Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter.
Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 214
~ -
Scientific Issues of Data Collection,
Distribution, and Analysis
Hydrologic processes are highly variable in space and time, and this
variability exists at all scales, from centimeters to continental scales, from
minutes to years. Data collection over such a range of scales is difficult and
expensive, and so hydrologic models usually conceptualize processes based
on simple, often homogeneous, approximations of nature. Hence a 2,000-
km2 river basin is commonly modeled as a lumped system that responds as
a point with average representative properties. Ground water flow is commonly
treated as one-dimensional or two~imensional. Rainfall is expressed as a
mean over large areas, and as depths over periods of a day. Snowmelt
runoff volumes are forecast from averages of snow accumulation at a few
index plots. These generalized conceptualizations reflect the normal dearth
of data, which lack the temporal and spatial resolution to support more
detailed modeling. This forced oversimplification is impeding both scientific
understanding and management of resources.
In the history of the hydrologic sciences as in other sciences, most of the
significant advances have resulted from new measurements, yet to-
day there is a schism between data collectors and analysts. The pio-
neers of modern hydrology were active observers and measurers, yet
now designing and executing data collection programs, as distinct
from experiments carried out in a field setting with a specific research
question in mind, are too often viewed as mundane or routine. It is
therefore difficult for agencies and individuals to be doggedly persis-
tent about the continuity of high-quality hydrologic data sets. In the
excitement about glamorous scientific and social issues, the scientific
community tends to allow data collection programs to erode.
214
OCR for page 215
DATA COLLECTION, DISTRIBUTION, AND ANALYSIS
215
Such programs provide the basis for understanding hydrologic systems
and document changes in the regional and global environments.
Modeling and data collection are not independent processes. Ide-
ally, each drives and directs the other. Better models illuminate
the type and quantity of data that are required to test hypotheses.
Better data, in turn, permit the development of better and more
complete models and new hypotheses. If we accept this synergism,
the hydrologic sciences will be well situated for progress, which is
needed because recent developments in spatial and temporal models
and new data acquisition technology require a rethinking of many of
the traditional hydrologic problems. We must, however, reemphasize
the value and importance of observational and experimental skills.
To address many of the issues described in Chapter 3, we need
new observations of hydrologic phenomena. Some of the current
uncertainties in our knowledge of the hydrologic cycle require better
understanding of hydrologic processes, but progress in the hydrologic
sciences will also depend on improved methods for collecting hydrologic
data, more complete and better-organized archives of already-available
information, and better mechanisms for distribution and exchange of
data, particularly in developing countries and in the international
arena. This chapter describes some requirements for and characteristics
of hydrologic data, assesses the current hydrologic data base, and
then discusses some opportunities to improve hydrologic data and
their use.
NEED FOR COLLECTION OF HYDROLOGIC
DATA AND SAMPLES
Hydrologic data are needed to measure fluxes and reser-
voirs in the hydrologic cycle and to monitor hydrologic
change over a variety of temporal and spatial scales.
Historically most hydrologic data have been collected] to
answer water resources questions rather than scientific
ones.
Hydrologists use information obtained in laboratories, such as soil
particle size, solute concentrations, or electromagnetic spectra, but
OCR for page 216
216
OPPORTUNITIES IN THE HYDROLOGIC SCIENCES
:~:~i~y~nR~o~Lo~G~1c~l~MpL~l~cAT~l~oNs~ OF:~AL~W AlC~M~I~r~Qi~:~:~:~:~
:~ : ~
:~:~:~: ~~:~::::
~ ~ ~ ~ ~ ~~ ~~ ~~ ~~ 0
:: ~ ~ : ~ ~ ~:~:~ :: ~:~ :~ : ~ ~ :~: :::: : ~~ ~ ~~::~ :~ ~ ~ ~::~: ::: ~ :~: ~ ~ F~ : T:~
:~:~c u~rlng~t 1~e:~spr~l~n~g:~:~montus~ot;;~Ap~r~i~ A l~a~gh~::~J~u~:n~e:.~:~:~The~ ~:net~:~res~u~l~t::~ot~ An
- :~:~::4::~::::: ~::~ :::::: :::: ::: :::~:::::~t: ~ ~ ~~ ~~ ~~ ~~ ~ ~~ ~~ ~ ~ ~~.
~n~£reas~e~ ~:~1~n~g~ nodal If::: tem~p~e~r:a:tu~re~wo~u~ tic ~:~:~be~a~ ~~d~ec~re~a~se~:~:~i~n~ stream flow ~:~d~u~r:~i~n~g~
:~ :: ~: :~:~ :~: ~: ~ ~ ~ ~ ~: ~ ~:~: ::~: :~ ~ ~ ~:::~:~::~::~: :~ ~ ~ ~:: ~: ::: ::~: ~: ~ ~ ~ ~ :~:~ :~:~ ::~:~ ~ ~:: ~ ~r:~ ~: ~:: ~-~ ~ :~ ~ ~: ~:~ ~: ~:~:~:: ~ ~ ~
~ : : ~ ~ ~ ~ ~ : ~ t n e ~ ~ ~ ~ n o ~ r m a ~ : l ~ ~ ~ ~ : ~ s ~ n ~ o w m e I ~ ~ t ~ r u ~ ~ n ~ r ~ ~ : ~ ~ ~ o ~ n ~ t h ~ s ~ ~ ~ ~ ~ o r ~ ~ ~ n e : ~ ~ ~ ~ ~ s p ~ r ! ~ ~ ~ n ~ g ~ ~ ~ ~ ~ a ~ ~ n ~ d ~ ~ ~ s u ~ m ~ m e r . : ~ ~ : ~ ~ T h ~ ~ ; ~ s : : ~ ~ h ~ a n ~ g e ~ ~ ~
· ~: ~::~: :~:~::: ~ ::::~:~:::: ~ ~ :~:~ ~ :~.~:~::: ~ ~:~:~.~ :~: ~ ~:~ :~:~:~ ~:~ ~ ~ ~
n the: rn:ont~h~ly~ :t~m~l~n~g~ o~f~ ~av~a ~ I~ab~le ~streamflow: ~co~u~ld~ ~ d:ra~m~ati~c:a~l l~y ~ ~n-c:re~a:se: :~:~ i ~:::
~I~:~w~i~nter~:~:f:loodi~:ng~la~n:d~ ~:si~gn~ifica~n~tly~decFeas~e:~lthel~l~w~ater~:~a~va~i:labl~e:l~d~u~Fing: :~the~::
c~rit~ical~A~pr~i~l~-th~ug~h-Se~ptem~r~season~ of ~soi~l-m~o;~s~t:ure~repl:e:n~is~h:me~:nt~via
. . ~ ~ ~ ~ ~ ~ ~ ~ :. ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
~ ~'rr~gat~Qn~.~T~ ~e~:~c~ 1~an~g~:ng~:~:rea~mfl~ow:~:~pat~r~n~$~:~wo~u~1d~affect~existing~1su~£e~:~l:ll~
~ ~ , ~ ~ ~ ~ ~ ~, ~ ~ ~ ~ ~ ~: ~ ~ ~
~ ~ : wam~r~:~:~se:rvo~' rs ~w lose~storage~ ~ca~,Dac~ities~ :~were: o~e~sl~g:n~ec :: to~: acco~mmo~c ~am:
.:::~ ~ ~ : : ~ ::~: : :: :: ~ :: : ~ ::
~:~: ~: ,~l~storic::~ ~yc ~ro og~'c:: runott~::~pa~erns~. :: :T ~e~:~li~m~paci: :of:~:g~lo:bal:::~c:~t~i~mat:ic~ c~h::a~nge~::~:
~ : ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
~ ::r r^~l:lr-~ ~~ t~t~` ~IS potent~~a ~ y :~ve~ ~se~riou~s~.: ~: T 1e~ costs of ~ tak~i~n~g:::: ~ ~
~~ ~: ~ ~ ~ ~ ~ ~ ::~: ~ ~: ~ ~ ,~ ~: ~:: ~ ~ ~ ~ ~ :~:~ ~ ~- ~ ~:: :~: ~: ~ ~ ~ ~ ~
I I l: I :I::::I~ I l~ lU I I t1 1~L ~ ~. . . 1 l l lO ~ ~ ~ Y ~ ~ e
~: ~ : ~ :: ~ ::: ~ ::: :: :: ~: ::
:: :: ::::~:::: : ~ :::: : : ~ : :
} : ~
on~: wa=~e:~ reso~u:rce:s~ sy~s e~ms::
: :::: :::: ~: ::::::::: ~:~:~::::: ~:::::: ::: :: :~:::: ::~ :::~ :~ ::: : ::::::~: :: :~:: : ::~:~:: :: ::: ~ ~ :::: :: .
~co~u~nterm~e~as~:w:res~ ~::~a~n~d~ rep:a~i~ri~ng ~d~amagel: co~u:ldl~:h~ave~a ~ ~si~g:n:iti~can~t ::e~ct :~on :~:~
:: ::::::: : ~:: : :~: ~ ::: : : :~: :::: ~ ::::: : : ~
: th:e~::~e~con:om~ies~:~:of individual::: s~te~s::~a~n~d~ofl~:t;~h~e ~l3:n~:ited~::~::~Stat:es~i~ ~ ::~ ~: ~ ~:::~:: ~ ~:~:: ~:~ ~: ~::~::;;~
. ~ ~ . , ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ,J, :: ~ ~ ~ . ~ ~, .
: ~ ~ ~:~ 1e ~ key:: ~qu~est~lQn:~: Tor 1~yc £0o~g~:~sts~ ~- - -
. . ~ ~ . . . . ~
~ :St eps ~ ~ ~sn~o~:~lo ::~De ~ ta~:~:en~ ~m:~: t~n~e~ t~a~c~e~ ~::~of:~:
. . ~ ~
:~ ~
~ ~ :
~now ~ ~ecotm~es:~ ~ W:ha:t~ ~i:m~med~i~ate~
~s~u~c:h~ ~:1 on~g-te~rml~::~u n:certai~n~ty:?~:~ Cer-~ ~ ~:~
. , ~ . . ~ ~ ~ ~ ~
;: ta~n~l~y ~one: :~a~sw:er::: ~I~ICS: ~n~: :a~:~co~m~m~:l:t~ment:~ ~:to ::~c~ontin~u:~ ~:a~nd to: ~ :::ex~p~a:nd ~:
~progr:ams::of~ca:refu~l:::~s~c::ie~n~tific:l~m~ea~s~u~re~m~en~ts~.l~:~T~h~e:~;~:~:~re:su:llti~ng~l:~lh~ydro:l~:ogic~:
:: ::: ::: :~ : :: :::::: :: :::: ~:::: ~ ::: :~ ~ : : ~: :: ~ ~ ~ :
~ ~:~:~ d~ata~l:ba~se$~:~:: wi~l~l~:~en~ab~le~: :hydrol~:ogi~sts~l~:to~stu~dy::~ :and~l~p~red~:~i~:ct~th~e ~ effects~:~of~:~
,~ ~ , , ~ ~ .~: :~ ~ ~ ~, ~ , ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
:::giocal::warm:lng~ anu~ m~a~e :a~pp~:ropriate recommend~at:ion~s::~r~ soc~iaJ p~o:;t:i:cy ::~
: :: ~: ~:~: : : :::: ~ ~::: ::::~:: :::: ~ : ::: ::: ~: ~ :
:: :::
~: ~ ~:~
::::
::
::::
:::
::::
:: ~
::
:::::: ~
OCR for page 217
DATA COLLECTION, DISTRIBUTION, AND ANALYSIS
217
most of their data must be obtained under field conditions. The
reason is that, in addition to the elucidation of microscale processes,
hydrologists are concerned with processes that have meaning only at
the field scale or over long time scales, such as runoff and sediment
yield from drainage basins or continental-scale drought. These re-
quirements make hydrologic data collection complicated and expen-
sive. Despite large financial investments, there remain important
questions about hydrologic fluxes and reservoirs that are unlikely to
be answered by the incremental growth of instrument networks. Technical
and analytical innovations are necessary to overcome the paucity of
useful hydrologic data now being collected and collated.
To better characterize the hydrologic cycle requires data in several
categories, and the choice of what to measure and where and when
to measure influences what hydrologic questions we can investigate.
1. We require information about the fluxes and storage of water in
its various phases as it moves through the components of the hydro-
logic cycle. These include precipitation, snow accumulation and ab-
lation, glacier flow and mass balance, discharge in rivers and streams,
movement of ground water, and evapotranspiration. Also needed is
information about the transport of solutes and sediments as well as
the fluxes of energy that drive the hydrologic processes.
2. Hydrologic data are needed to monitor change, or lack of change,
in the quantity and quality of water. The major effect of climatic variability
on the humans, plants, and animals that inhabit the earth is felt through
the hydrology. Similarly, changes in water chemistry cause concern
among users of a water resource and can dramatically affect the fish
and other biota that live in lakes and streams (Figure 4.1~. Thus we
need baseline data, especially in tropical and semiarid areas.
3. In the traditional scientific sense, hydrologic data are needed to
test hypotheses and models, and for exploration, to formulate new
hypotheses. Hydrologic science can advance as a discipline only if
measurement and theory evolve together. Sometimes the mechanisms
that govern a complicated hydrologic process are known so poorly
that precise data are needed simply to explore the phenomenon; then
the next generation of measurements awaits conceptual developments
that show which data are essential for testing ideas about how hydrologic
phenomena occur. We know only what we measure, and we know
what to measure only after some unifying conceptualization of the
existing data has pointed the way.
Finally, the measurement of hydrologic variables is a scientific
endeavor itself. Future progress in hydrologic data collection should
result from:
OCR for page 218
218
o
c 3
4>
o
-
I
80
-
~ 40
. _
OPPORTUNITIES IN THE HYDROLOGIC SCIENCES
~ 25
._ 10
, c~~~..o....~.~- ~cl ~
0 ~ - 5
80 82 84 86 88 ~
6
Yea r
50
40 ~ ~ ·_
~ ~,.,~ e Go Lo,.,,. ,, o l
30
80 82 84 86 88
Yea r
A:
^~?
0' . . . _
80 82 84 86 88
Yes r
30
._
~ 25
-
0
E
25
o
co
to' ~ ~ ~ ~
D
80 82 84 86 88
Yea r
..'~'...f~N if
o'/~'
. .
80 82 84 86 88
Yea r
120
R·~
0 - ~~
_ 95
cn ~.,,.aC, 4,. o,,.1
._
~0
ID
A
_
70 . . .
80 82 84 86 88
Yes r
60 .
50 "~i
40
80 82 84 86 88
Yea r
15 . . . .
80 82 84 86 88
Y e a r
-
O 25
-
In.,,. D . ~ ~ ~ ~—o
_
80 82 84 86 88
Yea r
FIGURE 4.1 Changes in ionic concentrations in two streams in Shenandoah National
Park, Virginia, 1980 to 1988. Units are microequivalents per liter for concentrations
and centimeters per year for discharge. SOURCE: Reprinted, by permission, from
Ryan et al. (1989). Copyright @) 1989 by the American Geophysical Union.
· coordinated experiments where diverse efforts are pooled;
· technological advances in such fields as remote sensing, instru-
mentation, and information systems;
· new forms of analysis such as isotope geochemistry, paleohydrology,
and improved models of spatial and temporal processes; and
· intensified efforts in design of monitoring networks and exami
nation of data quality and compatibility.
Need to Collect Data at Varying Scales
Hycirologic processes operate over a range of temporal
and spatial scales, and important questions exist at time
scales from seconds to millennia and space scales from
the molecular to the global.
OCR for page 219
DATA COLLECTION, DISTRIBUTION, AND ANALYSIS
219
Hydrologic processes operate over a continuum of space and time
scales, from those of laboratory experiments to global transport of
water and nutrients and from short-lived, transient phenomena to
gradual secular variations. Some important questions studied in the
laboratory or at scales of a field plot involve interactions between
solutes and water or between water vapor, liquid water, and ice. For
example:
· The rate of elusion of chemical impurities from seasonal snow
depends on interactions between solid, liquid, and vapor phases.
· The cycling of hydrogen ions plays a critical role in determining
the effects of acidic deposition on wild land and agricultural ecosys-
tems. The largest components of hydrogen ion cycling are consumption
by mineral weathering and production by plant roots. These components
are difficult to estimate at field scales because we do not know enough
about the kinetics of mineral dissolution reactions and biological release
processes at these scales. Errors in estimating annual hydrogen ion
consumption and production rates can be as large as the estimated
annual input rate from acidic deposition.
At the same time, our current knowledge of major fluxes of water
in the hydrologic cycle involves large uncertainty. For example:
· The mean annual discharge of the Amazon River is about 200,000
m3/s. Typical error estimates for the measurements are 8 to 12 per-
cent, i.e., 16,000 to 24,000 m3/s, a rate slightly higher than the mean
annual discharge of the Mississippi River.
· Data show that sea level is rising slightly, but our investigations
into the source of this rise, and the accuracies of our predictions of
the future, are hampered because our measurements of the snowfall
and iceberg calving from the Antarctic ice sheets do not tell us definitely
whether the Antarctic ice volume is growing or shrinking. Thus the
proportion of the water attributed to each of the sources causing this
rise in sea level is not confirmed. How much comes from Antarctica
and Greenland, from thermal expansion of the ocean waters, from
shrinking alpine glaciers, and from depleted ground water reservoirs?
Data are needed at a variety of scales, and the spatial and tempo-
ral scales of available data restrict the questions that can be investigated.
As is described in detail later in this chapter, the information is better
for some hydrologic fluxes and reservoirs than for others. For most
fluxes, however, a fundamental block to progress is our poor knowl-
edge of how to interpolate between measurement points and how to
extrapolate from few data points. For example:
OCR for page 220
220
OPPORTUNITIES IN THE HYDROLOGIC SCIENCES
· Depths and water equivalences of a snowpack are measured at
many snow courses in cold regions, but it is possible to use these
data only as crude estimates of the water content of a regional or
basin-wide snowpack.
· Topographic influences on rainfall, evaporation, and soil moisture
are poorly documented at scales varying from individual hillslopes
to entire mountain ranges.
An additional important issue in the sampling of hydrologic pro-
cesses is the structure of the statistical fluctuation that the processes
have at different scales of measurement. How do the mean and variance
of annual rainfall change as a function of the area over which the
estimation takes place? How do the mean and variance of evapo-
transpiration depend on the time scale considered? What is the com-
bined effect of time and space scales on the statistical properties of
hydrologic variables?
There is an urgent need to
1. quantitatively characterize the fluctuations of hydrologic vari-
ables at different time and space scales; and
2. design data collection programs that will allow the study of
theoretical constructs, described in Chapter 3, to structurally link the
fluctuations at different scales.
Need to Develop Accurate Hydrologic Data Bases to
Improve Scientific Understanding
Detection of hydrologic change requires a committed,
international, long-term effort and requires also that the
data meet rigorous standards for accuracy.
Synergism between models and data is necessary to de-
sign effective data collection efforts to answer scientific
questions.
Development of scientific theory in the absence of supporting facts
does not lead to understanding and can result only in conjecture.
The primary sources of facts for the hydrologic sciences are the mea-
OCR for page 221
DATA COLLECTION, DISTRIBUTION, AND ANALYSIS
221
surements that are made of hydrologic and ancillary variables. Most
hydrologic data are collected by government agencies for a variety of
purposes only one of which is the development of hydrologic un-
derstanding. Historically, the major providers of funding for hydro-
logic data collection have expected that the resulting data would be
useful in setting water policies, developing water resources plans,
designing water resources systems, operating the structures that make
up such systems, and monitoring the management of water resources.
Increased hydrologic understanding can and does contribute to im-
proved information for these utilitarian purposes, but the design of
hydrologic data networks seldom has as a primary objective the bet-
terment of basic hydrologic understanding. Therefore, the data needs
of the hydrologic scientist almost certainly will not be fully satisfied
by the existing data networks that are supported primarily for operational
and accounting purposes.
The existing data networks should be viewed by hydrologic scien-
tists as opportunities upon which they can build. To optimize these
opportunities, it is first necessary to define the characteristics of the
data sets that hydrologic scientists need. These characteristics include
the variables to be measured and the locations, frequencies, durations,
and accuracies of the measurements. They should be derived from
knowledge about the hydrologic phenomena to be explored and from
the hypotheses to be tested. Allocation of the resources available for
data collection must seek complementarily between the scientific and
operational data sets. However, the operational networks often change
in character because of changing operational demands for data or
because of budgetary pressures on the financial sponsors of the data
networks. These changes most often are manifested as discontinuities
in the time series of measurements, as shifts in the location of the
measuring site, or as changes in the accuracy of the data. Thus a full
measure of complementarily is an illusive objective but a worthy one
that can be approached by adequate communication between research
scientists and managers of data collection programs.
Important hydrologic changes may be subtle or may be difficult to
detect because of large interannual or inter-event variation, and spa-
tial and temporal scales of available data restrict the questions that
can be investigated. Some important processes are transient short-
lived but repeated. Fluxes and reservoirs of water, energy, solutes,
and sediment are monitored most intensively over those parts of the
world that are humid-temperate, densely populated, and industrialized,
but measurement networks are particularly sparse over the oceans
and in regions that are subhumid, tropical, at high elevation, or lightly
populated.
OCR for page 222
222
OPPORTUNITIES IN THE HYDROLOGIC SCIENCES
Need to Collect Long-Term Hydrologic Data
Long-term monitoring and the use of paleohydrologic records
are fundamental to understanding the role of extreme events
in hydrologic systems.
The need for long-term measurements is becoming clearer in our
investigations of environmental change, including hydrologic change.
Some disciplines, such as paleontology and historical geology, have
depended for their existence on the availability of data spanning long
periods, from 100 million to 2 billion years. Other disciplines that
have traditionally focused research over shorter time frames, such as
the environmental sciences, now stress the critical importance of long-
term records.
Despite the increasingly recognized importance of data records of
long duration, only a few dedicated research organizations have suc-
cessfully maintained high-quality data collection efforts over periods
of 50 to 200 years. Furthermore, these organizations have experienced
difficulty in committing limited research monies year after year to an
activity that is frequently termed "monitoring," often with pejorative
overtones.
Dams have been built in many areas of the world and the water
has been allocated for power generation or irrigation based on only a
few years of data, with the too frequent result that the anticipated
volumes of water have been available only in years of above-normal
runoff. But many scientific questions justify the collection of long-
term data. Two general areas for which long-term hydrologic data
are specifically needed are discussed briefly in the remainder of this
section, but these examples are not meant to be exclusive or exhaus-
tive.
Understanding Hydrologic Behavior and Hydrologic Change
Long-term data are required to understand the basic hydrologic
behavior of natural landscape units. In most humid areas, we do not
understand well enough the relationships between rainfall, evapo-
transpiration, streamflow, and long-lived vegetation such as forests.
Research efforts have typically focused on only a short segment dur-
ing the life span of forest stands that may exceed a century. Moreover,
in areas of low rainfall, where the occurrence of rain exhibits high
OCR for page 223
DATA COLLECTION, DISTRIBUTION, AND ANALYSIS
223
variability over time and space, understanding of such fundamental
relationships is even less complete because sufficiently long data records
are not yet available to separate the
inherent spatial and temporal variability of the processes involved.
We do not fully understand, for example, how evaporation and soil
moisture are regulated in such situations. The need for long-term
data is particularly acute for analyses that focus on hydrologic be-
havior at the continental spatial scales and on long time scales.
Detection of hydrologic change requires long-term data sets of greater
quality and reliability than are normally needed in the investigations
of processes. When we measure rainfall for such purposes as flood
forecasting, modest changes in the techniques, such as movement or
redesign of the gage, do not affect the usefulness of the data for
telling us whether to expect a flood on the river. However, when we
try to use the same data to identify a long-term trend that is superimposed
on the natural year-to-year variability, movement or redesign of the
gage may introduce artifacts into the data set, and these may be
falsely identified as trends or may disguise hydrologic change.
Identifying Extreme Events
Identification and analysis of hydrologic extremes, such as floods
or droughts, are needed to understand the functioning of human
societies as well as natural and managed ecosystems. In most hydrologic
processes the extreme events often have the greatest effects on both
systems and humans. Because they are infrequent in occurrence,
they are poorly represented in all but the longest hydrologic records;
only a few data sets contain enough extreme events to allow a precise
estimation of their return periods. Moreover, the dynamics of extreme
events are hard to measure; stage versus discharge relationships for
gaging stations are usually not calibrated for high stages, and scour-
ing of the channel during such flows makes extrapolation of rating
curves for lower stages prone to error.
Flood frequencies and drought recurrences may be well defined
for mid-range events, but the tails of the distributions are poorly
quantified, both in temporal distribution and magnitude. A series of
extreme events may represent just that, a combination of unlikely
probabilities, or it can show a change in climate, whereby the events
are no longer extreme but merely normal events within a new popu-
lation.
A good example is provided by analysis of the 1985-1986 drought
in the southeastern United States. Estimation of the severity and
interval of likely recurrence for this drought was made possible by
OCR for page 224
224
OPPORTUNITIES IN THE HYDROLOGIC SCIENCES
:: W~ER~RESOUR(:ES~IAGEMENT~i~ i ~~ A A ~~ ~ A i ~~:~ ~~ ~~.
H:u~nd~reds~of~bi I lilions~ot Eli den l~ars~ha~tei~b~et~:nikested :~s~~nce ~ lain i i
I: ~~ ~ ter Or: ~ ~ ~~sta~ em: :~ ~~£~ hi. ~~arI~c ~~::~::~:n::r~l~:VAt~::~WAt~~:~::nro:~:ec~ts id: Jiiam~i.~:~i 1~—irate it ~~:~:i
: ~: : : ~ ~ ~ ~ : : ~ ~: ~ ~ , ~: ~ ~ ~ : ~ ~ : ~ If:: : ~ ~ ~ If: : ~ ~ : : :-: i::: ~: ~ :~ i: ~ ~ ~ ~ ~ ~ : ~~: ~ ~ ~ ~ ~
~ ~ po~r~1~0ns~ ~evees,~c~ ~a~nrte ~i~lmp~~; ~ gatio~ ~—~ i i
~ ~ ~ ~ ~ ~ ~ : ~ ~ ~ ~ ~ ~
~ ~ : ~ ~ ~ w a ~ ~ ~ : t r : ~ a t m e R t : ~ ~ ~ i F , | ~ a n i t S ~ ~ ~ ~ i a ~ r e ~ ~ ~ e ~ p ~ S : . ~ ~ ~ ~ i i ~ ~ t ~ i b e s ~ ~ ~ : r e s e r ~ r ~ ~ ~ ~
~ ~ ~ ~ :~ ~ ~ :~ ~ :~ ~ ~ i ~ i ~: ~ ~ ~ ~
::~ h~ave:~been~:~sed,~: ~: are~: rn:ore::~:seinsitiv~e~ to:~ envimn n~n~i~esi~ :~a~ ~id:~t - . ~
~i~ ain~n~ iiniveStme~ ~t:~i~n: n~eW~ n^e—: ~s ~ on~l~i~sr~al ~tti~ik~ tt~a~bast~i~
:~:~ ~i~:~:vea' s.:~ :~on~:haLs~: ~i~hNi~i~tfu~l aoi~t h~ikimianaa-~: ~:~
~: In~ourwaterre~sources~ :~ ~ ~ :~ :~: ~: ~ ~ ~
~ :: ~ ::: :::~:: :~::::::: :::::: :::: ~:~
~:~ ~ ~i~:I~o~y~a~n~i~m~r~ta~nit~c~haHe~e~ ~ tacesi:~thls~nian~i~:l~ - : C~:I~S:~ffi~
~: ~ ~ ~impro:ve~how~rs~: vast~nvest~; is~tak~;n~g ~:n1D~accaunt~eco-: i~
:: : ~:~ ::: :~:~: :~.:~::~ ~:~ :::~::~::~ ~:~:~.~:~:: ~:::~ ~::~ ~::~ :~::::~: ~:~::~ ::~::: ~::::~:~::~: ~:~::~:~ ::~::::~::~ :~::~:~:~ ~: ~:
~: ~ no~m~lc: ~env~l~mn~m~ental:~,~ ~a~n~d~sociial i valuies.~i~ l~creases~:~i~n :~a~n~nioal be~nefi~i~i~i~
~ ~ ~ ~ ~ ~ ~i ~ ~ ~: i~ ~ ~ i~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ i ~ ~ ~ ~i~ ~ ~ ~ ~ ~ i~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ :~i i ~: ~ ~ ~i ~ i~ ~ ~
::~ ~o~nly~a~i:~few~;percenti~ld ibe~i~wor~ ~ite~n~si~of~b~llil~=si~iof do4~Nia~i~a~coi~ld ~i ~
:: : :~:i::~: ~::~ ~:::~ ~ ~:~::::~ ~::~: ~ ::::~ ~:~:::~ ::~:~:~ ~:~:~:::~ ~:~:~:~ ~:~:~ ~:~:~::: ~:~ :~::: ~:~:~ ~:~:~::~::~:::~::::
~:~ ~preiserve~e:nv~ir~onmenta~l val~ues~ ~that~i~re~i 3gno~red~ ~6ecades~iago.~i~ H~f-~ ~:~
~ ~ ~ ~: . ~ ~ :~ ~ ~ ~ ~ ~ ~ ~ ~ ~ , ~ ~ ~
ca ~ Y'~ wate~r~p~ro~c~ ~ave~ ~een ~p a~n:~r,:ec ~::t~O~seirve~: oine~:set~pur~p oses -i:~ut:~i:~
~::~ operat~d~to se~rw:~ano~er.:~O~n~e e~le~i~:i~s~i~Tenr e~ssee~l~ei:~. ~i
~Origi~:n~al Iy~developied~ ~r~i~:od ~c=oint~ ~ i~: in~p~ro~ites~mele~ir~ic~
:: ~ : ~ :: ::: ::: ~ ::: ~ : ::: ~ : :::: ::: :~ : ~ :~: :::: :: ~:: : ~:: ~ :: : ::~:: :: ::::: : :~: ~ : ~: : ::~ ::: ~ :: :: :: :::: : ~ :::::~: :::: ::::::::: : ::: ::: : ::::::: :: :::: :: ::::: ::::: ~:: :::: :::: :~ ~ : :~:::: ::~:~ : :~ ::: : ::~:: ~::: :~ ::::
Dower ~:wit: 1~ ben:ef~i~ts~:~fa~r~:~ar~eater:~t ~:an ~It ~nise~resu ~n~:~t~om ~ll~di~:nt~.~ ~:~i~:~
Th~i~s :~:~:~ren~ce~b~ee~n~:w~h~i~:i~s:~i~ pl~:r~i~a~i - ~:~i~s~a,~u~W c~:~sug- ~ ~ ~
~ ~ ~ ~ ~ · ~ ~ ~ ~ ~
~ gests~a~ma~nage~ment~o~ortu~n~i - , ~e~3~il~a~ Y~s~l~n-~:ei ~m~uc l~:t se~p~ ~a~n~nI`~ ~
:: :::: :: ~ :~:::: ::~:: :: :: ~ :::: ~ :::: :: 1:::: ::::: :::~: ::: :~::: ::~:: ::: :::
~:occurreC~::oe£aGes: ago.::
~:~ ~:~ ~A~:~m~ajor~dieraT~w~ate:r:~:resources~ agen~cy~ t:h6~U .S. ~B~ureau~of ~Rec~larn:a ~::~
~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~, ~ ,, ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ . :~ ~ ~ ~ ~ ~ ~ :: . .~ ~ ~ ~ ~ ~ ~ . ~ ~
~ ~ t~iio~n'~:rec=~ogni:z~en~ ~th~iiS~i~ ch~a~l~lenge:~anO~mrm:al~ly~re2~lign~m~sssfon~iaMf~ ~
~ ~ ~ ~ ~ ~ ~ ~ · ~ ~ ~ ~ ~ ~ ~ ~ ~ T~ ~ ~ ~ ~ ~ ~. ~ ~ ~ ~ ~ T~ ~ ~ ~ ~ ~ ~ ~ T~ ~ ~ ~
::~: ~t;:o~m~ co~n~str:uct~ oa~ iaina :~towa~m~prove-~ ~mana~mein:it~a~n:~a~at~mn ~iot~i :i
::: : :: ~ : ::
~. ~ ~ ~ :: ~ ~ i: . ~ ~ ~ ~ . ~ ~ ~ ~ ~ ~ ~ ~ T~ ~ ~ ~ :~ ~ ~ ~ ~: ~ ~ _
~ ~ts:~vast~ ~prior~:nv~es - ~em~i~t~ne~:~ il:7~mr~n:~states ~I~ne~n~atioi~n:~S~i~:a:rgest~ii~
:: ~ ~ ~: ~ ~ ~: ~ ~ ~: ~ ~ :: ~ :: ,: ~,~ ~ ~ .~ ~ ~ ~ ~ ~ ~ ,~ - ~: ~ . :~ ~ ~ ~ ~
~wate:~:~resiolo~ll-c~eS~agelelcy~i~tne~:u~.~.i~i~Amy~i~rpS~or~ng~:J~n~eer~S;~:nais~i~nm~ir ~i~
:::~ ::::~:::: : :~::: :::: :::: :: ~ :::::::::: :::: : ::~::: ::::::~ : : ::~::: : :::~:: ::::~ :: :::~:: :::: ::~::::: :~:: ::::: ~::~: : ~: : ~:~ ~:::: :~:~ :~:
::::~::::: ~ ~ ::: ::: :1: :: ::: :1 :::: :~::: :::: ::::f: :~: ::::: ~: :: ::::~::: ::~: :: r:: :.:: :: :: ::f:. :::: :: .~ : :::: :~::: ::::: l :::::: ~:.: :~.:: .~:: :::~: ~:~:::: ::~. .:::::::::: :::: :::: ::~: :::: ~
~ ~sucn ~a~:~to.-ma~l~:~s~p,~ou~t~: i~i~m~air.~y:~or~i~:l~ts~:lstr,ict :~an~i~a::lv~lsil~:o~n~rices~ ::~are~
:~fin~d~i~rg~new~n~on~strucitu~ral:~:s:ol~Won~s.~l~nagemebt~carifli~i:~Warise~among~:~
~ ~ ~ ~ ~ ~ · ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~, ~ ~ . ~ ~ :: . ~ ~ ~ ~ .
~ :~:~ ~co~mnet~ ~u~se~s~ ~ wate~ ~ ~s~c.~i ~as~ ~lirr~lR~atIon~ ~m~uin~c~! Da]~ siU=,hm~~ ~-~i~
:~d~iropow~er~ ~::~:~,=~wer ~cool ing ~ wi~ifNe~prese~rvafion,~ fisl~i - ~:: q~.iW~i~:~
:~ control: ~'no.~ust;rla~l ~u~se~ ~:~:nav~f~gat:l~o~n,~ano recreat~:~.~ ~:~t~l:mm~:~ mana~g;em~ent-~:~:~
~ ~ , ~ ~ , ~
:~ ::~:of :the :~n:at:ion'~s~ ~water:~ re~sourc~es~ :is:::~:e~ ~key~:~:to ::~reso.iiv~ing::~:se~cai~fli~c~ts.~::~:~::
:::: ::: :::::: ::~ :::: : :: :: ~::: :: :: :: :~::: : :: ~:~:: ~:::: : ::::: :~::::::: :: ~ :: :: :::~: : ::: :~::: ~ :::~::: ~:::::: ::~ ::: ~ ::~: ~ : ~ ~ ~ :::~ ~:i::::: :::: ~ : :~:~:~::~ ::
:~: :~ ~better~ i~nform:atior.~:lis~the::i~key~to ~imp~£oved:i~:water ~imarlagem - ~:~ ~ ~ ~ ~:~ ~
j ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~: ~ ~ ~ :~ ~ ~ ~ ~ ~ ~ ~ ~ :~
T~he '.n~porta~nce ~ot water: re~soUrces~lnana~g~en'~ent ~was: ~.~::~t~g~h~l~lgh~ted~ ~by~ ~:~
: , :::~ : ::: , :: :: :~: :~ ~, ~, ::~ ~: ~ ~ :~ ~, , :~ :~ ~: ~: ~ :: ~ ~: ~ ~ ~ ~:~ ~: ~:~
~ ~ the 1:98~8::droL~t~g~h~t. ~Print:~ano~ ~e~l~ectronic ~',neDia~oroug~t~; u~s~ ~cl~ear~ l~m~age~s~ of~: ~:
~ ~ , . ~ ~ ~ ~ ~ :: ~
the consequences o t ~d~rowght~ba~rges ~stranded~on~ sand~ba~:rs~ ~in~ th~e~t\A.isi-~ ~ ~
::: ~ :~ :: ::: :~ ::: : :~ :: ~ : :::: ~ :: :::: :~ ~ :: ::::::: :::::: : :::: ::::::::::
B~ ~ ~ ~ ~ ~ - ~ : ~ ~ . 1 :::~ ~:: ~ ~ ~ :~ ~ ~.: ~1 ~ 1 : ~ :: _~~ :: :: ~ : :~:~ : ::~: ::1:~ ~ ::
sissippl Klver empty re~servo'~rs ~ano witnerea.~ corn~Tl~e~tG.s~.~ ~l~e~oro~t~grit~
::i l I us:trated: d ramatic~al ly ~::~that:~th~e~ ~ avai~l~a:~l iW~l:~of ~water~:: i r~vo~Ives~:~r~i~sk :~a~:nd:~::~
t~hat th~e risk see~ns ~t o be~ increasing.~ ~ ~H~ow~ wil I in~creases in~ :atm~ospheric~
caroon~ U~ox~tUe~al~lu otner Ereenhou~se~aa~ses~chan~e:~ the cl~matet :~ W:~hat:~:~:~:
~ ettects w~ll I such ~ch~ar.ges~h~ave~ on~ou~r~wa~r~ ~r~eso~u~rc~es .li~ ~
bom~e ot the ~ntormat~on nee~nen~ to '~mp,rove:~:wa~ter ~reso~.~r~ces ~mana~ge-~
~ ~ , . ~ ~ , ~ . ~ , . . ~ , ~ , ~ _
~mer`.t ~:w' 1~.~ co~me~:~ tro~m~ ~ l~mprovec~ ~ ~a~pp~l~cat'~o:~.?.~o:J:~ex~::s~t~l~n~g~ ~tec~nn:ology~. ~ ~:~:tor:
~::: exa~mple~,~ ~new~d~ata: ~sy~stems~u~sing~ a~ut ~mated~:sur~ce~ :6bservat~ons:)~sa~te~l-:~:
:~:Iite c:ommun:ications, :w:eather rada:r ~and:~satell:~ite::i:m~agery~re~beg~inning::
:: : : ~ :: :~ : : :: ,:: ::~ ~:: ~ : : :~ ~ ~:
; ::~:: ::::
::
::
:~:::~
:::~:
: ::::
:::
::
::
~: :~::
:::~ ::~:
OCR for page 264
264
o
~ LL
Z
> cn
~ '_
Z
Z
~ c-)
<3 m
G O
> 0
<: —
)
1
a'
-
a
o
Q
a
>
n5
a'
1 ~
a)
~n
1 c
8
a'
~n
ra
c
-
~:
o
~ —
CE
IIJ
> C~
~ a)
o—
tr
o~
J ~
OCE
()o
1 1
. .
d. ~ O CO CO ~ C~ O
HV3A U3d SH313W OlEn
CE
6
C~ LIJ
>~
I
C~ I
l
~ /1
~ I
._
CD
J O)
~ '
Z C'
Z
~ _
I1, Q
CE
OZ
C)
LL
~n
~ l
m a
.
Z
— C
>
C)
LL
C~
~ m
cn
Z
8 o
LU —
CE
,. ..
o
d C~ o CO CO ~ C~ o
. .
g ~
oo
O
U)
o 60k ~V3A 83d SH313W 019nO 60l
(~all6M) (J8!JP)
H3GlM H3MOUU~N
SONIU 3381
o
o
o
C~
C'
\ ~
._
CD
._
s~
o _
"c" t~
o ~
_ _
.
U —
~ ,1:,
U, o
U
o
U
b4 ra
= ~
~ o
1 . -
,Y
U
o
C~
o C
d' ~o
:^
UO
_`
OCR for page 265
DATA COLLECTION, DISTRIBUTION, AND ANALYSIS
265
other proxy studies could be applied. Sometimes they provide infor-
mation where there are no instrumental records even today, for ex-
ample, in coral reefs and ice cores. To execute the application requires
interdisciplinary science. There must be expertise in the particular
proxy area e.g., palynology, dendrochronology, sedimentology and
expertise in hydrology itself to develop sensible and significant results.
Although it has been used successfully in many areas of the world,
there are now more opportunities than accomplishments in paleo-
hydrology.
Data Accessibility and Management
Advances in the hydrologic sciences depend on how well
investigators can integrate reliable, ~arge-scale, ~ong-term
data sets.
Storms, floods, and droughts are natural events that can be mea-
sured just once, whereas laboratory experiments can be repeated.
Instruments used in hydrology must be reliable and operated such
that data captured are of known standards and precision.
On rivers, measured stage data must be transformed to discharge.
The stage-discharge relationships, commonly called rating curves, typically
must be extrapolated to extreme stage values and may require adjustment
as new "agings at the extremes become available. This adjustment
may apply retrospectively for rating curves that have been used for
many years, and so a data archive should store original stage mea-
surements and rating curves separately, to allow this retrospective
examination and adjustment.
The data sets required to answer many of the open research ques-
tions in hydrology will be complex. Inevitably, many scientists from
a variety of disciplines and backgrounds will be involved in data
collection and analysis, over a significant period of time. How can
diverse investigators and investigations produce compatible data sets,
assure their quality, and confidently assemble them for larger, indeed
public, use and access? Creating effective data systems for assembling
and distributing scientific data sets is not trivial and depends heavily
on the personal efforts of active scientists. If the data systems are
constructed within the scientific community by scientists themselves,
rather than by independent data "experts," there will be many scientific
opportunities as well as technical and political challenges.
OCR for page 266
266
OPPORTUNITIES IN THE HYDROLOGIC SCIENCES
Data Management in the First ISLSCP Field Experiment
The First ISLSCP Field Experiment has wrestled with these issues.
FIFE was not designed exclusively as a hydrologic experiment, but it
included collecting a comprehensive set of hydrologic and related
data sets for a 15 x 15 km2 experimental site in central Kansas. The
studies of some 30 cooperating principal investigators were actively
supported by a data system built by participating staff scientists at
NASA's Goddard Space Flight Center.
The immediate goals of the data system were to capture and preserve
the data and distribute them as rapidly as possible. After the conclusion
of the field sampling, the system was converted into an open, widely
available archive. The technical elements of data distribution were
easily supported: magnetic media via mail for large volumes and an
on-line access via electronic networks for browsing and routine data
extraction. Equally important, however, was a user support staff.
Technical and scientifically competent people were required to simplify
access, prepare adequate documentation, and teach novice users about
both the system and the data.
Assessing data quality can be difficult enough for a single investi-
gator working with his own data. The problem is compounded when
the data are required quickly by cooperating investigators and distributed
through a data system to people who may be unfamiliar with the
technical details (or difficulty) of the measurement. The FIFE solution
was to use the data system as the focus for a cooperative assessment
of data quality.
Data Storage and Access
Optical disks and compact disks have become an attractive alternative
to traditional magnetic tape or disk storage media, because they offer
the capacity and security necessary for hydrologic archives, and because
multiple copies of large archives can be made cheaply. For example,
the entire daily stream gaging record of all gaging stations for one
year is stored on optical disks for such countries as the United States
and New Zealand. The need to publish expensive yearbooks of data
disappears.
Issues to Resolve
The evolving requirements characteristic of active research demand
a data system with real-time adaptability. This can be achieved by a
scientifically involved data system team that puts a priority on service.
OCR for page 267
DATA COLLECTION, DISTRIBUTION, AND ANALYSIS
267
This is no small effort: it requires direction on a day-to-day basis by
active scientists. In particular, for large projects, the role of a project
information scientist must be recognized as critical and must be re-
warded appropriately.
In addition to personnel issues, there are several political aspects.
Three legal categories of data can be defined: data that are acquired
OCR for page 268
268
OPPORTUNITIES IN THE HYDROLOGIC SCIENCES
from public sources with no distribution restrictions, data that are
collected by publicly funded principal investigators, and data that
are acquired with public funds from private sources, corporations, or
individuals with specific legal rights to restrict their distribution.
The FIFE approach was to recognize a data collection and analysis
phase in which data sets were exchanged and revised and their quality
controlled, but during which general access was restricted. The experience
suggests that the more direct control scientists have over the data
system, the better it will serve science. In particular, direct control
makes possible rapid adaptive responses to the unexpected opportu-
nities that can develop in any experiment. A control data base can be
a tool and focal point for cooperatively assembling and checking the
data sets. Advances in computer technology make it possible to link
such a data base electronically to field sites and investigator laboratories,
where a common set of hardware and software tools can be inexpensively
supported. These would allow each scientist to create his or her
portion of the data base, in near real time.
Challenges in Measuring Water Quality
Public concern with pollution of water resources, as well
as its effects on human health and the environment, is
widespread and occasionally intense.
Investigations of water quality must be designed accorcl-
ing to sound scientific principles.
In response to public concern, many studies are being conducted
to monitor and assess the amount and distribution of pollutants entering
the hydrologic cycle. If these studies are to be useful to understand
the causes of observed conditions, and thus provide a foundation for
cost-effective amelioration of water quality problems, they must address
scientific principles as well as practical ones.
Water Quality Monitoring and Assessment
Data for water quality monitoring and assessment may be divided
into three types:
OCR for page 269
DATA COLLECTION, DISTRIBUTION, AND ANALYSIS
269
· data collected to characterize ambient concentrations in lakes,
rivers, and ground water;
· data collected to monitor effluents; and
· data collected to monitor water quality for a specific use.
The remaining discussion focuses on ambient water quality data.
However, managers of data collection programs for all three types of
data need to become more aware of ways that data from individual
programs can be made more useful for addressing issues that are
beyond the immediate program objectives. These include:
· the need to collect important ancillary data, to place the water
quality data in the context of the natural and cultural setting;
· the need to carefully document sample collection and labora-
tory analysis procedures; and
· the need to archive the raw data in easy-to-access computer
files.
Scientific Issues and Challenges
Past experience shows that water quality data collected for utilitarian
purposes are either difficult or impossible to use for scientific purposes.
It is seldom appreciated that science-oriented designs not only contribute
to advancing science but also significantly improve the process of
attaining many practical goals.
Water quality is threatened by thousands of potentially harmful
substances. Developing effective evaluations of water quality for so
many chemicals is an imposing challenge, requiring continued devel-
opment of screening techniques and broad-spectrum analytical pro-
cedures. We also need better ways to link contaminant selection to
the physical-chemical properties of different substances, to the behavior
of different substances in surface and ground water, sediments, and
plant and animal tissues, to chemical usage estimates, and to the
relative health and ecological risks associated with different pollutants.
A related issue has been the failure of traditional monitoring pro-
grams to identify emerging water-quality problems, possibly because
of the lack of a significant link between these programs and scientific
inquiry. For example, most water quality sampling in the United
States has been targeted exclusively at substances for which regulations
already exist, leading to a focus of effort on selected constituents—
priority pollutants that occur infrequently, and often to a disregard
for more important contaminants.
Future data collection programs need to provide explicit flexibility
to enable adjusting to changing environmental concerns and incorporating
OCR for page 270
270
OPPORTUNITIES IN THE HYDROLOGIC SCIENCES
exploratory aspects into the design. Frequent interpretation of data
also is required to identify emerging issues; the data should not be
simply collected and archived for future analysis. The integration of
biological measurements with physical and chemical measurements
also can significantly strengthen the utility of a data collection pro-
gram to help identify emerging problems. For example, biological
properties may be more sensitive to water quality than are chemical
or physical measurements. Too often chemical and biological mea-
surements are considered competitive rather than complementary as-
pects of water quality characterization.
The design of water quality monitoring and assessment programs
usually does not reflect consideration of the issue of scales. Yet the
scale of focus will constrain the issues that can be addressed, for
example, in providing information on non-point-source contamina-
tion. Simple use of highly intensive area sampling will not produce
significant results within the limits set by realistic funding and the
human resources available. Instead, innovative designs must be developed
that fully use the existing understanding of the physical, chemical,
and biological processes that determine water quality.
A major deficiency in environmental data collection programs has
been the inadequate development of information useful for defining
long-term trends in water quality. Part of the problem is simply that
data collection programs are too easily abandoned when funding problems
occur or in the excitement of responding to newer, more glamorous
social or scientific issues. A greater commitment to continuity is
needed. Moreover, a key challenge is to carefully balance long-term
consistency with inevitable changes in hydrologic knowledge, the
technology available for field and laboratory measurements, and the
types of contaminants extant. To the extent possible, long-term pro-
grams should rely on repetition of measurement, but they must also
document carefully the criteria for site selection, the characteristics of
sampled sites, and the methods of data collection and analysis. When
changes in measurement techniques occur, the old and new techniques
should be applied in tandem as long as is necessary to determine the
relationships between them.
Interrelationships among components of the hydrologic cycle must
also be considered. Understanding of the connections among the
atmosphere, surface water, and ground water needs to be incorporated
into the design of environmental monitoring programs for these dif-
ferent media. For example, atmospheric cycling can be critical to the
transport of major and trace constituents of terrestrial waters. So in
some circumstances, a basic understanding of atmospheric processes
OCR for page 271
DATA COLLECTION, DISTRIBUTION, AND ANALYSIS
271
and appropriate atmospheric monitoring may lead to more effective
collection of data describing water quality.
Use of Biological Methods in Water Quality Analysis
Biological information can complement chemical analysis
to improve the measurement of water quality.
Physical and chemical properties of water may vary rapidly, and
intermittent or infrequent "grab" samples may give misleading indi-
cations of prevailing water quality. The native biota may be better
indicators of water quality and human effects because of their prolonged
exposure, integrated response, and differing sensitivity to all the varying
conditions of their environment. Indeed, organisms provide the only
direct measure from which ecologically significant impacts can be
deduced. All levels of biological organization molecular, cellular,
tissue, organ, individual, population, and community have been used
or proposed for use in water quality interpretation. The methods
may or may not identify a particular cause of change, but a measurable
biological response may help to identify physical or chemical tests
that should be used in the search for a cause or causes.
The first biological methods used in connection with water quality
assessment were based on the observed presence or absence of species.
Characteristic native species were used to demarcate zones of decreasing
concentration downstream from a point of heavy organic loading.
Particular species were thought to show the pollution condition in
each zone. However, the supposed indicator species also occurred in
unpolluted environments, and the zonation varied with the type and
intensity of pollution and other hydrologic properties. Further work
on human effects resulted in methods based on analysis of assemblages
of species. The relative dominance of tolerant and intolerant species
or of functional feeding groups in a biotic community is sensitive to
water quality. These methods are successful when enough ecological
knowledge exists about the species used, as is the case for most fish
(although fish may be impractical to sample). They are less success-
ful when the ecological requirements of the species are poorly known,
as is usually true for algae and benthic invertebrates. In the absence
of detailed information for the species of interest, effective ecological
methods are available based on resemblance between biotic commu-
OCR for page 272
272
OPPORTUNITIES IN THE HYDROLOGIC SCIENCES
nities in hydrologically similar streams, with and without human
impacts. The selection of suitable reference streams is crucial to the
success of this approach.
The occurrence of one type of effect, sewage contamination, has
traditionally been determined using as tracers microorganisms indigenous
to the gut of humans and other warm-blooded animals. Bacterial
density in laboratory cultures inoculated with water samples is inter-
preted to show the degree of fecal contamination and the potential
occurrence of associated human pathogens. Escherichia cold is replac-
ing fecal coliform and fecal streptococcus in these tests as a more
specific indicator of human effects.
The sensitivity of organisms to target contaminants or the concen-
tration of contaminants in living tissue can be used to detect the
spatial distribution or biological availability of contaminants. The
method samples native species or introduced, caged species. It is
limited by differences in sensitivity or in uptake of contaminants
among species, by lack of suitable widely distributed sentinel species
in continental waters, and by effects of enclosure on caged organ-
isms.
Laboratory bioassays using sensitive organisms are performed to
determine biological effects of specific environmental characteristics.
Responses, usually from short-term tests, are measured as bioaccumulation
or as changes in behavior or physiology. Although test conditions
are standardized, thus far the results cannot be extrapolated to other
test conditions or species. In particular, bioassay results do not directly
provide adequate information about an effect on the long-term structure
and functioning of ecosystems.
Limitations of single-species bioassays have led to the use of laboratory
or field-emplaced microcosms to determine the effects or the fate of
contaminants. The sizes of such microcosms range from less than a
liter to many cubic meters. Microcosms contain important components
and exhibit important processes of natural ecosystems. They simplify
environmental variability while exhibiting multispecies phenomena
under controlled and replicable conditions. The results obtained from
experimental microcosms are empirical analogues of whole-ecosystem
functioning but require great care in broad extrapolation to the field.
Methods based on levels of organization below the individual level
are applied in the field or laboratory to detect, quantify, or determine
possible human effects. Techniques based on enzymes, antibodies,
tissue cultures, and gene probes are being used or actively developed.
The degree of sensitivity and specificity possible with these methods
suggests that their use in water quality analysis will increase.
Clearly, biological data can supplement physical and chemical data
OCR for page 273
D~ COLE DlS~lBU~^ ~~D APSIS
2~
to provide more holistic understanding of Me Coning and of the
natural evolutionary bends of hydrologic system as emu as human
effects on such systems To accomplish this in detain major advances
are needed for determining the hydrologic implications of ecological
results Also needed are improvements aimed at increasing the sen-
si~iV, subdue ~ ~~ of biological methods Ed at decrease
costs and analytical bme. To date, only for indicator bacteria have
procedures been adequately st~dard~ed Ed ~ result made accessible
in water quality data banks. Other biological data relevant to water
quality assessment are scattered and are based on diverse methods of
sampling and analyst. Standardized methods Would enhance the
sciendRc value of biological ~krmabon by providing a refile baseline
fir making Choral Ed spatial co~adso=. Proved cocoon
of ecological results and their significance is also needed' in forms
useful to other scientists and to the public.
Biology can furnish uncommon insights for hydrologic science, in-
~ghts not achievable solely Tom a Edge of physics and chewy.
For example, organisms are involved in the transport and cycling of
elements in water and sediments. Organisms are targets of scientific
Earl to preserve rare Ed Educed species. Pop~abo~ of orgasms
are ~tendonaDy added by management programs and u~nhonaDy
affected by natural and anthropogenic environmental effects These
and other issues often require studies on large spatial and temporal
scales. Saw sages Cold be ~o~ora~d Ho national Ed ~temabonal
mater quality monitoring systems to provide the means to evaluate
Ed Prove Completely developed but potendaNy valuable biological
methods for understanding the organization and functioning of hy-
drologic systems.
SOURCES AND SUGGESTED READING
Andre, ~ Cal ~ P. Goutorbe, and A. Perrier. 1986. HAPEX-~OBILHY: A hydrologic
atmospheric experiment for me study of water budget and evaporation flux at
the climatic scale. BulL Am. ~eteorol Sac. 67:138-144.
Baumgarmer, A., and E. ReicheL 1975. The World Water Balance. Elsevier, Amsterdam,
179 pp.
Bernabo, C. 1978. Proxy Data: Natured Records of Past Climates. NOAA Environmental
Data Service ReporL U.S. Department of Commerce, Washington, D.C.
Earth System Sciences Committee, NASA Advisory Council. 1988. Earn System ScF
ence: A Closer View. National Aeronautics and Space Administration, Washing-
ton, D.C.
EOS Science Steering Committee. 1987. Earn Observing System. Vol. IL From Pattem
to Processes: The Strategy of the Earth Observing System. National Aeronautics
and Space Administration, Washington, D.C., 140 pp
OCR for page 274
274
OPPORTUNITIES IN THE HYDROLOGIC SCIENCES
Haeni, P. 1983. Sediment deposition in the Columbia and Lower Cowlitz rivers, Wash-
ington-Oregon, caused by the May 18, 1980, eruption of Mount St. Helens. U.S.
Geological Survey Circular 850-K, 21 pp.
Kinter, J. L., and J. Shukla. 1990. The global hydrologic and energy cycles: Suggestions
for studies in the pre-global energy and water cycle experiment (GEWEX) period.
Bull. Am. Meteorol. Soc. 71(2):181-189.
Krishnaswami, S., W. C. Graustein, J. F. Dowd, and K. K. Turekian. 1982. Radium,
thorium and radioactive lead isotopes in groundwaters: Application to the in
situ determination of adsorption-desorption rate constants and retardation factors.
Water Resour. Res. 18(6):1663-1675.
Robertson, W. D., and J. A. Cherry. 1989. Tritium as an indicator of recharge and
dispersion in a ground water system in central Ontario. Water Resour. Res.
25:1097-1109.
Ryan, P. F., G. M. Hornberger, B. J. Cosby, J. N. Galloway, J. R. Webb, and E. B.
Rastetter. 1989. Changes in the chemical composition of stream water in two
catchments in the Shenandoah National Park, Virginia, in response to atmo-
spheric deposition of sulfur. Water Resour. Res. 25:2091-2099.
Skinner, B. J., and S. C. Porter. 1989. Physical Geology. John Wiley & Sons, New York.
Smith, R. A., and R. B. Alexander. 1986. Correlations between stream sulphate and
regional SO2 emissions. Nature 322:722-724.
Stockton, C. W., and G. J. Jacoby. 1976. Long-term surface water supply and streamflow
trends in the upper Colorado River basin. Lake Powell Research Project Bulletin
Number 18. University of California, Los Angeles.
U.S. Committee for an International Geosphere-Biosphere Program. 1986. Global Change
in the Geosphere-Biosphere: Initial Priorities for an IGBP. National Academy
Press, Washington, D.C.
Representative terms from entire chapter:
hydrologic data
