Transcript
2D or not 2D,
that is the Hydraulic
Question!
Reinaldo Garcia, PhD
Webinar – 10/06/2011
• Model vs Reality
• 1D and 2D hydraulic model assumptions
• When the 1D approach is not enough
• Myths about 2D models
• 2D model advantages and limitations
• When 2D?
Webinar Outline
Models only approximate reality
Models perceive only their own reality
All models are not created equal
There is nothing wrong
with the 1D approach
If it is properly used
The BIG problem is that almost always models give an answer regardless of the incompatibility between model
assumptions and reality
• On a given cross section:
– Constant Water Surface Elevation
– Uniform (almost) velocity
• Unidirectional flow
• Negligible exchange of momentum between main channel and floodplain
• Many unsteady 1D models created primarily for subcritical flows (e.g. HEC-RAS)
1D model assumptions
1D Equations
1D model perspective
• Confined flow
• Floodplain storage is negligible
• Flow is mostly unidirectional
• No need of detailed velocities
When the 1D approach is fine?
• Threaded rivers and poorly confined flow
• Floodplain storage is important
• Non-uniform flow patterns
• Complex channel-floodplain interaction
When the 1D approach is not OK?
Make everything as simple as possible, but not simpler…
• Negligible vertical accelerations
• Uniform velocity throughout depth
2D model assumptions
2D Shallow Water Equations
2D model perspective
• Accounts for lateral variation in WSE’s without Ad hoc artifacts
• Better schematization of distributed flow in threaded rivers or unconfined flows
• More realistic computation of velocities and water surface elevations
2D model advantages
No-secondary flow (bends)
2D model limitations
2D model limitations No-vertical accelerations No-vertical velocity
1D vs 2D
• Require more data
• Only used for elaborate and detailed models
• Difficult to understand
• Difficult to use
• Slower than 1D models
Myths and half truths about 2D models
“If the flow pattern is truly two-dimensional, then a 1D approach will be useless as a predictive tool.”
“In such cases 1D models are unable to predict unobserved events.”
J. Cunge
Require more data?
Require more data?
Data 1D Model 2D Model
Elevations Surveyed cross sections LiDAR, DTM, DEM
Manning’s n 1D Spatial distribution 2D Spatial Distribution
Boundary Conditions Inflow Hydrograph Outflow WSE or rating curve
Inflow Hydrograph Outflow WSE or rating curve
• Difficulties lie in the physical processes
• 2D model may only be a better approximation to complex hydraulics
Difficult to understand?
YES!
Are 2D models slower than 1D models?
• +43,000 elements
• 1 hour simulation
• 3 minutes runtime
2D or RiverFLO-2D…
How difficult is it to setup a 2D model?
Simplicity or computational economy should not justify using a modeling approach that
ignores fundamental aspects of the flow hydraulics!
• If the flow is truly 2D, you will not get good results, no matter how much you tweak a 1D model.
• Use 2D when:
– Storage areas are important
– WSE is not constant on cross sections
– Threaded rivers
– Flow attenuation is important
2D or not 2D?
Hydronia.com
Next webinar:
Effective and practical transition from 1D to
2D modeling October 27th
Where do we go from here?
top related