3fflg.nt§i£f£te.£fen..-	....	...Qffiee °f Policy- Planning
                            July 1996
An Asses:
         leal Impacts of






                  Cheryl Grantham
        Idaho Water  Resources  Research  institute
                 University of  Idaho
              Moscow,  Idaho  83844-3011
                     May,  1996


                          Table of  Contents
            Acknowledgments and Disclaimers .......................Hi
            Abstract.	 1		v

Section 1  -  Introduction			 1

Section 2  -  Methods	..............		 3

Section 3  -  Historical Background......................		 .5

            Wetlands Protection.	,	 5
            Ground Water Use.			,6
            Ground Water Overdraft...........	.....	 8

Section 4  -  Overview  of the  Ecology of Wetlands and Riparian Areas . . 13

            Ecological  Importance of Wetlands and Riparian Areas. ... .13
            Types of Ecological Impacts Resulting From Overdraft ....  15

Sections-  Hydrogeologic Mechanisms Capable  of Causing
            Ecological  Impacts  from Ground Water Overdraft  . . .  . . ... 19

            Ground Water Level Fluctuation.	 20
            Reduction of Ground Water Discharge.	22
            Induced Recharge			 .23
            Land Subsidence. — .	...-.-.	.24
            Changes in River Channel Morphology, and Stability. ...... .25
            Salt Water Intrusion.	 27
            Changes in Ground Water Geochemistry..-.._	 28

Section  6  -  Locations of  Ecological Impacts of Ground Water ,
            Overdraft on Wetlands and Riparian Areas............... 31

            International	 .32
                  England ..	.		 r.	.32
                  Northwest European Lowlands, the Netherlands .. , . 32
              ..   Spain		...	.......,.'...	34

            United States.....	"..;...,		........ 37
                  Arizona.			37
   •               California..............		 .48

                   Florida	-....„	.. 56
                   Idaho ..	..	62
                   Indiana	„.		.63
                   Nevada		... 64
                   Oklahoma	^	69
                   South Carolina	70
                   Texas	...	.71
                   Wisconsin	.		....  73

 Section  7 -  Evaluation of the Adequacy of Existing  Information  on
             Ecological Impacts of Ground Water Overdraft	75

             Technical Perspective	 76
             Regulatory Policy Perspective	78
             Legal Perspective	79

 Section  8 -  Recommendations	81

 Section  9 -  Conclusions	 .85

Section 10 - Bibliography		........87

                  Acknowledgments  and  Disclaimers
 Funding  for this project was provided  by the  Office of  Policy,  Planning, and
 Evaluation   of  the   United   States   Environmental   Protection  /Agency
 (Cooperative  Agreement   CX  822008-01).     The  author  gratefully
 acknowledges the assistance .of EPA  Project Offiqer Rodges Ankrah  in the
 conceptualization and  initiation  of  this project.  Completion of the project
 was accomplished  with assistance from John Simons,  EPA Project Officer
 with  the  Office  of  Water.    Administrative  assistance was  provided
 throughout the project  by the staff of  the  Idaho  Water Resources  Research
 Institute  of the  University of Idaho.

 The author appreciates the  comments and  suggestions  made by Paul  Jehn
 (Ground  Water  Protection  Council), Sydney Bacchus (Institute of Ecology,
 University of  Georgia),  Harriet Hill (U.S.  EPA, Region IX),  and John Simons
 (U.S. EPA).   Assistance from many  of the authors whose work is cited in this
 report  also is appreciated greatly.

 Contents  of   this  publication  do  not  necessarily  reflect the  views and
 policies of the Idaho Water  Resources Research  Institute,  nor does mention
 of trade  names or  commercial  products  constitute their endorsement  by the
 Idaho Water  Resources Research Institute.

The views expressed  by  the  author are her own and do  not  necessarily
 reflect those  of  the   United   States  Environmental  Protection  Agency.
Mention of trade names, products,  or services does riot convey, and  should
not  be  interpreted  as  conveying,  official  EPA approval,  endorsement,  or
                      -  Printed on Recycled Paper -


                         IN  THE UNITED STATES        -


                             Cheryl  Grantham

 One potential,  consequence  of ground  water overdraft  that  frequently is
 overlooked in the allocation of ground  water resources  is the  impact of
 pumping  on  surface  ecosystems dependent on  ground  water.    Many
 researchers have noted the almost anomalous dearth  of information  on the
 linkage  between hydrogeological factors and ecosystem  impacts.  As  ground
 water use continues to increase and the health  of the remaining  wetlands
 and riparian habitat comes under increasing scrutiny,  there is a  growing  need
 for an  adequate technical understanding  of this emerging issue to  form the
 basis of consistent public policy.  The-purpose of this report is to collect and
 evaluate information on  the location,  nature, and extent of  ecological  effects
 that have  been  shown to have  occurred in  wetlands  and riparian areas as a
 result of ground  water pumping.

 As transition  environments between  aquatic  and  terrestrial  ecosystems,
 wetlands and  riparian areas are among  the most spatially and temporally
 complex natural  systems on earth.  These habitats  provide food web support
 for all trophic levels as  well as sites  for breeding, nesting, rearing,  resting,
 refuge;  feeding,  and  overwintering.  It is estimated that over 70%  of the
 original  area of  riparian  ecosystems in the United  States  has been  cleared
 and  much  of the rest has been  affected by a myriad  of land and water uses
 during this century.     .

 Ground  water  overdraft  is capable  of causing ecological  impacts  in wetlands
 and  riparian areas in  a  variety  of ways.   Because the stage and duration of ,
 the natural hydroperiod  are  among the most crucial aspects affecting the
 location,  composition,  and overall health of these ecosystems,  ground water
 drawdown  that  results  in  significant hydroperiod   perturbation can  have  a
 major impact.   Examples include ground water level fluctuation in excess of
 species'  limits  of tolerance,  reduction  of ground water discharge on  either a
seasonal or long-term  basis, and induced recharge of underlying saturated
zones derived  from drainage, of surficial aquifers on which ecosystems  are

 dependent.   Overdraft  can result  in  geomorphological changes that impact
 wetland hydrology such as land subsidence and  alteration  of  river  channel
 morphology  and  stability.    Lastly,  changes  in  the  natural  geochemical
 environment  in excess of species'  tolerance  may  result  from  overdraft.
 Examples  include  salt water  intrusion as well as  other modifications of the
 chemical characteristics of  the ground water and the  associated solid media.

 Literature  on ecological  impacts from ground  water  overdraft was found for
 several  countries  as well  as ten  states.  The  nature and  extent of these
 impacts are  described  as well as pertinent geological  and hydrogeological
•characteristics of the sites.   Some impacts are localized  while others  have
 occurred on  a  regional basis  throughout an entire drainage  basin or aquifer

 The extent to which  these impacts  have  been studied  varies widely. In a few
 cases, impacts  are  reasonably well  documented.    For  the most part the
 potential  for ecological impacts  from   ground  water overdraft  is   largely
 overlooked  in   the  context  of water  allocation as  well  as  ecological
 assessment.   The  locations  of impacts described   in  this  report   are  not
 hydrologically or  ecologically  unique.   Therefore it is  probable that  the
 existing  body of literature  on  this subject significantly underestimates  the
 extent of  ecological  impacts that result  from ground  water overdraft.

 To more adequately and accurately address this  issue, it is recommended
 that studies to  identify  and quantify ecological consequences from overdraft
 be undertaken  more frequently  in  a broad array of  academic  and resource
 management settings*  including  regional  water  use  planning  programs,
 wetland  surveys,  and  regional hydrological  studies.   A  major objective
 should  be an enhanced predictive  capability for early identification of "high
 risk"  hydrologic  settings and  "high risk" wetland  and  riparian  communities
 and species.   Efforts  to assess  and mitigate impacts from ground  water
 overdraft must be  approached with a  full understanding of the  complexity of
 hydrologic  systems and  wetland ecosystems.    Many human activities and
 natural  processes  can  result  in  ecosystem changes and  identification of
 causal  relationships  requires  detailed,  site-specific,  and  usually long-term

 With  care  to avoid  erroneous oversimplification, this information  should be
 used  to  improve planning and  policy regarding  ground  water use  and  wetland
 protection.   Permitting  processes  pursuant  to  applicable federal,  state, and
 local  statutes should include an assessment of the potential  for impact  from
 subsurface drainage  for projects  that   may  affect   wetlands  and   riparian
 areas.   A major objective of the activities should be an ecosystem approach

to overall water  management in  which there is scientifically  valid feedback
between the occurrence  of ground  water overdraft, water  stress tolerance
limits  of affected species,  and allowed pumping pates.      -


                                Section  1

 Utilization of ground  water resources typically  results in  increased  capture
 of  local  recharge  and/or  withdrawal  of ground  water  from  storage.
 Withdrawal-of ground water  from  storage  is commonly  called ground  water
 mining or overdraft (Bredehoeft et al.,J982).   It is manifested  by a decline
 in the  water table,  a  reduction in  artesian  pressure or  an  irreversible
 compaction of certain  fine grained  sediments (U.S. Geological  Survey, 1984).

 There is controversy  regarding ,the definition of  overdrawing  and whether it
 should  necessarily be  avoided (Smith,  1989).    One  perspective  is  that
^overdrafting  or ground water  mining  is  "no.more unsafe than the  mining  of
 any  other  mineral  resource,  provided  it's recognized  and  planned"  (U.S.
 Geological Survey, 1984).   However,  it is  frequently the  case  that the extent
 of drawdown  is'not  planned  and  the  associated consequences  are not
 accurately anticipated.   The impacts of some pumping  may not be noticeable
 until  they are irreversible  or the consequences may  be  borne by  parties  other
 than  those receiving the benefits  of  the  water which was pumped  (House
 Committee  on Natural Resources, 1994).   At  present  many of our  legal
 institutions do not adequately address these situations pertaining to  ground
 water pumping (Bredehpeft et al.,  1982).

 One  potential  consequence of ground  water  overdraft  which is frequently
 overlooked  in  the allocation  of ground  water resources  is  the  impact of
 pumping on surface ecosystems which are  dependent  on ground water.   For
 example,  many wetlands  are  located  in ground water discharge zones  and
 many  riparian  communities  are dependent  on  shallow alluvial aquifers.   In
 arid environments  in  particular, ground  water  may be  the  only perennial
water  source  available  to  some  wetland  and  riparian  communities.    In
locations  throughout the country where  water level  decline  has  occurred in
the  course. of ground^ water  development, the  potential for   large-scale
impacts in wetland habitats is great.

In response  to a growing body of information  on  water level  decline,  the
National  Water Commission referred to ground  water  mining as^one  of the
"three  principal   problems  of  ground   water law,  management  and
administration" (Fort et  al.,  1993).    Predicting  and  preventing adverse
impacts from  ground water  im^ is  a complicated task.   Habitat  impacts

are varied  and the remedies  are  often poorly defined and diverse.   A broad
group  of participants is  involved in developing public policy  including parties
with interests in  water allocation as  well  as habitat  protection.

Many researchers have noted the almost anomalous  dearth of information on
the  linkage  between  hydrogeological   factors  and  wetland  community
functions (for example Fort et al., 1993; Busch et al., 1992; Llamas, 1989).
As  ground  water use  continues to increase and  the  health of the remaining -
wetlands and  riparian  habitat  come  under  increasing  scrutiny,  there  is a
growing need for an adequate technical understanding of this emerging issue
to form the basis of consistent public  policy.

                                Section  2

The  purpose  of  this  report is to collect  and  evaluate information on the
location,  nature and extent of  ecological impacts which have  been shown to
have  resulted from ground water; overdraft.  It  is hoped that this document
will assist those  who seek  to predict or remedy  these  impacts  in  their

There  is considerable  controversy ovqr  the legal definition of wetlands when
used  in the context of. implementing various statutes.   Wetlands as described
in this report  are  broadly and  scientifically  defined and no attempt has  been
made  to  conform  to an  individual  statutory definition.   As  used herein,  a
wetland  is  an  area  "where  saturation  with  water  (either  permanent  or
intermittent)   is  the  dominant  factor determining  the   nature  of   soil
development  and  the  types of plant and  animal communities living in the
soil,  on  its surface,  and  in  the  overlying  water"  (U.S.  Geological  Survey,
1984).   As sudh,  wetlands are transitional between  terrestrial  and aquatic
systems  where the water table is usually at or  near  the  surface or the land
is at  least intermittently  covered  by  shallow  water.   Considerable difficulty
in determining  the location  and extent  of  some  wetlands  results  from
situations in  which human activities  have  either permanently or temporarily
created an artificial hydrological regime (Tiner,  1990).

Riparian  ecosystems  are  those  flobdplain,  bottomland  and  streambank
communities  which  occur   along   watercourses,   both   perennial  and
intermittent.   Riparian areas  generally occur  entirely  within the  100-year
flopdplain of  streams  and  rivers and are  characterized by vegetation  types,
which  are adapted to  and  tolerant of relatively  high  soil moisture  conditions
(Swift,  1984).
                i                               '           ,-...-•
For this report,  existing  literature was reviewed to  collect  information on
sites  at which ecological  impacts  have  occurred  as a result of ground  water
pumping.   Locations  included in this  study  are limited  to those where
pumping of.  water supplies is,known  to  be the, primary  causal factor of
ecological  impacts.   Suspected  impacts  or sites  at  risk  generally  were
excluded  as  the  determination of causation is  a  complicated  task.   As  a
result of this  limitation and the fact that many potential sites have not  been
investigated  by , ecological  researchers,-"'ti's number  of  sites listed  in  this

report  undoubtedly  underestimates  the  extent  of  ecological  impacts
resulting from ground  water pumping.

Sites described  in  this  report are  limited further  by  excluding  ecological
impacts  from  intentional  water table  suppression  such  as  reclamation
drainage to make land suitable  for agriculture  or development.   Over the last
century many millions  of acres  have been deliberately drained for cultivation
and  development.   These activities are  not  the focus  of  this  report.   In
adddition,  wetland  and  riparian  impacts from  other sources of hydrological
modification  such as  surface water  impoundment and  channel  modification
are excluded where  a  differentiation  can be made.  For this  study, discussion
is  confined to impacts  incidental to  ground water resource development.

This  report  is organized  to  give the  reader an historical, ecological  and
hydrogeological context with regard  to  this issue.   This  information is found
in  Sections  3  and 4.   For those  interested  in  a  predictive  capability,
hydrogeological  mechanisms  capable  of  causing  ecological  impacts in
wetlands and riparian areas are  discussed  in Section  5.   Section 6  contains a
summary of documented sites at which  impacts have occurred.  An evaluation
of  the adequacy of the existing  information  as a basis for development of
public policy  is found  in Section 7.   Lastly, recommendations for future  work
are made in Section  8.

                                 Section  3

             Historical  Background  on Wetlands Protection,
             Ground Water Use,  and  Ground Water  Overdraft
 Wetlands   Protection     ,                                    •

 For much of our nation's history, wetlands of all types have been  regarded as
 areas to be  converted into  some other "more  productive use" (Wentz,  1988).
 Drainage  of  wetlands has been  seen as  a progressive, public-spirited
 enhancement  of the  natural environment designed  to  alleviate flood danger
 and reclaim land for agriculture.  Likewise,  wetland  loss has  been viewed as
 a  relatively minor consequence  of receiving  the  benefits of large surface
 water storage projects (Dugan, 1990).

 In  arid  areas, the close association  of phreatophytic plant species  wjth the
 availability  of shallow ground  water has been  understood  since the  early
 part of  this century  (Meinzer,  1927).   Because most  phreatophytic species
 are of low  economic value, the  water they transpire  has been  defined  as
 "consumptive  waste"  for  most  of this  century  (Robinson, 1958).   The
 scientific literature  on wetlands  prior  to  the  early  1970s is  filled  with
 water  conservation studies describing the advantages of removal  of wetland
 and riparian vegetation  for  the purpose  of  "salvaging evapotranspiration" for
 human  needs  (Winter,  1988).   Replacement  of  native  vegetation  with
 agricultural  crops  or  grasslands lor  grazing  was  widely undertaken,
 particularly  in  the southwest (for  exiample 'Culler et al., 1970; Heindl,  1961;
 van der Leeden, 1991).                             ,

 It  is  estimated that  more than  70% of the  original  area  of  riparian
 ecosystems in the United  States has been  cleared and  less than  5% of the
 original  riparian vegetation remains  in the  southwest (Johnson and Hajgnt,
,1984).   Wetland loss  in the midwestern  farm belt  states of Illinois,, Indiana,
 Iowa,  Michigan, Minnesota,  Ohio, and Wisconsin accounts for approximately
 one-third of all  wetland loss  in the  history  of the nation.   The  highest
 percentage  loss, 91%,  has taken  place  in-California  and the  highest loss  of
 acreage  (9,286,713 acres) has  occurred  in  Florida  (a 46%  loss).   All states
 except for Alaska, Hawaii, and  New Hampshire  have lost more than  20%  of
 their  original  wetland acreage.    The  most  significant  historical  loss  of
 wetlands  has  resulted  from agricultural practices  (87%) (National Research
 Council,  1991).

Wetlands and riparian areas were  not widely regarded as ecological  systems
with essential functions for  supporting indigenous  flora  and fauna  until  the
1960s (Wentz, 1988).  Increasingly, scientific literature has  documented  the
importance  of  wetlands  and  riparian   areas  as  integral  parts  of  the
surrounding watersheds and stream  corridors.  Their importance for primary
production  and  nutrient  cycling   for  associated  terrestrial   and  aquatic
ecosystems  has been demonstrated repeatedly  as  well  as   their  role  in
providing spawning,  nesting, rearing and  refuge  habitat for  many  species
(Crance  and Ischinger, 1989).   Management  agencies  now also consider  the
value of riparian areas in providing a vegetated buffer  against erosion and
flooding  as well  as irreplaceable recreational and aesthetic  values (Harrison
and Kellogg, 1989).                           ,

Multiple  objective water resource planning  and management decisions  today
have replaced the  more single-minded objectives of water  conservation and
land reclamation which predominated earlier in the century (Wentz,  1988).
Recent   statutes  including  reauthorized  versions  of the Water  Resources
Development  Act,  the Clean  Water Act, and  the  1990  Farm  Act which
established  the Agricultural Wetland Reserve  Program have  provided  impetus
to achieve a "no net  loss"  of  remaining wetland acreage (National Research
Council,  1991).   Regulatory efforts  have  improved  in evaluating  cumulative
impacts  on wetland  ecosystems.   Although  wetland  restoration  is  a
technically  and   politically  elusive  goal,  for the  most  part,   efforts  have
improved in recent  years  (National  Research  Council,  1991).   However,
because  wetland preservation  and  restoration frequently  are impediments  to
development, ongoing controversy  surrounds  the  desirability  of protection  in
some  circumstances  (Lehr,  1991).    For  example,  recently   proposed
Congressional  bills  seek  to  weaken the wetlands  protection  provisions
currently  included in  the Clean Water Act.   At  present, the future  direction
of wetlands  protection is  difficult  to predict.
Ground  Water  Use

The  United  States is  fortunate to  have a  vast- ground water resource.   By
volume  it is estimated that over 90% of the fresh water in the United  States
is  in  the form of ground water, a  volume equivalent  to  about 35  years  of
surface  runoff  nationwide.  Of this volume,  about  half is considered to be
extractable if  no  consideration is given to changes in stream  flow, effects
on the environment, and the cost of extraction (U.S.  Water Resources Council,

 The United States  is also a major user of ground water.   In  an international
 survey  of  countries for which  data were available,  the  United States  ranked
 second in  terms  of volume of  ground  water used (Llamas et al.,  1992).  The
 pumpage of fresh ground water in  1980 was estimated to be  about 88 billion
 gallons  per day.   This  amounted to about  10%  of  the total  natural flow
 through  all of the nation's  ground water  systems  (U.S.  Geological  Survey,
 1984).   Ground water supplies drinking water for about half of the country's
 population and about 35%  of  the water used for irrigated agriculture  (Smith,
 1989).   It is  also  a  major source  of  the  water used in  industrial processes
 and power generation.

 Ground  water withdrawals  have increased steadily and  significantly  during
 most of the twentieth  century.  As   overall  water  utilization  has increased,
 ground  water_ has  supplied  an increasingly larger  portion  Tof  total  water
 needs.    Between  1950  and  19?5  surface water  withdrawals increased
 annually at a  rate of 2% compared to  a 4%  annual increase in ground water
 pumped  (U.S.  Water  Resources  Council,   1978).    Total  ground   water
 withdrawals more  than  doubled between  1950 and  1980  (Solley  and   Pierce,
 1988).   Among  the factors responsible for the sustained  increase in   ground
 water  use are significant expansion of irrigation, water supply  requirements
 of  growing  urban  areas, particularly  those in arid  areas of the west and
 southwest, water demands  associated with energy  production, objections  to
 the  construction  of  surface  reservoirs,  and  the  fact  that  ground   water
 usually  requires  less  treatment than surface, water as a  potable   water
 supply (U.S. Geological  Survey, 1984).

Agriculture is  the  largest consumer  of water, acccounting for 83%   of the
total water consumed in 1975 (U.S.  Water Resources Council, 1978).  Perhaps
the key  driving force for the development  of  ground  water resources  during
this century has been  the development of  irrigation technology which has
permitted  a  dramatic  increase in   irrigated  acreage  nationwide  (Smith,
 1989).   Centrifugal pumps jn shallow dug wells  were commonly used  on
small tracts in the  1920s.   The more  sophisticated turbine pumps, deep well
drilling technology  and  rural electrification  which followed by the middle  of
the century brought  many more acres  into  irrigated agriculture.  Development
of  wheel  line   and  center  pivot   application  systems  reduced   labor
requirements,  further encouraging  expansion of irrigated acreage  (Collins
and Cline,  1991).  Consequently, ground water useage has increased 5  to  10
fold from 1950 to_ the  present in most heavily irrigated areas of the  nation
(Smith, 1989).   A major exception is the recent reduction in pumpage in some
areas of extreme water level decline such  as parts of the High Plains  states
which  depend  on  the  depleted  Ogallala  aquifer.    Rising  energy costs  of
greater  lifts have  provided  an  economic  incentive  for water preservation,  a

return to dryland farming  and withdrawal  of  land  from agriculture (Kromm
and White,  1986).                  -

Many states are looking to ground water to meet most of their future growth
in water use.  In a recent  survey, the most frequent ground water,availability
issues identified  by states  were water level  decline in response to intensive
pumping  (35  states) and  legal disputes arising  from increasing  competition
for available  ground  water  supplies  (26  states)  (U.S.  Geological Survey,
Ground Water  Overdraft

Declines in  water tables  and  potentiometric surfaces have  occurred in all
states to some  extent.  Areas  with declines  in excess of  40  feet in at least
one  aquifer are depicted in Figure 1 (U.S. Geological Survey, 1984).

In terms  of volume,  the  areas of greatest overdraft tend to be  relatively
localized regions where water availability falls short of water demand.   It  is
estimated that 61% of the overdraft in the western  states occurs in Arizona,
California,  Texas,  and Nebraska (U.S. Water Resources Council, 1978).  For
example, about  two-thirds  of  the ground  water withdrawn  in Arizona in 1985
was  pumped from storage (U.S.  Geological  Survey, 1990).   Similarly about
77% of the ground water  pumped in  central and  coastal  Texas was  derived
from storage.  Other  areas  in  which  ground water  is withdrawn  significantly
in excess of recharge are parts of  Oklahoma, Kansas, New Mexico, Nevada,
South Dakota, and  eastern  Colorado (U.S. Water Resources  Council, 1978;  U.S.
Geological  Survey, 1985).

Most regional generalizations  about the  extent of ground water  overdraft and
the  potential for  ecological  impacts are  inappropriate for several  reasons.
First it  is  difficult to summarize the overall  extent  of water level decline  in
an area because  many areas  are  underlain by more than one aquifer and
declines have not  occurred,  or  have not occurred to the same extent,  in all
aquifers present.  In most areas, the  most significant  declines have occurred
in a semiconfined  water bearing zone  and  not in  the  overlying  unconfined
aquifers  (U.S. Geological  Survey,  1984).  Under  some circumstances, these
instances of overdraft are less likely to result in ecological  impacts  as the
extent of  interconnection  with  surface  water bodies may be  more  limited
than with some shallow water  table aquifers.

Secondly,  extensive  water  level  decline  is not  necessary  for ecological
impacts  in  wetlands  and  riparian  preas.   Any  perturbation  of the  natural


 a) "55
,TJ o
 _ O)
 (D O;

 0)  .
 •Jo w
 •55  2
 O £
 i_  0)
 0  E
 0)  0)
 Oi ' ^B

5  a

-  8
 to  PS

 g  0)
  .  to
v-  OJ
 0) —

hvdroperiod  to which a  hydric  ecosystem  is adapted  may be  sufficient to
result in  adverse  impacts.  Three important  aspects of a wetland hydroperiod
are-  1) the depth or stage  of the fluctuating ground and surface water, 2) the
duration   of   the  fluctuating  water  levels,  and   3)  the  periodicity  or-
seasonality   of  the  water  level   fluctuations   (Bacchus,  in  press).
Consequently, determination  of the affects  of ground  water  pumping  must
consider  ecosystem  sensitivity  to alteration  of  rates  of change and  duration
of the natural hydroperiod as well as the overall  magnitude of  drawdown.

Thirdly  many of the instances of  water level decline  depicted  in  Figure  1
took  place many years  ago  during the  initial  stages of water development.
Today water  levels  in those areas are  stable and associated ecosystems  have
adapted  accordingly.  Examples  include parts of South Dakota and declines in
Iowa and the Chicago-Milwaukee area (U.S.  Geological Survey, 1985).
                                   i|  •     •       •
A  fourth  and   extremely  important  caveat  regarding identification  of
ecological impacts  in wetlands and  riparian areas resulting from  ground
water overdraft  pertains  primarily  to   the  southwest.   Interpretation of
water level  decline data is complicated  by- a widespread  phenomenon called
channel  incision  or arroyo cutting which is  believed to  be independent of
ground  water  pumping.    These dramatic  changes  in  stream  channel
morphology  have had a major impact on  wetland and riparian communities
throughout the  southwest.

Prior to  the  mid-nineteenth century,  many streams in the  southwest  were
associated  with wide  floodplain  aquifers.  These shallow  water  sources
supported extensive riparian  habitats  and numerous  headwater  wetlands.
Frequent floods  resulted  in  adequate  seed  dispersal  and   constant
replenishment of fertile alluvium.   Mature  hardwoods  forests  as  well as
extensive marsh  vegetation  were   dependent  on these  shallow  alluvial
 aquifers (Hendrickson and Minckley, 1984).

 Between 1865 and 1915,  a regional decline in water tables which  is thought
 to be predominantly unrelated  to  ground water  pumping  occurred throughout
 the  southwest  (Betancourt,  1990).    A  combination   of  human  impacts
 Including stream ditching and draining, timber harvest from riparian  zones
 and uplands, and  excessive cattle grazing  is believed to  have interacted with
 drought  and floods to cause rapid  and  widespread  channel incision and
 headward erosion of watercourses throughout the  region (Stromberg,  1994).
 With smaller flows  confined to  vertical walled   channels,  alluvial  water
 tables declined  many  tens  of feet in  some  locations.   Many  valleys  which
 were previously swampy  or which had ground water  within 10  feet  of the
 surface  were drained  (Bryan,  1928jr  -Marshes and grasslands were  widely

eliminated  and  ultimately  riparian   forests  were  diminished  as   welJ
(Hendrickson and  Minckley, 1984).

In many cases  these ecological changes resulting  from channel incision  were
initiated  or  became  evident during periods of development  of  alluvial  ground
water   resources  for   agricultural   and   Domestic  water  supplies.
Differentiation of causal factors  is  difficult  in  many  regions.

In summary, it is  clear that in most cases regional generalizations about the
impact of ground water overdraft are not advised.   The  hydrology,  water and
land  use history,  and  ecological characteristics of  each  location  must  be
evaluated  individually,    Many instances  of ecological impact  are  highly
localized and  result from  very  minor perturbations of  natural ground  water
levels,  On  the other hand, some occurrences of water level  decline may  have
little or,no  impact on wetland and  riparian ecosystems.   Sections  4  and  5
provide more  detail  on the hydrological  and ecological  considerations which,
contribute  to  an  increased  potential  for  wetland  and  riparian  impacts.
Detailed  descriptions of the  locations and  the ecological  consequences of
ground water overdraft are found in  Section  6.


                                Section  4

                 Overview  of  the  Ecology  of  Wetlands
                           and  Riparian Areas

 Ecological  Importance  of Wetlands  and  Riparian  Areas

 An  understanding  of the  ecological processes  and  functions that control
 wetland ecosystems is  necessary to provide protection  for  these critical
 areas,     As  transition   environments  between  aquatic  and  terrestrial
 ecosystems,  wetlands are among the most  spatially and temporally complex
 natural  systems  on earth (Richter and  Richter,  1992).  And  perhaps  most
 importantly, the ecological importance of an  individual wetland is  a function
 of  the  presence of other wetlands  (Swanson, 1988).   Therefore,  in many
 regions in which  wetlands are naturally limited  or in  which  extensive  loss
 and alteration  have taken a toll on  predevelopment acreage,  the remaining
 wetland areas  are  absolutely critical for  many plant and animal species.

 Freshwater  wetlands,  although  subject to controversy  with   regard  to
 definition, generally may be  divided into three categories.   Riverine wetlands
 include  those  associated  with  both  perennial and  seasonal  watercourses.
 Lacustrine  wetlands are associated  with permanent and seasonal  lakes  and
 ponds.   Palustrine  wetlands include emergent ecosystems such as marshes,
 wet meadows,  springs,  potholes,  and  fens  as  well  as  forested  wetland
 ecosystems (Dugan, 1990).   Examples range  from small isolated depressional
 wetlands  such as  glacial potholes or karst sinkholes to  regional  features
 such  as poorly drained  low relief  areas  like the  Florida Everglades  (Brown
 and Sullivan, 1988).

 Recent  research  on  wetland  ecology  has  stressed the  importance  of
 understanding the  functional values  of wetlands.   Going beyond  traditional
 areas of  interest such  as  species  composition  and  community  structure,
 emphasis  on wetlands  functions  enables researchers  to evaluate  impacts  if
 wetlands  are  eliminated or disturbed  (National  Research Council, 1991;
 Brinson, 1993).   From an  ecological  perspective, the most important  function
 is  providing "food  web   support"  for  associated  terrestrial  and  aquatic
ecosystems  including  primary  production  and  nutrient  cycling.   Many
wetlands are among  the  most productive  of natural ecosystems,  exceeding
the  best  agricultural  lands and  rivaling the production  of  tropical  rain
forests  (National Research Council,  1991).    Other  important  functions for
higher trophic  levels include sites for  breeding,  nesting,  rearing,  resting,

refuge, feeding, and  overwintering.    Healthy  wetlands are  important in
maintaining regional biodiversity.                   ,           -

Beyond  a  biological  significance, wetlands  are  important for hydrologic
functions   including flood  conveyance,  erosion  protection, ground  water
recharge,  and potential  water supplies.   They also can contribute to  improved
water   quality  by  removing  excess   sediment,   nutrients  and  other
contaminants  from surface  runoff and  ground  water  recharge.    Lastly,
wetlands  are  of significant economic  value for  timber harvest, development
sites  and intrinsic  aesthetic  characteristics.

It  is beyond the scope of this  report to provide a thorough  review of the
broad  topic of the  ecological importance of wetlands.   Instead some of the
more  crucial roles  will  be highlighted for the  purpose of illustration.

One of the most   important types of  wetlands  is forested  areas in regions
which  otherwise have few  trees.   Riparian areas are the  only native  forested
environments in the Great Plains  (Segelquist  et al.,  1993)  and  most desert
areas  (Walters et al., 1980).   Overstory canopies provide perches,  nest  sites
and protection for  birds in these  locations  where they  would  not otherwise
be  available.   Foliage,  flowers, seeds, and  fruits support insects, birds,  and
mammals  '(England  et al.,  1984).   Therefore, loss of forested  riparian  areas is
significant for  all   trophic  levels.

Waterfowl rely on wetlands for breeding  grounds,  winter  feeding,   and
feeding and resting sites in  migration  corridors.   Decline  of duck populations
has been  linked directly to  drainage  and  degradation of wetlands  (Bellrose
and Trudeau, 1988).

In  general, the ecology of  mammals within  wetland ecosystems is poorly
known relative  to that in other environments.  Mammals  are  less likely to be
obligatory  inhabitants   of  wetlands   than  birds  and   other  vertebrates.
However,  it is clear that  a  diversity  of mammals engages in opportunistic
exploitation  of wetlands  for diet,  cover,  and  travel  corridors   (Fritzell,

Southwestern  riparian  systems support  some  of the  richest biotas  in  North
America.   At the  same time, these  areas  are some of the  world's  most
endangered ecosystems because 70  to  95%  have been lost  (Johnson  and
Haight, 1984).   Plant species found in these systems  are much less drought
resistant  than  surrounding   desert flora and  as  such  are  vulnerable to
hydrologica!  disturbance  (Walters  et al.,  1980).  -In  turn,   southwestern
riparian   habitats   support  relatively  large  and  diverse  populations  of

mammals,  other vertebrates,  flowering  plants  and  insects  (Johnson  and
Haight,  1984) as well as the highest density of noncolonial  nesting birds in
the United States (Carothers et  al.,  1974).   In  particular, cotton wood/willow
habitat  and  mesquite bosques  (forests) are  extremely important for birds
and other animals (England  et al., 1984).   Mesquite  bosques  were formerly
the most  abundant riparian  type in the southwest  and now are reduced  to
relatively  small  isolated  remnants,  virtually  none  9*  which  remain  in
pristine  condition  (Stromberg,  1993b)~                       •  ].

One of the  most  interesting  facts  which  emerges  from a  review  of  the
literature  on  the ecological  significance  of  wetlands  and riparian  areas is
their inordinate importance for rare and endangered species.  Almost 35% of
all rare and  endangered  animal  species  either inhabit wetland  areas  or  are
dependent on  them,  although  wetlands  constitute only about 5%  of  the
nation's lands (National Research Council, 1991).  .Over 50% of the federally
designated animal species are wetland  related and 28% of the listed plant
species are  wetland  dependent (Niering,  1988).
        i        .      •                          .          " (    •   • '
Reasons  for this relationship between  endangered  species and  wetlands
pertain  to the  nature  of wetlands  as well as the extensive  loss of wetland
acreage.  Many wetlands  are small in size and  lack surface water connection
to other bodies  of flowing water.   Thus small populations  of endemic species
have  a high degree  of  specialization and  are vulnerable  in  the  event  of
wetland disturbance.  Second,  wetlands in arid  areas,  particularly  in  the
southwest, are  refugia for Tertiary  and Recent  species which evolved  during
a  much wetter climate.  These isolated  wetland areas are often  remnants of
pluvial lakes  and support species  which  are unable to survive  elsewhere  in
current  post-pluvial  conditions  (Hendrickson and  Minckley, 1984).  As a
result,  many  fish and mollusk species depend  entirely on isolated  wetlands
in the southwest.  In addition to  development-related habitat loss,  this  role
of wetlands as  refugia accounts  for the fact that  about 60% of the federally
listed  fish  species  are found in the  desert  southwest (Williams  and Sada,
1985; Williams etal., 1989).
Types  of  Ecological  Impacts  Which  May Result  from  Ground  Water

Riparian wetlands are now considered  to  be the most modified land type in
the western United States  and have undergone major changes in most of the
other regions of the country as well (Jackson  and Patten, 1988).  Historically
the  most  destructive alterations  of  wetlands  have been  associated  with
changes  in  the  hydrologic  characteristics which support  the  wetland

ecosystem (National Research  Council,  1991).   In  many  wetlands  the
hydroperiod is  the single  most  important factor which  determines  species
distribution and  ecosystem health.   Systems supported  by ground water for
some or all of the year are therefore  extremely susceptible  to  impact  from
unnatural  changes in  water level.

A  broad   array  of  ecological  impacts  may  result  from  ground water
withdrawals.  Generalized examples are  discussed  here for  the  purpose of
illustration  as it  is  beyond the  scope of this report  to review this subject in
species-specific detail.  It is important  to  remember that ecological changes
occur along a continuum (Stromberg, 1994).  Some absolute thresholds  can be
recognized such as maximum .ground water  depths  beyond which a species
will not grow.  But most changes are gradual such as the ongoing decline of a
species  or a  gradual   reduction  in   disease  resistance  of  individuals
experiencing water  stress.  In addition,  there is often  a  time lag of as much
as a  decade before the effects of water stress are  noticeable.  Furthermore,
many of the impacts discussed  below can be caused by the other sources of
anthropogenic disturbance  which are common in many wetlands today as  well
as natural  cycles of drought  and disease.   Consequently it is often far  from
straightforward  to  identify  impacts from  ground water  overdraft.

Ground  water  is  a  major  determinant  of  riparian  vegetation  abundance,
community structure,  species  composition,  and   population   health
(Stromberg, 1994).   Riparian and  wetland ecosystems  undergo  changes in
response  to water stress in a hierarchical  fashion.  High levels of stress
cause ecosystem and community  level  changes and  less severe stress  can
evoke reponses  in individuals of  the least  tolerant  species  (Stromberg,
1992).  Among the most common impacts  seen in wetland areas in response
to hydroperiod  perturbation  are changes in  species composition,  species
distribution and  wetland extent (Stomberg, 1994; Harding,  1993).  A typical
response  may involve the  loss of obligate wetland  species followed  by  loss
of facultative  riparian vegetation  species.   Hydroriparian  and  mesoriparian
species may be replaced by  xeroriparian and upland species.   Nuisance
species which are  more tolerant  of  abnormal  water  fluctuations may  become
established.  On the  other hand, in some ecosystems water stress results in
ongoing  impoverishment  of the  whole  community  rather than  promoting
succession to  drought tolerant  species  (Walters et al., 1980).

As the  competitive balance  is disrupted in  water stressed wetlands, native
species will frequently be replaced by opportunistic  drought tolerant exotic
species which  may be less desirable  with  respect  to ecosystem function.
The  extensive  proliferation  of saltcedar coupled with  reduced  survivorship

 of  cottonwood and  willow  seedlings  in water  stressed  riparian  areas in  the
 southwest is an  example (Stromberg, 1994).

 Structural characteristics  of most plant  species  are  influenced  by  water
 level fluctuations.   Tree height and  morphology can  be altered  drastically.
 When ground  water  is  deeper  or not  available,  trees must  invest  more
 resources into  root  production  and  therefore  are  shorter  with  less dense
 canopies  (Stromberg,  1994).   For example,  mesquite will occur as a tree  if
 shallow ground ^ water is available  but  populations will  tend  toward a  shrub
 morphology if  ground water declines below  about  5 to  15  m  for  significant
 periods of time (Stromberg, 1993b).  In fact, most major attributes of plants
 can be  influenced  by  water availability  including  total  biomass,  lifespan,
 vegetation volume, leaf size, basal  area, root mass and depth of penetration,
 investment  in  reproductive structures,  and susceptibiity to  disease.    Since
 many of these parameters are also indicators of  insect  and avian abundance,
 impacts  from  water  stress  are felt  throughout the ecosystem (Stromberg,

 Other ecosystem  responses to  anthropogenic ground^ water  level fluctuations
 result from changes  in soil  parameters.   Of  primary  importance in  some
 areas is  an increased susceptibility to fire  (Rochow,  1994; Harding, 1993)
 Wetland  soil which  is  high in  organic  matter is  very  combustible  when
 desiccated.   Furthermore,  with  reduced  soil  moisture,  mineralization  of
 nutrients  from  decaying organic  matter  is  drastically  reduced causing  a
 decline  in soil fertility and  productivity (Lieuranpe et al., 1994).   In areas-
 where water level decline  has resulted  in  surface soil subsidence,  tree  roots
 may become exposed.  In the case  of pondcypress roots  which are adapted to
 anaerobic  conditions  characteristic of  organic  soil  and surface saturation,
 exposure  to air can result in death of  the  tree (Bacchus,  1995).

 One of  the key parameters which is indicative of the health  of an  ecosystem
 is  the age class diversity.   Seed germination  and seedling survival  are  often
 indicators  of  the ability of populations  to  sustain  themselves.   In some
 areas, seedlings will not germinate without alluvial  ground water  as rainfall
 alone is  insufficient  (Segelquist et  al.,   1993).   Frequently,  seedlings  of
 riparian  species  are  more sensitive  to  water level  decline  than  mature
 individuals  with  well  developed  root  systems.     Alternatively,  some
 streamside seedlings  have been shown to utilize surface water while  mature
 individuals have  evolved to use  a presumably  more seasonally dependable
 ground  water  supply  (Dawson  and  Ehleringer,  1991).  Thus  perturbations in
 ground  water  levels  can  be differentially detrimental  at various  stages  of
development for certain  species.  Knowledge of these variables is  necessary
to  interpret observed  impacts.

The  seasonal cycle  of  anthropogenic ground  water fluctuations-can  have a
great influence  on the extent of ecological impacts.   Spring drawdown can be
detrimental   to  seed   germination   as   well  as   populations   of
macroinvertebrates  and  their  associated  predator  species  (Riley  and
Bookhout, 1990).  Summer drawdown can  be harmful as alternative water
sources  such  as surface  water  or  precipitation  are  least  available to
compensate  for  the loss  of  ground  water  during  the summer  months
(McKnight, 1992; Lewis and  Burgy, 1964).

Much of the  preceding  discussion has focused on  impacts to plant species.
Impacts of water level decline are  found  at all trophic levels (Loftus et a!.,
1992).  Often the elimination or  degradation of one  plant species  will have
an impact on the entire  floral and faunal assemblages.

                               Section  5

                Hydrogeologic Mechanisms  Capable of
                   Causing  Ecological  impacts  from
                        Ground  Water .Overdraft

Wetlands  occur  in  geologic  and  hydrologic  settings which  enhance the
accumulation or  retention  of  water.   Water sources  may  include  surface
water,  ground  water,  precipitation  or  anthropogenic  sources  such  as
irrigation or  wastewater disposal.   Two  key aspects of a  landscape need to
be considered to understand wetland hydrology: 1) the shape  and hydraulic
characteristics of the land surface which  affect movement of  water across
it  and  2)  the  geologic  boundaries and  hydraulic   characteristics  of the
subsurface  which affect ground water flow systems (Winter,  1988).   Ground
water dependent wetlands result from the  interaction of  these two  sets of
surfaces.   Examples include  locations  were the  water table  intersects the
land  surface  such as at breaks in surface slope,  surface watercourses and
water table mounds as well as artesian discharge at springs and seeps.

The   natural hydroperiod  is  among  the  most  important factors  which
determine the  location,  composition  and  health  of  wetlands   and  riparian
ecosystems.  To preserve  the natural  hydroperiod, the sustainable yield  for
allowable  pumping  must  be  based on  the season,  rate,  and location of
pumping as  well  as  the  magnitude  of  the withdrawal  (Dingman,  1994).
Ground  water pumping which  results in changes in excess of  species'  limits
of tolerance in any of these  parameters will result in some  degree of
ecological impact.

In this  section  the  various  hydrogeological  changes  which result  from
pumping will  be described in terms of  the  ways in which  they  may cause in
ecological   impacts  in  wetlands  and  riparian   areas.     Hydroperiod
perturbations may result from changes  in the seasonally,  rate,  and extent of
natural  fluctuations  in the following:  1)  the water  table and  capillary rise in
the vadose  zone, _2)'surface  water  flow,  and 3) artesian  discharge.  In
addition,  alteration of  natural  hydrologic conditions  by ground  water
pumping may cause or  be accompanied  by other changes  in the physical
environment.   Overdraft can  result  in  geomorphological  changes which
impact  wetland  hydrology  such  .as land  subsidence  and alteration  of river

channel morphology and stability.  Lastly, changes in the natural geochemical
environment  in  excess of species'  tolerance may occur  including alteration
of equilibria  between  the  ground water and the associated solid  media as
well as salt water intrusion.
Ground  Water  Level Fluctuation

In  riparian settings,  alluvial  soil  moisture  is determined  by  a  complex
•interaction  of  channel geometry, river stage and discharge, precipitation and
alluvial ground water dynamics  (Segelquist et al.,  1993).   A growing body  of
evidence suggests that riparian  ground water is  a primary source  of  water
for many riparian  plant and tree species (Busch et al., 1992).   Ground  water
pumping may  lower  the  water level beneath  the  depth  of root penetration
either  temporarily or  permanently, thereby subjecting  riparian  vegetation  to
lethal or sublethal water stress.

Riparian vegetation  may  be  particularly  vulnerable  to   impact  from   water
level  decline   for several  reasons pertaining  to  the hydrolbgic  variability
characteristic  of  riparian  settings.   In  arid and  mesic environments, alluvial
ground water  may be the only  water supply available during summer periods
of  base flow.    Consequently,  even  a  small  change   in riparian   water
availability can  have  a   pronounced   impact   on  riparian  ecosystems
(Stromberg, 1993a).   Similarly, periodic flood flows deposit seeds  of  some
riparian species  such as Fremont cottonwood  and Goodding  willow in  high
fioodplain  settings distant from the active river channel.   Seedling survival
may thus  become  entirely dependent on, shallow  riparian  ground   water
(Stromberg, 1993c).   In  addition,  the  coarse -alluvium which  is  common  in
many riparian  environments  has a low water retention capacity  and reduced
capillarity,  predisposing these  environments to water  stress (Mahoney and
Rood, 1992).

The rate,  duration,  seasonality and  magnitude of  hydroperiod perturbations
are all  important in  determining  the  effect of ground  water  withdrawal on
riparian and  wetland ecosystems.   Frequently,  ground  water  pumping  is
greatest during  the  dry  summer  months,  particularly if  ground   water
withdrawal  is  needed for agricultural  supplies.   When pumping  ceases  or
declines  after the growing season, water levels may recover.  This  seasonal
pattern  can  be  detrimental  to riparian ecosystems because  recharge from
base  flow and  precipitation  are  at an  annual low   and   plant   water
requirements are highest  during periods of greatest pumping.

 Many phreatophytjc species  are  capable of rapid  root growth in  response, to  a
 seasonally declining water  table.   Water level  decline  within these species-
 specific limits may not be  harmful and  in  fact,  may promote  extensive root
 growth.  For example, seedling  survival  of  plains cottonwoods  was stiown to
 be highest at a drawdown  rate  of  0.4 cm/day and decreased with  increasing
 rates  of  water  level  decline  (Segelquist  et  al.,  1993).    Likewise,  the
 maximum  depth  of  root penetration varies by species and is a major factor in
 determining  the  ability of a species to  withstand  a  declining water table.  A
 related variable  in determining  riparian  survivorship  is  the  soil, texture.
 Coarse alluvial soils  enhance  root  penetration but are also readily  drained
 (Mahoney  and Rood, 1992).

 Time  lags oh the order  of  years to decades  may occur as impacted  riparian
 ecosystems adjust to reduced  soil  moisture  conditions.    This is because
 mortality  often  occurs  episodically  and  because  individuals  within  a
 population may  vary  in  their tolerance  to  water stress (Stromberg,  1993a).
 Thus  the  effects of water level decline  may  be  unnoticed until considerable
 impact has  occurred.

 Numerous examples of riparian  and wetland impacts from  ground water level
 fluctuation are described in  Section  6.  One brief example  will  be  mentioned
 here  as an illustration.  Portions  of  the alluvial  aquifer of the  Carmel  River
 Valley (California)  have experienced as  much  as 10 m of  drawdown resulting
 from pumping for municipal  water  supplies.  When combined  with a two  year
 drought   which   virtually  eliminated  annual  recharge   Trom  river  flow,
: phreatophyte  mortality  in  the  vicinity  of the  well fields  was  extensive.
 Downstream  of  the  well  fields,   aquifer  drawdown  was  minimal  and
 vegetation  experiencing the  same  weather-related  conditions  remained
'healthy (Kondolf and Curry,  1986).

 Adverse impacts  also may occur when ecosystems have successfully adapted
 to anthropogenic  ground water level  decline and pumping  is  then stopped  or
 reduced.   As the water  level  recovers,  surface or subsurface inundation  may
 be  injurious to , root  tissue adapted to  an aerobic, lower  moisture  regime.
 Groeneveld  (1989)  identified root  tissue  damage resulting from rapid water
 table  fluctuations  associated with  changes in ground  water pumping rates.
 Likewise,  when  pumping ceases in dewatered  mining  and  construction  pits,
 recovery  of water  levels, has  the potential for  similar adverse impacts on
 any wetlands in  the impacted area.

Reduction of  Ground Water Discharge

Ground water discharges to the land surface as a result of gravity,  artesian
pressure, topography,  and structural  and  stratigraphic geologic  features.
Ground water discharge zones  typically support important  ecosystems  such
as gaining reaches of  rivers and-their associated riparian areas.  Other  more
localized  discharge  features include a  wide variety  of  wetlands,  marshes,
springs,  and seeps.   Ground water discharge may be  the  sole  means of
sustenance  for  some of  these  settings,  particularly  in arid  environments
where  precipitation  and  surface   discharge  may  be   intermittent  or
nonexistent during much of the growing season.

Anthropogenic withdrawal  of ground  water may reduce natural discharge to
these types of ecosystems  in several  ways.   Excess pumping of ground water
in floodplain  aquifers may  reduce  the baseflow in  rivers to the detriment of
aquatic and  riparian ecosystems.    (Baseflow  is that portion  of the  annual
discharge of a watercourse which  is derived from  ground  water storage or
other delayed sources  (Hall,  1968)).  The  hydraulic  gradient may even  be
reversed resulting in a previously gaining reach  becoming a  losing reach.

Such is  the case in portions  of the  lower Carmel  River in California.   After
spring runoff  discharge events, the river  is  sustained primarily  by  bank
storage  from the  alluvial  aquifer.   Timing  of baseflow  contributions is
critical to downstream  migration of  steelhead trout  smolt and success of
willow seedlings.   In recent years,  localized pumping  of the alluvial  aquifer
for municipal water supplies  has  been  shown to  reverse  the  hydraulic
gradient  in  some reaches  and  deprive the river of  a volume of  baseflow
which was directly proportional to the rate of pumping  (Kondolf et al., 1987).
Furthermore, in  late  summer in  1982  when  pumping was  coupled  with
prolonged drought, the river dried up  completely in the pumped reach and  re-
emerged downstream.

Ground water pumping also may  reduce or eliminate discharge  at  springs,
seeps, and wetlands.  This has occurred  on a widespread  basis throughout the
southwest at localized  headwater  wetlands  called  cienegas  (Stromberg,
1994)  and at many former sites  of artesian  springs.   Reduction of spring
flow  is a serious  concern  at the major springs fed by  the Edwards Aquifer in
Texas (Longley,  1992) and at  springs  in similar  settings  in  the  southeast
(House Committee on Uatural Resources, 1994).  Likewise, seeps, springs and
marshes in the .Humboldt  River Basin  (Nevada) are threatened by  ground
water pumping  for dewatering  gold  mining pits (Bureau of Land Management,

 As was mentioned  in the  previous section,  the  extent of ecological  impacts
 from a reduction  of ground water  discharge  is determined by  the rate,
 duration,  seasonally and  magnitude of the  reduced  flow  as well  as various
 ecological  characteristics of the  wetland and  riparian communities.   In  many
 cases,  hydrophytic vegetation  is rapidly replaced by  species which  are more
 tolerant of water stress,  resulting in a  loss  of food sources, shelter and
 nesting sites for animals which rely on wetland areas.-
 Induced  Recharge

 In  many  areas ground  water  in  confined  or  semiconfined  aquifers is under
 sufficient  artesian  pressure  to  discharge  vertically  to  ah   overlying
 unconfined  aquifer or  to  discharge  to the surface in the  form  of springs,
 wetlands  or a surface water  body.   If  pumping  of the  confined  aquifer
 exceeds recharge, the  hydraulic pressure  in the  confined  system  may  be
 reduced to  the point that surface discharge ceases.   Further pumping may
 reverse the vertical gradient causing  the  unconfined  aquifer to recharge the
 underlying  pumped  aquifer.   If recharge to  the   unconfined  aquifer  is
 insufficient,  the water  table will  decline.   This  sequence is called induced
 recharge  (Fetter,  1988).   Clearly any riparian  or wetland ecosystems which
 are  dependent on the  artesian discharge  or  the  unconfined shallow  ground
 water will be  adversely impacted if  induced recharge  is  significant.

 Certain  geologic  settings  predispose  wetlands  to  being vulnerable  to
 hydroperiod  alteration  from induced  recharge.   In  general,  any discontinuity
 in a confining  layer will convey downward  recharge  in the  event  of  excess
 pumping  of  the  confined  aquifer.    Examples  include  lenses  of   highly
 permeable materials such_ as glacial  or alluvial  sand  and  gravel.   Structural
 discontinuities  such as fractures,  faults or karst features such  as sinkholes
 also serve as  conduits for  downward  recharge.  Geophysical methods such  as
 ground  penetrating radar  can  be  used to identify structural features  which
 present a risk  to  wetlands in the event of  excess pumping  (Bacchus,  1994;
 Bacchus,  1995).                                          ,

 Documented  examples  of wetland impacts from  induced recharge are  not
 abundant  although  the phenomenon  is  not  uncommon.   Pondcypress  in
 depressional  wetlands" in  the  vicinity  of  municipal well fields in southwest
 Florida  have  been  adversely  impacted  by induced  recharge derived from
surficial saturated  zones.  Other  ecological  impacts identified in  wetlands in
this  area  include  succession to  upland  plant species, soil  subsidence and
increased  susceptibility  to  fire  (Rqchow and Rhinesmith, 1991).

Las Vegas Valley  (Nevada)  provides another example.  Development of ground
water  resouces has resulted  in over 300 ft of water level decline.   Artesian
springs which were  abundant in the nineteenth century have been eliminated
and the vertical hydraulic gradient has been reversed in  some  areas.  As a
result, the shallow aquifer  which previously had  supported  marsh  vegetation
in  portions  of  the valley now recharges the underlying  semiconfined aquifer
(Katzer and Brothers,  1988).
Land  Subsidence

Ground water  decline  can result  in  land  subsidence which may adversely
affect  riparian  and  wetland ecosystems in several  ways.   Subsidence  can
occur  either surficially  or in  the  subsurface.   Surface  subsidence  typically
occurs  when  moist highly  organic soils are  subjected  to  drying  conditions
such as would result from water table decline or loss  of artesian flow.   With
exposure  to the  atmosphere,  the organic matter is oxidized and  surface
compaction may  result.   Wetland tree species  such as  pondcypress  are
particularly vulnerable to  damage from surface subsidence.   Tree bases  and
roots which are adapted to  an anaerobic soil environment become exposed to
air  and susceptible  to fungal  pathogens.   The resulting  decay may result in
the death of the tree (Bacchus,  1995).  Loss of organic soils at rates of up to
15  cm  within  the  first year  of ground  water withdrawal  is  not  uncommon
from this type of  subsidence in the Southeastern Coastal  Plain  (Bacchus,

Subsurface  subsidence is  by far  more  areally extensive.   It occurs  when
ground  water  is  withdrawn from  unconsolidated  or  poorly  consolidated  fine
grained sediments, typically clay.   Collapse of  the molecular structure of the
clay minerals  results in  permanent compaction of the dewatered  materials.
Although  irreversible once it  has occurred,  compaction will  cease  when
further ground  water decline  is halted (Freeze and Cherry,  1979).   Another
type  of  subsurface subsidence  occurs  in  the form  of  collapse   from
dissolution of  karst  features (Bacchus,  1995).

The areal  extent  and elevation  loss  resulting from land  subsidence  can be
considerable.  The maximum  subsidence recorded  in the United States  is 29
ft  measured  on  the  west side  of  the  San  Joaquin   Valley  (California).
California  also  ranks  first in  terms  of statewide  total  area affected by
subsidence induced  by  ground water withdrawal with over 6,200 rni2.  Texas
and Arizona follow with  4,600  mi2 and  1,000  mi2 respectively  (Poland,
1981).    For perspective, it is  interesting  to  note that  the  total volume of
aquifer storage lost as  a result of aquifer  compaction in California's Central

Valley alone  is: half the  man-made surface  storage capacity  statewide
(Conniff,  1993).  Management of this source of impact is difficult because
laws  governing  liability for  subsidence are  not  settled in  most  jurisdictions
(Kopper and Finiayson, 1981).

Many areas  of subsidence  are  in  coastal  regions  where  ground  water  is
derived  from alluvial  and shallow marine sediments  (Johnson,  1981).   As a
result, coastal  wetland ecosystems  are at  increased risk due  to salt  water
intrusion  and  erosion and  inundation from  tides and storms  (U.S.  Water
Resources  Council,  1978).   For example,  since  1943, several thousand  acres
of bayfront property  have  been  submerged into  tidal reaches of  Galveston
Bay, Texas (Neighbors, 1981) and similar  impacts  have been observed .around
southern  San  Francisco Bay, California (Fowler, 1981).  Subsidence of  about
450  mi2  in the Delta area  formed at the  confluence  of  the San Joaquin and
Sacramento Rivers  (California) has  resulted  in  submergence  of  islands to a
depth of 10 to 20  ft below  sea level (Bertoldi, 1992).   Specially adapted
intertidal communities  such as  salt  marsh  vegetation can  be  eliminated
when subsidence alters tidal elevations (National  Research  Council,  1991).

Subsidence also  can  alter or  reverse  surface  water drainage  patterns,
adversely affecting  riparian  and  wetland  ecosystems which  are  adapted to
the characteristics  of a particular watercourse.   Examples have  been cited in
Galveston Bay (U.S. Water Resources Council, 1978) and the  San Joaquin River
basin (Kopper and Finlayson, 1981).   Subsidence can lead to  reduction  of the
gradient  of a  drainage  basin.   Riparian ecosystems are  then  exposed to
repeated  flooding such as has occurred in the lower Santa Cruz River basin in
Arizona  (Schumann et al., 1986).  Where drainages  cross the periphery of a
subsiding  basin,  the  river  channel  gradient  is  increased  resulting  in
accelerated erosion.   Removal of  topsoil and deepening of the stream channel
may also adversely affect riparian ecosystems (Schumann et al., 1986).

Lastly, subsidence can result in  surface  features  such as earth  fissures and
sinkholes.   These depressions can accelerate  erosion  and  capture surface
flow  and  thereby   alter  the  hydrologic  characteristics to  which riparian
ecosystems are  adapted (Newton, 1981).
Changes  in River  Channel Morphology  and  Stability

River channel  morphology  and stability  are  determined by  a complex and
dynamic equilibrium  between many aspects  of a drainage basin  including the
soil and bedrock characteristics, the  slope  of the watercourse, and seasonal
discharge  p£i'ems  including  -flooding  cycles  (Bloom,  1978).   Riparian

 vegetation is adapted to these factors and  changes  outside species' specific
 ranges of tolerance can  adversely impact entire  riparian ecosystems.

 Ground  water flow in shallow riparian aquifers plays an  important role in
 riverine  hydrology  and geomorphology.   Excess  pumping of alluvial  aquifers
 can deprive  riparian vegetation of an adequate  water supply, particularly in
 the hot summer months during periods of  low river flow.   Loss  of  the
 stabilizing  effect of plant roots can greatly increase  erosion of river  banks
 resulting  in  temporary or permanant  loss of  riparian  habitat.   Subsequently
.eroded sediment can  be redeposited  farther downstream with adverse  effect
 on plant species which may not be adapted to depositional  environments.  In
 general, the  relative  importance  of  vegetation  for bank stability  is greater
 for smaller  streams but can  be critical  for  all watercourses (Kondolf and
 Curry, 1986).

 The Lower Carmel River (California) is an example of the scenario described
 above.   Extensive pumping  from  the alluvial  aquifer in recent  decades has
 resulted in  massive  death  of riparian vegetation.    Subsequently,  several
 relatively  minor discharge  events  widened the  river channel from 60  to over
400 ft  in  just  6  years (Groeneveld  and Griepentrog,  1985).   The overall
character of the  river in  the  affected  reach  has  changed from "a  narrow,
stable meandering-channel to  a wide shifting  channel  with  braided reaches,
with  obvious  effects  on the  success  with  which  the  previous  riparian
community  can  re-establish  itself.    Downstream  reaches  unaffected  by
pumping have  maintained  healthy bank  vegetation and have experienced no
major erosion (Kondolf and Curry, 1986).

Channel stability and  morphology also can  be  affected when ground  water
discharge to a  watercourse  is anthropogenically  increased.   In the case  of
Las Vegas Wash (Nevada), ground water pumped from  a deep aquifer is used
extensively to irrigate lawns and golf courses in the area.  This has caused a
considerable  increase in  recharge to  the  shallow  aquifer  system.   The
resulting increased  discharge  from the  shallow subsurface  to  Las Vegas
Wash  has  been  partially  responsible  for  changing the  wash  from  an
ephemeral to  a perennial  stream.    As  a  consequence of  this  increased
discharge, erosion and headcutting have had a  major  impact on  the  riparian
vegetation.   As the channel has  been lowered  by about 15 ft,  water levels
have  declined in the riparian communities.   Plant communities have changed
from swamp  and marsh vegetation to  saltgrass and salt cedar.   Headcutting
has resulted  in  the  upgradient migration of  the  hydrophytic  species (Burbey,
                                   .  26

Salt  Water  Intrusion

When ground  water is pumped from  fresh water aquifers that are in  hydraulic
connection  with saline water, the resulting  gradient  may  induce a flow  of
salt  water toward  the  well.  Overpumping of a  well  or  well  field  in  close
proximity  to the  salt water/fresh  water interface  can result  in  the  interface
being drawn toward the well to the extent that  salt water  intrudes into the
fresh aquifer  (Freeze  and Cherry,  1979).    This  can occur under several
conditions.    In  coastal   areas,  sea  water  of  greater  density frequently
underlies  surficial  fresh   water  aquifers.    In  addition,   an estimated two
thirds of the continental United States is underlain at  some  depth  by saline
ground  water,  with the majority of .the  shallowest  located in  the Central
Plains and  Midwestern states (Atkinson et al., 1986).   Excess pumping  in
these areas can > draw underlying salt water  upward toward  or into  a  fresh
water zone,  a  phenomenon known as upconing (Fetter, 1988).

Salt  water intrusion is a  widespread problem in  many parts of the country.
Atkinson  et al.  (1986) provide a  comprehensive summary  of  the  locations
where  salt  water  intrusion, is  threatening  or  contaminating  fresh water
supplies.  As  an overview, salt water Intrusion has occurred in each of the
21 coastal states and  has  the potential to worsen  as  water demand increases
in coastal urban areas.   Among the more critical problem areas  are  Long
Island (New York), the Biscayne Aquifer  (Florida),  and  several basins  in
California  and Georgia.  In total, only 8  states throughout the country  have
not reported any instances of salt, water intrusion (Atkinson  et al., 1986).,

Riparian areas  and wetlands which are supported  by fresh  ground water
discharge  can be  impacted  when  overpumpihg  results in  intrusion of highly.
mineralized  water.    Riparian  apd  wetland  communities  are  adapted  to
specific ranges  of  salt tolerance.   Increasingly  saline  conditions  interfere
with   water  and  nutrient   uptake by  plants  (Bolen,  1964).    As  salinity
increases, plant  growth usually is reduced and succession  to  more halophytic
species may occur.  Such changes  will typically  be accompanied by impacts
to native  animal  species,  particularly waterfowl,  which depend  on wetland
vegetation (Bolen,  1964).   Evaporation and evapotranspiration accelerate salt
accumulation  in  the soil  surface, exacerbating  salinity  impacts during the
growing season.    In contrast, one  potentially benejicial  but relatively minor
effect  of  increased salinity has been  noted in  fine  grained wetland  soils.
Increased salinity enhances the internal  cohesiveness of clay particles and
results in  reduced  erodibility of  these soils (Jenkins and Moore,  1984).

Although  salt  water; intrusion  is   geographically  widespread,  and  the
potential for ecological invents j& great,  research  on the nature and extent

of such  impacts  is very limited.   Attention is focused  more frequently on  the
impacts  on  drinking water  supplies, water quality .for irrigated agricultural
supplies, and costly impacts  borne by industrial users  (Atkinson  et  a!.,

One  comparatively well documented  example  of  potential  ecological impact
from  salt water  intrusion is occurring  in portions of the Edwards  Aquifer
(Texas).   Overpumping for public water supply  draws  highly  saline water
toward  freshwater springs which  threatens  several  endangered species  of
amphibians- (Longley, 1992).
Changes in  Ground Water Geochemistry

Studies  on  the interaction between ground water  and surface water  have
more commonly focused on the  hydrologic balance in  the  system.   As noted
above, alteration of  the  rate and direction of ground water flow can  result  in
ecological impacts  to ecosystems.   However, many geochemical aspects  of
the surface water/ground water  system and the  associated  solid media can
profoundly affect the biological communities which inhabit  them.   In  general,
geochemical effects  associated with changes in ground water discharge  have
received  relatively   little  attention compared  to  the  ecological  significance
of surface water chemical parameters (Hagerthey and  Kerfoot, 1992).

Nutrient   availability  can  be   influenced  significantly   by  ground  water
discharge in a wetland.  Reduction of ground water discharge can reduce the
rate of  mineralization  of  detrital  organic matter, resulting in decreased
delivery  of essential nutrients such  as carbon,  nitrogen  and phosphorus  to
the root  zone.  Alternatively, reduced  ground water discharge may  result  in
increased nutrient availability if  ground water discharge  is  not  available  to
dilute  nutrient concentrations and transport  nutrients from  the root  zone
(Harding, 1993).  Overall nutrient ratios  as well  as nutrient availability are
important in determining  species composition  and health  in   many wetlands.
Selective removal of the  more  soluble nutrients will  alter optimal  ratios  to
which species  are adapted.

Reduction of  ground  water discharge to  wetlands  can  alter several other
important geochemical parameters of ecological  significance including the
dissolved oxygen  concentration,  pH, redox potential,  salinity and  alkalinity
of the soil/water environment.   Redox potential controls the solubility  and
bioavailability   of  redox  sensitive  elements  such  as   iron,   manganese,
nitrogen, sulfur, and chromium  (Stumm  and Morgan,  1981).  Reduction  of
       water  discharge to a wetland also can affect physical features of the

soil  such  as  the  temperature  and  the  degree  of  aeration  of  the  soil
(Grootjans and Ten  Klooster,  1980). - Furthermore, without ground water
input  to  certain  wetlands,  surface  outflow  may   be  eliminated  and
evapotranspiration  may  become  tie  major  hydrologic  output,  thereby
promoting solute  buildup in  the  soil (Davis, 1993).

Wetland  ecosystems generally are  adapted to a specific  range  with  respect
to each  of these parameters.   Changes  that  exceed the  tolerance of the
individual species will  result  in  a  loss  of  species  and succession  to  more
tolerant  species.   For  example,   calcium-rich   ground  water  discharge
traditionally has  supported  many rare and  endangered  plant  species in  a
nature reserve in the  Netherlands.   Increased ground  water  pumping for
drainage  and drinking water  supply over a period of 40 years has reduced the
ground water  discharge.  As a  result, the extent  of the  calciphilous  marsh
species has been reduced and succession to woody  species such as Alniis has
been  promoted (Wassen et  al., 1989).   Another  example  is  provided  by
changes  in salinity in  Las Vegas Wash (Nevada).   Increased  land application
of pumped ground water  has resulted in greater flow through the salt-laden
shallqw  alluvial  deposits along the  banks  of  the wash.   The  resulting
increase  in salinity has adversely  affected  riparian  vegetation  in this area
(Burbey,  1993).                .     .


                               Section  6

                  Locations  of Ecological  Impacts of
                      Ground  Water  Overdraft on
                     Wetlands and Riparian  Areas
The  following  section is a compilation of the literature which was found on
the location and extent of impacts on  wetlands and  riparian  areas.   The
majority  of the  information  is  from  the  United  States  although  some
international  examples  have  been   included   where  documentation  is
sufficient.   Most of  the  locations are confined to  an  individual drainage  basin
or aquifer  system.    However, several instances of more  regional  impacts
were  identified.

This  compilation is  intended to be  as-comprehensive as  possible.   Because
many sites  are not well researched or  well documented,  it  clearly is  not
indicative  of the magnitude  of  this issue in the United States.   In  some
cases, extensive  hyd/ological  modifications resulting from ground  water
drawdown  are described  in  the  literature  but  no  followup  investigations
regarding  the  potential  for impacts  in associated wetland ecosystems  have
been conducted.

Discussions of some locations of  impacts  have  been  developed  into  case
studies  because  detailed  information  was available.   In  some cases,
considerable  information on geological  settings  and local hydrogeology has
been included.  It is hoped that  this information will be  of use in a predictive
sense for other as yet unrecognized locations with  similar  characteristics.

                       INTERNATIONAL STUDIES
Redgrave and Lopham  Fens, East  Anglia

Botanical  and zoological data  document  ecological  changes  occurring  over
the past 30 years in  several valley wetlands in a national nature  reserve in
East Anglia, England (Harding, 1993).   Beginning  in 1957, nearby pumping of
ground  water  for public  water supplies  was determined  to  be the primary
cause of impacts occuring  at  the species, community and ecosystem levels.
Because East  Anglia  has the  highest  concentration  of such  wetlands in
Britain  and also has a substantial water supply deficit, there is  the potential
for widespread and  long  term  impacts (Harding, 1993).

Prior to  the  late 1950s, calcium-rich, nutrient-poor  water discharged  under
artesian  pressure  from  a  semiconfined aquifer  and  supported  largely
herbaceous wetlands in the Redgrave  and Lopham fens in East Anglia.  With
the onset of pumping  in 1957,  artesian pressure was reduced  to" the  point
that  surface  discharge  was  eliminated  and   plant  communities   were
sustained  exclusively by  precipitation.   This change  in hydrology resulted in
a  drastic  alteration  of  the  competitive  balance  of  the dominant   plant
species.   By the 1970s,  herbaceous species  rapidly were being replaced by
scrub species such as Salix cinerea and Betula pubescens.  The loss of ground
water  discharge  eliminated  the specialized  environmental   conditions
required by semiaquatic and surface rooted species which  can  be sensitive to
reductions in  water levels of as little as 2 to 3  cm.

Invertebrate species  dependent on  spring-fed  and calcareous wetlands also
have declined.  Seventy seven percent of the fen  and bog species  have  been
lost.   The spider Dolomedes plantarius currently  faces  extinction in  this
area.  An associated impact of the loss of ground water discharge  is that the
wetlands are  now susceptible  to frequent  damage  from fires.
                     INCLUDING THE NETHERLANDS
The  hydrology of the vast  northwestern European  lowlands has been altered
drastically by dish -age,  primarily for land  development and agriculture.  Wet

.meadows have become rare  as they have  been converted  to highly productive
pastures.  Remnants  of  native  types  of  meadows  are  under protection  as
nature reserves (Grootjans and  Ten Klooster, 1980).

Vegetation changes associated  with altered hydrology were  investigated  in
Germany as  early as the  1930s.  However, the  objective of these studies was
to  measure  changes  in  agricultural productivity  to  estimate  compensation
payments to farmers  (Jalink, 1994).   Beginning  in  the  1970s, considerable
research  has been undertaken  by Dutch scientists but studies limited to the
impacts  of ground water  pumping have not been performed because of the
widespread  alteration, of  the  natural   hydrology  for  drainage.    Currently,
research  is  directed  increasingly toward identifying  ways  in  which changes
in  surface water management  and  .land  use  can  minimize  the  ecological
impacts  of ground water  pumping  (Jalink, 1994).   Regulations for well siting
and ground  water allocation  in  the Netherlands currently do address  the
importance of  conservation  of  the remaining  natural wetlands  (Jansen and
Maas, 1993).

In very low  relief landscapes such as  are  typical in  the  Netherlands,  ground
water discharge zones  called seepage  faces are common  (Jansen and Maas,
1993).  Natural vegetation around  these zones  may be dependent entirely  on
the  soil  chemistry and water  availability  resulting  from  the  ground  water
discharge. Excess pumping  of  ground  water readily can threaten  or eliminate
vegetation adapted, to these conditions  as was  shown  by Grootjans and Ten
Klooster (1980) for three Dutch wetland reserves and by  Jansen and Maas
(1993) for the  Punthuizen wetland sanctuary.

Similarly, Wassen et  al.  (1989)  have investigated  the  impacts of  ground
water decline in  the Naardermeer  nature reserve  in the Netherlands over the
past 40  years.  The seepage areas have  long  supported T/7e/yper/s-reediands
and many rare and endangered  plant species. During the  last 40 years ground
water has been  pumped  for drainage and  drinking water.  This has  reduced
ground water discharge in many  seepage areas.  Because  the  regional ground
water  is calcium-rich, the  distribution of  endangered   calciphilous plant
-species   including  Caricion  davallianae   has  been  restricted  severely.
Diminished   fresh  seepage  flow  also  has resulted  in  the  acidification,
salination, and eutrophication of the  studied marshes.   Succession to woody
species such as filnus is widespread and has  resulted in accelerated  loss of
additional marsh  species.

Spain is the most arid  country in Europe.  Ground  water demand for various
uses  is large and  has increased  rapidly  during  the  last  two decades.
Similarly, the great ecological value  of the wetlands in Spain did not  become
appreciated widely until the 1970s.   This  growth of water  demand  already
has led to  some  serious conflicts between wetland conservation and ground
water development (Llamas,  1989).  The problem is excerbated by a lack of
knowledge  of the  hydrology  of many wetland ecosystems (Suso and  Llamas,
Douro River  Basin

Among  the more systematically  studied  ecological  impacts resulting  from
ground water  extraction  are the ecosystem changes  occurring in the Dquro
River Basin in central Spain (Bernaldez et  aL, 1993).  Because the basin is  a
somewhat  geographically isolated semiarid enclave, many species  including
several endangered  species  of  birds of prey,  depend  almost entirely on the
dispersed wetlands.   Several different  types  of  wetlands occur  in  the  basin
including wet  meadows,  sedge  meadows,  marshes, phreatophytic woodlands,
ponds and sloughs.   They include ground  water  discharge and recharge  sites
and  have  differing  degrees of interconnection with the regional and shallow

Over 40  years of declining water table levels  are  the  result  of ground  water
pumping, primarily for  crop irrigation.   The average  decline from  1970 to
1987 was  1 m/year.   In the four subareas studied in  detail  by  Bernaldez et
al., 39 to 82% of the wetland area has been lost (i.e. is now  area which is no
longer classified  as  a wetland).

The  following  impacts were noted.   1)  Non-phreatophytic annual  species  such
as  Trisetaria  panicea,  Bromus  tectorum  and  Vulpia  spp.  have increased,
indicating  dryness  and  increasing  nitrification  due to  the  mineralization of
labile soil organic matter  in  the reduced  moisture  regime.   2)  Perennial
plants such  as Festuca  rothmaleri,  Phleum  pratense  and  others   have
disappeared.  Mosaics of xerohalophytes  have proliferated.   3) Desiccation of
slightly sandy soils  has resulted  in  wind  erosion  of the A  soil  horizon  and
enhanced  surface salination which further impedes growth  of vegetation.

Because  many of the wetlands in this basin are interspersed  in vast  expanses
of irrigated cropland, they are  reserves for a diverse  flora and fauna  which

 may,be dependent entirely  on an individual isolated locality.  Food webs  of
 plants, insects, aquatic invertebrates,  reptiles, waterbirds and, mammals are
 being  impacted by the loss of relatively small wetland  areas (Bernaldez  et
 a!.,  1993).
 Donana  National Park

 The Donana National  Park (DNP) is recognized as  one of the  most important
 natural environments  of the  European Community  (Llamas,  1989; Suso and
 Llamas,  1993).  The  area  first received legal  protection in 1969 and has been
 designated'a Reserve  of the Biosphere by the United Nations.

 Donana  National  Park is  situated  in  a  tectonic  basin  filled  with  Plio-
 Quaternary sediments in the Lower Guadalquivir Valley.  Three  ecosystems
 can be distinguished  in the DNP:  stabilized  eolian  sands and  moving coastal
 dunes on the  periphery and the central marshlands.  An extensive partially
 confined  aquifer is located  beneath thick clay deposits  in the central  area.
 Ground water  is  recharged  by rainfall on  the sands  and discharges at the
 ecotone  between the sands  and marshes as  well as contributing base flow to
 surface streams.  The permanent wetness of the ecotone renders  it the most
 productive and fertile  zone of the  DNP.

 In  1979  the Spanish government  approved  and  principally subsidized  a
 massive  irrigation, project  utilizing  ground  water from this  aquifer (Llamas,
 1988).  Since then the drinking  water demands in the area have increased as
 well,  primarily  due to expansion  of the tourist  industry.   The ground water
 models used to project allowable  pumping  rates, were   developed primarily
 from   the  perspective  of  technological   feasibility  with  little   or  no
 consideration of ecological suitability (Suso and Llamas,  1993).  More recent
 modelling  by leading  Spanish  researchers  predicts the   following  ecological
 impacts from the  approved pumping scheme:   1) desiccation of the  ecotone
 wetlands   in  some   locations  leaving  the  abundant  soil  organic  matter
 vulnerable to fire, 2)  considerable reduction  of surface water influent to the
 park as gaining reaches of streams become losing  reaches, and 3) degraded
.ground water quality, primarily by nitrate (Suso and Llamas,  1993).

 As  a  result of  international  public  protest,  the  size of  the originally
 approved  irrigation, project has been  reduced.  However,  the  researchers
 cited   above  continue to  predict  that  ecological impacts to   the  park's
 wetlands  are inevitable with the  timeframe  being dependent only on the
 amount of natural precipitation  available  in  the interim.

Tabias de  Daimiel  National  Park

One of the  most extreme  examples of ecological impacts from ground water
pumping is found in the Tabias  de Daimiel National  Park.  The internationally
recognized park is  located on the Central Plateau of Spain.   Prior to ground
water  development, the  park  consisted of an  approximately 20  km2  marshy
area around the confluence of the Guardiana and Giguela Rivers.  The area
was the natural discharge zone for the extensive underlying aquifer  system
which  is composed of  calcareous and detrital  material of continental origin
(Llamas, 1988).   Prior  to development, the  swamps  of  the Tabias were
covered by  about 1 m of water except during extremely dry  periods in  the

The major land use in  the surrounding region was  dryland farming until  the
hydrogeology of the extensive La  Mancha aquifer became understood more
widely about 20 years  ago.    From 1974  to 1987, ground  water  irrigated
acreage  increased  from  30,000  to 130,000 ha and  annual  ground  water
pumping increased  from  200 hm3 to 600 hm3 (Llamas,  1989).  The  average
annual recharge rate  is  only  about 260 hm3 from a combination of  sources
including ground  water, surface  water and rainfall.

As  result of prolonged overpumping, the water table  has fallen to as much  as
20  m  beneath the land  surface.   This depletion has caused the total
disappearance  in this area of the Guardiana River since 1984 in addition to
the progressive desiccation of  the Tabias.  Phreatophytic vegetation has been
lost totally  and the  highly organic soil  is  undergoing  a slow  process  of
spontaneous combustion.  In  1986 and  1987, two large  fires burned over one
third of the  national park (Llamas* et al., 1992).

In  1987, under pressure from  Spanish and  international ecological  groups,
the aquifer  officially  was  declared "overexploited"  in  accordance with  the
1985  Spanish  Water Law.  This  required the  preparation of  a management
plan and  the  creation  of  a  local Users Committee to  attempt  to  mitigate
these  impacts.   About  $10 million (US) from the. Spanish government was
used to attempt to regenerate  the wetland.   The primary approach  was  to
supply water from  other  sources including  land applying pumped  ground
water,  importation  of  surface  water  via  aqueduct  and  impoundment  of
surface water to  retain  it for  the park  (Llamas et al.,  1992).  However,  the
reallocation  of  water  resources  for regeneration was  met with considerable
local  opposition and ultimately  only  10%  of the  water deficit  was  made
available during the first three  year trial  period.

According to some  researchers,  the changes in the vegetation  and fauna are
so significant that the area appears to be an "ecological desert" even  though
there sometimes is water on the land surface.  It has been suggested that the
government's  investment  in  regeneration'would  be  spent  more  wisely  by
promoting reconversion to crops requiring less water (Llamas et al., 1992).
                            UNITED  STATES
In many  areas  in  Arizona,  particularly  in  the southern part of the state,
ground  water  pumping  has increased  dramatically  since the  1940s to meet
the  needs  of  irrigated  agriculture, industry  and  rapid  population  growth.
Ground  water supplied  48% of Arizona's large water demand  in  1985  and
agriculture accounted for 87% of the total water use (U.S. Geological Survey,

The  imbalance between  the  quantity of water  consumed in Arizona and the
long-term  dependable  suppy is a  major problem.  Overall about  two thirds  of
the ground water pumped  in recent years was withdrawn from storage (U.S.
Geological Survey,  1990).   Annual   pumpage  rates  may  exceed  natural
recharge rates by more than  500.times in  some  areas (Schumann  et al.,
1986).   Ground .water levels  have  declined by 50 to 300 ft in several  basins
throughout the state including the Gila, San  Simon, Avra Valley, and  others
and by 400 ft  or more in several additional basins such as the Santa Cruz and
Salt (Schumann et al., 1986).

As  transition  zones  between aquatic  habitat and  surrounding  terrestrial
habitat,  riparian  and   wetland   ecosystems   are the  most   biologically
productive areas  in the  arid Southwest.   Over 85% of wildlife species  are
dependent  upon wetland  and riparian  areas  for some  aspect  of their
existence.  These areas  serve  as  important  breeding areas,  refuges from the
desert  heat,   important  corridors  for  animal   movement  through  the
surrounding  desert, and critical sites  for  forage production  (Davis,  1993).
The  highest known breeding bird  densities  in  the  United States have been
recorded in  the desert riparian  habitat  (Carothers et al., 1974).

 Historically  many  of  Arizona's  rivers  and  streams  were  perennial  and
 supported  large  expanses  of  wet meadows,  marshes, swamps,  and dense
 mesquite  bosques  (woodlands) (Davis,. 1993).  These wetlands and riparian
 areas  have been impacted  extensively and severely in  the last century.   Only
 about  15% of the  riparian  areas  which were present in the  early 1800s still
 remain and the  percentage is  even lower when only Sonoran Desert riparian
 areas  are considered  (McNatt et al., 1980).   Impacts result  from ground
 water  pumping  as well as other causes such  as  arroyo cutting,  livestock
 grazing,  land  clearing   for  agricultural   and  urban  development,  and
 hydrological  modification  of   surface  flows  (Stromberg,  1994).    It  is
 frequently difficult to  identify which  factors  are responsible for  riparian
 damage or loss.

 Gallery forests  of Fremont cottonwood  (Populus  fremontii)  and  Gooddirig
 willow  (Salix gooddingii)  historically covered  hundreds of  kilometers along
 the floodplains  of  many  of Arizona's  rivers  and  similar low-elevation rivers
 in California,  Utah,  and northern Mexico (Stromberg,  1993c).   Today these
 Sonoran riparian cottonwood-willow forests  are among the most  threatened
 forest  types in  the United  States  (Swift,  1984).   Excessive  ground water
 pumping is one  of several  causes for the  decline of these forests.

 Similarly,  mesquite (Prosopis  velutina) woodlands were the  most abundant
 riparian type in  the southwestern United  States (Klopatek  et  al.,  1979).   In
 locations  where ground  water is  less  than  50  ft  such  as the  alluvial
floodplain  aquifers which  were  historically  common  in  Arizona,  mesquite
 can grqw  to tall  dense canopy forests called bosques.  Deep root systems and
symbiotic  associations  with nitrogen-fixing  bacteria contribute  to the  high
 productivity  of this  species.  In turn,  bosques support diverse and abundant
ecosystems including one  of the  highest densities  of  breeding  birds of  any
southwestern habitat.  Again, as a result  of ground water decline and several
other land uses such as those listed above, these bosques now are  reduced  to
relatively  small  isolated   remnants,   virtually none  of  which  remain  in
pristine condition.   Attempts to  restore  degraded  mesquite  bosques have  had
limited success  (Stromberg, 1993b).

Other  ecosystem impacts  which commonly occur with the loss of mesquite
flowers and fruits include  fewer insects  and  insectivores.   Avian  abundance
and diversity decline  with  reduced canopy volume.  The activity of  nitrogen-
fixing  bacteria   is  decreased thereby  reducing  the  soil  nutrient   pool
(Stromberg et al., 1992).   Furthermore, as has been observed in many other
drainages,  death of riparian vegetation results  in  increased flood  flows  and
increased  erosion and channel  widening (Groeneveld and Griepentrog, 1985).

Another  ramification of the extensive ground  water depletion  is  widespread
land subsidence which is manifested  as generally lowered land elevation  as
well as sink holes and  earth fissures.   Subsidence has  affected  .more than
3,000  mi2  in southern  Arizona  alone  (Schumann et al., 1986).   Subsidence
was detected initially in  1948 in the lower Santa Cruz  basin and  since then
subsidence  of up  to  12.5 ft has been measured in many southern drainage

In  addition to  the economic  costs  of  impacts to man-made  structures,
subsidence  and fissures can  result in costly environmental impacts as well.
Fissures  transect natural drainage patterns and can capture large  volumes of
surface runoff  which may deprive  downstream  alluvial  aquifers of  recharge
vital to  riparian communities.   Accelerated  erosion  along  fissures  forms
gullies which exacerbates the  impact on  natural  drainage channels over time.
Accelerated erosion also  occurs in  natural drainages along the periphery of
subsiding  basins   where the  gradient   between  the  basin  floor  and *fne
surrounding mountains is increased.   In  constrast, subsidence  decreases the
gradient  of streams and  rivers which  traverse subsiding  basins,  thereby
reducing surface water  flow  rates  and  increasing  sediment deposition and
flooding.   The  Combination  of these effects can  have  major impacts on the
natural hydrology and ecology  of an. area.  These  effects are most  pronounced
in the  Salt  River and lower Santa Cruz River basins  (Schumann  et al., 1986).
Santa  Cruz  River  Basin

The  13,790 mi2 .watershed of the Santa Cruz River is located  in  southern
Arizona and northern Mexico.  The  Santa Cruz River rises in the mountains of
southeastern  Arizona and after  a  short loop south into  Sonora, Mexico,  it
flows generally  northwest.   It is. an  intermittent  desert drainage containing
interrupted  perennial  and  effluent  dominated  reaches  and  regions  of
subsurface flow.   Perennial flow  is absent  except in short  reaches.   Primary
drainage is to the  Gila River near Phoenix in  south central Arizona.

The  floodplain of  the Santa  Cruz is alluviated deeply.   These sands and
gravels generally  are  unconsolidated  and  are major'water bearing  units.
Some  wells drawing  from these  deposits yield over 1,000 gpm  (Stromberg,
1994).   At the time of  early  settlement, this  unconfined alluvial aquifer
supported  extensive riparian  communities  including many gallery  forests.
Numerous marshes  and  springs  were present where  structural  features
forced underflow  to  the surface and  these areas  also  supported  wetland
ecosystems.    A  combination of  factors  .primarily  including  ground  water

 pumpage and  arroyo  cutting  have eliminated  or drastically reduced  most of
 these wetland  features (Hendrickson  and Minckley,  1984).

 The natural vegetation of the  low  floodplain  of  the  Santa  Cruz River is
 dominated  by  Fremont cottonwood  (Populus fremontii)  and Goodding willow
 (Sa//x gooddingii)  and  dense  bosques  (woodlands)  of  velvet mesquite
 (Prosopis velutina), netleaf  hackberry (Celtis reticulata)  and  Mexican elder
 (Sambucus mexicana) on the  river  terraces.    Other  riparian  vegetation
 associations include cienegas (marshes), sacaton grasslands,  and shrublancls
•of   seepwillow   (Baccharis   salicifolia),   rabbit   brush  (Chrysothamnus
 nauseosus), and burro brush (Hymenoclea spp.) (Stromberg, 1994).

 The Santa  Cruz basin has been inhabited  and  cultivated continuously since
 the  seventeenth century  which  has  resulted in extensive hydrological  arid
 ecological  changes  (Hendrickson  and  Minckley,   1984).    Early  irrigation
 required diversion of surface  flows from the river.  For the last few decades
 most of the  water needs of the extensive irrigated  croplands,  the  mining
 industry, and  the  rapidly  growing population have been met by ground water
 (Stromberg,  1994).   Total pumpage  in this hydrologic  basin  has increased
 greatly and often has  exceeded that of any other basin in southern Arizona by
 nearly an order of magnitude (Hendrickson and Minckley,  1984).   Ground
 water  levels have declined throughout the  basin with  the  maximum  decline
 of 460 ft measured in the lower Santa Cruz basin and  150  ft in the upper
 Santa Cruz basin (Schumann et al., 1986).

 The following  examples of the impacts in  riparian  ecosystems In the  Santa
 Cruz basin serve  to  illustrate  the  effects of  ground  water decline.   In
 general, the extent and severity "of ecological-impacts  increases downstream
 in the drainage.   In many cases, impacts  are  regional in extent and  native
 riparian  vegetation  has been  eliminated totally.
 In the lower Santa Cruz basin,  ground water is the only source of water for
 municipal,  industrial,  and agricultural  use  (Schumann  et al.,  1986).   Water
 and land uses in  the  relatively  recent past practically  have elimated riparian
 vegetation from the lower reaches of  the river  (Stromberg,  1994).

 In  the  upper  Santa Cruz basin irrigation and mining  consume the first  and
 second  largest quantities of ground  water respectively.  During  the  past
 several  decades, ground  water pumping  by  mines and  pecan growers  has
 caused  massive .ground water declines in  Pima  County.   This  has resulted in
 total  elimination  of  riparian  habitat  from  central  portions of the  river
 (Stromberg, 1994).   Obligate phreatophytic species  have been replaced by
 scrub  species  such as desert broom, burro  weed and burro brush.  Mesquite

 bosques  and   sacaton  grasslands  historically  supported  by  subflows
 essentially  are  no longer  present.

 Ground water pumpage  in  the  Avra Valley, a  major tributary  to  the  Santa
 Cruz, has  been extensive and  water tables have  undergone major  declines
 (White et  al.,  1966).   Farther  upstream, recent ground  water withdrawals
 from the floodplain aquifer in Santa Cruz  County have caused localized water
 table  declines and reduced  abundance  of cottonwood-willow forests.

 Ground  water  pumping  from  the floodplain  aquifer  for  the   growing
 populations  of  Nogales,  Arizona (population  20,000) and  Nogales, Sonora
 (population  250,000)  is creating  cones of  depression  which  have caused  low
 growth rate, low tree density and low canopy cover of Fremont cottonwoods
 and other  riparian populations  (Stromberg et al., 1993a).   In  areas where
 ground water depths are  greater than about 25  ft, cottonwoods and willows
 have  been  lost.   Effluent released  into  the  Santa  Cruz  channel from  the
 Nogales International  Wastewater Treatment  Plant  is increasing recharge to
 the  alluvial  aquifer to the benefit of riparian vegetation.

 In  comparison,  where   ground water  levels  re-main  shallow,   riparian
 communities are  healthy.   At a site near the Mexican  border where  a shallow
 bedrock layer  serves to  minimize ground water  level  deline (7  to 10  ft),
 populations  of  cottonwoods and willows  are in .relatively  good  ecological
 condition  arid  survival  of  seedlings  is  high.  .Similar  geological  conditions
 exist  farther downstream at the   Guevavi  Narrows  and   again   riparian
 communities are  able  to thrive (Stromberg, 1994). >       ""  '  '
Tanque  Verde  Creek

Tanque Verde Creek is an ephemeral  river in the Sonoran  Desert near Tucson.
It  flows  for  16  miles from its  headwaters  in the  Rincon Mountains  to  its
confluence with  Pantano Wash in  south  central Arizona,  in  some  areas, the
alluvial  floodplain  aquifer  supports large  mesquite  (Prosopis  velutina)
bosques.   Increased  ground water pumping in  this area has had  a severe
impact on these ground water  dependent riparian woodlands (Stromberg  et
al., 1992).                                              ......

Regional ground water decline in  this area  already  had resulted in sublethal
stress  to  mesquite  bosques in  the  early 1980s as  measured  by low stem
water  potential,  reduced leaflet  size,  and  canopy  mortality of .over  45%
(Stromberg et al., 1992).  Depth to  ground  water ranged  from  1  to  46 ft in
1986.  When the City of 
1988, ground water began declining at an unprecedented  rate of  12 to 21
ft/year,  reaching  depths of up  to  105  ft  in 1990  (Stromberg -et al.,  1992).
Water  depths of this  range  are  typically  lethal  to mesquite, particularly  in
coarse  alluvium with low  water retention  capacity.   Trees farther  upstream
m areas where  ground water levels remained  at about  10 to 16 ft were  in
good  ecological  condition  with  tall stature  and  no  canopy  dieback
(Stromberg,  1994).
San Pedro River  Basin

Roughly  parallel to, and 40 to 70 miles to  the  east of,  the  Santa Cruz River,
the  San  Pedro River  flows  northwest  from Sonora,  Mexico  for about  150
miles to  join the Gila River at Winkelman.  The 16,635  km2 drainage basin is
similar to others in the  Basin  and Range province.  Thick floodplain alluvium
of  gravel,  sand,  and silt  overlays   basin  fill  materials throughout  the
drainage.   Irrigation and municipal wells  in  the  valley  obtain  water from
both  the  basin fill  aquifer  and the  overlying  alluvial  floodplain  aquifer
(McGlothlin et  al.,  1988; Stromberg,  1994).

The San  Pedro River is largely perennial in the upper portion of the  basin and
intermittent in  most  other reaches.   Baseflow in the river  and the water
table beneath  the riparian  zone are  maintained  almost  entirely  by inflow
from the regional  basin fill. aquifer.  At times  of low flow, the entire flow is
diverted  in lower  reaches (Stromberg,  1994).

Irrigated  agriculture began in the latter part of the nineteenth century and
increased steadily until the  lateM960s.   Ground water  pumpage  roughly
parallels  growth in irrigated  acreage (Hendrickson and  Minckley, 1984).

As recently as a century ago, the San Pedro River was unincised and marshy
along much of its length (Hendrickson  and Minckley,  1984).   Ground water
withdrawal for  agriculture, mining and  municipal supplies,  as well as  dam
construction,  overgrazing and  clearing of riparian vegetation  for  pasture
drastically  have altered  the  hydrology of this area  (Richter  and  Richter,
1992).  Surface flows have been reduced or eliminated in some areas.  Ground
water levels  have declined  and  springs, wetlands,  and" cienegas  have been
reduced to isolated remnants, several of which  are located in  the headwaters
of tributary  valleys  such  as  the Aravaipa  and  Babocomari.   Cienega
vegetation has been  replaced widely by  riparian,  scrub species (Stromberg,

The  most severe  water level declines in the upper  San  Pedro basin have
occurred in  wells  near the expanding  population and agricultural centers  of
Sierra Vista  and  Hauchuca  City.   Water  levels  in  this  area  declined
approximately  1.4 ft/year between.._1966  and 1986  (Stromberg,  1994),
Ground  water  pumping in  this area also has  reduced the  baseflow in the river
(McGlothlin  et al.,  1988).   Riparian  mesquite  bosques  along  the river have
experienced, sublethal stress,as..a result of ground  water decline (Stromberg,
1993b).          ;

Portions of the upper  San  Pedro River in  the far  southeast corner of  the state
contain  some  of the  healthiest remaining desert riparian ecosystems  in the
southwest United States.   Included  are cienega  plant associations,  mesquite
bosques,  and cottonwood-willow forests as  well  as   the  most  extensive
remaining sacaton  grasslands  in  Arizona.   Also indicative  of  the  health  of
this  ecosystem  is  the  low abundance  of the  exotic saltcedar  (Stromberg,

About 40  perennial miles  of the river in this area and its associated riparian
zone were acquired in 1986 by the  Bureau of Land Management to be  managed
as  the  San  Pedro  Riparian  National  Conservation Area.   The primary
management objective  is  the  protection  of  the  remaining  riparian  habitats.
Cattle  grazing and sand  and gravel mining  are  restricted and agricultural
lands have been retired from farming.

Although ground water  pumping is not allowed  in  the Riparian  Conservation
Area  itself,  there  is  concern  that  the  cone  of depression from  increased
ground  water  overdraft  in  the  vicinity of  nearby Sierra  Vista may extend  to
the  Riparian  Conservation Area to  the detriment  of  native  vegetation
(Richter  and  Richter,  1992).    Also there  is high  potential  for  future
development elsewhere within this  portion of the  basin (Stromberg,  1994).
Ecological modelling  predicts that ground water decline of only 3  ft  would
result in the loss of many  marsh species  including the Huachuca water umbel
(Lilaeopsis  schaffneriana  var.  recurva)   which  is  a Federal  Endangered
Species Category 1 listing  candidate  and  only  recently has  been rediscovered
along the  San  Pedro  River  (Stromberg,  1994).    in,  some  areas in the
Conservation Area, seasonal  ground water fluxes  of  this magnitude already
have  eliminated obligate   wetland  species and-threaten facultative wetland
plants.   Other  impacts which  have been  noted  in  some areas  are low
survivorship of  cottonwood  and  willow  seedlings,  increased  establishment
of saltcedar and riparian scrub species such  as burro  brush  and rabbit  brush,
and sublethal  stress in sacaton, grasslands and mesquite  bosques (Stromberg,

Gila  River Basin

The Gila River is the major drainage  of southern Arizona.  The basin  extends
eastward into the mountains  of  New Mexico, crosses the  full  width  of  the
state  and joins the  Colorado at the western  state border.   Topographically
the basin is typical of .those in the Basin  and. Range physiographic province.
It  is  a  comparatively  flat, wide, sediment  filled  valley  between  narrow
rugged  mountain  ranges.   Basin  fill materials  and  terrace  and floodplain
alluvium  have been  significant ground water bearing  units in the past  (Culler
et al.f 1970).

Prior  to  the twentieth century, there  were extensive  marshes, swamps, and
floodplains along  much  of the  river.  Dominant vegetation included cattail
(Typha  domingensis),  bulrush  (Scirpus olneyi),  giant  reed  (Arundo  donax),
commonreed  (Phragmites communis),  arrowweed  (Pluchea sericea), and  many
cottonwood and willow trees.  The dense  vegetation of these well-developed
riparian communities often  reached  10 to  15 ft in height  and supported large
and  diverse  wildlife populations.   The river gradient  was  shallow  and  the
floodplain was  so  level that marshy lagoons formed in  places along the main
channel (Rea,  1983).

By the end of  the nineteenth century the  perennial flow of surface water was
reduced  drastically and  only in  a few  locations  did  surface  flow  continue
until the 1950s.  Deprivation of surface water recharge coupled with channel
incision and  ground water  exploitation caused the floodplain  water  table  to
decline.    As  a  result,   the  entire riparian community  of  willows  and
cottonwoods was eliminated and  replaced  in many areas  by exotic saltcedars
(Tamarix spp.)  (Rea, 1983).      '

One locality in this drainage provides an  example  of the extent of ecological
impacts  which have  resulted from the  major hydrological  alterations which
have  occurred  in this drainage.  As late, as 1940  an  extensive bosque known
as the New York Thicket existed at the confluence of the Santa Cruz  and Gila
Rivers and three major washes (Vekol, Green and  Santa  Rosa).   The bosque
was up  to 6  miles wide  in areas  with  mesquite and screwbean mesquite
(Prosopls pubescens) reaching heights up  to 40 ft.  By the late 1970s ground
water pumping had  caused the  water table to decline  to about  100  ft below
the land surface resulting  in the death of  90% of the  mesquite (Rea, 1983).
Although  mesquite  have  deep  roots and tolerate  moderate  water  stress,
lowering the  ground  waterJevel  below  about  50  ft results  in  death  of
riparian  mesquite trees or  in  conversion of mesquite  from  a dense  tree
community to  a sparse  shrub community (Stromberg, 1994).

 Regional habitat loss such as that .described above  has had a major impact on
 wildlife  in  this area.   Twenty-nine species of  birds  have been  completely
 extirpated  from  the middle Gila and  lower Santa Cruz  Rivers and several
 other avian species have  declined  noticeably  in population size because  of
 habitat  deterioration.   Loss of  nesting  sites  and food  sources  are  major
 factors affecting species numbers (Rea, 1983).      '
 Casa Grande  National Monument

 Casa Grande National  Monument is situated 1.5 miles south  of the Gila River
 and  approximately  50  miles west  of its confluence  with the  Salt  River  in
 south central Arizona.   The 480 acre site is  located  in the  floodplain  of the
 Gila  in Final County.   Designated a  national  monument in 1918, it  has had
 restricted  access to people  and livestock  for  over 70 years  (Judd  et al

 Ground  water withdrawal for  nearby agricultural  development began  in the
 early twentieth  century  and  increased in the  1940s,  causing  the  water table
 to  decline from about 6 ft  to  about  43 ft (Stromberg, 1993b).  An extensive
 bosque of large mesquite trees which utilized alluvial ground water  survived
 during this initial period of decline.   However, all  of the trees  died  when
 pumping increased and the  water table dropped about  3 ft  per year to depths
 of  40 to  150 ft  below  the  land surface (Judd  et  al.,  1971).   Continued
 increases in  pumpjng in the latter half of this century had  lowered the  water
 table to  as  much as 650 ft  as of  1970  and the site now is dominated by
 upland desert shrubs  (Stromberg,  1994).   The area is  littered  with  large
 deformed stumps of dead mesquite trees  (Judd et al., 197.1).
Verde  River Basin

The Verde  River  watershed  lies in the  Central  Mountains  physiographic
province in central  Arizona arid drains  17,218 km?.   The river rises in  the
-mountains  to the west  of  Flagstaff  and flows south  to join  the  Salt  River
east of  Phoenix.   The  watershed is semiarid and  has an  average annual
rainfall  of  30 cm  (Stromberg,  1993a).

The floodplain of the lower Verde River contains  thick deposits of alluvial
silt,  sand,  and  gravel  that support  stands of mesquite  (Prosopis juliflora),
interspersed  with  arrow-weed  (Pluchea  sericea),  seepwillow  (Baccharis
glutinosa),  and  saltcedar (Tamarix pentandra).    Fremont  cottonwoods
(Populus fremontii)  and Goodding willow (Salix gooddingii) are  common  m

 the  floodplain and occur  in  stands  along  the  existing  channel  or along
 sections of old river channels (McNatt  et at.,  1980).          ..-._.

 Where  it is healthy, this riparian habitat supports a large  and diverse fauna
 including  big game (mule deer and  javelina),  waterfowl and wading birds,  arid
 small mammals including beaver, muskrat,  and rabbits.   Higher than average
 densities  of coyotes, bobcats, skunks, and raccoons utilize  this  area.  Over
 160 species of  birds  are  known  to  frequent  this riparian  habitat  including
 two  endangered species, the bald eagle and the Yum a clapper  rail.  A portion
 of the  Verde  River riparian area has  been  identified  as having  the  highest
 bird population density in North America  (McNatt  et al., 1980).

 The riparian  ecosystems   in the  Verde  River floodplain  are experiencing
 considerable impacts due to water  table decline.  In central Arizona,  ground
 water  pumping  for  agriculture  reduces water  available to the Verde River
 riparian zone  and  its  tributaries (Stromberg,  1994).   Pumpage from the  Big
 Chino  Valley  aquifer  also  could pose a  threat to the upper Verde River
 riparian ecosystems (Stromberg,  1994).

 In the  lower  Verde drainage the  riparian community  is  "on  the  verge of
 collapse"  as a result of a  combination. of  natural and  anthropogenic  factors
 (McNatt et al., 1980).   Prior to  1977  less  than 4% of  the cottonwood trees
 were dead along  the Fort McDowell  reach of the lower Verde. Since  1977, 46
 to 84% have died.   In addition,  many of the  remaining  trees are approaching
 maturity  and   very few seedlings  are regenerating  successfully  in  the

The  cause of  this  mortality  is  a   combination  of  hydrological  factors.
 Limited water releases from Bartlett  Dam, located  upstream  of  this reach,
 have deprived  the alluvial aquifer of  recharge needed  to  maintain  the
floodplain water  table.   Further, natural floods which  assist  in  cottonwood
regeneration have been  eliminated.   Lastly, as of 1983, the City  of Phoenix
operated  an  infiltration  gallery and  14  wells  in  the  area of  greatest
mortality.    These  well  fields  withdraw 20,000 acre-ft of  alluvial ground
water per year for municipal use and are capable of lowering the  water table
over  10  ft.    In  1977,  the  year of greatest  cottonwood  loss,  drought
conditions  resulted   in  almost  no  releases  from   Bartlett   Dam  and
considerable water table decline.

In comparison,  Sycamore   Creek,  a nearby  tributary  to  the  Verde,  is  not
controlled or pumped but did experience similar  drought conditions  in  1977.
Riparian vegetation  including cottonwoods  in this drainage does  not  exhibit
increased mortality  or  impaired  seedling regeneration (McNatt et  al.,  1980).

 Loss of cottonwoods  is  very  detrimental to populations of nesting  birds  in
 this area.   Reduction  of  cottonwood  densities  from 46  trees/acre  to "10
 trees/acre  resulted in over a  50% reduction  in  the  number of nesting  bird
 pairs per 100  acres  (McNatt  et al.,  1980).   These  data are derived  from  a
 study  of  intentional phreatophyte  removal  for  water  conservation  but  are  a
 reasonable  estimate  of  avian impacts  from  cottonwood  mortality due to
 water  level  decline as well.
Salt  River  Basin

The Salt  River drainage is a major tributary  to the Gila and  passes through
the growing  metropolitan  area  of Phoenix.   Ground  water pumping in  the
basin has increased steadily over the  past  50 years.  About two-thirds of the
total ground  water withdrawal in  Arizona occurred in  the combined areas of
the Salt and  lower Santa Cruz basins  (Schumann et al., 1986).  Ground water
level decline  has exceeded 300 ft in many parts of the  Phoenix, Tempe, Mesa
and Scottsdale areas.  Subsidence of 3 to E> ft has been  measured in an area
of  over 500   mi2.   Fissures  resulting  from this subsidence capture surface
water flow in some  locations (Schumann et al.,  1986),  creating the potential
for detrimental ecological  impacts associated with surface dewatering.

From  the  time of earliest  record  of European,explorers  until  the  1920s,  the
Salt River was  a  perennial stream   lined with   cottonwoods and  willows.
Large sections of  the channel consisted of  sand  bars  which were  exposed at
low water flow and  were  colonized by seepwillow and  arrowweed.  By  the
1950s  the river  was  a conduit  only  for  flood  flow  as upstream  irrigation
dams  impounded all of  the normal discharge.   Native vegetation  was
eliminated  and many  miles of  impenetrable thickets  of exotic saltcedar
(Tamarix chinensis]  overgrew the dry channel banks.   The  saltcedar  was
sustained  by  a  sufficiently shallow water-table of  approximately 23 ft (Graf,
1982).  , Without river recharge,  intensified ground  water  pumping  by Tempe,
Scotttsdale and Mesa in south central Arizona caused ground  water levels to
decline to over 220  ft in the 1960s.  This resulted in the elimination of  the
saltcedar  thickets,   Despite ground  water level  rises of  more than  115 ft
after recent  floods,  the  ground  water remains too deep and  fluctuates  too
erratically  to  sustain  riparian   vegetation  except  for  a few  ribbons  of
tamarisk  growth supported by irrigation return flows and sewage  effluent in
some  locations (Graf, 1982).


California  consistently  leads all  states in volume  of  surface  and  ground
water  withdrawals.    The  state  has  retained this  position for 40 years
primarily because of  the large volume used  by  irrigated  agriculture  (U.S.
Geological  Survey, 1990).  During the past 100 years, the population and  the
.associated  industrial and agricultural  demand increasingly  have gravitated
toward  the more arid areas  in the  southern  portion  of  the  state.   Massive
projects have,been undertaken to  export surface and  ground  water from  areas
of relative  abundance to  areas  were it  is needed.   However,  available  water
supplies   are  insufficient  to   meet   current needs  without  substantial
depletion of ground water from storage.  Furthermore, population projections
predict  an  average annual  increase of more  than  330,000  people  for  the
remainder  of this century (U.S. Geological  Survey, 1990).   Periodic  drought
further  exacerbates  this water shortfall.

Ground water  overdraft  is an increasingly critical .problem in  many areas of
the state.   Statewide, ground water pumping exceeds recharge by an average
of 2.0  million  acre-ft/year.   Eleven basins in   California have  been identified
by the  Department of Water  Resources as being subject to  critical conditions
of overdraft, based on  problems  of salt water  intrusion,  deterioration in
quality, land subsidence,  and prohibitive pumping costs (Fort et al.,  1993).

California has a  wide range of  water rights laws.   Ground water  regulation is
undertaken primarily at the  local- level and overdraft generally  is  permitted
unless  curtailed  through  adjudication  of a basin or administration by a local
management entity  (Fort  et  al.,  1993).   Prevention  or  minimization  of
overdraft  is difficult  under this system.
 Owens  Valley

 Owens  Valley is  a  long  narrow  closed  basin  located  in  east-central
 California.   The  3,300  mi2 valley  is  bounded  longitudinally  by the Sierra
 Nevada on the west and the White and Inyo Mountains on the east. One major
 river, the Owens  River, flows south  through the  valley.   Numerous tributaries
 drain the east face of the Sierra Nevada and have  formed extensive alluvial
 fans  along the west side of the valley  (Hollet  et  al.,  1991).   Historically,
 springs and wetlands  were found throughout the  valley (Rogers  et al., 1987).

The  valley  is  filled  with  uncdnsolidated   to   moderately  consolidated
sedimentary deposits and intercalated  volcanic flows and ash.  Nearly all the
recoverable ground  water in the valley is found  in these  valley-fill materials
(Rogers et ah,  1987).  An unconfined  aquifer is present throughout the valley.
Depth to  water ranges from the land  surface to more than 15 ft below the
surface of the valley (Sorenson  et a!.,  1989).  The  saturated thickness of this
unit ranges from  30 to  100 ft.   An  underlying confined  water  bearing  unit
also is present throughout most of the valley.   The degree of  confinement  is
negligible  in many areas as the clay beds of the intermediate  confining  layer
are  discontinuous.   Virtually  all, of the, ground water in  the  Owens Valley
aquifer system is derived  from precipitation that falls  in  the  Sierra Nevada
and infiltrates through the alluvial fans (Hollet  et al.,  1991).

As  a  result  of  the  rain  shadow effect  caused  by the Sierra Nevada, the
climate in the  Owens Valley is semiarid to  arid.   Most of the land in the
valley is  covered  by native vegetation.  The  communities which  occupy the
greatest land area are 1) shallow ground water alkaline meadow,  2) shallow
ground water  alkaline scrub,  3)  dryland  alkaline scrub,  and  4) dryland
nonalkaline scrub (Sorenson et al.,  1991). The three shrubs Nevada saltbush
(Atriplex  torreyi),  greasewood  (Sarcobatus  vermiculatus),  and  rabbitbrush
(Chrysothamnus  nauseosus ssp.  viridulus)  in combination  with   two, grass
species   saltgfass  (Distichlis  spicata  ssp.   stricta)  and  alkali  sacaton
(Sporobplus airoides) comprise more than 90% of the vegetation  growing  on
shallow ground  water zones of the Owens Valley  floor (Groeneveld, 1989).
Because  of the  availability  of  shallow  ground   water,  the valley  floor
supports  about  73,000  acres of  phreatophytic   vegetation  (Dileanis and
Groeneveld, 1989).
                             *      *             ••
In the  early 1900s  the City of Los   Angeles  recognized the  value of the
relatively  abundant surface  and ground water supplies in  the  Owens Valley
and  acquired much of the land.  A 233 mile aqueduct was  completed in  1913
to divert water from  the  Owens River and modest quantities of ground water
(generally  less than  10,000 acre-ft/year).  Prior to  this diversion  the Owens
River flowed into the 100 mi2 Owens  Lake.   Evaporation now exceeds inflow
and, except in very wet years, Ihe lake is dry (Hollet et al., 1991).

In  1970  a second  aqueduct was  completed which increased the  average
capacity for exporting  water by 50% (Rogers et al., 1987).  The majority of
the additional export has  been  ground water.   Ground  water export accounts
for  over 50%  of the  discharge from the valley's aquifer system  (Hollet et al.,
1991) and  evapotranspiration  by phreatophytic vegetation accounts for the
majority of the remainder  (Duel!,  1990).                            7

Extensive  export  of  surface and  ground  water from  this arid region has
resulted in widespread  impacts to  the  shallow ground water  vegetation
(Rogers et al., 1987).   In the early  1970s,  phreatophytic plants  covered  about
the  same  acreage and conditions  were similar to those observed  between
1912 and 1921.  In 1981, a loss of 20 to  100% of the plant cover on  about
26,000  acres  was noted.  This reduction was postulated to be  a response  to
the  increased pumpage of ground water and changes in  surface water use.
Considerable  public concern was  expressed regarding these  environmental
impacts and  the related loss  of recreational  activities and  wildlife habitats
(Hollet et al.,  1991).

Historically,  springs between  the  communities of Big Pine and  Independence
in central Owens Valley discharged  the largest  quantities  of water.   Included
are  Fish,  Big  Seeley,  Little  Seeley, Mines, Little Black Rock  and Big  Black
Rock Springs.   A direct and immediate effect was measured  in the quantity
of spring  flow when  nearby deep wells were  pumped for  export.   Springs
ceased  to flow with continued  pumping  and  flowed  again when pumping
stopped or was minimal (Rogers  et  al., 1987).   As this occurred, aquatic and
riparian  habitat was lost.   Effects  were particularly  severe  on  the  four
native fish species in  the Owens River system.  Two species, the  Owens
pupfish  (Cyprinodon radiosus) and the Owens chub (Gila bicolor snyderi), are
listed as  endangered  and   one,  the  Owens  dace  (Rinichthys osculus),  is

Fish  Slough  is a  remnant  of a once widespread  shallow aquatic/riparian
wetland in central Owens Valley.  It supports a variety  of rare plant species
and  the endangered Owens pupfish  (Pister and Kerbavaz, 1984).   Three
springs  provide  the flow in  this slough.  Declining ground water levels have
reduced  spring flow since  1971  and  resulted  in  a reduction of riparian
wetland acreage.   The area currently  is  designated as  the Owens  Valley
Native  Fish   Sanctuary which  protects  the refuge  under  an interagency
management plan.   However, further  agreements  to avoid additional  ground
water drawdown are needed  (Pister and  Kerbavaz,  1984).
Little Black Rock Spring also is located in  central Owens  Valley about  9
miles north of Independence  and  is  used to support a local fish  hatchery.   In
1971, when the discharge from jme spring began declining, a nearby well was
pumped to replenish the water supply for the hatchery.   Soon  after pumping
began, the spring ceased to  flow.  Additional surface  water was used to meet
the needs  of  hatchery  (Perkins et al., 1984).   Not only was the attempt  to
mitigate the loss of spring flow unsuccessful  in  terms of  supplying  adequate
volume  to  sustain phreatophytic  vegetation, the surface  water  was lower  in
alkalinity,  salinity and nutrients.  The result was a significant loss  in  marsh

area as well as a change in species composition.  The area once  inhabited  by
marsh  vegetation  has  been  invaded , by more drought  tolerant perennials
(Perkins et ai., 1984).             -

Fish Springs Lake  and  the  Springfield  are  two other wetland environments
which have  undergone similar changes as a result, of ground water pumping.
The lake  is  now ephemeral  and artesian  flow at the  springs has  ceased.
Marsh  vegetation has been lost entirely from these areas and  plant diversity
ultimately  has been  reduced  to  include only  those  annuals which  can survive
on  infrequent precipitation.

As  part of ongoing  litigation  between  the City  of Los Angeles and-the  local
Jnyo County  government, a ground  water  monitoring  program has  been
designed  to curtail  pumping  when  potential,  ecological impacts  are predicted.
Provisions for well  shutdown  have  been  included  as the  basis for  a
permanent agreement for ground water  management to  preserve  the  existing
vegetation cover  (Groeneveld,  1989).   It  is  essential  to  prevent further
vegetation loss  because attempts  at revegetation  have  been unsuccessful.
Poor soil  aeration limits  the  invasion of xeric shrub  species from the nearby
alluvial  fans  (Groeneveld,  1989).
Carmel  River  Valley

The Carmel  River  drains  an area of 255 mi2 in the  northern Santa  Lucia
Mountains  of the  central California coast range.  The upper  21  miles flow
through  steep canyons with  little  alluvium  and  the  lower  15  miles flow
through  an alluvial valley  known as the  Carmel Valley.   The^ alluvial  fill is
typically  15 to 30 m thick and  consists of  sand and gravel  with some silt and
clay  interbeds.    Stream  flow  in the  Carmel  River is in  response  to
precipitation  with   high  flows  during  the rainy season from  November  to
April.  The  river stage declines in late spring and  summer:   In its  upper
reaches  the  Carmel  River  is  perennial  but  in  the  alluvial  valley flow  is
intermittent, typically drying up in late  summer (Kondolf and Curry,  1986).

The alluvial  valley fill  is a  generally unconfined  aquifer,  although  localized
areas  of confinement  exist.   Seasonal fluctuations of the  water table  result
from  recharge of the  aquifer by winter stream flow and  subsequent decline
of  the water table by drainage of bank  storage  and, more  importantly,  by (
ground water withdrawals  primarily during the dry  season (Kondolf  et  al.,
1987).                                                  /

Because  of  the  low water retention  capacity of the  poorly developed and
coarse  textured floodplain soils,  plants in the riparian zone  are-either xeric
or phreatophytic in habit due  to the long summer dry period.  In undisturbed
locations, the mature  riparian forest  is composed  of  approximately 60% red
willow  (Salix laevigata),  30% black  cottonwpod  (Populus trichocarpa),  and
10%  California  sycamore (Platanus  racemosa)  and white  alder  (Alnus
rhombifolia) combined  (Groeneveld and Griepentrog, 1985).

The  upper  watershed  is settled  sparsely.    Extensive commercial  and
residential development has occurred  in the last three decades  in the  lower
Carmel  Valley, especially  near the river  mouth  on the Monterey Peninsula.
Prior  to the 1960s, the  Carmel  River supplied most of the water  for the
peninsula.   Because  little surface water  storage  capacity is available, the
increasing  demand for municipal water supply has been met by ground water
withdrawn  from  the  alluvial aquifer.    Production from  the aquifer reached  a
relative  peak in 1976  and then decreased  in 1977 because the  aquifer was
depleted locally  (Kondolf and  Curry, 1986).  However,  throughout the 1980s,
additional wells  have been drilled in  the Carmel Valley for  increased export
of ground water to the Monterey Peninsula downstream (Kondolf et al., =1987).

This extensive  ground water  withdrawal has altered  the  riparian  ecosystems
as  well  as the  general  hydrology and geomorphology of the Carmel  River
itself  (Groeneveld and Griepentrog,   1985).   Beginning  in the  late 1960s
residents noticed that trees  in  the  vicinity  of  the  municipal  wells  were
dying.   Analysis  of  aerial  photographs taken at  intervals  between 1956  and
1980  confirmed  progressive  loss  of   riparian forest  cover over  that period
(Groeneveld  and Griepentrog, 1985).    Downstream of the reach which  was
impacted  by  ground  water pumping phreatophytes  remained  healthy (Kondolf
and Curry, 1986).

During two years of drought  in  1976  and 1977, the alluvial aquifer received
almost  none of  its  usual  annual recharge from  river flow.    Water  table
drawdown of over 10 m was measured along 4'km of the river in the vicinity
of the  well  fields.   Downstream of  the  pumped  reaches,  drawdown  was
minimal.  The  low water  retention capacity of the coarse alluvium  resulted
in a rapid decline  of  the water table.   This  prevented  phreatophytes  from
extending their root  systems to  follow the declining water table even  though
poplars,  of which the black cottonwood is  a member,  have been observed to
achieve daily root growth rates of up  to  5 cm  (Groeneveld and  Griepentrog,
1985).    Coupled  with  negligible   summer   precipitation,   the  riparian
vegetation was  eliminated rapidly.

 Pumping of the municipal" wells was sufficient to  have a  major effect  on
 several  other aspects of the regional  hydrology.  Drawdown_Jn  the  pumped
 reach reversed  the  hydraulic gradient in the upstream direction in this area.
 Stream  flow became  influent to  the  banks  in the  reach  of  major pumping
 which had  previously been  a gaining reach  of the  river.   In  fact  by  mid-
 August 1982, the Carmel River entirely dried  up in the pumped area but re-
 emerged downstream of  the pumped reach (Kondolf et al., 1987).  The loss of
 bank storage to sustain the base flow in the summer months was extremely
 detrimental  to  the summer  downstream  migration of steel head  trout smolts
 (Sa/mo gairdneri) as well as the rest  of the  riverine aquatic  life  (Kondolf et
 al.,  1987).                                                           ,

 Channel geomorphology  has also  been impacted by the loss of the stabilizing
 effect of the roots of  riparian vegetation.  The channel of the Lower Carmel
 River  had been  essentially stable from 1939  to 1977,  Since.  1978-the reach
 in which the majority of the phreatophyte  dieoff  occurred has  experienced
 extensive bank  erosion.  Aerial photographs  depict the channel widening from
 60 ft  in 1976  to over 400  ft in 1982  (Groeneveld  and Griepentrog, 1985).
 The flows that  produced this erosion  were not unusual  events (two five year
 recurrence   interval  flows)  but  the -impact  on  the  river  channel  was
 equivalent  to that of the 100  year recurrence interval event  which  occurred
 in 1911.   Downstream  reaches  unaffected  by  pumping  maintained  healthy
 bank vegetation and experienced no major erosion  (Kondolf and Curry, 1986).
 Recent bank erosion along the Carmel River has caused  property losses in
 excess of $1.5  million.   Further losses can be expected  because the present
 channel is  near the  threshold  of  meandering and braided characteristics
 rendering it inherently unstable (Kondolf and  Curry, 1986).
 Lower  Colorado River  (Needles  and  BIythe)

 Indirect  documentation was  found on  the  impact of anthropogenic lowering
 Of the riparian water table in the vicinity of Needles and  BIythe.   These two
 communities are near the Arizona -  California border on  the lower Colorado
.River.  The climate is extremely arid.   The river supports a shallow alluvial
 aquifer with depths to ground water ranging from 2 to 4 m  throughout the
 growing  season (Busch  et  al.,  1992).  The natural riparian  vegetation  is
 phreatophytic   forest  species  including   Fremont  cottonwood  (Populus
 fremontii),  Goodding  willow  (Sa//x gooddingii)  and the  Eurasian  native
 saltcedar  (Tamarix  ramosissima).   The  first  two species  , are  obligate
 phreatophytes  and as such  are  highly vulnerable to impact  if ground  water
 pumping  sev_ers their  connection to  their water supply.   This has  been
 observed in  ths  lower Colorado  floodplain.   With  the reduction or elimination

of the  mature canopy species, saltcedar can  competitively exclude  native
species and dominate the  plant community.   Similarly,  understory  shrubs
such  as screwbean mesquite  (Prosopis pubescens) and  arroweed  (Tessaria
ser/cea) have become dominant (Busch et al., 1992).
Central   Valley

The  Central Valley  of California occupies about 12% of  the total land area of
the state.  In the past  century, it has become one of the most hydrologically
and  ecologically altered regions  in the country.   Ground  water pumping  has
played a major although not exclusive  role in these changes.

Central  Valley  is  a   long   alluvium-filled  structural  trough  occupying
approximately 20,000  mi2  of  relatively  flat  land lying between  the  Coast
Ranges on the west and the Sierra Nevada to the east.  The  climate  is arid to
s.emiarid  with  precipitation decreasing  to the south.  The  northern  half  of the
basin is  drained by  the  Sacramento  River and  the southern  portion is the
drainage  basin of the San Joaquin River.  The  area surrounding'the confluence
of the two rivers  is called  the Delta.   The most southern  portion of the
valley, the Tulare Basin,  has  no perennial  surface outlet.  However, there is
considerable underflow of ground  water from the Tulare Basin to the  San
Joaquin drainage basin  (Katibah, 1984).

Lenses  of gravel,  sand,  silt,  and clay of  predominantly fluvial origin fill the
entire valley.    Most lenses are  not  widespread  with  the  exception of the
Corcoran  Clay Member which  forms  an extensive confining  bed between the
overlying  semiconfined  aquifer and  the  underlying confined  water bearing
zone.  The degree of vertical leakage is highly variable (Bertoldi, 1992).

In  1850  the Central  Valley  contained  an  estimated 4  million  acres of
wetlands  (Peters,  1989).  Prior to the construction of over 100 dams on the
two  main  rivers and their tributaries,  seasonal flooding formed vast  flood
basins and large  shallow seasonal  lakes which supported marsh vegetation.
Dense  riparian forests of Fremont  cottonwood, California sycamore,  and
willow and associated  intermediate  and  undergrowth  species utilized the
riverine silt of the natural levees.  Along  the Sacramento  River, the  levees
frequently  prevented  mountain streams  from  reaching  the  main  river
resulting  in  a  network of distributaries  ending  in sinks  of  tule  marshes.
Because  of the lack of surface drainage in the  Tulare  Basin,  seasonal lakes
with abundant marsh vegetation were common  (Katibah, 1984).   The  Delta
was an area  of convoluted inlets and islands supporting  wetland  ecosystems
(Conniff,  1993).

 Because  these wetlands occurred  in  an otherwise  arid area, they long have
 been important wintering areas for Pacific Flyway  waterfowl.  About  60%  of
 the  ducks,  geese, and  swans of this flyway  use the valley wetlands during
 the  winter  (Peters,  1989).              .

 Beginning around  1850, fertile soil,  flat  land,  and abundant surface water
 provided  the  incentive  for  the  development of one  of  the  nation's  most
 productive agricultural areas.  Today more than $15  billion worth of crops
 are  produced  annually  on the approximately  7.3  million  acres  of  irrigated
 agricultural  land in  the  valley (Bertoldi,  1992).

 To support  this lucrative industry,  massive flood control and water diversion
 projects  have  been constructed  and  hundreds of thousands of  acres of
 wetlands  have  been drained in the  past  100  years.  Two large  aqueduct
 projects were  built  in the middle  of  this century to convey water from the
 northern part of the  basin to the San Joaquin Valley.

 Ground water resources  have been developed simultaneously  with the peak of
 development occurring  in the 1960s  and 1970s.   Over  100,000  irrigation
 wells have  been  constructed.  Pumpage increased from  362,000 acre-ft in
 1912 to  about 15  million acre-ft  in  1977 (Bertoldi, 1992).   Ground water
 levels  have been  altered significantly  throughout the valley.   Most long-term
 declines have  been  less, than 100  ft  except, in the  southern  part  of the San
 Joaquin Valley  were heads have  declined from 100  to 400  ft (Bertoldi, 1992).
 In  many  areas horizontal  hydraulic gradients have  been  reversed  and
 downward gradients have been created as the  deeper aquifer is frequently
 the most  heavily pumped.  As  a result of -increased pumping  costs,  ground
 water withdrawals  have  leveled  off  or declined since  1967 and additional
 water  needs are  being  met with increased  surface  water delivery.   The
 recovery  of  the potentiometric  surface from  1967 to 1984 in  the most
 heavily  pumped areas  averages  approximately one  half of  the  previous
 drawdown (Llamas et al., 1992).

 Of the  4  million predevelopment acres of  wetlands in  the valley, less than
400,000 remain today.   It is difficult to  determine the role  which ground
water pumping  has played  in  this 90%  decline  because other  hydrologic
 modifications such  as  surface water impoundment for  flood control  and
irrigation  also  have  been extensive.   Today  the  majority  of the  remaining
wetlands  in  the Central  Valley are managed  for waterfowl  habitat and sport
hunting  (Peters, 1989).   Recent  inventories of remaining  riparian  vegetation
indicate that most  of  the acreage is  in a disturbed or degraded condition
(Katibah,  1984).               •  :                      v  ,  '  • .

Ground water pumping  in the Central Valley  also has  resulted in  the  largest
volume of  anthropogenically  induced land subsidence  in the  world (Bertoldi,
1992).  Subsidence has been most extreme in  the heavily pumped San Joaquin
Valley with  about 5,200 mi2  having subsided more than 1  ft and localized
regions of  20 to 30 ft.  The loss of 20 million acre-ft of  aquifer storage is
about  half  of all the manmade  surface water  storage capacity in the state
(Conniff, 1993).    Islands in the  Delta area  which  were  originally at  or
slightly above sea level are now 10 to 20 ft below  sea level (Bertoldi,  1992),
undoubtedly resulting in impacts to  all levels  of the wetland ecosystems.
Wetland  vegetation throughout  extensive portions of Florida has  died  or is in
a  state  of  premature decline due to excessive  ground water withdrawals
(Bacchus, 1994).  Many of Florida's  freshwater wetlands are associated  with
shallow  unconfined  aquifers  perched  above  confined  regional aquifers.
Ground water withdrawal from shallow saturated zones can  cause extensive
impacts  in these  wetlands.   In  addition,  where confinement is intermittent,
pumping of  underlying aquifers has resulted  in increased downward recharge,
thus  draining  wetland  ecosystems  of  the  water  which  sustains them.
Prevalent fractures  in the  extensive  karst  regions  in  Florida  promote  this
induced recharge (Bacchus,  1995).
Many  land  use  activities  have the  potential to  alter the   hydrology  of
wetlands  in Florida.   Included  are  ditching for drainage  for  agriculture  or
other development,  cattle grazing,  silviculture,  mining,  dredging and filling
operations, and  other land uses.  Occasionally it may be difficult to separate
the  impacts  of  these activities  from  the  impacts  of  ground water
withdrawal  for   consumptive  uses.    However,   numerous  instances  of
ecological impacts  from  ground water withdrawal  are  sufficiently  well
documented around  the  state  for this activity to be considered  a significant
threat to  wetlands in Florida.   Furthermore,  population growth in many areas
of the state continues to increase the demand for ground water.  Maximum
pumping for  irrigation_of crops,  lawns and  golf courses often  occurs during
the summer, the "season of greatest vulnerability  for wetlands  (Ormiston ef
al., 1994).

 Dade  County            •  '• '.

 Municipal well fields  in  Dade  County;  in  southern Florida  are increasingly
 susceptible  to  contamination from  salt", water  intrusion and urban  land  uses
 in the Miami  metropolitan area. ; To minimize the  risk of  contamination, the
 Northwest Well  Field was  developed;in  1983  in  an  inland  well  field
.protection zone.   This 65 mi2  site is  comprised  primarily  of undeveloped
 wetlands  and  rock  pits.    The hydrologic effects  of  this well field  are  of
 major  interest  because  of  the  increasing preference  for siting  municipal
 supply wells in such inland areas  (Sonenshein and Hofstetter,  1990).

 The  Northwest  Well  Field is located in the Everglades, a large  wetland  area
 which covers most of southern Florida.   The hydrology of the  Everglades has
 been  altered drastically throughout this century for  flood protection and  land
 development.  Many canals and levees have  been  constructed.

 The  Northwest  Well  Field  is operated by  the, Metro-bade  Water and  Sewer
 Authority.  The system consists of  15  supply wells  with a total capacity  of
 165 mgd and an  average  production of 132 mgd for the  period 1984 through
 1987.  Water is withdrawn from the  Biscayne  aquifer,  which is an  unconfined
 aquifer  composed  primarily  of  a very porous  sandy limestone  with
 interbedded sandstone  (Sonenshein and Hofstetter,  1990).
                  '     !         ..*.,-•'                  ,

 Continuous  water  level  data  beginning   in  1960 are  available at  five
 observation wells.   Two additional observation wells  were installed, one  in
 1970 and  another in  1982.  .Monitoring  data show that  water levels in the
 center of the well field have declined 6  to  7  ft during  both the  wet and dry
 seasons since  pumping  began.   Water  levels  in  a 10  mi2 wetland  area
 surrounding the  well field have been  below  the land surface  100% of the time
 since pumping  began.  Sonenshein and Hofstetter (1990) concluded that  this
 dewatered area was no longer considered a  wetland.   In another 10  mi2
 wetland  area surrounding  this core area,  water levels  have been  below  trie
 land surface as  much as 99% of the  time since pumping began.   Water  levels
 at four  monitoring wells outside  the cone of depression  of  the  well  field
 showed  no effect  from  pumping.  While  no references  describing the nature
 and   extent  of  ecological  changes  in  this • area  were  found,  impacts
 undoubtedly are extreme given the extensive alteration ,of the  hydrology.

Jensen  Beach  Peninsula,  Martin  County

Jensen  Beach Peninsula is a  long narrow spit along the southeastern coast  of
Florida.   The North Martin County well field is located on the peninsula in the
vicinity  of significant  wetland  areas.   Average daily  withdrawal  is 1.65 mgd.

A  long-term  ground  water  monitoring  network  has  been  established
primarily to  monitor  salt water intrusion.   Data from inland locations show a
gradual decline  in  water levels  resulting  from  a  variety of possible  causes
including  increasing  well  field withdrawals,  surface drainage  modifications
or below average  precipitation  over  the monitoring  period.    Ecological
investigations in the  area have shown  cause for concern that  increased
pumping may adversely affect wetland vegetation (Shupe and Gleason, 1989).
Loxahatchee  River  Basin, Palm  Beach  and  Martin  Counties

The  Loxahatchee  River  Basin  is  an area of considerable wetlands near the
southeast coast of Florida.  Historically, water was retained in the wetlands
throughout the basin  during the  wet season.   During  the dry  season, this
water  provided  fresh  water discharge  to  the Northwest Fork  Loxahatchee
River (Birkitt and Gray,  1989).  These ecosystems have undergone extensive
hydrological alteration as a result  of several  factors, the individual effects
of which  are difficult  to distinguish.   Channelization  and  drainage  for
development and ground water  pumping in  municipal well fields have reduced
water storage in  the basin.  This has resulted in excessive flows  in the wet
season  and very  low to  non-existent flows in the  dry season.
                                • '
The  reduced  dry season  flows  have  resulted  in  severe alteration of the
natural  hydroperiod  of  the basin wetlands.   Water  levels -in surface  water
bodies  have been lowered as  well.   Of particular concern is the fact that
decreased discharge in  the Northwest  Fork Loxahatchee River has allowed
salt  water to  intrude farther into  the drainage, damaging wetland  vegetation.
Cypress trees increasingly are  being replaced  by mangroves in a significant
portion  of the river  basin including  the reach which has  been designated a
federal  Wild and Scenic River  and  in Jonathan  Dickinson State Park  (Birkitt
and  Gray, 1989).
                                       • ^
As of 1989,  the  Northern Palm Beach County  Water Control District was
developing a  major  Water Resources Plan  to reallocate  water  in the basin  in
an attempt to restore a more  natural hydrologic  regime.  Specific objectives
were to enhance  the  quality of the wetlands, to restore  historic hydroperiods

  and water flow patterns, and  to increase ground water recharge to  provide a
  sustainable long-term  municipal water supply.

  When  fully  implemented, the  Water Resources  Plan  Was designed  to  create
  30,000 acre-ft  of  surface  water storage capacity to store  water during the
  wet season and redistribute it during the dry  season.   The creation of these
  reservoirs  was expected to destroy  2,358  acres of  wet  prairie or  marsh
  habitat and 330 acres of low  pine  flatwoods.   However 4,855 acres of open
  water  habitat and  650 acres of  newly created wet  prairie  and freshwater
  marsh  habitat were expected to replace formerly degraded wetland areas.  In
  addition,  4,200 acres of wetlands  were  expected  to  be restored  (Birkitt and
  Gray, 1989).
 Pasco,  Pinellas,  and  Hillsborough  Counties:
 Tampa  -  St.  Petersburg  Area

 Wetlands  such  as marshes,  wet  prairies,  pondcypress domes  (Taxodium
 ascendens), and swamps  are common  in southwest  Florida and typically
 comprise 20 to 30% of the  landscape (Rochow,  1993).  They serve as nesting
 and feeding sites  for Florida  sandhill  cranes,  bald eagles,  wood  storks  (a
 federal endangered species), and  an array  of amphibians, as well as provide
 habitat for other wildlife (Bacchus, 1995).
       ••;           ,            -            •              ''  •
 Throughout most of southwest Florida ground water is the primary  source  of
 drinking  water  as well  as water for agricultural  and for  industrial  uses.   To
 meet the needs  of the rapidly growing  Tampa - St. Petersburg area, ground
 water  pumping  has  increased extensively in  recent  years  (Rochow  and
 Rhinesmith, 1991).   At least 75%.of the  public water supply  is derived from
 concentrated  regional  well  fields  (Southwest  Florida. Water  Management
 District,  1996).

 The Southwest  Florida  Water  Management  District  has  maintained an
 extensive program  of  hydrological  and  biological  monitoring in  regional well
 fields  for approximately twenty years  (Rochow,  1994).   Observed impacts
 from pumping include  lowering of lake  levels, reduction in stream flow, and
 destruction  of  wetland  habitat.   In some  areas along  the  coast,  lowered
 ground water levels have caused  water  quality degradation  as a  result of
 saline  intrusion  (Southwest  Florida  Water Management  District,  1996).
 Rochow  and Rhinesmith (1991) report  that  detrimental  irnpacts observed in
wetlands include  replacement  of  aquatic plant species  by  upland  species,
decreased   abundance  of  wetland-dependent  wildlife,  increased   wetland

susceptibility  to  fire  damage,  increased soil  subsidence, and excess wetland
tree  mortality.

Based on this monitoring, the  District has found that water table drawdowns
ranging from  0.5 to 3.0 ft can  be  expected within a distance of approximately
one  mile from most production wells (Rochow,  1993).  The District also has
determined that  any ground water development which lowers the water table
in the surficial  aquifer by as  little  as 1  ft can  adversely  impact  wetlands.
Consequently,   in  1989  the  District  enacted  a  rule  requiring  detailed
environmental  review for any water  use  permits  where  drawdown  was
predicted  to  exceed  1  ft (Rochow and  Rhinesmith,  1991).   Long-term  site-
specific ecological monitoring  is required of permitees as  the  full  impact  of
ground  water decline may not become evident  for one to two  decades after
pumping begins  (Rochow,  1994).

Studies  conducted  at specific  well fields  under  the  jurisdiction  of the
Southwest Florida Water Management  District provide  examples  of the
wetland  impacts  which  result  from excessive  ground  water withdrawal
(Southwest  Florida Water Management  District, 1996).    The  Starkey  well
field  consists of 14  production wells located  about 30  miles north  of St.
Petersburg in Pasco County.   The well field boundaries encompass abut 8,200
acres  of  predominantly  natural  and undisturbed  pine  flatwoods, cypress
domes, marshes, wet  prairies,  and sandhills (Rochow and Rhinesmith,  1991).
Pumping began  in the 1970s at a  rate of  1  to 3 mgd and increased  beginning
in 1983 to an  average of 13  mgd  in 1989.   At this rate  of withdrawal, the
water table was predicted to decline at least 1 ft in an area of about 4 ml2
in the central part  of  the well field.

Beginning  shortly  after  the  increase in  pumping  in  the  mid-1980s, the
following  impacts  were observed  at several  sites in the Starkey well 'field:
1)  extensive invasion  of  upland  weedy  species  including dog fennel  and
broomsedge, 2) destructive  fires,   3)  abnormally  high  tree  fall, and  4)
excessive  soil  subsidence/fissuring  (Rochow,   1994).   Subsidence results
primarily from the  oxidation  of the  highly  organic soil.  This  contributes  to
the. loss  of  tree  species,  particularly  cypress, by  reducing  root support.
Abnormal hydroperiods  also  render trees  more  susceptible to pathogens.
Although  present at nearby unaltered control wetlands, fish  were absent and
amphibians  were absent  or  in very  low  abundance  at  sites associated  with
the well field.   The limited food resources at these  lower  trophic levels was
thought  to   curtail the  utilization  of these  sites  by wetland  reptiles  and
wading birds (Rochow and Rhinesmith, 1991).

 Impacts  of pumping at  the  the-Eldridge-Wilde  well  field  also have been
 extensively studied.   This well field is  located about  26 miles north of St.
 Petersburg in  Pinellas and Hillsborough Counties and consists of 3,500 acres
 of pine flatwoods, cypress domes, and marshes as well as improved pastures
 and  citrus  groves.  Production began in  1956 and averaged  about 16 mgd in
 1965,  after which it increased steadily to an  average of 29  mgd since  1972.
 Water  table decline of up to 3 ft  is predicted for about 50% of the area of the
 well field.   Vegetation, changes indicative of severely altered  hydroperiods
 were observed  on aerial  photographs  beginning  in  the mid-1970s  (Rochow
 and  Rhinesmith,  1991).   Cypress,have  been  almost   completely eliminated
 from some monitoring sites.    Recent assessments estimate that 85% of the
 wetlands  on well field  land  have been  moderately to severely affected in
 terms  of subsidence, loss of  canopy species, and invasion of upland species
 (Rochow,  1994).

 Similar impacts have  been documented  at  the  Cypress  Creek well field in
 Pasco  Cpurity.   Pumping began in 1976  and reached an annual rate of 30 mgd
 which  has  been sustained  since 1979.  A water table decline of 3 ft has been
 modelled  for about  60% of  the  site (about  6  mi2)  and  the, 1  ft drawdown
 contour encompasses about  12  mi2.   Biological  monitoring  since  1975  has
 documented  minor to  severe impacts  to wetlands in  this  area.    Although
 some  hydrologic  modifications undoubtedly  have occurred  as a   result  of
 agricultural and  residential   development in  this area,  detailed  long-term
 monitoring has  documented   that some  of the observed  ecological  impacts
 result  from ground water pumping.  Wetlands with  a  predicted water table
 drawdown  of  1  ft or  greater have hydroperiods about 50%  shorter  than
 control sites outside the  area of water table  drawdown.   This coincided with
 12 to 28% mortality of tree species such as-pop ash and cypress within  the 1
 ft drawdown area as well as  other impacts similar to those  mentioned above
 at other well fields (Rochow,  1994).

 Other  nearby  well  field sites monitored by the  Southwest Florida Water
 Management District  and  its  permitees  show similar results.   At the Cross
 Bar  well field the remains of 400 dead  gopher tortoises  were  found  at  Big
.Fish Lake  which was dry during  1991  in  spite of above average local rainfall
 for  that year.    Extensive replacement  of  wetland  plant species  such as
 floating hearts  and waterlily  by drought  tolerant  species such as dog fennel
 and  maidencane was  also noted.  Trends of wetland  alteration beginning in
 the late  1970s  at the Morris  Bridge well field were found to  have  stabilized
 when  pumpage was reduced  by 40%  from  1986  to  1993  (Rochow,  1994).
 Lastly, in  addition to several  tens of  thousands of acres of wetlands which
 have  been irreversibly altered  in this  area,  ground water  pumping has
 contributed  to the failure of  more  than 1,000 private wells which have been

repaired  or  redrilled at a cost of $3.5 to $4 million,  has drained lakes  and
streams,  and has increased the  rate of sinkhole formation in the-area (House
Committee on Natural  Resources,  1994).

Because these types of wetlands and hydrologic conditions  are  not  confined
to southwest Florida,  but extend  throughout the Southeastern Coastal Plain
physiographic province, it  is  likely that wetland  impacts such as  those
which  have  been  documented  carefully  by the  Southwest  Florida Water
Management District and its permitees are occurring elsewhere, either on  a
local  or  potentially regional scale  (Bacchus,  1994;  Bacchus,  1995).   For
example,  depressional  wetlands  typical   of  southwest  Florida  occur
throughout Florida, southern Georgia, North  and South Carolina, Alabama,
Mississippi and a portion of Louisiana.   The range  of pondcypress wetlands,
which   are  extremely  vulnerable  to  alteration   of  hydroperiod,  is
approximately coincident with  the limits  of the  Floridan aquifer  in Alabama,
Georgia  and South   Carolina  and continues  throughout  North Carolina,
Mississippi and the southeastern  portion  of Louisiana (Bacchus, 1995).
Bruneau  River,  Owyhee  County

Geothermal  discharge  at  Hot Creek  and 128 small  flowing thermal springs
and seeps along a 5.3 mile length of the Bruneau River in southwestern Idaho
has decreased  significantly  over the last 25 years  threatening  the  organisms
which inhabit the springs  and the outflow  (Federal Register, 1993).  Ground
water in  this  area  flows  northward  through volcanic rocks from  areas of
recharge  along the  Jarbidge  and Owyhee  Mountains and  is  discharged  as
spring flow  or leaves the  area as  underflow.  Prior  to  development,  natural
recharge and discharge from  the regional  geothermal aquifer  underlying the
600  mi2  Bruneau area were estimated to  be  approximately  in balance at
22,800  acre-ft/year. .

Ground  water  development began in  the late 19th century  and discharge to
wells  increased  throughout  this century,   primarily  to  meet the growing
demand from irrigated agriculture.  Maximum discharge was reached in the
early  1980s at  49,900 acre-ft resulting in  annual deficit  pumping  of  over
26,000  acre-ft.   In  large part  due  to the Conservation Reserve Program
administered by  the  U.S.  Soil Conservation  Service, pumping has declined to
the 1991  level  of  withdrawal of  34,700  acre-ft  in spite  of a  prolonged
drought  throughout the hie 1980s.   Pumping has caused water levels in the

 volcanic  portion of  the  geothermal aquifer to decline more than 30 ft  in
 much of the Bruneau area (Federal  Register,  1993).   ,    -.._.',.•

 As a result, geothermal spring  discharge also has declined.   For example,  in
 1965 spring  discharge at the Indian  Bathtub spring  was  2,400 gpm  and
 declined to about 130  to 162 gpm  during the summer of 1978 (Young  et al.,
 1979). 'By 1985 the spring had  ceased to flow during the irrigation season.
 A similar trend of  reduction of discharge was noted at other springs  in the
 area (Federal  Register, 1993).

 The  Bruneau  Hot Springsnail (Pyrgulopsis bruneauensis) is found only  in the
 springflow  of Hot Creek and the springs and  seeps of this area.  Most  of the
 springs are small and  highly vulnerable to impacts from water  level  decline.
 As a result of habitat threats from declining spring  flow,  the  springsnail
 was  determined by the U.S. Fish and Wildlife Service  to meet the criteria of
 an endangered species  in 1993.  Population estimates made in 1982 and 1992
 show a 50% reduction in  the number of  individuals in many springs.  In some
 local  areas such as  the  Indian Bathtub spring  the species has  been  totally
 eliminated  as  a  result  of spring  flow decline  and sedimentation.  Common
 aquatic associates of  the  springsnail -which are  also  at risk include  three
 mollusks,  the  creeping water bug  (Ambrysus mormon minor) (also endemic to
 the  Hot  Creek  thermal   spring  complex),  and  the  skiff beetle  (Federal
 Register,  1993).                                        .

 If  water levels in  the geothermal aquifer  continue to decline, the  U.S. Fish
 and  Wildlife   Service  anticipates  that  all   remaining  thermal springs
 containing  Bruneau Hot Springsnails eventually could  cease to flow, causing
 the extinction  of the  endemic species (Federal  Register, 1993).
Indiana  Dunes  National Lakeshore

The  Indiana  Dunes  National Lakeshore  is a  federally protected natural area
managed by  the National  Park  Service.   It is  a  12 mile segment of lakeshore
and  dunes located  on  Lake Michigan  in  northwest  Indiana.   In the middle
1970s  the  Bailly  Nuclear  Generator was constructed on a  7 acre site about
800  ft  west  of the  Lakeshore  boundary.  During construction,  ground water
levels  under  the nuclear site  were drawn down 20 to  30 ft  for  about  18
months.  With this drawdown, the westernmost  ponds in  the Lakeshore  were
predicted  to  dry up almost completely.   Less  than  0.5 ,;ft 5! water was

expected  to  remain  in  about 1% of the  pond  under average  conditions.
Sustained  dewatering would  obviously  be  damaging to  the  aquatic and
surrounding riparian  ecosystems (Marie,  1976).
Nevada  is  one  of  the  nation's fastest  growing  states.   Its  population
increased 50% from 1980  to 1990.  Nevada is also the  most arid state in the
nation  (U.S. Geological Survey, 1995a).  Because surface water resources are
scarce to nonexistent in many areas and  most  often fully appropriated,  new
development frequently  relies on  ground water.

The  combination  of  these factors creates  an ongoing  need to balance the
allocation of scarce  water  resources  between  humans  and  aquatic  and
wetland  ecosystems.   These isolated  habitats  are typically critical  refuges
for the plant and  animal species which are  entirely dependent upon them.
Ash Meadows and  Devils  Hole, Nye  County

Ash Meadows is  a  unique  and  biologically rich  hydric  ecosystem  in the
Amargosa Desert in southwestern Nevada.  Within the 162  km2 area, over 30
springs  and  seeps discharge a  total of  17,000 to 20,000  acre-ft of  water
annually,  sustaining  an  oasis  in* the otherwise  arid region.  Many  of the
springs are large with headwater pools of 6 to 10 m in diameter.  The area is
a  discharge  point  for  several  thousand  square miles of  a regional  flow
system developed in carbonate  rocks (Westenburg, 1993).

Like much of the Great  Basin, the Ash Meadows area was much wetter during
the late Pleistocene, supporting Lake Ash  Meadows as an  ephemeral feature.
As  the climate  of  the post-pluvial period  became drier,  Ash  Meadows
developed as an isolated remnant environment.  For this reason, the flora and
fauna of Ash Meadows represent a  unique and threatened  biota, including
two extinct and over  25 endemic species.  The  eight plants, two insects, ten
mollusks, five fishes and one mammal species constitute the highest  amount
of biological endemism of any area in the United  States (Williams, 1984).  As
of  1984,  eight species  were candidates  for listing  as federal  endangered
species  and two others  officially had been  designated as endangered species,

 the  Devils  Hole  pupfish  (Cyprinodon' disibolis) and the  Amargosa pupfish
 (Cyprinodon  nevadensis pectoralis)  (Williams, 1984).

 Dense to. moderate growths of mesquite and saltbush occur at  the springs and
 along  the outlet  channels.   Extensive  saltgrass  marshes  cover the poorly
 drained  flatlands  which receive the  spring  discharge  (Dudley and Larson,
 1976).   The large spring  discharge supports waterfowl and  other  migratory

 Several  factors have  contributed to extensive alteration  of this ecosystem.
 Carson Slough, the major discharge channel for the region was drained for
 mining  peat  and  clay  from  1910  to  1930.   During the mid-1940s several
 exotic  predators were  introduced to the  detriment  of the  native fish species.
 Most recently,  further alteration  of  the  natural  hydrology  has  occurred.
 Prior to  1960,  the small  number of local residents  used  natural spring  flow
 for  irrigation  downslope.   Beginning  in  1961,  development  of  the ground
 water resource  increased  steadily and about 40  wells had  been drilled as of
 1976.   In 1967,  a large ranching  corporation  began acquiring  extensive
 acreage  and  water rights  to  develop  about  12,000 irrigated  acres  in crops
 for  cattle feed (Dudley  and  Larson, 1976).   Pumping  for the  ranching
 operation began in May  1969.

 Beginning in 1969, spring  discharge throughout the  area  began  to  decline  and
 reached, a record low in 1972.  Annual overdraft in 1971 was estimated to. be
 1,500 acre-ft   Discharge was reduced by as much-as 50% in some springs
 (Dudley and Larson, 1976).

 Of particular concern was the'2.5 ft decline in water level in  Devils^ Hole, a
 warm pool in a  collapsed depression in  the limestone hills  of  a  40  acre
 disjunct  part of  Death Valley National Monument.  This pool is the  sole known
 natural  habitat  of  the  Devils, Hole  pupfish  (Cyprinodon  diabolis),  a  federally
 designated endangered species  (Williams,  1984).    The pupfish  feeds  and
 reproduces on a slightly submerged rock ledge which is highly susceptible to
 exposure  under conditions of  ground water  decline.   In  light of  declining
 spring  discharge, in 1976  the United States  Supreme  Court  established  a
 minimum water  level to ensure  the  survival of the pupfish.   Pumping rates
 were reduced  and eventually all  pumping for  irrigation  ceased  in  1982.
 Water  levels  recovered  quickly  (Westenburg,  1993).    In  1984,  the  Ash
 Meadows National  Wildlife Refuge was created to protect  remaining  endemic
 species (Graham, 1992).

-Subsequent to  the termination  of  pumping  for irrigation  for  the  ranching
 operation,  a  large  tract was  sold to a developer.   In the  mid-19eSs  plans

were  completed for development of  over 33,000  homes  in the area, requiring
additional ground water  withdrawal  (Williams,  1984).    It is clear that  the
feasibility of this development  must be  assessed  carefully  to  avoid future
impacts on  the spring ecosystems.
Las  Vegas Valley and Las  Vegas  Wash, Clark County

Increased  ground water  pumping  to meet the demand  of  the burgeoning
population of the  Las Vegas  valley as well as  for  crop  and  landscape
irrigation  needs  has contributed to a complex  alteration of the hydrology  of
this  region in the last fifty years.   This  frequently has been to the detriment
of the  wetland  ecosystems  for  which this city in the  Mojave Desert  was
named.  Ironically, Las Vegas  means "the  meadows" (Graham, 1992).

Water  use in Las Vegas is about 350 gallons per day  per  person, about twice
the national average and  almost  three times that of other arid  cities such as
Los Angeles and  Tucson (Egan, 1994). While undoubtedly inflated by the large
tourist  population,  water  consumption  is  still  high  in  this  city  which
receives only 4  inches of precipitation per year.   The  majority of the water
usage  is for  irrigation of urban landscapes  (Katzer and Brothers, 1988).

The  demand  for  water  in Las Vegas  Valley  has increased from that  obtained
from a few domestic wells in the early part of the century to  an annual use
of more than  200,000  acre-ft.  About  one third of the demand  is supplied by
ground  water and the rest is delivered from Lake Mead in  the Lower Colorado
River  system (Burbey,  1993).    Fine to  coarse  grained alluvial fan  and
floodplain  deposits  comprise the  principal aquifers  within the  valley.    In
many  areas  the water table is within  1  to  10  ft below the  land  surface
(Emme and  Prudic, 1991).   Along principal drainage courses such  as Las
Vegas  Wash, hydrophytic vegetation such as reeds  and cattails  abound.
Phreatophytic species such  as  mesquite, saltgrass,  and  saltcedar are  also
abundant in these areas.

Pumping  of ground  water has led to water level decline  of up to 300  ft  in
parts of  the  valley (Katzer  and Brothers,  1988).    Artesian   springs which
were scattered  throughout the  valley around the  turn  of the century  have
ceased flowing and their associated  marsh ecosystems have  been lost.   In
addition to wetland impacts, induced recharge from the overlying  shallow
aquifer,  which is high in natural  and anthropogenically  introduced  solutes,
threatens  the quality of the drinking water supply.  Dewatering  of  fine-
grained sediments has  caused subsidence over 400 mi2 in the valley (Katzer
and  Brothers, 1988).

 When used to irrigate lawns,  golf courses and crops,  ground water withdrawn
 from the  deeper  aquifer  typically  recharges the  shallow aquifer in  the
 central  part  of  the  valley.   This water ultimately  discharges into ephemeral
 channels, primarily Las Vegas Wash.   Excess irrigation recharge and treated
 sewage effluent have combined to change the  previously ephemeral wash into
 a perennial  stream.   As  a result,  the wash has  experienced considerable
 erosion and  headcutting in recent years.  The channel level has been lowered
 as much as  15  ft, isolating  adjacent  wetlands dependent on bank storage.  As
 a  result  of  headcutting,   hydrophyte-dominated   vegetation   has  been
 eliminated  in  lower  reaches  and  has  re-established  in  areas farther
 upstream where  ground  water  levels are sufficiently  shallow.   Continued
 headcutting  will ultimately  threaten  these riparian  areas as  well  (Burbey,
 1993);                                •".

 Other irnpacts from the  loss  of  wetland  vegetation along the  major washes
 include diminished  habitat  for bird  species  such  as phainopeplas, cactus
 wrens and  crissal  thrashers  (Graham,  1992).    These  species depended
 primarily on  the dense mesquite  stands the washes previously supported.   In
 addition, the  endemic fish species, the  Las Vegas dace (Rhinichthys deaconi),
 became  extinct when a  creek in  what  is now downtown  Las"  Vegas  was
 dewatered  by declining ground water levels (Graham,  1992).
         '  •                       '                 •              -i
 As the  population  of Las Vegas continues to increase at a rate of almost
 1,000, new  residents  each week,  additional sources  of water  are being
 sought.  In  addition  to pursuing increased allocation of surface water from
 the  Lower  Colorado and  the  Virgin  Rivers, Las Vegas is considering a  vast
 new  ground   water  project.  'Utilizing 1,200  miles  of pipelines  to  convey
 ground  water pumped  from more than  20,000  mi2  in  central  Nevada, the $2
 to $5 billion  project would  be one of the biggest and most expensive water
 projects undertaken  in  the West  (Egan,  1994).   The  environmental
 consequences of  this  pumping, including potential impacts on  ground water
 flow and springs in 4 counties, are being  evaluated.
Humboldt River  Basin

Within 'the Humboldt River  Basin  in  north  central Nevada is a large gold
deposit called  the Carlin  Trend.   Advances in ore  extraction  technologies
have allowed  the  profitable  development  of  over 15  mines  in  this area.
These mines have or are  proposing  to  dewater local aquifers to enable ore
extraction from deep pits.   Water  table  suppression  is desired at depths as
great as 1,700 ft and pumping rates  necessary  to accomplish this range from

a few hundred to  over 70,000 gpm.  This  magnitude  of ground water pumping
is  unprecedented  worldwide  (Miller, 1993;  Manning,  1994).

The  cumulative  impacts  of  all  the projected  ground  water  pumping
operations  on the  Humboldt River Basin have never been evaluated thoroughly
(U.S. Geological Survey, 1995b).   Ecological impacts are expected during the
active dewatering  phase  as well as  after dewatering  ceases  and ground
water levels recover.

During  the mining  operations,  many  companies  are  disposing of pumped
ground water  by  discharging  it  directly  to  the Humboldt River and  its
tributaries.    This  will  provide additional  flow temporarily for  irrigators and
riparian  habitat along  the river.   Some  of  the  pumped  ground  water is
reinjected or applied to the  land  and eventually reaches the Humboidt River
as  well.  As a result, wetlands  and springs  have  been created which  are
dependent  on  this temporary water  source.   If not actively sustained  after
the  mining  phase  of these  operations is  completed, these  habitats will  be

The major impact  is expected  after  pumping  has stopped  and the affected
aquifers re-equilibrate  with respect  to  the  natural  regional hydrology
(Meyers, 1994).   When pumping  is discontinued, the  cones of  depression
around  the pits and the  pits themselves  will  fill by diverting ground  water,
by  reducing surface water discharge, or by reversing hydraulic  gradients and
drawing water  from the Humboldt  system.   It  is estimated that  in excess of
3,000,000  acre-ft  will  be  drawn  into the cones  of depression  and  pit  lakes
from  both ground  and  surface water  sources.  (The  relative contribution  from
each source is unknown  (Miller, -1993; Miller, personal  communication)).   In
comparison, the annual flow  of  the  Humboldt River at Winnemucca is  only
around  150,000 acre-ft.   Predictive  models suggest that pit  refilling  could
eliminate   most surface   baseflow from   two  Humboldt  River tributaries,
Maggie  and Boulder Creeks, and reduce the flow in  the Humboldt  River by 6 to
66% (Meyers, 1994).  Additional mines and proposed  expansions of several of
the  existing  mines are not included in this estimate and would exacerbate
further  these  impacts.
                                                            i   , v „,
Detailed descriptions  of   the  anticipated  ecological impacts  of the  post-
mining phase have not been provided in the Environmental  Impact Statements
for  these  projects.  It is  evident  that many wetland  and  riparian  habitats in
this  arid basin  would be impacted  to some extent.   Rough estimates for just
two  of the pits project the loss of about 2,000 acres of springs, seeps, and
wetlands.   Several  creeks in the  area have been identified preliminarily  by

 the  U.S.  Fish  and Wildlife  Service as  potential  recovery  sites  for. the
 federally listed threatened  Lahorttan  cutthroat trout.
 Honey Lake Valley,  Washoe  County

 Developers in  the  Reno  area  in .west  central Nevada have sought  to  utilize
 ground water to meet  increased water demand,  lh'1991, Washoe County was
 granted rights  to pump 13,000  acre-ft of ground water  from the  Honey Lake
 valley and import it to the Lemmon and  Spanish Springs valleys via a 39 mile

 It was anticipated that this pumping would eliminate  or degrade  185 to 485
 acres of springs, seeps, and wet meadows as the  water table declined  to 61
 to 78 ft in the  area of withdrawal (Bureau of Land Management,  1993b).  Even,
 with  proposed  mitigation  measures such  as  pumping  additional ground  water
 to sustain wetlands, the  impact in the draft  Environmental Impact Statement
 was  considered "significant and adverse".  As. of  1994, the project was on
 hold  pending  resolution  of legal  matters  unrelated to water  or wetland
 issues (Hill,  personal  communication).
Beaver  River,  Texas  County                         ,   "

The  Ogallaja Formation  underlying  most of  the Oklahoma Panhandle  is part of
the  High Plains regional  aquifer  system  which extends  174,000  mi2  from
southern South  Dakota  to  northwestern Texas.  This aquifer  (also known  as
the  Ogallala aquifer)  is  primarily a water  table  aquifer deriving  recharge
from precipitation (Wahl and Wahl, 1988).   The average annual precipitation
in the Oklahoma Panhandle is about 17 inches, average  lake evaporation is  62
inches,  and  average annual  evapotranspiration is about  16 inches.  Thus
annual recharge to the aquifer in this region  is low.

The  primary  land  use in  the Panhandle  is  agriculture,  with  the  land  about
evenly  divided  between  farming  and  ranching.   The introduction  of the
center-pivot  sprinkler system  in  the early  1960s resulted in a ca'pid  increase
in the  use  of  ground water for irrigation.   In  1963 there were  about 450
large capacity" wells  in  the  Panhandle.   The  number had  risen to  2,500  by
1984 (Wahl and  Wahl, 1988).

 Because the rate of withdrawal from the High Plains aquifer is  much greater
 than the rate  of recharge,  widespread  water level  declines have  occurred
 (U.S. Geological  Survey, 1990).  In western Oklahoma declines of 25  to 100 ft
 are  common.   This  has resulted in a substantial reduction in discharge in the
 Beaver  River  which drains  most  of  the Panhandle.   The 10-year moving
 average discharge was relatively stable  from 1950  to  1965 at about 25 to
 30 cfs.  By 1986, the  10-year moving  average discharge had  decreased to
 about 7 cfs.   Prior  to  1971,  the  river was generally  perennial,  ceasing to
 flow less than  15% of the year.   By the mid-1980s the river was dry about
 85% of the year.  Changes in  river discharge are not correlated  with changes
 in precipitation or surface  water diversions  (Wahl and Wahl, 1988).

 The  long  annual  periods without surface  discharge  and the  associated
 depletion of bank storage undoubtedly have  been  detrimental  to  riparian
 ecosystems  in  this  semiarid  area.    Ground water  overdraft resulting  in
 reduced stream discharge  and  adverse  impacts  to  riparian  communities  also
 has  occurred  in  other states  where  ground  water from  the  High  Plains
 aquifer  is utilized in  excess of recharge (Wahl and Wahl, 1988).
                            SOUTH  CAROLINA
Savannah  River Site,  Barnwell  County

The  Savannah  River Site is a nuclear production facility operated by  the U.S.
Department  of Energy (DOE). The 300 mi2 site was purchased by DOE in 1951
and  the P Reactor began operating in 1954.  "Throughout its  operation, ground
water has been withdrawn from  6  production wells  within  a 1 mile radius of
the P Reactor.  Although the primary source of cooling water for the reactor
is the nearby Savannah  River, some  ground  water has been used for nuclear
production.    Since  1985,  ground  water  withdrawal has  been  reduced
drastically and 4 of the 6 wells have been abandoned (Bacchus,  1994).

The  Savannah River Site contains  numerous seepage wetlands  which are
typical of the  Southeastern  Coastal  Plain physiographic province.   Shallow
ground  water  which flows  laterally in  response  to  a  small gradient is
essential to  sustain these wetlands.  Highly  permeable  sandy soils underlain
by lower  permeability clays  sustain  this  lateral  flow.

The  aquitard  separating  the  surficial aquifer from the underlying  confined
systems  is  locally  permeable and discontinuous in extent.   Ground water
pumping  for the  site increased  downward leakage  from th-T-STirficial aquifer

 through the calcareous semiconfining unit  to recharge  the  confined  system.
 As  a result, the  stage  and duration  of the hydroperiod of the wetlands was
 altered  significantly  (Bacchus, 1994).

 Aerial  photographs  taken  between 1943  and  1992 were  analyzed  to
 investigate  changes in vegetation cover (Bacchus, 1994).  Observed changes
 in  vegetation  are at least in part  a  result of  these  perturbations in the
 ground  water  hydrology (Bacchus, 1994).   Forested areas  in  their natural
 state contain dense canopies of pondcypress (Taxodium  ascendens)  and  pond
 pine (Pinus serotina).   Vegetation changes included  the  loss of approximately
 2,825 acres of forested wetlands coupled with  an increase  of approximately
 2,183 acres of forested  uplands and  141  acres  of shrub wetlands including a
 predominance of fetterbush (Lyonia  lucida).  Only a small  portion of the  loss
 in forested wetlands  can  be attributed  to conversion of wetlands along the
 perimeter to planted  loblolly  pine  by the U.S.  Forest Service  after the  site
 was purchased by DOE.

 Ground  reconnaissance revealed that the  portion  of  the  wetland  complex
 closest to the  ground water  withdrawals  exhibited standing  dead and dying
 trees, canopy  dieback-of  pondcypress,  and tree  pathogens.   Under natural
 conditions,  cypress are  relatively  free  of  disease  and  pests.   However,
 cypress  associated with excessive subsurface  drainage are  susceptible to
 internal  fungal  pathogens  and  bark  and  leaf  beetles.    In  addition,
 encroachment  of  wetland   tree , species  more   tolerant  of  abnormal
 hydroperiods such as tulip poplar, sweet gum, wax myrtle,  and American
 holly was observed suggesting that  ground water levels had  been  lowered.
 Lastly,  subsidence  of surface  soil of  approximately 30  cm  had  occurred
 exposing cypress roots (Bacchus, 1994).   Because cypress roots  are  adapted
 to anaerobic conditions,  exposure is potentially lethal (Bacchus,  1995).
Harris  (Houston)  and  Galveston Counties

Rapid  population and  industrial growth since World ,War II in the Houston  -
Galveston area have combined with agricultural  demand to  increase the need
for  additional water_supplies.  Prior to 1954  when Houston  began augmenting
its  water supply with  surface water,  it  was the  largest city  in  the  nation
utilizing only  ground water for1  public supply.  Over  80%  of the ground water
demand  in some  of the outlying  areas  of  this region  is for  rice irrigation
(Neighbors,  1981).'     > ,

 As a result of large amounts of  water having been  pumped from the  ground,
 water  levels in the artesian  aquifers  have declined  by as muph as 200  ft in
 the  Chicot  aquifer and  325  ft in  the  Evangeline aquifer.   Associated  with
 these  water  level declines  is  extensive subsidence of the  land  surface.
 Subsidence was  first noticed  in this area in 1938.  Between 1940 and 1952
 subsidence averaged  0.4 ft/yr.   After  the  utilization  of ground water  was
 curtailed beginning in  the mid-1950s, subsidence  rates declined to about 0.1
 ft/yr.  Cumulative subsidence of over  9 ft has occurred near  the Houston  Ship
•Channel  and  subsidence  of  at least  1 ft has affected  over 4,500   mi2
 (Neighbors, 1981).

 Land  surface  subsidence  has become  critical  to  parts of the Houston  -
 Galveston  area  causing  permanent  inundation  or  increased  exposure  to
 flooding.   Since 1943, several thousand acres  of  bayfront land have been
 permanently submerged and if uncontrolled, subsidence would have  increased
 that total to  20,000  acres.   In  Harris  County,  945  mi2 have  been  heavily
 Impacted by permanent  submergence  and  increased flooding  (Neighbors,
 1981).   Presumably much  of this nearshore environment  has  or had some
 wetland values which have  been  irrevocably lost due to subsidence.

 Although  subsidence is not reversible, the rate  of subsidence can be slowed
 or stopped  with  reduction or cessation  of ground water extraction.   Several
 decades  of supplementing  ground water  with  surface  water  to meet  the
 Houston - Galveston region's growing demand have resulted in  reduced rates
 of compaction and  subsidence.   However,  current projections indicate  that
 all readily  available water  supplies will  be  exhausted during  the remainder
 of this century (Neighbors,  1981)'.  Conservation or  increased  ground water
 pumping  and  renewed loss  of  nearshore  acreage, will  be  among  the   few
 alternatives remaining.
 Balcones Fault  Zone  Edwards  Aquifer

 The Balcones  Fault Zone Edwards Aquifer is located in south central Texas
 and parallels  the  Balcones Escarpment..  It  consists of massive  limestone
 deposits averaging 400 to 500 ft  thick (Longley, 1992).   It is the sole source
 of  water for  San  Antonio's population  of about 1,000,000 (U.S. Geological
 Survey,  1990).

 Water levels  in the  aquifer and the  numerous  prolific springs which  are
 supported  by the  aquifer  are at risk of serious decline. The average recharge
 to  the  San Antonio  portion  of  this  aquifer  is  637,000 acre-ft  per year.

 Annual  recharge  is highly  variable and directly related to annual precipation.
 Pumping  is also  directly  related to annual rainfall.   Pumpage in 1989 was
 542,000 acre-ft.   During  the  summer of this  year, aquifer  levels decreased
 by more than  1 ft per day,  an occurrence which  has become commonplace in
 the spring and summer of recent years (Longley,  1992).  In 1991, the "largest
 well  in  the world" was drilled  in  this  area (Swanson, 1991).   Capable  of
 flowing  under artesian  pressure at a  rate  of 27,000  to  35,000  gpm, it
 potentially could produce an amount equal to 111%  of the annual recharge to
 the aquifer in  1956, the low year of record  (Longley, 1992).

 The  Balcones  Fault Zone Edwards Aquifer supports a diverse  assemblage  of
 over 40 species of aquatic  organisms.   Some  species are subterranean and
 others  inhabit  the many springs.   Several  species are extremely limited  in
 distribution.   For  example,  the  Texas  blind  salamander  (Typhlolmolge
 rathbuni) is found  only in a limited area around  the  San Marcos springs.  This
 highly adapted amphibian was the first species to  be placed oh  the federal
 Endangered  Species list.  Eleven other species of invertebrates, salamanders,
 and fish are being considered for  federal listing (Longley, 1992).

 This  unique community is at risk  of being impacted severely by declining
 water levels in the aquifer  (Longley, 1992).   In addition  to  reduced spring
 flow,  aquifer overdraft  promotes  the encroachment of highly  saline  water
 which threatens the fresh water biota of this  system.  A test well  located
 less than 100  yards from  major San Marcos springs was found to be highly
 saline.   Without  immediate  improved water management,  this ecosystem is
 likejy to  be impacted  severely (Longley,  1992).
Yahara River, Dane  and  Columbia  Counties

The City of  Madison  is  situated  on a  series  of  lakes in south  central
Wisconsin which are part of the Yahara River  system.   The river,  lakes,  and
ground water  form  interdependent parts of the regional  hydrologic  system.

The ground water bearing units in the 323 mi2 drainage basin of the upper
Yahara consist of two  aquifers.  The lower sandstone unit  is confined by an
overlying  heterogeneous  sandstone   layer (Fetter,  1977).   There is  a
considerable amount of leakage between these two units

The  majority of the municipal  and industrial  water  supply  comes  from deep
wells drawing  from  the  confined aquifer.   These wells  are  the  primary
source of ground water discharge from the  basin.  As 'a result of  pumping,  the
water level has been  lowered as much as 70 ft  in the confined aquifer and as
much as 20 ft in the  upper  unit.   Downward leakage from the upper aquifer
has  been enhanced.   An estimated 5.6 billion  ft3 of the upper  aquifer have
been dewatered.  The water table decline  in the upper aquifer has created a
gradient  toward the  well field  areas causing  ground  water which formerly
flowed into the lakes and streams to be intercepted (Fetter, 1977).

Prior to  1958  treated sewage effluent was discharged into  the Yahara River.
This  effluent was a  significant component  of  the  annual discharge  of  the
Yahara  River, particularly in  the summer months of low flow.   Beginning in
1958 the effluent was exported  from the  river basin  as  part  of a water
quality improvement  program for the  lower Madison  lakes.

The  loss of  ground  water  discharge and the  effluent  in  the Yahara  River
resulted  in a 50%  reduction of streamffow during periods of low  flow.  The
river  now is predicted  to be dry for periods in  some  years  (Fetter,  1977).
Undoubtedly  the combination of increased water table  depth,  loss of bank
storage  and reduction of stream flow to the point of an occasional dry river
bed  will  have  a detrimental  impact on the riparian ecosystem.

                                 Section  7

           Evaluation  of  the  Adequacy  of  Existing  Information
            on  Ecological  Impacts  of  Ground  Water  Overdraft

  Based  on the literature  and data reviewed  for this  study,  the  investigation
  and  documentation  of  ecological impacts  of ground  water  overdraft is
  clearly  not a major focus of attention for most researchers and regulators at
  this time.   Compilations  of  more  than  6,300 citations  published primarily in
  the  last  decade  on  ground  water  hydrogeology,  hydrology,  and  wetland
  ecology were  reviewed for  this  study  (examples  include  van der  Leeden,
  1991; Atkinson et  aL,  1986;  Emery,  in  preparation;  Hood,  1988;  Jalink,
  personal  communication;  Fjsk,  1989;  Stanford and Simons, 1992;  Stanford
  and Valett,  1994).   Fewer than 30 papers dealt directly  with  the  ecological
  impact   of   ground  water  overdraft.     Applicable  citations  were  also
,  infrequenty found in government publications  and refereed journals.

  As was  described in Section 6, ground water overdraft can be extensive in
  many diverse areas.  In some regions, research  is  occurring  to accurately
  evaluate  the nature  and extent  of  ecological  impacts  of  ground water
  overdraft.  Notable  examples  include  several  basins in.  Arizona,  California
  and  Florida.  Other researchers present  sufficient data to document the need
  for further study to understand this important  aspect of  wetland hydrology.
  However, this  topic  definitely  is  'addressed  inadequatey  in  the  technical
  literature  at  this time.
                            i       '   '       .       '   ~;               .
 Clearly,  investigation of the  ecological  impacts of  ground  water  overdraft is
 an  emerging  issue.   Historically, studies .on ground  water  decline  have
 focused  primarily on  the economic impact of water  shortage  and the  costs of
 developing new water  supplies.   Impacts of  ground  water decline  which
 affect people's  use  of  the land  and water such as  saline  intrusion  and land
 subsidence have been well documented.  .Investigation of  these impacts  has
 centered on  the resulting  geologic hazards, structural damage, and economic
 impacts  of abatement with little or no  attention being focused  on the  actual
 or  potential ecological  impacts  of these processes  (for  example  Katzer  and
 Brothers, 1988;  Atkinson et al.,  1986; Schumann et  al., 1985).

Another  example of the shortcoming  of the available  literature  in  addressing
the  ecological impacts of ground water overdraft  ist  that major- changes  in
surface hydrology are frequently  documented  without  noting the unavoidable
impacts  on  associated riparian  and aquatic  ecosystems.   For  example,
abundant  literature is available  on  springs,  lakes,  and  rivers  which are
almost  or completely  dewatered  without  noting the  associated  ecological
consequences (for example Wahl and  Wahl, 1988; Fetter, 1977).  Therefore it
is undoubtedly the case that the extent of ecological impacts resulting from
ground water overdraft exceeds those instances  currently documented  in the
available  literature.
Technical   Perspective

A  major  reason  for the  paucity of  information  on  this  subject  is  the
technical   complexity  of   performing  meaningful  studies.     Site-specific
studies are  needed.  Adequate data must be available on surface and ground
water  hydrology  as well  as sufficient ecological  data  to  measure  valid
trends  and physiological effects  of  sublethal  stress.   Studies  must  be of
sufficient  duration  to  identify  long-term  and cumulative impacts.

Most wetlands and riparian  areas are subjected to a variety of anthropogenic
ar\d  natural  perturbations.  In some  cases it  may  be difficult to distinguish
ecological  impacts  due  to. ground  water  pumping from  other causes of
impacts.   For example,  water level  decline in  riparian aquifers can  result
from  pumping as well as upstream surface  water impoundment,  stream
channel incision,  and natural  drought cycles (Swift, 1984; Scurlock, 1987).
Likewise,   livestock grazing,   watershed  degradation, recreation,  and  the
introduction of exotic species may  result in vegetation  changes similar to
those   associated  with   ground  water  overdraft  (Stromberg,   1993c;
Stromberg,  1994;  Hendrickson and Minckley, 1984).  Accurate  identification
of causal relationships requires  extensive  knowledge  of the  history of land
uses and carefully designed investigations.

Another, reason for the dearth of  studies on this subject is that the  current
scientific  foundation  for  understanding  wetland hydrology is weak.   Until
recently, there  have been  relatively  few  studies done on  the  relationship
between ground water and  wetlands  (Carter and Novistzki, 1988).  According
to  Winter (1988),  "Most hydrologic information  relative to wetlands has been
based  largely  on theoretical  studies  of  generalized  settings,  on scattered
field studies,  and  on hydrologic  intuition".   Calculation of  a water  budget
may be fraught with inaccuracies  (Siegel,  1988).  Particularly problematic is
the quantification  of  evaporation and  transpiration from  wetlands as well as

the seasonal and  annual changes  in ground  water flux in wetlands (Winter,
1988).   Without  an accurate  understanding  of the  spatial and  temporal
changes in water  movement in  wetlands and  riparian  areas, determination of
the magnitude of  impact of ground water pumping is  difficult.

Adequate surface  and ground water  monitoring data are  essential  but  often
are unavailable  and difficult to  obtain.   For example,  the fact that springs,
wetlands, and  seepage faces  may be  supported  by  shallow,  localized
unconfined aquifers  can  complicate efforts to obtain sufficient data in some
locations (Winter,  1988).    Such  systems  may  be  characterized,  by
considerable  seasonal and  annual  fluctuation in water  level  independent of
pumping.   For example,  Gerla  (1992)  noted  that  in  some  intermittent
wetlands, the water table  rose over  ten  times the  amount  predicted  by
infiltration due  to  physical  processes in  the  capillary fringe.   Furthermore,
while  of essential  ecological significance, hydrologic systems such,  as these
may be  unimportant  as  major water supplies  for human  needs  and  of low
economic value.   Consequently,  adequate long-term  water  level data may not
be  readily available.

Ecological .monitoring for impacts from-ground  water  overdraft  also may  be
problematic.   To be  of use  in preventing irreversible damage to  wetlands and
riparian  areas,  investigations  must  focus on  sublethal  stress responses
rather  than   mortality.   According  to  Bacchus  (1995), "Field  methods for
monitoring wetlands  were  designed  primarily  to  estimate  dominance and
(community  composition)  rather than to  detect  and measure responses  of
natural  systems to imposed stresses.   Although standard approaches may
provide  some   insight  into  stress  responses,  such interpretations  are
difficult because the  majority of monitoring  is short-term  and  there  is  a
dearth  of data on stress responses  in wetland species."   Factors such  as
changes  in  species composition  and  density,  reproductive  success,
susceptibility to pathogens, and  introduction  of new species  should  be
considered (Stromberg, 1994; Bacchus,  1995).

Inadequate information on  several key aspects of wetland ecology  further
hinders researchers' ability  to  develop  predictive models  of the  impacts  of
ground water drawdown.   In many  instances, it  is not known whether  certain
plant  species are  obligate  or  facultative phreatophytes in riparian  settings.
Consequently the  impact of water level  decline  beneath  the  depth of root
penetration may be  difficult to evaluate (Busch et al.,  1992).   Furthermore,
studies of wetland ecosystems are  rarely  designed to identify the earliest
indicators of   stress   resulting  from  water  level  decline,     Lastly,
ecophysiological  studies have  focused  more  heavily on the  effects  of
Hooding  on  root  tissue,  leaving  greater  uncertainty regarding  the

physiological tolerance of  short-term and long-term desiccation (Busch  et
al., 1992).
Regulatory  Policy  Perspective  .  ' -

It is not within the scope of this report  to  thoroughly explore the adequacy
of current  regulatory  policy in identifying and addressing instances in  which
ground  water pumping  results in  ecological impacts in wetlands.  However,
several  key factors are  readily apparent  in  reviewing  the  literature available
on this  subject.   Foremost  is the fact that federal agencies  and most  states
have  very  limited  regulatory  controls on  ecological  impacts  from ground
water overdraft (Smith,  1989).  While  the  ecological values associated with
maintaining surface  water  flows  are well  recognized,  the concept  is  riot
reflected in ground  water management in  most instances (Fort et al.,  1993).

At present,  most  approaches to management of ground water quantity are for
the^purpose of fulfilling  senior water  rights and not for the  protection  of
associated  ecosystems.    In  some states,  utilization of  ground water  is
unlimited as long as  the water is put to  beneficial  use  (Smith,  1989;  Parfit,
1993).  Limitations to this  broad  right  may be technically difficult  to invoke.
For example, in Texas, ground water drawdown  may be  limited  if the ground
water  meets  the definition of an "underground stream"  (Longiey,  1992).
Furthermore, the ongoing evolution of the  regulatory definition  of wetlands
complicates the issue (Lehr, 1991).  For  example, a decision  by the U.S.
Supreme Court was  required to  include  wetland areas  saturated  by ground
water (as opposed  to surface water) under protection by  the  Clean Water Act
(Want,  1988).

Other  regulatory  provisions which  provide  some measure  of  protection  to
wetland  ecosystems  may not be applicable to  impacts  from ground  water
drawdown.   Permits issued  under  Section 404 of the Federal  Clean Water Act
are required for  surface  water diversions which  involve fill  or excavation of
waters  of  the United States  including ditching,  channelization  and discharge
of dre'dged  materials.   However,  permits  generally are  not  required for
dewatering wetlands  and fiparian  areas through  ground water pumping.  This
is particularly ironic  as  the  areal  extent  of  ecological impact  may  be far
greater from dewatering  than  from  placing  fill for a project  such  as a road
crossing  (Bacchus,  1994;  Hill,  personal  communication).   Furthermore,
because the permits are issued  for individual  wetland  sites,  the permitting
process is  not conducive to the evaluation of cumulative impacts in an entire
hydrological  system  (National   Research  Council,  1991).    Likewise,
Environmental  Impact  Statements may  be  inadequate  in evaluating  the

  cumulative impact of pumping operations  on an aquifer a'nd  also greatly may
  underestimate  the  actual  drawdown  which  occurs  (Bureau   of  Land
  Management,  1994).    Finally,  mining  reclamation  plans  may" include
  revegetation  but  not the replenishment of  ground water pumped  during, pit
  excavation (Ross, 1992).

  Improved regulation  of  ground water  quantity to  manage the impacts  of
  water  level  decline  including  rising pumping  costs  has riot  been  readily
  accepted.  For example, in parts of the  High Plains dependent oh the Ogallala
  aquifer,  water  level decline is extensive  and rising pumping costs threaten
  the profitability of agricultural  water users.   However, in a  broad  survey  of
  water  users  conducted  in  this  area,  regulatory  controls  to  manage this
  shared  water resource  were  found to  be  unpopular  and  voluntary use  of
  conservation   measures  in-response to  economic necessity  was favored
  (Kromm  and  White, 1986).  However, while market mechanisms "have been
  successful in some  instances,  they may  be  ineffective in prevention  of
  impacts  to  natural resources of  limited monetary value  such  as  some
 wetlands  (Smith,  1989).

 Another  impediment  to  improved  management  of depleted ground  water1
 resources is  the  multijurisdictional nature of many ground  water  systems.
 Where  aquifers  cross state or  local boundaries,  individual  pumpers  have
 little  incentive   to  unilaterally   restrict  withdrawals  of   ground  water.
 Without  cooperative agreements, a "use it or lose  it"  mentality  may prevail.
 The need for cooperation between  states is  large.   In a  survey  to determine
 the  extent of  interstate  competition for  ground  water resources  in  the
 western  continental  United  States,  interstate   competition  was  found
 somewhere on the borders of all  states except Oregon (Smith, 1989).

 Water use planning  may be further complicated by the  absence of a detailed
 understanding  of  local hydrogeology.  Unlike surface water  users, individual
 ground  water  pumpers may  not recognize their co-dependence on a shared
 water resource.   Consequently, the  effects  of overexploitation  may not be
 recognized until they are considerable (Llamas  et al., 1992).
Legal   Perspective

There is no single  legal approach to ground water overdraft in water law in
the various  states.   State laws often do not explicitly address overdraft so
policies  may have to be inferred from other management  mechanisms such as
special  management  districts or  provisions  for  well  interference.   Among
the Western  states,  many  states  have a  de  facto policy  of  permitting

 unlimited ground water  mining (Fort  et al., 1993).   Furthermore, there is
 frequently poor  integration  between  the legal  framework  for regulating  the
 utilization of  surface  water and  ground water.  While  these two  resources
 are often very  different  in terms  of their  development and distribution,  they
 are hydrologically highly interconnected  and  in  some  instances must be
 managed as a unified system (Llamas et al., 1992).  A noteworthy example of
 a statutory  initiative  in this regard is the  Arizona  Riparian Protection  Act of
 1992.   The  act directs the Arizona  Department of Water Resources  to
 "evaluate  the   effects  of  ground  water  pumping  and  surface  water
 appropriations on riparian areas in the  state" (Stromberg,  1992).

 One of the  primary challenges with regard to improved management  of ground
 water overdraft  stems from the  fact that legal controls  on ground water
 quantity frequently are in the  private  domain.   The  present use  of  ground
 water  has  resulted   largely from the  accumulation  of   individual  private
 operations rather than larger scale, often  publicly  funded,  impoundment  and
 distribution  systems  more  commonly associated with 'utilization of  surface
 water  (Llamas et al.,  1992).   Water  rights often  are attached to property
 rights provided pumped water is used for a beneficial  purpose.  This has  led
 to  the  practice  of water farming  in some areas.   For   example, tens of
 thousands of  acres of irrigated farmland in rural Arizona have recently been
 purchased-by municipal water purveyors.  Their  intention is to retire the land
 from  agriculture  and  export  the  ground water  pumped on the property to
 rapidly  growing  nearby cities (Checchio and  Nunn, 1988).   If not carefully
 managed, such  export may result in environmental impacts  in  the area of
 origin.   It  is  ironic that  state  statutory  provisions  intended  to  curtail
declining ground  water levels in urban areas have  provided the  incentive  for
this extensive water farming  (Checchio and Nunn, 1988).

                                Section  8

                                                         ,   .   -  v
 The primary recommendations from this  study  fall into two categories  which
 will  be discussed  below.   First, under  certain circumstances,  the potential
 for ecological impacts from  ground water overdraft  is significant.  However,
 this phenomenon  commonly is  not investigated  in  most ground water or
 wetland studies.   Therefore, it is recommended that the  issue  receive more
 research and regulatory  attention in a broad array of contexts.  Second, one
 primary  focus  of  this  attention  should  be  on  an  enhanced  predictive
 capability  for early detection  of "high  risk"  hydrologic  settings and  "high
 risk"  wetland and  riparian communities  and species.   With care to  avoid
 erroneous  oversimplification, this  information  should be used to improve
 planning and  policy regarding water use and wetland  protection.

 As was described  in  Sections.6  and 7, the existing literature is sufficient  to
 conclude'that adverse ecological impacts from ground  water overdraft have
 occurred  in  many diverse  ecosystems.   The potential  exists for  similar
 impacts  to be  occurring  in  many ecosystems  which have not  yet  been
 investigated.   Studies to  identify and  quantify consequences from pumping
 should  be  conducted  more  frequently  and more  comprehensively.    The
 potential for  wetland  and  riparian .impacts  should be considered in regional
 water  use  planning  programs  where  water  level   decline   is  a  concern
 Similarly, this issue should be adequately addressed  in regional hydrological
 studies and  wetland  surveys.    Investigations  regarding   other  impacts
 associated  with  ground  water  overdraft  such  as  saline  intrusion  arid
 subsidence should  be  coupled with  or  should  include an  assessment of the
 potential for  ecological  impacts when  wetlands are  involved.   Permitting
 processes  pursuant to federal, state, and local environmental protection and
 natural resource management statutes  should include ah  assessment of the
 potential  for  impact from subsurface drainage for any development projects
 which may affect wetlands and riparian  areas.   In particular, a more rigorous
 and  comprehensive approach .should be employed in environmental  impact
 statements.   To  the  maximum  extent  feasible,  emphasis   should  be  on
 cumulative  impacts of regional  ground water  resource  development  rather
than on individual wells  or well  fields  or individual development projects.

 Because  hydrologic settings  and  wetland ecosystems are individually  unique
and  complex, site-specific studies will  be essential.   Therefore,  achieving

 the preceding  recommendation will- be  time-consuming and costly.    To
 maximize  the efficiency  with  which  impacts can be  identified,  or ideally
 prevented, additional  research effort  should  be  focused  on  developing
 predictive  capabilities  for  identifying  "high  risk" hydrologic  settings  and
 "high risk" ecosystems and  species.  This  approach can  then  be used  as a
 starting point to  improve planning and policy.  A  goal of these efforts should
 be prevention instead  of  restoration.   Restoration  of wetlands  is technically
 limited  at  this time and often  politically difficult and  costly.   Therefore,  use
 of  a  more proactive predictive  approach is likely  to achieve  better .results at
 a lower cost.

 As a starting point, examples  of high  risk  hydrologic settings might include
 locations  with  shallow ground water,  rapidly  drained  soil,  discontinuous
 confining  layers,  or  a  large  seasonal  difference  in  precipitation which
 results  in ground water dependence during the dry season.  For plant species
 which   may  be  particularly  sensitive  to hydroperiod  perturbations,  optimal
 rates of ground water  drawdown should  be  investigated to allow maximal
 use of the  resource  with minimal  ecological  impact.   Soil  type  and
 meteorological factors would  have to  considered (Mahoney  and Rood, 1992).

 Another predictive tool  which  should  be  developed  is the  species  specific
 indicators  of earliest  water  stress.    Impacts   must  be identified  before
 structural degradation occur in plants.   In some  cases  this  will  involve a
 better   understanding  of the  physiological effects of water  deprivation  on
 plant tissues.    In  other  cases, research  will be required to identify  the
 source  of  water* utilized by plant species,  which  may vary  seasonally,  by
 location or by age of the plant.  Understanding seasonal dependence on ground
 water,   precipitation and surface 'water  is  essential to  establish safe  limits
 on  drawdown.   Tracer studies using  the stable isotopes deuterium  and
 oxygen-18  to distinguish facultative versus  obligate phreatophytes appear to
 have  promise  (Dawson and Ehleringer,  1991; Busch et al.,  1992;  Dysart,
 1988) as  opposed to earlier studies  in  which plants were extracted from  the
 soil (for example Gary, 1963).

 In addition to an improved  understanding of the response  of  individuals and
 species to water  level  decline, it is  important to  consider that wetlands and
 riparian areas are  among  the  most  spatially  and  temporally  complex natural
 systems on earth (Richter and  Richter, 1992).  Riparian systems respond to
 water stress  by undergoing changes that occur  in  a  hierarchical fashion.
Whereas low  levels of stress cause  changes at  the  level of the individual,
 high levels  of  stress  result  in ecosystem  and   community  level  changes
 (Stromberg,  1992).   Few  studies  have   quantified  relationships  between
water stress  and  riparian ecosystem  response,  particularly in arid regions.

Although  highly complex,  enhanced predictive  capabilities are  needed  at
these levels  as well.   Linking ground  water and  ecological models  is one
potential  approach (Richter and Richter,  1992).

Predictive information  of this type should  be used  more  widely  in  ground
water management  programs.   A major  objective should  be to  create  a
management  approach in which there is  feedback between the occurrence of
hydroperiod  perturbations,  water  stress  tolerance  limits,  and allowed
pumping  rates.  For wells sited, in  highly sensitive areas,  permit  conditions
could require pumping to  be slowed  or  stopped when  plant  limiting soil
moisture  or  water level depths were reached (Groeneveld, 1989;  Stromberg
et al.,  1993bj.   The  long-term monitoring of  ground  water levels and
ecological impacts  required  by  the Soutwest  Florida  Water  Management
District is a  useful example which could  be utilized more widely.


                                 Section  9

 A  review of the  existing  literature on the environmental  impacts resulting
 from  ground water  overdraft revealed  that localized  wetlands and riparian
 areas throughout  the United States have undergone changes consistent with a
 loss  or reduction  of the ground water discharge which  previously had  been
 available.   While  many instances  are  localized, in  some  cases these impacts
 have  occurred on a  regional  basis  throughout entire  drainage  basins or
 aquifer  systems.          <

 The  locations in  which diseernable impacts have been  documented  in  this
 study are not hydrologically or ecologically unique.   It  is likely that  similar
 impacts are occurring  in  other  locations which  have not been investigated,
 particularly  in the arid  west  where  ground  water  is  frequently  the  only
 perennial water source available.  It  is  probable  that the  existing body of
 literature  significantly  underestimates  the  extent of  ecological impacts
 which are a result of ground water overdraft.

 As the demand for ground water continues to grow  and the remaining wetland
 and riparian  areas are subjected to  an increasing  array of developmental
 pressures, the importance of this issue is clear.   Left  unaddressed,  it is
 probable  that ecological  impacts of ground water  overdraft increasingly  will
 undermine  other  environmental  efforts such  as  wetlands restoration  and
 endangered  species protection.

 Efforts  to  assess  and  rectify impacts  from ground  water overdraft  must be
 approached  with   a full  understanding  of the complexity  of  hydrologic
 systems  and  wetland ecosystems.    Many  human  activities  and  natural
 processes  can result in ecosystem changes.   Causal"relationships  are  often
 difficult  to  identify.    Detailed,,  site-specific,  and  often  long-term studies
 are needed  to determine  pumping  rates which can be sustained without
 undesirable  impacts;   A  better predictive  capability  for  optimal rates of
 drawdown  and early indicators  of  ecological stress will be needed to  assist
water resource planners in being more  responsive to this  issue.

A  deliberate "effort should  be  made  to incorporate this  issue  into existing
environmental assessment and  water resource planning processes.  To the-

maximum  extent  possible, emphasis should be  on  cumulative  ecological
impacts and regional water allocation.

                                Section  10

 Atkinson, S.F., G.D. Milter, D.S. Curry and  S.B. Lee, 1986.  Salt Water intrusion
     -  Status  and  Potential  in  the  Contiguous  United  States.   Lewis
     Publishers  Inc., Chelsea, Michigan, 390 pp.
    •           '         .      v  '      '      '  •.   •          '•'•''-•'•
 Bacchus, ST.,  1994.   Initial  use  of  potential  ecological indicators to detect
     subsurface  drainage in  wetlands of  the  Southeastern  Coastal Plain,
     United States,  jn: J.A. Stanford and H.M. Valett (eds.), Proceedings of the
     Second  International Conference  on  Ground  Water  Ecology,  American
     Water  Resources Association, Herndon,  Virginia, pp. 299-308.

 Bacchus, ST., 1995.  Improved assessment of baseline  conditions  and change
    . in  wetlands  associated  with  groundwater  withdrawal  and  diversions.
     in:  K.J. Hatcher (ed.), Proceedings -of the 1995  Georgia Water Resources
     Conference,  Institute  of Government,  University   of Georgia, Athens,
     Georgia, pp.  158-167.

 Bacchus, ST., in press.  Determining sustainable yield for karst aquifers in
     trje southeastern  Coastal Plain:  a need for new approaches,   in: J.W.
     Borchers and C.D.  Elifrits (eds.), Current Research  and Case  Histories of
     Land Subsidence, Proceedings of the Dr. Joseph F.  Poland  Symposium on
     Land   Subsidence,  Association   of  Engineering   Geologists,  Special
     Publication, Star  Publishing Co.,  Belmont,  California.

Bellrose, F.C. and N.M.  Trudeau, 1988,   Wetlands  and their  relationship to ,
     migrating and  winter populations of waterfowl.   In:  D.D.  Hook (ed.), The
     Ecology and  Management of Wetlands  Vol. I: The  Ecology of Wetlands.
     Timber Press,  Portland, Oregon, pp. 183-194.

Bernaldez, F.G.,  J.M. Rey Benayas and A. Martinez, 1993. Ecological  impact of
     groundwater extraction on wetlands  (Duoro  Basin,  Spain).   Journal  of
     Hydrology 141: 219-238.

Bertoldi,  G.L., 1992.  Overview of Phase I Regional  Aquifer System  Analysis,
    Central  Valley, California,  in: K.R.  Prince  and  A.I. Johnson  (edsi),
    Regional Aquifer  Systems of the  United States -  Aquifers  of  the  Far

     West.   American Water  Resources Association  Monograph Series #16,
     AWRA, Bethesda, Maryland, pp.  15-28.

 Betancourt,  J.L., 1990.  Tucson's Santa  Cruz River and  the  arroyo legacy.
     Doctoral thesis, University of Arizona Press,  Tucson, Arizona, 239  pp.

 Birkitt, B.F. and  S. Gray, 1989.  Wetland impact evaluation  and  mitigation.  In:
     D.W. Fisk (ed.), Proceedings of the Symposium on Wetlands: Concerns and
     Successes.   American Water Resources Association, Bethesda, Maryland,
     pp.  259-267.

 Bloom, A.L,  1978.   Geomorphology - A Systematic Analysis of  Late Cenozoic
     Landforms.   Prentice-Hall,  Inc., Englewood Cliffs,  New Jersey, 510  pp.

 Bolen, E.G.,  1964.   Plant ecology of spring-fed salt marshes  in western  Utah.
     Ecological  Monographs 34(2): 143-166.

 Bredehoeft, J.D.,  S.S. Papadopulos and H.H.  Cooper, Jr., 1982.   Groundwater:
     the   water-budget  myth.    in: Scientific  Basis  of  Water-Resource
     Management,  Studies in  Geophysics,  American  Geophysical  Union,
     National Academy Press, Washington DC, pp. 51-57.

 Brinson,  M.N.,  1993. , A hydrogeomorphic  classification for  wetlands.  U.S.
     Army Corps of  Engineers,  Wetlands Research  Program Technical Report

Brown, M.T.  and M.F. Sullivan, 1988.  The value of wetlands in  low  relief
     landscapes,   in:  D.D. Hook  (ed.),  The Ecology and Management of Wetlands
     Vol. I: The  Ecology of Wetlands.  Timber Press, Portland, Oregon, pp.

Bryan, K., 1928.  Change in  plant  associations  by change  in ground  water
     level.  Ecology 28: 474-478.

Burbey,  T.J.,  1993.   Shallow ground  water  in the  Whitney Area, Southeastern
     Las  Vegas   Valley,  Clark  County, Nevada,   Part  II:  assessment  of  a
     proposed strategy to reduce the  contribution of salts, to Las  Vegas Wash.
     U.S.  Geological Survey Water-Resources investigations  Report 92-4051,
     58 pp.

Bureau of Land  Management,  1993a.  Final  Environmental Impact  Statement,
     Newmont Gold  Company's South Operations Area  Project.  U.S. Bureau of
     Land Manageiner*, Elko, Nevada.

  Bureau of  Land Management,  1993b.  Draft Environmental  Impact Statement,
      Bedell  Flat  Pipelines Rights-of-Way,  Washoe  County,  Nevada.   U.S.'
      Bureau of Land Management, Carson City, Nevada.

  Bureau  of  Land  Management,  1994.    Notice  of  intent  to  prepare  a
      supplemental  environmental  impact  statement and  to  conduct scoping
      for the Barrick Gqldstrike Mines Inc., Betze Project  in Elko and  Eureka
      Counties, Nevada.  Federal Register 59(168):  44999-45000.

 Busch, D.E., N.L  Ingraham  and S.D. Smith, 1992.  Water uptake  in woody
      riparian  phreatophytes  of the  southwestern-United  States:  a  stable
      isotope study.  Ecological  Applications 2(4): 450-459.
                                 .             - '     '             '  '

 Carothers, S.W., R.R. Johnson  and S.W. Aitchison,  1974.  Population structure
      and  social organization  in  southwestern  riparian birds.    American
      Zoologist  14:  97-108.                  .

 Carter, V. and R.P.. Novitzki,  1988.  Some comments  on the  relation between
     ground  water  and  wetlands,   in: D.D.  Hook (ed,), The Ecology  and
     Management of Wetlands Vol. I: The Ecology of Wetlands.  Timber Press
     Portland, Oregon,  pp. 68-86.                                         '

 Checchio, E. and S.C.  Nunn,  1988.  Water transfers in Arizona: assessing the
     adverse effects on areas  of origin,  in: M. Waterstone and RJ. Burt  (eds.),
     Proceedings of the  Symposium  on  Water-Use Data for Water  Resources
     Management,  American  Water  Resources  Association,   Bethesda
     Maryland, pp. 547-559.  ••

 Collins, C.A. and D.R.  Cline,  1991.  Ground-water  pumpage in the Columbia
     Plateau, Washington and Oregon, 1945-1984.  In:  K.R.  Prince  and  A:l;
     Johnson  (eds.),  Regional  Aquifer  Systems  of  the United  States  -
     Aquifers  of  the  Far West.   American  Water  Resources Association
     Monograph Series #16, AWRA, Bethesda,  Maryland, pp. 99-107.

 Conniff, R.,  1993.  California:  desert in disguise.  National  Geographic Special
     Edition: Water.  November 1993, pp. 38-53.

Crance, J.H. and L.S.  Ischinger,  1989.   Fishery  functions  and  values  of
     forested riparian  wetlands,  in:  F.E. Davis  (ed.), Water:  Laws and
     Management.   American  Water  Resources  Association,  Bethesda,
     Maryland, pp. 14A9-13.

Culler, B.C. et al., 1970.  Objectives, methods and environment - Gila River
     Phreatophyte  Project,  Graham County, Arizona. •  U.S.  Geological  Survey
     Professional Paper 655A, 25 pp.

Davis, J.C., 1993.   The hydrology and  plant  community relations of Canelo
     Hills  Cienega,  an -emergent  wetland in  southeastern Arizona,  Masters
     thesis,  University  of Arizona,  Tucson,  Arizona, 212 pp.

Dawson, T.E.  and J.R. Ehleringer, 1991.  Streamside trees that  do  not use
     stream  water.   Nature 350: 335-337.

Dileanis, P.O.  and  D.P. Groeneveld,  1989.  Osmotic  potential  and projected
     drought tolerance of four  phreatophytic  shrub species in  Owens  Valley,
     California.  U.S. Geological Survey Water-Supply Paper  2370-D, 21  pp.

Dingman,  S.L., 1994.   Physical Hydrology.  Prentice  Hall, Englewood Cliffs,
     New Jersey, 575 pp.

Dudley, W.W.,  Jr., and J.D. Larson, 1976.   Effect of irrigation pumping on
     desert  pupfish  habitats  in Ash  Meadows, Nye  County, Nevada.    U.S.
     Geological Survey Professional Paper 927, 52 pp.

Duel), L.F.W., Jr.,  1990.  Estimates of evapotranspiration  in  alkaline scrub and
     meadow  communities of  Owens Valley,  California  using  Bowen  ratio,
     eddy-correlation  and  Penman-combination  methods.   U.S.  Geological
     Survey  Water-Supply Paper 2370-E,  39 pp.

Dugan,  P.J., 1990.  Wetlands conservation: a review of current issues and
     required action.   International  Union for Conservation of Nature  and
     Natural  Resources, Gland,  Switzerland, 96 pp.

Dysart, J.E., 1988.  Use of oxygen-18 and deuterium mass-balance  analysis to
     evaluate induced  recharge to stratified-drift aquifers,   in:  A.D.  Randall
     and A.I. Johnson  (eds.), Regional Aquifer  Systems of the United  States -
     The Northeast Glacial Aquifers.  American Water Resources Association
     Monograph Series #11, AWRA, Bethesda, Maryland, pp.  133-154.

Egan, T.,  1994.  How  the  well  was won  -  Las  Vegas dukes it out over wateir.
     In: The Idaho  Statesman, April  10, 1994,  pp. 1.'

Emery,  S.H.,  in  preparation.   Bibliography  on  Hydrology  and Ecology.
     Environment and Health  Integrated Inc., Tampa, Florida, 30 pp.

 Emme, D.H. and D.E. Prudic, 1991.  Shallow ground water in the Whitney Area,
     Southeastern Las Vegas Valley, Clark County, Nevada, Part I: Description
     of  chemical  quality,  1986-1987.    U.S.  Geological  Survey  Water-
     Resources  Investigations Report 89-4117, 47 pp.

 England, A.S., LD.  Foreman and W.F. Laudenslayer, Jr., 1984,  Composition and
     abundance  of bird  populations  in  riparian systems of  the  California
     deserts,  in:  R.E. Warner :and  K.M. Hendrix (eds.), California Riparian
     Systems: Ecology,  Conservation, and Productive  Management.  University
     of California Press, Berkeley, CA,  pp.  694-705.

 Federal Register Vol. 58,  #14,  January 25,  1993, pp. 5938-5946.
                                1                            /
 Fetter,  C.W.,  Jr., 1977,  Statistical analysis  of  the  impact  of  ground water
     pumpage on low-flow hydrology.   Water Resources Bulletin 13(2): 309-

 Fetter,  C.W.,  1988.   Applied Hydrogeology.   Merrill Publishing Company,
     Columbus, Ohio, 591 pp.                           ,

 Fisk, D.W.(ed.),  1989.   Proceedings of the Symposium on Wetlands: Concerns
     and  Successes.   American Water  Resources  Association,  Bethesda,\
     Maryland, 568  pp.

 Fort, D.D., V.L..  Gabin,  E. Pinnes, 1993.  Managing  Groundwater Quality and
     Quantity  in .the  Western States.   National  Heritage Institute,  San
     Francisco, CA,  139- pp.

 Fowler, L.C.,  1.981.  Economic  consequences  of  land surface subsidence.
     Journal of the Irrigation and  Drainage Division  of the American Socfety
  v   of  Civil Engineers  107: 151-159.

 Freeze, R.A.  and  J.A. Cherry,  1979.   Groundwater.   Prentice-Hall,  Inc.,
     Englewood Cliffs, New Jersey, 604 pp.   '

 Fritzell,  E.K., 1988.  Mammals and wetlands. In: D.D.  Hook (ed.), The Ecology,
     and Management of Wetlands Vol. I: The Ecology of Wetlands.  Timber
     Press, Portland, Oregon, pp. 213-226.

Gary, H.L, 1963.   Root distribution  of  fiverStamen  tamarisk,  seepwillow, and
     arrowweed.   Forest Science X9(3): 311-314.

 Gerla, P.J., 1992.   The relationship  of  water-table  changes to the capillary
     fringe, evapotranspiration,  and  precipitation  in  intermittent  wetlands.
     Wetlands 12: 91-98.

 Graf, W.L., 1982.  Tamarisk and  river-channel management.   Environmental
     Management 6: 282-296.

 Graham, F. Jr., 1992. Las Vegas runs dry - gambling on water.  Audubon July-,
     August 1992, pp. 57-69.

 Groeneveld,  D.P.,   1989.    Shrub rooting  and  water acquisition on the
     threatened  shallow  groundwater   habitats  in   the   Owens  Valley,
     California,  in:  Proceedings of the  Symposium  on Cheatgrass Invasion,
     Shrub Die-off and Other Aspects  of  Shrub  Biology and Management. Las
     Vegas, Nevada, pp. 221-237.

 Groeneveld, D.P. and T.E. Griepentrog, 1985.  Interdependence of groundwater,
     riparian  vegetation  and streambank stability: a case study,  in: R.R.
     Johnson  et  al. (eds.),  Riparian  Ecosystems and  Their  Management:
     Reconciling   Conflicting  Uses.     First  North  American  Riparian
     Conference,  Tucson, Arizona,  USDA  Forest  Service General  Technical
     Report RM-120, pp.  44-48.                                      ,

 Grootjans, A.P. and  W.P.  Ten Klooster, 1980.  Changes of ground water regime
     in wet meadows.  Acta Bot. Neerl.  (the  Netherlands) 29(5/6): 541-554.

 Hagerthey, S.E. and W.C.  Kerfoot, 1992.   Groundwater influences  on the
     littoral communities of lakes/   in: J.A.  Stanford and J.J.  Simons  (eds.),
     Proceedings  of the  First  International  Conference  on Ground  Water
     Ecology,  American Water Resources  Association Technical Publication
     Series TPS-92-2, AWRA, Bethesda,  Maryland,  pp. 165-177.
Hall,  F.R.,  1968.   Base-flow recessions -  a review.
     Research 4:  973-983.
Water  Resources
Harding, M., 1993.  Redgrave and Lopham Fens, East Anglia, England: a case
    study  of  change  of  flora and  fauna due to groundwater  abstraction.
    Biological Conservation 66: 34-45.

Harrison, B.V. and  G.F. Kellogg,  1989.   Mapping  riparian/wetland habitats of
    the Nez Perce National Forest - a cooperative approach.  in: F.E. Davis
    (ed.),  Water:  Laws  and  Management.   American Water Resources
    Association, Bethesda, Maryland,  op.  6A15-23.

 Heindl, LA,  1961.   Groundwater in the  Southwest' - a perspective,   in: J.E.
      Fletcher  and  G.L.  Bender  (eds.), Ecology  of  Groundwater in  the
      Southwestern United States,  Proceedings of the thirty seventh annual
      meeting   of  the Southwestern "and Rocky  Mountain  Division of  the
      American Association for the Advancement of Science, pp. 4-26.

 Hehdfickson,  D.A. and  W.L. Minckley,  1984.  Cienegas - vanishing climax
      communities  of  the American  Southwest.   Desert  Plants 6(3): 131-175.

 Hill,  H.L., personal  communication.  Life Scientist,  Environmental Protection
      Agency,  Region  IX,  San Francisco, California.

 Hollett, K.J., W.R.  Danskin, W.F. McCaffrey and C.L. Walti, 1991.  Geology and
     water resources of  Owens  Valley,  California.  U.S. Geological Survey
     Water-Supply Paper 2370-B, 77  pp.

 Hood, D.D., (ed.),  1988.  The Ecology and  Management of Wetlands, Vol. 1: The
     Ecology of Wetlands and Vol. II: Management,  Use and Value of Wetlands.
     Timber Press, Portland,  Oregon.
                              •-,/,'          .
 House  Committee on Natural  Resources, 1994.   Analysis and  modeling of
     water supply  issues for the region  bounded  by Hiilsborough, Manatee,
     Pasco, and  Pinellas Counties: first year report.   Florida  House of
     Representatives,  Tallahassee, Florida, 110  pp.

 Jackson,  J.M.  and D.T.  Patten,  1988.    Plant-soil-water relationships in  Las
     Vegas Wash.  U.S. Bureau of Reclamation Technical Report REC-ERC-88-
     4,  Colorado River  Basin Salinity. Control Project, Boulder City,  Nevada,
     30 pp.

 Jalink,  M.H.,  1994.   Personal communication.   Ecologist,  Water and Soils
     Department, Kiwa N.V. Research  and  Consultancy, the Netherlands.
    •   '       .    •    /'            -..-•',                  • >
 Jarisen, A.J.M. and C. Maas,  1993.   Ecohydrological  processes in almost flat
     wetlands.   Proceedings  of  the  Engineering.  Hydrology  Symposium,
     Hydraulics  Division  of the American •', Society of  Civil  Engineers,  San
     Francisco, CA, July  1993, pp. 150-155.
                          •                                         /      ,
Jenkins,  S.R.  and R.K.  Moore,  1984.  Effect of  saltwater intrusion  on  soil
     erodibility  of Alabama  marshlands.    Water  Resources  Research
     Institute,  Auburn  University, Auburn, Alabama,  37 pp.

Johnson, A.I., 1981.  Foreword to five papers on land subsidence.  Journal of
     the  Irrigation  and Drainage  Division of the  American  Society  of  Civil
     Engineers 107: 113-114.

Johnson, R.R. and  L.T. Haight, 1984.  Riparian  problems and initiatives in the
     American Southwest: a  regional  perspective,   in: R'.'E. Warner and  K.M.
     Hendrix (eds.), California  Riparian Systems:  Ecology,  Conservation, and
     Productive  Management.  University  of  California  Press, Berkeley, CA,
     pp.  404-411.
Judd, B.I., J.M. Laughlin, H.R. Guenther and R. Handegarde, 1971.  The lethal
     decline of mesquite on the Casa  Grande National Monument.   Great Basin
    .Naturalist 31:  153-159.
Katibah,  E.F., 1984.   A brief history  of  riparian forests  in  the Central Valley
     of California,  in:  R.E. Warner and K.M. Hendrix  (eds.), California Riparian
     Systems: Ecology, Conservation,  and Productive  Management.   University
     of California Press, Berkeley, CA, pp. 23-29.

Katzer, T. and K. Brothers, 1988.  Water  conservation,  myth or  mandate, an
     arid Sourhwest perspective,   in: M. Waterstone  and  R.J.  Biirt  (eds.),
     Proceedings of the Symposium  on Water-Use  Data for Water Resources
     Management,  American  Water  Resources   Association,  Bethesda,
     Maryland, pp.  273-281.

Klopatek, J.M., R.J. Olsen,  C.J.  Emerson, and  J.L. Jones, 1979.  Land use
     conflicts  with  natural  vegetation in  the  United States.   Environmental
     Conservation  6:191-200.

Kondolf, G.M. and R.R. Curry, 1986.   Channel erosion along the Carmel River,
     Monterey County,  California.   Earth Surface Processes and Landforms 11:

Kondolf, G.M., L.M. Maloney  and  J.G. Williams, 1987.  Effects of  bank storage
     and well  pumping on  base flow,  Carmel  River,  Monterey  County,
     California.  Journal of Hydrology 91:351-369.

Kopper, W.  and  D.  Finlayson, 1981.   Legal aspects of subsidence due to well
     pumping.    Journal  of  the  Irrigation  and  Drainage Division  of the
     American Society of  Civil  Engineers 107: 137-149.

  Kromm,  D.E.  and S.E. White,  1986.   Variability  in adjustment preferences to
      groundwater depletion  in  the  American High  Plains.   Water  Resources
      Bulletin 22(5):  791-801.

  Lehr, J.H.,  1991.  Wetlands: a threatening  issue.  Ground Water 29(5V 642-
      645.                             ,;       .                  •

  Lewis, D.C.  and R.H. Burgy,  1964.  The relationship"between oak tree roots and
      groundwater in fractured rock as  determined by tritium tracing.  Journal
      of Geophysical Research 69(12): 2579-2588.

  Lieurance,  F.S.,  H.M.  Valett,  C.S. Crawford  and J.C.  Modes; Jr.,   1994.
      Experimental  flooding  of  a  riparian  forest:  restoration of  ecosystem
      functioning.  In: JA Stanford and H.M. Valett  (eds.), Proceedings  of the
      Second International Conference on  Ground Water  Ecology, Atlanta,
      Georgia.   American Water  Resources  Association, Bethesda,  Maryland'
   -   pp.  365-374.                                                        '

 Llamas,  M.R.,  1988.    Conflicts between, wetland conservation   and
      groundwater  exploitation:  two  case  histories in  Spain.   Environmental
      Geology and Water Sciences 11(3): 241-251.

 Llamas,  M.R.,  1989.    Wetlands  and  groundwater:  new  constraints in
      groundwater  management   In:  Proceedings  of the symposium held at
      Benidorm,   Spain,  October   1989.    International   Association  of
     Hydrological Sciences Publication #188,  IAHS, Washington DC, pp.  595-

 Llamas, M.R., W. Back  and  J.  Margat, 1992.   Groundwater use: equilibrium
     between social benefits and  potential environmental  costs.   Applied
     Hydrogeology 1(2): 3-14.
     i       •••..''                                ,'         •
 Loftus, W.F.,  R.A, Johnson and G.H. Anderson, 1992.  Ecological impacts of the
     reduction of  groundwater levels  in  short-hydroperiod  marshes of  the
     Everglades.  In: J.A. Stanford and J.J. Simons (eds.), Proceedings of the
     First  International Conference on Ground Water  Ecology, American Water
     Resources  Association  Technical  Publication  Series   TPS-92-2
     Bethesda, Maryland, pp.  199-208.

Longley, G., 1992. The subterranean aquatic  ecosystem of the Balcones  Fault
     Zone  Edwards Aquifer in Texas -  threats from oyerpumping.  in:  J.A.
     Stanford and  J.J.  Simons  (eds.),  Proceedings of the  First International
     Conference on  Ground  Water  Ecology,  American Water   Resources

     Association Technical  Publication  Series TPS-92-2,  AWRA,  Bethesda,
     Maryland, pp. 291-300.

Mahoney, J.M. and S.B. Rood, 1992.  Response of a hybrid poplar to water table
     decline in different  substrates.   Forest  Ecology and Management 54:

Manning, R., 1994.  Going for the gold.  Audobon, January-February, 1994, pp.

Marie, J.R., 1976.  Model analysis of effects on water levels at Indiana Dunes
     National Lakeshore  caused by construction  dewatering.   U.S.  Geological
     Survey Water-Resources Investigations 76-82, 32 pp.

McGlothlin,  D., W.L Jackson and P. Summers, 1988. Ground water, geomorphic
     processes,  and  riparian  values:  San Pedro River,  Arizona.  In: M.
     Waterstone and  R.J. Burt (eds.), Proceedings of the Symposium on Water-
     Use Data for Water Resources  Management, American Water  Resources
     Association, Bethesda,  Maryland, pp. 537-545.

McKnight,  S.K., 1992.  Transplanted  seed bank response to drawdown time in
     a created wetland in east Texas.  Wetlands 12(2): 79-90.

McNatt,  R.M., R.J. Hallock  and A.W. Anderson,  1980.   Riparian habitat and
     instream  flow studies,  Lower Verde River:  Fort  McDowell  Reservation,
     Arizona.  Riparian Habitat  Analysis Group, U.S. Fish and Wildlife Service
     Region 2, Albuquerque,  New Mexico, 52 pp.

Meinzer, O.E., 1927.   Plants as, indicators of, ground  water.   U.S.  Geological
     Survey Water-Supply Paper 577, 95  pp.

Meyers,  T., 1994.  Cumulative  hydrologic effects of open pit  gold mining the
     Humboldt River drainage.  Report to the Sierra Club,  21 pp.

Miller,  G.C.,  1993.   Potential  dewatering  impacts on  the Humboldt  River
     Basin:   Briefing  paper to  the Cooperative  Extension  Faculty, University
     of Nevada, Reno, Nevada, 9 pp.

Miller,  G.C.,  personal   communication,  Department   of Environmental and
     Resource Sciences; University of Nevada, Reno, Nevada.

 National  Research  Council,  1991.    Restoration  of  Aquatic  Ecosystems:
      Science,  Technology, and  Public Policy.   National  Research  Council,
      Washington DC, 485 pp.

 Neighbors, R.J., 1981.  Subsidence in Harris and GalvestdrNGounties, Texas.
      Journal of the Irrigation and  Drainage Division of  the American Society
      of  Civil Engineers  107: 161-174.            ~-

 Newton, J.G.,  1981.  Induced  sinkholes: an engineering  problem.   Journal of
      the  Irrigation  and  Drainage Division  of  the American  Society  of  Civil
      Engineers  107:  175-185.

 Niering,  W.A.,  1988.   Endangered, threatened and  rare wetland  plants and
      animals of the continental United States, in: D.D, Hook (ed.), the  Ecology
      and Management of Wetlands Vol. I: The  Ecology  of  Wetlands.  Timber
      Press, Portland, Oregon, pp. 227-238.

 Ormiston, B.G., S. Cook, K. Watson and C. Reas, 1994. Annual comprehensive
     report:  ecological and  hydrological monitoring  of  the  Cypress  Creek
     wellfield and  vicinity, Pasco County, Florida;  October 1992 through
     September 1993.  Prepared for the West  Coast Regional  Water Supply
     Authority,  Clearwater, Florida, 27  pp.                    '

 Parfit, M., 1993.   Sharing the wealth  of  our  water.   National  Geographic
     Special  Edition: Water. November  1993, pp. 20-37.

 Perkins, D.J., B.N. Carlsen, M. Fredstram, R.H. Miller, C.M.Rofer, G.T. Ruggerone
   \ and, C.S. Zimmerman, 1984.  The effects ^of groundwater  pumping  on
     natural spring  communities in Owens Valley.   In: R.E.  Warner and  K.M.
     Hendrix  (eds.), California Riparian Systems:  Ecology, Conservation,  and
     Productive  Management.    University  of  California  Press,   Berkeley,
   ,  California,  pp.  515-527.

Peters, D.D.-, 1989.   Status and  trends  of  wetlands in the  California Central
     Valley.  In:  D.W. Fisk (ed.), Proceedings of the Symposium on Wetlands:
     Concerns  and  Successes.   American Water  Resources Association,
     Bethesda, Maryland, pp.  33-44.

Pister, E.P. and J.H. Kerbavaz, 1984.  Fish Slough: a case  study in management
     of a desert wetland system.   In:  R.E.  Warner and  K.M. Hendrix (eds.),
     California  Riparian  Systems:  Ecology, Conservation,  and  Productive
     Management.    University of California  Press,  Berkeley, California   pp

Poland,  J.F., 1981.   Subsidence in the United States  due to ground-water
    withdrawal.    Journal  of  the  Irrigation  and  Drainage Division  of the
    American Society of Civil Engineers 107: 115-135.

Rea, A.M.,  1983.  Once A River - Bird Life and Habitat Changes on the Middle
    Gila River.  University  of Arizona Press, Tucson, Arizona, 285 pp.

Richter, B.D.  and H.E.  Richter, 1992.   Development  of groundwater  and
    ecological models for  protecting a southwestern riparian ecosystem,   in:
    J.A.  Stanford  and  J.J.  Simons  (eds.), Proceedings   of  the  First
    International Conference  on  Ground  Water  Ecology,  American Water
    Resources Association Technical  Publication  Series TPS-92-2, AWRA,
    Bethesda, Maryland,  pp. 231-245.

Riley, T.Z.  and T.A. Bookhout, 1990.  Response of aquatic macroinvertebrates
    to  early-spring  drawdown  in nodding  smartweed 'marshes.    Wetlands
    10(2):   173-185.

Robinson,  T.W.,  1958.   Phreatophytes.  U.S. Geological Survey Water-Supply
    Paper  1423, 84 pp.

Rochow, T.F., 1993.   The effects of water table level changes on fresh-water
    marsh  and cypress  wetlands in the northern Tampa  Bay Region.  Update
    of  a  memorandum  dated  February 9,  1989,  Southwest Florida Water
    Management District, Brooksville, Florida, 17 pp. plus appendices.

Rochow, T.F., 1994.   The effects-of water tabte level changes on fresh-water
    marsh  and  cypress  wetlands in the  northern  Tampa  Bay  Region.
    Southwest  Florida  Water  Management District Environmental Section
    Technical   Report   1994-1,  Southwest  Florida  Water  Management
    District,  Brooksville,  Florida, 21 pp.

Rochow, T.F.  and P. Rhinesmith, 1991.  Comparative  analysis of  biological
    conditions in  five cypress  dome wetlands  at  the Starkey  and Eldridge-
    Wilde   well   fields  in southwest  Florida.    Southwest  Florida Water
    Management  District  Environmental Section  Technical  Report 1991-1,
    Southwest Florida  Water Management  District, Brooksville,  Florida,  67

Rogers, LS. efal.,  1987.   Overview of water resources in  Owens Valley,
    California.    U.S.  Geological   Survey  Water-Resources  Investigations
    Report 86-4357, 39  pp.                        •'..'•

 Ross, C., 1992.  Gold mining affects ground water.  Ground Water Monitoring
     Review, Fall  1992:  100-102.

 Schumann, H.H., R.L Laney and LS. Gripe, 1986.  Land subsidence and earth
     fissures caused by ground-water depletion in southern Arizona,   in: T.W.
     Anderson and A.I. Johnson (eds.), Regional "Aquifer Systems of the  United
     States  -  Southwest  Alluvial  Basins  of  Arizona.   American  Water
     Resources Association Monograph Series #7, AWRA, Bethesda, Maryland,
     pp.  81-91.

 Scurlock,  D., 1988.   The Rio Grande Bosque: ever changing.   New Mexico
     Historical  Review 63: 131-140,

 Segelquist, C.A., M.L. Scott and G.T. Auble, 1993.  Establishment of Populus
     deltoides  under simulated  alluvial  groundwater  declines.   American
   ,  Midland Naturalist  130:  274-285,

Shupe,  M.G. and  P.J.  Gleason,  1989.    Wetlands  impacts  resulting  from
   .groundwater  withdrawals in  the  Jensen  Beach Peninsula Area.   In: RE.'
     Davis (ed.), Water: Laws and Management.   American Water Resources
     Association, Bethesda, Maryland, pp.  7B-5.

Siegel, D.I., 1988.   A review  of the recharge - discharge function of wetlands.
     In:  D.D. Hook (ed.), The  Ecology and  Management  of Wetlands Vol.  I: The
     Ecology of Wetlands.  Timber Press, Portland, Oregon, pp. 59-67.

Smith, Z.A.,  1989.  Ground Water in the West.  Academic Press Inc., San Diego,
     California,  308 pp.

Solley, W.B. and R.R. Pierce, 1988.  Trends in water use in the United States,
     1950-1985. In: M. Waterstone  and R.J. Burt  (eds.), Proceedings of the
     Symposium on  Water-Use  Data  for  Water  Resources  Management,
     American Water Resources Association, Bethesda,  Maryland,  pp.  31-39.

Sonenshein,  R.S.  and R.H.  Hofstetter, 1990.   Hydrologic effects of  well-field
     operations  in  a wetland, Dade County,  Florida.  U.S. Geological Survey
     Water-Resources Investigations  Report  90-4143, :16 pp.

Sorenson,  S.K., F.D. Dileanis and  F.A.  Branson, 1991.   Soil  water and
     vegetation  responses to precipitation and changes in depth to ground
     water in Owens Valley, California.  U.S. Geological  Survey Water-Supply
     Paper 2370-G, 54 pp.

Sorenson, S.K., R.F. Miller, M.R. Welch, D.P. Groeneveld and F.A.  Branson, 1989.
     Estimating  soil  matric  potential  in  Owens Valley,  California.    U.S.
     Geological Survey Water-Supply Paper 2370-C, 18 pp.

Southwest Florida Water  Management District, 1996.   Northern  Tampa Bay
     Water Resources Assessment Project, Volume  I:  Surface-Water/Ground-
     Water Relationships  and Volume  II:  Saline  Water  Intrusion  and Water
     Quality  (plus  appendices).   Southwest Florida  Water Management
     District,  Resource  Evaluation Section,  Brooksville, Florida.

Stanford,  J.A. and  J.J.  Simons  (eds.),  1992.   Proceedings of the  First
     International Conference on  Ground  Water  Ecology.   American Water
     Resources Association, Bethesda, Maryland, 420 pp.

Stanford,  J.A.  and H.M.  Valett (eds.),  1994.   Proceedings of  the  Second
     International Conference on  Ground  Water  Ecology.   American Water
     Resources Association, Herndon, Virginia, 390 pp.

Stromberg, J.C.,  1992.   Effects  of groundwater  pumpage and surface  flow
     reduction  on  riparian  .ecosystems  in  Arizona.    Unpublished  report
     submitted to the  Arizona Department of Water Resources,  17  pp.

Stromberg, J.C.,  1993a.  Instream flow  models for  mixed  deciduous  riparian
     vegetation within a semiarid region.   Regulated Rivers  Research  and
     Management 8:  225-235.

Stromberg, J.C., 1993b.  Riparian- mesquite forests: a review of their ecology,
     threats^  and, recovery   potential.    Journal  of the Arizona - Nevada
     Academy of  Science  27: 111-124.

Stromberg,  J.C.,  1993c.    Fremont  cottonwood-Goodding  willow riparian
     forests:  a  review of their  ecology,  threats,  and recovery  potential.
     Journal of the Arizona -  Nevada Academy of  Science 26: 97-110.

Stromberg, J.C.,  1994.  Draft  Riparian  Protection Program -  Legislative
     Report.   Arizona Department of Water  Resources,  Phoenix,  Arizona, 2

Stromberg, J.C., J.A.  Tress, S.D. Wilkins  and S.D. Clark, 1992.   Response  of
     velvet mesquite  to groundwater decline.   Journal of Arid Environments
     23: 45-58.

 Stromberg, J.C., M.R. Sommerfeld, D.T. Patten, J. Fry, C. Kramer, F. Amalfi,  and
     C.  Christian, 1993a.   Release  of  effluent  into the Upper  Santa Cruz
     River,  Southern Arizona: Ecological  considerations,  in:  Proceedings of
     the Symposium  on Effluent  Use Management, Tucson, Arizona.  American
     Water  Resources Association, Bethesda, Maryland,  pp. 81-92.

 Stromberg,  J.C.,  S.D. Wilkins,  J.A.  Tress, 1993b.   Vegetation-hydrology
     models:  implications for  management of  Prosopis  velutina  (Velvet
     mesquite) riparian ecosystems.   Ecological  Applications 3(2):  307-314.

 Stumm, W., and J.J.  Morgan, 1981.  Aquatic Chemistry.  John Wiley  and  Sons,
     New York, 780 pp.

 Suso, J. and M.R. Llamas, 1993.  Influence of groundwater development on  the
     Donana National Park ecosystem  (Spain). Journal of Hydrology 141: 239-

 Swanson, G.A.,  1988.  Aquatic habitats of breeding waterfowl,  in:  D.D.  Hook
     (ed.), The Ecology and Management of  Wetlands Vol. I: The Ecology of
     Wetlands.   Timber Press,  Portland, Oregon, pp. 195-202.

 Swanson, G.J., 1991.  Super Well is deep in the  heart of Texas.  Water Well
     Journal  45(7): 56-58.

 Swift, B.L,  1984.  Status  of riparian ecosystems in  the  United States.  Water
     Resources  Bulletin 20(2): 223-228,

Tiner,  R.W.,  Jr.,  1990.   Use of   high-altitude  aerial  photography  for
     inventorying forested wetlands in  the  United States.  Forest Ecology and
     Management 33/34: 593-604.

United States Geological  Survey,  1984.    National Water  Summary  1983 -
     Hydrologic Events and Issues. USGS Water-Supply Paper 2250, 243 pp.

United States Geological  Survey,. 1985.    National Water  Summary  1984 -
  .   Hydrologic  Events, Selected Water-Quality  Trends and  Ground-Water
     Resources.  USGS Water-Supply Paper 2275, 467 pp.

United States Geological  Survey, 1990.    National Water  Summary  1987 -
     Hydrologic Events and Water Supply and  Use.  USGS Water-Supply Paper
     2350, 553 pp.

United  States  Geological Survey, 1995a.   Water-Related  Scientific Activities
    of the U.S. Geological Survey in Nevada, Fiscal Years 1993-1994,  USGS
    Open-File  Report 95-126, Carson City, Nevada, 69 pp.

United  States  Geological  Survey, 1995b.   Programs in  Nevada - Fact Sheet.
    U.S.  Department of  Interior, FS-028-95.

United  States Water  Resources Council, 1978.  The Nation's Water  Resources
    1975-2000:  Water  Quantity,  Water  Quality  and   Related  Land
    Considerations,  Vol.  2: Second National  Water  Assessment.   United
    States Water Resources Council, Washington DC, 531 pp.

van der Leeden,  F., 1991.  Geraghty and Miller's Groundwater  Bibliography,
    fifth  edition.   Water  Information   Center,  Lewis  Publishers  Inc.,
    Plainview, New York, 507 pp.

Wahl,  K.L.  and T.L Wahl,  1988.    Effects of regional  ground-water  level
    declines on streamflow  in the  Oklahoma Panhandle,  in:  M. Waterstone
    and RJ.  Burt (eds.), Proceedings of  the Symposium on Water-Use  Data
    for  Water  Resources  Management,  American  Water  Resources
    Association, Bethesda,  Maryland, pp.  239-249.

Walters, M.A., R.O.  Teskey  and T.M. Hinckley, 1980.  Impact of water level
    changes  on  woody riparian  and  wetland  communities,  Volume  VII:
    Mediterranean region, western  arid and semi-arid region.  U.S. Fish  and
    Wildlife Service Biological  Services Program FWS/OBS-78/93, 83 pp.

Want, W.L., 1988.   Recent development in federal wetlands litigation.  In: D.D.
    Hook (ed.), The  Ecology and Management of Wetlands VoL II:  Management,
    Use  and  Value of Wetlands.   Timber Press,  Portland, Oregon, pp. 382-

Wassen, M.J.,  A. Barendregt, M.C. Bootsma and P.P. Schot, 1989.  Groundwater
    chemistry and vegetation of gradients from  rich fen  to  poor  fen  in the
     Naardermeer  (the Netherlands).  Vegetatio 79: 117-132.

Wentz, W.A.,  1988.  Functional status of the nation's wetlands,  in: D.D. Hook
     (ed.), The Ecology and  Management of Wetlands Vol. II: Management, Use
     and Value of  Wetlands.  Timber Press, Portland, Oregon, pp. 50-59.

Westenburg, C.L., 1993.   Water-resources data for  the Devils Hole area, Nye
     County, Nevada, July 1978 -  September 1988.  U.S. Geological Survey
     Open-File Report 90-381, 13 pp.                                 ~

White, N.D.,  W.G.  Matlock  and H.C. SchWalen,  1966.   An appraisal  of the
     ground-water  resources  of  Avra  and  Altar  valleys,  Pima  County,
     Arizona.   Arizona Land  Department Water Resources Report 25: 1r66.

Williams,  C.D.,  1984.   The  decline of  Ash  Meadows, a  unique  desert
     ecosystem,  In: R.E. Warner and K.M. Hendrix (eds.), California Riparian
     Systems: Ecology, Conservation, and  Productive  Management.   University
     of California Press,  Berkeley, California,  pp. 716-719.

Williams, J.E.  and  D.W. Sada, 1985.  America's  desert fishes:  increasing their
     protection under the  Endangered  Species  Act.   Endangered  Species
     Technical Bulletin 10:  8-14.

Williams, J.E., J.E. Johnson,  D.A.  Hendrickson,  S.  Contreras-Balderas, J.D.
     Williams,  M. Navarro-Mendoza, D.E.  McAllister,  and J.E.  Deacon, 1989.
     Fishes of North America:   endangered, threatened or of special concern,
     ,1989.  Fisheries  14: 2-20.

Winter,  T.C., 1988.   A' conceptual framework for  assessing  cumulative
     impacts   on  the hydrology  of  nontidal  wetlands.    Environmental
   ,  Management 12(5): 605-620.

Young, H.W., R.E. Lewis and R.L Backsen,  1979.  Thermal  groundwater
     discharge and associated convective heat flux,  Bruneau - Grandview
     area,  Southwest  Idaho.   U.S. Geological  Survey Water-Resources
     Investigations  Open File Report 79-62.