Can Drones Fly New Orleans Levee Crests to Flag Seepage?

Can a drone flag seepage on a New Orleans levee crest?

After a heavy rain, a levee district crew walks the crest to look for seepage, wet boils, soft ground and slumped slopes. On a saturated crest the walk is slow, footing is poor, and the people doing it stand exactly where a failure would show first. A drone can fly the same crest from above, returning imagery and thermal data before anyone steps onto it.

The research supports the idea, with limits. A field study in Scientific Reports in October 2024, by Wang and colleagues, found that infrared thermal imaging can accurately identify potential levee seepage leakage points. The same study reported that detection was less effective in daytime than with visible light methods, and that drone altitude above the leak points significantly affected accuracy.

So a drone is a screening tool. On a 40 mile crest, narrowing the walk to the spots that look wrong is valuable, but a drone does not produce a verified seepage finding by itself.

What does a drone see on a levee crest that a walker misses?

Thermal cameras read surface temperature differences, and water moving through soil can show up that way at the surface. Visible cameras catch cracks, scarps, bulges near the toe and new wet patches that were dry at the last survey. A 2025 ISPRS Annals paper described a lightweight drone carrying thermal, visible and LiDAR sensors for dike leakage identification.

A 2025 Drones journal study tested a multi-sensor levee inspection in heavy rain in Heilongjiang, China. It examined thermal resolution at several flight altitudes and built a dataset of 739 low-temperature target sets. The lesson for a district is that altitude is a setting you test, not a detail you assume.

A 2026 ISPRS Archives paper by Hu and colleagues describes a multimodal drone method for levee piping, which is internal erosion. Infrared saliency first extracts suspect areas, and those suspects are then verified. That two step design is the right model for a crest survey: flag first, then confirm.

Is a drone allowed to fly a 40 mile levee crest?

A 40 mile crest is a beyond visual line of sight problem. Under Part 107 as it stands, BVLOS generally requires an individual FAA waiver, applied for through the DroneZone portal with an operational plan, a safety case and detect-and-avoid capability. Those applications take time, so the district should start the paperwork long before the first storm.

A separate Certificate of Waiver or Authorization pathway covers public operators, including infrastructure surveillance. A public drainage district may fit that pathway, so ask the FAA which route applies before you file. The answer decides who holds the authority and what the paperwork looks like.

The rule is moving. The FAA and TSA published a proposed Part 108 on August 7, 2025, with comments due October 6, 2025, and it would remove the need for individual waivers for routine BVLOS. The final rule went to the White House OIRA on July 10, 2026, and the latest reporting we found said it was still under review. The statutory deadline of January 16, 2026 and the executive order deadline of February 1, 2026 both passed.

According to an FAA official quoted by UAV Coach, the agency hopes to publish before the end of 2026. Until a rule is published and effective, routine BVLOS still requires a Part 107 waiver or a Part 135 certificate. Check the FAA's own status page before you plan the season.

• Part 107 today: individual FAA waiver generally required for routine BVLOS
• Public operator pathway: Certificate of Waiver or Authorization, to be confirmed with the FAA
• Part 108 proposed: would remove individual waivers for routine BVLOS once final
• Part 135 certificate: the other route until a rule is effective

How are federal levee managers using drones now?

The Army Corps of Engineers has used drones around levees for some time. Its Omaha District lets staff fly over affected levees and flooded areas without putting employees in danger, and it used drone imagery for levee rehabilitation after the 2019 Midwest flooding.

The Corps also tested a drone at the Bonnet Carre Spillway in St. Charles Parish, near New Orleans. Imagery from the flight could be processed into a 3D model, and a thermal camera could help detect seepage between levees, according to the Herald Guide report on that test.

The wider picture points the same way. An ASDSO webinar described FAA rule revisions and gains in personnel safety, speed, data volume and cost driving growing drone use on dams and levees. A USACE-sponsored Texas A&M ARIS project is developing automated and robotic inspection of earthen flood control systems, and proposes a team of heterogeneous robots, including UAVs and ground vehicles, to cover large levee areas.

Be careful with vendor material. Some vendor pages claim USACE-compliant workflows and validation figures, and we could not verify those against an official USACE source. We also found no official USACE levee drone policy document, so ask for any claim of federal approval in writing.

Where does a drone fit in a levee inspection routine?

A drone is best at the first pass. It covers crest length quickly, records the whole line, and returns a list of spots worth a closer look. The crew then walks those spots, and the rest of the crest does not have to be walked in the rain.

A drone is not the right tool for everything. It cannot probe soil, judge how much water is moving, or decide whether a wet patch is seepage or a runoff pool. Those calls stay with an experienced inspector on the ground.

The comparison below shows where each approach wins. The rows are general, so run your own tests before you decide.

How does New Orleans weather affect drone flights on levees?

New Orleans is wet. The National Academies climate summary gives a 1991 to 2020 annual average of 60.2 inches of rain, with a loss of over 6 inches compared with the earlier period. Axios reported on April 22, 2025 that New Orleans rainstorms are getting more intense, which makes the day after a storm the walk that matters most.

Heavy rain is a practical limit. The Heilongjiang study ran its tests in heavy rain, which shows the sensors can work in wet conditions, but it does not set the limits for your aircraft. Wind, rain and light limits come from the aircraft maker and your pilot, and you should get them in writing for your crest route.

Storm recovery is when the crest is most likely to need a look, so keep a flight window ready for the days after a storm. For official monthly rainfall, check the NOAA climate normals for the New Orleans airport station, because the figures we found were inconsistent.

How would we scope a drone job on your levee?

We are an OEM-neutral robot integrator. For a levee job we would pick the aircraft and sensor pairing that fits your crest, lease the unit through financing, deploy and integrate it with your data workflow, and train your staff to fly it and read the results. The unit is then serviced under our service plan, backed by regional service engineers in every state we service.

The lease payment comes in a quote after a free site assessment. In that assessment we would map the crest route, note where the line of sight breaks, and set out the flight plan an FAA application would need. We would also tell you if a ground vehicle or a contracted pilot fits your levees better, since drone flight is one option among several.

The FAA application and the pilot's authority stay the district's responsibility to confirm. We would not want you to commit to a flight program on the assumption that a waiver will come through.

What should a levee district check before deciding?

Most of the answer depends on a few facts you can check this month. Start with flight authority, because it sets the timeline. Then run a sensor test on your own levee, since research results depend on soil, vegetation and flight altitude.

The checklist below lists what changes the answer. Each item should have a named owner on your side before any contract is signed.

What would change this answer over the next year?

Two things would move the answer toward yes. A final Part 108 rule that removes individual waivers for routine flight would simplify the crest route. A sensor test on your own levee, where flags match what crews later confirm on foot, would show that screening works on your soils.

Two things would keep the answer at partly. If the crest route still needs a waiver after the rule changes, flights will be scheduled rather than immediate. And if the sensor test misses slumps your crews already know about, the drone is a supplement, not the first pass.

Until then, use the aircraft to screen the crest after storms, keep the crew walk for confirming flagged spots, and log both so each storm starts with better records than the last.