Drone Erosion Monitoring for Hillslopes

ISEF Category: Engineering Technology: Statics and Dynamics

Subcategory: Civil Engineering  ·  Difficulty: Advanced  ·  Setup: Home Setup  ·  Time: Full Year

The Hook

A small slope can lose more soil than you think after each storm. If you can map that change in 3D, you can turn a backyard hill into a real erosion lab. Drone photos, terrain models, and rainfall patterns can reveal how water carves land over time.

What Is It?

This project studies erosion, which means the wearing away and movement of soil by water. You build a small hillslope model, then use repeated drone photos to measure how its shape changes after each rain event. Think of it like taking before-and-after topographic scans of a sandbox hill, except your data comes from photogrammetry, a method that turns overlapping photos into a 3D surface.

Open Drone Map helps convert those photos into a digital surface model. That model lets you compare volumes before and after rain, so you can estimate how much soil moved. You can then compare your results with a USLE-style model. USLE stands for Universal Soil Loss Equation, a common way to estimate erosion from slope, rainfall, cover, and soil factors.

Why This Is a Good Topic

This is a strong science fair topic because you can change one variable at a time, like slope angle, ground cover, or rainfall pattern, and measure the effect on soil loss. You also get a clear real-world connection, since erosion affects farms, roads, and stormwater systems. The project teaches imaging, 3D modeling, data cleaning, and model comparison, which are all useful research skills.

Research Questions

• How does slope angle affect the volume of soil lost from a scaled hillslope under repeatable artificial rainfall? ?
• What is the effect of ground cover type on erosion volume measured by drone photogrammetry? ?
• Does rainfall intensity change the agreement between measured soil loss and a USLE-style prediction? ?
• To what extent does surface roughness reduce runoff paths and lower erosion volume? ?
• Which slope position, upper, middle, or lower, shows the largest change in elevation after repeated rainfall? ?
• How does the accuracy of Open Drone Map reconstructions change when image overlap or flight altitude changes? ?

Basic Materials

• Consumer drone with a stable camera and manual photo settings.
• Backyard slope model frame or landscape tray large enough for repeated tests.
• Soil with a known, repeatable texture.
• Measuring tape or ruler set.
• Digital kitchen scale with 0.1 g accuracy.
• Garden sprinkler or adjustable hose nozzle for artificial rainfall.
• Waterproof tarp or drainage catch basin.
• Ground markers for photo alignment.
• Tripod or fixed camera mount for ground reference photos.
• Notebook or spreadsheet for measurements.

Advanced Materials

• Higher-resolution drone camera with RAW or fixed-exposure capture.
• Ground control points with measured coordinates.
• Surveying grade scale bars or coded targets.
• Soil moisture probe.
• Laser distance meter or total station access.
• Infiltration ring set.
• Sediment collection trays.
• Drying oven or food dehydrator for sediment mass measurement.
• Reflectance panel for image calibration.
• GIS software for terrain change analysis.

Software & Tools

• Open Drone Map: Builds orthomosaics and 3D terrain models from overlapping drone images.
• QGIS: Compares elevation rasters, measures volume change, and maps erosion patterns.
• ImageJ: Checks image quality, calibration targets, and color consistency across flights.
• Python: Automates volume calculations, summary plots, and regression tests.
• LibreOffice Calc: Organizes measurements and calculates erosion rates without paid software.

Experiment Steps

1. Define one erosion factor you will change first, such as slope angle, ground cover, or rainfall intensity.
2. Design a scaled hillslope that stays physically stable and can be reset between trials.
3. Plan a photo capture method that gives repeatable overlap, lighting, and camera geometry.
4. Build a baseline surface model so you can compare pre-rain and post-rain terrain in the same coordinate system.
5. Choose an erosion metric, such as volume loss, elevation change, or sediment mass, and match it to a statistical test.
6. Decide how you will compare measured erosion against a USLE-style prediction and report any mismatch.

Common Pitfalls

• Letting the soil compact differently between trials, which changes runoff and erodibility without you meaning to.
• Capturing drone photos under shifting sunlight, which can break photogrammetry alignment and confuse surface matching.
• Skipping ground control points, which makes small elevation changes hard to trust.
• Using a rainfall setup that sprays unevenly, which creates fake erosion hot spots.
• Comparing models built from different camera angles or scales, which makes volume change look larger or smaller than it really is.

What Makes This Competitive

A class-level version measures before-and-after change once. A stronger version tests several slope designs, checks uncertainty, and compares measured erosion to a model prediction. You can raise the level further by validating your 3D reconstructions with ground control points, then analyzing where the USLE-style model misses local behavior. Strong entries usually pair good engineering control with careful statistics and a clear reason the result matters.

Project Variations

• Test how mulch, grass seed, or gravel changes erosion on the same slope model.
• Compare drone photogrammetry with ground-based photos or phone photogrammetry for volume accuracy.
• Add a runoff collection step and compare sediment mass to the 3D volume loss estimate.

Learn More

• USGS Water Resources: Search for erosion, sediment transport, and watershed monitoring resources and fact sheets.
• NOAA National Weather Service: Use rainfall data and storm terminology to connect your tests to real weather patterns.
• NASA Earthdata: Explore remote sensing, terrain mapping, and elevation datasets relevant to surface change.
• PubMed: Search review articles on soil erosion measurement, photogrammetry, and agricultural runoff.
• MIT OpenCourseWare: Look for open classes in GIS, remote sensing, and environmental engineering for background on mapping methods.