Google Earth Pro
Introduction
Google Earth Pro is the desktop version of Google Earth, offering advanced features, view historical imagery, and create high-resolution screenshots. It is a crucial tool within the context of open source visual investigations, allowing for the analysis of terrain, measurement of distances and areas, and verification of locations using historical satellite data. It offers other types of features than QGIS, making it a complementary tool in geospatial workflows.
Official Website: https://www.google.com/earth/about/versions/#earth-pro
Download
Download Google Earth Pro for desktop (Windows, Mac, Linux).
3D View
To see the full 3D mesh, you must ensure two specific settings are active:
1. Turn on the Layer
Look at the Layers panel on the bottom left.
- Find and check the box for 3D Buildings.
- Note: You can expand this folder (click the + or arrow) to ensure "Photorealistic" is checked, but usually checking the main "3D Buildings" box is sufficient.
2. Enable 3D Imagery
Go to Tools > Options (on Windows) or Google Earth Pro > Preferences (on Mac).
- Click the 3D View tab.
- Ensure "Use 3D Imagery" (sometimes labeled "Use high quality 3D imagery") is checked.
- If this is unchecked, Earth Pro may revert to "Legacy 3D," which displays simplified grey or user-generated models instead of the realistic photogrammetry mesh.
3. Check "Terrain"
Ensure the Terrain layer (bottom of the Layers panel) is checked. Photogrammetry relies on the terrain mesh to place objects correctly.
Understanding the 3D Models
1. The "Melted" Look: Photogrammetry (Automated)
This is the modern standard. It is not hand-made.
- How it works: Google flies planes in a grid pattern capturing millions of overlapping aerial photos. Algorithms then mathematically stitch them into a single continuous 3D mesh (like a blanket thrown over the city).
- Characteristics: Everything is 3D (trees, cars, terrain), but it often looks "melted" or "blobby" up close because it is an automated scan, not a drawing.
2. The "Clean" Look: Legacy Models (Hand-Made)
If you see buildings with sharp 90-degree corners, flat roofs, and clean textures (or just plain grey boxes), these are from an older era.
- Are they hand-made? Yes. From roughly 2006 to 2012, Google allowed users to build 3D models in SketchUp and upload them to the "Google 3D Warehouse." If approved, they were added to the map.
- Current Status: Google stopped accepting these user submissions years ago. They have largely replaced them with the automated photogrammetry mesh because it covers entire cities rather than just individual landmarks.
Measurements
Google Earth Pro offers a Ruler tool for measuring distance, area, and heading.
- Access: Click the "Ruler" icon in the top toolbar.
- Modes:
- Line: Simple point-to-point distance.
- Path: Continuous measurement of multiple segments.
- Polygon: Calculate perimeter and area.
- Circle: Measure radius and area from a center point.
- 3D Path/3D Polygon: Measures the actual ground or building surface geometry (requires "3D Buildings" and "Terrain" layers), capturing elevation changes.
Measurement Accuracy & Resolution Analysis
Accuracy in Google Earth Pro is not absolute; it is dependent on the resolution of the imagery and the geometry of the target.
Example Scenario using an 8-meter Rooftop
For an 8-meter rooftop measurement using high-resolution residential imagery (15 cm pixels), the typical error margin is:
- Linear Measurement: ±0.20 meters to ±0.30 meters.
- Slope Error (ignoring pitch): > 1.0 meter.
Here is the breakdown of these error sources:
1. Pixel Selection Error (Precision): ±0.21 m Because the image is composed of discrete pixels (e.g., 15 cm blocks), one cannot select a point with infinite precision.
- The Math: A linear measurement requires two clicks (Start Point, End Point). If you are off by just one pixel (0.15 m) on both ends, the combined random error is √(0.15² + 0.15²) ≈ 0.21 meters.
- Practical Limit: This ~20 cm "fuzziness" is the hard limit of the imagery's resolution.
2. The "Slope" Trap (Geometry Error): > 1.0 m This is the most common source of significant error for rooftops. The standard ruler tool typically measures horizontal distance (top-down 2D projection), not the slant distance.
- The Scenario: You measure a roof section that appears to be 8 meters in 2D.
- The Reality: If the roof has a standard 30° pitch, the physical length is 8 / cos(30°) ≈ 9.24 meters.
- The Error: Your measurement is short by 1.24 meters, far exceeding pixel error.
- Mitigation: Use the "3D Path" tool or manually calculate the hypotenuse if the pitch is known.
3. Relative vs. Absolute Accuracy
- Absolute Accuracy: Where the house sits on the planet (Global coordinates). This may be shifted by meters.
- Relative Accuracy: The distance between point A and point B on the same roof. This is generally much higher. Even if the map is shifted 0.30 m North, the scale of the house usually remains consistent.
Resolution Spectrum: Identifying Quality
Image quality varies drastically by location. Use this guide to estimate your resolution and corresponding error.
| Tier | Source | Resolution (Pixel Size) | Typical Location |
|---|---|---|---|
| Best | Aerial Photography | 15 cm – 30 cm | Urban/Suburban US, Europe, Japan. |
| Mid-Tier | High-Res Satellite | 50 cm – 60 cm | Developing cities, rural US/EU. |
| Low-Tier | Older Satellite | 1 m – 2.5 m | Older archives, remote areas. |
| Worst | Landsat / Sentinel | 15 m – 30 m | Deep wilderness, open oceans. |
How to Identify Resolution (The "Windshield Test")
Data providers are listed at the bottom-center of the 3D viewer.
- "State of Texas", "Sanborn": Likely Aerial (15 cm).
- "Maxar", "CNES/Airbus": Likely Satellite (30-50 cm).
- "Google", "USGS", "NASA": Likely Low-Res (15-30 m).
The Visual Heuristic: Zoom in on a car.
- 15 cm: You can see the windshield and rear window clearly.
- 50 cm: The car is a rectangular "blob"; windshield is indistinguishable from the roof.
- 15 m: Cars are invisible; roads are blurry lines.
Impact on Measurement Confidence
If you are measuring an 8-meter rooftop, the resolution defines whether the measurement is accurate, rough, or impossible.
-
At 15 cm (Best Case):
Visual: The roof spans roughly 53 pixels.
Result: You can click the corners precisely.
Error: ± 0.30 m (approx. 3.75% error). Use this for estimates.
-
At 50 cm (Satellite Average):
Visual: The roof spans only 16 pixels. The edges will be blurry (aliased), making it hard to see exactly where the gutter ends and the ground begins.
Result: Your clicks will be approximations.
Error: The "rounding error" of one pixel at each end is 0.50 m. Combined with georegistration issues, your error jumps to ± 1.0 m – 2.0 m (approx. 12-25% error).
-
At 15 m (Worst Case):
Visual: The entire 8-meter house fits inside half of one pixel (8 m < 15 m).
Result: Measurement is Impossible. The house will not be resolved; it will blur into the color of the surrounding yard. You cannot measure an object smaller than the pixel size of the imagery.
Your error margin scales with resolution.
| Resolution | Typical RMSE (Positional) | Measurement Limit (1 Pixel) | Impact on 8m Measurement |
|---|---|---|---|
| 15 cm | ± 0.30 m | ± 0.15 m | Excellent. Roof spans ~53 pixels. Error ~3.75%. |
| 50 cm | ± 2.0 m – 5.0 m | ± 0.50 m | Approximate. Roof spans ~16 pixels. Edges are aliased. Error ~12-25%. |
| 15 m | ± 30 m – 50 m | ± 15.0 m | Impossible. Entire house fits in half a pixel. |
Dated Satellite Images
Streetview
Save Image
Documentation for Methodology
To ensure your work aligns with the Berkeley Protocol and the Guide for Judges, use the following points to document your use of Google Earth Pro.
Methodology: How to document Google Earth Pro
In Your Method Section
Guidance: Describe the purpose, the tool version, and the input data.
- Investigative Objective: To analyze historical satellite imagery and terrain context of [Location].
- Software Version: Google Earth Pro [Version] (e.g., 7.3.6).
- Input Data: Satellite imagery dated [YYYY-MM-DD] provided by [Data Provider] (e.g., Maxar/Airbus).
In Your Decision Log
Guidance: Record the specific procedural parameters and integrity checks.
- Specific Settings: Imagery Date ([YYYY-MM-DD]), Eye Altitude ([Meters]), Terrain Exaggeration (Set to 1.0 or 0 for accuracy).
- Assumptions: Assumed the satellite imagery was correctly orthorectified, acknowledging potential registration errors.
- Integrity Check: Verified that "Terrain" was enabled/disabled as appropriate for the specific measurement task.
- Date Performed: [YYYY-MM-DD].
Verification & Mitigation
Guidance: How did you mitigate bias or verify accuracy?
- Verification: Cross-referenced features with other satellite providers (e.g., Sentinel Hub) or ground-level imagery.
- Bias Mitigation: Checked multiple dates of imagery to distinguish between permanent features and transient artifacts (e.g., clouds, shadows).
Common Limitations
- Imagery resolution varies by location and date; not all areas have high-resolution coverage.
- Terrain data (DEM) is smoothed and may not reflect recent changes or small-scale topography.
Example Methodology Statement
"To verify the presence of the structure in 2020, Google Earth Pro (v7.3.6) was used to access historical imagery dated [YYYY-MM-DD]. Terrain exaggeration was set to 0 to avoid distortion. The location of the structure was cross-referenced with Sentinel-2 imagery from the same period to confirm its existence."