Spill Impact Modeling with ArcHydro?

[BrodySunderland]

Hi All,

I'm researching methods for modeling oil spills on land using ArcGIS. I've looked into ArcHydro and it seems to have some tools I can utilize for the movement of oil over terrain. I've also looked briefly into HEC-GeoRAS / HEC-GeoHMS for water movement in river systems. Can these programs be used to model spills at a given point outside of a river system? Any information will be helpful.

Are there other tools that would help model this process?

Thanks,
Brody

[MarkBoucher]

If you are looking for a way to click on a point where a spill occurs and see a line that indicates where the spill may go downstream based on topography inside or outside of a river system, the flow direction grid produced by Arc Hydro (or tools in Spatial Analyst) will provide that. There is a flow path trace tool for that. You would need to process the DEM up to the point of the flow direction grid for that function.

If you wanted to have a layer of flow paths, you could take the process further to create the steam grid or stream links (rasters) and then drainage lines (polylines).

In the DEM processing, you can "burn in" the storm drain pipes so they are like trenches in the DEM. If they are burned in correctly, they can help you trace the subsurface flow path if the oil got into the storm drain system.

GeoHMS has the same functions as Arc Hydro, but is focused on getting the data ready to export to a HMS hydrology model. Probably not what you want. GeoRAS is like GeoHMS in that it is used to get data ready to export to a hydraulic model. Also probably not what you want.

[BrodySunderland]

If you are looking for a way to click on a point where a spill occurs and see a line that indicates where the spill may go downstream based on topography inside or outside of a river system, the flow direction grid produced by Arc Hydro (or tools in Spatial Analyst) will provide that. There is a flow path trace tool for that. You would need to process the DEM up to the point of the flow direction grid for that function.

If you wanted to have a layer of flow paths, you could take the process further to create the steam grid or stream links (rasters) and then drainage lines (polylines).

In the DEM processing, you can "burn in" the storm drain pipes so they are like trenches in the DEM. If they are burned in correctly, they can help you trace the subsurface flow path if the oil got into the storm drain system.

GeoHMS has the same functions as Arc Hydro, but is focused on getting the data ready to export to a HMS hydrology model. Probably not what you want. GeoRAS is like GeoHMS in that it is used to get data ready to export to a hydraulic model. Also probably not what you want.

Thank you for the response Mark!

I've tested out some of the functions you mentioned and like what I am seeing so far. Also, thanks for the heads up on the storm drain burning in; I wasn't sure of how I was going to account for storm drain routing and it appears your suggestion is the solution.

In ArcHydro, can I assign a spill volume to a spill point in order to model the flow and final destination of the spill? Basically, I want to know if X amount of fluid spilled where would it end up and what areas would it have covered along the way?

Thanks again Mark for your help,
Brody

[MarkBoucher]

I'm pretty sure Arc Hydro will not take spill volume and do an analysis like that for you. You would have to do that another way.

I assume that an oil spill would coat the flow path (stream, pipe, bowl) and there would a loss of volume along the way reducing how far the spill would travel (assuming it is not carried by water). That kind of analysis would likely have to be done outside of Arc Hydro and possibly GIS.

There might be some way to measure the surface area of a creek along a path using another extension (3D analyst using a TIN?), but that would be outside my experience in GIS.

[BrodySunderland]

I assumed this would be the case, but thought I'd consult the ArcHydro veterans before I marked it off the list. I think it's still possible, but I'll probably be creating some custom tools to do it.

One more question: Does ArcHydro account for precipitation when modeling water flow, or it that done in a different program?

Thanks Mark,
-Brody

[MarkBoucher]

Precipitation and rainfall/runoff transformation modeling is done by another program. GIS is excellent for extracting data from geographic data, but is does not do the actual hydrology calculations.