Satellite Imagery, LiDAR, or Drone? Choosing Site Data for Solar Design

A solar design is only as dependable as the site information beneath it. The challenge is that “site data” can mean several very different things: a satellite image, recent aerial photography, a LiDAR-derived elevation model, a drone reconstruction, field measurements, or some combination of them.

Each source answers a different set of questions. A clear overhead image may be enough to qualify a lead and sketch an early layout. It may not be enough to confirm roof pitch, elevation changes, hidden obstructions, or the dimensions an installer will rely on. A detailed surface model can support much more precise work, but it still needs to be current, correctly aligned, and reviewed by someone who understands the site.

The right question is not “Which data source is best?” It is “Which source is appropriate for this project, at this stage, for this decision?”

The short answer

• Satellite or aerial imagery is usually the fastest starting point for lead qualification, a preliminary layout, and customer conversations.
• LiDAR adds elevation and surface shape, making it useful for roof geometry, height changes, trees, terrain, and obstruction context where suitable coverage exists.
• Drone photogrammetry can create a current, high-detail orthophoto and 3D surface from overlapping site images, making it valuable when the final design must reflect present conditions.
• Field verification remains important for conditions that remote data cannot reliably show, including electrical equipment, structural details, access constraints, labels, hidden damage, and construction readiness.

Most strong workflows use more than one source. They begin with fast remote data, increase certainty as the opportunity matures, and verify what matters before construction.

What each source actually gives you

Satellite and aerial imagery

These are overhead photographs of the property. People often use “satellite imagery” as a general label, but many high-resolution basemaps are captured from aircraft rather than satellites. Either way, the image is primarily a visual layer. It can show roof edges, vents, skylights, surrounding buildings, trees, driveways, and property context.

Its usefulness depends on resolution, capture date, viewing angle, seasonal conditions, stitching quality, and georeferencing. A sharp image can still be outdated. A current image can still hide a roof edge under a tree canopy. It is excellent for seeing what appears to be present, but an image alone does not necessarily provide trustworthy elevation or pitch.

LiDAR

LiDAR measures distance by sending laser pulses toward surfaces and recording their returns. The result is a point cloud that can be processed into elevation products. For solar work, LiDAR may reveal roof planes, ridges, height changes, trees, nearby structures, and terrain that a flat image cannot describe.

Coverage, point density, classification, age, and processing all matter. Some datasets are intended for broad terrain mapping rather than detailed roof design. LiDAR should not be treated as a magic accuracy label. The actual dataset and the surface derived from it determine what it can support.

Drone photogrammetry

Photogrammetry reconstructs a site by matching common features across many overlapping photographs. With a well-planned flight, suitable positioning, and sound processing, the output can include a corrected overhead image, a point cloud, a textured 3D model, and a digital surface model.

Its main advantage is recency and detail. The capture can reflect the roof, additions, equipment, trees, and obstructions as they exist now. Its limits come from capture quality, legal and safety constraints, reflective or textureless surfaces, movement, poor overlap, weather, positioning, and anything the camera cannot see.

Manual and field measurements

Onsite work can answer questions no overhead dataset should be expected to answer. A technician can document the main service panel, conductor routes, attic or structural conditions, roofing material, access, fall hazards, labels, equipment clearances, and apparent damage. Tape, laser, inclinometer, photos, and notes may also be used to check critical dimensions.

Field information is not in competition with remote data. It closes the gaps left by remote data and creates a record for engineering, permitting, and installation.

Site-data comparison for solar teams

Choose the source by project stage

1. Lead qualification

The goal at qualification is speed. You need to decide whether the property appears suitable enough to justify more effort. Recent overhead imagery, parcel context, a rough roof outline, visible shading risks, and customer usage can support that decision.

Do not force permit-level certainty into a stage where the customer may never proceed. Record assumptions clearly, especially if trees cover roof edges, the image date is unknown, additions are visible, or the roof geometry is difficult to interpret.

2. Preliminary sales design

At this stage, the design must be credible enough to discuss system size, placement, energy production, price, and financing. Imagery may be sufficient on a simple, clearly visible roof. Available LiDAR or another elevation layer can improve roof planes, pitch, obstructions, and shade context.

The proposal should still distinguish estimates from confirmed conditions. If a change in pitch, usable area, service capacity, or obstruction placement could materially change the offer, flag it for verification rather than burying the assumption.

3. Detailed design and permit preparation

As the project approaches contract, engineering, and permit documents, the cost of a bad assumption rises. A current drone survey or disciplined field measurement can reduce uncertainty around roof geometry, setbacks, obstructions, height changes, equipment placement, and array fit.

This is also the point to reconcile site evidence with the electrical design, structural requirements, applicable code, utility rules, and authority having jurisdiction. A geometrically accurate roof model does not replace engineering judgment or local requirements.

4. Construction handoff

Installers need more than a clean array rendering. They need a design tied to current conditions, equipment selections, attachment assumptions, wire paths, safety constraints, and approved documents. Any difference between the model and the field should have a defined escalation path.

A strong handoff keeps the original imagery, survey outputs, field photos, notes, design revisions, proposal, single-line diagram, and permit plan set connected to the same project record.

How to evaluate remote imagery

Before tracing a roof, check the image as evidence rather than decoration:

• Capture date: Has the roof, addition, tree line, or surrounding construction changed?
• Resolution: Are roof edges and obstructions actually visible, or are you interpreting blurred pixels?
• Viewing geometry: Is the image close to straight down, or are tall objects leaning because of perspective?
• Alignment: Do imagery, parcel lines, roads, and other layers agree, or is one layer shifted?
• Occlusion: Are trees, shadows, snow, or adjacent structures hiding important surfaces?
• Completeness: Can you see the entire project area, including ground equipment and access?

If the answer to one of these questions undermines a decision, upgrade the data source or arrange verification.

When LiDAR adds meaningful value

Aerial imagery tells you how a site looks from above. LiDAR can help describe its shape. That difference matters when you need to separate roof planes, estimate pitch, understand height changes, model nearby trees, or represent surrounding terrain.

LiDAR is especially useful when it is recent, dense enough for the features being modeled, and processed into a surface appropriate for the task. It may be less useful where coverage is old, sparse, missing, or too coarse to capture small roof obstructions.

Ask a provider what dataset is being used, when it was captured, whether the product is a point cloud or a derived surface, and how its accuracy and density relate to roof-scale decisions. “Uses LiDAR” is not enough information on its own.

Digital surface model versus digital elevation model

The terminology is easy to blur, but the distinction matters. A bare-earth elevation model is intended to represent the ground after buildings and vegetation are removed. A digital surface model represents the upper surfaces detected at the site, which can include roofs, trees, equipment, and other objects.

For rooftop solar, a surface model is often the more relevant input because the roof and obstructions are exactly what must be modeled. For ground mounts, terrain may be central, but vegetation and structures still affect access, grading, shade, and layout. Always confirm what the elevation layer represents before using it.

When a drone survey is worth the visit

A drone survey becomes valuable when current geometry can change the design or when sending a person onto the roof would create unnecessary risk. Good candidates include complex roofs, multiple additions, dense obstructions, flat commercial roofs, sites with meaningful tree interaction, and properties where available imagery is old or unclear.

The capture must be designed for reconstruction. That normally means sufficient overlap, consistent exposure, complete coverage, stable flight, and useful positioning data. Ground control or high-accuracy positioning may be needed when the required absolute accuracy is greater than ordinary onboard GPS can provide.

The result should also be inspected. Look for warped roof edges, gaps, duplicated objects, noisy vegetation, poor alignment, and surfaces that appear smoothed or distorted. A reconstruction is evidence for a designer, not permission to stop applying judgment.

Powerlily’s drone site survey workflow accepts supported GPS-tagged photo sets or DJI video with telemetry, then returns a current orthophoto and measured 3D surface to the project for review and design.

Why “high resolution” is not the same as accurate

Resolution describes how much detail an image or surface can represent. Accuracy describes how closely coordinates or measurements match the real world. Precision describes how consistently a process repeats. These qualities can overlap, but they are not interchangeable.

A detailed-looking model can be shifted from its true position. A well-positioned dataset can still be too coarse to show a small vent. A measurement can be internally consistent while using the wrong roof edge. Evaluate the complete chain: capture, overlap, camera calibration, positioning, control, processing, alignment, interpretation, and verification.

The strongest approach is usually hybrid

Efficient solar teams do not order the most expensive survey for every lead, and they do not rely on a convenient basemap after the project demands more certainty. They increase data quality as risk and commitment increase.

1. Start with available imagery and customer information.
2. Build a preliminary model and record important assumptions.
3. Add available elevation or LiDAR when it materially improves geometry and shade context.
4. Use a drone survey or targeted field visit when present conditions or detailed dimensions matter.
5. Verify electrical, structural, safety, access, and code-specific conditions onsite.
6. Carry the accepted site model into the proposal, engineering documents, permit set, and field handoff.

This staged approach preserves sales speed without pretending that early information has final-design certainty.

Questions to ask before choosing a source

• What decision will this data support?
• What happens if a roof edge, pitch, tree, or obstruction is wrong?
• How current must the site representation be?
• What level of relative and absolute accuracy is actually required?
• Can the source show elevation, or only appearance?
• Will trees, shadows, snow, water, or reflective roofing interfere?
• Is an onsite visit already required for electrical or structural verification?
• Can the data be reviewed, corrected, and traced back to its source?
• Will the finished design continue into proposals and permit documents, or be redrawn elsewhere?

What good solar design software should do with site data

The software should let the team choose a source that fits the stage, preserve the source and its date, compare imagery with elevation, edit the interpreted geometry, and document assumptions. It should not trap the project in a single basemap or treat automatic detection as unquestionable.

More importantly, the site model should remain useful downstream. In a connected workflow, a refined roof or ground model updates panel placement, shade, production, equipment, price, proposals, electrical documents, and plan sets instead of forcing each department to rebuild the property.

See how Powerlily connects those steps in Designs & Proposals, or read the broader guide to running a solar EPC from lead through operations.

Final takeaway

Satellite and aerial imagery are fast visual starting points. LiDAR adds three-dimensional context where appropriate data exists. Drone photogrammetry can provide a current, detailed site reconstruction. Field verification resolves the conditions remote sensing cannot see.

The best choice depends on the project stage and the consequence of being wrong. Use the lightest source that can responsibly support the current decision, then increase certainty before assumptions become purchase orders, permit documents, or work instructions.