Barton Creek Report
EXECUTIVE SUMMARY
Disclaimer: This excerpt from the report is a draft and is subject to
revision
Purpose
This report presents the Austin City Council, the Environmental Board, and
the community at large a comprehensive account of the water quality investigations
completed by the City's Environmental Resource Management staff and provides
a "state of the creek" report for its most popular natural resource,
the Barton Creek/Barton Springs system.
The studies described in this report attempt to diagnose the level of impact
on Barton Creek's water quality from urbanization activities such as large
scale land developments, wastewater disposal alternatives, golf course and
residential landscape maintenance, and cattle ranching. The primary approach
in most of the studies is to monitor ground water, surface water, sediments,
and biology of Barton Creek and compare urban or developed sites with rural
or undeveloped sites. The primary pollution impairments of concern include
sedimentation, nutrient enrichment, bacterial contamination, and toxic substance
contamination. These constituents are measured directly from the water and
sediments, and their impact is measured using various ecological methods
assessing the diversity and abundance of aquatic plant and animal communities.
Elements of this report include both comprehensive narratives of previously
unreleased studies as well as summaries of studies which have recently been
released.
Directives
Espey Huston and Associates (EHA) provided a baseline of data on the Barton
Creek watershed in 1979 with their Barton Creek Watershed Study, written
for Austin's Department of Environmental Protection (EHA, 1979). They identified
gaps in the ecological information and described studies needed for better
understanding of Barton Creek, its protection, and management. EHA stated
that the existing data base was not adequate to determine the effects of
development. A resolution passed by the Austin City Council on October 15,
1987 directed the City's Department of Environmental Protection to assist
the Environmental Board in a short term review and analysis of existing
data on the Barton Creek watershed. The resulting 1988 City Manager's Barton
Creek Policy Definition Report was endorsed by the Environmental Board,
and the recommendation for Action Group III-2 was -- "continue to monitor
and report upon changes in baseline conditions of land and water resources
in the watershed which are attributable to urban development." Such
is the essential focus of this report. Although this directive was made
in 1988, funding, staff, equipment, and monitoring plans were not solidified
until 1990, and monitoring for several study elements did not begin until
1992 or 1993. The following projects are documented in this report:
Report Element
Overviews and Findings
- Barton Creek Watershed Ground Water Monitoring Program: 1993 - 1996
Purpose:
- Determine impact of urbanization on ground water quality and quantity
for both baseflow and stormflow.
- Determine impacts of on-site and wastewater irrigation on ground water
quality and flow.
- Identify characteristics of water quality and flow in Barton Springs
and the Edwards Aquifer.
Overview:
Ground water monitoring in the Barton Creek watershed focuses on ground
water issues in both the Contributing Zone and Recharge Zone of Barton Springs.
Contributing Zone efforts examine local ground water resources and problems
because of the nature of the shallow water tables in the Contributing Zone.
Recharge Zone or Edwards Aquifer studies are more area-wide in scale due
to the complex nature of ground water recharge and movement in the Edwards
Aquifer and the associated difficulty of identifying local development impacts.
Current ground water monitoring efforts related to Barton Creek include:
- Water samples are collected and analyzed for nutrients, physical parameters,
ions, and selected heavy metals from selected springs in the Barton Creek
watershed two times a year.
- Water samples are collected quarterly from five springs discharging
from the Barton Springs segment of the Edwards Aquifer (Barton, Eliza, Old
Mill, Backdoor, and Cold Springs) and analyzed for physical parameters,
nutrients, ions, and selected heavy metals.
- Water samples are collected every two weeks from Barton Springs and
analyzed for nutrients and total suspended solids.
- Ground water flow paths are identified in the Edwards Aquifer from
specific points in Barton and Williamson Creeks through an interlocal agreement
with BS/EACD.
- In situ data recorders are used continuously in Barton Springs
measuring water pH, temperature, dissolved oxygen, specific conductance,
turbidity, and depth. Data recorders are used periodically in other Edwards
springs.
- Through a cooperative program with the U.S. Geological Survey, water
samples are collected annually from numerous wells in the Barton Springs
segment and Barton Springs.
Findings:
Ground water quality is generally good in springs monitored in the Glen
Rose Formation and in the Contributing Zone of the Barton Springs segment
of the Edwards Aquifer. However, statistically significant water quality
differences in total dissolved solids, total Kjeldahl nitrogen, calcium,
potassium, nitrate, sodium, chloride, sulfate, alkalinity, and total organic
carbon have been identified in springs located in urban areas versus rural
areas. One spring on the mainstem of Barton Creek, below the Lost Creek
Blvd. bridge, has a significant effect on the nitrate concentrations in
a localized area of the mainstem of Barton Creek. The only identifiable
source of the nitrates is leakage from effluent holding ponds and effluent
irrigation in the immediate area. Evidence of this source is similar nitrogen
concentrations and isotope ratios at the holding pond and spring. Increases
in springflow resulting from wastewater irrigation on a tributary in Barton
Creek West have also been indicated from monitoring data.
Barton Springs water remains relatively unchanged by urban activities since
monitoring was initiated. Nutrient and metal concentrations do not show
clear time trends that appear related to urban development. However, the
presence of tetrachloroethylene in Barton Springs water in the late 1980's
and early 1990's indicates that the results of some urban activities can
be seen in the springs. Several heavy metals, including arsenic, cadmium,
copper, lead, nickel, and zinc, as well as sediment of possible anthropogenic
origin also have been detected at Barton Springs. Old Mill and Cold Springs
also appear to be affected by urbanization as indicated by heavy metals,
pesticides, and total petroleum hydrocarbons. Comparisons of ground water
chemistry made among five discrete spring sources (Barton, Backdoor, Eliza,
Old Mill, and Cold) within the Barton Springs segment of the Edwards Aquifer
show that Barton Springs and Backdoor Springs have the highest nitrate levels.
These and other water quality differences among the five springs are commonly
attributed to differences in recharge areas, land use, and flow paths to
each spring. Barton, Eliza, and Old Mill Springs discharge into Barton Creek
near its confluence with the Colorado River. They appear to discharge water
recharged throughout the Edwards Aquifer. Cold Springs discharges into the
Colorado River downstream of Red Bud Isle and receives water recharged in
the Rollingwood area and Barton Creek. Backdoor Spring discharges to Barton
Creek in the upper end of the Recharge Zone and appears to receive water
recharged between Barton Creek and U.S. Hwy 290.
Many constituents like nitrate and various ion concentrations are inversely
related to discharge rates at Barton Springs due to dilution with less concentrated
recharge water, while suspended solids and bacteria concentrations are directly
related to discharge rates due to contributions from surface runoff. Impacts
to Barton Springs from rainfall in the Barton Creek watershed generally
have a lag time of approximately 14 hours, indicated by declining values
in specific conductance and pH, and increases in turbidity and dissolved
oxygen. Based on this lag time, ground water velocity of recharged storm
water is estimated to average around 860 ft/hr. A data point indicating
possible recharge from Williamson Creek reaching Barton Springs in sixty-five
hours indicates ground water average velocity for storm water of approximately
400 ft/hr.. Analysis of Edwards wells indicates that seven wells, in addition
to Old Mill Spring, may be affected by urban development, based on nitrate,
sulfate, and chloride concentrations. These sites are all in developed areas
of the Aquifer. Barton Springs water quality is representative of the overall
good quality of water recharging the Edwards Aquifer. Nevertheless, transient
impacts to Barton Springs are affected most strongly by water quality changes
in Barton Creek as shown by Datasonde parameter changes in water quality
at Barton Springs following stormwater runoff to the creek.
The Barton Creek Pools Study: November 1990 - November 1995
Purpose:
The Barton Creek Pools Study was initiated to document existing ecological
or water quality impacts to perennial pools due to current levels of development.
The study is a comparison of baseflow water quality and an ecological assessment
of nine pools along the mainstem of Barton Creek from the headwaters to
the Edwards Aquifer Recharge Zone. Analyses of data are made to determine
if statistically significant differences exist between pools for various
water chemistry parameters and percent cover of filamentous algae, and determine
if any trends in water quality degradation exist between developed and undeveloped
reaches of the creek or if any water quality degradation is measurable over
time.
Overview:
Since November of 1990, the City of Austin has monitored baseflow water
chemistry and percent cover filamentous algae growth at nine natural pool
sites on the mainstem of Barton Creek, from the headwaters, upstream of
Dripping Springs, to the Edwards Aquifer Recharge Zone, upstream of the
Loop 360 bridge in Austin.
Aquatic vegetative cover, nutrients, suspended and dissolved solids, bacteria,
and additional chemical and physical parameters were measured quarterly
in each of the nine pools.
Findings:
Comparisons made between pools in this study illustrate some small but statistically
significant spatial differences in water quality along Barton Creek's mainstem;
however, no temporal trends over the monitoring period were determined to
be significant.
Surface water comparisons made among nine perennial pools over a five year
period on the mainstem of Barton Creek indicate that the lower three study
pools, all below Barton Creek Blvd. and along the most highly developed
reach, are each impacted by either significantly higher nitrates, TDS, TSS,
turbidity or algal growth. The other six pools upstream of Barton Creek
Blvd. show no significant degradation with the exception of significantly
higher fecal coliform at the most upstream headwater pool. It is important
to note that impacts to each of the lower three pools are localized and
not ubiquitous along this lower reach of the creek. Water quality impacts
seen at one study pool are remediated before reaching the next study pool,
only to be replaced by other impacts related to local land use or construction
activities.
Baseflow water quality above Barton Creek Blvd. is fairly homogeneous, and
from the data available the water chemistry along this reach of the mainstem
has not deteriorated substantially since the 1988 Barton Creek Policy Definition
Report was written. The baseflow water chemistry throughout the study area
is still excellent compared to other urban streams contributing to Town
Lake studied by the City's Water Watchdog Program and to least-disturbed
streams studied by TNRCC in the Central Texas Plateau ecoregion.
The highest nitrogen and TDS concentrations are found in one pool located
below Lost Creek Blvd. Bridge. This elevated nitrogen and TDS is a result
of contributions from a spring, possibly enriched through leaks in effluent
holding ponds and effluent irrigation in the area. Similar stable nitrogen
isotope ratios and nitrogen concentrations link the spring and effluent,
but continued investigations, including dye tracing, would be necessary
to verify effluent as a source.
The pool below Lost Creek Blvd., downstream of residential and golf course
land uses, is significantly higher than all other sites in percent cover
of filamentous green algae, principally due to reoccurring Cladophora
sp. blooms there. Higher nitrates and conductivity correlate positively
with higher filamentous algae at this site,. ERM staff have also observed
that massive Cladophora blooms can result from nutrient surges caused
by accidental spills or mismanagement of domestic wastewater effluent used
for irrigation.
Significantly high turbidity is measured at two sites, one just below Barton
Creek Blvd. and one just above the Recharge Zone. The Recharge Zone site
is also significantly higher in TSS. Intense local construction activity
and upstream impoundments which trap and concentrate the fine sediments
from construction sites are the only unique observable source for these
elevated TSS concentrations. In general, higher TSS values were caused by
an increase in mineral sediment load rather than organic sediment load as
observed through VSS to TSS ratios.
Fecal coliform is significantly higher at the most upstream rural site (Pool
1); however, bacteria counts are still very low there compared to other
urban creeks and normally within safe limits for recreational contact. Fecal
coliform is of an animal, not human, origin throughout the watershed (as
indicated by coliform/streptococcus ratios), including this site, where
the source is most likely the cattle ranching operations upstream and adjacent
to the sampling pool.
At present, these significant water chemistry differences are rather small
and localized. During periods of good flow, enough pristine waters are still
contributed from Barton Creek's rural and undeveloped areas to dilute or
remediate impacted discharges from developed tributaries and springs located
lower in the watershed. As Barton's watershed develops and more impacted
discharges are added, water quality degradation in Barton Creek will likely
be more widespread and conspicuous.
Further development in the Barton Creek watershed will most likely be associated
with the following impacts observed in the pool study sites during baseflow
periods: (1) diminished water clarity in impounded and slower-moving waters,
resulting from cumulative impacts of construction-related runoff; (2) replacement
of a relatively diverse aquatic flora with a monoculture of Cladophora
algae below lands where there is potential for mismanagement of treated
sewage effluent used for irrigation; (3) maintenance of heavier filamentous
algae cover in the mainstem owing to nutrient-enriched waters draining to
Barton Creek from developed tributaries and springs.
Barton Creek Canyons Study: 1993 - 1995
Purpose:
The Canyons Study was initiated to compare water quality impacts to tributaries
of Barton Creek from different land uses and methods of wastewater disposal
in their contributing watersheds.
Overview:
Data was collected from 38 sites on tributaries to Barton Creek. Three tributaries
representative of each major land use are monitored monthly for baseflow.
Tributaries were categorized according to the dominant land use in their
drainage area: golf course, high density residential, or rural (ranching
and low density residential). Tributaries were further characterized according
to the predominant method of wastewater disposal used in their drainage
areas: golf courses using treated wastewater effluent for irrigation, residential
areas irrigating with wastewater effluent on native land, residential areas
on septic systems, residential areas on central sewage systems and rural
areas with little or no commercial or residential development.
Parameters measured in the laboratory included nutrients, bacteria, and
physical parameters.
Findings:
There are significant differences in baseflow nitrate, ammonia, TDS, TSS,
and turbidity concentrations between watersheds draining golf courses, residential,
and rural land uses. Under most analysis groupings, golf course tributaries
have higher constituent concentrations than residential tributaries, and
both golf course and residential tributaries have substantially higher concentrations
for these five parameters than rural tributaries.
Baseflow data, as indicated by antecedent dry conditions, suggest that nitrate
nitrogen is the most variable parameter measured in the Barton Creek watershed.
A comparison of tributaries characterized by various wastewater treatment
strategies reveal that golf course watersheds using sewage effluent irrigation
and fully developed residential watersheds on central wastewater systems
generate significantly higher nitrate concentrations in their baseflow than
residential watersheds irrigating native grass areas with sewage effluent,
residential neighborhoods on septic systems, or undeveloped rural watersheds.
Buffers associated with residential areas using septic systems appear to
be functioning to keep excess nutrients and bacteria from reaching surface
waters. This finding may also be related to the lower impervious cover associated
with larger lot sizes in residential areas on septic systems.
When water samples are collected simultaneously during storm events from
the three selected tributaries representing residential, golf, and rural
land use, the representative golf course site is significantly higher in
nitrates and ortho phosphorus, while the representative residential site
is significantly higher in pH and lower in TDS. The residential site's lower
TDS illustrates the effect of heavier storm runoff experienced in land uses
with more impervious cover.
Baseflow water quality samples collected contemporaneously from two adjacent
residential canyons on central wastewater collection systems indicate that
the size of the undeveloped buffer zone around a stream may be related to
water quality. Median nitrate concentrations in these two canyons indicate
that water quality improves as buffer zone size increases. Furthermore,
impacts to pH are mitigated by larger buffer zones.
In summary, when compared to streams representing rural land use, various
parameters indicate statistically significant water quality degradation
for streams representing golf or residential land use categories.
Barton Creek Sediment Quality Studies: 1991 - 1995
Purpose:
Barton Creek sediment quality was assessed from a composite of various studies
and investigations made by Austin's Environmental Resource Management staff
to examine trends and compare contaminant levels to regulatory criteria.
Overview:
Sediment samples were gathered by five different project teams, each attempting
to detect short and long term trends in the accumulation of heavy metals,
organic pesticides and other organic constituents.
Findings:
Concentrations of sediment constituents throughout the Barton Creek watershed
are not at levels of concern with the exception of the area in and around
Barton Springs. Polycyclic aromatic hydrocarbons were detected at levels
which may have biological effects in the Barton Springs area. Although many
potential pesticides were analyzed for, very few were found in detectable
concentrations. However, one sample, immediately above Barton Springs, contained
several organochlorine pesticides above the TNRCC 85th percentile, which
is a regulatory screening level used in assessing sediment contaminants.
Observed copper, lead, and zinc concentrations are elevated in the Barton
Springs area relative to upstream sites; however, the highest chromium,
cadmium, and zinc concentrations occurred in one sample taken at an upstream
rural site. Grain size distribution indicates that higher concentrations
of constituents at downstream sites could be attributed to the deposition
of a larger percentage of fine-grain material.
Bioassessment Strategies for Nonpoint Source Polluted Creeks, Grant
Funded Project: June 1993 - August 1996
Purpose:
The major goals and objectives of the study included investigation and documentation
of current levels of physical and biological impairment in two watersheds
(Barton and Onion Creeks) with varying degrees of development, correlation
of various biological community conditions with physical and chemical indicators
of nonpoint source pollution, and development of effective long-term biological
monitoring and assessment techniques for the Central Texas region.
Overview:
Aquatic biological communities are typically sensitive to water quality
and habitat degradation. "Bioassessment" methodologies have been
developed and are now widely used which analyze these communities for use
as indicators of stream health. For the purposes of this study, "benthic
macroinvertebrates" an aquatic assemblage of snails, mayflies, stoneflies,
blackflies, caddisflies, dragonflies, etc., were examined as well as a community
of periphytic algae, the diatoms.
Following initial protocol development, project staff cataloged potential
study sites by identifying reaches with appropriate habitat and substrate
for benthic communities. After site selection, water quality, habitat, and
biological data were collected at Barton and Onion Creek study sites on
a quarterly basis for three years. Biological data were analyzed with corresponding
water chemistry and land use attributes in order to document the relationship
between the data sets.
Findings:
Development in Barton Creek is still in the early stages, with current impervious
cover estimates in the bioassessment study reach at 6 percent. Onion Creek
has impervious cover estimates of 10 percent in the study reach. The findings
of this report suggest that the macroinvertebrate community is responding
more dramatically to the water quality variation on Onion than on Barton
Creek. It is likely that creeks with higher mean levels of water column
nutrients than Barton may have a more consistent response to chemistry by
the macroinvertebrate community.
Overall, the diatom community metrics are better than the benthic macroinvertebrate
metrics at differentiating between different water chemistries and land
uses. Consistent site level variation is more common in Onion Creek than
in Barton Creek, suggesting that there is a minimum level of chemical constituent
concentrations beneath which these biological metrics cannot effectively
differentiate.
On both Barton and Onion creeks, diatom community changes are related distinctly
to watershed changes due to levels of development as indicated by land use
breakdown. On Barton Creek the diatom community is significantly responding
to the land use change from undeveloped to golf course and residential land
uses which begin downstream of Barton Creek Blvd. and continue down to Lost
Creek Blvd.
From data collected in this study, the chlorophyll a mean concentrations
are different between the land use groups on Barton Creek. Sites adjacent
to and downstream of Barton Creek Blvd. with higher levels of residential
housing and golf course land use had significantly higher chlorophyll a
and pheophytin values than sites with lower levels of each of these land
uses. However, the relationship of chlorophyll a and its surrogates
to water chemistry data was not significant, suggesting that the measure
of algal biomass through chlorophyll a is a more sensitive indicator
of nutrient enrichment from non-point source pollution than routine chemical
water quality sampling.
The radical fluctuation in flow rates during this study emphasized temporal
variation in water chemistry concentrations and minimized the influence
of spatial, or land use, differences between sites. Nonetheless, consistent
relationships were identified between developed land use and two important
water chemistry parameters. - Total Dissolved Solids and Nitrate-Nitrite
Nitrogen.
Although overall nutrient concentrations on Barton Creek were not significantly
different from upstream to downstream due to high standard deviations, all
of the highest values and highest means were recorded between Barton Creek
Blvd. and Lost Creek Blvd. In general, the macroinvertebrate data from the
Bioassessment Grant indicate that current levels of biological impairment
in Barton Creek are extremely low.
PART 2 of the Barton Creek Report Executive Summary
Report Publication Date: April 1997, 460 pages.
|
