What Does A Healthy Rainforest Sound Like?
Today we're speaking with Sarah, Seti about ecosystem health monitoring using sound. But before we dig into it, let's first look at one traditional method for evaluating the health of an ecosystem. Say you're interested in measuring bird biodiversity. For instance, you might use the point count method where you stand outside for hours on end with a lot of patience and a talented pair of ears. Every single bud you have a closing will you see visually united down what species that was what time you saw it? You kind of repeat that thing over the twenty, four hours a day different house at different locations it's. A super thorough process for monitoring ecosystem health, but incredibly tedious. So Sarah and his colleagues thought you know with all this modern technology, we have sensors, wireless networks. Solar panels there has to be a more efficient way to do this. Can we get something this? So approximately as good as this kind of data, but is completely yours mason methods, recorders uploading audio to the internet straight from the field allowing them to potentially track ecosystem health. In real time, they've set up this acoustic monitoring network in Borneo part of the safe project which records audio continuously, and it is a staggering amount of data. About seventeen thousand dollars so far from the network seventeen thousand. One seven zero. Gosh but it's not just background noise housed in those seventeen thousand hours is a treasure trove of. Impossible for us, mere humans to listen through. But fortunately, the folks Google have figured out a way to sort through all that audio Sarah and his team turned to Google's set a massive data set of sounds that was developed using machine learning Odier says done is it has labeled data for. kind of almost every type of sound that you can imagine that being an so from that a cannon knows the amongst all dog box. There is something that is consistent about old dogs that makes it dog and so it knows the this is one fingerprint, and then amongst all of gloss smashing into, it knows that says finding things the kind of we as humans perceptually consistent in touch the sound. And then fixing them down to one type of fingerprint taking Google Technology, they applied it to their forest recordings, training their machine to create an audio fingerprint a way to kind of identify that forest brew, it sound, and the algorithm they've developed can potentially predict important indicators of a forest health like habitat quality and biodiversity based on it soundscape alone, and it didn't just work in one particular kind of Forest Sarah and his co authors analyzed the audio recordings of forests around. The world they published their findings this summer in proceedings of the National Academy of Sciences What did you and your team show with these audio recordings beyond the fact that yes, the technology worked, what did it reveal about the character and what's happening in these forests? What you see quite nicely full out from all of this is we looked at really Clinton is diurnal patterns so that so of how they annoy a different in how odier consistently follows the same kind of trajectory. Fingerprints when you can start see West season change in weather day. So how evolve through the day as the sun comes off goes down and how that changes the species communities of licensing. That's lovely because we think about days and years and months mostly in relation to light like the sun coming up the sun coming down. Yeah. But you're saying there's like a rise and fall of sound exactly throughout the twenty four hour day that you. You measure? Yeah. Exactly. Unto the point where you can and we did this analysis within the papers that you can just take random piece of audio and you can gas with pretty good accuracy. Our the rodeos recorded that again, is questionable. What's what's the point in time I recall it Kinda. Just shows you the amount of information that's like temporary coded in this audio as well, and you can guess what month is recorded from so. We're GONNA actually listen to some of the sounds that your team has recorded from the safe acoustics website, acoustic dot, safe project, dot net. So these recordings they're all the all uploaded like wirelessly. Yeah. Wow Cova did not live right now, but they would normally be recorded in real time and uploaded. So you'd be able to listen to the forest sounds like an all these different locations right now. So, this very mood setting. The rain at night in an old growth forest. In Borneo. Yeah. He spent some nights under a tent in these conditions. Yeah. I mean this kind of rain is like Gold Senate because most of the time spent doing fieldwork sweating. So when the rain comes in, it's Nicer Wendy and COO. Yeah Music