by Brian Woods
A major component of the work performed by the Lowlands Bigfoot Research Group involves analyzing many hours of audio captured in the field. We are also sometimes asked to evaluate and comment on audio clips that others have recorded. Recently, for example, I received a short sound clip via email from a property owner who wanted help determining the source of repeated howls.
Whether the recording is my own, or from someone else, I follow a procedure each time. The process can be adjusted depending on the length and quality of the original audio, but the overall goal of documentation and analysis remains the same.
The audio clip that I received came with no accompanying claims or preliminary guesses of what had been recorded. The question was open, which was my starting point. My next step was to listen to the file in original form, with no enhancements or filters applied. A good set of noise-cancelling headphones is essential for audio analysis. Not only do I want to focus on the audible elements in question, but I also want to hear any subtle sounds that could help identify a cause or sequence of causal events. I always use headphones, and you should too. Before I even think about applying any high/low pass filters or enhancements, I always make a copy of the original file, and name it in a way that indicates that it's a duplicate. This reduces the chance that I'll overwrite the original and allows me to revert back to it if necessary. I also suggest creating a separate folder for originals for the same reason.
Once I have a copy to work with, I open the file in Audacity, which is a free audio program that I use on my PC. https://www.audacityteam.org/ I won't get too technical with the various features and abilities of Audacity here, as there are already some great tutorials available on the internet and YouTube. The audio I received was good enough quality that only a high pass filter was applied to mute a few bumps from handling the recorder. Viewing the audio as a spectrogram made the repeated howls very easy to identify..
Looking at this spectrogram, it's possible to quickly count the howls in this timeframe, see their similar qualities and lengths, and determine the frequency range they all occur in. If this had been a lengthy file, I would be able to scroll along and scan for other howls based on appearance alone.
The next step, after becoming familiar with the howls, was to begin comparing them against known, identified recordings of animals that could be the source. This is where an internet or YouTube search can be valuable. Consider your sources carefully, though. Not everything you'll find is labeled correctly, and excessive or unnecessary enhancements made by others can transform audio into an inaccurate representation of the original sound. I already had my suspicions about what animal was making the sounds, so I listened to several examples and compared them to my file. It didn't take me long to feel satisfied that the mystery was solved. I concluded that that the howls were produced by a wolf or coywolf. Listen below and see if you agree.
This is the original:
Now compare it to a few examples found on YouTube:
To further support my explanation, I read more about the vocal qualities of wolves and coywolves. Of relatable interest is this:
"And even their cries blend those of their ancestors. The first part of a howl resembles a wolf’s (with a deep pitch), but this then turns into a higher-pitched, coyote-like yipping." - Roland Kays, North Carolina State University
If you listen again to the original, you'll hear this exact sequence of a howl followed by several barks. If I had not been able to find audio examples that sounded so similar, I would have continued to search for other clips, and branched out into audio of other canids. Some, such as fox, clearly wouldn't have sounded like the howls, and would be quickly eliminated as possibilities. This process would have continued until the list of candidates was as short as possible.
When evaluating potential animal sources for recorded sounds, you should also consider whether such animals are known to inhabit the research area. State DNR and wildlife studies remain the best sources for this information. While certainly not perfect, they can be valuable for migratory trends, population estimates, and recent studies or reports.
In summary, audio plays a significant role in the research performed by the Lowlands organization. Analyzing the audio by using a systematic approach can provide valuable information to the researcher. External sources such as information found on the internet can be of great assistance but should be used carefully.
I hope this has been helpful and informative for others interested in identifying animal sounds recorded during time spent in the great outdoors!
Brian Woods
Lowlands Bigfoot Research Group