United States
Environmental Protection
Agency
Corvallis Environmental
Research Laboratory
Corvallis, Oregon 97330
60019782
BIOLOGICAL SIGNIFICANCE OF FLUVIAL
PROCESSES IN THE LOTIC ENVIRONMENT
CERL - 042
-------
BIOLOGICAL SIGNIFICANCE OF FLUVIAL PROCESSES IN THE LOTIC ENVIRONMENT
By: Frederick B. Lotspeich
INTRODUCTION
It has long been recognized by aquatic biologists that the physical
nature of a stream bottom directly influences its productivity as an anadro-
mous fishery and the benthic communities comprising the bulk of fish food.
Experts concur that optimum fish and fish-food production requires clear, cold
water and a gravel substrate free of fine sediments to permit maximum oxygen
transfer, provide for removal of metabolic wastes, and allow sufficient pore
size for developing embryos and movement of benthic organisms. Aquatic biolo-
gists have also observed that within a given area encompassing separate water-
sheds, some streams consistently produce more fish.
Recent legislation mandates that future attention be given to the effects
of nonnoint source pollution - primarily sedimentation - on low-order streams,
the same streams that are of prime concern to aquatic biologists. All of
these requirements concern the biologist who is asked to devise ways to evalu-
ate stream systems with respect to relative productivity and document how low-
order streams respond to sedimentation from nonpoint sources. A distinction
must be drawn between low-order streams that are most amenable to control
measures within each watershed and those less so but where biological produc-
tivity is of prime consideration. In mountainous terrain, first ana second-
er -der streams are steep, bedrock is near the surface, and clastic spdiments
are thin and intermittant. Such streams cannot support a high biological
productivity, but because their aggregate linear length constitutes 80 to 90%
of total stream mileage, they act as collectors of sediment from nonpoint
sources. Sediment from these headwater reaches then moves downstream to third
and fourth order streams where productivity is high and excessive sediment
degrades the aquatic environment. By controlling sediments in unproductive
headwater streams, we can maintain optimum production in higher-order streams
1
-------
which do not act as sediment collectors but are impacted by erosion products
from careless management practices in headwater basins.
Therefore, assessing stream quality and evaluating best management prac-
tices requires expertise in physical as well as biological sciences. Any
evaluation of stream quality demands an understanding of stream dynamics and
mechanisms of sediment transport over a wide range of hydrologic conditions.
Only with such knowledge, can an environmental scientist provide advice in
developing a program of habitat evaluation as best management practices are
introduced to control erosion products from nonpoint sources.
For many years, physical scientists have generated information that, when
properly applied to aquatic biology, could help biologists evaluate many
sedimentation problems associated with nonpoint sources. Geologists special-
izing in sedimentation have relevant expertise, as do specialists in certain
fields of engineering, soil physics, hydrology, and ceramics, to name a few.
Collectively, these researchers have amassed a wealth of knowledge on proc-
esses and properties of clastic materials that can help solve many problems
facing lotic biologists. However, biologists often may not be aware of the
related work in other disciplines and the rich literature dealing with nat-
ural, clastic materials. This is not to imply that river mechanics are tho-
roughly understood, especially for low-order streams with gravel beds. Em-
phasis is, and has been, on sand and silt-bedded streams which are moderately
well understood. Gravel-bottomed streams are more difficult to study because
current methods of measuring bed movement and grain-size sampling leave much
to be desired. A more thorough understanding of these stream types requires
improved sampling methods for these two parameters.
However, with present knowledge and application of a few straightforward
measurements, streams can be readily evaluated with sufficient accuracy to
describe the habitat requirements for a particular use of a stream. In this
discussion, only fluvial processes relating to sediment movement will be
described; other processes such as meandering, riffle-pool formation, sinu-
osity, etc., will not be considered.
Recognition of the interdisciplinary nature of lotic biology requires a
melding of biological and physical sciences if aquatic ecology is to move
effectively to solve environmental problems. Moreover, this can only be
attained by establishing a professional rapport among scientists, gaining a
-------
better understanding of physical processes and use of conventional physical
measurements, using a common classification, and improving communication
between biological and physical scientists. The multidisciplinary approach
will achieve these goals more efficiently and with greater satisfaction than
if each scientist works in isolation.
The primary intent of this paper is to describe the origin, physical
processes, and properties of clastic sediments to help biologists understand
the fluvial environment and how these factors relate to lotic habitat re-
quirements. In an attempt to provide succinct, brief descriptions, the author
uses a minimum of technical "jargon" and no equations. Another objective is
to define those essential terms, commonly used by physical scientists, that
will enable an multidisciplinary group to share a common vocabulary and thus
facilitate communication. A third objective is to describe several methods of
stream bed evaluation to help biologists interpret physical data from the
field and laboratory in relation to lotic biology when assessing stream qual-
ity.
MATERIALS
As used here, materials will include sediments processed while using
flowing streams as the transporting agent. The unit controlling the func-
tioning of any stream is the watershed which provides sediments and water to
move them and which should be considered as an ecosystem because of the in-
teractions of the physical environment and the life contained within it.
Biologists unfamiliar with terms used by physical scientists frequently misuse
or coin new terms when describing sediments. To establish a common termin-
ology, several terms that will be used throughout this paper are defined.
They are included in the text because it seems appropriate to incorporate them
early in the discussion and avoid the inconvenience of searching for a defin-
ition in a glossary.
-------
Definitions
Bed transport: The movement of grains or particles (too large to carry in
suspension), by rolling, skipping, or sliding on or very near the stream
bottom.
Clastic: Fragment of any earth material and present as discrete grains rang-
ing in size from clay-size to boulders; clastic sediments (in our discussion,
alluvial) are thus distinguished from other sediments such as chemical, i.e.
limestone, cherts, etc.
Competence: The ability of a stream to move the largest grain, i.e. particle,
in a given reach.
Fines: "Fines" as used by sedimentologists and soil scientists refer to
grains smaller than 100 microns (0.1 mm), anything larger would be sand,
gravel, cobbles, etc.
Fluvial: Relating to rivers, conforming to changing course of a stream,
produced by river action.
Gradient: The degree of inclination, in the case of a stream it is the water
surface or that of specified reaches of the stream bed. Usually expressed as
slope or the ratio of vertical distance divided by horizontal distance.
Hydrology: The study of the interrelationships and reactions between water
and its environment in the hydrologic cycle.
Lotic: Relating to or living in actively moving waters as in stream currents
as opposed to limnic or standing body of water (lakes, ponds).
Median size: Relating to the size-percent accumulation curve; the median size
is the grain size (in millimeters) where the 50% accumulation (by weight)
intercepts the curve.
Order of stream: A systematic way to classify streams from their headwaters
downstream; a first-order stream is the uppermost headwater stream without a
tributary, after 2 first-order streams join they are classified second order
and so on downstream.
Permeability coefficient: A dynamic property of a clastic sediment that
expresses its ability to transmit fluids. The rate of flow depends on the
hydraulic gradient, the area of the cross section, the length of the trans-
mitting column, the nature of the porous media, and the viscosity of the
fluid. When all factors, except the media, are held constant, the rate is a
-------
function of the transmitting sediment and is termed the "permeability coef-
ficient."
Pore space: That space among the grains of a clastic sediment filled with
water or air; usually expressed as a percent of the total bulk volume.
Pore space efficiency: A term coined for this discussion to denote the rela-
tive efficiency of the interstitial voids of a substrate to support vital
biological functions which are corollary to the quantifiable hydraulic proper-
ties, permeability coefficient, and specific yield. Pore space efficiency is
envisioned as a field habitat description derived from field/laboratory esti-
mates of a specific yield over a time span.
Potential energy of a stream: The energy of a mass of water owing to its
position above other similar masses; this is the source of all energy pos-
sessed by flowing waters.
Sediment: Geologists define sediments as deposits of solid material on the
earth's surface from any medium (air, water, ice) under normal conditions of
the surface. Here we will only be concerned with clastic sediments deposited
in streams (alluvial sediment); textures range in size from clay to boulders.
Sediment discharge: A time rate of movement of dry weight of sediment through
a cross section.
Sorting coefficient: An index derived by sedimentary geologists to describe
the size distribution of grains about a median; defined as the square root of
the quotient of the grain size at the 75% level divided by the grain size at
the 25% level - a perfectly sorted sediment has a coefficient of one.
Specific yield: That amount of water yielded by a saturated sediment under
the force of gravity; usually expressed as a percentage of a unit volume of
sediment, coarse sediments yield more than fine sediments in a shorter time
span, this is a common term used by groundwater hydrologists.
Suspended transport: Fine grains carried by flowing water held in suspension
by turbulence or colloidal processes; size of grains held in the water column
is primarily a function of velocity.
-------
Clastic materials originate by physical and chemical processes of weath-
ering of all types of rocks. Properties of the elastics are controlled to a
large extent by the lithology from which they are derived as weathering prod-
ucts. Processes of erosion and mass wasting move these materials downslope
until they reach a stream which then modifies them by fluvial action and they
become true alluvial sediments. Fluvial processes continue to modify the
elastics by reducing them in size through mechanical abrasion, impact, and
hydraulic sorting as they are carried downstream.
Two properties of clastic sediments, porosity and permeability, are of
fundamental importance to aquatic biologists and will be discussed later.
These two derived properties can in turn be described by five properties of
grains comprising a sediment. They are (1) composition, (2) size, (3) shape,
(4) roundness, and (5) packing.
(1) Composition is a property derived from the original source rock
that determines the grain's mineralogy and how it will behave during
fluvial processes. Thus, a grain weathered from igneous bedrock will
have a different shape and resistance to abrasion than clastic material
from soft sediments or metamorphic rocks. Clastics from massive igneous
rocks (chiefly granitic types) tend to be equi-dimensional whereas those
from gneisses or schists tend to be tabular or platy.
(2) Size is controlled by source rock, weathering processes, how
far the grains have been transported, and degree of sorting. Clastics
from tabular rocks (schists, gneisses, laminated sediments) cleave at
linear planes and are reduced to smaller grains in a shorter distance
than are the massive rocks. Sand-sized and smaller grains tend to be
monomineralic whereas larger grains are composed of several minerals
derived from the parent rock.
Although sedimentologists have been using a standardized classifica-
tion scheme for decades, biologists have never seemed to be aware of it
during their discussions of sediments. The most widely used scheme is
that of Udden-Wentworth, which is presented in this paper, and is based
on the millimeter as the central unit. Sediments are usually classified
by dry sieving for coarse separates (> 0.062 mm) silt size and by some
wet procedures for < 0.062 mm. Grains larger than 1 mm are classed by
-------
doubling each sieve opening and those less than 1 mm by halving each
class interval as shown in Table 1.
(3) Shape of grains in a sediment is a property controlled pri-
marily by grain lithology. The shape of clastic grains causes them to
behave in a variety of ways during transport and imparts various proper-
ties to the resulting sedimentary deposit. Four basic shapes describe
clastic grains:
(a) equant or grains whose three "principal or orthogonal"
axes are nearly the same length (tend to be spherical).
(b) tabular, or disc shape, where two axes are near the same
length with the third much shorter; such a shape characterizes
elastics from schists and gneisses.
(c) bladed, where all three axes vary considerably in length
from one another giving a rectangular shape in two directions with a
short axis at right angles.
(d) cylindrical, where two short axes are about the same
length with the third much longer giving a rod shape to the grain.
(4) Roundness, as a descriptive property of clastic grains, is not
related to shape. Roundness is that condition that gives curvature to
angles and tends to obliterate surfaces caused by cleavage or fracture
planes; thus a blade-shaped clastic can be just as rounded as an equant
grain but will behave differently during transport and deposition.
Roundness, or its opposite, angularity, imparts certain characteristics
to a sediment that controls derived properties. Orientation influences
derived properties and is strongly influenced by shape during the depo-
sition process. Clastics with similar shapes tend to take the same
orientation during deposition.
(5) Packing refers to the arrangement and closeness of grains in a
sediment. This property is strongly influenced by shape which, in turn,
strongly influences porosity, pore size distribution, and permeability.
The total appearance and relationship to one another of all structural
elements of a clastic sediment is its "fabric." Derived properties will
be discussed under processes and properties because these are of prime
importance to life in the lotic system.
-------
SEDIMENT TRANSPORT
Flowing water derived from a watershed through groundwater and surface
runoff is the transporting agent that moves clastic sediment during fluvial
processes. The nature of flow, such as seasonal distribution, quantity, and
rate of flow, depends on the local climate, geology, and stage of watershed
maturity. The only energy source of flowing water is the potential energy of
gravity as water moves from higher to lower reaches. Thus, energy to trans-
port sedimentation in a given reach is a function of its gradient. Fine
sediments are transported by suspension which is a function of turbulence or
eddies as velocities change in various reaches and cross sections of a flowing
stream. Coarse grains move as bed load by skipping, rolling, or temporary
suspension during high water and the processes of moving depends on stream
power which is a combination of downstream force and hydrodynamic lift as
water flows past large grains. The relative proportion moving by either
mechanism is primarily a function of grain size although velocity and turbu-
lence are also important.
Suspended fine materials move with about the same speed as the flowing
water whereas grains moving along the bed move at a slower rate and may be
stationary at times, moving as pulses. Engineers and hydro!ogists have devel-
oped equations to estimate sediment discharge for both transporting modes but
all are qualitative, especially those for the coarse sediments moving on the
bed. Sediment discharge is also a function of available supply and, for fine
grains, the supply is usually much less than the stream can transport. On the
other hand, the supply of coarse grains is usually greater than the stream can
transport, hence they accumulate as gravel deposits. Velocities in a stream
are not uniform through the vertical water column but are highest at the water
surface and decrease to near zero at the bed surface, hence, the ability to
move grains is lowest on the stream bed. Competence of a stream is its abil-
ity to move the largest grain in a particular reach, for large grains can only
occur during flood or other times of high discharge.
Clastic sediments are coarser in the upper reaches of a stream, even
though its competence is higher because of steeper gradients, because most of
the sands and finer grains have been flushed out to reaches where gradients
are lower. As flood stage decreases most of the coarse material lags behind
as finer material passes through a given reach and thus is called lag gravel
-------
which generally contains very small amounts of sand and fines. As gradients
decrease downstream, clastic sediments decrease in grain size primarily be-
cause of hydraulic sorting but also as a result of continued grinding action
of large grains as they tumble and roll during transport.
THEORETICAL PROPERTIES OF CLASTIC SEDIMENTS WITH SPHERICAL GRAINS
Discussion of the two derived properties of clastic sediments, porosity
and permeability, was deferred to examine some relevant work demonstrating how
they are affected by the five fundamental properties described in the previous
section. Since these two properties determine the quantity of fluid in a
deposit, and how it will yield on pumping, petroleum engineers and groundwater
hydro!ogists have a vast interest in them. Ceramic and highway engineers use
this information to obtain maximum compaction with minimum porosity and, by
using spheres as an ideal standard, have developed appropriate criteria.
These criteria also are useful to lotic biologists. Porosity is that percept
of a volume of sediment that is not occupied by solid grains and which, when
under water, is filled with water or air and water.
Spheres of uniform size can be arranged in space by six packing arrange-
ments, ranging from cubic with most open packing (porosity of 47.6%), to the
rhombohedral with closest packing (porosity of 26.0%). Thus, merely by ar-
rangement in space, porosity can be altered by 21.6% without changing grain
shape or size. Theoretically, the same porosity is possible, with the same
systematic packing, regardless of the size of grain, assuming all are the same
size. Although the grain shape is uniform, the shape of pores is highly
irregular, from wide interstitial spaces to very thin channels or throats, but
is repetitive in space for a given packing. For closest packing of a given
grain size, it is possible to calculate grain sizes that will just fill inter-
stitial pores to introduce secondary, tertiary, quaternary, and quinary grains
that fill voids to produce a total pore space of only 14.9%. It is such
calculations that permit engineers to select optimum grain-size distribution
to give minimum porosity with maximum strength of material.
Using spheres, these interstitial grain sizes are given as ratios of the
radius of the primary or framework grain. The sizes to just fill voids with-
out forcing primary grains apart is called the "critical ratio of occupancy."
Thus in any clastic sediment with known size of primary grain, a grain size to
-------
fit the voids can be estimated. Another size, called "critical ratio of en-
trance," permits a determination of grain size that can enter interstitial
spaces without disturbing the packing. Such calculations permit estimating
what size of clastic grain may freely enter a deposit if the size of primary
grain is known. Adding interstitial grains without disturbing the primary
grains always decreases porosity. Although in systematic packing of spheres,
total porosity is independent of grain size, in random packing (such as occurs
in clastic sediments) total porosity always increases with decrease in grain
size because of bridging, but individual spaces decrease in size. This rela-
tionship has a significant influence on biological "pore space efficiency" and
on corollary permeability which is the property that defines intragravel flow.
Unlike porosity, which is a static property dependent on size and ar-
rangement of grains, permeability is a dynamic property dependent, in addition
to size and arrangement of grains, on potential force measured by hydraulic
head. Permeability may be defined as the ease with which a fluid moves
through a porous material, in this case clastic sediments. It has been found
experimentally that for a given cross section of material, the rate of flow is
directly proportional to the hydraulic head in the direction of flow and
inversely proportional to the viscosity of the fluid. Thus, in stream depos-
its, intergravel flow (permeability) is directly dependent on stream gradient,
the only energy source available to move the fluid. Sediments with large pore
spaces, i.e., coarse grained, have high permeabilities under a given head. As
pore size decreases, by having interstitial spaces filled with smaller grains
or an overall decrease in primary grain size, permeability decreases. Shape
and size of pores control permeability because it is the minimum size of pore
that determines the rate. A pore that varies in shape and size is controlled
in its rate by the smallest cross section of pore.
FLUVIAL PROCESSES IN SEDIMENT FORMATION
Obviously, stream gravels are not spheres nor are they of uniform size.
However, even with these limitations, concepts of porosity, pore shape, and
how these affect the really basic parameter, permeability, are useful to
biologists working in natural habitats. Many of the fluvial processes, acting
on clastic sediments, form structural units that have properties similar to
those described for the ideal case.
10
-------
Although stream gravels can never reach the ideal as to shape (spheres)
and sorting (one size class), processes in the stream cause elastics to trend
toward ideal shapes with properties that are predictable enough to be useful
to aquatic biologists. Size of the elastics ranges from coarse (e.g. boul-
ders) in the headwaters where competence is high to finer sand and silt as
stream order increases in lower reaches with lower gradients and velocities to
decrease competence. A given large grain becomes rounded in a few miles of
downstream travel but its size tends to be rather stable. Few streams have
competence enough to move boulders through reaches with low gradients.
Smaller grains are flushed out by flow that cannot move the largest grains and
are distributed downstream by hydraulic sorting. A perfectly sorted sediment
is composed of grains of one size and is extremely rare in nature; most depos-
its contain several size classes. Upper reaches contain the coarsest grains
because most fines are flushed out by hydraulic sorting in these steep
reaches. Intermediate reaches frequently have a mixture of grains too large
for the normal competence of that reach; many times these are added to the
master stream from tributaries with higher competence. Fines contained in
gravels are usually added after coarser grains are deposited as flow decreases
during hydrograph recession. The higher the degree of sorting for a given
size range, the higher the permeability.
Shape and roundness affect permeability because certain shapes tend to be
oriented differently than others and roundness tends to cause fewer eddies as
water flows past individual grains. Oblate grains tend to be deposited with
the long axis parallel to the flow and, if one end is blunt, the larger end is
upstream because this orientation offers the least resistance to flow. Flat
grains tend to take an imbricated or shingled orientation with their flat
surface facing upstream, dipping into the current at right angles to flow
direction. The downstream edges of these grains are higher than upstream
edges. When viewing imbricated deposits from upstream all grains appear flat
with the upstream edge lower than that downstream; as viewed from downstream
one is looking at the upper edges of individual grains. Deposits with this
orientation indicate the direction of flow at the time of deposition. Permea-
bility of such deposits depends on the direction of flow with maximum flow
parallel to the bedding and minimum flow across the flat surface. Highly
angular (low roundness) grains tend to collect fines at lower flows but not at
11
-------
higher flow because of turbulence caused by sharp angles. Such grains also
tend to interlock with adjacent grains and resist further movement.
Packing of clastic sediments in a river is far more complex and diverse
than for the ideal case because of the wide range in grain size and conditions
under which deposition occurs. One physical fact important in the study of
this phase of river phenomena is that most grains, when immersed in water,
weigh about 38% less than their weight in air. Density of most rocks and
rock-forming minerals is near 2.65; since water is close to 1 at all tempera-
tures, immersion allows such material to be moved with less energy than if
dry. Such a reduction in immersed density results in a more open packing
arrangement, even with uniform shape and size of grains, because gravity is
less effective as a packing force.
With an ample supply of elastics of all sizes, reaches with various
competencies will have gravel sizes that reflect the competence of each
individual reach. As river competence decreases, larger grains will drop out
and, if in ample supply, will form a more or less random packing arrangement
that constitutes the basic framework of the deposit for that reach. In
nature, packing is somewhere between cubic (very unstable) and closest packing
which is most stable. Smaller grains continue downstream until they encounter
a reach with low velocities where they drop out. If a supply of smaller
grains, upstream from a coarse deposit, continues to be available, many of the
void spaces of the coarse framework will be filled with these smaller grains
if they are not too large to enter the structure. Thus, the natural hydraulic
sorting action forms a framework of coarse grains with subsequent filling of
voids as competence decreases. It is at this second stage that the "critical
ratio of entrance" may become important; if the size of these secondary grains
exceeds the pore space available, they cannot enter the framework, and must be
carried further downstream or be deposited on top. Thus, many voids of the
primary framework may remain unfilled, resulting in higher permeability for a
given grain size. Since riffles fit these conditions, high competence with a
rapid decrease as the flowing water enters a pool, they tend to have random,
open packing with maximum permeability. Because hydraulic sorting is contin-
uously operating, elastics initially tend to be deposited with a narrow range
of grain size; finer materials enter later to form a matrix which decreases
permeabi1ity.
12
-------
Biologists frequently refer to a given grain size as being critical to
emergence or survival of young fish. This statement is only true for a given
size of primary grain with whatever packing is caused by shape and grain
distribution. Using the reasoning developed in this paper, based on findings
from related disciplines, other grain sizes may offer similar restrictions,
depending on the size of primary grains. Sizes of critical space needed can
be estimated for a given framework of known grain size distribution and
whether the "critical ratio of entrance" allows these sizes to enter the
structure. Thus, the filling of "living room or space" by one size of grain
without considering the entire deposit, or different deposits with another
grain size distribution, may unduly restrict options or alternatives in the
management of these resources. The coarsest gravel in which a given species
of salmonid can prepare a redd, bearing in mind the reduced density of natural
grains when immersed, offers the optimum permeability to give adequate reaera-
tion, provide living space, and have sufficient pore size for emergence. A
coarse deposit can tolerate a higher percentage of the "so called" critical
size than a finer grained deposit and still provide living space. Each de-
posit has its own limiting grain size depending on the grain size distribution
from the primary framework through other class sizes.
PHYSICAL PARAMETERS AS BIOLOGICAL INDICES
The preceding section dealt exclusively with physical processes in the
lotic environment. This section will describe some physical measurements and
observations that should assist aquatic biologists in evaluating stream qual-
ity. It was suggested at a sedimentation workshop—held at Seattle, WA in
March 1977—during a discussion of nonpoint sources of sediment, that stream
bed composition be used as criterion of stream quality; the exact measurements
to make and interpretations of the data were not explained. A simple tech-
nique to classify stream beds has been used by aquatic biologists at the
University of Idaho. The method consists of four elements from which an index
number is derived to classify the bed surface as it appears to an observer.
The method is simple, rapid, and, from personal communications with those who
have used it, surprisingly accurate when applied by an experienced observer.
Stream beds deemed desirable for aquatic life using this method have also been
13
-------
observed to reflect the underlying sediments so its utility extends to some
depth in assessing habitat conditions.
The four elements observed in using this tool consist of estimating the
size of the predominant material, the next dominant material, material sur-
rounding the dominant grains, and the degree of embeddedness of each of the
predominant grains. Size of grains is divided into several classes each of
which is assigned a number. The same process is followed to determine degree
of embeddedness. The sum of these numbers constitutes a suitability index
of the observed site as a habitat for aquatic life. An experienced observer
can classify a site in several minutes, hence, several sites on transects
across a stream or longitudinal traverses in a stream can be quickly assessed
to obtain reliable samples of a given reach.
Two additional observations are suggested when using this subjective
tool. One of these is shape of grain since shape is important as a factor in
intragravel flow and also gives some indication of source of sediment. The
second is roundness, a factor in intragravel flow and stability of sediments
in place. It also gives some indication of sediment source. By using these
six parameters when observing a stream bed, an experienced biologist can
quickly gain considerable insight on classifying a stream for its potential to
produce fish or fish food. If the water is clear, this method can be used to
a depth of two to three feet without additional equipment. However, because
this technique is subjective, relying on individual judgment, it alone cannot
form the basis for legally accepted habitat criteria, regulation of watershed
management practices, or scientific communication via the literature.
More quantitative evaluation of stream quality requires gravel sampling
by more sophisticated methods than visual observation. To get a quantitative
estimate of stream bed quality, a sample of the stream bed must be physically
removed and analyzed for grain size distribution. It is at this stage that
some differences of opinion may arise on what constitutes a representative
sample for the parameter or variable being tested. Statisticians hold that
sampling should be completely random. This requires an unrealistic number of
samples to meet the criterion for a representative sample. Some compromise
must be reached between the statistical significance of a sample and physical
significance of a relatively few samples whose numbers are determined by phys-
ical constraints on sample collection and analysis. The following discussion
14
-------
will point out that completely random sampling is unrealistic and unnecessary,
except in a restricted sense.
A fundamental point must be made here: the distribution of sediment
grains in a stream is not random but is the result of cause and effect, as
discussed previously. Therefore, reliable sampling of a stream depends more
on the investigator's knowledge of the sampling objective and the fluvial
mechanisms of grain movement, deposition, and entrainment than on absolute
randomization to obtain a representative sample. These concepts become es-
pecially important when sampling low-order streams (1 through 3) that are
narrow, steep, and have a very coarse substrate. These low-order streams are
of prime importance when dealing with nonpoint sources of stream degradation
because it is on these headwater basins that best management practices can
control erosion and prevent sedimentation of higher order streams with their
greater biological productivity.
When assessing the need for sampling these streams, some judgment is
required to decide what and how intense an effort is needed to meet the stated
objective. It may be that bed sampling is not necessary; visual evaluation
may be sufficient. On low-order streams, it is impossible to collect a sta-
tistically sound sample without disrupting a significant part of the stream.
The alternative is to select sampling sites that correlate with the biological
objectives of a study, accept the samples as representative, replicated if
possible, and make as careful an analysis as is justified. Usually the ana-
lytical portion of the process is the most accurate part of the operation.
With these headwater streams, it is better to systematically sample a site by
collecting several smaller samples, such as with a freeze-core method, than
one or two larger samples that disrupt a major portion of a given site in the
small streams.
To meet the objectives of a given study, sampling site selection and
determining the number of samples needed to represent the site requires the
expert judgment of those who fully understand stream mechanics. During the
sedimentation workshop at Seattle, the point was made that considerable varia-
tion among gravel samples can be eliminated by sampling at the same environ-
mental site by persons familiar with given stream systems. In other words, do
not mix samples from different environmental sites (pools - riffles - runs,
etc.) because the physical processes at each are different. Stream bed compo-
15
-------
sition at a cross section is a continuous variable resulting from hydraulic
events that form a continuous series in terms of both time and spacial effect.
Stratified sampling is one technique for dealing with heterogeneous natural
materials such as stream gravels. By selecting strata that are fairly homo-
geneous from a heterogeneous population, greater precision is accomplished
over simple random sampling and results can be treated statistically.
Statistical analyses are useful in evaluating results of a samplng pro-
gram. However, it must be remembered that the ultimate objective is not to
achieve maximum accuracy in a sampling program, but make it "sufficiently
accurate" to meet the stated objectives. It can well be that certain varia-
bilities are tolerable because the errors do not skew data or alter decisions
based on data interpretations. Assuming that good judgment is used when
selecting a sampling site, statistics is a valuable tool in establishing
variability among individual samples, detecting errors in the analytical
procedures, and providing some measure of whether results from one site can be
extrapolated to a similar site with a predicted degree of reliability.
Once sites have been selected and samples collected, the next step in
achieving credible results is to analyze the samples using the best standard-
ized procedures. Dry sieving is recommended for classifying stream gravels
using size classes based on the Udden-Wentworth scale (Table 1). Grains
bigger than large cobbles can be hand-picked and weighed if the sample is not
large; grains less than 0.062 mm can be reported as all material passing this
sieve, even though some finer classes will be present. Because of its chemi-
cal activity, clay as a component of the < 0.062 mm separate can be important
even when present in low percentages. However, in most low-order streams,
clay percentage is usually very low and need not be separated from silt, a
procedure requiring specialized equipment.
One of the most meaningful methods of displaying results from gravel
analyses is by a size-percent accumulation graph where the logarithm of the
size (in mm) is plotted on the "X" axis and percent accumulation by weight is
plotted on the "Y" axis. From such a graph, the median grain size of the
sample can be determined and a measure of the size distribution around the
median size calculated, i.e., sorting coefficient. Thus, with two small
numbers derived from the overall analysis the biologists can, with some prac-
16
-------
TABLE 1. SCALE SIZE FOR CLASTIC SEDIMENTS (BASED ON THE UDDEN-WENTWORTH
SCHEME). After Colby; 1963.
Class
Millimeters (mm)
boulders
large
smal 1
cobbles
cobbles
gravej
very coarse gravel
sand
silt
clay
coarse gravel
medium gravel
fine gravel
very fine gravel
very coarse sand
coarse sand
medium sand
fine sand
very fine sand
256
128
64
32
16
8
4
2.00
1.00
0.50
0.250
0. 125
> 256
- 128
- 64
- 32
- 16
- 8
- 4
- 2
-1.00
- 0.50
- 0.250
- 0.125
- 0.062
< 0.062
< 0.004
Inches (in. )
>
10 -
5 -
2.5 -
1.3 -
0.6 -
0.3 -
0.16 -
0.078 -
0.039 -
0.020 -
0.0098 -
0.0049 -
<
<
10
5
2.5
1.3
0.6
0.3
0. 16
0.078
0.039
0.020
0.0098
0.0049
0.0024
0.0024
0.00015
17
-------
tice, interpret the graph in relation to the biological significance of the
substrate.
Figure 1 displays five curves from grain size analysis ranging in size
from 0.1 to > 100 mm. Median size is shown to the left of each curve and the
sorting coefficient on the right. These curves illustrate a few of the many
possible conditions that may be found in natural streams with substrate tex-
tures ranging from sandy to coarse gravels.
Starting with curve #1, it can be seen that the median size is 0.21 mm
and the sorting coefficient is 1.3 with a maximum grain size of 0.5 mm and a
small percent < 0.1 mm. This sand is very well sorted, as shown by the small
coefficient and the steepness of the curve. Biologically this sediment has a
high, uniform porosity but permeability is low because pores are uniformly
very small and stream gradients, which provide energy for flow, are low with
such streams. Thus, merely with this one curve, an experienced field scient-
ist can draw several, biologically significant, inferences about the substrate
environment and the nature of the streams in which it is found.
Curve #2 shows that texture is becoming coarser but is still dominated by
small gravel and coarse sand. Median size is 1.5 mm (coarse sand) with a
sorting coefficient of 2.0, indicating that there is a wider range in grain
sizes than for curve #1. This in turn means that many large pores contain
smaller grains that impede intragravel flow. Maximum grain size exceeds 10 mm
(0.5 in.) and very small grains (< 0.2 mm) are essentially absent. Although
this sediment is much coarser than that shown in curve #1, biologically it is
still not a desirable habitat for salmonids or other benthic life because
pores are small and permeability is low because of poor sorting. Such a
deposit also indicates moderate to low gradients which also tend to cause low
rates of intragravel flow. Comparing curves 1 and 2, it may be inferred that,
although 2 is much coarser than 1, its biological environment is not too much
different because of its poor sorting which causes many larger pores to be
filled with interstitial grains.
Curve 3 illustrates a substrate with a well-sorted, medium texture whose
median size is 14.5 mm with a coefficient of 1.1. Although the maximum size
is only 50 mm (about 2 in) its minimum grain size is 2 mm which constitutes
very little of the total sediment. Moreover, only 6% of the total is < 9 mm
in size. The biological significance of such a size distribution curve
18
-------
I/)
•(J
c
OJ
e
0)
l/l
Ol
>
s_
e
3
U
a
1/1
c
ro
s-
cu
s_
3
O)
o
o
o
CO
o
10
o o
19
-------
implies high porosity with uniform pores and high permeability because the
extensive sorting eliminates interstitial grains that might fill large pores.
Such a substrate should provide an optimum environment for all forms of
aquatic life and, because gradients are moderately high, intragravel water
flow should be nearly ideal to sustain life.
Curve 4 illustrates another set of grain-size distribution. In this
example, median grain size is only 5.2 mm even though maximum size is > 100
mm (about 4 in) because of the very poor sorting with a coefficient of 3.
Total porosity of this sediment is high but, because of the wide range in
grain size, many pores are filled with interstitial grains of diminishing size
because grains are available to fill even small pores. Although no grains
< 0.2 mm are present in this sediment, 14% of the total is 1 mm or less in
size. Such a sediment should have a moderate to good biological potential,
but permeability will only be moderate to low for the reasons given above.
Such a grain-size distribution implies moderate to high gradients and an ample
source of sediment of all grain sizes.
Curve 5 portrays the coarsest sediment of the entire group and illu-
strates a somewhat different set of environmental conditions. Although
sorting is poorer than for curve 3 (1.6 compared to 1.1), porosity, size of
pores, and permeability should provide a substrate that is biologically as
good or better than 3 because of the greater overall grain size. Only 4% of
the grains are < 4 mm in size and almost none are less than 1 mm. About 10%
of the larger grains are > 100 mm in size and the curve suggests that there
are some larger grains, probably being supplied from a different source than
the main stream. The larger grains are probably being supplied by a tributary
with a higher gradient giving it competence to transport larger grains than
can the main stream, itself a high gradient stream.
The five curves presented here represent a range of potential biological
productivity from low (curve 1) to median (curves 2 and 4) to good for curves
3 and 5. No conclusions seem final without evaluation by biologists to attain
a stated objective. Since it appears that potential productivity increases as
grain size increases with good sorting, it might be argued that the coarser a
sediment the higher its biological productivity. This probably is not true.
A stream substrate composed of 10 in. boulders is not as likely to be as
productive as one with smaller grains, lower gradients, and some interstitial
20
-------
grains. Some combination of median grain size, sorting coefficient, gradient
porosity, and permeability in the median range for all parameters ought to
give optimum productivity, and not extremes at either end of the spectrum for
these factors. An assessment process is suggested in the next section.
Specific yield of a sediment, defined earlier, can be used to estimate
the efficiency with which a substrate yields water under the force of gravity.
The time needed to drain a given volume of sediment from saturation to equil-
ibrium with gravity is an estimation of the permeability coefficient which
depends on the nature of the porous medium, here stream gravels. Combining
the two is a measure of "pore space efficiency." A sediment that yields a
large volume of water in a short time has a high pore space efficiency. A
high efficiency implies that pores are large, fine grains are minimal, and
permeability high. Such a sediment would provide an optimal environment for
ail forms of aquatic life, i.e. pores large enough for movement in any direc-
tion and intragravel movement of water for reaeration and removal of metabolic
wastes.
Specific yield can be quickly estimated in the field by saturating a
known volume of stream gravel, weighing it, draining the gravel from below
under gravity until it reaches a defined equilibrium, and then weighing the
drained sediment. The difference in weight is the volume yielded to gravity
from which specific yield can be calculated. If the time elapsed from start
of draining to equilibrium is noted, a measure of the permeability coefficient
is obtained. As an aquatic scientist gains experience with a range of sub-
strate textures he can soon learn to equate biological indices which, in some
fashion, measure productivity with those physical indices that control or
influence biological activity. By combining those various, relatively simple,
measurements an aquatic scientist can quickly evaluate a stream for probable
biological potential with respect to the substrate environment.
DISCUSSION
The significance of fluvial processes and physical indices to lotic
biology have been discussed and the utility of this understanding in evalu-
ating stream systems has been emphasized. During their training aquatic
biologists usually are not exposed to physical sciences that explain the
physical processes involved with stream mechanics. This deficiency becomes
21
-------
evident when reading biological reports on how physical parameters affect
aquatic life, especially grain size and grain-size distribution. Often the
terminology is different than that used by physical scientists to describe the
same process or measurements. This frequently creates confusion and illu-
strates the need for a common vocabulary. These observations apply equally to
physical scientists who must expand their knowledge of biology to meet the
common objective of both disciplines. An understanding of stream mechanics
and evaluation of stream quality can best be attained by small interdisciplin-
ary groups working toward common objectives in an atmosphere structured to
foster group excellence.
Three methods have been described for evaluating the biological signif-
icance of the physical environment. They may be used separately or in combin-
ation. For a quick survey, visual observation might be used. However, for a
more rigorous evaluation, all three methods will probably be needed, even
though this requires considerably more resources. Early assessment efforts
usually result in a more extensive sampling program than may be needed.
However, as an evaluation group, or individuals, gain experience in an area
and become familiar with the various classes of streams, relatively few sam-
ples may verify or authenticate visual observations or quick methods of meas-
uring "pore space efficiency." It is always wise to collect more than enough
samples to meet an objective rather than have gaps that can only be filled
with conjecture.
The overall objective of stream quality assessment is to provide guide-
lines for developing regulations that will implement a system of best manage-
ment practices to benefit life in a stream and those who make use of its
watershed. Although problems of stream degradation are in the field, not in
an office or laboratory, standards regulating watershed management practices
must be based on criteria that can only be acquired by field and laboratory
research. Sound criteria must have broad, universal application because every
stream cannot justify a separate study. Reliable scientific data from which
to develop criteria for best management practices must be based on the premise
that a watershed and its waters constitute an ecosystem. If credible results
are to be achieved and accepted by those who regulate and enforce standards
for watershed use, research to establish these criteria must be designed and
22
-------
conducted as an integrated effort of those with the best scientific expertise
avai Table.
ANNOTATED BIBLIOGRAPHY
This annotated bibliography replaces the usual text references and gives
selected sources of the rich literature dealing with sedimentary processes as
described by professionals in the earth sciences. Most entries are descrip-
tive with enough equations for an interested biologist to benefit from their
findings and conclusions and apply them to the stream environment.
1. Bjornn, T. C. , M. A. Brusven, M. P. Molnau, J. H. Milligan, R. Klamt, E.
Chacho, C. Schaye. 1977. Transport of Granitic Sediment ui Streams and
its Effect o_n Insects and Fish. Research Tech. Comp. Rpt. , Proj. B-036-
IDA., Univ. of Idaho, Moscow, Idaho.
This report describes the use of the embeddedness and substrate
index in relation to aquatic life and also shows that the Meyer-Peter,
Muller equation was reliable in predicting bed movement of bottom sedi-
ments derived from the Idaho batholith as verified by the Helley-Smith
sampler.
2. Blatt, Harvey, Gerald Middleton, and Raymond Murry. 1972. Origin o_f
Sedimentary Rocks. Prentice-Hall, Inc., 634 pp.
This recent text brings together in a readable manner, the current
thinking of most authorities in the field. Chapter 3, on the Sedimentary
Textures, and 4, on Sediment Movement by Fluid Flow, are especially
useful to the nonspecialist.
3. Colby, Bruce R. 1963. Fluvial Sediments - A Summary p_f Source, Transpor-
tation, Deposition, and Measurement o_f Sediment Pi scharge. U.S. Geolog-
ical Survey Bull. 1181A, 47 pp.
Although concerned primarily with sand and silt textured sediments,
this summary acts as a primer describing the theory and observations of
fluvial processes involved in stream sedimentology.
4. Fraser, H. J. 1935. Experimental Study p_f the Porosity and Permeabi 1 ity
of Clastic Sediment. Journal of Geology 43:910-1010.
23
-------
5. Graton, L. C. and H. J. Fraser. 1935. Systematic Packing of Spheres -
With Particular Relation Jto Porosity and Permeability. Journal of Geol-
ogy 47:785-909.
Although more than 40 years old, these two works are still quoted
and their findings are still valid; they present equations and discuss in
detail packing, porosity, and permeability of clastic sediments for the
ideal case.
6. Koski, K. Victor. 1975. The Survival and Fitness o_f Two Stocks of Chum
Salmon (Ohcorhynchus Keta) from Egg Deposition to Emergence _i_n a Control-
led-Stream Environment at Big Beef Creek. Ph.D. Thesis, University of
Washington, Seattle, Washington.
This thesis is a thorough study of the controls over emergence and
the effects of grain size on productivity; includes a thorough literature
review that includes nearly all published works dealing with gravel
substrates from the biological requirements; no geological evaluations
included.
7. Leopold, Luna B., M. Gordon Wolman, and John P. Miller. 1964. F1uvial
Processes i_n Geomorphology. W. H. Freeman and Co. 522 pp.
Probably the most modern treatment of fluvial processes. Chapter 6,
on Water and Sediment in Channels, and Chapter 7, Channel Form and Pro-
cess, are especially pertinent to the fishery biologist.
8. Pettijohn, F. J. 1957. Sedimentary Rocks. Harper and Brothers (2nd
Ed.) 718 pp.
This is one of the bibles of sedimentary rocks, the chapter on
texture is especially informative to the nongeologist.
9. White, H. E. and S. F. Walton. 1937. Particle Packing and Particle
Shape. Journal of Amer. Ceramic Soc. 20:155-166.
These men developed the equations to calculate secondary, tertiary,
quaternary, and quinary sized particles to fit interstitial spheres that
can be used for clastic grains of any size class.
24
------- |