TABLE OF CONTENT

I - STATEMENT OF THE PROBLEM... 1

II - MY OBJECTIVE.. 2

III - MY SOURCES.. 4

IV - MY RESULTS.. 5

1) The first step: the terrain processing using ARCVIEW 5

2) Second step: the pre-processing using HEC-GeoRAS 8

I - STATEMENT OF THE PROBLEM

Brownsville and Matamoros are two cities located along the Rio Grande.

Brownsville / Matamoros

Because of the climate and the particular hydrologic landscape of this region, heavy floods often touch Matamoros and Brownsville and that is why it seems necessary to build a model that could predict the floods. One of the difficulties of the project is that the region is very flat and using elevation data to build a model is a kind of challenge. The elevation data that will be used will be LIDAR data because this technology offers a very high precision. This project is a two-year project since it is my Master project and I propose to begin this research work as follow.

II - MY OBJECTIVE

My work is going to be organised as follow (see the schema bellow):

· The first step is the terrain processing using ARCVIEW which consist on:

- Collecting LIDAR data and to convert them into a 1 m DEM grid.

- Building a Hydro Network and applying the ArcHydro Schema.

- Building a TIN and make a list of the problems I am going to face during this step.

· The second step is the pre-processing using HEC-GeoRAS. At the end of this step I should:

- Have the stream centrelines and the riverbanks.

- The flood plain boundaries.

- The cross section along the stream.

· The third step is to build the Hydrologic modelling with HEC-HMS. I am going to use some flow data and the land use to build the hydrograph.

· The fourth step is to run the HEC-RAS steady state simulation.

· The fifth step is the HEC-GeoRAS posprocessing.

· The last step is the flood map visualisation with ARCVIEW.

My objective for this semester is to concentrate on the first step and to start working on HEC-GeoRAS for the pre-processing.

III - MY SOURCES

Actually I have:

- Some elevation data that have been collected by satellite. The precision is 7.5 meters. I got these data from the Bureau of Economic Geology.

- I also have some LIDAR data form a small part of Brownsville. These data have auto filtered to remove vegetation and buildings in it.

- The Hydro Edge, the Water Body of the region.

- The land use.

I am going to collect:

- LIDAR data for the Rio Grande in Brownsville and Matamoros. I am going to put some GPS marks in December to be able to collect elevation data in January.

IV - MY RESULTS

1) First step: the terrain processing using ARCVIEW

- The first thing I did was to find information about the LIDAR technology and to understand precisely the process which is used to collect elevation data. Therefore I met Sir Roberto Gutierrez who is working in Aerospace engineering. He explained me the major problems encountered with the use of LIDAR such as the removal of vegetation, the removal of buildings or the bridges which are seen as dams. The quality of the data depends on the efficiency of these operations.

- Then I wanted to build a hydro Network for Brownsville and Matamoros. It seems obvious that because of the flatness of the region and the bad precision of the data I had at the beginning (7.5 m), the use of Arc Hydro Terrain Processing to build the Network couldn’t give me a good result but I wanted to check it (see the result here).

I also tried to build the Hydro network with the rivers features but as you can see, the network is huge and I got many errors (see the result here). Many of these errors were due to loops and I have corrected them but I still can’t make out where the other errors come from. I spent a lot of time on this without any success so I decided to concentrate my work on the Rio Grande only. This new orientation makes sense when we have a look at the inundation map. We see that one of the biggest part of the inundate zone of Brownsville is linked with the Rio Grande.:

Later in the semester I got some LIDAR data for a small part of Brownsville. I used these data as an input in Arc Hydro Terrain processing and this time I think that, because the region is too flat we can’t use LIDAR Data to define the streams and the flow direction. When I used these data I had a problem because the dispersion of the elevation data was not homogenous (data were missing in some areas). I had to interpolate the raster with Splin function to obtain something usable. (See the result here).

Conclusion: I am going to focus my work on the Rio Grande because the Network is too complicated and this restriction doesn’t seem too restrictive for the flood mapping.

- The second step was to build a TIN with different data. A TIN is required is input for HEC-GeoRAS.

My objective was to see which precision I needed to define the cross section of the river (I need the cross section in HEC-GeoRas). I started by building a TIN using the less precise data I had (7.5 m resolution). It appears that this data are not precise enough to define the cross sections of the Rio Grande. Then I used LIDAR data but I got an error. I think that it may be because the data are too heavy.

I have also tried, with a given TIN to make out how to deal with bridges since they are seen as dams. From pictures I have noticed that there were two bridges across the Rio Grande so I will have to go throw this problem. I took the cross sections before and after the bridge and I linked these cross sections with straight lines. Then I interpolated this with the TIN and I have broken the bridge:

Conclusion: I think that I have a good overview of the problems we can have when we work on LIDAR data and their solutions. What I did manually is being programmed by Venkatesh Merwade. It is also clear that I need LIDAR data (high precision) to build the TIN with which I can work on.

3) Second step: the pre-processing using HEC-GeoRAS

- I have defined the stream centrelines and the river banks. The left bank is defined as the left side when looking downstream. It is important to separate from the stream channel proper because of differences in resistance values.

- I still have to define the Overbanks flowpaths and the cross section lines.

Conclusion: LIDAR data are going to be collected in January and I will be ready to use them to go throw the two first steps of the building of my model. This term project has been an opportunity to discover GIS, Arc Hydro and HEC-GeoRAS pre-processor and to get use to these software. The most frequent problem I had was a problem of coordinate system and to link data from the Mexican side to data from the American side.

River Banks of the Rio Grande