I recorded the sound of a glacier calving in Iceland at 4 AM in March. The temperature was minus 12 degrees Celsius, the wind was gusting at 35 kilometers per hour, and I was lying on my stomach on a sheet of sea ice with a pair of DPA 4017 shotgun microphones pointed at a glacier face that was actively shedding chunks the size of apartment buildings. The resulting recording -- a deep, resonant crack followed by 45 seconds of cascading ice and a final bass boom that peaked at 118 dB SPL at 50 meters -- became the foundation for a creature roar in a film that premiered at Sundance two years later.

That recording session taught me everything I needed to know about field recording for sound design: preparation is everything, the best sounds happen when you're not expecting them, and the difference between a great recording and a useless one is often the quality of your wind protection. The glacier session was the product of three weeks of planning, and I got exactly 23 minutes of usable audio from four hours of recording. That's the ratio that defines professional field recording -- you spend far more time setting up and waiting than you do capturing the moment.

Field Recording Equipment Essentials

The minimum viable field recording kit for sound design work consists of a portable recorder, a pair of microphones, wind protection, and headphones. The recorder handles preamplification and analog-to-digital conversion. The microphones capture the acoustic signal. The wind protection prevents wind noise from destroying your recordings. The headphones let you monitor what's actually being captured versus what you think you're hearing.

For recorders, the Sound Devices MixPre-6 II is the current industry standard for professional field recording. It provides 6 channels of recording with 32-bit float preamplification, which means you never have to worry about clipping -- the 32-bit float format captures signals from -190 dBFS to +40 dBFS without distortion. The Zoom F6 is a popular alternative that also offers 32-bit float recording at a lower price point. Both record to SD cards and can output via USB to a laptop for extended sessions.

For microphones, the choice depends on what you're recording. Shotgun microphones (Sennheiser MKH 416, DPA 4017) are directional and excel at capturing specific sound sources from a distance. Stereo pair microphones (Schoeps CMC 6 with MK 4/MK 41 capsules, DPA 4006) capture a natural stereo image and excel at recording environments and atmospheres. Contact microphones (Barcus-Berry 4000 series, JRF piezo) capture structural vibration and excel at recording sounds that air microphones can't reach -- the resonance of a metal structure, the vibration of a tree trunk, the sound of ice cracking beneath your feet.

Wind Protection Strategies

Wind is the single greatest threat to field recording quality. Even a light breeze of 5 km/h creates turbulence around the microphone capsule that produces low-frequency rumble and broadband noise. At 20 km/h, unprotected recordings are typically unusable. At 35 km/h, even basic foam windscreens are overwhelmed.

The standard wind protection system is a two-stage approach: a foam windscreen (blimp) as the first stage, combined with a furry windshield (deadcat/windjammer) as the second stage. The blimp creates a still-air cavity around the microphone using a rigid frame and acoustic fabric. The deadcat's synthetic fur disrupts the wind flow, converting turbulent energy into heat before it reaches the blimp surface. A properly configured blimp-and-deadcat system reduces wind noise by 20-30 dB, making recordings usable in winds up to 50 km/h.

For extreme conditions -- the glacier session, for instance, or recording in a hurricane -- I use a third stage: a hard shell windshield (like the Rycote Nano Shield) that provides physical protection against rain and spray in addition to wind reduction. The Nano Shield adds another 5-8 dB of wind noise reduction and protects the microphone from moisture, which is critical when recording in environments where water damage is a real risk.

Recording Format and Bit Depth Selection

Field recordings for sound design should be captured at 96 kHz / 24-bit minimum. The 96 kHz sample rate provides headroom for pitch-shifting and time-stretching in post-production -- you can pitch a 96 kHz recording down two octaves (to 24 kHz effective sample rate) and still retain frequencies up to 12 kHz. The 24-bit depth provides a theoretical dynamic range of 144 dB, which captures the full range from the quietest ambient detail to the loudest transient without quantization noise.

If your recorder supports 32-bit float recording, use it. The 32-bit float format eliminates the need to set input gain levels, because the format's dynamic range exceeds any real-world acoustic scenario. In practice, this means you can record a whisper and a jet engine in the same file without adjusting gain, and both will be captured cleanly. This is particularly valuable for field recording, where you can't predict the dynamic range of what you'll encounter.

Planning a Field Recording Expedition

Every successful field recording session starts with planning. The planning process involves identifying the target sounds, researching the location, checking weather conditions, securing permissions, and preparing backup equipment. I typically spend 2-3 days planning for every day in the field.

Location research is the most important planning step. I use a combination of Google Earth for visual reconnaissance, weather APIs for wind and precipitation forecasts, and local contacts (park rangers, facility managers, residents) for practical information about access, noise sources, and safety hazards. Before recording in an industrial location -- a factory, a power plant, a construction site -- I contact the facility's safety officer and arrange supervised access. Industrial environments present real safety risks, and working alone in an active industrial zone without safety clearance is both dangerous and potentially illegal.

Recording Permits and Legal Considerations

Field recording in public spaces is generally legal, but there are important exceptions. Recording on private property requires permission from the property owner. Recording in national parks and protected areas may require a special use permit -- the US National Park Service requires permits for commercial recording activities, with fees ranging from $50 to $500 per day depending on the park and the scope of the activity.

Recording people requires additional consideration. In the United States, recording audio of conversations in public spaces is governed by federal and state wiretapping laws. Federal law requires one-party consent (meaning you can record a conversation you're participating in without notifying other parties). Some states -- California, Florida, Pennsylvania -- require all-party consent, meaning everyone in the conversation must agree to the recording. When in doubt, avoid recording identifiable speech and focus on environmental sounds.

"The best field recordings come from places where you've spent time listening before you start recording. Sit in the location for an hour without your equipment. Learn what the place sounds like at different times, with different weather, with different human activity. Then set up your microphones knowing exactly what you're trying to capture and why." -- Chris Watson, field recordist and founding member of Cabaret Voltaire, interviewed by Sound on Sound, 2019

Recording Techniques for Different Source Types

Different sound sources require different recording approaches. The techniques that work for recording a waterfall will produce poor results for recording insect activity. Understanding the acoustic characteristics of your target source helps you choose the right microphone, position, and recording strategy.

Water and Hydrophone Recording

Underwater sound recording uses hydrophones -- microphones designed to capture pressure waves in water rather than air. The Aquarian Audio H2a and the Cetacean Research C54 are popular hydrophones for field recording. Hydrophones capture a dramatically different sonic world than air microphones -- the speed of sound in water is approximately 1,480 meters per second (compared to 343 m/s in air), which means wavelengths are 4.3 times longer at the same frequency, and the frequency response is shifted accordingly.

On a recording trip to the coast of British Columbia, I deployed a pair of Aquarian H2a hydrophones at 3 meters depth in a kelp forest. The resulting recordings captured the cracking and popping of shrimp (the "snapping shrimp" phenomenon, which produces transient peaks up to 200 dB re 1 microPascal at 1 meter), the low-frequency rumble of distant boat engines (propeller cavitation produces a distinctive broadband noise centered around 50-200 Hz), and the subtle swishing of kelp fronds in the current. These recordings became the underwater ambience for a documentary about Pacific Northwest marine ecosystems.

Wildlife and Biological Sound Sources

Wildlife recording requires patience, distance, and an understanding of animal behavior. The most productive wildlife recording sessions happen during dawn and dusk -- the crepuscular periods when many species are most active. I typically arrive at the recording location 45 minutes before dawn and remain until 45 minutes after sunrise, which captures the dawn chorus of bird activity that provides the richest source material.

For distant wildlife, parabolic reflectors extend the effective range of shotgun microphones by 3-5x. A 22-inch parabolic dish with a boundary microphone at the focal point provides approximately 20 dB of gain for sounds originating on-axis, which makes it possible to record bird calls from 100+ meters away. The trade-off is extremely narrow directivity -- the parabolic dish picks up almost nothing off-axis, so you need to be precisely aimed at the target.

Organizing and Cataloging Your Field Recording Library

A field recording library is only useful if you can find the sounds you need. My personal library contains over 40,000 individual recordings accumulated across 12 years of field work, and the metadata system I use to organize them is as important as the recordings themselves. Every recording file is named using a consistent convention: date_location_source_description_take, followed by the file extension.

For example: "2024-03-15_iceland-jokulsarlon_glacier-calving_01.wav" tells me the date (March 15, 2024), the location (Jokulsarlon glacier lagoon, Iceland), the source (glacier calving), and the take number (01). The file's embedded metadata (BWF broadcast wave format metadata) includes GPS coordinates, microphone type, recorder model, sample rate, bit depth, and descriptive notes about the recording conditions and content.

Post-Processing Field Recordings for Sound Design

Raw field recordings rarely go directly into a final mix. They require post-processing to remove unwanted content, enhance desirable characteristics, and prepare them for the specific context where they'll be used. The post-processing chain for a field recording typically includes noise reduction, EQ, dynamics processing, and time/pitch manipulation.

Noise reduction is the first step. I use iZotope RX's Spectral De-noise to remove constant background noise (HVAC hum, electrical noise, distant traffic) while preserving the transient content of the recording. The key is to capture a noise profile from a section of the recording that contains only the unwanted noise -- no target sounds -- and then apply the reduction conservatively. Over-processing creates the characteristic "underwater" artifact of aggressive noise reduction, which is more distracting than the original noise.

EQ comes next. I typically apply a high-pass filter to remove subsonic content (below 20 Hz) that consumes headroom without contributing audible information, and I may apply narrow-band cuts to remove tonal interference (a specific frequency of electrical hum, for instance). Then I may apply a broad boost to enhance the frequency range that carries the most important information -- the 2-5 kHz range for detail and presence, or the 60-200 Hz range for low-frequency weight and impact.

References: Chris Watson, "Outside the Rainbow: Field Recording as Creative Practice," Sound on Sound, November 2019 | Bernie Krause, "The Great Animal Orchestra: Finding the Origins of Music in the World's Soundscapes," Little, Brown (2012) | Paul Virostek, "Field Recording for Fun and Profit: A User's Guide," Creative Field Recording (2015) | iZotope, "RX Audio Repair Handbook," 5th Edition (2023)