#76 Modelling with HEC-HMS

Video Statistics and Information

Video
Captions Word Cloud
Reddit Comments
Captions
subscribe to this youtube channel by clicking the subscription button below welcome to this australian water school webinar we're so glad you could join us today to discuss modeling with heck hms presented by thomas brower michael bartles craig price stephen joins this is the hydrologic modeling system designed to simulate complete hydrologic processes well let's get on with this my name's trevor filler i'm the national partnership manager here at ice warm and the webinar chair but it's been really good to see the response of this webinar today you can see on the screens there right today four presenters uh thomas brower michael bartles craig price and stephen joynes we're so glad you could join us gentlemen um you can see a fantastic amount of experience here welcome to you all i want to ask you just one question each before we got this off what drives you to this what motivates you to um to get right into this hms maybe we'll start with um ukraine cry well it may be no surprise that i'm more of a hydraulics guy than a hydrology guy but uh steve and i have been uh team teaching a hms and hec raz course once a year in new zealand um which has been a face-to-face course and now we're venturing out doing this on our own uh with with ice warm now as a partner to do it through the online system so we're pretty excited about that um it's been very good to be able to do this um with uh you know and and to get involved in this but really frankly um i won't uh try to pull the wool over anybody's eyes uh frankly i i'm i use uh hms for boundary conditions so that's what i use it for so um i'll pass along to everybody else i need the uh hydrographs to put into my hecaras models um and uh but i am excited uh because um to we we use those grass models um as rainfall runoff models as well and i am really excited to find out what's new in the upcoming version 4.4 so i don't want to take much more time than that pass it off to the other guys you sound like you're you should have the background that i've got in my screen but you're like you're not in the us yes i was just i was just there actually in san francisco a couple months ago actually at the hec offices and um you know i've been in australia about 12 years now and um but i do get back to the us occasionally that's uh i went to berkeley for my uh schooling so just across the bay i think you might actually see it on the left there yes not far from here yeah right yeah yeah steve yourself from the beautiful new zealand yeah hi everybody this is steve jones here been teaching akhms for at least 10 years now i've been using software for 20 years at least out of necessity but over time it's really interesting to use it's so simple and to be honest i think i practically use it every single day of my working life so i'd like to think and know it inside out but you never know everything about software so it's going to be an interesting thing for me but um always happy to answer questions as you need it thank you very much for joining us today and uh craig and steve will be on the uh q a chat lines right through this whole hour so do write your questions they'll be answering them they'll be talking with you about that right up to you michael what what brings you to heck hms and this whole area of hydrologic modelling so um i really enjoy the big problems right so i enjoy working on the models that are hundreds of thousands of square miles or should i say square kilometers we don't see what the mattress is very often here in the us but i like the very very large problems um it's always been attracted to them and that is mostly in the realm of hydrologic modeling though i do dabble in hydraulic modeling quite a bit as well yeah yeah i guess for me i once again like mike said i was always attracted to the big problems and the army corps is usually doing the big problems and you come to a place like hec and you're doing the hardest of the big problems and so that's all always been fun for me and i'll say the venn diagram between computer science and hydrology is kind of the world i operate in and that's a really fun world for me it was a combination of my interests yep well uh that's this is a good start we'll get we'll get stuck into this now the way we're going to do this is crow's going to pick up going to kick off with a couple of interest slides and and just talk about the overall picture then we're going to hand over to tom uh and michael and then we'll have time for q a c will also be on board actively working with the q a line but my background today is really dedicated to you two gentlemen because you're not far from here i understand you actually live quite close to this that's true uh we work out of the hydrologic engineering center davis california so we're about two hours east of where this picture is taken one of my favorite places in the world actually is where you're supposedly standing finally no adelaide south australia not quite the same but pretty good place to live though i can tell you okay no more of this let's get stuck into this uh right over did you crave missed anything let me know otherwise uh you you lead us away craig excellent um first thing is always an i.t check can you see my screen with the uh uh presenters fantastic excellent okay and i want to keep my remarks as short as we possibly can because we want to dive right into tom and michael's uh comments um what i wanted to do though is just uh kind of introduce where we're at with the hydrologic engineering center we have had probably six or eight of these webinars um on uh hec ras so on the hydrologic engineering center the hec and the purist will say hec instead of heck um we all uh at least on this webinar there's a lot of people who have sat in on hecraz um discussions um we've had over fifty thousand views of our uh hecarim's webinars as of tomorrow when this one is posted we'll probably have our first hms view on youtube but i wanted to just um set the stage for what we're looking at here um hec-ras came from uh the hec 2-pack it's a fortran package which computed water surface profiles and some of you may be my age or older and remember that but there's also heck three heck four heck five heck six um if you want to learn a little more about those um we can i'll show you a link here in just a minute but um what was heck one what was the predecessor to this heck rad we can answer chris question i see where craig was going with this so the predecessor to hms was actually hec1 so if anybody's used hec1 then hec hms a lot of it will make sense to them and if not we'll clarify what hms is to the world today so as was introduced earlier my name is tom my name is mike bartles and we are hydraulic engineers from the hydrologic engineering center as we said in davis california so hec is a part of the army corps of engineers within the united states underneath the department of defense so today thank you all first and foremost for joining on certainly appreciate the opportunity to talk with you about htc hms and without further ado let's get into it so here's an overview of our presentation today we're going to talk about hms a little bit of an overview and its various capabilities we're also going to talk quickly about the differences between hydrologic and hydraulic modeling since that was brought up earlier then through demonstration we're going to show how an hms model can be created parameterized calibrated and calibrated using observed data quickly and then finally we're going to close with a quick discussion of enhancements that we plan on including within hms in the future and then so do some q a afterwards okay first let's start with hydrology so hydrology is the study of water or more formally the scientific study of the movement distribution and quality of water we have a nice picture here that i think captures hydrology well we see precipitation falling from some clouds we see a land surface that has a potential for infiltration we see some vegetation with the capacity to transpire water and we see a lake a river system here that it probably flows downstream to the ocean at some point so all of this is captured in this picture and this is hydrology now let's take the same thing we just looked at in an image form and let's look at it in more of a chart so we in this chart we see some hydrologic processes well all of these processes are what we're modeling in hms so we think see things like precipitation we have methods to model precipitation we see surface runoff this is the major thing we're capturing in hms we have lost rates that subtract from the precipitation excess we have things like lakes and reservoirs and hms rivers even things like evapotranspiration and evaporation so a few bullet points on hms it's applicable for most regions of the world it's the way it's designed it can be it can and has been applied all around the world it's both we have and so for each process i guess one point of distinction that i often come to is people sometimes will say oh hms and call it a model and the point of distinction i like to make is that hms is actually a program that you do hydrologic modeling in so with each hydrologic process we have multiple approaches for modeling that process and so the end model the the result model in hms is actually it's nearly always a custom model that the user has built so we'll talk about things like we have empirical methods and physically based methods for for each hydrologic process people will talk when they talk modeling they tell is it event or continuous well hms can do either depending on how which modeling methods you select and how you parameterize it and then one another that comes up is it gridded or lumped those terms fly around in the modeling world and hms can do either of those we have gridded methods we have lumped parameter i would sometimes you can you could call that uh semi-distributed so hms we have approaches for doing both of those some of the strong suits of hms is data handling so temporal aggregation and disaggregation hms handles all that automatically so if you have daily precipitation data and run your model at a six hour time step hms handles that another thing that hms does well is handling unit systems so in the states we often work in u.s customary units around the world it's often metric and so hms handles the conversions between those unit systems without automatically without any input from the user uh and so another beneficial thing about hms is we have some compute types that are oriented toward hydrologic studies so the most common of which is a simulation run we also have an optimization trial where you're able to optimize parameters we have a forecast alternative that's set up for forecasting we do depth area analyses which are typically paired with a frequency storm for designing hydrologic and hydraulic structures and then we do an uncertainty analysis where we can start to consider things like parameter uncertainty so to have a quick discussion about the differences between hydrologic and hydraulic modeling um so in the past htc one hec2 world uh there were very well defined boundaries between hydrologic models now within the hms and htc raspberry world that we live in today those boundaries have started to become a little bit more blurred for instance hydraulic models have started to allow for the simulation of rainfall on the overland plain as you guys have seen in various other presentations also hydrologic models have started to become capable of including hydraulic routing but before choosing a hydrologic model or a hydraulic model for use in study it's important to take into account the considerations of the study itself so that includes things like run times so for instance hydrologic models tend to require less time to complete a simulation due to their inherent simplifications whereas hydraulic models take a bit more time because they're inherently more complicated the nature of the system you should also take that into account for instance very very flat channel slopes so in the us we usually apply a rule of thumb about two feet per mile anything less than that hydraulic routing tends to be the best choice that's usually the best choice anyways but hydrologic routing really starts to fall apart and doesn't do a great job of capturing true attenuation and translation effects of flood waves through systems so if you're dealing with systems like that you might want to be looking more towards the hydraulic modeling realm and also level of detail and desired outputs so if you're interested in inundation extents or arrival times hydraulic models are going to be your best bets to determine those those outputs as a quick example here's an example of hydrologic and hydraulic models being used in tandem with one another so this is a screen shot of the upper susquehanna river watershed in new york and pennsylvania in the u.s within the core water management system that we use within the army corps of engineers to manage our water management projects like i said manage a whole bunch of times in there but either way so in this watershed we have two usace owned and operated projects that are in working in tandem to reduce flooding risk to populated areas so we run hydrologic simulations on a daily basis to forecast inflow hydrographs to both these dams as well as hydrographs for downstream areas this is done within our hydrologic model due to time constraints desired level of accuracy and desired outputs that we need but hydraulic simulations are computed only when necessary usually when stages approach or exceed some defined flood threshold downstream and populated areas this is done because the hydrologic or hydraulic simulations take longer to complete and inundation extents they're not always required during normal operations so we hand output from one to the other using our data storage system hc dss from one model to the other right all right at this point we're going to leave the powerpoint presentation we're actually going to do a little bit of demonstration so what i'm going to do is i'm going to open htc hms now i'm going to begin a brand new project so i want to place this project in this location and i want to call it punxsutawney so if anybody has ever seen the awesome movie groundhog day you'll understand where this place is if you don't definitely look it up it's a great movie to watch so now that i have a brand new project within hcc hms i want to create and import a terrain data so i'm going to give it a name let's call it nad 10 meter nad stands for national elevation data set and 10 meters is the horizontal resolution so this data actually covers the entirety of the continental united states i want to get this data set so it's a geotiff raster now that i've imported that i can look and see that i have this terrain file that's been imported and i can now create a basic model so i'm going to call this base and model punks just a shorthand name [Music] and i want to associate the previously adjusted terrain data i'm going to select it i'm going to say skip because i just want to transfer the coordinate system from the terrain data now that i've done that you can see my terrain data in the background and i can begin doing all sorts of things one of those things being i want to pre-process syncs so what this is doing is creating a hydrologically corrected dem so one that doesn't include artificial pits you'll see that i've created two new layers one is a location of all the sinks within the watershed that have been filled and also a resultant dem that has been created over that next step i want to pre-process the drainage what's happening in this process as i'm computing flow directions and flow accumulations so if you're used to how geo hms and or archive used to work you know that those are both created as part of that process as well so let me turn off the flow accumulation raster and show you what the flow direction raster looks like so this is just computing which way water were to flow and the flow accumulation takes into account everything upstream of that to show you the drainage network so you can start to see dendritic channels and where water would actually flow if it were flowing on the overland plain so i'm going to turn that one off i'm going to go back to my base terrain and now i want to identify streams so this is asking for a threshold to say this is a stream or not and i'm going to input 70 square kilometers and now it's going to determine where enough drainage area is accumulated that exceeds 70 square kilometers and i'm going to call that a stream and now i want to add a break point and i'm going to go to a location where a stream gage exists so at this location the united states geologic survey is capturing essentially stream flow we'll just call it that at every single time step so 15 minute time step in this case or hourly depending upon your specific location now that i have a breakpoint identified i can say delineate me these elements and in very very short order i will have created a basin model that contains sub basins and routing reaches so all these guys are sub basins these are your catchments where the precipitation runoff process is modeled and these are streamflow routing reaches where channel routing processes are simulated now that i have this stuff i can start to get specific for instance this tributary right here is actually called stump creek so i'm going to rename that sub basin i'm going to call it steep creek this one is also called east branch mahoney creek so i'm going to simplify that i'm going to call it east branch mahoning creek and i also want to merge these sides together so i just want to make one single sub basin here so i select those two i'm going to merge those elements i'm going to select those two i want to merge those elements and now i can rename this guy this is actually my homing creek [Applause] and i want to make another junction right here and i'm going to call this confluence and i want to connect these sub basins to that junction so i can rename or re-select their downstream linkage and now that they're selected i can also select downstream here for reach to and finally i want to select these two routing reach elements and i'm going to merge them together i should say one more thing i want to rename this guy and i want to call that sync gauge and one more thing i lied again i'm going to call this money creek and i'm going to give it an r1 designation to mean reach so in one fell swoop well one a couple swoops i have created a complete basin model um discretized everything that goes on in here one last step i want to create a grid region and i want to call this shg which stands for standard hydrologic grid which is the option here you'll see what happens with this one in the next step i'm going to click finish and i'm going to set that as the default grid region all right okay get over tommy all right so now that mike's created a basin model i'm going to prepare a meteorologic boundary condition so we can dump some precip on this basin and we're also going to apply some temperatures so we're going to do a short continuous simulation to import gridded precipitation data we have this vortex import utility so this is a standalone set of small utilities it's kind of a test bed for features that will probably incorporate into hms in the future but for now we have this little wizard and you'll see it's pretty easy to use i'm going to select these grid files the metdata importer all it does is it reads common gridded data formats one of which is grid netcdf is another hdf5 is another you might be familiar with so we'll just select all of these files and now the program is just going to loop over the files and take a look inside and see what variable is in there so right now i'm importing mrms data it's a pretty good data set for north america if you're not in north america you might look out to some other data sources see what what is available in your neck of the woods another one that you might consider using is nasa's gpm data so i'm going to select that variable and now i have the option to clip these grids to my basin and so right here i just have a shapefile that is my base and extent i'm also going to set the projection uh the target projection for the grids i want these to be an shg if you're working in hms i would strongly suggest using utm or shg if you're working in states but you can use any projection you want i'm just suggesting these and then i'm going to select a target cell size so that's going to be i'm going to use a cell size of 2 000. and now we just need to say where do i want to write these grids we're going to write these to hec dss which is the data storage format that all hcc tools use so i have an empty dss file i'm just gonna tell the grids to right there so these grids are gonna it's quite a bit of data to churn through so each grid it's going to go through read the data it's going to clip it reproject it resample it and then write it so this will take a minute i've already created some grids so we'll just move forward with the grids that i've created and then we'll come back around at the end and take a look at some of the data so what i'm going to do now is add gridded data so i'm going to add a precipitation grid set i'm going to name this qpe because that's the data we're using now i'm going to link the dss file that's associated with this so i've linked the dss file now i just need to link a dss path you can select any but i usually go with the first so i've linked the precip data now i'm going to create a temperature grid i'm going to name this rtma that's the temperature data that i'm using now i'm going to link the dss file and i'm going to select a record okay now that i've added my data i'm going to create a meteorologic model so to do that i do components met model manager and create a new met model i'm going to name this september 2018. now on this model i have the opportunity to select the different methods that i want to use so one thing i'm going to do is i'm going to use gridded preset another thing i'm going to do is use grid hammond evapotranspiration now on this gridded precipitation method i can select my precipitation grids and in this evapotranspiration method i can use rtma temperature grids another thing that i'm going to do with this met model is link it to the basin model that i'm using so i'm just going to tell it to use the subbasins from the punxsutawney basin model okay i think i have one last thing and that's the link a gauge here so i'm going to bring in streamflow data that's observed and to do that i'm going to create a time series gage this is a discharge gauge and i'm going to name this usgs funks for punxsutawney so i've added a time series gauge now i'm going to link this to the observe flow record in dss which looks something like that now i just want to link this sync i want to set this observed gauge so now when we run a simulation we'll have observed flow to compare to okay so back over to vortex we see that our import has completed so i'll just close the wizard and now let's take a look inside of dss so here's all our records imported into dss and if i go to the 9th of september was the most active period so let's just select a handful of records and then we can graph them and we start to see that it looks like we have for this hour we have point four millimeters in that particular cell and i'll just step through a couple so you can see what this looks like so this is we've created a precipitation boundary condition for our basin so that's what the data looks like inside of gss right and now i'm going to quickly parameterize this model so first things first i want to set my processes for each sub basin and routing reach element i want to use deficit and constant losses i want to give parameter values to each sub basin of 50 millimeters of initial deficit maximum storage of 200 millimeters and a constant loss rate of 2.5 millimeters per hour i also want to set the canopy method to use the simple canopy method and parameterize that to use an initial storage of 0.1 percent or zero percent and a max storage of 0.1 millimeters so it's essentially nothing and simple uptake so what this will do is allow me to actually extract water from the soil for transform i want to set this to use the modified clark method and give that parameter 6 7 8 9 10 values for each individual sub basin so time of concentration and watershed storage coefficient for base flow i want to use the linear reservoir method so this guarantees conservation of mass and i want to use two layers for each sub basin set the discharge the initial type to discharge per area initially have zero coming out of the first layer and .015 cubic meters per second per square kilometer out of the second layer 50 of infiltrated water should go to the groundwater one layer and use a coefficient a routing coefficient of 20 hours 50 goes to groundwater 2 with a storage coefficient of 200 hours that's all i've got to do for that one more thing i want to set my routing reach to use the musking and routing method and i want to have a k value of 0.05 and a mustang x of 0.25 and a number of subregions of 5. now i want to create a control specifications for the simulation at hand i'm going to call that september 2018. now i specify that i want to start on the 1st of september 2018 zero zero zero zero and i want to end on the 28th of september 2018 zero zero zero as well and a time step of one hour i'm going to set that to be a computation point and now i can run i can create a simulation run i'm going to call this september 2018 i need to select a base and model i only have one i need to select a met model i only have one and i need to select a control specification i only have one so those choices are easy and now i can make a simulation run first time i run since i'm using the modified clark method it needs to compute time area histograms essentially and travel time indices for each individual sub basin so that's what this is doing this will only happen one time so it needs to do this one time and then never again unless i make modifications to the basin model itself you'll see subsequent simulations go much much much faster so now that simulation is complete i'll kick it back over to tommy okay so we've ran our first simulation and i think we're about 10 15 minutes into the demo so that's pretty quick let's take a look at some of the results so one of the opportunities you have to view results is straight out of this basin map so i just right clicked on the sync element we have here and i have the opportunity to view graphical results so here we see our graphical results not looking bad for our first run another type of result we have is a table so i can go in here and i can view a summary table and i see things like a peak discharge a volume i can change the units on this i also see things like an ash cliff efficiency so 0.95 hey we can just stop now that's pretty good uh percent bias so that gives an indication of volume so we have some performance metrics and these standard summary tables another way i can access the results is from the results tab so i've just switched from the components tab which is an outline of the watershed and now i have results here on this tab so i can just start to click through element nodes and view different types of results so here's just the results for east branch mahoney creek and if i scroll down we can go back to our gauge and access the same results that we just saw from the basin map let's keep this here for a second and do a brief run it calibration so one of the things i might change here you can see um my peak is a little low so one thing i might do is go in what i'm going to do is go back to components tab select the basin model node and now i'm going to globally edit my loss parameters so one of the things i can do here is i can start to change my loss rates so let's go with something sorry that's initial deficit [Music] here we go this is what i'm looking for so i'm going to bump these to 2.75 that's the wrong way 2.25 okay and then i can go ahead and compute from this menu see what my graph looks like okay now my peak's looking a little better another thing i might want to do is bring my base flow focus on the recession limb and edit my baseline parameters so i'll just globally edit those as well i'll try groundwater 1 coefficient of 30 here and a groundwater 2 let's try 250. okay let's compute that and let's take a look okay so now my recession ones look a little better but i'm missing on peak flow so this is calibration we're going through we're iterating we're calibrating the model so one last thing i'll do is these are these watersheds have some urban area so i'll go ahead and add a little bit impervious area in them and that'll likely bump up our peak okay so we've gotten a little closer on our peak and this just demonstrates the iterative process of calibration how you might go about it i would probably work on this some more but for the sake of time we'll keep moving so let's jump back into the slide presentation great thank you tommy so hopefully you guys saw how fast you could build a model parameterize it input boundary conditions simulate and calibrate you know you can always keep working on calibration until the end of time right but you can see how fast you can do all that stuff with an hms hopefully you guys get a lot of benefit out of that and can get a whole lot of bang for your buck and using this uh software application in your day-to-day life and your job however um now we're going to talk about some future enhancements that are planned for inclusion within hms within future versions so first and foremost everybody's favorite uh we want to include the 2d flow overland flow solver specifically from htc ras for use within highly detailed studies where unit hydrograph transform isn't going to work for you that's planned for future inclusion we also have simple reservoir operations that are planned for inclusion within future releases of hms so for instance i might want to simulate gated operations at a flood control reservoir so in this simulation here i'm showing results uh the natural flow and the dashed redline that is coming into the project and what i'm actually putting out of the project is in blue that's realized through gated operations throw so some logic with an hms would actually allow us to compute that blue hydrograph given the inflow in red to minimize damages downstream so another thing we're planning to work on is parameter estimation so we recently added a lot of these gis features and the next step is to do things like import a soils database and start to analyze those databases and make it easier for the modelers to quickly parameterize their models so this is just an example where someone's imported a soil database we've done some processing on it to come up with a notion of what the constant rate parameters loss parameters might be another area for future work is met data processing so i showed you vortex and how we built a meteorologic boundary condition based on gridded data and we plan to continue working with vortex and building out that functionality like i said a lot of it will potentially be moved into hms as it matures and develops the biggest thing we need right now is people to test it with their data sets let us know how it goes this is actually an open source project and it's available on the hydrologic engineering center's github page so if you want to go there you can download it test it let me know how it goes if you're inclined you can even go help develop on it finally we want to make improvements to the existing optimization uncertainty analysis within hc hms so within the current release you have access to the markov chain monte carlo optimization trial and that can be linked to an uncertainty analysis to investigate the uncertainty as a flow or stage at a specific location solely due to parameter uncertainty specifically we want to make further reductions in runtimes to make things faster since we have the wrong potentially millions of simulations to do this we also want to add new objective functions to evaluate goodness of fit we also want to improve the control of output variables to minimize disk space that we need to take up so that i'll say thank you for your time for more information you can visit the hydrologic engineering center's web page we have the hms page that's a good place to start from there you can find downloads and documentation and then i want to just point at some of the training materials we publish so every time we put on a class we'll publish some of the training materials and you might find that helpful to get started on your projects so thank you everybody for spending the time again truly truly appreciate the opportunity to present hsc hms to you all thanks tom thanks michael it's it's been a total feast um simulating hydrologic processes it's been fantastic we can see the enormous amount of experience behind it thank you everyone for your questions and you're i i haven't broken into the the the um the presentation simply because it was just so gripping and engaging and it was good to get the whole picture uh craig can i hand back to you and sorry but that went it went dead what chances are there that that are an outage should happen right in the two minutes when you're doing your presence i know exactly exactly can you hear me okay now though very clear very okay good um it's it's back on yeah somebody at the street cut the power right in the middle of my talk so that that's fine um i saw you coming right with something yes yes so um hopefully uh um uh joel will work his magic on trimming up the video for uh for youtube viewers but yeah that i i did get to get to see most of that um and very interesting i've been trying to answer as many questions as we can there are a few actually still on there that um uh steve and i are still getting to so tom and michael feel free to uh pop in there and hit the q a as well while we're in the session live you could also have a look at some of the text ones we'll we'll try to address a few of those and um and i think a lot of the questions that were coming in um as the questions came in you guys covered that um as far as the the future work so so all right we can probably dive right into those let's see i'll the way we'll do this i'll read them out because the recording doesn't actually see the questions if we pick the questions that have been upvoted i'll read them out and we can get stuck into them before i start um you can see how many people are on this webinar across the world it's going to be a fairly full class there's only 30 so i'm not doing a big sell here but i probably am but um but if you want to get in on this course and i can only say start registering as soon as the webinar finishes gentlemen do it i think uh joel yeah thank you for that uh joel's put a registration up on the chat line there you'll see everybody uh steve steven craig will be running that course but with tom and michael's backing has been fantastic let's get going uh first question from professor francisco in mexico is there any perspective to the heck hms to be an open source in future and is there any risk that it won't be available for free in the future good question okay so this is uh i guess i can speak to that i don't see hms being open source in the new near future i do seeing it being free uh for for the foreseeable future so one of the most difficult things working here is yeah we're tied in with dod there's some restrictions there but there's also just our capacity at hcc a lot of people think we're this big software development firm of 300 or 1000 but it's really the hms team is five to six people i would say and so open source is great and and little projects like vortex are open source so we we're open sourcing bits and pieces but the ability to manage a full open source project we would truly need more bodies to appropriately manage that but yeah for the foreseeable future hms is free it's been kind of one of the founding principles is to distribute the free software so yep yep that's right you crack a lot of uh you you you hit well above your weight without only half a dozen people everybody everybody's very keen about it uh right should we go on the next question uh tahira is asked what's the difference difference between heck what heck hms those two and applications of heck what versus hms who wants to take that one on so hcc watt thank you this is an awesome question and this is very very cool my buddy will will definitely love this who's the team lead of hcc watt htc watt stands for the watershed analysis tool so this is actually one of our integration pieces of software so you saw earlier a slide where i included a screen capture from our core water management system swims this is very similar to that but for a planning mode so htc watt allows for the integration of hcc hms hsc resim agency raz and hccfia to do the full analysis suite of hydrologic simulation reservoir analysis simulation hydraulic simulation and consequences estimation as well all in one fell swoop so htc watt allows you to do not just deterministic computes where you know exactly what everything is but also stochastic computes so we can sample from the uncertainty space for parameter inputs for boundary conditions and also fully analyze uncertainty uh all the way through to consequences so that is a very very cool question hsc hms is a component model of hcc what or can be a component model of hc1 awesome question uh thanks a lot and thanks for that question here uh where do we go next crowd with these questions what do you think uh joel please question yeah well one one question that seems to have come up in a couple of different forms and uh this this is one that i think you guys will have a standard answer for when is uh 4.4 coming out uh and and i know uh you know on on the hec razz side gary always tells us not to not to even ask and even if he answers don't quote him on it so um yeah when might 4.4 be expected yeah so if you want to get your hands on 44 today and help us test it we have a beta testing program so the way to do that will be to email the hms inbox you can find it on the hms webpage i would try to say it but i wouldn't quote me on it i believe it's hms usac dot army dot mil but except the atc.hms anyhow find that email send an email to that requesting 44 beta uh that'll eventually get to me one way or another and then i will send you a non-disclosure agreement which just kind of tells you what's appropriate for testing the data software like don't do any official studies with this because it's not officially released things like that and if you agree to the non-disclosure agreement we send you the test software and you can start to test with it uh formal release is hard to say like craig mentioned gary's gary's very dodgy about putting down firm dates and we've learned that in the software development world you can never be too sure so i i yeah i would anticipate uh how can i see this generally soon as we can as soon as we can yeah let's go with that yeah great thank you excellent okay well jalva's question is can the hydrologically corrected data be imported from arc map who wants to take that one on yes so we do have plans for that please view these gis tools as alpha versions right so this is our first rollout of integrating gis functionality in a holistic sense within hms we are going to make things better in the future we're going to continue to make improvements so if you have suggestions like this or things that you wish it would do chances are we probably are going after that in the future but we would love to hear from you so please send that to the hms inbox as well yeah and keep that in mind going forward i'll add to what mike says with the so uh on the meteorologic side all those dss records we have the ability there's a vortex utility to export those met records to geotef so you can export to geotef and view in any gis the gis features in hms those live in a sqlite database inside of the project folder so if you poke around in there you can find the sqlite database you probably have trouble viewing those you have to do some special importing steps to view those in esri products if you're using qgis you can drag and drop and they open right there well should we keep moving yeah i mean there's there's quite a few that we're not going to be able to get through all of these so um mike and uh and tommy if you want to again just if you see any of these on there that you could uh answer quickly or wanted to highlight just have a have a brief look at it because we need to kind of pick and choose which ones to um to handle and uh yeah i mean a lot of people have asked about the integration with hecraz which i think you've already covered um and uh when that might be happening uh one one that i thought um might be worth uh covering now the the an api for uh web applications is there uh is that something that's on the horizon yeah i'll talk about api a little bit that is on the horizon i guess i should clarify too all of the hms development you know a lot of people think oh we're us government we get a big pot of money and we go to town with it which isn't actually the case we work closely with development partners who have specific goals in mind and so we usually partner with them through things like memorandums of mousse memorandum of understanding so that's how we've partnered with other countries and things like that so the api we actually have a few development partners that are keen on this technology and this this capability we've had a kind of a course unpublished api in the past so that functionality is there there's some basic documentation in the user's manual but one of the things that we'll be working on in this next week actually today is the the beginning of our fiscal year so we just kind of reset the books and we're working on new features and so one of the things we have on tap for this next fiscal year is that api so if it's not and it likely won't be in four or four but uh in subsequent releases you'll probably see some basic api functionality and with that most importantly is the documentation on how to interact with the api excellent um so yeah as i guess i'm trevor i'll just uh keep having a look at some of these ones here um and and just kind of group them into uh into groups and again feel free um steve or uh mike or tommy to answer these but again uh those who are not on this session live and we'll be watching on on youtube um won't see these questions so we can uh we can kind of handle uh handle this one um live so that they can see the hear the answers right now um does hcc have a tool for rainfall disaggregation um that's i think i think you may have covered part of that but um want to address that uh yeah so uh you can do that with an hms so you can input a daily time series of precipitation and run at a one minute time step if you would like is the smallest time step that you can use however um that's still not going to apply a specific pattern so it's just going to interpolate from there so if you're specifically asking if you can apply a pattern to that um that is something that we haven't quite done yet though that is a great question and a great thing to tackle in the future excellent yeah i mean it sounds like there's a lot of things coming up uh in the in the future that um uh that we could look forward to um we may be looking well into the future for some of these questions that have come up about the integration some people have asked um are we going to be able to get this to run together with groundwater and with sediment transport and you know sediment yield and i would love to have a grand unified model where i can just open it up and in one fell swoop uh get everything modeled all at once um for now i think we're still going to be stuck with different models and using boundary conditions from one to feed into the other and quite a manual process and you know in the end we are stuck with models and all these models are wrong maybe i'll cover a quick one minute of what was going to go into my intro really quickly because i i put i brought some toys along here as well to show uh maybe i'll i'll share my screen real quick just to show you what these uh some of these other models might be hopefully i hate to do this during the q a part but the question has come up a number of times are you able to see my screen there absolutely this is where we left off with the prequel and i will just uh go heading back in time and covering the integration between heck hms and hec-ras that question came up a few times and just realizing that hydrology when we're doing uh you know the quantification of the flow and getting these hydrographs and the characterization of that flow in a hydraulic model that there is a bunch of overlap what comes into the hydrological models has to be routed using open channel flow methods and now with hecarius and other software you can use grain on grid and use that as a hydrological tool but they ought to be calibrated against each other just so that you get reasonable results um even when you're using these rain on grid results now a model itself um this guy is not the model he's got a model here that you can see i've got a few of these as well in front of me that i use as classroom tools and um in in some of these ones um i guess i just wanted to show you a couple of these um as as as kind of video displays and i'll put some of these up and and you can you can actually have your fifth grade project um if you want looking for a science project uh this is actually a very easy one you drop the rain in and uh you have a quantity of rainfall going in and you see what the runoff is coming out and a lot of these models are used as uh examples of how much sediment comes out and i wish we could do that all at once and determine all the sediment yield and the sediment transport and the scour um but for now it's it's kind of different different models um but what i wanted to do here is just again show you a couple of the ones that um that i've got here i'll just uh stop sharing my screen here um just losing your face a bit there oh yeah i'm i'm gonna actually point this down now and show you the ones that i've got here um just so that we can get the difference between uh heck as a rainfall uh a rain on grid model and um and uh the the heck hms and so i don't know if you can see this thing right here i always use these toys um i'm gonna have this digital elevation model that i'll build a little river and watershed into that's what you've got in a rain on grid model what i do in these ones then is put that into my basin right here and i overlay it with a computational grid which i'm going to put right over the top of it and that computational grid then will get rain on every single one of these grids so i've got this little toy that i use here which gets these syringes and as i push this thing down right now in heck razz you can change the tempo of this i can push down fast and push down slow and squirt water into each one of these grids i can't as yet change the temporal pattern and push one to get a storm moving one direction or another that's coming in the next version but what we've got then with rain going on every single one of these grids in hecara's compare that to hms which is like those bottle models that i showed you and i'll put those up here as well when you drop water into this uh dirt basically have dirt and water coming together and you measure how much is coming out we're quantifying the flow going in and we're quantifying the flow going out based on these parameters soil losses curve numbers um infiltration rates and runoff coefficients this this is mathematically based versus a rain on grid model which many of you have attended those rain on grid webinars in the past um that is just looking at the topography and we try to slow the uh slow the water down with roughness and we try to do all these things that don't quite simulate the same thing you can get the same results out you can calibrate them to be giving you the same results but they're two completely different physical mechanisms and having these models talk to each other is a very challenging process but you can do it either way you can do it as rain on grid or you can do it as a rainfall runoff routing model and you can get results out but what we need to realize is we don't actually know the results the those results are are the uncertainties that go into the rainfall and the uncertainties that come out of what's going on in the soil are still very very great and so we're getting better and better at modeling it and getting these temporal patterns and spatial spatially variable patterns coming in from radar data and everything else but there still is a lot of uncertainty in these models and calibration data is key absolutely key when you work in an area that has no calibration data just remember your model is probably wrong it actually is wrong and uh you know we we do our best to get it as right as we can but it's never going to be absolute so anyway that was my lengthy little explanation on a few of these questions that i've already already answered that were between uh comparing rain on grid in hecara's versus rainfall runoff uh modeling in hms and uh obviously mar mike mike and tommy have done a much better job of explaining the hms piece of that but for our hec-ras fans who have tuned in on this one i just wanted to make sure we highlighted those differences thank you questions then i'll go ahead and put some comments here just let me make some comments stephen hi um been a great explanation from michael and tom and really top-end a lot of the work i do is a lot simpler than that a lot of small projects a lot of flood studies and when i do my course live courses as i will do in the webinar in the in the future course we'll really get down to basics and keep things simple so you can build up from you know two plus one plus one equals two and two plus two equals four and and what we've seen this this um this morning or this evening depending where you are in the world we can you can see the potential of what it can do when you've got the technology and data but at the end of the day if you don't have the data it's going to be very hard to build the models but i can build a lot of models and get quite good flows to put in for example a necroz model that will enable you to do very easy jobs very quickly and i know that michael and tom like to do the big projects um massive square kilometers where for a lot of engineers and i could be wrong that could be it in the wrong place here are dealing with very small catchments trying to design pipe systems trying to understand flooding of streams where they're building um commercial properties for example so my comment is i'll pare it down to be a lot easier for people to understand when it comes to the course now that sounds great and look um we're gonna have to wrap this up here we've just hit the hour mark and it's been wonderful having your time gentlemen and everyone on board uh with all your questions we haven't uh hit all of them i understand that um but i do appreciate that we've got the high science we've got the high modeling and we've got the high practice oh and the stuff you were showing just a moment ago craig just makes it so so uh obvious to everyone and and like you say steve um it's the it's the large uh broad area uh right through to the smaller um acreages you know that that need to be managed here look why don't we have a a a quick go around of a last word from each one uh on this topic then we'll close down the webinar and i'll invite um each one of you now just to have a bit of a one-tenth or two-sentence um wrap-up um who wants to kick off or well maybe we'll kick off with our guests michael and tom okay yeah i'd just like to say thanks for this opportunity and i'm enthused when modelers and users do really cool stuff so if you do cool stuff uh let us know about it send it to us and feel free to reach out and contact us if you have any questions with the software anything like that thank you so much for the opportunity to discuss today i'm super super happy to see everybody from around the world it's always so flattering and awesome to see people using uh our software all over the place and all over the world please send us some use cases let us know how you're using it yep and yourself steve me next yep okay and now it's been really good you've got to remember as a modeler that um hms like any tool is only at all and you've always got to have your backup we're engineers or scientists we've always got to know and understand what we're doing we don't throw numbers in and it's a great way of doing like i say i use it every day practically and it's a great way of producing some hydrographs and volumes of discharge into basins etc and but you've always got to have a backup you've always got to understand what you're doing i like that it takes the dark art out of it it's not a black box it's actually got real data and it's got it you've got to have real ground ground truthing uh and yourself cray oh yeah um i don't want to take too much more time just uh but i don't want to leave anybody hanging everybody who's got these questions and uh asked here um even those who that have been answered and uh dismissed um if we can just collect those and then give tom and michael a chance to weigh in on those and then we can respond individually so we don't want to leave anybody hanging if you asked a question uh we do have it recorded we um on our ends and so we'll make sure we get you taken care of i'll show you yeah if you're able to do that that'd be great absolutely yep no thank you for that that's that's been really good gentlemen um and as craig said i'll i'll collate those questions and shoot them out to every one of the presenters uh so you will hear from us uh well thanks for participating this this feedback going to appear in a moment uh on your screen uh the recording will email to you uh there's the three the webinars coming up in the next four weeks uh and online courses please please do join us for the um uh the hms course coming up in november there's also one as you can see there in october on hec ras and others there i won't go through each one go to our website go look on the youtube and you'll have all the details once again tom michael steve craig it's been a feast it's been a fantastic hour we should go on for another two or three it's absolutely brilliant appreciate all your efforts and time so for now it'll be a goodbye to yourselves and to everybody watching here today thanks very much bye for now subscribe to this youtube channel by clicking the subscription button for future webinars and online short courses please visit our website at australianwaterschool.com
Info
Channel: Australian Water School
Views: 35,530
Rating: undefined out of 5
Keywords:
Id: _cw28gRwxeY
Channel Id: undefined
Length: 59min 29sec (3569 seconds)
Published: Wed Oct 02 2019
Related Videos
Note
Please note that this website is currently a work in progress! Lots of interesting data and statistics to come.