1. watershed analysis of Oak Ridge
By: Nick Sisco

2. Objectives Delineate streams and tributaries Delineate watersheds
Create streamlined process for future analysis
Map outfalls remotely to compare with GPS data
Compare watershed boundaries

3. Work Flow Obtain and create DEM
Flow Direction
Fill
Sink
Flow Accumulation
Threshold
Stream Order
Stream polylines
Vertices to points
Watershed
Raster to polygons
Create Model and Instructions for future analysis
Compare Outfalls with GPS field data
Compare watershed boundaries with existing data

4. Watershed Analysis
Using the ArcGIS Hydrology toolset

5. Digital elevation Models
Roane County
Anderson County
USGS quarter quadrangle DEMs
10ft resolution
Mosaic To New Raster
LiDAR points converted into DEMs
Natural Neighbor
Focal Statistics
Con(IsNull("myraster"),FocalStatistics("myraster",NbrRectangle( 2,2),"MEAN"),"myr­aster")
1ft resolution
Mosaic to New Raster

6. Resolution Differences 1:8,000 Aerial reference
Aerial reference for next 5 slides.

7. Resolution Differences 1:8,000
Roane County
Anderson County
Comparison of pixel resolution of flow direction rasters.

8. Roane County qquad dataset. Flow direction raster. 10x10ft resolution.

9. Anderson County LiDAR set. Flow direction raster 1x1ft resolution.

10. Roane qquad dataset DEM with Hillshade

11. Anderson county LiDAR set. DEM with hillshade.

12. Flow Direction
Calculates the direction of flow by assigning separate cell values for each direction.
1 – east
2 – southeast
4 – south
8 – southwest
16 – west
32 – northwest
64 – north
128 – northeast

13. Flow Direction Roane County Anderson County
Also denotes spatial extent of the two datasets

14. Flow Accumulation
Calculates the theoretical accumulation each cell would accrue
A threshold is required to delineate streams and to determine level of tributary detail
The Con tool is used to determine threshold by using a conditional statement
“Value <= (threshold #)” is null

15. Threshold Comparison
Value <= 15
Value <= 200

16. Stream Order Strahler Method of stream ordering
Distinguishes tributaries from streams and rivers
The number of orders depends upon threshold value and the stream/tributary detail. The Anderson County LiDAR dataset has more stream orders than the Roane county. More stream orders means smaller watershed delineations.

17. Feature Vertices to points
The line end-points are converted to a point shapefile
These points will act as watershed outlets, pour points, or stream convergence points
These outfalls will act as the exit points for water for each watershed.

18. Watershed delineation
Ridgelines are delineated using the flow direction raster and pour point shapefile
The watershed raster is converted polygons

19. Automating the analysis
Model building for future analysis

20. Building the models Two models were built to expedite analysis process
These models can be used for future analysis with updated DEMs
The first model completes the entire analysis from DEM to watershed polygons
The second allows the user to experiment with different threshold values

21. Watershed analysis model

22. Set null to watershed model

23. Comparing the data
Field data comparison and analyzing results

24. Data clean up
Some streams were delineated in the river, which need to be removed.

25. Data Clean Up

26. Errors Actual Outfall Errors
Of the 7,166 stream vertices, 1,527 intersected the river/waterbody shapefile.
571 points were deemed erroneous and removed.
Leaving 956 potential river outfalls in the Anderson County side of Oak Ridge.

27. Data Comparison Blue Points: GPS Outfall
Orange Points: River Outfall Via Analysis
Purple Points: Stream Convergences

28. Data Comparison
With river outfalls and stream convergences mapped and compared to observed data…
GPS data acquisition can be planned
Blue lines can be reassessed for accuracy
Outfalls can be reported to the state in compliance with EPA NPDES Phase II MS4

29. Watershed comparison
Allows for more detailed watershed delineation

30. Future Considerations
New Lidar Implementation

31. Terrain Dataset Benefits to Terrain Datasets
Can store large quantities of data
Break lines can better delineate road cuts, curbs and shorelines
Higher accuracy and precision of surface modelling
Multi-resolution data display and interpolation
Potential for 3D Analysis