Kang et al..fm - Kansas State University

3 downloads 1 Views 8MB Size Report
Binford, Chris Ennen, Mike Othitis, and Kate Lehmert for assisting with field work. .... May, C. W., Horner, R. R., Karr, J. R., Mar, B. W., and Welch, E. B. (2002) ...

DOWNSTREAM EFFECTS OF URBANIZATION ON STILLWATER CREEK, OKLAHOMA Ranbir S. Kang Department of Geography Western Illinois University Tillman Hall 312 Macomb, Illinois 61455 Daniel E. Storm Department of Biosystems and Agricultural Engineering Oklahoma State University Stillwater, Oklahoma 74075 Richard A. Marston Department of Geography Kansas State University Manhattan, Kansas 66506-2904

Abstract: Geomorphic effects of urbanization vary according to local conditions and with different ecoregions. This project evaluates the effects of urbanization on Stillwater Creek, located in central Oklahoma. The upper section of this basin is predominantly rural, while the downstream section is experiencing urban expansion. It was hypothesized that the channel morphology of the lower section would differ significantly from that of the upper section due to the location of the confluence of Boomer Creek, which brings urban runoff from the city of Stillwater. Statistical analysis of downstream trends revealed no significant change in the majority of response variables between upstream and downstream sections. However, local conditions (i.e,. riparian trees, cohesive bank materials, occasional woody debris jams, and entrenched nature) in this basin counter the possible effects of urbanization on channel morphology. Increasing urbanization was expected to reduce the sources of woody debris to stream channel and affect channel morphology. However, Stillwater Creek had thick riparian corridors dominated by trees, which helped protect stream banks. A downstream parabolic channel cross-sectional shape also helped explain why this stream channel did not change radically due to urbanization. [Key words: urbanization, channel morphology, fluvial geomorphology.]

INTRODUCTION Urbanization has transformed fluvial landscapes in different parts of the world (Wolman, 1967; Leopold, 1968; Arnold and Gibbons, 1996; Booth and Jackson, 1997; Chin, 2006; Urban et al., 2006; Keen-Zebert, 2007; O’Driscoll et al., 2009). The expansion of impervious surfaces, a commonly used measure of urbanization, reduces the infiltration capacity of land and leads to higher runoff compared to areas not affected by urbanization (Douglas, 1974; May et al., 2002; Li and Wang, 2009). Because water runs faster over impervious surfaces, construction reduces the 186 Physical Geography, 2010, 31, 2, pp. 186–201. DOI: 10.2747/0272-3646.31.2.186 Copyright © 2010 by Bellwether Publishing, Ltd. All rights reserved.

DOWNSTREAM EFFECTS OF URBANIZATION

187

Fig. 1. Study area in the Central Redbed Geomorphic Province of Oklahoma (produced from data provided by the U.S. Geological Survey and the U.S. Environmental Protection Agency).

lag time of surface runoff and increases debris production as well as flood peaks; this affects channel morphology in different ways, including alterations in channel cross-sections, types of bed materials, types of channel units, organic debris, and riparian vegetation (Orme and Bailey, 1971; Morisawa and Laflure, 1979; Nanson, 1981; Booth, 1990, 1991; Johnson, 2001; Jeje and Ikeazota, 2002; May et al., 2002; Avolio, 2003; Brierley and Fryirs, 2005; Gurnell et al., 2007). Charbonneau and Resh (1992) noticed that impacts of urbanization lead to enhanced downcutting, stream bank erosion, and modification of the natural pool-riffle sequence. Such effects of urbanization, however, vary locally with the degree of imperviousness (urbanization) and are determined by basin and adjacent riparian conditions (Kang and Marston, 2006; Marston, 2006). The effects of urbanization on channel morphology are not well understood (Hammer, 1972; Morisawa and Laflure, 1979; Booth, 1990, 1991; Arnold and Gibbons, 1996; Booth and Jackson, 1997; Trimble, 1997; Chin and Gregory, 2001; Chin, 2006; Kang and Marston, 2006). Such a lack of understanding is especially evident in the Central Redbed Plains geomorphic province of Oklahoma. This paper, part of a larger project (Kang and Marston, 2006) presents a detailed investigation of Stillwater Creek, located in the Central Redbed Plains geomorphic province, which is transforming from a rural to an urban basin with extensive impervious growth in the downstream section (Fig. 1). It also evaluates whether the channel response to urbanization conforms to that found in similar studies conducted in other regions. Therefore, it was anticipated that expansion of impervious surfaces in the lower section of Stillwater Creek would change the channel

188

KANG ET AL.

Fig. 2. Aerial photograph shows Stillwater Creek flowing through the urbanizing areas of Stillwater, Oklahoma (MRLC Consortium, 2001).

morphology of downstream reaches, as compared to the upstream section (Gregory and Park, 1976; Paul and Meyer, 2001). The confluence of Boomer Creek, a tributary of Stillwater Creek, was used to divide this river into two sections. The upper section is rural, whereas the lower section is urban and includes the City of Stillwater (Fig. 2). The objective of this research was to identify any differences in channel morphology of the downstream section as compared to upstream. If there was a significant difference between the two sections, could this difference be explained by urbanization? Finally, what factors other than urbanization might explain observed changes in Stillwater Creek? Six variables (channel width, mean depth, width-depth ratio, bankfull area, sinuosity, and gradient) were compared between the two sections. It was hypothesized that channel width, width depth ratio, bankfull area, and gradient are significantly greater downstream of Boomer Creek than upstream, as tested at the 0.05 level of significance. It also was hypothesized that mean depth and sinuosity are significantly less downstream as compared to upstream. STUDY AREA Stillwater Creek Basin is located in Payne, Noble, and Logan counties in central Oklahoma and has a drainage area of 733 km2. This basin is characterized by a humid subtropical climate and Red Permian shales and sandstones as main bedrock types, dominated by red iron oxides (Johnson, 1996). Lake Carl Blackwell, Lake McMurtry, and Boomer Lake are three reservoirs located in this basin. Two of these

DOWNSTREAM EFFECTS OF URBANIZATION

189

reservoirs are located upstream of the urban area of the City of Stillwater. The largest (14.2 km2) of these reservoirs is Lake Carl Blackwell, which was built in 1932 for recreation and as a secondary source of water for Oklahoma State University. Lake McMurtry (5.26 km2) was built for fishing, flood control, and recreation. Boomer Lake (1 km2 area), named after Boomer Creek, primarily serves for recreation and cooling an electricity plant. Stillwater Creek basin is an urbanizing basin that supports agricultural land (pasture, grassland, and crops) and expanding impervious area (4%). It is dominated by the Mollisol soil order (MRLC Consortium, 2001). A significant portion of the watershed includes grassland/pasture, followed by deciduous forest dominated by elm, oak, pecan, and cottonwood. In 1889, Stillwater was established in a fertile valley at the confluence of two streams now known as Boomer and Stillwater creeks (Bivert, 1988). What impressed the settlers the most was the fact that these two streams never ran dry and were surrounded by fertile land (Cunningham, 1979; Bivert, 1988). At that time, this basin was completely rural, with substantial area under cropping systems (Fitzpatrick et al., 1939; USDA, 1969). Since then, the population of Stillwater increased from 300 in 1890, to 5962 in 1920, and to 41,320 (estimated) in 2003 (U. S. Census Bureau, 2007). This increase in population served as the primary reason for the rural-to-urban transformation in this basin. The expanding campus of Oklahoma State University in Stillwater is another factor responsible for the increase of impervious surface area in this basin. Runoff generated from impervious areas enters Stillwater Creek through Boomer Creek. Therefore, the confluence of Boomer Creek makes a good dividing point for comparing upstream and downstream effects on the main channel. METHODS Field Data The main channel was divided into 30 reaches according to sinuosity and confluence of new tributaries. Channel cross-sections and riparian vegetation were measured at the beginning of each reach along Stillwater Creek. The channel cross-section measurements included channel width and depth at bankfull stage. Channel cross-section measurements also included identification of channel bank materials (by visual observation), the presence or absence of woody debris jams, and channel type according to the Rosgen Classification of Natural Rivers (Rosgen, 1996). The Rosgen classification provided a common language for describing the two sections of this river. Channel morphology data were used to calculate hydraulic variables, such as mean bankfull depth and bankfull area. Other stream variables, such as sinuosity and gradient, were calculated from U.S. Geological Survey digital elevation models (DEMs) using ArcGIS 9.3. An inventory of riparian vegetation also was prepared that consisted of a transect perpendicular to the channel, located near the beginning of each reach (Moore et al., 2002). These transects were 5 m wide and 30 m long, divided into three 10-meter long sections. Measurements within each riparian transect recorded area under grass, shrubs, percent canopy cover, and number

190

KANG ET AL.

of trees. Riparian surveys also were conducted at 30 sites along the stream. In addition, a small airplane was used to get a better view of contemporary land cover in the study area. Sub-basin Delineation and Land Cover Data Boundaries of various sub-basins contributing water to surveyed reaches were delineated using ArcView Soil and Water Assessment Tool (AVSWAT©), which is an ArcView extension and a graphical user interface for the Soil and Water Assessment Tool (SWAT©) (Luzio et al., 2002). The National Land Cover Dataset (NLCD) for the year 2001 was used to map and measure the area under impervious cover and other types of land cover, such as open water, pervious though developed, impervious cover, barren land, deciduous forest, grassland, pasture/hay, and cultivation for each sub-basin. Shapefiles of impervious areas from the City of Stillwater were used to validate the NLCD-generated impervious area estimates. Aerial photographs for 1973 and 2003 were compared to calculate change in total impervious area and riparian corridor with the help of raster calculator tool in ArcGIS. Statistical Analysis The statistical analysis was completed in two steps using an α = 0.05. Step one compared variables measured in upstream and downstream sections, and step two explained differences in any variables from upstream to downstream sections. Step one applied an ANCOVA (Analysis of Covariance) to compare channel morphology variables for the two sections. In such an analysis, differences in channel morphology may result from increasing runoff due to increasing drainage area downstream (Downs and Gregory, 2004). To avoid that problem, channel morphology variables were normalized based on drainage area by using drainage area as the covariate in the ANCOVA test. Since ANCOVA is a parametric test based upon an assumption of normality, eight levels of transformation (original units, square root, cube root, logarithm, reciprocal root, reciprocal, cube, and square) were used for each geomorphic variable to select the most normal level as suggested by Helsel and Hirsch (2002). Table 1 shows the transformation selected for each variable. Step two developed a multiple linear regression model for each variable that differed from upstream to downstream sections. The backward elimination method was used in developing these regression models (Helsel and Hirsch, 2002). RESULTS Based on the field surveys, glide appeared to be the main channel unit type and bank materials were predominantly fine-grained (Table 2). The channel bed and bank materials did not change over the entire length of Stillwater Creek; they consisted of 95–100% silt-clay throughout. The shape of channel cross-sections was parabolic with low gradient. At the same time, there was no difference in the bedrock from upstream to downstream. According to the Rosgen Classification, Stillwater Creek was classified as an E6b channel, which is a very stable channel type with

DOWNSTREAM EFFECTS OF URBANIZATION

191

Table 1. Transformations Selected for Comparing Upstream and Downstream Channel Sections of Stillwater Creek Variable Width

Transformation used in ANCOVA Natural log

Mean depth

Reciprocal root

Width–depth ratio

Reciprocal

Bankfull area

Reciprocal

Sinuosity

Reciprocal root

Gradient

Square root

Drainage area

Square root

slight entrenchment (Rosgen, 1996). Field observations also revealed an occasional woody debris jam dominated by large trees. Stillwater Creek was characterized as having a thick riparian corridor primarily bordered by agricultural fields. Each riparian transect was dominated by trees with a thick canopy, followed by grass and shrubs (Table 3). Major riparian tree species included American elm (Ulmus americana), cottonwoods (Populus sp.), and green ash (Fraxinus pennsylvanica). Field work also revealed the presence of old trees with trunks over 100 cm in the riparian corridor. Most of the riparian corridor was more than 30 m wide. A study completed by Cross (1950) found a thick riparian corridor dominated by elm, oak, pecan, and cottonwood in Stillwater Creek basin. The comparison of aerial photographs for 1973 and 2003 revealed minor changes in the riparian corridor. Therefore, the riparian corridor in this basin has not experienced any significant change during last few decades. In the case of land cover, grassland covered more basin area (55.5%) than any other land cover, followed by deciduous forest (22.2%) in this basin (Fig. 3, Table 4). During the 1979–2003 period, impervious area increased by 65%. The expanding campus of Oklahoma State University was a major factor in such increase. Currently, 3.9% of the total watershed area is under impervious cover, most within the city limits of Stillwater. The impervious area outside the city limits was primarily road network. The upstream section of Stillwater Creek Basin was rural, with most of the area covered by grassland, deciduous forest, cultivated land, and pasture. Based on the ANCOVA results (Table 5), width, bankfull area, sinuosity, and gradient did not differ significantly between the downstream section of Stillwater Creek and the upstream section. Therefore, the null hypotheses for these variables were not rejected. The two variables that exhibited significant differences were mean depth and width–depth ratio. The hypothesis for decreasing mean depth from upstream to downstream was based on the argument that the process of urbanization would increase sediment production and aggrade the channel, leading to a decrease in mean depth. This anticipated change in mean depth was the primary

192

KANG ET AL.

Table 2. Watershed Characteristics for Stillwater Creek Reach Areaa Channel Width number (km2) unit (m)

Mean Width– depth depth Bankfull Rosgen Bank ratio area (m2) Sinuosity Gradient type materialb (m)

1

197.3

Glide

29.6

1.1

32.9

1.3

0.1

C6b

100

2

201.2

Glide

11.2

1.4

7.96

15.6

1.2

0.1

E6b

100

3

221.8

Glide

12.2

2.1

5.9

25.3

1.6

0.1

E6b

100

4

325.4

Glide

65.4

27.6

2.4

1802.0

1.1

0.1

E6b

100

5

330.5

Glide

17.3

4.21

4.1

72.9

1.5

0.0

E6b

100

6

332.1

Glide

21.5

2.2

9.7

48.0

1.3

0.1

E6b

100

7

335.1

Glide

18.8

2.8

6.6

53.0

1.0

0.0

E6b

100

8

337.4

Glide

12.2

2.4

5.0

29.8

1.2

0.0

E6b

100

9

338.3

Glide

64.8

30.9

2.1

2003.0

1.3

0.1

E6b

100

10

340.8

Riffle

15.8

2.4

6.6

37.8

1.9

0.1

E6b

100

11

375.1

Glide

64.0

26.1

2.5

1670.0

1.3

0.1

E6b

100

12

392.1

Glide

17.2

1.7

9.9

30.0

1.1

0.1

E6b

100

13

399.7

Glide

20.2

8.5

2.4

172.0

1.0

0.1

E6b

100

14

418.0

Riffle

8.3

3.3

2.5

27.5

1.5

0.1

E6b

95

15

420.9

Glide

19.9

3.5

5.7

69.8

1.0

0.0

E6b

100

16

423.0

Glide

65.0

27.3

2.4

1772.0

1.1

0.1

E6b

100

17

424.7

Glide

63.8

23.8

2.7

1519.0

1.3

0.1

E6b

100

18

473.6

Glide

18.0

2.0

8.9

36.0

1.6

0.1

E6b

100

19

475.5

Run

13.8

3.1

4.4

43.1

1.2

0.1

E6b

100

20

563.3

Glide

47.1

7.2

6.6

338.0

1.5

0.0

E6b

100

21

570.6

Glide

31.4

5.0

6.3

156.0

1.2

0.1

E6b

100

22

577.1

Glide

63.7

32.5

2.0

2070.0

1.2

0.1

E6b

100

23

586.3

Glide

22.4

3.3

6.9

73.0

1.4

0.0

E6b

100

24

591.0

Run

18.2

6.6

2.8

119.0

1.0

0.0

E6b

100

25

637.1

Glide

8.2

2.9

2.8

24.0

1.4

0.0

E6b

100

26

637.1

Riffle

33.0

11.1

3.0

366.0

1.2

0.0

E6b

100

27

706.0

Glide

31.0

10.2

3.0

316.0

1.0

0.0

E6b

100

28

706.5

Glide

33.0

10.3

3.2

338.0

1.0

0.0

E6b

100

29

732.8

Glide

44.8

13.1

3.4

587.0

1.2

0.0

E6b

100

30

733.1

Glide

34.2

11.0

3.11

378.0

1.1

0.1

E6b

100

a

Watershed area upstream of transect. Silt + clay %.

b

26.9

193

DOWNSTREAM EFFECTS OF URBANIZATION

Table 3. Riparian Vegetation Characteristics for Stillwater Creek Reach number

Shruba (%)

Grassb (%)

Canopy coverc (%)

Number of trees in riparian transect

1

0

30

23

24

2

6

56

26

32

3

16

16

30

10

4

0

53

53

7

5

3

60

33

13

6

16

73

46

19

7

0

96

40

23

8

5

25

60

34

9

33

26

53

3

10

1

98

36

12

11

0

0

0

0

12

26

36

90

40

13

15

28

60

21

14

16

56

26

13

15

38

45

66

35

16

13

53

40

3

17

0

80

40

11

18

10

43

70

13

19

3

90

60

9

20

20

60

53

3

21

26

36

73

8

22

26

56

26

6

23

0

90

93

15

24

3

90

30

6

25

0

93

63

9

26

13

56

33

3

27

0

13

0

2

28

3

56

6

2

29

41

83

26

6

30

16

56

46

20

a

Percent area of riparian transect under shrub. Percent area of riparian transect under grass. c Percent area of riparian transect under canopy cover. b

reason for hypothesizing the increasing width–depth ratio. The null hypotheses for these two variables were rejected—mean depth and width–depth ratio both differ significantly between upstream and downstream sections of Stillwater Creek. Differences in mean depth and width–depth ratio were explained with the help of multiple linear regression (Table 6). In case of mean depth, R2 was low (0.51), which reflected unexplained variance (Table 6). However, R2 was higher (0.61) in

194

KANG ET AL.

Fig. 3. Land cover in Stillwater Creek Basin (produced from data provided by the U.S. Geological Survey).

Table 4. Land Cover in Stillwater Creek Basin Derived from the 2001 National Land Cover Data Land cover type

Area (%)

Open water

3.1

Pervious though developed

6.4

Impervious

3.9

Deciduous forest

22.2

Grassland/herbaceous

55.5

Pasture /hay

2.6

Cultivated

6.3

Source: MRLC Consortium, 2001.

the case of width–depth ratio. Upstream-to-downstream differences in these variables were not completely explained by urbanization alone. The presence of riparian trees, and of deciduous forest, in this basin were two other factors that may have contributed to this trend. According to regression models, these trends were due to multiple factors, such as urbanization along with riparian trees and deciduous forest in this basin. Research in other basins has yielded similar results (Leopold, 1972; Hollis, 1976;

195

DOWNSTREAM EFFECTS OF URBANIZATION

Table 5. Upstream and Downstream Comparison of Stillwater Creek

Channel variable compared

Null hypothesis

Actual result

Status of null hypothesis (α = 0.05)

Width (natural log)

No difference

No difference

Not rejected

0.29

Mean depth (reciprocal root)

No difference

Decrease

Rejected

0.01

Width–depth ratio (reciprocal)

No difference

Increase

Rejected

0.03

Bankfull area (reciprocal)

No difference

No difference

Not rejected

0.54

Sinuosity (reciprocal root)

No difference

No difference

Not rejected

0.16

Gradient (square root)

No difference

No difference

Not rejected

0.07

p-value

Table 6. Multiple Linear Regression Models Used to Explain Change in Mean Depth and Width–Depth Ratio from Upstream to Downstream Channel Sections in Stillwater Creek Dependent variable, Y Independent variable, Xi

Coefficient

Mean depth (m)

Width–depth ratio 549

b0

0.50

Impervious areaa

b1

–0.000125

Area under deciduous foresta

b2

1.78

Area under grassland/herbaceousa

b3

b

Riparian trees

a

–134

b4

0.080

17.4

R2

0.51

0.61

Percent of total area. Number of trees in riparian transects.

b

Nanson, 1981; Montgomery, 1997; Booth and Henshaw, 2001; Hession et al., 2002). As an urbanizing basin with active construction, the downstream section of Stillwater Creek was characterized by substantial sediment production and runoff, while imperviousness did not appear as a significant factor explaining differences in channel morphology from upstream to downstream. As one moves downstream from the confluence with Boomer Creek, the tributary that delivered urban runoff and sediment, none of the downstream factors showed a statistically significant change that could be attributed to urbanization alone. However, local conditions, such as riparian trees, deciduous forest, and cohesive bank materials, provided possible explanations for the lack of difference in the majority of the response variables.

196

KANG ET AL.

Fig. 4. Images of upstream and downstream sections of Stillwater Creek showing similar riparian vegetation and geomorphic characteristics.

DISCUSSION Despite a 65% increase in the impervious area in 24 years (1979–2003), the majority of variables do not show any significant change in the downstream section of Stillwater Creek. Mean depth and width–depth ratio are the only variables showing significant differences between the upstream and downstream sections. As revealed by multiple linear regression, differences in these variables can be attributed to a combination of land cover types, which rule out imperviousness as a major explanatory variable. The presence of riparian trees (Fig. 4) can provide some explanation for differences in mean depth and width-depth ratio. Tree density, in combination with cohesive bed and bank material, may have helped stabilize the banks against erosion and provided woody debris for trapping and depositing sediments, thus leading to a decrease in mean channel depth downstream. Informal discussions with Mr. Bud Payne, who spent most of his life in Stillwater, revealed that the riparian corridor and channel geometry of Stillwater Creek have not changed substantially in the last five decades, despite urban growth. The intact stream banks of Boomer Creek, the main tributary bringing urban runoff into Stillwater Creek, showed no structural change after the severe flooding of summer 2007 (Fig. 5). Such geomorphic characteristics support the decisive role of local conditions, such as riparian trees, cohesive bank materials, occasional woody debris jams, and entrenchment (Montgomery, 1999; Kang and Marston, 2006). Although Fryirs and

DOWNSTREAM EFFECTS OF URBANIZATION

197

Fig. 5. A thick riparian corridor dominated by trees on the banks of Boomer Creek helps protect the stream bank from erosion.

Brierley (2000) suggested that urbanization leads to irreversible alterations in stream channels, such alterations were not evident in this stream. Overlaid on the sandstone bedrock with some shale, Stillwater Creek is conveying runoff and sediment without any notable change in channel morphology; this type of channel response is unlike the findings of Hession et al. (2002), Pizzuto et al. (2000), Trimble (1997), or Fryirs and Brierley (2000). Another reason this channel has not responded dramatically to urbanization could be related to the parabolic channel cross-sections of the downstream section. A parabolic cross-section has been shown to be the equilibrium shape based on threshold theory (Stevens, 1989), models of lateral diffusion (Parker, 1978), minimum stream power (Chang, 1980), and minimum variance (Langbein, 1965). This stream experienced entrenchment during the early 20th century for reasons other than urbanization. At present, the entrenched parabolic cross-sections carved into the cohesive shales and clay and combined with the soil-binding effect of streamside vegetation, appear insensitive to the hydrologic and sediment impacts from urbanization. These results lay a foundation for understanding this unique geomorphic behavior. These findings also present a solid base for future research to develop generalizations about the geomorphic response of urbanizing streams in the Central Redbed Plains of Oklahoma.

198

KANG ET AL.

CONCLUSIONS Local conditions play a decisive role in countering the effects of urbanization in this basin. The soil-binding effect of streamside vegetation, banks comprised of cohesive clays, and a parabolic cross section have combined to create a channel morphology (as measured by width, gradient, bankfull area, and sinuosity) that is stable despite the increased runoff and sediment supplied by the Stillwater urban area. The finding that channel morphology does not change downstream from the Stillwater urban area reminds us that place matters when understanding the impacts of urbanization on stream channels. One must be mindful of the resisting framework as well as the driving forces when analyzing urban impacts on streams. In the Central Redbed Plains of Oklahoma, streams are able to counter the impacts of increased runoff and sediment due to urbanization. Contrary to the findings of Fryirs and Brierley (2000), urbanization within in the Stillwater Creek basin has not led to anticipated dramatic changes in the geomorphic system. These findings are consistent with Montgomery (1999) and Kang and Marston (2006), who argued that local conditions must be considered in any such analysis. Klein (1979) argued various measures to limit the adverse effects of urbanization on streams; however, in this geomorphic province, the decisive role of local conditions in countering such effects of urbanization advocates the place dependency of such measures. This research offers a unique, detailed data set in the south-central United States. The observed site-specific geomorphic response of Stillwater Creek to imperviousness can provide guidance in devising river management practices in this geomorphic province. Acknowledgements: The authors sincerely thank Dr. Carol Harden, Dr. John C. Dixon, and anonymous reviewers for providing constructive comments on this manuscript. We also thank Brandon Binford, Chris Ennen, Mike Othitis, and Kate Lehmert for assisting with field work. Earlier review of this manuscript by Dr. Thomas Foggin is sincerely appreciated.

REFERENCES Arnold, C. L. J. and Gibbons, C. J. (1996) Impervious surface coverage: The emergence of a key environmental indicator. Journal of American Planning Association, Vol. 62, 243–258. Avolio, C. M. (2003) Local Impacts of Road Crossings on Puget Lowland Creeks. Unpublished master's thesis, Department of Civil and Environmental Engineering, University of Washington, Seattle. Bivert, R.E. (1988) My Town Stillwater: A Quick Fact Review of Stillwater, Oklahoma on Its 100th Anniversary. Stillwater, OK: Copy & Art. Booth, D. B. (1990) Stream channel incision following drainage basin urbanization. Water Resources Bulletin, Vol. 26, 407–417. Booth, D. B. (1991) Urbanization and the natural drainage system: Impacts, solutions, and prognoses. The Northwest Environmental Journal, Vol. 7, 93–118. Booth, D. B. and Henshaw, P. C. (2001) Rates of channel erosion in small urban streams. In S. W. Mark and J. B. Stephen, eds., Landuse and Watersheds: Human

DOWNSTREAM EFFECTS OF URBANIZATION

199

Influence on Hydrology and Geomorphology in Urban and Forest Areas. Washington, DC: American Geophysical Union, 17–38. Booth, D. B. and Jackson, C.R. (1997) Urbanization of aquatic systems: Degradation thresholds, stormwater detention, and the limits of mitigation. Journal of the American Water Resources Association, Vol. 33, 1077–1090. Brierley, G. J. and Fryirs, K. A. (2005). Geomorphology and River Management. Malden, MA: Blackwell. Chang, H. H. (1980) Geometry of gravel streams. Journal of Hydraulics Division of American Society of Civil Engineers, Vol. 106, 1443–1456. Charbonneau, R. and Resh, V. H. (1992) Strawberry Creek on the University of California, Berkeley Campus: A case history of urban stream restoration. Aquatic Conservation: Marine and Freshwater Ecosystems, Vol. 2, 292–307. Chin, A. (2006) Urban transformation of river landscapes in global context. Geomorphology, Vol. 79, 460–487. Chin, A. and Gregory, K. J. (2001) Urbanization and adjustment of ephemeral stream channels. Annals of the Association of American Geographers, Vol. 91, 595–608. Cross, F. B. (1950) Effects of sewage and of a headwaters impoundment on the fishes of Stillwater Creek in Payne County, Oklahoma. American Midland Naturalist, Vol. 43, 128–145. Cunningham, R. E. (1979). Stillwater: Where Oklahoma Began. Stillwater, OK: Evans Publications. Douglas, I. (1974) The impact of urbanization on river systems. In Proceedings of the International Geographical Union Regional Conference and Eighth New Zealand Geography Conference, 307–317. Downs, P. W. and Gregory, K. J. (2004). River Channel Management: Towards Sustainable Catchment Hydrosytems. London, UK: Arnold. Fitzpatrick, E. G., Boatright, W. C., and Rose, L. E. (1939). Soil Survey: Garfield Country, Oklahoma. Washington, DC: U.S. Department of Agriculture. Fryirs, K. and Brierley, G. (2000) A geomorphic approach to the identification of river recovery potential. Physical Geography, Vol. 21, 244–277. Gregory, K. J. and Park, C. C. (1976) Stream channel morphology in northwest Yorkshire. Revue de Geomorphologie Dynamique, Vol. 25, 63–72. Gurnell, A., Lee, M., and Souch, C. (2007) Urban rivers: Hydrology, geomorphology, ecology, and opportunities for change. Geography Compass, Vol. 1, 1118– 1137. Hammer, T. A. (1972) Stream channel enlargement due to urbanization. Water Resources Research, Vol. 8, 1530–1540. Helsel, D. R. and Hirsch, R. M. (2002). Statistical Methods in Water Resources. Reston, VA: U.S. Geological Survey. Hession, W. C., Pizzuto, J. E., Johnson, T. E., and Horwitz, R. J. (2002) Influence of bank vegetation on channel morphology in rural and urban watersheds. Geology, Vol. 31, 147–150. Hollis, G. E. (1976) The response of natural river channels to urbanization: Two case studies in southeast England. Journal of Hydrology, Vol. 30, 351–363.

200

KANG ET AL.

Jeje, L. K. and Ikeazota, S. I. (2002) Effects of urbanization on channel morphology: The case of Ekulu River in Enugu, southeastern Nigeria. Singapore Journal of Tropical Geography, Vol. 23, 37–51. Johnson, K. S. (1996) Geology of Oklahoma. In K. S. Johnson and N. H. Suneson, eds., Rockhounding and Earth-Science Activities in Oklahoma, 1995 Workshop. Oklahoma City, OK: Oklahoma Geological Society, 1–9. Johnson, M. P. (2001) Environmental impacts of urban sprawl: A survey of the literature and proposed research agenda. Environment and Planning A, Vol. 33, 717– 735. Kang, R. S. and Marston, R. A. (2006) Geomorphic effects of rural-to-urban land use conversion on three streams in the Central Redbed Plains of Oklahoma. Geomorphology, Vol. 79, 488–506. Keen-Zebert, A. (2007) Channel response to urbanization: Scull and Mud creeks in Fayetteville, Arkansas. Physical Geography, Vol. 28, 249–260. Klein, R. D. (1979) Urbanization and stream quality impairment. Water Resources Bulletin, Vol. 15, 948–963. Langbein, W. B. (1965) Geometry of river channels: closure of discussion. Journal of the Hydraulics Division of American Society of Civil Engineers, Vol. 91, 297– 313. Leopold, L. B. (1968) Hydrology For Urban Land Planning—A Guidebook on the Hydrologic Effects of Urban Land Use. Washington, DC: U. S. Department of Interior, U.S. Geological Survey Circular 554. Leopold, L. B. (1972) River channel change with time: An example. Geological Society of America Bulletin, Vol. 84, 845–860. Li, Y. and Wang, C. (2009) Impacts of urbanization on surface runoff of the Dardenne Creek watershed, St. Charles County, Missouri. Physical Geography, Vol. 30, 556–573. Luzio, M. D., Srinivasan, R., Arnold, J. D., and Neitsch, S. L. (2002). ArcView Interface for SWAT 2000: User’s Guide. College Station, TX: Texas Water Resources Institute. Marston, R. A. (2006) Ecoregions: A geographic advantage in studying environmental change. AAG Newsletter, Vol. 41, 6. May, C. W., Horner, R. R., Karr, J. R., Mar, B. W., and Welch, E. B. (2002) Effects of urbanization on small streams in the Puget Sound ecoregion. In T. R. Schueler and H. K. Holland, eds., The Practice of Watershed Protection. Ellicott City, MD: Center for Watershed Protection, 87–98. Montgomery, D. R. (1997) What’s best on the banks? Nature, Vol. 388, 328–329. Montgomery, D. R. (1999) Process domains and the river continuum. Journal of the American Water Resources Association, Vol. 35, 397–410. Moore, K., Jones, K., and Dambancher, J. (2002) Aquatic Inventories Project, Natural Production Program. Corvallis, OR: Oregon Department of Fish and Wildlife. Morisawa, M. and Laflure, E. (1979) Hydraulic geometry, stream equilibrium and urbanization. In D. D. Rhodes and G. P. Williams, eds., Adjustments of the Fluvial System. Dubuque, IA, USA: Kendall/Hunt, 333–350. MRLC Consortium (2001). National Land Cover Dataset. Reston, VA: U.S. Geological Survey.

DOWNSTREAM EFFECTS OF URBANIZATION

201

Nanson, G. C. (1981) Downstream reduction of rural channel size with contrasting urban effects in small coastal streams southeastern Australia. Journal of Hydrology, Vol. 52, 230–255. O’Driscoll, M. A., Soban, J. R., and Lecce, S. A. (2009) Stream channel enlargement response to urban land cover in small coastal plain watersheds, North Carolina. Physical Geography, Vol. 30, 528–555. Orme, A. R. and Bailey, R. G. (1971) Vegetation conversion and channel geometry in Monroe Canyon, Southern California. Yearbook of the Association of Pacific Coast Geographers, Vol. 33, 65–82. Parker, G. (1978) Self-formed straight rivers with equilibrium banks and mobile bed. Journal of Fluid Mechanics, Vol. 89, 127–146. Paul, M. J. and Meyer, J. L. (2001) Streams in the urban landscape. Annual Review of Ecology, Vol. 32, 333–365. Pizzuto, J. E., Hession, W. C., and McBride, M. (2000) Comparing gravel bed rivers in paired urban and rural catchments of southeastern Pennsylvania. Geology, Vol. 28, 79–82. Rosgen, D. (1996). Applied River Morphology. Pagosa Springs, CO: Wildland Hydrology. Stevens, M. A. (1989) Width of straight alluvial channels. Journal of Hydraulic Engineering, Vol. 115, 309–326. Trimble, S. W. (1997) Contribution of stream channel erosion to sediment yield from an urbanizing watershed. Science, Vol. 278, 1442–1444. U. S. Census Bureau (2007). American FactFinder Dataset. Suitland, MD: U.S. Census Bureau. Urban, M. C., Skelly, D. K., Burchsted, D., Price, W., and Lowry, S. (2006) Stream communities across a rural-urban landscape gradient. Diversity and Distributions, Vol. 12, 1–14. USDA (1969). Soil Survey: Oklahoma County, Oklahoma. Washington, DC: U.S. Department of Agriculture. Wolman, M. G. (1967) A cycle of sedimentation and erosion in urban river channels. Geografiska Annaler Series A, Vol. 49A, 385–395.