For passive acoustic monitoring during the memory chip shortage

If you've had trouble sourcing SD cards lately, you're not imagining it. Multiple supply-chain watchers have warned that AI-driven demand is squeezing memory chip availability and driving prices up.

The good news: the new Song Meter SM5 and SM5BAT (and many of our recorders) give you a lot of control over how fast you fill cards. Here we outline practical, field-tested ways to make limited card inventory go farther.

1. Start with the right cards (and avoid “fake capacity” disasters).

Use SDXC (64 GB–2 TB) full-size cards.

• Buy from reputable sources. Counterfeit cards are common on large multi-seller marketplaces. They can misreport capacity or underperform, resulting in negative consequences for unattended deployments.

• Tip: standardize one or two reputable models your team can actually source. Treat everything else as “needs testing” before it goes in the field.
  
• Tip: Don’t spend more money than you need. Unlike video cameras or game consoles, Song Meter recorders gain no benefit from the fastest or most expensive SD cards. Even the slowest cards from major brands have all the performance you need.

• Avoid SDHC (4–32 GB) cards. They are not recommended and can cause performance issues (including gaps) when writing higher data rates (e.g., two-channel ultrasonic and/or simultaneous ultrasonic + acoustic).

2. The biggest storage wins come from recording with W4V compression (NEW with SM5 and SM5BAT).

• W4V compression drastically reduces file size, but can introduce additional background noise in some scenarios. Both ultrasonic and acoustic recordings can be compressed in W4V:
  • W4V-8: ~50% smaller
  • W4V-6: ~62.5% smaller
  • W4V-4: ~75% smaller (most compression; most added noise)

• For many real-world outdoor sites, the environment dominates the noise floor anyway, so modest W4V compression often has minimal practical impact—but you should validate using your habitat.

• Workflow tip: Use the free Kaleidoscope Lite software to convert old WAV data to W4V, then re-run your analysis on the compressed files to test the differences before you commit to compressed recording.

3. Reduce the amount of sound you record (without missing target species).

Use the lowest sample rate that meets your needs.

• Our ultrasonic recorders offer a few full-spectrum sample rates: 256 kHz (default), 384 kHz, and 500 kHz. Many projects can meet objectives at 256 kHz, depending on regional species and protocol requirements.

• Our acoustic recorders offer several sample rates: 8, 12, 16, 22.05, 24 (default), 32, 44.1, 48, 88.2 (SM5 and SM5BAT, only), and 96 kHz.

• Higher sample rates allow you to record higher-frequency sounds, but they also produce larger files. In either case, remember: always choose a sample rate that is at least 2X the highest frequency you intend to record.

IF RECORDING BATS:

Record only one ultrasonic channel unless you truly need two.

• While two channels can be valuable (e.g., vertical stratification, dual-mic studies), it costs you storage fast. If your goal is standard presence/activity monitoring, one channel (as is typically used) is the most storage-efficient approach.

Keep triggered recording “tight.”

• Triggered settings affect how many files you write and how long each file is. Storage-friendly (but still biologically useful) levers:
  • Save noise files? → No (default) to prevent rain/insects/leaf noise from clogging cards.
  • Minimum trigger frequency: raise it just enough to cut non-bat triggers while staying below your lowest target call frequencies.

4) Don’t record what you won’t analyze (scheduling is your friend).

When cards are scarce, treat schedule design as part of study design. High-impact scheduling tactics:

• Set→Rise (or set-30→rise+30) instead of 24 hours, unless you’re intentionally capturing non-bat ultrasound.
• Add a duty cycle (e.g., 10 min on / 20 min off) if your objective is occupancy/activity trends rather than maximum call capture.
• Use date ranges (e.g., only run during the survey window) so you don’t waste storage outside your sampling period.
• Use Delay Start when you must deploy early but don’t need data yet (this protects both storage and battery).

5) Avoid “phantom full cards”: always format in the recorder.

• A classic gotcha: deleting files on a computer doesn’t necessarily restore the recorder’s usable contiguous space display. The recorder is optimized to write to contiguous free space, so a card can look “still used” even after you delete data.

• Fix: copy off what you need, then use Utilities → Format SD card 1/2 on the recorder before redeploying. This one step prevents a lot of mid-season surprises.

Quick takeaways for maximizing SD capacity

• Use SDXC (64 GB–2 TB), full-size cards from a reputable source
• Format in the recorder before deployment
• Use one ultrasonic channel unless your study design demands two
• Choose the lowest sample rate your protocol allows
• Use W4V compression when appropriate, starting with WV4-8
• Save noise files = No unless you have a specific reason
• Tighten triggers to reduce non-bat recordings
• Use set→rise + duty cycles + date ranges to avoid collecting data you won’t use

For product user guides, troubleshooting instructions, and other product documentation, be sure to check out our Knowledge Base.