Decide on hydrologic model

[ththelen]

3Di website

[ththelen]

@anardek @caseydietrich I think 3Di looks like an excellent model choice for our application. It is capable of simulating pipe flow, surface water flow, rainfall/runoff, and groundwater flow - all process that we care about for our research application (https://3diwatermanagement.com/for/hydrologists/). The subgrid model should allow use to simulate hyper-local floods with greater precision (see Subgrid versus depth averaging section in previous link). Finally, I found the attached paper from 2017 that used 3Di to map inundation from SLR and storm events in California. The 3Di model inputs for this study were time-series water level data and ground surface data, both of which we will have available for Carolina Beach. Ultimately, I think 3Di will be a nice package for both simulating our hydrologic processes of interest and producing visualizations of flood extents.

Let me know if you have any questions or comments.

Ju2017_Chapter_PlanningForTheChangeMappingSea.pdf

[ththelen]

Modeling takeaways from 3Di meeting with Nici Volp

• How structures are represented in the elevation DEM is really important. If structures are erased from the elevation model, we will not be correctly overland surface flow. Options to correct for this include editing them DEM or adding an obstacle item in 3Di with a constant above flooding elevation and boundaries around all the structures.
• To make flood extent maps subtract the DEM raster elevation from the water level elevation raster. This is the same process that the subgrid model uses in its calculation process.
• I should use a variable friction raster. At a minimum I should probably pick one value for elevations below NAVD88 (generally wet areas - same as ADCIRC maybe) and another for elevations about NAVD88 (generally dry areas).
• 3Di will be rolling out functionality for pipe/groundwater interaction and vegetation routines (for simulating wetlands) within the next year.
• There is no conversion of vertical datums within 3Di. The vertical datum you use for DEM and water level inputs is carried through the simulation to water level results.
• We should rethink the ADCIRC / 3Di coupling point. Coupling the two models so close to our area of interest does not allow for any overlap to compare model results. It also introduces an artificial boundary (the coupling point) near to where we care about water levels. This is not a good thing. Nici suggests the tidal inlet and the Cape Fear River as better coupling locations.
• Nici suggests a simulation time step of 15-30 second in 3Di. Model output results can be tracked for an duration longer than the simulation time step.
• Trying a running the same simulation on 3-4 different grid sizes is the best way to find out what 3Di resolution works best. The grid must be large enough to capture hydrodynamics. 10-20 meters is a common grid size for urban flooding applications.

[anardek]

Is it free?

On Friday, July 29, 2022, Thomas Thelen @.***> wrote:

@anardek https://github.com/anardek @caseydietrich https://github.com/caseydietrich I think 3Di looks like an excellent model choice for our application. It is capable of simulating pipe flow, surface water flow, rainfall/runoff, and groundwater flow - all process that we care about for our research application ( https://3diwatermanagement.com/for/hydrologists/). The subgrid model should allow use to simulate hyper-local floods with greater precision (see Subgrid versus depth averaging section in previous link). Finally, I found the attached paper from 2017 that used 3Di to map inundation from SLR and storm events in California. The 3Di model inputs for this study were time-series water level data and ground surface data, both of which we will have available for Carolina Beach. Ultimately, I think 3Di will be a nice package for both simulating our hydrologic processes of interest and producing visualizations of flood extents.

Let me know if you have any questions or comments.

Ju2017_Chapter_PlanningForTheChangeMappingSea.pdf https://github.com/NCSU-CHAZ/Sunny-Day-Flooding/files/9220514/Ju2017_Chapter_PlanningForTheChangeMappingSea.pdf

— Reply to this email directly, view it on GitHub https://github.com/NCSU-CHAZ/Sunny-Day-Flooding/issues/7#issuecomment-1199319390, or unsubscribe https://github.com/notifications/unsubscribe-auth/ANXYL57HLBRTPS2WCRECPIDVWPPPDANCNFSM55AY6N2A . You are receiving this because you were mentioned.Message ID: @.***>