Q: "I'm looking for a microphone to record sounds in nature. Since many of these sounds aren't very loud I've determined that I want to get a mic with very low self-noise. Most good mics have a rating for this, and I assume lower numbers are better, but I don't really understand the specification so I thought I should check."

A: If the listed spec is for self-noise then yes, lower numbers are better. In many cases it will be listed as referenced to dB-A, which is a common standard for self-noise measurements. Let's say a mic's self-noise is rated at 15 dB-A. Basically this just means that the noise produced and output by the mic is equivalent to its output from a sound at 15 dB SPL (which is 15 dB above the commonly agreed upon threshold of human hearing). The A means the measurement is "A-weighted," which takes into account human hearing's sensitivity to various frequencies. With the specification listed this way you can theoretically get a pretty good idea how the mic will perform in certain environments and what kind of signal to noise ratio you can expect. In the case above if you're recording sounds that are 30 dB SPL A-weighted, then the absolute best-case signal to noise ratio is going to be 15 dB.

There are, however, several things that contribute to noise in recordings. Even the self-noise spec of a mic can have multiple sources. A lot of self-noise, particularly in very sensitive mics, is caused by molecules randomly hitting the diaphragm. On top of that there is the self-noise of the wire used in the coil of moving coil mics, or the noise of the electronics in condenser mics. Though minor, the extreme amplification required to bring the low level mic signals up to line level can make them pretty significant. Moving outside of the microphone itself is the noise of the mic preamp being used. Most high quality mic amps in use these days have low enough self-noise not to be much of a factor, but if a mic has low sensitivity, or the sounds being recorded are really low in level, more gain will be required of the mic amp, which could cause its noise to come more in to play. Plus, even though a mic amp may have noise specs much lower than the mic it can still add some noise to the signal. Noise accumulates geometrically so it's not as if the noise of the amp simply adds to the noise of the mic, but it also isn't true that your signal will only be as noisy as the one weakest link. It's somewhere in between.

Some mics may only provide a specification for signal to noise ratio instead of self-noise. Higher numbers are better on this one. Fortunately you can figure out the self-noise from this if they provide enough info. Typically microphone signal to noise ratio is related to a reference acoustic level of 1 Pascal (Pascal is a unit of pressure and relates to "Sound Pressure" [SPL]). A typical rating might be 70 dB, 1 Pa @ 1 kHz. 1 Pascal is equivalent to 94 dB SPL. So, for a S/N ratio of 70 dB you can assume the self-noise to be 24 dB. There are a variety of testing methods, standard, and units of measurement that can come in to play to further muddy the waters, but you will find that most reputable microphone manufacturers will generally give you enough information in a standard enough format to make a comparison.

In general large diaphragm mics tend to have lower self-noise characteristics relative to their sensitivity than small diaphragm mics due to a phenomenon called Browniam Movements. They also tend to have greater sensitivity than their small diaphragm counterparts simply due to the larger surface area (for capturing sound pressure) of the diaphragm. Unfortunately, though there are advantages for large diaphragm mics with respect to noise, there are some potential disadvantages for doing recordings of things like nature sounds (transient response is one thing that comes to mind), many of which are beyond the scope of this Tech Tip (but have been discussed previously in others) so as always your final decision will include evaluating what your needs are and weighing the various characteristics.