-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and
Monitoring, Environmental Protection Agency, have
been grouped into five series. These five broad
categories were established to facilitate further
development and application of environmental
technology. Elimination of traditional grouping
was consciously planned to foster technology
transfer and a maximum interface in related
fields. The five series are:
1, Environmental Health Effects Research
2, Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
This report has been assigned to the ENVIRONMENTAL
PROTECTION TECHNOLOGY series. This series
describes research performed to develop and
demonstrate instrumentation, equipment and
methodology to repair or prevent environmental
degradation from point and non-point sources of
pollution. This work provides the new or improved
technology required for the control and treatment
of pollution sources to meet environmental quality
standards..
-------
EPA-R2-T2-083
October 1972
CONTROLLING THERMAL POLLUTION IN SMALL STREAMS
By
George W. Brown
Jon R. Brazier
Project 16130 FOK
Project Officer
Dr. Bruce A. Tichenor
National Environmental Research Center
Corvallis, Oregon 97330
Prepared for
OFFICE OF RESFARCH AND MONITORING
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20^60
For sale by the Superintendent of Documents, U.S. Government Printing Office
Washington, D.C. 20402 - Price $1.25
-------
EPA Review Notice
This report has been reviewed by the Environmental Protection Agency
and approved for publication. Approval does not signify that the
contents necessarily reflect the views and policies of the Environmental
Protection Agency nor does mention of trade names or commercial products
constitute endorsement or recommendation for use.
ii
-------
ABSTRACT
Buffer strips have been proposed as a method for controlling
temperature changes in streams adjacent to clear-cuttings. Nine
small mountain streams in Oregon's Coast Range and Cascade Mountains
were studied to determine the influence of buffer strips on water
temperature. Timber volume in the strip, strip width, and canopy
density perpendicular to the sun's rays were compared to the effectiveness
of the strip in controlling temperature change. This effectiveness
was not well correlated with timber volume or strip width. The density
of the canopy in the path of the sun is the most important buffer
strip characteristic for water temperature control.
A method for measuring the density of the canopy in the path of
the sun is described. The use of this method in the design of buffer
strips will provide protection for the stream and maximum harvesting
of the timber resource.
This report was submitted in fulfillment of Grant Number 16130
FOR under the sponsorship of the Office of Research and Monitoring,
Environmental Protection Agency.
iii
-------
CONTENTS
Section Page
I CONCLUSIONS 1
II RECOMMENDATIONS 3
III INTRODUCTION 5
Purpose 5
Scope 6
IV REVIEW OF LITERATURE 7
Undesirable Effects of High Water Temperatures 7
Stream Temperature and Logging 8
Canopy Density 10
V STUDY AREAS AND METHODOLOGY 13
Description of Study Sites 13
Study Methods !4
Discharge 15
Travel Time 15
Water Temperature 16
Surface Area 17
Net Heat 17
Buffer Strip Volume 17
Angular Canopy Density 17
VI ANALYSIS OF DATA 21
VII RESULTS 27
VIII DISCUSSION 41
IX ACKNOWLEDGEMENTS 49
X BIBLIOGRAPHY 51
XI APPENDIX 55
v
-------
FIGURES
Figure Page
1. The angular canopy densiometer. 19
2. A cross sectional view of the channel shapes of the 24
streams in the study: a. V-shaped channel; b. broad-
flat channel.
3. The hypothetical relationship between buffer strip 28
volume and AH.
4. The observed relationship between buffer strip volume 30
and AH.
5. The hypothetical relationship between buffer strip width 33
and AH.
I
6. The observed relationship between buffer strip width 35
and AH.
7. The hypothetical relationship between angular canopy 37
density and AH.
8. The observed relationship between angular canopy 39
density and AH.
9. A comparison of the predicted maximum and observed 42
maximum stream temperatures.
10. The relationship between buffer strip width and 43
angular canopy density.
11. The decline in importance of buffer strips for 46
temperature control with increasing stream size.
vi
-------
TABLES
Table Page
1. A comparison of the commercial volume of the buffer 29
strips and the percent of shade contributed by the
conifers.
2. A comparison of buffer strip width with the amount 34
of heat blocked by the strip (AH).
3. A comparison of the angular canopy density (ACD) of 38
the buffer strips with the amount of heat blocked by
the strips (AH).
Appendix
Table
A. A comparison of the various measured and calculated 55
parameters for the study streams.
vii
-------
SECTION I
CONCLUSIONS
1. Large temperature changes often observed after clear-cut logging
alongside small streams in the Pacific Northwest can be controlled
using buffer strips of vegetation between the clear-cut boundary
and the stream.
2. The density of the forest canopy in the path of the sun is the
most important characteristic for water temperature control.
3. The effectiveness of buffer strips for temperature control is not
well correlated with the volume of commercial timber remaining in
the strip or the width of the strip.
4. The canopy density of the strip can be planned prior to logging so
that it minimizes the temperature change resulting from logging
and also minimizes the volume of commercial timber required for
this protection. This minimum timber volume can be determined
with a tilting mirror that can be used in the stream channel.
-------
SECTION II
RECOMMENDATIONS
The results of this study indicate that the efficiency of a buffer
strip for controlling stream temperature is most closely related to the
density of the canopy shading the stream and not the volume of commercial
timber in the strip nor strip width. Maintaining natural stream
temperatures during logging requires only that vegetation actually
providing shade for the stream during the critical mid-day hours be
left alongside the stream. This vegetation can be readily identified
using the canopy densiometer and streamside vegetation survey technique
described in this report. It is recommended that such surveys be made
as part of the normal pre-logging planning procedure where timber
harvesting will be conducted adjacent to valuable streams. It is also
recommended that the angular canopy density be maintained at 80%
coverage or at the pre-logging level where the canopy density is less
than 80% before logging.
-------
SECTION III
INTRODUCTION
Purpose
Stream temperature is an important criterion for water quality. The
temperature of the stream affects its use for municipal consumption as well
as for a desirable fish habitat. High water temperatures, which often are
a result of land use activities, adversely affect the quality of the water
for both of these uses.
In recent years, there has been an increased interest in water quality
by the public. Certain segments of the public have demanded that land use
activities be conducted in such a manner as to have no effect on the quality
of the water in streams. Public pressure has been especially intense in the
Pacific Northwest where both commercial forests, the principal land use,
and high quality water are important resources.
The adverse effects of logging on stream temperature have been known
for many years. Ways of preventing these effects have also been known.
Several researchers in the early part of this century reported on the value
of strips of vegetation along streams to provide shade and to control the
water temperature (1, 27,- 30, 31, 34). In recent years, guidelines for the
protection of watersheds during logging have recommended similar strips
(15, 20, 29). The problem with these guidelines is that they tend to be
too general concerning desired properties of these buffer strips. The
closest they come to listing desirable characteristics of buffer strips is
to specify a minimum width for the strip. This type of specification usually
results in a less than optimum utilization of the timber resources by creating
larger buffer strips than are necessary for protection of the stream.
The purposes of this study are to determine which buffer strip
characteristics are important in regulating the temperature of small streams
and to develop a method for designing the minimum buffer strip that will
5
-------
provide adequate temperature control.
Buffer strips have been proposed as a solution to many water quality
problems. It has been suggested that accumulation of slash in stream channels
and destruction of channel banks can be prevented if buffer strips are left
between the logging unit and the stream. Buffer zones are usually required
along streams during the application of chemicals to forest lands. All of
these benefits are important. However, this study is limited to determining
the relationships between buffer strips and temperature control on small
forest streams.
The study did not consider the silvical aspects of managing buffer
strips. These include such considerations as individual tree removal within
the strip, complete harvesting, or regeneration of the strip. These problems
must be considered on a site by site basis according to existing environmental
and operational conditions. Problems attendant with sudden exposure to full
sunlight, insect infestations, or blowdown are also not considered. However,
these problems do not appear to be of much concern in the Pacific Northwest
(13, 24).
-------
SECTION IV
REVIEW OF LITERATURE
Undesirable Effects of High
Water Temperatures
High water temperatures can have undesirable effects on the quality of
water for fish production and for human consumption. Excessively high water
temperatures can be lethal to fish. The lethal temperature varies among
species and with the acclimation temperature of the individual fish. Brett
(4) reported on the tolerance of various species of fish to warm water
temperatures. He found the lethal limit for several species of salmon to be
in the range of 75° to 77°F. Research in the Alsea Logging-Aquatic Resources
Study revealed no significant mortality to coho salmon (Oncorhynchus kisutch)
fingerlings when the water temperature exceeded 80°F (19). This indicates
that the fish are more tolerant of high temperatures than was previously
thought. However, little is known about the effect of the temperature on
the vitality of the fish or of their ability to survive at sea and return
to spawn.
High water temperatures affect the aquatic environment by limiting the
amount of dissolved oxygen in the water. At 32°F water can hold 14.6 milligrams
of oxygen per liter at saturation. The saturation level decreases to 7.8
milligrams of oxygen per liter at 80°F (7). The latter concentration of
oxygen approaches the minimum value recommended by water quality standards.
Growth of bacteria and parasites which can cause disease and death
in fish is accelerated in warm water. Brett (4) reported on the devastation
of a run of blue-back salmon (Oncorhynchus nerica) in the Columbia River in
1941. At the time of the run, high temperatures in the river promoted the
growth of a myxobacterium, Chrbndococcus columnaris, lethal to salmon. The
bacteria infected the salmon and almost destroyed the run.
Warm water allows "trash" fish such, as dace (Rhinichthys sp.) to become
7
-------
established. These fish increase competition for the available food for
the sport and commercial fish. These warm-water fish are better adapted
for this competition, causing the cold-water fish to decline in numbers.
On a warm trout stream in Michigan, trout made up 14.3 to 19.7% of the total
population. At the same time on a nearby cold-water stream, trout comprised
93.8 to 98.7% of the population and accounted for over 99% of the total
weight (33).
The growth of algae is stimulated by warm water. The algae produce
an undesirable taste, odor, and color in the water, making it less agreeable
for human consumption. Also, algae lower the dissolved oxygen content of
the water through respiration at night and by increasing the biochemical
oxygen demand upon decomposition after death (33).
Cold, clear streams are a great asset to the Northwest. Not only are
they an inexpensive source of high-quality drinking water, but they are also
a source of income for the region from the fishing and recreation industries.
Activities which raise the water temperature in these streams lower the
quality of their water and lower their value for several uses.
Stream Temperature and Logging
The adverse effects of logging operations on stream temperatures have
been known for a long time. Several intensive studies of this relationship
have been conducted. Reports from these studies have indicated that the
water temperature can be expected to increase 6° to 28°F if the streams are
completely exposed to the sun after logging (7, 8, 14, 22, 31, 34). Changes
of these magnitudes can have significant effects upon the aquatic ecosystem.
Although the general impact of logging on stream temperatures was known,
the physical processes which caus.ed this impact were not well understood.
It was recognized that the source of the heat was solar radiation. However,
it was not known how the radiation acted to increase the water temperature.
Eschner and Larmoyeux (14) attempted to describe this processes. They
8
-------
attributed the rise in stream temperature after logging to the entry of
warm runoff water into the stream rather than to direct insolation of the
stream surface. They theorized that increased soil temperatures after
logging produced high ground-water temperatures. This hypothesis must
be rejected. The subsurface soil does not become hot enough to have an
appreciable effect on the ground-water temperature. Even if the ground-
water were heated sufficiently, there is not a large enough volume of
runoff in the summer in the Northwest to produce the temperature changes
observed within the clear-cuttings.
Energy budgets for exposed and protected streams illustrated why
streams increased in temperature after clear-cut logging. The net radiation
on exposed streams is up to five times as great as that on protected
streams and is responsible for the rise in temperature (5). It can be
concluded from this research that temperature problems on small streams
can be prevented by maintaining a forest canopy above streams to intercept
this radiation.
Water quality standards for small streams created the need for a model
that would allow foresters to predict the magnitude of the change in
temperature following exposure. A temperature prediction model was developed
by Brown (5) using energy budget techniques to assess the heat absorbed
by the exposed stream. The model can be written as:
A (Ha)
AT = —5— (C) (1)
where:: AT = the predicted temperature change in °F.
A = the surface area of the section of the stream exposed by
clear-cutting in ft.2.
H/L = the net radiation absorbed by the stream in BTU/ft.2-min.
D = the stream discharge in cubic feet per second.
C = 0.000267. This constant converts the discharge measurement
to pounds of water per minute. AT is then expressed in BTU/
pound of water which is equivalent to °F.
9
-------
A subsequent publication by Brown (6) included figures for the
calculation of HI based on net radiation assuming complete exposure of the
stream. This eliminated the need for expensive equipment necessary for
the measurement of solar radiation.
The water temperature in any stream is important for the influence
it might have on larger streams to which it is tributary. This influence
can be calculated with an equation described by Brown (6). This can
be written as:
Dt(Tt) + Dm(Tm)
Adjusted Temperature = —-r-j^ (2)
t m
where: D and D = the discharge of the tributary and main streams
t m
respectively in cubic feet per second.
T and T = the temperature of the tributary and main streams
t m
respectively in °F.
The effect of a proposed clear-cutting on the temperature of a tributary
system can be predicted with equation 1. The predicted temperature can
be used in equation 2 to determine the effect of the proposed clear-cutting
on the temperature of the main stream. This approach was used in a U. S.
Forest Service study to determine the effects of logging along the tributaries
of Steamboat Creek (9).
Methods to calculate the effect of logging on the water temperature of
a stream and of the stream system are available. Knowledge of how to prevent
these changes is also available. This study was conducted to provide a
method for designing minimum sized buffer strips for adequate stream
temperature control.
Canopy Density
Buffer strips protect streams from temperature changes by intercepting
the incoming solar radiation. The more radiation the canopy is able to
intercept, the more shade it can provide for the stream and the better able
10
-------
it is to control the stream temperature.
Not all of the buffer strip canopy provides shade for the stream. Only
that portion of the canopy in the path of the solar radiation shades the
stream. Therefore, when measuring the canopy density, it is necessary to
measure only this portion of the canopy.
Several types of instruments were considered for this measurement. The
first was the "moosehorn" (2, 17, 26). This instrument is a hand-held
device designed to provide an estimate of vertical canopy density. The
instrument is not satisfactory for measuring angular canopy density since
it has no means of angle determination nor any way to hold it steady during
measurement. The design of this instrument also makes it impractical for
use in buffer strips composed of low growing or bushy vegetation such as
salmonberry.
The second instrument considered was the hemispherical densiometer.
Two types of densiometers are available; one consists of a curved mirror
(21) and the other of a camera fitted with a "fisheye" lens (10, 12). Both
are designed to provide estimates of total canopy density. The problem with
densiometers is that they cover too much of the canopy; it is too difficult
to pick out that portion of the canopy which provides shade to the stream
during the critical part of the day. In addition, using the camera proved
to be too time consuming. Its use would necessitate two trips to the site
instead of one; one trip to take the pictures and a second to set the
boundaries for the buffer strip.
The third type of instrument considered was the photoelectric meter
(25). The photoelectric meter is designed to measure light intensity. While
this would give an indication of the amount of shade the buffer strip was
providing, it would not yield any readings helpful in determining the
boundaries of the buffer strip.
Because none of the instruments described above were acceptable for
11
-------
use in this study a new instrument was designed. This is the angular
canopy densiometer described in the following section.
12
-------
SECTION V
STUDY AREAS AND METHODOLOGY
DATA COLLECTION
Description of Study Sites
Study sites were located on nine small streams in Oregon. Three
sites, Little Rock, Reynolds, and Francis Creeks, are in the Umpqua
National Forest in the southern Cascade Mountains. Five others, Deer,
Lake, Grant, Griffith, and Savage Creeks, are in the Siuslaw National
Forest in the Coast Range. The remaining stream, Needle Branch, is on
land owned by the Georgia Pacific Corporation in the Coast Range.
The streams all flow through or adjacent to clear-cuttings. All
have a buffer strip of vegetation which isolates them from the clear-
cuttings. All are valuable for fish production and have a potentially
large temperature problem.
The main difference between the streams in the Cascades and those in
the Coast Range is the amount of vegetation growing around them; the
Cascade streams are less densely vegetated. Species common to both areas
are: Douglas-fir (Pseudotsuga menzlesil [Mirb.] Franco), western hemlock
CTsuga heterophylla [Raf.J Sarg.), western redcedar (Thuja plicata Donn),
red alder (Alnus rubra Bong.), bigleaf maple (Acer macrophyllum Pursh),
and vine maple (Acer circinatum Pursh). The Coast Range streams are also
bordered by salmonberry (Rubus spectabilis Pursh), and Northwest nettle
(Urtica gracilis Ait.).
The shade for all of the streams, except for Savage Creek, is provided
by the buffer strips; there is no topographic shading. Savage Greek
receives its shade from the uncut side of the stream. For this reason,
Savage Creek was not included in the analysis.
Four of the streams were divided into two stretches because of
13
-------
inconsistencies in either the buffer strip or the stream itself. Reynolds
Creek was divided at the junction of two distinct types of buffer strips.
The strip on the upper stretch was composed almost entirely of scattered
clumps of alder. The other contained Douglas-fir, western hemlock, and
western redcedar in a continuous strip.
Deer Creek was divided on the basis of both strip and stream characteristics,
The upper stretch flows through a broad, flat meadow and is buffered by
alder. The lower stretch flows through a steep, V-shaped trough. The
principal species in the lower buffer strip is Douglas-fir.
The two stretches of Grant Creek are separated by a large beaver pond.
It was not possible to trace the path of the water across the pond. This
prevented calculation of the surface area for one long stretch for use in
equation 1. Therefore, only stretches above and below the pond were used
in the study.
Needle Branch was divided into two 1000-foot sections. This was
done in an attempt to obtain an accurate predicted temperature change
with equation 1. Examination of the data later revealed a significant
inflow of ground-water into the lower stretch. This inflow altered the
actual temperature change across the section and had an adverse effect
on the analysis of the data.
A more complete description of the study sites is contained in the
appendix.
Study Methods
The study methods and instruments were chosen on the basis of
availability to the forester. Every instrument is either easily found or
constructed; all charts and tables are readily available in published form.
14
-------
Discharge
Discharge measurements, in cubic feet per second, are necessary to
obtain temperature predictions using equation 1. Discharge is a measurement
of the volume of water affected by the net energy flux.
Three methods were used to obtain discharge of the streams. On Needle
Branch and Deer Creek broad-crested, V-notched weirs maintained by the
U. S. Geological Survey were used. Stage and corresponding discharge
readings were obtained in the gage house. In both cases, the readings
were assumed to be accurate. The influence of tributary streams and
ground-water inflow was neglected. This is not a serious omission since
the tributary streams were dry and the ground-water inflow was minimal.
The low discharge on Needle Branch enabled the ground-water to have a
significant effect on the temperature of the stream, but its effect on the
discharge was not measurable.
Discharges of the remaining streams, with the exception of Francis
Creek, were measured with a Gurley #625 pygmy current meter. Suitable
sites were chosen so that an adequate number of velocity readings could
be taken across the stream and so that no water could pass the meter
unmeasured.
Francis Creek posed a special problem for the measurement of discharge.
It was too small to allow the use of the current meter and there was no
weir on the stream. Discharge was obtained by measuring the time it took
the water, as it exited a culvert, to fill a container of known volume.
The measurement was then converted into cubic feet per second.
Travel Time
Travel time is an important component of both the predicted and observed
temperature change. Travel time, as used in this study, is the amount of
time a parcel of water is exposed to solar radiation. It is used with Figure
2 in Brown (6) to obtain the solar radiation absorbed by the stream. This
15
-------
is the value H^ used in equation 1 to predict temperature change.
Travel time was measured by using DuPont Rhodamine B dye. This dye
was mixed with acetic acid to a specific gravity of 1.05. The dye was
introduced into the stream as it entered the clear-cutting and its
movement through the study reach was timed.
Francis Creek again presented a problem. The stream flowed underground
in several places in the lower stretch and the dye was filtered out. The
travel time of this stretch was calculated as a proportion of that of the
upper stretch.
Water Temperature
The actual change in stream temperature as it flowed through the clear-
cutting was measured for comparison to the predicted temperature change and
for use in later analyses. The water temperature at the top of the reach
was used for the equilibrium temperature prediction, as discussed in a
later section of this report.
Two methods were used for temperature measurement. Continuous
measurements were made using Partlow model TR-112KL thermographs. The
sensor for these instruments is a mercury-filled bulb. This bulb is
connected to a movable arm by a mercury-filled capillary. A pen attached to
the arm records the temperature on a chart connected to a clock. The chart
is graduated in 1°F increments, allowing for interpolations to the nearest
half-degree. The temperature change observed in a clear-cutting was
calculated as the difference between the temperatures recorded on the
thermographs at the top of the reach and at the bottom of the reach based
on the travel time between them.
The second method made use of a hand-held thermometer which, also
allowed for interpolation to the nearest half-degree Fahrenheit. Readings
were taken at the points where the stream entered and exited the clear-cutting
16
-------
according to the travel time across the logged unit.
Surface Area
Surface area, as used here, is the amount of the stream exposed to the
incoming solar radiation and is the value A used in equation 1. Surface area
is the product of the length of the study stretch and the average width of
the stretch. Measurements of stream width were taken every 15 feet in order
to account for variation encountered in the study streams.
Other temperature studies now in progress have indicated that only the
flowing portion of the pools should be included in the estimation of surface
area. This portion of a pool was obtained by measuring the path of dye
as it moved through the pool.
Net Heat
No instrumentation was used to obtain these data. Net heat was
obtained from charts and tables. The heat absorbed by the stream was
obtained from Brown (6) using maximum solar angles from List (23) and travel
times measured for each study stretch.
Buffer Strip Volume
Timber volume estimates on Reynolds and Lake Creeks were made from
diameter and height data taken with a diameter tape and an Abney level.
The data were converted into commercial timber volume with the aid of
tables found in Forbes (16).
The volume for the remaining strips were supplied through the courtesy
of the U. S. Forest Service and the Bureau of Land Management.
Angular Canopy Density
The concept of angular canopy density, ACD, was conceived as a method of
17
-------
measuring that portion of the buffer strip canopy that actually shades the
stream. It is a measure of the percent of the sky between the stream and
the sun covered by the canopy at solar noon.
Since none of the instruments described in the literature were satisfactory
for the type of measurement required, a new instrument, the angular canopy
densiometer, was designed (Figure 1). This instrument consists of a one-foot
square plane mirror marked with a three inch grid. The unique feature of the
instrument is that it can be tilted so that the observer, looking down
vertically on the mirror, will see the canopy along a predetermined angle.
The mirror is canted to an angle equal to the complement of the maximum
angle of the sun for the period when the temperature problem is greatest.
The density provided by the instrument is a measure of the shading ability
of the buffer strip. In addition, the forester can see which trees are
providing the shade for the stream and make provisions for leaving only
these trees or shrubs in the buffer strip.
The angular canopy densiometer was placed in the stream channel at
100-foot intervals. It was pointed south, leveled, and tilted to the
proper angle. The ACD was determined by counting the nuinber of squares
and fractions of squares covered by th.e canopy. This was converted into
percent. The type of vegetation providing the shade was also recorded
for future reference.
1.8
-------
... •" *•
PI
<*'•*-.
Figure 1. The angular canopy densiometer.
19
-------
SECTION VI
ANALYSIS OF DATA
A method for evaluating the effectiveness of the buffer strip was considered
first in the analysis of data. Several attempts were made to describe the
effectiveness of the buffer strip. Most of these were unsuccessful.
Stream temperature alone could not be used because of the effects of
clear-cuttings upstream from the study sites. On several streams, these
clear-cuttings raised the stream temperatures above acceptable levels before
they reached the study areas.
The difference between the observed temperature change and that predicted
by assuming complete exposure also proved to be of little value. It is
not possible to determine from the difference alone whether a particular
buffer strip is effective or not in controlling the water temperature.
For instance, a 3°F difference has little meaning unless it is referenced
to some specific temperature change. It could be the difference between
an observed temperature change of 1°F and a predicted temperature change
of 4°F as well as the difference between an observed change of 11°F and
a predicted change of 14°F. The two situations are obviously different,
but they would appear to be the same with regression analysis.
The same problems occur when buffer strip efficiency is used. The
efficiency is calculated as follows:
T - T
Ef = -Ej X 100% (3)
P
where: E. = the buffer strip efficiency in percent.
T = the predicted temperature change in °F.
P
T. = the observed temperature change in °F.
A
The efficiency has little significance unless it is referenced to the
stream temperature. Regression analysis cannot distinguish between two 50%
21
-------
efficiencies even though one may occur when the predicted change is 2°F and
the observed change is 1°F and the other when the predicted change is 20°F
and the observed change is 10°F. As in the example above, the two are clearly
different conditions physically, but not statistically.
Both the difference between the observed and predicted temperature changes
and the buffer strip efficiency could be used if all of the streams were
equal in size. Then the predicted temperature change would be approximately
equal for all streams. The differences in the observed temperature changes
along the streams would then be the result of buffer strip differences.
Attempts to equate the streams or standardize according to size, met with
little success. Length, width, and discharge of each stream would have to
be standardized to some common unit. This would require the development of
a complicated set of constants which would create an artificial system not
well adapted for use by field personnel. This was not done for this reason
and because the constants developed would apply only to this set of streams.
The factor decided upon as a measure of buffer strip effectiveness was
the amount of incoming radiation which the buffer strip prevents the stream
from absorbing. This factor was calculated in two ways. One was to use a
rearrangement of equation 1 so that:
(T - T )D
where: AH = the heat blocked by the buffer strip in BTU/f t .2-min.
Inadequacies in equation 1 prevented the use of equation 4 for all
of the streams in the study. For extremely small streams, such as Francis
Creek, equation 1 yields unreasonably high values. Predicted temperature
changes ranging from 33° to 626°F were calculated for these streams. These
same limitations caused equation 4 to yield extremely low values for AH.
It was necessary to find a method for calculating AH which was independent
22
-------
of stream discharge. The method chosen used the equilibrium stream temperature
calculated from equations developed by Brady, Graves, and Geyer (3) for
predicting equilibrium temperatures in cooling ponds for thermal power
plants. The equations are:
T + T,
T = JL__! (5)
B = 0.255 - 0.0085T + 0.000204 T2 (6)
f(U) = 70+0.7 U2 (7)
K = 15.7 + (B + 0.26) f(U) (8)
Et - Td + IT (9)
where: T = the average between the surface temperature at the beginning
of the reach, T , and the dewpoint temperature, T , in °F.
s a
B = the slope of the saturated vapor pressure curve between T
S
and T, in mm Hg/°F.
U = the wind speed in mi./hr.
f(U) = the evaporative wind speed function in BTU/ft.2-day-mmHg.
H = the gross solar radiation in BTU/ft.2-day.
K = the heat exchange coefficient in BTU/ft.2-day-°F.
E = the equilibrium temperature in °F.
Since the time period for calculation was considerably less than a day,
the gross solar radiation, H , was adjusted so that it was in terms of
S
BTU/ft.2 during the period of time required for the stream to flow through
the clearcut. Calculations using this method are in the appendix.
The equilibrium temperature obtained above was then used to obtain
the heat blocked by the buffer strip. The heat blocked by the buffer
strip can be computed as a percentage of the heat received. This percentage
is the difference between the expected and observed temperature change
23
-------
Figure 2. A cross sectional view of the channel shapes of the streams
in the study: a. V-shaped channel; b. broad, flat channel.
24
-------
divided by the expected change. The equation can be written as:
[(E - T ) - T ] [H-J
AH = 1 ~ (10)
Et " Ts
where the quantity E - T represents the hypothetical or expected
L S
temperature change based on the concept of equilibrium temperature.
There are two physiographic problems associated with estimating AH.
One is the result of ground-water inflow into the stream. The cool
ground-water reduces the actual temperature change across the study
stretch, giving too large an estimate of the effectiveness of the buffer
strip. This is a problem with the lower reaches of Needle Branch and
Francis Creek. For this reason, these two study reaches were not used
in any of the regression analyses performed in the study.
The second problem stems from the shape of the terrain surrounding
the stream. The streams in this study flow through two types of terrain.
Most flow through V-shaped troughs. This channel shape makes the forest
canopy very effective by aligning more of the canopy in the path of the
incoming solar radiation (Figure 2a).
Two study reaches, Little Rock Creek and the upper stretch of Deer
Creek, flow through broad, flat areas. This land form often limits the
effectiveness of the buffer strip. The canopies in the buffer strips of
Little Rock Creek and upper Deer Creek are two-layered. The gap between
the layers often permits radiation to strike the stream surface directly
(Figure 2b). Also, there is often more of the sky visible directly above
the stream in the flat channel, providing a larger source of diffuse
radiation. These two factors combine to increase the net radiation
available at the stream surface, causing an increase in the temperature
change across the study reach. This results in a lower calculated value
for AH on Little Rock and upper Deer Creeks than on the other streams.
These two streams appear to be samples from a different population of
conditions. However, the two streams do not provide a large enough sample
25
-------
to test this hypothesis. All of the figures showing the data have curves
calculated without these two streams. The two streams appear on the
figures, however, to illustrate these anomalies.
The heat blocked by the buffer strip was compared to the measured
buffer strip characteristics. The methods of comparison were both linear
and non-linear regression analyses. The analyses were not intended to
develop predictive equations. Rather, they were designed to show which
of the buffer strip characteristics had the greatest effect on AH. There
are non-linear, hypothetical models presented along with the actual data.
These hypothetical models are based upon ideas developed during the
course of this study.
26
-------
SECTION VII
RESULTS
The influence of buffer strips on temperature control was determined
by comparing the heat flow blocked by the strip to the volume of commercial
timber within the strip, average strip width, and angular canopy density.
The results are presented in three sections, one for each of the buffer
strip characteristics studied.
Buffer Strip Volume
One would expect almost no relationship between the volume of the
commercial timber in the buffer strip and the ability of the strip to
control the temperature of small streams. Dense vegetation with no
commercial volume, such as alder or salmonberry thickets, can shade some
streams as well or better than a strip which has a large commercial volume.
Volume therefore provides no assurance of an effective buffer strip.
This expected lack of relationship can be shown by a hypothetical
plotting of AH against buffer strip volume (Figure 3). Line A represents
some maximum amount of heat the buffer strip can prevent the stream from
absorbing. This is the maximum value determined by subtracting the maximum
value for net radiation under a closed canopy over a stream from the
maximum net radiation above a fully exposed stream. Line A is horizontal
since a dense strip with no commercial volume may provide the maximum
protection by completely shading the stream. In contrast to this possibility
are buffer strips with several thousand board-feet of timber which offer no
protection. In these strips, the trees may be too widely spaced to
provide an adequate amount of shade or they may be ineffective because
of the aspect of the stream or because of their location with respect to
the stream. The latter condition is the reason why the buffer strip
on Savage Creek is ineffective, even though it contains almost 200,000
board-feet of timber. It is on the east side of the stream. There is,
however, some volume at which a buffer strip has a positive effect on the
27
-------
3
OJ
m
#*
x
B
BUFFER STRIP VOLUME, BOARD-FEET
Figure 3. The hypothet±calrelationship between buffer strip
volume and A H.
28
-------
Table 1. A comparison of the commercial volume of the buffer strips
in conifers and the percent of shade contributed by the
conifers.
Stream
Little Rock
Lower Reynolds
Upper Francis
Lower Francis
Lower Deer
Upper Grant
Lower Grant
Griffith
Savage
Commercial
volume in
conifers
(Bd.-Ft.)
75.000
25,118
187,885
55,145
138,830
36,073
36,073
411,625
194,980
Shade
contributed
by conifers
(%)
87.5
33.0
79.2
83.3
25.0
10.0
10.0
74.2
0.0
The other buffer strips were composed entirely of hardwood and
brushy species of vegetation.
29
-------
z 3
5
CM
U.
\2
m
0
AH = 0.69+ 0.4 log VOLUME
_R2 = 0.3627 D
D INCLUDED
O OMITTED
I
I
10'
I02
10'
BUFFER STRIP VOLUME, BOARD-FEET
Figure 4. The observed relationship between buffer strip
volume and A H.
30
-------
water temperature, assuming that the strip is positioned to shade the
stream. At this point, the trees in the strip are close enough together
to prevent further increases in the water temperature. This situation
is represented by line B.
A listing of the streams which have buffer strips containing some
commercial volume in conifers and the percent of shade provided by these
conifers is presented in Table 1. Buffer strips composed exclusively of
hardwoods are not considered in this study because of the low value of
Northwest hardwoods. It can be seen from this table that the shading
provided by the commercially valuable trees in the strip varies widely.
The relationship between AH and volume found in this study is presented
in Figure 4. Superimposed is the hypothetical relationship described in
Figure 3. A linear regression analysis of the data is shown in the figure.
The R2 value, 0.3627, is very low indicating poor relationship between
the two terms. The strip on the lower stretch of Reynolds Creek illustrates
a point made earlier. It has a volume of 25,118 board-feet, yet it provides
no protection for the stream. This is because the buffer strip is not
uniform. There is almost no protection for the stream for the initial
200 feet of the stretch. Most of the shade is provided by the next
200 feet with the remaining 190 feet being open to the sun again. The
shaded portion of the stream is not large enough to prevent the stream
temperature from rising the predicted number of degrees.
Buffer Strip Width
Width is the most commonly used term in the design of buffer strips.
Guidelines for the management of watersheds frequently specify widths for
buffer strips along streams. While these guidelines recognize the need
for buffers, they do not recognize the reason buffer strips control
temperature. Strip width alone has very little to do with the ability
of the vegetation in the strip to shade the stream. Strip width is
related to the effectiveness of buffer strips through a complex inter-
31
-------
relationship of canopy density, canopy height, stream width, and stream
discharge.
On very small streams such as those included in this study, the
relationship between AH and strip width can be viewed as asympototic
in nature. The quickness with which the relationship approaches some
asymptote is a function of the type of vegetation contained within the
strip. Two curves representing a hypothetical relationship between
AH and strip width for different types of vegetation are shown in Figure
5. Curve A represents very dense, bushy vegetation. These plants do not
grow very tall; only those plants adjacent to the stream provide any
shade at all. However, their dense foliage is an excellent source of
shade. They require only a narrow space along the stream in which to
provide the maximum amount of shade. Strips wider than this narrow
section should not improve in effectiveness.
Curve B represents a buffer strip composed of trees. Trees have
canopies of lower densities than the type of vegetation described above.
More space is required to provide an equal amount of protection. This
protection cannot be achieved in an equal area because of tree size and
spacing. A wider strip is, therefore, required to attain maximum
effectiveness when the strip is composed of trees.
The width of the buffer strips on the study streams and the heat
blocked by them are presented in Table 2. Only seven streams were used
in the statistical analysis of the data. The Reynolds Creek reaches were
omitted because of the difficulty in defining strip width. Little Rock
Creek and upper Deer Creek were omitted because the shape of the surrounding
terrain influenced the heat received at the stream surface. These data
are included in Figure 6, however.
Non-linear regression analysis was used to analyze the data. The
curve was forced through the origin. The hypothetical maximum, noted
in Figure 5, was determined on the basis of a set of conditions not met
32
-------
4 -
A (SHRUBS)
B (TREES)
BUFFER STRIP WIDTH >
Figure 5. The hypothetical relationship between buffer strip
width and A H.
33
-------
Table 2. A comparison of buffer strip width with the amount of heat
blocked by the strip (AH).
o
Stream
Width
(Ft.)
AH
(BTU/ft.2-min.)
Remarks
Little Rock
Upper Reynolds
Lower Reynolds
Upper Francis
Upper Deer
47
10
40
50
100
1.4
0.0
0.0
3.8
2.0
omitted from the
analysis because
of channel shape
omitted from
analysis because
of strip width
same as upper
stretch
omitted from
analysis because
of the channel
shape
Lower Deer
Lake
Upper Grant
Lower Grant
Griffith
Upper Needle Branch
100
30
60
60
50
8
3.7
3.1
2.3
3.2
3.5
2.8
Lower Francis and Lower Needle Branch omitted from the table and
analysis because of the inflow of ground-water.
34
-------
- 3
CsJ
OD
••
I
d INCLUDED
O OMITTED
H
AH = 3.267 - 3.267 e~°-371 sw
R2 = 0.8749
/k I i /K I
20 40 60 80 100
BUFFER STRIP WIDTH, FEET
Figure 6. The observed relationship between buffer strip
width and A H.
35
-------
on all streams. This is why the curve in Figure 6 is asymptotic at a
lower level than hypothesized. However, the curve has a high R value
(0.87) indicating that it is a good approximation of the actual process.
Angular Canopy Density
Angular canopy density, ACD, is a measure of the shading ability of
the vegetation. By knowing the ACD of the buffer strip, the forester
can estimate its effectiveness in controlling water temperature. Design
of buffer strips on the basis of ACD assures adequate protection for the
stream while allowing for the maximum harvest of timber.
The hypothetical relationship between AH and ACD can be considered
logistic in nature (Figure 7). Low canopy densities, while reducing the
solar radiation incident to the stream in direct proportion to the percent
of sky covered, do not provide sufficient shade for the effect to be
measurable. Thus, the value for AH is zero until some measurement threshold
is reached. Above this value, there should be a direct, linear relationship
between AH and ACD until the canopy approaches full closure. As the
canopy density approaches 100 percent, additional increments of density
should block less radiation than the previous increment. This is because
that at high canopy densities, the possibility for reflection and absorbtion
of the incident radiation increases allowing mostly diffuse radiation
to reach the stream. The level of this diffuse radiation is controlled
by factors other than canopy density. The volume of vegetation in the
canopy influences the amount of transmission. Thicker canopies, such as
conifers provide, are more efficient radiation traps than thin canopies,
such as hardwoods provide, even though the canopy density may be the
same. Thus, with greater canopy density, the relationship between AH
and ACD should approach some asymptote at a level less than complete
blockage of incident radiation. Values of AH for undisturbed canopies
are in the range of 3.0 to 3.6 BTU/ft2min (5, 9). This corresponds
with values calculated in this study-
36
-------
i
CM
I-
00
•»
X
<
ANGULAR CANOPY DENSITY
Figure 7. The hypothetical relationship between angular
canopy density and A H.
37
-------
Table 3. A comparison of the angular canopy density (ACD) of the
buffer strips with the amount of heat blocked by the strips
(AH).
a
Stream
Little Rock
Upper Reynolds
Lower Reynolds
Upper Francis
Upper Deer
Lower Deer
Lake
Upper Grant
Lower Grant
Griffith
Upper Needle Branch
ACD
73.6
18.3
46.9
75.9
80.3
78.3
77.7
59.1
65.2
79.1
55.6
AH
(BTU/ft.2-min.)
1.4
0.0
0.0
3.8
2.0
3.7
3.1
2.3
3.2
3.5
2.8
Remarks
omitted from
analysis because
of channel shape
omitted from
analysis because
of channel shape
«a
Lower Francis and Lower Needle Branch omitted from table and analysis
because of influence of ground-water on temperature.
38
-------
AH = - 1.61 + 0.066ACD
R2 = 0.8939
D INCLUDED
O OMITTED
20 40 60 80
ANGULAR CANOPY DENSITY, %
100
Figure 8. The observed relationship between angular
canopy density and A H.
39
-------
The ACD's for the study streams are presented in Table 3. Little
Rock and upper Deer Creeks are again omitted from the analysis because
of the surrounding terrain. Problems with the computer programs prevented
fitting a logistic curve to the data. For this reason, a straight-line
approximation to the logistic curve was used (Figure 8). Segment A
represents the ACD values below the threshold level, which occurs at
about 23% with these data. This point was determined by the linear
regression analysis used for segment B. Line segment B is the section
of increasing buffer strip effectiveness with increasing ACD. The line
fits the data well with an R2 value of 0.89. Segment C is the area of
maximum protection. The maximum was determined from the net radiation
over protected and exposed streams as explained earlier. Once the maximum
protection has been reached, increases in ACD offer no greater protection.
The lower reach of Reynolds Creek is anomalous to the relation
mentioned above. Although the buffer strip completely shades 200 feet
of the reach and has an average ACD of 46.9% over the entire reach,
it is ineffective in controlling the stream temperature. The observed
temperature change (3°F) exceeds that predicted (1.6°F). However, the
two changes must be considered similar because of the accuracy of equation
1. The reach was resurveyed in an unsuccessful attempt to explain the
anomaly.
40
-------
SECTION VIII
DISCUSSION
Buffer strips are an effective means of controlling stream temperatures
The predicted maximum and observed maximum temperatures for all study
streams are presented in Figure 9. With the exception of Reynolds Creek,
all of the observed water temperatures are lower than the predicted
temperatures. Nine of the streams have predicted maximum temperatures
above 65 °F, but only four have observed maximums that high. Three of
the four, Little Rock Creek and both reaches of Reynolds Creek, had
temperatures above 65°F when they entered the study sections. The upper
reach of Needle Branch was the other stretch that exceeded a temperature
of 65°F. Its temperature rose only to 67°F. This is a full 12°F below
the predicted maximum temperature for this stream. This difference is
due primarily to the effect of the buffer strip.
It has been known for a long time that the presence of a well-designed
buffer strip will control stream temperatures. What has not been known
is which characteristics of buffer strips are important in this control.
The data indicate that the commercial timber volume contained in the
buffer strip is not an important criterion for temperature control.
The effectiveness of the strip is not well related to timber volume.
Strip width beyond some minimum value (about 30 feet in this study)
is not an important criterion for stream temperature control. The use of
a 100 to 200-foot standard width for all buffer strips will assure adequate
protection for most streams. However, this practice usually results in
more trees than necessary to provide adequate protection. The width of
each buffer strip studied and its canopy density are plotted in Figure
10. These data indicate that for the average buffer strip, the maximum
ACD, and hence the maximum shading ability, is reached within a width of
80 feet. Moreover, 90% of that maximum is reached within 55 feet.
The maximum effective width presented here differs from that in
Figure 6 for several reasons. The most important of these is that the
41
-------
80 \-
or
S70
cc
UJ
0_
UJ
60 h
50
10
—
-
rn
>
rrr
9!
5 MAX
P
V*T
IMUM TEMPERATURE c
REDICTED ED 90 ,
OBSER
m
tTfl
VED 0
rr
1
r
ll
TTT
PW
i
)2
rrr
•TT
j:
rrr
i s
5 S
£ IS s s
*
0
o
cr
LkJ
_l
(—
1-
_l
tn
w f
O
z
<
cr
u.
cr
UJ
0.
0.
3
CO
o
z
<
cr
u.
cr
Ul
$
0
_i
cr
UJ
UJ
o
cr
UJ
Q.
0.
:D
cr
Ul
UJ
0
cr
LU
£
q
_i
i-
z
cr
o
uj 2]
* Q.
< Q.
-I ID
1-
z
cr
CD
cr
UJ
£
0
_J
I
1-
u.
u.
cr
CD
UJ
cr
cr
UJ
Q_
o_
^
to
V /
o
_l
o
z
UJ
cr
cr
UJ
o
_i
Figure 9. A comparison of the predicted maximum and
observed maximum stream temperatures.
42
-------
100
80
UJ
o
60
Q.
O
40
o
20
20 40 60 80
BUFFER STRIP WIDTH, FT
100
Figure 10. The relationship between buffer strip width and
angular canopy density.
43
-------
relationship described in Figure 6 is based on the interrelationship
between buffer strip width and canopy density and height, and stream width
and discharge. The relationship in Figure 10 is based only on strip width
and angular canopy density. In addition, all of the study streams were
included in Figure 10, whereas several were omitted in Figure 6. The large
amount of variance exhibited by the buffer strips limited the effectiveness
of regression analysis in describing the relationship presented in Figure
10. For this reason, a hand-fitted curve has been used to describe the
relation between the two factors.
Angular canopy density is the only buffer strip parameter which is
strongly correlated with stream temperature control. It is the only
measure the forester can use that will assure him of providing enough
shade for the stream without overdesigning the buffer strip. ACD
incorporates the variation imposed by different vegetation types and
stream configurations. It is for this reason that the design of buffer
strips must be done site by site. In this way, vegetation providing the
maximum ACD can be preserved in each buffer strip. Design of the strip
on this basis allows for a consistent level of protection on each stream.
The normal temperature regime of each stream should not change when this
method of planning is used because the natural surroundings of the stream
which influence its temperature are not changed. This is in line with
existing water quality standards. At the minimum, the ACD should be
maintained at 80% on all fully covered streams; for streams with pre-
logging ACD less than 80%, ACD should not be reduced below the natural
condition.
When designing a buffer strip, the forester should be aware of several
other factors. He should be cognizant of the aspect of the stream and the
placement of the clear-cutting with respect to the stream. Since the sun
never shines from the north in this region, there is no need for a buffer
strip for temperature control on the north side of the stream.
There is also little need for a buffer strip on the east side of a
44
-------
stream. Stream temperatures do not peak until afternoon when the sun
shines from the southwest. The buffer strip on the east side of the
stream would offer no protection. This is the case on Savage Creek.
Streams that flow north-south present a special problem. At solar
noon the sun can shine straight up the stream. The only protection for
these streams comes from overtopping vegetation on both sides of the
stream. Side shading will have no effect until the sun is past its
zenith.
Buffer strips should vary with the size of the stream. The forester
should take advantage of all of the brushy vegetation available; in
many instances such vegetation is sufficient to provide temperature
control.
Foresters should design buffer strips with a uniform canopy density.
Large gaps, such as those on Reynolds Creek, lower the effectiveness of
the strip. The reason for this lack of effectiveness seems to be that
even with 100 percent canopy density, isolated patches of trees provide
much less than complete blockage of the incident radiation. Transmission
through the canopy and reflection from the unforested zones at each end
of the isolated strip of trees combine to reduce the heat it blocks.
Thus, the heat blocked by a buffer strip that is half completely
exposed and half completely shaded should be much less than that blocked
by a continuous canopy with 50 percent density.
Buffer strips decline in effectiveness as the streams increase in
size (Figure 11). Small streams have the greatest temperature problem.
This is because of the inverse relationship between temperature change
and discharge (equation 1). Buffer strips are very effective for the
control of water temperatures on these small streams. On larger streams,
such as Steamboat Creek, buffer strips have little effect on the water
temperature (9). One reason for this is the volume of water in the
stream. Another reason is that it is physically impossible for the
45
-------
STREAM SIZE
Figure 11. The decline in importance of buffer
strips for temperature control with
increasing stream size.
46
-------
vegetation to shade more than a small portion of the stream during the
periods of peak radiation density.
For most of the small forest streams adjacent to clear-cuttings,
temperature problems can be controlled by buffer strips. This study
has presented a method for designing buffer strips to provide the
maximum protection to the stream with the minimum amount of timber.
This information will help preserve the water quality of a stream while
allowing for the optimum utilization of the timber resource.
47
-------
SECTION IX
ACKNOWLEDGEMENTS
This research was sponsored by the Environmental Protection agency under
project number 16130 FOK, and Oregon State University. Dr. Bruce A.
Tichenor, Pacific Northwest Water Laboratory, National Environmental
Research Center, Corvallis, served as project officer.
49
-------
SECTION X
BIBLIOGRAPHY
1. fielding, D. L. Water temperature and fish life. Transactions of the
American Fisheries Society 58:98-105. 1928.
2. Bonner, G. M. Estimation of ground canopy density from ground
measurements. Journal of Forestry 65:544-547. 1967.
3. Brady, D. K., W. L. Graves, and J. C. Geyer. Surface heat exchange
at power plant cooling lakes. Publication number 69-901. Edison
Electric Institute, New York City, New York. 1969. 153 numb, leaves.
4. Brett, J. R. Some principles in the thermal requirements of fishes.
Quarterly Review of Biology 31:75-81. 1956.
5. Brown, G. W. Predicting temperatures of small streams. Water
Resources Research 5:68-75. 1969.
6. . Predicting the effect of clearcutting on stream
temperature. Journal of Soil and Water Conservation 25:11-13. 1970.
7. and J. T. Krygier. Changing water temperatures in small
mountain streams. Journal of Soil and Water Conservation 22:242-244.
1967.
8. and . Effects of clearcutting on stream
temperature. Water Resources Research 6:1133-1139. 1970.
9. , G. W. Swank, and J. Rothacher. Water temperature in
the Steamboat drainage. USDA Forest Service Research Paper PNW-
119, Pacific Northwest Forest and Range Experiment Station, Portland,
Oregon. 1971. 17 p.
51
-------
10. Brown, H. E. and D. P. Worley. Some applications of the canopy
camera in forestry. Journal of Forestry 63:674-680. 1965.
11. Chapman, D. W. Effects of logging upon fish resources of the west
coast. Journal of Forestry 60:533-537- 1962.
12. Clark, F. G. A hemispherical forest photocanopymeter. Journal of
Forestry 59:103-105. 1961.
13. DeWitt, J. W. Streamside vegetation and small coastal salmon streams.
In: Proceedings of a Forum on the Relation Between Logging and Salmon.
American Institute of Fishery Research Biology, Juneau, Alaska. 1968.
14. Eschner, A. R. and J. Larmoyeux. Logging and trout: four experimental
forest practices and their effect on water quality. Progressive Fish
Culturist 25:59-67. 1963.
15. Federal Water Pollution Control Administration. Industrial waste
guide on logging practices. U. S. Department of the Interior,
Northwest Region, Portland, Oregon. 1970. 40 p.
16. Forbes, R. D. Forestry handbook. New York, Ronald, 1955. 1153 p.
17- Garrison, G. A. Uses and modifications of the "moosehorn" crown
closure estimator. Journal of Forestry 47:733-735. 1949.
18. Greene, G. E. Land use and trout streams. Journal of Soil and Water
Conservation. 5:125-126. 1950.
19. Hall, J. D. and R. L. Lantz. Effects of logging on the habitat of
coho salmon and cutthroat trout in coastal streams. In: A symposium
on salmon and trout in streams, T. G. Northcote, ed. University of
British Columbia, 1969. 388 p.
52
-------
20. Lantz, R. L. Guidelines of stream protection in logging operations.
Oregon State Game Commission, Portland, Oregon, 1971. 29 p.
21. Lemmon, P. E. A spherical densiometer for estimating forest over-
story density. Forest Science 2:314-320. 1956.
22. Levno, A. and J. Rothacher. Increases in maximum stream temperature
after logging in old-growth Douglas-fir watersheds. USDA Forest
Service Research Note PNW-65. Pacific Northwest Forest and Range
Experiment Station, Portland, Oregon, 1967. 12 p.
23. List, R. J. Smithsonian meterological tables. Smithsonian Institution,
Washington, D. C. 1966. 527 p.
24. Marlega, R. R. U. S. Forest Service District Ranger. Personal
communication. Gardiner, Oregon. February, 1972.
26. Robinson, M. W. An instrument to measure forest crown cover. Forestry
Chronicle 23:222-225. 1947.
27. Roth, F. The fisherman and reforestation. Transactions of the
American Fisheries Society 35:164-168. 1906.
28. Sheridan, W. L., S. T. Olson, and T. C. Hoffman. Monitoring certain
land use effects on salmon spawning environment. Society of American
Foresters Proceedings, September, 1966. pp. 49-52.
29. Society of American Foresters Columbia River Section, Water
Management Committee. Recommended logging practices for watershed
protection in Oregon. Journal of Forestry 57:460-465. 1959.
30. Surber, T. Biological surveys and investigations in Minnesota.
Transactions of the American Fisheries Society 52:225-238. 1922.
53
-------
31. . Methods in restocking Minnesota lakes and streams, with
comments on the dangers attending overstocking of certain waters.
Transactions of the American Fisheries Society 55:63-71. 1925.
32. Swift, L. W. and J. B. Messer. Forest cuttings raise temperature
of small streams in the Southern Appalachians. Journal of Soil
and Water Conservation 26:111-116. 1971.
33. Tarzwell, C. M. and A. R. Gaufin. Some important biological effects
of pollution often disregarded in stream surveys. In: Proceedings
of the Eighth Industrial Waste Conference, Lafayette, Indiana, 1953.
pp. 295-316.
34. Titcomb, J. W. Forests in relation to fresh water fishes. Trans-
actions of the American Fisheries Society 56:122-129. 1926.
54
-------
SECTION XI
APPENDIX
55
-------
Table A. A comparison of various measured and calculated parameters for the study streams.
Stream Predicted
tempera-
ture
change
Little Rock
Upper Reynolds
Lower Reynolds
Upper Francis
Lower Francis
Upper Deer
Lower Deer
Lake
Upper Grant
Lower Grant
Griffith
Upper Needle Branch
Lower Needle Branch
10.1
4.8
1.6
41. 9b
35. 7b
7.6
9.0
12.5
4.4
4.3
21.7
32
32
Observed AH
tempera-
ture
change BTU/f t .2-min.
6.0
7.5
3.0
2.0
1.0
4.0
1.0
3.0
2.0
1.0
3.0
9.0
3.0
1.
0.
0.
3.
3.
2.
3.
3.
2.
3.
3.
2.
2.
4
0
0
8
9
0
7
1
3
2
5
8
8
ACD Strip
width
% Ft.
73.6
18.3
46.9
75.9
55.3
80.3
78.3
77.7
59.1
65.2
79.1
55.6
47.6
47
10
40
50
50
100
100
30
60
60
50
8
8
P-Aa
°F
4.0
2.7
1.4
39.9
34.7
3.6
7.9
9.5
2.4
3.3
18.7
23
29
Volume Observed
maximum
tempera-
Bd.-Ft. °F
75,000
0
25,118
187,885
55,145
18,745
138,830
4,392
38,245
38,245
413,260
0
0
72
74.5
71.5
62
60
57
56
61
55
55
62
67
63
Predicted
maximum
tempera-
°F
76
72
70
101
94
60
64
70
57
58
80
90
92
.9
.7
.5
.5
.5
.5
.5
£1
The predicted temperature change - the actual temperature change.
Predicted from the equilibrium temperature calculation.
-------
EQUILIBRIUM TEMPERATURE CALCULATION
T + T.
(5)
B = 0.255 - 0.0085 T + 0.000204 T2 (6)
f(U) = 70 + 0.7 U2 (7)
K = 15.7 + (B + 0.26) f(U) (8)
E - Td + T- W
For the upper stretch of Francis Creek:
Travel time = 3 hours
T . = 95°F T = 71°F
air d.
RH = 50% T = 60°F
s
U = 2 mph H =4.0 BTU/ft.2-min.
S
= 720 BTU/ft.2-3 hours
J. ry \J _/ • -/ J-
B = 0.255 - 0.0085 (65.5) + 0.000204 (65.52) = 0.573 mm Hg/°F
f(U) = 70+0.7 (22) = 9.09 BTU/ft.2-3 hours-ram Hg.
K = 15.7 + (0.573 + 0.26) (9.09) = 23.27 BTU/ft.2-3 hours-°F.
E = 71 + ||^27 = 101.9°F
Predicted temperature change = 101.9 - 60 = 41.9°F.
57
-------
SITE DESCRIPTIONS
Little Rock Creek
Aspect: east-west flow, buffer strip on south side.
Vegetation: alder, Douglas-fir, western redcedar, bigleaf maple
Travel Time: 1-1/3 hours.
Discharge: 1.16 CFS.
Stream Width: 8.7 feet.
Stream Length: 1500 feet.
Surface Area: 13050 square feet.
Strip Width: 47 feet.
Strip Volume: 75,000 bd.-ft.
Angular Canopy Density: 73.6%.
Heat Load: 4.2 BTU/ft.2-min.
Predicted Temperature Change: 10.1°F.
Actual Temperature Change: 72° - 66° = 6°F.
AH: 1.4 BTU/ft.2-min.
Reynolds Creek, upper stretch
Aspect: northwest-southeast flow, strip on the southwest side.
Vegetation: alder, western hemlock, vine maple. (Primarily
regrowth following logging.)
Travel Time: 1 1/4 hours.
Discharge: 2.51 CFS.
Stream Width: 7.3 feet.
Stream Length: 1850 feet.
Surface Area: 13505 square feet.
Strip width: 10 feet.
Strip Volume: 0 bd.-ft.
Angular Canopy Density: 18.3%
Heat Load: 4.2 BTU/ft.2-min.
Predicted Temperature Change: 4.8°F.
Actual Temperature Change: 74.5° - 67° = 7.5°F.
AH: 0 BTU/ft.2-min.
58
-------
Reynolds Creek, lower stretch
Aspect: northwest-southeast flow, strip on the southwest side.
Vegetation: alder, Douglas-fir, western redcedar, western hemlock,
vine maple.
Travel Time: 1/2 hour.
Discharge: 2.51 CFS.
Stream Width: 7.7 feet.
Stream Length: 590.feet.
Surface Area: 4543 square feet.
Strip Width: 40 feet.
Strip Volume: 25,118 bd.-ft.
Angular Canopy Density: 46.9%.
Heat Load: 4.2 BTU/ft.2-min.
Predicted Temperature Change: 1.6°F.
Actual Temperature Change: 70° - 67° = 3°F.
AH: 0 BTU/ft.2-min.
Lake Creek
Aspect: east-west flow, strip on the south side.
Vegetation: alder, bigleaf maple, salmonberry, Northwest nettle.
Travel Time: 2-1/6 hours.
Discharge: 0.71 CFS.
Stream Width: 4'.5 feet.
Stream Length: 1800 feet.
Surface Area: 8100 square feet.
Strip Width: 30 feet.
Strip Volume: 4392 square feet.
Angular Canopy Density: 77.7%.
Heat Load: 4.1 BTU/ft.2-min.
Predicted Temperature Change: 12.5°F.
Actual Temperature Change: 61° - 58° = 3°F.
AH: 3.1 BTU/ft.2-min.
59
-------
Francis Creek, upper stretch
Aspect: east-west flow, buffer strip on the south side.
Vegetation: vine maple, western redcedar, western hemlock,
Douglas-fir.
Travel Time: 3 hours.
Discharge: 0.01 CFS.
Stream Width: 1 foot.
Stream Length: 2400 feet.
Surface Area: 2400 square feet.
Strip Width: 50 feet.
Strip Volume: 187,885 bd.-ft.
Angular Canopy Density: 75.9%.
Heat Load: 4.0 BTU/ft.2-min.
Predicted Temperature Change: 41.9°F.
Actual Temperature Change: 62° - 60° = 2°F.
AH: 3.8 BTU/ft.2-min.
Francis Creek, lower stretch
Aspect: east-west flow, buffer strip on the south side.
Vegetation: vine maple, western redcedar, western hemlock,
Douglas-fir.
Travel Time: 2 hours.
Discharge: 0.01 CFS.
Stream Width: 3.1 feet.
Stream Length: 1700 feet.
Surface Area: 5720 square feet.
Strip Width: 50 feet.
Strip Volume: 55,145 bd.-ft.
Angular Canopy Density: 55.3%.
Heat Load: 4.1 BTU/ft.2-min.
Predicted Temperature Change: 35.7°F.
Actual Temperature Change: 60° - 59° = 1°F.
AH: 3.9 BTU/ft.2-min.
-------
Deer Creek, upper stretch
Aspect: north-south flow, buffer strip on the west side.
Vegetation: alder, Douglas-fir, salmonberry.
Travel Time: 2-3/4 hours.
Discharge: 0.74 CFS.
Stream Width: 3.1 feet.
Stream length: 1650 feet.
Surface Area: 5115 square feet.
Strip Width: 100 feet.
Strip Volume: 18,745 bd.-ft.
Angular Canopy Density: 80.3%.
Heat Load: 4.1 BTU/ft.2-min.
Predicted Temperature Change: 7.6°F.
Actual Temperature Change: 57° - 53° = 4°F.
AH: 2.0 BTU/ft.2-min.
Deer Creek, lower stretch
Aspect: north-south flow, buffer strip on the west side.
Vegetation: alder, Douglas-fir, salmonberry.
Travel Time: 1-1/3 hours.
Discharge: 0.74 CFS.
Stream Width: 4.4 feet.
Stream Length: 1350 feet.
Surface Area: 5940 square feet.
Strip Width: 100 feet.
Strip Volume: 138,830 bd.-ft.
Angular Canopy Density: 78.3%.
Heat Load: 4.2 BTU/ft.2-min.
Predicted Temperature Change: 9.0°F.
Actual Temperature Change: 56° - 55° = 1°F.
AH: 3.7 BTU/ft.2-min.
61
-------
Grant Creek, upper stretch
Aspect: north-south flow, buffer strip on the west side.
Vegetation: alder, Douglas-fir, salmonberry.
Travel Time: 5/6 hour.
Discharge: 1.12 CFS.
Stream Width: 4.9 feet.
Stream Length: 900 feet.
Surface Area: 4410 square feet.
Strip Width: 60 feet.
Strip Volume: 38,245 bd.-ft.
Angular Canopy Density: 59.1%.
Heat Load: 4.2 BTU/ft.2-min.
Predicted Temperature Change: 4.4°F.
Actual Temperature Change: 55° - 53° = 2°F.
AH: 2.3 BTU/ft.2-min.
Grant Creek, lower stretch
Aspect: north-south flow, buffer strip on the west side.
Vegetation: alder, Douglas-fir, salmonberry.
Travel Time: 2/3 hour.
Discharge: 1.12 CFS.
Stream Width: 4.8 feet.
Stream Length: 900 feet.
Surface Area: 4320 square feet.
Strip Width: 60 feet.
Strip Volume: 38,245 bd.-ft.
Angular Canopy Density: 65.2%.
Heat Load: 4.2 BTU/ft.2-min.
Predicted Temperature Change: 4.3°F.
Actual Temperature Change: 55° - 54° = 1°F.
AH: 3.2 BTU/ft.2-min.
62
-------
Needle Branch, upper stretch
Aspect: north-south flow, buffer strip on west side.
Vegetation: alder (regrowth after logging).
Travel Time: 3 hours.
Discharge: 0.08 CFS.
Stream Width: 2.4 feet.
Stream Length: 1000 feet.
Surface Area: 2400 square feet.
Strip Width: 8 feet.
Strip Volume: 0 bd.-ft.
Angular Canopy Density: 55.6%.
Heat Load: 4.0 BTU/ft.2-min.
Predicted Temperature Change: 32°F.
Actual Temperature Change: 67° - 58° = 9°F.
AH: 2.8 BTU/ft.2.-min.
Needle Branch, lower stretch
Aspect: north-south flow, buffer strip on west side.
Vegetation: alder (regrowth after logging).
Travel Time: 3 hours.
Discharge: 0.08 CFS.
Stream Width: 2.4 feet.
Stream Length: 1000 feet.
Surface Area: 2400 square feet.
Strip Width: 8 feet.
Strip Volume: 0 bd.-ft.
Angular Canopy Density: 47.6%.
Heat Load: 4.0 BTU/ft.2-min.
Predicted Temperature Change: 32°F.
Actual Temperature Change: 63° - 60° = 3°F.
AH: 2.8 BTU/ft.2-min.
6.3
-------
Griffith Creek
Aspect: east-west flow, buffer strip on the south side.
Vegetation: alder, Douglas-fir, western hemlock, vine maple,
bigleaf maple.
Travel Time: 2-1/3 hours.
Discharge: 1.3 CFS.
Stream Width: 8.6 feet.
Stream Length: 3000 feet.
Surface Area: 25,800 square feet.
Strip Width: 50 feet.
-Strip Volume: 413,260 bd.-ft.
Angular Canopy Density: 79.1%.
Heat Load: 4.1 BTU/ft.2-min.
Predicted Temperature Change: 21.7°F.
Actual Temperature Change: 62° - 59° = 3°F.
AH: 3.5 BTU/ft.2-min.
Savage Creek
Aspect: north-south flow, buffer strip on the east side.
Vegetation: alder, Dotxglas-fir, salmonberry.
No other information was taken on Savage Creek because the location
of the buffer strip prevented it from having any effect on the
temperature of the stream.
64
-------
SELECTED WATER
RESOURCES ABSTRACTS
INPUT TRANSACTION FORM
/. Report Ho.
4. Title CONTROLLING THERMAL POLLUTION ON
SMALL STREAMS
7. Author(s)
George W. Brown and Jon R. Brazier
9. Organization
School of Forestry
Oregon State University
Corvallis, Oregon 97331
12. Sponsoring Organization
is. SUpp/&ffiy#W8ftfal Protection Agency
Environmental Protection Agency report
number EPA-R2-T2-083, October 1972.
3. Accession No.
W
5. Report Date July, 1972
6.
8. Performing Organization
Report No.
10. Project No.
11. Contract/Grant No.
16130 FOK
13. Type of Report and
Period Covered
16. Abstract
Buffer strips have been proposed as a method for controlling temperature changes
in streams adjacent to clear-cuttings. Nine small mountain streams in Oregon's
Coast Range and Cascade Mountains were studied to determine the influence of buffer
strips on water temperature. Timber volume In the strip, strip width, and canopy
density perpendicular to the sun's rays were compared to the effectiveness of the
strip in controlling temperature change. This effectiveness was not well correlated
with timber volume or strip width. The density of the canopy in the path of the sun
is the most important buffer strip characteristic for water temperature control.
A method for measuring the density of the canopy in the path of the sun is
described. The use of this method in the design of buffer strips will provide
protection for the stream and maximum harvesting of the timber resource.
This report was submitted in fulfillment of Grant Number 16130 FOK under the
sponsorship of the Office of Research and Monitoring, Environmental Protection
Agency.
17 a. Descriptors
Thermal pollution, water temperature, lumbering, forestry
lib. Identifiers
17c.CO WRR Field & Group Q5G
IS. Availability
19. Security Class.
(Report)
20. Security Class.
(Page)
Abstractor George W. Brown^
21. No. of
Pages
22. Price
Send To:
VATER RESOURCES SCI ENTIFIC INFORMATION CENTER
u.s. DEPARTMENT OF THE INTERIOR
WASHINGTON, D. C. 20240
Institution
Oregon State University
WRS1C 102 (REV JUNE 1971)
913.20!
-------