Hydrologic Change: A formal line of scientific inquiry

1
Hydrologic Change A formalline of scientific
inquiry

• Upmanu Lall
• Columbia University

2
Hydromorphology Evolution of Hydrologic
Systems at the Scale of Years to Centuries

• Chris Duffy, Upmanu Lall, Dennis Lettenmaier,
• Jose Salas, Jery Stedinger, Richard Vogel

System Natural Human Infrastructure ?human
watershed dynamics
3
Sources of Hydrologic Change

• Climate Variability
• Climate Change
• Human Modifications in the River basin
• Dams and diversions
• Aquifer Pumping and Compaction
• Land Use changes
• Drainage of wetlands
• Consumption
• Channel Modifications
• Feedbacks to Regional Climate

4
Main points

• Hydrologic change may be more important
  /interesting than climate change both as a line
  of scientific inquiry and as an avenue for
  societal response
• Need Detection, Attribution and Prediction
• Local hydrologic changes are often documented
  with some attribution
• However, no systematic efforts have evolved to
  document and understand hydrologic change or
  evolution at regional to global scales.
• Prediction is consequently not possible. Requires
  the development of an understanding of long term
  system evolution
• There continues to be a tendency to view human
  effects as exogenous and very limited work has
  gone towards scientifically addressing the human
  component of change
• Climate variability and change need to be
  understood as a bi-directional process linked to
  human activity mediated by the hydrologic cycle
  hydrologic change ? climate change impacts

5
Machu Picchu's agricultural terraces play a key
role in stabilizing slopes and controlling
erosion. A quarry, upper right, provided granite
for construction.
6
Satellite Image of Tamil Nadu State in India
showing Water Tank. These tanks date back to 150
BC to 200 AD
Human modification of the landscape - small
scale storage/harvesting vs large dams
7
WATER RESOURCES RESEARCH, VOL. 35, NO. 4, PAGES
1305-1311, APRIL 1999Dam nation A geographic
census of American damsand their large-scale
hydrologic impactsWilliam L. Graf
75,000 dams in the continental United States Av.
Storage mean annual flow West dams gt 3 years
runoff N. East and N. West 25 of ann. flow.
Area per dam varies from 44 km2 (17 miles2) per
dam in New England to 811 km2 (313 miles2) per
dam in the Lower Colorado basin. Storage
volumes, range from 26,200 m3/km2 in the Great
Basin to 345,000 m3/km2 in the South Atlantic
The nations dams store 5000 m3 (4 acre-feet)
of water per person.
.., general statements about the fragmentation
of rivers and the large-scale hydrologic
consequences of dams at a national scale are not
available.
8
Smaller storage ? daily to seasonal storage Large
Storage ? interannual storage How does dam
storage relate to frequency modulation of
downstream flows? Are certain frequencies
amplified (relatively) leading to higher
vulnerability/sensitivity and impact
downstream? What are the cumulative effects of
dams, in series and in parallel, on flows,
ecology and biogeochemistry?
9
Constanz, 2003
Some structural or conceptual representations of
what is impacted exist, but we have no
documentation of the actual cumulative or
collective impacts or how to predict them
10
Vitousek, 1997
11
The scale of hydrologic modification by humans
has continued to increase exponentially,
consistent with population trends. A
sustainability perspective requires human
dimensions to be considered as an integral part
of hydrology particularly at long time scales

• At least seven major continental rivers no
  longer reach the sea for at least parts of the
  year -- the Ganges, Indus, Amu Darya, Syr Darya,
  and Chao Phraya (Asia), the Nile (Africa), and
  the Colorado (North America). Associated changes
  in land cover (due to irrigation, and changes in
  streamside vegetation resulting from changes in
  discharge regimes downstream of dams) have
  doubtless had large effects, at least locally, on
  evapotranspiration. Yet the effects of these
  activities on global and regional water balances
  are essentially unknown all coupled
  land-atmosphere-ocean models represent, at best,
  a natural land surface hydrologic condition,
  perhaps with vegetation prescribed to current
  conditions.

12
Vorosmarty et al 2000
13
Vorosmarty et al 2000
14
Vorosmarty et al 2000
15

• Some examples

16
Consider an urbanizing watershed Aberjona River
near Winchester, MA
17
Hydromorphology attempts to improve our
understanding of the impact of anthropogenic
influences, in this case, on streamflows in an
urbanizing watershed. Note the apparent increase
in floods. Is this due to urbanization or due to
climate change?
18
Decadal Daily Flow Duration Curves 1940-1999.
Note how nearly all streamflows have increased
over time, particularly the lower streamflows
due to increases in impervious areas resulting in
reductions in evapotranspiration. It is not
only floods which are influenced by urbanization.

19
From J. Salas
Decadal Variability Overwhelms secular trend
20
Model Streamflow Variability e.g. persistence
(Salas)
Persistence in annual streamflows arises from two
main sources          Land surface and
subsurface storages -   lakes, wetlands,
glaciers, etc. -   soil and groundwater
         Oceanic and atmospheric processes
-   abrupt (sudden) changes and trends in
precipitation (e.g.) - periodic
effects on precipitation
-   pseudo-periodic effects on precipitation
Example of Natural System Dominating Mindset
21
A Perspective on Hydrologic Change in the
Columbia River Basin

• David A. Jay
• OGI School of Science and EngineeringOregon
  Health Science University, Portland, OR

22
Changes in CR Mainstem Hydrology

• Causative Factors --
• Climate cycles/change
• Human intervention in hydrologic cycle
• flow regulation andflood control (FCRPS)
• irrigation depletion
• tributary dams
• Loss of flood plain/marshes
• Increased fine sediment input to tributaries
• Sediment trapping by dams

Observed Columbia River flow at The Dalles
23
Impacts Important to Salmonids --

• Freshets are earlier, smaller and last longer
• Overbank flow and flood-plain processes altered
• Estuarine habitats displaced and altered
• Flow regulation constrains habitat restoration
• Plume is smaller, timing relative to upwelling
  has changed
• Less sediment delivered to estuary and plume

Human-induced reduction in Columbia River flow
at The Dalles
24
(No Transcript)
25
(No Transcript)
26
(No Transcript)
27
(No Transcript)
28
(No Transcript)
29
(No Transcript)
30
(No Transcript)
31

• Changing Climate and Floods

32
(No Transcript)
33
(No Transcript)
34
Non stationary Flood Series
35
Probability distribution of the number of
anomalous exceedances of the flood series based
on a quantile threshold. (a) lt10, (b) lt33, (c)
gt67, (d) gt90. Quantiles are computed using a
30-year time window, and exceedances of each
quantile are computed for the next 30 years on
record.
36
Wavelet Analysis of 1000 year sample of annual
maximum NINO3 from a 110,000 year integration of
the Cane-Zebiak Model with stationary forcing (
Clement and Cane, 1999)
37
Probability distribution of the number of
anomalous exceedances of the 90th percentile of
the ZC model NINO3 series, for two successive n
year periods using block or random sampling,
where n is (a) 50 years, (b) 200 years. - based
on 1,000 years of control run data
38

• Groundwater system response and modifications

39
Alley et al 2002
40
Alley et al 2002
41
(No Transcript)
42
(No Transcript)
43

• Climate Projections Forward and Backwards

44
Allen and Ingram, 2002
45
Wehner 2004
46
(No Transcript)
47
Fye et al 2003
Reconstructions assessing climate variability
48
How does hydrology influence climate variability
and change?
49
Marshall et al, 2001
50
(No Transcript)
51
(No Transcript)
52
kl
Questions
Is the hydrological cycle changing? If so, is
it accelerating or decelerating?
53
le
Conventional Wisdom Acceleration
Model predicts
With climatic warming comes more oceanic
evaporation and greater lower tropospheric water
vapor concentrations.
It has been proposed that the upper troposphere
will dry, and provide a negative feedback to
warming. But models predict increased upper
tropospheric water vapor.
Aerosols may slow the warming by reflecting
sunlight, but they will likely decline in
concen-trations while CO2 continues to rise.
Rind, D. 1998. Just add water vapor. Science 281
1152-1153.
54
el
Our Most Reliable Data?
55
le
Petersons Pan Paradox Evidence for Change
Evaporation rates from open pans of water across
the Northern Hemisphere have generally steadily
decreased over the past 50 years
due to a weakening hydrological cycle?
Peterson, T.C., V.S. Golubev, and P. Y. Groisman.
1995. Evaporation losing its strength. Nature
377687-688.
56
eb
what do the models say??
57
bb
A Major Wild Card
Human-made aerosols increase scattering and
absorption of solar radiation, and produce bright
clouds that dont rain as easily.
Anthropogenic sources contribute almost as much
as natural sources to global aerosol optical
depth.
Ramanathan, V. 2001. Aerosols, climate, and the
hydrological cycle. Science 294 2119-2120.
58
bb
Radiative Forcing
Aerosols overpower greenhouse forcing.
By reducing irradiance, aerosols reduce
evaporation. Precipitation rates shift to match
lower evaporation rates.
The hydrological cycle appears to be weakening
(spinning down).
Ramanathan, V. 2001. Aerosols, climate, and the
hydrological cycle. Science 294 2119-2120.
59
bk
The Cause of Decreased Pan Evaporation

• Pan evaporation has decreased globally (in both
  wet and dry environments), but water vapor
  deficit has remained nearly constant, despite
  temperature increases. How?
• Decreased evaporation is caused by decreased
  irradiance, at the scale of a single pan, as well
  as at the global scale.
• Decreased irradiance is caused by increased
  aerosol concentration and cloud cover.
• Additional evidence for decreasing irradiance
  reduction of diurnal temperature range (DTR)
  coinciding with reduced pan evaporation.
• The power of aerosols DTR increased markedly
  over the U.S. from September 11-14, 2001.

Roderick, M.L. and G.D. Farquhar. 2002. The cause
of decreased pan evaporation over the past 50
years. Science 298 1410-1411.
60
kk
Summary

• The conventional wisdom that evaporation trend
  would be determined by temperature is wrong (and
  this has been known since 1963 !).
• We need to determine trends in actual
  evaporation--pans aren't the whole story, and
  land and ocean evaporation trends must be
  separated.
• A trend toward increasing global cloudiness
  reversed in 1990, which may cast some doubt on
  the indirect forcings caused by aerosols proposed
  by Ramanathan, which might mean reversals both in
  the trend toward decreased evaporation and in the
  negative aerosol forcing which has been proposed
  to counter greenhouse warming.

Ohmura, A. and M. Wild. 2002. Is the hydrological
cycle accelerating? Science 298 1345-1346.
61
Peterson et al, 2002
62
Observations

• Water related uncertainties may be a major factor
  in predicting future climate. The extent of human
  influence in this regard is unknown
• Climate variations and change confound
  attribution of human changes to water systems. A
  framework for understanding the role of
  variations could be developed under retrospective
  analyses and reconstructions
• There is currently no framework (data or theory
  that links climate, societal infrastructure and
  hydrosystem evolution or assessment)