From:
from lower altitudes (Earman and Phillips
2003). Geochemical hydrograph separa-
Focus on tions in the highest basins of the Rockies
demonstrate that, even during the peak
snowmelt, 60% or more of the streamflow
is supplied by ground water (Liu et al.
Climate Change 2004). On the other hand, recharge from
streamflow infiltration through fans and
basin floors depends on large, rapid, but
Western Ground Water and Climate generally infrequent, outflows of runoff
from the mountains onto surrounding fans
Change—Pivotal to Supply Sustainability and basins. The mix of mountain vs. fan
and basin recharge presumably varies from
basin to basin and from year to year, but
these variations are poorly understood in
or Vulnerable in Its Own Right?
most of the West.
by Michael Dettinger, U.S. Geological Survey, La Jolla, California; As the region has warmed in recent
decades (Cayan et al. 2001), precipitation
has occurred more frequently as rain rather
and Sam Earman, Desert Research Institute
The NGWA policy statement on ground the same tendency for ground water to than snow (Knowles et al. 2006) and snow-
water and global climate change respond more to slow climate fluctuations packs have thinned (Mote 2003). If warm-
(www.ngwa.org/ngwainwashington/ than to fast ones (Changnon 1987, ing continues, snowline elevations are
climatechange.cfm) outlines several cur- Dickinson et al. 2004) makes ground water expected to rise. If this happens, mountain
rent and potential impacts of climate an attractive option during droughts. This recharge may also be expected to decline,
change on water supplies and concludes could also mean that ground water will because recharge areas will shrink and the
that ground water will be pivotal to sustain- respond more strongly to the comparative- amount of snowmelt available to infiltrate
able water supplies because of its capacity ly slow trends associated with climate at any one time will dwindle. Using the
to balance large swings in precipitation and change than to isolated droughts. The link- University of Washington’s VIC model
demand. The impacts discussed include ages between climate and ground water are (Maurer et al. 2002), recent simulations by
reductions in snowpacks, changes in pre- inherently complex. In anticipation of the authors of near-surface water budgets
cipitation, sea level rise, warming surface unprecedented times ahead, some of those in the western mountains, with and without
waters, and changes in water demand. intuitions need to be reassessed. warming, suggest that the partitioning of
However, in the western United States, There are likely many different ways net precipitation between surficial runoff
other important challenges also face that warming may influence ground water and subsurface runoff may respond to
ground water science and management supplies. One particular concern in the warming with declines in subsurface runoff
communities because climate change may western United States is warming’s effects of as much as 50%. We conjecture that the
impact ground water supplies directly and on snowpacks. In the West, because moun- declines would, most likely, be reflected
indirectly, in ways that have, to date, been tains are generally wetter and cooler than also in ground water recharge declines.
largely unexplored. adjacent basins, most ground water is On the other hand, another simulated
Current climate change projections are derived from mountain precipitation. Some result of warming in the western mountains
unanimous in predicting warming across recharge occurs in place in mountain is relatively more surficial runoff. Declines
the United States, with most ranging from recharge areas, sustaining important sur- in mountain recharge triggered by loss of
about +3º to +6ºC. Projected precipitation face water base flows and cooler water snowpack would have immediate impacts
changes are more complicated, with some temperatures. Recharge also occurs by on mountain water resources, including
consensus developing toward modestly infiltration where runoff crosses from low flows and stream temperatures, and
wetter conditions in the northern tier of mountain blocks onto abutting fans and may also have serious impacts on long-
states and modestly drier conditions in the alluvium or onto basin floors. term ground water supplies in surrounding
southwesternmost states, but with relative- A growing tide of studies is showing basins through reductions in mountain
ly small changes of uncertain sign in the that snowmelt provides more mountain recharge. Although recharge that supplies
regions between these projections (conclu- recharge than rain does in western settings mountain ground water may decline, much
sions by the authors from analyses of pro- (Earman et al. 2006). This is probably of this unrecharged water may run off onto
jections used in IPCC 2007). Based on because large and intense infiltration is the region’s fans and basins and potentially
these projections, the impacts of warm- required to break through the region’s thick increase recharge on fans and basin floors.
ing—the more confident of the projec- unsaturated zones. Additionally, snow- However, if the unrecharged water is
tions—on water supplies, including ground packs store and then release precipitation instead mostly evapotranspired from the
water, are urgently in need of assessment. from several storms at once. Studies in the mountain soils, the overall recharge
Warming will likely reduce runoff genera- Great Basin indicate that 50% to 90% of (mountain plus basin) may decline.
tion, whether precipitation increases or the recharge there originates from At present, whether the overall recharge
decreases, because of attendant increases snowmelt (Winograd et al. 1998). will increase, decrease, or stay the same is
in the potential for evapotranspiration. Recharge-temperature analyses based unknown at any scale in the West.
Due to the parallels between such runoff on dissolved gases show that recharge in Similarly, the impacts to ground water sup-
declines and historical droughts, it has been the central part of the Chiricahua plies due to changed conditions of the loca-
natural to envision ground water as a safe- Mountains, Arizona, is derived only from tion and timing of recharge are also insuf-
guard against climate change. However, altitudes above seasonal snowlines and not ficiently understood.
continued on page 5
4
Cayan, D.R., Kammerdiener, S., Dettinger, Intergovernmental Panel on Climate
M.D., Caprio, J.M., and Peterson, D.H. Change. 2007. Climate change 2007,
Western Ground Water
and Climate Change 2001. Changes in the onset of spring in The physical science basis. Summary
the western United States. Bulletin, for Policymakers: IPCC Secretariat,
continued from page 4 American Meteorological Society, 82, http://www.ipcc.ch, 21 p.
Given ground water’s crucial role, the 399-415. Knowles, N., Dettinger, M., and Cayan, D.
potential impacts of warming on recharge Changnon, S. 1987. Detecting drought con- 2006. Trends in snowfall versus rain-
deserve more attention than it has received ditions in Illinois. Illinois State Water fall for the western United States.
to date. It is possible that ground water sup- Survey, Circular 169, 36 p. Journal of Climate 19(18), 4545-4559.
plies will fare well, overall, in a warming Dickinson, J.E., Hanson, R.T., Ferre, Liu, F., Williams, M.W., and Caine, N.
world, but they may also fare poorly. The T.P.A., and Leake, S.A. 2004. Inferring 2004. Source waters and flow paths in
projected climate changes are unprecedent- time-varying recharge from inverse an alpine catchment, Colorado Front
ed in the modern era, and we lack the tools analysis of long-term water levels. Range, United States. Water Resour.
and data (e.g., long-term continuous moni- Water Resources Research 40. Res. 40, 10.1029/2004WR003076.
toring of recharge processes) to confident- doi:10.1029/2003WR002650, 15 p. Maurer, E.P., Wood, A.W., Adam, J.C.,
ly detect or predict ground water responses Earman, S., Campbell A.R., Phillips F.M., Lettenmaier, D.P., and Nijssen, B.
to climate. and Newman B.D. 2006. Isotopic 2002. A long-term hydrologically-
In a world with a changing climate, our exchange between snow and atmos- based data set of land surface fluxes
intuition and ground water science’s tradi- pheric water vapor: Estimation of the and states for the conterminous United
tional focus on long-term average condi- snowmelt component of groundwater States. Journal of Climate 15(22),
tions may fail to recognize, predict, and recharge in the southwestern United 3237-3251.
alert us to significant changes in ground States. Journal of Geopysical Research Mote, P.W. 2003. Trends in snow water
water supplies. Monitoring commitments 111 (D9), http://dx.doi.org/10.1029/ equivalent in the Pacific Northwest
like the USGS’s national-scale Ground- 2005JD006470. and their climatic causes. Geophysical
Water Climate-Response Network are a Earman, S., and Phillips, F.M. 2003. Research Letters 30. DOI 10.1029/
start (http://groundwaterwatch.usgs.gov/), Groundwater recharge and movement 2003GL0172588.
but that network’s focus on ground water in the central Chiricahua Mountains, Winograd, I.J., Riggs, A.C., and Coplen,
level variations from days to years needs to Arizona. Geological Society of T.B. 1998. The relative contributions
be extended beyond ground water level America Annual Meeting Abstracts of summer and cool-season precipita-
fluctuations to the full range of ground with Program, abstract 201-15. tion to groundwater recharge, Spring
water issues and processes on multi- Mountains, Nevada. Hydrogeology
decadal time scales. Otherwise, changes Journal 6, 77-93.
may come from unexpected directions to
ground water managers and users.
Ground Water, Climate,
and River Basins
continued from page 3
so that it includes the role of ground water
resources.
Acknowledgements
Research was funded by the National
Oceanographic and Atmospheric
Administration, NOAA-GAPP Program
(NA04OAR4310085) and the National
Science Foundation (NSF ER0326064).
References
Duffy et al. 2002. Ground water modeling
for wellhead protection. In-house report
to sponsor. Pennsylvania Dept. of
Environmental Protection. 10p.
Qu, Y., and C.J. Duffy. 2007. A semi-dis-
crete finite-volume formulation for
multi-process watershed simulation.
Water Resources Research doi:10.
1029/2006 WR005752, in press.
(accepted 9 May 2007)
5
Figure 2b. The eigenspectrum for one observation well plotted versus frequency in cycles/7 days.