Blake Downey

Class: GIS in W.R.

Prof: Christopher Neale

Assign: Term Project

Re-mapping the Logan River Flood Plain to Estimate Potential Damages and Create an Easy to Read map

Table of Contents:

                        1.  Introduction

2.  Objectives

3.  Background

4.  Data Sources

5.  Methodology

6.  Results

7.  Conclusion

8.  Suggestions

Introduction

For my term project I have chosen to re-map the Logan River flood plain within Logan City’s Urban areas.  I have fished this river many times, and have always enjoyed it’s scenery.  I also enjoy looking at hydrological aspects of rivers such as flood potential, sources of water, and flow regimes.  This is why I chose this river for my study area.  The only current maps of the flood plain were created by FEMA in March of 1984.  With their map it only shows roads and blobs of color where flooding should occur.  I found this to be hard to interpret because of the lack of natural detail.  Also, in their report they did not mention and damage estimates if floods did occur.  So in my project I wanted to create a detailed map of the flood plain that an average person could read easily, and be able to locate there house.  Moreover, I also wanted to give some estimate of monetary cost in damages created by flooding.

Objectives

My objectives are relatively simple:

1.  Gather GIS data to create a map of Logan City’s flood plain along the Logan River.

2.  Gather hydrological data to extrapolate flood magnitude (plus probability of occurrence) and elevation of water surface during a flood event.

3.  Conduct a structure count and give an estimate of flood damages.

4.  Post easy to read map on my website that people can easily find out if their house is in risk of being flooded.

Background

The Logan River is located in the Bear River Basin, and in the Great Salt Lake Sub Basin.  The Drainage area of the river is 524 square miles at the Mendon road site, and 214 square miles at the mouth of Logan Canyon (where my water gage is also located).  The elevation of this river ranges from 9,000 ft. in the mountains to 4,500 ft in the valley.  The main source of runoff in the area is snow melt.  The average snow pack ranges from 6 ft. to 8 ft. in depth.  There are three dams located up the canyon to help mitigate flood risk.  There are two tributaries that enter the Logan river in my study reach.  The Blacksmith Fork River (peak flow of 1,900 cfs) and Spring Creek (peak flow of 1,980 cfs).  Which can contribute to higher flood potential in the lower portion of my study reach.  There are three different events that can create floods in my reach: 1. Snow melt runoff events  2.  Cloud burst events  3.  Rain on snow events which is the most likely to cause flooding. The Island area in Logan has made housing developements within the flood plain of the Logan River.

Data Sources

I gathered much of my GIS data from the Utah AGRC website.  I obtained a 10 meter DEM from this site as well as my Ortho images of Logan City with 1 ft. resolution.  In addition I also gathered my streams file from this website.  I also obtained a river reach file (to compare to the ARGC file) and an elevation dataset from the NHDPLUS website.  I got my hydrological flow data for my recurrence intervals and exceedence probabilities from the USGS website.  In addition I gathered other hydrological data from the Logan City website that provided longitudinal profile of the water surface elevations for different magnitudes of floods along the river, manning’s n values etc.  Moreover, I also walked the river to collect digital images to compare it’s geometry from the upstream to the downstream sections of my study reach.  However, I could not find cross sectional data for the river.

Methodology

I started by clipping my DEM to contain my region of interest in the attempt to reduce time in processing.  With this DEM I used Arc GIS applications to create a hillshade layer as to illustrate the change in relief within my study reach, and added my stream layer as well.  From the hillshade layer I then created a contour layer so I could reference my change in elevation easier when mapping (refer to Fig. 1).  With these layers done I then overlaid the ortho images of Logan City on the hillshade layer.  My images came in six panels that I then performed mosaics to combine them.  This gave me an 1 ft. resolution image of Logan City, with this I could now start to calculate the hydrological data to begin mapping. 

Now I got the peak flow data for gage # 10109000 from the USGS server.  I then used excel software to create the reoccurrence intervals of floods based on the historical data (refer to Table 1).  I did this by using the Weibull equation of    “ RI = (n+1) / M  ”, where n is the number of events, and M is the rank in descending order.  I needed the reoccurrence interval to determine the 100 year flood, with is the standard flood per which FEMA uses to map flood plains.  Also, with this data I created the an exceedence probability table and a graph (refer to Table 1 and Fig. 2).  Using this equation to obtain these values    “ P = 100*(M / (n+1)  ”.

 Now I had my 100 year flood value of 2,480 cfs.  FEMA reported a 100 year flood of 2,380 cfs. This was a lower value because they had less data when they derived their reoccurrence interval, and since then there has been some larger floods observed. I then used this value to get the water surface elevation from my longitudinal profile at different areas along the Logan River. The maps from FEMA had 19 benchmark point elevations with a description of the location.  There elevations were based off of the National Geodetic Vertical Datum of 1929.  FEMA also used a manning’s n of 0.033-0.045 for the main channel and 0.035-0.080 outside the channel in the COE HEC-2 program to make the water elevations.  I used these to make sure my elevation data correlated with their elevations so I could use the longitudinal water elevation profile FEMA made. I then would identify where the water level would reach outside of the river with the contours and the identify tool in Arc GIS.  When I found it I would mark it with a dot from the draw tool.  I did this all the way around my study reach.  When I completed all the dots, I then created a new shapefile polygon by connection the dots.  I created a main layer that is outline in red on my maps.  In addition, I also created another layer showing where minimal flooding would occur outlined in green. 

Once these polygons were created I then conducted a count of the structures located with in the boundaries.  I counted 182 structures within my flood plain.  I used this number to estimate the monetary damages caused by any flooding.

FIGURE 1.

TABLE 1. FLOW REOCCURRENCE % EXCEEDENCE

                             2,480 cfs                  100 yrs.                             0.9 %

                             2,100 cfs                    50 yrs.                             1.8 %

                             1,980 cfs                    20 yrs.                             4.5 %

                             1,720 cfs                    10 yrs.                             10  %

                             1,080 cfs                      2 yrs.                             50  % 

Figure 2.

Results

I created my maps into 7 high resolution panels to get greater detail when viewed (refer to maps below).  When creating my maps and walking through my study reach.  I also noticed that many structures  are built on mounded areas in the flood plain.  This means the actual flooding that would effect the houses is pretty minimal due to island like effects.  I did not clip out all the island areas in the flood plain because it would have taken a lot of time.  Almost all the areas would be flooded less than one foot, and in many areas just one inch of flooding would occur.  However, if a basement is present seepage could occur into the basement to cause damages.  Moreover, as you moved down stream the flood plain widen a lot because of the decrease in relief (flatter refer to Pic. 1 & 2 for differences in geometry upstream vs. downstream).

I made estimates of damage based on a one inch to one foot flood of $4,04,000.00 (for a one inch flood everywhere) to $10,010,000.00.  These estimates are base on a average home of 2,000 square foot home (which may be an underestimate) being flooded based on the Floodsmart.gov damage estimator.          

Uptsream Pic. 1

Downstream Pic. 2

Conclusion and Suggestions

I did also meet with the GIS department of Logan to see if they had done any GIS layers for the flood plain, but they had not.

With these maps I hope to provide an easy to read reference for those who live near the Logan River that they may be in a flood plain.  Also, I hope to provide some public awareness of how much damage flooding can do. With this study it is apparent that the upstream geometry of the river compared to the downstream geometry are extremely different. Moreover, the longitudinal profile is much steeper in the up stream than the downstream.  The upstream geometry is a lot deeper, while the downstream geometry is shallow and wide.  This makes faster waters upstream which means a larger volume of water can pass through with out flooding.  While when downstream lower increases in water elevation can create flooding.  In addition the Blacksmith Fork and Spring Creek enter in the lower end of the reach, which can increase the flooding significantly.  The average difference in water elevation from a 10 year flood to a 100 year flood as around 2 ft.

Suggestion, people that live near a river should probably buy some flood insurance policies!  Below is a sample of the old FEMA flood maps to compare with my maps above.