#113 Coastal Modelling 101- Oceans, coasts and estuaries

[Music] welcome to the australian water school the home of demand-driven industry design training for the global water sector well welcome everybody to this australian water school webinar so good to have you all here coastal modeling 101 this is going to be good mitchell michael we're so glad to have you also we want you to sit back relax get your coffee this is a great way to do webinars it's a fantastic way to do webinar in fact let's see who's here look at this unbelievable over 50 countries represented it's terrific we're going to have a great discussion i reckon all right mitchell michael are you there that's fantastic yes they are wonderful hi hi michael hi mitchell how are you today yeah good thanks trevor and uh yeah good great to be here hey everyone it is today in your in your area of the world is it i think a lot of people are watching today it's not today it's it's like midnight or something where are you from michael um in brisbane brisbane and mitchell yeah same place brisbane brisbane the wonderful warm part of australia called queensland that's fantastic that's right that's excellent yeah which was going to be presenting michael's going to be doing the um uh answering on the q a panel and clarifying any questions come on screen whatever you want michael correct your colleague mitchell whenever you feel like it although you won't like that very much [Laughter] it's pretty standard that's good i don't think so this is looking does it look a very non-standard the way i look at your powerpoint mitchell it's fantastic mitchell's an associate principal engineer and leads two flows software development twoflow fv's software development and michael is the lead water quality development person within two flow bmt australia software business so looking forward to this today what else was i going to say oh yeah that to flow was awarded most innovative company of the year is that the way to say this that's the right way to say this in the australian financial review most innovative companies 2020 we were awarded a place yes and two flow was a uh was the content of our submission fantastic a fantastic award and it shows what's been happening for to flow bmt all right we had a poll wonderful should we look at that poll here we go thank everyone for sharing your your responses on that poll uh did we expect commercial consulting contracting to be as high as it is i guess we did yeah coastal modeling has got to be high up on planning as well as consulting academics any comments across these questions number two in experience i think they're pretty pretty interesting results around a good spread of the experience of people and that's that's great to see a good good level of mixture with the different models as well interesting that a lot of people have said you know none for the types of models that they've you know used coastal processing well in fact that's the highest category of nuns the highest occurring of what modules to use yeah this too just going back to that other question about how many years the highest one is five to ten so they're right in the middle of their careers that's great everybody thank you for that uh skip down number four how important ongoing sustainable very important here yeah that was definitely the biggest category [Music] and the last one's interesting as well in terms of what people want to investigate um the 2d currents and water levels being being a key one but a reasonably good good spread across a lot of the different categories there so it highlights the diversity of coastal modelling and coastline yeah sediment transports right behind 2d current water levels in what typically do you want to investigate tsunami's in there of course yeah yeah all right ever happy with that and i think we've done enough to get ourselves rip roaring ready got the coffee in hand and i can't wait to see how this is going mitchell where in your hands michael you're going to be hammering away with q a so right over to you mitchell uh so yeah coastal modeling 101 it's a really an introduction to some of the coastal modding that you can do and it's really interesting to see the poll question there about a lot of people that maybe haven't done any modeling before so this should be right up your alley this presentation so that's quite good today i'm going to talk about coastal modeling and compare it quite a bit to flood flood modeling we're going to go through a challenge where we build up a hydrodynamic model trying to solve a problem and as we move through building that model we'll work through what the astronomical tides are how the climate weather and the ocean uh can affect the the ocean water levels uh spectral wave modeling and then we'll wrap wrap things up now this presentation i'll be talking for about 40 minutes is really a taster to coastal modeling okay you're not going to go away today uh you know being able to run or build all these things straight away but what i really want you to take away from this is that you have a better understanding of the the scale and the size of the processes that that are involved in coastal modeling and how they compare to flood flood modeling so we'll be using uh two flare fee as a hydrodynamic model and swan as the spectral wave model today it's really aimed at new coastal modelers um flood modelers and those interested in understanding more about coastal modeling as well so managers and things so there's it's very much aimed at those target audience members i've really kept the theory to a minimum but i'd encourage you to go and have a look at this free coastal wiki on the astronomical tide and waves if you're interested in knowing knowing more because we could talk about that for for a long time to flow fee is the hydrodynamic model we're using today we'll be basically doing some 2d 2d modeling but there are lots of options to look at sediment transport and water quality a lot of the the pictures that i'll show you today are two flower fee results but we're using qgis and the to flow the free to flow viewer plug-in for qgis there's a few definitions i'll run through just to start with rivers and flood modeling okay a lot of us are used to these types of studies but we're really interested in how rainfall runs off a catchment and to a lesser extent how groundwater might affect the catchment as we move downstream we have estuaries and this is a region where rivers will start to mix with the ocean and on the right hand side this is myanmar and shout out to the myanmar people i saw there's a couple on the uh on the list of attendees um this is near yangon and we have a river system moving up this is salinity that's shown in the plot and this is a long section of salinity moving up up the river so we have salt water down near the coast this is a 3d long section and we have fresh water moving down to to meet that i've nominally called coastal the area seawater of the coastline for about 100 or 250 kilometers and this is where we're actively using the coastal zone beaches and ports and shipping um once we move further offshore into the ocean we're really talking about you know the deepwater ocean and where the global and offshore processes are important so just keep that in mind so how do flood models and coastal models affect each other or the processes that we're looking at on the left hand side i've got coastal processes that might affect how you set up a flood model now the most obvious one that comes to mind is that we're building a flood model down near the coast and we've got to assign some sort of tail water level to the downstream boundary now that might be in the form of the astronomical tide it might be a storm tide boundary or it might be some static water level that we need to to ascertain if we move further up into the catchment you know the ocean effects don't really play a role so they're not as important but certainly down near the coast they are the river mouth morphology so if there's a lot of sediment built up at the river mouth or in the lower parts of the river that can lead to issues or increases in flood upstream and that can be important and also the interaction of salt and fresh water so as per the myanmar picture i just showed you we can get these interactions that change the density of the estuary however if we're looking at big flood events the fresh water tends to move move down and and basically blast all of that out to sea so they're not so important now in terms of river processes that affect coastal modeling freshwater river flows can be important in the estuary area but flood flows don't really tend to affect the ocean at all really the volume of water out in the ocean uh compared to what's coming out of a river is you know very limited and any any increases in level are very localized to around the river mouth what rivers are important for uh delivery of coastal sediments okay in the sediment budget and also nutrients that can be important for you know fisheries and a whole range of other ecological processes offshore just a few applications of coastal modeling that we can look into for our projects on the left-hand side we mainly have you know water levels and current related activities in the middle we have port structures and sediment transport and dredging and on the right hand side it's more environmental considerations aquaculture ocean outfalls and to to an extent animal migration and salvage and rescue the big difference that we see when we start to model in the coast and the ocean is related to scale so in terms of temporal scale that is the time time frames that these things are occurring over in flooding we tend to have events that might happen for for hours or days you know and that's it's usually the type of time frame we're looking at as we move into the coasts we're looking at hours to years and as we move off into the deep ocean days to hundreds of years these these scales or these processes can occur over in terms of the distances that these things are happening over you know in the catchment we're worried about things that may be meters to kilometers and they're rarely bigger the coast coastal area meters to thousands of kilometers and ocean we're looking at global scales so we need to think bigger when we're moving out into the the ocean in terms of the processes that we need to to look at and include in our models rainfall is no longer the star of the show you know it's been out in the limelight there for flood modeling and once we move across into coastal space we're not really that interested in it anymore sorry rainfall but what is important though is the astronomical tides and the interaction between the atmosphere and the ocean coastal model boundary conditions we tend to use gridded outputs or gridded boundary conditions which you can see on the right hand side here which is gridded wind speed and pressure because of the space and time scales that are being operated on 3d effects often need to be considered so you can get you know the situation where the water's moving one direction at the surface and then it's moving the other direction at that depth but also density so the effect of salinity and temperature and how they move water around can also be important and rarely do we need to look at salinity and temperature in flood modeling and the last one ocean waves okay generated by wind and breaking on the shore is not something that we we typically look at for flood modelling all right so to work through these processes we've got this modeling challenge we've had a client come to us and say oh we've had this event occur at our tide gauge and so this black dash line represents an actual real water level measurement that occurred during a real storm and we're going to build a model to try and match that observed level and this is uranium from the 24th to the 30th of january 2013. now where is this well it's over on the east coast of australia on where that blue dot is and as we move in move into harvey bay area and we move right into where the urangan tide gauge is so in order to try and reproduce this water level we need to firstly develop a model domain then we're going to run that model with the astronomical tide only as boundary conditions then we're going to increase our inputs to the model and start to include meteorological influences primarily air pressure means sea level pressure and wind and as we build these models i'm going to walk you through the various data sets that we need to build the model but also the physical processes that we are trying to to represent in a model so step one develop the hydrodynamic model now you might be a bit surprised but really coastal and flood models you know they're not not that different really we still have to demarcate where the model boundary is we still need to make decisions about that we then have bathymetry okay or topography we then need to have a computational mesh to solve equations of momentum and mass conservation similar to a flood model we have boundary conditions okay in a flood model we're usually looking at at flow rates or water levels downstream in the ocean okay water level variations are important to capture and you know we rarely put flows in except maybe at a river location then we need to make some decisions around bed roughness models and turbulence and viscosity just like flood modelling and we also need data points to calibrate our models too and so in our study area we have the urangan tide gauge which is the one that we've just been looking at with that observed measurement and we've also got a location at bundaberg which is just slightly up the river that's important for something i'll discuss so this is a twofold fv mesh and what we do is you'll see you move with quadrilateral elements and triangular elements so it's a flexible mesh we move from quite low resolution offshore where the water's deep into finer resolution as we come come on shore we spend quite a bit of time demarcating certain parts of the mesh so that we can define the right sort of size mesh cells that we want for the water depth and the processes and as we move further ensure we increase the complexity as well so that's a very different process from flood modeling where you might just set a set water sorry a set grid resolution for the whole model we're really trying to upfront save a lot of time and that's because of the scale you know if you tried to model at 10 meters celsius all the way out to the other end of the ocean your computer would explode basically and we don't we just don't need to model at that resolution out in the deep water we need to apply bathymetry and in the same way we do to a flood model so the data comes from a different place there's soundings out in the ocean from admiralty charts it might be analog charts or digital online data services in estuaries there may be soundings that have been collected or sections that we can use and if we're moving out of the the tidal range and onto the land for example looking at storm tides or tsunamis then lidar can be important so for this study i've been very lucky in that there's some online data i've grabbed this jebco 2020 which is quite good in deep water and in the shallow water i've moved in and grabbed some of this elvis data this is all freely available online data again if you're in a different part of the world there may be some online data for you or you may need to source some up for yourself now i'm going to talk about putting the astronomical tide onto this model that we've just built but before i do that i want to walk you through what the tides are and why they're important so the astronomical tides they drive the water levels and the currents you know out in the ocean out in the tidal ranges of the ocean largely they are responsible for the intertidal areas so mangroves and mud flats and seagrass and they can be really important ecological areas so where we have very large tides you have a larger area of intertidal and you can have small tides where you have have less and that can affect things like how long the mud flats or the intertidal areas wet for if there's only one high tide or one low tide a day that can affect the types of ecology that grows there the tides can modulate the the wave attack on the beach the size of the waves and can be responsible for erosion and deposition and at the kind of pointy end of the stick when we're starting to try and protect our coastal infrastructure or design how high it needs to be we need to have a look at how high the tides get we also from a navigational perspective and that's what the dredge is doing right here is we need to understand when the tides goes out how much water is under a ship and is it going to be stuck there now most tidal so most flood modelers will have some experience with the astronomical tidal plane chart so you need to go in and have a have a look at things this might be where where the story ends for you essentially they're just apart from the h18 and lat they are just mean levels that have been derived from tidal observations and i'll go into this in a bit more detail so the astronomical tides are long period waves that are driven by the differences and that's important to note here the differences in gravitational attractive attraction across the planet as our rotating earth moves and that's mainly responsible because of the moon and the sun so the moon and the sun affect the gravitational forces on the earth the time of the year and the changes in the moon's orbit and earth's orbit the earth's axial tilt all affect the tides and so each of these different gravitational pulls the moon and the sun and the earth's rotation can be all broken down into a series of partial tides called tidal constituents these forces are influenced by the continents and the ocean bathymetry and the coriolis force and tides differ greatly across the world so this animation on on the right hand side here is the main moon constituent the m2 constituent the white colours are higher areas of water elevation and the black is lower and what we tend to see is that these waves set up as standing waves in the ocean basins and they rotate around nodes or amphidromic points and so there's a lot of different uh tide heights and tide directions across the world now if we look at look at the tides on the eastern side of australia and the western side of australia these are both tied predictions and you can see that there's some really different behavior going on first thing you might notice is that on the east coast in our study area actually we get a lot higher amplitude so the height of these tidal waves and they're also semi-diurnal that means every sort of six hours now when we go down to the bottom you see very different the amplitudes aren't as great and the actual period how often they occur is quite different as well so these numbers up here are these different components of gravity that i was talking about these different gravitational waves that move around the the oceans and so if we have a look at the the top uh picture here we can break that water level prediction down into a series of constituents and so we start with the moon okay and here's the effect of of the sun and when they're when they're in phase so full moon or new moon we get a resonance an amplification of the tide and when they oppose each other we get a reduction so you sort of spring and neap cycles as we add more of these gravitational features if you watch down the bottom what we're doing is we're adding each one of the constituents and you're starting to see something that resembles more of the tide that we observe at the coast so the ocean water level that you see offshore is really a combination or an addition of all of these different tidal constituents how do we know what these are though well we can put tide gauges in on the coast okay and they might be there for 20 30 years and when we take those water level measurements we can analyze them and we can pull out the different wave periods and wave amplitudes for these different constituents so that's why it's really important that we have these monitoring stations around the country if we go over to western australia you know it's quite a different story okay the actual m2 constituent doesn't do a lot and some of these diurnal like k10201 constituents do do more so that's a little bit about about the tide and how it works what i want to talk about now is how we might go about getting some of this information into a model now our our period that we're looking at was at the end of january okay this this time when our challenge is being investigated so what we do is we run the model for a month for january 2013. that allows the water to slosh around a bit in the model and warm up then we need to source and apply astronomical tide boundaries then we're going to calibrate it to the available points that we have in terms of tied data we've grabbed some gridded title predictions we have also got tools to do this if you're trying to do it yourself but there's global global title models that you can get this information from now why we need to to grab these gridded data sets is this picture here shows a long section of water level pretty close to my my boundary so i've got a boundary that moves along here and it's a sloping water surface boundary at one end of the boundary here we've got water levels of about 0.9 and at the other end we've got values of 0.5 so that water is sloping all the way along there okay and that's the way that the astronomical tides work if we change this from being in in the x direction so at the moment in meters along the line to time and we took at a point you can see over time for the whole of january that point there is moving around like the tide and if we took another one here it would be similar but slightly different now i'm going to show you some model results what we're showing on here is the water level from the tide and these are the tidal currents and this is the the blue is the model results and the red dashed line is the predicted tide at urangan so same as those predictions i showed you before from all the constituents so i'll play through that just step through and what you can see here is that as the tide comes in okay here the water level's going up the velocities are moving in towards the coast and then we get to the top of the tide and then we move move back offshore and we have our ebb tide and water moves away and what this allows us to do is we've got a pretty good match down here at uranium but if we wanted to look at tidal currents or tidal water levels anywhere away from that point we can be reasonably confident that they're looking looking okay this is the astronomical tide calibration at urangan and also bundaberg so the blue is the model the red is the predicted tide okay so just the gravitational forces and if we go to our our challenge we can see that the blue line which is the model tide is not really anywhere near the the black line which was the the measured water level so even though we've done a really good job of of modeling the tide okay in the top here we're still still missing something and so it's a bit of head scratching so going on not sure what what to do or what what it is um so so we need to have a bit of an understanding of what what could be causing this process i'm going to talk you through the climate the weather and the ocean and just to give you an understanding of the again the scale of the processes that we're looking at and some of the things that you might need to consider for a coastal modeling project now the ocean and the atmosphere are intimately coupled together okay they they need each other they drive each other and a lot of our climate and weather is driven by the differential heating of the earth so at the equator there's an excess of heat to reach equilibrium that heat moves away and moves towards the poles and essentially this redistribution of heat drives atmospheric circulation we have coriolis deflection in both hemispheres which leads to a tendency to have trade winds okay in the mid latitudes and in down in the sorry in the subtropical areas we have the trade winds in the mid latitudes we tend to have westerly westerly winds that occur now as all of this air is moving around it essentially pushes the ocean and can affect the ocean circulation and then we have storms that are embedded in these larger scale systems that occur as weather events so this animation here uh let's get rid of my here it just shows the the air temperature and the wind wind speed so this is from the knoll earth null school awesome go check it out this is just from a few days ago and you can see the wind speed at the surface moving around we have all these embedded storms and this is happening all the time um if we take that a bit further and look at the ocean okay so this is the temperature if we then look at the ocean we can see that let's give it a moment but there's also currents in the ocean so what's being shown on the screen right now is ocean currents and they are also redistributing heat mixing water and having eddies in some places those eddies lead to upwelling and down dwelling and differences in sea surface temperature and so a lot of these ocean circulation processes can increase or decrease weather events so just if you take anything out of these slides just be aware that you know there's a lot more happening on a ocean basin and synoptic scale than just your flood catchment that you might need to be aware of so how do we go about actually including some of this stuff you know in our models it's a bit overwhelming right well we're actually really lucky in 2021 in that we have a lot of data that we can we can use both from meteorological perspective but also from an ocean modeling perspective and so we have these global weather models that we can go and freely download data from i'm going to today grab some mean sea level pressure data and also wind speed if we're looking at a 3d assessment which we won't cover today we might need to get things like air temperature and radiation there's also models of the of the ocean and these can be important for the 3d structure of the ocean but also wind waves things like the easter strand current and so we need to sometimes grab solid and temperature data which i uncovered today as well so this is awesome you can go on to get a copernicus account and go and download some some wind data and pressure data and then you get it and then you can drop it straight into the two flow viewer here's our model domain we can add this in and so what i'm about to show you here is actually the gridded wind speed surface wind speed the colors are the mean sea level pressure and so we have that that data it extends far enough to cover our whole model domain and so we can apply that and if we look at a time series of wind speed which is blue and the pressure in red you see there's a bit of a bit of a weather event potentially occurring around the 25th of january and so maybe maybe that's what's going on with our water levels back to the model we've gone in and we've applied those wind speed and pressures into the model as a boundary condition and that's applied across the surface what i'm showing up the top here is the wind speed which is the white arrows contours of the water level values and down here we have the predicted tide which is red which i was showing you before nothing's changed there the blue is the model tied plus surge and the black is the same measurements from from our challenge and so if we play through this what we'll see is that the wind is pushing the water down into harvey bay and this is known as a storm surge so both the wind and the pressure is really pushing that water down and when you look at tropical cyclone modeling these sort of storm type events can be fairly important so back to our challenge how are we doing now so again the black is the dashed line the measurements the model with the wind and pressure in there is the blue and look i'd say you know that's that's pretty good that's pretty good you know it's a lot better than where where we were and so we can go in and we can write up our report if we're trying to match the peak level and happy happy days so what happened well there was a passage of a storm that created a storm tide at urangan and even though we had a really good astronomical tide calibration it wasn't enough to reproduce what was actually being measured at that location and so we did need to include wind and pressure for for this site in other places it might be important to include 3d ocean effects where you have different types of stratification but for this case it was okay just to have these inputs so because we did such an awesome job and we're all happy and everyone's happy the client has actually come back to us and said all right well i've actually had a bit of a doozy here and during the same storm there was a fishing boat out and it capsized and sank at this location and we need to have an investigation understand what the wave conditions were at the rec site so maybe we can you know try and avoid these types of things in future if we can forecast the waves and see what's happening and so we've been asked to reproduce a significant wave height and the peak wave period at this location now what i want you to do is take off your your two flower fee hydrodynamic hat so up until now we've been talking about water levels and currents and hydrodynamics to solve these questions around waves and wave propagation we need to move towards how to model wind wind waves and we do that via spectral wave modeling so wind waves are a complex combination of different waves in the ocean if you look at this picture here you know you you don't have one particular wave moving from one direction you have many different different waves that overlay on each other to give you this kind of chaotic looking picture and they have different directions heights and speeds they're generated in quite a complex process by shear and resonance by the by the wind but essentially the longer and the harder that wind blows over a given area of the ocean known as the fetch the larger the waves will grow until an equilibrium is is reached and there's two main types of wind driven waves in the ocean sea waves which are sort of locally locally generated waves and then we can get swell which can be generated by local storms but can also travel vast thousands of kilometers across across the ocean so why are ocean waves important well they they have to distribute energy across across the ocean so as i said you could have a storm that's down in antarctica the waves from that storm can end up in alaska and and those waves can have a tremendous amount of of energy such as port ferry recently where some waves threw a few rocks around and look this shouldn't be all that surprising to people these rocks aren't that large anyway they the waves are extremely powerful and can throw through much larger things around they are responsible for you know in a part to upper ocean mixing but where they're really important for us as engineers and scientists and managers in this space is around structural design okay am i going to have my structure destroyed by by waves at that site navigation is also important so ships might go you know off off their normal um direction just to go around large waves that are not nice um also for coastal structures to protect ports from from large waves they're important for sediment transport and coastal morphology so they drive a lot of that that process and also important for recreation surfers come come to mind there so in terms of estimating the wave conditions what we need to do is is build a spectral wave model and what this does is it it models the evolution of wave energy across the surface they look at wave dissipation so how waves might be affected by white capping or frictional effects wave generation by the wind interaction between different waves waves set up and there's a lot more that you can you can do with them we'll be investigating the coastal wave model swan here built by the delft university technology you can go and grab it from there if you if you want to and it has a lot of good local coastal process shallow wave processes in it there's also offshore wave models as well that can be used further offshore this is a bit of a plug for our newly freely available swan gis tools so these will be ported with the twofold qgis plugin they're not actually the swan model but they're tools to help you develop the swann model domain so now this is a two-fold fee model here the red grids are the swan domains that we're building so we have a large model we have a smaller one and we're nesting into smaller models here so anyone who's used swan before this can be quite a cumbersome process to do this so what the swan gis tools will do will help you to build your control files wave spectra uh nesting the symmetry and wave wind files so if you're interested in any of that please let me know and we can send you an example and some um some doco to try it out this is just the the tool in process to go the model builder brings up this process here panel sorry got the the model folder the configuration we the key thing here is that we can build our domains dynamically with gis polygons so we can draw those in and modify things like the rotation and the length of the cells and put our bathymetry in and the same wind data and wave data that we had before from the copernicus site we can drop drop into here so this will just show you quickly you know we can come in here we can change the rotation of the grid and see the rotation in the gridded box there we can reverse the rotation okay and we can change the the cell sizes i'll just skip forward a bit on this and we can also have you know nested nested grids as well so so i'll just show you quickly so you can have nesting and what the tool will do is it will actually generate all of the files that you need for that so so helps a lot in terms of actually solving the dilemma of this capsize ship what we need to know is what was the significant wave height at a particular point and the significant wave height is the average of the highest one-third of the waves that we see at that site a particular period of time and the wave period peak wave period and that's the time between successive wave crests for the waves that had the highest energy in the wave spectrum and these can be output directly by swan and there's lots of other other options what's quite nice is that we can also bring that into the twofold viewer plugin and so up the top here we've got again the wind speed during this event so this is that just before 25th of january 2013. we've got our site here that we're interested in and the colors are the significant wave height contours so so just the first thing that's interesting right is that you've got the winds coming across and this is quite shallow up here so there is some shoaling or breaking of waves in that area but you can see the sheltering effects of of the island okay as you move down here down down on the time series in blue we have the peak wave period which is on secondary axis in seconds and the black which is a significant wave height at this point and so if we play through this i'll sort of jump jump to different parts of the storm so as we move across we start to see the wind moving further from the north or the northeast and the way that that island is shading the more so harvey bay is different if we go further into the storm to the peak of the event you really see the wind starting to drive the waves down and and the actual wave height and the direction of attack uh changes as well in that the waves are coming from the north and they're affecting greater areas as well so in terms of solving the mystery of the boat and what the wave parameters were at that location we can pick them off the the time series here so we're talking about a four four minute significant wave height okay and around nine seconds which is at the lower end of the swell wave period spectrum so again very quick uh run through of some spectral wave modeling just to give you a taste there if you're interested in knowing what else you can do with them you can integrate waves and hydrodynamic models together and so you can apply those wave outputs from swan to two flow fe or another hydrodynamic model and they can affect the currents in the water levels waves can be really important for sediment transport modeling and you can also run a wave and hydrodynamic model together in two-way coupled bone if you're interested in seeing an application of that have a look at the melula bar structural design example from from this previous aws webinar so just to wrap wrap things up so really i know i say it but the flood and ocean models really aren't aren't too different okay we're modeling water moving around what is different though and very different is the scale and the processes that we need to include okay noting that we didn't even go to 3d today and so the water levels and the currents due to the astronomical tides they vary across the globe in in time and space and the currents and water levels at the coastline you know aren't just due to the astronomical tides because the ocean and the atmosphere are intricately linked to each other we also often have to include the effect of of the weather so often each year the water levels that you see at the coast will exceed the predicted astronomical tide due to these things we call tidal residuals which is contributions of ocean processes climate and weather on top of the astronomical time and ocean wind waves are also important to consider in coastal space that we don't consider in the flood world i guess finally models can help us to extend our knowledge away from the measurements okay so we had those those calibration points but because we built those models we can have some confidence that the water levels and the velocities away from those points are reasonably well represented they also allow us to model scenarios and designs as well calibration is really important obviously not always possible but where possible you should really try to calibrate your model to give yourself more confidence and there's some great models out there and there's some great data sets out there and ultimately it's how how we use them and i guess just the final closing remark again if you can take anything out of the presentation today it's i hope that you have a better understanding of some of the coastal processes that occur the scale and how we need to to be aware of what's happening you know just away from our river system if we if we're in the coast so thanks thanks very much for that again michael's going to answer all your questions um but um and i'd just like to say thanks again to the australian water school if you haven't checked it out go have a look at the website and all the previous webinars it's a really great great resource so so thanks very much great mitchell that was absolutely wonderful we enjoyed that i did hugely i mean to go from global to um to local like you have and yet not uh tie us down in vast amount of detail you've only got this 40 30 40 minutes and you did that with aplomb you did that so well thanks so much for that it is incredible i mean i'm a surfer from way back to watch those wave diagrams i just wish i had a handle on them when i'm on my board you know where's the next next next hit going to come from a good surge here beautiful no that was fantastic and michael hammering away in those questions i'm watching him going back and forth uh why don't you take us through the next few minutes and what you think we should do next so up to you where you went ahead to from here yeah yeah there's a couple of there's a common theme running through a few questions mitch i was wondering if you could comment on what a few people have asked how we might integrate a coastal model and a flood model to look at a combined storm surge and flooding impacts inland sort of questions how often we do that what sort of model configurations would we use to do so all those sorts of things have come up about that that linkage or interaction between the two some comments on experience or examples around that yeah often often in the the tidally affected areas of a river we we want to include some you know coastal processes so we can go go offshore a little bit and and use the title predictions offshore to really come come into the the coast um there's there's increasing um benefit i suppose to having combined modeling systems and i think that over time they will be more prevalent um it's there's still value in having specific models to answer specific problems you know like if you've got a flood flooding issue and you're running a flood model you know that's that's good often you don't need to consider what's happening on the coast or if you are you can run the coastal model separately and apply the boundary conditions down downstream um having said said that there's this these models i've showed you today there's no reason why we couldn't go up up the river have the floodplain incorporated have the structures all incorporated if if that was important for the study and have a fully uh integrated flooding and coastal model um where you need to be careful about that though is you know that always comes at a at a cost and so if we're trying to run a flood event um you know we we may be restricted in how fast the model can can run through such a large system as opposed to running them separately so i don't know if you had anything more to add to that yeah i think the message is that it can be done but everything comes at a cost and that's usually a computational cost tooflow fe for example certainly has all the structures and things that it needed for bridges and culverts and whatever to simulate those inland flooding processes but it comes back to me to what are we using the models for what questions are we trying to answer and therefore how do we need to configure models do we really need to look at detailed coastal processes and inland flooding processes at the same time or can we break it up and those sorts of questions are really beginning of study questions that need to be addressed it's there's another question here which is uh is the swan model here was it in stationary mode or in stationary mode yeah it was is in stationary dynamic um yeah changing uh meteorological conditions so the wind and the was was changing over over time yeah another question similarly was hs and tp are sensitive to return period how long a return period is specified for this challenge or what's this challenge about um i guess the concept of having a return period assigned to to this is it was a one one event okay if we needed to try and assign a return period so term being the you know the average likely period that it might occur or be exceeded by we typically run a very long long set of simulations over a long period of time to look at that so so i wouldn't say that the hs and tps relate to their the return period i say that that we have a storm that's occurred it's it's got a certain wind speed it's caused water levels and sorry it's got caused wave heights and periods and they can be basically referenced relative to how often that event might happen at that location there's another question here i've seen during the presentation that era 5 were used to force the model wind and atmospheric pressure how confident are you with using era 5 in terms of its accuracy look i think what wasn't done today just through time is that you always want to try and ground truth this stuff so if you've got measurements of wind speed and mean sea level pressure on the ground at your site you can compare the global model data to what you have um so that's that's what i recommend that you do era 5 seems to be great there are for this particular site it might not be so good for other sites and so it's good to look at a range of data and to ground truth that data another one any comments on approaches and modelling limitations relating to breaking waves and breakwater design in terms of energy dissipation yeah i think if you're getting right down and dirty into how the waves may affect a structure with the spectral wave modeling i've showed you today you can get keyway parameters like wave height and wave period um and you can put those into coastal engineering guidelines to pull out you know likely stresses but if there's a specific structure that you're looking at you might need to go to like a cfd type model and run the waves waves through that so um hopefully that helps with with that question yep now is uh there's a question here which i can probably answer how can we do wind and storm surge in one model software for the arabian sea along the pakistani coastal area we've actually done that in two flow fe and we have a model of the arabian arabian sea and the gulf of oman and we look at those those things right now we're doing some work for the abu dhabi environment agency on that but certainly the model extends out there and can be used for those sorts of processes that mitch has been talking about today along the similar lines that mitch has has outlined [Music] mitch a good question for you a coastal process is different on the east and west coast yeah yeah they certainly certainly are different um one of the contributors to to that different i guess we're talking about australia specific here is the difference in the the astronomical tides you know we have a lower i want to say lower tide down in down in southern parts of western australia have very low tides but if you go to the kimberley region you have these these massive massive tides um the exposure of the western australian coastline uh to the indian ocean southern ocean is different so you get different different types of waves that approach the coast on the east coast um you know down around brisbane um we have big sandy beaches and uh yeah waves tend to drive a lot of that we have um as we move north they've got the great barrier reef and um they tend to dissipate a lot of wave energy you get mud flats and things like that so fundamentally coastal processes are very different uh in a lot of different different places and um yeah the vegetation the wildlife the sediment types that you get there will be reflective of those coastal yep um an interesting one here mitch how many tight constituents should we use to produce a reliable type prediction is that there is depending on where you are in the world yeah exactly so some some places can be incredibly complex and you might need you know over 100 tidal constituents to accurately model model the tide because there's a lot of interactions happening as the tidal waves move across shallow water we we based ours on on eight there's not really any any answer what you want to do most likely is to go and grab your water level recordings from a site and you can understand what the different constituents are for that site and that should show you reasonably quickly how important the different constituents can be yep okay one that's that's related to the first question that we talked about which was the interrelationship between catchment or land-based processes around flooding and coastal processes a really good question here um do you have any recommendations or experience in modeling entrance morphodynamics for coastal like summer games storm events would you need to integrate models together eg ex beach and healthcare yeah look the only experience that i i have is looking at uh twofold fees sediment transport module morphological integration with with two flow and in that case you you know run the hydrodynamics and it scours out the the bed and sediment transport can can move um i haven't done i've personally done a lot of like long long shore long term sediment transport but i can see that there are you know advantages of of doing doing that so but they are quite um i guess unless you're doing longer term processes you might be quite separate exercises that you do yeah i think the short term events can be handled in a model like two flow fe with morphological dynamic bed changes in time um and we have icos in australia intermediately closing and opening lakes and lagoons and they're all about that sort of process and that's an example where i think you really do need the catchment modeling to play a role because often water levels will build up in these lakes inland and then drive the process of erosion once the initial breakage occurs of the of the sand gene or whatever it is that's close in the mouth so that is an example i think of overlap but it's typically short-term events we look at in that sort of modelling um and one of those reasons is again because there's a large computational overhead associated with modelling morphological processes over extended periods take a long time to do that we would therefore look at longer term models like ex-speech or similar empirical models hey what do you reckon we leave at that we're up up to nearly the hour mark what do you think everyone don't you glean so much from hearing a couple of professionals toss ideas back and forth i do it's just a fantastic uh it's an incredible medium to play with and we're talking to people from 50 countries all at once you know just with the with the um graphics and the and the audio all happening together thanks so much mitchell michael um we do hugely appreciate uh you you've taken the time to do this today um and thank you everyone for participating in the discussion like you have um it's been a fantastic hour i i'm i've never i never cease to be anywhere i won't run on anymore that's enough of that it's been great uh recording link we're going to email it to you uh we'll upload it to our website and youtube channel watch it for freely with your colleagues after for as long as you want it'll be there's hundreds of webinars up there now and there'll be a short one minute survey popping up at the web at this webinar conclusion in just a minute or so it'll come up automatically and if you don't get to see it hit hit put your phone on that qr code down there and you'll be able to go straight to the strategy survey i think that's it any final comments mitchell michael oh good for me trevor thanks very much and thanks everyone for attending it's great to see you yeah that's a great pleasure and the the comments coming through on the chat line right now uh highly appreciative of the time today it's been fantastic look at those uh on-demand courses webinars and live courses all coming up there stay on the on the website uh and get your friends to come to the website and you'll see what's going on it's huge well done thank you everybody thank you joel uh and australian water school it's been terrific see you mitchell see you michael thanks everyone i'll catch you next time bye for now everyone thanks for watching subscribe by clicking the link below and click on the notification bell to stay up to date with new releases for the latest in significant innovative and critical advances in water science technology and management subscribe now to build your skills enhance your technical knowledge and learn from leading experts in water visit the australianwaterschool.com.au and discover our online training courses both live and on demand

Channel: Australian Water School

Views: 860

Rating: 5 out of 5

Keywords:

Id: dMNsKKJchbQ

Length: 58min 18sec (3498 seconds)

Published: Thu May 20 2021