United States
Environmental Protection
Agency
National Risk Management
Research Laboratory
Ada, OK 74820
Research and Development
EPA/600/SR-97/023
July 1997

                            of
                    to
Mark E. Jensen, Mike Lowe, and Michael Wireman
  Methods commonly used to delineate
protection zones for water-supply wells
are often not directly  applicable  for
springs.  This investigation focuses on
characterization of the hydrogeologic
setting using hydrogeologic mapping
methods to  identify  geologic and
hydrologic features that control ground-
water flow to springs to aid in delineating
protection zones. Thesetechniqueswere
applied at two public-supply springs
selected  to represent diverse geologic
settings.  One spring discharges from a
fractured carbonate system and one from
a clastic-rock aquifer.  Results of this
investigation allowed development of the
conceptual model for site hydrology and
identification of potential constraints on
ground-water flow and protective zones
at each site.    The  report discusses
results from these case studies and a
general methodology for applying these
techniques to delineation of protection
zones   around   springs.     The
hydrogeologic  conceptual  model
resulting from the  use  of these tools
provides the framework  for identifying
significant data gaps and implementing
studies  to  obtain  any  additional
information required for reliable and
practicable protection zone delineation.
  This Project Summary was developed
by  EPA's National Risk Management
Research Laboratory's Subsurface
Protection and Remediation Division,
Ada, OK,  to announce key findings of the
research projectthat is fully documented
in a separate report of the same title (see
Project Report ordering information at
back).

Introduction
  In the United States, over 3400 public
water-supply systems obtain part or all of
their drinking  water  from springs. These
systems provide water for more than seven
million  people.   One of the  six critical
elements of a wellhead protection  program
is delineation of a  scientifically  valid
protection zone around public water-supply
wells and springs. Hydrogeologic mapping
may be used  at many sites to identify
geologic and hydrologic controls on ground-
water flow to  springs and to  locate flow
boundaries.   Protection  zones  may be
specified based on flow boundaries at sites
where the zone of contribution to the spring
is practicable to protect.
   Hydrogeologic mapping,  as applied in
this project, refers predominantly to geologic,
hydrologic, geochemical, and geophysical
techniques  for characterizing  subsurface
features using the surface expression and
geophysical or geochemical signatures of
such features.   These techniques  include:
  «  fracture-trace analysis,
  «  analysis of land surface topography,
  «  geologic mapping,
  «  potentiometric surface mapping,
  •  geophysical surveys,
  •  tracer studies,

-------
  •  stable and radioactive isotope studies,
    and
  •  geochemical characterization.

Objective      Methodology
  The objective of this cooperative research
project  was  to  investigate the  use of
hydrogeologic mapping to determine the
zones of contribution to two springs and
application of this information to delineation
of potential protection zones at the field
sites.  Hydrogeologic mapping was chosen
as the primary focus forthis project because
of  its   applicability  as   an   initial
characterization step in many settings that
include strongly anisotropic aquifers, such
as  fractured  bedrock.     Additional
considerations included the relatively low
cost and low technological requirements of
many of these tools.
  The methodology applied  in  this study
integrated information  from  multiple
hydrogeologic mapping methods. An initial
conceptual model for each  site was
developed through review  of site-specific
literature, principally, documents obtained
from state and federal agencies and local
universities. Based on this model, potential
ground-water flow controls were identified.
Applicable hydrogeologic mapping methods
were chosen  to  test assumptions  of the
conceptual  model and delineate geologic
and hydrologic controls. Results from these
studies were used to locate potential ground-
water flow boundaries defining the zones of
contribution to the springs and evaluate the
effects of potential controls.  The results
were then evaluated for use in delineating
potential protection zones.  Monitoring wells
were installed at each site to provide litholog-
ic/stratigraphic information and allow limited
hydraulic testing.   Results of  these
subsurface  investigations were compared
with  information  inferred from  the
hydrogeologic mapping.
  The objective of this report  is to evaluate
the utility and limitations of the hydrogeologic
mapping  studies performed at these sites
for delineating the zones of contribution and
protection zones around the springs.  A
general methodologyforapplying such basic
characterization  techniques  is  also
discussed.  The report is  designed to aid
investigators  involved  in  planning
characterization studies  leading  to the
establishment of protection zones  around
springs.

Results and Discussion
  Two study sites were chosen to represent
different geologic settings  for springs:
fractured carbonate aquifers and  clastic-
rock aquifers.   Hydrogeologic mapping
methods  applied at these sites included
geologic mapping (stratigraphic and structur-
al), topographic analysis,  fracture-trace
analysis,  geochemical  characterization,
isotope  studies,  and  catchment  area
estimation. Geologic mapping, fracture trace
analysis, and  topographic  analysis  were
applied in an attempt to  infer the extent of
aquifers, identify lithologic units that may
act as flow boundaries or pathways, map
joint patterns or karst features that may act
as flow paths, identify faults that may act as
flow barriers or flow  paths, and  identify
potential  recharge areas.   Analyses of
ground-water chemistry and tritium activity
were used to inferaquiferlithology, potential
seasonal fluctuations in ground-water flow
paths, and relative ground-water residence
times. Estimates of the required catchment
area were calculated from spring discharge
measurements  and  potential  recharge
estimates for comparison  with other
estimates of the zone of contribution.

Carbonate Aquifer
  Olsens Spring, located in  Mantua Valley
in the Wasatch  Range  of northern Utah,
discharges about 1,700 l/min (1.0 ft3/s) from
jointed Cambrian dolomite.  West Hallings
Spring, which discharges about 6,400 l/min
(3.8 fP/s) from a faulted limestone, is located
approximately  275  m (900 ft) southwest of
Olsens  Spring within the  same  surface
drainage basin. These springs are a source
of potable water for Brigham City, which has
a population of about 17,000.  The area
surrounding the springs is mostly hilly and
mountainous, underlain by a Paleozoic-age
stratigraphic section  of interbedded
limestone, dolomite,  shale, and quartzite
that has been  tilted, faulted, and fractured
by thrust faulting and normal faulting. This
study focused on Olsens Spring. However,
West  Hallings  Spring and other springs in
Mantua Valley were considered during these
investigations, as appropriate.
  Geologic mapping  was  conducted
throughout the  surface drainage basin
surrounding West Hallings Spring  and
Olsens Spring  following analysis of fracture
traces and lineaments visible in aerial photo-
graphs.   No faults were mapped in  the
immediate area of Olsens Spring. However,
West  Hallings Spring did  appear to  be
located on afault. Ground-waterflowtothis
spring may be influenced by this feature.
Extensive north to northeast trending joints
were  observed in outcrops throughout  the
area indicating that ground-water flow may
be predominantly controlled by secondary
porosity,  transmissivity may be relatively
high,  and that the area surrounding  the
spring may be vulnerable to contamination
from  surface sources.  No  karst features
were observed during the geologic mapping.
Based on these  results,  no geologic units
sufficiently competent to behave  as
boundaries to ground-water flow could be
identified.
  Monthly analyses of physical/chemical
parameters (i.e., temperature, pH, specific
conductance, and turbidity) in water from
Olsens  Spring revealed  no significant
fluctuations and very low turbidity during a
one year period.  This indicates ground-
water residence time was sufficient to mask
any variations due to seasonal variation in
recharge. Quarterly analyses of major ions
were  used to  determine  hydrochemical
facies,  mineral saturation indices,  and
calcium/magnesium molar ratios.   No
significant shifts in hydrochemical  facies
that could be related to temporal fluctuations
in ground-water flow paths were observed.
Calcium/magnesium molar ratios were
stable  and indicated  ground-water flow
through formations composed largely  of
dolomite.  Calcite saturation indices were
generally  positive,  indicating sufficient
residence time  for saturation to occur.  A
potential temporal trend  in the calcite
saturation  index  suggested  a possible
seasonal variation in saturation that may be
related  to decreases in residence time
associated  with increased recharge.
Analyses of tritium activity indicated the
average ground-water age was less than
approximately  40 years and, potentially,
less than 20 years.
  Estimates of the catchment area required
to  support the  combined discharge from
Olsens Spring  and West Hallings Spring
ranged from 10 km2 (4 mi2) to 54 km2 (21 mi2).
The upper limit of this range is significantly
largerthan the area of the surface drainage
basin (approximately 17 km2 [6.5  mi2]),
indicating that the zone  of contribution  to
these springs   may  be  greater than the
drainage basin. However, there is significant
uncertainty in estimation of the catchment
area.
  In  order to   obtain  direct information
concerning subsurface conditions and initial
estimates of hydraulic  parameters,  two
monitoring wells were installed topographi-
cally upgradient of Olsens  Spring.  Short-
term, single-well  pumping tests were
conducted to obtain  preliminary estimates
of transmissivity nearthe wells. Transmiss-
ivity estimated from these tests ranged from
270 m2/d (2900 ft2/d)to 550 m2/d (5900 ft2/d),
supporting the conceptual model of a
relatively transmissive aquifer  near the
springs.  Results of  a tracer study also
indicated that ground-water velocity near
the springs was relatively high.
  These data  appear  to support  the
hydrogeologic conceptual model of a system
that  may be  dominated by flow through
interconnected fractures.  However,

-------
additional studies would be required to better
define  potential conduit flow components.
The evidence forthis model includes highly
fractured  bedrock  units,  stable field
parameters,  relatively  stable  water
chemistry, very low turbidity, and  relatively
young  ground water.   As  indicated by
hydrogeologic mapping and in situ studies,
transmissivity and ground-water velocity
near the springs  appear to  be  relatively
high. Significant anisotropy may  exist due
to the dominant north to northeast trending
fractures.
  Although potential  controls on ground-
water flow were  identified,  definite  flow
boundaries  were  not  located  using
hydrogeologic mapping techniques.  The
minimum zone of contribution to Olsens
Spring  and West Mailings  Spring  was
estimated to be the surface drainage basin.
The shallow flow  system within the basin
may potentially be bounded by ground-water
flow divides coincident with the topographic
divides. However, the zone of contribution
for these springs  may be larger than the
surrounding surface drainage basin.  These
data suggest that possible protection zones
around  these springs should encompass
the entire surface  drainage basin and  may
extend  beyond  the  basin  boundaries.
Additional  studies,  potentially including
installation  of a piezometric  network and
extensive aquifertesting, would be required
to better define the zone of contribution and
more reliable protection zones around these
springs.

Clastic-Rock Aquifer
  Sheep Spring is located in southwestern
Utah and discharges 7.5  l/min (2 gal/min)
from interbedded siltstone, sandstone, and
shale. This spring  is part of the Santa Clara
city water system, which serves about 1520
people,  and is representative of many low-
flowrate springs used for drinking water in
this area.   Inflow  to the  collection  tunnel
appears to discharge from  a sandstone
layer approximately 0.5 m (1.5 ft)  thick and
from joints  in units below this layer.  The
sandstone  layer consists of very fine-  to
fine-grained silty sandstone and  interbed-
ded siltstone.  Ground water discharges
both from joints  and pore spaces  in the
sandstone matrix.  This layer appears to be
transmitting much of the  ground water  to
Sheep  Spring.  Minor seeps, other small
springs, and vegetation are localized near
outcrops of this unit.
   Based on  topography and  review  of
previous regional studies,  the  zone  of
contribution extends north of Sheep Spring.
Thetopographicdrainage divide, which may
coincide with a ground-water flow divide
forming a hydrologic boundary to flow, is
approximately 24 km (15 mi) north of the
spring. An analysis of fracture traces from
aerial photographs was conducted prior to
detailed geologic mapping  of an  area of
about 4.5 km2 (1.75 mi2) within the potential
zone of contribution to Sheep  Spring.
Although  significant,  individual fracture
traces were not observed near the spring;
aerial photographs indicated  a  pervasive
north-trending fracture system exists in this
area. The bedrock is highlyjointed at Sheep
Spring and throughout the zone of contribu-
tion.  The joints strike  approximately north
and dip steeply to the east and west. These
data  indicate that  ground water may be
vulnerable to surface contamination sources
with fractures serving as pathways for rapid
transport to the aquifer.  The system  may
also  be  highly  anisotropic  due to  the
predominant north-trending joint orientation.
  A  seasonal variation in water tempera-
ture of approximately 4.5° C (8.1° F), which
appears to be related to air temperature,
was observed during  monthly analysis of
physical/chemical parameters in waterfrom
Sheep Spring. Other physical/chemical field
parameters,  including  discharge  rate,  did
not vary significantly. Water from the spring
is  a  calcium-sulfate  type  ground water
without  significant seasonal  variation to
indicate fluctuations in ground-water  flow
paths.  Average ground-water  residence
time based on  analysis of tritium activity
probably  was  greater than 40 years.
Constant discharge rate, relatively constant
physical and geochemical parameters, and
the  relatively  low tritium activity   are
indications that ground water discharging at
Sheep Spring may have a relatively  long
residence time.
  Three  monitoring wells were  installed
about 26 m (85 ft) north of the Sheep Spring
collection  tunnel.  Two of the wells were
cored through the sandstone layer identified
during the surface mapping to provide strati-
graphic control and obtain a sample for
porosity  measurements.    Hydraulic
conductivity of the formation estimated from
rising head slug tests ranged from about
0.04 m/d (0.1 ft/d) to 0.2 m/d  (0.6 ft/d). A
study conducted by previous investigators
estimated a hydraulic conductivity of 0.3 m/d
(1  ft/d) based on aquifer characteristics of
the same formation and the specific capacity
of a well located approximately 4 km (2.5 mi)
from Sheep Spring. Results from this study
and the previous work indicated hydraulic
conductivity in this formation may be relative-
ly  low.  However, more extensive testing
would be required to better define hydraulic
parameters and potential anisotropy at the
site.
  The  hydrogeologic conceptual  model
developed from these studies is one of a
spring discharging from a fractured clastic-
rock aquifer of low hydraulic conductivity
with ground water  moving through  both
primary and secondary porosity. The system
may be highly  anisotropic with preferential
flow along the north-trending joints.  Sheep
Spring appears to be a local discharge point
with a potentially large zone of contribution
for  ground water with a moderate to  long
residence time.
  This information was evaluated for use in
delineation  of  potential protection  zones
around the  spring.  Ground-water  flow
boundaries near the spring  could  not be
inferred from these studies. Hydrogeologic
mapping  indicated that the distance to the
hydraulically upgradient limit of the potential
zone of  contribution was approximately
24 km  (15 mi)  based on analysis  of  land
surface topography.  However, a relatively
low ground-water seepage velocity may limit
the zone of contribution within specified
time periods  to  a much smaller area.
Protection zones that cover the entire zone
of contribution may be too large to be
effectively managed by the water supplier.
Other approaches (e.g., ground-watertime-
of~travel estimates) may be useful in refining
the size  of the protection  zones.  For
example, using the limited hydraulic  data
that are available, ground water discharging
at the spring may have traveled less  than
approximately  660  m  (2200 ft) within a
15-yeartime frame. Additional studies would
be required to  delineate reliable protection
zones based on such techniques.

Conclusions
Recommendations
  Hydrogeologic   mapping  provided
information on aquifer characteristics
including  aquifer lithology, vulnerability to
surface  sources   of  contamination,
boundaries on average ground-water
residence times, transmissivity, potential
recharge area, and potential  geologic/
hydrologic controls on ground-water flow.
Specific findings and recommendations from
these studies are:
  1. Hydrogeologic mapping techniques are
    relatively  low cost characterization
    tools. These methods commonly will
    provide information essential for
    conceptual  model development,
    planning subsurface investigations, and
    defining protection  zones based on
    ground-water flow  boundaries.  It  is
    recommended that the utility of such
    techniques be evaluated early in all
    spring  protection  zone delineation
    projects.

-------
2.  Development of  a  conceptual  model
   for site hydrogeology is an essential
   element in the delineation process. The
   model  serves  as the focus  for
   characterization  efforts,  allowing
   important assumptions  and potential
   ground-water flow controls  to  be
   identified and studied.
3.  Applicable  hydrogeologic  mapping
   techniques depend on site conditions
   and the information gaps that are
  identified. Availabletechniques include
  geologic, hydrologic, geophysical, and
  geochemical  characterization tools.
  Mapping  of the ground-water  flow
  system using potentiometric information
  is an extremely powerful tool that should
  be used whenever data are obtainable.
4. Ground-water flow boundaries that may
  serve  as boundaries of practicable
  protection zones around springs  may
  not   be   definable  using   only
hydrogeologic  mapping techniques.
Incorporation of additional delineation
tools, such  as  ground-water time-of-
travel estimation,  and performance of
detailed subsurface investigations may
be required to define more reliable and
manageable protection zones at some
sites.
Mark E. Jensen is with the Utah Department of Environmental Quality, Salt Lake City, UT
84114-4830; Mike Lowe is with the Utah Geological Survey, Salt Lake City, UT 84109-1491;
and Michael Wireman is with the U.S. EPA, Denver, CO 80202-2405.
Steven D, Acree is the EPA Project Officer (see below).
The complete report, entitled "Investigation of Hydrogeologic Mapping to Delineate Protection
Zones Around Springs Report of Two Case Studies," ( Order No. XXXX-X; Cost: $X.OO,
subject to change)  will be available only from:
      National Technical Information Service
      5285 Port Royal Road
      Springfield, VA 22161
      Telephone: 703-487-4850
The EPA Project Officer can be contacted at:
      U.S. Environmental Protection Agency
      National Risk Management Research Laboratory
      Subsurface Protection and Remediation Division
      P.O. Box 1198
      Ada, OK 74820

-------