Every year on July 4th, the city of Orlando puts on the ‘Fireworks at the Fountain’ show over Lake Eola — one of the largest municipal fireworks displays in Florida, with shells launched from a barge in the middle of the lake and roughly 50,000 people packed into the surrounding park. It’s the kind of event you can’t recreate, can’t reschedule, and can’t fly without doing the homework first.
This is the flight-notes article on how the 2025 footage in the catalog actually got captured — what I planned for, what surprised me, and what I’d do differently next time. If you’re a pilot considering shooting your own city’s fireworks show, this might save you a few mistakes. If you’re a buyer evaluating fireworks footage for licensing, this’ll give you context for what you’re looking at.
The TFR situation
The first thing to know about flying any major municipal fireworks display: there’s almost always a Temporary Flight Restriction (TFR) over the launch site for the duration of the show. For Lake Eola in 2025, that was roughly a one-mile radius cylinder up to 1,000 ft AGL, in effect from about 8 PM to midnight on July 4th. The TFR exists to keep aircraft (including drones) out of the path of the shells, which can climb to several hundred feet before bursting.
What this means in practice: you cannot legally fly a drone inside the TFR during the show without specific FAA authorization, and you almost certainly will not get authorization as a freelance pilot. The good footage of municipal fireworks isn’t shot from inside the TFR — it’s shot from a position outside the TFR ring, with telephoto compression bringing the show in tighter on the frame.
The launch position
I scouted three positions in the week leading up to the 4th. The criteria: outside the TFR ring; line-of-sight to the lake; preferably elevated to put the shell bursts at or near the horizon line; reasonable LOS for the drone with no major obstructions; and somewhere I could legally launch from (no private property without permission, no city parks with their own restrictions).
The position I ended up using was a private parking deck about 0.6 miles south of Lake Eola, with the property owner’s permission to launch from the rooftop. From that elevation the lake reads as a depression in the foreground, the downtown towers stack behind it, and the show happens in the middle of the frame. The sight line cleared all the major obstructions on the south side of downtown.
Exposure: the part everyone gets wrong on the first try
Fireworks are one of the hardest exposure scenarios for any video camera. The shells themselves are essentially point sources of intense light against an essentially black sky. If you expose for the sky, the shells blow out and read as white blobs. If you expose for the shells, the rest of the frame goes black and you lose the city context.
The setup that worked: ISO 100, shutter 1/60 (a touch slow for normal video, but it lets the burst trails register as streaks rather than discrete points), aperture wide open at f/2.8 to keep the ambient skyline visible, and a +1 stop overall exposure compensation pegged to the bright skyline lights rather than the burst flash. Manual white balance set warm (5,200K) so the shell colors stay saturated rather than going flat-white. No auto-exposure — auto would chase the bursts and turn the rest of the frame to garbage.
The other thing that worked: shooting D-Log color profile and grading in post. Burst colors are extremely saturated in raw capture and you can pull a lot more detail out of a flat profile than you can rescue from over-baked Rec.709 footage.
The composition
The hero composition for any city-fireworks shoot is the wide static frame with the skyline grounding the bottom third, the lake or bay or river in the foreground, and the bursts owning the upper two-thirds of the frame. The drone hovers in a fixed position, locked-off, no movement — the show provides the motion.
I also captured a slow pull-back during the finale and a slow lateral slide partway through the second movement of the show, but the static wide is the one that ends up in 90% of the use cases. Movement is a distraction during a fireworks show; the show is the movement.
What almost didn’t work
Two things almost killed the shoot.
First: the wind picked up about 20 minutes before showtime — not dangerously, but enough that the static-hover frame was visibly drifting. I had to shift to actively pilot-correcting against the drift to keep the framing locked, which is more attention than I wanted to spend during a 22-minute show. Future-proof: bring a more wind-resistant airframe for July events. The summer storm-front conditions in Central Florida are real.
Second: the show ran about three minutes shorter than the published schedule, which meant my finale capture started almost immediately rather than after a pause. I had budgeted a buffer for the finale and ended up burning that buffer instead on a recovery pull-back. Lesson: don’t trust the published runtime to the minute; have your finale move ready from the moment the second movement ends.
Both fixable. Neither catastrophic. The footage came out clean.
What I’d do differently in 2026
Three things on the list for next year:
- Two airframes. One locked off in the wide hero composition for the entire show. A second (with a second pilot) capturing the alternate angles — pulls, slides, finale. Single-airframe coverage means you’re always trading one shot for another. Two airframes means you don’t.
- Earlier scouting. The position I used was good. The position I would’ve picked with another week of recon would’ve been better — specifically, an elevated position northwest of the lake that would’ve put the bursts in front of the western face of the skyline at sunset rather than against the southern face at full dark. The light at the start of the show is the most cinematic light of the whole evening.
- A backup launch position. The week of the show, the property manager at my primary position got nervous about liability and almost pulled my permission. I had a backup ready about a quarter-mile away, but it would’ve compromised the framing. Lock down two locations next time.
For buyers using this footage
The 2025 Lake Eola fireworks footage in the catalog is captured under all the conditions described above — legal launch, outside the TFR, fully Part 107 compliant, with the licensing chain clean for commercial use. The standard tier covers the typical use cases: editorial b-roll, civic-pride pieces, social-media content, regional broadcast, festival film, real-estate creative for downtown properties.
Use cases I’ve seen this footage licensed for: news editorial coverage of the holiday, downtown-Orlando real-estate marketing, a regional tourism campaign, a non-profit fundraising appeal, and one international fireworks-industry trade publication. Each one used different cuts; one license covered every angle for each.
License the fireworks footage
2025 Lake Eola July 4th coverage is available now. See the full listing →