a.m. | 4–8% | ~10,100–10,600 kWh | Acceptable for most roofs |
| One tree, midday shade on part of one string, string inverter | 15–25% | ~8,300–9,400 kWh | Payback math changes noticeably |
| Same midday shade, with microinverters/optimizers | 8–15% | ~9,400–10,100 kWh | Electronics recover a real chunk |
| Heavy multi-tree canopy over peak hours | 30–50%+ | ~5,500–7,700 kWh | Trim, remove, or reconsider the project |
Two honest takeaways from that table: timing dominates (morning shade is cheap, midday shade is expensive), and panel-level electronics soften but don’t erase a real shade problem.
Mitigation Option One: Manage the Trees
The highest-leverage fix is usually the least technological one. Selective trimming — raising the canopy, thinning branches that cross the sun’s midday path — can transform a roof’s solar access without removing a tree you love. An arborist and a shade report together tell you exactly which branches matter; you often don’t need to touch most of the tree.
Things worth weighing before firing up the chainsaw:
- Trees have value too. Shade on your roof in a hot climate reduces cooling costs; mature trees add property value. The right answer is sometimes “trim modestly and accept a small solar loss.”
- Neighbors’ trees are their trees. Rules vary by state and municipality. A friendly conversation beats a legal one, and some neighbors will split trimming costs.
- Trees grow back. Budget for maintenance trimming every few years, not a one-time fix. A shade study done the year of installation goes stale.
Mitigation Option Two: Hardware That Tolerates Shade
If the trees are staying, system design can limit the damage:
- Microinverters (Enphase is the best-known maker) convert DC to AC at each panel, so every panel produces independently. A shaded panel loses its own output and nothing else.
- DC power optimizers (SolarEdge is the common example) achieve a similar per-panel independence while still using a central inverter.
- String layout matters even without panel-level electronics. Grouping the shade-prone panels onto their own string quarantines the loss instead of letting it infect clean panels. This is a design decision — raise it with your installer explicitly if part of your roof has known shade.
- Modern panel features like half-cut cells and better bypass-diode arrangements reduce how much a partial shadow costs within a single panel. Worth having, not sufficient alone for serious shade.
None of this is free, and panel-level electronics add cost and (arguably) more components that can fail. On a genuinely unshaded roof they’re optional; on a partially shaded one they usually earn their keep.
Keep Watching After Install
Shade management doesn’t end at commissioning. Trees add height every year, and a system that modeled clean in year one can be losing real money by year six. Make it a habit: once a season, on a clear day, look at your production curve. A healthy curve is a smooth arc; a bite taken out of the same spot every day is a shadow. Catching that early — while it’s one branch and not a canopy — is the cheapest mitigation there is.
If you’re still at the shopping stage, the order of operations is simple: run PVWatts for your baseline, insist on a real shade measurement in every quote, and make installers show you the layout and electronics choices they made because of your shade, not despite it. Shade is a solvable problem — but only for people who measure it.