During the height of the COVID pandemic, human-generated ocean noise from freight shipping and recreational boating nearly stopped. Because some locales measure oceanic noise with hydrophones, open-source data offered lots of useful raw data. Artash Nath, a student researcher, analyzed some of this data to demonstrate how ocean noise impacts marine mammals and by what measure the noise was reduced when human activity stopped. He shares these findings on his website, Monitor My Ocean. We spoke with Artash about these insights in episode 26 of Soundproofist.
During the height of the COVID pandemic, human-generated ocean noise from freight shipping and recreational boating nearly stopped. Because some locales measure oceanic noise with hydrophones, open-source data offered lots of useful raw data. Artash Nath, a student researcher, analyzed some of this data to demonstrate how ocean noise impacts marine mammals and by what measure the noise was reduced when human activity stopped. He shares these findings on his website, Monitor My Ocean. We spoke with Artash about these insights in episode 26 of Soundproofist.
Cary (00:06):
This is episode 26 of Soundproofist. And my name is Cary.
Phill (00:11):
And this is Phill.
Cary (00:12):
And today we're talking with a student researcher from Toronto named Artash Nath. In fact, when Phill and I talked with him, Artash was still in the 10th grade. He's been analyzing ocean noise during the pandemic and its effects on marine mammal life. Artash has been using open-source data from different locations around the world, and we're going to learn more about him, what he's discovered, and what he hopes to do next with this information.
Artash (00:46):
Hi, I'm Artash, and I'm a grade 10 student from Toronto, Canada. And I'm really passionate about using my skills in science, big data, and machine learning on intergenerational and interspecies challenges. I've worked on several projects over the past few years, tackling problems from climate change and space exploration to COVID 19 .. . and most recently ocean noise pollution. I'm also fluent in English and French.
Cary (01:10):
Tell us more about the project you founded: "Monitor My Ocean."
Artash (01:14):
Sure. Well, the COVID 19 pandemic affected me greatly, as my school became virtual for almost two years. And during this time I noticed how human activity slowed down around me while urban biodiversity started to thrive. I decided to use this anthropause as an opportunity. Last year, I worked in measuring the sounds of the earth by analyzing seismic vibrations substations across Canada. And I enjoyed this project so much, I decided to do it for global oceans. And the fact that Canada was coming up with its ocean noise strategy in late 2022 meant that it would be the most opportune time to work on this issue.
So a bit about my project. Well, as for the background, low-frequency noise for marine shipping is an acoustic pollutant. And it overlaps with frequencies that marine mammals use to communicate and navigate this leads to stress and an increase in ship-mammal collisions. My research project, Monitor My Ocean establishes a model to measure the contribution of anthropogenic activity to the underwater soundcape, using several use of cumulative hydrophone or underwater microphone data from global ocean observatories.
Cary (02:14):
Just a quick note here, in case some of you don't know what a hydrophone is. It's a type of microphone that you can put in the water to collect acoustic signals and analyze them later.
Artash (02:26):
I program functions to extract ocean noise levels in the low frequency band affected by commercial and passenger ships. I applied statistical tools to calculate quantiles and means of anthropogenic activities and compare them at different times.
Cary (02:38):
You did mention that you had different anthropogenic activities that you had selected, and that included passenger ships and cargo ships. Of the locations that you chose, which were which? Both are anthropogenic activity, but one might be different than the other in terms of frequency or sound level.
Artash (02:56):
That is correct. So I actually looked at seven stations or hydrophones across global oceans. And all of them had slightly different sources of activity. So for example, Ontario being the Pacific, Stellwagon Bank in the Atlantic and Barcelona Coast were predominantly affected by only commercial shipping. On the other hand, Georgia Strait and Rangitoto Channel were also affected by recreational and passenger shipping and [...] shipping. And recreational and passenger shipping also creates noise in the higher frequency levels. Lastly, Glacier Bend, Cambridge Bay were frequented exclusively by cruise ships or tourism vessels. And for this reason, the decrease in ocean noise for these locations happened later, during the summer tourism season.
Phill (03:36):
And are those locations, did you notice any like pile driving or like industrial work for oil exploration or building offshore wind farms? Or were all of these locations too far from such industrial activities?
Artash (03:50):
That's an interesting question. So while these may not be located close to that kind of activity, I was looking at a station in the Gulf of Mexico. And the Gulf of Mexico is, you know, widespread with energy exploration and other such activities. And while these have not been published yet, my research has shown that noise levels there actually increased during that year. And this was driven because of the increased energy exploration that was done in the Gulf of Mexico during that time.
Phill (04:14):
And do you have access to hydrophone arrays in the Gulf of Mexico?
Artash (04:17):
Fortunately NOAA -- or the National Oceanic and Atmospheric Administration of the US -- runs arrays of hydrophones all along the US coasts. So there is one hydrophone located the Gulf of Mexico, near the Florida keys.
Cary (04:30):
So are you still retrieving data from some of these same sites now in 2022?
Artash (04:35):
Where possible I've continued getting noise levels for those sites as well. For example, in Monterey Bay, I was able to get comparison to the year 2021 as well. And it showed that noise actually increased in 2021 in that area because of the resurfacing of shipping, trafficking, global trade. However, for some stations, it was harder because due to the pandemic servicing these hydrophone stations became harder, which means there were wider gaps in data and availability. So I wasn't able to do this for all those stations.
Cary (05:02):
I see. So some of them went offline for a while or people were pulling in hydrophones and not putting them back out again, or...?
Artash (05:08):
That's correct because not all hydrophones are permanent installations. Some of them are set and then retrieved after a certain period of time.
Cary (05:15):
So yeah, I would imagine all activity, probably -- I'm guessing -- got lower in 2020 because of supply chain slowing down, obviously tourism crawling to almost a halt. Was there a particular time period? I don't know if you did it like for the entire calendar year, or did you just look at like X number of months into the pandemic? What was this time span that you looked at?
Artash (05:42):
Well, I looked at the widest time possible. But in particular for each of the stations I looked and analyzed the period, which was affected by the COVID 19 pandemic. So one thing interesting that I found for my research is that the stations that they did in the Pacific Coast decreased first in terms of ocean noise. And this was because the pandemic started in China, right? So the first to go down was the trade between China and North America and China to Europe. And the Pacific route is heavily used for trade between Asia and North America. So that was the first place where ocean noise decreased. And this was followed closely by the Atlantic shipping routes, like in Stellwagon Bay from my analysis or the Barcelona Coast hydrophone. So these decreased next, and this happened during May and June, and obviously the last decrease was Glacier Bay and Cambridge Bay because these ones were affected by cruise vessels. And that means the decrease happened during the summer tourism months, which was later than any of the other issues.
Cary (06:32):
I'm trying to remember, what month was it where Spain was in a huge lockdown in 2020? I think it was Italy first, but I think it was right around possibly June or July where Spain had a really huge spike in COVID. And I think a pretty serious lockdown. So that probably coincides when you talked about Barcelona.
Artash (06:52):
Correct. And also ambient ocean noise is an amalgamation of ocean noise from several countries because a lot of European countries use the Mediterranean Sea for trade and so on. So the decrease in ocean noise was representative of the decrease in trade from all these countries at the same time.
Cary (07:07):
And so of that, was there any -- besides what you've already mentioned -- was there any really interesting data either that you didn't expect, or just really interesting data that you uncovered from all of this research?
Artash (07:18):
Yeah, of course. So while the main purpose of my project was to evaluate and monitor anthropogenic ocean noise. I was also able to look in more detail in terms of some of the hydrophone data sets. And in terms of raw audio, I was also able to learn to identify the different signature sounds of marine mammals. So throughout my entire data set, I was able to detect the whistles of dolphins and the echolocation clips of these marine mammals and even the low frequency roots and pulses of different types of whale species. So looking at these signals was really interesting, but what was even more fascinating was that I could clearly see that the noise from these marine mammals overlaps with the sound of a passing ship in several instances. So this really emphasized, you know, the problem that we're facing here and the problem that my project aims to solve. That this ocean noise from shipping traffic is overlapping the sounds of biophony.
Cary (08:06):
Do you think that the shipping noise, when you say "overlapping" is causing the animals to vocalize more or that it is drowning out their ability to communicate with each other that's going on all the time?
Artash (08:21):
Well, what's happening here is the process called "acoustic masking." So acoustic masking is...but in the presence of shipping noise, marine mammals must vocalize louder or shift to different frequencies to hear each other over all this. And the problem worsens in the fact that some marine mammals in the presence of shipping noise may completely stop foraging for food, especially the females. So in this case, they're not able to use their echolocation abilities to properly define and locate prey and understand their environment. So this has a negative impact on the whale and dolphin species.
Cary (08:51):
So in other words, there could be malnutrition, lack of being able to feed their young and so on all caused by this anthropogenic noise. I think I've asked you, we've sort of covered already. I was gonna ask you: what's your assessment of the acoustical impact before and after the lockdowns? I think, I mean, we already know: less noise during the lockdown increasing noise now. And I guess what I'm wondering is where we're going right now and, and how much of that would show up in the current data coming from hydrophones. Because it seems like around the globe, we're making every attempt to kind of return everything back to the way things used to be. And that brings with it a lot of the bad, but I don't know whether or not we actually are returning to the way things used to be or something potentially even worse.
Artash (09:39):
Yeah, there are a few things to say here. First, to be more exact about the measurements, during the lockdown, the exact figure was that ocean noise decreased by 4.5 decibels during the lockdown on average for all the locations. The highest degree was around seven decibels in Georgia Strait. And as for how things are going now, in most locations, ocean noise has resurfaced, right? Because trade has gone back to pre-pandemic levels, and so on. However, this is not the end to this issue because the European Union recently has set up its MSFD -- the Marine Strategy Framework Directive for ocean noise. And Canada, US are expected to follow in their footsteps. And these strategies are aimed towards decreasing ocean noise in marine sanctuaries, and other protected ocean areas along these countries, exclusive economic zones. So that's why my project comes at a very opportune time, because my project can evaluate whether these policy changes are actually driving lower ocean noise instances.
Cary (10:34):
It's interesting. I don't know what the specifics are about the European Directive. I also wonder if -- since oceans are connected -- if the European directive, if everybody else doesn't get on board, you have a ship that's going between two different continents. If one has a directive and the other one doesn't... how does that work? I mean, specifically, can you cite, what are some of the parameters around ocean activity under the European Directive? Is it a frequency? Is it a decibel level? What exactly are they capping or stating?
Artash (11:07):
Well, this is still an ongoing policy,. But right now, some of the clear steps they've done is that they've mandated the monitoring of ocean noise levels up to the one kilohertz frequencies. And then they'll be put in place on, you know, restrictions and using this data that they're collecting to improve policies. So Canada and the US are doing similar things. And there's several projects that have already been set up in place as pilot projects to reduce ocean noise during the calming months of whales, such as the ECHO program near the port of Vancouver. So the ECHO program -- what it encompasses -- is that it recommends that ships slow down to around 10 knots while the calming grounds of the [...] in that season. And for doing this, they're recompensated a certain amount of money for the birthing fees. So this policy has been a success. And lower ocean noise levels were recorded in the Georgia Strait during this period. So a similar strategy are being set up in the US like the Go Slow policy. So it's still an ongoing project, and I'm excited to see where all these policies go next.
Cary (12:07):
And all of this data is helpful in making a case for improving and strengthening and enforcing these policies. Are there other people working with you on this project -- instructors, classmates, other fellow researchers --or is this mostly a solo project on your part?
Artash (12:23):
Well, this is an independent project. There's several other researchers around the world who I've talked to and I have shared hydrophone data for my study. So while I have a network of people around me that I've interacted with, this project has been independent.
Cary (12:36):
Do you have any suggestions of how people can use the data you've collected right now? And are you already talking to different organizations about using that data? I know you have a website, but how are you getting people to your data and information, or are you reaching out to other people... or exactly how?
Artash (12:57):
I've been sharing this project, presenting at different conferences, and so on to reach out to more people. Also youth programs. So I can get more youth interested in this issue. I did make all the...so the source code and analysis behind that project are open source, so that students around the world can replicate and adapt by project to monitor ocean noise levels at their locations. And hopefully together, once enough researchers and students are able to measure noise levels at their location, we can come up with like an annual report on ocean noise and global oceans.
Cary (13:24):
Yeah. I'm wondering also about, you know, there's different citizen science organizations where you can establish a project. I don't know if you've done that. Some of them are regional. Sometimes people want to get involved in a citizen science project and they don't know how. And I wonder if you've looked into any of that, any of the different organizations and setting it up so that people can sign on and get involved.
Artash (13:46):
Yeah, well, this project is not near complete. There's several other steps. I'm hoping to continue with the right project. And one of them does include, you know, sharing my project and make it more widespread. So I'm currently in the process of taking steps so that other students can find it easier to find this project and work on it as well.
Phill (14:03):
Have you encountered the DOSITS organization -- Discovery Of Sound In The Sea - dot org? Are you aware of them?
Artash (14:10):
DOSITS has been a very useful resource to read up on different aspects of ocean noise. So it's definitely something I've learned a lot from.
Phill (14:16):
So, okay. That's great to hear. And your project you mentioned is open source, this data analysis. Are you doing this manually or is this an automated process where you're actually analyzing the wave form programmatically?
Artash (14:29):
Well, I've coded all the functions behind my project myself. So, you know, these things involve frequency, computative analysis to extract noise in the lower-frequency band affected by shipping, as well as statistical analysis to calculate quantative levels and effective measurements across different years. So all these functions are custom made. I have used Python, which is an open-source coding language, as well as several other libraries like Node PI and SciPi and so on to this project. Yeah.
Phill (14:54):
Okay. And so if you're using Python and this is primarily ...the interface that I encountered was a website. So do these Python scripts run offline, and then you upload the data to the website or do these run on the server of the website so that additional data can be added.
Artash (15:08):
Well, the analysis of this hydrophone data is very computation heavy because audio data comes at a very high sampling rate, right? And so if the hydrophones that came up to 64 kilohertz, so there was almost 10 terabytes of data to analyze. So this obviously can't happen on the server. So all the data was pre-crunched on my computer. The analysis ran for several weeks and then the results of my project were posted on that web app in interactive way. Right? So the website runs off data sets that I've combined all my data from and made it into like a much simpler and lower form, so it can be added on a server. So the Python script and so on, and the analysis can be found on a Github repository, which will be linked to the web.
Phill (15:48):
That's great. And these different features that you extracted from the audio data, how did you determine which features were the most salient for your purposes? As you mentioned some low-frequency analysis among other things, how did you decide which features you would extract?
Artash (16:03):
Well, I looked at a lot of literature in other previous research to build up my model. So one of the most important things were research studies that were visualized as a Wenz curve. So the Wenz curve shows the contribution of different activities to the underwater soundscape at different frequencies. So for example, it shows that the effect of geophony or the natural sounds of the earth happened from zero to 10 Hertz. Anthropogenic noise from shipping traffic happens mainly between 10 and a hundred Hertz, while noise from wind and etcetera happened at the higher frequencies above one kiloHertz. So based on the studies by the Wenz curve, I was able to load frequencies to look at for an anthropogenic activity. And I ended up settling mainly on the 63 Hertz one tape band. This band runs from 55 to 71 Hertz and is most representative of the noise on traffic. However, I've also done analysis on the 125 Hertz and one kiloHertz tape band. So I can look at activities from different noise sources.
Phill (16:58):
And when you're analyzing noise, do you ever measure the quality of the noise or the more subjective impact of the noise using something like kurtosis for example, have you ever encountered that concept?
Artash (17:11):
Uh, no, I haven't.
Phill (17:13):
Okay. Well, I'll send you a paper by Michael Stocker -- - one of the first folks we interviewed here. But basically the, the quick point is that kurtosis is a statistical measurement you could do, and basically is like the smoothness or peakiness of a spectrum. And this can kind of subjectively be correlated to the noisiness of a noise itself and is a pretty good indicator of biological impact from a noise source. I can forward you that paper. But that might be an additional interesting statistical thing that gives you more of a subjective evaluation of noises, if that makes sense.
Artash (17:48):
Sure. That sounds interesting. So the first phase of my project focused more on the energy levels, different activities on a wider scale and not as high for resolution. While in the next phase of my project, I'll be focusing on the much higher resolution data to work more on the sounds of biophony.
Phill (18:03):
That's really exciting. It's very impressive, this work you've done. That's pretty cool. So I'm curious, you know, you mentioned that this was a consequence from doing school from home. Do you think that if you had just been in school in the normal historical way, that you would've had the opportunity to work on this project or the inspiration to do this project?
Artash (18:22):
Well, I have been in school for the past six or seven months. So the only time I wasn't in school was in 2020 and part of 2021. So while in the start of my project, I was at home the COVID 19 pandemic more than giving me more time, also gave me a rare research opportunity, because COVID 19 pandemic looking at it, it''s really interesting. It's one of the biggest stoppages of a anthropogenic activity, you know, in almost in human history, because so many things stopped at once all around the world. So more than the time I was given was how interesting the scenario was. And it motivated me to want to use a scenario to some advantage and to make some interesting work and studies type of thing.
Cary (19:01):
Yeah. I definitely think it was both a global opportunity, but also, like you said, I think you probably had more time outside the normal routine to take a look at it. And you're absolutely right. I don't think -- definitely not in my lifetime --there's never been anything like this that happened around the world that would have this kind of impact simultaneously. Hopefully we won't have another.
Phill (19:25):
So Artash, I think you already answered this question, but you've mentioned that you want to take a look at the more higher resolution data. Do you have any other future plans for the work you're gonna do with this project?
Artash (19:37):
Yeah. You previously mentioned somewhere that this is a global problem. It's not enough for the European Union to decrease ocean noise that's there. That's not gonna solve the problem because ocean noise does not respect national boundaries. It's a global problem. So all countries need to come together and agree on ocean policies. So we can actually take a stab towards this problem. That's why some countries have very advanced ocean noise strategies, but some have barely any. So I will be working with countries, especially in the coast of southern coast of Africa and South America. I'll be working and reaching out to these governments to help establish better ocean noise laws. And one of the things I'll be doing towards that is looking into low cost hydrophones because hydrophones are currently very expensive and difficult to maintain. So working towards lower-cost hydrophones will make it easier to see the ocean with ocean noise monitoring systems, to get a better idea of what policy we can take and evaluate these policies.
Cary (20:27):
Yeah. So have you ever, just out of curiosity, have you ever created or built your own hydrophone?
Artash (20:32):
In the earlier phases of my project, I did want to run some tests and see how a hydrophone works. So I built a very basic hydrophone at my home at Lake Ontario, which is pretty close to my house. So this was a very simple design, right? All you do is take a microphone, dip the end into a balloon and fill the balloon with oil. And then you tie the balloon. So you make it sealed. Then when you drop this into the water, it almost acts like a hydrophone that's able to, you know, to some extent measure underwater resonance.
Cary (21:01):
Interesting. So you, there really is a very low-cost alternative. And when you say a microphone, what exactly kind of microphone are you talking about? Sort of something that you buy that...
Phill (21:12):
A Radio Shack karaoke microphone?
Cary (21:15):
Yeah. Right. Maybe something like that. Yeah.
Artash (21:17):
Honestly like just a low-cost hydrophone. You know, at conferences you wear a microphone on your shirt, right? Just really small. So something like that. Obviously this isn't like an actual solution. This was just a pilot test that I had done. So hydrophones can be made for under a hundred dollars that are much professional than putting a microphone in a balloon. But this was a really cool idea. Just the test of the concept for me.
Cary (21:38):
When you mentioned like low-cost alternatives to try to engage areas that might think the cost of entry is too high. I wanted to know just how simple could it potentially be. And that's actually pretty darn good, a balloon and a cheap microphone and a little bit of oil.
Phill (21:55):
If you can solder, and you can get some piezo microphone buzzers and make some for probably less than $10 as well. But often the problem with these is you're not gonna get the sensitivity that you need over different frequency ranges, and you're gonna need a really good preamp to keep the noise down. And that's, you know, the really expensive hydrophones take care of all of those problems. And that's, I think that's the biggest challenge with the cheap hydrophones is that finding that middle ground between affordable and getting the fidelity that you want is kind of the holy grail on hydrophones from what I know.
Cary (22:30):
Yeah. And I don't even, I can't visualize myself, like where's the preamp in all of this. I mean, you've got the hydrophone out in the water, where's the preamp?
Phill (22:39):
You just have a really long cable, which is another reason you need a good preamp. So in my homemade hydrophone I probably have 20 feet of wire, so you can throw it down into the deep and then you have...and so you're gonna need a preamp to boost that signal when it gets there, but mine's a piezo buzzer in epoxy, inside of like a rubber gasket. And you hear stuff. You get a lot of noise, too.
Artash (23:04):
All the modern hydrophones, you know, the ones that are operated by NOAA and so on, do use a piezo kind of interface to get the best resolution. Obviously they're very advanced, but hydrophones can be pretty low-cost nowadays, especially with improving technology. So they can be pretty affordable. The problem apart from just the cost of the hydrophone is also maintaining them. Because they do need to be set out in the ocean. Some of them like the one Stellwagon Bank are like almost 50, 60 kilometers from the shore, so they need to be serviced sometimes. And there's a pretty long cable running back to shore. So obviously that can be a problem, but placing lower cost hydrophones closer to shore can be much more affordable.
Phill (23:40):
It occurs to me now ...you mentioned you tested your balloon microphone hydrophone in the lake. Is there any interest in measuring that traffic or understanding the impacts on the lake? And I understand those lakes have separate issues from noise noise might be, that might be the highest concern there. But the more inland, is there any interest there around anthropogenic noise?
Artash (24:02):
Well, yeah, as you mentioned, Lakestone really has bigger marine mammals usually. In the St Lawrence River, however, there's a large population of North Atlantic pike that are frequented. In fact, they started entering the St Lawrence River just a couple weeks ago. So now it's come to the season. So there is an interest in noise monitoring there. And there are also other policies that been implemented. Like for example, when a whale is spotted, they're given like, I think a two-week period in which they're not allowed to conduct fishing activities there. Because these whales, you know, they feed on different types of fish and so on. So that's another thing that has to be tackled there, but ocean noise remains large as well in the trading in the ocean.
Cary (24:39):
Definitely a use case for not just shipping routes, but all kinds of other water activities. I mean, all bodies of water, ultimately most of them are connected. So each one of them deserves a close analysis of how we're impacting them. So from here, as you mentioned, like maybe using some of your data to influence policy, where do you want to go? Where do you see yourself a couple years from now? You know, when you're moving on to university or onto... where do you wanna go?
Artash (25:10):
Well, I don't wanna limit myself to solving one issue. So I've always explored like a variety of issues, all other intergenerational and interspecies. So I'd like to continue using my skills in technology, machine learning, and big data analysis to continue working on big problems, whether it be ocean noise, exoplanets, pandemic, and so on. So I kind of work on these projects basically.
Phill (25:31):
So just this occurs to me as well. If ...I don't know if internships are a thing in your future, but there's a local organization in Santa Cruz, California called Conservation Acoustics [Metrics]. And they specialize in creating machine learning algorithms for autonomous microphone arrays. So they have a lot of microphone arrays in rainforest examples. And they're do using machine learning to identify gunshots by poachers or, you know, backhoes destroying habitat, or...and also identifying different species, getting a biomass count of different species, doing population monitoring, things like that. So that's like a pretty, you know, that's one of the first organizations outside of academia that I know working on those kind of things. And that seems to me to at least check the interspecies and intergenerational box of, of your interest. I guess all of that that's one recommendation, but then it also leads me to the question: have you ever considered applying these same concepts to terrestrial microphone arrays and not ocean hydrophones?
Artash (26:32):
Well, it's a similar process of analysis, right? Breaking, you know, into frequency competence, looking at spectrograms and so on. So it's definitely transferable technology, right. From either whether it be ocean hydrophones or surface microphones. So it's definitely somewhere where the same type of source code could be applied to, but looking at different things.
Phill (26:50):
No, I think that's great. I'm really impressed. I'm excited to see where this project goes and what you do in the future, man.
Cary (27:03):
So first of all, I'd like to thank Artash for speaking with us on the Soundproofist podcast. I'm super impressed with his work. And I look forward to seeing where he goes next. Just a quick summary of what we learned since the COVID pandemic started, Artash analyzed open-ource ocean noise, data collected by hydrophones in these locations in north America, it was Stellwagon Bay on the Atlantic coast, the Florida keys in the Gulf of Mexico, Georgia Strait on the Pacific coast of Canada, Glacier Bay and Monterey Bay on the Pacific coast ,and Cambridge Bay in the north of Canada. In Europe, he looked at the Barcelona Coast, and in New Zealand, the Rangitoto Channel. And during the lockdown, the noise in the oceans dropped just like noise decreased in the cities. On average, ocean noise decreased by 4.5 decibels and in the Georgia Strait it decreased by seven decibels.
Ocean noise disrupts marine mammal life, especially at certain frequencies. This noise impacts their ability to communicate with each other or to find food. Artash looked at previous research and determine that human generated noise in the oceans had a frequency range from 10 Hertz to 100 Hertz. So he focused mostly on noise data in the 55 to 71 Hertz range. The data he collected can help support policymaking decisions that will help to lower ocean noise and protect marine mammal life. One example of how this can work is to slow down the speed of ships to 10 knots during certain times of the year. You can find more information on Artash's website. The URL is monitormyocean.com. We'll also post a summary and some links on Soundproofist.com. Thanks for listening. And thank you, Artash for everything you've done.