WATER POLLUTION CONTROL RESEARCH SERIES
16010 EXE 12/71
             THE EFFECTS
                 OF
          ARTIFICIAL AERATION
            LAKE ECOLOGY
U.S. ENVIRONMENTAL PROTECTION AGENCY

-------
          WATER POLLUTION CONTROL RESEARCH SERIES
The Water Pollution Control Research Series describes the
results and progress in the control and abatement of pollution
in our Nation's waters.  They provide a central source of
information on the research, development and demonstration
activities in the Environmental Protection Agency,  through
inhouse research and grants and contracts with Federal,
State, and local agencies, research institutions, and
industrial organizations.

Inquiries pertaining to Water Pollution Control Research
Reports should be directed to the Chief, Publications Branch
(Water), Research Information Division, R&M,  Environmental
Protection Agency, Washington, D.C. 20^60.

-------
  THE EFFECTS  OF ARTIFICIAL AERATION

             ON LAKE ECOLOGY
             Arlo Wade  Fast
        Michigan State  University
          East  Lansing,  Michigan
                  for the

    ENVIRONMMTAL PROTECTION AGENCY
          Project No.  16010 EXE


              December 1971
For sale by the Superintendent of Documents, U.S. Government Printing Office
             Washington, D.C. 20402 - Price $4.25
                Stock Number 6601-0232

-------
                  EPA Review Notice
This report has been reviewed by the Environmental Protection
Agency and approved for publication.  Approval does not
signify that the contents necessarily reflect the views and
policies of the Environmental Protection Agency nor does
mention of trade names or commercial products constitute
endorsement or recommendation for use.
                            ii

-------
                         ABSTRACT

            THE EFFECTS OF ARTIFICIAL AERATION
                     ON LAKE ECOLOGY

                            By

                      Arlo Wade Fast
     Two northern Michigan lakes were artificially aerated

using compressed air.  Hemlock Lake, a eutrophic lake, had

only its hypolimnion aerated while thermal stratification

was maintained.  A special hypolimnion aeration device was

used.  Section Four, an oligotrophic lake, was completely

destratified by releasing air from a perforated pipe at the

deepest point in the lake.  Both lakes were studied during

1969 under normal conditions, and during 1970 under test

conditions.

     Artificial hypolimnion aeration of Hemlock Lake caused

oxygen concentrations to increase from 0.0 mg/1 to over

10.0 mg/1 while thermal stratification was maintained.

Zooplankton, zoobenthos and fish distributed throughout the

lake after aeration, while limited to shallow depths before.

Midges emerged from the deepest point following aeration.

Aeration apparently reduced anaerobic nutrient regeneration,

but increased nutrient regeneration through aerobic decompo-

sition of the profundal sediments.  These sediments were

                            ill

-------
highly organic and incompletely decomposed due to the



previous anaerobic conditions.



     Artificial destratification of Section Four Lake



greatly increased the minimum temperatures and heat budget.



Although zoobenthos and surface phytoplankton standing crops



were reduced,  destratification had little apparent effect on



the biota.  Midges emerged from greater depths during



aeration but depth distributions of most organisms,  other



than the crayfish, were not greatly altered.   Crayfish



distributed evenly throughout the lake during aeration.



Changes in their distribution suggests that the thermal



gradient and aggressive behavior of the male are the most



important factors determining their normal depth distribu-



tions.
                            IV

-------
                    TABLE OF CONTENTS


                                                       Page

LIST OF TABLES	    vii

LIST OF FIGURES	      x
INTRODUCTION	.	 .      1

METHODS	     21

     Physical-chemical .....  	     21
     Phytoplankton 	 .......     28
     Primary Production	     29
     Periphyton	     33
     Zooplankton 	  ...........     33
     Zoobenthos	     34
     Crayfish	. .     36
     Emergent Insects. ...... 	  . .     36
     Rainbow Trout .......  	  . .     40
     Statistics.	     50

DESCRIPTION OF THE LAKES	     51

HEMLOCK LAKE	     72

     Hypolimnion Aerator ..............     72
          Description	     79
          Compressor 	 .........     90
          Operation. ................     90
          Aeration efficiencies.  ..... 	     92

RESULTS	     97

     Physical-chemical Parameters. .... 	     97
          Thermal stability. ............    115
          pH,  alkalinity and conductivity	    120
          Phosphorus	    127
          Ca,  Na,  K,  Mg,  DOM, POM	    129
     Primary Production	    129
          Phytoplankton	    129
          Periphyton	    133
                            v

-------
TABLE  OF  CONTENTS—continued                           Page


     Zooplankton  .........«......=.    138
     Zoobenthos.  .............  	    143
     Crayfish.	     193
     Rainbow  Trout	    196

DISCUSSIONS AND CONCLUSIONS.  .....  	    211

     Physical-chemical  Parameters.  .  .  	    211
     Primary  Production.  ..............    218
     Zooplankton  ..................    228
     Zoobenthos.  ............  	    235
     Crayfish.  ...................    255
     Rainbow  Trout  ........  	    257

SECTION FOUR  LAKE.  ........  	    270

     Destratification System ......  	    270
           Compressor ................    275
           Compressor operation ...........    275

RESULTS.  ...........  o  	  ....    276

     Physical-chemical  Parameters.  .........    276
           Temperature and oxygen	    276
           pH,  alkalinity and conductivity.  .....    290
           Phosphorus ................    297
           Ca,  Na, K,  Mg,  DOM,  POM.  .........    297
     Primary  Production.  ..............    301
           Phytoplankton.  ..............    301
           Periphyton ................    305
     Zoobenthos.  ..................    305
     Crayfish.  ...........  	  .    363
     Rainbow  Trout  .................    366

DISCUSSION AND  CONCLUSIONS .............    372

     Physical-chemical  Parameters.  .  .  	    372
     Primary  Production.  ..............    375
     Zoobenthos.  ..................    385
     Crayfish.	    389
     Rainbow  Trout  .................    396

AERATION  TO PREVENT  WINTERKILL ...  	    401

LITERATURE CITED  ...........  	  .    415

APPENDIX  ......  	  .....  	    426

ACKNOWLEDGEMENTS .  .   .  .   .  .  .  .   „  „  „  „  „   „  „  „

                             vi

-------
                      LIST OF TABLES
TABLE                                                  Page

   1.  Section Four Lake limnological data collected
       August 1,  1932 by the Institute for Fisheries
       Research,  Michigan Department of Natural
       Resourceso   Secchi disc depth was 6.8 meters
       on this date. .................   63

   2.  Hemlock Lake limnological data collected July
       28, 1932 by the Institute for Fisheries
       Research,  Michigan Department of Natural
       Resources.   Secchi disc depth was 4.2 meters
       on this date. .................   69

   3.  Hemlock Lake total phosphorus and total dis-
       solved phosphorus collected July 22,  1969.  Two
       water samples were collected from each depth. .  128

   4,  Hemlock zooplankton collected June 11 and July
       15, 1970.   Three samples were collected from
       each two meters depth interval.  Totals repre-
       sent the sum of the average number of zooplank-
       ters per liter from each depth.  Total samples
       on each date = 27 ...............  142

   5.  Hemlock Lake zoobenthos collected during the
       summers 1969 and 1970 with an Ekman dredge.
       125 dredge samples were taken each summer.  Wet
       weights are shown ...... o ........  144

   6.  Emergent midge adults collected from 600
       samples during 1969,  and 650 samples during
       1970.   All  specimens are from Hemlock Lake and
       were collected in emergent insect traps ....  147

   7.  Section Four total phosphorus and total dis-
       solved phosphorus collected July 22,  1969.  Two
       water samples were collected from each depth
       interval.  ...................  300

   8.  Section Four Lake zoobenthos collected during
       the summers 1969 and 1970 with an Ekman dredge.
       125 dredge  samples were taken each summer.  Wet
       weights are shown ...... 	 ....  310
                            vn

-------
LIST OF TABLES—continued

TABLE                                                  Page

   9.  Emergent midge adults collected from 600 sam-
       ples during 1969,  and 650 samples during 1970.
       All specimens are from Section Four Lake and
       were collected in emergent insect traps. ...   319

 A-l.  Hemlock Lake 17 day periphyton weights during
       1969 and 1970.  Samples were incubated on
       plastic slides during June,  July and August
       each year.  Four slides were incubated at each
       of five depths.  Ash-free dry weight is shown
       for each sample. ...............

 A-2.  Hemlock Lake accumulative periphyton weights
       during 1969 and 1970.  Samples were incubated
       starting June 15 each year and a portion was
       retrieved at different times during the summer.
       Samples were incubated on plastic slides.
       Three slides were incubated at each of five
       depths.  Ash-free dry weight is shown for each
       sample ....................

 A-3.  Section Four Lake 17 day periphyton weights
       during 1969 and 1970.  Samples were incubated
       on plastic slides during June, July and August
       each year.  Four slides were incubated at each
       of five depths.  Ash-free dry weight is shown
       for each sample. ...............

 A-4.  Section Four Lake accumulative periphyton
       weights during 1969 and 1970.  Samples were in-
       cubated starting June 15 each year and a por-
       tion was retrieved at different times during
       the summer.  Samples were incubated on plastic
       slides.  Three slides were incubated at each of
       five depths.  Ash-free dry weight is shown for
       each sample. .......... 	

 A-5.  Hemlock Lake zoobenthos collected with an
       Ekman dredge during 1969 and 1970.  Numbers
       and wet weights for the seven most abundant
       taxa are shown in this table for each sample.
       125 samples were collected each summer.  The
       less abundant taxa are listed in Table A-6.
       To verify the total organisms for a given sam-
       ple,  consult both tables.  Depth is in meters
       and weight is in grams ............
                           Vlll

-------
LIST OP TABLES—continued

TABLE                                                  Page

 A-6.  Hemlock Lake zoobenthos collected during 1969
       and 1970.  The less abundant organisms are
       listed in this table.  To obtain a total for
       a given sample, add the values for each sample
        (check each organism) to the values for that
       sample given in Table A-5.  Wet weights are
       shown. ............ 	 , .

 A-7.  Section Four Lake zoobenthos collected with an
       Ekman dredge during 1969 and 1970.  Numbers
       and weights for the seven most abundant taxa
       are shown in this table for each sample.  125
       samples were collected each summer.  The less
       abundant taxa are listed in Table A-8.  To
       verify the total organisms for a given sample,
       consult both tables.  Depth is in meters and
       weight is in grams 	

 A-8.  Section Four Lake zoobenthos collected during
       1969 and 1970=  The less abundant organisms
       are listed in this table.  To obtain a total
       for a given sample, add the values for each
       sample (check each organism) to the values for
       that sample given in Table A-7.  Wet weights
       are shown. ...................

 A-9.  Area-capacity table for Hemlock Lake based on
       January 1957 survey of the lake. .......
A-10.  Area-capacity table for Section Four Lake
       based on January 1957 survey of the lake .  .  .

A-ll.  Hemlock Lake calcium, sodium, potassium, mag-
       nesium, dissolved organic matter  (DOM) and
       particulate organic matter  (POM) collected
       during 1970.  Samples were collected from six
       depths seven times during the summer.  The
       mean concentration for the entire lake is
       shown.  These analyses were made by R. G.
       Wetzel ...............  	

A-12.  Section Four Lake calcium, sodium, potassium,
       magnesium, dissolved organic matter  (DOM) and
       particulate organic matter  (POM) collected  dur-
       ing 1970.  Samples were collected from six
       depths seven times during the summer.  The  mean
       concentration for the entire lake is shown.
       These analyses were made by R. G. Wetzel .  .  .

                            ix

-------
                     LIST OF FIGURES


FIGURE                                                 Page
   1.  Hypothetical phosphorus cycle within a lake
       showing the three main compartments.  The comr-
       ponents of the littoral compartment are also
       shown	     7

   2.  Hypothetical destratification patterns caused
       by diffuse aeration system.  Figure A illus-
       trates a stratified lake, and Figure B a lake
       being destratified	    13

   3.  Hypolimnion aerator of Bernhardt (1967).  From
       Fast (1968)	,	    17

   4.  X-sectional view along principle sample tran-
       sect.  Raft,  emergent insect traps, gill nets
       and transect float are shown	    23

   5.  View of Section Four Lake taken from basin rim.
       Emergent insect traps are stacked on the raft.
       The periphyton float is on the left of the
       transect barrel,  and the gill net rollers are
       to the right of the barrel.  (Photo by Dr.
       0. E. Kurt.)	    25

6 (a).  Phytoplankton incubation chamber.  Four sub-
       merged sample bottles are on the rotating
       wheel.   (Photo by author.)  	    31

  (b).  Periphyton ring-   Five plastic periphyton
       slides are visible clamped to the ring.
       (Photo by author.)	    31

7(a).  New emergent insect trap.  (Photo by Dr. 0. E.
       Kurt.)  ..........  	    38

  (b).  Styrofoam trap used in collection jar on the
       emergent insect trap.  (Photo by Dr. O. E.
       Kurt.)	    38

  (c).  Replacing collection jar on emergent insect
       trap.  Trap is suspended from bracket on the
       raft and not taken out of the water during
       transfer.   (Photo by Dr.  O. E.  Kurt,)	    38

-------
LIST OF FIGURES—continued

FIGURE                                                 Page

8(a).  Vertical gill net and roller.  Gill net is
       suspended between raft pontoons as during
       sample collection process.   (Photo by author=)    43

 (b) .  Robert Hoffman removing fish from vertical
       gill net.  (Photo by author.)	     43

   9.  Acclimation cage used to hold rainbow trout at
       specific depths in Hemlock Lake.  Rubber hoses
       led to water pumps on a raft.  Cage is covered
       with polyethylene plastic and chicken wire.
       (Photo by author.)	     46

  10„  Configuration of Hemlock Lake acclimation
       cages.  Fin clip of rainbow trout held in each
       cage is shown.  The oxygen and temperature
       profiles during the acclimation period are
       also shown	     48

  11.  Contour map of Hemlock Lake showing sample
       transects and aerator.  Depth intervals are in
       meters	     53

  12.  Contour map of Section Four Lake showing
       sample transect and air line.  Air was re-
       leased from the dashed section of the air
       line.  Depth intervals are in meters	     55

  13.  Hemlock and Section Four relative irradiance
       measurements on August 11, 1969. .	     59

  14.  Section Four transect profile illustrating
       percent organic carbon and percent CaCOa-car-
       bon at different depths	     61

  15.  Hemlock transect profile illustrating percent
       organic carbon and percent CaCOs-carbon at dif-
       ferent depths	     66

  16.  Hemlock Lake carbon dioxide, alkalinity, pH and
       conductivity profiles on August 13, 1969.
       This is representative of pre-aeration condi-
       tion.  Chemocline of monimolimnion is evident
       below 12 meters	     68

  17.  Cross-sectional view of Hemlock Lake hypolim-
       nion aerator.  Dotted lines represent projected
       edges.  Tower is tilted toward the viewer, and
       parts are drawn approximately to scale ....    74

                            xi

-------
 LIST OF  FIGURES—continued

 FIGURE                                                 Page

   18. Hemlock hypolimnion aerator  in operating posi-
       tion.   Only the upper  three  meters  are visible.
       Air  supply  line enters the scene from the left.
       Above water styrofoam  flotation units are a
       safety feature to prevent tower from sinking if
       submerged units should fail.   Ladder on side of
       aerator permits access to top of tube.   (Photo
       by author.)  ...........  	    76

19(a). Arrival of  Hemlock hypolimnion aerator tubes
       from the  factory.   Top three meters section of
       aerator is  separated and located next to the
       truck cab.   The tower  was unloaded  on the
       wooden cradle and logs in the foreground and
       fittings  were attached before it was shoved in
       the lake.  (Photo by author.)	    78

  (b). Hemlock hypolimnion aerator  floating horizon-
       tally- Temporary floats kept the lower end up
       while the current deflector  was attached.   The
       lower end is closest to the  viewer.   Deflectors
       could not be added until the aerator was float-
       ing horizontally in the lake.   (Photo by
       author.)	    78

  (c). Hemlock hypolimnion aerator  tilting into sampl-
       ing position.   The temporary floats have just
       been removed by the author using SCUBA.  (Photo
       by Robert Hoffman.)  r  .  .  .	    78

   20. Cross-sectional view and parts of hypolimnion
       aerator.  A.  Cross-section of aerator taken
       near the  top.   Two styrofoam flotation units
       and one barrel flotation unit are shown.
       B.  Styrofoam flotation unit.   C.  Barrel flota-
       tion unit showing the  tee structure used to
       attach it to the aerator.  D.  Cross-section of
       tee  inside  the slot structure.   The slot is
       welded to the outside  of the aerator	    81

   21.  Hemlock hypolimnion aerator  current deflector
       before they were attached to the aerator.
       Anchors in  foreground  were used to  anchor the
       tower  in  its operating position.   (Photo by Ed
       Schultz.)	    83
                             xn

-------
LIST OF FIGURES—continued

FIGURE                                                  Page

22(a).  Styrofoam flotation unit used on the Hemlock
        Lake hypolimnion aerator.  The tee structure
        used to attach the unit to the aerator is
        shown on top.  (Photo by author.)	    86

  (b).  Slots for flotation unit tee's being welded on
        the side of the aeration tower.  The iron
        plates used to position the inner tube are
        shown projecting through the outer tube to the
        left of the workmen.  (Photo by author.)  ...    86

  (c).  Sliding styrofoam flotation unit into slot on
        side of aeration tower.   Logs and "runway" are
        shown leading into the lake.   (Photo by Ed
        Schultz.)	    86

   23.  Air diffusor used on the hypolimnion aerator.
        The hole site spacing along one arm are
        shown.  Three holes were drilled at each site
        as shown in the cross-sectional view of one
        arm		    89

   24.  Water flow rates through the hypolimnion aera-
        tion tower as a function of air input and
        tower level.  The (0)  level is when the top of
        the inner tube is level with the lake1s sur-
        face.  (+)  level is with the inner tube's top
        0.5 m above the lake's surface, and the (-)
        level is with it 0.5 m below the lake's sur-
        face.  See text for discussion of true flow
        rates	    94

   25.  Hemlock Lake's isotherms during the summer
        1969, before aeration began.  Isotherms are
        in °C. .  .	    99

   26,  Hemlock Lake's top,  bottom and average oxygen
        concentrations during the summers 1969 and
        1970.  Continuous aeration occurred between
        June 14 and September 7,  1970.	   101

   27.  Hemlock Lake's total chlorophyll A,  phaophytin
        A,  oxygen and temperature profiles during
        August 13,  1969.   These are representative of
        values before aeration began 	   103
                             Xlll

-------
LIST OF FIGURES—continued

FIGURE                                                 Page

  28.  Hemlock Lake selected oxygen profiles during
       the summers 1969 and 1970.  Continuous aera-
       tion occurred between June 14 and September
       7, 1970	   105

  29.  Hemlock Lake's isotherms during the summer
       1970.  Continuous aeration occurred between
       June 14th and September 7th.,  Isotherms are
       in
                                                        108
  30.  Hemlock Lake selected temperature profiles
       during the summers 1969 and 1970,,  Continuous
       aeration occurred beeween June 14 and
       September 7,  1970.		   110

  31.  Hemlock Lake hypolimnetic oxygen isopleths
       (mg/1) along the air line transect one day be-
       fore aeration began and after one day of
       hypolimnion aeration ..... 	   112

  32.  Hemlock Lake hypolimnetic oxygen isopleths
       (mg/1) along the air line transect one day be-
       fore aeration began,  and after nine days of
       continuous hypolimnion aeration	   114

  33.  Hemlock Lake's maximum, minimum and average
       temperatures ( C)  during the summers 1969 and
       1970.  Continuous aeration occurred between
       June 14 and September 7,  1970. ........   117

  34.  Hemlock Lake's stability values during the
       summers 1969 and 1970.   Continuous aeration
       occurred between June 14 and September 7,
       1970 .........  	 .......   119

  35.  Hemlock Lake's bottom,  top and average pH
       values during the summers 1969 and 1970.
       Continuous aeration occurred between June 14
       and September 7,  1970.	   122

  36.  Hemlock Lake's bottom,  top and average alka-
       linity values during the summers 1969 and 1970.
       Continuous aeration occurred between June 14
       and September 7,  1970.  ..... 	   124

  37.  Hemlock Lake's bottom,  top and average conduc-
       tivity values during the summers 1969 and
       1970-  Continuous aeration occurred between
       June 14 and September 7,  1970. „	   126

                            xiv

-------
LIST OF FIGURES—continued

FIGURE                                                 Page

  38.  Hemlock Lake secchi disc transparencies,  sur-
       face primary production potentials,  surface
       phytoplankton densities and surface production
       efficiencies during the summers 1969 and 1970.
       Continuous aeration occurred between June 14
       and September 7,  1970. .	   131

  39.  Foam spilling over the top of the hypolimnion
       aerator during August 1970.  (Photo by author.)   135

  40.  Hemlock Lake periphyton standing crops based
       on 17-day incubation periods and continuous
       incubation.  The 95% confidence interval is
       shown about each average value.  Continuous
       aeration occurred between June 14 and Septem-
       ber 7,  1970.	   137

  41.  Hemlock Lake zooplankton depth distributions
       three days before aeration began,  and after
       one month of aeration.  Oxygen and tempera-
       ture profiles are shown for each date.
       (x" = average depth.)	   141

  42.  Hemlock Lake zoobenthos percent composition
       during the summers 1969 and 1970.   Percent of
       wet weight and percent of number are shown for
       each taxa.  Total weights and total numbers
       collected each summer are also shown.  Samples
       from dredge collections only ..,.„....   146

  43.  Hemlock Lake Chironomid larvae depth distribu-
       tion as percent of number during each sampling
       period during the summers 1969 and 1970.
       Shaded histograms represent aerated periods. .   150

  44.  Hemlock Lake Chironomid larvae depth distribu-
       tion as percent of wet weight during each
       sampling period during the summers 1969 and
       1970.  Shaded histograms represent aerated
       periods		   152

  45.  Hemlock Lake Chironomid pupae depth distribu-
       tion as percent of number during each sampling
       period during the summers 1969 and 1970.
       Shaded histograms represent aerated periods. .   155
                            xv

-------
LIST OF FIGURES—continued

FIGURE                                                 Page

  46.  Hemlock Lake Chironomid pupae depth distribu-
       tion as percent of wet weight during each
       sampling period during the summers 1969 and
       1970.  Shaded histograms represent aerated
       periods.	   157

  47.  Total midge emergence from Hemlock Lake by
       depths during the summers 1969 and 1970.
       Aeration occurred continuously between June 14
       and September 7, 1970.  Total include Chaobor-
       inae and Chironomid midges from emergence
       traps only	   159

  48.  Total estimated weekly midge emergence from
       Hemlock Lake during the summers 1969 and 1970.
       Totals include Chaoborinae and Chironomid
       midges from emergence traps only.  Aeration
       occurred continuously between June 14 and
       September 7,  1970. ..............   161

  49.  Total estimated Chironomid larvae number and
       wet weight in Hemlock Lake during the summers
       1969 and 1970.  One standard error is shown
       about each estimate.   Aeration occurred con-
       tinuously between June 14 and September 7,
       1970.  Totals from dredge samples only ....   164

  50,  Total estimated Chironomid pupae number and
       wet weight in Hemlock Lake during the summers
       1969 and 1970.  One standard error is shown
       about, each estimate.   Aeration occurred con-
       tinuously between June 14 and September 7,
       1970.  Totals from dredge samples only ....   166

  51.  Depth emergence of selected insects from Hem-
       lock Lake during the summers 1969 and 1970.
       White areas during the sampling periods repre-
       sent no observed emergence.  A= Frocladius
       denticulatus,  B= Tanypus, C= Dicrotendipes,
       D= Mayflies (Ephemoreptera) .  Aeration occur.-
       red continuously between June 14 and September
       7, 1970.  Totals from dredge samples only.  .  .   169

  52.  Total estimated emergences from Hemlock Lake.
       Samples from emergence traps only.  Aeration
       occurred continuously between June 14 and
       September 7,  1970. ..............   171
                            xvi

-------
LIST OF FIGURES—continued

FIGURE                                                 Page

  53.  Total estimated Chaoborus larvae number and
       wet weight in Hemlock Lake during the summers
       1969 and 1970.  One standard error is shown
       about each estimate.  Aeration occurred con-
       tinuously between June 14 and September 7,
       1970.  Totals from dredge samples only ....   175

  54.  Total estimated Chaoborus Pupae number and
       wet weight in Hemlock Lake during the summers
       1969 and 1970.  One standard error is shown
       about each estimate.  Aeration occurred con-
       tinuously between June 14 and September 7,
       1970.  Totals from dredge samples only ....   177

  55.  Total estimated emergences of Hemlock Lake
       Chaoborus flavicans and C_. punctipennis during
       1969 and 1970.  Total estimated larvae and
       pupae are also shown.  All samples are from
       emergence traps.  Aeration occurred continu-
       ously between June 14 and September 7,  1970. .   179

  56.  Depth distribution of Chaoborus during the
       summers 1969 and 1970.  All samples were col-
       lected by emergence insect traps.  A=
       Chaoborus f lavicans emergent adults,  B= C_.
       punctipennis emergent adults,  C= Chaoborus
       larvae,  D= Chaoborus pupae.  White areas dur-
       ing emergence periods represent no observed
       specimens.  Aeration occurred continuously
       between June 14 and September 7, 1970.  ....   182

  57.  Hemlock Lake Chaoborus larvae depth distribu-
       tion as percent of number during each sampling
       period during the summers 1969 and 1970.
       Shaded histograms represent aerated periods. .   184

  58.  Hemlock Lake Chaoborus pupae depth distribu-
       tion as percent of number during each sampling
       period during the summers 1969 and 1970.
       Shaded histograms represent aerated periods. .   186

  59,  Total estimated Mayfly (Ephemeroptera)  number
       and wet weight in Hemlock Lake during the sum-
       mers 1969 and 1970.  One standard error is
       shown about each estimate.  Aeration occurred
       continuously between June 14 and September 7,
       1970.  Totals from dredge samples only ....   189
                            xvii

-------
LIST OF FIGURES—continued

FIGURE                                                 Page

  60.  Hemlock Lake Mayfly (Ephemeroptera)  depth dis-
       tribution as percent of number during each
       sampling period during the summers 1969 and
       1970.  Shaded histograms represent aerated
       periods. .......... 	   192

  61.  Hemlock crayfish depth distributions during
       the summers 1969 and 1970.  Total numbers dur-
       ing each sample period and their average
       depths are shown.  The shaded area represents
       the 1969 distributions.  Aeration occurred
       continuously between June 14 and September 7,
       1970. ................ 	   195

  62.  Hemlock Lake Rainbow trout depth distribution
       during 1969.  These fish were stocked during
       June 1969 and marked with a right-abdominal
       fin clip.  Each square represents one fish .  .   198

  63.  1970 depth distribution of Hemlock Lake rain-
       bow trout stocked during June 1969.   Each
       circle represents one fish-  These fish were
       marked with a right-pelvic fin clip.  Aeration
       occurred continuously between June 14th and
       September 7th. ................   201

  64.  1970 Hemlock Lake depth distribution of right-
       pectoral clipped rainbow trout stocked during
       June 1970.  These fish were held in the 3 m
       covered cage which received 12 m water for one
       week before their release.  Each circle repre-
       sents one fish.  Aeration occurred continuous-
       ly between June 14 and September 7,  1970 . .  .   203

  65.  1970 Hemlock Lake depth distribution of left-
       pectoral clipped rainbow trout stocked during
       June 1970.  These fish were held in the 12 m
       covered cage which received 3 m water for one
       week before their release.  Each circle repre-
       sents one fish.  Aeration occurred continuous-
       ly between June 14 and September 7,  1970 . .  .   205

  66.  1970 Hemlock Lake depth distribution of left
       pelvic clipped rainbow trout stocked during
       June 1970.  These fish were held in a screened
       cage at 12 m for one week before their re-
       lease.   Each circle represents one fish.
       Aeration occurred continuously between June 14
       and September 7,  1970. ............   207


                           xviii

-------
LIST OF FIGURES—-continued

FIGURE                                                 Page

  67.  1970 Hemlock Lake depth distribution of anal-
       clipped rainbow trout stocked during June
       1970.  These fish were held at 3 m in a
       screened cage for another week before their
       release.  Each circle represents one fish.
       Aeration occurred continuously between June
       14 and September 7,  1970	    209

  68.  Hypothetical changes in Hemlock Lake limiting
       nutrient, phytoplankton, zooplankton and
       Chaoborus densities during 1970.  Aeration
       began June 14th and continued through Septem-
       ber 7th.  Major food chain relationships are
       also shown	    230

  69.  Hemlock Lake oxygen and temperature condi-
       tions for trout during August 1969 and August
       1970.  Adequate temperature is temperature
       less than 24 C,  and adequate oxygen is values
       of 5 mg/1 or more	    266

  70.  Cross-sectional view of Section Four diffuse
       aeration system.  The air was released from
       the last 10 meters of pipe,  situated near the
       deepest point in the lake	    272

  71.  View of Section Four Lake taken from the
       basin rim.  Rising air and water is seen near
       the center of the lake  (Photo by Author.)  . .    274

  72.  Section Four isotherms during the summer,
       1969, before aeration began.  Isotherms are
       in  C 	  .............    278

  73.  Section Four selected oxygen profiles during
       the summers 1969 and 1970.  Aeration occurred
       between June 16 and September 7, 1970 ....    280

  74.  Section Four maximum,  minimum and average
       temperatures ( C)  during the summers 1969 and
       1970.  Aeration occurred between June 16 and
       September 7,  1970	    283

  75.  Section Four top,  bottom and average oxygen
       concentrations during the summers 1969 and
       1970.  Aeration occurred between June 16 and
       September 7,  1970	    285
                            xix

-------
LIST OF FIGURES--continued

FIGURE                                                 Page

  76.  Section Four selected temperature profiles
       during the summers 1969 and 1970.  Aeration
       occurred between June 16 and September 7,
       1970	    287
  77.  Section Four isotherms during the summer
       1970.  Aeration occurred between JuneQ16th
       and September 7th.  Isotherms are in  C .  .  .     289

  78.  Section Four stability values during the
       summers 1969 and 1970.  Aeration occurred
       between June 16 and September 7,  1970 ....     292

  79.  Section Four's bottom, top and average pH
       values during the summers 1969 and 1970.
       Aeration occurred between June 16 and Septem-
       ber 7, 1970 .	     294

  80.  Section Four bottom,  top and average alkalin-
       ity values during the summers 1969 and 1970.
       Aeration occurred between June 16 and Septem-
       ber 7, 1970	     296

  81.  Section Four bottom,  top and average conduc-
       tivity values during the summers 1969 and
       1970.  Aeration occurred between June 16 and
       September 7, 1970 ..............     299

  82.  Section Four secchi disc transparencies, sur-
       face primary production potentials,  surface
       phytoplankton densities and surface produc-
       tion efficiencies during the summers 1969 and
       1970.  Aeration occurred between June 16 and
       September 7, 1970	     303

  83.  Section Four periphyton standing crops based
       on 17-day incubation periods and continuous
       incubation.  The 95% confidence interval is
       shown about each average value.  Aeration
       occurred between June 16 and September 7,
       1970	     307

  84.  Section Four zoobenthos percent composition
       during the summers 1969 and 1970.  Percent of
       weight and percent of number are shown for
       each taxa.  Total weights and total numbers
       collected each summer are also shown.
       Samples from dredge collections only	     309

                            xx

-------
LIST OF FIGURES—continued

FIGURE                                                 Page

  85.  Section Four oligochaete (microdriles)  depth
       distribution as percent of numbers during the
       summers 1969 and 1970.  Shaded histograms
       represent aerated periods .... 	    313

  86.  Section Four oligochaete (microdriles)  depth
       distribution as percent of wet weight during
       the summers 1969 and 1970.   Shaded histograms
       represent aerated periods .  .  .  .......    315

  87.  Section Four total estimated oligochaete
       number and biomass during the summers 1969
       and 1970.  One standard error is shown about
       each estimate.  Aeration occurred between
       June 16 and September 7,  1970 ........    318

  88.  Section Four Chironomid larvae depth distri-
       bution as percent of number during each
       sampling period during the summers 1969 and
       1970.  Shaded histograms represent aerated
       periods . 	 ............    322

  89.  Section Four Chironomid larvae depth distri-
       bution as percent of wet weight during each
       sampling period during the summers 1969 and
       1970.  Shaded histograms represent aerated
       periods ......... 	 ....    324

  90.  Total estimated Chironomid larvae number and
       wet weight in Section Four during the summers
       1969 and 1970.  One standard error is shown
       about each estimate.  Aeration occurred be-
       tween June 16 and September 7,  1970.  Totals
       from dredge samples only.	    326

  91.  Section Four Chironomid pupae depth distribu-
       tion as percent of number during each sampl-
       ing period during the summers 1969 and 1970.
       Shaded histograms represent aerated periods .    329

  92.  Section Four Chironomid pupae depth distribu-
       tion as percent of wet weight during each
       sampling period during the summers 1969 and
       1970.  Shaded histograms represent aerated
       periods 	  . ......    331
                            xxi

-------
LIST OF FIGURES—continued

FIGURE                                                 Page

  93.   Total estimated Chironomid pupae number and
       wet weight in Section Four during the summers
       1969 and 1970-  One standard error is shown
       about each estimate.  Aeration occurred be-
       tween June 16 and September 7,  1970.  Totals
       from dredge samples only, ..........    333

  94.   Total midge emergence from Section Four by
       depths during the summers 1969 and 1970.
       Aeration occurred between June 16 and Septem-
       ber 7, 1970.  Totals include Chironomid
       midges from emergence traps only. ......    336

  95.   Total estimated weekly midge emergence from
       Section Four during the summers 1969 and 1970.
       Totals include Chironomid midges from emer-
       gence traps only-  Aeration occurred continu-
       ously between June 16 and September 7,  1970      338

  96.   Total estimated midge emergences from Section
       Four.  Samples from emergence traps only.
       Aeration occurred between June 16 and Septem-
       ber 1, 1970 .................    340

  97.   Depth emergence of selected Section Four
       midges during the summers 1969 and 1970.
       White areas represent no observed emergence.
       A= Procladius, B= Ablabesmyia mallochi,  C=
       Clinotanypus thoracicus.  Aeration occurred
       between June 16 and September 7,  1970 ....    342

  98.   Depth emergence of Section Four midges and
       mayflies during the summers 1969 and 1970.
       White areas represent no observed emergence.
       A= Lauterborniella spp., B= Chironomini, C=
       Mayflies (Ephemeroptera).  Aeration occurred
       between June 16 and September 7,  1970 ....    344

  99.   Section Four Mayfly (Ephemeroptera) depth
       distribution as percent of number during each
       sampling period during the summers 1969 and
       1970.  Shaded histograms represent aerated
       periods	    347

 100.   Total estimated Mayfly  (Ephemeroptera)  number
       and wet weight in Section Four during the
       summers 1969 and 1970.  One standard error is
       shown about each estimate.  Aeration occurred
       between June 16 and September 7,  1970.
       Totals from dredge samples only .......    349

                            xxii

-------
     OF FIGURES—continued

FIGURE                                                 Page

 101.  Total estimated Amphipod number and wet weight
       in Section Four during the summers 1969 and
       1970.  One standard error is shown about each
       estimate.  Aeration occurred between June 16
       and September 7,  1970.  Totals from dredge
       samples only .  .  .	   352

 102.  Section Four Amphipod depth distribution as
       percent of number during each sampling period
       during the summers 1969 and 1970.  Shaded his-
       tograms represent aerated periods. .  	   354

 103.  Section Four Trichoptera depth distribution as
       percent of number during each sampling period.
       during the summers 1969 and 1970.  Shaded
       histograms represent aerated periods .....   356

 104.  Total estimated Trichoptera number and wet
       weight in Section Four during the summers
       1969 and 1970.   One standard error is shown
       about each estimate.   Aeration occurred be-
       tween June 16 and September 7,  1970.   Totals
       from dredge samples only ...........   358

 105.  Section Four Heleidae (=Ceratopogonidae)  depth
       distribution as percent of number during each
       sampling period during the summers 1969 and
       1970.  Shaded histograms represent aerated
       periods. ............. 	   360

 106.  Total estimated Heleidae (=Ceratopogonidae)
       number and wet weight in Section Four during
       the summers 1969 and 1970 .  One standard error
       is shown about each estimate.  Aeration occur-
       red between June 16 and September 7,  1970.
       Totals from dredge samples only. 	   362

 107.  Section Four crayfish depth distributions dur-
       ing the summers 1969 and 1970.   Total numbers
       during each sample period and their average
       depths are shown.  The shaded area represents
       the 1969 distributions.   Aeration occurred be-
       tween June 16 and September 7,  1970	   365

 108.  Section Four rainbow trout depth distributions
       during 1969.   Open Squares represent fish
       stocked during June 1969 and marked with a
       right-pectoral  fin clip.  Solid squares are
       fish stocked during 1964-65 and lack fin clips.
       Each square represents one fish.	   368

                           xxiii

-------
LIST OF FIGURES —continued

FIGURE                                                 Page

 109.  Section Four rainbow trout depth distribution
       during 1970.  Open circles represent fish
       stocked during May 1970 and marked with a
       left-pelvic fin clip.  Solid circles represent
       fish stocked during June 1969 and marked with
       a right-pectoral fin clip.  Solid squares
       represent fish stocked during 1964-65 and lack
       fin clips.  Each symbol represents one fish. .    371

 110.  Hypothetical residence times for a passive,
       neutral buoyancy object with the photic and
       aphotic zones of a stratified and unstratified
       lake ................ 	    378

 111.  Ratios of Section Four surface/bottom phyto-
       plankton concentrations during 1969 and 1970.
       The lake was destratified during 1970	    381

 112.  Section Four oxygen profiles during January
       1970 and 1971.  The 1970 profile is after a
       summer of normal stratification,  while the
       1971 profile is after a summer of artificial
       aeration ...................    405

 113.  Hemlock Lake oxygen profiles during December
       1969 and 1970, and January 1970 and 1971.  The
       December 1969-January 1970 profiles are after
       a summer of normal stratification, while the
       December 1970-January 1971 profiles are after
       a summer of artificial aeration.  .......    408

 114.  Artificial aeration of Hemlock Lake du£ing
       January 1971.  The compressor was towed onto
       the lake and run for two days.  A rubber air
       line leads to the aeration tower .......    411

 115.  Effects of artificial aeration on the oxygen
       regime during January 1971.  The January 22nd
       figure shows the oxygen profiles before winter
       aeration began,  but after a summer of artifi-
       cial aeration.  The January 23rd profile is
       after 24 hours of air injection and the
       January 24th profile is after 48 hours of air
       injection.  ...........  	    413
                           xxiv

-------
LIST OF FIGURES—continued

FIGURE                                                 Pag-e

 A-l.  Length histograms of hatchery reared rainbow
       trout at time of stocking in Hemlock Lake dur-
       ing June 6,  1969 and June 25, 1970.  Only one
       lot of fish were stocked during 1969, whereas
       four lots were stocked during 1970.  Each lot
       received a separate fin clip.  Total numbers
       (n),  average fish lengths (x) and fin clips
       for each lot are shown	  465

 A-2.  Length histograms of hatchery reared rainbow
       trout at time of stocking in Section Four
       Lake during June 6, 1969 and May 23,  1970.
       One lot was stocked each year.  Total numbers
       (n),  average fish-lengths (x) and fin clips
       for each lot are shown.	467
                            xxv

-------
                       INTRODUCTION








     Essentially all temperate lakes follow thermal strati-



fication cycles.  Yearly cycles are the most important and



have the greatest influence on lake processes.  Chemical



and biological stratification are related to thermal



stratification.



     Typical temperate lakes are isothermal in the spring.



As the seasons progress, the upper lake water is heated



faster than the lower.  Wind generated water currents



distribute this heat to the lower depths.  Wind mixing



efficiency decreases greatly with increasing depth and the



lake is divided into three thermal zones by early summer:



(i) Epilimnion.  The warm water zone.  It is circulated by



the wind and oxygen concentrations are generally near satura-



tion.  Temperature and chemical properties are nearly



homogeneous throughout.  Daily and seasonal temperature



variation is much greater here than in the other zones.



Most of the biota is restricted to this zone in eutrophic



lakes.  (ii)  Metalimnion.  A zone of rapid change in tempera-



ture,  density and chemical properties.  Characterized by a



1.0 C or greater change per meter change in depth.



(iii)  Hypolimnion.   The cold water zone of the lake.  This



zone is sealed off from the surface.  It characteristically

-------
stagnates by early summer in organically rich, eutrophic


lakes.  Stagnation results in an oxygen deficit, buildup of


hydrogen sulfide, increased iron, phosphate and manganese


concentrations, anaerobic conditions and overall deteriora-


tion of water quality-  Aerobic biota are often excluded


from the hypolimnion of eutrophic lakes.  The rate of hypo-


limnetic oxygen depletion is a measure of the lake's organic


richness.  Organically poor,  oligotrophic lakes characteris-


tically do not develop oxygen depletions.  Oligotrophic


lakes differ markedly in their chemical and biological prop-


erties although their temperature regimes are similar to


eutrophic lakes.  Chemical concentrations are often uniform


throughout oligotrophic lakes and restrictions to biotic


distributions are minimal.  Anaerobic biota are either


conspicuously absent, or restricted to subsurface mud.


Hypolimnetic biotic concentrations can be greater than


epilimnetic concentrations.


     As the seasons progress,  the epilimnion cools and


approaches the hypolimnion temperature during late summer


and fall.  The lake becomes isothermal again during the fall.


Most temperate lakes develop ice cover and inverse thermal


stratification during the winter.  The coldest water, at


0.0 C,  is just under the ice,  whereas warmer water at about

. o
4 C is near the bottom of the lake.  Chemical and biological


values often stratify in response to this winter thermal


stratification.  Oxygen may be completely depleted from

-------
eutrophic lakes during the winter with a resultant loss of



aerobic biota.  Oligothrophic lakes generally have high



oxygen concentrations all winter at all depths.



     Lake stratification often conflicts with man's exploi-



tation of the water body.  Thermal stratification of



eutrophic lakes often results in deterioration of drinking



water quality, anaerobic and corrosive conditions, increased



evaporation rates, reduced heat budgets and other undesir-



able properties within the lake.  Stratification of oligo-



trophic lakes is generally not undesirable.



     In general, lakes are thought to undergo a succession



from oligotrophy through mesotrophy and eutrophy to dystrophy



(Lindeman, 1942).  This natural process is called eutrophica-



tion.  Accelerated and undesirable eutrophication due to



man1s activities is termed organic pollution.  Although



lakes may enter this scheme at any point based on their



origin, morphometry,  and other factors, they generally prog-



ress towards dystrophy and extinction.  Productivity per unit



area increases at least through the eutrophic stage and may



decline at dystrophy due to volume reduction and unfavorable



conditions,  such as low pH.  Increased productivity is



associated with higher nutrient levels and oxygen depletion



of deep water during periods of stagnation.  Indeed, this



loss of oxygen is seen as a pivot point between oligotrophy



and eutrophy.  Aging is greatly accelerated once eutrophy



is attained.   Oxygen depletion is often credited with greatly

-------
accelerating eutrophication since anaerobic-sis results in



accelerated generation of nutrients from profundal sediments.



These sediments are rich in essential nutrients, especially



phosphorus.  Although many factors may limit production in a



given system at a given time,  phosphorus is generally the



one most limiting factor in a cosmopolitan spatial and



temperal sense; especially of primary production.  Considerable



effort has been expended trying to define its behavior in



lakes, but its movements and forms are still very poorly



understood.  Although our understanding is imperfect, certain



characteristics are apparent.   Much of the following phosphorus



discussion is based on a conversation with Dr. F. H. Rigler



of the University of Toronto.   The fact that lakes act as



nutrient traps is of special importance when considering the



phosphorus cycle.  Only 20 to 70% of the phosphorus entering



a lake will leave the lake (Figure 1; B = (0.2 to 0.7) x A) .



The remainder, 30 to 80%, remains in the lake and most of



this ultimately resides permanently in the sediments.



     The behavior of phosphorus before its eventual sedimen-



tary incorporation determines  the level of eutrophication.



In general, the greater the amount of phosphorus in circula-



tion,  the more eutrophic the lake will be.  Our conceptual-



ization of phosphorus cycling  in a lake is facilitated by



envisioning compartments that  roughly correspond to the major



lake zones.  Phosphorus behavior within each compartment is



somewhat independent of its behavior within the other

-------
compartments, but the compartments are very much interrelated.



The three general compartments are limnetic, littoral and



profundal.  Phosphorus is exchanged between these compart-



ments, as well as within components of each compartment.



Relationships within the limnetic and littoral compartments



are probably the most complex, and those within the former



are best understood.  The components, or forms limnetic phos-



phorus assumes can be categorized as sestonic and soluble



 (Figure 1)«  Each category is again divisible into subcate-



gories.  These categories appear discrete, but in fact there



is much overlap.  In a static sense, about 70% of the total



phosphorus is found within the sestonic forms.  In a dynamic



sense this picture may be reversed since the movements be-



tween components is very rapid.  For example, although solu-



ble inorganic phosphorus probably represents less than 6% of



the total limnetic phosphorus at any given time, its turnover



rate is from 1 to 7 minutes.  The small algae and bacteria



are most important in this movement.  The main pathway



within the limnetic compartment is from soluble phosphorus



to the bacteria and in algae  (Li).  Phosphorus then moves



back to the soluble form (L2) and into the zooplankton  (LS)



in about equal proportions (L2 2rLs)°  Phosphorus is also



directly taken up by larger limnetic algae and zooplankton,



but at a lesser rate.  Zooplankton excrete and secrete phos-




phorus,  with secretion exceeding excretion  (Johannes and



Satomi,  1967; Rigler, 1964).  Most of the zooplankton-voided

-------
Figure 1.  Hypothetical phosphorus cycle within a lake showing the'three
           main compartments.   The components of the littoral compartment
           are also shown.

-------
Limnetic Phosphorus
I Seston
  (a) >70ju.  zooplankton
  & large phytoplanton (25%)
  (b).45ju.-70xL bacteria
   & aigae (35%)
  Soluble
  (a) Inorganic orthophosphate (6%)
  (b) .1—.45ju.  non-living particles (9%)
  (c) Other organic phosphorus (25%)
fl

-------
                            8
phosphorus is orthophosphate.  The dynamics of these exchanges



and those between lake compartments is such that the concen-



tration of the soluble component does not fluctuate widely.



The rates of movement within compartments and between com-



partments is probably the key factor controlling eutrophica-



tion.  The vast majority of studies have measured only static



phosphorus concentrations within certain compartments.  These



measurements have contributed something to our understanding



of the role phosphorus plays, but all too often they have led



to confusion and conflicting conclusions.  Rates of movement



in a total lake system have not been thoroughly analyzed.



     We know that ultimately most of the phosphorus ends up



in the profundal compartment, and there is a net movement



out of the limnetic compartment.  Hutchinson and Bowen (1947)



indicate the net movement is from the littoral to the lim-



netic to the profundal.  Rigler (1956) and Coffin et al.



 (1949) later found a net movement during the summer from the



limnetic to the littoral with less than 5% from the limnetic



to the profundal.  Presumably there was a net movement from



the littoral to the profundal especially during the fall and



winter breakup of littoral vegetation.  McCarter et al.



(1952)  have shown that very little phosphorus moves from the



profundal to the other compartments during thermal strati-



fication,  but no studies to date have evaluated movements



between compartments on a yearly basis.  They are all incom-



plete and taken singularly often lead to misinterpretations.

-------
     The movement of phosphorus between the profundal compart-



ment and the trophogenic zone is especially important in



understanding eutrophication in general, and the specific



applicability of artificial aeration as a corrective measure



for eutrophication.  As discussed, the onset of profundal



oxygen depletion is considered a most important event in the



eutrophication process.  Anaerobiosis may result in a sig-



nificant increase in nutrient movement from the profundal



compartment back into the limnetic and littoral.  This move-



ment occurs mostly during spring and fall overturns.  Before



anaerobiosis, phosphorus was tightly held by the aerated



sediments and little was returned during overturns.  After



anaerobiosis, a large, but undetermined quantity is returned



to the trophogenic zone.  What fraction this represents of



the total input to the profundal compartment is unknown.



If this fraction is relatively small, then the onset of



anaerobiosis may represent only a signpost on the road to



dystrophy and our efforts to retard eutrophication by aera-



tion will be ineffective.  In this case, the total input of



phosphorus to the lake is seen as the dominant factor.



Studies on artificial fertilization infer this may be the



case.   Artificially fertilized lakes typically return to



their former level of production soon after nutrient input



ceases,  even though prodigious quantities of nutrients were



added and a high degree of anaerobiosis was attained.



     On the other hand,  if anaerobiosis results in the return



of a significant fraction of nutrients to the trophogenic

-------
                            10
zone, then artificial aeration may be a useful method of



reversing eutrophication.  This would be achieved by keeping



the profundal sediments well oxidized and thus preventing



the release of nutrients once they were incorporated in



these sediments.  This also assumes that the nutrient input



into the lake is not accelerated.  If it is, this may more



than compensate for reduced regeneration from the profundal



compartment and many of the characteristics of eutrophication



would not diminish.  In any case, the anaerobiosis would be



diminished or eliminated by artificial aeration.  This is



not  always the most important characteristic, however.  This



in essence forms the basis of my artificial hypolimnion



aeration of Hemlock Lake.  Rather than attempt to measure



changes in phosphorus transfer rates from the profundal zone



associated with artificial aeration, I chose to measure cer-



tain biotic parameters and thereby infer changes in the phos-



phorus transfer rates.  This approach has two important



advantages:   (1) it is probably impossible to accurately



measure phosphorus transfer rates with our present technology,



while biota parameters are estimable in many cases, and



(2) the effects of aeration on the biota are ultimately the



most important factors.   Although we may not materially



affect phosphorus regeneration from the profundal sediments,



some other event related to the aeration may result in re-



duced biotic productivity.  Thus the net effect might be



desirable,  although the effect on regeneration might be in-



consequential.  This is in fact how many important scientific

-------
                            11
 advances occur.  We  first find out what will work, and later



 possibly how  it does work.  The major disadvantage of moni-



 toring only the biota is that results not conforming to the



 theory are very difficult to explain.  To be sure, direct



 measurement of phosphorus movements is desirable, but impos-



 sible from a  practical standpoint.



     Until recently, little was done to economically reduce



 stratification.  Artificial destratification was attempted



 with mechanical pumps  (Hooper et al., 1952; Irwin et al.,



 1969), but this method is generally slow and relatively inef-



 ficient.  Several artificial destratification techniques



 using compressed air have been developed.  Fast  (1968) re-



 viewed several of these techniques in greater detail.  Their



 common principle is that compressed air is released near the



 bottom of the thermally stratified lake.  The rising air



 generates vertical water currents that diverge horizontally



 upon reaching the lake's surface  (Figure 2).  This upwelled



 water is much colder and denser than the surface water.



 Upon converging with the warm surface water and sinking, the



 cold water mixes with the epilimnion and metalimnion water



 along its periphery to form water of intermediate tempera-



 ture and density.




     This mixed water now spreads out horizontally at levels



 of equal density.  The depth of outflow depends on the degree



 of mixing and initially may be confined mostly to the



metalimnion.   As the mixing process progresses, the shape of

-------
Figure 2.  Hypothetical destratification patterns caused by diffuse
           aeration system.  Figure A illustrates a stratified lake,  and            £
           Figure B a lake being destratified.

-------
D, < D2 < D3 < D4 < D5


      D= den sity
^-s*+**^*~*~*-s*~***

-m
D,     D7
 -}! t
                D4 •" I 1 ,
                Ail H
^a/
                          3/
     B.
  D7

-------
                            14
the iso-density strata changes.  Concomitant with this change



is a change in the rate of mixing.  The rate of destratifica-



tion is greatest when air injection begins and approaches



zero apparently as an exponential function as the lake ap-



proaches isothermy (Koberg and Ford, 1965).  The lake will



approach either an isothermal or a steady-state condition as



the mixing continues.  The time it takes to reach this condi-



tion depends on the time of year, size of lake, and method



of injection.  The best method of injection and equipment



specifications for a given lake situation is not well defined.



     Artificial destratification of a lake by compressed air



is commonly called "lake aeration."  The reasons for this



are two-fold:   (1) Eutrophic lakes experience an oxygen



deficit below the metalimnion during the summer.  By artifi-



cially circulating the lake, the oxygen deficit is reduced



or eliminated, and (2) compressed air is used to circulate



the water.  While the compressed air adds oxygen directly



to the upwelled water, oxygen is also gained from contact of



the water with the atmosphere and by photosynthesis of aquatic



plants.  The term "aeration" applied to oligotrophic lakes



is somewhat of a misnomer since these lakes typically have



adequate oxygen levels.  Destratification and aeration are



generally used synonymously.




     The purpose of artificial destratification is to reduce



the density barrier to complete circulation.  After artifi-



cially induced circulation of a reservoir, the water

-------
                             15
 temperature  is  about  equal  from top to bottom and many of



 the  chemical properties are likewise homogeneous.  The whole



 water mass and  bottom area  is theoretically habitable by



 the  aerobic  biota which otherwise could only inhabit the



 epilimnion and  metalimnion  of a eutrophic lake.



     Artificial destratification also increases the lake's



 heat budget.  After destratification, the entire lake is



 about as warm as the  epilimnion before aeration began.  This



 warming occurs  with continuous air injection  (Fast, 1968),



 as well as intermittent air injection  (Fast and St. Amant,



 manuscript in preparation).  Although an increased heat



 budget generally benefits eutrophic lakes, it may have seri-



 ous  repercussions in  oligotrophic lakes by eliminating those



 cold water organisms  inhabiting the hypolimnion.



     A new method of  aerating the hypolimnion of eutrophic



 lakes was more  recently developed (Bernhardt, 1967).  Using



 Bernhardt's  aerator (Figure 3) the hypolimnion is aerated



 but  not greatly heated.  Hypolimnion aeration is only applic-



 able to eutrophic lakes, since oligotrophic lakes already



 have high oxygen concentrations.  Bernhardt's aerator con-



 sists of a large diameter pipe extending from the lake bottom



 to above the lake surface.   Inlet ports are located near the



 bottom of the pipe and outlet ports are located below the



metalimnion.   The top of the pipe is open to the atmosphere.



Air  is released and passes  through a diffusor near the bottom



of the pipe.   As air rises  in the pipe, water is drawn in

-------
                       16
Figure 3.  Hypolimnion aerator of Bernhardt  (1967)
           From Fast (1968).

-------

-------
                            18
 through the bottom ports and rises.  Oxygen diffuses into



 the water as it rises.  Water and air are carried to the top



 of the pipe where the air escapes to the atmosphere.  The



 water, however, cannot escape at the surface and sinks to



 the outlet port level where it flows back into the hypolimnion.



 Once  a hydraulic head is established in the pipe, water may



 flow  directly  from the inlet to outlet ports without rising



 to the top of  the pipe.  Consequently, hypolimnion water is



 aerated, but not significantly heated or mixed with epilim-



 nion  or metalimnion water.  Thermal stratification is not



 affected by this technique if the outlets are below the



 metalimnion.



      Bernhardt used his hypolimnion aerator in Wahnbach



 reservoir near Siegberg, Federal Republic of Germany.  This



 domestic water supply reservoir is about 37xl66 cubic meters



 volume with 43.4 meters maximum depth.  His main objective



 was to supply  cold, well-oxygenated water for domestic and



 industrial uses.  Previously, he used a diffused air injec-



 tion  system to aerate the water, but this technique increased



 the water temperatures to undesirable levels.



     Bernhardt estimates 0.167x106 cubic meters per day are



 aerated by the hypolimnion aerator using about 117 cfm of



 compressed air.  Contact time within the aerator is about 54



 seconds and increases the oxygen concentrations from about



5 mg/1 to about 10-11 mg/1 at steady state.



     Oxygen  transfer occurs primarily within the bottom 7.5



to 20.0 meters  of pipe and decreases rapidly thereafter.

-------
                            19
 This  is  primarily  a  function of:  hydrostatic pressure,



 oxygen saturation  of the water, oxygen content of the air



 bubbles,  and bubble  size.  As water and air rise higher in



 the pipe,  conditions for oxygen transfer become progressively



 less  favorable.  It  may be possible to increase the oxygen



 concentration of the hypolimnion to greater levels than



 found in the epilimnion or metalimnion.  This is possible



 because  of the greater hydrostatic pressures and lower



 temperatures of the  hypolimnion water.



      My  present investigation actually consisted of two



 separate,  but related aeration studies.  One oligotrophic



 lake, Section Four Lake, was thermally destratified.  This



 study was conducted  principally to determine the effects of -



 an increased heat  budget on the coldwater biota and the other



 characteristics of the lake.  This is one of the first studies



 to evaluate the influence of continuous summertime destratifi-



 cation on a oligotrophic lake.  Most destratification studies



 were  conducted on  eutrophic lakes where the benefits from



 and needs for artificial destratification are greatest.



      Hemlock, a eutrophic lake, was aerated with a new hypolim-



 nion  aerator of my own design.  This study was conducted



 principally to determine the value of this aeration system



 for eutrophication control and to evaluate its effect on the



 vertical  distribution of rainbow trout (Salmo gairdneri).



While Bernhardt (1967) was mainly interested in improving



water quality by artificial hypolimnion aeration, I am more

-------
                            20
interested in the specific effects of this aeration system



on the biota.



     The experimental design is essentially the same for both



lakes.  The lakes were studied under natural conditions



during 1969 and artificially aerated during 1970.   Special



emphasis was placed on the oxygen,  temperature,  zoobenthos



and rainbow trout depth distributions.

-------
                         METHODS







     The transect method was used to collect most samples.



A rope extended from a post on shore to a steel barrel



anchored near the center of the lake (Figures 4 and 5).



A pontoon raft was hand-pulled along the rope and samples



collected as desired.  Unless otherwise indicated, samples



were collected at the barrel.  The summer sample periods



extended from June 15, 1969 through September 5, 1969 and



from June 7, 1970 through September 7,  1970.  In addition,



one set each of oxygen, temperature, pH, alkalinity and



conductivity measurements were made during December 1969,



January 1970, December 1970 and January 1971.







Physical-Chemical





     Water samples for chemical determination were collected



each two meter depth interval from each lake with a PVC



plastic water sampler. Samples for pH,  alkalinity and con-



ductivity were placed in plastic bottles at the lake and



taken to the laboratory for analysis.  Total alkalinity and



pH were measured with a glass electrode pH meter,  A 4.4 pH



endpoint was used for total alkalinity.  Specific conductance



was measured on a Type RC,  Industrial Instruments conductivity
                            21

-------
Figure 4.  X-sectional view along principle sample transect.  Raft, emergent

           insect traps, gill nets and transect float are shown.
                                                                                     to
                                                                                     to

-------
Transect Post
                                                                                                                           Transect  Float

-------
                            24
Figure 5.   View of Section Four  Lake  taken  from basin rim.
           Emergent insect traps are  stacked  on the raft.
           The periphyton float  is  to the left of the
           transect barrel,  and  the gill net  rollers are
           to the right  of the barrel.   (Photo by Dr.
           0. E.  Kurt.)

-------
,,

-------
                            26
bridge during 1969, and on a different Type RQ bridge during
1970.  A correction value was determined for each instrument.
     Temperature-depth measurements were made with a re-
sistance thermometer at weekly intervals or less.  Thermal
stability was calculated from temperature data as described
by Fast  (1968) using Schmidt's (1915) and McEwen1s (1941)
formulations.
     Most oxygen determinations were made from water samples
collected with the PVC water sampler.  A few sets of deter-
minations during June 1970 were taken with a Precision
galvanic cell D.O. analyser.  Oxygen values were determined
by the Alsterberg modification of the Winkler method, except
that Phenylarsene oxide  (PAO) was substituted for thiosulfate
and  thyodene was substituted for starch solution.  Samples
were taken weekly during the summer 1969, most of the summer
1970, and each winter from one location, but were taken more
often from both lakes during June 1970.  In addition to the
standard sampling location in Hemlock Lake, depth samples
were taken at several locations along a transect from the
aeration tower to a point near the compressor.  Most of the
Hemlock Summer 1970 and January 1971 oxygen and temperature
measurements were made along this transect.
     Transparency was generally measured daily with a standard
black and white 20-cm secchi disc.  Only one set of light
transmittance measurements were made in each lake with a
subsurface photometer.

-------
                            27
     Carbon dioxide concentrations were estimated using a



monograph from pH and total alkalinity values  (Moore, 1939).



     Total dissolved organic matter, total particulate



organic matter, Ca, K, Na, and Mg were analyzed by R. G.



Wetzel.  He used slight modifications of Stickland and



Parsons'  (1965) methods for DOM and POM; and simple atomic



absorption for the cations (Jarrell-Ash Model 82-700).



     Relative irradiance measurements were made with  a sub-



marine photometer.  Both deck and submerged cells were used



without filters.



     Chlorophyll analyses were made only once.  Water was



filtered through Whatman GFIC glass filters.  A small quantity



of magnesium carbonate was placed on the filter to prevent



acidification and the filters were stored at about -23 C.



Samples were macerated with a tissue grinder and chlorophyll



extracted for 24 hours in 90% acetone.  Optical densities



were measured over a 1.0 cm pathlength.  Chlorophyll  values



were computed using formula of Parsons and Strickland (1963)



and Lorenzen (1967).



     Area-capacity values are based on topographic maps



constructed by the Institute for Fisheries Research,  Michigan



Department of Natural Resources.  They constructed these



maps from surveys made of each lake during January 1957.



I constructed area-capacity tables  (Tables A-9 and A-10) from



these maps as described by Welch (1948).

-------
                             28
      Phosphorus measurements were made  in both lakes  during



 1969.   Samples were collected  in glass  bottles and a  portion



 of each was filtered through HA millipore filters.  The



 samples were then  acid-digested and  analyzed  using the



 stannous chloride  method  for orthophosphate  (American Public



 Health Association,  1965).  Total phosphorus  and total solu-



 ble phosphorus were calculated from  these measurements.



      Sediment total carbon  content was  measured using a new



 method developed by Dr. Frank  D1Itri of Michigan State



 University.  Briefly,  a finely ground sample  of dry sediment



 (dried at 105°C) is oxidized by heating with  potassium di-



 cromate and concentrated  sulfuric acid.  Carbon dioxide is



 produced in proportion to the  amount of carbon present.



 The carbon dioxide is absorbed by sodium hydroxide and de-



 termined gravimetrically.   Sediment  carbonate content was



 measured using an  acid neutralization method  (Allison and



 Moodie,  1965).  Organic carbon is estimated by the difference



 between total carbon and  carbonate carbon.








 Phytoplankton






     Brian  Moss identified  and counted  the phytoplankton.



 Three 500-ml  samples were collected  twice a week from each



 lake during 1969.   These  samples were collected from  0,  5  and




 15 meters depths.   One ml of Lugol's  solution was added and



 the samples were shipped to East Lansing for  examination.



During 1970,  11 samples were collected  each time from Hemlock

-------
                            29
and 9 from Section Four.  One thousand ml were collected for



each sample, and collections were taken once or twice a week.



These samples were placed in plastic graduated cylinders,



four ml of Logol's was added and they were allowed to settle



for two days.  After two days, the supernate was drawn off



through a small glass tube inserted in the side of the



cylinder.  About 30 ml of concentrate remained after decant-



ing.  This was drawn off through a larger glass tube inserted



at the bottom of the cylinder.  The cylinder was rinsed with



distilled water and added to the concentrate.  About 60 ml



of concentrate was thus obtained.  This was shipped to East



Lansing for examination.



     A subsample of each sample was filtered through a HA



millipore filter.  After filtering, a drop of oil was added



and the sample examined at 150x to 600x magnifications.



From 10 to 50 fields were examined, depending on the algal



concentrations, to give an estimated ± 5% error.







Primary Production





     Primary production estimates were made using the carbon-



14 technique in a constant light intensity chamber (Figure 6



(a).  Water samples were collected with a PVC water sampler



and transferred to 125 ml Pyrex bottles.  One light bottle



and one dark bottle were collected from 0.3m.  These samples



were immediately placed in a light tight box and taken to



the incubation chamber.   About 25 minutes passed between

-------
                            30
Figure 6.  (a)  Phytoplankton incubation chamber.  Four
              submerged sample bottles on the rotating
              wheel.   (Photo by author.)
          (b)  Periphyton ring.   Five plastic periphyton
              slides  are visible clamped to the ring.
              (Photo  by author.)

-------
   Fig.  6a
Fig. 6b

-------
                            32
collection and incubation.  Each Hemlock Lake sample was
inoculated with 1.0 uc 14C in 1.0 ml distilled water both
years.  Each Section Four sample was inoculated with 1.0 uc
in 1.0 ml distilled water during 1969, but 2.0 uc in 20 ml
distilled water during 1970.  The samples were incubated
in the light chamber for four hours.  They were placed on
their side on a revolving horizontal wheel.  The wheel was
45 cm in diameter and turned at 5 revolutions/min.  The water
was  usually kept within 5°C at the lakes' surface tempera-
tures.  The bottles revolved 33 cm from 6 Sylvania Gro-Lux,
20-watt fluorescent bulbs and 20 cm  (minimum distance) from
two  G.E., 150-watt spotlight bulbs.  After incubation, 50 ml
of Hemlock sample was filtered through a 25 cm diameter HA
millipore filter  (0.45 u).  Fifty ml of each Section Four
sample was filtered during 1969, but 100 ml was filtered
during 1970.  Each filter was rinsed with three percent
formalin after the sample water had passed through.  After
drying in the dark, the filters were exposed for 15 minutes
to HC1 fumes (Wetzel,  1965).  Sample activity was measured
using a Tracerlab Omniguard Low Background, internal gas
flow Counting System;  the efficiency of which was determined
from samples of known activity.  Efficiency averaged about
12%.   Total inorganic carbon available for photosynthesis-
and the rate of carbon fixation was estimated using Saunders
et al.  (1962)  method.

-------
                            33
Periphyton





     Periphyton standing crop was measured as ash-free dry



weight.  The periphyton grew and collected on 6mm x 50mm x



127mm  plexiglass slides that were suspended horizontally in



the lake at six depths.  The slides were fastened with metal



clips to a plastic ring (Figure 6 (b)).  The ring was sus-



pended by rope from a horizontal bar held above the lakes'



surface by floats.  One series consisted of four slides at



each depth.  These were usually incubated 17 days before



collection.  The second series consisted of three slides at



each depth.  These samples were incubated 34,  51 or 68 days



and represented total, undisturbed accumulations during the



sample period.  Periphyton samples were scraped from the



slides and stored in 90% ETOH at -15°C until dried.  They



were dried at 105°C four days and held in a dessicator until



incinerated.  They were incinerated at 550 C for 10 minutes.



They were cooled in the dessicator before weighing.  Weights



to the nearest 0.1 mg were measured.  All periphyton weights



collected are presented in the Appendix (Tables A-l through



A-4) .





Zooplankton





     Two series of zooplankton samples were collected during



1970.  Three samples from each depth were collected by fil-



tering 22 liters of water through a Wisconsin plankton net.



The net was suspended on the pontoon raft and water was

-------
                            34
pumped via a garden hose into the net.  About 8 liters of



water per minute were pumped.  The concentrated zooplankton



were preserved in 3% Rose Bengal solution made from 90%



ETOH.  The stained specimens were later filtered through a



HA millipore filter with a printed grid.  The filter was



then placed over a drop of glycerin in a plastic millipore



petri dish.  This gave a permanent, mount and the grid



facilitated counting.








Zoobenthos





     Zoobenthos samples were collected along one transect



in  each  lake with a screened Ekman 15-cm square dredge



 (Welch,  1948).  The screen was No. 30 brass sieve.  Five



samples  were collected from each depth interval each sample



period.  Five collections were made in each lake each year,



and the  depth intervals in meters were:  0.0-3.7, 3.7-7.3,



7.3-11.0, 11.0-14.6, and 14.6-maximum depth.  Maximum depth



was 18.6 meters in Hemlock and 19.1 in Section Four.



Collections were made at three-week intervals beginning in



mid-June each year.   A total of 125 samples were collected



from each lake each year.  Samples were sifted through a



No. 30 brass sieve (0.59 mm opening)  at the lake and later



preserved with 20% formalin.  Reisch  (1959) found that



about 93% by biomass of his marine zoobenthos were retained



by a No.  20 sieve.   The No. 30 sieve used in this study



probably retained comparable biomass,  but Jonasson (1955)

-------
                            35
has shown that zoobenthos population estimates can be biased



by sieve size.  These samples were later stained with a 3%



Khodamin B solution in 90% ETOH and sorted into 90% ETOH



using Anderson's  (1959) sugar flotation technique.  An il-



luminated magnifier (2x) lamp was used in the sorting.  Fast



 (1970) found the  efficiency of sugar .flotation is high for



such groups as midge larvae and pupae, but much lower for



oligochaetes.  I  feel these efficiencies were somewhat



greater in my present study because the present samples were



stained and a better magnifier was used to sort.  Sixteen



zoobenthos categories were identified, and organisms were



sorted into these categories.  These include:  oligochaetes



 (microdriles) , oligochaetes (megadriles), Chironomid larvae,



chironomid pupae, amphipods, dragonflies  (Anisoptera),



damselflies  (Zygoptera), mayflies  (Ephemeroptera), Chaoborus



spp. larvae, Chaoborus spp. pupae, clams, Heleidae  (=Cera-



topogonidae)larvae, Trichoptera larvae, Tabanid larvae,



Megaloptera and leeches.  No attempt was made to determine



relative species  compositions of each category, but represen-



tative specimens  of certain categories were sent to taxon-



omists  for the respective group.  The number of organisms



in each category  were counted for each sample, and their wet



weight measured to the nearest 0.001 mg.  Before weighing,



excess moisture was removed using King and Ball's  (1964)



technique,  but were centrifuged for 4 minutes instead of 30



seconds.   All numeric and weight data are presented as wet



weights in the Appendix (Tables A-5 through A-8).

-------
                            36
Crayfish





     Crayfish (Orconectes virilis)  were collected with modi-



fied wire minnow traps (Momot and Gowing,  1970) suspended



from the transect line at ten depths.  After collection they



were either removed from the lake,  or released near the



center of the lake.  Four collections were made from each



lake between August 12 and August 30, 1969,  but 22 collec-



tions were made from each lake between June 7,  1970 and



September 4, 1970.  Times between collections varied from



one day to one week.  Crayfish were sexed upon collection.








Emergent Insects





     Emergent insects were captured using a new half-square



meter submerged trap  (Figure 7a) .  The trap has a steel frame



on which clear polyethylene plastic is attached.  On hard



substrate, it is held off the bottom by legs, thus permitting



water to circulate between the trap and the lake.  A glass



jar is attached to the top of the trap and collects the in-



sects as they rise to the surface.   A removable funnel and



perch apparatus is situated in the jar mouth (Figure 7b).



The funnel is made from two styrofoam drinking cups.  The



bottom cup has a small hole in its top to permit entry of the



insects and fine mesh netting on one side.  The other styro-



foam cup has most of its sides cut away and is glued to the



first cup to form a perch for emerged insects.   Water is added

-------
Figure 7.   (a) New emergent insect trap.   (Photo by Dr. 0. E. Kurt.)

            (b) Styrofoam trap used in collection jar on the  emergent
               insect trap.   (Photo by Dr. O  E. Kurt.)

            (c) Replacing collection jar on emergent insect trap.
               Trap is suspended from bracket on the raft and not
               taken out of the water during transfer.   (Photo by
               Dr. O. E. Kurt.)

-------
Fig. 7a
Fig. 7b
Fig. 7c

-------
                            39
to the jar as needed to assure that the water level will be



between the top of the first cup and bottom of the perch



when the trap is in sampling position.  The amount of water



added will depend on sampling depth.  At 20 meters,  for



example,  no water is added beforehand since hydrostatic



pressure will reduce the air space to 1/3 its surface volume.



On the other hand,  the shallow water bottles must have 2/3



their volume filled to assure an adequate level.  As the deep



water traps are lowered,  water will flow into the bottle



through the hole on the top, and the netting.  As they are



raised, the water will flow out these same apertures.  If



the netting were not present, the water would be forced to



flow between the jar lip and the cup.  The netting prevents



this while retaining organisms and exuvia that are still in



the water.



     Five pair of insect traps were used in each lake; two



at each of the same depth intervals indicated for the zoo-



benthos.   The traps were used to collect insects five days



each week.  They were attended and moved daily,  and allowed



to dry out during the other two days each week to prevent



insect attachment.   The first day of each week the traps



were placed in the water and suspended from the raft by a



hook.  The sample jar was added and the trap lowered to the



bottom by a rope.  The next day the trap was raised to the



surface and again suspended from the hook while the sample



jar was removed and another jar added (Figure 7 (c)).

-------
                            40
The trap was then lowered to the same depth interval, but at



a slightly different location.  The body of the trap was not



taken out of the water during the sampling period.  It was



not taken out for two reasons:  (1)  to prevent entrapment of



insects on the surface of the lake in the trap.  This was



sometimes a problem during the first sampling period of each



week.  Aquatic and terrestial insects will accumulate on the



lake1s surface and be entrapped as the trap is lowered into



the lake; and (2) it is more efficient and easier not to re-



move the trap.  The collection jars were taken back to the



laboratory where the insects were removed.  At the laboratory,



a small amount of tap water was added to the jar if needed



and the jar was vigorously shaken to disorient the adult



insects.  The top and funnel were then removed and the water



passed through a small fine mesh net into a white enamel tray.



The funnel and jar were rinsed into the net.  Insects and their



exuvia were preserved in 90% ETOH vials for later identifica-



tion and counting.  Water was added to the jars as needed and



the cap secured to be used for the next day's set of samples.



Emergent insect samples were collected from June 15,  1969



through September 5,  1969 and from June 7, 1970 through



September 4,  1970.  Six hundred and 650 samples were collected



from each lake during 1969 and 1970 respectively-








Rainbow Trout






     One thousand right-pectoral fin clipped rainbow trout



(RBT)  were stocked in Section Four on June 6",  1969,  and

-------
                            41
1,002 right-pelvic fin clipped RBT were stocked in Hemlock



on June 6, 1969.  These fish were mostly one-year-olds raised



at the Michigan Department of Natural Resources  (DNR) trout



rearing ponds at Wolverine, Michigan.  Both lots averaged



7.3 inches  (188 mm) fork length  (Figures A-l and A-2).



The fish were released at one point on the shore of each



lake.  We measured each fish at the rearing ponds after they



were anesthetized with MS-222.  We observed only a few dead



fish in each lake after their release.



     One thousand seventy-one left-pelvic fin clipped RBT



were released in Section Four on May 23, 1970.  These fish



were one-year-olds raised at the DNR's Wolverine rearing



ponds.  The fish averaged 7.9 inches  (200 mm) fork length



(Figure A-2).  They were measured and handled the same as



during 1969, except that they were sorted at Wolverine with



wooden sorting trays.  This was necessary to assure larger-



sized fish than were stocked during 1969.  Only a very few



fish of less than 8.0 inches (215 mm) were captured by our



gill nets.  Two vertical gill nets similar to those described



by Horak and Tanner (1964)  were used in each lake (Figure 8



(a)).  The stretched mesh sizes were 3.4 inch (19.0 mm) and



1.0 inch (25.4 mm).  The nets were tied to the transect line



at its deepest point (Figure 4).  The nets were pulled once



a day and the depth of capture and fin clip of each fish was



determined (Figure 8 (b)).



     The RBT stocked in Hemlock during 1970 were handled



much differently than during 1969.  About 2,000 one-year-olds

-------
                            42
Figure 8.   (a)  Vertical gill net and roller.  Gill net is
               suspended between raft pontoons as during
               sample collection process.   (Photo by author.)
           (b)  Robert Hoffman removing fish from vertical
               gill net.   (Photo by author=)

-------
Fig. 8a
Fig. 8b

-------
                            44
were transferred from the DNR's Wolverine rearing ponds to



their hatchery at Grayling about June 1, 1970.  These fish



were sorted through wooden sorting trays at Wolverine.  On



June 25, 1970 these fish were divided into four lots of over



500 fish each.  Each lot was measured as before and clipped



with a different fin clip.  These fish averaged 8.0 (203 mm)



and 8.1 inches (206 mm) fork length (Figure A-l) .  The four



lots were transported to Hemlock June 26, 1970 and each lot



was put in a separate cage.  The tops and bottoms of the



cages were hexagons with each segment being 2.2 meters



 (Figure 9) .  One meter separated the top and bottom of each



cage and they each contained 11.6 cubic meters of water.



Two of the four cages were completely covered by wire and



plastic, while the other two cages had only the top and



bottom covered with plastic and wire and the sides were



covered with aluminum window screen.  One covered cage with



the right-pectoral clipped fish was held at 3m depth



 (Figure 10).  About 4,000 gallons/hour  (15 m3/hour) were



pumped from 1 meter for the first day, and from 12 meters



for the remainder of the acclimation period.  The temperature



within this cage was measured with a resistance thermometer



and found to be the same temperatures as water at 12 meters.



One screened cage with the anal-fin clipped fish was held



at 3 meters depth.   The other screened cage with the left



pelvic clipped fish was lowered to 5 meter depth for one day



and then to 12 meters.   The other covered cage with the left

-------
                            45
Figure 9.  Acclimation cage used to hold rainbow trout at
           specific depths in Hemlock Lake.   Rubber hoses
           led to water pumps on a raft.  Cage is covered
           with polyethylene plastic and chicken wire.
           (Photo by author.)

-------

-------
Figure 10.  Configuration of Hemlock Lake acclimation cages.   Fin clip of
            rainbow trout held in each cage is shown.  The oxygen and
            temperature profiles during the acclimation period are also
            shown.

-------
                                                                                      0
    Oxygen


4      8	]2
    0-






    2-






    4-






? 6-
 a
 a>
a
    14






^' "


-I
^r>



Covered Cage
Rt. Pec. Clip

1




.
T
i—> '•
«MH

\



Screened Cage
Left Pel. Clip
-^

L

^^^^-^

1
Screened Cage
Anal Clip
1

Covered Cage
Left Pec. Clip
                                                                                                 Temperature


                                                                                             10     15     20
    184

-------
                            49
pelvic clipped fish was lowered to 7 meters for one day and



then to 12 meters.  About 15 m3/min. of water from one meter



depth was pumped into this cage during the entire acclima-



tion period.  Water left the two cages through 5-inch diam-



eter  (127 mm) irrigation pipe.  The pipes were screened to



prevent fish passage.  The pipe extended from 12 meters to



6 meters in the case of the 12 meter covered cage, and from



3 meters to 9 meters in the case of the 3 meter covered cage.



     On July 1, 1970, the covered 3-meter cage and 12-meter



cages were opened and left at their respective depths.



Water was pumped into each covered cage as usual for another



8 hours.  The 12-meter screened cage was opened at 12 meters



and then floated to the surface.  Over 200 RBT remained in



the cage as it surfaced.  The screened sides were then en-



tirely removed, but the fish were reluctant to leave.  Rather



than open the screened surface cage at the surface as orig-



inally planned, I lowered this cage with the anal-fin clipped



fish to 12 meters on July 1,  1970.  They remained there until



July 7, 1970 at which time I opened the cage at the 12-meter



depth and left it there.



     A few left-pectoral and left-pelvic fish were caught in



the gill nets between June 25th and July 1st.  These fish



may have escaped from the cages, or escaped when the fish were



being placed in the cages from the planter truck.  The cages



were later inspected but no obvious openings were present.



The length,  weight and scale samples were recorded for each

-------
                            50
fish collected in the gill nets,  and gross estimates of their



stomach contents were made.



     Only a few dead fish were observed in Hemlock following



stocking during 1970 but 15  dead fish were observed in



Section Four after they were released in May.








Statistics





     All confidence estimates placed on means,  or totals



were computed using the appropriate t and standard error



values.  Non-homogeneous and non-rectifiable variances negated



the use of analysis of variance tests and other parametric



statistics.  For these reasons these tests were not applied.



The graphical method (Dice and Leraas,  1936)  is used to



compare appropriate means in many cases.   Most of the calcu-



lations were performed on Michigan State University's CDC



3600 computer.   Most figures were drawn using this computer



and their CalComp plotter.

-------
                 DESCRIPTION OF THE LAKES








     Hemlock and Section Four lakes are located in the



Pigeon River State Forest, about 85 km south of the Straits



of Mackinac.  The Pigeon River Trout Research Area includes



these lakes, four other lakes and 8.7 mg of the Pigeon River,



and has special use restrictions.  All the lakes were closed



to public angling and certain other activities during our



study.  Hemlock Lake is in Cheboygan County and Section Four



is in Otsego County.  These two lakes are only 3 km apart,



but possess much different properties.



     Earlier observers (Eschmeyer, 1938) thought these lakes



were glacial pit lakes (Scott,  1921),  but later evidence



indicates they are actually lime sinks  (Tanner, 1952, 1960).



This latter hypothesis was partly substantiated when the



west shore of Section Four collapsed during May 1950.  The



lakes are apparently enlarging in this manner,  and scallop-



ing of their margins is evident from aerial observation



(R. C. Ball, personal communication).



     The lakes are nearly circular in outline with concentric



depth contours (Figures 11 and 12).  Although not shown on



Hemlock's contour map,  a marshy area is situated just north



of and confluent to the lake.  This area is in direct contact
                            51

-------
                            52
Figure 11.   Contour map of  Hemlock  Lake showing sample
            transects  and aerator.   Depth  intervals
            are in meters.

-------
                         S 33320
  Hemlock  Lake
Scale: 25.6mm = 30.0meters
    0  10  20 30
       scale

-------
                            54
Figure 12.   Contour map of Section Four Lake showing
            sample transect and air line.   Air was
            released from the dashed section of the
            air line.   Depth intervals are in meters.

-------
                Sample Transect
     Section Four Lake
Scale: 31.6mm = 30.0 meters

-------
                            56
with the lake, but contains only a few inches of water.



Very dense vegetation undoubtedly restricts communication




between the lake and the marsh.



     Tanner (1952) indicates that during his September 1950



surveys of the lakes Hemlock had a maximum depth of 19.5



meters and surface area of 2.4 hectares,  whereas Section



Four was 22.8 meters and 1.0 hectares.  Based on the 1957



surveys and my depth soundings during January 1970, Hemlock



was 18.6 meters maximum depth and 1.8 hectares,  while Section



Four was 19.1 meters and 1.2 hectares.  My Hemlock surface



area value does not include the marsh area to the north of



the lake.  The 1957 surveys measured 18.1 meters maximum



depth in both lakes.  These data indicate that Hemlock water



level has been relatively stable, but Section Four's has



changed as much as 4.6 meters between 1950 and 1957.  This



change may have been due in part to the basin collapse



during May 1950 (Tanner, 1952).  The collapse could have



caused greater siltation of the basin, or increased seepage



from the basin.  Tanner's September 1950 lake survey did not



indicate an immediate depth change due to the collapse.



     Section Four's water level was stable during the summer



1969,  fluctuating only a few centimeters.  Its early summer



1970 level was within a few centimeters of the 1969 level,



but decreased about 0.3 meters by the end of the summer.



The greatest rate of decrease occurred during August 1970.



Hemlock's water level was very stable both summers and fluc-



tuated only a few centimeters during the study period.

-------
                            57
     Although both lakes are designated marl lakes, they



differ greatly.  Section Four is a typical marl lake.  It is



oligotrophic with plentiful oxygen at all depths all year.



During 1969-70 Chara spp. was found at all depths.  It was



sparse in the shallow depths, but formed large beds in



deeper water.  Other rooted plants were sparse except at the



shoreline.  Phytoplankton was very sparse,  with the result



that secchi disc transparencies often exceeded 12 meters.



On August 11, 1969, more than 15% of the surface irradiance



was still present at 12 meters  (Figure 13).  Secchi disc was



10.25 meters on this date.  The water had the greenish-blue



coloration typical of marl lakes.



     Section Four's sediments are mostly calcarious.  They



range between 1.2% and 12.2% as CaCOa-C on a dry weight



basis  (Figure 14).  Organic carbon ranges between 0.0% and



6.4%.  The profundal sediment measured 3.4% organic carbon



and agrees well with Barrett's  (1952)  post-collapse data.



Before the collapse of the west shore during May 1950, Sec-



tion Four's profundal sediments average 41.2% as organic



matter.  After the collapse they averaged 6.2%.  This indi-



cates that the profundal sediments were blanketed with a



layer of sand and silt which in effect sealed in the rich



organic matter.  This occurrence is also evidenced by changes



in the phosphorus content of Section Four's profundal sedi-



ments.   These sediments averaged 37.2 mg PO4/kg air-dry soil



before the collapse,  but only 1.7 mg/kg after.  The removal

-------
                            58
Figure 13.   Hemlock and Section Four  relative irradiance
            measurements on August 11,  1969.

-------
     Percent of Surface Irradiance
 20      40
i  I  i  i  i  I
                          0      80
                          I  i  i  i  I  i i
100
 6-
 9-
12-
        /' Hem.
15-

-------
Figure 14.  Section Four transect profile illustrating percent organic
            carbon and percent CaCO3-carbon at different depths.

-------
   2-



   4-



   6-
Q.KH
0)
Q
  12-J
  14 -



  16-



  18-

                  Distance  From  Shore (m)
   1	I I I  I I f I  I
TO
i 1 i
                         20
   I I I I I I
30
j I i
,5f
                          HU CaCO3-C%


                          I   1 Organic-C%
                        20


                        10


                        - 0
                                                 C
                                                 0)
                                                               0)

-------
                            62
of this nutrient-rich material from the lake ecosystem should



decrease productivity.  This was not immediately evident



following the collapse because of artificial fertilization



during 1949 and 1950  (Tanner, 1952).  Long-term comparisons



indicate productivity has decreased.  Survey data from August



1932  (Table 1) indicates a sharp oxygen reduction below the



metalimnion, approaching zero at the bottom.  Carbon dioxide



and total alkalinity  increased markedly below this depth.



Secchi disc was only  6.8 meters.  During August 1969, oxygen



values were more than 4.5 mg/1 at all depths.  Carbon dioxide



was less than 8 mg/1  and alkalinity was not stratified.



Secchi depths were generally greater than 10 meters.  These



comparisons indicate  that primary productivity decreased as



a result of the basin collapse and organic profundal sedi-



ments entrapment.



     Hemlock Lake is not a typical marl lake.  Although it



is highly alkaline,  and has marl deposits in shallow water,



these sediments are covered with a thick layer of peaty



organic material,  The organic material originated mostly



from the surrounding forest and the marsh bordering the lake.



It is very loosely compacted in shallow water,  but gelatin-



ous in deep water.   These profundal sediments float to the



lake's surface when broken loose from the bottom by sampling



gear.   The surrounding forest contains mostly deciduous



species which shed much of their leaves into Hemlock Lake.



Tree leaf remains are evident in the profundal sediment.

-------
Table  1  .   Section Four Lake limnological data collected
            August 1, 1932 by the Institute for Fisheries
            Research, Michigan Department of Natural
            Resources.  Secchi disc depth was 6.8 meters
            on this date.

Depth
(m)
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
Temp.
C°c)
24.4
22.8
22.5
22.2
21.7
20.6
16.1
13.3
12.2
11.1
9.4
8.3
7.8
7.2
6.7
6.1
6.0
5.6
5.6
5.6
5.2
5.0
5.0
Oxygen
9.1

- _

—



11.1
—

—


1.8



- -
_ _

Trace
- _
Free
co2
0.0
_ -
_ _
- -
- -


- -
4.0
—
—
- -
—
_ -
13.9
- -


- -
_ -

22.0
_ _
pH
7.9


- _
—

—
- -
7.6




- -
7.3

- -

- -
- -

7.2
_ _
Total
Alkalinity
154


_ _

_ _

- -
174




- -
187


- -

_ _

198
_ _

-------
                            64
Section Four Lake, on the other hand, is surrounded mostly



by evergreens.  Hemlock's sediments range from 0.0% to



43.1% as organic carbon, and from 0.0% to 26.1% as CaC03



carbon  (Figure 15).  During 1949-50 Hemlock's profundal



sediments averaged 53% as organic matter, compared to 43%



for  1968-69  (Barrett, 1952).  This difference is probably



due  to analytical and sampling differences rather than



changes in the sediment.



     Unlike  Section Four, Hemlock Lake is eutrophic and



meromictic.  Prior to aeration, the bottom few meters were



amictic and  formed a monimolimnion that did not circulate



following spring and fall turnovers.  This zone was anoxic



and  contained high concentrations of carbon dioxide and



other gases.  Water samples drawn from this depth effervesced



due  to the release of dissolved gases.  These gases were held



in solution  by hydrqstatic pressure.  Carbon dioxide,



alkalinity,  and conductivity values increased sharply in this



zone (Figure 16).  Oxygen values were always 0.0 mg/1 and pH



was  about 6.6.  A small unidentified bacterium  (Brian Moss,



personal communication) , was very abundant in this zone and



was  apparently responsible for the greenish tint of the moni-



molimnion water.  Fine green needles appeared suspended in



the water.   Abundant filamentous algae or bacteria have often



been found in the monimolimnion of other meromictic lakes.



Hemlock's meromixis developed after 1932, since survey data



collected during July 1932 indicate no monimolimnion  (Table 2)

-------
Figure 15.  Hemlock transect profile illustrating percent organic
            carbon and percent CaC03-carbon at different depths.
                                                                                     cn

-------
            Distance  From Shore (m)
i i i I i ........ I i i I i t I
         .
Q— ,-t M ' I I » I I I I I I I I I I I I I i i I I I I I I I I i i I i i , i i i 1
                                  mm Caco3-c%

                                      Organic-C%

-------
                            67
Figure 16.   Hemlock Lake carbon dioxide alkalinity,  pH
            and conductivity profiles on August 13,  1969.
            This is representative of pre-aeration con-
            dition.  Chemocline of monimolimnion is
            evident below 12 meters.

-------
                         -6
          Conductivity (10 mhos)

9         300         600         900





















x-^
E
\^

1L
0)
Q





<


6


1
f\
0—


2

4


6-



8-


10 -

12-
14-
16-
18-
— i I i i — i — i i i i i — I—I — l — i — 1 — i — I — I — r—
Carbon Dioxide (mg/l)
3 100 200
i i 1 1 1 1 1 1
1 1 1 1 i 1 i 1
0 70 Ph 8.0 9


Total Alkalinity (mg/l)
200 400 6V


J
\
Alk. > ,,
|PH
\" , Cond. \
. x
• \?

i /
i /
i /
\ 1 ^
^ ! ^^
\ i ^-^'
\ i /'
' \ • /
. V, \\ /
Co\v Vy'
/'^^^^£H£^



-------
Table  2  .   Hemlock Lake limnological data collected July
            28,  1932 by the Institute for Fisheries  Research,
            Michigan Department of Natural Resources.  Secchi
            disc depth was 4.2 meters on this  date.

Depth
(m)
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
Temp.
(°C)
21.1
20.6
20.6
20.0
19.4
14.4
13.3
10.0
6.7
6.7
6.1
6.1
6.1
6.1
6.1
5.6
5.6
5.6
5.6
5.6
Oxygen
8.7

- -




_ -


6.0






_ _
_ _
2.3
Free
co2
10
. _
_ _
- .
_ _
_ _
_ _
_ _
_ w
_ _
12
_ _
_ w
_ _
_ _
_ _
_ _
_ .
w _
39
pH
8.1
- -
_ _

- -
_ _
_ _
_ _
_ _
_ _
8.0
w •
. .
_ _
. .
. .
. -
* *
— —
7.4
Total
Alkalinity
160
_ _
.. .
«. _
_ _
M M
_ _
_ _
— _
_ .
156
• •
. -
w .
• •
w .
• ••
• -
— —
198

-------
                            70
Oxygen was 2.3 mg/1 at the bottom compared to zero during
1969.  Carbon dioxide and alkalinity increased below 10
meters, but not dramatically.  pH was 7.4 at the bottom
compared to 6.6 during 1969.  The 4.2 meter secchi disc read-
ing  is comparable to 1969 values.
     Due to its organic richness, Hemlock secchi disc trans-
parency seldom read more than five meters prior to aeration.
On August 11, 1969 less than 1% of surface irradiance was
still  present at 9 meters (Figure 13) .  Secchi disc was 6.25
meters on this date.  The water supported abundant phytoplank-
ton  and zooplankton standing crops.  Submergent aquatic plants
were very sparse, while emergents were abundant only along
the  shore and in the marsh.
     During 1969-70, Hemlock Lake contained bluntnose minnows
 (Pimephales notatus) and redside dace (Gila elongata).  Both
species were abundant and reproduce in the lake.  They are
primarily littoral in habit, but are sometimes found in open
water near floating objects.  Neither apparently inhabit depths
greater than a few meters,  even when the lake is isothermal.
One  brook trout  (Salvelinus fontinalis)  was caught on the
first net set in 1969,  but none were found thereafter.
     Section Four Lake contained a residual population of
rainbow trout at the time my study began, but no other fish
species.   These fish were stocked during 1964 and 1965.
Three thousand trout,  at 2,200/kg,  were stocked each year
(Carl Latta,  personal communication)•  These fish averaged
about 33  cm FL during 1969-70,  but were very emaciated.

-------
                            71
They typically had disproportionately large heads and "slab""



sides.  The lack of abundant,  large zooplankters or other



desirable forage undoubtedly accounts for their condition.



They fed almost exclusively on adult insects that fell on



the lake1s surface.  Detritus and benthic filamentous algae



were often present in their stomachs.  They very seldom fed



on crayfish or zoobenthos,  although both were relatively



abundant.   Trout mature sexually,  but do not reproduce in



either Hemlock or Section Four lakes.



     Tanner (1952,  1960),  Barrett (1952) and Siler (1968)



describe in greater detail the history and artificial ferti-



lization of these lakes.  Both lakes were fertilized during



1949 and 1950.  Their productivities were greatly increased



and they became more eutrophic.  They apparently returned to



their prefertilization conditions by 1967,  however-

-------
                       HEMLOCK LAKE








Hvpolimnion Aerator





     The aerator used in Hemlock Lake is a new design.  It



differs significantly from one used by Bernhardt  (1967) and



one designed by R. E. Speece (Fast, 1968).  It is only the



second hypolimnion aerator to ever have been used success-



fully.



     The aerator free-floats in the center of the lake



(Figures 17 and 18) .   Styrofoam and steel barred floats sus-



pend it off the bottom,  and air is delivered through a 38 mm



I.D. plastic pipe from a shore compressor.  The aerator is



held in place by four anchors and ropes.  The aerator was



partly fabricated by Armco Steel Corporation at their Indiana



plant.  It was trucked to Hemlock Lake where it was unloaded



onto a wooden cradle (Figure 19 (a)) .   The 3.1m section that:



extends above the lake's surface was then banded to the lower



section.   The wooden cradle provided support for the aerator



while the floats and hardware were attached, and made it



easier to slide the aerator into the lake.  The cradle



rested on 27 round wooden posts.  The post, in turn, rested




on two wooden "rails" that ran into the lake.  After the



floats and hardware were attached, the aerator rolled down
                            72

-------
                            73
Figure 17.   Cross-sectional view of Hemlock Lake
            hypolimnion aerator.  Dotted lines repre-
            sent projected edges.  Tower is tilted
            toward the viewer,  and parts are drawn
            approximately to scale.

-------
                            , Air Line
Styrofoam Safety
    Flotation Unit
           Barrel
    Flotation Unit
       Styrofoam
   Flotation Unit
    Current
  Deflector
       Air Diffusor —
Epilimnion
                                     Hypolimnion

-------
Figure 18.  Hemlock hypolimnion aerator in operating position.  Only the
            upper three meters are visible.  Air supply line enters the
            scene from the left.  Above water styrofoam flotation units
            are a safety feature to prevent tower from sinking if sub-
            merged units should fail.  Ladder on side of aerator permits
            access to top of tube.   (Photo by author.)

-------
—"-—="—— -   -	—£

-------
                            77
Figure 19.  (a)  Arrival of Hemlock hypolimnion aerator tubes
               from the factory.   Top three meters section
               of aerator is separated and located next to
               the truck cab.   The tower was unloaded on
               the wooden cradle and logs in the foreground
               and fittings were attached before it was
               shoved in the lake.  (Photo by author.)
           (b)  Hemlock hypolimnion aerator floating hori-
               zontally.   Temporary floats kept the lower
               end up while the current deflector was
               attached.   The lower end is closest to the
               viewer.  Deflectors could not be added until
               the aerator was floating horizontally in the
               lake.   (Photo by author.)
           (c)  Hemlock hypolimnion aerator tilting into
               sampling position.   The temporary floats
               have just been removed by the author using
               SCUBA.   (Photo by Robert Hoffman.)

-------
Fig. 19a
Fig- 19b
 Fig. I9c

-------
                            79
the rails on the posts and into the lake.  Temporary styro-



foam floats were attached to the bottom of the tower so the



tower floated horizontal in the water (Figure 19  (b)).  This



was necessary since the current deflector had to be attached



after the tower was in the lake.  After the current deflector



was attached, the temporary floats were cut loose and the



tower swung into its vertical operating position  (Figure 19



(c)).





Description.  The aerator consists of two corrugated, 14-



guage,  galvanized iron tubes (Figure 17).  One 1.85 meters



diameter tube extends 3.1 meters above the lake surface, to



9.2 meters below the surface.  The other 1.38 meters diameter



tube is partly located inside the larger tube and extends



from the lake's surface to the 15.5 meters depth.  The smaller



tube is attached to, and positioned within the larger tube



by eight 13 mm x 0.31 m x 0.49 m iron plates.  These plates



are welded to the outside of the smaller tube, extend through



slots cut in the larger tube and are also welded to the larger



tube around the slot (Figure 20).   Four plates are thus



located near the top of the small tube and four near the bot-



tom of the large tube.   The plates are spaced 90° apart



around the circumference of the tubes at each site and are



positioned with their long axis vertical.



     A current deflector is attached to the small tube by



clamps  and is located one meter below the bottom of the large



tube (Figures 17,  19 (b)  and 21).   It is 6.3 meters in

-------
                            80
Figure 20.   Cross-sectional view and parts of hypolimnion
            aerator.

              A.  Cross-section of aerator taken near the
                 top.   Two styrofoam flotation units and
                 one  barrel flotation unit are shown.

              B.  Styrofoam flotation unit.

              C.  Barrel flotation unit showing the tee
                 structure used to attach it to the aerator.

              D.  Cross-section of tee inside the slot
                 structure.   The slot is  welded to the out-
                 side of the aerator.

-------

-------
Figure 21.  Hemlock hypolimnion aerator current deflector before they
            were attached to the aerator.  Anchors in foreground were
            used to anchor the tower in its operating position.
            (Photo by Ed Schultz.)

-------

-------
                            84
diameter.  The deflector was constructed in halves.  Each



half has a central axis of 14 guage corrugated iron.  The



iron is curved with the same radius as the small tube and



is 0.31 m wide.  Seven 13 mm I.D. pipes are welded perpen-



dicular to the axis.  These pipes are 2.5 meters long.



A 6 mm x 51 mm piece of flat iron is welded to the ends of



the pipes to form a large semi-circular rim of radius 3.15



m.  Two additional 2.5 m length of flat iron extend from



the axis to the outer flat iron rim.  Galvanized fencing



 (2 cm x 2 cm holes)  is wired to the pipes and flat iron.



A nylon parachute covers the top of the fencing and is drawn



tight about the axis.  Chicken wire (6 cm diameter holes)



covers the parachute and holds it firmly against the fencing".



     The aerator is free-floating.  It is buoyed up by six



styrofoam flotation units and four 220 liter steel drums



 (Figures 19 (b) and 20) .  Each drum has a 12 mm aperture on



its bottom that is open to the water.   A 3 mm copper tube



and globe valve extends from the top of the drum to above



the water level.  Air may thus be let out of the drum by



opening the valve.   The buoying of the aerator is thus



adjusted with these drums.  Each styrofoam flotation unit



consists of four 0.15 m x 0.46 m x 2.5 m pieces of styroform



(Figure 22 (a)).  These are sandwiched between two pieces of



19 mm marine plywood and banded by 6 mm x 51 mm flat iron



strips.   Both barrel and styrofoam units are attached to the



aerator via a 6 mm thick,  51 mm x 38 mm iron tee (Figures 20



and 22  (a)).   The tee is 1.8 m long and slides into an iron

-------
                            85
Figure 22. (a)  Styrofoam flotation unit used on the Hemlock
               Lake hypolimnion aerator.  The tee structure
               used to attach the unit to the aerator is
               shown on top.   (Photo by author.)
           (b)  Slots for flotation unit tee1s being welded
               on the side of the aeration tower.  The iron
               plates used to position the inner tube are
               shown projecting through the outer tube to
               the left of the workmen.  (Photo by author.)
           (c)  Sliding styrofoam flotation unit into slot
               on side of aeration tower.   Logs and "runway1
               are shown leading into the lake.  (Photo by
               Ed Schultz.)

-------
                         o
                         CM
                         CM

                         O)
 Fig.  22 b
Fig.  22 c

-------
                            87
slot welded to the side of the large tube  (Figures 22  (b)



and 22  (c)) .



     Not shown in Figure 17 are two small styrofoam units



that were attached on opposite sides of the tower to the lower



steel plates that connect the inner and outer tubes.  These



floats are attached with block and tackle to the portion of



the plate that extends through the outer tube.  These styro-



foam units were used to adjust the vertical position of the



tower in the water.  Also not shown in Figure 17 is a 220-



liter steel drum filled with cement and attached near the



bottom of the tower.  This was added after the tower was in



its operating position and I discovered there was too much



buoyancy.  Parts of the submerged styrofoam units were cut



away using SCUBA to further reduce buoyancy.



     Air was released from an air diffusor located one meter



from the bottom of the aerator (Figure 23).  A PVC plastic,



38 mm-diameter air line passed over the top rim of the



aerator and down to the diffusor.  The air line was fastened



to the walls of the aerator with clamps and sheet metal



screws.  The diffusor consisted of four 61 cm long,  38 mm



diameter steel pipes joined by a side-armed cross fixture.



Four 6.8mmx58.8mmx30 cm pieces of flat iron were welded



over the outer ends of the pipes.  A hole was drilled in each



piece of flat iron and a bolt used to fasten them to the



walls of the aerator.   Each of the four diffusor arms con-



tained 24 3.18 mm diameter holes.  Eight sets of holes were

-------
                            88
Figure 23.   Air diffusor  used on  the  hypolimnion aerator.
            The hole site spacings  along  one arm are
            shown.   Three holes were  drilled at  each
            site as  shown in  the  cross-sectional view of
            one arm.

-------

         CO
                 •no— •ooo-—
                 n  if   >o i
distances  (cm) to holes

-------
                            90
unevenly spaced along each arm.  At each site, one hole was



drilled on the top of the pipe, while the other two were



drilled on opposite sides, 45° from the top.  The holes



were spaced in this manner to assure an even distribution



of air and to impart an upward thrust to the water.





Compressor.  A Jenbach air-cooled JW 156 diesel compressor



supplied air to the tower during the summer 1970.  The com-



pressor had three set speeds:  (a) 1500 rpm delivering 147



Cfm at 100 psi; (b) 1200 rpm delivering 125 cfm at 100 psi;



and  (c) 1000 rpm delivering 100 cfm at 100 psi.  The pres-



sure in the system was only about 28 psi.  An 8,000 liter



fuel tank was connected to the compressor.  The compressor



used less than 4 liters of diesel fuel per hour at 1500



rpm.



     During January 1971,  I used a Jaeger air compressor.



This has a maximum air output of 75 cfm at 100 psi.  It ran



at about two-thirds maximum capacity.  This is a very rough



estimate since we did not measure air output.





Operation.  I began aerating June 13, 1970, but only ran



the compressor for 10 minutes.  A great deal of water and



air leaked through the tower,  especially where the two sec-



tions were banded.  The tower was then raised so that the



inner tube was 0.5 m above the water level and we plugged



some of the holes with an epoxy material.  The epoxy hardened



overnight.  June 14th we  resumed air injection.  The tower



was kept at its elevated position and the compressor run at

-------
                            91
1000 rpm.   Water rose in the inner tube and cascaded over



one side of the tube.  The tower tilted at an 80  angle to



the lake surface.  We tried to correct the tilt with floats/



but without success .  Air rising in the tower displaces



water and decreases the specific density of the tower-water



system.  Water is also elevated 0.5 m above the lake level



within the tower.  The center of gravity is thus raised and



the tower tilts.  This condition could be corrected with



proper anchoring, and/or addition of ballast.



     Water and air continued to leak through the tower,



although less than before.  This was due in part to the tilt



of the tower,  but in larger part to improper construction of



the culvert tubes.  The tubes are constructed of galvanized



iron plates.  The plates were bent and riveted together.



Caulking should have been applied between the plates before



they were fastened,  but was not.



     June 17,  1970 the tower was lowered 0.5 meters and the



compressor run at 1500 rpm.  The compressor was always run



continuously,  except when shut down for maintenance or



repair.  The aerator was run in this manner until July 15th.



At this time it was lowered another 0.5 meters and run at



1000 rpm.   This schedule was maintained until September 7th,



at which time operation was discontinued.



     I injected air through the aeration tower for 48 con-



tinuous hours beginning January 22,  1971.  The tower was in



its lowermost position,  such that the inner tube was 0.5m

-------
                            92
below the lake's surface level.  Holes were chopped in the



ice inside the aerator to vent the air.  All the ice in the



tower melted quickly after aeration began.





Aeration Efficiencies.  We wished to determine the most



efficient means of aerating,  in terms of air input, as well



as the maximum aeration rate with our aerator.  To do this,



we had to determine the relationships between the operating



variables.  These variables include water flow rates through



the tower, air input to the tower, elevation of the tower



and oxygen absorption.  Water oxygen concentrations at the



top of the tower were always at or near saturation.  This



indicates the most efficient level of air input was below



our range of input, and therefore I will not include oxygen



absorption in the following analysis.



     We may consider water flow rate our dependent variable,



with air input and tower elevation our independent variables.



Tower elevation is the more important independent variable



over the range of air input values tested, since its influ-



ence on water flow rate is much greater (Figure 24).  At low



air input almost twice as much water flows through the tower



when the inner tube is 0.5 m below the lake's surface level,



compared to when it is 0.5 m above.  This difference dimin-



ishes at the higher air input levels, but the lower tower



elevation is still the most effective level.  On the average,



water flow rates are 1.3 times as great as high air input,



compared to low air input.  A more thorough analysis of the

-------
                            93
Figure 24.   Water flow rates through the hypolimnion aera-
            tion tower as a function of air input and tower
            level.   The (0)  level is when the top of the
            inner tube is level with the lake's surface.
            (+)  level is with the inner tube's top 0.5 m
            above the lake's surface,  and the (-)  level is
            with its top 0.5 m below the lake's surface.
            See text for discussion of true flow rates.

-------
 60-1
        Air Volume (m3/min.PZO kg/cm2)


          3.0         3.4        3.8         4.2
   504-
 x
 D
0)

£


0)
404-
   30 4-
             I    I   T
                                             4-69
                                             4-63
                                            "   1
                                                   "
                                             +45
                                             4-39
                                             4-33
                                                     (U
                                                     E
                                                     0)

                                                    I
         100
                    120
140
           Air Volume (c.f. m. @ 100 p. s. i.)

-------
                            95
flow characteristics should include a wider range of air



input.  We were limited in this analysis to only three levels



because we could not accurately determine air input other than



at the standard compressor speeds.



     The efficiency of air input can be measured by water



volume/day divided by air volume/min.  Comparing air input



efficiencies, we see that the efficiency decreases for the



 (0) and (-) tower levels as the air input is increased.  The



efficiency is relatively unchanged for (+) tower level,



however.  This indicates that more water is moved per unit



of air input at the lower air input levels.  Tne most effi-



cient level of air input is undoubtedly below our range of



values.  However,  as air input is decreased, the water oxygen



concentration may decrease below saturation.  If this occurs,



then the measure of efficiency must also account for oxygen



concentration as well as water flow rate.



     Our measured -water flow rates may be excessive.  The



current meter does not measure the direction of flow, only



the rate.   It measured horizontal as well as vertical flow.



The horizontal component may be very important,  since the



flow pattern was eccentric.  Because the tower tilted, air



tended to rise along the higher side of the tube.  In con-



trast, the return water flow was much greater on the lower



side than the upper at all three tower elevations.  As much




as a seven-fold difference in return flow rates was observed



between the upper and lower side.  This disparity is un-



doubtedly due to water flow around the tower as well as

-------
                            96
through it.  The horizontal flow was measured as vertical



flow and resulted in excessive flow-through estimates.



The measured flow rates are probably relative to each other,



however,  and therefore can be used to compare tower eleva-



tion and air input rate efficiencies.

-------
                         RESULTS





 Physical-Chemical Parameters



 Temperature and Oxygen.  Hemlock Lake stratified normally


 during 1969.  By early June the metalimnion extended from


 3  to 8 meters  (Figure 25), and the monimolimnion began at


 14 meters.  Temperatures during early June 1969 ranged from

   o                       o
 18 C at the surface to 4.5 C at the bottom.  During this


 same period, oxygen values averaged 8.0 mg/1 at the surface


 and 0.0 mg/1 at the bottom (Figure 26).  Metalimnion oxygen


 maxima of as much as 15.0 mg/1 were often observed through-


 out the summer.  These oxygen maxima were caused by high


 photosynthetic efficiencies and reduced mixing and diffusion


 rates.  Chlorophyll concentrations were not very great in


 either the epilimnion or metalimnion  (Figure 27),  indicating


 the oxygen maxima were not caused by algal concentrations.


 As the season progressed, the metalimnion depth increased


 slightly.  By late August it extended from 4 to 10 meters.


 The 7 C through 15 C isotherms have a nearly parallel decline


 throughout the summer.   Maxima surface temperatures of over


 25 C were observed during mid-July 1969 and again during


mid-August.   Minimum temperatures were nearly constant at


4.5 C.   The oxygen depletion depth was nearly constant at 10.5


meters during the entire summer 1969  (Figure 28).



                            97

-------
Figure 25. Hemlock Lake's isotherms during the summer 1969, before
           aeration_began.  Isotherms are in  C.

-------
  July
10    20
  August
10     20      1

-------
Figure 26.  Hemlock Lake's top, bottom and average oxygen concentrations             ,_,
            during the summers 1969 and 1970.  Continuous aeration occurred          o
            between June 14 and September 7, 1970.                                   °

-------
 12-
  9-
O)
£
  6-
C
0)
O)
  0-
      1969
                             BOTTOM
                 1970
                                                          r
                                -—\
   1   10  20   1   10   20
      June       July
1   10  20
  August
11   10  20
     June
                                                        I  '  I ^  T  ' I
1   10  20
   July
1   10  20   1
  August

-------
                            102
Figure 27.   Hemlock Lake's total chlorophyll A,  phaophytin
            A,  oxygen and temperature profiles during
            August 13,  1969.   These are representative of
            values before aeration began.

-------
          Chlorophyll (mg/m3)
        20
        I
40
 i
60
80
           TemperaturefC)

         6       12        18
                   24
            Oxygen (mg/l)
 0    2    46    8    10   12    14
    i  i  .  i   ,  I  i  i  i  i  i   i  i  i
2-
4-n*ot.Chl. A
                                It

-------
                            104
Figure 28.   Hemlock Lake selected oxygen profiles during
            the summers 1969 and 1970.   Continuous aera-
            tion occurred between June  14 and September
            1,  1970.

-------
                    Oxygen  (mg/|)
  o-
          .  i .  .
8
I
 12
.  I .  .
             VI-18/
                  *"*. \VI-13
  5—
  10-
                      1970
                      1969
  15-
/•^


£ 0-
a
0)
a   _
                             VIM
   5-
           VI I- 30,
                    IX-2
  10-,'
                       VII-29
                                              IX-2
   15-

-------
                            106
     Before aeration began during June 1970, the thermal



regime was similar to that of June 1969 (Figure 29).



Temperatures ranged from 21°C at the surface to 4.5 C at



the bottom (Figure 30).   Although the surface temperature



was warmer during June 1970,  the thermal profile was other-



wise similar,  Oxygeri values before aeration began during



1970 were likewise very similar to the 1969 values.  Oxygen



depletion began at 12 meters, and a maxima occurred within



the metalimnion.



     Continuous artificial hypolimnion aeration began June



14, 1970 and caused significant alterations of the physical-



chemical regime.  A tongue of 1.0 to 4.0 mg/1 oxygen extended



from the aeration into the hypolimnion after one day of con-



tinuous aeration  (Figure 31).  After nine days of aeration,



much of the hypolimnion had over 8 mg/1 oxygen (Figure 32) .



Shortly thereafter the entire hypolimnion had more than 10



mg/1 oxygen.



     Bottom oxygen concentrations at maximum depth increased



from 0.0 mg/1 to 9 mg/1 during the first week of aeration,



and remained above 10 mg/1 most of the summer (Figures 26  and



28).   Oxygen maxima still occurred within the metalimnion,



but they were not as distinct as during pre-aeration periods.



Average oxygen concentrations were always greater during



aeration,  and surface values were generally greater.  Average



oxygen concentrations during mid-July 1970 were 12 mg/1 com-



pared to 8.5  mg/1 during mid-July 1969 (Figure 26).

-------
Figure 29.  Hemlock Lake's isotherms during the summer 1970.   Continuous
            aeration occurred between June 14th and September 7th.
            Isotherms are in °C.

-------
                                               August
a
Q>
a

-------
                            109
Figure 30.   Hemlock Lake selected temperature profiles
            during the summers 1969 and 1970.  Continu-
            ous aeration occurred between June 14 and
            September 7,  1970.

-------
                     Temperature (°c)
Q.
0)
Q
    16-
    0-
    4-
    8-
    12-
    16-
             7-30-69
                  '7-29-70
                                        9-2-69
                                                   9-2-70

-------
Figure 31.  Hemlock Lake hypolimnetic oxygen isopleths (mg/1)  along the
            air line transect one day before aeration began and after one
            day of hypolimnion aeration.

-------
70
60   50   40   30    20   10
10    20    30  40   50   60   70
                          Distance From Aerator (m)

-------
Figure 32.  Hemlock Lake hypolimnetic oxygen isopleths  (mg/1) along the                   i-
            air line transect one day before aeration began, and after nine
            days of continuous hypolimnion aeration.

-------
70
60   50   40    30    20    10
10    20    30   40   50   60   70
                          Distance From Aerator (m)

-------
                           115
     Another conspicuous alteration of Hemlock's physical-



chemical regime is the gradual warming of the hypolimnetic



waters.  Their temperature increased more than 2 C/week



 (Figures 29 and 33).   This increase is marked by the gradual



and continued extinction of 18 C or less isotherms,  "into



the bottom of the lake."  The minimum temperature increased



greatly to over 17°C during September 1970,  compared to 4.5°C



the previous September.  The average temperature increased



to a maximum of 19.5°C during August 1970,  compared to 17.5°C



summer maximum during August 1969.  Surface temperatures were



slightly cooler during July 1970,  and several degrees cooler



during late August 1970 compared with 1969.   The hypolimnetic



warming is attributed to heat conductions throughout the



tower, whereas the surface cooling is attributed to mixing



of hypolimnetic waters with surface waters.   This led to a



gradual destratification of Hemlock Lake in disaccordance



with my experimental design.  Although the lake did destratify



almost completely by September 1970,  I did maintain a distinct



thermal gradient during most of the summer.





Thermal Stability.  Thermal stability increased from 7x108



kg-m during June 1969 to 18x108 kg-m during July and August.



These two peaks correspond to surface temperature maxima



(Figure 34).   The increased stability reflects the general



surface warming trend,  increased epilimnetic volume and



thermal gradient maintenance during 1969.  Thermal stability



followed a much different pattern during 1970.  It increased

-------
Figure 33.  Hemlock Lake maximum,  minimum and average temperatures (  C)
            during the summers 1969 and 1970.  Continuous aeration
            occurred between June 14 and September 7,  1970.

-------
                                Hemlock
4-
  1   10  20  1   10  20   1   10  20   1]  10  20   1   10  20   1  10  20   1
     June      July     August         June      July     August

-------
                            118
Figure 34.   Hemlock Lake stability values during the
            summers 1969 and 1970.  Continuous aera-
            tion occurred between June 14 and September
            7,  1970.

-------
    20-
 0>
•s

 i


 2
.•=  10-.
=   5-
ja
 (T3
        11 ii 11 ii ii |in i nil i [i i iii71 riTfTTTii i ii 11 ii [rmrjnrrnil riT|M 1111 up IITI UN p u MI n ni| n 111

         1     10    20     1     10     20      1     10    20       1
              June
July
August

-------
                            120
from 8x108 kg-rn during early June 1970 to a 14.5x108 kg-m



maximum during mid-June.  Thereafter it decreased gradually



to almost zero by September 1970.  This gradual decrease in



stability reflects the gradual destratification of the lake.





pH, Alkalinity, and Conductivity.  During 1969, pH values



were consistently low within the monimolimnion, but about



8.0 near the surface (Figure 35) .  The low monimolimnetic



values are attributed to high carbon dioxide concentrations



associated with decomposition of seston and sediments.



After aeration began during 1970, the bottom pH values in-



creased and eventually equaled the surface values.  Surface



values were greater during early 1970 than during 1969, but



gradually declined.  The high surface pH values are attributed



to photosynthesis, whereas the increases in bottom pH values



are attributed to carbon dioxide elimination.  Carbon dioxide



was removed from the bottom water as it passed through the



aerator tower.



     Surface and bottom total alkalinity values differed



greatly during 1969.  Surface values were relatively constant



at 120 mg/1,  but bottom values were about 500 mg/1 (Figure



36).   Aeration caused a marked decrease in bottom alkalinity,



but a gradual increase in surface alkalinity.  The average



alkalinity was relatively constant and about the same as



during 1969.




     Conductivity followed the same general pattern as total



alkalinity (Figure 37).   Bottom conductivity during 1969

-------
Figure 35.  Hemlock Lake's bottom, top and average pH values during the
            summers 1969 and 1970.  Continuous aeration occurred between
            June 14 and September 7,  1970.

-------
 8.5-
 8.0-
7.0-
6.5-
6.0-
        1969
  TOP

    **»,.
      "x

  MEAN
                     BOTTOM .
                            1970
      June
   1   10  20   1    10   20    1    10  20    1
                           10  20   1    10  20    1    10   20    1
July
August
June
July
August

-------
Figure 36.  Hemlock Lake's bottom, top and average alkalinity values
            during the summers 1969 and 1970.  Continuous aeration
            occurred between June 14 and September 7, 1970.

-------
   500-
   400-
   300-
O 200-
I-
   100-
          1969
                          BOTTOM
                          MEAN
                          TOP
                                           1970
                                               ~~l—'—\—
                                                10   20
                                                June
                                                        ~i—'—r
1   10   20
   June
10   20    1    10   20
July       August
i  i
1   10  20
   July
~i—'—r
 1   10
     20
August

-------
Figure 37.  Hemlock Lake's bottom, top and average conductivity values
            during the summers 1969 and 1970.  Continuous aeration
            occurred between June 14 and September 7,  1970.

-------

1H""Oo-
o
_c
£
0
^ _
*•
o
£800-
>»
«w
^^^
'>
+<
u
3 500-
"O
e
O-

Ofift
xVJU
1

1969
.•
BOTTOM .

-
AAPAKI _—•—•"""'' -
IVlCAM" 	 -.^^^^^ ^
TOP 	 	 --'''
' i ' i ' i • i ' i ' i i i ' i i i
10 20 1 10 20 1 10 20 1
June July August
1970





-^^^ | * * " * ' 	
x* **" "" "^ 	
"* **-* J — "
' 1 ' 1 ' I 1 f " 1 1 1 I 	 1 	 1 	 1 	 1 1 	 1 	 '
10 20 1 10 20 1 10 2b i
June July August

-------
                            127
averaged 950 micromhos compared to 250 at the surface.



Following aeration during 1970, the bottom conductivity



decreased rapidly to 400 micromhos and then gradually con-



verged with surface conductivity.  When the lake destrati-



fied,  the surface and bottom conductivity were essentially



equal.





Phosphorus.  Total phosphorus concentrations during July



1969 were ten times as great in the monimolimnion as in the



holomixion (Table 3).  They ranged from 0.020 mg/1 at the



surface to 0.280 mg/1 at the bottom.  Total dissolved phos-



phorus did not follow this same pattern.  It was more vari-



able,  but about twice as great below the metalimnion.



Surface to 3.7 meter samples averaged .011 mg/1 whereas



those below 10 meters averaged .021 mg/1.  Phosphorus deter-



minations were not made during 1970, but changes in their



concentrations can be inferred.  Oxygen depletion leads to



low redox potentials in the mud and a net movement of phos-



phorus from the mud to the water (Mortimer,  1941, 1942).



This situation was evident before aeration began.  After



aeration began the redox potential increased rapidly as the



oxygen increased.   This caused a re-absorption of phosphorus



by the aerated mud (Fitzgerald, 1970)  and precipitation of



phosphorus.   The net concentration of phosphorus should thus



decrease.   Subsequent accelerated biodegradation of the



sediments may have caused a regeneration of phosphorus.   The



net result of the processes is speculative.

-------
Table 3.  Hemlock Lake total phosphorus and total dissolved
          phosphorus collected July 22, 1969.  Two water
          samples were collected from each depth.

Total Phosphorus
(mR/1)
Depth
0.0
1.9
3;7
5.6
7.4
9.2
11.0
12.8
14.6
Xl
0.020
0*015
0.025
0.008
0.008
0.040
0.025
0.020
0.300
X2
0.020
0.020
0.015
0.010
0.020
0.040
0.030
0.030
0.250
Mean
0.020
0.018
0.020
0.009
0.014
0.040
0.028
0.025
0.280
Total Dissolved
Phosphorus (rng/1)
xl
0.010
0.008
0.020
0.002
0.005
0.008
0.025
0.020
0.020
X2
0.015
0.008
0.008
0.005
0.020
0.040
0.015
0.025
0.020
Mean
0.012
0.008
0.014
0.004
0.012
0.024
0.020
0.023
0.020

-------
                            129
Ca, Nat K, Mq, DOM and POM.  These constituents were measured



only during the summer 1970.  The average concentrations of



sodium and magnesium were relatively constant during the



summer (Table A-ll).  Particulate organic matter was especial-



ly variable.  Calcium, potassium and dissolved organic matter



had intermediate variability.







Primary Production





Phytoplankton.  Relative to 1970, Hemlock's phytoplankton



parameters were relatively stable during 1969*.  Secchi disc



measurements during 1969 ranged between 2.5 and 6.5 m  (Figure



38) .  Surface phytoplankton densities were always less than



4,000 cells/ml,  and surface 14C production estimates were



always less than 50 mg C/m3/4 hrs.



     A small,  unidentified bacterium was very abundant within



the monimolimnion during the summer 1969.  Brian Moss found



that its density often exceeded 300,000 cells/ml.  Water from



this region had a distinct green tinge, although the bacter-



ium was pink when preserved.  Gallionella was occasionally



collected from 15 m while not found at 0 m or 5 m.  Other



phytoplankters were infrequently collected at 15 m,  but were



presumably produced at shallower depths.



     After aeration began during June 1970,  dramatic changes



occurred.   Secchi disc measurements decreased to 1.7 m after



a week of  aeration.   Concurrently,  phytoplankton densities



increased  to over 30,000  cells/ml.   Primary production also

-------
                            130
Figure 38.   Hemlock Lake secchi disc transparencies, surface
            primary production potentials,  surface phyto-
            plankton densities and surface production
            efficiencies during the summers 1969 and 1970.
            Continuous aeration occurred between June 14 and
            September 7,  1970.

-------
    0-
    2-
H   6-1
 0)      '
U)
    8—
 2 300-
 E 200-
U"
 o>
 E 100-
    0-

   30

1
> 20-
o
^   10
    0-
   45
VI
_£~30
    15 H
gs
    o-
                      1970
      1970
                             1969
           1969 --
1970
           10    20
          June
         i
        10
       July
20     1     10    20

          August

-------
                            132
increased, but did not reach its maximum of almost 400 mgC/



m3/4 hrs. until mid-July, at which time phytoplankton densi-



ties had declined.  Following the tremendous increase in



phytoplankton production, standing crops declined to an all



time low by the first week of August.  This is exemplified by



the Secchi disc measurement of over 9 meters.  This is the



deepest Secchi measurement ever recorded for Hemlock Lake.



Surface phytoplankton standing crop and 14C production



approached zero during the first week of August.



     Following this phytoplankton decline,  there was a gradual



increase that continued throughout August.   Secchi disc



measurements declined to about 2 meters and 14C estimates



increased to over 300 mgC/m3/4 hrs. by September.  Phytoplank-



ton densities also increased to about 10,000 cells/ml,  which



is about one-third their June maxima.



     The bacterium that was so abundant in the monimolimnion



during 1969,  disappeared soon after aeration began during



June 1970.  Several phytoplankton species were found in the



hypolimnion during the summer 1970, but these were typically



more abundant at shallower depths and were probably produced



at shallower depths.   Densities near the bottom, at 16.5 m



followed the same seasonal pattern as found at the surface.



Densities were greater than 3,000/ml on June 26th, July 2nd



and July 9th,  but declined to only I/ml on July 31st.




Densities gradually increased thereafter to 1,700/ml by



September.

-------
                            133
     Associated with changes in the plankton, an unusual



event occurred.  Beginning about August 1,  1970, voluminous



quantities of foam were generated in the aerator.  This



foam was tan-colored and had a musty odor.   Algal cells and



Daphnia ephypia were mixed with it.  It was sticky, but



easily dissolved in water.  When dried, it was a dark green



or black color, presumably due to entrapped chlorophyll.



Large quantities were generated during the night and spilled



over the top of the aerator (Figure 39).   Large amounts of



foam floated about the lakes'  surface early in the morning,



but were gone by mid-day.  The foam was apparently "melted"



by the sunlight and/or increased temperatures.  From midday



on, it completely filled the tower, above the water level, but



did not cascade over the rim.   It again resumed overflow



starting sometime during the night.



     The foaming ceased from about August 15th through August



18th.  This coincides with a dip in the surface Secchi and



14C measurements.  Foaming started again August 19th and



continued until we terminated operations.  This second time



foaming began coincides with an upturn in 14C production at



the surface and secehi decreases.





Periphvton.  Average periphyton standing crop during 1969,



based on 17-day incubations,  ranged between 0.004 and 0.020



gm/day (Figure 40).   Minimum values occurred during June and



maximum during late July and August.  Periphyton accumula-



tions in Hemlock were most abundant between the surface and

-------
Figure 39.  Foam spilling over the top of the hypolimnion aerator during
            August 1970.  (Photo by author.)

-------

-------
                            136
Figure 40.   Hemlock Lake periphyton standing crops based
            on 17-day incubation periods and continuous
            incubation.   The 95% confidence interval is
            shown about  each average value.  Continuous
            aeration occurred between June 14 and September
            7,  1970.

-------
   .04-
 0)

£  .03-
. i
 « o>
a- -
i-S
    .01-
   .00-


    1.2-
 V)
    0.9-

•c .g>0.6-
 3


 I   Q3-
   0.0—
          Hemlock Lake

           • 1970

           01969
1
            T
            10
    20


June
I  '   I-  '	T
1     10     20
                              July
                        August

-------
                            138
2 meters both years.  Five meter slides often contained many



Hydra, especially during June and early July 1970.



     Periphyton standing crop during early June 1970 was



significantly less than during the same period 1969.  After



aeration began during 1970, periphyton standing crops in-



creased to over 0.030 gm/day during July.  It then declined



to 0.016 gm/day during early August,  and increased a second



time to 0.025 gm/day by late August.   These changes closely



approximate changes in the phytoplankton, both in changes



of relative abundance, as well as times of maxima and minima.



     Total accumulative periphyton standing crop increased



throughout the summer both years.  These increases were al-



most linear between June and September (Figure 40).  An



average maximum value of over 0.7 gm was attained by August



21, 1969.  Although accumulated periphyton was less during



June 1970 compared to June 1969,  it was significantly greater



during the remainder of the summer 1970.   Total accumulated



periphyton reached a 1»2 gm maxima by August 21,  1970.  This



was almost twice the 1969 maximum value.








Zooplankton






     Only two sets of zooplankton samples were collected.



One set was collected just before aeration began,  June 11,



1970.   The other set was collected after one month of aera-



tion.   Both sets were collected during daylight hours.

-------
                            139
     Before aeration began, zooplankters were mostly limited



to depths above the oxygen depletion  (Figure 41).  Diaptomas



adults, Bosmina,  and Diaphanosoma were most abundant within



or below the metalimnion.  Over 85% of Diaptomas nauplii



were found within one meter of the surface, and Daphnia pulex



were scattered throughout the water column.  After one month



of aeration,  the depth distributions of Diaptomas adults,



Diaphanosoma and D. pulex were much changed.  Diaptomas adults



extended their maximum depth from 9 m to 18.6 m and their



average depth from 4.5 m to 6.0 m.  Diaphanosoma were most



abundant after aeration near the surface and just off the



bottom.  Their depth distribution was much altered by aera-



tion, but their average depths were about the same.  D. pulex"



increased their average depth from 4.6 m before to 13.8 m



after aeration.  Over 80% of the Diaptomas nauplii were still



found in the upper meter after aeration,  and Bosmina had



essentially the same depth distribution as before aeration.



     More astonishing than the changes in depth distribution



were the changes in population numbers.  All zooplankters



increased significantly (0.05 level)  except Diaptomas



nauplii (Table 4).   Diaptomas nauplii decreased from a total



of 71 to 13.   Increases in E>. pulex,  Diaphanosoma and Bosmina



were 88x,  21x and 3.Ix respectively-   No further information




is available concerning the seasonal patterns of these zoo-



plankters,  other  than for D.  pulex.   D. pulex became an



important rainbow trout food item during the first week of



July 1970 and remained so throughout the summer.  It was also

-------
Figure 41.  Hemlock Lake zooplankton depth distributions three days before
            aeration began, and after one month of aeration._ Oxygen and
            temperature profiles are shown for each date.   (x = average
            depth.)

-------
        Temp. (°C)     Oxygen
                                             Percent  of   Numbers

                              Diaptomas  spp. Adults    Diaptomas  spp.  nauplii
      5  10   15  20  0
10   15
1111 11111
4   8   12   45  30   15   0    15   30  45  75  50  25   0  25  50   75
I I I I I I I I ll I I  I I I  I l  I I  I I   I  I I  I I  I I  I 1  I II I I I I I I II I I I I M I I I I I I I I I I i I i i i I I I
a.
a>
a
                                                                                  7-15-70

                                                                                  x-0.8
                              6-11-70

                              x=4.5
                                            6-11-70

                                            x=0.6
                                      6-11-70

                                      x-8.0
                                            7-15-70

                                            x=78
                                            6-11-70

                                            x=4.6
   7-15-70

1  x=13.8
      45   30  15
               15   30   45    45  30  15
                             15  30  45    45  30   15    0   1
               Bosmina spp.              Diaphanosoma spp.            Daphnia pulex

                                                      Percent  of   Numbers

-------
Table
Hemlock zooplankton collected June 11 and July 15, 1970.  Three
samples were collected from each two meters depth interval.
Totals represent the sum of the average number of zooplankters
per liter from each depth.  Total samples on each date = 27.

June 11, 1970
Organism
Diaptomas spp.
acTults
Diaptomas spp.
nauplii
Bosmina spp.
Diaphanosoma spp.
Daphnia pulex
Total
162.
71.
1,366.
7.
2.
951 C.I.
156.
61.
1,254.
6.
1.6
on Total
168.
81.
1,478.
8.
2.4
Total
193.
13.
4,228.
157.
176.
July 15, 1970
95% C.I.
188.
11.
4,100.
143.
160.
on Total
198.
15.
4,456.
171.
192.

-------
                            143
commonly collected in the emergent insect traps after the



first of July-   These observations are not quantitative,  but



give some idea  of D.  pulex relative abundance during the



summer.   It seemed to be most abundant about mid-July-  On



July 28th/  over 75% of the individuals carried ephypia.



This indicates  adverse conditions, such as those associated



with the large  decline in the phytoplankton population at



that time.   Ephypia were not common thereafter,  and D. pulex



seemed moderately abundant the remainder of the summer.







Zoobenthos



     The three  most important zoobenthos taxa in Hemlock,



not including craybish,  are the Chironomidae,  Anisoptera and



the Chaoborinae (Table 5 and Figure 42).  Together they com-



prise more than 75% of the biomass and more than 95% by



numbers of the  benthic macro-organisms.   Seven other taxa



comprise the remainder.   Chironomids are numerically the



most abundant/  but Anisoptera have the largest biomass.



These estimates of relative abundance are based on static



measures,  namely standing crop.  If production rates were



known,  the relative composition of these groups may be much



different.



     At least 12 species of Chironomid midges were identi-



fied by D.  R. Oliver  (Canada Dept. of Agriculture),  from



emergent adult  specimens (Table 6) .   Only four species were



relatively  abundant in the emergent samples:  Procladinus



denticulatus. Tanypus,  Tanytarsus, and Dicrotendipes.

-------
Table 5.
Hemlock Lake zoohenthos collected during the summers 196? and 1970 with an Fkman dredge.
taken each summer.  Wet weichts arc shown.
                                                                                                    125 drectce samples were
Total
1969

Oligochaetes
(microdriles)
Chironomid L.
Chironomid P.
Amphipods
Dragonf lies
Damself lies
Mayflies
Chaoborus spp.
L.
Chaoborus spp.
P.
Heleidae
Trichoptera
Tabaniid
Leeches
Total
Grams
0.00347
0.32719
0.03928
0.00044
0.83401
0.01013
0.04548
0.040378
0.13232
0.05514
0.01608
0.00000
0.54629
2.41360
Percent
0.1
13.6
1.6
0.0
34.6
0.4
1.9
16.7
5.5
2.3
0.7
0.0
22.6
100.0
Weight
1970
Grams
0.13473
0.49652
0.02301
0.00069
0.97448
0.00380
0.02483
0.31419
0.04911
0.02370
0.00008
0.00024
0.02961
2.07497
Percent
6.5
23.9
1.1
0.0
44.7
0.2
1.2
15.1
2.4
1.1
0.0
0.0
1.4
100.0
Total
1969
Number
4
1,459
32
3
17
IS
115
535
91
96
28
0
1
2,396.
Percent
0.2
60.9
1.3
0.1
0.7
0.6
4.8
22.3
3.8
4.0
1.2
0.0
0.0
100.0
Numbers
1970
Number
267
2,401
80
3
14
3
110
1,336
77
96
1
1
1
4,390.
Percent
6.1
54.7
1.8
n.i
0.3
0.1
2.5
30.4
1.8
2.2
0.0
0.0
0.0
100.0
Number of
Samples Found In
1969
4
72
19
3
12
3
18
78
41
25
1
0
1

1970
33
98
32
3
8
3
26
88
39
32
1
1
1

Mean Number of
Individuals /gram
1969
1,153
4,459
815
6,818
20
1,481
2,529
1,325
688
1,741
1,741
--
2

1970
1,982
4,836
3,477
4,348
14
789
4,430
4,252
1,568
4,051
12,500
4,167
34


-------
                            145
Figure 42.   Hemlock Lake zoobenthos  percent  composition
            during the summers  1969  and  1970.   Percent
            of wet weight and percent  of number are
            shown for  each taxa.   Total  weights and  total
            numbers collected each summer are  also shown.
            Samples from dredge collections  only-

-------
   70-
   60-
 fl)
JJ 50-
 E
 3

Z 40H
 x
CO


 c 30-
 o>
 u
0.  20-
    10-
   60



   50-
 O)
 x
CO


 c 30-
 4)
 u

 0)
o. 20-
    10-
    0-
              o.
              o
              >o
                  K
                               Total Number: 1969=2,396

                                            1970 = 4,390


                                            I	11969
                                o
                                co
                             CO

                             CN
                             CN
                              Total Weight: 1969=2.41360

                                           1970=2.07497


                                           I11969

                                           r*^^^n 107^
                o

                co
               •o
>q
•O
                                                      CN

                                                      CN
                             o  — .
                             •-  10
                                  CX  CN
                                               >o
       0*  >




-------
Table  6 .   Emergent midge adults collected from 600 samples during 1P69,  and 650 samples during 1970.
        *    are from Hemlock Lake and were collected in emergent insect traps.
All specimens
1969

Family - Chironomidae (Tendipedidae)
Subfamily - Pelopiinae (Tanypodinae)
Procladius (Psilotanypus) bellus
P., (s.s.) denticulatus
Tanypus spp. I/
Subfamily - Chironominae (Tendip«dinae)
Endochironomus spp. I/
Zavrelia spp.
Chironomus spp. 2/
C. (s.s.) staegeri
Polypedilum (Pentapedilum) sordens
Cladotanytarsus viridiventris
Tanytarsus spp.
Dicrotendipes spp.
Paratendipes albimanus
Family - Culicidae
Subfamily - Chaoborinae
Chaoborus flavicans
C. punctipennis
Totals
Total
Number


27
76
43

0
2
1
0
4
2
117
105
6

47
0
4.30
Percent
of Total
Number


6
18
10

0
1
1
0
1
1
27
24
1

11
0
100
No. of
Samples


20
34
26

0
2
1
0
4
2
53
42
6

35
0
225
No. of
Dates


16
26
19

0
2
1
0
4
2
37
31
6

21
0

Total
Number


30
261
86

2
4
7
1
44
0
86
231
2

84
212
1,050
1970
Percent
of Total
Number


3
25
8

1
1
1
1
4
0
8
22
1

8
20
100
No. of
Samples


21
80
S7

2
3
6
1
14
0
32
37
2

51
57
363
No. of
Dates


19
48
35

2
3
5
1
11
0
27
24
2

29
22


I/ Probably a new species
2/ C. tentans identified from larvae

only















-------
                            148
Procladius and Tanypus are generally thought to be pre-



daceous while the other two genera are generally thought to



be omnivorous, feeding mostly on plant material and



detritus.  One species each of Tanypus and Endochironomus



may be new species.  Positive identification of many of the



species was impossible because of the small sample sizes



with few males.  Only 430 adult midges were collected during



1969, and 1,050 during 1970-  These values are very low con-



sidering 1,250 emergent trap sets were made.  Only 225 of



600 samples contained any emergent midges during 1969, and



363 of 650 samples contained emergent midges during 1970.



     The relative abundance of the emergent midge species



may not reflect their actual relative abundance in the lake.



Many species have peak emergence in May  (Miller, 1941),  but



we did not begin our collections until mid-June.  Chironomus



tentans was collected as larvae,  but not as adults.  This



species appeared relatively abundant in the hypolimnion after



aeration.



     During 1969,  most of the Chironomid larvae were re-



stricted to 9 meters or less by hypolimnion stagnation



(Figures 43 and 44).  Less than five percent were found be-



tween 9 and 15 meters,  and none were ever collected between



15 meters and maximum depth.  Larvae living in deep water



were generally larger than the shallow water larvae.  For



example,  on September 6,  1969 five percent of the larvae by



number were found between 9 and 15 meters, but 15 percent by



weight were in this interval.

-------
                            149
Figure 43.   Hemlock Lake Chironomid larvae depth distribu-
            tion as percent of number during each sampling
            period during the summers 1969 and 1970.
            Shaded histograms represent aerated periods.

-------
o-
60 -:
40-
20 —
0-
60 -_
0 40 —
.Q :
3 20-;
z :
^_
o :
^60-
C
0
O 40 —
LM —
m
°- 20 "
jg -
CO
5 -:
-J 60-
."5 :
*E 40-
o :
c
O 20— '
.E "
^> 0
60-
40-^
20-^
o
c















6rl3-69
x=5.0


7-
x=


4-69
3.0

7- 25-69
x- 3.1

8-5-69
lf=4.7





9
jf=

III I 1 1 [ 1 I
) 4 8

•6-69
4.1



S'"- '

- '



v1- *"\ "
~A









-
6-12-70
x= 5.3

7-3-70
x=3. 2
7-24-70
x=3.1
.* .™"i
-------
                            151
Figure 44.   Hemlock Lake Chironomid larvae depth distribu-
            tion as percent of wet weight during each
            sampling period during the summers 1969 and 1970
            Shaded histograms  represent aerated periods.

-------
 o-
60-:
40-
    0-

   60-^

+- 40-
 O)

|2°
M-  °~

H- 6°~
 0)
 O 40-
   20 —
8
a
    0-
   60
•D
'E 40 H
 O
 o 2°-
!S    I
O  0-
   60 -
   40-
  20-
   O-l
                      6-13-69
                      x-5.4
                     7-4-69
                     *= 3.9
                      7- 25 -69
                      8-5-69
                      x=6.7
                            1
                    9-6-69
                    x"=7.0
                     T~T
                        12
                           I I  I I  I I
                              16
                                                   6-12-70
                                                   1(=6.0
                                                 7-3-70
                                                •Jf-5.6
                                                7- 24-70
                                                T?=4.8
                                                 8-4-70
                                                 x= 9. 1
                                                9-4-70
'<'"'  I '  I 'I
   4     8


                                                     12
                                                         16
                          Depth  (m)

-------
                            153
     Although substantial numbers and biomass of Chironomid



larvae were present below 7 meters during 1969, very few



emerged from these depths.  Pupae were collected below 7



meters only during August 5, 1969, and then only between



7.5 and 11 meters (Figures 45 and 46).  Emergent adults were



most abundant above 4 meters depth and were collected below



8 meters only during late July (Figure 47).   Emergence in



the 0 to 4 meter interval ranged between 70% and 95%.



Relatively more midges emerged from 0 to 4 meters during



June than during August.  Miller {1941)  reports this same



situation in Costello Lake, Ontario,  Canada.



     Total emergence during 1969 was generally less than



3. x 10s individuals/week  (Figure 48).  Emergence does not



have any obvious pattern,  due in part to its composite nature.



The species composition changes from week to week,  but



Tanytarsus and Dicrotendipes were generally most abundant.



These two midges comprised 51% of the total emergence during"



1969.



     Ghironomid larvae gradually extended their depth dis-



tribution after aeration began June 14,  1970 (Figures 43 and



44) .   Less than 1% by number or 2% by weight were found below



11 meters on July 3rd.   By September 4th more than 15% by



number and 40% by weight were found below 11 meters.



Chironomid larvae were first taken from dredge samples near



maximum depth on August 4th.  Larvae were observed in the



emergent trap samples before this date,  however.  By September



4th 5% by number and 15% by weight were present between

-------
                            154
Figure 45.   Hemlock Lake Chironomid pupae depth distribu-
            tion as percent of number during each sampling
            period during the summers 1969 and 1970.
            Shaded histograms represent aerated periods.

-------
••
75-:
so-:
25~.
0~
•
75 —
0) 50-
1 l
= 25-:
Z :
o
H- U -
o :
+* 75 —
g :
S501
°-2^

2 :
rt ft -
a ° -
* » J
•D :
E 50-
o -
P 25-
tm -
• 1^ ••
On '
u—
75 -_
fft
50 —
25-
.
O-—

easaBSB

—



















1 1 1






















• 1 *
6-13-69
if- 1.8
7-4-69
if. 3.0


7-25-69
y. 9 f>



8-5-69
x-5.0




1
9-6-69
x-3.9





I , . , 1 , . . 1 , .


1
S|l Hi
llti;; Si;-;


1










f|S|fi*:1flll
H


lM| ...njp,- |
6-12-70
"x-5.5
7-3-70
i<=2.9

fci6»j%4^^^si
7-24-70
v. 3 3



8-4-70
ic- 75


$&''••'?''''&•*£%'•,<

lmKM__
9-4-70
Tc- 5.9



';; '_.',?£ ,'//^,
f?m??i'
1 * i r j i i i i | i i i j i • j
12
16
8
12
16
    Depth (m)

-------
                            156
Figure 46.   Hemlock Lake Chironomid pupae depth distribu-
            tion as percent of wet weight during each
            sampling period during the summers 1969 and
            1970.   Shaded histograms represent aerated
            periods.

-------
75 -:
50 -_
25-
0~
75-
£ 50-
O) -
|25^
> j
u_ n
O ;
•£'»-
o> :
£ 50J
£ :
25 "

g =
Q. 0 "
3 ° -
Q. :
s75i
c :
§50-
1 2sl-
JZ
° o •
u—
75-
50 -;
25 -_
0 ~
° 1
C




















1 1 1
)




















1 1 1
4
6-13-69
X-1.8
7-4-69
x-2.7


7-25-69
x> 2.6





8-5-69
7-5.5



9-6-^9
•JT-3.3



| I i I | i i . | i i
8 12 16 C




",










r., 	 j



IK *''??:'
i" f 	 I"
)



.:f





'





' ,, '/




ri
4
6-12-70
X-5.4
7-3-70
x-6.3
* J ><
/• v
^^v,
*< <
7-24-70
x-- 3.5



T, """ 1,.,.,...., .„.,.,».

8-4-70
lf= 11.9
*
W" ' ; "", *"'!
-*:; • £l
9-4-70
"X-10.9
.,
' s ^
1 > ' .' i / ,^
j I 1 1 j I 1 . 1 I J
8 12 16
Depth (m)

-------
                            158
Figure 47.   Total midge emergence from Hemlock Lake by
            depths during the summers 1969 and 1970.
            Aeration occurred continuously between
            June 14 and September 7,  1970.  Totals include
            Chaoborinae and Chironomid midges from emerg-
            ence traps only.

-------
CO
h.
0)
-D

E
3

z
c
0)
o


£
80-
40-
0_
80-
40-

—
80-
40-

80-
40-

80-
_
A(\


-
80—

40-
-
0-
0







	 •,,











iiil
6-15-69
6-27-69
x=1.9
6-30-69
7-11-69
— f* f\
x=2.0

7-14-69
7 9*> f\Q
x=2.9

7-28-69
8 ft A9
x=2.6

8-11-69
8-22-69
x=3.0



8-25-69
9-6-69
X=2.9

| I i I r p r i i | i i •(
5 10 15 0



















iiil
6-15-70
6-26-70
x=2.1
6-29-70
7-10-70
T~ ^"t ^
X=3. 7
	 1 	 1
7-13-70
7-94-7O
x=3.2
i
7-27-70
8-7-70
x=3.0

8-10-70
8-21-70
x>3.7



8-24-70
9-5-70
x-3.5
1 	 1 	 ,
| T 1 1 1 | 1 1 1 1 | 1 1 1 '1
5 10 15
                         Depth (m)

-------
                            160
Figure 48.   Total estimated weekly midge emergence from
            Hemlock Lake during the summers 1969 and
            1970.  Totals include Chaoborinae and Chirono-
            mid midges from emergence traps only.
            Aeration occurred continuously between June 14
            and September 7,  1970.

-------
0)
O

0)

S>
0)


LU

0)
O)
•o
 0)

I
•o

-§
   12—
    9-
    6-
    3-

       _^  i	,	!	,	q—,	T—,	j—

      1    10    20     1     10    20
                                      T-T—i—i—i—'—H"
                                       1     10    20     1
         June
                         July
August

-------
                            162
14.5 m and maximum depth.  Total numbers of Chironomid  larvae



increased greatly during aeration.  An estimated 2.5x107



maximum number were present during 1970, compared to a



l.SxlO7 maximum for 1969 (Figure 49).  Biomass also increased



during aeration to a 44 gm maxima during July I960, compared



to a 38 gm maxima during 1969.  There was a 65% increase in



total number collected during 1970 compared to 1969, and a



52% increase in weight (Table 5).



     Chironomid emergence during 1970 also gradually extended



into deeper water at about the same rate as the larvae.



Pupal concentrations by numbers were always greatest above



7 meters, but much larger pupae were collected below 11



meters on August 4th and September 4th than in shallower water



(Figures 45 and 46) .  Total numbers of chironomid pupae were



much greater during 1970.  An estimated maxima of 12xl05 were



present during July 1970 compared to a 3x105 maxima for 1969



(Figure 50) .   Interestingly enough,  total biomass was essen-



tially the same both summers,  indicating the 1970 pupae were



smaller.  Total number of pupae collected during 1970 in-



creased 250% compared to 1969, but total weight decreased 41%



(Table 5).   Total emergent adults were always most abundant



between 0 and 4 meters (Figure 47).   More than 5% of the



total emergence between June 29th and July 10th occurred in



the 12 m to 16 m interval.   Emergence from maximum depth



occurred during July,  but was not abundant until mid-August.



Adults almost never emerged below 8 meters during 1969.

-------
                            163
Figure 49.   Total  estimated  Chironomid  larvae  number  and
            wet weight  in Hemlock  Lake  during  the  summers
            1969 and  1970.   One  standard  error is  shown
            about  each  estimate.   Aeration  occurred con-
            tinuously between June 14 and September 7,
            1970.   Totals from dredge samples  only-

-------
ts
o
 a, 4-
 o
72 3-
 E
 o
 c
 o
U
    2-
 3
z
 o>
 O
                                1970

                                1969
    6-
 O
 E
 o
 eo

 *  >
 o  4-
 o
 _i
 •o
    2-
          10    20


            June
10
~20~
  July
10    20


August

-------
                            165
Figure 50.   Total  estimated Chironomid pupae number  and wet
            weight in  Hemlock Lake during the summers  1969
            and 1970.   One  standard error is shown about
            each estimate.   Aeration occurred continuously
            between June  14 and September 7,  1970.   Totals
            from dredge samples only.

-------
0)
o
a.

r°
'i
 o
 ciO
u

M-
 O 5
_Q

 E
 3

Z 0
                                                      £ 1970


                                                         1969
o
   6-
 V)
 o

 E
 o
   4-
 a>
 o
 a
 E
 o
 c
 o
    2-
   0-
                                 -O
                           loT
                 10  '  20
           June
July
August

-------
                            167
     Procladius denticulatus nev:er emerged below 12 meters



during 1969 (Figure 51).   After aeration it emerged below



16 m in early July and was more numerous below 8 m during



late July and early August.  During 1969 its peak emergence



occurred during August,  with no individuals observed before



July 10th (Figure 52).   During 1970 its peak emergence



occurred about July 15th and individuals were observed all



summer.  Total individuals observed in the traps was 261 for



1970 compared to 76 for 1969.



     Tanypus never emerged below 4 m during 1969, but



emerged below 16 m during August 1970 (Figure 51).   Its emer-



gence pattern was rather uniform during 1969,  but during



1970 it reached peak values during late July and late August.



About twice as many Tanypus emerged during 1970 compared to



1969 (Figure 51) .



     Dicrotendipes modestus emerged from slightly deeper



depths after aeration began (Figure 51).  During 1969 its



emergence was limited to less than 8 m,  but during August



1970 it emerged below 8 m.  Almost twice as many emerged



during 1970,  compared to 1969 (Figure 52).



     Tanytarsus were almost entirely restricted between 0



and 4 m both years.   Only one emergent adult was collected



between 4 and 8 m during 1969,  and none were collected below



8 m.  During 1970 one emergent adult was collected between



4 and 8 m and one between 8 and 12 m.  No adults were col-



lected below 12 m during 1970.

-------
                            168
Figure 51.   Depth emergence of selected insects from
            Hemlock Lake during the summers 1969 and 1970.
            White areas during the sampling periods repre-
            sent no observed emergence.  A= Procladius
            denticulatus,  B= Tanypus,  C= Pierotendjpes.
            D= Mayflies (Ephemeroptera).  Aeration occurred
            continuously between June 14 and September 7,
            1970.  Totals from dredge samples only.

-------
90-
   1  10 20  1  10  20  1  10 20   11  10 20  1  10 20   1  10  20  1
      June     July    August       June      July     August
               1969                          1970
                     4-8
8-12   :yv:^'  12-16
16-18.6

-------
                            170
Figure 52.  Total estimated emergences from Hemlock Lake.
            Samples from emergence traps only.  Aeration
            occurred continuously between June 14 and
            September 7,  1970.

-------
a
a>
(A
I

E
3
Z
        Procladius denticulatus (]Q4
        Tanypus spp. (10)
         Dicrotendipes modestus (10)
         Mayflies Adults (104)
August
         June
July

-------
                            172
     Total weekly emergence was much greater during  1970



compared to 1969.  A maximum of over 12x105 individuals



emerged about August 1, 1970 compared to a 3x105 maximum



for the summer 1969.  One thousand fifty adults were col-



lected from the traps during 1970 compared to 430  for  1969.



The greatest percentage increase occurred with the Chaoborus.



Chaoborus comprised 11% of the total emergence during  1969,



but 28% during 1970 (Table 6).



     Two species of Chaoborinae were identified by B.  V.



Peterson and D. M. Wood (Canada Dept. of Agriculture).



Chaoborus flavicans was present both years, comprising 11%



of the emergence during 1969 and 8% during 1970.   C_. puncti-



pennis was not observed during 1969, but comprised 20% of the



emergence during 1970.  This large increase in emergent adults



during 1970 and species shift is reflected by an increase in



numbers of larvae and decreased average size.  Chaoborus



larvae collected by the dredge totaled 535 during  1969 and



1,336 during 1970 (Table 5).  Average size of those  larvae



decreased from 1,325/gm during 1969 to 4,252/gm during 1970.



This decrease is probably due to two factors:  increased



proportions of C_. punctipennis and increased population size.



C- punctipennis is smaller than C. flavicans.  More  than



twice as many C_. punctipennis adults were collected  during



1970 compared to C_.  f lavicans.  No C. punctipennis adults



were observed during 1969.




     Increased population size will result initially in larger



numbers of the smaller instars.  Instars three and four are

-------
                            173
essentially the only ones collected by the dredge.  Instars



three and four were usually the only ones collected in the



emergence traps,  but instars one and two were also occasion-



ally captured.  Chaoborus larvae collected by the dredge



reached a maximum estimated population size of 20x106 indi-



viduals during July 1970 (Figure 53).   Their 1969 maximum



was 6x106.  Maximum total biomass was only slightly greater



during 1970,  however.  Total numbers increased 250% during



1970 compared to 1969,  but total weight decreased 22%.



Chaoborus larvae comprised 22.3% of the macro-zoobenthos by



number during 1969 and 30.4% during 1970 (Figure 42).



However, the percent of Chaoborus larvae in the zoobenthos



biomass decreased from 16.7% during 1969 to 15.1% during



1970.



     Much lower Chaoborus larvae and pupae population sizes



were estimated from the emergent insect trap samples  (Figures



53, 54 and 55).   A maximum larval population size of 48xl05



was estimated from the traps during July 1970,  compared to



20x106 from the dredge samples.   By their very nature, the



emergent trap samples are not expected to quantitatively



sample the larval population.  Larvae that live in the mud



part of the day,  but do not migrate during the sampling



period will not enter the traps.  Work by Roth (1968) indi-



cates that some larvae do not migrate everyday-  The propor-



tion migrating may also be related to temperature (Stahl,



1966) .   Furthermore,  in lakes with pronounced oxygen deficits

-------
                            174
Figure 53.   Total estimated Chaoborus larvae number and
            wet weight in Hemlock Lake during the summers
            1969 and 1970.   One standard error is shown
            about each estimate.   Aeration occurred con-
            tinuously between June 14 and September 7,  1970,
            Totals from dredge samples only.

-------
   25-
•o
 O
 0>

 >
 1_
 o
 o
 _Q
 O
 O
 .C
 U
 J5
 E
20-
    15-
 10-
                                                         1970

                                                         1969
              Q--

                                      XV-
n»  5
"b
    A
 vi   *
 O
 E

 i   3

 0)
 o

 t   2
 o
     1-
  o
 .C
 u
            10
             ~20~
10
             June
                            July
10     20

 August

-------
                            176
Figure 54.  Total estimated Chaoborus pupae number and wet
            weight in Hemlock Lake during the summers 1969
            and 1970.  One standard error is shown about
            each estimate.  Aeration occurred continuously
            between June 14 and September 7,  1970.  Totals
            from dredge samples only.

-------
  20-
o
a
O
-Q
O
O
   15-
   10-
 o
 •-  5-
 0)
_a
 E
    0-
                                                            1970


                                                            1969
 E
 o>
 in

 S
 E
 o

CD
   40-
   30-
 0)
 o
 a
 D
a.
   20-
 3
 i_
 O

 0
 O


u
   10-
    0-
  -r,	,	,	-p-

1     10     20
            June
                                                       T
1     10    20


       July
                                                lo
     20


August

-------
                            178
Figure 55.   Total estimated emergences of Hemlock Lake
            Chaoborus flavicans and C. punctipennis dur-
            ing 1969 and 1970.   Total estimated larvae
            and pupae are also shown.  All samples are
            from emergence traps.   Aeration occurred con-
            tinuously between June 14 and September 17,
            1970.

-------
   18-
   12-
   18-
   12-
    6-
U
a>
c   o
 ^haoborus fqvicans (104)
        	1969

        	1970
        Chaoborus  punctipennis  (105)
O 40-



aJ 30
.Q


3 20
   10-
        Chaoborus  Larvae (105)
         Chaoborus Pupae (105)
    6-




    4-




    2-
    0-
I  '—I—I—r~
1     10    20
                       1—'	1	'	r-
                        1     10    20
         June
                     July
1     10    20


   August

-------
                            180
in the deep water,  all four instars may remain in the water



and not nestle in the mud during the day (Northcote, 1964;



Teraguchi and Northcote,  1966).   In other lakes,  first and



second instar are limnetic throughout the day, while the



third and fourth instars typically nestle in the profundal



sediments during the daylight hours.



     Even though estimates of larval abundance in the emer-



gence traps are much lower than dredge estimates, the esti-



mates of relative changes in population sizes are very



similar from the two methods (Figures 53 and 55).  Estimates



of Chaoborus pupal population size are not only much lower



from the emergence traps, but relative changes in pupal



population size are also dissimilar (Figures 54 and 55).



These data indicate that emergence traps and dredge samples



provide comparable estimates of relative changes in the



larval population,  but not the pupal population.   The reason



for this difference is not obvious.



     During 1969, Chaoborus larvae were never collected



above 4 m in either the emergence traps or by the dredge



(Figures 56,  57 and 58) .   They were most abundant in the 8



to 12 m depth interval and about 20% were generally present



in the 16 to 18.6 m interval.  Pupae were present in all



depths during 1969,  but were generally most abundant between



4 and 12 m (Figures 56 and 58) .   C. flavicans only emerged



from the 0-4 m interval during 1969 (Figure 56).



     Chaoborus depth distributions changed greatly following



aeration.   Larvae concentrations in the 4 to 8 m interval

-------
                            181
Figure 56.   Depth distribution of Chaoborus during the
            summers  1969 and 1970.   All samples were col-
            lected by emergence insect traps.   A= Chaoborus
            flavicans emergent adults,  B= C_.  punctipennis
            emergent adults,  C= Chaoborus larvae,  D=
            Chaoborus pupae.   White areas during emergence
            periods  represent no observed specimens.
            Aeration occurred continuously between June 14
            and September 7,  1970.

-------
9(H
0
  1 10 20  1   10 20   1  10 20   11  10  20  1  10  20  1  10 20   1
     June     July     August





               1969




     |$ 0-4    :^9 4-8    H
   June     July      August




             1970




8-12   §:^ 12-16        16- 18.6

-------
                            183
Figure 57.   Hemlock Lake  Chaoborus  larvae depth distribution
            as  percent  of number  during each sampling period
            during the  summers  1969 and 1970.   Shaded histo-
            grams  represent aerated periods.

-------
75 i
50-
25 -:

75^
o> 5°^
.Q :
£ 25-^
ft ~"
u
H-
O 75-
*- I
O 50 —
O
i.
fli
£L 25-

k.
jj 75~
3 50—
O J
•9 9^
O •'*)
^
Ort
u _
75 T
50^
25-
-









6-
1T=

13-69
10.4



7-4-69
x= 1 1. 7






7-25-69




x- /. o
8-5-69






9-6-69
x= 11. 2







1 i i i i i | i i ill
04 8 12 1
6-12-70
x=8.3






	 1
7-3-70
"x= 9.4







'/ '
7
"x



>» '
-24-70
= 10.3
' ' . " ' '<•>'<
8-4-70
x=13.0


ii i i r
6 0

;;; ...U.,:' .


9-4
"x-15

%

- 70
.2
i
(••••• 	
•pi" r ' " \~ 	 i i 'i 	 i i
i 8 12 16
Depth (m)

-------
                            185
Figure 58.   Hemlock  Lake  Chaoborus  pupae  depth  distribution
            as  percent  of number  during each  sampling period
            during the  summers  1969 and 1970.   Shaded histo-
            grams represent  aerated periods.

-------
75-E
50-
25 -:
0"
—
75-
S^
& 1
£0e -1
/ J
3 :
Z P:
0-
»»-
o
75-
** 1
°J 50 "
Q OU
1_
« :
0- 25-
o> :
<« o-
a
£«-!
(/> :
3 50-
O _
•S 25^
1
U 0-j
75-
50-
25 -_
0-
(

6-13-69
*= 8. 4





7-4-69
TT=11.6









7-25-69
x-5.5
8-
X

-5-69
= 10-4

9-6-69
IT- 12.9





i


ill i i i | i i | i i i | i i
) 4 8 12 16 <
6-12-7
x»6.3





" • -•- -;"--:'-"-:i:-----'










7-3-70
x=5.5
,i,-iS"':i7

7-24-70
x=8.2



_ -:L .^..
8-4-70
x- 1 1.0


::i; ,





' ' ^', < ^r
' '"';' f'--'^
9-4-70
x=9.2
III II | 1 T'l | 1 | 1 | I 1
D 4 8 12 16
Depth (m)

-------
                            187
increased steadily until about July 20, 1970 at which time



about 60% were collected from this interval.  About this



same time,  larval numbers reached a peak.  Thereafter larval



numbers declined and pupal numbers increased and C.. puncti-



pennis emerged in large numbers.  There was also a sharp



reduction in number of larvae in the 4 to 8 m interval and



a large increase in larvae in the 16 to 18.6 interval.  Pupal



depth distributions in the emergence traps were similar to



larval distributions,  but dredge sample distributions are



spurious.



     Ephemeroptera (mayflies) constituted less than two per-



cent of the macro benthos total weight each year, and less



than five percent of the numbers (Table 5; Figure 42).



T. Wilson Britt (Ohio State University) identified at least



three species of mayflies from Hemlock Lake based on nymphs



and emergent adults:   Caenis simulans McDunnough (most abun-



dant) ,  Stenonema tripunctatum (Banks)  and Callibaetis (least



abundant) .   Mayfly standing crops were about the same each



year.   Estimates of total numbers in the lake ranged between



zero and 2x106 individuals,  and estimates of total biomass



ranged between zero and 900  grams (Figure 59) .  One hundred



fifteen individuals were collected by the dredge during



1969,  and 110 individuals during 1970  (Table 5).  Even though



standing crops were essentially the same each year, more




mayflies emerged during 1970 than during 1969.  Peak emer-



gence was over 8xl05/week during August 1970,  compared to



5xl05/week  during June 1969.   This greater emergence rate

-------
                            188
Figure 59.  Total estimated Mayfly (Ephemeroptera)  number
            and wet weight in Hemlock Lake during the
            summers 1969 and 1970.  One standard error is
            shown about each estimate.   Aeration occurred
            continuously between June 14 and September 7,
            1970.  Totals from dredge samples only.

-------
    6-



•o""*
O
~  4-


.2
M-
 X
 0)  0
 -Q
 E
 3
 z

   -2-
 o>
CN
O
    20-
    15-
 o

 o  10H
 _x  5-
 M-
 X
 o
   -5-
      1     10    20      1     10    20       1

             June               July
ib    2'o      I

 August

-------
                            190
may reflect increased rates of production not obvious from



estimates of standing crop.  An interesting feature of



mayfly emergence is the similarity of peak emergences each



year.  Both years peak emergences occurred about July 1st and



August 20th.  Maximum emergences differed by about 15 days



between the two years.  These emergence patterns may be



caused by different species emerging at different times, but



this was not evaluated.



     Mayflies were always most abundant between 0 and 4 m



depth.  Over 85% were always found in this interval (Figure



60).  A few individuals were occasionally collected below 8



meters, but never below 11 meters.  This distribution per-



sisted both years.   Emergence during 1969 was mostly between



4 and 8 m (Figure 51).  About 20% of the total emergence 1969



occurred between 0 and 4 m.  During 1970,  all mayflies



emerged from less than 4 m.



     One Heleidae was identified by J. A.  Downes (Canada



Dept. of Agriculture).  It belonged to the tribe Stenoxenin;



and was probably Jenkenshe1pa maqnipennis (Johannsen).



Heleids were not common in the emergents,  but comprised about



2% of the biomass standing crop and about 3% of the numeric



standing crop.




     T. Wayne Porter (Michigan State University) identified



only one species of Amphida as Hyalella azteca.  Only three



amphipods were collected each year.



     Jarl K. Hiltonen (United States Bureau of Commercial



Fisheries)  identified three species of Oligochaeta from

-------
                            191
Figure 60.   Hemlock Lake Mayfly (Ephemeroptera)  depth
            distribution as  percent of  number  during each
            each sampling period during the summers  1969
            and 1970.   Shaded histograms represent aerated
            periods.

-------
75-_
25-
0~
75-
Number
KJ Ol
:> 0» O
1 1 ! 1 1 1 1 1 1 1 1 1
**- "
0 75^
c :
aJ 50-
o
0 9-5-
CL 25 :
tn 0-~
fi\ ~"
• ••
«j E
« 50-^
25-^
o
75-
50 -:
25 -_
0 "
(
















6-13-69

7-4-69
•sr-2.1

7-25-69
x-1.8


8-5-69
x.1.8


9-6-69
x- 2.0





-
,->-












6-12-70
X- 2.4

| 	 ,
7-3-70
If. 1.8

7-24-70
x-1.8


8-4-70
x=2.4


t- :z~.-.: \
9-4-70
x=>2.4



1 ' i i i i iii i i i | i i i i i i i i ' i i i i i i i i i |
) 4 8 12 16 0 4 8 12 16
Depth (m)

-------
                            193
Hemlock:   Tubifex tubifex,  Limnodrilus hoffmeisteri and



Ilyodrilus templetoni.   L.  hoffmeisteri was most abundant



and T_.  tubifex least abundant.  Only 4 oligochaetes were



collected during 1969,  but  267 were collected during 1970.



Oligochaetes were found at  all depths.  They were more abun-



dant at intermediate depths,  but no persistent pattern was



obvious.



     One species of Damselfly (Zygoptera)  Enallagma,  and



two species of Dragonflies  (Anisoptera),  Gomphus spicatus



Hagan and Ladona julia  (Uhler) were identified by Leonora



K. Gloyd (University of Michigan).   Anisoptera comprised



over 35% of the biomass standing crop each year, but less



than 2% of the total numbers.  Odonata were always most



abundant between 0 and  4 m.  Naiads were frequently collected



in the emergence traps,  but adults  never emerged in the



traps.   A very large emergence of G. spicatus occurred June



1, 1970,   No estimate of its  magnitude was made, nor were



emergences observed at  any  other time.



     Trichoptera,  Tabaniids and leeches were not identified



to lesser taxonomic categories because of their small con-



tribution and difficulties  in identification.








Crayfish






     Crayfish were collected from Hemlock Lake  during 1969



only from August 12th through August 30th.  furing this



period,  they were found between 0 m and 8 m  (Figure 61) .

-------
Figure 61.  Hemlock crayfish depth distributions during the summers 1969
            and 1970.  Total numbers during each sample period and their
            average depths are shown.  The shaded area represents the
            1969 distributions.  Aeration occurred continuously between
            June 14 and September 7, 1970.

-------
                                       Percent
a
0)
Q

0.
-
5-
10-
-
-
5-

10 —


15-


60 40 20
• i i i i j



n = 8
7=1.9
6-7-70
6-16-70



n=32
x=3.9
p



7-31-70
8-21-70
C














20 40 60
i l i l i l i


J
cf



n=89
x=4-l
d*




60 40 20 C


L
7=1.9
6-16-70
7-3 -70



n=25
7=6.2
9


r
8-21-70
9-4-70
) 2,0 40 60


J
c*



-,- n=Hl
P 7=6.7
Cf*




6,0 4,0 20 C
• i i i i i i



n=10
7=3.2
7-3-70
7-31-70
*> «',
: s
^
n-27
7=4.0
9
T


-12-69
8-30-69
) . 2P . 4P , 6I° ,



C/1
•f^'
« f > ":
- '
n-53
7=4.3
cf





-------
                            196
Oxygen and temperature at 8 m were greater than 5 mg/1,



and about 8°C.  Oxygen was not limiting, but temperature is



colder than normal for sustained activity of crayfish  (Momot,



1967).  Females average depth was 4.0m, and males had a



4.3 m average.  Their sex ratio was 1:2 in favor of the



males.



     Before aeration began during 1970, the crayfish were



confined to 6 m depth or less.  The females had an average



depth of 1.9 m and the males 1.5 m.  Oxygen and temperature



at 6 m were 6 mg/1 and 7°C respectively.  Continuous aera-



tion began June 14, 1970 and continued throughout the summer.



After one week of aeration,  there was always more than 5



mg/1 oxygen in all parts of the lake.  The hypolimnetic



temperature increased more than 2 C/week..  Concurrent with



the continuous aeration,  the average crayfish depth increased



gradually.  By the end of the summer, crayfish were found at



all depths.  The depth distribution for males and females



are similar all summer.   During August 12th to September 4th



females had an average depth of 6,2 m and males 6.7 m.  The



female to male sex ratio during 1970 was 1:3.6.








Rainbow Trout






     Rainbow trout ranged between the surface and 10 meters



during June and early July,  1969 (Figure 62).  Their lower



depth range was undoubtedly defined by the anoxic condition



of the deep water.  Those fish captured below 7 meters were

-------
Figure 62.  Hemlock Lake rainbow trout depth distribution during 1969.               i-
            These fish were stocked during June 1969 and marked with a               -j
            right-abdominal fin clip.  Each square represents one fish.

-------
             June
                                                 August
            10
    0-
IIII111 I ll 111 11 111111111 11
                                                            nn-n	cmp
                                                             aaa    ati
•*-  10  --

-------
                            199
always dead and moribund.   During late July and August 1969,



the trout generally ranged between the 7°C and 21°C iso-



therms.   Their lower range was related to the anoxia, but



their upper extent appeared limited by warm water.



     Before June 14,  1970,  the 1969 stocked rainbow trout



were again limited to shallow depths.  They ranged from the



surface to 6 meters (Figure 63).   Their lower depth range is



again related to anoxia.   Continuous artificial hypolimnion



aeration began June 14 and continued through September 7,



1970.  The trout gradually extended their depth distribution



after June 14 and eventually distributed throughout the lake



by July 4,  1970.  The remainder of the summer 1970, they



were found at all depths,  with an apparent preference for



the bottom during late July and early August.



     The rainbow trout held in cages during 1970 distributed



throughout the lake when released (Figures 64,  65, 66 and 67)



They were most abundant in the 2  to 6 meter zone soon after



release (between 10°C and 17°C isotherms),  but their range



extended from the surface to the  bottom.  A slight concen-



tration also initially existed at 12 to 15 meters.  This



general pattern existed for all four lots and indicates that



acclimatization did not greatly affect their distribution.



Their concentration at 2  to 6 meters persisted for several



weeks;  thereafter their distribution was more uniform.



Throughout the summer they avoided water 21°C or warmer-



This same avoidance was also apparent both years with the



1969 stocked trout.

-------
Figure 63.  1970 depth distribution of Hemlock Lake rainbow trout stocked            N>
            during June 1969,  Each circle represents one fish.   These               §
            fish were marked with a right-pelvic fin chip.   Aeration
            occurred continuously between June 14th and September 7th.

-------
 June
10     20
 July
10     20
        August
1     10    20

-------
Figure 64.  1970 Hemlock Lake depth distribution of right-pectoral clipped           NO
            rainbow trout stocked during June 1970.  These fish were held            S
            in the 3 m covered cage which received 12 m water for one week
            before their release.  Each circle represents one fish.
            Aeration occurred continuously between June 14 and September 7,
            1970.

-------
   August
10     20       1

-------
Figure 65.  1970 Hemlock Lake depth distribution of left-pectoral clipped
            rainbow trout stocked during June 1970.  These fish were held
            in the 12 m covered cage which received 3 m water for one week
            before their release.  Each circle represents one fish.
            Aeration occurred continuously between June 14 and September 7,
            1970.

-------
  August
10     20      ]

-------
Figure 66.  1970 Hemlock Lake depth distribution of left pelvic clipped
            rainbow trout stocked during June 1970.  These fish were held
            in a screened cage at 12 m for one week before their release.
            Each circle represents one fish.   Aeration occurred continuously
            between June 14 and September 7,  1970.

-------
       August
1     10     20

-------
Figure 67.  1970 Hemlock Lake depth distribution of anal-clipped rainbow             o
            trout stocked during June 1970.  These fish were held at 3 m             °o
            in a screened cage for one week,  and then at 12 m in a
            screened cage for another week before their release.  Each
            circle represents one fish.  Aeration occurred continuously
            between June 14 and September 7,  1970.

-------
   August
10     20      1

-------
                            210
     Rainbow trout grew rapidly in Hemlock Lake both years.



Fish less than 200 mm were seldom caught in the gill nets.



Although the 1969 stocked fish averaged only 188 mm and the



1970 about 200 mm, many of the smaller fish grew sufficiently



during the summers to become vulnerable to the nets.  RBT



growth rates will not be presented in this thesis.

-------
                DISCUSSION AND CONCLUSION








Physical-Chemical Parameters






     A lake1s oxygen and temperature regimes are very useful



indicative parameters.  Yearly extremes, and the distribu-



tion of oxygen and temperature at any given time reveal much



about a lake.  Indexed, the kind of life and its spatial and



temporal distribution is usually determined in part by the



oxygen and temperature regimes.  Normally,  aerobic biota



were excluded from Hemlock's profundal zone by anoxia.



Continuous anoxia existed in the monimolimnion,,  Oxygen was



typically depleted in the hypolimnion below 12 meters.



This combination of anoxia and cold water undoubtedly con-



tributed to the accumulation of organic debris within the



lake,  and its enriched condition.



     Artificial hypolimnetic aeration greatly altered Hemlock



Lake's limnology in general,  and its oxygen and temperature



regimes in particular.  Although the lake gradually destrati-



fied during 1970,  a well-defined meltalimnion existed most



of the summer.   As a result of aeration, the monimolimnion




was eliminated by mixing with the hypolimnion, and oxygen




concentrations were greatly increased throughout the hypolim-



nion.   Hypolimnetic oxygen values often exceeded surface
                            211

-------
                            212
values.  This hypolimnetic super saturation relative to the



surface was possible due to hypolimnetic low water tempera-



tures and greater hydrostatic pressures.



     Based on hypolimnetic oxygen concentration, Hemlock



lake was eutrophic prior to aeration,  but "oligotrophic"



after aeration began.  Oxygen values increased from zero



before to over 11 mg/1 after aeration.  This alteration



greatly affects the chemical and life processes within the



lake.  Before aeration,  the hypolimnetic waters and profundal



sediments were characterized by anaerobic decomposers.



These forms are metabolically less efficient than aerobic



forms.  As evidenced by the highly organic nature of the



profundal sediments,  these decomposers could not break down



the input of organic detritus.   The profundal sediments were



gelatinous and adhesive before aeration.  After aeration



began, these sediments readily fell apart when handled.



I would attribute this change in character to decomposer



changes from anaerobic types to aerobic, and invasion of the



profundal zone by macrozoobenthos.  The latter include midge



larvae and oligochaetes.  These zoobenthos accelerate the



aeration of the sediments by burrowing and circulating



aerated water through their burrows.



     Increased hypolimnetic temperatures during aeration



must have also affected chemical reaction rates and the




growth of microorganisms and zoobenthos.  By September



1970, aeration increased hypolimnetic temperature more than

-------
                            213
12°C above its usual level.   Chemical reactions and metabolic



processes generally double with every 10°C temperature in-



crease.



     Increased hypolimnetic temperature was partly caused



by heat conductance through the aeration tower.  As the water



flowed through the tower and back into the hypolimnion, it



absorbed heat through the metal walls of the outer tube



(Figure 21) .   This heating occurred over that region of the



outer pipe where the outside water temperatures were greater



than the inside temperature.  This heating also occurred



through the pipe above the lake1s surface where the inside



water was elevated due to the rising air.  During most of



the summer this heating through the pipe occurred between 8



meters depth and 0.5 meters above the lake's surface.  This



heating can and should be virtually eliminated in future



designs by insulating the outer pipe.  Polyeurothane foam



sprayed on the sides of the outer pipe would provide such



insulation and also contribute to the tower's buoyancy.



This heating should be eliminated if cold aerated water is



desirable.   If warm aerated water is desired,  then an arti-



ficial destratification system employing the free release



of air is more desirable,  since such a system is much more



efficient than the one used.



     The compressed air temperature is not important as a




factor affecting the water temperature.  This is true because



of low mass  and specific heat of the air relative to that of



the water,  even though the air temperature may exceed 100°C.

-------
                            214
Based on my conservative calculations, the compressed air



did not raise the hypolimnetic temperature more than 0.5°C



during the entire summer of continuous aeration.



     The slightly lower epilimnetic water temperatures,  and



the altered metalimnion profile,  were caused mainly by hypo-



limnetic water mixed with epilimnetic and metalimnetic



waters.  This mixing was caused by two factors:   (1) Leakage



of hypolimnetic water through the upper walls of the aerator,



and (2) Leakage of air through lower walls of the aerator.



The escaped hypolimnetic water mixed with epilimnetic and



metalimnetic waters to form water of intermediate tempera-



ture and density.  This water flowed into the metalimnion



causing it to increase in volume.  Air escaping from the



lower section of the tower caused hypolimnetic water to upwell



around the tower and mix with surface and metalimnion water.



This mixed water also flowed into the metalimnion and in-



creased its volume.  The metalimnion volume increase should



lead to lower epilimnetic water temperatures as normal wind-



driven currents erode the metalimnion from above.  Higher



hypolimnetic temperatures and aerator-driven hypolimnetic



currents also erode the metalimnion from below.  Heat loss



from the surface waters into the tower also contributed to



this process, but is probably of less importance.  Water



and air losses from the tower were caused by faulty design.



The riveted plates forming the pipe walls should have been



caulked,  but were not.  Water and air passed through the

-------
                            215
pipes at the joints.   This fault can easily be overcome in



future designs by either caulking between the plates, or by



using welded pipe which lacks overlapping plates and is



watertight.



     Schmidt (1915)  defines thermal stability in terms of



energy required to change a thermally stratified lake to



one of equal temperature throughout.  A thermally stratified



lake has a shallower center of gravity than an isothermal



lake.  This  fact arises from water1s specific density proper-



ties as a function of temperature.  Pure water is at maximum



density at 3.94 C and less dense above or below this tempera-



ture.  The less dense water is near the surface.  Thermal



stability is zero when a lake is isothermal and reaches



maximum value when the lake is well stratified.  Stability is



therefore a  measure of destratification.  It is not an ab-



solute measure,  since Fast (1968)  found it partly a function



of water volume.   Hemlock's thermal stability was greater



during June  1970 than during June 1969 because of warmer



surface waters during June 1970.  It gradually decreased to



near zero during the summer 1970 as the lake gradually de-



stratified.   It thereby reflects the gradual converging of



maximum-minimum water temperatures during 1970.  An opposite



trend occurred during 1969 when the stability was near maxi-



mum by the end of the summer.




     I had expected an increase in average pH and a decrease



in average alkalinity following aeration.  As carbon dioxide

-------
                            216
was removed from the hypolimnetic water by the injected air,



I expected the following net reaction:






     Ca++ + HC03    v  CaC03Y  + H2C03     * C02 I  + H20





This should result in a net decrease in calcium ions in solu-



tion, an increased pH and decreased total alkalinity-  Average



alkalinity was slightly higher after aeration  (Figure 36) ,



while average pH was slightly lower (Figure 35).  Calcium



showed the most variability (Table A-ll).  It decreased from



an average 47.7 mg/1 on June 13, 1970 to a 40.8 mg/1 average



minimum during July 17th.  The greatest decreases occurred in



the deep water.  This suggests that during this period CaC03



was precipitated within the hypolimnion as the carbon dioxide



was driven  off. From July 17th to August  15th the average cal-



cium concentration increased greatly to 55.4 mg/1 maximum.



The greatest increases occurred in shallow water.  The lake's



gradual destratification in late August may have caused this



increase.  The free carbon dioxide concentration of shallow



waters may have been increased and caused a re-solution of



shallow water CaC03 deposits.   Surface C02 concentrations



were 1.4 mg/1 on June 13th, but 5.4 mg-1 on September 2nd,



although the average COa concentrations for the lake on these



dates were 22.0 mg/1 and 5.4 mg/1 respectively.  The shallow



water sediments were much richer in CaCO3 than the profundal



sediments  (Figure 15).




     I had expected greater changes in the dissolved organic



matter.  The average concentrations varied between 7.04 and

-------
                            217
10.46 mg/1.   For any given date surface concentrations were



similar to bottom concentrations.  I anticipated precipita-



tion of DOM as it passed through the aeration tower.  H. B.



Hynes (personal communication,  University of Waterloo),



observed such precipitation when DOM from leaf extract was



mechanically agitated in a flask.  This process should have



caused a reduction in DOM within the lake.  Concurrently I



expected an increase in DOM as the profundal microbiota



changed from anaerobic to aerobic forms and accelerated the



decomposition of the highly organic profundal sediments.



Possibly these two processes did occur to a significant but



nearly equal extent and balanced each other's effects.



R. G. Wetzel (personal communication,  Michigan State Univer-



sity) feels that a more likely explanation is that the DOM



fraction is composed mostly of biologically highly refrac-



tory organic compounds that are not subject to much change



and there was little net input from the sediments.  The



particulate organic matter varied greatly during the summer.



This variation is related to changes in organic production



rates.  R.  G. Wetzel plans to evaluate the effects of aera-



tion on DOM,  POM,  Ca,  K,  Na and Mg in much greater detail in



a later publication.



     We originally planned a more intense evaluation of



phosphorus.   However,  due to the inherent variability and



limited resources,  this plan was aborted.  Barrett  (1952)



studied the  effects of artificial fertilization on the

-------
                            218
phosphorus contents of these lakes in greater detail than



we intended.  His results were often incongruous with ex-



pected results.  To be sure, phosphorus is an important



nutrient and often the limiting factor, or is associated



with the limiting factor of primary production.  However,



other nutrients such as Mo, Co, Zn, or Mn can also be limit-



ing  (Goldman,  1962).   Many other conditions or substances



can also affect production.  As Rigler  (1964) quite skill-



fully points out,  we do not know very much about the cycling



of phosphorus with a lake ecosystem.  The forms phosphorus



takes,  its concentrations with ecosystem "compartments" and



especially the rate functions between compartments are



unknown and not easily measured.  The best we could hope for



in our study was to determine the changes in the vertical



concentrations of total and dissolved phosphorus; but this



could readily be inferred from changes in oxygen concentra-



tion.  With these facts in mind, I decided to discontinue



phosphorus determinations after 1969 and concentrate my



efforts on evaluating the effects of aeration on production



and standing crops of the biota.








Primary Production






     Primary production changes are sensitive indications of




changes in nutrient availability,  as well as many other



basic conditions in a lake.  Unicellular primary producers



respond very rapidly to these changes.  They integrate

-------
                            219
changes in a large number of variables to produce a given



response,  a feat not fully understood.  Responses include



changes in species composition as well as production rates.



Happenings at the primary producer level are ultimately



passed along to higher trophic levels.  Although the trans-



mission route is not fully understood, high primary pro-



duction generally leads to high production at higher trophic



levels.  High primary production also leads to eutrophy and



hypolimnion stagnation, conditions I intended to eliminate



by artificial hypolimnion aeration.



     Although aeration did eliminate hypolimnion stagnation,



it did not reduce primary production during the entire



summer.  My hypothesis is that if nutrient regeneration due



to hypolimnetic anoxia constitutes a significant input to



the nutrient cycle,  then hypolimnion aeration should reduce



this input and lead to reduced nutrient and primary produc-



tion levels.  Unless a net nutrient flow exists from the



anoxic hypolimnion to the trophogenic zone,  then primary



production reductions should not occur until the fall turn-



over following aeration,  and during subsequent periods if



high profundal oxygen concentrations are maintained.



     Almost immediately following the beginning of artificial



aeration during June 1970 an intense plankton bloom devel-




oped.   I attribute this bloom to leaks in the aeration



tower.   When aeration began,  hypolimnetic water was especial-



ly rich in nutrients.   Significant quantities of this water

-------
                            220
leaked through the tower into the epilimnion.  These nutrients



undoubtedly led to the intense plankton bloom.  The bloom



subsided to a very low level by the end of July 1970.  At



that time,  the lake was clearer than at any known time, while



planktonic primary production and standing crops were less



than during 1969.  I attribute this decline primarily to



nutrient deprivation.  Although hypolimnetic water leaked



through the tower all summer,  I believe many of the nutrients



became oxidized and effectively removed from the hypolimnetic



water soon after aeration began.  The profundal mud surface



was oxidized by the artificial aeration.  Such a surface has



a large capacity to remove phosphorus from the water by



sorption (Fitzgerald, 1970) .   Nutrients such as iron, phos-



phorus and manganese were probably precipitated directly by



the advent of aerobic conditions,  higher redox potentials



and higher pH.



     Another indication that nutrient limitation contributed



to the plankton decline during July 1970 is reduced primary



production efficiencies.   Efficiency is measured as the rate



of primary production in mgC/ms/4 hours per phytoplankton



cell.  A more appropriate measure of phytoplankton biomass



could be used, such as cell volumes or chlorophyll concen-



trations,  but these measures were not available.  The



Observed production rate is actually a measure of the



"production potential" and not the true in situ production



rate.  As will be demonstrated later for Section Four Lake,

-------
                            221
these two properties can be quite different.  The production



rate was measured in a constant light-intensity chamber and



many of the variable lake influences are thereby eliminated



or diminished.   In any event,  this production rate is a



valuable measure of production potential,  and the efficiency



ratio provides  a measure of the ability of each cell to grow



and reproduce.   High efficiencies imply an adequate environ-



ment and vigorous population.   Low efficiencies imply nutri-



ent limitation  or some other deleterious condition.



     We observed a nearly uniform efficiency value of less



than 5 mgC/m3/4 hrs/ln cell during most of 1969.  This ef-



ficiency increased markedly after aeration began,  to more



than 45 and subsequently declined to less than 5 by late



July.  The efficiency increase implies more desirable con-



ditions for phytoplankton growth,  such as provided by in-



creased nutrient concentrations.   Decreased efficiencies could



be attributed to the removal of the same factor.  It should



be noted that maximum efficiencies do not coincide with



maximum phytoplankton standing crops during June and July



1970.  A number of factors could contribute to this situa-



tion, such as:   (1)  changes in phytoplankton species composi-



tion, with different species having different growth poten-



tials and nutrient requirements,  (2)  change in size of the



phytoplankters.   Since we have essentially equated all cells



in our efficiency parameters,  cell size alone could influ-



ence the parameter,  and (3)  zooplankton grazing.  Standing



crop reduction  without a corresponding production efficiency

-------
                            222
reduction may be due to intensive zooplankton grazing„



Nutrient limitation should not produce these effects, but on



the contrary cause a corresponding reduction in efficiency,



Harvey et al. (1935)  have demonstrated that zooplankton can



exert a substantial influence on the phytoplankton popula-



tion.  Intense grazing could possibly reduce the plankton



population without reducing the production efficiency.



Large zooplankton population sizes probably occurred in Hem-



lock during early and mid-July,  coinciding with maximum



primary production efficiencies.  Brooks and Dodson  (1965)



suggest that the kind and quantity of zooplankton can effect



the kind and quantity of phytoplankton.  Large standing crops



of large zooplankton should lead to reduced phytoplankton



standing crops,  whereas large standing crops of small zoo-



plankters should favor large phytoplankton standing crops.



HrbacJ-ek et al. (1961)  found that dominant small zooplankters



favor the abundance of nannoplankton, as well as larger



phytoplankters-   Small zooplankters presumably cannot harvest



small phytoplankton as efficiently as large zooplankters,



nor can they handle large phytoplankters.  The large zoo-



plankter Daphnia pulex became very abundant in Hemlock during



July, and a large increase in the smaller zooplankter Bosmina



was also observed.  The kind of zooplankton that is present



is largely controlled by fish predation.  Changes in phyto-



plankton species composition caused by zooplankton grazing



could account in part for the lag in production efficiency



relative to standing crop in Hemlock Lake, but probably not

-------
                            223
the large decrease in efficiency during late July.



     The very low phytoplankton production and standing crop



values during late July and early August suggest that there



could ordinarily be a net movement of nutrient from the



hypolimnion during hypolimnetic stagnation.  This suggestion



is supported by the lower plankton populations during this



period of 1970 when the hypolimnion was aerated,  than during



the same period of 1969 when the hypolimnion was anoxic.



There was probably a net movement of nutrients from the



limnetic compartments into the littoral and profundal com-



partments during June and July 1970 (Figure 1)„   Aerobic



hypolimnetic conditions prevented nutrient regeneration from



the hypolimnion.  Aerobic littoral muds and plant growth



prevented regeneration from the littoral zone.  These net



losses would result in lower phytoplankton production.



     Following the 1970 phytoplankton minima,  there was a



gradual increase in standing crop, production potential and



production efficiency during the remainder of the summer.



I believe this recovery can also be attributed to changes



in nutrient levels,  but by a different mechanism than



associated with the earlier plankton bloom.  I believe there



were two main sources for the nutrients that sustained the



August bloom:   nutrients released from the littoral compart-



ment,  and nutrients released from the profundal muds.



Hutchinson (1941)  found large increases in phosphorus con-



centrations associated with temperature increases.  He con-



cludes,  "...  that this phosphorus can only have come from

-------
                            224
the marginal sediments.  Such a sudden excess of the element



is undoubtedly due to an increased rate of organic decomposi-



tion in the shallowest mud and in the organic debris in the



weed beds of the lake."  (Hutchinson, 1957.)  If this is



true, then this same mechanism could have resulted in aerobic



nutrient regeneration from the profundal sediments.  These



nutrients would have become available to the limnetic algae



from leaks in the tower, but more importantly by the in-



creased thermal destratification that occurred during August



1970.  This destratification rate was accelerated as tempera-



ture differences diminished between shallow and deep water.



Increased oxygen and temperature content of the profundal



sediments should have greatly accelerated their decomposition.



These sediments are very high in incompletely decomposed



organic matter.  Before aeration,  decomposition was hindered



by anaerobiosis and low temperatures.  Aeration increased



oxygen concentrations from zero to over 10 mg/1, and tempera-



tures from less than 5°C to 19 C.   Aerobic bacteria are more



efficient decomposers and undoubtedly accelerated decomposi-



tion rates.  Macrozoobenthos such as Chironomids and oli-



gochaetes undoubtedly accelerated this process by burrowing



through the sediments.  Increased decomposition rates and



water circulation through the sediments may have caused



aerobic nutrient regeneration.  This process was probably



greatest during late summer and probably accounts in large



part for the August plankton bloom.

-------
                            225
     It may seem  incongruous  to credit nutrient depletion



for cessation of  the bloom during July,  and nutrient regen-



eration for the August  bloom.   I am guessing that the



nutrient  depletion  was  caused by precipitation and sorption



of nutrients by the sediments.   A net influx of nutrients



to the profundal  sediments could have occurred during early



summer, before they were greatly heated and before microbial



and macrozoobenthic populations were established.  A net



output from the profundal sediments could have occurred



later in  the summer due to increased decomposition.  The



latter hypothesis may be partly supported by the foam produc-



tion.  Foam first became noticeable about August 1, 1970.



This coincides with the commencement of the second phyto-



plankton  bloom.   Although the composition and source of the



foam is unknown,  it is  probably a dissolved organic substance.



It could  have been  generated  by the accelerated decomposition



of the profundal  sediments.   H. B. Hynes (personal communi-



cations)  observed the precipitation of DOM from leaf litter.



This precipitation  was  promoted by shaking.  Particles of



larger size were  formed by longer periods of agitation.



Increased profundal decomposition could have released DOM



as well as other  nutrients.   Our estimates of DOM concentra-



tion are  too incomplete to shed much light on this hypothesis.



Furthermore,  complete concentration estimates might not pro-



vide the  answer since the rates of generation and utilization



and the nature of the material is of greater importance.

-------
                            226
R. G. Wetzel (personal communications) feels that most of



the DOM generated may be utilized very rapidly by the



microbes,  such that essentially only the refractory fraction



remains.  Static measures,  such as concentration, would



therefore not give an accurate picture of its role.



     A minimum DOM value of 4,68 mg/1 at 15.6 m was observed



August 15,  1970.  This occurred during a 4 or 5 day cessation



in foam production.  Foaming began again on August 19th and



continued until we stopped aeration.  A maximum DOM concen-



tration of 10.89 mg/1 was observed September 5th.  If foam



is formed from DOM produced by decomposition of the profundal



sediments,  then it also indicates that other nutrients are



also being generated by this decomposition.  This supports



my hypothesis that the second plankton bloom was promoted by



nutrient regeneration from the profundal sediments.



     If primary production is to be reduced, then it appears



essential that nutrient regeneration from the profundal



sediments be greatly reduced.  Whether this regeneration is



by anaerobic decomposition,  or by aerobic decomposition,  the



results could be the same;  but it is too early to tell.



Continued hypolimnion aeration could lead to a well-oxidized



and mineralized "crust" on the profundal sediments.  This



could occur due to microbial decomposition and continuous



oxidation of the sediment surface.  Once this crust is formed,




aerobic nutrient regeneration might be minimal, or at least



less than the former anaerobic nutrient regeneration.  Even



on a short-term basis,  a leakproof aeration tower will

-------
                            227
greatly reduce the impact of nutrient regeneration.   These



nutrients may mostly remain in the hypolimnion during the



warm summer months and be essentially unavailable for plank-



ton growth until turnover,  sometime in the fall,,



     Periphyton standing crop may be taken as a measure of



attached algal net production.  This is not strictly true



since many kinds of biota are included in this measurement,.



In most cases,  it probably affords a simple measure of



attached algal relative production.



     Long-term periphyton accumulations were significantly



greater following aeration compared to comparable non-



aerated periods.   Likewise,  short-term accumulations were



significantly greater except during late July and early



August 1970.   During this later period,  accumulation rates



were similar both years.   Even long-term accumulations had



a slight slump during late July 1970=  These changes in



periphyton accumulation rates indicate that both populations



were affected by the same variables,  the most likely being



nutrients.   It seems unlikely that herbivore grazing on



both kinds of primary producers could cause such similar



response patterns.   Under some circumstances,  increases in



phytoplankton production could lead to reductions in peri-



phyton production due to shading or nutrient deprivation.



Conversely,  periphyton growth could limit phytoplankton by



nutrient deprivation,  but not by shading.  Neither event was



apparent in Hemlock Lake.   This was probably due in part to

-------
                            228
the vertical distribution of the periphyton.  It was mostly



concentrated in the upper few meters and thus not greatly



affected by shading due to dense phytoplankton populations.



Likewise, both populations seemed to "share" the nutrients



and thus responded in a similar manner.



     Periphyton never appeared overly abundant on logs or



other suitable substrate.  Submerged aquatic plants and algal



mats were likewise sparse.  Tanner (1952) observed similar



responses at moderate fertilization rates.  He also observed



a primary response by the phytoplankton.  Ball (1949) and



Ball and Tanner (1951)  on the other hand observed no more



than a moderate increase in plankton production at heavy



fertilization levels.  Their primary response was by attached



forms and floating algal mats.  Heavy fertilization led to



winterkill conditions,  whereas winterkill conditions were



approached with moderate fertilization.








Zooplankton



     After phytoplankton, the zooplankton appeared to give



the greatest response to artificial aeration.  Daphnia pulex,



Diaphanosoma and Bosmina had especially large increases in



total numbers.  Unfortunately, we do not have estimates of



their seasonal abundance patterns.  Figure 68 illustrates



the hypothetical changes in limnetic zooplankton, phytoplank-



ton, Chaoborus and limiting nutrients during the summer 1970.



It also shows the major food chain relationships.  D. pulex1 s

-------
                            229
Figure 68.  Hypothetical  changes  in  Hemlock  Lake  limiting
           nutrient,  phytoplankton,  zooplankton  and
           Chaoborus  densities during  1970.  Aeration
           began June 14th  and continued  through September
           7th.  Major food chain relationships  are  also
           shown.

-------
 C/5
 C
 o
 o
 c
 o
o
                           Phyfroplankton

                                Chaoborus
                                          Zooplankton
                                                     Limiting
                                                     Nutrients
                I
June I              July
     Aeration begins
August
    Phytoplankton
     Zooplankton
      Chaoborus
                  Phytoplankton
                   Zooplankton
   Phytoplankton
   Zooplankton
                Chaoborus   RBT      Chaoborus—> RBT
         RBT

-------
                            231
abundance seems related to predation by trout, phytoplankton



abundance and hypolimnion oxygen depletion.  D. pulex was



scarce before aeration began.  I attribute this scarcity to



predation by rainbow trout.  Before aeration, D. pulex were



mostly limited to depths above 9 meters by hypolimnetic



oxygen depletion.  In this zone they were especially vulner-



able.  After aeration began,  they inhabited the dimly lit



depths of the lake.   This sanctury,  in addition to increased



primary production,  allowed their population to increase very



rapidly.  By the second week of July they were very abundant.



About the end of July they formed ephypia and their popula-



tion appeared to decline.  This reduction is associated with



the reduction in phytoplankton abundance.  After early July,



D. pulex constituted the major food item of the rainbow



trout.  Before then and all during 1969, the trout fed almost



exclusively on Chaoborus spp. larvae and pupae.  From early



July through mid-August, Chaoborus were a secondary trout



food item.  By the end of August,  trout ingested D. pulex



and Ghaoborus in about equal proportions.  This is especially



interesting since Chaoborus also became more abundant after



aeration began,  and Chaoborus are larger than D. pulex;.



     The small cladocera, Bosmina spp., and medium-sized



cladocera, Diaphenosoma spp.  and the copepod Diaptomas spp.



were virtually untouched by rainbow trout at any time



although they were very abundant.   These observations agree



well with those of Galbraith (1967) .   He found that rainbow



trout in two Michigan lakes feed almost exclusively on

-------
                            232
Daphnia greater than 1.3 mm.  This predation led to the near



extinction of Daphnis pulex in one lake and apparently



favored the survival of smaller species of zooplankton.



Hall  (1964) demonstrated that fish predation accounted for



as much as a 25% per day loss from the Daphnia galeata



mendotae population in a southern Michigan lake.  Predation



on sub-adult individuals is critical to their survival and



is probably the major factor leading to the extinction of



D. pulex and other zooplankters that mature at a larger size.



     My observations conflict somewhat with Brooks and



Dodson's (1965) size efficiency hypothesis:  namely, that



planktiverous fish generally decimate the larger zooplankton



species and thus favor the survival of smaller species.  In



Hemlock, this hypothesis may effectively apply to the



D_. pulex,  Bosmina,  Diaphanosoma and Diaptomas group, but not



to p. pulex and Chaoborus.  Chaoborus is much larger than



D. pulex,  but D. pulex appears to be the much preferred food



item, even though Chaoborus were much more abundant after



aeration than before.  Chaoborus larvae and pupae consti-



tuted over 90% by volume of the trouts'  diet before aeration



began, but about only 5% to 50% after aeration.  D. pulex



constituted 50% to 95% after aeration.



     Chaoborus were much larger on the average before than



after aeration began.  This change in average size could be



due to species shift (Table 6)  and/or to increased produc-



tion rates.  Increased production following aeration would



skew  the  size distribution, with the smaller instars

-------
                            233
becoming mote abundant.  Even though the Chaoborus average



size decreased following aeration, they were still much



larger than D. pulex.



     These findings suggest that factors other than absolute



size are directing food selection by the trout.  Chaoborus



are more transparent than D. pulex and may therefore have a



smaller "effective" size.  Likewise/  their particular diel



migration pattern and habit of nestling in the mud may reduce



their vulnerability.  Rainbow trout very seldom feed on the



benthos.  Galbraith concluded that rainbow trout feed



selectively,  rather than merely straining plankton from the



water.  My findings tend to substantiate his conclusion.



     Cladocera and other zooplankters commonly exhibit diel



vertical migrations  (Gushing,  1951; Hardy,  1956; Wynne-



Edwards, 1962).   I did not try to measure their migration in



Hemlock Lake.  The typical pattern is for the zooplankter to



inhabit the dimly lit depths during the day,  and the surface



waters at night.   Some zooplankton have just the opposite



behavior,  or do not migrate according to any fixed pattern.



In our case,  ID.  pulex and Bosmina appear to conform to the



typical pattern.   Since both collections were made during



the day, we would expect the zooplankton that migrate in the



usual pattern to be concentrated near the bottom of the lake,



or at some gradient barrier (Harder,  1968).  D. pulex was




most abundant near the bottom of the lake after aeration,



whereas Bosmina was most abundant within the metalimnion,



during both collections.  ]D. pulex' s distribution was not

-------
                            234
related to the temperature gradient, but Bosmina's distribu-



tion apparently was.  The distribution of the other zoo-



plankters after aeration did not appear to be related to the



temperature gradient.



     One common hypothesis to explain vertical migration is



that the zooplankton migrate to escape predation by sight



feeders.  Those inhabiting the depths during the day are



less conspicuous than those in shallow water and predation



is thus reduced on the deep-living individuals.  They migrate



to the surface under the protection of darkness to feed.



Their upward, movement generally coincides with sunset,  while



their downward movement is initiated or timed by sunrise.



In our case,  rainbow trout and Chaoborus are probably the



two most important planktivores.  Of these two, the trout



are probably the most important and feed largely on the



D. pulex.  Furthermore,  trout feed on Chaoborus and thereby



reduce the latter's predatory impact.  Since the other zoo-



plankter species were not preyed upon by the trout, there



was probably no incentive for them to migrate.  They could,



therefore,  inhabit the surface waters during the day and



thereby prolong their grazing time.  If anything, it might



have been advantageous to disperse throughout the lake or



migrate to the depths after dark to avoid possible predation



by Chaoborus.




     In El Capitan Reservoir, California, copepods responded



to artificial destratification by concentrating along the



bottom during the day (Fast,  1971) .  They were formerly

-------
                            235
restricted within or above the metalimnion by oxygen deple-



tion.  The threadfin shad (Dorosoma petenensa) was very



abundant in this lake and large zooplankters were scarce.



Small copepods constituted about 90% of the net zooplankton.



The Hemlock copepod adults did not concentrate at the bottom



after aeration.   The reason for this is not clear.  It could



be related to the temperature gradient, or some other factor



such as lack of significant predation on their population.



Predation pressure could be an important factor affecting



zooplankton migration behavior.  It could partly explain why



a given species may migrate in one situation,  but not in



another.



     Any future investigation of this aeration system should



include a more thorough evaluation of the zooplankton re-



sponses.  A detailed investigation was not included in this



case because I felt responses in the zoobenthos and other



components of the biota would be more important,  and require



less effort to measure.  Although significant changes did



occur in these other components,  they were not all as rapid



or spectacular as those changes in the zooplankton.







Zoobenthos





     Benthic organisms are sensitive indicators of changes



in their environment.   These organisms may respond in one



or more ways depending on the nature and intensity of the



environmental change.   These responses are interrelated and

-------
                            236
include changes in:  standing crops, growth or metabolic



rates,  species composition, species dominance, depth or



spatial distribution, survivorship, reproductive rates and



behavior.  Changes in standing crop, species composition and



distributions are most easily measured.  Most studies have



concentrated on these parameters and at best tried to infer



changes in the other parameters.  Changes in the other para-



meters are presently difficult or impossible to measure



accurately.



     Zoobenthos analyses are notoriously laborious.  On the



average,  each sample required over two hours of laboratory



and field work.  This includes:  collecting, sieving, sort-



ing from the sediments,  sorting each taxa into separate



vials,  counting each taxa,  weighing each taxa and statistical



analysis.



     Probably the single most important zoobenthic parameter



is its yearly production rate.  From this you may infer how



much energy passes through this ecosystem component.  This



information relates directly to how much energy was fixed



by the plants,  and how much energy was made available to



higher trophic levels such as fishes.  Ball (1948) found that



invertebrate production is related to fish production.



     Good estimates of total zoobenthos production are im-



possible with our present knowledge and technology.  Most



studies estimate zoobenthos standing crop,  and possibly some



parameter of production such as adult emergence.  From this

-------
                            237
they infer rates of production.  A few studies attempt to



measure the productivity of a given species of zoobenthos



 (Cooper,  1965;  Hilsenhoff,  1967)  but overlook changes in



the total benthic fauna assemblage.  From the standpoint of



total energy flow,  total zoobenthos production is much more



important and the production of particular species may be



relatively inconsequential.  At present we cannot accurately



measure production rates of some organisms such as oligo-



chaetes or nematodes.  We have methods for measuring pro-



duction of some organisms such as the midges, but the



techniques are greatly complicated by multivoltine species



populations,  different generation times for each species or



even with a given species,  rarity of some species and inef-



fectual sampling methods.  Crayfish production is relatively



easy to measure,  but requires different sampling techniques



than are generally used to measure zoobenthos standing crops.



Crayfish are omnivorous,  feeding mostly on plant material



and detritus.  Their trophic position is similar to other



zoobenthic organisms, but their production rates or standing



crops are seldom measured because of sampling problems.



Momot (1967)  found that their production may be more than



eight times as  great as the total production of other zoo-



benthos.   Even  for species where techniques of estimating




production are  available, the techniques differ greatly.




Therefore,  separate sampling programs,  etc. may be necessary



to measure production of more than one taxa.  This leads to

-------
                            238
exceptionally large expenditures of time and money and has



not yet been attempted.



     Because of the foregoing considerations, I chose to



measure zoobenthos standing crops,  and emergence rates of



certain taxa.  From these data I hoped to infer in a larger



sense what changes occurred due to artificial aeration.



In fact, most of our time was devoted to the analyses of



the zoobenthos.



     Between-sample variances are generally very large and



non-homogeneous.  With the possible exception of the



Chaoborus larvae,  there were not significant differences



between numbers or biomasses of specific taxa on any given



date.  Even though trends are evident,  their statistical



verification is not possible because of sample variance.



Variances were not only large in most cases, but non-



homogeneous and not a function of sample size.  This problem



invalidated the use of certain powerful statistical tests.



However, I still feel we can draw some conclusions from



trends in the data,  keeping in mind that their conclusions



are not statistically valid in all cases at the usual sig-



nificance levels.




     Increased primary production generally increases zoo-



benthos production and standing crop.  Tanner (1952) found



a large increase in Hemlock Lake1s phytoplankton and zoo-



benthos standing crops during artificial fertilization.  He



added 71.5 kg of inorganic fertilizer to Hemlock during 1949,



and 56. kg during 1950.  Zoobenthos total numbers  (standing

-------
                            239
crop)  increased 980% in 1949 compared to 1948, and 400% in



1950 compared to 1948.   Chironomidae accounted in large part



for this increase.   They increased 1078% the first year.



All other zoobenthos taxa also increased,  but not at the



same rate.   This led to changes in zoobenthos species compo-



sition.  Before fertilization,  Anisoptera were dominant in



terms of biomass.  After aeration, Trichoptera,  Ephemeroptera



and Chironomidae were more abundant.  Tanner attributed



these changes to increased primary production.  Secchi disc



transparency decreased from between 3.1m and 4.6 m before



fertilization to less than 1.5 m after.  In this lake,  secchi



measurements are representative of phytoplankton density.



     Zoobenthos standing crop and production rates apparently



increased during artificial aeration of Hemlock Lake.  Total



numbers of zoobenthic organisms almost doubled.   Total bio-



mass decreased slightly,  but not for the more important



species.  Relatively unimportant,  but large leeches comprised



25% of the biomass  during 1969 but less than 1% during 1970.



Only one leech was  captured each year.  Total midge emer-



gence more than doubled during aeration,  indicating increased



production rates.  I attribute these increases to increased



sediment decomposition and primary production, and increased



"lebensraum."  These increases are directly attributed to



artificial hypolimnion aeration.




     Hemlock Lake's zoobenthos populations were mostly re-



stricted to the epilimnion,  in depths less than 4.6 m during



1948-50 (Tanner,  1952).   Artificial fertilization did not

-------
                           240
greatly affect their summer depth distribution although



metalimnion and hypolimnion oxygen concentrations were



greatly reduced and the metalimnion depth became shallower.



Depth of oxygenated water decreased to about 75% of its



former value following fertilization,  and the average



metalimnion depth decreased from 6.8 m before 4.5 m after



fertilization.  This increase in anoxic conditions and



shoaled metalimnion should lead to decreased living space



and increased rates of organic sedimentation.  Organic matter



settling within the anaerobic zone does not readily decom-



pose and is essentially unavailable to higher trophic levels.



Artificial fertilization also increases the availability of



organic production within the epilimnion and more than com-



pensated for the potential losses from the system due to



anaerobiosis.



     Artificial hypolimnion aeration not only increased



primary production,  but it increased the availability of



former organic production.  Increased primary production



occurred inadvertently due to technical oversights.  This



response may be avoided by use of different aerator designs



and construction.  Increased availability of previously



produced organic matter,  on the other hand,  was a primary



goal.   Before aeration much of the lake was uninhabited and



underexploited by the zoobenthos due to anaerobiosis.  This




accounts in part for the highly organic profundal sediments.



The peat bottom developed in part because the browsers and



decomposers could not utilize all the allochthonous and

-------
                            241
autochthonous inputs of organic matter.   McConnel (1968)



found that alloehthonous organic litter  contributes a



significant food input to Pena Blanca Lake,  Arizona.  Based



on an estimated ecological efficiency of 0.56%,  he estimates



that this input may account for as much  as 16.7% of the



yearly largemouth bass (Micropterus salmoides)  harvest.



The lake1 s bottom,  ana other eutrophic lake bottoms like



it, have become depositaries for organic materials and



nutrients.  Anaerobiosis accelerates the rate of accumulation



and filling of the lake.   After aeration,  the entire lake



bottom was again accessible to aerobic decomposers and



macro-zoobenthos.  Midges rapidly invaded the profundal zone



and capitalized on the rich supply of organic materials.



Certain other benthic organisms such as  the mayflies and



Odonata did not invade the hypolimnion after aeration.  This



may be due to low water temperatures, positive phototaxis,



preferred food concentrations,  shelter,  sediment composition



or some other factor.



     Artificial hypolimnion aeration should lead to acceler-



ated decomposition of the profundal sediments.   Increased



oxygen will favor aerobic decomposers which are more effi-



cient than anaerobic,  and increased temperatures will



accelerate decomposition.   Increased bacterial production and



concentration,  plus their increased availability may lead to




increased production of macro-zoobenthic organisms such as



chironomidae and oligochaetes that feed  on micro-organisms.

-------
                            242
Only four oligochaetes were collected during 1969 compared to



267 during 1970,  and many of these latter worms were in the



profundal zone.  Chironomids were also more abundant in the



profundal zone after aeration.  Wirth et al. (1970)  and



Ogborn (1966)  reported increased biodegradation of organic



sediments following artificial destratification of eutrophic



reservoirs.  Their observations are qualitative rather than



quantitative,  however, since biodegradation of sediments is



very difficult to measure.  Increased biodegradation and the



subsequent incorporation of this energy into the higher



trophic levels is a form of energy "recycling".  Energy that



was formerly stored via organic compounds and lost to the



system is brought back into the system by artificial aeration.



This process may increase the overall lake productivity to a



level that cannot be sustained on a long-term basis.



Productivity may therefore decrease as the organic materials



are oxidized.



     Artificial aeration may cause a net energy loss from



the profundal sediments on a short-term basis,  but not on a



long-term basis.   Nutrient exchanges with the sediments may



follow a similar pattern.  Under anaerobic conditions,  phos-



phorus and other nutrients are solubilized and a net movement



occurs from the sediment to the water (Mortimer, 1941).  At



the spring and fall turnovers, these nutrients are distributed




throughout the lake.  These turnovers also distribute oxygen



throughout the lake, oxidize the mud surface layers and at

-------
                            243
least temporarily cause a net influx of nutrients into the



mud.  Fitzgerald (1970)  found that aerobic muds have a con-



siderable capacity to remove phosphorous from the water by



sorption.  Four tenths g (dry wt)  of mud sorbed 0.05 mg



P04-P in less than 30 min.   Artificial hypolimnion aeration



should not only oxidize the mud surface and increase its



sorption capacity,  but it will pass water over the mud and



thereby increase contact between mud and water.  Macro-



zoobenthos also increase the oxidation rate by burrowing



into the mud and pumping water through their burrows.  This



will greatly increase oxidation over that expected from



diffusion and also increase biodegradation.  Chironomid



larvae may burrow 50 cm or more into the sediments (Hilsen-



hoff,  1966).   Chironomus riparius increased the oxygen supply



to the sediments and increased the redox potential by circu-



lating water through its burrow (Edwards, 1958).  This circu-



lation is maintained to provide oxygen for respiration and



food.   Oligochaetes are also active burrowers and probably



exceed that of the midges in both extent and duration.



     The effect, of artificial aeration on the profundal



nutrients is open to question.  Hasler (1963)  found that cir-



culating water in an aquarium over mud,  increased the phos-



phorus content of the water.  Furthermore,  the artificial



destratification of El Capitan Reservoir, California did not



appear to reduce phytoplankton primary production (Fast,



unpublished data),  indicating that nutrient availability was

-------
                            244
not decreased.  Other factors such as water volume increase



complicated this evaluation,  however.  The mechanism by



which phosphorus,  or certain other nutrients,  would be trans-



ferred from the profundal muds is unknown.  It may well be



that significant quantities of nutrients may be regenerated



under both anaerobic and aerobic conditions.  In the first



case regeneration is a function of redox potential.  In the



latter case regeneration may be related to the amount and



kind of organic matter and its rate of decomposition.  The



former process could account for the June 1970 plankton



bloom, and the latter process for the August 1970 bloom.



It is my opinion that long-term artificial aeration will



result in the net transfer of nutrients into the mud as their



energy source is depleted, and a commensurate decrease in



the primary productivity of the lake.  Mechanisms affecting



nutrient transfers and rates of transfers within lakes are



presently very poorly understood.  Until we understand and



evaluate rates of transfer, we will be greatly hindered in



our understanding of lake ecology.



     The primary method by which the Chironomidae invaded



the hypolimnion is unknown.  They probably arrived by two



methods:  Dispersion of eggs and just-hatched larvae, and



active migration of the late instar larvae.  Eggs are gen-



erally broadcast over the entire lake1s surface  (Hilsenhoff,




1966).   They may settle and concentrate in certain depth



zones (Bardach,  1955; Gleason, 1961), but their development

-------
                            245
at the settling site depends on environmental factors.   Even



the eggs of low oxygen tolerant species such as Chironemus



decorus (Gleason,  1961) or C. plumosus  (Augenfeld,  1967;



Dugdale, 1955)  will not hatch in the absence of oxygen.



Newly hatched larvae of C_. plumosus are free-swimming and



strongly phototropic in the absence of a suitable substrate



 (Hilsenhoff,  1966).  In the presence of suitable substrate



their phototropism diminishes and they remain in the mud.



Furthermore,  the growth and density of the Chironomids



depends on the quantity and quality of the food supply.



Gleason found the greatest concentrations of C. decorus  in



the zone receiving the greatest concentration of fresh sedi-



ments including a high percentage of phytoplankton.



Jonasson and Kristiansen  (1967)  found that availability  of



fresh phytoplankton and oxygen concentration were the two



most important factors affecting the growth of profundal



.C. anthracins in Lake Esrom,  Denmark.  Their growth was



mostly limited to times of spring and fall turnover when



phytoplankton production was maximum and oxygen plentiful.



Growth during the summer was inhibited by low oxygen levels



and poor quality of the food.  Food reaching the bottom  was



partly decomposed,  and under low oxygen levels the  larvae



either spent much time in respiratory activity, or became



lethargic.   Dugdale (1955) also found that Chironomus plumosus



growth and emergence is mostly restricted to spring and  fall



when profundal oxygen and temperatures are greatest and



presumably food is  optimal.

-------
                            246
     Later  instar  Chironomid larvae may actively migrate



to the profundal zone.  Dugdale (1955)  found mature larvae



swimming between the surface and bottom shortly after sunset.



Mundie (1959)  found Chironomid larvae swimming at the surface



of Lac La Ronge, Saskatchewan,  Canada.   Early and late in-



stars were found mostly over 5 m of water or more and con-



sisted of 60% Chironomini,  25% Tanypodinae and 15% Ortho-



cladinae.  Active migration of Chironomid larvae could lead



to a rapid invasion of a new habitat.  There is some evidence



that late instar larvae invaded Hemlock's profundal zone soon



after aeration began.   Procladius denticulatus emerged from



between 16 and 18.6 m during the first week of July 1970,



after two weeks of aeration (Figure 50)„  Their emergence



during 1969 was restricted to 12 m or less.  Tanypus spp.



emerged from between 8 and 12 m by mid-July 1970 and from



between 16 and 18.6 m during August.  During 1969 Tanvpus



emergence was restricted to less than 4 m.  Both these spec-



ies are predaceous,  and migratory by nature (Miller, 1941).



Miller found that Procladius emerged from all depths of



Costello Lake,  Ontario at about the same time during June



and July.  He suggests this syncronous emergence is due to



larvae moving back and forth through the metalimnion in



search of prey.  Larval development is partly related to



temperature.  Larvae moving back and forth would thereby be




exposed to about the same average temperatures, develop at



about the same rate and emerge at about the same time.  Other



non-predaceous midges emerged from the profundal zone, but

-------
                            247
not as soon as Procladius.   Third and fourth instar larvae



of Ghironomus tentans were observed below 18 m during July



even though their adult form was never collected, even in



shallow water.



     Hemlock1s zoobenthos standing crop is impoverished con-



sidering the level of primary production in this lake.



motal numbers and biomass collected during 1969 were 2,396



and 2.41360 gms,  respectively (Table 5).  Although Section



Four Lake appears much less productive,  total numbers and



biomass collected during 1969 were 17,609 and 11.09227 gms.



Before fertilization,  Tanner (1951)  found averages of 83.5



organism/m2 in Section Four.  During this same period in



Hemlock,  he found only 48.5 organisms/m2.  The reasons for



this impoverishment are not obvious,  but are probably due in



part to fish predation and the nature of the substrate.  Fish



predation can greatly limit the zoobenthos standing crop



(Ball and Hayne,  1952; Wilkins,  1952).  Tanner (1951)  observed



a sharp decrease in the zoobenthos standing crop during the



second year of fertilization of Hemlock and three other



neighboring lakes.   He attributes these decreases to enhanced



survival of minnows during the first year of fertilization.



The bluntnose minnow and redside dace were very abundant in



Hemlock during 1969 and 1970.  They were restricted to



shallow depths,  generally less than 4 meters,  and mostly near



shore.   I concur with Tanner's conclusion that minnow preda-



tion on the zoobenthos greatly reduces zoobenthos standing



crop in shallow water but not in deep water after aeration.

-------
                            248
Some other factor is more important in the profundal zone.



The small profundal standing crop after aeration is partly



due to migration rates.  These apparently are slow.  We



witnessed a gradual increase during the summer,  but profundal



standing crop probably did not reach its maximum value by



September 1970.  Even with complete destratification of



El Capitan Reservoir,  California profundal standing crop



did not appear to maximize until the second summer of



destratification (Fast, unpublished data).   If we had started



aeration much sooner during 1970,  it might have increased



more rapidly.  Many species reproduce and become established



during May and early June.  The hypolimnion was anoxic dur-



ing this period,  and the organisms were probably well-



established in shallower depths before we began aeration



June 14th.



     The nature of the sediments may be another reason for



the general impoverishment of Hemlock's zoobenthos standing



crop,  and the slow profundal invasion rate.  Bog lakes, or



lakes with abundant peat deposits,  typically have impover-



ished zoobenthos populations.  Hilsenhoff and Narf (1968)



correlated 14 physical-chemical parameters with standing



crops of 13 species of midges,  Ostracoda and Copepoda in the



profundal zone of 14 Wisconsin lakes.  They found significant



negative correlations between organic matter in the mud and



standing crops of copepoda,  ostracoda,  Palpomyia and



Chironomus plumosus.  The reasons for this are unclear.

-------
                            249
In theory,  this mass of organic matter should provide an



abundant food supply.  In practice,  it seems to inhibit



zoobenthic organisms.  Three likely explanations are:



(].) The rich organic matter has a high carbon dioxide con-



tent which inhibits oxygen utilization by the midges.  High



carbon dioxide concentration sharply inhibits oxygen utiliza-



tion by fishes (Black et al.,  1954).  This may also apply



to certain zoobenthic species.  Midges collected from an-



oxic hypolimnions are typically relaxed and extended, which



is a sign of respiratory distress; (2) The accumulated plant



material possesses some inhibitory or antibiotic substance



that either directly affects the macro-zoobenthos or bacteria



used as food by the macro-zoobenthos.  The plant detritus



accumulation is evidence that bacterial decomposition is not



optimum.  This may be due to low nutritional value of the



detritus,  or due to inhibitory substances.  Hilsenhoff and



Narf also found a highly significant negative correlation



between organic matter in the mud and pH of the mud.  Highly



organic muds had low pH,  some being below 6.0.  A highly



significant correlation existed between organic matter in



the mud and water content of the mud, and a highly significant



positive correlation between pH of the mud and pH of the



water.   Low pH may be due to carbonic acid,  humic acid or



some other acid associated with the organic matter.  Low pH



in itself  could deleteriously affect zoobenthos; and/or



(3)  highly organic bottoms are often flocculant.  Midge are



positively thigmotactic,  requiring contact with a substrate.

-------
                            250
The high water content may not provide adequate contact for



the midges or facilitate tube building.



     Differences in average Chironomid larval size at dif-



ferent depths was most pronounced.  Shallow water larvae



were much smaller than deep-water larvae.  On September 4,



1970, only 5% of numbers of Chironomid larvae were found



below 14.5 m, but these comprised over 15% of the biomass



(Figures 43 and 44).   This condition is undoubtedly a result



of different species complexes at different depths.  This



gradation in size from small species in shallow water to



large species in deep water is often found in moderately and



highly productive lakes (Brundin, 1951 and 1953).  Large



larvae can better cope with oxygen microstratification at



the mud-water interface because of their greater capacity



for circulating water.  This circulation disrupts the micro-



stratification and provides oxygen to the larvae.  Further-



more, large larvae of the Chironomus type contain hemoglobin



and can withstand anaerobiosis for extended periods should



the oxygen microstratification lead to more general hypo-



limnion stagnation.



     Although oxygen microstratification was not investigated,



it seems unlikely that this developed to any appreciable



extent after aeration began.  Water currents, especially near



the aerator should have kept it to a minimum.  If this is



true, then the carbon dioxide and BOD of the sediments could



have been the main factor.  In either case,  large larvae



would be favored because of their greater respiratory activities

-------
                            251
     The extent of oxygen depletion greatly affects the kinds



of midges found in a lake.  Indeed, lakes may be classified



according to their midge populations (Lenz, 1925, 1927;



Lundbeck, 1926,  1936;  Brundin,  1953).   Unproductive lakes are



characterized by Tanytarsus midges.  These midges are small



and lack hemoglobin.   They are very intolerant to low oxygen



concentrations,  and are apparently better adapted for survival



in oxygen-rich lakes than are Chironomus midges.  Chironomus



midges are typical of rich lakes that often develop oxygen



deficits within the hypolimnion.  Bryce (1965)  found differ-



ent species complexes in 17 shallow English acid peat pools.



Nine pools were dominated by Tanytarsus, while seven other



pools were dominated by Chironomus.  One pool was intermediate



with about equal proportions of both species.  All but one



pool was less than one meter deep.  pH ranged between 4.0



and 6.0.  Bryce did not indicate what factors seem to be



responsible for the species complexes.



     Interestingly enough, oxygen concentration of the water



does not appear related to organic content of the sediments.



Hilsenhoff and Narf (1968) found no correlation.  Hargrove



(1969)  also found no significant relationship between these



variables.   He found that sediment oxygen consumption is



by the microbiota and that the consumption rate is tempera-



ture-dependent.   The rate was also accelerated by stirring



when oxygen concentrations fell below 6,0 mg/1.



     Hemlock Lake's position on the Tanytarsus-Chironomus



scale is unclear.   Tanytarsus accounted for 27% of the

-------
                            252
emergence during 1969, but only 8% during 1970.  Most of



the other species,  and certainly Chaoborus,  are indicative



of rich Chironomus type lake.  Indeed, different analyses



of the water indicate oxygen-limited conditions in the



profundal zone and high BOD of the sediments at all depths.



The true picture may be obscured by incomplete data.  We



probably missed much of the total emergence because of our



sampling schedule.



     Chaoborus are often categorized as benthic organisms,



but at other times as planktonic.  In fact,  they are both.



The first instar larvae are strictly planktonic and typically



inhabit the lower epilimnion and metalimnion.  They are non-



migratory-  Second and third instars are also planktonic



and exhibit weak diel migration.  Some third instar larvae



inhabit the profundal muds.  Fourth instar larvae have a



definite diel migration pattern (Teraguchi and Northcote,



1966; Roth,  1968).   Typically they nestle in the profundal



muds during the day and migrate to near the surface after



dusk to feed.  About dawn they descend into the mud again.



This migration is apparently temperature dependent.  They



do not migrate during the winter.  During early spring a



small percentage migrate,  but during mid-summer migration is



maximal.  However,  even at that time a certain percentage



of the population may remain in the mud at night.  Laboratory



tests indicate this diurnal rhythm is endogenous, but partly



controlled by light and temperature (LaRow,  1968).

-------
                            253
The diurnal activity pattern persisted as long as 10 days



under total darkness in the laboratory-



     Although Chaoborus larvae are relatively insensitive to



low oxygen concentrations,  they will avoid highly anoxic



conditions such as those found in the monimolimnion of



Hemlock during 1969.  They are often found below the level



of oxygen depletion, during summer and winter, and are ap-



parently unaffected by winterkill conditions  (Northcote,



1964).   During the summer,  fourth instar larvae may remain



in the deep water during the day rather than nestle in the



deep profundal muds (Teraguchi and Northcote, 1966) ,  presum-



ably because of highly anoxic conditions.  In other lakes no



fourth instar larvae are found in the water during the day



(Roth,  1968) .



     Their total distribution in Hemlock Lake is not known



since we did not adequately sample their planktonic distri-



bution.  It is likely that a larger percentage of the fourth



instar larvae remained in the water during the summer 1969,



compared to 1970.  Dredge and emergence trap samples indicate



relatively few larvae below 14.5 m during 1969.  Some of



those supposedly captured below 14 m may actually have been



entrapped by the samplers as they were lowered.  In any



event,  Chaoborus emergence was restricted to 4 m or less



during  1969,  but extended to all depths during 1970-



     If fourth instar larvae were more planktonic during 1969



than during 1970,  we may have underestimated their relative

-------
                            254
abundance during 1969.  Benthic samples indicate a much



larger population during 1970.  Emergence data also indicates



a much larger population during 1970.  Chaoborus constituted



11% of the total midge emergence during 1969, but 28% during



1970.  Their total numbers increased from 47 during 1969



to 296 during 1970.  These data indicate a significant in-



crease in Chaoborus standing crop and production rate during



aeration.



     Increased Chaoborus production may be related to in-



creased food, less severe environmental conditions, and re-



duced predation by fish.  Chaoborus'  preferred prey are



copepods, but they also feed on cladocera,  chironomid larvae,



mosquitoes and other Chaoborus larvae (Stahl, 1966).



Increased densities occurred in all these categories (except



mosquitoes)  during artificial aeration.   Although they can



tolerate anoxic conditions,  they probably are less restricted



under aerobic conditions.  Under aerobic conditions a larger



percentage may have nestled in the mud and thus reduced



their vulnerability to predation.  There was a large increase



in the number of larvae found below 14 m after aeration.



Over 90% of the larvae were found below 14 m during late



August and September 1970,  compared to less than 20% during



1959 (Figures 56 and 57).  As noted,  Chaoborus avoid highly



anoxic conditions.  Highly anoxic conditions were eliminated



by aeration during 1970.  Chaoborus were by far the most



important trout food item before aeration began.  After

-------
                            255
aeration,  trput fed more on Daphnia pulex.  This reduced



predation rate undoubtedly contributed to their population



increase.



     The shift in Chaoborus species composition from C_.



flavicans to C. punctipennis is dramatic.  C. flavicans is



much larger than £. punctipennis and was the only Chaoborus



adult found during 1969.  Almost twice as many C_. flavicans



emerged during 1970 compared to 1969, but almost three times



as many £. punctipennis emerged during 1970 compared to



C_. flavicans.  Some environmental change could have favored



this shift.  C_. punctipennis,  being smaller, may have fed



more efficiently on the small zooplankters such as Bosmina



that increased significantly during aeration.







Crayfish





     Qrconectes virilis depth distributions during August



1969 differ markedly from those described by Momot  (1969)



and Momot and Gowing (19¥0)  for nearby lakes, and from the



August 1969 Section Four crayfish distributions.  In these



other lakes,  the females were found in deeper water than the



males.   In Hemlock Lake, the females were more abundant in



shallow water.  Momot (1967)  proposed that the females'



migration  to deeper water is associated with their sexual



maturation.   Low temperatures and light intensity are associ-



ated with  sexual maturation.   It is difficult to draw definite



conclusions from my data because of small sample sizes and

-------
                            256
lack of information on age distribution.  Yearlings appar-



ently respond differently than mature animals.  However,



these data indicate that there is an avoidance of the males



by the females.  The males have moved into deeper water,



possibly causing the females to move into shallow water.



The females apparently avoided moving into still deeper



water than the males because of the low temperatures.  Less



than 10% of the crayfish were found below 6 m at temperatures



less than 13°C.  Momot (1967) found that 0. virilis also



avoided temperatures of less than 13 C.  In Section Four



Lake,  10 C was the lowermost temperature limit.



     Depth distribution changes during 1970 are most diffi-



cult to explain.  Before aeration began, their distributions



were as expected.  Over 95% were in 0 to 4 m depth at



temperatures above 10 C.   This distribution is similar to



those found in the other lakes at this time of the year.



After aeration began,  oxygen concentration did not limit



their maximum depth distribution nor did temperature seem



to limit them to shallow depths.  Hypolimnetic temperatures



warmed to over 16°C by August 20,  1970, but crayfish were



not found below 10 m.   Some factor other than oxygen or



temperature seemed to limit their depth distribution.  After



the lake destratified in late August,  they distributed to



the bottom.   This indicates that the inhibitory factor was



either distributed evenly throughout the lake, or eliminated



by destratification.  There  are also no obvious differences



in distribution between males and females during 1970.

-------
                            257
Rainbow Trout



     Rainbow trout were restricted to levels above 8 meters



during 1969 by hypolimnion stagnation.  Their upper limit was



related to the 21°C isotherm.  Over 90% of the fish were



captured between these two restraints.  Before aeration began



during 1970, these fish were again restricted to shallow



water by anoxia,  but quickly extended their depth distribution



as the anoxia was eliminated.  An oxygen deficit as well as



other chemical gradients are associated with anoxia.  Hydro-



gen sulfide, carbon dioxide,  ammonia and other detrimental



substances are present in high concentrations below the oxygen



depletion depth.   The trout undoubtedly reacted to these



factors and others in avoiding the anoxic water.  Black et al.



(1954) have shown that carbon dioxide above certain levels



sharply inhibits oxygen utilization by fishes.  The carbon



dioxide concentration at which this inhibition becomes most



apparent is species specific.  Fish adapted to low oxygen



levels,  such as the brown bullhead (Ictalurus nebulosus) are



not as readily affected by carbon dioxide.



     The depth preferences of the 1969 stocked trout are not



well defined after mid-July 1970.  They seem to prefer the



bottom during late July and early August 1970, and then dis-



tribute throughout the lake after late August.  However, this



pattern may be an artifact of small sample size.



     The results  of the acclimatization tests during 1970



are puzzling.   The rainbow trout distributed throughout the



lake very soon after release and did not show preferences

-------
                            258
based on their acclimatization history.  While these  fish



were held in the cages they should have been acclimating to



several factors.  These factors include temperature,  light,



pressure,  oxygen, carbon dioxide, and possibly scent.



Intuitively I expected temperature, light, pressure and



scent to be the most important factors.  Many people  have



demonstrated species specific temperature preferences in



both laboratory and field tests.  Although a given species



will occupy a wide range of temperatures, it generally will



prefer and seek a given temperature.  Bardach and Bjorklund



(1957) found that fish could detect temperature changes as



low as 0.05°C/min.  Changes of 0.2°C/hr. were apparently not



detected.   They trained fish to respond to temperature



changes and thereby demonstrated that fish can detect smaller



changes than they generally respond to.  Various researchers



have found that light (Sullivan and Fisher, 1953; Brett,



1952; Pearson,  1952),  feeding activity  (Brett, 1952;  Pearson,



1952) and social behavior (Pearson, 1952) can interfere with



temperature selection.  Hasler (1966) clearly demonstrates



fishes ability to detect and respond to scents.  Pressure



acclimation can also severely limit short-term vertical



movement of fish responding to temperature differences.



Phyoclists,  fish with closed swimbladders, are especially



restricted against rapid upward movement above their  equi-



librium depth.   Jones (1952) estimates it would take  Perch



(Perca flavescens)  50 hours to adjust to a 90% hydrostatic

-------
                            259
pressure reduction.  Downward movement would be less re-



stricted.  Physostomes,  fish with an open duct connecting



the swimbladder with the alimentary canal, are much less



restricted in their vertical movements.  The expanding air



associated with rapid ascent can be readily vented through



the mouth.  The re-secretion of air following descent is a



slow process however (Ledebur,  1937).   Hodgson and Richardson



 (1949)  report the ascent of prichards from 15 to 7 meters in



3 minutes.  Prichards are marine Physostomes.  Northcote



et al.  (1964)  observed marked diel migrations of sockeye



salmon (Oncorhvnchus nerka) and peamouth chub (Mylocheilus



caurinus) in a British Columbia lake.   The fish migrated



over 10 meters and through a 4°C to 6°C temperature differ-



ence in less than 6 hours.  Their estimates of fish depth



distribution from gill nets agreed well with their echo



sounding observations.   Although they could not document



explanations for these migrations,  food selection appears to



be an important factor.



     Rainbow trout are physostomes and presumably capable



of rapid vertical movements.  This ability should allow them



to rapidly respond to vertical temperature differences or



other factors affecting their depth distribution.  This



ability allowed them to rapidly disperse throughout Hemlock



lake soon after release.   Unfortunately we have no estimate



of their  diel depth distribution.  I would expect their diel



distribution to agree closely with the diel distributions of



their prey.

-------
                            260
     Garside and Tait (1958) and Ferguson  (1958) demon-



strated by laboratory tests that temperature alone can



determine the depth distribution of fishes.  Their tempera-



ture preference is related to their recent thermal history



as well as genetically defined factors.  Fry (1947) defines



their preferred temperature, or short-term preference, as



the "region,  in an infinite range of temperature, at which



a given population will congregate with more or less pre-



cision."  This short-term preference is related to their



recent thermal history.   He also defines the final prefer-



endum temperature as "a temperature around which individuals



will ultimately congregate,  regardless of their thermal ex-



perience before being placed in the gradient."  The final



preferendum temperature,  or long-term preferred temperature,



is largely species specific, although the final preferendum



temperature varies greatly with size and age for a given



species.  Younger fish usually have a higher final prefer-



endum temperature.  The final preferendum temperature is



usually measured as the point at which the preferred tempera-



ture equals the acclimation temperature.  Fish acclimated



above the final preferendum will generally prefer cooler



temperatures,  whereas those acclimated below the final prefer-



endum will generally prefer warmer temperatures.  Ferguson



(1958)  observes that the laboratory determined final prefer-



endum temperature is usually greater than the temperature



selected by most fish species in nature.  He attributes

-------
                            261
these discrepancies to use of younger fish in the laboratory



studies than were observed in nature.  Fry (1937) found in



nature that the young cisco (Leucichthys artedi), tended



to remain in the warm shallow water, while the older fish



moved into deep cooler water.   The selected temperature in



nature can also be altered by physical or chemical limita-



tions such as oxygen depletion (Bendy, 1946,  1948; Botges,



1950) .



     The degree of thermal gradient may also be  an important



factor affecting response to temperature differences.



Although this has not been well documented for fish,  Beeton



(1960) found that vertically migrating Lake Michigan Mysis



relica would not penetrate a thermal gradient of 1.67 C to



2.0°C/m,  but some would penetrate a 0.66°C to 0.94°C/m



gradient.  The latter individuals soon returned below the



gradient, however.  During isothermal periods they would



migrate uninterrupted from the bottom to the surface.



Whether this response is due to preferred temperature,



temperature gradient or density discontinuity,  is unclear.



Harder (1968)  found-many marine zooplankters reacted to dis-



continuities in stratified laboratory cylinders.  These



discontinuities included density gradients due to temperature,



salinity and density without temperature or salinity grad-



ients.   Some organisms react more to these discontinuities



than others.   Littorina spp. veliger larvae were distributed



almost entirely at the discontinuity, while young mysids did

-------
                            262
not significantly change their distribution after a discon-



tinuity was imposed.  The mysids1 negative phototaxis prob-



ably outweighed their possible positive response to the



discontinuity-



     Garside and Tait (1958) ,  using 10 to 15 cm fish in



laboratory tanks, observed a 13 C final preferendum for rain-



bow trout.  This does not agree with the rainbow trout



temperature selections found by Horak and Tanner (1964) in



Horsetooth Reservoir, Colorado.  There the trout were most



numerous in 18.9 to 21.1 C water.  Although they ranged



throughout the reservoir, in temperatures from 7.7 to over



22 C,  over 93 percent were found in or above the metalimnion



in temperatures above 10 C.  Hypolimnetic oxygen,  pH,  carbon



dioxide or bicarbonate alkalinity did not appear limiting.



These trout were larger than those used by Garside and Tait,



ranging in length from 15.7 to 62.5 cm.  Larger fish usually



prefer cooler temperatures.  Although many factors undoubtedly



affected their depth distribution,  Horak and Tanner felt that



food selection was a major factor affecting the trout depth



distribution.  Although copepods were more numerically



abundant than cladocera  (54.5% to 45.5%), the trout fed



selectively on cladocera.  The cladocera were mostly distri-



buted above the thermoclirie.




     Hemlock Lake rainbow trout fed almost exclusively on



Chaoborus larvae and pupae and D. pulex.  Both organisms are



known to exhibit diel vertical migrations.  They inhabit the

-------
                            263
deep,  dimly lit regions of the lake during the day, and the



surface waters at night.  They are presumably most vulnerable



to trout predation during dawn and dusk, the periods of their



vertical migrations.



     The vertical migration of the trout as response to the



vertical migration of their prey could largely explain the



widespread distribution of the trout after aeration began.



Burbige (1969)  found that the American smelt (Osmerus mordax)



underwent a vertical migration within the hypolimnion in



response to their major prey species, Chaoborus.  Although



Chaoborus migrated from the bottom of the lake to the surface,



the smelt did not penetrate the thermocline.  Galligan (1962)



found that lake trout (Salvelinus namaycush) greatly altered



their depth distribution in response to their prey's depth



distribution.  Normally the trout preferred 24 to 31 meter



depths,  but invaded 6 to 24 meter depths in response to spawn-



ing alewife  (Alosa pseudoharengus)  on which they preyed



heavily.  It is my conclusion that as long as certain physical,



chemical,  or behavior factors do not absolutely limit the



distribution of fishes,  then the fishes will distribute in



response to their prey's distribution.



     We can define adequate oxygen and temperature conditions



for trout as oxygen concentrations above 5.0 mg/1 and



temperatures less than 24°C.  Trout can withstand tempera-



tures as great as 26.7 C for a few days, but prolonged



temperatures above 24°C lead to high mortality  (Eipper, 1960).

-------
                            264
Likewise,  they may also withstand oxygen concentrations of



less than 5.0 mg/1 at low temperatures,  but 5.0 mg/1 is



generally considered a safe lower limit.  Within these con-



straints we can clearly see that artificial hypolimnion



aeration greatly increased the "lebensraum" available to



the rainbow trout (Figure 69).   During early August 1969,



24% of the lake had adequate oxygen,  but the temperature was



too great.  Forty-one percent had adequate temperature,  but



the oxygen was too low.  Thirty-five percent had both ade-



quate oxygen and temperature.   During the same period 1970,



over 97% of the lake had both adequate oxygen and temperature



due to artificial hypolimnion aeration.   If the aeration tower



had operated as intended,  the surface water would still be



too warm,  but in our case,  the surface waters were apparently



cooled slightly because of the faulty design and construc-



tion.  In many lakes, especially in the American southwest,



the portion of the lake with adequate oxygen and temperature



approaches zero due to very high surface temperatures and



hypolimnetic oxygen depletion within and below the metalim-



nion (Fast,  1968; Fast and St.  Amant,  in preparation) .   Many



of these lakes are eutrophic and owe their richness in large



part to the high nutrient content of their basins,  water-



sheds and water sources.  This high nutrient content causes



plankton blooms,  hypolimnetic oxygen depletion and many



other conditions deleterious to water management.  Because of



water shortages,  these lakes are managed intensively for

-------
Figure 69.  Hemlock Lake oxygen and temperature conditions for trout during
            August 1969 and August 1970.  Adequate temperature is tempera-
            ture less than 24 C,  and adequate oxygen is values of 5 mg/1 or
            more.

-------
^•^:-}i\\-:'< Adequate
^:M£.f£ Temperature


        Adequate Oxygen
        And Temperature
Adequate
Oxygen
  70    60   50   40    30    20   10
                10    20   30    40   50   60   70
                          Distance  From Aerator (m)

-------
                            267
multiple uses.   Many water managers have attempted to allevi-



ate the deleterious eutrophic conditions by artificially



destratifying their lakes with compressed air.  Artificial



destratification of eutrophic lakes is a useful fisheries



management technique (Fast, 1968; 1971).  It increases the



habitat and food available to many fishes.  It reduces the



probability of oxygen depletion by algal decay and respira-



tion.  It is most widely used to improve domestic water



quality.  With continued complete mixing, it also increases



the heat budget and eliminates the deep, cold hypolimnetic



water.  This water is generally anoxic.  While artificial



destratification eliminates this anoxia, it also greatly



increases the bottom temperature.  After continued destratifi-



cation,  the entire lake is about the same temperature as the



surface before destratification began.  These reservoirs,



whether destratified or not,  presently support year-round



warmwater fisheries,  but coldwater fisheries only when sur-



face temperatures are less than 24 C.  Development of such



reservoirs for year-round coldwater fisheries will greatly



increase their fisheries potential.  High angler demand for



trout coupled with the usual location of such reservoirs



near populous areas would result in manyfold increases in



angler patronage.



     Presently,  the only known method of creating suitable



coldwater habitat in an eutrophic lake is by artificial



hypolimnion aeration.   This system of aeration can result in

-------
                            268
adequate oxygen values throughout the lake without intoler-



able increases in hypolimnion temperatures.  Oxygen can be



added to  the hypolimnion without mixing it with, or heating



epilimnetic or metalimnetic water.  This may be accomplished



through modifications of my basic hypolimnion aerator



design,  or use of other designs.



     Artificial destratification of oligotrophic lakes may



result in the elimination of the coldwater species.  The



oxygen concentration may be little affected,  but the bottom



temperature will be greatly increased.  Although destratifi-



cation will not greatly affect water quality, it may be con-



ducted to reduce evaporation rates.  Annual evaporation may



be reduced from 4 to 10% (Koberg,  1964; Koberg and Ford,



1965; and personal communication).  From the fisheries stand-



point,  artificial destratification of oligotrophic lakes



can be deleterious.  An exception is when stratified surface



temperatures normally are not limiting to the cold water



biota.   In this case destratification may increase fish food



and trout production.



     After artificial hypolimnion aeration is initiated,  a



system for stocking coldwater fish into the cold,  aerated



hypolimnion must be developed.  Stocking the trout at the



lake's surface,  in the usual manner, will cause mortality



due to thermal shock.  Sharpe (1961) transported rainbow



trout in plastic bags to the hypolimnion of a Tennessee



reservoir.  Some of these fish made short excursions towards

-------
                            269
the surface,  but  soon  returned to the hypolimnion.   Trout



stocked  at  the  surface were greatly distressed and  dis-



oriented; many  died.   A more appropriate method of  stocking



could be devised.   A simple method would use a length of



irrigation  pipe extending from the shore,  along the bottom



into the hypolimnion.   Water could be pumped from the hypo-



limnion  through a separate hose and used to flush the fish



into the hypolimnion.   Fish could then be stocked directly



into the pipe from a hatchery truck.

-------
                    SECTION FOUR LAKE








Destratification System





     The Section Four Lake artificial destratification system



was very similar to the diffuse aerator described by Fast



(1968).   A 38 mm I.D. plastic pipe conducted air from the



rim of the lake's basin to the deepest point in the lake



(Figures 70 and 71) .   The distal 19 meters were perforated



with 48 holes.  The holes were 3.2 mm in diameter and located



at 12 sites.  Four holes were located at each site,  posi-



tioned at 90  intervals around the pipe.  The hole sites were



unevenly distributed along the 19 meter section.  From the



distal end, they were positioned in meters as follows:  0.0,



0.6,  1.2,  2.5, 3.7, 4.9, 7.4, 9.8, 12.9, 16.0, and 19.0.



The air line portion leading from the compressor to the shore



was covered with soil to prevent damage by porcupines.



Porcupines chewed the tires on the compressors and the plywood



decks on the barges at both lakes, but were not a problem



after we began aerating.  The sound of the compressors appar-



ently kept them away.  The air line portion in the lake was



anchored with several cement blocks.  These were tied



directly to the air line.
                            270

-------
Figure 70.  Cross-sectional view of Section Four diffuse aeration
            system.  The air was released from the last 10 meters of
            pipe, situated near the deepest point in the lake..

-------
Compressor
                                                                         Cement  block  Anchors

-------
Figure 71.  View of Section Four Lake taken from the basin rim.                      tj
            Rising air and water is seen near the center of the lake.                ^
            (Photo by author.)

-------

-------
                            275
Compressor.   We used a Jenbach JW78 diesel compressor.  It



was always run at maximum speed and delivered 78 cfm at 100



psi.   It was fueled with five gallons of diesel each day and



ran about five hours.   On a few occasions it was fueled two



or three times on a given day.





Compressor Operation.   Aeration began June 16, 1970.



Destratification was rapid but the bottom 3-4 meters were



not being circulated.   Two factors contributed to this situa-



tion:   (1)  The end of the air line was not adequately



anchored; the air line bowed up and was four to six meters



off the bottom.  Most of the air escaped at the top of the



bow due to reduced hydrostatic pressure, and  (2)  The air



was not released over the deepest point in the lake due to



the air line bow.  Knoppert et al. (1970)  produced a similar



result by purposely positioning their air line at an inter-



mediate depth.  Their lake was totally mixed at depth



shallower than their air release,  but not deeper.



    On June 20,  1970,  I added additional anchors to the air



line using SCUBA.  Thereafter the air release was more uni-



form along the perforated section, and extended to the



deepest point in the lake.  The entire lake was then circu-



lated.

-------
                         RESULTS








Physical-Chemical Parameters





Temperature and Oxygen.  Section Four Lake stratified nor-



mally during 1969.  By early June, a thermal gradient



extended from 3.5 to 12 meters  (Figure 72).  Strictly speak-



ing  (i.e.,  a thermal gradient of l°C/m)  the metalimnion



extended from 3.5 to 7.5 meters.  This definition seems



somewhat arbitrary in this case since the gradient decreased



uniformly with depth.  Although the gradient between 7.5 and



12 meters was less than 1°C,  it appears to represent stable



stratification.  Temperatures ranged from 7°C at the bottom



to 16 C at the surface.  During this same period oxygen



concentrations were 8.7 mg/1 at the surface and 9.5 mg/1 at



the bottom (Figure 73).  Metalimnion oxygen maxima occurred



during the summer 1969,  but were not as pronounced as in



Hemlock Lake (Figure 36) .   As the summer progressed,  the



thermal gradient region increased in depth.  By late August



it extended from 5.0  to 14.5  meters.   The isotherms are



nearly parallel during the entire summer.   Maximum surface



temperatures of over  23°C were observed during mid-July and



again during mid-August.   These surface temperature maxima



coincide with maxima  for  Hemlock Lake.   Minimum bottom
                               276

-------
                                                                                      to
Figure 72.  Section Four isotherms during the summer, 1969, before                    
-------
 July
10     20
10
August
    20

-------
                            279
Figure 73.   Section  Four  selected oxygen profiles  during
            the summers  1969  and 1970.   Aeration occurred
            between  June  16 and September 7,  1970.

-------
               Oxygen  (mg/|)
a
a>
O
03690369
n i i 1 i i 1 i i ii i i 1 i i 1 t i 1 i i

-
-
-
-
5-
-


10-
_
-
15-
-






5-


10-



15-



/
j
	 1970 f
___1969



VI-7




J
/I
\
\
,
/
1

i
, VI-18
1

-1

\
1
I
VII-30!
v
X^

i
|
1
I
'
1
1














f /
VII-30
\ "X^^
\ \
V |
\
\
\
I
\
1
1
VIM,
1
/
/ VII-30

\ /
y
j.
1

/
fjX-2
\ V.
IX-2\ ;'
1
/ 1
( 1
\ I
I /
L'
s 1
/ 1


-------
                            281
 temperatures were stable, but increased about 1 C during



 the summer 1969 (Figure 74).  Average and surface oxygen



 concentrations varied considerably during the summer 1969,



 but were always above 6.0 mg/1  (Figure 75).  Bottom oxygen



 concentrations decreased during the summer 1969 but were



 always above 4.0 mg/1.



     Before aeration began during June 1970, the thermal



 regime was similar to that of June 1969 (Figure 75).



 Temperatures ranged from 21 C at the surface to 5.5 C at



 the bottom.  The thermal profiles were similar although the



 surface was warmer,  and the bottom slightly cooler during



 1970.  Oxygen values were likewise very similar to 1969



 values.  Oxygen concentrations ranged from 8.0 mg/1 at the



 bottom to 9.2 mg/1 at the surface (Figures 73 and 74).



     Artificial air injection began June 16, 1970 and caused



 immediate and significant changes in the temperature regime



 (Figures 76 and 77) .  After four days of aeration the surface



 temperature decreased from 21°C to 14 C and much of the lake



 was isothermal.  By July 1st,  the lake was nearly isothermal



 at 16 C except for the upper meter which approached 25 C.



 By July 10th the entire lake was isothermal at 18.5°C.  The



 average temperature increased gradually and reached a 23.3 C



maximum,  compared to a 19.7°C maximum average during 1969.



Surface temperatures were not greatly altered by artificial



elestratification,  but bottom temperatures reached a 23.3 C



maximum during 1970,  compared to 8°C during 1969.  With

-------
Figure 74.  Section Four maximum, minimum and average temperatures ( C)
            during the summers 1969 and 1970.   Aeration occurred between
            June 16 and September 7,  1970.

-------
27-




244
                                     Section Four
         1969
u
   20--
5  16-+
0)
a

£
1970
     1   10  20   1   10   20   1   10  20   11   10  20   1   10  20   1  10  20    1


        June      July      August         June       July     August

-------
Figure 75.  Section Four top, bottom and average oxygen concentrations               oo
            during the summers 1969 and 1970.  Aeration occurred between             *"
            June 16 and September 7, 1970.

-------
  10.0-
O)

£8.0-

c
0)
O)
X
O
  6.0-
  4.0-
        1969
                                 MEAN
                                 BOTTOM
10   20    1   10   20    1   10  20

June
                                         T~
                                         1
 1970
10   20    1   10   20    1   10  20    1
                    July       August        June
            July
August

-------
                            286
Figure 76.   Section Four selected temperature profiles
            during the summers 1969 and 1970.  Aeration
            occurred between June 16 and September 7,
            1970.

-------
                      Temperature  (°C)
    0-
    12H
0)
Q
           8    12    16   20   24 4    8   12   16   20
          _j	I	I	i	i  I	I	i	i     i
                       7-29-70
9-2-69
   /
  /
 /
/
                                                  9-2-70

-------
Figure 77.  Section Four isotherms during the summer 1970.  Aeration                  r>o
            occurred between June 16th and September 7th.  Isotherms                  °°
            are in  C.

-------
         June
         10      20
  July
10      20
         August
1       10     20
0-
                                i Mijm=MJjiLU4-f-niiifim 11 inuiii 11 in

-------
                            290
continuous aeration,  the entire lake became about as warm



as the surface waters were during stratified periods  (Figure




74).



     Thermal stability was 7x108 kg-m on June 10, 1970,



compared to 5x108 kg-m the previous year on this date



(Figure 78).  After 4 days of artificial aeration stability



neared zero.  It was always near zero after July 29, 1970,



whereas it reached a maximum value of over 12x10® kg-m on



August 20, 1969.  These changes reflect the degree of strati-



fication and indicate the lake was nearly isothermal during



most of 1970.



     Oxygen values were more uniform after aeration began.



Surface values were about the same,  but bottom values were



always above 7.0 mg/1 compared to 4,5 mg/1 during 1969.



Oxygen maxima at intermediate depths were not present during



1970, and the oxygen profiles were nearly vertical.





pH, Alkalinity and Conductivity.  During 1969, pH values



were variable,  but always above 7.5 (Figure 79).  Bottom



values were always lower than surface values,  but on July



14, 1969 the average pH was higher than the surface pH.  The



average pH increased from 7.7 on July 14, 1969 to 8.0 by



September 1969.  After aeration began the pH was very uni-



form throughout the lake and nearly constant at 7.9 through-



out the summer 1970.




     Alkalinity was quite consistent during 1969.  Bottom



values averaged about 190 mg/1 and surface values about 170



mg/1 (Figure 80).  After aeration began alkalinity was nearly

-------
                            291
Figure 78.   Section Four  stability  values  during  the
            summers 1969  and  1970.  Aeration  occurred
            between June  16 and  September  7,  1970.

-------
£
Q>
         10
          June
August

-------
Figure 79.  Section Four's bottom, top and average pH" values during
            the summers 1969 and 1970.  Aeration occurred between
            June 16 and September 7, 1970.

-------
   8.2-
X
Q,
   7.4-
   7.0-
        1969
                                   TOP
                                /   MEAN
                           BOTTOM
                 1—i—I
                                       1970
                                              —i—'—|—i—i—i—i—i—|—i	1-
                                              10  20    1   10  20    1

                                              June       July
10   20    1   10   20    1   10  20

June       July       August
August

-------
Figure 80.  Section Four bottom,  top and average alkalinity values
            during the summers 1969 and 1970.  Aeration occurred
            between June 16 and September 7,  1970.

-------
  225-
*T 200-
C
<
15
H
         1969
                       BOTTOM.

                       MEAN
        —i—i—i—i—i—i—i—'—i—'—i—'—i—'—i—
         10  20   1    10  20    1   10   20
         June        July       August
1970
10   20    i   l6  ' 20    1   l6 ' 2b   f
June       July       August

-------
                            297
 constant at all depths.  Average alkalinity reached a maximum



 of over 200 mg/1 during 1970 compared to a 185 mg/1 maximum



 during 1969.



     Conductivity followed the same general pattern as



 alkalinity during 1969 (Figure 81).  Surface values averaged



 about 350 micromhos and bottom values about 410 micromhos.



 After aeration began,  conductivity was somewhat erratic



 until mid-July.  It was nearly uniform on any given date



 after aeration, but average values ranged between 365 and



 415 micromhos during 1970, compared to 370 to 385 micromhos



 during 1969.





 Phosphorus.  Phosphorus concentrations were measured only



 during 1969.  On July 22,  1969 total phosphorus ranged from



 0,002 to 0.062 mg/1 (Table 7).  Its distribution is erratic.



 Total dissolved phosphorus was more uniform, but near the



 limit of resolution for the analytical method used.  It



 ranged from 0.002 to 0.010 mg/1.





 Ca, Na,  K,  Mg, DOM and POM.  These constituents were measured



 only during the summer 1970.  Average calcium decreased from



 56.0 mg/1 during mid-June to 52.6 mg/1 by August (Table A-12).



 Calcium,  like the other constituents,  does not exhibit much



 vertical variability,  even before aeration began.  Average



 sodium and potassium concentrations were nearly constant all



 summer.   Average magnesium concentrations varied from 16.0



mg/1 to  9.8 mg/1.  Dissolved organic matter and particulate

-------
Figure 81.  Section Four bottom, top and average conductivity values
            during the summers 1969 and 1970.  Aeration occurred between             [^
            June 16 and September 7, 1970.                                           oo

-------
  420-
E

2 390-|


£
>
+-

'>

"o 360-
3
•o
C
O
O
  330-
        1969
            BOTTOM
             MEAN
             TOP
                          N
                            \
                                               1970
        TO '  20    i
        June
                    10 ' 20   \  ' 10   20

                    July      August
10   20

June
T  ' ~| i   1	1	1	1	1	1	1—
10   20    I    10  20    1


July      August

-------
Table 7.  Section Four Lake total phosphorus and total
          dissolved phosphorus collected July 22, 1969,
          Two water samples were collected from each
          depth interval.

Total Phosphorus
(ng/1)
Depth
0.0
2.8
5.6
8.3
11.0
13.7
16.5
17.8
Xl
0.002
0.003
0.003
0.025
0.025
0.003
0.100
0.025
X2
0.017
0.003
0.003
0.002
0.025
0.002
0.025
0.000
Mean
0.010
0.003
0.003
0.014
0.025
0.002
0.062
0.012
Total Dissolved
Phosphorus (mg/1)
xl
0.002
0.003
0.002
0.002
0.002
0.003
0.003
0.002
X2
0.017
0.003
0.002
0.002
0.001
0.002
0.002
0.002
Mean
0.010
0.003
0.002
0.002
0.002
0.002
0.002
0.002

-------
                            301
organic matter were most variable.  Average DOM ranged be-



tween 2.74 mg/1 to 6.18 mg/1, whereas DOM ranged between



117.1 mg/1 to 436.2 mg/1.





Primary Production





     Section Four appears very unproductive.  Phytoplankton



and attached algal densities are sparse.  Secchi disc read-



ings typically exceed 10 meters.  Chara may be the main



source of primary production since dense beds extend over



as much as 10% of the lake's bottom.  These plants were most



abundant below 5 meters and extended to maximum depth both



years.  No estimates of the abundance or production rates



were made because of the obvious technical problems in-



volved.  Instead, I concentrated on the phytoplankton and



periphyton in hopes that relative changes in these two



components would represent relative changes in the lake's



primary production as a whole.





Phytoplankton.  Phytoplankton standing crop and production



rates were relatively low both years.  Standing crop was



about one-tenth that found in Hemlock Lake,  and production



ranged between one-fifth and one-thirtieth of Hemlock's.



Primary production potential during July and August 1969 was



fairly constant at 6 mg/m3/4 hrs. (Figure 82).  Surface



standing crop averaged about 200 cells/ml during June and



early July,  but increased to about 600 cells/ml during August



1970.  Secchi measurements reflect this change by decreasing



from about 12 m to 10 m.

-------
                            302
Figure 82.   Section Four secchi disc transparencies,
            surface primary production potentials,
            surface phytoplankton densities and sur-
            face production efficiencies during the
            summers 1969 and 1970.   Aeration occurred
            between June 16 and September 7,  1970.

-------
-C
«J
u

CO
o-

2-

4-




8-

10-

12
   30-
   20-
    10-
CM
o
 0)
u
0-

8-



6-



4-



2-



0-
    3-
    2-
    1 -
    o-
        1970
                   1970
                            1969
          1969
                  1970
                            1969
                        1 -  i
                                              i     i
      1     10    20    1    10   20     1     10   20     1

         June           July            August

-------
                            304
     Artificial destratification during 1970 resulted in
increased primary production potentials.  Potentials during
August 1970 were about three times as great as during
August 1969.
     Phytoplankton standing crops did not reflect the in-
creases in production potential.  On the contrary, average
1970 standing crops were less than 1969.  Phytoplankton
standing crops follow a similar pattern during June and July
both years.  Concentrations increased from 200 cells/ml
during June 1970 to a maximum of over 600 cells/ml during
the first week of August.   A sharp reduction back to 200
cells/ml followed.  Secchi disc measurements responded some-
what differently during 1970.  Secchi measurements were about
the same during June both years, but decreased to 6 meters
by August 1970,  compared to 10 meters during August 1969.
This greater decrease occurred during 1970 even though sur-
face phytoplankton densities were somewhat less than during
1969.  Furthermore,  Secchi measurements showed only a slight
increase following the phytoplankton population decline dur-
ing August 1970.  The lower transparency during 1970 is
probably due to suspension of detritus, sediments and other
materials by the compressed-air-generated water currents.
Brian Moss, who examined the samples, indicated that larger
quantities of such materials were present in the 1970
plankton samples than in the 1969 samples.

-------
                            305
Periphvton.  Seventeen-day periphyton accumulation rates



during June 1970 were less than during June 1969  (Figure 83).



Thereafter, the 1970 values were always larger, with the



possible exceptions of July 15th and August 30th.  Average



17-day accumulation rates were about 0.004 gm/day during



1969 and about 0.007 gm/day during 1970.



     Differences in 17-day periphyton accumulation rates



were not reflected in total accumulation rates.  Total accu-



mulation rates were lower during June 1970, greater during



early July 1970, but not different during late July and



August.  As with Hemlock, these total accumulations were



almost linear during both summers.  Accumulations increased



from near 0.0 gm to about 0.5 gm both years.





Zoobenthos



     Oligochaetes and Chironomids dominated the Section Four



benthic fauna assemblage, exclusive of the crayfish.



Together,  they comprise about 93% of the numbers and 90%



of the biomass during 1969 (Table 8 and Figure 84).  During



1970,  these percentages were 94 and 84 respectively.  Ten



other taxa comprised the remaining macrobenthic fauna.



Chironomids were numerically more abundant than Oligochaetes



during 1969,  but this situation was reversed during 1970.



Oligochaetes had more biomass than any other taxa both



years.



     Zoobenthos biomass collected during 1969 totaled 11.9



gm and 7.93 mg during 1970 (Table 8).  This is a 28% decrease

-------
                            306
Figure 83.   Section Pour periphyton standing crops based
            on 17-day incubation periods and continuous
            incubation.   The 95% confidence interval is
            shown about  each average value.  Aeration
            occurred between June 16 and September 7,
            1970.

-------
   .015-
 «
 
u_
 I



.
JT 0.2—

3
E
3

-------
                            308
Figure 84.   Section Four zoobenthos percent composition
            during the summers 1969 and 1970.   Percent
            of weight and percent of number are shown
            for each taxa.   Total weights and total
            numbers collected each summer are also shown.
            Samples from dredge collections only.

-------
70-,
                                     Total Number: 1969-17,609
                                                   1970-11,979
                                                   I11969
                                                   FT""] 1970
                              _  _ CN CTi -«r "O o O «N  {K •— CO O -^
                              d  d d d d ° d o --'  d d d d d
                                      Total Weight:  1969=11.09227
                                                   1970= 7.93214
                                                   I11969
                                                   FZ31970
                              d CH odd do d  odd  odd

-------
Table 8.  Section Four Lake zoobenthos collected durinp the SUPPCTS 1P6P and 1970 with an Ekman dredge.
          taken each sunnier.  I'.'et weights are shown.
                                                                                                        125 dredee samples were

Total
1969

Oligochaetes
(microdriles)
Oligochaetes
(megadriles)
Chironomid L.
Chironomid F.
Amphipods
Dragonf lies
Damself lies
Mayflies
Chaoborus spp.
L.
Clams
lleleidae
Trichoptera
Tabaniid
Megaloptera
Grams
6.75189
1.54189
2.16182
0.05490
0.26445
0.09587
0.03211
0.05025
0.00000
0.03355
0.05561
0.02433
0.02278
0.00280
Percent
60.9
13.9
19.5
0.5
2.4
0.9
0.3
0.5
0.0
0.3
0.5
0.2
0.2
0.0
Weight
1970
Crams
3.86228
2.19364
0.59028
0.63301
0.25269
0.16457
0.05024
0.02288
0.00075
0.05209
0.02733
0.02772
0.05348
0.00118
Percent
48.7
27.7
7.4
8,0
3.2
2.1
0.6
0.3
0.0
0.7
0.3
0.3
0.7
0.0
Total
1969
Number
7,506
27
8,817
336
583
11
31
62
0
1
203
26
5
1
Percent
42.6
0.2
50.1
1.9
3.3
0.1
0.2
0.4
0.0
0.0
1.2
0.1
0.0
0.0
Numbers
1970
Number
6,684
12
4,041
163
771
13
32
61
1
8
102
38
45
5
Percent
55.8
0.1
33.7
1.4
6.4
0.1
0.3
0.5
0.0
0.1
0.9
0.3
0.4
0.0
Number of
Samples Found In
1969
89
10
121
74
60
7
18
35
0
1
54
20
4
1
1970
110
6
123
71
72
10
23
34
1
7
45
27
18
5
Mean Number of
Individuals/gram
1969
1,112
18
4,078
6,120
2,205
115
965
1,234
--
30
3,650
1,068
219
357
1970
1,730
5
6,846
257
3,051
79
636
2,666
1,333
154
3,752
1,371
341
4,237
Total
                 11.09227   100.0
                                     7.93214
                                               100.0
17,609.   100.0   11,976
100.0

-------
                            311
in total biomass during the summer of destratification.



Total numbers collected decreased 32% during this period



from 17,609 during 1969 to 11,976 during 1970.  Decreases



in oligochaetes (microdriles)  and chironomid larvae account



for most of this change.  Interestingly, associated with



the decreases in biomass and numbers during 1970, the



average size of oligochaetes,  Chironomid larvae, amphipods,



mayflies, Trichoptera, tabaniids and megaloptera also de-



creased substantially.  Many of the minor species became



relatively more abundant following destratification



 (Table 8) .



     Jarl Hiltunen could identify only two oligochaete



microdrile species from Section Four:  Limnodrilus hoffmis-



teri and Ilyodrilus ternpletoni.  L. hoffmeisteri was much



more abundant in the samples from which the specific identi-



fications were made.  Several large oligochaete megadrile



specimens were collected,  but not identified.  These speci-



mens closely resembled the common earthworm, and were only



found in a gravel outcropping at 10 to 15 meters depth.



Hiltunen tentatively identified these as Lumbricidae.



     Oligochaetes were very abundant both years at depths



below 14 meters (Figures 85 and 86).  Their average depth



for both numbers and biomass always averaged between 15 and



16 meters.  There appears to be a slightly greater concen-



tration of worms,  especially small individuals, in shallow



water after early July 1970 compared to 1969.  Thermal



Stratification during 1969 may have led to their migration

-------
                            312
Figure 85.   Section Four oligochaete (microdriles)  depth
            distribution as percent of numbers during
            the summers 1969 and 1970.   Shaded histograms
            represent aerated periods.

-------
°— 1

-
75—
50^
25~
0-
—
-
75~

-------
                            314
Figure 86.   Section Four oligochaete (microdriles)  depth
            distribution as percent of wet weight during
            the summers 1969 and 1970.   Shaded histo-
            grams represent aerated periods.

-------
75—
5°-=
25-=
0 ~
75-=
50^
25—
r~-
£ o =
O) -
« 75-^
M- 50—
o -
*•* T* -
c 25—
0)
0 ~
»- 0 —

-------
                            316
into deeper water as the epilimnion warmed.  The June 15,



1969 and July 15, 1969 samples contain some worms at all



depths,  but on July 25th and thereafter none were collected



above 7 meters.  When the lake was completely mixed after



mid-June 1970, their depth distribution during the entire



summer resembles the June distribution.



     Total oligochaete (microdriles)  biomass during 1970



decreased 43% from the 1969 total, and total numbers de-



creased 11% during this period (Table 8).   The indication is



that this decrease occurred after aeration began during 1970



(Figure 87).   Total estimated oligochaete numbers and biomass



were greater during mid-June 1970 than during mid-June 1969,



but the 1970 totals decreased after aeration began such that



by mid-July 1970 and thereafter both numbers and biomass



were less than the 1969 totals.  Large variances obscure the



significance of these trends,  but intuitively I believe they



represent the true population responses.



     Oligochaete (microdrile)  average size decreased during



1970 compared to 1969 (Table 8).   They averaged 1,112 indi-



viduals per gram during 1969,  but 1,730/gm during 1970.



     D.  R. Oliver identified more than 11 chironomid midges



based on adult emergence (Table 9).  Four species, Lauter-



borniella. Prpcladius.  Ablabesmyia mallochi and Clinotanvpus



thoracicus accounted for 67% of the total emergence during



1969 and 77% during 1970.  Unidentified adult Ghironomidae



accounted for 10% of the total emergence during 1969 and 13%

-------
                            317
Figure 87.   Section Four total estimated oligochaete
            number and biomass during the summers 1969
            and 1970.   One standard error is shown
            about each estimate.   Aeration occurred
            between June 16 and September 7,  1970.

-------

-------
Table  9  .  Emergent midge adults collected from 600 samples during 1969, and 650 samples during 1970.
             are from Section Four Lake and were collected in emergent insect traps.
                                                 All specimens
1969

Family - Chironomidae (Tendipidae)
Subfamily - Chironominae (Tendipedinae)
Lauterborniella spp.
Tany tarsus spp.
Polypedilum spp.
Chironomus spp.
Paratanytarsus spp.
Pagastiella spp.
Unidentified Chironomini (Tribe)
Subfamily - Pelopiinae (Tanypodinae)
Procladius spp.
Ablabesmyia mallochi
Clinotanypus thoracicus
Subfamily - Hydrobaeninae
Cricotopus spp.
Unidentified Adult Chironomidae
Total
Number


128
15
38
13
2
3
116

217
140
59
2
79
Percent
of Total
Number


16
2
S
2
1
1
14

27
17
7
1
10
No. of
Samples


51
13
18
11
2
2
57

70
73
35
2
47
No. of
Dates


32
12
17
8
2
1
38

38
41
24
2
34
Total
Number


96
5
3
22
5
0
34

57
307
101
0
98
1970
Percent
of Total
Number


13
1
1
3
1
0
5

8
42
14
0
13
No. of
Sampl es


42
5
3
14
5
0
28

47
130
69
0
65
No. of
Dates


32
5
2
11
5
0
26

31
53
34
0
40
                                Totals
300
                                                                   381
728
100
                                                                                                             408

-------
                            320
during 1970.  These adults included Pseudochironomus netta,



Endochironomus,  Tanypus,  Dicrotendipes fumidus, Cryptochirono-



mus fulvus,  Ablabesmyia monlis,  Procladius bellus and possibly



other species.  We had some problems of identification, and



added our counts to the "unidentified" category when there



was doubt as to the true identity.



     Chironomid larvae were abundant at all depths both years



(Figures 88 and 89) .   Both years they were numerically most



abundant in shallow water during June, but more abundant in



deep water by September.   Biomass distribution follows a



similar trend, but the shallow water shift during June is not



as pronounced.  This indicates that the deep water larvae



are much larger individuals,  either because they are in later



instars and/or different species.



     Total chironomid larvae collected decreased from 8,817



during 1969 to 4,041 during 1970, and biomass decreased from



2.16182 gm during 1969 to 0.59028 gm during 1970.  These are



54% and 73% decreases respectively.  Average size also



decreased from 4,078/gm during 1969 to 6,846/gm during 1970.



Contrary to changes in oligochaete population trends, Chiron-



omid larvae were less abundant during the entire summer 1970,



compared to 1969 (Figure 90).   The mid-June samples were taken



before aeration began.  These first samples indicate that



the larvae were numerically less than 50% as abundant during



1970,  but almost as abundant in regards to biomass.



     Chironomid pupae show the same numerical shift from



shallow to deep water during the summer 1970 as do the

-------
                            321
Figure 88.   Section Four Chironomid larvae depth distribu-
            tion as percent of number during each sampling
            period during the summers 1969 and 1970.
            Shaded histograms represent aerated periods.

-------
45
                i l l  I l I l1  i ill   i |" I i
                                        15
                   Depth (m)

-------
                            323
Figure 89.   Section Four Chironomid larvae depth distribu-
            tion as percent of wet weight during each
            sampling period during the summers 1969 and
            1970.   Shaded histograms represent aerated
            periods.

-------
  60
  40-
  20-
   0

  60-1
       6-15-69
                          6-12-69
                          X=9.4
       7-5-69
       x=n.5
C
0>
o
       7-25-69
       X=12.6
a, 40-

s20-
•o o-
'E
*" 60
O

J4°-

020-
                          7-3-69
                          X=9.5
                          7-24-69
                          X=11.7
8-15-69
X=14.6
                                8-14-69
                                X=12.5
  60-


  40-


  20-
9-6-69
X=13.3
                                 9-4-69
                   1 I ' i  '
              12    16    o
                                          8
                                               12    16
                       Depth  (m)

-------
                            325
Figure 90.   Total estimated Chironomid larvae number and
            wet weight in Section Four during the summers
            1969 and 1970.   One standard error is shown
            about each estimate.   Aeration occurred be-
            tween June 16 and September 7,  1970.   Totals
            from dredge samples only.

-------
o
-^ 60-

 o>
 o
-o 45-

'i
 o
 c
 o

;! 30-

U
   15-
 E
 =>
Z
                                                           1970


                                                           1969
o
•^ 25-
 O

 E
 o 20

 CO

 (1)
 o

 t 15
 o
 o
 c
 o
    10-
    5-
              T
                      \
           10    20      j     ib  '  20      110 ''  20  'T
            June
July
August

-------
                            327
larvae (Figure 91).   Over 55% by numbers were collected from



the shallowest interval during June 1970 and 5% from the



deepest.   By September,  only 20% were collected from the



shallow interval and over 30% from the deepest.  The June



1970 pupal biomass distribution is more uniform than their



numeric distribution,  but by September 1970 over 80% were



collected from the deepest intervals  (Figure 92).   Pupal



distributions during 1969 are not as definite as during 1970,



but indicate greatest concentrations between 4 and 11 meters



during most of the summer.  Few pupae were collected below



15 meters at any time and none were collected below 15



meters during July and August 1969.



     Total pupal biomass collected increased to 0.63301 gm



during 1970 from 0.05490 gm during 1969.  However, total



numbers decreased from 336 during 1969 to 163 during 1970



(Table 8) .  This represents a considerable increase in average



size of the pupae during 1970.  They increased from 6,120/gm



during 1969 to 257/gm during 1970.  Total estimated pupal



population trends are less spectacular  (Figure 93).  Both



numerically and biomass-wise pupae tended to be more abundant



during early summer 1969 than during 1970,  and then tended



towards equal or slightly higher values by late summer 1970.



     Total adult emergence reflects the decreases in larval



and pupal standing crops.  Only 72 adults emerged during




1970 compared to 812 for 1969 (Table 9).  Although total



emergence was less during 1970,  they occurred in more samples

-------
                            328
Figure 91.   Section Four Chironomid pupae depth distribu-
            tion as percent of number during each sampling
            period during the summers 1969 and 1970.
            Shaded histograms represent aerated periods.

-------
        I I I I I I I I I I I I I I I
o
                  Depth  (m)

-------
                            330
Figure 92,   Section Four Chironomid pupae depth distribu-
            tion as percent of wet weight during each
            sampling period during the summers 1969 and
            1970.   Shaded histograms represent aerated
            periods.

-------
                    12   16
Depth  (m)

-------
                            332
Figure 93.   Total estimated Chironomid pupae number and
            wet weight in Section Four during the summers
            1969 and 1970.   One standard error is shown
            about each estimate.   Aeration occurred between
            June 16 and September 7,  1970.  Totals from
            dredge samples  only.

-------
0)
D
Q.
 E
 o


I   2-
_c
U
 0)
J2

 E
 3

Z
                                                             1970

                                                             1969
    9-
 IO


 I  6'
 O

CD

 0)


 I  3
Q.

Tl

'i

 g  o
 O
—1—'	1
 10     2i


  June
                       1—'—T-
                        1    10
20
                               July
10    20

 August

-------
                            334
during 1970 than during 1969,  408 compared to 381.  This



change is reflected in a more uniform depth emergence follow-



ing destratification (Figure 94).   During 1969, almost all



emergence was limited to less than 4 meters.  No midges



emerged below 12 meters after mid-July 1969.  During mid-



June 1970,  emergence was again largely limited to less than



4 meters, but as aeration continued emergence increased at



the greater depths.  Between 20% and 40% of the total emer-



gence occurred below 12 meters during August and September



1970,  whereas none occurred in this interval during the



same period of 1969.



     Total seasonal emergence was generally greater during



1969 (Figure 95).   Two large peaks occurred during 1969,



one during mid-July and another during mid-August.  These



coincide closely with periods of intense heating  (Figure 73).



Emergence during 1970 has several peaks,  but these coin-



cidences with periods of heating are obscure after mid-July-



     Of the more abundant Chironomid adults, only Ablabesmyia



mallochi and Clinotanypus thoracicus showed much increase in



emergence during 1970 (Table 9; Figure 96) .  The other



species either remained about the same or declined.  These



two species belong to the predaceous sub-family Pelopiinae.



Of the more abundant adults,  only C. thoracicus showed much



change in its emergence depth  (Figures 97 and 98).  Over 50%



of this species emerged below 16 meters during 1970 during



its peak emergence period.  Its emergence during 1969 was

-------
                            335
Figure 94.   Total midge emergence from Section Four by
            depths during the summers 1969 and 1970.
            Aeration occurred between June 16 and
            September 7,  1970.  Totals include Chironornid
            midges from emergence traps only.

-------
60
40-
9fi -

—
60-
40-
mergence
o J^ o K:
3000o
L I I ii
LU
_ 0-
(9
"S 60-
40-
»*-
0 20-
*-*
c 0-
Q, 0
u
0) 60-
0.
40-
20-
0_
60-
40 -
20 -
0





o-io-ov
6-27-69
x=4.0
1 1 	 1
6-30-69
7-11-69
x=4.4
	 , 1 	 1
., ._ 	 	 T 1 t if\


r >•* 
-------
                            337
Figure 95.   Total estimated weekly midge emergence from
            Section Four during the summers 1969 and
            1970.  Totals include Chironomid midges from
            emergence traps only.   Aeration occurred
            continuously between June 16 and September 7,
            1970.

-------
    30-
O
I  25
0)
JJJ
8,  2°-
|  15H
I
1  ion
UJ
—   5-
|2
    0-
        1970
            1969/
           I   '  !	r
      1    10    20
                     -7—I	1	1	1	r
                      1     10   20
_r_T__1	,	r	,	r_r
 1    10    20     1
         June
                        July
  August

-------
                            339
Figure 96.   Total estimated midge emergences from Section
            Four.  Samples from emergence traps only.
            Aeration occurred between June 16 and
            September 7,  1970.

-------
   45-
   30-
   15-
    0-
   12-
   9-
      Chironomini  (10 )
Proclodius  spp.  (10 )
      Ablobesmyia  mallochi (104)
      Clinotanypus thoracicus (10 )
     1    10    20      1    10    20     1     10    20
           June
                       July           August

-------
                            341
Figure 97.   Depth emergence of selected Section Four midges
            during the summers 1969 and 1970.   White areas
            represent no observed emergence.   A= Procladius,
            B= Ablabesmyia mallochi,  C= Clinotanvpus
            thoracicus.   Aeration occurred between June 16
            and September 7,  1970.

-------
90-
   1  10 20  1   10  20   11" ' 10' 20   i    10 20  1   10  20  1   10 20
       June
July
August
June
July
                                                         August
                1969
                               1970
       : 0-4
     4-8
        8-12
   v^>: 12-16
            16-19.1

-------
                            343
Figure 98.   Depth emergence of Section Four midges and
            mayflies during the summers 1969 and 1970.
            White areas represent no observed emergence.
            A= Lauterborniella spp.,  B= Chironomjni,
            C= Mayflies (Ephemeroptera).   Aeration occurred
            between June 16 and September 7,  1970,

-------
  90
                           10 20 1 io ^
110201
   June
          Jul
          1969
                 Au9ust
                                   JulX    August
                                   1970
             4-8
                      8-12
                            #££12-16
                                          16-19.1

-------
                            345
limited to less than 12 meters during the entire summer.



A. mallochi also increased its emergence depth substantially



during August 1970, but not as much as C. thoracicus.  Changes



in the other species were erratic.



     N. Wilson Britt identified four species of mayflies



from Section Four:  Caenis spp., Callibaetis spp.  (probably



C. brevisostatus), Neocloeon alamance and Stenonema spp.



Caenis was the most abundant.  Mayfly nymph biomass collected



by the dredges decreased by 55% during 1970 compared to 1969,



but total number collected were almost identical both years



 (Table 8) .  Total estimated nymph biomass was always less



during 1970 (Figure 100).  Total numbers were less during



early summer 1970 but exceeded the 1969 values during August



1970.  Nymph depth distribution was very similar both years



 (Figure 99) .  Nymphs were seldom collected below 11 meters



depth.  They were never collected below 14.5 meters, and were



only abundant between 11 and 14.5 meters during September



1970.  The emergence depth distribution largely reflects the



nymph distribution (Figure 98).  Adults never emerged below



8 meters during 1969.  Their peak emergence occurred during



early July 1969,  at which time most emerged from between 0



and 4 meters depth.  By the end of the summer 1969, all were



emerging from between 4 and 8 meters depth.  Their 1970



emergence pattern differed greatly from the 1969 pattern in



both time and depth of emergence.  The 1970 peak emergence



occurred during August,  with almost no emergence before the

-------
                            346
Figure 99.   Section Four Mayfly (Ephemeroptera)  depth
            distribution as percent of number during
            each sampling period during the summers 1969
            and 1970.   Shaded histograms represent
            aerated periods.

-------
  50
0)

A
z

»^
o
o

0)



~
—
)—

-
>—
)-
) —
> -
s ~
)-
5-
» "
>—
)-
>—
1 -

6-15-
x=4.0

7-5-^
x=4.7

7-25-
x=5.2

8-15-
9-6-d
x=4.9

69

>9

69

69

9









ill i ii


1



l l i

i i i i
6-12-70
x=5.9



7-3-70
1
11







7-24-70
X=6.5


•
1

8-14-70
x=7.0
»*
m

9-4-70
x=8.2
mil

(III III II

,


I




11

1
1 1 1 1 1
  75-
          5     10     15
0
10
                       Depth (m)

-------
                            348
Figure 100-   Total estimated Mayfly (Eph enter opt era)
             number and wet weight in Section Four during
             the summers 1969 and 1970.   One standard
             error is shown about each estimate.  Aeration
             occurred between June 16 and September 7,
             1970.  Totals from dredge samples only.

-------
    8-
    6-

 
 0)
 >s
 O

3


<+-
 O

 k.



 E


Z
    4-
    2-
    0-
    6-
 E
 O)
CM
 O
    4-
 o
 E
 o

 a  2-

 x
 x
 O
    0-
   -2-
           10    20      1     10     20      1      10    20      T


             June              July              August

-------
                            350
end of July.  The early summer emergence was entirely from



between 0 and 4 meters,  but the August 1970 emergence



occurred from between 0 and 12 meters.



     Two species of Amphipods were identified by T. Wayne



Porter:  Hyalella azteca and Gamarus.  H. azteca was by far



the most abundant.  Total amphipod biomass was about the



same both years, but about 32% more individuals were col-



lected during 1970 (Table 8; Figure 101).  Their depth



distributions are about the same both years (Figure 102),



with almost no specimens collected below 14.5 meters.



     Four Trichoptera were identified by Glenn B. Wiggins



(Royal Ontario Museum):   Mystacides spp., Oecetis spp.,



Polycentropus and Oxyethira in order of relative abundance.



Trichoptera numbers, biomass and depth distribution were



about the same both years (Table 8; Figures 103 and 104).



     Leonora K. Gloyd identified six Odonates from Section



Four:  Dragonflies:  Gomphus spp.  (probably G. spicatus) ,



G. quadricolor  (or possibly G. lividus);  Damselflies:



Ischnora verticallis.  Enallagma hageni, Argia spp. (probably



A. fumipennis violacea)  and E. ebrium  (or possibly E. hageni).



Odonata numbers and biomass were about the same both years-



(Table 8) .




     Heleidae (=Ceratopogonidae )  larval biomass and numbers



decreased by about 50% during 1970 compared to 1969  (Table



8; Figure 105).  Their depth distributions were erratic, with



no obvious trends (Figure 106).

-------
                            351
Figure 101.   Total estimated Amphipod number and wet weight
             in Section Four during the summers 1969 and
             1970.   One standard error is shown about each
             estimate.   Aeration occurred between June 16
             and September 7,  1970.  Totals from dredge
             samples only.

-------
•o
o
 in
•o
 O
 Q.
 CL

 E
JQ

 E
 3

Z
9-
     6-
     3-
     0-
 E
 O)
CO

 2   2-1
 O

 E
 o

 OQ
 -o
 O
 a

 IE
 a
 E
1-
     0-
                                                        -  - -O
            10  '  2b     i    10  '  20
                                             —r-
                                              10
     —I—•-
     20
             June
                            July
August

-------
                            353
Figure 102.   Section Four Amphipod depth distribution as
             percent of number during each sampling
             period during the summers 1969 and 1970.
             Shaded histograms represent aerated periods.

-------
rri i i i i i i i' i i i i i i i i
              Depth (m)

-------
                            355
Figure 103.   Section Four Trichoptera depth distribution
             as  percent of number during each sampling
             period during the summers 1969 and 1970.
             Shaded histograms represent aerated periods.

-------
0
                Depth (m)

-------
                            357
Figure 104.   Total estimated Trichoptera number and wet
             weight in Section Four during the summers
             1969 and 1970.   One standard error is shown
             about each estimate.   Aeration occurred
             between June 16 and September 7,  1970.
             Totals from dredge samples only.

-------
    0-
 O>
O
I/)
O
E
o

03
 a
 o
    4-
3-
 •-   2-
    1-
    0-
                             c
            June
                         July
10    20

 August

-------
                            359
Figure 105.   Section Four Heleidae (=Ceratopogonidae)  depth
             distribution as percent of number during each
             sampling period during the summers 1969 and
             1970.   Shaded histograms represent aerated
             periods.

-------
   45-
   30-
   15-
6-15-69
x=4.1
                                               6-15-70
       7-5-69
   45- x=6.4
   30-
(f)
   15-
                              7-3-70
                              x=7.3
       7-25-69
       x=9.6
g
O
0) 15-^
O.
 (C
                             ! 7-24-70
                             !x=4.6
   45-
            8-15-69
            x=4.0
 0) 30-1
 (0
 0)
I  0
                             18-14-70
                             Ifcll.l
                            1
   45-
   30-
    15-
      9-6-69
      x=5.0
 9-4-70
]x=9.8

-------
                            361
Figure 106.   Total estimated Heleidae (=Ceratopogonidae)
             number and wet weight in Section Four during
             the summers 1969 and 1970.   One standard
             error is shown about each estimate.   Aeration
             occurred between June 16 and September 7,
             1970.  Totals from dredge samples only.

-------
-o
 o
   25-
   20-
ju  15-

 0)
I
 »*-  10-
 o
 1  s-
    0-
    8-
 E
 D)

-------
                             363
 Crayfish






     During 1969, Section Four  crayfish were  only  collected



 between August 12th and August  30th.   During  this  period,



 the males were largely confined to  shallow water,  and  the



 females to deeper water  (Figure 107).  The average depth for



 males was 4.4 m and 8.8 m for females.  Three times as many



 males were captured as females.  No crayfish  were  collected



 below 14 meters depth.  Their maximum  depth during 1969



 corresponds closely with the 10°C isotherm.   Although  some



 chemical stratification was present, it did not  appear to be



 an important barrier to crayfish depth distribution.



     During early June 1970, both males and females were



 largely confined to shallow water.   Their average  depths were



 2.6m and 3.3m respectively, with  no  animals below 9  meters.



 The lake was well-stratified thermally during this period,



 but chemical stratification was  not evident.   As during



 August 1969, their maximum depth distribution during early



 June corresponds to the 10°C isotherm.  After aeration began,



 the crayfish very rapidly distributed  to maximum depth.



 Aeration began June 17, 1970 and the lake was almost com-



 pletely destratified within a few days.  After destratifica-



 tion,  the minimum temperature was always greater than  14 C.



 The females'  average depth between  June 16th  and June  30th



was 9.5 m and the males' 11.8 m.  Their depth  distributions



varied somewhat during the remainder of the summer,  but they



were always distributed to maximum  depth.  No conspicuous

-------
Figure 107.  Section Four crayfish depth distributions during the summers            u>
             1969 and 1970.  Total numbers during each sample period and             ^
             their average depths are shown.  The shaded area represents
             the 1969 distributions.  Aeration occurred between June 16
             and September 7,  1970.

-------
 5-
10
15-
10-
15-
¥,

4P

2i°


0 20
i ill


n=8
7=3.3
6-7-70
6-16-70











n=22
x=72
7-24-70
8-14-70








40




J












60 60 40

n=39
x=2.6
n=118
7= 12.0
CJ
jj

>T

n 90
x=9.0
6-16-70
6-30-70
n=31
x=9.5
8-14-70
9-4-70
Percent
2p 0 2jO 4p 6p 6(0 4p 2p 0 2p ^ 4p 6(0

















L


























1











n A")
7=75



0*
n=273
x=11.8




1

J
n=52
7=8.6
6-30-70
7-24-70
n=40
x=8.8
8-12-69
8-30-69










v






-x
'~f '
s""*

% . ,
?














~ f. A
IT-O4
7=9.3
rf>
\
£ ^ / ':
" ^





n=120
x=4.4

-------
                            366
concentrations in response to environmental factors was



obvious.  The sex ratio was always in favor of the males.



During 1969 this ratio was 1:3.0 and during 1970 it was




1:3.8.



     From August 14,  1970 through September 4, 1970, the



distributions were much changed from the previous year.



The males were most numerous at maximum depth during 1970.



Their average depth was 11.8 m during 1970 compared to



4.4 m during 1969.  The females were also distributed to



maximum depth (19.6 m)  during 1970,  compared to 14.0 m during



1969.  Their average depth was only slightly greater during



1970; 9.5 m compared to 8.8 m.





Rainbow Trout





     Rainbow trout ranged between the surface and bottom



during June 1969 (Figure 108).  Later in the summer they



were found mostly below 8 meters depth.  At no time were



they caught in water warmer than 19°C.  The gill net samples



may give a distorted picture however,  because of small sample



size.  Only thirty-two 1969 stocked RBT and nineteen 1964-65



stocked RBT were caught during 1969.   This low capture rate



is partly due to the small size of the 1969 stocked fish and



small population size of the 1964-65 RBT.  The 1969 stocked



RBT averaged 188 mm when stocked,  but fish less than 200 mm



were seldom caught in the nets.  Only 10% of the 1969



stocked fish were 200 mm or larger when stocked.  Section



Four fish have always had very slow growth and few attained

-------
Figure 108.  Section Four rainbow trout depth distributions during 1969.             w
             Open squares represent fish stocked during June 1969 and                o\
             marked with a right-pectoral fin clip.  Solid squares are               ^
             fish stocked during 1964-65 and lack fin clips.  Each square
             represents one fish.

-------
August
    20

-------
                            369
net vulnerable size during 1969.  The clear water of Section



Four Lake also contributed to low catch rates.



     During early June 1970,  the rainbow trout were mostly



distributed between the surface and 12 meters (Figure 109).



Soon after aeration began they distributed throughout the



lake.  By early July they were mostly distributed along the



bottom of the lake.  Ninety-four 1970 stocked RBT, eleven



1969 stocked RBT and eleven 1964-65 stocked RBT were cap-



tured during 1970.  The increased capture rate of the 1970



stocked fish is attributed to their larger size; they



averaged 200 mm when stocked.

-------
Figure 109.  Section Four rainbow trout depth distribution during 1970.
             Open circles represent fish stocked during May 1970 and
             marked with a left-pelvic fin clip.  Solid circles repre-
             send fish stocked during June 1969 and marked with a right
             pectoral fin clip.  Solid squares represBnt fish stocked
             during 1964-65 and lack fin clips.  Each symbol represents
             one fish.

-------
June
10     20
 July
10     20
  August
10     20

-------
                DISCUSSION AND CONCLUSIONS








Physical-Chemical Parameters





     As discussed earlier,  a lake1s oxygen and temperature



regimes are its most important parameters.  In Section Four,



oxygen concentrations were generally quite high and pre-



sented no biotic distributional barrier.   In fact,  oligo-



chaetes and midge larvae were most abundant in the profundal



zone.  Temperature exerted a greater influence than oxygen.



Artificial destratification greatly altered the temperature



regime in Section Four Lake.  Although the maximum average



temperature was increased 3.6 C by destratification,  maximum



bottom temperatures were increased more than 15.3°C.   Since



the dominant biota live in the profundal zone,  this repre-



sents a very significant alteration of their environment.



Many metabolic and other chemical processes double for every



10 C increase.  This sort of temperature change could lead



to changes in species composition,  growth rates,  reproduc-



tive patterns and distribution within the lake.  When



temperature is no longer an important barrier,  the organ-



isms should distribute according to other conditions such as




substrate,  light,  plankton,  rooted aquatic plants and pre-



dator-prey relationships.  Certain coldwater species may be
                            372

-------
                            373
 eliminated from the lake because of the absence of cold,



 aerated water during the summer.  I had expected this sort



 of problem with the rainbow trout, but it did not occur.



 However, this could be a serious consequence of destratify-



 ing oligotrophic lakes that normally have greater surface



 temperatures than Section Four.



     I did not expect any great changes in the chemical



 conditions in Section Four as a consequence of destratifica-



 tion.  Changes did occur, but they were generally not as



 striking as when eutrophic lakes are destratified (Fast,



 1968).  Oligotrophic lakes are typified by their chemical



 homogeneity.  Since the water chemistry did not appear to



 limit the biotas'  distribution before aeration,  we should



 not expect much distributional change in the biota in re-



 sponse to changes in the water chemistry after destratifi-



 cation.



     The thermal profiles and thermal stability values indi-



 cate the lake was nearly isothermal after June 20,  1970.



 Slight temperature gradients often occurred at the surface



 as it was warmed faster than the heat could be redistributed



 throughout the lake.  During periods of intense heating, it



 is difficult to prevent this thermal microstratification.



This is due in part to the fact that the relationship between



rate of destratification with constant air input is not



linear.  The closer thermal stability is to zero,  the greater



the energy required to cause a unit decrease in thermal

-------
                            374
stability.  It may require relatively little energy to reduce



thermal stability by 75%,  but much more energy to reduce it




to zero.



     The destratification system used in Section Four was



much larger than necessary.   This situation was purposely



arranged to assure complete mixing.  As in Hemlock Lake,  the



important factor was measuring the effects of this system of



mixing on the biota.  We did not want to measure the effects



of partial mixing,  but of complete destratification.  The



proper size compressor,  distribution and duration of air



input,  etc. is basically an engineering problem,  and one



that has not yet been adequately solved.



     There were large changes in dissolved organic matter



and particulate organic matter.   DOM changes may be related



to changes in primary production.  These changes will be



discussed later in greater detail by R. G. Wetzel.



     As with Hemlock Lake,  phosphorus concentrations were



measured only during 1969.   They were not continued for the



reasons given.  Because of the oligotrophic condition of



Section Four,  I expected even less measurable changes in



phosphorus concentrations than we expected in Hemlock.



Phosphorus concentrations before aeration were quite variable



and low.




     There is usually not much economic incentive to arti-



ficially destratify an oligotrophic lake.  In fact this is the



only case that I am familiar with.  Destratification is usual-



ly instigated to improve water quality by oxidizing and

-------
                            375
 otherwise eliminating anoxic conditions.  This  is  not  the



 case with oligotrophic  lakes.  The major  reason oligotrophic



 lakes may be destratified  is to reduce  evaporation rates.



 This is also an important  reason  for destratifying eutrophic



 lakes.  Destratification may reduce annual  evaporation by



 4  to 10 percent  (Koberg, 1964, personal communications).



 This is accomplished by a  slight  reduction  in surface  tempera-



 tures during the summer.   Surface temperatures  are slightly



 warmer during the cooling  period, but the periods  of above-



 normal and below-normal surface temperatures are such  that



 a  net evaporation reduction is realized.  Although the lake's



 heat budget is greatly  increased  by destratification,  this



 extra heat is lost to a greater extent  by infra-red radiation



 from the lake, than by  evaporation.  Reduced evaporation rates



 can result in a substantial savings.  An  estimated #10,000



 worth of water was saved from evaporation  during one year's



 destratification of El  Capitan Reservoir, California  (Fast,



 1968).  The annual cost of running the  compressor,  amortiza-



 tion of initial investment and maintenance  was  only about



 #3,000.  These savings  are most important,  and  most likely,



 in water-starved regions such as  our American southwest.



 Water is in short supply,  expensive, and  evaporation rates



 are high.





 Primary Production





     Because of Section Four's oligotrophic condition  and



marl deposition,  I did  not expect much  change in primary

-------
                            376
productivity during destratification.  Although I expected



profundal sediment temperature increases to regenerate some



nutrients,  I expected these nutrients to be readily bound by



the carbonate complexes and therefore effectively unavailable



for plant growth.



     The evidence indicates that this did not occur.  Surface



primary production potentials were about three times as



great during August 1970 compared to August 1969,  and produc-



tion efficiencies were likewise increased from 1 mgC/m3/4 hrs/



In cells during August 1969 to between 3 and 4 mg during



August 1970.  These changes strongly suggest that nutrients



were more available following aeration.   However,  as dis-



cussed for Hemlock Lake,  other factors could also account for



these apparent increases.



     Although production potentials and efficiencies increased



following aeration,  phytoplankton standing crop did not.



Average surface values were actually lower during 1970 com-



pared to 1969.  Zooplankton grazing could be flaunted as a



possible explanation for lower standing crops, but I feel



that was not the case.  I believe that increased mixing rates



and increased mixing volume caused the reductions in standing



crop.  This mixing in effect prevented the phytoplankton



from realising their growth potential.



     During normal,  stratified periods phytoplankters may



spend most or all of their time within the photic zone



(Figure 110).  The ratio ti/t2 is relatively large and

-------
Figure 110.   Hypothetical residence times for a passive,  neutral buoyancy
             object with the photic and aphotic zones of a stratified and
             unstratified lake.

-------
t,
Depth of
Photic Zone
  Stratified
                Unstratified

-------
                             379
 approaches infinity where  tz  is  small.   This  latter  situa-



 tion probably occurs  in  a  thermally  stratified Section Four



 because of its clarity-  About 20% of  the  surface irradiance



 was still present  at  12  meters on  August 11,  1969 when the



 Secchi was 10.25 m (Figure 82).  Although  the entire lake



 may lie within the photic  zone,  light  intensity and  produc-



 tion rates are very low  in deep  water.   After destratification,



 phytoplankters were transported  into the deep water  to a



 greater extent than under  stratified conditions (Figure 111).



 Surface/15 meter phytoplankton density  ratios average 2.2



 during 1969, whereas  surface/16  meter  densities averaged 1.5



 during 1970.  This indicates  a more  even distribution during



 1970 compared to 1969 and  a shift  in densities from  the sur-



 face to deep water.   A 1.0  ratio indicates  equal densities



 at both depths.  Ratios  above 1.0  represent higher densities



 at the surface, and ratios  less  than 1.0 represent higher



 bottom densities.  Densities  were  always greater at  the



 surface both years, and  quite variable  from day to day.



 Bottom densities were greater than surface  densities in only



 2 of 24 sample sets during  1970.   These data  indicate that



 this system of artificial mixing redistributed a significant



 portion of the phytoplankton  population into  deeper  water.



 Time spent in the  aphotic,  or at least  dimly  lit deep water



 (t4)  was much increased  relative to  time spent at shallow



 depths (ts).  Thus the time available  for high production



was reduced (t3/t4 <  ti/t2) by destratification.   Increased

-------
                            380
Figure 111.   Ratios of Section Four surface/bottom
             phytoplankton concentrations during 1969
             and 1970.   The lake was destratified
             during 1970.

-------
June
July
1    10   20
  August

-------
                            382
nutrient availability was not adequate to compensate for



these changes and net in situ production decreased.  The



above conditions are likely to occur in relatively deep,



oligotrophic lakes with a large hypolimnetic volume and



nearly complete destratification.  If destratification is



not complete,  but stratification persists near the surface,



production rates may remain the same or be increased.  I ob-



served increased primary production rates in El Capitan



Reservoir,  California following incomplete destratification.



Even with continuous air injection,  the eutrophic reservoir



typically had microstratification of 2 to 3 C within the



upper few meters (unpublished data) ,  and almost all of the



primary production occurred within this region.  Phytoplank-



ton were apparently kept within this zone,  and not swept



into the dimly lit regions of the reservoir in sufficient



quantity to greatly reduce their population.  This stratifi-



cation was almost always present during the yearly heating



cycle and developed because the heat input from solar irradi-



ance was greater than could be distributed throughout the



reservoir by the aeration system.  This condition almost



never developed in Section Four Lake because mixing was



almost complete.  The air input was actually greater than



necessary to assure a thorough mix.




     Murphy (1962)  hypothesized that increased mixing depths



and/or turbidity acting alone would reduce primary produc-



tion.   His hypothesis predicts that increased mixing depths

-------
                            383
and turbidity such as occurred in Section Four during com-



plete destratification would lead to decreased primary



production, and that decreased mixing depths such as



occurred in El Capitan Reservoir during incomplete destrati-



fication would lead to increased production rates.  His



predictions in these two cases agree with observations.



Likewise his calculated production rates agreed well with



those he observed in 33 shallow California ponds.  He sug-



gests manipulating reservoir production rates by adjusting



metalimnion depths.  Decreased metalimnion depths could be



achieved by epilimnion withdrawals or incomplete destratifi-



cation/ whereas increased metalimnion depths could be



achieved by hypolimnion withdrawals or air injection at some



desired depth which is less than maximum depth.  Hooper et al.



 (1952) increased the metalimnion depth by mixing hypolimnion



water with epilimnetic.  This increased phytoplankton and



periphyton production because the hypolimnion was nutrient-



rich relative to surface waters.  This nutrient increase more



than compensated for possible increased mixing depths.



     Knoppert et al. (1970) found they could increase the



metalimnion depth by air injection at some intermediate



depth.  Their metalimnion usually began at about 5 meters,



but was lowered to the air input level of 19 m within three



weeks.  Maximum depth was over 27 m, but water below 19 m was



not upwelled by their aerator and a sharp thermal discontin-



uity developed at 19 meters.  Knoppert et al. conducted

-------
                            384
another experiment to determine the effect of increased mix-



ing on plankton populations.   They increased the mixing



rate of two shallow reservoirs without altering their thermal



regimes.  These reservoirs did not normally stratify thermal-



ly, but they had reduced photic zones.  Their results were



inconclusive,  but suggest that increased mixing alone did not



greatly affect the phytoplankton.



     Robinson et al. (1969)  evaluated the effects of inter-



mittent destratification on two northern Kentucky reservoirs.



Total phytoplankton standing  crops in these destratified



lakes responded about the same as  those in a control lake.



They concluded,  however,  that bluegreen algae declined faster



than green algae and the number of plankton species remained



the same or increased slightly following air injection.



Bernhardt (1967)  also reports a decline in the bluegreen



algae Oscillatoria rubescens  during artificial destratifica-



tion of eutrophic Wahnback Reservoir,  Germany.  Melosira



qranulata angustissima became abundant after 0^.  rubescens



disappeared.   Bluegreen algae were not conspicuous in Section



Four either before or during  destratification.  Most signifi-



cantly,  Bernhardt also found  "...  there was no increased bio-



production, "  due to artificial destratification.  Artificial



destratification was incomplete since a small thermal gradient



persisted near the surface part of the summer.  This condi-



tion is similar to that discussed  for El Capitan Reservoir,



California.

-------
                            385
     I am uncertain how to interpret periphyton events.  In



contrast to the phytoplankton, they were not swept out of



the surface waters by the currents.  Increased turbidity



could have deleteriously affected their growth, but increased



nutrient availability should have promoted periphyton growth.



Likewise, increased water currents due to air injection



should have favored periphyton growth.  Short-term periphyton



accumulations do show significant increases, but the long-



term accumulations are about the same both years.  This sug-



gests that colonization and the early "successional" stages



were accelerated by destratification but the maximum accumu-



lation levels were not affected.  In contrast,  both short-



term and long-term periphyton accumulations were increased



in Hemlock Lake.  I have no good suggestions to explain



these differences in Section Four Lake.





Zoobenthos



     The most obvious response to the Section Four zoobenthos



to destratification appears to be a reduction in numeric



and biomass standing crops.  Oligochaetes (microdriles)



and Chironomid larvae had especially large reductions in



standing crop as well as average sizes during 1970  (Table 8).



All organisms except the Oligochaete megadriles, Chironomid



pupae and Odonates were smaller during 1970, and all the



minor species,  except the Heleids were numerically more



abundant during 1970.

-------
                            386
     Although reduction in oligochaete standing crop appears



directly related to destratification,  Chironomid standing



crop reduction could be due to emergence.  Midge emergence



during the summer 1969 was closely related to periods of



rapid heating.  The spring 1970 temperatures appear to have



been warmer than during the spring 1969.   For example,  the



early June 1970 surface temperature was 19.5 C compared to



16°C the previous June.  These higher  temperatures could have



promoted a greater emergence rate prior to my sampling and



resulted in lower standing crops during June 1970.  This



could account for lower standing crops during June 1970,



but not necessarily during the remainder of the summer since



reproduction and growth should have occurred.



     The most likely explanation for the observed zoobenthos



reductions in standing crop is that destratification resulted



in a reduction,  or at best no change in primary production



and an increased heat budget.  The surface phytoplankton



standing crop appears less concentrated during most of the



summer 1970 (Figure 82),  and total accumulated periphyton



was about the same both years.  Total  phytoplankton standing



crop may actually have been greater during 1970 than during



1969 since aeration could have distributed it more uniformly



throughout the lake;  but this cannot be accurately demon-



strated from the data.  In any case,  phytoplankton surface



concentrations were less during 1970.   An increased heat



budget should have resulted in greatly accelerated metabolic



rates for certain organisms such as the oligochaetes.

-------
                            387
Almost all of the oligochaetes and a large percent of the



Chironomid larvae were found below 14 meters both years.



These deep-dwelling individuals experience maximum tempera-



ture increases of about 15°C due to destratification.  Many



biochemical reactions were doubled by every 10°C temperature



increase.  Increased catabolic rates without a commensurate



increase in anabolism should result in decreased biomass,



as observed.  Put more simply, respiration was increased by



temperature increases,  but synthesis either remained the



same or decreased due to an unchanged or decreased food



supply.  The net result was "negative growth," i.e.,  a de-



crease in standing crop.  If we agree that this is deleteri-



ous, then we could say that artificial destratification of



Section Four caused thermal pollution of the lake,  although



the results were not obvious to the casual observer.



     Unlike Hemlock Lake,  the Section Four's sediments were



well oxidized and contained sparse organic matter prior to



artificial aeration (Figure 14).   Oxygen concentrations were



always high over the sediments and decomposition was probably



almost complete.  Aerobic decomposer microbes, as well as



larger detritivores undoubtedly worked over these sediments



to a large degree.  Consequently,  destratification probably



did not greatly affect nutrient regeneration from,  or



decomposition of,  these sediments.  The availability of



organic debris to the zoobenthos was probably likewise un-



changed by destratification.  These conditions differ

-------
                            388
greatly from the profundal conditions of eutrophic lakes



where decomposition is retarded by anaerobiosis and organic



debris is unavailable to the detritivores for long periods.



     Reductions in average sizes of most organisms during



1970 is most puzzling.  It is generally agreed that smaller



organisms have greater metabolic rates.  If this is true of



aquatic invertebrates, and the aforementioned changes occurred



in the metabolic rates,  then I would expect larger species



or individuals to be favored during destratified periods.



In fact,  this does not appear to be the case.



     Although production rates were not measured for any



species,  midge emergence may be considered a rough estimate



of midge production.   The reduction in total emergence sug-



gests that total midge production was decreased by destratifi-



cation.  Production per individual may have actually in-



creased,  however,  since total midge emergence decreased 10%



between 1969 and 1970, but number of midge larvae collected



decreased 54% during these same periods.



     I was surprised by the almost indiscernible effect of



destratification on the zoobenthos depth distributions.



Midge pupae and emergence depth distributions were most



affected,  but not as much as expected.  Although oxygen and



presumably most other chemical parameters were not limiting,



I expected temperature to represent a real barrier to certain



organisms (e.g.,  see discussions of fish and crayfish depth



distributions).  I had expected a net movement of Ghironomids,

-------
                            389
mayflies, amphipods and possibly other taxa into deep water



and a net movement of oligochaetes into shallow water.  The



absence of these changes suggests that other conditions such



as sediment conditions, attached vegetation, and light are



more important factors.  Oligochaetes were almost entirely



limited to shallow water by anoxia in eutrophic El Capitan



Reservoir, California  (Fast, unpublished data).  When this



lake was destratified the oligochaetes distributed evenly



throughout the entire lake even though large differences



existed between sediment conditions at different depths and



at different points in the lake  (Fast, 1968).  I had expected



a similar net movement of oligochaetes into shallow water



after Section Four was mixed, but only a few individuals



moved into, or remained in shallow water after destratifica-



tion.





Crayfish





     Factors influencing Orconectes virilis depth distribu-



tions within lakes are not well understood.  Momot  (1967)



and Momot and Cowing  (1970) found that both sexes of



0. virilis, in nearby lakes were mostly found  in shallow



water during May and June.  Adult females moved to deeper



water first,  followed by young females and males as they



reached sexual maturity.  By August 70% of the adult females



were below 6.0 m and were concentrated at 7.6 m.  During



this same period,  65% of the yearling males and females were

-------
                            390
between 3.0 m and 6.0 m, while over 85% of the adult males



were above 6.0 m.  The lowermost extent of the metalimnion



during August was 9.1 m with a 13 C temperature.  Oxygen



and other chemical concentrations were not presented.  The



crayfishes' maximum depth distribution coincided with the



bottom of the metalimnion.  These authors feel that the



migration to deeper water is related to sexual maturation.



Aiken (1968)  also showed that mature 0_. virilis in an Alberta,



Canada stream also moved to deeper water in late summer.



The females preceded the males.  (This migration contributed



to overwinter survival of the species.)  He also believes



the migration is related to sexual maturation since matura-



tion is related to photoperiodism and temperature (Aiken,



1969) .



     Section Four 0_. virilis depth distributions during 1969



coincide well with those described by Momot (1967)  and Momot



and Cowing (1970).  During August the males were in shallow



water and the females were much deeper.  By June 1970,  both



sexes were concentrated in shallow water.  Both seasons



their lowermost limit was related to the 10°C isotherm and



not other chemical factors.  However,  almost immediately



after Deration began during June 1970,  both sexes distributed



throughout the lake.  Although they were not aged,  there was



no conspicuous size distribution of either sex as a function



of depth.   By early July and thereafter,  the males always



had a greater average depth than the females.  Both sexes

-------
                            391
were about evenly distributed throughout the lake after



destratification.  This even distribution seems to mitigate



against the sexual maturation hypothesis.  If light intensity



is important for maturation, then the mature females and



recently mature males and females still should have preferred,



and concentrated in, deep water.  The mature males distribu-



tion should have remained about the same.  If cold water is



the important factor affecting their depth selection,  then



an even distribution of the females could be expected; since



the lake was isothermal, they should have randomly searched



for cold water.  This does not necessarily explain the even



distribution of the mature males, however,  since the surface



temperatures were about as warm as during the preceding



summer.  If the males were independent of temperature above



10°C,  then their distribution should have been about the



same both years.



     I would like to suggest that O. virilis depth distribu-



tions are not directly related to maturation, but when other



factors such as oxygen are not limiting, mostly to water



temperature,  social aggression of the mature male, and



maturation of the mature female and yearlings.  The mature



males are highly aggressive and are known to repel the smaller



males and females (Abrahamsson,  1966; Camougis and Hichar,



1959).   These authors also suggest that the adult males con-



centrate in shallow water because this zone offered the best



conditions for food and shelter.  I suggest that the mature

-------
                            392
males usually select shallow water because this is the zone



with the highest temperature.  Although 0. virilis functions



above 10°C (Momot,  1967)  its metabolism should increase as



a function of acclimation temperature.  If increased meta-



bolic rate is advantageous,  then the crayfish should seek



warmer waters.  Not all individuals successfully inhabit the



warmer water, however,  since the mature males repel the



weaker,  less aggressive females and yearlings into deeper,



colder water.  Bovbjerg (1964)  found that both (). virilis



and 0. immunis prefer the same substrate when tested sep-



arately.  When tested together 0_. virilis drove 
-------
                            393
individuals during this period.  This is important for the



survival of the species.  If growth and molting is also



related to temperature, it is important that mature females



and yearlings remain in shallow, warm water as long as pos-



sible.  This will assure development of the eggs and matura-



tion of yearlings.  Once these functions are complete, it



may be immaterial where these members spend the rest of the



summer, as long as there are enough available for breeding



and reproduction the following year.  The newborn have



mostly left the mature females by June.  This is when the



mature females migrate to deeper waters, or are driven there



by the males.  The yearlings move into deeper water some-



what later than the mature females.  This exposes them to



predation by the mature males for a longer period,  but not



during the warmest part of the summer.  The advantages of



remaining in the shallow,  warm water to mature may outweigh



the detrimental influence of the mature males.



     Artificial destratification greatly altered the usual



distribution patterns.   Since the lake was isothermal, there



was no longer any advantage to select shallow depths.  The



water was about the same temperature at all depths and all



individuals were exposed to similar thermal conditions.



Under this situation,  the mature males no longer selected



the shallow depths,  but distributed more or less evenly



throughout the lake.  Several authors have shown that cray-



fish either have a very large home range or none at all

-------
                            394
(Abrahamsson,  1966; Hichar and Camougis, 1959; Penn, 1950)



and travel great distances in a short period.  During



stratified periods, they moved randomly about the shallow



depths,  limited by the cold temperature of the lower depths



and by the shore.  After destratification,  individuals un-



doubtedly wander throughout the entire lake.  During August



1970,  the entire lake was about as warm as the surface waters



were during 1969.  I would therefore expect similar behavior-



al activities and metabolic rates of the mature males at



all depths.  The effect of this on the females and yearlings



is unknown-  Also unknown is the effect of increased tempera-



ture on the maturation and metabolism of the females and



yearlings.  Increased metabolism may have permitted them to



better cope with the aggressive males.  Decreased densities



following dispersal should also have lowered encounter rates.



If destratification were continued several years, densities



of mature males throughout the lake might approach former



shallow water densities and lead to low survival of females



and young crayfish.



     If repulsion of females is the important factor affect-



ing their distribution during late summer,  then evidence



should exist for this both years«  The correlation between



male and female depth distributions during August 1969 was



-0.29 (F= 0.76,  d.f.= 8).  During August and September 1970,



when the lake was isothermal,  the correlation was -0.52



(F= 1.89,  d.f.= 5).  Although neither correlation is

-------
                            395
significant at the 0.05 level, they are indicative.  In both



cases low concentrations of females were associated with



high concentrations of males.  During 1969 the males were



most abundant in shallow water, but during 1970 they were



most abundant in deep water.  This evidence is rather weak



and indirect, but tends to substantiate the general hypothe-



sis.



     Another, somewhat overlooked possibility exists to



explain the observed distributions.  If the much touted



aggressive behavior of the male does greatly affect the



behavior of the female and yearlings, it is also possible



that this factor alone determines the distribution of females



and yearlings within traps.  If male aggression can be



credited for altering the preferred distribution of other



individuals within the lake, it seems logical that this ag-



gression could also affect their distribution within the



traps.  Traps with many mature males usually contain few



females or young crayfish.  Is this because the latter are



not present in that area,  or because they are driven from



or do not enter the trap?  If the actual distribution of



females was exactly the same as the males at all times, but



the females avoided high concentrations of males, such as



in certain traps,  then we would observe distributions such



as those discussed.  Likewise, the sex ratio would on the




average be in favor of the males and female catch rates



would be much lower.  These conditions are also universally



observed with (3. virilis.

-------
                            396
Rainbow Trout





     During 1969,  the Section Four rainbow trout ranged



between the 5°C and 19°C isotherms (Figure 108).  They



exhibit no preference for the 13°C final preferendum tempera-



ture indicated by Garside and Tait (1958).  Garside and Tait



conducted their laboratory experiments with 100 mm to 150 mm



length fish.  Although light (Sullivan and Fisher, 1953;



Brett,  1952; Pearson,  1952),  feeding activity  (Brqtt, 1952;



Pearson,  1952)  and social behavior (Pearson,  1952) are known



to affect fishes'  temperature selection,  it is uncertain



what the major factor was in Section Four.  These fish had



very slow growth rates.   Their stomachs were usually



empty when captured.  The usual stomach contents included



periphyton,  sticks,  marl encrustaceons, miscellaneous detri-



tus,  terrestrial insects, amphipods and cladocera.  No food



item predominated.  The nature of their diet and slow growth



indicate they were at near starvation levels.  The 1954-65



stocked trout were very emaciated.  These conditions could



lead to a general "search" behavior and thus account in part



for their scattered distribution.  This distribution is in



contrast to that found by Horak and Tanner (1964) in Horse-



tooth reservoir.  Horsetooth Reservoir is also oligotrophic.



As in Section Four,  oxygen,  pH,  carbon dioxide and alkalinity



gradients apparently did not present barriers to the rainbow



trouts'  depth distribution.   Nevertheless, the Horsetooth



Reservoir trout preferred water between 18.9°C and 21.1°C.

-------
                            397
Over 93% were found in or above the metalimnion.  The pre-



dominant factor affecting their distribution in Horsetooth



Reservoir appeared to be food distribution.  The trout pre-



dominantly fed on cladocera, and cladocera were most abundant



in the epilimnion.  We did not measure cladocera depth dis-



tribution in Section Four,  but trout food analysis indicates



they were not a major item.



     The intensity of the thermal gradient could be an



important factor.  In Section Four this gradient was rela-



tively weak.  This is typical of sheltered lakes.  Using a



definition of the metalimnion as a change of 1 C/m depth



increase, the metalimnion begins at the surface on some dates



(Figure 72)  and extended to near the bottom.  Although no one



has clearly described rainbow trout responses to different



temperature gradients, other organisms react abruptly to



strong thermal gradients.  Beeton (1960)  found that vertical-



ly migrating Lake Michigan Mvsis relica would not penetrate a



thermal gradient of 1.67°C to 2.0 C/m, but some temporarily



penetrated a 0.66°C to 0.94 C/m gradient.  During isothermal



periods they would migrate uninterrupted from the bottom to



the surface.  Harder  (1968) found many marine zooplankters



reacted to discontinuities in stratified laboratory cylinders.



These discontinuities included density gradients due to



temperature,  salinity and density without temperature and



salinity gradients.  Burbige (1969)  found that American smelt



(Osmerus mordax)  would not penetrate strong thermal gradients



in the metalimnion.

-------
                            398
     Although trout were not captured near the surface after



early July 1969,  they were seen to feed at the surface after



that period.   Likewise,  terrestrial insects and surface



living insects were found in their stomachs.  This indicates



that they avoided prolonged periods in shallow water„  They



appeared to reside mostly in the deeper water with feeding



excursions into shallow water.   Because of the water clarity,



they could undoubtedly see surface disturbances from some



depth.



     I was especially interested in the effects of artificial



destratification on the survival and distribution of the



trout.  Since artificial destratification increases the



lake's heat budget,  and eliminates the deep, cold water,  this



poses a threat to the trout (Fast,  1968; Fast and St. Amant,



in preparation).   Although trout can withstand 26.7 C tempera-



tures for a few days,  prolonged temperatures above 24°C lead



to high mortalities (Eipper,  1960),  During a summer of con-



tinued aeration the entire lake will become about as warm as



the surface waters during a summer of normal stratification.



If the surface waters normally attain 26 C or more during the



summer,  the entire lake may become this warm.  Normally



Section Four's maximum epilimnetic temperature is between



23 C and 25 C.  With continuous air injection during 1970,



the entire lake was over 23°C for about two weeks during



August.   No trout mortalities were observed during this



period,  nor did the fish appear otherwise adversely affected.

-------
                            399
The temperature increase apparently was not great enough



to be detrimental.



     Although they were distributed throughout the lake dur-



ing artificial destratification, they showed a preference



for the bottom.  Before aeration began during early June,



they were found mostly between 1 and 12 meters,  but soon



after aeration began, they were most abundant in the 17 to



19.6 meter interval.  Food does not appear to be an important



factor affecting this distribution, since their diet was



essentially the same as during 1969.  It could be a thwarted



attempt to seek cooler water, since before aeration cooler



water could be found by swimming downwards.



     Many trout concentrated in the rising air bubbles.



Fast  '(1971) describes this same response of threadfin



shad  (Dorosoma petenense)   in a southern California reser-



voir.  In the shad case, their aggregation was thought to



be a rheotactic response which was reinforced,  or rewarded,



by high food concentrations in the rising water.  If zoo-



plankton concentrated along the bottom of Section Four,



this rising water undoubtedly contained higher concentra-



tions than the surface waters..  Trout aggregated in the



rising water might thus be exposed to higher food concen-



trations than other surface-dwelling trout.



     From the fisheries standpoint, artificial destratifica-



tion of oligotrophic lakes is generally not advisable, since



such mixing will increase their heat budgets and eliminate

-------
                            400
the lakes'  cold water.  Such heating can lead to the elimi-



nation of coldwater species such as trout.  Oligothrophic



lakes with borderline surface temperatures,  such as Section



Four,  are possible exceptions.   Destratification of these



lakes may increase fish food and trout production.   There



is no evidence that artificial  destratification of Section



Four was eithsr beneficial or detrimental for the rainbow



trout.  No obvious increase in  mortality was observed,  nor



did their growth or general condition appear to improve.

-------
              AERATION TO PREVENT WINTERKILL








     Winterkill conditions often develop in small eutrophic



lakes subject to extensive periods of ice and snow cover



(Greenbank,  1945; Cooper and Washburn, 1946).  These lakes



are typically shallow with a high BOD.  The ice cover pre-



vents absorption of atmospheric oxygen,  while the snow cover



limits photosynthetic oxygen production.  As a consequence,



oxygen may be depleted below concentrations necessary for



fish and other biota and mass mortalities occur.



     There are basically two solutions to the winterkill



problem:  (1)  Snow removal to promote photosynthesis.  This



procedure should be started early and continued most of the



winter; and (2)  Artificial aeration of the water.  Several



techniques for artificial aeration have been developed.



Merna (1965)  and Flick (1968)  pumped water onto the ice and



allowed it to drain back into the lake through holes chopped



in the ice.   This procedure melted the snow cover and in-



creased oxygen concentrations within the lake to acceptable



levels.  Rasmussen (1960)  and Wood (1961) injected air under



the ice from perforated air lines.  This technique melted



large areas of the ice cover and greatly increased the oxygen



concentrations.   Patriarche (1961) unsuccessfully used a
                            401

-------
                            402
perforated air line system in a shallow southern Michigan



lake.  His system circulated the water over high BOD sedi-



ments which removed oxygen from the water.  His air input



was not great enough to compensate for this loss and the



total oxygen content decreased.  Halsey (1968) effectively



prevented winterkill conditions by artificially aerating



before an ice cover formed.  His lake usually did not



thoroughly mix following the fall overturn.  Consequently,



oxygen concentrations,  especially in deep water,  were low



when ice covered the lake.  Snow cover compounded this situ-



ation and led to near zero oxygen concentrations by spring.



By aerating before ice formed,  he raised oxygen concentra-



tions throughout the lake to near saturation,  oxidized much



of the organic matter and thus prevented oxygen depletion.



     As part of my study,  I was especially interested in



what effect summertime aeration has on winter oxygen concen-



trations.  As we have seen, primary production and plant



standing crops were greatly increased in both Hemlock and



Section Four Lakes by aeration*  These materials use oxygen



when they decompose and could contribute to winterkill con-



ditions.  Tanner (1952)  greatly reduced winter oxygen con-



centrations in these two lakes by summertime artificial



fertilization.  Fertilization increased primary production



and the rate of oxygen consumption.  Both summer and winter



oxygen concentrations were greatly reduced by this practice.



During the winter following fertilization, the depth where

-------
                            403
oxygen fell below 2.0 mg/1 decreased  from  6.8  to  3.4 meters



in Hemlock lake, and from over  15 meters to  1.2 meters  in



Section Four Lake.  Ball  (1948) also  reports the  winterkill



of fish and invertebrates following artificial fertilization.



Ball used much higher fertilizer concentrations and thereby



produced anaerobic conditions throughout the entire lake



during the winter.



     After a summer of normal stratification Section Four



had more than 5 mg/1 at all  levels on January  24,  1970



 (Figure 112).  About 37 cm of powdery snow and 31 cm of ice



covered the lake.  This much snow reduced  light penetration



by more than 99%.  The dates of ice cover  formation are not



known, but were thought to be during  early December both



winters.  Winter oxygen concentrations were  greatly reduced



following fertilization in 1949 and 1950.  Most of the  lake



had less than 2.0 mg/1 by February 26, 1951  (Tanner, 1952).



The increased oxygen concentration by 1969 suggests that



the effects of fertilization were not long lasting.  The



nutrients presumably were tied  up in  the sediments and



essentially non-cycling.



     On January 22,  1971,  following a summer of artificial



destratification, Section Four  had more than 7.0  mg/1 oxygen



at all depths.   Oxygen concentrations were from 1.0 to  2,0



mg/1 greater at all depths than during the previous year.




Snow and ice cover were less during January  22, 1971; 19 cm



and 16 cm respectively.  This indicates that snow and ice

-------
                            404
Figure 112.   Section Four oxygen profiles during January
             1970 and 1971.   The 1970 profile is after a
             summer of normal stratification,  while the
             1971 profile is after a summer of artificial
             aeration.

-------
           Oxygen  (mg/l)



     0    24    6    8    10
a
a>
a
   12-
   16-

-------
                            406
cover was not as great as during the previous year, and



could account in part for the higher oxygen concentrations.



Nevertheless, the increased productivity during artificial



aeration apparently did not deleteriously affect the winter



oxygen concentrations.  I believe that the decomposition



rates were also greatly accelerated by destratification,



and the increased plant biomass was more completely oxidized



before winter stratification began.  Destratification in-



creased the lakes'  average summer temperature, as well as



oxygen concentration.  Both of these conditions should pro-



mote rapid decomposition.



     Hemlock Lake has a much greater oxygen demand than



Section Four.  On December 21,  1969,  levels above 12 meters



were well oxygenated, but no oxygen was observed below 13



meters (Figure 113) .   Snow and ice cover were 18 cm and 12



cm respectively.  By January 24,  1970,  no oxygen was present



below 8 meters and only 3.5 mg/1 were present just under the



ice.  Snow and ice cover were 37 cm and 31 cm respectively.



Conditions undoubtedly became worse before the spring turn-



over,  but no appreciable winterkill occurred.   Many rainbow



trout overwintered without any known ill effects.



     After a summer of artificial aeration, oxygen levels



were much greater at all levels.   About 4 cm of ice and



scant snow covered Hemlock on December 12,  1970.  Oxygen



concentrations were about 8.0 mg/1 at all levels.  This



represents a large increase in oxygen content compared to the

-------
Figure 113.  Hemlock Lake oxygen profiles during December 1969 and 1970,
             and January 1970 and 1971.  The December 1969-January 1970
             profiles are after a summer of normal stratification, while
             the December 1970-January 1971 profiles are after a summer
             of artificial aeration.

-------
                             Oxygen  (mg/l)
 a
 0)
a
      4-
      8-
     12-
              2
              i
468
I  i i  i I  i i  I LL
                XII-21-69
                      XII-12-70
     16-
I  1 I I  I
6
i
                          ' 1-24-70
                                                1-22-71,
8
I

-------
                            409
previous winter.  By January 22,  1971, oxygen values ranged



from 4.5 mg/1 at the bottom to  7.5 mg/1 at the surface.



Snow and ice cover were  15 cm and 25  cm respectively.



Although less stringent  climatic  conditions could account



for some of the increased oxygen, they probably do not



account for much of it.  I attribute  this large increase in



winter oxygen concentrations to summertime aeration of



Hemlock.  Although more  plant biomass was produced during



the summer 1970 compared to the summer 1969, I believe it was



more completely decomposed because of the well-aerated condi-



tions.  These conditions, plus  increased temperatures, per-



mitted more efficient decomposition and oxidation of materi-



als before the onset of  winter  stratification.  The oxygen



concentration was not only increased  by aeration, but the



BOD was undoubtedly lowered by  continuous summertime aera-



tion.



     After the samples were collected January 22, 1971, I ran



the aerator almost continuously for two days  (Figure 114).



This was done to determine its  ability to aerate under the



ice.  Before aeration began, oxygen concentrations at the



lake's center ranged from 4.5 mg/1 at the bottom to 7.5



mg/1 just under the ice  (Figure 115J.  Oxygen isopleths were



not horizontal,  indicating nonuniform rates of oxygen con-



sumption and/or convection currents.  Oxygen concentrations



were above 4.5 mg/1 at all locations  however.  On January



23rd,  after 23 hours of  air injection, not less than 6.5 mg/1

-------
                                                                                     *>•
Figure 114.  Artificial aeration of Hemlock Lake during January 1971.                H
             The compressor was towed onto the lake and run for two
             days.  A rubber air line leads to the aeration tower.

-------

-------
                            412
Figure 115.   Effects of artificial aeration on the oxygen
             regime during January 1971.   The January 22nd
             figure shows the oxygen profiles before
             winter aeration began,  but after a summer of
             artificial aeration.   The January 23rd profile
             is after 24 hours of  air injection and the
             January 24th profile  is after 48 hours of air
             injection.

-------
         Pistance From  Aerator  (m)


        80   70   60   50   40  30   20  10
a
41
a

-------
                            414
oxygen was present at all locations.  A maximum concentra-



tion of 7.7 mg/1 was observed.  Air and water leaked through



the tower walls under the ice.  After four hours of aeration,



the ice was completely melted around the tower.  This melted



region measured about 20 meters by 10 meters after one day



of aeration.  The ice was melted from below by upwelled



water,  since water did not flow onto the surface of the ice.



     On January 24th,  after about 46 hours of aeration,



oxygen concentrations ranged between 6.5 mg/1 and 8.3 mg/1.



Oxygen concentrations inside the aerator were 9.3 mg/1 at



the top.



     These data indicate that this hypolimnion aerator can



be used to aerate under the ice.  If it had not leaked air



and water through its walls,  I would expect the ice around



the tower to remain intact and not melt.  In most situations



it is desirable to prevent melting since open water or weak



ice is a hazard.  Although the under-the-ice oxygen values



were increased substantially,  the relative increase would



have been even greater if conditions had been comparable to



January 1970.   Oxygen absorption is much more efficient when



the concentration is near zero,  than when it is near satura-



tion.

-------
LITERATURE CITED

-------
                     LITERATURE CITED
Abrafcamsson, S. A. A.  1966.
     Dynamics of an isolated population of the crayfish
     Astacus astacus Linne.  Oikos, 17: 96-107.

Aiken, D. E.  1968.
     The crayfish Orconectes virilis;  survival in a region
     with severe winter conditions.  Can. J. Zool. 46: 207-
     211.

Aiken, D. E.  1969.
     Photoperiod, indocrinology and the crustacean molt
     cycle.  Science, 164: 149-155.

Allison, L. E. and C. D. Moodie.  1965.
     Methods of soil analysis, part 2, chemical and micro-
     biological properties.  C. A. Black, ed., Amer. Soc.
     Agronomy, Inc., Madison, Wise., pp. 1387-1389.

American Public Health Association.  1965.
     Standard methods for the examination of water and
     sewage, 12th ed.  Amer. Pub. Health Ass. and Amer. Wat.
     Wks. Ass. and Wat. Poll. Cont. Fed., New York.  522 pp.

Anderson, R. O.  1959.
     A modified flotation technique for sorting bottom fauna
     samples.  Limnol. Oceanogr. 4: 223-225.

Augenfeld,  J. M.  1967.
     Effects of oxygen deprivation on aquatic midge larvae
     under natural and laboratory conditions.  Physiol.
     Zool.   40: 149-158.

Ball, R. C.  1948.
     Relationship between available fish food, feeding habits
     of fish and total fish production in a Michigan lake.
     Mich.  St. Coll. Agr. Exp. Sta., Tech. Bull. 206, 38 pp.

Ball, R. C.  1949.
     Experimental use of fertilizers in production of fish
     food organisms and fish.  Tech. Bull. 210, Michigan
     State College Agr. Exp. Sta.
                            415

-------
                            416
Ball,  R. C. and D. W. Hayne.  1952.
     Effects of the removal of the fish population on the
     fish food organisms of a lake.  Ecol.  33: 41-48.

Ball,  R. C. and H. A. Tanner.  1951.
     The biological effects of fertilizer on a warm awter
     lake.  Tech. Bull. 223, Mich. State College Agr. Exp.
     Sta.

Bardach, J. E.  1955.
     Certain biological effects of thermocline shifts.
     Hydrobiologia,  7s 309-324.

Bardach, J. E. and R. G. Bjorklund.  1957.
     The temperature sensitivity of some American fresh-
     water fishes.  The Amer. Naturalist.  91: 233-251.

Earnhardt, H.  1967.
     Aeration of Wahnback Reservoir without changing the
     temperature profile.  Amer. Wat. Wks. Ass.,  J.,
     (Aug.): 943-964.

Barrett, P. H.  1952.
     Effects of alkalinity on adsorption and regeneration
     of phosphorus in natural lakes.  Ph. D. Thesis,
     Michigan State College, E. Lansing, Michigan.

Beeton,  A. M.  1960.
     The vertical migration of Mysis relicta in Lakes Huron
     and Michigan.  J. Fish. Res.  Bd. Canada.  17: 517-539.

Black,  E. C., E. J  Fry and V. Black.  1954.
     The influence of COa on the utilization of oxygen by
     some freshwater fish.  Can. J  Zool.  32: 408-420.

Borges,  H. M.  1950.
     Fish distribution studies, Niaugua arm of the Lake of
     the Ozarks, Missouri.  J, Wildlife Manag.  16: 16-23.

Bovbjerg, R. V.  1964.
     Dispersal of aquatic animals relative to density.
     Verh. Internat. Verein. Limnol.  15: 879-884.

Brett,  J. R.  1952.
     Temperature tolerance in young Pacific salmon, genus
     Oncorhvnehus.  J. Fish Res. Bd. Canada, 9: 265-323.

Brook,  J. L. and S.  I. Dodson.  1965.
     Predation, body size and composition of plankton.
     Science, 150s 28-35.

-------
                            417
Brundin, L,  1951.
     The relation of 02-microstratification at the mud  sur-
     face to the ecology of the profundal bottom  fauna.
     Kept. Inst. Freshwat. Res. Drottingholm, 32? 32-42.

Brundin, L.  1953.
     The bottom faunistical lake type system and  its appli-
     cation to the southern hemisphere.  Moreover a theory
     of glacial erosion as a factor of productivity in  lakes
     and the oceans.  Verh. Internat. Ver. Limnol.  8:  288-
     297.

Bryce,  D.  1965.
     Notes on some Chironomidae  (Diptera) from acid peat
     pools.  Entomologist 98: 49-53.

Burbige, R. G.  1969.
     Age, growth, length-weight relationship, sex ratio and
     food habits of American smelt, Osmerus mordax  (Mitchell)
     from Gull Lake, Michigan.  Trans. Amer. Fish. Soc.
     98; 631-640.

Camougis, G. and J. H. Hichar.  1959.
     Some studies on crayfish distribution in a small pond.
     Amer. Midi. Nat., 62: 227-321,,

Coffin, C. C., F. R. Hayes, L. H. Jodrey and S. G. Whiteway.
     1949.
     Exchanges of materials in a lake as studied by the addi-
     tion of radioactive phosphorus.  Canad. J. Res.  27;
     207-222.

Cooper, W. E.  1965.
     Dynamics and productivity of a natural population of
     fresh-water emphipod, Hyalella azteca.  Ecol. Monog.
     35: 377-394.

Gushing, D. H.  1951.
     The vertical migration of planktonic Crustacea.
     Biol. Rev.  26s 158-192.

Dendy,  J. S.  1946.
     Fish distribution, Norris Reservoir, Tennessee, 1943.
     II.  Depth distribution of fish in relation to environ-
     mental factors, Norris Reservoir.  Rept. Reelfoot  LakB
     Biol. Sta., 9s 114-135.

	.  1948.
     Predicting depth distribution of fish in three TVA
     storage type reservoirs.  Trans. Amer. Fish. Soc.  for
     1945.  75: 65-71.

-------
                            418
Dice, L. R. and H. J. Leraas.  1936.
     A graphical method for comparing several sets of
     measurements.  Contr. Lab. of Vert. Gen. Univ. of
     Mich.  3s 1-3.

Dugdale, R. C.  1955.
     Studies in the ecology of the benthic Diptera of Lake
     Mendota.  Ph. D. Thesis, Univ. of Wisconsin, Madison.

Edwards, R. W.  1958.
     The effect of larvae of Chironomus rjparius Meigon on
     the redox potential of settled activated sludge.  Ann.
     Appl. Biol.  46s 457-464.

Eipper,  A. S.  1960.
     Managing farm ponds for trout production.  New York
     State Coll. Agri. Cornell Ext. Bull. 1036,  Ithaca.
     29 pp.

Eschmeyer, W. R.  1938.
     Experimental management of a group of small Michigan
     lakes.  Trans. Am. Fish. Soc., 67: 12O-129.

Fast, A. W.  1968.
     Artieifial destratification of El Capitan Reservoir by
     aeration, part I, effects on chemical and physical
     parameters.  Calif. Dept. Fish and Game, Fish Bull.
     141: 97 pp.

Fast, A. W.  1970.
     An evaluation of the efficiency of zoobenthos separation
     by sugar flotation.  Prog. Fish. Cult.  32s 212-216.

Fast, A. W.  1971.
     Effects of artificial destratification on zooplankton
     depth distribution.  Trans. Amer. Fish Soc.  2; 355-358.

Fast, A. W. and J. St. Amant.  (In preparation.)
     Nightime artificial aeration of Puddingstone Reservoir,
     Los Angeles County, California.

Ferguson, R. G.   1958.
     The preferred temperature of fish and their midsummer
     distribution in temperate lakes and streams.  J. Fish.
     Res. Bd. Canada, 15: 607-624.

Fitzgerald, G. P.  1970.
     Aerobic lake muds for the removal of phosphorus from
     lake waters.  Limnol. Oceanogr.  15: 550-555.

-------
                            419
Fry, F. E. J.  1937.
     The summer migration of the  cisco, Leuchichtys  artedi
     (LeSeur), in Lake Nipissing, Ontario.  Univ. Toronto
     Stud., Biol. Ser-, 44  (Pub.  Ontario Fish. Res.  Lab.,
     No. 55), 91 pp.

Galligan, J. P.  1962.
     Depth distribution of  lake trout  and associated species
     in Cayuga Lake, New York.  N. Y.  Fish  and Game  J.
     9s 44-68.

Garside, E. T. and J. S. Tait.  1958.
     Preferred temperature  of rainbow  trout (Salmo gairdneri
     Richardson) and its unusual  relationship to acclimation
     temperature.  Can. J.  Zool.  36s  563-567.

Gleason, G. R. Jr.  1961.
     Factors influencing the ecological distribution of
     Tendipes decorus  (Joh.) in Coldwater Lake, Isabella
     County, Michigan.  Ph. D. Thesis, Mich. State Univ.,
     E. Lansing, Michigan,  103 pp.

Goldman, C. R.  1962.
     Primary productivity and micro-nutrient limiting factors
     in some North American and New Zealand lakes.
     Comm. XV Internat. Congr. Limnol.

Hall, D. J.  1964.
     An experimental approach to  the dynamics of a natural
     population of Daphnia  qaleata mendotae.  Ecol.  45?
     94-112.

Harder, G. F.  1968.
     Reactions of plankton  organisms to water stratification.
     Limnol and Ocean.  13; 156-168.

Hardy,  A. C.  1956.
     The open sea.  Collins, London.   335 pp.

Hargrave, B. T.  1969.
     Similarity of oxygen uptake  by benthic communities.
     Limnol. Oceanogr., 14s 801-805.

Harvey, H. W., L. H. N. Cooper, M. V.  Leboor and F.  S. Russell.
     1935.
     Plankton production and its  control.   J. Mar. Biol.
     Assoc. U. K. 20: 407-442.

Hasler, A. D.  1963.
     Wisconsin 1940-1961.   In Limnology in  North America.
     D. G. Frey, ed., The University of Wisconsin Press,
     Madison, Wisconsin, p. 59«

-------
                            420
Easier,  A. D.  1966.
     Underwater guideposts, homing of salmon.  University
     of Wisconsin Press, Madison, Wisconsin.  155 pp.

Hilsenhoff, W. L.  1966.
     The biology of Chironomus plumosus  (Diptera:  Chirono-
     midae) in Lake Winnebago, Wisconsin.  Ann. Entomol.
     Soc. Amer.  59: 453-73.

Hilsenhoff, W. L.  1967.
     Ecology and population dynamics of Chironomus plumosus
     (Diptera:  Chironomiade)  in Lake Winnebago, Wisconsin.
     Ann. Entomol. Soc. Amer.   60: 1183-94.

Hilsenhoff, W. L. and R. P. Narf.  1968.
     Ecology of Chironomidae,  Chaoborinae and other benthos
     in fourteen Wisconsin lakes.  Ann. Entomol. Soc. Amer.
     61: 1173-81.

Hodgson, W. C. and I. D. Richardson.  1949.
     The experiments on the Cornish Pilchard Fishery in
     1947-48.  Fish Invest., ser. 2, 17(2).

Hooper,  F. F., R. C. Ball and H. A. Tanner.  1952.
     An experiment in the artificial circulation of a small
     Michigan lake.  Amer. Fish. Soc.,  Trans., 82: 222-241.

Horak,  D. L. and H. A. Tanner.  1964.
     The use of vertical gill nets in studying fish depth
     distribution, Horsetooth Reservoir, Colorado.  Trans.
     Amer. Fish. Soc.  93s 137-145.

Hrbacek, J., M. Dvorakova, V.  Korinek and L. Prochazkova.
     1961.
     Demonstration of the effect of the fish stock on the
     species composition of zooplankton and the intensity
     of metabolism of the whole plankton association.
     Verh. Internat. Verein. Limnol.  14: 192-195.

Hutchinson, G. E.  1941.
     Limnological studies in Connecticut.  IV.  The mechanism
     of intermediary metabolism in stratified lakes.  Ecol.
     Monogr., 11: 21-60.

Hutchinson, G. E. and V. T. Bowen.  1947.
     A direct demonstration of the phosphorus cycle in a
     small lake.  Natl. Acad.  Sci. Proc., 33: 148-153.

Hutchinson, G. E.  1957.
     A treatise on limnology.   Vol. i,  Geography, physics
     and chemistry.  John Wiley & Sons, New York. 1015 pp.

-------
                            421
Irwin, W. H., J. M. Symons  and G.  G.  Robeck.   1969.
     Impoundment destratification  by  mechanical pumping.
     Proc. Amer. Soc, Civil Engin.  92s  21-40,

Johannes, R. E. and M. Satoni.   1967.
     Measuring organic matter retained by  aquatic  inverte-
     brates.  J. Fish, Res. Bd.  Canada.  24;  2467-2471.

Jonasson, P- M.  1955.
     The efficiency of sieving techniques  for  sampling fresh-
     water bottom fauna.  Oikos  6s  183-207.

Jonasson, P- M. and J. Kristiansen*   1967.
     Primary and secondary  production in Lake  Esrom.  Growth
     of Chironomus'anthracinus in  relation to  seasonal cycles
     of phytoplankton and dissolved oxygen.   Int.  Revue ges.
     Hydrobiol.  52: 163-217.

Jones, F. R. H.  1952.
     The swimbladder and vertical  movement of  teleostean
     fish,  J. Exptl. Biol.  29: 94-109.

King, D. L. and R. C. Ball.  1964.
     A quantitative biological measure of  stream pullotion.
     J. Wat. Poll. Control  Fed.  36(5) s  650-653.

Knoppert, P. L., J. J. Rook, Tj. Hofker  and G. Oskam.  1970.
     Destratification experiments  at  Rotterdam.  J. Amer.
     Wat. Wks. Assoc.  July: 448-454.

Koberg, G. E.  1964.
     Elimination of thermal stratification in  Lake Wohlford,
     California.  U. S. Geol. Surv.,  Prof. Pap.,  (501-D) s
     D 190-D 192.

Koberg, G. E., and M. E. Ford.   1965.
     Elimination of thermal stratification in  reservoirs and
     the resulting benefits.  U. S. Geol..  Surv., Wat. Supp.
     Pap., (1809-M): 1-28.

LaRow, E. J.  1968.
     A persistent diurnal rhythmn  in  Chaoborus larvae.  I.
     The nature of the rhythmicity.   Limnol. Oceanogr., 13:
     250-256.

Lenz, F.  1925.
     Ghironomiden undseetypenlehre.   Die Naturwissenschaften
     13: 5-10.

Lindeman, R. L.  1942.
     The trophic-dynamic aspect  of ecology.  Ecology 23s
     399-418.

-------
                            422
Lorenzen,  C. J.  1967.
     Determination of chlorophyll and pheo-pigments:  spec-
     trophotometric equations.  Limnol. Oceanogr., 12: 343-
     346.

Lundbeck,  J.  1926.
     Die Bodentierweld norddeutscher Seen- Arch. Hydrobiol.
     Suppl,  7: 1-473.

Lundbeck,  J.  1936.
     Untersuchungen liber die Bodenbesiedelung der Alpenrand-
     seen.  Arch. Hydrobiol.  10s 207-258.

McConnell, W. J.  1968.
     Limnological effects of organic extracts of litter in
     a southwestern impoundment.  Limnol. Oceanogr.  13:
     343-349.

McCarter,  J. A., F. R. Hayes, L. H. Jodrey and M. L. Cameron.
     1952.
     Movement of materials in the hypolimnion of a lake
     studied by the addition of radioactive phosphorus.
     Can.  J. Zool., 30; 128-133.

McEwen, G. F.  1941.
     Observations on temperature, hydrogen-ion concentration
     and periods of stagnation and overturn in lakes and
     reservoirs of San Diego, California Scripps. Inst.
     Ocean., Bull., 4(9)s 219-260.

Miller, R. B.  1941.
     A contribution to the ecology of the Chironomidae of
     Costello Lake, Algonquin Park, Ontario.  Pub. Ontario
     Fish. Res. Lab. No. 60, U. of Toronto Press, Toronto,
     Canada, 63 pp.

Momot,  W.  T.  1967.
     Population dynamics and productivity of the crayfish
     Orconectes virilis in a marl lake.  Amer. Midland
     Naturalist,  78; 55-81.

Momot,  W.  T. and H, Cowing.  1970.
     Differential seasonal migration of the crayfish
     Orconectes virilis  (Hagen), in small lakes.  Michigan
     Dept. of Natural Resources, Inst. Fish. Res, Rept.
     1767: 14 pp.   (mimeo.)

Moore,  E.  W.  1939.
     Graphic determination of carbon dioxide and the three
     forms of alkalinity.  J. Amer. Water Wks. Assoc., 31;
     51-66.

-------
                             423
Mortimer, C. H.  1941.
     The exchange of dissolved  substances between lake mud
     and water.  J. Ecol.  29s  280-329.  1942.  Ibid.  30:
     147-201.

Mundie, J. H.  1959.
     The diurnal activity of the  larger  invertebrates at
     the surface of Lac La Ronge, Saskatchewan.  Can. J.
     Zool.  37: 945-956.

Murphy, G. I.  1962.
     Effects of mixing depth and  turbidity on the produc-
     tivity of fresh-water impoundments.  J. Amer. Fish.
     Soc.  91: 69-76.

Northcote, T. G.  1964.
     Use of high-frequency echo sounder  to record distribu-
     tion and migration of Ghaoborus larvae.  Limnol,
     Oceanogro  9s 87-91.

Northcote, T. G., H. W. Lorz and  J. C. MacLeod.  1964.
     Studies on diel vertical movement of fishes in a
     British Columbia lake.  Verh. Internat. Verein.
     Limnol.  15s 940-946.

Ogborn, C. M.  1966.
     Aeration system keeps water  tasting fresh.  Publ. Wks.
     (Aug.)

Parsons, T. R. and J. D. H. Strickland.  1963.
     Discussion of spectrophotometric determinations of
     marine plant pigments, with  revised equations for
     ascertaining chlorophills  and carotenoids.  J. Mar.
     Res., 21s 155-163.

Pearson, B. E.  1952.
     The behavior of a sample of  hybrid  trout  (Salvelinus
     fontinalis x Cristivomer namaycush) in a vertical
     temperature gradient.  (M. S. Thesis, in Ontario Fish.
     Res. Lab. Library, Toronto;  24 pp.)

Reisch, D. J.  1959.
     A discussion of the importance of screen size in wash-
     ing quantitative marine bottom samples.  Ecol.  40:
     307-309.

Rigler, F. H.  1956.
     A tracer study of the phosphorus cycle in lakewater.
     Ecology.  37s 550-562.

Rigler, F. H.  1964.
     The contribution of zooplankton to  the turnover of
     phosphorus in the epilimnion of lakes.  Canad. Fish
     Cult.,  32: 3-9.

-------
                            424
Robinson, E. L.. W. H. Irwin and J. M. Syraons.  1969.
     Influence of artificial destratification on plankton
     populations in impoundments.  Trans. Kentucky Acad.
     Sci.  30s 1-18.

Roth, J. C.  1968.
     Benthic and limnetic distribution of three Chaoborus
     species in a southern Michigan lake  (Diptera, Chao-
     borinae).  Limnol. Oceanogr.  13: 242-249.

Saunders, G. W., F. G. Trama and R. W. Bechmann.  1962,
     Evaluation of a modified C-14 technique for shipboard
     estimation of photosynthesis in large lakes.  Great
     Lakes Research Division, Publ. 8s 61 pp.

Schmidt, P- S.  1915.
     Uber den Energiegehalt der Seen, mit beispielen vom
     Lunzer Untersee nach Messungen miteinem einfacheu
     Temperaturlot.  Int. Rev. Ges. Hydr. Hydrog. Hydrograph.,
     Suppl.  6(1) s 1-25.

Scott,  I. D.  1921.
     Inland lakes of Michigan.  Mich. Geol. and Biol. Survey,
     Publ. 30, Geol. Ser. 25, xxi - 383.

Sharpe,  F. F.   1961.
     A midsummer plant of rainbow trout in a stratified
     Tennessee impoundment.  Prog. Fish-Cult. 23s 152-155.

Siler,  D. H.  1968.
     A comparison of invertebrate benthos populations in
     four sink lakes sixteen years after fertilization.
     M.  S. Thesis,  Michigan State University. E. Lansing,
     Michigan.

Stahl,  J. B.  1966.
     The ecology of Chaoborus in Myers Lake,  Indiana.
     Limnol. Oceanogr.  11s 177-183.

Strickland, J. D.  H. and T. R. Parons.  1965.
     A manual of sea water analysis.  Fish. Res. Ed. Canada,
     Bull. 125, end ed., 203 pp.

Sullivan, C. M. and K. D. Fisher.  1953.
     Seasonal fluctuations in the selected temperature of
     speckled trout Salvelinus fontinalis  (Mitehill).
     J.  Fish.  Res.  Bd. Canada.  10s 187-195.

Tanner,  H. A.   1952.
     Fertilization of Michigan trout lakes.  Ph. D. Thesis,
     Michigan State Univ.,  E. Lansing, Michigan.  186 pp.

-------
                            425
Tanner,  H. A.  1960.
     Some consequences of adding fertilizer to five Michigan
     trout lakes.  Trans. Amer. Fish. Soc.  89: 198-205.

Teraguchi, M. and T. G. Northcote.  1966.
     Vertical distribution and migration of Chaoborus
     flavicans larvae in Corbett Lake, British Columbia.
     Limnol. Oceanogr.  11: 164-176.

Welch,  P- S.  1948.
     Limnological methods.  McGraw-Hill Book Co., New York.
     381 pp.

Wetzel,  R. G.  1965.
     Necessity for decontamination of filters in C14
     measured rates of photosynthesis in fresh water.
     Ecology, 46: 540-542.

Wilkins, L. P.  1952.
     Relationships of fish populations to the invertebrate
     fauna in two small ponds.  Thesis, Michigan State
     College, E. Lansing, Michigan.

Wirth,  T. L., R. C. Dunce, P. D. Uttormark and W. Hilsenhoff.
     1970.
     Manipulation of reservoir waters for improved quality
     and fish population response.  Wisconsin Dept. of Nat.
     Resources, Madison, Wise., Res. Rept. 62, 23 pp.

Wynne-Edwards, V. C.  1962.
     Animal dispersion in relation to social behavior.
     Oliver and Boyd, London.  563 pp.

-------
APPENDIX

-------
Table A-l.
Ilerilock Lake 17 day peri nhyton weights  during  1969
and .1970.  Samples were  incubated  on  plastic slides
durinp June,'.July and August  each  year.   Four  slides
were incubated at each of"  five depths.  Ash-free  dry
weight is shown for each sanple.

0.0
0.9
Date Collected
0.0140
0.0106
0.0132
0.0103
0.0184
0.0159
0.0155
0.0068
Date Collected
0.0062
0.0075
0.0091
0.0112
0.0118
0.0186
0.0122
0.0133
Date Collected
0.0122
0.0145
0.0146
0.0305
0.0159
0.0213
0.0305
0.0214
Date Collected
0.0254
0.0317
0.0206
0.0225
0.9366
0.0496
0.0489
0.0318
DEPTHS IN METERS
1.8 2.7
7-2-69
0.0177
0.0144
0.0161
0.0172
7-19-&9
0.0141
0.0108
0.0149
0.0072
7-28-69
0.0223
0.0277
0.0282
0.0219
8-5-69
0.0341
0.0440
0.03S 2
O.Gv-'+O

0.0116
0.0114
0.0126
0.0103

0.0176
0.0132
0.0172
0.0165

O.C176
0.1451
0.0470
0.0167

O.C236
o.o '319
0.0394
O.C437
Date Collected 8-12-69
0.0553
0.0811
0.0604
0.0620
0.0416
0.0541
0.0472
0.0671
0.0337
0.0289
0.0345
0.0327
0.0467
0.0564
0.0713
0.0581
3-7
Date Set
0.0090
0.0084
0.0125
0.0115
Date Set
0.0027
0.0035
0.0100
0.0091
Date Set
0.0187
0.0152
0.0098
0.0231
Date Set
0.0061
0.0134
0.0108
0.0143
Date Set
0.0340
0.0465
0.0341
0.0 '507
4.5
6-15-69
0.0059
0.0043
0.0068
0.0091
7-2-69
0.0039
0.0063
0.0031
0.0059
7-9-69
0.0033
0.0063
0.0040
0.0047
7-19-69
0.0095
0.0172
0.0159
0.0110
7-28-69
0.0495
0.0140
0.0217
0.0263

-------
Table A-l (Continued)

0.0
DEPTHS IN
0.9 1.8
METERS
2.7
Date Collected 8-22-69
0.0374
0.026?
0.0337
0.0416
0.0426 0.0349
0.0514 0.0491
0.0460 0.0429
0.0576 0.0416
0.0378
0.0414
0.0462
0,0528
Date Collected 8-29-69
0.0522
0.0411
0.0321
0.0362
0.0618 0.0461
0.0365 0.0428
0.0450 0.0491
0.0324 0.0375
0.0390
0.0435
0.0474
0.0328
3.7
Date Set
0.0479
0.0466
0.0407
0.0486
Date Set
0.0318
0.0267
0.0258
0.0285
4.5
8-5-69
0.0253
0.0275
0.0365
0.0301
8-12-69
0.0226
0.0198
0.0198
0.0245

-------
Table A-l (Continued)

0.0
D3PTHS IN METERS
0.9 1.8 2.7
Date Collected 7-2-70
0.0096
0.0138
0.0194
0.0124
0.0172 0.0091
0.0119 0.0164
0.0102 0.0118
0.0070 0.0094
0.0123
0.0166
0.0038
0.0048
Date Collected 7-19-70
0.0462
0.0385
0.0444
0.0452
0.0503 0.0161
0.0306 0.0161
0.0347 0.0188
0.0337 0.0213
0.0449
0.0691
0.0861
0.0658
Date Collected 7-26-70
0.0366
0.0374
0.0387
0.0422
0.0338 0.0273
0.0326 0.0353
0.0297 0.0287
0.0371 0.0290
0.3219
0.2668
0.1658
0.2223
Date Collected 8-5-70
0.0534
0.0526
0.0510
0.0439
0.0514 0.0740
0.0597 0.0583
0.0550 0.0575
0.0296 0.0498
0.0800
0.1037
0.0918
0.1063
Date Collected 8-12-70
0.0581
0.0639
0.0520
0.0578
0.0564 O.OS13
0.0579 0.0560
0.0619 0.0608
0.0572 0.0622
O.Of^07!
0.0654
0.0727
0.0639
3.7
Date Set
0.0229
0.0029
0.0030
0.0021
Date Set
0.0393
0.0426
0.0367
0.0467
Date Set
0.1157
0.1083
0.1187
0.0660
Date Set
0.0345
0.0591
0.0422
0.0363
Date Set
0.0333
0.0323
0.0367
0.0404
4.5
6-15-70
0,0033
0.0015
0.0016
0.0040
7-2-70
0.0219
0.0266
0.0109
0.0203
7-9-70
0.1451
0.0808
0.1168
0.0816
7-19-70
0.0241
0.0240
0.0286
0.0221
7-26-70
0.0192
0.0290
0.0270
0.0138

-------
Table A-l (Continued)
0.0
DEPTHS IN
0.9 1.8
METERS
2.7
Date Collected 8-21-70
0.0669
0.0636
0.0714
0.0748
0.0645 0.0626
0.0664 0.0645
0.0648 0.0698
0.0650 0.0662
0.0902
0.0789
0.0851
0.0745
Date Collected 8-29-70
0.0492
0.0559
0.0643
0.0431
0.0648 0.0718
0.0676 0.0669
0.0653 0.0669
O.C520 0.0878
0.0429
O.C493
0.0595
0.0455
3.7
Date Set
0.0612
0.0578
0.0642
0.0592
Date Set
O.OR14
0.0893
0.0883
0.0926
4.5
8-5-70
0.0482
0.0358
0.0398
0.0463
8-12-70
0.1153
0.1256
0.1282
0.0897

-------
  Table A-2.
   Hemlock  Lake  accumulative periphyton weights
   during  1969 and  1970.   Samples were incubated
   starting June 15  each year and a portion was
   retrieved at  different  times during the summer.
   Samples  were  incubated  on plastic slides.  These
   slides were incubated at each of five depths.
   Ash-free dry  weight  is  shown for each sample.
0.0
0.9
DEPTHS II
1.8
T METERS
2.7
3.7
4.5
Date Collected   7-19-69
0.0142
0.0549
0.0278
0.0474
0.0383
0.0322
0.0518
0.0463
0.0338
0.0216
0.0198
0.0301
                                Date Set   6-15-69
0.0224
0.0179
0.0104
0.0136
0.0069
0.0122
Date Collected   3-5-69
0.0859
0.0494
0.0354
0.1486
0.1127
0.0809
0.1652
0.1179
0.0899
0.0953
0.0712
0.0822
                                Date Set   6-15-69
0.0685
0.0340
0.0245
0.0282
0.0282
0.0423
Date Collected   8-22-69
0.1773
0.0841
0.0982
0.1855
0.1427
0.2042
0.1902
0.1217
0.1257
0.1266
0.2471
0.1032
                                Date Set   6-15-69
0.0790
0.0898
0.0748
0.0722
0.0506
0.0849

-------
Table A-2 (Continued)
0.0
DEPTHS IN METERS
0.9 1.8 2.7
Date Collected 7-19-70
0.0861
0.0953
0.0619
0.07^6 0.0753
0.0785 0.0410
0.0626 0.0890
0.0497
0.1180
0.1491
Date Collected 8-5-70
0.0631
0.0706
0.0571
0.1425 0.1187
0.1338 0.1072
0.1460 0.1511
0.1571
0.1464
0.1301
Date Collected 8-21-70
0.244-9
0.2262
0.2066
0.2712 0.2796
0.2598 0.2360
0.2806 0.2515
0.2468
0.2351
0.1801
3.7
Date Set
0.0338
0.0527
0.1237
Date Set
0.0532
0.0690
0.0874
Date Set
0.1200
0.1110
0.1202
4.5
6-15-70
0..0217
0.0480
0 . 0424
6-15-70
0.0393
0.0473
0.0337
6-15-70
0.0949
0.0851
0.1043

-------
Table A-3.
Section Four Lake 17 day periphyton weights durinp
1969 and 1970.  Samples were incubated on plastic
slides durinp June, July and Aupust each year.   Four
slides were incubated at each of five depths.   Ash-
free dry weight is shown for each sample.
0.0
DEPTHS IN METERS
1.8 3.7 5.4
Date Collected 7-2-69
0.0110
0.0120
0.0118
0.0155
O.C1J3 0.0220
0.01 £-9 0.0173
0.0137 0.0199
0.0129 0.0135
0.0118
0.0141
0.0075
0.0075
Date Collected 7-19-69
0.0129
0.0108
0.0107
0.0226
0.0061 0.0235
0.0123 0.0129
0.0112 0.0107
0.0145 O.OC83
0.0103
0.0177
0.0202
0.0210
Date Collected 7-28-69
0.0066
0.0114
0.0078
0.0095
0.0105 0.0090
0.0109 0.0114
0.0105 0.0137
0.0084 0.0168
0.0079
0.0118
0.0150
0.0103
Date Collected 8-6-69
0.0060
0.0070
0.0074
0.0161
0.0103 0.0118
0.0117 0.0122
0.0127 0.0069
0.0101 0.0079
0.0077
0.0101
0.0144
0.0098
Date Collected 8-12-69
0.0071
0.0070
0.0066
0.0119
0.0065 0.0102
0.0147 0.0125
0.0095 0.0089
0.0275 0.0099
0.0235
0.0142
0.0115
0.0191
7.3
Date Set
0.0074
0.0058
0.0103
0.0083
Date Set
0.0186
0.0190
0.0177
0.0196
Date Set
0.0068
0.0084
0.0039
0.0087
Date Set
0.0084
0.0137
0.0057
0.0087
Date Set
0.0063
0.0017
0.0052
0.0157
9.3
6-15-69
0.0092
0.0072
0.0049
0.0072
7-2-69
0.0111
0.0114
0.0127
0.0180
7-9-69
0.0104
0.0081
0.0016
0.0088
7-19-69
0.0080
0.0086
0.0092
0.0107
7-28-69
0.0027
0.0045
0.0061
0.0197

-------
  Table A-3 (Continued)
                    DEPTHS IN METERS
0.0 .1..8 3.7
Date Collected 8-22-69
5.4

7.3 9.3
Date Set 8-6-69
0.0020
0.0145
0.0030
0.0064
0.0057
0.0180
0.0152
0.0112
0.0217
0.0175
0.0094
0.0179
0.0184
0.0108
0.0047
0.0146
0.0099
0.1176
0.0071
0.0125
0.0084
0.0102
0.0088
0.0088
Date Collected   8-29-69
0.0100
0.0059
0.0146
0.0207
0.0070
0.0109
0.0034
0.0042
0.0162
0.0147
0.0194
0.0160
                                .Date Set   8-12-69
0.0145
0.0167
0.0127
0.0141
0.0157
0.0120
0.0198
0.0093
0.0114
0.0084
0.0046
0.0071

-------
Table A-3 (Continued)
 0.0
1.3
DEPTHS IN METERS
    5-7        5.4
7.3
9.3
Date Collected 7-2-70
0.0031
0.0017
0.0046
0.0018
0.0056 0.0025
0.0091 0.0061
0.0066 0.0063
0.0049 0.0084
0.0075
0.0083
0.0077
0.0078
Date Collected 7-19-70
0.0068
0.0076
0.0062
0.0069
0.0149 0.0172
0.0196 0.0252
0.0201 0.0305
0.0165 0.0207
C.0198
0.0242
0.0166
0.0290
Date Collected 7-26-70
C.0141
0.0103
0.0120
0.0150
O.C305 0.0980
0.0285 0.0334
0.0254 0.0415
0.0667 0.0378
0.0244
0.0307
0.0276
0.02^6
Date Collected 8-5-70
0.0121
0.0182
0.0179
0.0154
0.0230 0.0195
0.0228 0.0219
0.0258 O.G280
0.0242 C.0191
0.0210
0.0177
0.0269
0.0138
Date Collected 8-12-70
0.0114
0.0118
0.0155
0.0189
0.0300 0.0344
0.0339 0.0111
0.0192 0.0194
0.0339 0.0130
0.0203
0.0399
0.0279
O.C181
Date Set
0.0078
0.0094
0.0124
0.0135
Date Set
0.0268
0.0178
0.0201
0.1231
Date Set
0.0195
0.0291
0.0187
0.0281
Date Set
0.0243
0.0193
0.0159
0.0169
Date Set
0.0189
0.0246
C.021C
0.0210
6-15-70
0.0070
0.0044
0.0041
0.0054
7-2-70
0.0133
0.0178
0.0168
0.0110
7-9-70
0.0259
0.0233
0.0247
0.0123
7-19-70
0.0114
0.0126
0.0110
0.0136
7-26-70
0.0155
0.0156
0.0162
0.0167

-------
Table A-3 (Continued)
0.0
1.8
DEPTHS IN METERS



    3.7        5.4
7-3
9-3
Date Collected
0.0192
0.0196
0.0177
0.0268
0.0270
0.0310
0.0246
0.0272
Date Collected
0.0148
0.0153
0.0182
0.0145
0.0287
0.0359
0.0381
0.0378
8-21-70
0.0246
0.0241
0.0262
0.0269
8-29-70
0.0288
0.0229
0.0336
0.0253

0.0272
0.0251
0.0210
0.0221

0.0352
0.0356
0.0369
0.0189
Date Set
0.0178
0.0208
0.0180
0.0183
Date Set
0.0268
0.0277
0.0346
0.0286
8-5-70
0..0110
0.0101
0.0144
0.0084
8-12-70
0.0141
0.0166
0.0140
0.0199

-------
Table A-4.
Section Four Lake accumulative periphyton weiphts
during 1969 and 1970.  Samnles were incubated
starting June 15 each year and a portion was retrieved
at different times during the summer.   Samples were
incubated on plastic slides.  These slides were incu-
bated at each of five depths.  Ash-free dry weiqht is
shown for each sample.
  0.0
 1.8
DEPTHS IN METERS

    3.7        5.4
7-3
9.3
Date Collected
0.
0.
0.
0393
0152
0107
0.
0.
0.
0252
0185
0228
Date Collected
0.
0.
0.
0318
0356
0304
0.
0.
0.
0482
0536
0606
Date Collected
0.
0.
0.
153^
0299
2914
0.
0.
0.
0757
1378
1517
7-19-69
0.
0.
0.
8-6-69
0.
0.
0.
8-22-69
0.
0.
0.
Date Set
0166
0253
0184
0.
0.
0.
0184
0228
0111
0.
0.
0.
0204
0141
0185
Date Set
0472
0735
0390
0.
0.
0.
0327
0282
0357
0.
0.
0.
0247
0335
0657
Date Set
1010
2195
0683
0.
0.
0.
0575
2635
0511
0.
0.
0.
0315
0335
0304
6-15-69
0
0
0
.0120
.0178
.0139
6-15-69
0
0
0
.0237
.0264
.0283
6-15-69
0
0
0
.0311
.0304
.0303

-------
Table A-4 (Continued)
0.0
D3PTH3 IN METERS
1.8 3-7 5.4
Date Collected 7-19-70
0.0142
0.0107
0.0092
0,0323 0.0355
0.0366 0.0453
0.0309 0.0332
0.0383
0.0429
0.0425
Date Collected 8-5-70
0.0291
0.0325
0.0293
0.0630 0.0648
0.0652 0.0427
0.0633 0.0701
0.0575
0.0564
0.0577
Date Collected 8-21-70
0.0^90
0.0536
0.0535
0.1083 0.1256
0.1214 0.1090
0.0890 0.1355
0.1000
C.0653
0.1278
7.3
Date Set
0.027«
0.0301
0.0015
Date Set
0.0277
0.0425
0.0350
Date Set
0.0969
0.0748
0.0772
9.3
6-15-70
0.0583
0.0433
0.0281
6-15-70
0.0195
0.0211
0.0404
6-15-70
0.0405
0.0340
0.0558

-------
Table A-5.  Hemlock l,a-ke  loobenthos  collected  with  an  Fl-.man dredpe durinc I960 and 197n.  Numbers
            and wet weights  for  the  seven  most abundant taxa are *hown in this table for each
            samnle.   125  sannles  were  collected each simmer.  The less abundant taxa are listed
            in Table  A-6,  Tn verify the  total orpanisms for a piven sample, consult both tables.
            llenth  is  in meters and weipjit  is  in
Sample
Number
6-13-69
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
7-4-69
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
Depth

15.6
16.0
15.5
15.1
15.5
14.4
14.2
13.5
12.8
12.2
10.5
9.6
7.5
9.2
8.3
6.9
6.2
4.6
6.4
4.6
3.0
2.3
1.6
1.0
0.3

16.5
16.9
16.0
15.5
15.5
14.4
13.2
12.3
13.5
11.4
10.7
10.1
8.7
9.4
7.8
7.1
6.0
5.7
5.0
Oligochaeta
No. Wt.

0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000

0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
o o.nooooo
0 0.000000
0 0.000000
0 0.000000
0 0.000000
o o.nooooo
0 0.000000
1 0. 000807
1 0. 1101200
0 0.000000
0 0.000000
0 0.000000
Chironomid L.
No. Kt.

0
0
0
0
0
0
0
0
0
0
0
2
23
14
32
14
41
22
5
19
14
14
9
17
51

n
0
0
0
0
0
0
1
0
1
4
1
0
0
0
8
1
0
74

0.000000
0.000000
0.000000
0.00(1000
0.000000
0.000000
0.000000
0.000000
0.000000
n. oooooo
0.000000
0.001062
0.004367
0.003997
0.009144
0.007068
0.040451
0.008452
0.001444
0.008793
0.003842
0.004136
0.005714
0.002697
0.004725

0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000164
0.000000
0.00049S
0.001055
0.000160
0.000000
n. oooooo
0.000000
0.004809
0.000563
0.000000
0.020104
Chironomid P.
No. Wt.

0
0
n
0
0
0
n
0
0
0
0
0
0
0
0
0
0
0
0
0
0
i
0
0
1

0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
n
0
2

0.000000
0.000000
n. oooooo
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000194
0.000000
0.000000
0.000088

0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0,000204
Mayflies
No. Wt.

0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
2
0
1
4
1

0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0

0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000798
0.000000
0.000560
0.001520
0.000629

0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000348
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
o.oooono
0.000000
Chaoborus L.
No. Wt.

6
3
3
9
8
1
14
11
18
29
40
2
20
10
4
10
14
18
1
8
0
0
0
0
0

11
11
9
11
3
2
15
16
3
8
15
5
9
12
5
13
5
4
1

0.002399
0.002742
0.002443
0.007949
0.009722
0.000418
0.010680
0.007672
0.011638
0.017525
0.024427
0.002856
0.015112
0.007879
0.003355
0.006468
0.006407
0.011059
0.000316
0.005127
0.000000
0.000000
0.000000
0.000000
0.000000

0.007579
0.009043
0.008026
0.008793
0.003142
0.001055
0.010921
0.012084
0.002552
0.006492
0.013125
0.004211
0.009479
0.011695
0.003100
0.009874
0.003972
0.002097
0.000613
Chaoborus P.
No. Wt.

2
1
0
1
0
0
1
1
0
2
10
0
6
3
2
2
4
11
1
.3
1
0
0
0
0

2
2
0
1
0
0
0
0
0
0
0
1
1
1
0
1
0
n
i

0.013074
,0.001037
0.000000
0.000945
0.000000
0.000000
0.000874
0.001240
0.000000
0.002777
0.009640
0.000000
0.010270
0.005007
0.002676
0.003642
0.012160
0.013505
0.000976
0.002355
0.000730
0.000000
0.000000
0.000000
0.000000

0.002270
0.002203
0.000000
0.000653
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.001246
0.000969
0.001175
0.000000
0.000845
0.000000
0.000000
0.000709
Heleidae
No. Wt.

0
0
0
0
0
0
0
0
0
0
0
0
1
0
1
2
0
2
1
2
0
0
0
0
0

0
0
0
0
0
0
0
0
0
0
0
0
0
0
2
1
2
10
0

0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.001182
0.000000
0.000912
0.000342
0.000000
0.002971
0.000653
0.002524
0.000000
0.000000
0.000000
0.000000
0.000000

0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000558
0.000214
0.000608
0.003064
0.000000

-------
Table A-S  (Tont inueil ]
Sample
Number
120
121
122
123
124
125
7-25-69
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
8-15-69
301
302
303
304
305
306
307
308
309
310
311
312
Depth
4.3
3.7
2.3
2.9
1.4
0.7

17.8
16.5
16.5
16.0
15.3
14.2
13.7
13.2
12.8
12.2
10.5
8.7
9.6
7.8
8.5
7.3
6.9
6.0
5.0
4.6
3.2
2.7
1.8
1.4
0.7

17.8
16.9
16.9
16.5
16.2
14.6
13.7
13.2
12.6
12.0
10.1
9.6
Oligochaeta
No. Kt.
0 0.000000
o n.nooono
0 0.000000
0 0.000000
o o.oooooo
o o.oooono

o o.oooooo
0 0.000000
o o.onoooo
0 0.000000
0 0.000000
o o.oooooo
n n. oooooo
o o.onoooo
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0'. OOOOOO
0 0.000000

0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
o o.nooooo
0 0.000000
o o.ooooon
o o.onooon
0 0.000000
0 0.000000
0 0.000000
Cbironomid I..
So. Kt.
i:
20
1
3(1
89
90

n
0
0
c
0
n
n
n
0
0
2
->
0
1
n
4
5
13

17
4
23
6
39
39

0
n
0
0
0
0
0
0
s
r)
n
n
n. 004593
0. 004953
0. 000626
0.003803
0. 010616
0. 008971

0.000000
o.ooooon
o.oooooo
0.000000
o.oooooo
0.000000
0.000000
o.oooooo
n. oooooo
0.000000
0.000574
0.00676S
0.000000
0.000030
0.000000
0.0037RS
0.001082
0.00236T
0.002(129
0.005782
0.000218
0.001429
0.000253
0.002314
0.003501

1.000000
O.OOOOnn
O.OOOOOn
1.000000
n.oononr
o.oooooo
0.000000
o.oonoon
0.001162
0.001494
0.000000
n. oooooo
Chironomid P.
No. Wt.
0
1
0
1
1
1

0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
1
0
0
0
3
4

0
0
0
n
0
0
0
0
0
0
0
0
0.000000
0.000283
0.000000
0.000123
0.000060
0.000190

0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000073
0.000000
0.000066
0.000000
0.000000
0.000000
0.000267
0.000276

0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
Mayflies
No. Wt.
0
0
0
1
4
19

0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
43

0
0
0
0
0
0
0
0
0
0
0
0
0.000000
0.000000
0.000000
0.000791
0.002167
0.013419

0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
o.ooonoo
0.000000
0.000000
0.000000
0.000000
0.000000
0.000051
0.002655
0.017489

0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
Chaoborus L.
No. Wt.
1
0
0
0
0
0

0
0
0
0
0
0
0
0
0
0
0
1
2
1
0
1
2
0
0
0
0
0
0
0
0

4
2
2
1
3
1
9
6
7
4
13
6
0.000365
0.000000
0.000000
0.000000
0.000000
0.000000

0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000767
0.002268
0.000436
0.000000
0.000762
0.000754
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000

0.003367
0.000769
0.001700
0.001494
0.003742
0.000940
0.009658
0.006078
0.007245
0(003729
0.015894
0.005475
Chaoborus P.
No. Wt.
1
0
0
0
0
0

0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0

1
0
0
0
0
5
2
0
1
1
3
0
0.000570
0.000000
0.000000
0.000000
0.000000
0.000000

0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000885
9. OOOOOO
0.000000
0.000000
0,000000
0.000000
0.000000
0.000000
0.000000
0.000000

0.001791
0.000000
0.000000
0.000000
0.000000
0.006079
0.002257
0.000000
0.001325
0.001514
0.004065
0.000000
Heleidae
No. Wt.
0
0
0
0
0
0

0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
3
0
0
0
1
1
0

0
0
0
0
0
0
0
0
0
0
0
0
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000

0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
o.eooooo
0.000000
0.000000
0.000117
0.000000
0.002063
0.000000
0.000000
0.000000
0.000039
0.000120
0.000000

0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000

-------
Table A-5 (Continued)
Sample
Number
313
314
315
316
317
318
319
320
321
322
323
324
325
9-6-69
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
6-12-70
1001
1002
1003
1004
1005
Depth
8.7
8.3
7.3
7.3
6.9
6.0
5.0
4.8
3.2
2.S
1.8
0.9
0.5

17.8
17.4
17.1
16.5
15.8
14.6
12.8
13.7
12.0
11.4
10.5
7.8
8.7
9.4
8.7
6.9
7.3
5.0
6.0
4.3
3.4
2.1
1.8
0.9
0.5

0.0
0.0
17.4
16.5
16. n
Oligochaeta
No . Wt .
0 0.000000
0 0.000000
1 0.000442
0 0.000000
n o.oooooo
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000

0 0.000000
0 O.OCOOOO
0 0.000000
0 0.000000
o o.nooooo
ft 0.000000
0 0. 000(100
0 0.000000
0 0.000000
o o.oonooo
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
1 0.001023
0 0.000000
0 0.000000
n o.oooooo
P 0.000000
o o.ooonoo
0 0.000000
0 0.000000
n n. oooooo
n o.oooooo

0 0.000000
0 0.000000
0 0.000000
0 0.000000
o o.noonoo
Chironomid L.
No. Wt.
2
28
30
5
4
0
21
26
11
14
4
63
64

0
0
0
0
0
0
2
0
8
7
3
17
5
8
6
48
34
18
1
20
15
16
9
152
29

0
0
0
0
0
0.001487
0.018956
0.012203
0.001011
0.000812
0.000000
0.008194
0.007209
0.00'2274
0.001255
0.000538
0.005374
0.006092

0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000574
0.000000
0.006546
0.001930
0.000613
0.011584
0.003208
0.006087
0.002373
0.004146
0.005071
0.002413
0.000147
0.001984
0.001502
0.001885
0.000913
0.008952
0.002256

0.000000
0.000000
0.000000
0.000000
0.000000
Chironomid P.
No. Wt.
0
0
1
0
0
0
2
3
0
0
0
0-
2

0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
3
0
0
1
2
0

0
0
0
0
0
0.000000
0.000000
0.000192
0.000000
0.000000
0.000000
0.004295
0.001834
0.000000
0.000000
0.000000
0.000000
0.000182

0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.001935
0.000000
0.000000
0.000000
0.010601
0.000000
0.000000
o.ooonn
0.018306
0.000000

0.000000
o.nooooo
o.oonooo
o.oooooo
o.ooonoo
Mavflies
No. Wt.
0
0
0
0
0
0
0
0
0
1
2
9
3

0
n
0
0
0
0
0
0
n
0
0
0
0
0
0
0
0
0
n
1
4
n
0
17
0

0
0
n
0
0
o.oooono
0.000000
o.oooopo
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000103
0.000505
0.001182
0.001010

0.000000
0.000000
0.000000
0.000000
P. OOOOOO
0.000000
0.000000
n. oooooo
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
o.oooooo
n. "oooooo
n. 000176
n. 000405
0.000000
n. oooooo
0.001671
0.000000

o.onnooo
0.000000
o.ooonoo
0.000000
o.ooonon
Chaohorus L.
No. Wt.
3
1
1
4
6
0
1
0
0
0
0
0
0

1
3
3
1
2
3
3
4
1
6
20
2
2
14
1
0
0
0
0
n
n
p
p
n
0

0
0
3
2
1
0. 002953
0.000479
0. 000594
0.001634
0.003496
0.000000
0.000746
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000

0.000923
0.002929
0.001823
0.000718
0.000844
0.001230
0.002643
0.002263
0.000903
0.006074
0.013576
0.000752
0.001923
0.007097
0.000687
o.oonooo
0.000000
0.000000
o.noooon
0.000000
0.000000
O.OOOhOO
0.000000
0.000000
0.000000

n. oooooo
n.nnonno
n. 001574
0.001630
n. 001161
Chaoborus P.
No. Wt.
3
3
1
1
2
0
0
0
0
0
0
0
0

0
0
2
0
0
1
0
0
0
0
0
0
0
1
1
0
n
0
0
0
0
0
0
n
n

n
n
0
0
0
0.003871
0.004301
0.002051
0.002074
0.002404
0.000000
0.000000
0.000000
0.000000
0.000000
o.nooooo
0.000000
0.000000

0.000000
0.000000
0.001660
0.000000
0.000000
0.003538
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.001228
0.001728
0.000000
0.000000
n. oooooo
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
O.ROOOOO

o.ooooon
n.onnooo
0.000000
0.000000
o.oo.nono
Heleidae
No . Kt .
0
2
0
31
0
0
6
7
2
0
0
1
0

0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
2
5
0
7
0
0
2
0
0
0

0
p
0
0
n
0.000000
0.000442
0.000000
0.018306
0.000000
0.000000
0.005091
0.007542
0.000165
0.000000
0.000000
0.000082
0.000000

0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000316
0.000000
0.000000
0.000608
0.001536
o.ooooon
0.005464
O.OOQOOO
0.000000
0.000228
0.000000
o.oooooo
0.000000

0.000000
0.000000
0.000000
P. OOOOOO
o.ooooon

-------
Tnble A-S (Continued)
Sample
Number
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
7-3-70
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
Depth
0.0
13.2
0.0
0.0
12.0
11.0
10.5
10.1
8.7
7.8
7.3
6.9
5.3
4.6
3.9
3.2
1.8
1.4
0.5
0.3

17.8
17.4
16.7
16.0
15.8
14.6
13.7
13.5
12.6
12.0
11.0
10.3
9.6
8.0
8.5
7.1
6.9
4.6
5.0
4.3
3.2
2.5
1.6
0.9
0.3
Oligochaeta
No. Wt.
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
JO 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000

0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
4 0.001083
88 0.018548
7 0.001453
2 0.000124
0 0.000000
0 0.000000
4 0.004159
2 0.003062
0 0.000000
0 0.000000
2 0.000259
0 0.000000
n o.oooooo
0 0.000000
1 0.000160
Chironomid L.
No. Kt.
0
0
0
0
4
4
6
11
3
6
22
3
48
52
24
15
12
9
10
5

1
0
0
0
0
0
1
0
0
1
7
28
1
11
2
12
13
36
20
42
10
Q
53
131
260
0.000000
0.000000
0.000000
0.000000
0.000457
0.001170
0.005489
0.006766
0.001525
0.003202
0.011157
0.001426
0.028546
0.016277
0.003680
0.002233
0.002606
0.001168
0.000553
0.000328

0.000193
0.000000
0.000000
0.000000
0.000000
0.000000
0.000606
0.000000
0.000000
0.000564
0.004382
0.012183
0.000214
0.01063!)
0.000999
0.008927
0.007622
0.039185
0.017539
0.008507
0.003254
0.000729
0.002923
0.006770
0.012253
Chironomid P.
No. Wt.
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0

0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
1
0
0
0
0
0
3
5
o.oooono
o.onoooo
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.001078
0.000000
0.000000
0.000000
0.000000
0.000000

0.000000
0.000000
o.oooono
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.001110
0.000000
0.000000
0.000000
0.001110
0.000000
0.000000
0.000000
0.000000
0.000000
0.000150
0.000378
M
No.
n
0
0
0
0
0
0
0
0
]
0
0
1
1
3
1
19
5
6
11

0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
2
2
6
J
13
avflies
Wt.
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000050
0.000000
0.000000
0.000066
0.000275
n. 000075
0.000257
0.005706
0.000942
0.001575
0.001661

0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000666
0.000551
0.002970
0.002788
0.003023
Chaoborus L.
No. Wt.
n
i
0
0
2
1
4
7
9
3
3
13
0
0
0
0
0
0
6
0

0
2
3
3
5
0
1
1
4
0
1
4
8
4
3
5
3
6
8
6
0
0
0
0
0
0.000000
0.000535
0.000000
0.000000
0.002212
0.002586
0.002576
0.006233
0.006715
0.002793
0.001824
0.001685
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.004714
0.000000

0.000000
0.000986
0.002013
0.000163
0.000234
0.000000
0.000043
0.000132
0.001263
0.000000
0.000136
0.000524
0.004832
0.001427
0.000113
0.000788
0.000343
0.001083
0.000787
0.001461
0.000000
0.000000
0.000000
0.000000
n. 000000
Chaoborus P.
No. Wt.
0
0
0
0
0
1
0
1
0
1
2
1
2
0
0
0
0
0
1
0

0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
4
1
6
1
9
1
0
0
0
0
0.000000
0.000000
0.000000
0.000000
0.000000
0.000852
0.000000
0.001118
0.000000
0.000774
0.001939
0.000928
0.001822
0.000000
0.000000
0.000000
0.000000
0.000000
0.000559
0.000000

0.000000
0,000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000810
0.000000
0.000000
0.000000
0.000000
0.003438
0.000819
0.004837
0.000701
0.005280
0.000388
0.000000
0.000000
0.000000
0.000000
Heleidae
No . Kt .
0
0
0
0
0
0
0
1
1
0
9
0
2
0
1
2
1
0
0
0

0
0
0
0
0
0
0
0
0
0
0
4
0
4
0
0
0
1
1
2
0
0
0
0
0
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000222
0.000062
0.000000
0.001004
0.000000
0.000388
0.000000
0.000037
0.003432
0.000215
0.000000
0.000000
0.000000

0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000341
0.000000
0.001010
0.000000
0.000000
0.000000
0.000267
0.000359
0.000617
0.000000
0.000000
0.000000
0.000000
0.000000

-------
Table A-5 (Continucdl
Sample
Number
7-24-70
3001
30(12
3003
3004
3005
3006
3007
3008
3009
3010
3011
3012
3013
3014
3015
3016
3017
3018
3019
3020
3021
3022
3023
3024
3025
8-14-70
4001
4002
4003
4004
4005
4006
4007
4008
4009
4010
4011
4012
4013
4014
4P15
40J6
4017
4018
4019
Depth

18.1
17.6
16.7
16.7
16.2
14.6
14.2
12.3
11.4
12.6
11.0
10.5
9.6
7.8
8.5
7.3
6.7
5.0
5.5
4.1
3.4
1.8
2.5
0.9
0.3

18.1
17.4
16.9
16.1
15.1
14.6
13.7
12.6
11.7
i:.o
11. n
9.6
8.7
10.1
S.3
" . 3
6.9
5.3
4.6
OliRnchaeta
No. Kt.

0
0
0
0
0
0
0
6
2
0
2
1
7
0
0
1
2
1
0
0
1
0
0
0
0

0
0
0
0
n
0
c
11
n
7
5
2
4
1
n
1
p
r
n

0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.001384
0.000337
0.. 000000
0/000457
0.001051
0.004039
0.000000
0.000000
0.000117
0.002728
0.000240
0.000000
0.000000
0.000074
0.000000
0.000000
0.000000
0.000000

r. oooooo
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.002900
o.oooono
0.002178
0.003687
0.000776
0. 003144
0.000141
o.oonooo
n. 000934
O.OPPPOO
O.noooon
o.onooon
Chironomid 1, .
No. Wt.

0
1
0
0
0
0
0
4
6
0
10
11
0
~l
10
45
31
1
15
45
172
17
73
78
147

0
1
0
0
5
1
1
M
5
8
^
27
4
10
13
C4
SR
31
60

0.000000
0.000113
0.000000
0.000000
0.000000
0.000000
0.000000
n. 000885
0.005058
0.000000
0.004102
0.005610
0.000000
0.000931
0.004292
0.011427
0.014501
0.001617
0.001494
0.004718
0.018816
0.000701
0.008740
n. 007059
0.012557

n. OOOOOO
0.000159
0.000000
0.000000
0.000993
0.000494
0.000192
0.023224
0.004328
0.010667
0.000587
n. 010915
n. 002437
0.005115
o.oioooi
n. 003236
n. 003820
0.006249
n . 0 0 4 4 3 1
Chironomid P.
N'o. >Vt.

0
0
0
0
0
0
0
1
0
0
-)
0
0
0
0
2
0
0
0
4
3
0
10
4
7

0
0
0
0
1
0
0
1
1
3
0
3
0
2
n
2
0
0
i

0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000107
0.000000
0.000000
0.000306
0.000000
0.000000
0.000000
0.000000
0.000347
0.000000
0.000000
0.000000
0.000860
0.000296
0.000000
0.001464
0.000855
0.000509

0.000000
0.000000
0.000000
0.000000
0.002072
0.000000
0.000000
0.000330
n.oooson
0.006105
0.000000
0.001004
0.000000
0.000973
0.000000
0.000246
0.000000
0.000000
0.000159
Mavf lies
No. ' Wt.

0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
2
.2
7

0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0

0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000074
n. oooooo
0.000563
0.000312
0.001542

0.000000
0.000000
o.oooeoo
0.000000
0.000000
0.000000
0.000000
0.000000
P. OOOOOO
0.000000
0.000000
0.000162
0.000000
P.OPOOOO
0.000000
O.OOPOOO
0.000000
0.000000
O.OPOOOO
Chaobnrus L.
No. Wt.

18
15
31
21
45
14
34
43
25
21
100
46
16
68
48
33
31
25
66
4
6
1
0
0
0

26
15
31
45
11
35
26
10
7
6
21
8
13
16
11
5
12
3
2

0.002534
0.002945
0.007430
0.002454
0.006981
0.002142
0.007453
0.007448
0.004476
0.004675
0.016980
0.008715
0.004008
0.011200
0.010063
0.007372
0.010005
0.005735
0.018560
0.000440
0.001905
0.000530
0.000000
0.000000
0.000000

0.003743
0.004067
0.005504
0.010303
0.002567
0.007870
0.005705
0.002281
0.002226
0.002014
0.004824
0.001388
0.002659
0.005156
0.002385
0.000915
0.001997
0.000799
0.000303
Chaohorus P.
No. Wt.

1
0
1
0
0
0
0
0
1
1
2
0
1
0
1
0
1
2
4
2
1
0
0
0
0

2
1
1
0
1
5
0
0
2
0
2
3
3
0
2
0
3
1
P

0.000960
0.000000
0.000614
0.000000
0.000000
0.000000
0.000000
0.000000
0.000513
0.000645
0.001046
0.000000
0.001062
0.000000
0.000515
0.000000
0.000607
0.001276
0.002521
0.001779
0.000736
0.000000
0.000000
0.000000
0.000000

0.001387
0.0001D9
0.000453
0.000000
0.000119
0.002344
0.000000
0.000000
0.001099
0.000000
0.001167
0.001304
0.001169
0.000000
0.001001
o.nooooo
0.001381
0.000580
0.000000
Heleidae
No. Kt.

0
0
0
0
0
0
0
0
0
0
0
0
0
0
2
7
2
0
3
0
0
0
0
1
1

0
0
0
0
0
0
0
0
0
0
0
0
0
2
0
0
8
0
1

0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
O.OOOtJOO
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000269
0.000406
0.000586
0.000000
0.001106
0.000000
0.000000
0.000000
0.000000
0.000161
0.000113

0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000322
0.000000
0.000000
0.000832
0.000000
0.000615

-------
Table A-5 (Continued)
Sample
Number
4020
4021
4022
4023
4024
402S
9-4-70
S001
5002
S003
5004
5005
5006
5007
5008
5009
5010
5011
5012
5013
5014
5015
5016
5017
5018
5019
5020
5021
S022
5023
5024
5025
Depth
3.9
3.0
2.1
1.6
1.2
0.3

18.3
17.8
16.9
16.2
15.1
13.9
12.6
11.4
12.0
12.8
11.0
9.8
9.2
8.0
8.3
6.9
5.7
4.6
4.8
3.7
3.0
2.1
1.4
0.7
0.3
Oligochaeta
No. Wt.
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000

0 0.000000
0 0.000000
0 0.000000
0 0.000000
1 0.000198
0 0.000000
74 0.055148
3 0.000632
4 0.002107
1 0.010180
3 0.001317
3 0.003327
8 0.006156
0 0.000000
6 0.002028
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000
Chirononid L.
No. Wt.
136
6
18
8
42
59

2
0
4
3
1
1
10
3
6
5
3
2
14
13
9
1
17
38
12
40
41
24
9
IS
63
0.011041
0. 000450
0.000760
0.000886
0.002043
0.002828

0.003847
0.000000
0.002151
0.001267
0.002180
0.000089
0.007225
0.001256
0.005148
0.002651
0,000768
0.001666
0.004711
0.002222
0.003552
0.000037
0.000786
0.001862
0.000332
0.003022
0.002620
0.001584
0.000151
0.010543
0.003172
Chironomid P.
No . Wt .
8
1
2
0
0
1

0
0
0
1
0
0
0
0
0
1
0
0
0
1
0
1
0
0
0
1
0
0
0
1
4
0.000747
0.000086
0.000176
0.000000
0.000000
0.000119

0.000000
0.000000
0.000000
0.000249
0.000000
0.000000
0.000000
0.000000
0.000000
0.000932
0.000000
0.000000
0.000000
0.000046
0.000000
0.000030
0.000000
0.000000
0.000000
0.000150
0.000000
0.000000
0.000000
0.000049
0.000179
Mavflies
No . Wt .
0
0
2
0
1
9

0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
1
2
0
(1
3
0.000000
0.000000
0.000151
0.000000
0.000150
0.000683

0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000-00
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000084
n. oooooo
0.000000
0.000118
0.000090
0.000000
n.onoooo
0.000292
Chaoborus L.
No . Wt .
4
0
0
0
0
0

10
25
90
16
12
0
2
6
3
6
4
0
3
3
6
5
0
0
1
0
0
n
0
i
0
0.000321
0.000000
0.000000
0.000000
0.000000
0.000000

0.001336
0.005454
0.029378
0.004247
0.002245
0.000000
0.000319
0.001675
0.000418
0.000927
0.000900
0.000000
0.000473
0.000424
0.001338
0.000947
0.000000
0.000000
0.000155
0.0t)0000
0.000000
0.000000
0.000000
0.000307
0.000000
Chaoborus P.
No. Wt.
0
0
0
0
0
0

0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000

0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000366
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
Heleidae
No. Wt.
4
0
0
0
0
1

0
0
0
0
0
0
0
0
0
0
0
0
1
10
14
2
2
1
2
0
0
2
0
0
1
0.002745
0.000000
0.000000
0.000000
0.000000
0.000025

0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.001240
0.001430
0.002777
0.000399
0.000601
0.000831
0.000743
0.000000
0.000000
0.000524
0.000000
0.000000
0.000020

-------
Table A-6.
Hemlock Lake zoobenthos collected during 1969 and
1970.  The less abundant organisms are listed in
this table.  To obtain a total for a given sample,
add the values for each sample (check each organism)
to the values for that sample piven in Table A-5.
Wet weights are shown.
Sample
Number
Amphipods
224
325
424
1024
1025
2004
Dragonf lies
22
23
24
124
125
223
225
321
324
325
424
425
1022
2024
2025
3025
4022
4025
5020
5025
Damselflies
118
424
425
1025
4002
5025
Depth
On)

1.37
0.46
0.92
0.46
0.23
16.02

2.29
1.60
0.92
1.37
0.69
1.83
0.69
3.20
0.92
0.46
0.92
0.46
1.83
0.92
0.23
0.23
2.06
0.23
3.66
0.23

5.72
0.92
0.46
0.23
17.40
0.23
Date

7-25-69
8-15-69
9- 6-69
6-12-70
6-12-70
7- 3-70

6-13-69
6-13-69
6-13-69
7- 4-69
7- 4-69
7-25-69
7-25-69
8-15-69
8-15-69
8-15-69
9- 6-69
9- 6-69
6-12-70
7- 3-70
7- 3-70
7-24-70
8-14-70
8-14-70
9- 4-70
9- 4-70

7- 4-69
9- 6-69
9- 6-69
6-12-70
8-14-70
9- 4-70
Number

1
1
1
1
1
1

1
2
1
1
1
1
5
1
1
1
1
1
2
1
2
1
2
3
1
2

4
9
2
1
1
1
Weight
CRN)

0.000148
0.000058
0.000229
0.000012
0.000454
0.000220

0.043686
0.076687
0.008999
0.020766
0.057300
0.169671
0.055579
0.225487
0.000118
0.152927
0.000233
0.022SS7
0.151958
0.081826
0.005540
0.049046
0.003524
0.079121
0.202513
0.400948

0.004010
0.004751
0.001369
0.003350
0.000409
0.000038

-------
Table A-6 (Continued)
Sample
Number
Trichoptera
115
5022
Tabanid
5022
Leeches
323
5017
Depth
(m)
7
2
2
1
5
.78
.06
.06
.83
.72
7 _
9-
9-
8-
9-
Date
4-
4-
4-
15-
4-
69
70
70
69
70
Number
28
1
1
1
1
Weight
(Rm)
0.
0.
0.
0.
0.
016076
000082
000243
546294
029607

-------
Section Pour Lake inobenthos collected  with  an  Fkman  dredce during 1969 and 1970.
Numbers and weinhts for  the seven most  abundant taxa  are shown in this table for
each sample.  125 samples were  collected  each  summer.   The less abundant taxa are
lisli:.' iii T.ible A-S.  To verifv thr  toL.'l  irmnisms  for a oivcn s.innlo, consult
both tables.  Depth is in neters and  wciclit  is  in  °rams.
Sample
Number
6-15-69
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
7-5-69
151
15?
151
ZS4
155
156
157
158
159
160
161
162
163
164
165
166
167
168
Depth

18.7
17.8
14.6
16.9
16.0
13.0
13.7
12.3
12.8
11.7
11.7
8.9
7.8
7.5
9.2
6.4
3.7
4.6
6.2
5.0
3.2
2.5
2.1
0.9
1.4

19.7
19.2
18.3
17.4
16.0
14.2
13.0
12.8
12.3
11.2
10.1
8.3
8.3
9.2
8.5
5.0
5.0
4.8
Oli^ochaeta
No. Wt,

12S 0.
263 0.
216 0.
112 0.
20 0.
35 0.
29 0.
6 0.
34 0.
3 0.
2 0.
1 0.
6 0.
16 0.
0 0.
9 0.
4 0.
18 0.
4 0.
10 0.
1 0.
11 0.
0 0.
11 0.
4 0.

398 0.
358 0.
156 0.
32 0.
45 0.
6 0.
70 0.
30 0.
18 0.
3 0.
0 0.
2 0.
7 0.
2 0.
20 0.
2 0.
5 0.
4 n.

037484
129977
113488
036015
006093
008992
022968
001411
004647
000785
000589
000207
001358
016207
000000
001384
000680
003336
000558
002172
OOOS41
001344
000000
003405
001753

394441
469270
055653
028483
010068
,- --750
017975
006122
005887
000272
000000
000782
005178
001123
018710
000^25
002251
001672
Chironomid L.
No . Wt .

0
2
11
40
53
8
1
5
4
1
8
22
51
258
53
84
89
164
72
225
142
112
184
205
200

0
0
54
59
77
13
39
32
17
9
183
78
180
119
192
132
57
155

0.000000
0.001303
0.001349
0.030825
0.007247
0.007369
0.000102
0.001807
0.000495
0.000510
0.001563
0.002000
0.005433
0.030392
0.006443
0.006158
0.004919
0.008330
0.005483
0.012618
0.000500
0.006432
0.010559
0.012364
0.042538

0.000000
0.000000
0.134584
0.092093
0.012093
0.001*33
0.014097
0.012732
0.007923
0.003595
0.056019
0.015103
0.038387
0.007244
0.027144
0.017560
0.012130
0.012R60
Chironomid P.
No. Wt.

0
0
1
1
1
1
0
1
0
0
0
1
1
53
1
23
4
7
9
8
3
2
3
11
16

0
0
0
1
2
0
0
0
i
0
4
1
0
4
3
3
3
7

0.000000
0.000000
0.000062
0.000286
0.000124
0.000091
0.000000
0.000146
0.000000
o.oooooo
0.000000
0.000163
0.000099
0.006324
0.000088
0.002117
0.000139
0.000670
0.001031
0.000728
0.000597
0.000214
0.000112
0.000540
0.002381

0.000000
0.000000
0.000000
0.000920
0.000389
n. Trioooo
0.000000
0.000000
0.000406
o.ooaooo
0.001597
0.003525
o.oononn
o.ooii,,:
0.00162S
0.000601
0.00052R
0.001003
Amphinoda
No. ' Wt.

1
0
2
0

-------
Table A-7 (Continued)
Sample
Number
169
170
171
172
173
174
T - :
7-25-69
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
8-15-69
351
352
353
354
355
356
357
358
359
360
361
362
Depth
4.1
3.7
2.7
2.3
3.0
1.2
•".7

19.7
18.7
19.4
18.5
17.4
13.2
13.7
12.0
11.0
11.4
10.3
9.2
8.3
9.2
8.3
7.3
5.5
4.8
5.0
4.3
3.2
3.7
2.3
1.2
0.5

19.7
19.7
19.4
18.5
16.5
14.2
11.0
11.0
11.2
11.0
S.7
9.2
01 igochaeta
No . Wt .
1 0.000604
13 0.005265
3 0.001496
3 0.002026
3 0.001165
22 O.OOS603
2 1.000874

1123 0.947312
480 0.259524
438 0.587135
286 0.154376
16 0.002096
46 0.005866
33 0.020125
17 0.009033
1 0.000531
1 0.000570
0 0.000000
3 0.002475
17 0.006339
0 0.000000
1 0.000080
0 0.000000
0 0.000000
0 0.000000
o o.ooooon
0 0.000000
0 0.000000
0 0.000000
1 0.000415
0 0.000000
0 fl. 000000

360 0.346086
280 0.341950
146 0.078877
112 0.085044
493 0.495488
74 0.013434
0 0.000000
0 0.000000
o o.ooooon
o o.onnon.i
5 0.003183
o o.ooonon
Chironomid L.
No. Wt,
153
109
43
58
81
228
93

3
2
14
375
109
30
381
185
150
324
52
150
33
24
46
43
13
13
4
8
110
26
7
21
19

1
20
62
128
147
349
18-1
52
25
15
30
72
0.018628
0.025384
0.003544
0.010751
0.007529
0.019800
0.006263

0.003898
0.002852
0.067439
0.049310
0.011697
0.001467
0.051125
0.040806
0.028433
0.026003
0.007057
0.026280
0.008026
0.006246
0.009995
0.006635
0.003826
0.001632
0.000855
0.003904
0.012074
0.006374
0.001342
0.003564
0.002602

0.001577
0.007445
0.148463
0.160487
0.087398
0.087005
0.050390
0.015136
0.005971
0.003621
0.009534
0.008541
Chironomid P.
No. Wt.
12
14
4
4
4
18
4

0
1
0
1
0
1
0
0
0
1
0
0
1
0
0
5
3
5
1
1
2
1
0
1
0

0
0
0
0
0
1
4
1
0
0
0
3
0.002326
0.002677
0.000595
0.000586
0.000629
0.001913
n.0006"-!

0.000000
0.000215
0.000000
0.000481
0.000000
0.000322
0.000000
0.000000
0.000000
0.000120
0.000000
0.000000
0.000160
0.000000
0.000000
0.000868
0.000441
0.000431
0.000091
0.000591
0.000471
0.000114
0.000000
0.000099
0.000000

o.nooooo
0.000000
0.000000
0.000000
0.000000
0.001065
0.000941
0.00018(1
o.ooooon
0.000000
0.000000
0.000412
Anphipoda
No. Wt.
0
11
1
1
0
11
-1

0
0
0
0
0
0
2
0
0
0
0
0
1
1
0
5
1
1
3
0
8
5
0
0
9

0
1
0
0
0
15
1
0
0
0
0
0
0.000000
0.007166
0.001035
0.001208
0.000000
0.009274
0 . 0 0 2 1 8 5

0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.002589
0.000000
0.000000
0.000000
0.000000
0.000000
0.000836
0.000291
0.000000
0.004608
0.016308
0.000672
0.001130
0.000000
0.001677
0.002387
0.000000
0.000000
0.001688

0.000000
0.000407
0.000000
0.000000
0.000000
0.019278
0.000520
0.000000
0.000000
0.000000
0.000000
0.000000
Mavflies
No. Wt.
1
2
2
0
1
1
1

0
0
0
0
0
1
0
0
0
0
0
1
0
0
0
0
2
3
1
0
1
2
0
1
0

0
0
0
0
0
0
0
0
0
0
n
0
0.000204
0.003410
0.004788
0.000000
0.000340
r. "00655
0.000803

0.000000
0.000000
0.000000
0.000000
0.000000
0.000159
0.000000
0.000000
0.000000
0.000000
0.000000
0.000252
0.000000
0.000000
0.000000
0.000000
0.000709
0.002148
0.000173
0.000000
0.001265
0.001160
0.000000
0.000921
0.000000

0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
llelcidae
No. Wt.
7
1
1
0
0
0
7

0
0
0
0
0
1
0
4
1
2
0
1
0
0
3
0
0
0
0
0
0
2
0
1
0

0
0
1
0
0
0
0
0
0
0
1
0
0.002687
0.000476
0.000602
0.000000
0.000000
0.000000
o.ooime

0.000000
0.000000
0.000000
0.000000
0.000000
0.000010
0.000000
0.001830
0.000214
0.000149
0.000000
0.000077
0.000000
0.000000
0.001206
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000447
0.000000
0.000165
0.000000

0.000000
0.000000
0.000424
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000114
0.000000
Trichoptera
No. Wt.
1
0
0
0
0
0
0

0
0
0
0
0
0
0
0
0
0
0
1
1
0
0
0
0
0
0
0
0
0
0
0
0

0
0
0
0
0
0
1
1
0
0
0
0
0.000446
0.000000
0.000000
0.000000
0.000000
o.oooono
0.000000

0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.004088
0.005864
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000

0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.001711
0.000637
0.000000
0.000000
0.000000
0.000000

-------
TaMc  V7 fCniitiiiucUl
Sample
Number
363
364
365
366
367
368
369
370
371
372
373
374
375
9-6-69
475
474
473
472
471
470
469
468
467
466
465
464
463
462
461
460
459
458
457
456
455
454
453
452
451
6-12-70
1051
1052
1053
1054
1055
Depth
8.3
7.5
9.2
6.9
6.0
5.3
4.6
3.9
3.2
2.5
1.2
1.6
0.5

20.1
19.9
18.1
19.7
IB. 3
11.4
11.4
12.2
12.2
12.2
10.5
10.1
9.2
9.6
8.0
6.4
6.0
4.1
6.0
5.3
2.7
1.4
1.4
2.7
0.3

20.1
20.1
19.4
18.5
16.7
Oliqochaeta
No. Wt.
15 0.007483
2 0.000960
4 0.002165
3 0.000369
0 0.000000
0 0.000000
1 0.000154
1 0.000013
0 0.000000
0 0.000000
0 0.000000
0 0.000000
0 0.000000

512 0.411180
342 0.775198
88 0.025160
219 0.632133
137 0.070083
0 0.000000
4 0.002229
32 0.004395
28 0.014446
34 0.005916
0 0.000000
0 0.000000
1 0.001245
1 0.000552
0 0.000000
1 0.000406
0 0.000000
0 0.000000
o o.onoooo
0 0.000000
1 0.000632
0 0.000000
o o.ooonon
0 0.000000
1 0.002280

433 0. H3696
476 (1.394387
124 O.OS7330
125 O.H37617
195 n. 048141
Chirnnomid L.
No. Wt.
76
30
153
35
13
5
46
77
23
12
38
51
35

0
2
71
57
44
57
16
68
218
29
75
17
19
17
12
12
3
23
14
f.
3
30
6
7
21

1
2
11
5
40
0.016883
0.007216
0.017192
0.005275
0.005913
0.000202
0.003389
0.005515
0.001465
0.000751
0.1)03347
0.002407
0.001618

0.000000
0.003283
0.073864
0.019344
0.053595
0.017942
0.010740
0.021900
0.064953
0.014170
0.032766
0.004612
0.007020
0.003493
0.004235
0.002295
0.000148
0.006938
0.002253
0.002592
(1.000446
0.005540
'•-. 000798
0.003316
O^-'l'S:

0.000235
0.000279
0.018985
0.000480
0.008197
Chironomid P.
No. Wt.
6
3
1
3
0
1
2
7
1
2
4
6
5

0
0
0
0
0
0
0
2
8
0
3
6
2
1
0
0
0
n
0
0
1
i
n
0
2

0
0
n
i
i
0.000394
0.000390
0.000043
0.000456
0.000000
0.000138
0.000279
0.000335
0.000064
0.000086
0.000100
0.000222
0.000526

0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000379
0.001723
0.000000
0.001103
0.001557
0.001755
0.000276
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000561
0.000194
n. oooooo
o.o'ooooo
0.000283

0.000000
n. oooooo
n. oooooo
n. 000144
0.002021
Amphipoda
No. Wt.
1
2
11
1
4
2
1
5
1
1
0
2
37

0
0
0
0
0
0
0
0
0
0
0
1
1
0
3
20
0
8
18
5
1
5
0
3
89

0
0
0
1
0
0.000577
0.001673
0.016288
0.000245
0.001816
0.000952
0.000279
0.000396
0.000093
0.000096
0.000000
0.000241
0.003858

0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000391
0.000173
0.000000
0.000804
0.004973
0.000000
0.003159
0.005197
0.001814
0.000212
0.001433
0.000000
0.000975
0.016186

0.000000
0.000000
0.000000
0.001444
0.000000
Mayflies
No. Wt.
1
0
0
2
0
0
0
0
0
0
0
0
0

0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
2
2
0
0
0
0
0
1

0
0
0
0
0
0.001607
0.000000
0.000000
0.008254
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000

0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000482
0.000000
0.001926
0.000728
0.000000
0.000000
0. OOOOOO
0.000000
o.onorno
0.000106

0.000000
0.000000
0.000000
0.000000
0.000000
Heleidae Trichoptera
No. Wt. No. Wt.
0
0
0
0
1
0
6
1
1
0
8
4
1

0
0
0
1
0
0
0
0
1
1
0
0
0
0
0
0
0
1
0
1
1
2
2
4
0

0
0
0
0
0
0.000000
0.000000
0.000000
0.000000
0.000388
0.000000
0.000649
0.000134
0.000091
0.000000
0.000994
0.000126
0.000076

0.000000
0.000000
0.000000
0.0003S3
0.000000
0.000000
0.000000
0.000000
0.000470
0.000316
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000501
0.000000
0.000145
0.000156
0.000675
0.000376
0.000805
0.000000

0.000000
0.000000
0.000000
0.000000
0.000000
0
0
1
0
0
1
2
1
0
0
0
0
0

0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
0
0
0
0
0

0
0
0
0
0
0.000000
0.000000
0.000037
0.000000
0.000000
0.000289
0.000289
0.000708
0.000000
0.000000
0.000000
0.000000
0.000000

0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000200
0.000227
0.000000
0.000000
0.000000
0.000000
0.000000

0.000000
0.000000
0.000000
0.000000
0.000000

-------
Table A-7 (Continued!
Sample
Number
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
7-3-70
2051
2052
2053
2054
2055
2056
2057
2058
2059
2060
2061
2062
2063
2064
2065
2066
2067
2068
2069
2070
2071
2072
2073
2074
2075
Depth
12.3
12.6
12.0
11.2
11.0
10.5
9.8
8.9
8.0
8.0
6.2
7.3
3.7
4.1
4.6
2.5
1.8
2.7
1.4
0.7

19.7
19.9
19.4
18.5
17.6
12.0
11.9
11.9
11.9
11.7
10.1
8.7
9.6
8.5
7.3
5.7
6.0
4.8
4.1
4.6
3.7
2.7
1.8
0.9
0.3
Oligochaeta
No. Wt.
17 0.003504
17 0.002013
10 0.003760
0 0.000000
4 0.000696
0 0.000000
9 0.001766
0 0.000000
5 0.003505
1 0.001205
1 0.000895
0 0.000000
2 0.000806
0 0.000000
0 0.000000
0 0.000000
9 0.003465
1 0.000938
4 0.000771
1 0.000378

473 0.249827
476 0.363800
205 0.095493
142 0.029062
40 0.012842
11 0.000745
4 0.000672
5 0.000961
2 0.000164
16 0.002047
20 0.003617
11 0.006321
7 0.002160
21 0.001965
6 0.002499
5 0.002070
9 0.002428
0 0.000000
4 0.000880
2 0.000270
24 0.011530
3 0.001-19
3 0.000253
15 0.010221
10 0.004249
Chironomid L.
No. Wt.
2
10
5
17
37
14
11
10
30
48
125
30
27
SI
13
32
188
8
33
78

26
62
42
58
62
7
17
21
1
3
46
78
14
100
39
60
25
6
11
27
9
14
6
40
113
0.000969
0.003160
0.000595
0.002336
0.005432
0.002232
0. 002148
0.002073
0.006840
0.005799
0.007423
0.001600
0.004981
0.006111
0.001755
0.001492
0.011150
0.000257
0.002726
0.005612

0.002114
0.003808
0.005442
0.010911
0.002948
0.000830
0.003580
0.005144
0.000590
0.000199
0.006423
0.009408
0.001321
0.009743
0.005367
0.004884
0.007207
0.000314
0.000563
0.002702
0.0015V:
0.001804
0.000441
0.006086
0.009120
Chironomid P.
No. Wt.
1
0
1
0
0
2
0
3
0
1
6
1
0
1
0
4
11
2
4
4

0
0
0
0
0
0
0
1
0
0
0
2
0
0
3
2
1
0
1
0
0
1
0
0
15
0.000344
0.000000
0.000145
0.000000
0.000000
0.000403
0.000000
0.000374
0.000000
0.000254
0.000924
0.000210
O.OOOPfi'T
n. ''00153
0.000000
0.000334
0.000767
0.000007
0.000713
0.000171

0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000110
0.000000
0.000000
0.000000
0.000150
0.000000
0.000000
0.000370
0.000299
0.000103
o.ooonoo
O.OOOOS4
0.001000
0.000000
0.000038
0.000000
0.000000
0.000760
Amphinoda
No . Wt .
0
1
0
0
0
0
0
4
0
2
16
5
6
9
30
4
4
4
3
13

0
0
0
0
0
0
1
0
0
0
2
2
0
2
3
11
0
4
1
6
0
0
0
3
23
0.000000
0.000407
0.000000
0.000000
0.000000
0.000000
0.000000
0.001489
0.000000
0.000711
0.006418
0.001497
0.002295
0.003694
0.009409
0.001379
0.001945
0.001667
0.001094
0.006169

0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000717
0.000000
0.000000
0.000000
0.000823
0.000738
0.000000
0.000831
0.001214
0.005259
0.000000
0.002023
0.000405
0.002639
O.ooonnn
0.000000
0.000000
0.000930
0.006811
Mayflies
No . Wt .
0
0
0
0
0
0
0
1
0
1
1
0
1
1
3
0
0
1
0
0

0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
5
0
0
0
0
0
0
0
0
1
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000322
0.000000
0.000738
0.000178
0.000000
0.000841
0.000408
0.002254
0.000000
0.000000
0.000253
0.000000
0.000000

0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000068
0.000000
0.006054
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000990
Helcidae
No . Wt .
0
0
0
0
0
0
0
0
1
4
3
0
2
9
6
0
7
1
1
0

0
0
0
0
1
0
1
0
0
0
2
4
0
5
1
2
1
0
0
0
4
0
0
1
1
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000550
0.001591
0.000556
0.000000
O.OOOS2J
0.002798
0.001183
0.000000
0.001382
0.000116
0.000166
0.000000

0.000000
0.000000
0.000000
0.000000
0.000169
0.000000
0.000151
0.000000
0.000000
0.000000
0.000497
0.001607
0.000000
0.001738
0.000328
0.000630
0.000139
0.000000
0.000000
0.000000
0.001189
0.000000
0.000000
0.000159
0.000025
Trichoptera
No. Wt.
0
0
0
0
0
1
0
2
2
0
1
2
n
0
2
1
0
1
0
0

0
0
0
0
0
0
0
0
0
0
0
1
0
1
2
0
1
1
0
2
0
0
0
0
0
0.000000
0.000000
0.000000
0.000000
0.000000
0.000396
0.000000
0.000210
0.001863
0.000000
0.000132
0.000158
0.000000
0.000000
0.001456
0.000321
0.000000
0.000025
0.000000
0.000000

0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.001685
0.000000
0.000676
0.000462
0.000000
0.000238
0.000071
0.000000
0.002582
0.000000
0.000000
0.000000
0.000000
0.000000

-------
Table A-7 (Continued)
Sample
Number
7-14-70
3051
3052
3053
3054
3055
3056
3057
3058
3059
3060
3061
3062
3063
3064
3065
30fh
3067
•3068
3069
3070
3071
3072
3073
3074
3075
8-14-70
/•OS 3
4052
4053
4054
4055
4056
4057
4058
4059
4060
4061
4062
4063
4064
4065
4066
Depth

19.9
19.9
19.7
18.7
18.3
11.2
11.2
11.4
12.0
11.7
9.6
8.9
8.7
7.5
6.4
fi.2
5.5
5.0
5.5
3.7
3.2
2.5
2.5
0.7
0.3

20.1
19.7
19.7
19.0
18.1
11.0
11.4
11.2
12.0
11.0
10.1
8.9
9.6
7.8
8.7
7.3
Oligochaeta
No . Wt .

217 0.
468 0.
108 0.
474 0.
177 0.
10 0.
24 0.
12 0.
2 0.
14 0.
17 0.
12 0.
12 0.
8 0.
17 0.

6 Q!
10 0.
0 0.
2 0.
5 0.
2 0.
1 0.
4 0.
2 0.

178 0.
63 0.
140 0.
537 0.
18 0.
8 0.
32 0.
22 0.
1 0.
6 0.
4 0.
19 0.
5 0.
2 0.
2 n.
1 n.

146485
261606
033697
161224
059882
00145!)
003274
004448
000116
001423
004405
002601
004418
001540
008398
00
-------
Table A-7 (Continued)
Sample
Number
4067
4068
4069
4070
4071
4072
4073
4074
4075
9-4-70
5051
5052
5053
5054
5055
5056
5057
5058
5059
5060
5061
5062
5063
5064
5065
5066
5067
5068
5069
5070
5071
5072
5073
5074
5075
Depth
4.1
6.:
5.0
4.3
3.7
2.5
2.1
1.2
0.3

19.9
20.1
19.7
19.0
18.1
12.6
12.0
11.0
12.3
11.4
10.1
8.5
8.9
7.5
8.7
8.0
6.0
6.9
4.6
4.3
3.2
0.5
2.1
1.4
0.9
Oligochaeta
No . Wt .
3 0.001033
2" C. 004225
4 0.000682
8 0.001202
2 0.000064
3 0.000714
3 0.000726
7 0.001158
S 0.000992

318 0.318906
168 0.181866
186 0.181879
121 0.163979
103 0.121870
3 0.000588
6 0.001020
1 0.000018
5 0.000627
18 0.001878
0 0.000000
10 0.005546
12 0.002072
0 0.000000
6 0.001848
8 0.0012C2
12 0.003628
5 0.001647
4 0.002513
0 0.000000
0 0.000000
0 0.000000
0 0.000000
1 0.001053
21 0.-OOS237
Chironomid L.
No . Wt .
43
3
35
16
30
18
22
57
31

36
S3
57
43
69
3
0
14
18
9
10
52
19
18
6
13
1
9
16
i:
21
13
0
21
42
O.OCS801
0.000701
0.002083
0.000893
0.003245
0.001111
0.902142
0.004818
0.004376

0.013950
0.005296
0.009009
0.017090
0.008579
0.000702
0.000000
0.000707
0 . 0 0 0 6 S 9
0.000648
o . o n 7 ^ : '
0.002848
0.002122
O.nil] 112
0.000605
0.001^11
0.000072
0.000520
0.000043
0.0018?"
0.001947
0.00156'
o.ooopon
0. 00132 c
0.001393
Chironomid P.
No. Wt.
1
0
n
0
0
1
1
3
2

2
1
3
3
0
0
0
2
1
3
n
1
1
0
1
0
0
0
0
1
1
0
0
2
2
o.nooo?0
o.nooonn
o.oooooo
o.nnoooo
0.000000
0.000013
0.000126
0.000139
0.00012R

0.000609
0.000295
0.009677
0.000600
0.000000
0.000000
0.000000
o.onoi] 7
0.000121
0.000178
-.nni'ooo
0.000212
0.000089
0.000000
n. 000115
0.000000
0.000001
0.000000
0.000000
0.000171
n. onoois
0.000000
0.000000
0.000092
o.nnoo69
Amphipoda
No . Wt .

1
g
1
19
0
S
40
1R

0
0
0
0
0
0
0
10
0
27
1
18
15
91
1]
32
9
IP
S
in
14
69
0
14
0
.nni236
0.000037
0.000630
0.000754
0.004432
0.000000
0.002548
0.007304
0.002500

0.000000
o.onooon
o.noooon
0.000000
0.000000
0.000000
o.nnoooo
O.n04700
0.000000
0.025896
0.002827
0.002550
0.025109
0.005985
0.003880
0.006377
0.003883
0.002522
0.00200?
0.002782
n . o n 4 9 8 .1
o . o n s o 3 n
o.ooooon
0.003675
o.onnooo
Mayflies
No . Wt .
2
n
n
0
0
0
i
n
i

0
0
0
0
n
0
0
i
0
4
2
4
2
4
1
0
0
2
1
1
4
0
0
0
0
0.001773
o.onnoon
0.000000
0.000000
0.000000
0.000000
0.000064
0.000000
o.ooooso

0.000000
0.000000
n. oooooo
n. oooooo
0.000000
0.000000
0.000000
n. 000082
0.000000
0.000179
n. 000058
0.000145
0.000076
0.000491
0.000013
0.000000
0.000000
0.000459
0.000091
0.000059
0.000135
0.000000
0.000000
0.000000
0.000000
Heleidae
No . Wt .
1
0
0
0
0
0
0
0
0

0
0
0
1
0
2
0
0
3
1
0
0
0
0
0
0
1
0
2
1
0
1
0
0
0
o.opnrv
0.000000
o.onoooo
0.000000
0.000000
0.000000
o.oooooo
0.000000
0.000000

0.000000
0.000000
0.000000
0.000878
0.000000
0.000208
0.000000
0.000000
0.000026
0.000038
o.ooooon
0.000000
o.oooono
0.000000
0.000000
0.000000
0.000073
0.000000
0.000202
0.003446
o.ooonnn
O.OOOOR5
n.oonooo
0.000000
o.onoooo
Trichoptera
No. Wt.
0
0
0
0
0
1
0
0
0

0
0
0
0
0
0
0
1
0
2
I
1
0
2
3
2
0
0
1
1
1
0
0
1
0
P. OOOOOO
0.000000
0.000000
0.000000
0.000000
0.000222
o.ooooon
0.000000
o.ooooon

0.000000
o.oonooo
0.000000
0.000000
0.000000
0.000000
0.000000
0.000153
0.000000
o.ooisir,
o.-nf-nss
0.000259
0.000000
0.000410
0.006150
0.001937
0.000000
0.000000
0.000739
0.002432
0.000595
0.000000
0.000000
0.001209
o.ooooon

-------
Table A-8.
Section Four Lake zoobenthos collected during 1969
and 1970.  The less abundant organisms are listed
in this table.  To obtain a total for a given sample,
add the values for each sample (check each organism)
to the values for that sample given in Table A-7.
Wet weights are shown.
Sample
Number
OliEOchaetes
(Megadriles)
55
57
58
59
60
158
161
252
256
466
1056
1057
3061
3067
4071
5057
Dragonflies
71
73
170
173
174
369
374
1068
2075
4060
4067
4069
4070
4074
4075
5066
5074
Depth
On)


16.02
13.73
12.36
12.82
11.67
12.82
10.07
18.77
13.28
12.13
12.36
12.59
9.61
5.49
3.66
11.90

3.20
2.06
3.66
2.98
1.14
4.58
1.60
3.66
0.23
10.99
4.12
5.04
4.35
1.14
0.23
8.01
1.37
Date


6-15-69
6-15-69
6-15-69
6-15-69
6-15-69
7- 5-69
7- 5-69
7-25-69
7-25-69
9- 6.-60
6-12-70
6-12-70
7-24-70
7-24-70
8-14-70
9- 4-70

6-15-69
6-15-69
7- 5-69
7- 5-69
7- 5-69
8-15-69
8-15-69
6-12-70
7- 3-70
8-14-70
8-14-70
8-14-70
8-14-70
8-14-70
8-14-70
9- 4-70
9- 4-70
Number


1
11
2
2
4
1
1
1
1
3
2
5
1
1
1
2

2
4
i
i
i
1
1
1
2
1
1
1
1
1
2
2
1
1
Weight
(pm)


0.004690
0.229702
0.020609
0.029082
0.821241
0.053231
0.314032
0.037648
0.018359
0.013296
0.070588
0.091363
0.346751
0.533722
0.209138
0.942075

0.045291
0.00295.5
0.001424
0.011425
0.023480
0.006573
0.004726
0.018994
0.028874
0.000113
0.000409
0.000445
0.000168
0.097749
0.015670
0.001060
0.001092

-------
Table A-8 (Continued)
S amp 1 e
Number
Damself lies
58
63
65
161
163
266
268
363
364
365
368
369
461
457
456
453
1065
1067
1070
2066
2070
3070
3072
4063
4064
4066
4067
4070
4071
5058
5060
5062
5064
5066
5068
165
5069
5070
5072
5074
Depth
(m)

12.36
7.78
9.16
10.07
8.24
7.33
4.81
8.24
7.55
9.16
5.27
4.58
8.01
5.95
5.27
1.37
8.01
7.33
4.58
5.72
4.58
3.66
2.52
9.61
7.78
7.33
4.12
4.35
3.66
10.99
11.45
8.47
7.55
8.01
6.87
8.47
4.58
4.35
0.46
1.37
Date

6-15-69
6-15-69
6-15-69
7- 5-69
7- 5-69
7-25-69
7-25-69
8-15-69
8-15-69
8-15-69
8-15-69
8-15-69
9- 6-69
9- 6-69
9- 6-69
9- 6-69
6-12-70
6-12-70
6-12-70
7- 3-70
7- 3-70
7-24-70
7-24-70
8-14-70
8-14-70
8-14-70
8-14-70
8-14-70
8-14-70
9- 4-70
9- 4-70
9- 4-70
9- 4-70
9- 4-70
9- 4-70
7- 5-69
9- 4-70
9- 4-70
9- 4-70
9- 4-70
Number

1
1
1
2
2
1
1
1
2
">
*_,
1
1
1
2
9
1
2
1
1
1
1
2
]
2
1
1
2
1
3
1
1
3
2
1
1
2
1
1
1
1
Weipht
(gni)

0.001010
0.001154
0.000705
0.005000
.0.003253
0.002526
0.001913
0.000041
0.001018
0.001315
0.000148
0.000684
0.000173
0.000908
0.008156
0.003337
0.004232
0.002547
0.002735
0.002653
0.000677
0.010770
0.000063
0.000390
0.000444
0.000885
0.001389
0.003097
0.002754
0.000003
0.000004
0.005000
0.001896
0.006620
0.002077
0.000771
0.000695
0.000290
0.000046
0.000973

-------
Table A-8 (Continued)
Sample
Number
Chaoborus
Pupae
1067
Clams
161
1060
2057
2062
4058
4060
4065
4071
Tabanids
371
373
374
375
451
1065
2074
3067
3068
3070
3072
3073
3074
4067
4072
4073
4074
5066
5068
5070
5072
5074
5075
Depth
0")


7.33

10.07
10.99
10.99
8.70
11.22
10.99
8.70
3.66

3.20
1.14
1.60
0.46
0.23
8.01
0.92
5.49
5.04
3.66
2.52
2.52
0.69
4.12
2.52
2.06
1.14
8.01
6.87
4.35
0.46
1.37
0.92
Date


6-12-70

7- 5-69
6-12-70
7- 3-70
7- 3-70
8-14-70
8-14-70
8-14-70
8-14-70

8-15-69
8-15-69
8-15-69
8-15-69
9- 6-69
6-12-70
7- 3-70
7-24-70
7-24-70
7-24-70
7-24-70
7-24-70
7-24-70
8-14-70
8-14-70
8-14-70
8-14-70
9- 4-70
9- 4-70
9- 4-70
9- 4-70
9- 4-70
9- 4-70
Number


1

1
1
1
2
1
1
1
1

1
1
2
1
2
1
4
1
1
1
8
5
3
1
1
2
1
1
1
4
2
5
3
Weipht
(flro)


0.000749

0.033548
0.004714
0.004614
0.007890
0.001932
0.003084
0.022210
0.007647

0.000580
0.002152
0.001375
0.006476
0.013576
0.001660
0.000255
0.000652
0.008910
0.000661
0.000643
0.001622
0.001811
0.001217
0.000527
0.001474
0.001771
0.004368
0.003284
0.011280
0.004249
0.011580
0.008520

-------
Table A-8 (Continued)
Sample
Number
Megaloptera
452
3060
3063
3075
4056
4067
Depth
Cm)

2.75
11.67
8.70
0.23
10.99
4.12
Date

9- 6-69
7-24-70
7-24-70
7-24-70
8-14-70
8-14-70
Number

1
1
1
1
1
1
Weight
Om)

0.002803
0.000092
0.000083
0.000070
0.000411
0.000526

-------
Table A-9.
Area-capacity table for Hemlock Lake based
on January 1957 survey of the lake.

Max.
Depth
(m)
0.0
0.5
1.0
1.5
2.0
2.5
3.0
?•*
L.o
IN 5
5.0
5.5
6.0
6.5
7.0
7.5
3.0
3.5
9.0
9.5
10.0
10.5
11.0
11.5
12.0
12.S
13.0
13.5
lk.0
UN*
15.0
15.5
16.0
16.5
17.0
17.5
18.0
18.5
19.0
19.5
20.0
Area
(m2)
0
555
1110
1665
2220
2775
32^.6
3638
U030
4^22
^313
5205
&6?
55<6
56)12
572Q
5816
5902
6161
6782
7^03
8025
36k6
9267
9851
10 3 Ol|
10758
11220
11723
12225
12733
13280
13826
111390
1511^6
15903
16659
17ij-l6
18172
18928
19685
Volume
(m3)
0
0510
1010
1520
2030
25UO
3860
5950
flolj.0
101^0
12220
1^320
16720
19570
22l|20
25260
28110
30960
3ljl!.iO
38030
IJ.1920
1+5810
19690
53^0
5776o
62920
68090
73330
70260
85150
911^0
97810
lO^QO
111260
119020
126780
13^.0
li+2300
150060
157830
165590
Max.
Depth Area Volume
(m) (m2) (ra3)
20.5 20l|.ljl 173350
21.0 21198 131110
21.5 21954 188870
22.0 22711 196630
22.5 23l|67 20!|1)00


































-------
Table A-10
Area-capacity table for Section Four Lake
based on January 1957 survey of the lake.
Max.
Depth
(m)
0.0
i.'o
1.5
2.0
2.5
3.0
3.5
5.0
5.5
5.0
?•*
6.0
6.5
7.0
7.5
8.0
3.5
9.0
9.5
10.0
10.5
11.0
11.5
12.0
12.5
13.0
13.5
lfj.0
15.5
15.0
*%•*
16.0
16.5
17.0
17.5
18.0
18.5
19.0
19.5
20.0
Area
0
4U
832
1323
1764
2206
2610
2979
3343
3717
4o36
4455
4712
ij8i2
4912
5012
5112
5212
5371
5653
5935
6217
61(99
6762
7062
745 2
7303
314?
8409
8672
8932
9174
95-17
9060
9989
10311
10633
10955
11277
11599
11922
Volume
(m3)
0
0400
08 10
1210
1610
2020
3110
k86o
6600
3350
10090
11340
13890
16370
13860
21350
23330
26320
23990
32050
35110
38170
41230
44290
475io
512QO
55070
58930
63190
67440
71750
76390
8io4o
85710
90780
95850
100910
105980
111040
116110
121170
Max.
Depth Area Volume
(rn) (m2) (ra3)
20.5 12244 126240
21.0 12555 131300
21.5 12338 136370
22.0 13210 i4i44o
22.5 13532 146500


































-------
Table A-ll
Hemlock Lake calcium, sodium, potassium, magnesium, dissolved organic
matter (D.O.M.) and particulate organic matter (P.O.M.)  collected during
1970.  Samples were collected from six depths seven times during the
summer.  The mean concentration for the entire lake is shown.  These
analyses were made by R.G. Wetzel.
              0.0
         2.8
6.5
Depth (m)

 10.1    12.8
14.6
15.6
Mean
Calcium
VI-7-70
VI-13-70
VI-18-70
VI-27-70
VII-17-70
VIII-15-70
IX-5-70
Sodium
VI-7-70
VI-13-70
VI-18-70
VI-27-70
VII-17-70
VIII-15-70
IX-5-70

38.4
38.7
36.8
33.3
33.8
55.3
49.5

1.75
1.55
1.50
1.70
1.90
2.07
2.07

43.2
39.2
38.3
34.5
37.2
57.8
50.3

1.45
1.50
1.65
1.72
1.98
2.08
2.05

46.8
47.2
44.0
46.3
44.0
59.3
50.3

2.12
2.05
2.00
2.08
2.09
2.05
2.07

46.7
45.8
55.2
49.0
45.0
52.0
50.9

2.08
2.14
2.14
2.05
2.14
2.07
2.05

57.9
47.1
55.5
49.1
44.9
50.4
50.3

2.22
2.14
2.15
2.11
2.15
2.05
2.14

102.5
111.0
57.0
48.9
44.9
50.9
50.9

2.57
3.00
2.22
2.12
2.11
2.11
2.11

— —
47.7
44.9
41.8
40.8
55.4
50.3


1.90
1.87
1.92
2.04
2.07
2.07

-------
Table A-ll (Continued)
Depth (m)

Potassium
VI-7-70
VI-13-70
VI-18-70
VI-27-70
VII-17-70
VIII-15-70
IX-5-70
Magnesium
VI-7-70
VI-13-70
VI-18-70
VI-27-70
VII-17-70
VIII-15-70
IX-5-70
D.O.M.
VI-7-70
VI-13-70
VI-18-70
VII-17-70
VIII-15-70
IX-5-70
0.0

0.60
0.58
0.52
0.60
0.85
1.00
0.96

8.0
9.1
9.3
8.4
9.7
10.5
10.3

7.12
7.04
6.17
8.84
6.82
10.32
2.8

0.63
0.55
0.52
0.70
0.90
1.00
0.96

8.8
9.2
9.8
8.7
9.8
10.7
10.3

7.33
6.56
6.54
8.51
6.74
11.02
6.5

0.78
0.68
0.85
0.95
1.02
1.07
0.96

10.5
11.3
11.7
11.4
9.8
10.7
10.0

6.99
7.24
6.36
7.91
6.48
10.29
10.1

0.80
0.75
1.18
1.00
1.02
1.03
0.99

11.3
11.7
13.5
11.1
10.3
10.2
10.0

7.63
6.93
5.99
9.28
6.67
10.21
12.8

1.05
0.75
1.19
1.02
1.03
1.05
1.05

12.3
11.9
13.2
11.6
10.6
11.0
10.2

7.99
6.73
6.26
8.72
6.21
9.42
14.6 15.6

1.78
2.62
1.23
0.97
1.01
1.05
1.00

14.5
15.9
12.6
10.6
1003
10.9
10.6

8.39
9.14
6.53
9.24
4.68
10.89
Mean

_ .
0.78
0.81
0.84
0.96
1.03
0.98

_ _
10.8
11.3
10.1
10.0
10.6
10.2

„ _
7.04
6.33
8.60
6.49
10.46

-------
Table A-ll (Continued)
P.O.M.

VI-7-70
VI-13-70
VI-18-70
VI-27-70
VII-17-70
VIII-15-70
IX-5-70
              0.0
                                   Depth (m)
                       2.8     6.5     10.1    12.8
287.6
857.0
642.5
1619.1
693.2
148.3
591.1
756.2
893.3
1287.0
1915.6
993.4
156.2
444.8
981.6
759.9
641.3
1963.1
819.2
207.6
765.1
916.3
1580.1
1100.9
1635.9
566.4
128.5
377.6
1345.3
1408.6
1026.8
1774.3
554.5
164.1
319.7
14.6
                                                      2627.4
15.6
Mean
3661.1
1161.4
1774.3
643.5
243.2
1195.7
972.9
1818.3
776.7
170.0
                                                               306.4   513.2

-------
Table A-12.  Section Four Lake calcium,  sodium,  potassium, magnesium,  dissolved organic
             matter (D.O.M.)  and particulate organic matter (P.O.M.)  collected during
             1970.  Samples were collected from six depths seven times during the summer,
             The mean concentration for the entire lake is shown.  These analyses were
             made by R.G. Wetzel.
              0.0
3.7
7.4
Depth (m)

 11.0    14.6
17.4
18.2
Mean
Calcium
VI-7-70
VI-13-70
VI-18-70
VI-27-70
VII-17-70
VIII-15-70
IX-5-70
Sodium
VI-7-70
VI-13-70
VI-18-70
VI-27-70
VII-17-70
VIII-15-70
IX-5-70

51.5
53.0
55.0
55.7
55.0
52.8
52.8

2.88
2.95
3.04
3.15
2.81
3.17
3.16

55.0
56.5
55.6
56.0
53.4
52.6
52.6

3.25
3.05
3.00
3.15
3.20
3.10
3.15

58.4
58.0
55.6
56.0
54.5
53.2
52.8

3.28
3.22
3.06
3.13
3.20
3.11
3.17

57.5
56.5
55.7
57.0
54.8
49.8
51.6

3.20
3.08
3.13
3.15
3.09
3.17
3.21

57.5
58.0
58.8
56.9
55.0
52.8
52.8

3.08
3.16
3.22
3.17
3.23
3.08
3.17

57.5
58.2
58.5
56.2
55.0
52.8
53.2

3.20
3.18
3.13
3.13
3.21
3.12
3.13

— —
56.0
56.0
56.2
54.4
52.3
52.6

— —
3.09
3.07
3.15
3.12
3.12
3.17

-------
Table A-12 (Continued)

Potassium
VI-7-70
VI-13-70
VI-18-70
VI-27-70
VII-17-70
VIII-15-70
IX-5-70
Magnesium
VI-7-70
VI-13-70
VI-18-70
VI-27-70
VII-17-70
VIII-15-70
IX-5-70
D.O.M.
VI-7-70
VI-13-70
VI-18-70
VII-17-70
VIII-15-70
IX-5-70
0.0

0.58
0.63
0.67
0.59
0.57
0.60
0.60

15.2
15.1
16.0
13.9
10.3
10.3
10.4

6.63
3.45
3.09
3.49
3.15
6.75
3.7

0.69
0.63
0.63
0.60
0.58
0.60
0.62

16.4
15.7
16.5
12.9
10.1
10.3
10.2

2.57
2.77
2.80
3.78
2.53
5.81
7.4

0.64
0.64
0.63
0.57
0.57
0.60
0.60

18.7
17.0
14.9
11.9
10.3
9.2
9.9

2.42
2.48
2.67
3.34
3.09
6.93
Depth (m)
11.0 14.6

0.65
0.64
0.64
0.57
0.57
0.58
0.60

17.2
16.4
14.6
11.3
9.9
9.2
10.3

2.43
2.79
2.80
3.49
2.47
4.76

0065
0.65
0.69
0.59
0.64
0.57
0.61

18.7
16.0
15.1
11.1
8.9
9.3
10.6

2.20
2.68
2.56
3.54
2.32
6.90
17.4 18.2

0.65
0.64
0.64
0.58
0.55
0.59
0.59

16.8
15.9
14.1
10.4
9.2
10.3
10.4

2.57
2.23
2.42
3.66
2.90
6.26
Mean

_ _
0.64
0.65
0.58
0.58
0.59
0.61

_ _
16.0
15.5
12.3
10.0
9.8
10.2

._
2.79
2.78
3.55
2.74
6.18

-------
Table A-12 (Continued)
Depth (m)

P.O.M.
VI-7-70
VI-13-70
VI-18-70
VI-27-70
VII-17-70
VIII-15-70
IX-5-70
0.0

140.4
140.4
310.4
417.1
128.5
259.0
148.3
3.7

168.0
136.4
381.6
405.3
81.0
176.0
199.7
7.4

156.2
172.0
405.3
440.9
132.4
235.3
211.5
11.0

247.1
215.5
385.5
452.7
140.4
187.8
259.0
14.6 17.4

227.4 345.0
144.3
425.0
480.4
124.6
203.6
148.3
18.2

_ _
191.8
322.2
531.8
132.4
160.1
191.8
Mean

•» •>
161.4
378.0
436.2
117.1
207.4
198.0

-------
                           465
Figure A-l.  Length histograms of hatchery reared rainbow
             trout at time of stocking in Hemlock Lake dur-
             ing June 6, 1969 and June 25, 1970.  Only one
             lot of fish were stocked during 1969, whereas
             four lots were stocked during 1970,.  Each lot
             received a separate fin clip.  Total numbers
             (n) ,  average fish lengths (x\. and fin clips
             for each lot are shown.

-------
0-
 5--
O-.r
   10+
 c
 i 5:~-
a,  -
  0-
   5--
0-
10-
                                   1969
                                  R. Pel.
                                  n.= 1002
                                  X= 7,3
             ^JTf
                                     1970

                                    Anal
                                    n = 540
                                    7=8.1
                                      1970
                                     R.Pec.
                                     n=605
                                     x=8.0
             rn rT
                                    1970

                                    L. Pec.
                                     = 8.1
 11iii r j  i i i i  i j ii i i i
6.3     6.9     7.5     8'.1
          Length  (inches)
                                        1970

                                        L.Pel.
                                        Jl=560
                                        x= 8.1

                                   fe
                                8.7
                                          9.3
9.9

-------
                            467
Figure A-2.  Length histograms of hatchery reared rainbow
             trout at time of stocking in Section Four
             Lake during June 6, 1969 and May 23, 1970.
             One lot was stocked each year.  Total numbers
              (n) , average fish-lengths  (3c) and fin clips
             for each lot are shown.

-------
10-

5-
«**
c
0)
s°-
0,)
p. -
10-
~
5^
0 "




-i_


























-










-




1969





—i














R. Pec.
















™










n = 1000
7=73


TKn
1 rrK-^-f-i_ „

1970
L.Pel.




n=1071
x-7.9
TTh-i^
6-3      6i9      7.5       8.1      6.7
           Length  (inches)
9.3
9.9

-------
                     ACKNOWLEDGEMENTS

     The advice and assistance of my graduate committee is
much appreciated.  Drs. R. C  Ball, W.  E.  Cooper,  K.  W.
Cummins, G. E. Guyer and N. R. Kevern  served  on  this  commit-
tee.
     The Michigan Department of Natural Resources  generously
supported this project through its loan of equipment,  facili-
ties, records, hatchery fish and personnel.   This  project
could not have been conducted on the scale it was  without
their help.  Dr. Gerald Cooper, Jerry Myers,  Bob Barber and
Dr. Carl Latta were especially helpful.
     Several students assisted me with the work.   I cannot
say enough about their enthusiasm and support.  They in-
cluded:  Larry Waterworth, Bob Hoffman, Mike  Coney, Tawn
Jarvis, Kathy Hunter, Cheryl Bowden,  Scott Mosiman and Bob
McConnel.
     Dr. R. G. Wetzel assisted with water  chemistry.
Dr. Brian Moss analyzed the phytoplankton.  Drs. Frank D'ltri
and Marvin Stevenson advised on water chemistry and engineer-
ing aspects of the project respectively.  The help of all
these people is appreciated.
                             1+69

-------
     Martha L. Past assisted with much  of  the work.   Her



assistance is also greatly appreciated.



     Ivan Borton fabricated much of the equipment  used in



the study and gave valuable advice.  Here  again, this



project could not have been conducted on as large  a  scale



as it was without his valuable assistance.



     We acknowlege the support of the Water Quality  Office



of the Environmental Protection Agency for supporting much



of the work through Grant 16010 EXE.



     My personal support was through a Predoctoral Fellowship



5-F1-WP-26-292-01,03.



     Additional support was through the Michigan Agricultural



Experiment Station, Project 64.
                            14-70




U. S. GOVERNMENT PRINTING OFFICE : 1972 O - 456-249

-------