1. Floodplain Mapping using TINs
Triangulated Irregular Networks (TINs)
Representation of stream channels using TINs
Floodplain delineation using HEC-HMS, HEC-RAS and ArcView

2. TIN with Surface Features
Classroom
UT Football
Stadium
Waller Creek

3. A Portion of the TIN

4. Input Data for this Portion
Mass Points
Soft Breaklines
Hard Breaklines

5. TIN Vertices and Triangles

6. TIN Surface Model
Waller
Creek
Street and
Bridge

7. 3-D Scene

8. 3-D Scene with Buildings

9. Floodplain Mapping using TINs
Triangulated Irregular Networks (TINs)
Representation of stream channels using TINs
Floodplain delineation using HEC-HMS, HEC-RAS and ArcView

10. River Modeling
River hydraulic modeling provides a tool to study and gain understanding of hydraulic flow phenomena
Topographic data describe the geometry of the simulated river system and permit the establishment of model topology
HEC-RAS, MIKE 11 all hydraulic models require channel information for model development

11. River Morphology
One way to track the amount and location of erosion and deposition in a river valley is to establish a series of cross-sections through the valley and survey the cross-sections on a regular basis…. such cross-sections help to identify regions of sediment erosion and deposition.

12. Flood Inundation

13. Floodplain Delineation
The channel and floodplain are both integral parts of the natural conveyance of a stream. The floodplain carries flow in excess of the channel capacity. The greater the discharge, the greater the extent of inundation. …. Because of its devastating nature, flooding poses serious hazards to human populations in many parts of the world. “The Flood Disaster Protection Act of 1973” required the identification of all floodplain areas in the United States and the establishments of flood-risk zones within those areas.

14. Floodplain Delineation
Ideally should have Two D Models…... A floodplain delineation process determines inundation extent by comparing water levels with ground surface elevations. We start with a DTM or topomap. > Water levels are computed based on cross-sections. During normal condition, flow remains within the main channel, but > during flood, water spill over the bank. In these cases it’s important to extent cross-section over the floodplain. > We then bring back water levels on the topomap. > Extent water levels until hit contour of higher elevations. > Finally delineate floodplain following the contours. The accuracy largely depends on water levels and in turns on the cross-section used for computation.

15. Channel and Cross-Section
Direction of Flow
Cross-Section
Channel
Before we get into detail let’s quickly define a channel and a cross-section. A river of stream Channel is a conduit or water course carrying water flow under gravity.
A Cross-Section refers to the section of a channel taken normal to the direction of the flow.

16. ProfileLines
Types
1- Thalweg
2- LeftBank
3- RightBank
4- LeftFloodLine
5- RightFloodLine
It is important to relate other features such as embankments and the extent of inundation with the main channel itself and see everything as an integrated system rather than as individual components. An embankment profile provides useful information regarding breaches or identifies the location of possible overspilling. Historical flood extent information is very useful for checking the validity of a flood model. These longitudinal view of channel features are stored as ProfileLines in this model. These are linear features. Currently five different features are supported such as Thalweg, left and right banks, left and right flood lines. ProfileLines and CrossSections are linked through Channel_ID.
ProfileLines and CrossSections are linked through Channel_ID

17. TIN as a source of cross-sections
Advances in GIS have provided many tools for extracting cross-section data from DTMs. Since the TIN allows variable point density, it actually captures and represent surface features better than a Grid. Therefore, TIN is preferred as a cross-section data source over the Grid. The process involves defining the TIN as a surface and digitizing cross-section cutlines. The ground elevation beneath each vertex is then interpolated from the surface.

18. CrossSections
Even though a cross-section is usually plotted as distance vs. elevation graph. > However, we actually start with a cutline on the ground (xy plane) and extract elevation (z) for each of the points. > These points with x,y,z coordinates eventually proved a 3D view of a cross-section.

19. Floodplain Mapping using TINs
Triangulated Irregular Networks (TINs)
Representation of stream channels using TINs
Floodplain delineation using HEC-HMS, HEC-RAS and ArcView

20. Floodplain Mapping Approach
HEC-HMS
HEC-RAS
Flow
discharge
Water
surface
profiles
Parameters
Schematic
Geometric
data
HEC-GeoHMS
HEC-GeoRas
This slide presents the methodology followed for floodplain mapping.
The approach is based on the use of Geographic Information Systems (GIS), a hydrologic model and a hydraulic model.
An informal definition of GIS would be a software which allows you to have graphical representations of features with associated information stored in a database. This give you a powerful tool that can be used in very different fields including hydrology and hydraulics. A hydrologic model would convert a flow discharge along the streams existing in that area. Finally, a hydraulic model lets you determine the elevation of the water flowing through a channel/river given a certain discharge.
We start with some digital spatial data in GIS and use it – by means of the GIS-based application CRWR-PrePro – to delineate watersheds and develop watershed parameters for hydrologic modeling – in this case HEC Hydrologic Modeling System (HEC-HMS) –. After adding precipitation data, you run the hydrologic model and determine the flow values corresponding to different amounts of rain fallen in a given area. These flow values can be then entered into the hydraulic model – in this case HEC River Analysis System (HEC-RAS) – to generate water elevation values . But first you need to input some geometric data into the hydraulic model. This geometric information is extracted from a very detailed digital representation of the terrain in GIS. The application used to link GIS and HEC-RAS is called AVRas. AVRas extracts the information contained in the TIN, exports it into HEC-RAS, reads the results of the hydraulic model and represent the flooded areas.
ArcView

21. Purpose
Integrate/Validate existing tools for floodplain determination and visualization.
Reduce the dependence on field data.
Improve the floodplain analyses capabilities (lower costs and more accuracy).
As I mentioned before, computer models are assisting in the floodplain determination process. My research focused on the use of existing tools for floodplain analysis that sometimes are used separately, integrating them to create a single one.
As I said, unfortunately these computer models require extremely detailed terrain information that involves large amounts of fieldwork. Furthermore, water surface elevations maps are usually plotted by hand, resulting in a tedious and potentially inaccurate task. Solving these two limitations would dramatically improve the floodplain analyses capabilities, reducing costs and improving the accuracy of the results.

22. Digital Spatial Data Digital elevation model (DEM). Stream definition.
Digital spatial data is processed and used to determine some parameters and the schematic used by HEC-HMS (our hydrologic model). These data includes at least a digital elevation model (DEM) and a stream network. But you can also use LULC and STATSGO (calculate CN) and stream gages.
Austin Watersheds with Streams derived from Aerial Photographs. Streamlines generated by the aerial photographs are not always continuous (storm sewers). Some information is required for correcting stream networks, including DEMs, contours, storm sewers and DOQs.
HEC-HMS
HEC-RAS
ArcView

23. CRWR-PrePro Watershed delineation. Reach/Watershed
parameters determination.
The hydrologic model is going to require a series of parameters to accomplish the analysis. These are reach and watershed parameters.
Reach Parameters: Grid code, watershed code, length (m), velocity (m/s), routing method (Lag or Muskingum), lag time (min), Muskingum X, Muskingum K (hr), number of sub-reaches.
Watershed Parameters: Grid code, area (km2, unit hydrograph model (SCS), length of longest flow path (m), slope of longest flow path (fraction), average curve number, lag-time (min), baseflow (none).
Polygons and lines are attributed with hydrologic parameters and stored in attribute tables. The information is exported into HEC-HMS.
HEC-HMS
HEC-RAS
ArcView

24. HEC-HMS: Flow Determination
As I already mentioned, HEC-HMS determines the flow values corresponding to different amounts of rain fallen in a given area. The model divides a watershed into subbasins. It converts an amount of rain into runoff at the exist of each sub-basin, and routes this runoff along the reaches until the outlet of the watershed.
The model supports different methods for calculating infiltration/runoff, transforming this runoff into a flow at the exit of each sub-basin, and routing this flow. The required parameters were obtained from GIS. Watersheds and reaches are attributed with hydrologic parameters and stored in tables.
The model also requires information related to the precipitation (100 year hypothetical storm) and some to specify the duration of the simulation and also the time interval of the calculations.
You can calculate the discharge at a particular location and plot the corresponding hydrograph and use this information as the input flow data in HEC-RAS.
ArcView
HEC-RAS
HEC-HMS

25. HMS-RAS Connection HMS Junctions RAS Cross-sections
This slide illustrates the connection between HEC-HMS and HEC-RAS. A junction has its equivalent in a cross-section in HEC-RAS. The centerline used in both models is exactly the same (these windows are only schematic) ensuring that control points are located at the same point along the streams.
ArcView
HEC-RAS
HEC-HMS

26. HMS-RAS Connection HMS Hydrograph RAS Flow Data (0500, 3559.6)
Flow values (represented here as a hydrograph) are stored in tables in HEC-RAS. You can see how the point with the coordinates 0500, (meaning 5:00 AM and scf discharge) is translated to the profile #5 (PF#5) which corresponds to the values measures at 5:00 AM.
ArcView
HEC-RAS
HEC-HMS

27. Digital Terrain Model: TIN
Observed points and
breaklines for
constructing a
triangular irregular
network (TIN).
Following the methodology, GIS is used again to obtain the geometric information required - this time - by the hydraulic model HEC-RAS. This terrain information is stored in a digital terrain model called triangular irregular network. A TIN is an efficient way to represent continuous surfaces as a series of linked triangles generated according to the Delaunay criterion (making triangles as equi-angular as possible).
The information required to generate our TIN came from aerial images of the area. TINs (and also DOQs) are the result of processing these images using specific processing software and instruments as well as some ground control points. Observed points and breaklines can be extracted from the aerial pictures and be used for constructing a TIN.
ArcView
HEC-RAS
HEC-HMS

28. Digital Terrain Model: TIN
Embedding Buildings
into the TIN.
Using Avenue scripts (ReturnBuffered/BufferElev.ave) its possible to modify the TINs and embed any building or other structures that were stores as 2D polygons in other coverages. The tricky thing is that a single point cannot have the same x, y coordinates and to different z.
For instance, if there is a building close to a channel, the flow water could be affected by this building. By including the buildings into the TIN its effect can be considered.
ArcView
HEC-RAS
HEC-HMS

29. GIS-RAS Connection Stream centerline. Banks. Flow paths.
Cross sections.
To analyze stream flow, HEC-RAS represents the stream as a set of cross-sections along the channel. At each cross-section, bank stations define the location of the channel. Floodplains are used by HEC-RAS to identify the existence of bends along the streams.
Right now, centerlines, banks & Floodplain are delineated over the TIN. Extracting them from a DOQ would increase the accuracy.
ArcView
HEC-RAS
HEC-HMS

30. GIS-RAS Connection Location of cross sections.
It is important to locate the cross-sections properly in order to define the Channel and Bridges. HEC-RAS requires two cross-section before and after any bridge or hydraulic structure such a culvert that we want to define.
We can appreciate one of the limitations of the TIN here. The sides of the bridge are to perpendicular to the channel. Instead, some slope can be appreciated. I will comment this point later on when talking about the limitations of this methodology.
ArcView
HEC-RAS
HEC-HMS

31. Hydraulic Modeling with HEC-RAS
HEC-RAS schematic with the stream centerline and the cross-sections we defined in GIS. We can appreciate in each cross-section the banks and water elevation (once calculated).
The primary procedure for computing water surface profiles between cross-sections used for steady gradually varied flow is the standard step method (based on iterative solution of the energy equation from downstream -water elevation- to upstream).
Cross-section extracted
from the TIN.
RAS stream geometry.
HEC-HMS
HEC-RAS
ArcView

32. Hydraulic Modeling with HEC-RAS
Resulting water
elevations.
The output of the model comes in two primary forms: a graphical xyz perspective plot, and in ASCII text format.
At this point, the hydraulic engineer would typically return to the original contour maps showing the cross-sections, and plot the water elevations. In this manner, the floodplain extent is determined. This could quickly become tedious if the goal is to evaluate different flow scenarios. However, we can use AVRas and export the water elevations into GIS and accomplish the floodplain delineation automatically.
HEC-HMS
HEC-RAS
ArcView

33. Floodplain Mapping Floodplain for 500 cfs. HEC-HMS HEC-RAS ArcView
Overlapping Buildings and Transportation layers helps identify affected areas.
The user can click anywhere on the inundated area and determine water depth.
HEC-HMS
HEC-RAS
ArcView

34. Floodplain Mapping 2-D floodplain animation (500 – 5,000 cfs).
Or you can make animated gifs like this one showing the effect when the flow varies from cfs.

35. Floodplain Mapping 3-D floodplain animation.
Finally, you can also create 3D animations that give you a more realistic vision of the study area.

36. Limitations Bridges/culverts: - depend on field data. - data input
by hand.
The first limitation is the requirement of field work to obtain the data that defines structures such as bridges and culverts. This information cannot be obtained from a digital representation of the terrain right now. Furthermore, these data has to be input by hand.

37. Limitations The accuracy obtained from our TIN is not good enough.
The second and more important one is the degree of detail that you can obtain from the TIN. In this graph, yo0u can see that there is a difference of 2-3 ft. It can be explained by a not very exact location of the cross-sections corresponding to the field and to the TIN. The existence of water also induces some problems. The second one has a more difficult explanation. You probably require a more detailed TIN...

38. Solutions New technologies (i.e. LIDAR) are improving the
quality of the
digital terrain
representations.
New technologies
(i.e. LADAR) are
improving the
quality of the
digital terrain
representations.
These TINs have been obtained using new techniques like Laser Radar Systems (LADAR). You can see that the degree of detail is really good. The results from a TIN like this would be much more accurate.
Source: digital representation of NYC
generated by ASI and published by ESRI.

39. Michael Schultz
NWS River Forecast Center
Fort Worth, Texas
1998 Guadalupe
Flood

41. Cross-Sections developed
using HEC-GeoRAS and
National Elevation Dataset

42. Hydrologic simulation
Using NWS Fldwav model