THE UNIVERSITY OF NORTH CAROLINA WATER RESOURCES RESEARCH INSTITUTE
NORTH CAROLINA STATE UNIVERSITY THE UNIVERSITY OF NORTH CAROLINA at RALEIGH at CHAPELHILL
124 Riddick Building North Carolina State University Raleigh, North Carolina, 27607
June 7, 1971
TO : Whom It May Concern
F RON : David M. Howells SUBJECT: Institute Report E?o, 49 - "Migration and Metabolism in a Stream Ecosystem,'"y Charles A. S. Hall
lJhile this reports and interprets findings on fish migration and stream metabolism in New Hope Creek, it has a much broader application to Piedmont streams in general,
I4r. Hall's conclusions and recornmendatfons, pages xv to xix, relates to diurnal variations in dissolved oxygen and importance of pre-da-m sampling, barriers to fish migration, and stream classification for research and other scientific purposes.
Enclosure
MIGRATION AND JXIETABOLIISM IN A STREAM ECOSYSTEM
by
Charles A, S. Hall
A thesis submitted to the faculty of the University of North Carolina at Chapel Hill in partial fulfillment of requirements for the degree of Doctor of Philosophy in the Department of Zoology, September 1970.
TQe work upon which this publication is based was supported in part by funds provided by the Office of Water Resources Research, Department of the Interior, through the Water Resources Research Institute of the University of North Carolina as authorized under the Water Resources Research Act of 1966.
Pro j ect No. B-007-NC Matching Grant Agreement mo. 14-01-0001-1933
Professor H. T. Odum - Thesis Advisor Professor Charles M. Weiss - Project Director
Department of Environmental Sciences and Engineering School of Public Health University of North Carolina at Chapel
February 1971 c 'I
TABLE OF CONTENTS
LIST OF TABLES, ...... ix
LIST OF FICTURFS...... xi
INTRODUCTION ...... I. Theories for Migration...... 2 P4igratiion to ar~idunfavorable condition, Migration and reproduction, Migration and optimal use of fluctuating environments.
Role of Migrating Animals in Mineral Cycling...... 7
Previous Studies on the Movements of Fishes...... 8 Movements totally within one stream, Movements of fishes in streams with adjoining lakes, Movements of fishes be- tween fresh and salt water, Movements of fishes in the open sea.
Statement of Purpose...... 14 Description of Study Area...... 14 Qualitative energy flow diagram for migration in New Hope Creek.
MATERIALSANDMETHQDS...... 22 Physical and Chemical Data...... 22 Characteristics of the sampling stations, Discharge, Stream morphology - depth and width, Insolation, Stream temperature, Total phosphorus in water, Phosphorus in organisms, Total nitrogen in water, Stream conductivity, Discharge of leaves. MetabolicStudies...... 31 Dissolved oxygen, Winkler method; Dissolved oxygen, Gal- vanic probe method; Diffusion rates; Gross community metabolism: Two station analysis, Single curve method; Estimate of metabol.ism from pH changes; Computer program for estimating community metabolism from diurnal oxygen curves . Fish Sampling Procedures and Apparatus...... 53 Design of weirs and traps, Check on possible sampling bias in up a.nd down traps, Check on ra:e of fish escape from traps, Sampling modifications for low water, Sampling modifications during high waters, Methodology of handling species and species groups for analysis, baily fish sampling procedure.
Physical Data...... 7 4 Stream morphology, Stream level and discharge rate, Stream temperatures, Light intensity at the surface of the stream, Leaf discharge, Total phosphorus, Nitrogen, Stream conductivity. Metabolicstudies...... 90 Daily variations in oxygen, Annual variations in metabo- lism, Spatial variations in metabolism, Annual and spatial variations in P/R ratio. FishMovements...... 110 Analysis of all species considered together: Principal sampling station, April 1968 - June, 1970, Seasonal variations in movements, Cumulative occurrence of species vs. cumulative occurrence of individuals, Diversity of moving animals, Movements at other stations on New Hope Creek, Movements at Morgan Creek; Analysis by each spe- cies: Numerical and weight contribution of each species to migration, Seasonal patterns of movements for each taxonomic class, Evidence of spawning condition of fish at different times of the year, Recaptures of marked fish, Tagged fish returns analysis, Daily concentration of moving animals. DISCUSSION...... 174 Seasonal Patterns of Metabolism...... 174 P/R ratio and heterotrophic regime. Spatial Distribution of Metabolism...... 175 Dilution of resources with depth. Comparison With Some Other Studies...... 179 Patterns of Fish Movement...... 182 Movements of different species, Movements and floods, Movements of juvenile fishes, Differential movements of different-sized fish. Com~arisonsof Ewrgy Rwlgets ...... 182 Energy of ruaning water, Energy of bi~logicalmetabolism, Energy of insolation, Energy of fish metabolism, Energy of migration, Net Contributions of Migration to Headwaters and Turnover Rate...... I92 Comparisons of migration in New Hope Creek with salmon migration. Possible Adaptive Values of Migrations in New Hope Creek. . .I93 Migration as a coupling function, Interaction of yield and organization. Some Other Animal Migrations and Environmental Energy Patterns ...... 196 New Hope Creek Watershed Annual Phosphorus Budget...... 197 Analog Simulation of a Migration Model...... 202 Analog results and discussion.
LITERATURE CITED...... 212 APPENDIX A: DIFFUSION PROCEDURES USED IN NEW HOPE CREEK METABOLISMSTUDIES ...... 227
AESTRACT
Fish migration and total stream metabolism were studied in New
Hope Creek, North Carolina, from April, 1968 to June, 1970. Up-
stream and downstream movement of fishes was monitored using weirs
with traps. Most of the 27 species had a consistent pattern of
larger fish moving upstream and smaller fish moving downstream. Both
upstream and downstream movements were greatest in the spring. For
example, in the spring of 1969, a daily average of 7 fish weighing a
total of 1081 grams were caught moving upstream, and 17 fish, weighing
a total of 472 grams, were caught moving downstream. Although more
moved downstream than up, the larger average size of the fish moving
upstream resulted in a large transfer of fish mass upstream.
Diurnal oxygen series were run to measure the metabolism of the
aquatic community. Gross photosynthesis ranged from 0.21 to almost
9 g m-2 day-1 ~~/m~/da~), and community respiration from 0.4 to
13 g m-2 day-1 at the principal sampling station and both were highest in the spring. Area values of metabolism were similar for different parts of the stream, but both production per volume and respiration per volume were much larger near the headwaters than farther downstream. This was appareqtly due to the diluting effect of the deeper water downstream. Migration may allow populations to take advantage of such differences in productivity by maintaining young fish in areas of high productivity. An energy diagram was drawn comparing energies of insolation, currents, photosynthesis, respiration, fish populations, and migra- tions. Parts of this model were simulated on an analog computer. Input energies from insolation and stream flow were similar. About 0.14 percent of the total respiration of the stream was from fish populations, and over one year about 0.01percent of the total energy used by the ecosystem was used for the process of migration. If it is assumed that upstream migration is necessary to maintain upstream stocks, which may be periodically decimated by droughts, the migration energy has an amplifying value of 14. ACKNOWLEDGEMENTS
The dissertation was done under the supervision of Howard T.
Odum and Charles M. Weiss, with facilities at New Hope Creek pro-
vided through the courtesy of the Duke Forest Administration and
with cooperation of the North Carolina Wildlife Resources Commission
by courtesy of Harry Cornell. Elizabeth Mcblahan, Edward J. Kuenzler,
and Joseph Bailey served on the supervisory committee.
Financial support was provided by the Water Resources Research
Institute, University of North Carolina, Grant B-007-NC (Office of
Water Resources Research, United States Department of the Interior)
to Charles M. Weiss, 2nd AEC Contract AT-(40-1)-36666, H. T. Odum,
principal investigato~,and an allocation from the North Carolina
Computer Center,
Thomas P. Stevenson, Wayne Franklin, John Floyd and others
assisted in the often arduous field work and data processing. John
Gum and others at tl7e University of North Carolina Computation
Center aided with digital computer programs,and Larry Burns and Fred
Waf aided with an analog program. Tony Owens of the Department of
Environmental Sciences and Engineering, University of North Carolina did phosphorus and nitrogen analyses. Dennis Whigham provided S nso-
lation charts. Joseph Bailey of Duke University aided in the identification of fishes. Peter Larkin, of the Institute of Ecology, University of
British Columbia, and David Narver, of the Fisheries Research Board of Canada,provided suggestions and funds for salmon studies at Vancouver, Nanairno, and Babine Lake, British Columbia in the summer of 1969. My advisor, Dr. Howard T. Odum, was the impetus and nucleus for the excitement in ecology that I have experienced at the University of North Carolina for the past three years. I an grateful for having had this opportunity. LIST OF TABLES
Tables Page --, -
Some Characteristics of the Various Sampling Stations
Drift of Oxygen Recorder Over One or Several Days
Modifications in Basic Trapping Procedure
Catch of Fish in 'Sideways' Traps
Floods in New Hope Creek That Affected Sampling
Fishes Captured in New Hope Creek and Groupings Used to Simplify Analysis
Organisms Other Than Fish Captured in New Hope Creek
Depth and Width Profile for 300 m Below Concrete Bridge, May 23, 1970
Depth and Width Profile for 1.8 km Above Concrete Bridge Station, April, 1969
Depth and Width Profile for 900 m Above Wood Bridge Station, May 13, and 23, 1970
Depth and Width Profile For the Zone lOOOm Above Blackwood Sampling Station, May 18, 1970
Light Intensity at Surface of New Hope Creek
Total Phosphorus (Dis solved and suspended) in New Hope Creek at Concrete Bridge Station
Nitrogen Compounds in New Hope Creek
Total Community Metabolism for New Hope Creek, Concrete Bridge Station, April, 1968-~a~,1970
Total Community Metabolism for New Hope Creek, Wood Bridge Station, June, 1968-August, 1969 17. Total Community Metabolism for New Hope Creek, Blackwood Station, Febrxary, 1969-February, 1970
18a. Average Daily Fish Movements by Month
18. Summary of Trap Catches at Wood Bridge and Jungle Stations, New Hope Creek
19. Summary of Trap Catches at Morgan Creek
20. Minimum, Maximum and Total Mass and Total Numbers of Each Species or Group Sampled at Principal Station, Mew Hope Creek
21. Average Mass of Animals Moving at Principal Station
22. Evidence of Spawning Conditlcn
23. Recapture of Marked Fish
24. Recapture of Tagged Fish
25. Concentrat ion (~aily)of Moving Organisms
26. Concentration of Phosphorus at Different Stations on Same Dates
27. Metabolism in Some Other Unpolluted Streams
28. Metabolism of Some Selected Lakes and Maine Waters
29. Annual Movement of Phosphorus in New Hope Creek: June 14, 1968- ~une13, 1969
A-1. Diffusion Constants Derived from Diurnal Oxygen Data
A-2. Predicted Values for Diffusion Constant for Ne7~ Hope Creek Above Concrete Bridge Station Using Formula Based on Average Depth and Velocity
A-3. Basis for Calculations of Diffusion Coefficient from Dome Measurements
A-4. Estimates of Diffusion Constant (K) Obtained Using the Dome Method for Representative Pools and Riffles Above Concrete Bridge Station LIST OF FIGURES
--.--Page
1. Location of sampling stations on New Hope and Morgan Creeks, North Carol-ina.
2. A typical riffle stretch of New Hope Creek, located just above the Concrete Bridge Station.
3. Energy circuit diagram for migration in New Hope Creek.
4. Symbols used in energy network diagrams, from H. T, Odum (1967a) .
5. Water stage vs . discharge.
6. Total mass of leaves (dry weight) discharged per day in stream flow at Concrete Bridge Station vs. stage level (ordinate) in centimeters above zero flow.
7. Cork and tubing device to fill oxygen bottle with- out air mixing.
8. Comparison of probe and Winkler oxygen values over a 24 hour period, July 25, 1969, at Concrete Bridge Station.
9. Variation in oxygen meter readings with constant dissolved oxygen and varying temperature.
LO. Comparison of different diffusion constants obtained in this study.
11. Similarity of oxygen curves one hour's flow distance apart at Blackwood Station, February 14, 1-970.
12. Similarity of oxygen curves one hour's flow distance apart at Concrete Bridge Station, February 14, 1970. 13. A representative sample of single station analysis for community metabolism in New Hope Creek, February 14, 1970, as conducted and plotted by the UNC CALCOMF plotter.
14. Various lines drawn to represent daytime respiration.
15. Carbon dioxide titration of New Hope Creek water for metabolic studies.
16. Estimation of metabolism in New Hope Creek, Diurnal pH method.
17. Early design of fish weir (~ecember,1968).
18. Design of fish weirs used in New Hope Creek.
19. Big Pool sampling station, looking downztream during normal spring flow.
20. "Sid.eways" fish sampling arrangement.
2 Design of fish trap used in New Hope Creek.
22. a. "sideways" and b. "double reverse" weirs used. to test possible sampling bias in normal trap arrangement.
23. New Hope Creek d.uring drought (~eptember,1968) .
24. New Hope Creek at Concrete Bridge Station during flood.
25. Overrun of weir during severe flood at Big Pool Station on April 14, 1970.
26. Daily water stage level; in cm above zero flow, of New Hope Creek at Concrete Bridge Station.
27. Mean daily temperatures for New Hope Creek during this study.
28. Typical diurnal oxygen curve for spring, Concrete Bridge Station.
29. Typical diurnal oxygen curve for spring, Wood Bridge Station. xiii
30. Typical diurnal oxygen curve for spring, Blac1cwood Station.
31. Typical diurnal curve for late fall, Concrete Bridge Station.
32. Typical diurnal curve for winter, Wood Bridge Station.
33. Typical diurnal oxygen curve for late fall, Blackwood Station.
34. Annual- variation in metabolism, Concrete Bridge Station, New Hope Creek, April, 1968 - May, 1970.
35. Annual variation in metabolism, Wood Bridge Station, New Hope Creek, June, 1968 - August, 1969,
36. Annual variation in metabolism, Blackwood Station, New Hope Creek, February, 1969 - February, 1970. 37. Seasonal variation of photosynthesis respiration ratio at Concrete Bridge Station.
38. Average daily migration by month.
39. Cumulative species versus cumulative individuals trapped at principal sampling station; only fishes are included.
40. Upstream and downstreax movement of each species or species group in New Hope Creek by si~einterval.
41. Average daily movement for each number, by species.
42. Seasonal patterns of insolation under a hardwood canopy, Duke Forest, near New Hope Creek.
43. Growth of tagged fish, New Hope Creek. 44. Annual movement and metabolism of fish popula- tions in the headwaters of New Hope Creek above the Concrete Bridge.
45. Energy flow diagram for upstream (middle set of modules) and downstream (lowermost set of modules) of New Hope Creek. xiv +r
46. Diagram of phosphorus flows in New Hope Creek A watershed. 47. Energy flow diagram for analog computer model.
48. Analog symbols representing the energy pathways in Figure 47.
49. Analog output of energy pulse generator.
50. Analog simulation of annual energy accrual to populations of fishes in New Hope Creek.
A-1. Use of clear plastic dome to measure diff'usion constant .
-2 Use of plastic dome to measure diffusion. Mieion and Metabolism in a Stream Ecosystem - ---.. -"- - "- -" - ----
Preserving and Enhancing the Qualities of the Waters of North Carolina
In the period April, 1968 to June, 1970 an intensive investi- gation was made in New Hope Creek, in the stretch where it flows through Duke Forest, to establish the relationship between fish migration and the total stream metabolism. New Hope Creek at this particular point may be the only stream in the Research Triangle area of North Carolina where studies of relatively natural conditions can be carried out. Nearly all other streams in the region are either polluted or are too small for any extended studies. The location within Duke Forest, with controlled access made the region particularly desirable for studies in what is essentially a natural outdoor laboratory.
The basic investigation consisted of monitoring up and down- stream movement of fishes, using weirs with traps. Of the 27 species collected, most had a consistent pattern of the larger fish moving upstream and smaller fish moving downstream. Movement in both directions was greatest in the spring. For example, in the spring of 1969, a daily average of 7 fish weighing a total of 1087 gms were trapped moving upstream and 17 fish weighing a total of 472 gms were caught moving downstream. Although more fish moved downstrean
than up, the larger average size of the fish moving upstream resulted
in a larger transfer of fish mass upstrsam. Associated with the fish movement studies, the metabolism of the aquatic communiJiy was determined using the technique of diurnal oxygen measurements. Gross photo-
synthesis ranged from 0.21 to almost 9 g/m2/day and community respira- tion from 0.4 to 13 g/m2/day. All measurements of this nature were highest in the spring. Both production and respiration per volume were much larger near the headwaters than farther downstream. This was apparently a result of the diluting effect of the deeper dater
downstream. Migration appeared to allov .the fish population to ta,ke advantage of such differences in productivity by rnzlntaSning :joung fish in areas of high productivi5y.
A total energy diagram was devi-ed cornparin& energies of insola-
tion, currents, photosynthesis, respiration, fish gopula tions and migrations. When this mode? was simulated on an analog computer, it was determined that input energies frorr, insolation and streax flow were similar with about 0.14 percent of the total respiration of the
stream derived from fish ponulations. Over a period of a year a>out
0.01 percent of the total energy used by the ecosysiem Nas consumed
in the process of migration. It can be asamcd ihai th2 upstream migration is necessary to maintain upstream fish stocks xhich may be periodically decimated by drought conditions. Migration el?ergy appears to have an amplifying value of 14. There are certain lessons thag may be learned from the preceding investigation, which are of value for preserving and enhancing the qualities of the waters of North Carolina. As found in New Hope Creek and probably for most other streams of piedmont North Carolina, the diurnal dissolved oxygen variation may be quite large up to three or four mg/l. Thus, criteria for oxygen in any stream must be established as a minimum pre-dawn value since this could very well be the limita- tion for any aquatic organism requiring oxygen since they must live in the stream 24 hours a day. The daily fluctuations in oxygen were found to be greatest in shallow water. This characteristic may be of considerable significance in water quality decisions since:
(1) as streams become more shallow during summer low waters, the difference between day and night oxygen values become larger;
(2) upstream, there are generally more shallow regions of the stream with greater day-night differences in oxygen content.
Since aquatic organisms using dissolved oxygen require more at higher temperatures, suxmer conditions, therefore, may create a critical circumstance due to (a) lowering the solubility of oxygen in water and (b) increasing the oxygen requirements of organisms and
(c) increasing the daily fluctuation of oxygen as biotic components of the stream ecosystem become more crowded in shallower water. It is therefore indicated that the oxygen requirements for streams be set for minimum conditions at one hour before sunrise during periods of highest temperature and/or lowest waters, generally in August. It would thus be indicated that if any pollution is suspected in a stream, it becomes even more critical to establish the pre-dawn oxygen level with reference to the quality of the particular body of water.
' Regional planning of aquatic wastes disposal should take into account the potentially greater stress that is imposed on the shallower regions of streams and rivers. This implies that the establishment of regional plans for economic growth, a basic principle should be one of not introducing industries and waste disposal facilities on the headwaters of rivers.
It was also determined from the investigation on New Hope Creek that many fishes in Piedmont streams have distinct patterns of move- ment. These may be necessary for optimizing the reproductive potential fish populations that are available for restocking of an area that may naturally or otherwise loose fish population. It may be a wise manage- ment policy to aid this movement by removing unnecessary stream obstruc- tions. A localized area of pollution in a stream may be detrimental to more fish than just those in the immediate vicinity. The entire reproductive potential for a large area of a stream may be lost as migrating fish attempt to move through a polluted region. This con- sideration should be taken into account in stream pollution studies and may be particularly critical during the March to May period of fish migration. Utilizing information gathered in the study, predic- tions for a repopulation of an area that has been totally depleted xix
of a fish population due to pollution indicates that it would take
about 2$ years to re-establish the pre-pollution population. This
estimate could be used in the economic assessment of pollution damage.
The nutrient balance established for New Hope Creek as it flows
through Duke Forest, with particular emphasis on the cycling of phosphorus, indicated the value of a natural ecosystem for retaining
vital nutrients. Protection of water sheds have thus two values, one for maintaining stocks of nutrients in valuable locations such as forests and keeping the same nutrients out of oligotrophic streams where they might cause undesirable eutrophication if they should be released.
The value of ilTew Hope Creek to the studies in the basic metab- olism of a stream cannot be overemphasized since so few unpolluted streams are available for such studies. Maintenance of this stream in its natural state as an outdoor laboratory for the Triangle Uni- versities requires that it receive a stream classification under the
North Carolina system of stream classification which will give it ad.equateprotection.
INTRODUCTION
Animal migrations are a conspicuous and important phenomenon in many ecosystems of the world. Myriads of popular articles have been written about the migrations of fishes and birds, and the scientific literature is full of data on these and other migrants.
This study considers migration as a functional component of a stream ecosystem by relating fish movements to stream metabolism.
Seasonal patterns of metabolism and fish migration were measured in field studies in New Hope and Morgan Creeks, Orange and Durham Counties,
North Carolina, from April, 1968 to June, 1970. The results were compared with the movement patterns of some other species in the bio- sphere as reported in the literature.
What is the role of migration in the many and varied ecosystems in which it is found? Under what conditions do groups of animals that migrate have selective value over other groups that do not migrate?
How much energy is required to migrate, and can enerqy be gained by migration? What effect does migration have on the ecosystem of which it is a component and vice versa? What percentage of an ecosystem's energy budget is tied up in maintaining a migratory component? This study considers the above questions for a small warm-water stream in the piedmont region of North Carolina. Theories For hligration
There may be selective advantages for migration patterns which lead to success of the migrants and survival of the systems which support migrants. Consider previous studies which discuss migration as a mechanism for improving the chances of survival of the population.
Non-reproductory migratory movements may be undertaken for the sake of self or species preservation (Ijeape, 1931). Three principal types are: alimental, or having to do with food; climatic, or having to do with extremes in climate (particularly temperature), and gametic, or having to do with reproduction. Heape considered that these different migration types are often related: "In all animals which experience a gametic migration, a return journey is involved which is directly concerned with either climatic or alimental conditions." According to him, the return journeys often can be considered nomadic; and non-gametic migrations are considered, as a rule, spasmodic or due to exc'eptional conditions--although he discusses on the next page regular seasonal movements of arctic animals which move in response to "not cold so much as want of food." bligration to Avoid Unfavorable Conditions
Allee --et al. (1949, p. 539) state that an organism has but three choices when exposed to adversity: it may die, adjust, or migrate. 3 Thus,in their discussion of fluctuations in environmental conditions, migration is considered a ~echanismfor removing the organism from unfavorable circumstances. The reason for return during more favorable circumstances is not as clearly spelled out.
Migration and Reproduct ion ---.. "-,-- -- Migration may bring fishes back to areas in which their ancestral eggs developed. "In most instames there is also a seasonal or periodical (non-spawning or larval) migration affecting the immature and mature" (Meek, 1916). He considers migrations that occur from deep to shallow regions in a lake, movements up and down rivers, and, particularly, movements in various locations in the sea. In the ocean, he says, there is a general movement in toward shore for spawning, followed by dispersal seaward. This pattern recurs each year with increasing amplitude as the young mature. The result is that the oldest fishes disperse farther from shore during non-spawning times. Heape (1931) gives many examples of fishes, birds, and mammals with extensive migratory movements for reproduction without, however, saying why an animal should migrate to reproduce.
Migration and Optimal Use of Fluctuating Environments
Mayr and Meise (1930), as quoted in Cox (1968), suggested that competition for food, principally as a result of reproductive excess, is the factor favoring the development of mechanisms allowing seasonal occupation of areas with alterations of favorable and unfavorable con- ditions. A rigorous approach to this problem has been undertaken by students of S. C. Kendeigh (Siebert, 1949; West, 1960; Cox, 1961; 4 Zimmerman, 1965). These studies investigated the energy balance of migrating birds in terms of energy required to migrate and energy gained by being in different places at different times. Siebert concluded that southward migration for the slate-colored junco and the white-throated sparrow was a metabolic necessity. West came to the same conclusion for the tree sparrow, but did not find that the northern migration gained an improved energy balance. Cox (1961) found that resident tropical finches would gain little by northward migration. Zimmerman, however, concluded that the dickcissel gained an improved energy balance in both its northern and southern movements.
Cox (1968) suggested divergent adaptation by both morphological and ethological means. Given interspecific or intergeneric competition, animals may broaden their niche by exploiting, for example, different food sources; or, they may broaden their niche by exploiting spatially different environments. Cox showed that within taxonomic groupings (usually orders or families), culmen (a part of the beak) length variability among species was much greater for bird groups that did not have a high frequency of migratory members.
Thus some bird groups diverged by exploiting different food sources within a single environment, and oth.ers moved to different areas. There may be a limit to food niche divergence at which animals must begin to exploit new physical environments. llechanisms for this are discussed by Cox.
Migratory patterns of animals associated with Texas estuaries are considered in relation to the primary production and environmental food supply by Odum and Nosltins (1958), Simmons and Hoese (l959),
Hellier (1960), Odum and Yilson (1962), Copeland (l965), and Odum
(1969). These studies emph2size how the very large spring production
of these areas are utilizecl by migrating animals, especially during
their juvenile stages, and how the migratory patterns are such
that maximvm use is made of the pulse in energy in those ecosystems
during the late spring, The migrations themselves are seen as a
mechanism to even out the flow of energy in the system and distribute
energy and nutrients. Odum (1959) states that "Seasonal a.nd
diurnal migrations not only make possible occupation of regions
which would be unfavorable in the absence of migration but also
enable animals to maintain a higher average density and activity
rate."
Another attempt to explain the reproductive migrations of
animals in relation to selective advantages for the migrating
population and energetic characteristics of environments is by
Margalef (1963, 1968). He discusses different degrees of maturity
in ecosystems. Margalef defines maturity in terms of the degree
of organization of the ecosystem, which is not necessarily related
to chronological age. According to him, less mature ecosystems are
less efficient in their use of energy and support less biomass on
the same energy flow. Thus there is an excess of available energy
that may be exported. More mature systems, with a great com- plexity of biological interactions and resultant greater efficiency in
energy use, produce no, or at least less, excess energy.
Margalef continues with the argument that those individual or- ganisms that have developed behavior patterns leading to reproduction 6 8' in less mature ecosystems have left behind more offspring and, there-
fore, are selected for. He gives examples of animals that tend to &B spend their adult life in more mature areas and reproduce in less
mature areas or send larvae or reproductive elements into them.
Some examples are: migrating birds flying to polar regions to re-
produce, benthic animals sending larvae into less mature planktonic
environments, and clupeid fishes that spawn in less mature parts of
the coast of Spain and spend their adult lives in more mature
regions. Even the seemingly enigmatic situation of eels and salmon
can be explained in this way, he says, since the specific regions that
both adult animals inhabit are more mature than the specific habitat
of the larvae of the respective fishes.
McLaren (1963) found in models of migrating zooplankton that
the energy saved by living one-half of the day in colder water,
where metabolism was less, was greater than the energy used in the
process of migration. The energy gained by this process could
then be used for growth and reproduction.
Ricard (1968) considers animal migrations as an integral part
of biological rhythm. He suggests that migration serves as a
mechanism for regulating population numbers of species such as swallows, since many members of a population are lost during mi-
gration. This may also be true for lemmings, although their movement
cannot be considered a true migration since the lemmings do not
return. Ricard also suggests that animals move to different areas
where their food is seasonally more abundant. He generalizes:
"One must conclude, therefore, that migration is not the only solution 7
to the problem of the balance between animals and food resources, but
that it is the one that exists 8t the present time."
According to F. R, Harden Jones (1969) migrations are "an
adaptation for abundance by making the most of a varied environment."
In giving a detailed analysis of migratory patterns of five groups
of fishes, he considers how these patterns have evolved to aid in
the utilization 05 various food sources.
Foster (1969), in reviewing possible causes for the development
of migration in fishes, considers the possibilities of changes in
food availability, climate, salinity, and topography over geologic
time. The interaction of exploitation of new resources with the need
for the adults or eggs to stay within certain physiological limits
may have set the stage for the first fish migrations.
One common factor in all these previous studies is the role of
migration in increasing the flow of energy, or decreasing the energy
loss, to populations involved. Movements away from energy-consuming,
food-poor, cold regions in the winter, as well as to energy-rich
areas of high productivity, can be considered in these terms. The
energy cost of migration has been considered by Idler and Clemens
(1959) , McLaren (l963), and Brett (19 70) .
Role of Migrating Animals in hlineral Cycling
Among the first authors to consider the potential of migrating animals for recycling or important limiting minerals was Juday -et -al. (19521, who speculated upon the role of dead salmon in bringing phosphorus and other minerals to the stream-lake ecosystems of the F * salmon's early life history. Quantitative work on this was under+ I,e:l by Donaldson (1967) and Krokhin (1967) who demonstrated the very large b.. role dead salmon had in supplying sufficiently high levels of phosphorus to maintain productivity of sockeye lakes at a sufficient level to support large runs of salmon.
Many further examples may be present in other fishes, vertically migrating plankton, and migrating birds. With the tremendous im- portance of small amounts of some trace elements now being recognized
(Hutchinson, 1957; Goldman, 1969), possibilities do exist for migrations to control critical nutrients.
Previous Studies on the Movements of Fishes
Nearly all studies of fish migration within fresh water have occurred with species that are associated either with lakes or the ocean. Only a small amount of the total information available con- cerns fishes that spend all their time within one fresh-water stream.
Movements Totally Within One Stream
Many studies have been conducted over the years to study fish movements in streams. Bangham and Bennington (1938) report a re- capture of only about 11 percent of fishes seined and marked in a warm-water Ohio stream. Three centrarchids (smallmouth bass, green sunfish and rock bass) had much higher (19-20) percentages of tag returns than did other species. No marked fish were recovcr~c;In adjacent one-mile sections of streams located above and below the marking area. They concluded from these studies that fish in their streams moved about very little. Further evidence for this view, 9 most of it based on returns of tagged fish by sport fishermen, is supplied in Scott (??d.9$ for rock bass in Indiana and by Tate (19fl9) for smallmouth bass in some small streams in Iowa, Allen (1951) found little seasonal movement of trout in New Zealand, Gerking (1959) con- cluded that most fresh water fjshes had limited home ranges, and
Gunning and Shoop (1961) found little short range movement in stream dwelling American eels.
Other investigators have come up with other conclusions. Stefanich
(1952) found some fishes that had moved and some that were stationary in a Montana cold-water stream. Brown (1961) found similar results for warm-water fish in Ohio. Bjornn and Mallet (1964) found very distinct patterns of spring upstream movements and fall downsteam movements for native populations of cutthroat trout and Dolly Varden.
Some of these fish had traveled at least 50 to 60 miles. Considerably greater numbers of fish were recaptured in areas outside of the original capture area than within. Behmer (1964) found different patterns of movement for different warm-water fishes in Iowa, in- cluding some movements of 40 miles. Hunt (196A) reported that wild brook trout in Laurence Creel:, Wisconsin, used upstream reaches of the creek for spawning much more than they used downstream areas,with the inference that the trout moved upstream to spawn. He also found considerable dispersive movements of young trout, generally in a downstream direction. Shetter (1368) found complicated patterns of trout movement in the Au Sable River in Michigan. Many did not move; some move up and some moved down, with no particular seasonal pattern evident. The patterns differed from one watershed to another.
i C Shetterls study, as all those listed so far, is based on recapturt of marked fish either by seining, electric shocking, or angler re- turns.
One answer to these complicated patterns of movement (including no movement) is supplied by Funk (1955), who suggested that many stream fishes have both a mobile and a sedentary population of each species. As in other studies, his work indicated varied patterns of fish movements. Some fish moved up, some moved down,and some did not move at all. This was true both for species groups and for different individuals within a species. All important fish species showed a greater tendency to move in the spring than during the summer. Re- sults of fish movements in the fall were varied.
Unfortunately, almost all of these data are heavily biased by the sampling procedures. hfuch more field work was done in summers than at other times of the year. More sampling was done in areas readily accessible to vehicles, hence angling pressure also tended to be concentrated at these areas causing bias of results toward recaptures in the area of original capture. Some studies considered recaptures within the same pool as representative of no movement, others included all fish captured within one mile of the sampling site.
The overall picture for streams to date is confusing. A similar conclusion is reached in a literature survey by Carpenter (1967).
Movements of Fishes in Streams with Adjoining Lakes
On the other hand there is a fairly consistent pattern of fish spawning runs from lakes and ponds to inflowing or outflowing streams.
Stream dwelling brook trout moved upstream in the fall; brown and 11 4 rainbow trout were principally captured moving upstream in the spring and summer (Shetter, 19%). Suckers, which were the most important C fish captured during this study in terms of numbers and mass, were captured moving downstream in the spring and upstream in the fall.
Shetter suggests that this is probably a spawning run from lakes that are located above the counting weir. Northern pike generally moved downstream, and other fishes had less consistent patterns. During the one year of Shettervs study, approximately equal numbers of fish were captured moving upstream as down. No data were given as to the size of the fishes moving upstream and down.
Raney and Webster (1942) and Ra.yner (1942) found runs of spawning common white suckers and rainbow trout in an inlet to
Skaneateles Lake, New York. The suckers moved upstream in April and
May and back downstream sometime later in ?lay. The rainbow trout migrated into the stream during the second and third week in April and appeared to stay in the stream for five days to two months.
Other studies done in Michigan using two-way fish weirs (Carbine and Shetter, 1943) showed that tributary streams contributed many small brook trout to the main stream of Hunt Creek and that large spring runs of suckers and redhorses moved downstream from Houghton
Lake into Muskegon River. Some of the suckers and redhorses returned upstream; but the majority, apparently d.id not, and many dead spent fish were observed just after spawning. Similar results were obtained at Lake Gogebic. Large upstream runs of suckers and rainbow trout were captured in a two-way weir installed at the mouth of the Platte
River where it enters Lake Michigan. About 20 times more fish were captured moving up the Platte than down. Most of the fish movement occurred during the month of April and was apparently associated with spawning. In the Brule River of Wisconsin (Niemuth, 1967), heavy runs of exceptionally large brown trout moved out of Lake
Superior during the summer and fall for spawning. Large numbers of these fish died after spawning, although some returned to the lake the following spring. Young trout stayed in the river for about two years, then moved down to the lake. Warner (1959) found that landlocked salmon moved downstream from large lakes in Maine to spawn and that the majority returned to the lakes after spawning.
Perhaps the most intensive study of the relation of lake-dwelling fish and spawning streams has been conducted by Martman --et al. (1962) in Loon Lake, British Columbia, with naturally occurring rainbow trout.
Both inlet and outlet streams were used for spawning, although the inlet stream was used much more heavily. Both spawning runs apparently had a large mortality of spawning fish.
A common characteristic in most of these studies is that more fish are captured going from the lakes into the streams than vice-- versa. Since all weirs used for these studies had mesh sizes that allowed juvenile fish to pass, the movements for the total populations are unknown. There may be a substantial return of small fish. In addition, most of these studies indicate that the movements of fish into the streams were associated with spawning activities and that a large percentage of the spawning fish failed to return to the lake from which they originally came. Movements of Fishes Between Fresh and Salt Water
Some species of fish that move between fresh and salt water have been studied intensively. The movements of salmon a.nd eels have been reviewed by Harden Jones (190R), and the ge~esallife history patterns of these fishes is well known. Banks (1969) has reviewed the literature on the movement of salmon from the sea to their spawning grounds. Fishes such as salmon, that spend their adult life in salt water but spawn in fresh waters, are known as anadro- mous; while those that do the reverse, such as eels, are known as catadromous.
The life history of several other Atlantic anadromous fish, such as alewives, shad, and striped bass, are reviewed by Bigelow and Schroeder (1953), Talbot and Sykes (1958), and Miller (1969).
Studies of the'movements of brook trout between fresh and salt water have been done by Smith and Saunders (1958, 1967, 1968) on Prince
Edward Island. Sumner (1962) studied the movements of cutthroat trout between fresh and salt water in Oregon. Many other sea fishes, such as tarpon and snook, travel freely between fresh and salt water in movements apparently not directly connected with spawning (Breder, 1948).
Movements of Fishes in the Open Sea
The movements of Pacific salmon on the open sea have been summarized by Manzer (1960), Neave (1964), and Royce --et al. (1968). These papers present evidence for extremely far-ranging movements of some individual fish that may encompass almost the entire Pacific
Ocean. The fishes appear to follow fairly well defined routes, often in a broad circular pattern, and return to their parent streams from two to seven years after their entrance into the sea.
The movements of cthes fishes are in many cases not well known. Strasburg (1969) and Royce (1967) report that available evidence indicates a movement to the north of billfishes in summer and a return southward in winter. Neave and Hanavan (1960) found a northward movement of many species from May to August and
September. Mather (1969) reports east-west Atlantic migrations of bluefin tuna and seasonal north-south movements of white marlin.
Seasonal north-south movements for several species of northeast Pacific Ocean sole have heen reported by Alverson --et al. (1964). F. R. Harden Jones (1968) summarized much of the available evidence concerning movement of many North Sea fishes to and from breeding and wintering grounds.
Statement of Purpose
Many of these previous studies show migration of fishes to be prominent. Presumably, these movements involve considerable amounts of energy, possibly enough to be influential in controlling, directly or indirectly, the main flows of energy within their ecosystems. To study this possibility more fully requires measurements of migration and energy budgets in the same ecosystem, in order to determine their roles and relative magnitudes. This was done for a New Hope Creek, a small stream located in Duke Forest, North Carolina.
Description of Study Area
New Hope Creek is a relatively small piedmont stream located in
Orange, Durham, and Chatham counties, North Carolina (Figure 1). Its Figure 1. Location of sampling stations on New Hope and
Morgan Creeks, North Carolina. Each station was given a mnemonic name. Station 1 is "Way up"; 2 is "Horsefield"; 3 is lfBlackwoodw;
4 is "Weight limit 10"; 5 is "Wood Bridge"; 6 is "Jungle"; 7 is
'Toncrete Bridge", also "Big Pool" station is located about 100 meters upstream from "Concrete BridgeM; 8 is nP-66ff; and 9 is
"Pipeline." "Blackwood", "Wood Bridge", and "Concrete Bridge" -
"Big Pool" stations, numbers 3, 5, and 7, were most heavily sampled. M is the location sapled on Morgan Creek.
waters flow into New Nope River and then into Haw River and Cape
Fear River. The principal st~tdyarea is located in the Korstian
Division of Duke ForesC hetween Chapel Hill and Durham. The stream in this region is chamc?erized by a moderate gradient (3.96 m km-') and virtual lack of pollution. The average width is about 5 m and the average depth is about 0.4 m. Rocky rapids alternate with deep large pools (Figure 2). The water is normally clear, although the stream becomes turbid during floods.
New Hope Creek is relatively unaffected by man's activities and has the biological characteristics of a diverse and healthy stream. Larvae of mayflies, stoneflies, caddis flies and many other insects are abundant in the riffles and the fish life is diverse.
The North Carolina Division of Inland Fisheries has classified the creek as a "Robin-Warmou.th" stream (Carnes, Davis and Tatum, 1964) and considers the stream to be one of the best fishing streams in the Deep-Haw watershed. However, fishing pressure is light in the portion of the creek studied. Much of the watershed lies within the
Duke Forest and the rest runs through forested areas with an occasional farm. Very slight additions of domestic sewage enter from several sources near the headwaters. About 3.8 km below the study area, however, treated sewage from the town of Durham enters the creek. Very low oxygen (< 1.0 ppm) was occasionally found below the point of sewage addition during this study.
New Hope Creek, like many other piedmont streams of North Carolina, is subject to extreme fluctuations in water levels. During the two years studied summer water flows dropped to almost zero, although Figure 2. a. A typical riffle stretch of New Nope Creek, located just above the Concrete Bridge station. b. A typical pool of New Hope Creek, located just above the Big Pool sampling station. This particular pool is over 150 m long and 16 m wide.
numerous large pools remained. Fall, winter and spring floods were fairly frequent and raised the water flow to as much as 14.2 nS second-' (400 cu. ft. second-'). During these periods the stream expanded well beyond the banks and the water became quite muddy.
Morgan Creek, located just west of Chapel Hill, is a smaller stream which flows into University Lake,a 70 ha artificial im- poundment, about 3 km below the study site. .Morgan Creek above
University Lake is also virtually unpolluted and sha.res many physical characteristics and species of fish with New Hope Creek.
-.Qualitative Energy Flow Diagram for Migration in New Hope Creek Figure 3 shows a qualitative energy flow diagram for migration in New Hope Creek. The symbols used are those developed by H. T.
Odum (1967a, 1967b, 1969; Figure 4). Quantitative data on some of these flows are made available later in this thesis.
The ultimate source of the energy that runs the hydrology and the biology of New Hope Creek is, of course, the sun. Energy flows from the sun to green plants in the water, such as benthic algae, aquatic macrophytes, and pseudophytoplankton. Energy is then transferred through food chains to the fish populations. Sun energy also enters New Hope Creek in'directly through the Duke Forest trees, which drop their leaves into the water, and through the organisms that feed on these leaves. About one-half of all the energy required to run the biology of the stream enters in this fashion. The storage tanks represent the accumulation of organic material produced by the primary producers that is not immediately used by higher trophic levels. An obvious example is the accumulation of dead leaves on the bottom of the Figure 3. Energy circuit diagram for migration in New Hope
Creek. See text for explanation.
Figure 4. Symbols usccl in erxrgy network dia~ra~:;,F,-o;:I 1:. T.
Oduni, (1967a). ENERGY SOURCE PASSIVE ENERGY HEAT SINK S'TQRAGE
POTENTIAL PURE ENERGY &'%'OEM GATE GENERATING WORK RECEPTOR
SELF-MAINTAlfJ I NG . PLANT ECONOiiI f C CONSUMER POPULATIONS TEAf4SACTOR POPULATION stream, many of which are not eaten until the following spring.
Migration of fishes and other organisms is represented by the dashed lines connecting the populations of fishes. The usage of energy at any one place in the stream is to a certain extent dependent upon the relation of that part of the stream with currents and other parts of the stream. An indefinite number of such production-fish population sets could be drawn representing different parts of the stream. MATERIALS AND METHODS
The general plan for the study over a 27-month period
included studies of upstream and downstream migration at several
double-weir stations and measurements of photosynthesis and
respiration in various sections of the stream using changes in
oxygen concentration.
Physical and Chemical Data
Characteristics of the Sampling Stations
Nine sampling stations for oxygen and/or fish-movement analysis
were established on New Hope Creek and one on Morgan Creek. The
locations of the sampling stations are given in Figure 1, and some
characteristics of each station are given in Table 1.
Discharge
Current velocities were measured with an A. Ott (#I36241 'pigmy1
current meter, Total stream flow was determined by measuring the
rate of flow either in about. 12 points in a grid pattern in the stream
(during flood stages) or in the middle of four 38 cm pipes through which all water flows under the concrete bridge. The total discharge was calculated as the summation of each flow rate times the cross
sectional area represented by that flow rate. Daily stage measurements were maintained, and a graph of stage versus discharge was constructed Table 1. Some Characteristics of the Various Sampling Stations
Kilometers above Stream Bottom Station Erwin Road width Type
1. Way up 18.0 3.0 (a) sand and gravel
2. Horsefield 14.3 4.0 (a) sand and gravel
3. Blackwood 12.5 8.3 silt and sand and boulders
4. WL 10 10.6 lO.O(a) silt
5. Wood Bridge 6.0 boulders and gravel
6. Jungle 10.0 boulders and gravel
7. Big Pool 3.3 14.2 boulders and gravel
8. Concrete Bridge 3.2 5.1 hard rock
9. Pipeline (b) 5.0 (below) 15.0 silt and sludge
Morgan Creek ---- 4.0 sand and boulders
a. estimated
b. below outlet from Durham Sewage Treatment Plant 24
(Figure 5) from which daily flow rates were read. Discharge as m3 scwas computed as 0.02832 times discharge as cubic feet per second (cfs) .
Stream Morphology: Depth--- and Width".. Stream width and depth were measured at 50 or 100 m intervals for one or two km above each major oxygen sampling site. Each interval between the Concrete Bridge and the Wood Stations was marked off with a 50 m string. Intervals at other locations were determined by pacing off 100 m. A marked tape was stretched. across at each location; the width was measured and depths were taken at
1 m intervals. The average depth for each stream interval was computed as the arithmetic mean of all the depth measurements in that interval.
The average width and depth for the section of stream over which water flowed during one hour was calculated from:
where D is the discharge of the stream at that time in m3 hr-l, L is the length in m of each stream bed segment. Wn is the width in m of the stream at each successive sampling location (50 or 100 m),
Idn is the average depth in m at that location, and n is the total number of sample segments necessary for [(w,) (On) (L,)] to equal one hour's water discharge. The total length of n stream segments was the length of stream through which the water flowed in one hour.
Once this total length was found the average of all width and depth Figure 5. Water stage vs. discharge. Abscissa = stage level in inches (cm) above zero flow. Ordinate = discharge in 103 m3 day-l. The break before the last two values occurs as the stream overflows its banks. measurements in that interval was also found. These values weri used for calculations of stream metabolism.
Time intervals for water masses to flow between two points was computed from stream morphology as follows:
where t is the time, in hours, for the water mass to flow that distance, D is the discharge in m3 hr-l, is the mean width of the stream in the interval between the two points, and is the mean depth in that same interval. Time intervals were also, on one occasion, checked with dye. The turbulent and varied nature of the stream made this method difficult, since the mass of dye in the current traveled much faster than side eddies. The results were about
40 percent lower than the morphology method, but were not different enough to effect metabolic calculations.
Insolat ion
Estimates of relative amounts of insolation penetrating tree canopies at two regions of New Hope Creek were made with a Weston
M~del756 illumination meter'. This measured total incident sunlight in foot-candles.
Measurements were made on a completely cloudless day. Estimates of relative sun energy reaching New Hope Creek at different locations were made by sampling every 100 meters for a distance of 1 km above oxygen sampling sites. Each specific location was determined by pacing off approximately 100 m, then taking a reading at the center of the stream just above the surface of the water. Since the insolation was patchy, the light receptor was moved in an arc at armqs length and an average reading was taken.
In addition, total insolation, both in a clear field and under a hardwood canopy, was obtained with an Epply pyroheliometer from the International Biological Program site located in another section of Duke Forest (Blackwood division) which is about 200 m from the watershed of the headwaters of New Hope Creek (Figure 1, north of
Station 1).
Stream Temperature
Temperature on each sampling date was msasured, generally in the late afternoon, using a standard laboratory thermometer.
Diurnal temperatures taken with the oxygen-temperature recorder were corrected as explained in the section on "Metabolic Studies."
All diurnal temperatures above 5' C varied during the day. Since the later afternoon temperatures were, almost without exception, about
2" C warmer than the average temperature for the day, average tempera- tures for days on which diurnal temperatures were not run were com- puted as the late afternoon temperature minus 2.
Total Phosphorus in Water
All phosphorus and nitrogen analysis were based on FWPCA (1969).
Total phosphorus in stream water was determined using a Technicon
Auto-analyzer with a 660 mu filter and a 5 cm flow cell. Samples were collected in 100 ml polyethylene bottles to which 40 mg of Hg 2 F liter-' had been added as a preservative. The samples were frozen until analyzed.
Samples were digested in an autoclave with persulfate and sulfuric acid. Phosphorus analysis was by colorimetry following stannous chloride reduction and the formation of a phosphomolybdate complex.
Phosphorus in Organisms
Estimates of total phosphorus contained in several species of hardwood leaves following abscission (Woodwell, 1970) and in a mixed forest (Gosz --et al., 1970) were averaged to give approximate values (0.041 percent P dry weight) for leaves floating down New
Hope Creek.
Estimates of total phosphorus in fish were taken from values supplied by Vinogradov (1953) and Donaldson (1963). These were approximately 0.3 percent P by weight for many species of fish and
0.4 percent P for whole sockeye salmon, respectively. An approximate value of 0.35 percent wet weight was used for calculations in this thesis.
Total Nitroeen in Water
Total nitrogen was also analyzed on the Technicon Autoanalyzer.
Samples were taken from the same bottles as for P analysis, and analyzed colorimetrically following digestion with a sulfuric acid solution containing potassium sulfate and mercuric sulfate. The blue color measured results from the addition of alkaline phenol, sodium hypochlorite and sodium nitroprusside. Stream Conductivity The conductivity of water samples from New Hope Creek was determined with a Yellow Sprin.gs Instruments Company Model 31 conductivity bridge.
Discharge of Leaves Estimates of (dry weight) leaves flushed downstream were made from June 13, 1968 to June 12, 1969. During normal water levels the leaves that accumulated on the upstream side of the 1/4 inch (0.6 cm) hardware-cloth weir were removed every day or two. During flood levels, when it was impossible to maintain the weirs, leaves were sampled by holding a 50-foot (16.4 cm) fish seine with a mesh size of 0.4 cm across the stream for 15 minutes (sometimes less during exceptionally heavy flow). The weight of leaves moving downstream in 24 hours was calculated assuming constant flow. On days during which leaf discharge was not measured, estimates were obtained by reading values from the graph of water stage versus leaf discharge (Figure 6). This was possible because of the nearly linear relation of total leaves discharged to the water stage when plotted on semi-log paper. Although there was a tendency for greater leaf discharge for a given water level to occur during the fall, this was not sufficiently consistent to use seasonal correc- tions in reading the graph. The calculations made using these data did not require precise measurements. Figure 6. Total mass of leaves (dry weight) discharged per day in stream flow at Concrete Bridge Station vs. stage level
(ordinate) in centimeters above zero flow. 50 7 5 STAG E
Metabolic Studies
Entire ecosystems, like individual organisms, produce and use energy to maintain life. This process can be measured by determining the total amount of oxygen, or carbon dioxide, produced and consumed.
The following section describes how these gases rer re measured in
New Hope Creek and are used to estimate metabolism,
Dissolved Oxygen, Winkler Method
Estimates of community metabolism for New Hope Creek were made from diurnal variations in dissolved oxygen and pH. Oxygen was measured both by the azide modification of the Winkler method and by an automatic field temperature and oxygen recorder (Rustrak
Model 192) used with either a Yellow Spring Instrument #5419 probe or Rustrak #I921 probe.
The Winkler determinations were made following Standard Methods
(American Public Health Association, 1965). For the diurnal studies, samples of water were taken every two or three hours for
24 hours at standard stations. A simple tube device minimized oxygen diffusion from the air during filling of the 300 ml sampling bottle (Figure 7). The sampling bottle clamped to the end of a stick was held with incurrent tube about 15 cm below the water surface. Duplicate samples were taken within about two minutes of each other. All reagents were added in the field and titrated within
12 hours in the laboratory. Welch (1968) found no difference in duplicate oxygen samples when one was titrated immediately and the other 24 hours later. Nater temperatures were taken with a standard Figure 7. Cork and tubing device to fill oxygen bottle without air mixing.
3 3
laboratory thermometer. Percent saturation was calculated from the oxygen solubility values of Churchill --et al., (1962); they are intermediate to other values in the literature,
Dissolved Oxygen, Galvanic Probe Method
Oxygen concentration were measured at one station with an
automatic recorder installed in a streamside shed. The chief
advantage of this method was the tremendous savings in effort to obtain a diurnal curve. Only one hour or less per day was required to set up and standardize the instrument against Winkler determina- tions, read the chart, and enter the data on punch cards. A typical diurnal sequence using Winklers required about 28 hours. In addition, a continuous record was obtained so that non-typical water masses could be identified. One disadvantage of the probe was that only one station could be sampled on a given day with the equipment available. The membrane electrode may have been less accurate than
Winkler determination because of drift, however, since the probe averages oxygen values in various water masses flowing over it, it may be a truer representation of stream oxygen. Figure 8 shows typical results of O2 estimates using both methods. The maximum deviation in this case was only about 0.3 gm3, which is within extreme ranges of duplicate Winklers.
Particular care was necessary to avoid several sources of error inherent in the field recording unit. The probe was water- velocity dependent, and it was necessary to place the probe in water that had a velocity of at least 0.5 meter per second or the probe Figure 8. Comparison of probe and Winkler oxygen values over a 24 hour period, July 25, 1969,at Concrete Bridge Station. Open circles are average of duplicate Winkler samples, the range of which is represented by a vertical line. Data from galvanic probe and recorder are triangles connected by solid line. The maximum difference between the two determinations is 0.45 mg 1-I (g m-3), which is within the range of duplicate Winkler samples. would use oxygen faster than the water could resupply it. During periods of low flows rock jetties were constructed to concentrate the flow of the major part of the stream on the probe. The current velocity was assumed sufficient if manual movement of the probe in the water did not increase the reading on the meter. A mechanical agitator that increased flow over the probe was used during periods of extremely low stream flow.
Other potential sources of error were the corrections for effects of temperature on the probe. The manual that comes with the Yellow Springs Model 51 oxygen meter states that the temperature response of the oxygen probe is about 5 percent per degree centigrade. In other words, for each degree higher than the calibration temperature, the probe would read about 5 percent too high.
A check on this temperature effect was made by putting the probe and a thermometer into a container completely filled with water at several different oxygen levels. The container was then cooled with an ice bath while a magnetic stirrer kept a constant flow over the probe. Temperature and oxygen readings were recorded. As the temperature of the water dropped, the reading of the oxygen meter also dropped even though the oxygen content of the water remained constant, since the container was airtight and no diffusion of oxygen could occur. After the temperature of the water approached zero, warm water was put in the water bath and the temperature raised to the original value. Diffusion was checked by determining if the new oxygen reading at the original temperature was the same as the original oxygen reading, and small corrections were made. Sample results
obtained by this method are presented in Figure 9. Results fo-
changes in air readings with temperature were similar, The
temperature correction for the oxygen probe per degree change in
temperature is the slope of the line of the graph, which is ex- pressed as:
where Oc is the corrected oxygen reading in mg O2 rl, 0, is the uncorrected oxygen value, t is the difference in temperature from that at standardization, and S is the daily average saturation value.
This correction was entered into the computer algorithm.
Temperature effects on the recording unit were checked by putting the whole unit in a refrigerator while leaving the probe at room temperature. No changes in reading occurred. The chart was read as percentage of full scale, and the following equation was used to calculate oxygen concentrations (mg 1-I) where 0, is
1 oxygen concentration, in mg 1- at time t, Oc is oxygen concentration, in mg 1-I at calibration time c, as determined by Winkler titration,
Ct is chart reading, in percentage of full scale, at time t and Cc is chart reading, in prcentage of full scale, at 1-hr: time of c'yygen calibration. Since the reading of the oxygen recorder is linearly proportional to the concentration of oxygen in water (Gulton Figure 9. Variation in oxygen meter readings with constant dissolved oxygen and varying temperature. Laboratory determination was done May 11, 1969, The slope of this line represents the temperature correction necessary to get true readings at temperatures other than that at which the probe was calibrated. Calibration was at 24' C and at oxygen saturation for room temperature (8.33 mg 02
1)Triangles represent descending temperatures, circles represent rising temperatures, and points represent a later decrease to room temperature. The slope at oxygen values less than saturation was in- versely proportional to the percent saturation.
Industries, Bulletin no. M26802), the above simple proportion will give the oxygen concentration mg 1-I (g ~n-~).Comparison of temperature-corrected scale readings and IYinkler oxygen values over a period of several days indicates that drift was relatively small (Table 2). Table 2. Drift sf Oxygen Recorder Over One or Several Days
Numb er Observed reading, Reading if of temperature there were Diffesence days since corrected no drift initial -1 -1 Date calibration mg 1 mi3 1 mg 1 -1
June IS 1 7.47
July 21 2 6.73
July 25 4 5.67
April 23
April 25
May 29
Diffusion Rates
Oxygen movos from air to water and water to air according to
Dalton's law of partial pressures. Corrections must be made in aquatic metabolic studies for this. It is necessary to know both the percent oxygen saturation of the water and the diffusion constant to make these corrections.
Three months were tried for determining the diffusion constant on New Hope Creek: the diurnal curve method (Odum, 1956;
Odum and Hoskins, 1958), the stream morphology method (Churchill --et al., 1962) and the dome method (Hall and Day, 1970), The diurnal curve method gave results often much higher than the other two methods and was not used for this study. The reason for the high values was that nighttime respiration was not constant, a pre- requisite for the accurate use of this method (Odum and Wilson,
1962; Owens, 1969).
For this study the stream morphology method, which averages pool and riffle values, was used to determine the diffusion constant; this constant varied from day to day as the water level changed (Figure 10).
The dome method gave similar results, since New Hope Creek is about equally divided between pools and riffles. A more complete treatment of the diffusion studies is given in Appendix A. Figure 10. Comparison of different diffusion constants obtained in this study. Crosses represent estimates based on the stream morphology method; triangles represent estimates made with dome method; and circles represent estimates made from diurnal curves selected for substantial diurnal range. The stream morphology method was used for calculations made for this study. See text for further explanation.
Gross Community Metabolism
Primary production and total respiration of the living organisms in New Hope Creek were measured using diurnal variations in metabolic gases (Odum, 1956; Odum and Hoskins, 1958; Odum and Wilson,
1962; Beyers --et al., 1963). The basis for these measurements is the fundamental equation for photosynthesis (or respiration):
Thus, the total creation and utilization of organic compounds is pro- portional to the amount of C02 and 02 being produced and consumed.
Some variations in the relation of oxygen to energy occur when proteins or fats are being utilized instead of sugars, or when there are lags in one process relative to another. For these reasons it is most accurate to consider the metabolism in terms of oxygen without converting to carbon or caloric values.
Two Station Analysis
The most accurate estimates of photosynthesis and community respiration can be obtained for a stream with the "two station'? method of oxygen analysis (Odum, 1956; Owens, 1969). The two station analysis is based on the actual change in oxygen occurring as a water mass flows from one region of the stream to another. Thus, changes over a clearly defined area can be measured and rates of change determined from differences between the upstream and downstream oxygen measure- ments of the same water mass.
This method was generally impractical on New Hope Creek because of the necessity of sampling at three to five locations that were too far apart. However, on one occasion two station analysis was run run on New Hope Creek at both the Blackwood Station and the Concrete
Bridge Station. Tho results (Figure 11 and 12) indicate that in these short stretches of Mew Hope Creek, equivalent to the distance water flow in about one hour, the metabolism of one part is similar to another.
-.Single Curve Method Where upstream and downstream diurnal curves are similar, one may use a single station curve as a first approximation (Odum,
1956). This procedure was used for New Hope Creek. The basic procedure in estimating stream metabolism by this method is to measure oxygen and temperature in the field every two or three hours using either Winkler oxygen methods or a galvanic probe, either with or without recorder. The data are then plotted
(Figure 13). The first derivative is computed for these oxygen changes and again plotted using the same time scale. Corrections for diffusion are made by adding the product of the diffusion constant and the saturation deficit to the rate-of-change curve.
If there were no biological or chemical activity in the water being studied, there would b'e only the change in the oxygen con- centrations over the day due to temperature changes affecting saturation values. The rate-of-change curve would be near zero for the entire day. However, biological respiration tends to it,wilr the oxygen in the water throughout the day and night, and the photosyn- thesis of green plants raises the oxygen during the day. Thus, a Figure 11. Similarity of oxygen curves one hour's flow distance apart at Blackwood Station, February 14, 1970. Winkler determinations were done in duplicate. The triangles are oxygen concentrations at the upstream station; the solid line connects their averages. The circles are oxygen concentrations at the downstream station, and the broken line represents the average of these. MN is midnight.
Figure 12. Similarity of oxygen curves one hour's flow distance apart at Concrete Bridge Station, February 14, 1970. Symbols and lines used are same as for Figure 11.
Figure 13. A representative sample of single station analysis for community metabolism in New Hope Creek, February 14, 1970, as conducted and plotted by the UNC CALCOMP plotter. The upper graph is the average of duplicate Winkler determinations taken every three hours. Each triangle represents a single sample. The second graph is of temperature taken every three hours. The third graph is the percent saturation of the average of the two Winklers at the temperature of the sample. The lower graph shows the rate of change
(first derivative) of the oxygen samples. The line with the tri- angles shows the rate of change corrected for diffusion of oxygen across the surface of the water. The gross photosynthesis of the water mass represented by these water samples is indicated by the area stippled. Gross community respiration is estimated as the area cross-hatched. The diffusion constant is in g O2 rn-3 hr-I atmosphere-', and the depth is average depth in meters for one hour's flow distance above sampling station.
characteristic rate-of-change curve is produced, rising during day-
light and falling at night, often leveling as the amount of oxygen
that diffuses in equals the amount of oxygen being used by respiring
organisms (Figure 13).
Although daytime respiration tends to lower the amount of oxygen
in the wa?er, this is masked by the increase in oxygen caused by photosynthesis. Thus, a real mcasure of respiration during the day-
time is impossible by this method. In order to overcome this
difficulty, it was originally suggested (Odum, 1956; Odum and
Hoskins, 1958) that da.ytime respiration shoul d be approximately
equal to nighttime respiration, and that a line drawn on the rate-
of-change curve at the average nighttime respiration rate would
approximate daytime respiration (Figure 14a). Further refinement of this method (Odum and Wilson, 1962) takes into account the varying nature of daytime respiration which is greater toward the end of the
day when temperatures and oxygen levels are higher. Thus, a sloping
line drawn from the pre-dawn low point on the rate-of-change curve to the post-sunset minimum (Figure 14b) is probably a more accurate
representation of what is occurring in nature. In almost all curves
analyzed for this study, the.post-sunset rate-of-change point is lower than the pre-dawn point, indicating greater respiration during the latter part of the day.
Further studies (Sollins, 1969; Odum, Nixon and DiSalvo, 1970) have indicated that daytime respiration may be considerably higher due to higher oxygen levels and photorespiration. Thus, the actual daytime respiration curve may dip down considerably as suggested in Figure 14. Various lines drawn to represent daytime respiration: a. constant daytime respiration at the level of average nighttime rates (from Odum and Hoskins, 1958); b. varying daytime rates similar to varying nighttime rates (Odum and Wilson, 1962); c. hypothetical curve assuming respiration proportional to oxygen concentration (Sollins, 1965); and, d. hypothetical curve correcting for photorespiration (Odum, Nixon, and DiSalvo, 1970). Corrected rate-of-change curve from Wood
Bridge Station, New Hope Creek, October 4, 1968. Daytime respira- tion as represented by line b was used in the present study. G 12 TIME Figure 14 c and d. This oxygen consumption is obviously compen- sated for by a greater amount of oxygen being concurrently pro- duced by photosynthesis, as the oxygen level in the water rises during the day. Thus, community metabolism during the day may be considerably greater than during the night. However, until some adequate means for measuring photorespiration becomes available, the method of connecting the pre-dawn point by a straight line to the post-sunset point is, at least, objective and may closely represent all community respiration except photorespiration in green plants. This procedure has been used for all analysis of
New Hope Creek data.
The gross community respiration was estimated by integrating the area between the zero rate-of-change line and the diffusion- corrected respiration curve over 24 hours (Figure 13). Gross photo- synthesis was estimated by integrating the area between the daytime respiration line and the daytime rate-of-change curve (Figure 13).
The integrations can be done by counting squares on graph paper or by using a planimeter.
Estimate of Pdetabolism from pH Changes
An estimate of community metabolism was made using the diurnal pH method (Beyers --et al., 1963). The production of carbon dioxide by the respiration of living organisms produces carbonic acid by the following formula:
I C02 + H20 $ H2C03 ,(other carbon compounds) Thus, respiration lowers the pH of the water, and photosynthesis, by taking C02 out o-C fhe water, raises the pH. Since the inter-
action of various ca7:bon compounds in natural waters is extremely complicated and subject to unknown buffering, -a --priori coordination of pH and amounts of C02 produced or utilized is virtual-ly
impossible. However? this relation can be determined empirically by
titrating the water of interest with distilled water of known C02 concentration (Beyers --et al., 1963). A sample titration of New Hope Creek water with carbon dioxide-saturated distilled water is
supplied (Figure 15).
The change in relative amounts of carbon dioxide in the water can then be determined by reading the pH-CQ2 graph. To determine
absolute values of C02 in the water requires separate determinations of total C02 at the start of titration. However, this is not necessary since the metabolic determinations are based on changes in
C02, not on absolute values.
Estimates of total production and respiration from changes in
C02 were made by a procedure similar to that used for oxygen. Plots were made of relative amounts of CQ2 in the water over 24 hours
(Figure 16). The first derivative of this was plotted as a negative function to make the results compatible with oxygen data which, of course, behave in an opposite fashion. Daytime respiration was estimated according to the methodology discussed in the previLjli section, and total photosynthesis and respiration were determined by integrating the same areas as discussed for oxygen. A sample determination is included (Figure 16). No correction for diffusion Figure 15. Carbon dioxide titration of New Hope Creek water for metabolic studies. The abscissa represents the carbon dioxide in the water sample added to that present at the start of the titration.
Figure 16. Estimation of metabolism in New Hope Creek, Diurnal pH method. February 21, 1969. The upper graph represents pH over
24 hours in New Hope Creek, also changes in the carbon dioxide content
of the water. The lower graph is the rate of change based on the above
carbon dioxide values. Gross photosynthesis and community metabolism
are estimated as in Figure 13. No corrections were made for diffusion
of C02, but the results of this method (Gross production = 1.33
gm m-3 day-', Respiration = 2.0 gm m-3 day-l) agree fairly well with uncorrected-for-diffusion oxygen estimates of 1.5 and 2.2 gm mw3 day-l respectively. GO%/ LITER
C SUNRISE was made, but outward diffusion of C02 at night may increase the estimated community respiration. - Computer Program for Estimating Community Metabolism from Diurnal Oxygen Curves 4
Since the calculations involved in these determinations are long, tedious, and subject to human error, use was made of the International
Business Machine M~del70 digital computer and Calcomp plotter, located at the Tri-University Computation Center and the University of North Carolina, respectively. These are linked by cable and are completely coordinated. Appendix B includes a flow chart of the program, the computer program, in PL/1, and instructions for entering data.
Fish Sampling Procedures and Apparatus
Design of Weirs and Traps-
Weirs to measure movements of stream fishes were constructed following the basic plan of Shetter (1938). Since many different locations were sampled over a considerable time scale, variations in the basic setup occurred with different bottom types and evolution of design. A summary of the different modifications used is given in
Table 3.
The basic plan for the 'hardware-cloth" (wire screening) weir
(Figures 17 and 18) was to stretch a 0.6 cm (1/4") mesh barrier across the stream at an angle such that migrating fishes would be funneled into traps placed at either side of the stream. The lower 30 cm or so of the screening was bent at a 90 degree angle to the vertical and placed on the rock-cleared stream bottom. Rocks were then placed on Table 3. Modificatians in Basic Trapping Procedure
Date Locat ion New Modification
April 10, 1969 Concrete First day sampled. Downstream only.
April 15, 1968 Concrete Upstream and downstream traps installed.
June 5, 1968 Wood Station Upstream trap only installed
June 14, 1968 Jungle Station Upstream trap installed
October 22, 1968 Jungle Station Upstream and downstream traps installed
February 12, 1969 Big Pool Station ~arge-sized(66X 132 X 132 cm) trap inst a1led (down) . May 6, 1969 Wood Station Upstream and downstream trap installed
September 21, 1969 Big Pool Station Pipe-weir instal led
February 21, 1970 Big Pool Station Large-sized traps installed up and down
Figure 18. Design of fish weirs used in New Hope Creek. a. Arrangement of weir and traps. b. Hardware cloth weir. c. Steel pipe weir with wooden supports, plastic spacing collars and hardware cloth seal with bottom of stream.
the top of this screening, and the entire boundary of the screen and stream bottom Iws checked for holes through which fish miphf: pass.
In the summer of 1969, a more elaborate weir was cmstructcd
(Figures 18 and 19). T??is weir was designed to overcome the pro- blem of leaves accumula.tjng in the upstream side of the hardld~are cloth weir during high water. Steel electrical condirit pipes spaced at 1.5 cm intervals by plastic collars were used following the recommendations of Feimers (1966).
A fish-tight border with the bottom of the stream was created by bending a 1/2 m wide section of hardware cloth over the lowermost pipe and by piling rocks on either side of this. The pipes were held in position by cement-anchored wooden frames that allowed the pipes to be inserted and removed. Hardware cloth trap-entrance cones provided entrance to the actual traps.
Traps were placed at both ends of the weir; one designed to catch fish moving upstream and the other downstream (Figure 18).
Frames for the traps were constructedof2 b2cm by 66 or 132 cm pieces of aluminum (Figure 21). This was covered with 0.6 cm wire cloth to form a rectangular trap. A cone was constructed at one end that fitted in a corresponding cone in the weir to form a tight fit that could be easily separated for fish removal. The other end of the trap contained a 10 cm by 10 cm spout with door to facilitate fish removal. A larger door was also included. Figure 19. Big Pool sampling station, looking downstream during normal spring flow. The trap to the left center of the picture catches fish moving upstream and the trap to the right center catches fish moving downstream. The far right bank of pipes has been lifted to show underwater arrangement.
Figure 20. Y3idewaysf1 fish sampling arrangement. The entrance cone to the trap is visible in the center of the trap. Also visible are the plastic pipespacing collars and the method of anchoring the pipe supports with rocks and concrete.
Figure 21. Design of fish trap used in New Hope Creek. The frame is made from 2 by 2 cm by 66 or 132 cm sections of aluminum.
The screening is 1/4 inch (0.6 cm) hardware screen, heirl to the frame with aluminum "pop" rivets. A cone of wire screening on the weir fits inside the similar cone in the trap. The small box is a spout for removing small fish.
Check on Possib1.e Sampling- -..--- Bias in Up and Down Traps- The consistently greater catch in fish mass moving upstream com- pared to fish mass moving downstream raised the possibility of sampling bias--that fish enter the upstream trap more readily than the downstream trap. This was considered unlikely, as a certain additional amount of effort would have been needed by the fish to swim up into the trap; whereas, a fish entering the downstream trap could drift passively with the current. However, this possibility was checked by two means, the "sideways" trap and the "double reverse1' trap (Figures 20 and 22). The sideways trap was designed so that fish entering from either upstream or downstream would have to enter the actual trap sideways to the current. This was con- sidered the principal check on trap bias. In theory, if there were substantial error introduced by the facing of the weirs this would become apparent by extreme variations from the expected ratio of fish moving upstream to fish moving down. However, during the use of these traps about the same ratio of upstream to downstream move- ment occurred as during normal sampling at similar times of the year (Table 4).
A further check on trap bias was the use of the "double reverse" traps in which fish moving upstream would have to turn around and enter the trap moving downstream, and --vice versa. No large fish at all were caught moving ~~streamduring the use of these traps, and only a few were caught moving downstream even during periods of ex- pected fish movement. Thus, perhaps the ultimate test of trap bias Figure 22. a. "sideways" and b. ffdouble reverse" weirs used to test possible sampling bias in normal trap arrangement.
Table 4. Catch of Fish in 'Sideways1 Traps
Date Number up Number down bQass up Mass down
1970
April 23
April 24
April 25
April 26
May 18
May 19
May 20
May 26
Totals 8 9 61 3052 1674 failed and the possibility of trap bias remains, although the results of the l~sideways"traps indicates that this is unlikely.
Check on Rate of Fish Escape from Traps- The trapping procedure used in this study is based on the assumption that it is much easier for a fish to swim into the traps than to swim out of them, The cone entrances make this seem intuitively true. However, this was checked by a series of 18 sets of experiments in which fish that were already caught in the traps were left in them for an additional 24 hours. Thus the rate of trap escape was measured.
A total of 132 fish weighing 10,364 g was placed in the traps.
Of these, 96 fish weighing 7747 g were recovered. Thus, over the
24-hour interval, 36 fish weighing 2517 g escaped. Since each fish originally trapped would have been in the traps an average of 12 hours (one-half the trapping time interval) it was assumed that the escape rate for a normal day would be half of the 24-hour escape rate.
Correcting the above data for this gives an average daily escape rate of 13.6 percent for numbers of fish and 12.3 percent for mass of fish. No important difference in escape rates for different traps was noted except that the smaller (66 cm by 66 cm by 132 cm) trap, when used to catch fish moving downstream, had a greater escape rate than other traps. This trap was replaced in February, 1969, by a larger one from which virtually no fish escaped. Since the escape rates kqeie not large, no corrections were made in the data used for analysis.
Sam~lin~Modifications for Low Water
During the summsr of 1968 and the summer and fall of 1969 very low rainfall conditions existed in central North Carolina, and extremely low flows in New Hope Creek resulted (Figure 23). However, large pools remained throughout the stream; and if the run of fishes captured in the spring of 1969 is acy indication, the fish populations were not adversely affected. During these periods the Big Pool weir was no longer functional because the water dropped below the level of the trap entrances. Sampling was conducted only at the Concrete
Bridge site, where adequate water depth was present.
Sampling Modifications--.-- During High Waters New Hope Creek was subject to extreme flooding conditions during the fall, winter, and spring months (Figures 24 and 25). The magnj- tude of some of the flows made any sampling of migration impossible; however, over the course of this project various sampling modifications were made to increase the level at which fish counts could be made.
Table 5 gives the maximum flow at which sampling could be maintained during the study and the dates at which sampling was impossible be- cause of higher water.
Initially the height of the weir was increased with additions of wire cloth. This reached a practical limit at a stage level of about
46 cm above zero flow, and even less during periods of heavy leaf flow.
In February, 1969, the sampling station was moved to a wider portion of the stream so that a given increase in water flow would cause less of an increase in water height. Thus sampling was initiated at the Big Pool site, located about 100 m above the Concrete Bridge Figure 23. New Hope Creek during drought (September, 1968).
a. Riffle area at Wood Bridge sampling station. b. Pool area below Wood Bridge Station.
Figure 24. New Hope Creek at Concrete Bridge Station during flood. a. shows riffle area just above fish sampling station
(April 14, 1970). b. shows large pool between the concrete bridge and the Concrete Bridge sampling station (March 21, 1970).
Figure 25. a. Overrun of weir during severe flood at
Big Pool Station on April 14, 1970. b. Destruction of trapping apparatus, April 14, 1970.
Table 5. Floods in New Hope Creek That Affected Sampling
Date of start Maximum stage level Maximum stage that Day weir of flood in centimeters above could be sampled inoperable zero flow at that time in cm above zero flow
May 27 Oct. 19 Nov. 12 Nov. 19 Dec. 4
Jan. 21 Feb. 9 Feb. 28 March 3 March 18 March 24 June 16 Aug. 6 Oct. 2
Jan. 30 Feb. 2 Feb. 17 Feb. 25 March 5 March 21 March 31 April 14 site. This provided a means of sampling to a stage level of about
76 cm, if leaf flow was small. During high waters a series of rock and wire supports were utilized to maintain the weir in an upright position. Removal of fish and general maintenance were performed two, three, or four times each day during flows that approached the limit of sampling. During the fall of 1969 the entrance cone pointing downstream was lined with a stiff clear plastic sheet so that clogging of this cone with leaves would not occur.
The final major modification to aid in sampling during high water was the construction of the pipe weir discussed previously.
This allowed sampling to a stage level of about 76 cm, even during periods of moderately heavy leaf discharge. Sampling was impossible during higher stage levels, and the traps and pipes were removed to avoid total destruction of the apparatus. Floods of this magni- tude rarely happened during periods of expected heavy migration.
Methodology of Handling Species and Species Groups for Analysis
Due to some uncertainties in taxonomy during the beginning of this study, and to the necessity of simplifying the large amounts of data for analysis, all organisms were placed into one of 21 taxonomic groups as listed in Table 6. All taxonomic groups of more than one species are listed below.
pickerels sf^ included both chain pickerel (Esox-- niger) and redfin pickerel (Esox- americanus), but the redfin pickerel was captured only as a rarity. Flat bullhead (Ictalurus platycephalus) has recently been subdivided into two species, -I. platycephalus and -I. brunneus. They are considered as one species for this study, although both Table 6. Fishes Captured in New Hope Creek and Groupings Used to Simplify Analysis.
Analyzed as: Common name a Scientific name a
B. crappie Black crappie Pomoxis -nigromaculatus (~eseur)
Bluegill Bluegill Lepomis macrochirus Rafinesque
B. H. Chub Bluehead chub Hybopsis -leptocephalus (Girard)
Bullhead Flat bullhead Ictalurus platycephalus (Girard) Snail bullhead Ictalurus brunneus (Jordan) Chubsucker Creek dhubsucker Erimyzon oblongus (blitchi 11)
Creek chub Creek chub Semotilus atromaculatus (Mitchill)
Darter Johnny darter Etheostoma nigrum Rafinesque Piedmont darter Percina crass a (Jordan and Brayton) H.F. shiner Highfin shiner Notropis altipinnis (Cope) Madtom Margined madtom Noturus insignis (Richardson) Pickerel Chain pickerel --Esox niger (Le~eur) Redfin pickerel Esox americanus Gmelin Redhorses Smallfin redhorse Moxostoma robustum (Cope) V-lip redhorse Moxostoma collapsum (Cope) Pirate perch Pirate Perch Aphrododerus sayanus (Gilliams) a Nomenclature used follows A--- List of Common -and Scientific Names -of Fishes --from the United States -and Canada, 1960, American Fisheries Society publication No. 2., Waverly Press, Inc., Baltimore,l02 p. Table 6. Continued
Analyzed as: Common name a Scientific name a
Pumpkinseed Pumpkinseed Lepomis gibbosus (Linnaeus)
sunfish Redbreast sunfish Lepomis auritus (Linnaeus) shiner White shiner --Notopis albeolus Jordan Sandbar shiner Notropis scepticus (Jordan and Gilbert)
M. shiner Whitemouth shiner Notropis alborus Hubbs and Raney Others American eel --Anguilla rostrata (LeSueur) Bow fin --Pmia calva Linnaeus Gizzard shad Dorosoma cepidianum (LeSueur)
Green sunfish Lepomis cyanellus Rafinesque
Largemouth bass Micropterus salmoides (Lacepede)
Speckled killifish Fundulus rathbuni Jordan and Meek are common in New Hope Creek. 'Redhorses' include both v-lip redhorse (Moxostoma collapsum) and the smallfin redhorse (Moxo- stoma robustum). About two-thirds of the redhorses captured were v-lip. The patterns of movement of both species were not obvious- ly different.
'Larger Notropis~ncludedseveral species that were morph- ologically quite similar to an untrained eye. The most abundant of these was the white shiner, Notropis albeolus although the sandbar shiner, Notropis scepticus, and probably several other species, were also taken. Darters were generally Johnny darter,
Etheostoma nigrum, although some Piedmont darters, Percina crassa, were also captured.
'Crayfish1 included members of up to four species listed as being present in New Hope Creek (Hobbs, mimeographed). No attempt was made to separate these into species. 'Frogs' included several species, and 'Turtles' included four species. 'Others' included largemouth bass, Micropterus salmoides; green sunfish, Lepomis cyanellus; bowfin, --Amia calva; gizzard shad, Dorosoma cepedianum; and snakes, wooddrnks, muskrats, large bugs and various other organ- isms. Table 7 gives an analysis of 'otherlorganisms encountered.
Daily Fish Sampling Procedure
The fish were removed from the traps and individually weighed.
Smaller fish were marked by a fin clip distinctive for each station.
Fish larger than about 80 g were tagged with individually numbered
dart tags (Floy Tag and Manufacturing Company). The live fish were
returned to the water about 50 m upstream or downstream from the weirs,
in the direction they were moving. Table 7. Organisms Other Than Fish Captured in New Hope Creek
Common name Scientific name Number captured
Dragonfly nymph Order Odonata 1
Walking stick Anisomorpha.- --sp. 1 Water bugs 2 Crayfishes Cambarus -spp . 446 Procambarus -sp. Newt Notopthalmus viridescens 2 Toads --Bufo spp. Snapping turtle Chelydra serpentina
Painted turtle Chrysemys picta a Mud turtle Kinosternon -sp . Musk turtle Sternothaerus -sp. Black (Racer) snake Coluber constrictor
Common water snake Natrix sipedon
Wood duck Aix sponsa P
a. 62 turtles of the four types
RESULTS
Physical Data
Stream Morphology
New Hope Creek, in the one km stretch above Concrete Bridge
Station, averages 11.6 m in width and 0.45 m in depth in a normal spring. Measurements taken at this and other stations are given in Tables 8-11. From these measurements it becomes apparent that the greatest depth in New Hope Creek, at least in the areas sampled, is above the Wood Bridge, and the least depth is above Blackwood
Station. All depth values taken on different dates were corrected to 'standard water levelt (50 cm above zero flow), which was the nor- ma1 spring flow.
Stream Level and Discharge Rate
Stage level at the Concrete Bridge Station varied from a minimum of 0 cm above the level of no flow to a maximum of more than 100 cm above the level of no flow. In springtime normal stage levels were about 50 cm. Figure 26 gives daily water levels for the 27 months of this study.
Daily discharge rates varied from summer and drought level- ~)f
0 m3 daym1 to spring flood levels of at least 7 lo5 m3 day-' or
251 cubic feet second-l. Table 8. Depth and Width Profile for 300 m Below Concrete Bridge,
May 23, 1970
Meters below Mean station Width depth m m
50 14.9 0.55
100 10.9 0.37
150 8.4 0.26
200 4.3 0.23
250 5.7 0.37
300 11.8 0.45
Average 9.33 0.45
Correction to standard flow (See text for explanation) - Average depth at standard flow 0.54 Table 9. Depth and Width Profile for 1.8 km Above Concrete Bridge,
Station, April, 1969
Meters above Mean station Width depth m m
Average 0.445 at standard flow Table 10. Depth and Width Profile for 900 m Above Wood Bridge
Station, May 13, And 23, 1970
Meters above Mean station Width depth
Average
Correction to standard flow . 0.07 - Average depth at standard flow 0.68 Table 11. Depth and Width Profile For the Zone 1000 m Above Blackwood
Sampling Station, May 18, 1970
Distance above station Width Average depth m m
Average 7.1 0.23
Correction to standard flow 0.05
Average depth at standard flow 0.28 Figure 26. Daily water stage level, in cm above zero flow, of
New Hope Creek at Concrete Bridge Station.
Stream Temperatures
The minimum temperature recorded in New Hope Creek was 0" C during January, 1969 and 1970. A maximum temperature of 28" C was recorded on July 21, 1969. Average daily temperatures for the study period are presented in Figure 27.
Light Intensity At Surface of Stream
The intensity of light striking the surface of New Hope Creek was measured on a completely cloudless day at 100 m intervals for
1 km above both the Blackwood Station and the Concrete Station
(Table 12). The results indicate that when leaf canopy is full about 11 percent of the solar light energy enters the aquatic ecosystems at each station with ranges from 9 to 66 percent of that above the forest. The results at the Blackwood Station may be biased by the one very large value (6080 foot-candles) which was not representative for that stretch.
Leaf Discharge
Measured daily leaf discharge at the Concrete Bridge Station varied from zero to about 825,000 g. The amount discharged per day was linearly proportional to,water stage when plotted on semilog paper (Figure 6). Estimated total monthly leaf discharge is given in Table 29, found in the "Discussion."
Total Phosphorus
Water samples analyzed for total phosphorus indicate that New
Hope Creek has about the same amount of phosphorus (Table 13) as many other freshwater environments summarized in Hutchinson (1957). Values ranged from not detectable (less than 0.005 ppm) to 0.26 ppm (or mg 1-l) . Figure 27. Mean daily temperatures for New Hope Creek during this study. q- P- *,La *-4 LA-
ti:. ti:. Table 12. Light Intensity at Surface of New Hope Creek
Meters Light Date above Intensity and oxygen sampling foot- Time Location station candles
Sept. 26, 1969 Open Field: Start 11:40-12~18 Blackwood Stat ion
100
200
300
400
500
600
700
800
900
1000
Average
Open Field: Finish
Sept. 26, 1969 Open Field: Start 1*3:!?0-14:OO CxLcrete BY, lge Station
Table 13. Total Phaphorus (Dissolved and Suspended) in New Hope
Creek at Concrete Bridge Station
Date Water stage Total phosphorus (ppm) Manual Autoanalyzer
1968
Aug. 29 5
Sept. 18 1
Sept. 23 0
Oct. 25 18
Nov. 9
Nov. 24
Dec. 16
Dec. 22
Jan. 8 3 3
Jan. 20
Jan. 24
Jan. 28
Feb. 3
Feb. 4
Feb. 9 Table 13. Continued
Date Water stage Tot a1 phosphorus (ppm) Manual Autoanalyzer
1969 (Continued)
Mar. 3
Mar. 5 0.270
Mar. 7 0.050
Mar. 8 0.030
Mar. 10 0.020
Mar. 16 0.020
Mar. 19 0.060
Mar. 21 0.100
Mar. 26 0.030
Apr. 2 0.020
Apr. 6 Not detectable
Apr. 10 0.050
Apr. 11 0.020
Apr. 17 0.060
Apr. 20 0.020
Apr. 25 0.250
May 8 0.130
May 14 0.060
May 18 0.060
June 4 0.060
June 11 0.160
July 2 0.040
July 13 0.080
Normal values were in the range of 0.02 to 0.1 ppm and the average for all samples taken was 0.06 ppm. No seasonal trend was evi- dent and samples often varied widely from one sampling date to the next. Nor was there any consistent relation between river discharge and phosphorus concentrations.
Nitrogen
Several total nitrogen analyses were done and the data are presented here for possible use as base line data (Table 14).
The ratio of N to P varied from 6.5:l to 210:l.
Stream Conductivity
The results of three stream conductivity determinations at different times of the year were from 63 to 200 um ohms.
Metabolic Studies
The results of all metabolic studies are given in Tables 15 to 17 and in Figures 28 to 36 and in the Appendix.
Daily Variations in Oxygen
On all days on which oxygen was measured the oxygen showed some manifestation of the expected diurnal curve, that is, it rose during the daylight hours and dropped at night. Within this general pattern many variations were observed (Figures 28-33; Appendix C).
Annual Variations in Metabolism
Photosynthesis in New Hope Creek at the Concrete Bridge Station, which was the station most heavily sampled, varied from about 0.21 m N 0 ygggggg OPN.z. to 8.85 g 02 mm2 day-l (0.58 to 10.88 g O2 mm3 dayml). Gross community respiration varied from 0.39 to 13.40 g O2 mm2 day-l
(0.94 to 16.30 g O2 rn-3 dayw1). Typical diurnal curves for the
Concrete Bridge, Wood Bridge, and Blackwood stations for times of greatest and least metabolism (spring and late fall or winter) are given in Figures 28-33. Annual results in which the same months for different years are lumped, are plotted in Figures 34-36.
Thus New Hope Creek has an annual cycle of metabolism that repeats fairly consistently from one year to the next. Both photosynthesis and respiration are least in the winter. Primary production increases as the season progresses, reaching a peak in March and April when light striking the surface is also maxi- mal. Respiration follows a similar pattern but remains high throughout the summer. Very high respiration is associated with high productivity and/or high temperatures and low water. March
1970 had a greater metabolism than at any other time studied, and the highest metabolism recorded (March 13) was associated with a small flood. Tables 15-17 give values for each day sampled.
Spatial Variations in Metabolism
On all days studied, volume metabolism was greatest at the
Blackwood Station. Except during the later summer, the /'bod Bridge
Station has the least volume metabolism (Tables 15 - 17). Areal values were generally similar on any one date at all stations. The relatively high areal production and respiration at Wood Bridge
Station during the summer may be a result of an erroneous water Figure 28. Typical diurnal oxygen curve for spring, Concrete
Bridge Station.
Figure 30. Typical diurnal oxygen curve for spring,
Blackwood Station.
Figure 31. Typical dirunal curve for late fall, Concrete Bridge
Station. SU?iRlSE. * SUNSET .--,--- .--,--- - I k'-"- Y--f------F-- .00 5.00 9.00 33.0 17.0 21.0 Figure 32. Typical diurnal curve for winter, Wood Bridge
Station.
Figure 33. Typical diurnal oxygen curve for late fall, Blackwood
Station. ' CCR.
.O
N CO 0
G-l d 0
a Ln N
N Ln
A
N M 0
a
0, d Table 15. Continued
Wink1 er Gross Community Gross Community or Depth product ion respiration product ion respiration Date probe m g 0, m-3 day-' g O2 m-3 day-l g 07 m-2 day-' g 07 m-2 day-'
July 27 P 0.45 0.99
Aug. 25 W 0.37 0.77
Oct. 3 P 0.65 1.23 4.28
4 P 0.44 1.36 4.15
5 P 0.41 0.74 2.54
Nov. 16 P 0.35 1.22 3.34
21 W 0.36 0.60 1.58
1970
Feb. 14 W 0.40
Mar. 13. P 0.50 6.30 7.15 3.15 3.58
Figure 34. Annual variation in metabolism, Concrete Bridge
Station, New Hope Creek, April, 1968 - May, 1970. The solid line ' connects means of gross photosynthesis for each month and the broken line connects means of community respiration for each month. The vertical bars are 1 standard deviation from the mean. The horizontal axis is months.
Figure 35. Annual variation in metabolism, Wood Bridge Station, New Hope Creek, June, 1968 - August, 1969. The solid line connects measurements of gross photosynthesis and the other line connects measurements of community respiration.
Figure 36. Annual variation in metabolism, Blackwood Station,
New Hope Creek, February, 1969 - February 1970. The solid line connects measurements of gross photosynthesis and the other line
connects measurements of community respiration.
depth, since during low waters the water does not flow rapidly enough to be influenced by the deeper'upstream water which was included in estimates of average water depth. When no flow at all was present, mean depth measurements were taken in the pool sampled itself, eliminating this source of error.
Annual and Spatial Variations in P/R Ratio
New Hope Creek at the Concrete Bridge Station exhibits an annual variation in the ratio of photosynthesis to respiration
(Figure 37). The stream as a whole is more autotrophic in the
spring and becomes increasingly heterotrophic during the summer.
Only rarely was the stream running entirely upon energy produced totally within its boundaries.
Fish Movements
The data for fish movements are presented in Table 18 to 25 and Figures 38 to 41, and in Appendix D.
Analysis of All Species Considered Together: Principal Sampling Station
A~ril1968 - June 1970 During the 27 months (785 days) of this study, fish movement was sampled on 455 days. During this period 6,043 fish and other organisms were captured in the traps at the principal sampling station, 2,655 moving upstream and 3,379 moving downstream, for a daily average of
5.8 organisms moving upstream and 7,4 moving downstream. The live mass (live weight) of the organisms moving upstream was 187,927 g and moving downstream was 93,092 g for a daily average of 421 g and 209 g, respectively. Fish alone accounted for 170,229 g moving upstream llla
Figure 37. Seasonal variation of photosynthesis respiration ratio at Concrete Bridge Station. and 47,964 g moving downstream.
Thus for this study 1.27 times more organisms were captured moving downstream than upstream, and 2.02 times more mass of organisms was captured moving upstream. Although more animals moved down than up, the larger size of those moving up contributed to a net movement of mass upstream. For fish alone, 3.58 times more mass was sampled moving upstream. A number of very large snapping turtles moving downstream contributed heavily to the difference between the total mass of fishes moving downstream and the total mass of all organisms.
More organisms moved up than down on 183 days; more moved down than up on 172 days; and the movement was equal or zero on
94 days. A greater mass of organisms was captured moving upstream on 239 days; a greater mass was captured moving downstream on
145 days; and the movement was equal or zero on 65 days.
Seasonal Variations in Movements
Table 18a and Figure 38 summarize by month the movements of all organisms captured at the principal sampling station. The maximum number and weight of organisms sampled was in the spring months of the three years sampled. The greatest mass moving was, consistently, in March and April, and the greatest number of animals moving was in
April, May, and June. Movement was much less during low water in the late summer and during the winter months. The pattern of movements was quite similar from one year to the next, although movements in
1969 were greater than movements during 1968 or 1970. Table 18a. Average Daily Fish Movements by Month
Number Average Average Average Av erag e of number number mass mass Date days moving moving moving moving 1968 - 1970 sampled UP down UP g down g
April 2 12 4 347 3
June 25 14 5 3 27 130
July 17 11 2 111 5 8
August 15 5 1 74 39
September
October 15 8 4 184 6 4
November 20 3 3 170 42
December 17 0 1 2 0 13
January 19 0 1 6 1 11
February 18 1 3 301 29
March 16 4 9 1204 275
April 30 8 3 2 2280 927
Table 18a. Continued
Number Average Average Average Average of number numb er mass mass Date days moving moving moving moving 1968 - 1970 sampled UP down g down g
May 3 1 8 13 358 310
June 24 9 8 319 222
July 15 3 2 177 122
August
September
October
November
December
January
February
March
Apr i 1
May
June Figure 38. Average daily migration by month. Mass is in grams moving per day. Numbers are represented by lines and mass by bars.
Cumulative Occurrence of Species vs. Cumulative Occurrence of Individuals
New species were continually added to the total as sampling pro- gressed, including a warmouth which was captured on the last day that samples were run and which had not been encountered previously.
For fish the plot of cumulative number of species versus the log of the cumulative number of individuals (Figure 39) was remarkably straight and consistent with theoretical species organization suggested by Odum, Cantlon, and Kornicker (1960). The line may come up slightly as found in some studies (Preston, 1963).
Diversity of Moving Animals
A total of 44 species were encountered in the 6,034 animals trapped at the principal station during this study. Of these 4,416 were fish of 27 species. The diversity of these organisms as measured by D1 = (s-l)-l logeN (Margalef, 1968), where S is the number of species and N is the number of individuals, was 43/8.6052, or 5.0, for all animals and 26/8.3929 or 3.1 for fish alone.
Movements at Other Stations on New Hope Creek
Table 18 summarizes the fish sampling data for the Wood Bridge and Jungle Stations. The movement patterns are, in general, similar to those observed at the principal sampling station. However, the lack of downstream sampling at these stations before October,1968 makes analysis of some of these data less useful. When these statioAi5 avLLi. sampled simultaneously with the principal station, a movement of total mass at least as great as at the Concrete Bridge Station is apparent. Figure 39. Cumulative species versus cumulative individuals trapped at principal sampling station; only fishes are included.
The movements at the Jungle Station may have been influenced by
the position of the trap in the middle of a very large pool rather
than in an interpool riffle, as at the Concrete Bridge and Wood
Bridge stations.
Movements at Morgan Creek
Data for Morgan Creek are given in Table 19. Local sports
fishermen intefered with sampling at Morgan Creek during the
spring and almost no sampling was accomplished during periods of
expected large migration. Nevertheless, Morgan Creek also shows
a greater movement of animals upstream than down, although the
results are less pronounced than those for New Hope Creek. As in
New Kope Creek, the movements were greatest in the spring.
Analysis by Each Species
Twenty-seven species of fish and 16 species of other organisms
greater than 1 g were captured during this study. No species was
captured at other locations that was not also captured at the princi-
pal sampling station, with the exception of one small species of
shiner caught at Morgan Creek,thay may or may not have been Notropis
altipinnis. Most of the fish captured in New Hope Creek were also
captured in Morgan Creek. The only exceptions were bowfin, chain
pickerel, green sunfish, largemouth bass, speckled killifish,
threadfin shad and piedmont darter, All but the pickerel were en-
countered in New Hope Creek only as a rarity.
- Numerical and Weight Contribution of Each Species to Migration The maximum, minimum, and average weight, as well as the
total number and mass, of the more important species encountered M d M N M In OOlnN CO NrlU3V) *. 6, .I rl N N at the principal sampling station are given in Table 20. V-lip and smallfin redhorses together were, by far, the most important in terms of mass. Turtles, redbreast sunfish, flat bullheads, and chain pickerel also contributed heavily to the total mass. Frogs (including tadpoles), whitemouth shiners, white shiners, crayfish, bluehead chub, and redbreast sunfish were most frequently encountered. The average size moving upstream and downstream of each species is given in Table 21. A more detailed analysis of each species by size interval and by upstream or downstream move- ment at the principal sampling station for the entire sampling period is presented in Figure 40. It is apparent from this information that for almost all fish species, there is a tendency for larger individuals to move upstream and for smaller individuals to move downstream. This is particularly evident for black crappies, bluegill sunfish, flat bullheads, creek chubsuckers, pumpkinseed sunfish, redhorses, white shiners, and redbreast sunfish. Creekchubs and darters show no particular pat- tern, highfin shiners and whitemouth shiners had greater upstream movement for all sizes, madtoms and pirate perches of all sizes moved downstream more than up. All larger species showed the large fish upstream--small fish downstream pattern. Smaller clayfishes were captured moving downstream more f~.. quently than up. Larger crayfishes moved in both directions about equally. Turtles of all sizes were caught moving downstream more often than up. Table 20. Minimum, Maximum and Total Mass and Total Numbers of
Each Species or Group Sampled at Principal Station,
New Hope Creek
Minimum Maximum Total Total Average Species weight weight number weight weight
Black crappie 1.0 210 64 4457.5 69.6
Bluegi 11 0.5 194 266 1869.5 7.0
Bluehead chub 1.0 80 404 4204.6 10.4
Bullhead 1.0 689 2 03 14416.2 71.0
Chubsucker 1.0 345 202 4544.0 22.4
Creekchub 1.0 84 7 2 798.5 11.0
Darter 1.0 4 109 161.0 1.4
Highfin shiner 0.5 7 139 270.5 1.9
Margined madtom 1.0 3 2 120 1107.5 9.2
Pickerel 1.0 738 217 13394.5 61.7
Pirate perch 1.0 12 8 1 309.0 3.8
Pumpkinseed 1.0 287 120 1982.0 16.5 sunfish
Redhorses 1.0 1363 328 150799.7 459.7
Redbreast sunfish 0.5 167 394 15299.5 38.8
White shiner 1.0 35 543 361 2.9 6.6
Whitemouth shiner 1.0 6 868 960.5 1.1
Crayfishes 1.0 44 446 5242.0 11.7 Table 20. Continued
Minimum Maximum Total Total Average Species weight weight number weight weight
Frogs 1.0 500 881 5359.7 "6.0
Turt les 6.0 4000 62 35679.7 575.4
Table 21. Average Mass of Animals Moving at Principal Station
Average Average Species mass up mass down
Black crappie
Bluegill
B%uehead chub
Bullhead
Chubsucker
Creekchub
Darter
Highfin shiner
Madt om
Pickerel
Pirate perch
Pumpkinseed sunfish
Redhorses
Redbreast sunfish
White shiner
Whitemouth shiner
Crayfish
Frogs
Turt 1es
Others
&/UMBER DOWN MUT':BEB UP
fdUP?B ER DOWN I NUMBER DOWN NUMBER lJP
NUMBER BONK NUMBER UP %-a -,,--.-,.-. L o I 187 Ll______l_-..l .__-_-_--I--.. - NUMBER DOWN NUMBER L'F
loo NUMBER DOWN NUMBER lj?
Seasonal Patterns of Movements for Each Taxonomic Class
Each taxonomic group was analyzed for seasonal trends in move- ments (Figure 41 asd Appendix D). From these it is obvious that the overwhelming bulk of the movement for all taxonomic groups, with the possible exception of crayfish, occurs in the spring. There is in some fishes continued, although smaller, movements throughout the summer; and for chain pickerel a secondary series of movements for the fall. The pattern for most fishes is repeated from year to year. The centrarchids are almost never encountered during the colder months. Some important movements for each group are noted below:
Black crappie: Crappies had one of the latest movements of any species, generally not moving until late May or June; however, a few small individuals were caught moving downstream in the spring of
1969 and 1970. The movements in 1968 were larger than in either of the other two years.
Bluegill: Bluegills moved principally in April and May. Very large numbers of small fish were caught moving downstream in 1969 and 1970.
Bluehead chub: In the spring of 1968 and 1970 these fish were one of the most consistent upstream movers; but in the spring of 1969, the movements were much smaller and were not as distinctly upstream.
Creek chubsuckers: These fish were not distinguished from red- horses until March of 1969. Heavy movements of these fish upstream occurred in March and April of 1969 and 1970, and a large movement Figure 41. Average daily movement for each number, by species.
Full names and scientific name for each fish are given in Table 6.
Lines are numbers of fish and bars are mass, in g, months are April,
1968, through June, 1970. Each species is on a separate page.
Black crappie Figure 41 a
Bluegill Figure 41 b
Bluehead chub Figure 41 c
Creek chub Figure 41 d
Creek chubsucker Figure 41 e
Darters Figure 41 f
Flat bullhead Figure 41 g
Highfin shiner Figure 41 h
Mad t om Figure 41 i
Pickerel Figure 41 j
Pirate perch Figure 41 k
Pumpkinseed Figure 41 1
Redbreast sunfish Figure 41 m
Redhorses Figure 41 n
White shiner Figure 41 o
Whitemouth shiner Figure 41 p
Crayfish Figure 41 q
Frogs Figure 41 r
Turtles Figure 41 s
Others Figure 41 t
PIRATE PERCH
'm or-.
T- ry- WHITE
(V z. t-i .iITEMr!iJTH SHINER
is, w -,
t OTMERS
'II) or-7 downstream of small fishes occurred in 1970.
Creek chub: Creek chubs were most frequently encountered in
April and May of 1968, and were rarely sampled later. Nearly all movements were in April and May, with a slight upstream bias.
Darters: Darters moved upstream during the spring and rarely at any other time.
Flat bullhead: These fish were caught at nearly all times of the year. Peaks in movements occurred in the warmer months.
Highfin shiner: These little fish were the most numerous fish in this study. Movements in both directions were greatest in April, May, and June.
Madtom: Madtoms moved greatest in May and June. In 1968 and
1970, movements were more up than down.
Redhorses: The two species of redhorses completely dominated the mass of fishes in New Hope Creek migrations. Movements were large at almost all seasons of the year, with some diminuation during the summer and very large peaks in March and April. A small flood in December of 1969 caused heavy movements even in the winter. Small redhorses were not caught very often.
Redbreast sunfish: The,se colorful fishes were caught during all warmer months of the year. Heaviest movements occurred in
March, April, and May, Movements from year to year were similar in magnitude .
Pumpkinseed sunfish: These fish were rarely caught except in
April, May, and June, although some smaller fishes were captured moving downstream in March. Pirate perch: Pirate perch started their annual movements before most other fishes, as early as January in 1969. They were rarely caught at times other than the spring, although floods in the fall of 1969 may have stimulated the secondary movements noted then.
Chain pickerel: Pickerel moved throughout the year with peaks in both the fall and spring. Smaller pickerel were frequently
captured in the summer. Their predatory habits may influence the year-round movements noted.
White shiners: Movements were greatest in April and May, and
1968 and 1970 were more important than 1969. Smaller movements
continued throughout the year. The largest recorded movements were associated with the floods which occurred in October, 1968,
after a long drought.
Crayfish: Crayfish were active throughout this study with a peak movement during the spring of 1969. Movements were generally more upstream than down except during that time.
Frogs: Frogs were caught from time to time, most frequently
in the spring as tadpoles moving (or being swept downstream) down-
stream. Some large bullfrogs were also captured moving in both directions.
Turtles: Turtles were caught in the spring and summer, and not at all during the rest of the year. Some very large snapping
turtles contributed to a large transfer of mass downstream during the spring. 155
Others: Miscellaneous organisms also moved most heavily during the spring. Sometimes the capture of a large muskrat or a number of snakes contributed to heavy movement during other times of the year. In general there was greater movement downstream than up for these assorted creatures. Large-mouth bass, of sport fishing interest, exhibited movements similar to other centrachids but on a smaller scale; a few large bass moved upstream in the spring and small ones moved downstream at various times of the year, often in the fall.
Evidence of Spawning Condition of Fish at Different Times of the Year
Records were kept of signs of reproductive activity for the fish sampled. These signs include: breeding tubercles, seasonally bright colors, and the actual discharge of eggs or milt, Table
22 gives these results for all species where the information is available. Signs of breeding condition were only noted in the spring, and were invariably associated with heavy movements of that species. Fish in obviously ripe condition were taken almost invari- ably moving upstream, and spent fish were always taken moving down- stream.
Recaptures of Marked Fish
Of the 6,043 fish and other organisms captured at the principal sampling station, 417, or 6.9 percent, were marked from previous
Table 22. Continued
Species (number, if more than 1) Date Direction moving Condition noted
Creek chubsucker (Cont) 1969
Mar. 30 Tubercles
Mar. 31 Tubercles
(many 1 Apr. 6 Tubercles
Apr. 9 Tubercles
1970
Apr. 11 Discharged eggs
Creekchub Apr. 11 Tubercles, rosy-colored, discharged milt
Apr. 12 Tubercles
Apr. 19 Tubercles and rosy- colored
Apr. 20 Concave be1ly
Apr. 25 Tubercles
Table 23. Recapture of Marked Fish
Number of Number of Total Total Percent Percent marked fish marked fish number number recaptured recaptured recaptured recaptured moving moving moving moving Species moving up moving down up down UP down
Black crappie
Bluegill
Bluehead chub
Bullhead
Chubsucker
Creekchub
Darter
Highf in shiner
Madt om
Pickerel
Pirate perch
Pumpkinseed sunfish Table 23. Continued
I Number of Number of Total Total Percent Percent marked fish marked fish number number recaptured recaptured recaptured recaptured moving moving moving moving moving up moving down up down UP down
I Redhorses 16 13 248 82 6.5 15.9 Redbreast sunfish 15 21 212 181
White shiner 2 1 54 329 213
Whitemouth shiner 13 26 578 286
Crayfish 11 9 190 256
Frogs 0 0 18 863
Others 3 1 52 532 5.8 .2 This very low recapture rate indicates that, in general, 'home range1 movements (Gerking, 1959) were not being intercepted--or possibly that the fish became very trap-shy after one encounter.
Tagged Fish Returns Analysis
During this study larger fish were marked with numbered plastic dart tags (Floy Tag and Manufacturing Company), and 75 of these were recaptured (Table 24). Recapture patterns were varied, but many individual fish were captured moving upstream and recaptured moving downstream shortly thereafter. Fish were some- times recaptured at the same location and moving in the same direction without having been captured moving in the other direction.
These generally occurred only during intervals in which the weir was disassembled in the interim. Fishes marked at the principal station were rarely recaptured at another, although the other stations were sampled much less frequently.
Daily Concentration of Moving Animals
Each species was analyzed for number of individuals moving up- stream or downstream during a given day. This may be some indication of the tendency for the fish.to school, and may have theoretical im- plications as to the best way for a given mass of fish to be moved from one place to another.
All taxa traveled more frequently as individuals than in any other numerical association, and in groups of two more than in any larger groups (Table 25). Thus,all taxa generally appear not to travel in schools, During periods of heavy migration, however, some species were captured in numbers of 10 or more per day. This was true of bluehead chub, redbreast sunfish, bluegill sunfish, redhorses,
a 'a, a, M C, M cd cd n C, mom mmoo 44N \D**bM44N NNN k . %9$ &222222k k k k &EX. 32EZ Table 25. Concentration (Daily) of Moving Organisms
Times Organisms moving in groups of
0 1 2 3 4 5 6 7-9 10 - 19 20 or more
Black crappie
868
Bluegill
841
Bluehead chub
714
Bullhead
755
Chubsucker
810
Creekchub
848
Darter
825 a, k 8 k 0 0 N
0, d
I 0 l-4 cn 1 I-.
\O
In
d
M
c.l
d
v, a, v, k 0 C 5 0, a, v) 0 d I-.
DISCUSSION
This study of migration and metabolism of the New Hope Creek
stream system allows the two to be related so we may infer some of the roles that migration may play in stream metabolism and the ways
in which the migrations may take advantage of programs of life
support. These comparisons may be made by examining the seasonal timing of events, th'e spatial distributions, the nutrients processed,
and the energy involvements of each part.
Seasonal Patterns of bletabolism
New Hope Creek has a sharp peak in primary production in the early spring (Figures 34-36). This peak is associated with high levels of respiration that continue throughout the summer and early fall. There- fore, as the season progresses, the stream becomes increasingly dependent upon outside sources of energy and/or energy storages. This greater usage than production of energy constitutes a condition of heterotrophy.
P/R Ratio and Heterotrophic Reg'ime
The greater respiration than photosynthesis observed in New Hope
Creek (Figures 34-36) is often characteristic of woodland streams that are dependent on allochthonous detritus for some or a great deal o; thei. energy supply (Smith, 1966). Hoskin (1959) found similar patterns in other streams of North Carolina. The principal supply of this detritus to New Hope Creek is probably leaf fall and organic runoff from the surrounding forest. As shown by some experiments I made at the Pennsylvania State University, mayflies and stoneflies, both abundant in New Hope Creek, quickly reduced dead leaves of many species to skeletons of vascular tissue. Such skeletonized leaves were often observed in New Hope Creek during the late spring and summer months. Other sources of externally supplied energy may be forest insects dropping into the stream, organic substances in runoff, and several minor sources of domestic pollution that exist near the headwaters (Research Triangle
Regional Planning Commission, 1968). These may be less in summer when discharge is small
Spatial Distribution of Metabolism
As shown in Tables 15-17 and Figures 34-36 the productivity and respiration per unit area are fairly uniform in the zones of New A
Hope Creek studies although there was at least a three-f~ldrange of depth and volume metabolism. Table 11 suggests that the percentage + of sun energy reaching the creek surface is similar in the two stretches studied (Blackwood Station and Concrete Bridge Station).
Comparison of phosphorus by.station is made in Table 26, but no differ- ences were found that consistently correlated with differences in volume metabolism.
Dilution of Resources with Depth a Many previous authors have found an inverse effect of depth and the productivity of waters. Rawson (1952; 1960) with data from large boreal lakes in Canada indicated that the areal production
of net plankton, benthos, and fish was inversely proportional to the depth of the lakes. Shallower lakes were more productive in these higher trophic levels. Steeman Nielsen (1957) found an in- verse relationship of productivity and depth of the euphotic zone in the sea. Bailey (1967) found an inverse relation between depth and primary production in the Sacramento-San Joaquin estuary.
Demersal oceanic fisheries tend to be concentrated on relatively shallow banks and near shore areas (Bigelow and Welsh, 1924;
Alverson, 1964).
In New Hope Creek with similar metabolism per unit area, in- crease in water depth between stations diminished the concentration of metabolism per unit volume. It may be reasoned that food resources for fish were also diluted. If so the shallow zones may have more concentrated food for young fish.
In New Hope Creek there were two manifestations of this change in water depth. These are changes from deeper, downstream areas to more shallow upstream regions, and changes at any one place as the water drops during the summer. Much of the energy that enters a system remains the same no matter what the depth, for both light and leaves enter a stream on a square meter basis. In addition, the amount of energy available to benthic plants would be less in deeper regions because of extinction with depth. Thus the total energy to support organisms does not change much with depth. It does be- come more concentrated, however, and perhaps more available to food chains. This effect during summer low waters may be partially offset by a lessening of the total light energy input to the stream as the total water area becomes smaller.
Both photosynthesis and community respiration in New Hope Creek varied seasonally with a spring maxima and a secondary peak in the fall (Figures 38-40). The possible causes may be seasonal variations in minerals, solar energy, temperature and water level changes. Daily records of insolation under a deciduous forest canopy kept at the International Biological Program Site (Figure 42) showed seasonal patterns of insolation with a peak in early spring that corresponds with the peak of observed values of gross primary pro- duction (Figures 34-36). The peak of photosynthesis was in March rather than in June, due to the shading effect of overhead trees, which leaf during the middle of April. A second, smaller peak in primary production in the fall also corresponded with an increase in light following leaf fall. Neither dissolved phosphorus (Table 13) nor any of the important forms of nitrogen (Table 14) showed any consistent seasonal variations that were correlated with seasonal variations in metabolism. The seasonal variation of community respiration at the Concrete Bridge Station (Figure 34) showed two peaks, one during the high solar energy input in the spring and one during fall low waters. Therefore, apparently neither the primary production nor respira- tion was controlled by temperature which had maximum values in late summer.
Comparison With Some Other Studies
The areal metabolism was generally lower than values obtained in other studies (Table 27). The metabolism was within ranges of Figure 42. Seasonal patterns of insolation under a hardwood canopy, Duke Forest, near New Hope Creek. Pyroheliometer data is from the International Biological Program site located a few hun- dred meters to the north of the upstream watershed of New Hope Creek.
Table 27. Metabolism in Some Other Unpolluted Streams
Gross Total community Production respiration Location Time g m-2 day-l g m-2 day''
Birs, Switzerland a 1946, April 11-12
Kljasma, Russia a 1929, July 21
Itchen, England b April - October
IOrdinary1 stream in North Carolina (various times of year) 1956 - 1957
Ivel, England c 'typical1 single curve at two locations Elay 1964
Spring Creek, Pennsylvania d Entire year 1-17 1.5-13
This study Concrete Bridge Stat ion 0.21-8.85 0.39-13.40 a. Quoted in Odum, 1956 b. Hoskins, 1958 c. Owens, 1969 d. Cole, 1969 Pa.tterns of Fish Movements
The annual peak in fish movemelts was close1y cnrrcl ated w; tl~
the annual peaks in gmss phntosvnthcsis, community scspjration a~d
the end of winter floods (Figure 34-36, and 26) . These may be a
selective pattern in the fishes to schedule their own time of high
energy usage with the time of maximum total energy availability
in the environment. Among the large fishes caught in this study,
there was no clear-cut pattern for maximum fish growth apparent at
any one time of the year (Figure 43). Storage and lag processes in
the stream may smooth out the pulse in food availability at second
and third trophic levels over the season when the young fishes
are ready to tap the food chains. The continuing high levels of
respiration after the pulse in primary production indicate sus-
tained biological activity in summer and fall.
Movements of Different Species
Some smaller species, such as darters and pirate perch, with a high surface to volume ratio and resultant high friction, do not show a large upstream movement. The energy lost in rnigrati.cn may he greater than that gained.
Movements and Floods.- The current as an auxiliary energy source to moving animals both aids in the planned movements downstream and increases energy demands on animals holding their position. Observations during high water suggests actual washing of fish downstream is not important.
Table 28. Metabolism of Some Selected Lakes and Maine Waters
--a- Gross Coinmn i t y product ion respiration Location Time g m-2 day-l - 2 - 1 - J3 _m_md2~ -* _.-_- --. Eniwetok Atoll a Midsummer 1954
Texas Bays b Various times 1957
Stuart Farm Pond c Durham, N. C.
Lake P*lichigan d July 13-14 3.2 km from shore 1968 6.4 km from shore 11
Sacramento-San Joaquin delta e
i3rackish Ponds
blorehead City, I.?.C, f ' Average for year
a. Cdum and Odum, 1355 d. Planny and Hal 1, 1969 g. Sum of net productina and night time respi~z-ti?n b. Oduix and Wilson, 1958 e. Bailey, 1970
I c. Odum and Wilsor., 1958 f. Odum et al., 1970
I I Figure 43. Growth of tagged fish, New Hope Creek. The left of each pair of points represent weight at first capture, and the right point represents weight at second (or third) capture. Decrease in weight for some fish probably represents spawning losses.
bc = black crappie; ecp = chain pickerel; fb = flat bullhead;
rb = redbreast sunfish; rh = redhorses.
The greatest discharge occurred during the months of February and
March, but few animals were sampled in the downstream trap during this period. For example, on February 2, 1970, there was a medium- sized flood that raised the water from 50 to 65 cm above zero flow.
During the spring or summer months this would have been accompanied by an increased movement both upstream and downstream. However, on this date, when the water temperature was only 7" C, there was no recorded movement either upstream or downstream. This pattern was repeated on many occasions during cold weather. Apparently the fish move downstream only by internal program.
Movements of Juvenile Fishes
Many more very small fishes may move downstream than were measured in this study, since the mesh size on the weir and traps was large enough to let any fish smaller than 1 g pass through.
Plankton nets hung in the current on six separate days during periods of heavy migration of larger fish caught only one small darter. More extensive sampling could possibly give very different results. On
May 15, 1970, leaves plastered on the side of the downstream trap formed a barrier in which was observed a school of about 250 tiny (1.3 cm) fish. Complete keying was impossible but they appeared to be some species of Notropis. How often this occurs when the movement is not observed is a matter for another study.
Differential Movements of Different-Sized Fish
The generally upstream movement of larger fishes and generally downstream movement of smaller fishes observed in this study raise some interesting questions about usage of available energy by a population of animals. The upstream movements are obviously tied in with reproduction, which implies that there may be reasons to bring the potential progeny upstream. One reason would be to distribute the genetic stock over the stream. The very small fishes with large surface-to-volume ratios affecting friction cannot swim upstream against the current, but the large ones can and lo, and the small ones can and do move downstream with the current. Thus, the upstream migrations of the adults may be necessary as agents for stock maintenance and gene dispersal. Since the spring pulse in energy available per volume at the upstream station is con- siderably greater than the pulse at the downstream station, it would be more advantageous to have the most rapidly growing small stages located upstream.
This raises the question of why the fish move back domstream.
The large number of sub-one-year class fish moving downstream in spring indicates a dispersal of many fishes after spending one year upstrean. This may be an ad.aptation to prevent population pressures between the new year class of juveniles and other fishes upstream which increase their activities as the water warms. Because of geometric adaptation to rocks, currents, and microenvironments large fish may experience less stress in the deeper waters, Nellier (1962) found small fishes moving to deeper waters as they grow. Another possibility was considered by Margalef (1968), who commented on the movement of animals as they grow older from the highly productive regions of juvenile growth to more stable environments. Downstream 157 regions in New Hope Creek may be, to a fish at least, more stable, since they are not subject to the extreme diurnal variations in oxygen that occur in the upstream, more shallow regions durjng the low water stages of summer drought. In addition, the deep pools in the downstream regions provide insurance against complete annihilation during extreme droughts. There may be a tradeoff of high productivity versus a more stable environment that is best utilized by sending armies of young to the highly productive regions to get a quick start in life, followed by dispersal of those that survive to more stable regions. A single small fish is more ex- pendable than a larger one since there are many more of the former and an ecosystem has invested less of its energy resources in it.
Comparisons of Energy Budgets
Consider next the energy budgets of the stream, its metabolism, the fishes and their migrations. To relate energy budgets of the fishes to that of other parts of the system it is convenient to express work processes in their caloric form, since energy is a . common denominator for all processes.
Energy of Running Water
The physical potential energy released in turbulence (E) in a cubic meter of water flowing downhill between the Blackwood sampling station and the Concrete Bridge Station can be calculated as equations that follow: 3 E in kg-m m - = (mass in g) (acceleration due to gravity in n ~ec-~) (difference in height in m)
* E in Cal m-3 = (lo3) (9.8) (45.7) (2.34 X lom3 Cal kg-m)-'
E in Cal mm3 = 1040 Cal m-3
For the year a mean time of one day was required for water to flow from Blackwood Station to the Concrete Bridge station, there- fore the physical power dissipation is 1040 Cal m-3 day-1, or about
350 Cal mm2 day-'.
Energy of Biological Metabhlism
The mean biological metabolism in the same cubic meter of water flowing through the same zone over the entire year was found to be about 2.85 g oxygen per cubic meter per day (Table 15). Since about 3.5 Calories are released per gram of oxygen metabolized
(Brody, 1945), a total of about 10.0 Calories of energy were used in respiration per day. Thus the system receives about 100 times more energy from the work of currents as from organic fuels.
Energy of Insolation
The energy of insolation reading through the canopy to the stream was estimated to be 6.7 3 lo4 CaP m-3 day-l. Thus the solar energy budget is about60 times thewater current contribution.
Energy of Fish Metabolism .
A rough figure for the use of oxygen by fish under normal conditions is about 100 ml (0.143 g 02) hr-l kgq1 (Brown, 1957;
Brett, 1965). This is quite variable with temperature and activity rates but is probably close to mean values for New Hope Creek. There is about 18.3 g fish m-3 in New Hope Creek (Carnes et al., 1964) which -- would use (0.143 g 02) (365 days)(24 hours) (18.3 a kg) or 23 g O2 m-3 year-1. For the entire watershed above the Concrete
Bridge, this would be, including additions and losses by migration, about 4 g 02 m-2 year-1, or about 13.4 Cal m-2 year-l (Figure 44) .
Energy Used by Migrating Fishes -- .-- -- Consider three different ways to measure the energy used by migrating animals such as fjslles: (1) one nay calculate the tatzl work expended against f~lction;l.J.forces and/or that used in mnvjn~
the organisms against gravity. (2) One may convert the additima! oxygen used during mig~?t;on to Caloric values. ConsJderah;e data exists on oxygen use chiring diEferent levels ofaaci;vit~~(Grown, 1957;
Brett, 1965, 19701, and estimates have bee? made for use during mi- gration (Brett, 1970). These figures are about 100 ml O2 kg-I hr-1
for fish at Isw levels of activity and about twice this during migra- tions. (3) One may weigh fish and analyze them for different food reserves at the beginning and end of their movements. This type of work has been done for organisms that do not feed during their spawning migration, such as salmon (Idler and Clemens, 1959). The second method was used for these estimates. An additional
100 ml 02 kg hr-I was aBloted as the cost of migration. Multi- plying this by the annual biomass moving upstream at the Concrete Bridge (120 kg) by the period of major movements (three months or 2200hours) gives a total energy cost of 132,000 Cal, or 0.88 Cal
year-'. This is about 7 percent of the estimated fish metabo?'~~ and about 0.1 percent of the annual metabolism of the entire ecosystem. If it is assumed that the upstream migration is necessary to maintain the stocks of fish in the upstream position of the stream, this energy used for migration has a multiplying effect of 14. All energy relations are summarized in Figure 45. Figure 44. Annual imvc!:ient arid r~ct::holi s;; of fiili p+~llnt.i 011s
in the hcadiratrrs of Sew liopc Cree!< a!>ovc the Concrcip Cridge.
Numbers represcii: appr.oziii:~.te s:an,!i ng cro~r, ar:nii~ini gr2T i cw ,
7 respiration and food intake in Cnl n-- ycor - 1 . Food inti!:. is calculated to bal ance other energy flop,s .
Figure 45. Energy flo~diagram for upstream (nio'dle set of modules) and downstreas (loiiemost set of ~~~odules)of Scir iIo;~c
Creel:. Eietabolisrm is in Cal n-3 day-?, 3s VO~UIOCdi fie:-encc~ in metabolisin is suggos tcd as an j rnportant factor. Energy enters the systcm as sun energy, which passes throi~ghfood chains, 2nd sireaiii fluw energy which aids in the distribution of resources and dispersal of wastes. Upstream niigration requircs sdlitional energy to overcome this flow.
Net Contribution of Migration to
Headwaters and Turnover Rate T
During this study an estimated 119,400gEsh and other animals moved each year into the region above the Concrete Brid.ge, and 56,700 g moved out of this area. A net movement into this area of 62,700 g occurred. There is approximately 1.6 105 m 2 of stream above the
Concrete Bridge as determined by field measurements and topographic maps. Over this one year period there was a net addition of 0.39 g animals m-2 of water. This is about 14 percent of the estimated fish standing crop of 2.78 g mq2 (Carnes --et a1 . , 1964) . Summing the mass of animals leaving this area and the mass of animals entering the area gives 176,100 g, or 1.1 g m-2 of animals (about
0.8 g m-2 of fish alone) involved in migrating to or from the area.
This is 40 percent of the estimated standing crop of fish in that region (Figure 44 ) . The standing crop of a part of New Hope Creek would, according to these figures, be replaced in 3.7 years by upstream migration alone, 7.85 years by downstre& migration alone, or in 2.5 years by movement from above and below. This replacement rate has implica- tions for predictions in relation to pollution and fish kills. These relationships are summarized in Figure 44. Food is calculated to * balance respiration and migration. Comparison of Migration in New Hope Creek With a Salmon Migration
Two lakes in British Columbia, Owikeno and Long Lakes, were chosen for a comparison with New Hope Creek. The total runs of salmon are well known and have been virtually constant within the period of record, indicating that present average run figures are about the same as run figures before heavy exploitation
The mean annual catch and escapement of all salmon (mostly sockeye) has been about 500,000 fish. An estimate of the average weight for sockeye salmon is about 2.5 kg for each fish (Idler and
Clemens, 1959). Thus, about 1.25 - 109 g of fish entered the lake area of 97.1 km 2 or would have entered had there been no fishery.
This is about 12.5 g m-2, or 21.4 times the contribution of migra- tion to the upstream reaches of New Hope Creek. However, this mass is distributed throughout much deeper water, does not feed, and was produced from a much broader food base.
Possible Adaptive Values of iyiigrations in New !-Iope Creek
Migration As a Coupling Function .~' Various possibilities exist for the selective advantage of tying together various sections of the stream by animal migration: (1) Already discussed is the role of migration in the reproduction and dispersal of juvenile stages of the species. (2) Various areas ~f the stream may become devoid of fishes due to natural disasters, such as drought, summer low oxygen, severe predation, etc. Migra- tion provides a steady source of recolonizers that can occupy empty habitats: (3) The migration and reproductive system allows a population to be maintained in a current. Any downstream drift may be compensated for by migration. (4) Predators, moving through the stream tend to feed most heavily in areas with large numbers of prey species, and thus tend to control possible excessive increases in these species. (5) The contribution of minerals to upstream regions by migrating animals is discussed elsewhere in this section. (6) According to Levins (1964) migration has selective value in that it allows sufficient interchange between populations so that local adaptation for short-range environmental fluctuations will not become a very important factor which would reduce the overall fitness of the gene pool. However, this does not reduce the adaptability of the population as a whole to widespread changes in environment. This may be a factor in the selection of species fitness in streams such as New Hope Creek in that the fitness of the gene pool as a whole is maintained and not wasted on non- selective adaptation to short-term local events, such as stress during exceptional drought.
Interaction of Yield and Organization
New Hope Creek, like many other complex systems, can be arbi- trarily divided into a subsystem exporting energy and another sub- system receiving this energy, and in return supplying certain organizational or other services to the exploited system in a feedback loop. Examples of this relation would be: A prey Ifdonatingfta certain percentage of its energy resources to predators in exchange for population regulation; flowers providing bees with energy in return for pollination services; and farms su?plying cities with food and receiving in exchange fertilizers, farm machinery,and social services.
Such a system can be defined in New Hope Creek. The upper regions
of the stream export food energy to downstream regions and receive in return genetic informatio? resources, populations of higher trophic levels to utilize seasonal energy excesses, population
control, reseeding when necessary, and minerals. All of the above are concentrated biological control agents and are effective in relatively small amounts.
Margalef (1962, 1969; see also Deevey, 1969) has considered the energy information exchange between two systems or subsystems in some detail. A downstream system that has greater organization (which Margalef calls 'mature') may be more efficient in its use of energy. An upstream, less organized system (which Margalef calls
'imrnatureV)may not have the structural and organizational frame- work for using energy as efficiently as the more mature system, and as a result often loses much of its energy to export. If the more organized system is able to utilize this energy lost by less organized systemsit,inacertain sense, exploits the less organized system. In return, the more organized system gives v~ information to the lcs mature one, aiding it in becoming more efficient in its own use of energy, and increasing its organization. Although New Hope Creek is readily divisible into two segments the upstream one supplying energy to the downstream one, and the downstream one supplying information to the upstream one, further agreement with Margalef's theory is not substantiated. The indices of maturity should, according to Margalef, be higher in the downstream region and lower in the upstream region. This was not substantiated by investigations. The upstream region had higher volume production
(mean of 3.1 versus 1.4 g m-3 day'') and a lower ratio of production to respiration (0.4 versus 0.6), indicating that the biomass sup- ported per unit of photosynthesis was greater. Studies of pigment ratios (Motten, 1970) showed slightly greater D430/D665 for both pools and riffles in the upstream regions of New Hope Creek in oppo- sition to Margalef's theory . Thus, the energy-information inter- change theory may have validity apart from any consideration of relative 'maturity.'
Some Other Animal Migrations and Environmental
Energy Patterns
Migration in New Hope Creek may be an example of a widespread phenomenon. A number of examples of migration to regions of high productivity for reproduction were considered in the discussion--for example, the Texas Bays work of Odum and others. Other examples would be summer migration of many birds to arctic areas for reproduc- tion during the high energy input of very long daylight periods, and the heavy utilization of rich estuarine and nearshore regions by
small salmon, particularly chums and pinks. Further analyses of pre- sent migration and energy availability patterns, as well as post-
Pleistocene opening of niches and evolution of lakes using sockeye
salmon, may be fruitful areas for additional research. New Hope Creek Watershed Annual Phosphorus Budget
Fish migration upstream may partially offset the downstream
transport of minerals. It is important in the mineral budgets
of salmon lakes in Alaska (Donaldson, 1968) and in Russia (Krokhin,
1967); and in New Hope Creek, the relatively large mass of fish moving upstream qmarently also contrjbutes to the mineral balance.
Some data on tbe phosp5~rusbudget for New Hope Creek are calculated
in Table 29 and summarized. for the watershed ir? Figure 46.
Measurements were made from .Tune 14, 1968 to June 13, 1969 of phosphorus flows in the water in leaves and in fishes. The results
indicated that phosphorus discharged in suspension or soluti~nis, by far, the most important factor in the movement of that element.
Less than 0.2 percent of that amount is lost in leaf discharge and even less than that as fish and other animals moving downstream.
The amount of phosphorus brought upstream by fishes was about one-half of that lost by leaf discharge and less than 0.1 percent of that lost by stream discharge. Therefore, on an annual basis, the con- tribution of phosphorus to the headwaters by fish was small.
Some studies have suggested that upstream movements of inver- tebrates may help or entirely compensate for the doh-nstream drift
(Minckley, 1961; Ball and Hooper, 1963; Hughes, 1970). Others have emphasized the loss in drift. Insect drift was not formally sampled in New Hope Creek; however, casual observations and several 24 hour plankton net drift sam2les indicated that insect biomass in the drift recorded by Waters (1965) and Anderson and Lehmkhul (1968) and adjusting these to the volume of discharge of New Hope Creek. Table 29. Annual Movement of Phosphoms in New Hope Creek: June 14, 1968 - June 13, 1969.
Total Total Total Total Wet weight P Total P Month water dissolved and leaf P of animals in animals animals in animals discharge suspended discharge lost in-moving upC moving upd moving downC moving downd 104 m3 P dischargea g dry wt. leavesb g g g wet wt. g €! €! g
1968
June 14-30 14.04 7,570 478 0.2 5,560 19.46 2,210 7.84
July 14.06 . 7,990 971 0.4 3,440 12.09 1,795 6.28
August 1.93 104 764 0.3 2,290 8.02 1,218 4.27
September 0 0 0 0 480 1.68 300 1.05
October 15.8 8,540 20,025 82.9 3,440 12.05 1,985 6.95
November 81.6 44,100 851,760 349.0 5,140 18.00 1,259 4.40
December 77.3 40,500 38,475 6.6 620 2.16 403 1.41 a Monthly water discharge times mean total phosphorus concentrations (5.4 . 10- 8 g P g-l water).
Monthly leaf discharge times mean phosphorus content of leaves (4.1 . 10-4 g P g-l leaves) .
Monthly averages for each day times number of days in month.
Monthly movements times mean phosphorus content of fish (3.5 . g P g-l fish). 0 I-. In N I-. a cn 0 N
0 M N \L) 00 rl .-, .-, r. cn N Ln
0 v, "i N 01 N M M N Ln
0 0 0 M d M e, " CO N a m d
0 N cd a, c,CLtJ>M m I-. 0 mcd d w b oa, cn
d r. Ln 10 Ln -3 a M d- ri u3 m N
0 d 0 N 0 N " m d rl CO a m
N CO Ln CO m N rl N" rl
ri rl cd ri cd t' .rl 3 0 k a b 2 2 Figure 46. Diagram of phosphorus flows in New Hope Creek watershed. Flows are in grams of phosphorus per hectare of watershed per year. The watershed has about 6800 hectares above the Concrete
Bridge Station. Phosphorus added in rainfall was calculated as about *. 1 parts P in rainfall (Donaldson, 1967; Cooper, 1969) times lo4 in2 ha-l times 0.95 m rainfall for the one year period (U .S.
Weather Bureau, Raleigh-Durham Airport, N.C.), or about 95 g phos- phorus per hectare.
Annual cycling of phosphorus through leaf development and fall was obtained for deciduous forest in Duke Forest (Garrett, personal communication). Using conversions from Gosz --et al. (1970) and Rodin and Bazilevich (1965), this represents about 1960 g P per ha per year for leaf litter and approximately 2880 g P per ha for total litter.
Standing crop of phosphorus for fishes was estimated from stream sampling data done by the North Carolina Wildlife Resources Commission (Carnes --et al., 1964). Their value of 27.8 kg per ha (l758g for 0.154 acres) was multiplied by the approximate phospho- rus content of fish (0.35 percent).
Total phosphorus in the total organic standing crop of Duke
Forest was approximated by averaging values for 14 pine and mixed deciduous regions of the world given in Rodin and Bazilevich (1965).
A mean value of 63 kg phosphorus per ha was used, as the watershed of New Hope Creek is about one-half deciduous and one-half pine
(anonymous map supplied by the Duke University School of Forestry).
Even assuming ten catastrophic floods per year of the magnitude of the largest recorded in New Hope Creek, only about 0.2 of one percent of the mass of leaves lost would be lost as insect drift. Thus phosphorus loss by insect drift was considered small.
The results of these calculations computed for the entire watershed are given as a diagram using energy flow language (Figure
46). Standing crops from literature values are included. The loss of phosphorus by New Hope Creek (97 g ha-l) compared to the standing crop is small, and may be entirely replaced by amounts added in rainfall alone.
This loss of phosphorus from Duke Forest by New Hope Creek was about the same as values from other studies reported in Rodin and Basilovitch (1965). This compares with a general loss of from 0.9 to 21 kg per ha per year for Ca, Mg, K, and Na for four studies summarized by Likens --et al. (1967). Standing crops of phosphorus in fishes is an important reservoir during summer low water flows. Migration may be important in the mineral budget by maintaining this reservoir. . Analog Simulation of a Migration Model
An Electronic Associates, Inc. Model TR-20 analog computer .was used to simulate the process of migration in New Hope Creek. Figure 47 gives, in energy flow symbols( a model based on parts of
the energy diagram in Figure 45. Included are differentia-l equations
for the energy accrual to upstream and downstream populations of fishes and approximate coefficients for energy transfer based on New Hope Creek data.
The analog circuitry corresponding to the differential equations
is given in Figure 48.
Analog Results and Discussion
Annual results of the model are given in Figures 49 and 50. The former shows the input of energy into the upstream and downstream compartments of the model after the function generator and associated pots had been adjusted to give an energy input curve with a spring peak similar to such curves observed in New Hope Creek. Figure 50 shows the rate of energy accrual to upstream, downstream, and upstream and downstream combined populations of fishes with and without migration. Each population of fishes draws energy as a product of the energy available in the environment and the number of fishes available to exploit the energy source. As fishes move from one region of the stream to another their ability to add to their energy supplies changes, being greater in the upstream, more productive regions. Thus, as the upstream or downstream population of fishes gains or loses Figure 42. Energy flow diagram for analog computer model. The input of sun and organic energy is represented by the circle on the left -hand side of the page. The curve drawn in the circle repre- sents the annual energy input to upstream and downstream fish populations with a peak in the spring corresponding to peaks in photosynthesis and respiration that occur in New Hope Creek. The storage symbols represent fish population biomass in upstream and downstream regions which feed from the energy sources. The energy input to the downstream population is, on a volume basis, only one- third of the energy input to the upstream population. Arrows drawn to heat sinks represent metabolism, or energy loss by the second law of thermodynamics. The multiplier symbols represent rates of food flow from the primary production and other input sources to the fish populations. The lines connecting the fish populations are migrations. The energy drain on fish migrating against the current is represented by the heat sink attached to the upstream migration line. Current-assisted downstream movement is shown as a multiplier on the downstream migration route with an input from the upper circle. Differential equations describing the populations and transfer coefficients are included below the figure.
Figure 48. Analog symbols representing the energy pathways in Figure 47. Symbols are standard analog notation. Triangles represent summers (two inputs) or inverters (one input), triangles with rectangles are integrators,six-sided figures are multipliers, and the small circles are potentiometers, or lp0ts.l Lines drawn between symbols are electrical lflowsl (actually differences in potential), with an arrow pointing into a vertical line represent- ing one-way flow and diagonal bars,off-on switches. The letters VDFG stand for variable diode function generator. Numbers on each symbol refer to actual numbers used on the computer. The upper set of modules represents the energy-input pulse generator, the middle set represents the upstream fish population, and the lower set represents the downstream fish population.
Figure 49. Analog output of energy pulse generator. The
annual input of energy into New Hope Creek was simulated at the upstream section (a) and the downstream section (b) of the
stream, with a large spring pulse similar to one found in New
Hope Creek.
Figure 50. Analog simulation of annual energy accrual to populations of fishes in New Hope Creek. The annual energy flow to fishes in New Hope Creek at the downstream station region (a and b) and at the upstream region (c and d) is represented without migration (b and c) and with migration (a and d). Since a greater mass of fish moved upstream than downstream, there is a net gain to the upstream population and a net loss to the downstream popu- lation. The energy accrual to the total population of fishes is represented by the upper pair of lines. Line o is without migra- tion and line f is with migration. The tctal amount of energy flowing to the entire populati,on of stream fishes is greater with migration than without. individuals, the population's energy drawing power becomes greater or less. In Figure 50, the lower set of lines represents the annual energy accrual to the downstream population of fishes. The upper of these two lines (a) represents energy accrual without migration.
With migration (b), the energy accrual is less since a greater mass of fish movcsupstream than downstream. On the other hand, the upstream population gains energy with migration as a greater mass of fish move into the upstream region than moves out (line c, without migration; line d, with). The total energy accrual to the entire population of fishes is greater with migration (line f) than without (line e) since the energy gain to the fish moving into the upstream region is larger than the loss to the downstream popu- lation. This includes the loss of energy due to the cost of upstream migration.
Any model of nature suffers from simplicity, and this one is no exception. The greatest deficiency in this model is the failure to include provisions for reproduction and growth of young. The movement of a juvenile fish or egg mass may have more potential for drawing energy than an equivalent mass of larger fish. Another shortcoming is the Pack of data on population levels of fishes at the two different stations. Only one downstream value was available and had to be used for both compartments. However, this model shows some consistency with the hypothesis for an adaptive role of migration in complex stream ecosystems. SUMMARY
1. Patterns of fish and other aquatic animal movements were in- vestigated in two North Carolina streams from April, 1968 to June, 1970 using weirs with traps.
2. Community metabolism was measured during the same period using diurnal analysis of oxygen.
3. More animals were caught moving downstream than up, but more than twice as much mass (three times for fish alone) of or- ganisms was sampled moving upstream than down. This was true for all locations sampled and for virtually all months of the year.
4. A pattern of larger fishes moving upstream and smaller fishes moving downstream was observed for virtually all species of fish.
9. Migration was by far heaviest each year during the months of blarch, April and May. This was true for all species considered as a group and also for most species considered individually.
6. All physical evidences of spawning condition were observed during periods of maximum movement for each respective species, and ripe specimens (discharging eggs or milt) were almost invariably sampled moving upstream. Spent individuals were sometimes sampled moving downstream, but never upstream. Thus, it is assumed that the movement patterns are connected with spawning and that fish move upstream to spawn.
7. The very low recapture (6.9 percent) of marked individuals
indicates that the movements are true migrations and are not,
in general, merely intercepts of home-range movements. The
low returns also indicate that for many fish the upstream move-
ment is not accompanied by a return downstream movement.
The fate of these post-spawning individuals is not known.
8. Movement of fishes in the streams was increased by a rise in
water temperatures during the spring months and by a rise in
water level during all periodsof the year except during the
winter.
9. Each species was more Likely to travel one at a time than by
two's or more, but concentrations of many individuals traveling
together, or at least on the same day, occurred more fre- C-
quently than would be expected by chance alone.
10. Metabolism for the entire aquatic community of New Hope Creek
was moderately low compared with other areas studied. Gross
primary production at the principal sampling station ranged
from 0.2 to almost 9 g O2 m-2 day-l. Normal values were about
1 and 1.5 g O2 m-2 day-l, respectively. Both gross primary
production and community respiration were greatest in April
and May, wtih another smaller peak in the autumn. Respiration a was nearly always greater than photosynthesis, indicating that
New Hope Creek is dependent upon external sources for much
of its organic energy. *
11. Although volume metabolism was considerably greater in some regions than in others, areal metabolism (i.e.,volume metabolism corrected for depth) was remarkably constant throughout the stream. The generally shallower upstream reaches of the stream had the highest volume photosynthesis and respiration.
12. The hypotheses are presented that behavioral patterns of up-
stream migration for spawning have selective value in starting juvenile fishes in a region where food resources are concentrated. Later dispersal to other, often more stable and less stressful, areas may also have selective advantage. Other potential advantages accruing from migration include: recolonization of defaunated regions, such as those following a drought; overcoming displacement tencencies of a current; more efficient population control of prey; genetic interchange; and distribu- tion of minerals.
13. An energy diagram was drawn comparing calculations of metabolism and fish migration. About 0.01 percent of the entire eco- system's energy usage is contributed by the migrating animals.
14. About 40 percent of the estimated standing crop of fish above the
principal sampling station were involved in migration during one year. Upstream migration alone could replace the population
of part of New Hope Creek in 3.7 years. Downstream migration
alone would take 7.9 years, and replacement from above and
below would take 2.5 years. 15. The loss of phosphorus from the watershed abmc thc Coacretc
Bridge by all processes connected with New Hope Creek was small
compared with the standing crop of phosphorus, and may ha.ve
been replaced by rainfall. The mount of phosphorus replaced
upstream by migrating fish was small considered on an annual
basis, but during the summer months upstream contributions
of phosphorus from the spring and summer fish runs may be
important. In addition, the standing crop of fishes was
an important reservoir of phosphorus.
16. New Hope Creek can be considered an energy-information exchange
system similar to many other such systems. Upstream regions lose
a certain part of their energy reserves to downstream popula-
tions, and, in return, downstream populations supply genetic
information and control functions to the upstream reaches.
17. Analog simulation of fish populations in New Hope Creek with
and without migration indicate that observed patterns of
migration can increase the energy accrual to a population of
fishes. LITERATURE CITED
Allee, W.C. --et al., 1949. Principles of animal ecology. Saunders
Company, Philadelphia. 837 p. Allen, K.R., 1951. The Horokiwi Stream. Bull. Mar. Dept. New Zealand Fisheries, No. 10. Alverson, P., 1964. Demersal fisheries of the North Pacific. H.R. MacMillan Lectures in Fisheries. The University of British Columbia, Vancouver, B.C. American Public Health Association --et al., 1965. Standard methods for the examination of water and waste water including bottom
sediment and sludges. Boyd Company, Albany, New York. 742 p. Ammon, F.V., 1954. Die mathematische Erfassung der naturlichen
Selbstreinigung und der Abwasserbelastung in Fliessegewassern.
-In Biologie und Chemie des ungestauten and gestauten Stromes. Muncher Bietrag zur .4bwasser, Fisherie- und Flussbiologie,
Band 2. R. Oldenbourg, Munchen. Anderson, N.H. and D.M. Lehmkukl, 1968. ~atostrophicdrift and
insects in a woodland stream. Ecol. 49 (2): 188-206. Aro, . and M. Shepard, 1967. Spawning populations of North Pacific salmon: Pacific salmon in Canada. In Salmon of the North
Pacific Ocean, Part IV. Institute of North Pacific Fisheries
Commission. Bull. No. 23: 255-327. Ball, R.C. and F.E. Hooper, 1961. Translocation of phosphor~rsi.n a trout stream ecosystem. Radioecology: First National Symposium on Radioecology: 217-238.
Bangham, Ralph V. and Neville L. Bennington, 1938. Movement of
fish in streams. Transactions of the American Fisheries Society 68: 256-268.
Banks, J.W., 1969. A review of the literature on the upstream mi- gration of adult salmonids. J. Fish. Biol. l: 85-136.
Behmer, David J., 1964. Movement and angler harvest of fishes in the Des Moines River, Boone County, Iowa. Iowa Academy of Science. Vol. 71: 259-263. Beyers, Robert J. --et al., 1963. Directions for the determination of changes in carbon dioxide concentration from changes in pH.
Pub. Institute of Marine Sciences, University of Texas.
454-489. .4 Bigelow, Henry B. and Welsh, W., 1924. Fishes in the gulf of Maine. -In Bulletin of the United States Bureau of Fisheries. Vol. XL. Government Printing Office, Washington, D.C. 532 p. Bigelow, Henry B. and William C. Schroeder, 1953. Fishes of the
Gulf of Maine. U.S. Fish and Wildlife Service, Fishery Bulletin
74. 577 p. Bjornn, Ted C. and Jerry Mallet, 1964. Movements of planted and wild trout in an Idaho River system. Transactions of the
American Fisheries Society 93 (1) : 70-76. Breder, Charles M., 1948. Field book of marine fishes of the Atlantic coast. G.P. Putnam and Sons, London. 303 p.
Brett, J.R., 1965. The relation of size to rate of consumption and sustained swimming speed of sockeye salmon (Oncorhynchus
nerka). J. Fish. Res. Bd. Canada. 22 (6): 1491-1501.
Brett, J.R. and D.B. Sutherland, 1965. Respiratory metabolism of pumpkinseed (Lepomis gibbosus) in relation to the swimming
speed. J. Fish. Res. Bd. Canada. 22 (2): 405-409. Brett, J.R., 1970. Fish--the energy cost of living. -In William J. McNeil (ed.) Marine aquiculture, selected papers. Oregon State University Press, Corvallis, Oregon.
Brody, Samuel, 1945. Bioenergetics and growth, with special reference to the efficiencies complex in domestic animals. Reinhold Pub. Corp., New York. 1023 p. Brown, Edward H., 1961. Movement of native and hatchery-reared game fish in a warm-water stream. Transactions of the American Fisheries Society, 90: 449-456. Carbine, W.F. and David S. Shetter, 1943. Examples of two way fish weirs in Michigan. Transactions of the American Fisheries Society. 73: 70-89. Carnes, William C., James R. Davis, and Buford L. Tatum, 1964.
Appendices to the survey and classification of the Deep-Haw
rivers and tributaries, North Carolina. North Carolina Wildlife Resources Commission. Raleigh, N.C. 35 p. Carpenter, Edward John, 1-967. Movements of redbreast sunfish jn the Little River, I?iake %nd Johnston counties, North Caroljna. Unpublished Master's thesis, North Carolina State University at Raleigh. Chur~hill,M.A., R.A. Buckingham, and H.L. Elmore, 1962. The prediction of stress reaeration rates. Tennessee Valley Authority, Division of Health and Safety, Environmental Hygiene Branch. Chattanooga, Tennessee, 98 p.
Cole, Richard A., 1969. The effect of macrophytes on the abundance
and diversity of macroinvertebrates in an enriched stream. Un- published Ph.D. thesis. Penn. State Univ, Univ, Park, Pa. Cooper, Charles F., 1969. Nutrient output from managed forests.
-In Eutrophication: causes, consequences, corrections. Proceedings of a Symposium, National Academy of Science, Washington, D.C. 446-463. Copeland, B.J., 1965. Fauna of the Arkansas Pass Inlet, Texas. I. Emigration as shown by tide trap collections. Pub. Inst. Marine Sciences, University of Texas, No. 11. 9-21. Copeland, B.J. and W.R. Duffer, 1964. Use of a clear plastic dome to measure gaseous diffusion rates in natural waters.
Limnol. Oceanog. 9 (4): 494-499.
Cox, George W., 1961. The relation of energy requirements of tropical finches to distr5bution and migration. Ecology Cox, George W., 1968. The role of competition in the evolution of migration. Evolution 22: 180-192. Deevey, E.S., 1969. Book review of Ramon Margalef, Prospectives in ecological theory. Limnol. Oceanog. 14 (3): 313-315. Donalson, John Russell, 1967. Phosphorus budget of Iliamna Lake, Alaska as related to the cyclic abundance of sockeye salmon. Unpublished Ph.D. thesis, University of Washington, Seattle.
Federal Water Pollution Control Administration, 1969. Methods for chemical analysis of water and wastes. U.S. Dept. of the
Interior, FWPCA, Cincinnati, Ohio. 280 p. Foster, Neal, 1969. Factors in the origins of fish migrations.
Bull. of the American Littoral Society. 6 (1): 27-31. Funk, John L., 1955. Movement of stream fishes in Missouri. Trans- actions of the American Fisheries Socity. 85: 39-57. Gerking, Shelby D., 1959. The restricted movement of fish popula- tions. Cambridge Philosophical Society, Bio. Rev. 34: 221-242.
Goldman, C.R., 1969. Micronutrient limiting factors and their de- tection in phytoplankton populations. -In C.R. Goldman, Primary productivity in aquatic environments. University of California Press, Berkeley. 121-136. Gosz, J.R., G.E. Likens, and F.H. Bormann, Intrasystem nutrient
cycling from litter fall on the Hubbard Brook Experimental
Forest. -In Ecology and systematics. Cornell University, Ithaca, New York and Yale School of Forestry, Yale University, New Haven, Connecticut. Gunning, Gerald E. and Robert C. Schoop, 1961. Restricted move- ments of the american eel, --Anguilla rostrata,in freshwater streams with comments on growth rate. Tulane Studies in
Zoology. 9: 265-272.
Hall, Charles A.S. and John W. Day, 1970. A new means of measuring
the diffusion constant of natural waters using a clear plastic
dome. Unpublished manuscript. 12 p.
Haney, Paul D., 1954. Theoretical principles of aeration. J.
Amer. Water Works Assn. 46: 355-376.
Harden Jones, F.R., 1968. Fish migration. Edward Arnold, London.
271 p.
Hartman, G.F., T.G. Northcote, and C.C. Lindsey, 1962. Comparison of
inlet and outlet spawning runs of rainbow trout in Loon Lake,
British Columbia. J. Fish. Res. Bd. Canada. 19 (2): 173-200.
Heape, Walter, 1931. Emigration, migration and nomadism. W. Heffer
and Sons, Cambridge. 342 p.
Hellier, Thomas R., 1960. Fish production and biomass in relation
to photosynthesis in the 'Laguna Madre of Texas. Pub. Inst.
Marine Sciences, University of Texas. 8: 1-22.
Hoskin, Charles M., 1959. Studies of oxygen metabolism of streams
of North Carolina. Pub. Inst. Marine Sciences, University of
Texas. 7: 186-192.
Hobbs, Horton. Unpublished check list of fishes in New Hope Creek. 2 p. Hughes, D.A., 1970. Some factors affecting drift and upstream move-
ments of Gammarus pulex. Ecol. 51 (2): 301-305. Hunt, Robert L., 1964. Dispersal of wild brook trout during their
first summer of life. Transactions of the American Fisheries
Society. 94 (2) ; 186-188. Hutchinson, George E., 1957. A treatise on limnology. Vol. I.
John Wiley and Sons, New York. 978 p.
Idler, D.R. and W.A. Clemens, 1959. The energy expenditures of
Fraser River sockeye salmon during the spawning migration to
Chilko and Stuart Lakes. International Pacific Salmon
Fisheries Commission.25 p.
Juday, C. --et al., 1932. Limnological studies of Karluk Lake, Alaska, 1926-1930. U.S. Bureau of Fisheries Bulletin. 57:
407-336.
Krokhin, E.M., 1967. (Effect of size of escapement of sockeye salmon
spawners on the phosphate content of a nursery lake.) Izvestiya
Tikhookeanskogo Nauchno-Issledovatellskogo Instituta Rybnogo Kjozyaistva i Okeanografii. 57. Fisheries Research Board of
Canada. Translation Series No. 1186. 31-54.
Likens, G.E. --et al., 1967. The calcium, magnesium, potassium, and sodium budgets for a small forested ecosystem. Ecol. 48 (5):
772-7135. Levins, Richard, 1964. The theory of fitness in a heterogeneous
environment. IV. The adaptive significance of gene flow.
Evolution 18: 635-638. McFarland, W.N., 1063. Seasonal change in the nunher and the bimnass
of fishes from the surf at Mvstmg Island, Texas, Pub. Inst.
Marine Sciences, University of Texas. 9: 91-105.
McLaren, I.A., 1963. Effect of temperature on growth of zooplankton,
and the adaptive value of vertical migration. J. Fish. Res.
Bd. Canada. 20: 685-727.
Manny, Bruce A. and Charles A.S. Hall, 1969. Diurnal changes in
stratification and dissolved oxygen in the surface waters of
Lake Michigan, Proc. 12th Conf. Great Lakes Res. Internat.
Assn. Great Lakes Res. 622-634.
Manzer, J.I., 1960. The sea life of Canada's Pacific salmon.
Canadian Geographical Journal. LXXIL (1) : 2-11.
Margalef, Ramon, 1963. On certain unifying principles in ecology.
Am. Naturalist. 97: 357-374.
Margalef, Ramon., 1968. Perspectives in ecological theory.
University of Chicago Press, Chicago. 102 p.
Mather, Frank J., 1969. Long distance migrations of the tunas and
marlins. Bull. of the Am. Littoral Society. 6 (1.) : 6-14.
Mayr, E. and W. Meise, 1900. Theorestisches zu-,.Geschichte des
Vogelzuges. Der Vogelzuges (Berlin). 1: 149-172.
Meek, Alexander, 1961. The migrations of fish. Edward Arnold,
London. 414 p.
Miller, Robert V., 1969. Continental migrations of fishes. Bull.
of the Am. Littoral Society. 6 (1): 15-23. Minckley, W.L., 1964. Upstream movements of Gammarus (Amphipoda)
in Doe Run, Meade County, Kentucky. Ecol. 45 (1): 195-197. Neave, Ferris, 1964. Ocean migrations of Pacific Salmon. J. Fish Res. Bd. Canada. 21: 1227-1244. Neave, Ferris and M.G. Hanavan, 1960. Seasonal distribution of
some epipelagic fishes in the Gulf of Alaska region. J. Fish. Res. Bd. Canada. 17: 221-233. Niemuth, Wallace, 1967. A study of migratory lake-run trout in the
Brule River, Wisconsin. Fish Management Division of the Wisconsin Conservation Department. 80 p. Nixon, S.W., 1969. Some characteristics of hypersaline ecosystems. Unpublished Ph.D. thesis. University of North Carolina at Chapel Hill. Nixon, S.W. and H.T. Odum, 1970. A model for photoregeneration in
brines. ESE Notes. 7 (1): 1-3. Dept, of Envr. Sciences and Eng., University of North Carolina at Chapel Hill. Odum, E.P. in collaboration with H.T. Odum, 1959. Fundamentals of
ecology. 2nd ed. Saunders. Company, Philadelphia. 546 p. Odum, H.T., 1956. Primary production of flowing waters. Limnol.
Oceanog. 2: 85-97.
Odum, H.T., 1967a. Energetics of world food production. -In
The world food problem. Vol. 111. U.S. Govt. Printing Office, Washington, D.C. 55-94.
Odum, H.T., 1967b. Work circuits and system stress. -In Symposium on primary production and mineral cycling in natural ecosystems. University of Maine Press, Orono. 81-138. Odum, H.T. and E.P. Odum, 1955. Trophic structure and productivity
of a windward coral reef community on Eniwetok Atoll. Ecol.
Monogr. 25: 291-320.
Odum, H.T. and C.M. Hoskin, 1958. Comparative studies on the meta-
bolism of marine waters. Pub. Inst. Marine Sciences, University
of Texas. 6: 159-170.
Odum, H.T., John E. Kantlon, and Louis S. Kornicher, 1960. An organizational hierarchy postulate for the interpretation of
species -individual distributions, species entropy, ecosystem evolution and the meaning of a species -variety index. Ecol.
Odum, H.T. and F.R. Wilson, 1962. Further studies on reasration
and metabolism of Texas bays, 1958-1960. Pub. Inst. Marine
Sciences, University of Texas. 8: 23-55. Odum, H.T., R.J. Beyers, and N.E. Armstrong, 1963. Consequences of small storage capacity in nannoplankton pertinent to measure- ment of primary production in tropical waters. J. Mar. Res.
21 (3) : 191-198. Odum, H.T. --et al., 1969. Coastal ecological systems of the United States, a source book for estuarine planning. A report to
the Federal Water Pollution Control Administration. Contract
RFP 68-120. 1878 pp. Odum, H.T., S.W. Nixon, and L.H. DiSalvo, 1970. Adaptations for photoregenerative cycling. -In H.T. Odum (ed.) Systems of man and nature. Annual Report to the Atomic Energy Commission FY-70.
Owens, M., 1969. Some factors involved in the use of dissolved-
oxygen distributions in streams to determine productivity.
-In C.R. Goldman (ed.) Primary productivity in aquatic environ- ments. University of California Press, Berkeley. 209-224. Preston, F.W., 1962. The canonical distribution of commoness and rarity: Part I. Journal of Ecology. 43: 185-215. Raney, Edward C. and Dwight A. Webster, 1942. The spring migration of the common white sucker, Catostomus -c. commersonnii, in Skaneateles Lake Inlet, New York. Copeia. 139-148. Rayner, H.J., 1942. The spawning migration of rainbow trout at Skaneateles Lake, New York. Transactions of the American Fisheries Society. 71: 180-183. Rawson, D.S., 1952. Mean depth and the production of large lakes. Ecol. 33: 513-521. Rawson, D.S., 1960. A limnological comparison of twelve large lakes in Northern Saskatchewan. Limnol. Oceanog. 5 (2): 195-211. Reimers, Norman, 1966. A low-maintenance fish barrier with free-flow
characteristics. The Progressive Fish Culturist. 28 (2): 103-107. Remezov, N.P. --et al., 1963. Dynamics of the interaction between an oak wood and soil. Abstracts from a conference on methods for
research station studies of soil processes, Moscow. Research Triangle Regional Planning Commission, 1968. Map of area
of Korstian division, Duke Forest. 1 p.
Ricard, Matthieu, 1968. The mystery of animal migration. Hill and Wang, New York. 194 p.
Rodin, L.E. and N.I. Bazilevich, 1965. Production and mineral cycling in terrestrial vegetation. Oliver and Boyd Company,
London. 287 p.
Royce, William F., 1967. Observations on the spearfishes of the central Pacific. U.S. Fish and Wildlife Service. Fishery Bulletin. 57: 497-554.
Royce, William F., Lynwood S. Smith, and Allan C. Hartt, 1968. Models of oceanic migrations of Pacific salmon and comments on guidance mechanisms. U.S. Dept. of the Interior. U.S. Fish and Wildlife
Service. Fishery Bull. 66 (3): 441-462.
Scott, Donald C., 1949. A study of stream populations of rock bass.
Investigations of Indiana Lakes and Streams. 3: 169-234.
Shetter, David S., 1938. A two-way fish trap for use in studying stream fish migrations. Third North American Wildlife Con-
ference. 331-338.
Shetter, David S., 1968. Observation on movements of wild trout in two Michigan stream drainages. Transactions of the American
Fisheries Society. 97 (4J: 472-481.
Siebert, H.C., 1949. Differences between migrant and nonmigrant
birds in food and water uptake at various temperatures and
photoperiods. Auk. 66: 128-163. Simmons, Ernest G. and Hinton D. Hoese, 1959. Studies of the hydro-
graphy and fish migrations of Cedar Bayou, a natural tidal inlet
on the central Texas coast. Pub. Inst. Marine Sciences, Uni-
versity of Texas. 56-80.
Smith, M.W. and J.W. Saunders, 1958. Movements of brook trout,
Salvelinus fontinalis, between and within fresh and salt
water. J. Fish. Res. Bd. Canada 15 (6): 1403-1449.
Smith, M.W. and J.W. Saunders, 1967. Movements of brook trout in
relation to an artificial pond on a small stream. J. Fish.
Res. Bd. Canada. 24 (8): 1743-1761.
Smith, M.W. and J.W. Saunders, 1968. Effect of pond formation on
catches of brook trout from a coastal stream. J. Fish. Res.
Bd. Canada. 25 (2): 209-238.
Smith, Robert Leo, 1966. Ecology and field biology. Harper and Row,
New York. 686 p.
Sollins, P., 1970. Measurement and simulation of oxygen flows and
storages in a laboratory blue-green algal mat system. b1.S.
thesis, University of North Carolina at Chapel Hill.
Steeman Nielson, E. and E. Aabye Jensen, 1957. Primary oceanic
production. The autotrophic production of organic matter in
the oceans. Galathea Rep. 1: 49-136.
Stefanich, Frank S., 1952. The population and movement of fish in
Prickley Pear Creek, Montana. Transactions of the American
Fisheries Society. 81: 260-274. Strasburg, Donald W., 1969. Billfishes of the central Pacific
Ocean. U.S. Wildlife Service Bureau of Commercial Fisheries. 11 p.
Streeter, H.W. and E.R. Phelps, 1925. A study of the pollution and natural purification of the Ohio River. U.S. Public Health
Service, Washington, D.C. Bulletin 146. Sumner, F.H., 1962. Migration and growth of the coastal cutthroat trout in Tillaxook County, Oregon. Transactions of the American
Fisheries Society. 91 (1) : 77-83. Talbot, Gerald B. and James E. Sykes, 1958. Atlantic coast migrations of amesican shad. U.S. Fish and Wildlife Service, Fisheries
Bulletin. 58: 473-488. Tate, William H., 1949. Studies on smallmouth black bass in Iowa streams: stream dynamics and smallmouth movements. 11th Midwest Wildlife Conference. Mimeo 3 p. Vinogradov, Aleksandr Pavlovich, 1953. The elementary composition of marine organisms. Julia Efron and Jane K. Setlow (trans.) Yale University Press, New Haven, Connecticut. 647 p. Warner, Kendall, 1959. ~igrationof landlocked salmon in the fish
river lakes, Maine. The Journal of Wildlife Management 23:
17-27. Waters, T.F., 1965. Interpretation of invertebrate drift in
streams. Ecol. 46: 327-334. Welch, Harold C., 1968. Use of modified diurnal curves for the
measurement of metabolism in standing water. Limnol. and Oceanog. 13 (4): 679-687. West, G.C., 1960. Seasonal variation in the energy balance in
relation to migration. Auk. 77:306-329.
Woodwell, George M., 1970. Personal communication.
Zimmerman, J.L., 1965. Bioenergetics of the dickcissel (Spiza
americana). Physiol. Zool. 38: 370-389.
APPENDIX A
DIFFUSION PROCEDURES USED IN NEW HOPE CREEK
METABOLISM STUDIES
One of the most difficult tasks in determining community meta-
bolism from changes in oxygen is establishing the rate at which
oxygen enters or leaves the water if the water is not at 100 per-
cent saturation. Two papers in 1954 (Haney, 1954; Ammon, 1954)
summarized the knowledge about diffusion to date and concluded that
the amount of gas transfer could be calculated by equation (1):
where D is the diffusion rate per area, S is the saturation defi-
cit between water and air, z is the depth, and k is the gas transfer coefficient defined on a volume basis. Since the rate of diffusion
is linearly proportional to the saturation deficit, no correction
of the constant is needed at different oxygen concentrations. See
Odum (1956) and Owens (1969) for a more detailed consideration of
this.
Diurnal Curve Method for Determining the Diffusion Constant
In the study in New Hope Creek, three methods were used to esti- mate the diffusion coefficient. In the first, appropriate values were substituted into the expression (2) supplied by Odum (1956):
where k is the diffusion constant in g oxygen rn-3 hr'l, qm is the rate of change at a time before dawn in mg 1.' hr' 1 , qe is the rate of 1 change at a time after sunset in mg 1-I hr - , Sm is the percent ss turil- tion deficit before dz~nat the time chosen for ca.lc112atj.ng q,, end Se
is the percent saturatkm deficit aftm sunset at the time chcse.7 fo- calculating qe. (These terms must be chosen careftilly to avoLd non- representative water masses .) If respiration at night is constant, the change in dissolved oxygen will be proportional to the percent saturation of the water. However, as pointed out by Odum and Vilson
(l962), hielch (l968), and Owens (l969), the respiration of an aquatic ecosystem is not constant, especially in systems with many small algal cells. In general, respiration is considerably higher in the post-sunset hours than in the pre-dawn hours, due to daytime storages in the cells (Odum et al., 1963), higher temperatures, and higher oxygen values. Not correcting for different levels of respiration can introduce serious error into the determination of k and will, in gene- ral, give values that are too large. Also, as pointed out by Owens s,
(1969), the difference between the saturation values must be large to give meaningful results, because of statistical error. This method was used for New Hope Creek data on several days with smooth oxygen curves showing no abrupt changes caused by different water masses or sampling error. Table A-1 gives these results, which varied widely from day to day.
A possible correction for the differences in respiration bras given by Odum and Wilson (1962) :
where kl is the volume based diffusion constant, q is the rate of
change in g mm3 hr-I per gradient of 100 percent saturation deficit, .I r is the independently determined respiration also in g m'3 hr-1 , and S is the percent saturation deficit. Neither Odum and FTilson
nor this investigator was able to apply this correction because of
insufficient independent data on respiration.
In open waters, Copeland and Duffey (1964) and Owens (1969) used different diffusion constants at different times of the day
due to changes of circulation patterns with changes in wind and tem-
perature stratification. Since all of New Hope Creek flows through more or less protected areas, either through a gorge or in areas
completely surrounded by trees, there were few wind effects apparent
over most of the stream. Hourly corrections of k due to differences
in wind velocity were considered unimportant.
Stream Morphology Method -of Determining -the Diffusion ---.---Constant A second method of determing k uses the expression (4), originally developed by Streeter and Phelps (1925) and improved upon by Churchiil --et al. (1962)
The diffusion constant k2 in this expressi-on (4) is defined dif-
ferently from k of equations 1-3, k2 is the g m-3 hr-I diffusion when the saturation deficit is one mg 1 where k2 is in days-', R equals the hydaulic radius (approximately the depth), in feet, and V is the cross sectional mean velocity. Corrections can be made for other temperatures at 2.41 percent per degree C (Churchill p;t alL, 1962).
-Two Diffusion Coefficients---- -and --.-Conversion The diffusion equations for the two commonly used diffusion coeffi- cients may be compared in equations (5) and (6).
D = kS = k (100 - percent oxygen saturation of stream) . (5) Table A-1. Diffusicn Constants Derived from Diurnal Oxygen Data
Date Location Depth Diffusion constant m g rn-3 hr-l
1969 Mar. 29 Concrete Bridge 0.55 2.04
Apr. 1 11 t I 0.50 6-57
Apr. 25 Blac kwood 0.25 2.50
May 16 Concrete Bridge 0.41
June 2
July 1
July 25
1970 Feb. 14 I I I I 0.40
Feb. 14 23 1 where D is the total oxygen diffusion rate expressed as g 02 m-3 hr'l, k is the diffusion coefficient as g m-3 hr-I 100 percent saturation deficitmL (base e), and S is the percent saturation deficit. k is the slope of a semilog plot (natural log) of diffusion with time.
The total diffusion process using k2 is:
where k2 is the re-aeration rate coefficient (days-', base lo), Cs is the oxygen saturation concentration (mg 1-I or g m-3) at the prevailing temperature and pressure, C is the actual stream oxygen concentration, and D2 is expressed as g 02 m-3 daym1. k2 is the slope with time of a semilog plot (base 10).
Solving (5) and (6) for k and k2, respectively, gives:
k = D s-' in g O2 m-3 hr-I per 100 percent saturation deficit (7) - 1 -1 -1 ,-3 k2 = D (Cs - C) in g O2 m-3 day g (8)
The last formula reads "grams oxygen per cubic meter per day per grams per cubic meter difference between oxygen saturation and stream oxygen
values .I1
Dividing (8) by 24 gives diffusion per hour; thus, the time intervals used in each formula are made the same. One hundred per- cent saturation deficit expressed as grams per cubic meter is Cs.
Since the engineering formulation for re-aeration coefficients (k2) is defined in terms of logl0, k2 must be multiplied by 2.3 to yield the equation: 2.3 ' (k2) Cs k = in g m-3 hr-I per 100 percent 24 saturation deficit (9)
Thus, with a knowledge of the oxygen saturation value for the time in question, k2 can be expressed as k, and the formula of Churchill et al. can be used in diurnal oxygen studies. A further correction must he made for temperature, since Churchill's formula which is base.d only on differences in stream depth and flow is temperature-dependegt with different molecular activity rates. This correction: 2.41 percent Q increase or decrease per degree above or below 20O C as a geometric ratio, somewhat compensates for the increase in k as the saturation \i value of water falls with increased tem~erature. For purposes of cal- culation, the formula k2(T) = k2 (at 20' C) 1.0241 (T - 20) is used. Thus, k is not temperature-dependent due to changing saturation values, while k2 is.
The area based diffusion coefficients obtained by this method were divided by the average depth of the water reach being considered to give values for volume diffusion corrections (equation 1). Table
A-2 gives results of diffusion constant estimates for different mean depths, mean flow conditicns, and different stage levels for the stretch of creek above the Concrete Bridge station through which water flows in one hour.
-Dome --- Measurements -of Diffusion Coefficient The third method of determining diffusion constant is a direct empirical approach using methods derived from the work of Copeland and Duffey (1964), who determined k by comparing the changes in oxy- gen concentrations in the water with changes under a clear plastic dome. Similarly, k has been determined at the Institute of Marine
Sciences, University of North Carolina, by measuring the rate of transfer of carbon dioxide across the water surface using an open system with a plastic dome and an infrared C02 analyzer (Hall and
Day, 1970). Table A-2. Predicted Values for Diffusion Constant for New Hope
Creek Above Concrete Bridge Station Using Formula Based on Average
Depth and velocitya
Average K2 day-' K k depth (per area) g m-2 hr-I g ~n-~hr-l-l m at 20' C atmosphere-' atmosphere
a See equations (4 and 9).
In the present studies, the diffusion rate was measured directly by determining the rate of oxygen re-entry into a clear plastic domz after filling it with pure nitrogen. In theory, any gas can be used, since, according to Dalton's law of partial pressure, the oxygen would diffuse independently of the concentrations of other gases present. However, experiments with CO and methane were abandoned 2 after these gases diffused into the water, allowing water to rise too rapidly into the inside of the dome, changing gas values. Since ordinary air is 78.09 percent nitrogen by volume, water in contact with the atmosphere would be at about 78 percent saturation relative to an atmosphere of 100 percent pure nitrogen. This relatively small difference allows the diffusion of oxygen into the sphere to be stu- died without interference by substantial loss of the atmosphere within the sphere and the resultant changes of sphere volume.
The apparatus for these determinations were set up according to Figures A-1 and A-2. The clear plastic dome was attached to a plywood collar which floated the dome on the surface of the water.
An oxygen probe (Yellow Springs Instrument $15419) was inserted in the top of the dome with an airtight seal. Thus, tk probe was measuring the partial pressure of oxygen in the atmosphere within the dome. The dome was then sunk within the stream, tilted at an angle that allowed a substantial flow of stream water over the probe. The flow was considered sufficient if additional manual agitation of water over the probe did not give higher readings on the meter. This was necessary because the probe requires a minimum flow of water for accurate readings. Figure A-1. Use of clear plastic dome to measure diffusion constant. a. The dome is immersed into the water to equilibrate the probe with oxygen in the stream. b. The water-filled dome is turned upright, filled with N2, and floated on the water. The hole in the dome, open to the atmosphere, insures a pressure of
1 atmosphere inside the dome after gas flow is stopped. c. The diffusion constant, k, is computed from the rate of oxygen re-entry into the dome in the water.
Once the probe had reached equilibrium with the water with a constant reading for several minutes, the meter was adjusted to '10,' which represented 100 percent of the partial pressure of the oxygen in the water. Three Winkler oxygen determinations were taken at this time. Thus, the '10' on the meter scale was equivalent to the total concentration of oxygen in the water as determined by the average of the three Winkler tests.
While still submerged, the dome was turned collar-side down.
Nitrogen was introduced into the dome through a hole in the side; this procedure floated the dome on the water surface with an atmos- phere of pure nitrogen, After a short period, the oxygen meter would read near zero, as there was little or no oxygen within the dome. An additional hole in the dome insured that the nitrogen within the dome was at 1 atmosphere. The nitrogen was turned off, the holes plugged, and time and meter reading records were kept as oxygen diffused into the atmosphere of the dome from the water.
It is inaccurate to calculate rates from finite measurements of change as a simple quotient. Instead, the integral form must be used because the expression is non-linear. The derivation in Table
A-3, supplied by Dr. H.T. Odum, is required to provide the units necessary for easy computation.
Diffusion Constants Used for These studies
A11 diffusion results were converted to a volume basis for comparison. The results for diffusion constant determinations by each method varied from one to another and within the three methods used (Figure 10). The results of the stream morphology estimates Table A-3. Basis for Calculations of Diffusion Coefficient from
Dome Measurements
It is desirable to find the area-based diffusion coefficient for oxygen, K, in grams oxygen per square meter per hour per atmos- pheres gradient.
K as defined above relates the flow of oxygen into the dome to difference in partial pressure with equation (1).
3 = KA (pw - pd) (1) where J is in grams of oxygen per hour, A is the area of dome in square meters, pw is partial pressure of oxygen in water in atmos- pheres, and pd is partial pressure of oxygen in dome as percent of total pressure. Solving for K gives equation (2) and the desired units. J K = g m-2 atmosphere- 1 (2) A (P, - P~) The weight of oxygen (Q) in air phase in the dome is given by equation (3) which has geometrical density considerations. Q is in grams of oxygen
where v is the volume of dome above waver in ml, /J is the density of air at elevation and temperature, and p is partial pressure of oxygen in dome in percent of total pressure.
Next write the differential equation stating that the rate of change of the weight of oxygen in the dome's air phase is equal to the flux of oxygen across the surface (J). Table A-3 continued
Then substitute in equations (1) and (3).
Next integrate as in equation (7).
(7
Integral tables show the left hand expression to be of the form
1 = - In (a + bx) $ a d; bx b p is the variable x; substituting back in a, b, and p one finds d d the integral expression 2 + I
To evaluate the integration constant, substitute p as initial 0 pressure when time is zero at start. Then
The final integral equation becomes Table A-3. continued
Changing natural logs to base 10 logs and solving for K one
obtains
For 100 percent saturation with atmospheric oxygen as the gradient, multiply by 0.20s the oxygen fraction of atmosphere. The constant
in the above equation becomes 0.48 instead of 2.3. Where v is the volume of dome in ml,p is the density of air in mglml, A is the area of dome in square meters, t is time after start in hours, p, is percent saturation, and po is percent oxygen in dome at start.
Results for five estimates of diffusion constants using the dome method are given in Table A-4. Table A-4. Estimates of Diffusion Constant (K) Obtained Using the Dome Method for Representative Pools and Riffles Above Concrete
Bridge Station
Date Location Depth K k (average) g rnm2 hr-I g m-3 hr-I at 0 percent at 0 percent saturation saturation"
1970 May 13 Riffle 0.55 0.95 1.73
May 22 Riffle 0.58 0.49 0.83
June 14 Riffle 0.35 0.178 0.712
May 21 Pool 0.55 0.10 0.18
June 16 Pool 0.35 0.036 0.079
Based on average depth for stream at that time and place. were in between the pool and riffle estimates obtained with the dome.
The results of the diurnal curve method were scattered and often higher than estimates made using the other two procedures. As men- tioned in "Methods and Materials", the lack of correction for differences in respiration would tend to give high values for k obtained by the diurnal curve method where evening respiration is higher than pre-dawn respiration, as is the case for New Hope Creek.
The stream morphology estimates of diffusion, which were roughly substantiated by the dome measurements, were used for the estimates of metabolism for this study.
Bailey (1970) found similar variations within and between different diffusion determination methods, but concluded that for the shallow reaches of the Sacramento-San Joaquin region, Churchill's formula had reasonably good predictive value.
Diffusion Constants for Other Stations
A diffusion constant of 1 g m-3 hr-I was used for the Blackwood station. This was based on estimates by the stream morphology method of from 0.77 to 1.0 g mm3 hr-I and one estimate by the dome method 1 of 1.3 g m'3 hr - . The area.based diffusion constants were similar to the ones for the Concrete Bridge station, but the shallower nature of the stream gave larger volume values.
The diffusion constant used for the Wood Bridge station was 0.4 g m-3 hr-I at all depths sampled for oxygen changes except during very low water when a diffusion constant of 1 was used. These dif- fusion constants are based on the areal values from the other stations.
A very deep pool located above the shallows at the station gives this site the deepest average depth for all the oxygen stations, resulting in the relatively small volume diffusion constant.