WUH (0'an suum: AND SODIUM ARSENITE Thesis {or “a Doqm of M. S. MICHIGAN STATE UNEYERSITY Leonard Paul Sohacki 1965 T W?! W ‘mfllnflw WW 3 1293 00991 7943 ' ABSTRACT ECOLOGICAL ALTERATIONS PRODUCED BY THE TREATMENT OF POND ECOSYSTEMS WITH COPPER SULFATE AND SODIUM ARSENITE by Leonard Paul Sohacki The ecological alterations produced by the addition of herbicides to pond ecosystems were investigated. Three ponds were used in the study and each received treatment as follows: pond A, 8 ppm sodium arsenite; pond B, 8 ppm sodium arsenite and 2 ppm copper sulfate; pond C, 2 ppm copper sulfate. Changes in water chemistry, which included dissolved oxygen, alkalinity, pH, and carbon dioxide, were exhibited by all of the treated ponds. In general, the severity of changes produced by the treatments were related to the concentration of plant biomass killed. Dissolved oxygen and pH values of the pond waters decreased shortly after the herbicide treatments but returned to the normal range of values within two weeks. The alkalinity values of the ponds increased to as high as 20 ppm above the normal range of values and took as long as one and one—half months to attain normal conditions again. Carbon dioxide concentrations in- creased simultaneously with plant decomposition and returned to normal within two weeks after the herbicide applications. 1 Leonard Paul Sohacki Herbicide induced changes in the productivity of the macrophytes, phytoplankton and periphyton were observed. Copper sulfate effectively eliminated the macroscopic alga, Chara. Sodium arsenite selectively killed the higher aquatic plants but showed no toxicity towards Chaga. Once the higher aquatic plants were killed, they showed no signs of recovery for the remainder of the summer. Contrarily, Chara exhibited new growth within one month after herbicide treatment. The periphyton and phytoplankton productivity were altered by sodium arsenite as well as by copper sulfate. Phytoplankton productivity as measured by the carbon-14 method showed that the phytoplankton were inhibited immedi- ately after the herbicides were added and for five to ten days thereafter. No evidence of a phytoplankton bloom accompanied the release of nutrients from the decomposing vegetation. Unlike the phytoplankton, the periphyton showed little inhibition from the herbicide treatments, but re- sponded to the release of nutrients as shown by increased standing crop levels. A comparison of the pretreatment and post-treatment benthic invertebrate populations showed significant changes which were attributed to the toxicity of the herbicides. Both c0pper sulfate and sodium arsenite produced serious Leonard Paul Sohacki decreases in the invertebrate populations. However, the greatest decrease was shown in pond B where the bottom fauna was subjected to the combined toxicities of copper sulfate and sodium arsenite. Although all invertebrates were affected, the mayfly naiads showed extreme suscept- ability particularly to sodium arsenite. ECOLOGICAL ALTERATIONS PRODUCED BY THE TREATMENT OF POND ECOSYSTEMS WITH COPPER SULFATE AND SODIUM ARSENITE BY Leonard Paul Sohacki A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Fisheries and Wildlife 1965 ACKNOWLEDGEMENTS The writer wishes to extend his thanks to Dr. Robert C. Ball for his guidance and patience during the course of this study. Thanks is also extended to the graduate students of the Fisheries and Wildlife Department who offered assist- ance whenever it was needed. Finally I wish to thank my wife, Betty, for her constant encouragement and selfless devotion in assuming more than her share of family responsibilities while this work was being conducted. This study was conducted while the author was subsi- dized through a Graduate Research Assistantship sponsored by the Atomic Energy Commission Grant AT (11-1) 655, and administered by Drs. R. C. Ball and F. F. Hooper. ii TABLE OF CONTENTS Page INTRODUCTION . . . . . . . . . . . . . . . . . . . 1 DESCRIPTION OF STUDY AREA. . . . . . . . . . . . . 8 METHODOLOGY. . . . . . . . . . . . . . . . . . . . 10 Sodium Arsenite Application. . . . . . . . . 10 Copper Sulfate Application . . . . . . . . . 11 Copper Sulfate and Sodium Arsenite . . . . . 12 Physical Measurements. . . . . . . . . . . . . . . 12 Solar Radiation. . . . . . . . . . . . . . . 12 Chemical Measurements. . . . . . . . . . . . . . . 15 Dissolved Oxygen . . . . . . . . . . . . . . 15 Alkalinity . . . . . . . . . . . . . . . . . 15 Hydrogen Ion Concentration . . . . . . . . . 15 Carbon Dioxide Concentration . . . . . . . . 15 BIOLOGICAL MEASUREMENTS. . . . . . . . . . . . . . 14 Macrophytes. . . . . . . . . . . . . . . . . 14 Phytoplankton. . . . . . . . . . . . . . . . 14 Periphyton . . . . . . . . . . . . . . . . . 19 Bottom Invertebrates . . . . . . . . . . . . 21 MACROPHYTES. . . . . . . . . . . . . . . . . . . . 25 Copper Sulfate . . . . . . . . . . . . . . . 25 Sodium Arsenite. . . . . . . . . . . . . . . 26 Copper Sulfate and Sodium Arsenite . . . . . 28 DISSOLVED OXYGEN . . . . . . . . . . . . . . . . . 29 CARBON DIOXIDE, pH, AND ALKALINITY . . . . . . . . 55 iii TABLE OF CONTENTS - Continued Page PHYTOPLANKTON . . . . . . . . . . . . . . . . . . . 44 Immediate Response of Phytoplankton Productivity to Herbicides. . . . . . . 44 COpper Sulfate. . . . . . . . . . . . . . . . 47 Sodium Arsenite . . . . . . . . . . . . . . . 51 Copper Sulfate and Sodium Arsenite. . . . . . 52 Post-treatment Response to Increase in Nutrients . . . . . . . . . . . . . . . 55 PERIPHYTON. . . . . . . . . . . . . . . . . . . . . 57 Copper Sulfate. . . . . . . . . . . . . . . . 58 Sodium Arsenite . . . . . . . . . . . . . . . 62 COpper Sulfate and Sodium Arsenite. . . . . . 65 BOTTOM INVERTEBRATES. . . . . . . . . . . . . . . . 64 Copper Sulfate. . . . . . . . . . . . . . . . 65 Sodium Arsenite . . . . . . . . . . . . . . . 67 Copper Sulfate and Sodium Arsenite. . . . . . 69 Comparisons of Herbicide Toxicities . . . . . 72 SUMMARY . . . . . . . . . . . . . . . . . . . . . . 76 LITERATURE CITED. . . . . . . . . . . . . . . . . . 80 APPENDIX. . . . . . . . . . . . . . . . . . . . . . 85 iv TABLE 1. LIST OF TABLES Herbicide concentrations, treatment dates, and pretreatment macrophyte composition of the Lake City ponds. . . . . . . . . . . . . . . . Standing crop estimates of the macrophyte populations before and after herbicide treat- ments. . . . . . . . . . . . . . . . . . . . . Matched pairs test (mean of the differences) of the pretreatment and post-treatment stand- ing crop (mg/m2 dry weight) of Chara in pond A . . . . . . . . . . . . . . . . . . . . Phytoplankton productivity and solar energy determinations of the Lake City ponds on the dates of herbicide application . . . . . . . . Comparison of phytoplankton productivity of the pretreatment and post-recovery periods using the Student's t-test . . . . . . . . . . Comparison of the mean pretreatment and post- treatment invertebrate populations of pond C (CuSO4) via the Student's t—test . . . . . . . Comparison of the mean pretreatment and post— treatment invertebrate populations of pond A (NaA802) via the Student's t-test. . . . . . . Comparison of the mean pretreatment and post- treatment invertebrate populations of pond B (CuSO4+NaAsOZ) via the Student's t-test. . . . Per cent reduction in bottom invertebrates of the test ponds following herbicide treatment . Page 24 25 27 46 54 66 68 7O 71 FIGURE 2. LIST OF FIGURES Map of the Lake City ponds. . . . . . . . . . Dissolved oxygen concentrations in the test ponds during the summer of 1962 . . . . . . . Carbonate and bicarbonate alkalinity values of the Lake City test ponds during the summer Of 1962 0 O 0 O O O O O O O O O O O O O O O 0 Hydrogen ion concentration (pH) and free carbon dioxide concentrations greater than 1 ppm of the test ponds during the summer of 1962. . . . . . . . . . . . . . . . . . . . . Phytoplankton productivity of the test ponds during the summer of 1962 . . . . . . . . . . Periphyton standing crop of the test ponds expressed as phytopigment units per square decimeter . . . . . . . . . . . . . . . . . . vi 52 59 41 50 6O INTRODUCTION Aquatic weed control has become an indispensable practice for successful pond management because of the detrimental effects nuisance weeds exert upon fish popu- lations. Fish management is hampered by aquatic weeds in the following ways: (1) weeds protect small fish from predators and thereby enhance the development of over- crowded, stunted fish populations, (2) weeds compete for nutrients with the desirable primary producers which com- prise the base of the food chain, (5) heavy weed infesta- tions lead to winter and summer fish kills, and (4) weeds interfere with fish removal from the ponds. Weeds are removed from the ponds by mechanical or chemical means, however, chemical control is usually pre- ferred. Sodium arsenite and copper sulfate are the most commonly used chemical controlling agents; sodium arsenite is utilized primarily for higher aquatic weed control, whereas copper sulfate is used for algae. These two herbi- cides have remained popular for over fifty years despite the discoveries of newer organic herbicides. The lasting popularity of sodium arsenite and copper sulfate can be attributed to their lower cost, and of more importance, their effectiveness in removing common aquatic nuisance weeds. In spite of their extended use as aquatic herbicides, very little is known of the ecological alterations induced by the addition of copper sulfate or sodium arsenite. The more apparent changes such as reduction in the amounts of vegetation and depletion of dissolved oxygen concentrations are well-known. Investigators have devoted considerable study to the alterations which produce fish mortality, but the insidious changes which accompany herbicide treatment have been neglected. However, the absence of mortality doesn't necessarily mean the fishes escape all detrimental effects. Lawrence (1958) showed that bluegill production was reduced by sodium arsenite treatments even though the concentrations used were considered "safe" by herbicide specialists. Although Lawrence )op. cit.) did not attribute the reduced production to any specific factor, it is pre- sumed that a combination of herbicide toxicity and the accompanying ecological changes were responsible for the inhibition of fish growth, with the latter enhancing the toxic effects of the former. The objective of this study was to discover the magnitude and duration of the latent ecological changes which are caused by treatment of sodium arsenite, copper sulfate or combination thereof. Of the various latent changes initiated by the herbi- cide addition, the following are thought to most seriously impair the welfare of the aquatic community: (1) alteration of the water chemistry, and (2) inhibition of the primary producers. The chemical alteration of pond waters which follows herbicide treatment results primarily from the death and decomposition of the aquatic vegetation. The most obvious change is the reduction of dissolved oxygen in the water. Since severe oxygen depletion results in fish mortalities, this phase of water chemistry has received considerable attention from the investigators. However, other important chemical changes are known to accompany plant decomposition. Of particular significance is the increase in carbon dioxide concentrations. Because carbon dioxide, pH, and alkalinity are interrelated, a change in one is usually accompanied by changes in the other two factors. This is especially true in aquatic systems which are not well buffered. Although radical alterations in the pH can deleteriously influence the aquatic biota, its effects are not thought to be as toxic as changes in carbon dioxide and oxygen. Low oxygen concentrations are considered more toxic in the presence of high carbon dioxide levels (Welch, 1952). Such conditions often follow herbicide applications. Due to the non-specificity of sodium arsenite and copper sulfate, the desirable as well as the undesirable primary producers are killed or inhibited in growth. In pond ecosystems the phytoplankton and periphyton are principal contributors to the base of the food chain, and hence all aquatic heterotrophs are directly or indirectly dependent upon these algae for existence. Consequently, the absence or inhibition of these primary producers for an extended period of time would adversely influence all trophic levels. Yet little information can be found per- taining to the magnitude and duration of algal inhibition following herbicide applications. Smith (1940) reported that algal production was impaired for over a year in soft water lakes which received 5 parts per million (ppm) copper sulfate. On the other hand, Moyle (1949) stated that a decrease in undesirable forms is followed shortly by an increase in desirable forms. The information presented in these two articles dealt only with planktonic forms. No data pertaining to the influence of copper sulfate upon periphyton could be found. Although sodium arsenite is considered a higher aquatics-specific herbicide it also exhibits algacidal properties, and hence could also inhibit the beneficial primary producers. Surber (1945) and Lawrence (1958) demon- strated the effectiveness of sodium arsenite upon filament- ous algae such as Hydrodictyon, Spirogyra, and Pithoghora. Since these nuisance algal forms are inhibited, it seems reasonable to assume that the desirable forms are inhibited also. However, this has never been proven. A meager amount of information is available concerning the toxicity of copper sulfate and sodium arsenite treatment upon benthic fauna. Lawrence (1958) is the only investi- gator who demonstrated that sodium arsenite reduced bottom insect populations under natural conditions. Surber (1951) showed that low concentrations (2.5 ppm) killed susceptable insects under laboratory conditions. Even less is known about the influence of copper sulfate. Although Mackenthun and Cooley (1952) demonstrated that copper sulfate accrual in lake sediments from prior treatments had no effect upon the benthic fauna, no data are available on its direct influence. Because of the important role played by the bottom fauna in the aquatic community, further investigation was warranted. .mpcom xuflo mxmq mcu wo mm: .H wusmflm mQZOm A¢EZ§Hmmme MHO>mmmmm Mm08¢m0m¢q DESCRIPTION OF STUDY AREA This study was conducted at the Lake City Experi- mental Farm which is located approximately two miles south of Lake City, Michigan. The farm, owned and operated by Michigan State University, is used for agricultural. research projects conducted by the Agricultural College. The Department of Fisheries and Wildlife maintains four experimental ponds and a field laboratory on the premises where limnological research work is conducted during the summer seasons. The experimental ponds were constructed during the period 1945-1945 on a marshy area near Mosquito Creek, a small stream which rises on the farm property. A dam, located on the stream, forms a backwater area which is used as a reservoir for filling the ponds. Each pond has its own inlet and outlet and hence can be filled or drained independently of the others (Fig. 1). Downstream from the dam the stream bed passes behind the pond outlets and serves as a drainage channel when the ponds are emptied. From west to east the ponds are designated by the letters A, B, C, and D; only ponds A, B, and C were used in this project. The areas and average depths of the test ponds are as follows: Pond A - .56 acres, 5.2 ft. average depth Pond B - .46 acres, 5.5 ft. average depth Pond C - .17 acres, 5.4 ft. average depth. All ponds have a maximal depth of 6 ft. at the spillway. The ponds were originally built on a sandy base, but presently the sand is partially covered by a mucky deposit interspersed with small woody particles. METHODOLOGY Sodium Arsenite Application Sodium arsenite (NaAsOe) is a herbicide well-known for its toxicity to terrestrial as well as aquatic plants. Liquid sodium arsenite is recommended for treating aquatic habitats in preference to the dry forms (Hooper and Cook, 1957). The liquid form is manufactured by a number of companies under different brand names, but in each case the product contains four pounds of arsenic trioxide (A5203) per gallon. On July 20, 1962, pond A was treated with a form of sodium arsenite known as Atlas-A, which is manufactured by the Chipman Chemical Company. Enough sodium arsenite was added to the pond to bring the arsenic concentration up to 8 ppm A5203. Before being added to the pond, the concentrated sodium arsenite was thoroughly mixed with 150 gallons of pond water in a large plastic lined tank located at the pond's edge. A portable engine-driven pump was used to pump the diluted herbicide from the tank through a length of 2 inch diameter hose into the pond. An operator at the nozzle directed the stream of dilute herbicide across the surface of the pond. Immediately after dispensing the 10 11 herbicide, a small boat equipped with a 5 horsepower out- board engine was used to agitate the pond and mix the herbicide to all depths. Copper Sulfate Application COpper sulfate (CuSO4-5H20) is an algae specific herbicide which has been used for controlling algal blooms since 1904. With the exception of a few species, most algae readily succumb to low concentrations of copper sul- fate. As little as .5 ppm effectively control obnoxious phytoplankton blooms, but as much as 1.5 ppm are required to kill Chara. Since Chara was the nuisance alga in the test ponds, higher concentrations of copper sulfate were necessary. However, the susceptibility of the algae is not the only factor which determines the concentrations of cop- per sulfate to be used; the water chemistry of the system must also be considered. Bicarbonates and suspended organic matter react with copper sulfate to inactivate its algacidal properties. Consequently, adjustments must be made to compensate for these reactions and are described by Smith (1955). Powdered copper sulfate, which is the most soluble form, was used to treat pond C on July 12, 1962. The amounts necessary to bring the c0pper sulfate concentration of the pond water to 2 ppm was measured out to the nearest gram, placed in a fine meshed sack and towed through the pond behind an outboard powered boat. 12 Copper Sulfate and Sodium Arsenite Pond B received a treatment of 2 ppm copper sulfate and 8 ppm sodium arsenite on August 5, 1962. These herbi- cides were applied within one-half hour using the methods of application described in the preceding sections. Physical Measurements Solar Radiation The incident solar radiation in the vicinity of the Lake City ponds was measured by a pyrheliometer-recorder assembly. The pyrheliometer is a radiation detector of the thermopile type which consists of a number of series cone nected thermocouples that generate an electromotive force (emf) proportional to the intensity of the incident radi- ation. An Eppley pyrheliometer was mounted on a platform located between ponds C and D, and was connected by an under- ground cable to a Bristol strip-chart recorder located in the laboratory. The recorder converted the emf produced by the pyrheliometer into units of gram calories per square centimeter per minute. The gram calorie is a unit of pyrheliometry defined as the quantity of heat necessary to change the temperature of one gram of water from 5.5 to 4.5 degrees centigrade. 15 Chemical Measurements With few exceptions, chemical analyses were conducted daily throughout the summer. Surface water samples were collected between 1400 and 1600 hours, and transferred to the laboratory for immediate processing. Dissolved Oxygen The unmodified Winkler method (Standard Methods, 1961) was used to determine the dissolved oxygen content of the pond waters. Alkalinity Alkalinity, or the acid destroying capacity imparted to the water by carbonates, bicarbonates, and hydroxides, was determined by titrating the water samples with N/50 sulfuric acid using phenolphthalein and methyl orange as end point indicators. Hydrogen Ion Concentration The hydrogen ion concentration, or pH, was determined with a Beckman Model H2 pH meter. Carbon Dioxide Concentration The alkalinity and pH values were used in conjunction with Moore's nomograph to determine the carbon dioxide concentrations. 14 Biological Measurements Macrophytes The macroscopic bottom plants (aquatic macrophytes) of each pond were sampled twice during the summer, once before and once after the herbicide applications. 'Each sampling con- sisted of thirty randomly selected Petersen dredge samples (.802 M2). The individual samples were placed in a 50 mesh sieve and washed to remove the silt and sand particles. Then the plants were carefully separated from the material remaining in the sieve and washed again to remove any adhering particles of sand, detritus, etc.; the residue in the sieve was saved for invertebrate analysis. Subsequently, the plants were dried in a drying oven maintained at a temperature of 550C., and finally weighed to the nearest .001 gram. Phytoplankton The carbon-fourteen method, introduced by Steemann Nielsen (1952) with modifications by Rodhe (1958), was used throughout the summer of 1962 to study phytoplankton pro- duction. Rodhe (op. cit.) exposed his light and dark bottles from sunset of one day to sunset of the next. He maintained that quarter or half-day production values extrapolated for the whole day are subject to error because of unequal inso- lation. 15 Glass reagent bottles (250 ml.) were employed as light and dark exposure containers. Light was excluded from the dark bottles with a coat of black paint and a layer of black plastic tape, as recommended by Strickland (1960). Shortly after sunset water samples for the Cl4 method were taken from the 52 inch depth with a modified Meyer bottle and poured into three containers, a light and a dark bottle plus a sample bottle for alkalinity determination. Two microcuries of C14 as sodium carbonate (NaCl403) were injected into the contents of the light and dark bottles with a 2 ml. insulin syringe. (The NaC1403, contained in 2 ml ampoules, was purchased from the Volk Chemical Company.) Then the exposure bottles were stoppered, clamped to a removable pipe support and exposed at the 52 inch depth. The sample taken for alkalinity determination was immediately transferred to the laboratory and titrated. The exposed bottles were removed from the ponds on the sunset of the following date and processed immediately. A 100 ml aliquot was measured from each bottle and filtered separately through .45 micron Millipore Filters. Then 10 ml .005N HCl was passed through the filter to remove inorganic C14, and followed by 10 ml 10% formalin to kill and preserve the plankton. The filters were removed from the funnels, glued to stainless steel planchets and stored in a dessicator which contained silica gel. 16 All samples were counted in a Tracerlab low background Omni/Guard counter for ten minutes. Since a zero thickness was assumed, no corrections for self absorption were made. The basic principle underlying the Cl4 method is the photosynthetic conversion of carbon dioxide to organic car- bon by phytoplankton. Unlike the terrestrial environment, free carbon dioxide in many aquatic ecosystems is practically nonexistent. Instead, carbon dioxide exists in the chemically combined forms of bound and half-bound carbonates such as CaC03 and Ca(HC03)2. Aquatic plants possess physiological mechanisms which enable them to extract carbon dioxide from these carbonates. When a small tracer dose of radioactive carbon (C14) in a sodium carbonate (NaCl403) carrier is added to a sample of pond or lake water and exposed to favorable conditions, the phytoplankton convert the C1402 from the artificial carbonate into organic matter. Upon completion of the exposure period, the phytoplankters are filtered from the water sample and assayed for radioactivity. The ratio of Cl4 assimilated Cl4 available represents the proportion of Cl4 uptake from the original concentration. Since stable carbon (C12) from the natural occurring carbonates is assimilated simul- taneously with C14 in about the same proportions, the total original C12 content of the water sample multiplied by the preceding ratio yields the Cl2 assimilated during the ex- posure period. 17 Calculations: The presumed constant relationship between the assimi- lation of stable and radioactive carbon is expressed in the following equation: C12 Assimilated = Cl4 Assimilated (K) C12 Available C14 Available Cl4 assimilated from the total avail- For every fraction of able supply, a proportional amount of C12 is assimilated from the stable carbon supply. 1. Cl4 assimilated: Non-photosynthetic Cl4 assimilation by phytoplankton and bacterial physiological processes obscures the true photosynthetic values. Fortunately, the non-photosynthetic assimilation progresses at the same rate in the darkness as in the light, and hence this assimilation is equal in both, the light and dark bottles. The activity of the dark bottle in counts per second (cps) subtracted from the cps of the light bottles yields net photosynthetic activity. The counts registered by radiation detectors are not enumerations of particle emissions, but manifestations of particle initiated ionization events in the counting chamber. Because not all beta emissions initiate an ionization, not all particles are counted. The fraction of particles counted of the total emitted constitutes the counter efficiency, 18 a factor which must be multiplied times the cps of each sample to convert relative counts into absolute disinte- grations per second (dps). For example, if the counter efficiency is 25% all counts must be multiplied by a factor of 4. Correction must also be made for subsample size if the entire sample is not filtered. For example if a 100 ml aliquot of a 500 ml sample is filtered, the dps must be multiplied by a factor of 5. With all factors included, the Cl4 assimilated value of the preceding equation may be calculated as follows: C14 assimilated in dps = (cps light bottle-cps dark sample size ) subsample size ' bottle) (counter efficiency) ( 2. Cl4 available: This value represents the total activity in dps added to the sample bottle. In this study, 2 microcuries or 7.4 x 104 dps were injected into the sample. 5. C12 available: The C12 available is calculated from the alkalinity determination made at the time of sampling. Alkalinity titrations with .02N H2504 results in alkalinity values expressed as milligrams per liter (mg/l) CaC03; hence C12 = (.12) (mg/l CaCOa). The factor .12, which represents 19 the fraction of CaC03 weight contributed by carbon, was calculated as follows: Molecular Wt. (M.W.) Carbon = 12; 12 M.W. Caco3 = 100; 15 = .12. 4. K: K equals the discrimination constant. Nielsen (1952) and others demonstrated that Cl4 is assimilated slightly C12 slower than , hence a correction of 6% is added to the C12 value. Periphyton The autotrophs which are found growing on the surfaces of organic and inorganic objects in aquatic environments, and forming a slimy layer upon these objects are known as periphyton. These organisms, comprised mainly of algae, differ from other primary producers by their habit of attach- ing to but not penetrating the surface upon which they live. Various methods have been used for measuring the pro- duction of periphyton, however, the artificial substrate method was used in this study. Twenty-four plexiglass sub- strates (140 cm2 exposure area) were attached to an ex- posure rack and exposed horizontally 15 inches below the pond's surface. Periodically, four of the substrates were removed from each pond; three were processed for pro- ductivity estimations and the organisms from the fourth were preserved in a formal—alcohol solution for future microscopic analysis. 20 The substrates used for production estimations were processed in a manner similar to Grzenda's (1960). Each substrate was scraped of periphyton and rinsed with 95% ethanol. This mixture was then increased in volume to 50 ml with more ethanol, transferred to a 2 oz. bottle and placed in the dark for at least 24 hours for phytOpigment extraction. After extraction, 25 ml of the phytopigment solution was pipeted into a colorimeter cell and absorbency readings were made using a Klett Colorimeter equipped with a red filter (645-700 mu range). Special care was taken during the pipet- ing process to prevent the settled algal cells from being disturbed, since the turbidity caused by the suspended cellular material would interfere with the colorimeter read— ings. Any samples which were accidentally agitated were allowed to settle before the absorbency reading was made. Grzenda (1960) showed that ethanol-chlorophyll solution absorbency conformed to the Lambert-Beer Law up to a reading of 0.20 (Klett Colorimeter with 600-700 mu filter). Beyond this range the deviation from the Lambert-Beer Law increased proportionally with higher phytopigment concentrations, thereby requiring the use of a correction graph. In this study the absorbency readings never exceeded 0.20 and hence no correction of figures was necessary. Consequently, the absorbency figures presented here are directly equivalent to Grzenda's (op. cit.) phytopigment units. For convenience, the phytopigment units were multiplied by a factor of 103 for all graphical and tabular presentations. 21 Ultimately, each periphyton sample was transferred to a preweighed evaporating dish and completely dried at 55°C. After the dry weight was determined, the samples were placed in a muffle furnace and ashed at 550°C; the dry weight minus the ash free dry weight constituted the organic weight of the periphyton. Bottom Invertebrates The material remaining in the sieve after the plants were removed consisted of coarse bottom sediments, organic debris, and benthic invertebrates. This agglomeration was transferred to pint jars, preserved in formalin, and eventually inspected for invertebrates. A sugar flotation method was used to separate the insects from the bottom refuse. Sugar water, concentrated enough to float the invertebrates, was added to the samples and the mixture agitated. The agitating action caused most of the inverte- brates to float to the surface where they were easily re- moved with forceps; the residue was then thoroughly in- spected for non—floating invertebrates. After the invertebrates were separated into their respective categories, the wet weights were determined by the following method. First, these organisms were soaked in distilled water for 50 minutes. Then they were trans- ferred to special centrifuge screens, and centrifuged for 22 50 seconds to 1800 rpm. After centrifuging, the insects were immediately weighed on an analytical balance to the nearest .0001 gram. MACROPHYTES Pretreatment standing crop estimates showed Chara to be the sole macrophyte in pond C, and the dominating plant in the other ponds. In addition to Chara, ponds A and B also contained higher aquatics plants. The macrophyte composition of the test ponds along with the herbicide treatment data are included in Table 1. A comparison of the pretreatment and post-treatment standing crop estimates revealed the effectiveness of the herbicide treatments. This is shown in Table 2. All macro- phytes except the Chara in pond A were successfully elimi- nated or retarded by the herbicides. COpper Sulfate The Chara in pond C showed first signs of reacting to the copper sulfate treatment three days after the herbicide was added. At this time, these plants exhibited a pale yellow color. Within a few days after the abnormal colora- tion was exhibited, all of the gha£a_slumped to the bottom and decomposed. Very little growth activity was shown by the Chara for the remainder of the season. In fact, thirty post-treatment bottom samples yielded only a fraction of a gram of this macrophyte. 25 24 Table 1. Herbicide concentrations, treatment dates, and pretreatment macrophyte composition of the Lake City ponds. Treatment Pond Herbicide Ppm. Date Macrophytes C CuSO4 2 7-12 Chara pp. A NaAsOa 8 7—20 Chara pp. Elodea canadensis Najas flexilis Potamogeton pp, B CuSO4 2 8-5 Chara pp. Elodea canadensis NaA502 8 Najas flexilis Potamogeton pp. Table 2. Standing crop estimates of the macrophyte popu- lations before and after herbicide treatment. (9-12-62) Higher quatics Chara sp. Mean* Standard Mean* Standard Error Error Pond C Pretreatment 0.00 ---— 96.4 12.7 (7-10-62) Post-treatment 0.00 -—-- trace Pond A Pretreatment 52.0 11.9 49.6 9.1 (7-18-62) Post—treatment 0.0 ---— 55.8 9.5 (9-5-62 Pond B Pretreatment 57.5 9.8 61.7 14.7 (8—1-62) Post-treatment 0.0 ---- 6 9 1.7 * Grams per square meter dry weight. 26 Sodium Arsenite The plants in pond A exhibited a delayed response to the toxicity of sodium arsenite. Six days elapsed before the higher aquatic plants showed any signs of unhealthiness. Shortly thereafter, these plants slumped to the bottom. Mackenthun (1955), Hooper and Cook (1957) and others state that the plants usually slump to the bottom in approximately five days or less after herbicide treatments. Pond A was the only pond which required a longer time to respond to the herbicide action. It is believed that the cloudy con- ditions which prevailed on the date of herbicide treatment and the ensuing days were responsible for the delay (see Appendix for solar energy values). Low light intensities accompanied by reduced metabolic activities in the plants would result in a slower uptake of the herbicide and hence a delayed kill. Although the higher aquatic plants were completely eliminated by sodium arsenite (Table 2), the gpppp exhibited resistance. A slight yellowing of the Chara "stems" was the only visible sign of reaction to the herbicide. Inhi— bition of growth, if any, was slight, for a matched observations test revealed the post—treatment population to be significantly greater in grams per square meter than the pretreatment population (Table 5). This indicated that the Chara was actively growing during the period between 27 Table 5. Matched pairs test (mean of the differences) of the pretreatment and post-treatment standing crOp (mg/M2 dry weight) of Chara in pond A. 2a = 9.22 Zd2 = 21.85 2 _ —1§§l—-= 2.85 d = .307 n s- = 0.145 a t = 2.115 * *- Significantly different at the .05 level (d.f. = 50). 28 samplings. The resistance of this alga to the toxicity of sodium arsenite was demonstrated by Riggs (1955) and is mentioned in most herbicide manuals. Copper Sulfate and Sodium Arsenite The aquatic plants in pond B responded to the combined toxicities of copper sulfate and sodium arsenite within two days after application. And although the plants in this pond succumbed more rapidly than those in the other ponds, they were the first to recover. A new growth of Chara was evident at the post-treatment sampling, only forty days after the herbicide application (Table 2). However, the higher aquatic plants showed no signs of recovery during the remainder of 1962. DI S S OLVED OXYGEN The dissolved oxygen concentrations of any aquatic ecosystem are subject to diurnal fluctuations normally brought about by changes in the biological processes and physical components of the environment. According to Welch (1952) the two most important sources of oxygen in an aquatic system are diffusion from the atmosphere and photosynthetic production by plants. Deoxygenation is produced by the following: (1) animal respiration, (2) plant respiration, (5) bacterial decomposition, and (4) chemical oxidation of organic matter in solution (Hutchinson, 1957). Usually the oxygen depletion is "balanced" by replacement so the oxygen content of the water is seldom reduced to levels deleterious to the aquatic biota. Addition of herbicides to an aquatic en- vironment upsets this "balance" and tends to reduce the oxygen concentrations, sometimes to critical levels. The most important factors responsible for a herbi- cide induced oxygen reduction are the cessation of photo— synthetic mechanisms of the primary producers and increase in bacterial respiration. Undoubtedly the first reaction of the primary producers to the toxicity of the herbicide is the halting of photosynthetic oxygen production. 29 50 When this happens, the major portion of the oxygen enter- ing the waters is by direct diffusion from the atmosphere. However, the diffusion rates are not sufficient to satisfy the increased metabolic demands of the decomposer organ- isms, particularly the bacteria. As the primary producers succumb to the toxic action of the herbicides, their tissues are reduced to simpler compounds by oxidative decomposition of the aquatic bacteria. The greater the bulk of decaying plant tissues, the more numerous are the bacteria and the greater their oxygen demand. For this reason herbicide experts recommend repeated partial treat- ment of ponds heavily choked with weeds. Under such circum- stances, an entire treatment would result in a kill of the desirable fauna due to the oxygen deficiency caused by the decay of the large bulk of weeds. All of the test ponds exhibited a decrease in dis- solved oxygen concentrations shortly after the herbicide treatments. The herbicide induced alterations of dissolved oxygen are illustrated in Figure 2. As expected, the magnitude and duration of oxygen reduction was directly related to the amounts of vegetation killed in the pond. And hence it is not surprising to note that pond A exhibited the least reduction in oxygen concentrations. Since the Chara in pond A was apparently unaffected by the sodium arsenite, its production of phOtosynthetic oxygen alleviated the reduced oxygen condition and promoted rapid recovery of the pond. 51 .Nmma mo HmEESm wcu msflnsp mpcom ummu map CH mCOHumuusmuaoo cmmmxo Um>HommHQ .m mudmflm 52 N madman Hmnfimummm um505¢ wasp mCSb 0a m on mm ow ma 0a m on mm om we .04“ m on o m pcom m Q . Aoa m0m¢m2+¢omao ma 1.3.1:: __._.:j__:_j:____:31...__3___,______.____._I..._.a+_j.._s:14.:::-o ¢ 6 om m Q .3 . momamz 3:3:i._Tfl:j._:__1.242.422._.___:12_____.___24131221....5: o U U 0m HLJ I l 14 '- l l l L l acmso uabfixo peAIosst °mdd 55 More plant material was killed in ponds C and B than in A, and therefore the deflections in the oxygen curve are greater and longer in duration. Both, higher aquatic plants and Chara were killed in pond B by the copper sulfate-sodium arsenite combination. The bulk of decompos- able plant tissues exceeded the amounts in either of the other test ponds; this is reflected in the post-treatment oxygen concentrations. Dissolved oxygen concentrations in pond B decreased to 1.52 ppm four days after the herbicide treatment, and remained lower than the pretreatment levels for approximately two weeks. The oxygen concentrations in ponds B and C were re- duced to concentrations of less than 5 ppm. In general, 5 ppm oxygen at 250C. is critical to most fresh water fishes (Ellis, M. M., Westfall, B. A., and Ellis, M. D., 1948). Although no visible fish kill was noted in pond C, a considerable number of bullheads, green sunfish, and minnows were found dead or dying in pond B four days after the herbicide treatment. In addition, many invertebrates, mainly crayfish and dragonfly naiads were also found. Although the oxygen reduction was undoubtedly im- portant in killing the aquatic organisms, it most certainly was not the sole factor responsible. The direct toxicity of the sodium arsenite-copper sulfate undeniably had some effect. In addition, other factors such as changes in the pH and carbon dioxide are known to enhance the toxicity 54 of poisons. The alterations in the latter two factors will be discussed in the next section. CARBON DIOXIDE, pH AND ALKALINITY Carbon dioxide, pH and alkalinity are interdependent chemical properties of natural waters which are predominant factors in determining the composition of the aquatic com- munity. A change in one of these chemical properties is usually reflected in an alteration of the others. Any rapid increase or decrease, particularly in the pH and carbon dioxide can deleteriously effect the organisms of the aquatic environment. Ordinarily such drastic changes seldom occur, however, a disturbance such as the addition of an herbicide can produce such undesirable consequences. The addition of a herbicide to an aquatic ecosystem initiates a series of biological responses which alter the chemical composition of the water. Of the various chemical changes that occur, the increase in free carbon dioxide is of prime importance. Besides altering the pH and alkalinity of the water, excessive free carbon dioxide influences the respiration of fishes and other aquatic animals directly by reducing the affinity of blood for oxygen (Welch, 1952). In fact, high carbon dioxide tensions may result in the suffocation of fishes even though ample oxygen is available (Prosser and Brown, 1961). Although carbon dioxide increases may not always be severe enough to induce mortality, lesser 55 56 concentrations could possibly produce sufficient physio- logical stress to retard growth and/or impair the fertility of the aquatic fauna. Free carbon dioxide is produced as a waste product dur- ing the utilization of dead organic matter by decomposer organisms. All of the aquatic organisms which succumb to the toxicity of the herbicide immediately become available to the decomposers. During decomposition under aerobic condi- tions, 61 to 67 per cent of the carbon in algae and net plankton were found to be released as carbon dioxide within 51 days (Allgeir, R. J., Peterson, W. H., and Juday C., 1954). Once the free carbon dioxide is released into the water, some or all of it combines immediately with monocarbonates to form bicarbonates (Welch, 1952). In the absence of carbonates, carbon dioxide reacts with water to form carbonic acid (op. cit.). Consequently, the fate of carbon dioxide depends upon the chemistry of the aquatic environment. Chemical determinations made on the Lake City ponds in the undisturbed condition (summer, 1961) showed the waters to be moderately hard and alkaline. Total alkalinity values ranged between 55 and 75 ppm, and the pH varied between 8.0 to 9.0. Under these alkaline conditions little free carbon dioxide (less than 1 ppm) is found, instead it exists in the combined form of half—bound bicarbonates (HC03') and bound carbonates (C03=). Similar chemical conditions existed in the ponds during the summer of 1962, prior to the herbi- cide treatment. 57 Within 24 hours after the herbicide treatments, the water chemistry of the ponds began to exhibit changes. Carbon dioxide and alkalinity concentrations increased and the pH decreased. These changes are readily recognizable from Figures 5 and 4. Changes in pH and alkalinity were due to the increase in free carbon dioxide released from the decomposing organic matter. As mentioned previously, free carbon dioxide reacts immediately with carbonates to form bicarbonates. As long as the carbonates are available, the formation of carbonic acid is prevented and the pH of the water remains unchanged. Hence, the carbonates serve as a buffer substance; one which resists changes in pH. However, the buffer action is limited by the concentration of carbon- ates. Once the carbonates are depleted and free carbon dioxide continues to be released, a change is inevitable. The pH changes exhibited by the test ponds indicated that the carbonate concentrations were insufficient to buffer the water during the duration of carbon dioxide release from the decaying plants. In fact, the water chemistry data show that the carbonate alkalinity was completely absent from the water in pond C within 24 hours after the herbicide was added. Ponds A and B are assumed to have reacted similarly although alkalinity determinations were not made on the day following herbicide addition. Nevertheless, the carbonate alkalinity disappeared completely from these ponds within three days after the herbicide additions. 58 .Nmma wo HmEESm mnu mafinsp mpaom muflo mxmq mnu mo mmSHm> mpHCHmeHm mwmcoflumofln paw mumsoflgmu .m mndmflm 59 m musmflm mwumcoflumoflm WHHH mmumconnmu “ER HmQEmummw umsmsm mash ma 0a m on mm ON ma 0.“ m on mm om m.“ OH m on _dufid-du-uj_q‘4——-_._Jfi..—_.-—.~qq-_—‘44111441d.dq-«jujdqqaj—d«an—dqu-1¢¢q¢< O m pcom . : . m is. . . om o \~‘ SS} momwsmztomso lro om. Waif__flev___~___;a_q_fi.___;_%__;+#_.L1H__%4_—9__a___fi_a__L_.4__4a._____._.._.L_fimmw 4 pcom . .0m 0% \ i oh .om BIA? .Ahs&&wwansfip .kfinNNSV \SSRSO momfimz .____.___4__..__ ___._____.:__._ I _____J_._”_____4_._JJ._:J___L..4._____4__:1_.._.._._.__.___l_:1TO 0 com . Joe om \\\\.\\\§H MW 1 fl om . om «Omao uoTIIIm Jed salsa 4O .mmmfi mo umEEdm mzu mCHHSU mpcom ummp m3» Mo .Emm H cmnp Hmpmmum mCOHumuucmocou mpflxoflp coflumu mmum paw Ammv soflumuucmucoo COH cmmoupmm .w musmflm 41 Ppm carbon dioxide ¢ musmflm .Emm a many umumunm .osou NOU_S§E mm vllt Hmnfimummm umsmsd >H5b mash OH m Om. mm Om. ma O.“ m on mm Om m.“ OH m on . q‘€“ “‘ .—_+.4_..—._—dqufiq_—__..qu—u.._-...._..‘. \ . 0d Hm ._ , ® .3 m psom . . m0m4m2+¢omso Om.s_.___:____________]__:____22321..lgfisjljija:_.Z__..4gm mm. . Oar Lm {Ev D a. ANUCOAH Nomfiz owc.n. _____fia___s_w_kmnhfinnnfifinumwwwus______.m«‘.nwnmmmwummuwu$_________.____. m O OH . m © as U Ucom somso satun Hd 42 The reduction in pH which followed the depletion of carbonates was also accompanied by an increase in carbon dioxide. Free carbon dioxide concentrations (calculated from Moore's nomograph) in all ponds exceeded 1 ppm within three days after the herbicide treatments. A minimum pH of 7.5 and a maximum carbon dioxide concentration of 7.0 ppm were recorded in pond B. The pH and carbon dioxide changes of the other ponds are easily recognizable, but not as pronounced as in pond B. Of all chemical factors measured, the bicarbonate alkalinity exhibited changes which were greatest in magni- tude and longest in duration. Post-treatment maximum values exceeded the pretreatment values by 21 ppm in pond B, 14 ppm in pond A, and 12 ppm in pond C. The initial post-treatment increase was attributed to the conversion of carbonates to bicarbonates. However, it has been mentioned that the carbonate alkalinity was completely depleted from the pond waters within three days after the herbicide additions. And yet the bicarbonate alkalinities continued to rise for as long as eight days thereafter. Evidently a source of carbonates other than that contained by the water was being utilized. Since the only other available carbonates were contained in the bottom muds and plant encrustations, the reactions of the carbon dioxide and carbonic acid must have included these sources. 45 The most severe changes in the water chemistry of the ponds took place within two weeks after the herbicide applications. The time required to return to normal varied between ponds and between the various chemical factors. By the end of the summer, the carbon dioxide and pH levels of ponds A and B had attained the pretreatment range of values. Contrarily, the pH in pond C remained lower than the pretreatment values for the remainder of the season. In addition, the carbon dioxide levels of this pond continued to exhibit sporadic increases above 1 ppm. PHYTOPLANKTON Because of the non—specific action of copper sulfate and sodium arsenite, treatment to eliminate macrophytes may result in the elimination of phytoplankton and periphyton as well. Most heterotrophic forms of life in lentic systems depend directly or indirectly upon the phytoplankton and periphyton for their existence. Since these producers com- prise an important segment of the food chain, any prolonged decrease in their production would be reflected in the pro- duction at all trophic levels. Although copper sulfate and sodium arsenite exhibit algacidal properties, their long-term effects upon the phyto- plankton populations are not fully understood. During this study, the phytoplankton productivity was followed from the time of herbicide application until complete recovery was attained. Post-recovery productivity estimates were con- tinued to ascertain whether the productivity was increased by the release of nutrients from the decaying macrophytes. Immediate Response of Phytoplankton Productivity to Herbicides In order to detect the immediate influence of herbicide toxicity upon phytoplankton productivity, the routine C14 method was modified on the dates of herbicide application. 44 45 Instead of exposing the samples for an entire day, three series of exposures were placed in the pond to be treated for the following time periods: 0600-1000, 1000-1400 and 1400-1800. In each case, the herbicides were applied immediately preceding the 1400-1800 series. The phytoplank- ton productivity values for the days of herbicide application appear in Table 4 along with the incident light energy values of the exposure periods- All of the ponds exhibited a greatly reduced productiv- ity immediately after the introduction of the herbicides. These reductions were too drastic to be caused by natural limiting factors. Light is normally considered as the most important limiting factor to phytoplankton production. However, the sunlight energy values on the herbicide appli- cation dates show that no severe decrease in sunlight energy followed any of the herbicide applications. On the contrary, all of the sunlight energy values for the afternoon exposures were greater than those for the mornings. Yet, a tremendous difference in productivity is shown between the two periods. These data seem to indicate that the reduction in phytoplank— ton productivity could hardly have been attributed to low sunlight energy. Other limiting factors could haVe contributed to the reduction in phytoplankton productivity, but this is unlikely. Depletion of an essential nutrient or combination thereof would produce such a reduction, but certainly not within a 46 Table 4. Phytoplankton productivity and solar energy determinations of the Lake City ponds on the dates of herbicide application. Time Gram calories cm‘2 Assimilated Carbon (mg. m‘3) Pond C -- Julyg12L 1962 0600-1000 187.68 18.22 1000-1400 257.55 50.67 1400-1800 225.08 2.12 Pond A -- July 20Lg1962 0600-1000 22.7 7.67 1000-1400 55.5 15.44 1400-1800 50.6 5.96 Pond B -- August 5, 1962 0600-1000 125.70 10.06 1000-1400 252.95 18.21 1400-1800 150.65 2.10 47 few hours. Furthermore, the phytoplankters probably main- tain a reserve supply of nutrients which would sustain them for at least several hours after such a mishap. Although changes in water chemistry, particularly pH, could induce such an effect, no such change was detected. More- over, the change which reduced the phytoplankton production gives all indications of being immediate rather than gradual. Since the only radical change that occurred on these days was the addition of herbicides, any anomalous disturbance of the productivity must have been due to the herbicide toxicities. Additional evidence of phytoplankton inhibition is provided by the post—treatment productivity measure- ments. Copper Sulfate Copper sulfate is a very potent algacide. As little as .5 to .5 ppm will control undesirable plankton blooms, but concentrations as high as 1.5 ppm are required to con- trol the macroscopic alga, 92233, If low concentrations are inhibitive to phytoplankton growth, how much more ex- tensive is the inhibition when high concentrations are used? Smith (1940) reported that 5 ppm of COpper sulfate added to soft water lakes inhibited the phytoplankton population for over a year. Moyle (1949) implied that the phytoplankton populations of lakes are not seriously inhibited by standard concentrations of copper sulfate. However, most of the 48 reports dealing with the influence of copper sulfate upon phytoplankton have been casual observations; no specific production studies were conducted. The decline of phytoplankton productivity is quite evident in Figure 5. The pretreatment productivity values averaged 28.2 mg assimilated carbon/Ms/day, but decreased to 2 mg on the day following the copper sulfate treatment. This depressed condition continued for nine days before the pretreatment range of productivity values was once again attained. After recovering from the copper sulfate toxicity, the phytoplankton population maintained normal productivity except on two occasions, August 6th and 15th (Figure 5). On these dates the productivity decreased tremendously. In fact, the magnitude of decline in productivity was similar to that which took place during herbicide inhibition. At least one of these erratic values may be attributed to a physical change in the pond waters. After the dead Chara had decomposed, the waters of pond C became highly colored and turbid. This condition is assumed to have arisen from the decomposition products released from the plants. But regardless of the source, these factors were responsible for attenuating the sunlight in deeper areas of the pond. On one of the days of low production (August 15th), the incident sunlight energy was only 108.6 gram calories. It is assumed that the attenuation of the already low 49 .Nmmd wo umEESm may mcflusp mpcom ummu m3» m0 mufl>fluosponm COpxcmHmouxnm .m musmflm 50 m musmHm Hmflfimymmm umsms¢ wHSO @355 OH m on mm Om mH OH mm on mm ON mH OH m Om _._._,.AJ._..:;_.:.Haaa.___:.1:___. 44413.2:H::_..j+H«_..HsslH:n:_: l m UCOQ 4 momsmz U + aomso H 4 Ucom - D P NOmsmz . 1HJ‘..qa—JH_.1—Jj_.flq—fiu.-aa.— j4u-5fiddqdqqdl-qfi—dul—uu_a .q—afi—Jfiq_djfluaadL O pcom sow:o ON 0% Om Om OOH ON 0% Om Om ON 0% OO (I_£ep/8_m/6m) peqextmrsse UOQIEO 51 incident light resulted in sub-optimal light conditions at the level of Cl4 exposure. Hence, the extreme re- duction in assimilated carbon. Since the other ponds did not develop the turbid conditions, they did not react similarly. The other low productivity value shown in pond C (August 6th) cannot be explained as above for the light on this date was normal. Sodium Arsenite Little is known about the influence of sodium arsenite upon the phytoplankton even though its algacidal properties were demonstrated over thirty years ago. Surber (1951) warned against exceeding the recommended concentration of 2 ppm because of its toxicity towards microscopic plants and animals. Dupree (1960) discovered that plankton concen- trated tremendous amounts of arsenic from waters treated with 4 ppm sodium arsenite. Further studies on the algacidal properties of this herbicide dealt mainly with filamentous pond scum algae, although a few general comments on the reactions of the phytoplankton were usually included. But that is about the extent of information available. Phytoplankton productivity decreased following the sodium arsenite application (Figure 5), but the decrease was not as great as that produced by copper sulfate. In addition the phytoplankton recovered sooner from the toxic effects of sodium arsenite. Only six days elapsed before 52 the pretreatment range of productivity values were attained. The slight reduction in productivity and short recovery time demonstrates the resistance of algae toward sodium arsenite toxicity. After recovering from the toxicity of sodium arsenite, the phytoplankton productivity in pond A gradually decreased until a seasonal low of 8.2 mg carbon/m3/day was recorded on August 6th. From this date until the end of the summer the productivity showed a marked increase. The gradual de- cline in productivity was assumed to be part of a seasonal cycle and not associated with herbicide inhibition. Copper Sulfate and Sodium Arsenite The phytoplankton population of pond B reacted to the combined toxicities of sodium arsenite and copper sulfate by exhibiting post-treatment productivity reductions. But un- like the responses shown in the other ponds, those in pond B were highly erratic. Pond B exhibited an initial decrease like the other ponds, but only three days after herbicide application productivity increased abnormally (Figure 5). Productivity values again decreased, but repeated the unusual increase three days later, whereupon the anticipated levels were finally attained permanently. Judging from the post- treatment responses of the other ponds, the erratic values observed could hardly have been so great, eSpecially since there were two algae inhibitors present. The only possible 55 explanation for the extreme increases is human error. This error could have occurred during the processing of the Cl4 samples, and the following factors are suspected: (1) failure to rinse the Millipore Filters of inorganic C14 during processing, or (2) contamination of the filter with inorganic C14 during or after processing. If the data suspected of being in error are deleted, then the period of reduced phytoplankton productivity caused by the addition of the herbicide combination lasted ten days. Post-treatment Response to Increase in Nutrients Treatment of a pond or lake with herbicides usually results in a post-treatment production decrease followed by a tremendous increase (Mackenthun, 1955; Moyle, 1949). These increases are initiated by the release of nutrients from the decaying macrophytes and other primary producers which succumb to herbicide toxicity. In order to detect a possible increase in productivity, the pretreatment pro- ductivity values were compared to the post-recovery values (exclusive of the inhibited period). The comparisons for the three ponds are shown in Table 5. No significant dif- ferences were shown between the pretreatment and post- recovery values of pond B or pond C. A significant differ- ence did exist between these values in pond A, but was exactly opposite the results expected; the pretreatment 54 Table 5. Comparison of phytoplankton productivity of the pretreatment and post-recovery periods using the Student's t-test. Pretreatment* Post-recovery Dates Standard Dates Standard Pond included Mean error included Mean error 6-25-62 7—20-62 C to 28.18 5.55 to 40.76 4.42 0 7-11-62 9-9—62 6—25-62 7-26-62 . A to 56.55 5.20 to 29.71 4.60 X 7-17-62 9-9-62 6-26-62 8-15-62 B to 48.70 6.00 to 50.58 8.51 0 7-17-62 9-9-62 *- Productivity measurements in milligrams carbon assimilated per cubic meter per day. X = significantly different at the .05 level. 0 = no significant difference at the .05 level. 55 productivity was greater than that of the post-treatment period. From these data, no evidence of a post—treatment bloom is shown. The lack of response to the nutrient increase merits further discussion. Evidence that the nutrients, in fact, were released was shown by the increases in periphyton productivity in all test ponds. These results will be dis- cussed in the next section. However, the periphyton data offer proof that nutrients were not concentrated near the bottom sediments and unavailable to the phytoplankton, for the periphyton substrates were exposed only 15 inches below the surface. Any nutrients available at this depth would certainly be available for phytoplankton also. But for some reason the phytoplankters were unable to utilize these nutrients. A possible explanation of this phenomenon may reside in the reaction of these unicellular algae to intense sunlight conditions. Knight, Ball and Hooper (1962) observed low phyto- plankton productivity values in the Lake City ponds during the summer of 1960, and attributed the reductions to photo- inhibition effects produced by extended exposure to criti- cal light intensities. Supra-optimal light intensities can lead to decreasing photosynthetic activity and loss of chlorophyll. The phenomenon of photoinhibition has been observed in marine phytoplankton and is discussed in detail by Steemen Nielsen (1952). 56 Both, Steemen Nielsen's (loc. cit.) and Rhode's (1958) data, show that the greatest amount of inhibition occurs in the first meter below the ocean's surface. However, the phytoplankton in the ocean are not constantly subjected to the deleterious intensities for they are carried by the currents to more optimal light conditions at lower depths. In the test ponds, which average 1 meter in depth, the phyto- plankters are constantly exposed to intense sunlight without relief. If this inhibitive effect persisted during the post-treatment period, the phytoplankton could not use the nutrients even though they were available. PERIPHYTON Although more emphasis is usually placed upon the production of phytoplankton in lentic environments, the role of the periphyton producers cannot be ignored. This is particularly true in shallow lakes and ponds where the lit- toral zone encompasses all depths (Wetzel, 1964). In such systems, a major portion of the total pond production is due to the microscopic sessile algae. The greatest significance attached to these algal forms is the amount of organic matter contributed to the higher trophic levels as food. In fact, Lauff (1959) stated that a direct correlation exists between the periphyton and fish production in many lakes. If this is true, then a long term inhibition of the periphy- ton producers would ultimately be accompanied by a reduced production at all trophic levels. Despite the fact that periphyton, like all other primary producers, are susceptible to the toxicity of herbi- cides, no information is available concerning the extent and duration of inhibition initiated by these toxic sub— stances. Periphyton productivity is normally expressed as milligrams of carbon produced per unit area per day, and I originally intended to express my data as such. But the 57 58 extreme variability which existed between the subsample weights precluded such treatment. Unequal settling of disturbed bottom detritus upon the horizontally exposed substrates was suspected as the variability inducing agent. Fortunately, phytopigment readings were made on the pig- ment extracts within a few days after the samples were collected. And furthermore, the presence of organic detritus gave no indication of interfering with the phyto- pigment determinations. Since phytopigment concentration has been proven to be closely correlated with the organic weights of periphyton (Grzenda and Brehmer, 1960), these determinations were used to detect the herbicide induced responses of the periphyton communities. Copper Sulfate Because algae are extremely sensitive to the toxicity of copper sulfate, a serious reduction in periphyton stand— ing crop was expected after treatment of pond C. However, this did not occur. Instead, a very slight decrease in standing crop was elicited in response to the treatment (Figure 6). But the magnitude of this reduction could have been deceptive due to the method of measurement. Dead algal cells are known to retain detrital chlorophyll which when measured during pigment analysis contribute to the total chlorophyll concentration and limit the accuracy of the method (Strickland, 1960). The periphyton algae which 59 .HmumEHomp mumsvm Mom muHss psmEmHmouwnm mm pmmmmumxm mpsom ummu map mo mono OCHpcmum couwanumm .m musmHm 60 w musmHm Hmnfimumwm umsms< mHSO OH m Om mm ON mH OH mm on mm ON mH OH m aqdq—HH-fid-‘W-quju—-q—¢#-¢-—qu--d«Ha-<14‘—J1.1—..4-_-‘-uqaquqduauaq-dda o m0mH+¢F>HFA¢HD~Jm ~J®Cfi$WNDDO I NED F30 ENDODQCDGHPODmCDGHDODm I N N 8-25 8-24 9-8 9-10 9-11 9-12 9-13 9-14 UHfiUTPCHOHDODMCDPWU$>$WD~JWCNODN tvDJNPOUHVDDNPOFW4F>PF3H¢00HVF3H 91 Summary of the phytoplankton productivity as measured by the Cl4 method in the Lake City ponds during the summer of 1962. Date Pond A Pond B Pond C Mg.C/M3/day Mg.C/M3/day Mg.C/M3/day 6-25 49.19 19.21 6-26 72.25 42.29 22.74 6-29 51.90 67.11 17.64 7—2 67.70 14.71 26.64 7-5 57.16 46.75 25.47 7-8 52.91 81.89 59.15 7-11 47.15 65.02 26.42 7-12 51.01 7-15 1.96 7-14 58.05 86.24 5.59 7-15 4.45 7-16 6.61 7-17 44.50 57.88 24.12 7-19 16.84 7-20 27.07 20.50 15.22 7-21 29.05 7-22 12.28 55.55 7-25 17.82 54.68 58.76 7-24 50.46 48.81 7-26 42.01 58.55 40.94 7-50 55.74 40.51 45.50 8-2 45.60 59.55 59.57 8-5 50.56 8-4 1.81 8-5 2.21 8-6 41.12 102.88 2.21 8—7 4.19 8—9 71.58 8-10 22.76 12.60 40.68 8-15 27.55 26.74 0.26 8-16 15.29 11.89 72.40 8-19 15.80 105.17 28.54 8-22 11.09 54.15 48.86 8-26 8.25 26.90 20.85 8-28 78.76 57.50 72.28 8-51 19.58 65.29 48.95 9-5 50.44 52.72 9-6 50.79 70.15 56.15 9-9 25.06 55.54 55.54 9-15 27.66 41.62 55.48