1. Terrain Based Flood Inundation mapping
David Tarboton
Utah State University
Acknowledgements: David Maidment, Nazmus Sazib, Xing Zheng, Solomon Vimal, US Army Corps of Engineers (SWWRP, ERDC, HEC), National Science Foundation (XSEDE, CyberGIS, OpenTopography, CI-WATER), William Penn Foundation (Stroud Water Center Model My Watershed)

2. Concept WRF + NOAH-MP + RAPID/SPRNT produce flows at reach scale
Need a way to obtain reach level hydraulic properties for inputs to these models
Need a way to map from reach scale stage to flood inundation depth
Exploit high resolution topography and 1:1 relationship between reaches (Hydro) and Catchments (Ele)

3. Height above the nearest stream (HANS) flood mapping
Each stream reach has a water depth hw (e.g. from SPRNT)
Each stream reach has an ID
Each grid cell has the ID of the reach it connects to and the height above the nearest stream hs
Evaluate approximating inundation using hw-hs
If(hw(id) > hs(id))
Inundation depth = hw(id) - hs(id)
Else
Inundation depth = 0

4. Mapping Flood Extent and Depth by Reach
Height above stream (relative elevation of land surface cell above cell in stream to which it flows)
Flowlines
Reach Scale Flood Depth
Comid
Depth (ft) 8 9 10 15 14 12 11
Catchments
Inundation map

5. Height above the nearest stream background
TauDEM ( Tesfa, T. K., D. G. Tarboton, D. W. Watson, K. A. T. Schreuders, M. E. Baker and R. M. Wallace, (2011), "Extraction of hydrological proximity measures from DEMs using parallel processing," Environmental Modelling & Software, 26(12): , Nobre, A. D., L. A. Cuartas, M. Hodnett, C. D. Rennó, G. Rodrigues, A. Silveira, M. Waterloo and S. Saleska, (2011), "Height Above the Nearest Drainage – a hydrologically relevant new terrain model," Journal of Hydrology, 404(1–2): 13-29, Nobre, A. D., L. A. Cuartas, M. R. Momo, D. L. Severo, A. Pinheiro and C. A. Nobre, (2015), "HAND contour: a new proxy predictor of inundation extent," Hydrological Processes,

6. Reach Scale River Hydraulic Properties
Comid H A R P T V Ab As 3 4
A table with reach hydraulic parameters as keyed to hydrography
Derived from DEM, LIDAR or HEC RAS cross sections As
Surface Area
Wetted Bed Area Ab L V
Volume T
𝐴= 𝑉 𝐿
Cross Section Area
Depth h A
P= 𝐴𝑏 𝐿 P
Wetted Perimeter
T= 𝐴𝑠 𝐿
Top Width
R= 𝐴 𝑃
Hydraulic Radius

7. SPRNT Model – flow and water depth on large networks
Open source code in Github
Very Large Scale Integrated (VLSI) design of computer chips – solve 100 million equations each night to check on effects of design changes on electricity flow in chip
Dynamic wave routing
Compute water flow by analogy with electricity flow in chips
Ben Hodges, Slide from David Maidment

8. TauDEM http://hydrology.usu.edu/taudem/
4/26/2017
Stream and watershed delineation
Multiple flow direction flow field
Calculation of flow based derivative surfaces
MPI Parallel Implementation for speed up and large problems
Open source platform independent C++ command line executables for each function
Deployed as an ArcGIS Toolbox with python scripts that drive command line executables

9. Representation of Flow Field
Steepest single direction 48 52 56 67 D8 D
This slide shows how the terrain flow field is represented. Early DEM work used a single flow direction model, D8. In 1997 I published the Dinfinity method that proportions flow from each grid cell among downslope neighbors. This, at the expense of some dispersion, allows a better approximation of flow across surfaces.
Tarboton, D. G., (1997), "A New Method for the Determination of Flow Directions and Contributing Areas in Grid Digital Elevation Models," Water Resources Research, 33(2): )

10. TauDEM Parallel Approach
MPI, distributed memory paradigm
Row oriented slices
Each process includes one buffer row on either side
Each process does not change buffer row
Improved runtime efficiency
Capability to run larger problems

11. TauDEM generalized terrain flow surfaces
Dinfinity contributing area
Distance to stream (vertical)
Decaying accumulation
Distance Down and Distance Up hs hr vr vs
Point of interest
Ridge pr ps ss sr
Stream

12. Generalized Flow Accumulation (“Flow Algebra”)
Pki
Pki
𝐴 𝑖 = 𝑤 𝑖 ∆+ 𝑘: 𝑃 𝑘𝑖 >0 𝑃 𝑘𝑖 𝐴 𝑘
Pki
Generalized Flow Accumulation (upslope) i
𝜃 𝑖 =𝐹𝐴( 𝑃 𝑘𝑖 , 𝜃 𝑘 ,…
Generalized Flow Accumulation (downslope)
𝜃 𝑖 =𝐹𝐴( 𝑃 𝑖𝑘 , 𝜃 𝑘 ,…

13. TauDEM model for drop to streami
Pik
ℎ 𝑖 =0 for i on stream raster
ℎ 𝑖 =𝑑(𝑖,𝑘)+ 𝑃 𝑖𝑘 ℎ 𝑘
𝑑 𝑖,𝑘 = z i − z k for vertical drop
Could use D8 instead of Dinfinity
Could use Euclidean distances
Premised on good elevation model and consistent stream raster
Can be applied separately by catchment
parallelism
Incremental refinement of DEM

14. Drop to Stream from TauDEM for Onion Creek near Austin Texas
1/3 arc second NED DEM

15. Mapping from height above nearest stream to flood extent
Reach and Watershed id
Height above nearest stream raster hs
hw(1) = 1.5
Elevations are filled to that of the pools pour point
hw(2) = 2.5
hw(3) = 3.5

16. Reach based height above nearest stream flood map example
hw=8 ft
hw=9 ft
hw=11 ft
hw=12 ft
hw=14 ft
hw=10 ft
hw=15 ft

17. Terrain based derivation of “reach scale” hydraulic properties
For each Catchment
For each height h
Identify cells where hs < h
Wetted Bed Area L As Ab V
Surface Area
Volume
Single Cell Plan Area 𝐴 𝑐 =dx * dy
Surface area 𝐴 𝑠 = 𝐴 𝑐
Bed Area 𝐴 𝑏 = 𝐴 𝑐 1+𝑠𝑙 𝑝 2
Approximates each cell as sloping plane
Volume 𝑉 = 𝐴 𝑐 ℎ− ℎ 𝑠 T
𝐴= 𝑉 𝐿
Cross Section Area P A
P= 𝐴𝑏 𝐿
Wetted Perimeter
T= 𝐴𝑠 𝐿
Top Width
R= 𝐴 𝑃
Hydraulic Radius

18. Reach Hydraulic Properties Example
1 m inundation
3 m inundation
Height (m)
As (m2)
Vol (m3) Ab
L (m)
A=V/L (m2)
P=Ab/L (m)
T=As/L (m)
R=A/P (m) 1
129878
79466
129948
2975
26.7
43.7
0.612 3
319877
530378
320414
178.3
107.7
107.5
1.655

19. Terrain Approximated Reach Average Hydraulic Properties
Hydraulic Radius
Wetted Perimeter
Cross-sectional Area
Top Width
Need to evaluate in Hydraulic Model (e.g. SPRNT)

20. Terrain Catchments reconciled with NHDPlus by “seeding” with stream sources
The approach is predicated on a DEM stream raster consistent with DEM and NHDPlus reaches
Here stream raster computed using weighted flow accumulation starting from source points

21. DEM Flowlines challenged by road barriers
Need for hydrography conditioned DEM

22. DEM Conditioning Options
Fill pits
Carving
Optimal pit remove
Punch through using flowlines
Geonet

23. Pit Filling Original DEM Pits Filled
Elevations are filled to that of the pools pour point
Pits
Pour Points
Increase elevation to the pour point elevation until the pit drains to a neighbor

24. Carving Original DEM Carved DEM
Elevations are filled to that of the pools pour point
Pits
Carve outlets
Lower elevation of neighbor along a predefined drainage path until the pit drains to the outlet point

25. Filling
Minimizing Alterations
Carving

26. Minimizing DEM Alterations
Original DEM
Optimally adjusted
Carved
Elevations are filled to that of the pools pour point
Pits
Filled

27. Decompose computation by catchments or River Corridor
Parallel computation
High resolution in some areas
Incremental improvement

28. Conclusions
NFIE presents a use case that demands consistency between elevation and hydrography information at high resolution
The height above nearest stream approach suggested as way to rapidly approximate real time flood inundation
Includes an approach to approximate reach scale hydraulic properties
Need to evaluate versus more detailed (e.g. HEC-RAS) hydraulic model
Need to evaluate with HRT (LIDAR) vs general (NED) DEM as input
Need to align and reconcile elevation and hydrography

29. Are there any questions ?
AREA 1
AREA 2 3 12