Due to population growth, urban areas in Oregon have been expanding, leading to increases in impervious surfaces and net losses in wetlands, riparian vegetation, and forestation in the Northwest. Utilizing ArcGIS and NOAA’s C-CAP imagery, this study classifies and analyzes urban land use changes between 1996 and 2010. These findings shed light on the importance of land use management in urban settings and are being used by local watershed councils to advocate for changes within their stream basins.

1. Modeling the Effects of Land
Use Change on Flooding
Adam Nayak, Cleveland High School

2. Research Question
How will changes in wetlands and developed land affect future flooding in
rivers?
Aim: To predict the effects of changes in land use on flooding to advocate for future
land restoration and implementation

3. Background: Urban Development
● Causes losses of wetlands, forested land, and riparian vegetation
● Leads to limitations in native habitat growth, water passage, and water quality
● Developed land is mainly impervious, preventing rainfall and snowmelt from
infiltrating the soil
● Since human settlement, about 57% of total wetlands area has been lost in the
Willamette River Basin

4. Background: Wetlands, Basins & Flooding
Wetlands & Basins
● Support biodiversity key for rivers and streams
● Cleanse polluted waters by retaining sediments and collecting excess
nutrients
● Offer a source of baseflow for streams in dry summer months
● Provide a floodplain for rivers during the wet season
Flooding
● Discharges affected directly by basin topography, land use, precipitation, and
temperature
● Urbanization has been shown to have major influence on stream discharge
due to its effects on land use and basin topography

5. Study Area
Urban Streams Near Portland, OR
● Johnson Creek
● Tualatin River
● Pudding River
● Clackamas River

6. Methods
● Part I: Historical Evaluation
● Part II: Land Use & Changes
● Part III: Predictive Modeling

7. Methods: Part I
Part I: Historical Evaluation
1. Investigate wetlands and developed lands historically in four stream basins in
the Portland-Metro Area between 1996 and 2010 (Johnson Creek, Tualatin
River, Pudding River, and Clackamas River).
2. Use historical flood records to investigate the relationship between land use
and flooding.

8. Methods: Part II
Part II: Land Use & Changes
1. Track changes in wetlands and developed areas in each watershed using Geographic
Information Systems (GIS) and NOAA Coastal Change Analysis Program (C-CAP) Data
1996-2010 (Johnson Creek, Tualatin River, Pudding River and Clackamas River).
a. Classify palustrine wetlands in terms of forested, shrub/scrub, and emergent.
b. Classify developed land in terms of high intensity, medium intensity, low intensity,
and developed open space.
c. Calculate total area of each wetland and developed land type in each basin for
each C-CAP evaluation.
d. Track change over time in wetland area and determine percentage change
between periods.
e. Use trendline regression to predict future change.
2. Use NOAA’s Impervious Surface Analysis Tool (ISAT) to evaluate changes in
imperviousness

9. Methods: Part III
Part III: Predictive Modeling
1. Using the Community Hydrologic Prediction System (CHPS) to Predict the Effects of Land Use Changes on
Flooding
a. Based on annual percent change, input new percentage impervious constants into the Sacramento
Soil Moisture Accounting (SAC-SMA) Model based upon ISAT analysis.
b. Investigate how flooding/discharges would change based upon changes in wetland and developed
land coverage using historical data to run the SAC-SMA Model and then CHPS.
2. Applying the GIS Flood Tool (GFT) for Flood Inundation Mapping
a. Obtain four raster datasets for desired flood mapping area: Raw & Conditioned Digital Elevation
Maps (DEM), Flow Directions, and Flow Accumulation.
b. Map different projected floods to visually represent discharge rates.
3. Suggesting Land Use Changes with Green Infrastructure
a. Investigate potential solutions in land use changes to counteract development
b. Contact local watershed councils and isolate areas for potential renovation based upon community
recommendations and site availability
c. Use land use and flood maps to prioritize locations for green infrastructure implementation

10. Model of Experimentation

11. Tools & Models
● NOAA Historical discharge/flood records
● ArcGIS Mapping Software
● Impervious Surface Analysis Tool:
○ NOAA ISAT Software
● Remotely Sensed Imagery
○ C-CAP-- 30m res
○ DEM-- 30m res
● Hydrologic Models
○ NWRFC SACSMA & CHPS
● Flood Inundation Mapping
○ USGS GIS Flood Tool

12. Part I: Historical Evaluation
Johnson Creek Evaluation:
● Milwaukie Gauge (MIWO3)
○ Percent Basin Developed: 62%
○ Percent Basin Wetlands: 1.1%
● Sycamore Gauge (SYCO3)
○ Percent Basin Developed: 36.5%
○ Percent Basin Wetlands: 1.9%
● Historical data in Johnson Creek suggests that basins with larger percentages of
wetlands and smaller percentages developed land experienced less severe flooding
● Data suggested a significant relationship between percent basin wetlands and flood
chance percentage across basins (p<0.001)
*Outlier due to a broken gauge during evaluation of the flood in 2009

13. Part II: Land Use & Changes
● Each basin investigated experienced significant changes in imperviousness between
1996 and 2010
● Overall areas of developed land increased in all basins, and wetlands area decreased
across all basins
● Due to overall small percentages of wetlands within basins investigated, losses in
wetlands usually did not change imperviousness significantly
● Data availability limited the time period investigated

14. Part II: Land Use & Changes Cont.
Regression Analysis
● Changes in land use from 1996 to 2010 were processed and analyzed
● Using regression analysis of 26 different land classifications within all four
basins, predictions were made for future land use changes for developed and
wetland areas
● These changes were applied to ISAT, then converted into new impervious
percentages for each basin
Land Use Summary: Tualatin River Basin

15. Part III: Predictive Modeling
Using CHPS to Predict the Effects of Land Use Changes on Flooding
● Using the imperviousness percentages calculated with ISAT, new flood responses were modeled with
prior flood data
● Outdated impervious percentages in CHPS and SAC-SMA Models were updated using new accurate
percentages calculated with GIS and ISAT in Part II
● With increases in imperviousness, discharges increased
● Discharges varied depending on the basin and specific flooding event
● Factors such as snowmelt and precipitation patterns during each flooding event proved to create
variance in discharge changes between flooding events
● The effects of climate change and increased precipitation were not accounted for in modeling, leading
to limitations in flood intensity projections

16. Part III: Predictive Modeling Cont.
Applying GFT for Flood Inundation Mapping
● 30m DEM resolution led to limitations in
flood inundation mapping
● Due to low resolutions, very little change
was able to be seen visually in flood
inundation maps, despite large changes in
discharge rates
● Additionally, low resolutions prevented
floods from being mapped in smaller
basins
● LiDAR DEM was able to be applied within
the Johnson Creek Watershed and
produced much more accurate results
Conditioned DEM
Pudding River
Pudding River Basin
Flood February 1996

17. Johnson Creek Prediction, LiDAR DEM
Potential Stormwater Site
Initial Flood 1996
Predicted Flood 2045
ESRI World Imagery
Created Using ArcGIS 10 and USGS GIS Flood Tool
Depicted to the left is the flood inundation map
for Johnson Creek created using LiDAR DEM and
USGS GIS Flood Tool at the Sycamore Gauge.
Changes in flooding between the initial 1996
flood and the 2045 prediction are shown in red
based upon land use change prediction analysis.

18. Part III: Predictive Modeling Cont.
Suggesting Land Use Changes with Green Infrastructure
● Large land use changes must be implemented to account for urban growth
● Wetlands restoration proved to require very large sums of land to counteract developed land
changes across all basins
● Data suggests that prevention of future flooding is best mitigated through implementation of pervious
area within impervious regions
● Data serves as strong support of green infrastructure and helps to quantify areas of change
necessary to counteract changes in development in each basin

19. Part III: Predictive Modeling Cont.
Suggesting Land Use Changes with Green Infrastructure Cont.
● GIS Mapping serves as strong resource for local watershed councils to isolate locations for
green infrastructure implementation
● Potential areas for green infrastructure (stormwater) implementation were isolated based upon
recommendations by the Johnson Creek Watershed Council and impervious surface data
● Flood inundation maps within the Johnson Creek Watershed are to be used to advocate for
green infrastructure implementation
Stormwater sites in the Johnson Creek Watershed

20. Isolating Areas for Green Infrastructure
Implementation
To the left, potential areas for green infrastructure
implementation (stormwater sites) have been isolated
based upon recommendations by the Johnson Creek
Watershed Council and impervious land use mapping.

21. Major Takeaways
● Remote sensing technology serves as a strong tool for land use analysis
● Restoration of wetlands would be expensive
● Implementation of green infrastructure could mediate wetland loss and urban
development
● Impervious surface maps are useful to prioritize areas for future restoration

22. Future Directions
● Using LiDAR DEM to map floods in other basins
● Using ArcMap to isolate open areas for wetland implementation
● Approaching other local watershed councils to advocate for land use changes
within various basins
● Advocating for green infrastructure implementation using flood inundation
maps

23. Acknowledgements
I would like to thank first my mentors that have guided me in my scientific research for the past few years:
Ronda Royal, Kate Fickas, and Andy Bryant. Additionally, I would like to thank those who directed me
through problems we encountered while working with the different software utilized in this study: Bill
Eslinger, NOAA (ISAT), James Verdin & Kristine Verdin, USGS (GFT), and James Woolard, NOAA (ASD Field
Spectroradiometer options). Lastly I would like to thank the members of the Johnson Creek Watershed
Council, in particular Katie Songer and Danielle Miles for their support. Without the direction of these
professionals, I could not have completed my work; thanks again to all of these amazing scientists.

24. References
● “Effects of Urbanization on Stream Ecosystems." Effects of Urbanization on Stream Ecosystems. USGS, n.d. Web. 23
June 2016. <http://water.usgs.gov/nawqa/urban/html/flowtempmethods.html>.
● Fickas, Kate C., Warren B. Cohen, and Zhiqiang Yang. "Landsat-based Monitoring of Annual Wetland Change in the
Willamette Valley of Oregon, USA from 1972 to 2012." Springer Science+Business Media Dordrecht 2015.
DOI:10.1007/s11273-015-9452-0.
● "Functions and Values of Wetlands | Washington State Department of Ecology." Functions and Values of Wetlands |
Washington State Department of Ecology. Washington State Department, n.d. Web. 22 June 2016.
<http://www.ecy.wa.gov/programs/sea/wetlands/functions.html>.
● "Hydrologic Cycle." MetEd. COMET, n.d. Web. 23 June 2016.
<https://www.meted.ucar.edu/hydro/basic/HydrologicCycle/sectionmenu.htm>.
● United States Department of the Interior: FIM. Washington: Dept. of the Interior, 2015. USGS. Web. 23 June 2016.
<http://water.usgs.gov/admin/memo/SW/sw2015.03.pdf>.
● "USGS Flood Inundation Mapper." USGS Flood Inundation Mapper. USGS, n.d. Web. 23 June
2016.<http://wimcloud.usgs.gov/apps/FIM/FloodInundationMapper.html#app=605a&585c-selectedIndex=0&39fb-selec
tedIndex=1>.
● "Water Encyclopedia." Stream Hydrology. N.p., n.d. Web. 23 June 2016.
<http://www.waterencyclopedia.com/St-Ts/Stream-Hydrology.html>.
● Wetlands: Protecting Life and Property from Flooding. Washington, D.C.: U.S. Environmental Protection Agency, Office
of Water, 2006. EPA.