DHI webinar: Risk Assessment of Environmental Noise Impacts

Just another WordPress site

DHI webinar: Risk Assessment of Environmental Noise Impacts

hello everyone good afternoon and good morning wherever you are welcome to the webinar on risk assessment of environmental noise impacts my name is frank Thompson and beside me here is Nicola high school and we will give a joint presentation on on this risk assessment of noise impacts the way we have planned is that is that we give half an hour or so 40 minutes presentation divided in two parts you are able to address questions once we have started with the presentation in the chat function we will collect these questions at the end and then we will have 20 minutes a question and answer session in the talk we will talk about sounder marine life what are the issues we will also look at how can those impacts be investigated best we will then have some summary on what’s new on the regulatory front and then nuclei will check over to explain a little bit about the underwater acoustic simulator or new software and Mike and and then we will mention the causes we are planning in spring this year in March and April now before we go into detail we want to describe what is really sound now sound sir is a traveling wave in a medium like water and it’s defined in decibels of sound intensity and decibel is denoted a sound pressure level in 20 log p divided by p 0 and this p 0 value is quite important because that’s a reference pressure now what one has to remember is that the reference pressure on the water is different from an air underwater it’s one micro Pascal in here it’s 20 micro Pascal so basically that means at decibel values in water now I can’t be compared directly another important variable is pitched the cycles of the sound wave per seconds are denoted in Hertz or kilo hertz so both these variables will pop up during the presentation decibel and hurts or kilowatts why are we interested in sound underwater now it’s quite simple what is just an excellent medium for sound transmission sound is more than four times faster underwater compared to air and there’s much less and generation so consequently sound ranges can be very fond of water and this is also the reason why sound is really important for marine life marine life has evolved to receive and to emit sounds and many marine life have very interesting sounds and I want to play back one sound here this is from northern resident killer whales and recorded about 20 years ago during my PhD in British Columbia you heard long ranging chords that the animals use in sustained country with one another so the the baseline is submarine life uses sound for communicating for navigating for orientation and for finding prey and if you look at impacts then we also look at their sensitivity what sounds do they here and this is shown here in this graph on the x-axis you can see frequency on the y-axis sound pressure level in decibel and these are audiograms these are hearing course of some marine mammals and they go from very low sensitivity in this area here to very very good sensitivity to underwater sound in the higher frequencies and if you look at this is an example here I have to go back a slide and activate the pointer so you can see here the harbor focus for example has a very bad or very poor sensitivity in lower frequencies and the sensitivity is increasing as we go along this is a green line here and it’s a very good sensitivity to underwater sound at very high frequencies at around 100 kilo Hertz and this is typical for marine mammals they hear they have a very wide bandwidth of hearing but they hear really well in the higher frequencies compared to the lower frequencies now for fish it’s a different story fish is more sensitive to lower frequencies than has a more restricted hearing a bandwidth and this is shown here you can see here that that

the x axis extending from 10 hertz to 1 kilo hertz and that the god here for example is quite sensitive in this area between 10 hertz and around 1 kilo hertz so fish here very here well in lower frequencies a marine mammals here well in higher frequencies and marine mammals have compared to fish a wider bandwidth of hearing and this is important if you look at it impacts if you look at the marine sound sources bedside marine animals we can see that there’s a whole lot of sound out there that is produced by both marine life but also by by humans and this is shown here on the x-axis again you can see the frequency from one hurts to all the way to above 100 kilo Hertz and on the y-axis again at the sound pressure level and you can see here that shipping is in the lower frequencies also in this case low intensity then you got submarines a little bit higher in sound pressure levels more intense than drilling and and dredging this our source levels back hairy-legged levels at the source and you can see that they are relatively low in frequency here this dredging sound here and that they go out with between 150 and 180 decibel now higher than that above that you can perhaps see the seismic argon array and seismic air guns are used for geophysical exploration and they’re quite loud they are above 200 decibel and the same is true for pi driving most pi driving sound useful of shore wind farm construction goes out with much more than 200 decibel the frequencies of these sounds are usually low with most energy below 1 kilohertz so basically the oceans are not quiet the oceans are quite filled with sound both from from marine life and also from humans now for impact assessments we it’s important to consider that we are following a risk-based approach and this risk-based approach is comprised of several steps from identifying what is the problem this is really the first step and true then investigating how far does the sound spread and how many animals on the range of the sound and this is something that we approach with our underwater acoustic simulator which nicole i will talk allow about a little bit more in detail later the next step is then once we know where the sound is and what is a exposure to the elements we want to know how do they react with the sounds we call that the dose-response assessment and if we find an impact on significant impact we want to manage these impacts by for example mitigating adverse impacts perhaps reducing the sound levels and I will talk you through all these different steps now if you look at the first bit identifying the problem this has taken from richardson a 90 95 and this is the zones of noise influence model and this art zones out from the center which is denoted red in the in the center here this red area is a sound source and then out from the center out to a certain distance there are different zones where noise can have an impact now relatively far from the from the sound source as a possibility of detecting a sound without apparent reaction a little bit closer by there the sound from humans have the possibility to mask biological relevant signals and this can lead to communication difficulties in in marine life now another zone even closer by or actually also overlapping with masking as a behavioral response zone and this is a very variable song because behavior responses are very variable they can range from very subject reactions like safe looking to outright avoidance of relatively large areas by marine life closer to the sound source even there’s a potential of hearing loss either as temporary threshold shift which is a temporary shift in the hearing threshold that we have seen in these audiograms and/or a permanent threshold shift that is affecting certain frequencies permanently so a permanent hearing loss at certain frequencies you can say and then closer by there’s a potential of and usually a very high received sound pressure levels there’s a potential of injury of marine life so all these impacts are can have them of course depending on a variety of factors now if you look at detection this yone on the left is a picture of hypothetical pi driving scenario polish waters and in the eastern Baltic and you can see here in the middle this this red area is a sound source the PI driving source impact by driving in the red area goes out to a certain distance as well above ambient sound and then you can see further out yellow and green the

this levels are still above ambient sound and they are approximately 30 to 50 kilometer from the sound source you can actually postulate with that that pie driving is audible / quite wide range from the sound source but of course detection for other sound sources can be much less depending on the soil strengths the sound spread and different ambient noise conditions so it’s really difficult to make any generalizations of sound detection at this pi driving is just one example now for masking we have to consider the sounds the animals emit themselves and the range of impact sounds if we look at this guy below here you can see that it ranges from 1 1 Hertz to 100 kilo Hertz and shipping has its main energy between and 10 Hertz and one kilo Hertz and tues weds for example they have a quite wide bandwidth of sounds this has to be expected because they have also very wide bandwidth of hearing but there are their frequencies and we mainly use our shift into the higher spectrum analogue to their hearing so this overlap but not a really significant one now seeds and sea lions have more sounds at the lower frequency sources a little bit more overlap and if you look at the fish here this is quite prominent you can see that most of the sounds of fish have what we call high am asking potential with regards to shipping sound so they can be masked by it by shipping in the same holds true as somewhat for larger waves which have also a lot of sounds at the lower frequencies for them masking could potentially be a huge issue now one of our behavior reactions behavior reactions are diverse they depend on a variety of factors like properties duration transient continues of the sound source of course then the channel the propagation characteristics of the and of course also on a lot of internal variables like age condition social state season and behavior state of of the receiver so it’s very very difficult again to generalize on behavior response because they’re just so very variable I have picked one example in a from a quite famous study in done in 1996 by Angus they looked at a response of haddock and caught true seismic gurgaon as sounds during a seismic survey remember seismic airgun sounds are quite loud above 200 decibel and they’re low in frequency so you would expect that there’s an overlap with the sensitivity of fish and what they did is they did catch the rates that is a codfish at different positions from the center out to distances of about 20 nautical miles from the sound source and they did that before during and after the seismic surveys and they found that the catch rate was much lower during compared to before the seismic survey took place and that led them to conclude that there was a quite large scale effect behavior affect because they concluded that fish would simply leave the area during the seismic survey and it took five days to recover truth so that the koechers would return to normal so this example shows that there can be actually quite profound there can be the potential for quite profound behavior reactions due to sound exposure now what does it all mean if animals just swim out they move to another location might not mean very much for for an individual now the population love the consequences of acoustic disturbance have been put into a model in the p cad model that was developed by the National Research Council in 2005 and has undergone interation since then and it really is comprised of several different steps from sound over to behavioural change life functions by drains affected and then finally population level effects and all these different variables are in these boxes and and they are transfer functions from one of the books to another so what type of sound has what type of effect on the behavior then type of behavior might have an effect on a live function affected so a sound has a certain duration it can affect diving for example diving for example here then during a migration it can affect a vital a vital life function of an animal what is really important if you look at the pig head model is that these transfer functions of the model are not really well understood as of now we have very little evidence pointing at the later parts of the of the model we know a little bit about sound and behavior change but not very much about the letter parts so there are a lot of gaps to be filled and they are what is ongoing right now that are precisely attempting this gap filling exercise we have to remember the ways well that strandings or behavior reactions can lead to imminent really serious consequences we know that strandings of

somewhere species have been caused by behavioral changes due to a naval sooner exposure to naval sooner sounds so they have arranged behavior changes can lead to two strandings and and death of animals now what about TTS temporary through hearing threshold shift in this regard the acoustic dose concept is really important so if you look at and this Harbor bob was here this popper swims a billion in waters and here’s a pie driving operation and what i did is i just calculated an impact range based on quite realistic assumptions that are shown here and if you look at one strike and look at impact criteria sound impact criteria then you would arrive at a zone of run round seven hundred meter from the sound source where TTS were a cure for these species now this is for one strike but if you have several strikes like for example over an hour the situation changes because the acoustic energy from the singer strikes is actually adding up and supposedly also hits it’s a receiver so if the harbour porpoise doesn’t it’s the impact range is actually extended from seven hundred meter to 12.5 kilometer in this case so the animal that was safe for one strike isn’t really safe for a number of strikes and E longer but in reality of course what really happens with these elements is that they move they move in response of environmental variables and move in response to underwater sound to a whole lot of factors and any impact assessment that is actually not looking at these movements is truly very limited now at ehi we have developed what we call a dynamic risk assessment model for acoustic disturbance to analyze movements of wales to underwater sound in response to underwater sound and we are doing this together with star doyle and what we are using for these investigations as a technology called agent-based modeling which is basically modeling the movements of particles in a model environment giving an unlimited number of variables and I want to play back 11 simulation so what you can see in the simulation and I suggest big over it is that the these are two different stocks of beluga weds a green is the eastern should Cece stock the proper one is a Western stock and they move into the Bering Strait this is on below here is at the US on the left side is or above is Russia and you can see the seismic seismic puts in the center there’s a seismic survey and you can see the agents moving out of the behavior impact zone and this survey has now stopped and the ancients move back into the area this is of course over a long period of time over time over of a survey before and during and after so what you can actually do with these models is you can investigate long-term consequences of these behavior exposures and of course these models have to be validated with with with real data but they can be used in cumulative impact assessments for example for marine spatial planning purposes and the data is getting better as we speak because a lot of studies out there now that actually look at behavior response of weights to noise now last but not least if you find this an impact for example using your dynamic risk assessment or other impact assessment methodologies and you have to it’s or options for the risk management and there are basically three ways you can do that you can mitigate sounds at the source for example up here using using this pi driving caps you put on a on a pile driver to just dampening the source a second reduction technology is what I call channel mitigation you put something in between the sound source and the animal in the water column where the sound is trapped and this has happens when you put in Bubba curtains out Bubba curtains or bubble emissions air bubbles that are released into the water and the sound is reflected at these bubbles and entrapped and they seem to be quite effective and there are different technologies like nms noise mitigation system that is used by dong energy in German warner’s and the hamlets resonator that is denoted here by a chap called Mark Warner in the u.s he has developed that system and they’re based on all the same principle on this airfield structures where sound is trapped and they seem to be of course depending on on environmental conditions quite effective quite effective in some cases now the third way you can reserve mitigate sound is using what I call receive a mitigation you can put in acoustic devices into the water to chase the animals out of the zone of immediate

danger and this is these are these equal marks here below which we also call acoustic pingers so basically three ways source mitigation channel mitigation and receiver mitigation now let’s talk a little bit about about regulation there are basically two important things going on right now one is in Europe and the marine strategy framework directive now the Marines ready frame attractive has a goal to for the member states are to achieve good environmental status in their war status in their waters by 2020 the European Commission hopes that the member states do that along 11 descriptors for good environmental status one of these descriptors is covering underwater sound and other forms of energy and within that descriptor there are two indicators one is covering the issue of a good exposure to loud low and mid frequency in pulsar sounds like this pi driver here the other one is covering chronic exposure to continues low frequency sounds are due to shipping leading to for example communication difficulties so there there are two indicators addressing two different issues and at the moment the European Commission is setting up advice to monitoring programs where the Member States can actually develop monitoring programs to measure these two indicators it’s basically a pressure concept and we’re monitoring is used to monitor the pressure rather than the impact as of now a second huge development that is ongoing as a revision of exposure criteria noise exposure criteria are criteria that shouldn’t be sound level that shouldn’t be exceeded in order not to hurt marine life or to injure it or 22 behavior reactions and in 2007 southwell it all came out with with the paper publishing a variety of noise exposure criteria and this is currently under revision and new is also revising their guidance for assessing the effects of anthropogenic sounds on marine mammals as we speak in this under preparation and both papers are working in parallel and there will no doubt be having a very high attention when they come out so this is basically covering my part a little bit the generic things the risk assessment and I have mentioned i think the exposure assessment when you look at the sound ranges now we have done much more on this and all you need to know about this will be told to you by nikolay now thank you Frank and the next part of the this webinar will be on modeling of underwater acoustics and with the new were developed underwater acoustic simulator aim by Mike powered by DJ I mike is the commercial deej I software suite M which holds sir a number of modules for planning and marine operations the underwater acoustic simulator in Mike is as some propagation model and it accounts for a noise source of broadband nature it is range dependent and which means that it takes into account and number of a ambient conditions these conditions are for instance the bathymetry which you will see over to the right and by clicking here I hope oh just need the pointer so here you see this is the seabed and then the structure of the seabed is accounted for it accounts for a temperature variation in the water column it accounts for the saline change in salinity and they also the absorption in the seabed is modeled which means that the sound can travel in the seabed furthermore a volume attenuation the water column is is a modeled which is significant for high frequencies with the underwater acoustic simulator a focus is on the risk assessment of a marine life and by embedding it into the mic framework a will have the mic graphical user fees which is easy to use

furthermore you will have easy access to a hydro dynamics and hydrodynamic sir which means that they from a mic HD model you will be able to extract the environmental properties needed for you are underwater acoustic simulation it furthermore its automated in this in the sense of a that all the new miracle settings are given automated for the user which means that you can they focus on on setting up the properties of your underwater acoustics model M furthermore it is automatic in terms of computational grid which means that all the the grid where the where the calculations are calculated a are given their to the user without helping to think about this examples of that applications could be offshore wind turbines say installation seismic surveys shipping drilling dredging and number of a different sound sources can be modeled in the underwater acoustic simulator a the the underwater acoustic simulator is an integrated solution in the mic suite which means that a that you were able to directly transfer your sound maps and do like Frank told about talked about an a dynamic risk assessment using a the agent-based model lap in in the Mike sweet and this is an heir to the rider is an example of a computation now I will take you to through some of the steps for doing air underwater acoustic analysis using the newly developed the UAS to a for this purpose I have chosen to look at the pile driving example where you for instance a used when you install the offshore wind turbines so you have a hammer which a hammers down this pile and sounds are generated from this hammering and I are radiated away from the pile in the water column and also through the soil this is a the calculations are done in order to assess the impact ranges for marine life so first you will will have to set up your domain and by this say you need to have included a your bath imagery and this UAS allows you to directly import your Betemit free through your mic a hydrodynamic model in the USA you have a this tree structure which you see out here which is a common way to set up a simulation in the Mike sweet so where you go through all these steps and end out at the bottom where you start your simulation for instance then the next step in setting up you your simulation would be a to define your sound source and this UAS allows you an easy way to do that you can either easily plug in a constant value for your sound source you can give a location or you can read in the spectrum through a file which you can see over here it’s a broadband spectrum read in you can also scale your spectrum if you want if this is needed now a the second part that you need to do is say to define the environmental properties in your water phase in first you need to specify the speed of sound depending on the how complex your problem is you can either have a constant speed of sound maybe a you can use that in case of shallow waters or you can have a you can read in em and a variation of the sound speed in

the vertical a border column and as you see over here yeah the sound speed profile can vary do during a season em next you need to set up your water phase in terms of attenuation and this can be done a in a number of ways a we have a four different models here and if you are in doubt you can always seek online help by pressing f1 and this will give you an open up a window where you will get a short description of what the settings you go you have here and there then for instance you see that they you can either set the attenuation as a constant or you can read in your own definition file of a volume attenuation in the water column or you can let the computer calculate a volume attenuation based on the constant values of temperature salinity or acidity or you can have a profile a file of the temperature salinity and acidity imported in in and then a true a calculation based on third Francois and Garrison you will get a body imagine you Asian profile for for your simulation here we have chosen to wear have constant values for instance temperature of 13 degrees and salinity of 18.5 and that CDT of eight next you need to define a your properties in your seabed for this you take your geological profile and then you make an assessment of this and us then gives you the option of a defining a and a battery number of a seabed layers for instance in this case we have sand clay and chalk and you define these the acoustic properties by saying em for instance the density and the compressional wave speed and or n compressional attenuation needed for the computation finally you will have an easy access to where define your output which a you have two different kinds of output you have a 2d output and a 1 the output in this figure it’s shown the 2d our output which option you have for instance you can have you can specify transmission loss for each frequency you can have the sound exposure level for each frequency calculated as an output or you can have a the overall values of a transmission loss sound exposure level you can also choose if you want to have the output of the sound transmission in the seabed or not and this is helpful there to save a computational memory on your hard drive if you don’t really need the sound field in the seabed the common work for flow for a noise impact assessment is that you take your your marine filter for the species relevant for the area that you are looking at for instance here we have air pinnipeds and this is the M waiting filter then you filter you are calculations or your results using a threshold either it could be a pts permanent threshold ship or TTS depending on what kind of analysis you want to use and these of course depending on which species you’re looking at and then by applying this you will see some contours in your noise map and from these contours you can calculate the impact ranges which means that they when do these thresholds are they a are they not are they are they is seeded so we’re UAS provides an easy way

to way QA you are raw data through the mic sweet so way by clicking your result file you will be able to easily see the raw data and for instance here we see the transit plot for this calculation with them with the pile driving and you have the sound source in sound source at this side and then sound is radiated through the water column and through the seabed but I’ve chosen not to show a click of the sea bed too see the sound transmission in the seabed and furthermore you are also able to specify the resolution of your a of your resolve file which means that a you can also save memory and this is why this is like the Brazil you can see it has a very coarse resolution but a by defining a high-resolution you will get a smooth a better imagery game outline to the left you see the sound transmission loss and this is plotted against a geometrical spreading laws just to make a quality assurance of a your data that it’s not a way of after this accusing you can use a different tools to make your plot for your reporting for instance the mic suite offers a different kind of ways to do plotting or you can use a matlab tool to generate a moray an isolating plot of your lab results for instance over here you will see the sound transmission in terms of frequency content so you have a this the black line up here is so they just very very close to the source and as you go further away from the source you can see that the the the sound is attenuated and at the the at the farthest away you will see that the sum of the frequency totally drops out because it’s attenuated and you can do also a number of other analysis for instance of here are chosen heat show you a that you have you are results of the sunday a exposure level here in in the blue curve and this is the radial direction away from the source and by applying a mitigation measure you will be able to reduce your son sound in the way that it’s shown in the green line near you can also have a direct cumulative analysis which means that you will account for the number of of blues with your hammer during one hour and then by applying a a source mitigation measure you will be able to reduce the sound as shown here by the purple from unmitigated to mitigate it sound the US has been validated against to a two very common test cases the first one is the lloyd mirror test gate which is a test case where you apply a source at the sea surface and then you see reflections down the second one is the ideal which a test case both get test cases was successfully run and matches almost perfectly with the with the analytical and numerical results giving in the papers so summarizing for the first time ever we have now a unique integrated solution which means that you have together with your underwater acoustic simulator a an option to directly import hydro dynamic properties as shown here into your acoustic model and then when you have done your acoustic simulation you can combine it with an animal response simulator as the APM lab and then you will have a dynamic risk assessment of your of your a sound field furthermore a the u.s. provides a time-saving interface which means that

you can easily come you can very quickly do your acoustic simulation as it is automated as I told you in the sense of you don’t have to think about numerics or a jitter ingeneral it provides a basis for managing impacts of underwater sound on marine life now a Frank and I we can either help you do this kind of simulations or you have the option of a buying a software a la license for either the underwater acoustic simulator by itself or you can have like the whole package of integrated solutions going from the hydrodynamics all the way to the dynamic risk assessment tool now I’ll give the word to Frank again and he will conclude this is a webinar and also give you some information about an upcoming exciting a software course on the UAS and thank you very much Nikolai yes everyone a couple of slides to finish things off I think we have I’ve learned today that men made noise can impact marine life in quite various ways we believe that the sound impacts our best assessed using a risk-based approach and we also think that the AI has developed novel tools to assess sound impacts we have the agent-based modeling for cumulative impact assessments and looking at response movement of whales and then we have the underwater acoustic simulator or software true to actually measure sound ranges or model sound ranges are different distances for the impact assessment we go into much more detail very soon we have two causes coming up one in Denmark and the ninth of March and one in Singapore on the 13th of April and they will cover modeling of underwater acoustics we will go much more in detail into the risk assessment framework and we will also make you actually the experts and using the acoustic underwater acoustics in the later because we have some test case you can run so it’s one day course and it’s we hope it’s going to be very interesting for you now that’s about it when you have questions please address them the acoustic questions to Nikolai and their biological questions to mean there’s also of course our software support that them can help you with all software related questions and with that I conclude this webinar a talking session I should say and we’re up to the more interactive part now we have 43 minutes up so that’s we have roughly 15 minutes left for questions