A review of changes in the geomorphology and hydrology of Waller Creek (Austin, Texas) as a result of urban development.
Waller Creek drains a small basin in central Texas, and is a part of the larger drainage basin for the Colorado River. The basin for Waller Creek lies entirely within the city limits of Austin, Texas. In most places, Waller Creek cuts into the Cretaceous Austin Chalk, which is covered by thin clays and gravel terrace deposits. Downstream, near the Colorado River, Waller Creek is entrenched into these terrace deposits. According to Garner and Woodruff (1979), the creek is a typical urban stream because it is polluted, it contains considerable debris, and its discharge is highly variable. The climate of the Austin area is generally mild, although sudden temperature changes are common. Mean annual precipitation at the U.S. Weather Bureau station, approximately 1 1/2 miles east of the watershed, is 32.58 inches. Rainfall is generally well-distributed throughout the year; however, individual storms may cause serious flooding in any season. In general, the largest storms occur during the months of April-May and September-October (Twichell, 1967). These intense, irregular floods are usually associated with tropical storms from the Gulf of Mexico (Slade, 1986). As a result of these floods, the city of Austin has established a number of parks along the Waller Creek floodplain.
There are two streamflow gauging stations, formerly operated by the U.S. Geological Survey, along Waller Creek. These stations are now operated by the hydrogeology students at the University of Texas at Austin. One station is at 23rd Street, and the other is at 38th Street. Since at least the early 1960's the area around 23rd Street has experienced extensive urban development and channel improvements (Espey et al., 1965). The area around 38th Street, however, is mostly residential and appears relatively undeveloped when compared to that near 23rd Street. According to the U.S. Geological Survey (1981), the 23rd Street station gauges an area of 4.13 square miles (10.70 square kilometers), whereas the 38th Street station gauges an area of 2.31 square miles (5.98 square kilometers). An analysis of changing flow data from these two stations, together with information on geomorphic changes, should show how Waller Creek has responded to increased urbanization.
Murphey et al. (1977) studied the geomorphology of small, semiarid basins (up to 60 square miles) of southwestern North America. They concluded that for ephemeral streams, the watershed area is the most useful parameter for predicting the hydraulic response to precipitation. They also found that "the average rise time and duration of flow decreased as drainage density increased and as the slope of the main channel increased ... respectively." In another study, Shepherd (1979) related bedrock lithology to stream geomorphology in central Texas. He concluded that "channels dominated by limestone [such as the Austin Chalk] or schist bedrock characteristically have the thinnest, worst-sorted, and coarsest channel-bed sediments, the lowest width-depth ratios, and pool-riffle thalweg patterns." Unfortunately, neither Shepherd nor Murphey et al. examined the effects of urban development on the streams. They only considered some of the natural parameters that affect stream geomorphology.
A study of the impact of urbanization on stream geomorphology in a suburb of Denver, Colorado, was conducted by William L. Graf. He concluded that the suburban development caused expansion of the stream floodplains and subsequent downcutting of the stream channels. He found that "the rapid suburbanization had introduced so much sediment into the channel system that the streams could not carry the material (mostly sand) and so deposited it as floodplain alluvium. New areas of floodplain were created, and old areas were enlarged, primarily by vertical accretion ... After construction, increased amounts of impervious surface caused further increases in runoff but reduced the sediment load, which in turn caused the streams to erode through the newly accumulated deposits" (Graf, 1975).
In a similar study, Graf (1977) studied effects of "suburbanization" on the South Branch of Ralston Creek, Iowa City, Iowa. In this case, he found that suburbanization caused an increase in the number of channel drainages. He concluded that urban development reduces the time of overland flow. Thus, the runoff is both collected and transported more rapidly.
These results are similar to the conclusions of Espey et al. (1965), who studied some of the effects of urbanization on small watersheds in Texas. They included Waller Creek in their study, and predicted that "the time sequence of the discharge hydrograph will be shortened, the peak discharge will be increased, and the unit yield (in/m[i.sup.2]) will be increased" as a result of future increases in urbanization.
[FIGURE 1 OMITTED]
As mentioned above, Waller Creek lies entirely within the city of Austin, Texas. The main channel of the creek has been extended north by excavation just south of Croslin Street (Sutton, 1980). A secondary branch of Waller Creek ("The West Branch") begins near West 45th Street and Lamar, and joins the main channel just west of San Jacinto Boulevard.
Murphey et al. (1977) concluded that the size of the watershed area was useful for predicting hydrograph responses. Sutton (1980) recorded the Waller Creek basin as an area of 5.4 square miles (14.1 square kilometers). Tovar and Maldonado (1981) found that the basin had an area of 5.45 square miles. Just south of Airport Boulevard a drainage ditch formed by the Texas and New Orleans Railroad Track joins Waller Creek. The drainage ditch was created so that an area of 0.3 square miles that previously drained into Shoal Creek to the west would drain into Waller Creek instead (Espey et al., 1965). Thus, human development has increased the natural drainage area of the Waller Creek basin.
A more significant change associated with the Waller Creek basin is that of channel length. Although the original extent of Waller Creek before human development is not known, "an 1891 map of the city shows the creek extending only as far north as West 45th Street. However, 1959 plans for excavation of the channel between West 53rd Street and Koenig Lane indicate that a natural channel occurred there at that time" (Sutton, 1980). Sutton also noted that by 1980 the creek had extended to within "one block" of its drainage divide. As of 1988, Waller Creek was less than 1300 feet (396 meters) from its drainage divide (U.S. Geological Survey, 1988). This elongation of channel length with urbanization might be attributed to an increase in runoff within the basin.
Urbanization can lead to changes in channel width as well as changes in channel length. Sometime before 1962, the channel north (upstream) of Kenniston Drive was excavated, and by 1980 all of Waller Creek north of West 53rd Street had been excavated (Sutton, 1980). "Virtually all excavated channels undergo undesirable morphological changes. Reaches not excavated for flood control tend to erode and enlarge. Reaches excavated for flood control tend to aggrade, sometimes significantly reducing the capacity of the channel" (Sutton, 1980). Thus the excavation of channels to protect urban development changes channel morphology. An example of this is described by Sutton (1980). In 1968, the Waller Creek channel near East 9th Street was widened by 1.5 times its original width and was also deepened to some extent. "By 1974 the bed of the channel had aggraded by about 0.3 to 1.2 m (1 to 4 feet). Deposition affected all parts across the section. The greatest deposition was along the west bank, where deposits were thickest before excavation. Because aggradation of the bed built up the west bank along the bed, the channel is now narrower at low stages than it was following excavation" (Sutton, 1980). Thus, human excavation of the channel caused aggradation. As a result, within six years of this excavation, the channel began to approach its preexcavation dimensions.
In addition to changes in basin area and channel geometry, changes in channel slope may also be related to urbanization. Leopold and Maddock (1953) studied the physiography and geometry of stream channels, and concluded that the slope of a fluvial channel can be modified by processes within the fluvial system. Therefore, a study of changes in channel slope through time might show how Waller Creek has adjusted to increased urbanization. This information is shown below in Figure 1.
Different methods used by the various authors to calculate mean channel slope may account for some of the changes seen in Figure 1. Nevertheless, there does seem to be a correlation between slope and time (increased urbanization). This decrease in channel slope with time can be associated with elongation of channel length. Thus, the downstream portion of the channel may not have changed, but the overall length of the entire channel may have increased. In fact, below the 23rd Street station the morphology of Waller Creek is controlled by bedrock incisement, and "bedrock streamss are, by their very nature, less able to adjust to changing hydrologic conditions, and an attempt to relate their form to present hydrologic conditions may not be justifiable in all cases" (Patton, 1977).
[FIGURE 2 OMITTED]
The effect of urbanization on the channel networks also should be examined. Paved streets act as channels that convey runoff to the creek. Graf (1977) studied a location in Iowa and concluded that urbanization increased the number of channel drainages and affected discharge. Patton (1977) also determined that channel networks strongly influenced drainage of basins. In the case of Waller Creek, Sutton (1980) claimed that "the frequency of first-order channels, basin magnitude, drainage density, and basin ruggedness number has increased as a result of [increased] runoff." Furthermore, he added that paved streets act both as drainage channels, which transport runoff, and as impervious surfaces, which collect precipitation. Espey et al. (1965) also investigated the effect of impermeable surface cover on small Texas watersheds and decided that an increase in impermeable surface cover results in reduced infiltration and increased runoff. Although Sutton (1980) concluded that channel network modification of Waller Creek has a greater effect on runoff than does the increase in impermeable surfaces, he included a graph of impermeable surface area over time for the Waller Creek basin. A modified version of this graph is shown in Figure 2.
[FIGURE 3 OMITTED]
Another way to examine the effects of urbanization on a watershed is to consider changes in maximum discharge. The U. S. Geological Survey began keeping records of the gauging station at 23rd Street in 1951, and records from the 38th Street station in 1956. Using these records (U.S. Geological Survey, 1960, 1975, 1976, 1980, and 1981), graphs of maximum discharge can be constructed (Figs. 3 and 4).
Greater maximum discharges have been recorded at the 23rd Street station, which gauges the more urbanized location, than the station at 38th Street. This is to be expected, because the 23rd Street station is farther downstream. However, when examining individual records for either one of the stations, no definite correlations between maximum discharge and time (increased urbanization) can be made. It may be that more data are needed before such a trend can be found.
[FIGURE 4 OMITTED]
[FIGURE 5 OMITTED]
Veenhuis and Gannett (1986) also investigated discharge of Waller Creek. They examined peak discharge relative to recurrence interval for the period 1956-1962, and compared this to data for 1966-1980. They found increases in peak discharge for a given recurrence interval ranging from 33 percent (two-year recurrence interval) to 10 percent (100-year recurrence interval). Veenhuis and Gannett (1986) invoked changes in impervious cover of the Waller Creek basin to account for the increase in flood peaks of selected recurrence intervals. Figures 5 and 6 are graphs of their data for the 38th Street gauging station.
[FIGURE 6 OMITTED]
The Waller Creek drainage basin lies within the city of Austin, Texas, and has undergone geomorphic and hydrologic changes resulting from increased urban development. As stated by Garner and Young (1976), urban basins are more likely to flood than are nonurban basins because urban areas generally contain less vegetation (which inhibits runoff), and because pavement and buildings prevent water from soaking into the ground. From this study of the Waller Creek basin, it appears that these conclusions are true.
The following parameters of Waller Creek were examined for change: 1) basin area; 2) channel geometry; 3) channel slope; 4) channel network and impermeable surface cover; 5) discharge. The Waller Creek basin area has increased in size as a result of urban development. Channel length has increased and channel slope has decreased as urbanization has progressed. In addition, stream channels have been widened and deepened in an attempt to protect residential buildings. Although the effects of channel networks and impermeable surface cover are difficult to separate, it appears that an increase with urban development of both of these characteristics has led to an increase in runoff. Increase in runoff may be related to decrease in channel slope and to elongation of the channel. Finally, increased flood peak values for selected recurrence intervals correlate with increased urbanization.
Thanks to F. Perez and D. Trombatore for advice and encouragement. Early drafts of this manuscript benefitted from the constructive criticism of J. Sharp and C. Woodruff.
Espey, W. H., Jr., C. W. Morgan, and F. D. Masch. 1965. A study of some effects of urbanization on storm runoff from a small watershed. Tech. Rept. HYD 07-6501, Texas Water Comm., 109 pp.
Garner, L. E., and C. M. Woodruff, Jr. 1979. Urban hydrogeology and other environmental aspects of the Austin area. Austin Geol. Soc., Fall Field Trip, 8 December 1979, Austin, Texas, 39 pp.
Garner, L. E., and K. P. Young. 1976. Environmental geology of the Austin area: an aid to urban planning. Rept. Investigations, Bur. Econ. Geol., Univ. Texas at Austin, 86:1-39.
Graf, W. L. 1975. The impact of suburbanization on fluvial geomorphology. Water Resources Res., 11:690-692.
_____. 1977. Network characteristics in suburbanizing streams. Water Resources Res., 13:459-463.
Leopold, L. B., and T. Maddock, Jr. 1953. The hydraulic geometry of stream channels and some physiographic implications. U.S. Geol. Surv. Prof. Paper, 252:1-57.
Murphey, J. B., D. E. Wallace, and L. J. Lane. 1977. Geomorphic parameters predict hydrograph characteristics in the Southwest. Water Resources Bull. 13:25-37.
Patton, P. C. 1977. Geomorphic criteria for estimating the magnitude and frequency of flooding in central Texas. Ph.D. dissertation, Univ. Texas at Austin, 222 pp.
Robbins, W. D. 1970. Annual compilation and analysis of hydrologic data for urban studies in the Austin, Texas metropolitan area, 1969. U.S. Geol. Surv. Water Resources Div., Texas Dist. Open-File Rept., Austin, Texas, 46 pp.
Schroeder, E. E. 1974. Flood stages and discharge for small streams in Texas. U.S. Geol. Surv. Interim Rept. (Open File Rept.), 85-9:225-226.
Shepherd, R. G. 1979. River channel and sediment responses to bedrock lithology and stream capture, Sandy Creek Drainage, Central Texas. Pp. 255-275, in Adjustments of the fluvial system (D. D. Rhodes and G. P. Williams, eds.), Kendall/Hunt Publ. Co., Dubuque, Iowa, 372 pp.
Slade, R. M., Jr. 1986. Large rainstorms along the Balcones Escarpment in Central Texas. Pp. 15-20, in The Balcones Escarpment (P. L. Abbott and C. M. Woodruff, Jr., eds.), Geol. Soc. Amer., Comet Reproduction Serv., Santa Fe Springs, California, 200 pp.
Sutton, S. M., Jr. 1980. Urban fluvial geometamorphosis. M. A. thesis, Univ. Texas at Austin, 375 pp.
Tovar, F. H. 1973. Annual compilation and analysis of hydrologic data for urban studies in the Austin, Texas metropolitan area, 1971. U.S. Geol. Surv.-Water Resources Div., Texas Dist. Open-File Rept., Austin, Texas, 73 pp.
Tovar, F. H., and B. N. Maldonado. 1981. Drainage areas of Texas streams--Colorado River Basin. Texas Dept. Water Resources, LP - 145, Austin, 36 pp.
Twichell, T. (District Chief). 1967. Compilation of hydrologic data, Waller and Wilbarger Creeks, Colorado River Basin, Texas, 1966. U.S. Geol. Surv., Water Resources Div. Texas Dist. Open File Rept., Austin, Texas, 31 pp.
U.S. Geological Survey. 1960. Hydrologic studies of small watersheds--investigation of the effect of urban development on runoff, Waller Creek Basin, Texas. U.S. Geol. Surv. Water Resources Div., Surface Water Dist. Office, Austin, Texas, 8 pp.
_____. 1975. Water resources data for Texas, 1974. Pt. 1. Surface water records. U.S. Geol. Surv. Water Resources Div., Water-Data Report 1974: 454-455.
_____. 1976. Water resources data for Texas, 1973. Vol. 3, U.S. Geol. Surv. Water Resources Div., Water-Data Report TX, 75-1:156-157.
_____. 1980. Water resources data for Texas (Water Year 1979). Vol. 3. U. S. Geol. Surv. Water Resources Div., Water-Data Report TX, 79-3:195-196.
_____. 1981. Water resources data for Texas (Water Year 1980). Vol. 3, U.S. Geol. Surv. Water Resources Div., Water-Data Report TX, 80-3:193-194.
_____. 1988. Austin East Quadrangle, Texas-Travis County. U.S. Geol. Surv. 7.5 Minute Series Topographic Map, scale 1:24000.
Veenhuis, J. E., and D. G. Gannett. 1986. The effects of urbanization on floods in the Austin metropolitan area, Texas. U.S. Geol. Surv., Water-Resources Investigations Rept., 86-4069, 66 pp.
CHRISTOPHER S. SWEZEY
Department of Geological Sciences, The University of Texas at Austin, Austin, Texas 78713
|Printer friendly Cite/link Email Feedback|
|Author:||Swezey, Christopher S.|
|Publication:||The Texas Journal of Science|
|Date:||Aug 1, 1991|
|Previous Article:||Helminth parasites of unisexual and bisexual whiptail lizards (Teiidae) in North America. VI. the gray-checkered whiptail (Cnemidophorus dixoni).|
|Next Article:||First report of Mesocestoides sp. tetrathridia (Cyclophyllidea: mesocestoididae) in the Texas alligator lizard, Gerrhonotus liocephalus (Sauria:...|