Citation
Differential changes in groundwater quality due to urbanization under varying environmental regulations in Austin, Texas

Material Information

Title:
Differential changes in groundwater quality due to urbanization under varying environmental regulations in Austin, Texas
Creator:
Herrington, Chris
Hiers, Scott. E.
Pope, Sylvia
Clamann, Andrew
Publisher:
City of Austin
Watershed Protection and Development Review Department
Publication Date:
Language:
English

Subjects

Subjects / Keywords:
Avery Deer Spring (Travis County, Texas, United States) ( 30.2667, -97.7333 )
Avery Springhouse Spring (Travis County, Texas, United States) ( 30.2667, -97.7333 )
Barton Springs (Austin, Texas, United States) ( 30.263819, -97.771395 )
Canyon Creek Spring 1 (Travis County, Texas, United States) ( 30.2667, -97.7333 )
Fern Gully Spring (Travis County, Texas, United States) ( 30.2667, -97.7333 )
Hill Marsh Spring (Travis County, Texas, United States) ( 30.2667, -97.7333 )
Geology ( local )
Genre:
Technical Report
serial ( sobekcm )
Location:
United States
Coordinates:
30.2667 x -97.7333
30.263819 x -97.771395

Notes

General Note:
Karst springs in the Canyon Creek and Avery Ranch subdivisions were monitored before, during and after construction of residential homes with some commercial development. The objective of the monitoring program was to track trends in spring chemistry with changing land use in source area and compare the water quality impact of different water quality regulations after development. Groundwater chemistry, particularly ions, changed in correlation with increasing development. The data seems to suggest that with the exception of one spring, the enhanced water quality controls at Avery Ranch had some success in limiting nitrate enrichment of ground water during the construction of the subdivision. Ion concentrations including strontium are generally higher in Canyon Creek Subdivision springs than Avery Ranch, suggesting a potential difference in source water composition. Piper plots indicate few differences between sites that could identify source waters. A comparison was made between the spring chemistry data collected from a newly developing subdivision that was permitted under enhanced development agreement Planned Unit Development (PUD) and an older subdivision that was built-out under less restrictive Municipal Utility District (MUD) agreement to determine if differences in water chemistry could be seen. Spring data collected seems to indicate that only slight difference was seen in groundwater chemistry results between the two subdivisions. This suggests that water quality benefits provided by surface water quality controls had little effect on groundwater quality within subdivisions. Additional data is needed to test this hypothesis.
Restriction:
Open Access - Permission by Publisher
General Note:
See Extended description for more information.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
K26-01270 ( USFLDC DOI )
k26.1270 ( USFLDC Handle )
11531 ( karstportal - original NodeID )

USFLDC Membership

Aggregations:
Karst Information Portal

Postcard Information

Format:
Serial

Downloads

This item has the following downloads:


Full Text
Description
Karst springs in the Canyon Creek and Avery Ranch
subdivisions were monitored before, during and after
construction of residential homes with some commercial
development. The objective of the monitoring program was to
track trends in spring chemistry with changing land use in
source area and compare the water quality impact of different
water quality regulations after development. Groundwater
chemistry, particularly ions, changed in correlation with
increasing development. The data seems to suggest that with
the exception of one spring, the enhanced water quality
controls at Avery Ranch had some success in limiting nitrate
enrichment of ground water during the construction of the
subdivision. Ion concentrations including strontium are
generally higher in Canyon Creek Subdivision springs than
Avery Ranch, suggesting a potential difference in source
water composition. Piper plots indicate few differences
between sites that could identify source waters. A comparison
was made between the spring chemistry data collected from a
newly developing subdivision that was permitted under
enhanced development agreement Planned Unit Development (PUD)
and an older subdivision that was built-out under less
restrictive Municipal Utility District (MUD) agreement to
determine if differences in water chemistry could be seen.
Spring data collected seems to indicate that only slight
difference was seen in groundwater chemistry results between
the two subdivisions. This suggests that water quality
benefits provided by surface water quality controls had
little effect on groundwater quality within subdivisions.
Additional data is needed to test this hypothesis.



PAGE 1

SR-07-05 Page 1 of 14 June 2007 Differential changes in groundwater quality due to urbanization under varying environmental regulations in Austin, Texas. By Chris Herrington, P.E., Scott E. Hiers, P. G., Sylvia Pope, P.G., and Andrew Clamann Water Resource Evaluation Section, Environmental Resource Management Division Watershed Protection and Development Review Department, City of Austin SR-07-05 Abstract Karst springs in the Canyon Creek and Avery Ranch subdivisions were monitored before, duri ng and after construction of residential homes with some commercial development. The object ive of the monitoring program wa s to track trends in spring chemistry with changing land us e in source area and compare th e water quality impact of different water quality regulations after development. Groundwater chemistr y, particularly ions, changed in correlati on with increasing development. The data seems to suggest that with th e exception of one spring, the enhanced water quality controls at Avery Ranch had some success in limiting nitrate enri chment of ground water during the construction of the subdivision. Ion concentrations including strontium are generally higher in Canyon Creek Subdivision springs than Avery Ranc h, suggesting a poten tial difference in source water composition. Piper plots indica te few differences between sites that c ould identify source wa ters. A comparison was made between the spring chemistry data collected from a newly developing subdivision that was permitted under enhanced development agreement Planned Unit Development (P UD) and an older subdivision that was built-out under less restrictive Municipal Utility District (MUD) agreement to dete rmine if differences in water chemistry could be seen. Spring data collected seems to indicate that only slight difference was seen in groundwater chemistry results between the two subdivisions. This suggests that water qual ity benefits provided by surface water quality controls had little effect on groundwater quality within subdivisions. Additional data is needed to test this hypothesis. Introduction The effects of urbanization on water quality of surf ace and groundwater resources are well documented in the scientific literature. In general, as impervious cover increases water quality decreases. As development of land and water resources increase, it is apparent that development of either of these resources affects the quantity and quality of the other, because of the connectivity and interaction between surface-water features such as streams, lakes, reservoi rs, and wetlands with groundwater. In the Austin area, these interactions take the form of surface water recharging the karst aquifer. As a result, the loss of wetland habitat and degradation of surface water quality has an effect on groundwater resources. Thus, effective land and water management requires a cl ear understanding of the linkage between the ground and surface water. Over the years, the City of Austin has created polices governing the management and protection of aquifers and watersheds. These development ordinances have been modified over the years as our understanding of the interaction and interdepe ndency between surface water and groundwater has increased to where we no longer view surface water and groundwater as not simply two independent resources, but as integrated resources. The goal of this study is to measure the effectiveness of development policies and management practices by monitoring the rate and degree of water chemistry changes at karst springs located in two subdivisions within the Northern Segment of the Edwards Aquifer on the Jollyville Plateau in Central Texas.

PAGE 2

SR-07-05 Page 2 of 14 June 2007 Spring sites within the Avery Ranch and Canyon Cr eek subdivisions were monitored during and after construction of residential subdivisions (Table 1). Prior to development, the Canyon Creek and Avery Ranch areas were primarily hunting and ranching areas. The Canyon Creek subdivision was developed as a municipal utility district (MUD) and construction star ted in the early 1980s and most of the subdivision was built out by 2005. Avery Ranch subdivision began construction early in 2000 and there is current Figure 1 Source: City of Austin GIS based on Collins, E.W. 2005, Geologic Map of West half of the Taylor, Texas 30x 60 Minute Quadra ngle and Trippet, A.R. and Garner, L.E., 1986, Geol ogic Map of Austin, Texas 1:62,500 scale.

PAGE 3

SR-07-05 Page 3 of 14 June 2007 development underway. The Avery Ranch subdivision in cludes an 18-hole golf course and amenity center and was developed as a Planned Unit Development (PUD) district. Municipal Utility Districts (MUD) are a special pur pose governmental entity of the State of Texas. Regulated by the Texas Commission on Environmental Quality (TCEQ), the MUD's primary function is to provide water, wastewater and stormwater dr ainage service within its boundaries. A MUD may sell bonds, levy and collect taxes, provide and charge fo r water and sewer services, build infrastructure, condemn property, enforce restrictive covenants a nd make regulations to accomplish its purposes. All Municipal Utility Districts are managed by an elected Bo ard of Directors. Each Director on the Board is a resident or a property owner within the boundaries of the District. The Board of Directors conducts a meeting, usually every month to transact the business of the District. A Planned Unit Development (PUD) district is the designation for a large or complex single or multi-use development that is planned as a single contiguous project and that is under unified control. The purpose of a PUD district designation is to preserve the natu ral environment, encourage high quality development and innovative design, and ensure adequate public facilities and services for development within a PUD. A PUD district designation provides greater design flexibility by permitting modifications of site development regulations. Developmen t under the site development regulations applicable to a PUD must be superior to the development that would occur under conventional zoning and subdivision regulations. A PUD district must include at least 10 acres of land, unless the property is characterized by special circumstances, including unique topographic constraints. Development within the two subdivisions occurred under different water quality ordinances. The Canyon Creek subdivision was initially developed in accordan ce with the City of Austins Comprehensive Watershed Ordinance and later se ctions were permitted in accordan ce with the Save Our Springs Ordinance. The Avery Ranch Subdivision was devel oped as a PUD and provided water quality controls superior to the development that would occur unde r the conventional Save Our Springs Ordinance. The water quality controls for Avery Ranch are ma inly wet ponds, compared to the standard sedimentation/filtration ponds constructed in the Ca nyon Creek Subdivision. To determine if there are differential changes in groundwater quality due to ur banization under varying e nvironmental regulations, springs within each watershed were monitored for a period of five years from 2002 to 2007. Five karst springs; two springs in Canyon Creek and three in Av ery Ranch Subdivision were selected for this study (Table 1). All five springs discharge at or n ear the same geologic contact between the Edwards Limestone and Walnut Formation and have simila r discharges rates and springshed sizes (Figure 1). Groundwater movement within the Northern Edwards in the Jollyville Plateau is typically controlled by primary porosity flow through fractures, bedding plan es, and conduits. Spring recharge occurs in upland areas where karst recharge features allow for surface wa ter to infiltrate into the Edwards and migrate to the Walnut Formation through a network of voids, fracture, bedding plans and conduits. An example of this is the cave stream intercepted upstream of Avery Deer Spring during trenching activities for construction of a wastewater line. However, spring development occurs mostly in the underlying Walnut Formation where groundwater movement is more diffu se via secondary porosity, or water moving within the rock matrix. Table 1. Spring sites Site # Site Name Subdivision 504 Canyon Creek Spring 1 (Tubb Spring) Canyon Creek 1078 Fern Gully Spring Canyon Creek 1352 Avery Springhouse Spring Avery Ranch 1353 Hill Marsh Spring Avery Ranch 1355 Avery Deer Spring Avery Ranch

PAGE 4

SR-07-05 Page 4 of 14 June 2007 Except for site 504, Jollyville salamanders have been observed by City staff at all the springs (City of Austin field data, 1999-2006). Spring sites were sampled quarterly beginning in 2002, although some historical data collected prior to 2002 was considered for temporal trend analyses. Metals were only analyzed once annually at all sites. The contributing drainage area to each spring was estimated using surface topography derived from City of Austin Geographic Information Systems (GIS), see Figure 2. Impervious cover for each area was estimated from GIS landuse information and corresponding average percent of impervious cover (Figure 3). Impervious cover within the Avery Springhouse and Fern Gully springs rapidly increased over the 1997-2006 time period. Impervious c over has consistently been higher at Tubb Spring partially due to the fact that development occurred earlier than at Avery Springhouse or Fern Gully springs. The Canyon Creek subdivisions consists of a higher percentage of residential land use and a lower percentage of undeveloped land a nd parks based on 2003 aerial photography analysis (Table 2). Table 2. 2003 Landuse in the two subdivisions. Landuse Avery Canyon Creek Single Family Residential 16.6 19.6 Multi-family Residential 0.0 13.9 Commercial 0.1 1.5 Office 0.0 0.3 Industrial 0.0 2.0 Civic 5.1 0.4 Parks/Open Space 15.4 37.0 Transportation 12.5 8.2 Undeveloped 50.3 17.2 Figure 2: Changes in the amount of Impervious Cover within the springshed 1997 to 2006

PAGE 5

SR-07-05 Page 5 of 14 June 2007 Figure 2 (continued)

PAGE 6

SR-07-05 Page 6 of 14 June 2007

PAGE 7

SR-07-05 Page 7 of 14 June 2007 Figure 3 Impervious Cover Increase 1997-20060% 5% 10% 15% 20% 25% 30% 35% 40% 1996199820002002200420062008 YearPercent Impervious cover AveryDeer AverySpringhouse HillMarsh FernGully CanyonTubbs Figure 3. Percent impervious cover by springshed ba sed on City of Austin GIS land use information. Methods Outliers were identified and removed from the anal ysis. These datapoints included all potassium and sodium data collected on 6 March 2002. Iron data co llected on 5 October 2005 were also removed. One unusual ammonia value (0.7 mg/L) at Hill Marsh Spring on 12 December 2007 was excluded from the analysis. Differences within and between subdivisions were analyzed by the non-parametric Wilcoxon rank-sum test in SAS using PROC NPAR1WAY and confirmed by Ryan-Einot-Gabriel-Welsch multiple range test (SAS PROC GLM). Temporal trends were assessed using least-square linea r regression in SAS (SAS Institute, version 9.1) by PROC GLM. For datasets with censored values (i.e., less than detection limit), temporal trends were assessed by Cox proportional hazards regression (Allis on 1995) using SAS PROC PHREG. All temporal trends were verified graphically. Water quality measurements were compared to imperv ious cover by linear regression analyses. Because impervious cover information is cu rrently available only for three years, impervious cover change over time was assumed to be a linear function and the change in impervious cover in intervening years (without actual impervious cover measures) was estimated by linear regression. Negative values for impervious cover were censored at zero. This met hod is likely to over-estimate impervious cover in more recent years. More frequent annual estimates of im pervious cover, as can be derived from local county appraisal district records, would be more realistic to use in this type of analysis. (check to see if GIS aerial photo analysis updated the i.c. estimates or to see if TCAD and WCAD records were used) Piper plots were used to compare groundwater chemis try between sites. Unless specified otherwise, a critical value ( ) of 0.05 was used to determine statistical significance.

PAGE 8

SR-07-05 Page 8 of 14 June 2007 Results Temporal trend analysis yields significant change ove r time for most ion parameters at the majority of sites (Table 3). Nitrate increased in Canyon Creek s ites but yields conflicting results for sites in Avery Ranch as seen in nitrate values at the different Avery springs showing no change, increasing and decreasing trends in concentrations (Figure 4). The da ta seems to suggest that the enhanced water quality controls at Avery Ranch had some success in lim iting nitrate enrichment of ground water during the construction of the subdivision with the exception of one spring; Avery Deer, where nitrate levels increased. The decreasing nitrate trend at Avery Hill Marsh is surprising and seems to indicate a change in the source water recharging the springs. Perhaps, th e drop in nitrate concentrations is related to the enhanced Integrated Pest Management Plan that lim its fertilizer application a nd irrigation practices used for the Golf Course. Additional studies are needed to confirm this hypothesis. Development practices appear to result in differential water quality degradation in nitrate. Tubb 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 Feb-1995Nov-1997Aug-2000May-2003Feb-2006Oct-2008NO3+NO2-N (mg/L) Fern Gully 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 Feb-1995Nov-1997Aug-2000May-2003Feb-2006Oct-2008NO3+NO2-N (mg/L) Avery Springhouse 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 Feb-1995Nov-1997Aug-2000May-2003Feb-2006Oct-2008NO3+NO2-N (mg/L) Avery Deer 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 Feb-1995Nov-1997Aug-2000May-2003Feb-2006Oct-2008NO3+NO2-N (mg/L) Hill Marsh 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 Feb-1995Nov-1997Aug-2000May-2003Feb-2006Oct-2008NO3+NO2-N (mg/L) Figure 4. Nitrate plus nitrite as nitrogen (mg/L) over time for all five study springs with consistent axes. The temporal trend analysis indicates significant increasing cation and anion concentrations are occurring at all sites. Figure 5 shows a piper plot on cation and anion values over time. The piper plots, used to describe the hydrochemical facies of an aquifer, indicate that the groundwater is calcium /bicarbonate type, as anticipated for groundwater in limestone lit hology. The calcium, magnesium and bicarbonate are the dominant ions, as indicated in the plot of average dominant ion concentrations. However, a plot of all the major ion data indicates the hydrochemical fa cies in the springsheds is slowly shifting to magnesium/chloride type because of changes in the solution kinetics, flow patterns, or source recharge in the springsheds. Despite observed differences in l ong-term mean concentrations of ions between

PAGE 9

SR-07-05 Page 9 of 14 June 2007 subdivisions, piper plots indicate few differences be tween sites. However, groundwater chemistry, particularly ions, changed in corre lation with increasing development. Temporal trend analysis yields significant change ove r time for three metals; iron, nickel and lead. Iron values spiked in 2003, and then app eared to slightly decrease over the remainder of the period of record. Observed temporal trends in metals may be due to laboratory analytical problems or from the use of contaminated acid for sample preservation as indicated by metals values peaking in approximately 2003. An internal investigation of metal results by Water Resources Evaluation Section revealed that some type of preservation or analytical laboratory error resu lting elevated metal concentrations through 2004?. However, the more recent samples have been all nondetect or at low levels. Canyon Creek springs typically maintain higher ion concentrations than Avery Ranch springs (table 4), although nitrogen appears to be higher in Avery Ranch springs. Metals show no differences between subdivisions. Table 3. Summary of temporal trend analysis results over the period of record. Canyon Creek Avery Ranch = no trend Tubb Fern Deer Springhouse Hill Marsh Ions ALKALINITY Increasing Decreasing Increasing Increasing CALCIUM Increasing Increasing Increasing CHLORIDE Increasing Increasing Increasing Increasing CONDUCTIVITY Increasing In creasing Increasing Increasing FLUORIDE * * MAGNESIUM Decreasing Incr easing Increasing Increasing POTASSIUM Increasing1 Increasing1 Increasing Increasing SODIUM Increasing Increasing In creasing Increasing Increasing SULFATE Increasing Increas ing Increasing Increasing Nutrients AMMONIA * Decreasing1 * NITRATE/NITRITE Increasing Increasing Increasing Decreasing ORTHOPHOSPHORUS * * Decreasing Metals ARSENIC * * COPPER * * IRON * Decreasing Decreasing Decreasing1 LEAD Decreasing1 Decreasing Decreasing Decreasing NICKEL Increasing Increasing Increasing Increasing Increasing STRONTIUM * * ZINC * * Conventionals DISSOLVED OXYGEN Decreasing1 * E COLI BACTERIA Increasing * * FECAL COLIFORM * * FLOW * * ORGANIC CARBON Increasing Increasing Increasing PH * Increasing TURBIDITY Decreasing Decreasing Decreasing TEMPERATURE Increasing * 1. Only significant at the =0.10 level

PAGE 10

SR-07-05 Page 10 of 14 June 2007 Table 4. Differences between site groups by Wilcoxon rank-sum test and site group means. Mean Wilcoxon Rank-Sum Parameter Units Avery CC Pr > |z| Notes Ions ALKALINITY-CaCO3 MG/L 332.184 329.434 0.4292 CALCIUM MG/L 101.648 113.406 0.0000 CC>Avery CHLORIDE MG/L 14.802 21.684 0.0000 CC>Avery CONDUCTIVITY S/cm 690.722 739.027 0.0107 CC>Avery FLUORIDE MG/L 0.155 0.185 0.9263 MAGNESIUM MG/L 27.860 19.192 0.0000 Avery>CC POTASSIUM MG/L 1.085 1.006 0.0354 Avery>CC SODIUM MG/L 7.931 11.938 0.0000 CC>Avery STRONTIUM UG/L 106.917 379.833 0.0037 CC>Avery SULFATE MG/L 17.529 21.015 0.0001 CC>Avery Nutrients AMMONIA AS N MG/L 0.013 0.012 0.0460 Avery>CC NITRATE/NITRITE-N MG/L 2.893 1.535 0.0000 Avery>CC ORTHOPHOSPHORUS-P MG/L 0.009 0.009 0.1176 Metals ARSENIC UG/L 0.654 1.195 0.8680 COPPER UG/L 0.833 0.681 0.3727 LEAD UG/L 5.601 5.531 0.8878 NICKEL UG/L 2.161 2.206 0.7379 ZINC UG/L 4.109 4.147 0.8985 Conventionals DISSOLVED OXYGEN MG/L 6.200 6.885 0.0006 CC>Avery E COLI BACTERIA MPN/dL 66.020 15.426 0.0001 Avery>CC FECAL COLIFORM Colonies/dL 207.839 33.833 0.0656 FLOW CFS 0.036 0.038 0.0114 CC>Avery IRON UG/L 45.566 42.716 0.1910 ORGANIC CARBON MG/L 2.290 2.865 0.3954 PH Std Units 7.025 7.020 0.6745 TEMPERATURE Deg. C 20.757 20.676 0.7293 TURBIDITY NTU 2.851 2.937 0.5867 * = indicates no significant difference between site groups CC = Canyon Creek Sites within each subdivision were also assessed for pot ential differences. Avery Ranch sites yield higher concentrations of nutrients than springs in Canyon Cree k (Table 5). There were no statistically significant differences between sites for metals. Within Canyon Creek, Tubb Spring (site 504) had generally higher concentrations of ions than Fern Gully (site 1078) Strontium values in Canyon Creek are significantly higher than Avery Ranch, suggesting a potentia l difference in source water composition.

PAGE 11

SR-07-05 Page 11 of 14 June 2007 Table 5. Site means and differences between site s for each subdivision by Wilcoxon rank-sum test (pvalues < 0.05 indicate significant difference within site groups). Canyon Creek Avery Ranch Parameter Unit Tubb Fern Pr>|z| Deer Spg. House Hill Marsh Pr>|z| Ions ALKALINITY-CaCO3 MG/L 367.857 286.400 0.0000 307.433 339.844 352.080 0.0000 CALCIUM MG/L 117.915 108.175 0.0295 103.550 99.174 102.688 0.0612 CHLORIDE MG/L 23.000 20.209 0.0236 9.610 14.145 21.872 0.0000 CONDUCTIVITY S/cm 802.693 632.917 0.0000 627.120 694.042 756.391 0.0000 FLUORIDE MG/L 0.240 0.124 0.0044 0.129 0.166 0.171 0.1839 MAGNESIUM MG/L 26.224 11.035 0.0000 19.162 31.139 33.275 0.0000 POTASSIUM MG/L 1.045 0.987 0.6231 0.765 0.685 1.967 0.0000 SODIUM MG/L 12.593 11.173 0.0844 5.334 7.749 11.178 0.0000 STRONTIUM UG/L 501.667 258.000 0.0495 55.933 145.633 80.467 0.0222 SULFATE MG/L 22.411 19.452 0.0020 12.436 16.119 25.444 0.0000 Nutrients AMMONIA AS N MG/L 0.011 0.013 0.1880 0.013 0.015 0.015 0.2167 NITRATE/NITRITE-N MG/L 1.671 1.337 0.0007 3.292 2.874 2.443 0.0000 ORTHOPHOSPHORUS-P MG/L 0.010 0.021 0.1338 0.030 0.008 0.012 0.8245 Metals ARSENIC UG/L 0.657 3.934 0.3851 0.481 8.240 1.241 0.7167 COPPER UG/L 0.610 0.720 0.6264 0.328 1.030 1.282 0.5444 IRON UG/L 48.493 36.486 0.9342 39.107 48.382 49.335 0.8922 LEAD UG/L 4.602 6.706 0.5165 7.311 6.223 8.074 0.8992 NICKEL UG/L 2.320 2.149 0.2661 2.108 2.260 2.370 0.5161 ZINC UG/L 4.788 3.537 0.9372 3.725 4.620 3.933 0.9032 Conventionals DISSOLVED OXYGEN MG/L 6.651 7.257 0.0232 6.114 5.496 7.027 0.0000 E COLI BACTERIA MPN/dL 25.819 3.969 0.8331 103.250 17.450 81.000 0.9088 FECAL COLIFORM Col. /dL 26.889 41.667 0.0759 472.100 59.250 114.222 0.7470 FLOW cfs 0.021 0.059 0.4428 0.023 0.041 0.044 0.0693 ORGANIC CARBON MG/L 2.486 3.320 0.4851 2.040 2.169 2.834 0.0138 PH Std Unit 7.042 6.983 0.2020 7.054 6.937 7.086 0.0430 TEMPERATURE Deg C 20.865 20.361 0.2413 20.537 20.928 20.807 0.8437 TURBIDITY NTU 3.792 1.919 0.0976 3.482 2.133 2.910 0.4528 Piper plots indicate few differences between sites or subdivisions (figure 3).

PAGE 12

SR-07-05 Page 12 of 14 June 2007 Figure 5. Piper plots using site mean parameter concentrations. All Ion Concentrations 8 0 6 0 4 0 2 02 0 4 0 6 0 8 020 40 60 80 80 60 40 20 20 40 60 80 20 40 60 80Ca Na+KHCO3Cl Mg SO4 A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E A A A A A A A A A A A A A A A A A A A A A A A A B B B B B B C C C C C C C C C C C C C C CLegend LegendAAvery Deer SpringBAvery SpringhouseCFern GullyDHill Marsh Tubb SpringsE B A A A A A A A A A A A A A A A B Average Ion Concentrations 8 0 6 0 4 0 2 02 0 4 0 6 0 8 020 40 60 80 80 60 40 20 20 40 60 80 20 40 60 80Ca Na+KHCO3 Cl Mg SO4 A A A C C C D D D E E E B B B

PAGE 13

SR-07-05 Page 13 of 14 June 2007 Water quality data was compared to impervious cove r by linear regression analysis. Ions and nitrate generally increased with increasing impervious cover. Trends in metals may be affected by change in analytical laboratories and sample preser vation methods as discussed previously. Table 6. Summary of regression analys is with impervious cover by site. Canyon Creek Avery Ranch = no trend Tubb Fern Deer Springhouse Hill Marsh Ions ALKALINITY-CaCO3 Increasing Decreasing1 Increasing Increasing CALCIUM Increasing Increasing Increasing Increasing CHLORIDE Increasing Increasing In creasing1 Increasing Increasing CONDUCTIVITY Increasing In creasing Increasing Increasing FLUORIDE * * MAGNESIUM Decreasing Incr easing Increasing Increasing POTASSIUM Increasing1 Increasing1 Increasing SODIUM Increasing Increasing In creasing Increasing Increasing STRONTIUM * * SULFATE Increasing Increas ing Increasing Increasing Nutrients AMMONIA-N * Decreasing1 * NITRATE/NITRITE-N Increasing Increasin g Increasing Increasing1 Decreasing ORTHOPHOSPHORUS-P * * Decreasing Metals ARSENIC * * COPPER * * IRON * Decreasing Decreasing Decreasing1 LEAD Decreasing1 Decreasing Decreasing Decreasing NICKEL Increasing Increasing Increasing Increasing Increasing ZINC * * Metals DISSOLVED OXYGEN Decreasing1 * E COLI BACTERIA Increasing * * FECAL COLIFORM * * FLOW * * ORGANIC CARBON Increasing Increasing Increasing PH * Increasing TURBIDITY * Decreasing Decreasing TEMPERATURE Increasing1 * 1. Only significant at the =0.10 level Conclusions Since the late 1980s, the City of Austin has created development ordinances and polices governing development within its city limits and extraterritorial jurisdiction area to help protect the local environmental resources. As our understanding of th e interaction and interdependency between surface water and groundwater has increased over the years to no longer view surface wa ter and groundwater as not simply two independent resources, but as integrated resources, ordinances were modified with goal of improving the City protection of surface water resour ces. Although the effect of impervious cover on surface water and groundwater are well documented, the goal of this study was to determine if measures difference in groundwater chemistry can detected at spring within subdivision development under different and changing development ordinances and increases in impervious cover over time. The Avery

PAGE 14

SR-07-05 Page 14 of 14 June 2007 Ranch Subdivision was developed as a PUD and prov ided superior water quality controls to the development that would occur under the conventiona l Save Our Springs Ordinance. Water quality controls are mainly wet ponds, compared to the standard sedimentation/filtration ponds constructed in the Canyon Creek Subdivision. A comparison was made between the groundwater chemistry data collected from springs in a newly developing subdivision that was permitted unde r enhanced development agreement Planned Unit Development (PUD) and an older subdivision that was completely built-out under less restrictive Municipal Utility District (MUD) agreement. The comparison was to determine if differences in water chemistry could be seen. Although surface water quality might have benefited greatly improved water quality measures, the groundwater data collected seems to indicate that only slight differences were seen in groundwater chemistry results between the two subd ivisions. This suggests that water quality benefits provided by surface water quality controls had little effect on groundwater quality within subdivisions. Additional data is needed to test this hypothesis. Groundwater chemistry, particularly ions, changed in correlation with increasing development. These data seem to suggest that the enhanced water quality c ontrols at Avery Ranch had some success in limiting nitrate enrichment of ground water during the constr uction of the subdivision with the exception of one spring; Avery Deer, where nitrate levels increased. Io n concentrations including strontium are generally higher at Canyon Creek Subdivision than Avery Ranch, suggesting a potential difference in source water composition. Despite observed differences in longterm mean concentrations of ions between subdivisions, piper plots indicate few differences between sites. Water quality data was compared to impervious cove r by linear regression analysis. Ions and nitrate generally increased with increasing impervious cover. Trends noted in metals data may have been affected by change in analytical laboratories and sample preservation methods. References Allison, P. D. 1995. Survival Analysis Using the SAS System: A Practical Guide. SAS Institute. Cary, NC. 292 pp. Senger, R. K., Collins, E.W., Kreitler, C.W.. Hydrog eology of the Northern Segment of the Edwards Aquifer, Austin Region. University of Texas Bureau of Economic Geology. Report Investigation No. 192, Austin, TX, 58 pp. Acknowledgments This report was generated with the cooperation of Chris Herrington, Sc ott Hiers, Andrew Clamann, and Sylvia Pope.