Example Flooding Scenario

[BenMaynard]

I am a master's student working on my final project.  One of the nice add-ons I would like is just a flood model to show that, if the water is at a specific height above sea level, which roads on a road grid would be affected.  Seems pretty simple!

I have the road network, which I imported into ArcMap but have no idea where to start.  I also have read that the Army Corps of Engineers HEC-RAS tool (which I have been playing with) could help.

Does anyone have a high level list of steps or a general example that I may look at?  Thanks in advance for your help!

[MattMead]

I cant provide you with examples as that would require client release forms, but I can help get you started on your own. 

You are both lucky and unlucky.  Projects like this are already done for a number of reasons and a similar system is being included in FEMA's new RISK MAP program.  Does it have to be a coastal area?  That greatly complicates modeling(which is the unlucky part).  When you are dealing with coastal areas they are generally divided into relatively short stretches and a flood elevation is developed for just that section. Due to the wave runup, areas that are next to each other may have different flood elevations during the same storm event.  HEC-RAS is a River Analysis System and does not do coastal flooding.  There are models out there that do that, but being located in middle america I havent had to use one yet.

You are lucky though in that USGS has gages on rivers and NOAA has coastal gages (http://tidesandcurrents.noaa.gov/) that provide elevation and discharge data over time.  Included in the gage data are historical storms, ie hurricane Floyd hitting the Carolinas was captured by some gages so you can watch the storm surge come in and recede which is basically what you need).This would negate having to develop a flood model(unless its a requirement of your masters program)

Data needs:
topography-either contours or a digital elevation model(DEM) for the area of interest
water surface elevations-grab a gage and get some elevation data

Tools:
ArcGIS
3dAnalyst extension
Possibly Spatial Analyst

Process
1) Create a polyline shapefile and draw lines describing the area you think your gage data is accurate for and assign the appropriate elevations(since you want multiple gage events create an attribute field for each event).

2) using 3dAnalyst convert your polyline file to a TIN (3dAnalyst Toolbar->Create/Modify TIN select your polyline file, assign the output location, identify the elevation source, all else leave at default)

3)convert your topography to a TIN
  3a) if contours, do it the same way you did the polyline elevation file
  3b)if a DEM use 3dAnalyst toolbar-> Convert->Raster to TIN  (keep the Z value as low as possible)

4) If you just want a polygon to show the extents of the flooding use "TIN difference" which is located in the 3d Analyst Tools Toolbox under "Tin surface"

5) if you want a raster that shows the depths at each location convert your Terrain and Water Surface TINs to Rasters (3dAnalyst toolbar->Convert-Tin to Raster) and use Raster Calculator in the Spatial Analyst toolbar to subtract your terrain data from the water surface raster.

If you have any questions or need something explained more let me know and I'll do what I can.  Good luck with your masters degree.

[BenMaynard]

Matt, you rock.  Thank you.  I'll get started.

[ShawnKraft]

Matt, I was wondering if I could pick your brain about this.  I have 2 foot contours along the Laramie River.  Using the NOAA stream gauge, there is only one, I calculated water surface levels for the river in half foot increments beginning at 4.5 foot up to 7 foot. I then selected the contours that corresponded with my water surface level calculations.  In my calculations, I adjusted for the rise in elevation as I moved up river. Example, stream gauge is at 7129, I added 4.5 to that to get my first contour of 7131.5 water surface level and so on up the river.  I have each water surface level saved as a shapefile so I have one for 4.5 foot, one for 5 foot and so on. I used them all when I ran the surface difference and it generated the difference with Above, Same, and Below values for the data.  I am assuming that since my highest value was 7 foot that it based the Above on anything above 7 foot level and anything else below 7 as Same and Below.  Would that be correct? I would appreciate any help.  Thanks

[MattMead]

Shawn,
It depends on what order you put your files in, but yes, your "above" should consist of anything with a higher elevation than your water surface elevation.  The resultant "below" and "same" should be a rough floodplain for your stream.  If you reverse the order of your terrain and WS files it will give the reverse where "above" is actually everything below.  It doesnt really matter one way or the other as long as you know that its reversed.

Let me know if you need anything else.

[ShawnKraft]

I was wondering what you mean exactly by this 1) Create a polyline shapefile and draw lines describing the area you think your gage data is accurate for and assign the appropriate elevations(since you want multiple gage events create an attribute field for each event).

Could I just select the contours that correspond to my water surface elevations and use those?

[MattMead]

I was wondering what you mean exactly by this 1) Create a polyline shapefile and draw lines describing the area you think your gage data is accurate for and assign the appropriate elevations(since you want multiple gage events create an attribute field for each event).

Could I just select the contours that correspond to my water surface elevations and use those?

Shawn,
I'm sorry I am so late in getting back to you, work got busy. 

In rereading your original post I'm assuming you are trying to generate a library of flood maps that can be tied to AHPS data to create a real time flood map?  You wouldn't happen to work for USGS would you? Unfortunately we are starting to approach workline specific information and I'm not sure how much I can say/be specific about.  I'll do the best I can to help you though.

That statement was basically saying that he needed to draw cross sections in.  Since he was dealing with a coastal area, I wasnt sure how to phrase it.  Since you are dealing with a river, these polylines would be the cross sections used to create any modeling you have in reach.  I googled your river and it looks like you are probably up in the mountains or at minimum the foothills of them(very pretty looking too, I'm quite jealous).  Given this, I'd assume you have a pretty severe slope to your river which means the elevation recorded by the gage probably isnt accurate very far away from the recording location.  The elevations can be used to calibrate a detailed hydraulic model that can then be used to extend the gage results upstream and downstream.  We always use HEC-GeoRAS to create our models so we always already have cross section data.  Its pretty easy this way.

To answer your other question, maybe? 

If you are using the closest contour(rounding up) running parallel to the stream on both sides and interpolating a water surface between them it will work for a very short reach....that said, if you have contours that exactly match your elevation you dont need to do the difference as your polygon created by this process should exactly match your contours.

If you have additional elevation data for the same storm event(ei you have a detailed model that has been calibrated) then no, that would not work.  This way you would have variations in the elevations along the stream which your surface created by contours would not be able to represent.  You could create 3dpolylines but its kind of a pain and its much easier to use cross sections with the appropriate elevations.

Edit: if this doesnt answer your question could you post a screenshot of what you are trying to do, I might not be correctly understanding your question?

[NguyenMinh_Huyen]

I cant provide you with examples as that would require client release forms, but I can help get you started on your own. 

You are both lucky and unlucky.  Projects like this are already done for a number of reasons and a similar system is being included in FEMA's new RISK MAP program.  Does it have to be a coastal area?  That greatly complicates modeling(which is the unlucky part).  When you are dealing with coastal areas they are generally divided into relatively short stretches and a flood elevation is developed for just that section. Due to the wave runup, areas that are next to each other may have different flood elevations during the same storm event.  HEC-RAS is a River Analysis System and does not do coastal flooding.  There are models out there that do that, but being located in middle america I havent had to use one yet.

You are lucky though in that USGS has gages on rivers and NOAA has coastal gages (http://tidesandcurrents.noaa.gov/) that provide elevation and discharge data over time.  Included in the gage data are historical storms, ie hurricane Floyd hitting the Carolinas was captured by some gages so you can watch the storm surge come in and recede which is basically what you need).This would negate having to develop a flood model(unless its a requirement of your masters program)

Data needs:
topography-either contours or a digital elevation model(DEM) for the area of interest
water surface elevations-grab a gage and get some elevation data

Tools:
ArcGIS
3dAnalyst extension
Possibly Spatial Analyst

Process
1) Create a polyline shapefile and draw lines describing the area you think your gage data is accurate for and assign the appropriate elevations(since you want multiple gage events create an attribute field for each event).

2) using 3dAnalyst convert your polyline file to a TIN (3dAnalyst Toolbar->Create/Modify TIN select your polyline file, assign the output location, identify the elevation source, all else leave at default)

3)convert your topography to a TIN
  3a) if contours, do it the same way you did the polyline elevation file
  3b)if a DEM use 3dAnalyst toolbar-> Convert->Raster to TIN  (keep the Z value as low as possible)

4) If you just want a polygon to show the extents of the flooding use "TIN difference" which is located in the 3d Analyst Tools Toolbox under "Tin surface"

5) if you want a raster that shows the depths at each location convert your Terrain and Water Surface TINs to Rasters (3dAnalyst toolbar->Convert-Tin to Raster) and use Raster Calculator in the Spatial Analyst toolbar to subtract your terrain data from the water surface raster.

If you have any questions or need something explained more let me know and I'll do what I can.  Good luck with your masters degree.

Hi Matt,
I have quite the same problem. I have a DEM of the interested area, a vector layer of land use and of course a specific height of water level
Can I solve the problem just by using the Raster Calculator in the Spatial Analyst tools to visualize which area will be flooded with that specific height of water, then merge it with the land use to calculate the surface area of the flooded area.
Is it correct? and Would it be more simple doing this way?

Thanks in advance!

[anitaYeeting]

Hi,

I am trying to do a flood model as well with a contour line data of 1m interval. The problem is the lowest contour line started with 2m which is about the high water mark level in the island i am studying.

I have created a DEM on this contour line with a lowest raster value of 1.7m...in order to map a rise in sea level, MSL is always taken as a baseline, however, the mean sea level is lower than the contour lines therefore it can not be displayed on the DEM. In this case, I converted all my flood levels to be relative to the same datum as the contour lines or DEM.

I managed to delineate contours with flood elevations, however I do not know how to mark these areas as flood zones.

Another aim of my study is to determine the land area loss due to flooding.

I also realized that the selected contour lines with flood elevations are just line features and therefore can not produce areas because they are not polygons. I have tried to convert these lines into polygons using feature to polygon and then try to add an area field in the table, and calculate the geometry. The area produced is not what I am expecting because it is too small to the normal land area.

Any tips on this????I appreciate your ideas....

Thanks

Jay