r *-~" n7"! INTERNATIONAL I 1I H TECHNOLOGY r LJU CORPORATION L r ~ RESPONSIVE TO THE NEEDS OF ENVIRONMENTAL MANAGEMENT 1

DIAGNOSTIC/FEASIBILITY STUDY FOR ONOTA , PITTSFIELD,

i] u L r r I I I I I

• DIAGNOSTIC/FEASIBILITY STUDY FOR ONOTA LAKE, PITTSFIELD, MASSACHUSETTS

PREPARED FOR: THE MASSACHUSETTS DEP DIVISION OF WATER POLLUTION CONTROL

PREPARED BY: I IT CORPORATION ™ 165 FIELDCREST AVENUE EDISON, NEW JERSEY 08837

NOVEMBER 1991

. ENV/L22-frt I

I ABSTRACT I A Diagnostic/Feasibility Study of Onota Lake was conducted by IT Corporation I for the City of Pittsfield, Massachusetts. The purpose of the study was to identify factors which have contributed to the degradation of the lake, quantify their impact, and recommend strategies for their mitigation. Field I sampling took place from March, 1986 through March 1987.

I Problems identified during the course of the study included: I 1) Excessive growth of aquatic macrophytes 2) Severe dissolved oxygen depletion in the hypolimnetic waters 3) Sediment accumulation I 4) Impacted fishery resulting from dissolved oxygen limitations.

I These problems were attributed to urbanization of the watershed, faulty I agricultural practices and stormwater runoff. The management and restoration plan devised for Onota Lake incorporated both I long-term and short-term measures. Long-term watershed management measures were recommended to address the source of the problems in the lake. Included I in the recommendations were sewering, the implementation of more stringent zoning ordinances, and enforcement of soil erosion control ordinances.

I In order to realize an immediate tangible improvement, short-term, in-lake procedures were also recommended. These included weed harvesting, public I education, hypolimnetic aeration, dredging, and modification of the lake's I flow pattern. I I I I ENV/L22-rpt3 I

I TABLE OF CONTENTS

I 1,0 GENERAL DESCRIPTION 1-1 2,0 PRESENT AND HISTORICAL USES 2-1 • 3,0 SOILS AND GEOLOGY 3-1 4,0 MORPHOMETRIC DATA 4-1 • 5,0 LIMNOLOGICAL SURVEY 5-1 ™ 5.1 WATER QUALITY 5-1 5.1.1 Temperature 5-1 1 5.1.2 Dissolved Oxygen 5-2 5.1.3 Nitrogen 5-4 I 5.1.4 Phosphorus 5-5 5.1.5 pH 5-7 • 5.1.6 Alkalinity 5-7 5.1.7 SeccM Disc Transparency • 5-8 • 5.1.8 Chlorophyll 5-8 ™ 5.1.9 Total and Fecal CoHform Bacteria 5-9 5.2 WATER QUALITY - TRIBUTARY SAMPLING 5-10 I 5.2.1 Churchill Brook and Daniels Brook 5-11 5.2.2 Parker Brook 5-12 || 5.2.3, Blythewood Drive - Unnamed Stream 5-13 5.2.4 Storm Sampling 5-14 • 5.3 FLORA AND FAUNA " 5-15 5.3.1 5-15 • 5.3.2 Benthic Colonization 5-25 ~ 5.4 FISHERY 5-26 6.0 MACROPHYTE SURVEY 6-1 • 7.0 SEDIMENT CHARACTERISTICS 7-1 8.0 WASTEWATER DISPOSAL PRACTICES 8-1 | 9.0 ANNUAL HYDROLOGIC BUDGET 9-1 9.1 PRECIPITATION/EVAPORATION 9-1 • 9.2 TRIBUTARY INFLOW 9-2 9*3 OUTFLOW 9-3 • 9.4 GROUND WATER 9-4 • 9.5 HYDROLOGIC BUDGET 9-5

ENV/L22-frt I

I TABLE OF CONTENTS (Cont'd)

I PAGE

10.0 NUTRIENT LOADING AND LAKE TROPHIC STATE ANALYSIS 10-1 •^^^v 10.1 NUTRIENT/SEDIMENT BUDGET 10-1 10.2 CALCULATION OF THE NON-POINT SOURCE LOADS 10-1 1^^^v 10.3 CALCULATION OF THE SEPTIC LOAD 10-3 10.4 INTERNAL REGENERATION OF PHOSPHORUS 10-4 _ 10.5 ATMOSPHERIC CONTRIBUTIONS 10-6 10.6 CALCULATION OF THE ANNUAL NUTRIENT AND SEDIMENT BUDGETS 10-6 1 10.7 LIMITING NUTRIENTS 10-7 10.8 TROPHIC STATE ANALYSIS 10-8 I 10.8.1 Phosphorus Retention/Critical Loading 10-9 10.8.2 Nitrogen Retention/Critical Loading 10-9 I 10.9 ONOTA LAKE, TWO IN ONE 10-12 •• 10.10 TROPHIC STATE CRITICAL LOADING BOUNDARIES 10-15 11.0 RESTORATION/MANAGEMENT PLAN 11-1 _ 11.1 INTRODUCTION 11-1 11.2 DEVELOPMENT OF THE FEASIBILITY ANALYSIS 11-1 1 11.3 MANAGEMENT PLAN 11-2 11.3.1 Watershed Management 11-3 I 11.3.1.1 Stormwater Retention/Detention 11-6 11.3.1.2 Erosion Control 11-7 11.3.1.3 Land Use Ordinances 11-9 I^H 11.3.1.4 Stormwater Management 11-10 11.3.1.5 Product Modification 11-10 11.3.1.6 Sewering and Infrastructure Improvements 11-10 I 11.3.1.7 Diversion and Pretreatment 11-12 11.3.2 In-Lake Restoration 11-12 11.3.2.1 Short Circuiting of Flow 11-13 11.3.2.2 Weed Harvesting 11-18 / 11.3.2.3 Spot Dredging 11-21 11.3.2.4 Aeration 11-22 I 11.3.2.5 Macrophyte Barriers 11-24 ENV/L22-frt 11 I I 11.3.2.6 Algacide/Herbicide Application 11-24 11.3.2.7 Lake Lowering 11-24 I 11.3.2.8 Power Boat Limitations 11-29 11.3.2.9 Low Priority Options 11-30 11.3.3 Recommended Plan 11-30 I 11.3.4 Public Education and Involvement Program 11-32 11.3.5 Institutional Arrangements 11-33 I 11.3.6 Permit Requirements 11-34 11.3.7 Monitoring Program 11-35 I 11.4 ANTICIPATED IMPROVEMENTS 11-38 11.5 SCHEDULE OF ACTIVITIES 11-40 I 11.6 BUDGET 11-40 11.7 ADMINISTRATION OF FUNDING 11-47 I 11.8 ENVIRONMENTAL EVALUATION 11-48 12.0 LITERATURE CITED - 12-1

I APPENDIX A - LABORATORY DATA APPENDIX B - TRIBUTARY SUMMARY DATA I APPENDIX C - STORM EVENT DATA I I I I I I I I I ENV/L22-frt 111 LIST OF TABLES

2.1 SUMMARY OF RECREATIONAL ACTIVITIES ON ONOTA LAKE 3.1 LIMITATIONS OF SOILS FOR DEVELOPMENT 4.1 MORPHOMETRIC DATA OF ONOTA LAKE 4.2 AREA OF ONOTA LAKE SUB-WATERSHED BASINS 4.3 LAND USE PER SUB-BASIN FOR ONOTA LAKE 5.1 WHOLE PLANKTON % COMPOSITION BASIN 1A 5.2 WHOLE PLANKTON % COMPOSITION BASIN IB 5.3 NET PLANKTON % COMPOSITION BASIN 1A 5.4 NET PLANKTON % COMPOSITION BASIN IB 5.5 FISHERY SURVEY RESULTS - 1986 6.1 SPECIES LIST AND RELATIVE ABUNDANCES OF AQUATIC MACROPHYTES COLLECTED FROM ONOTA LAKE 6.2 TOTAL HARVESTABLE BIOMASS AND CHEMICAL CONTENT OF MACROPHYTON IN ONOTA LAKE 7.1 SURFACE AREA AND VOLUME OF MATERIAL TO BE DREDGED FROM ONOTA LAKE 7.2 ONOTA LAKE SEDIMENT SAMPLE ANALYSES 8.1 RESULTS OF SEPTIC SNOOPER SAMPLES 8.2 SUMMARY OF RESPONSES TO ONSITE WASTEWATER DISPOSAL SYSTEM 9.1 NOAA AVERAGE AND MEASURED PRECIPITATION RECORDED FOR ONOTA LAKE 9.2 TRIBUTARY-INFLOW CONTRIBUTION 9.3 VOLUME OF PRECIPITATION AND VOLUME OF LOSS FROM THE SURFACE OF ONOTA LAKE 9.4 FLOW/STAFF GAGE RELATIONSHIP FOR THE OUTFALL (SPILLWAY) OF LAKE 9.5 RESULT OF LINEAR REGRESSION CONDUCTED ON FLOW/STAFF GAGE DATA RECORDED ONOTA LAKE 9.6 GROUND WATER RECHARGE TO TRIBUTARIES OF ONOTA LAKE 9.7 HYDROLOGIC BUDGET OF ONOTA LAKE 10.1 LOADING COEFFICIENTS FOR VARIOUS LAND USE ACTIVITIES IN THE ONOTA LAKE WATERSHED 10.2 SUMMARY OF NON-POINT SOURCE NUTRIENT AND SEDIMENT LOADING TO ONOTA LAKE ANNUAL LOAD PER SUB-BASIN 10.3 RELATIVE ATMOSPHERIC CONTRIBUTIONS OF NITROGEN AND PHOSPHORUS TO QNOTA LAKE 10.4 NUTRIENT/SEDIMENT BUDGET OF ONOTA LAKE 10.5 PHOSPHORUS RETENTION ONOTA LAKE 10.6 TROPHIC STATE ANALYSIS OF ONOTA LAKE USING THE CRITERIA OF DILLON (1974)

ENV/L22-frt 1v I

I LIST OF TABLES (Continued) I 10.7 NITROGEN RETENTION ONOTA LAKE 10.8 TROPHIC STATE ANALYSIS OF ONOTA LAKE USING THE TN CRITERIA OF BAKER, ET AL. (1985). • 10.9 PERTINENT HYDROLOGIC AND MORPHOMETRIC CHARACTERISTICS OF ONOTA LAKE'S NORTH AND SOUTH BASINS • 10.10 ANNUAL TOTAL PHOSPHORUS LOADING TO THE NORTH AND SOUTH BASINS OF ™ ONOTA LAKE . 10.11 PHOSPHORUS RETENTION AS CALCULATED INDEPENDENTLY FOR ONOTA LAKE'S | NORTH AND SOUTH BASINS 10.12 LAKE TROPHIC STATE AS PER DILLON (1974) CRITERIA CALCULATED INDEPENDENTLY FOR ONOTA LAKE'S NORTH AND SOUTH BASINS. • 10.13 CALCULATION OF CRITICAL LOADING BOUNDARIES BETWEEN OLIGO-EUTROPHIC CONDITIONS • 11.1 WATERSHED MANAGEMENT FEASIBILITY MATRIX 11.2 IN-LAKE RESTORATION FEASIBILITY MATRIX 11.3 PREDICTED POST-RESTORATION TROPHIC STATE FOR THE NORTH AND SOUTH BASINS 1 OF ONOTA LAKE 11.4 ANTICIPATED COSTS OF THE ONOTA LAKE RESTORATION PROGRAM

_ ENV/L22-frt I

I LIST OF FIGURES I 1.1 SITE LOCATION MAP 3.1 SOIL ASSOCIATION MAP 4.1 WATERSHED AND SUB-BASINS OF ONOTA LAKE 4.2 BATHYMETRY MAP • 5.1 LOCATION OF IN-LAKE SAMPLING STATIONS 5.2 TEMPERATURE ISOPLETH FOR IN-LAKE #1 MAIN BASIN • ™ 5.3 TEMPERATURE PROFILE WINTER IN-LAKE #1 MAIN BASIN 5.4 TEMPERATURE PROFILE SPRING IN-LAKE #1 MAIN BASIN • 5.5 TEMPERATURE PROFILE SUMMER IN-LAKE #1 MAIN BASIN 5.6 TEMPERATURE PROFILE FALL IN-LAKE II MAIN BASIN j| 5.7 TEMPERATURE PROFILE WINTER IN-LAKE #2 SHALLOW BASIN 5.8 TEMPERATURE PROFILE SPRING IN-LAKE #2 SHALLOW' BASIN • 5.9 TEMPERATURE PROFILE SUMMER IN-LAKE #2 SHALLOW BASIN 5.10 TEMPERATURE PROFILE FALL IN-LAKE #2 SHALLOW BASIN • 5.11 DISSOLVED OXYGEN PROFILE IN-LAKE #1 WINTER ™ 5.12 DISSOLVED OXYGEN PROFILE IN-LAKE #1 SPRING 5.13 DISSOLVED OXYGEN PROFILE IN-LAKE #1 SUMMER • 5.14 DISSOLVED--OXYGEN PROFILE IN-LAKE #1 FALL "5.15 DISSOLVED OXYGEN PROFILE IN-LAKE #2 WINTER | 5.16 DISSOLVED OXYGEN PROFILE IN-LAKE #2 SPRING 5.17 DISSOLVED OXYGEN PROFILE IN-LAKE #2 SUMMER • 5.18 DISSOLVED OXYGEN PROFILE IN-LAKE 12 FALL 5.19 AMMONIA VS. TIME IN-LAKE #1 MARCH 86 - FEB 87 5.20 NITRATE VS. TIME IN-UKE #1 MARCH 86 - FEB 87 I 5.21 AMMONIA VS. TIME IN-LAKE #2 MARCH 86 - FEB 87 5.22 NITRATE VS. TIME IN-LAKE 12 MARCH 86 - FEB 87 I 5.23 TOTAL PHOSPHATE VS. TIME IN-LAKE #1 MARCH 86 - FEB 87 5.24 TOTAL PHOSPHATE VS. TIME IN-LAKE #2 MARCH 86 - FEB 87 I 5.25 PH VS. TIME IN-LAKE #1 MARCH 86 - FEB 87 5.26 PH VS. TIME IN-LAKE #2 MARCH 86 - FEB 87 • 5.27 SECCfil DEPTH AND TOTAL CHLOROPHYLL VS. TIME IN-LAKE II 5.28 SECCHI DEPTH AND TOTAL CHLOROPHYLL VS. TIME IN-LAKE #2 I I ENV/L22-frt v1 I

I LIST OF FIGURES (CONT'D)

I 5.29 BACTERIA COUNTS - IN-LAKE #1 5.30 BACTERIA COUNTS - IN-LAKE #2 • 5.31 BACTERIA COUNTS TRIB #1 DANIELS BROOK 5.32 BACTERIA COUNTS TRIB #2 CHURCHILL BROOK • 5.33 BACTERIA COUNTS TRIB #3 DAN CASEY ROAD ™ 5.34 BACTERIA COUNTS TRIB #4 CAMP WINADU 5.35 BACTERIA COUNTS TRIB #5 PARKER BROOK I 5.36 BACTERIA COUNTS TRIB #6 BLYTHWOOD DRIVE 5.37 BACTERIA COUNTS ONOTA BROOK | 5.38 TOTAL BACTERIA VS TIME TRIB #1 DANIELS BROOK 5.39 CHLORIDE VS TIME TRIB #1 DANIELS BROOK • 5.40 TSS VS TIME TRIB II DANIELS BROOK 5.41 NITRATE VS TIME TRIB #1 DANIELS BROOK 5.42 TOTAL PHOSPHATE VS TIME TRIB #1 DANIELS BROOK • 5.43 TOTAL BACTERIA VS TIME TRIB #2 CHURCHILL BROOK 5.44 CHLORIDE VS TIME TRIB #2 CHURCHILL BROOK • 5.45 TSS VS TIME TRIB #2 CHURCHILL BROOK 5.46 NITRATE VS TIME TRIB #2 CHURCHILL BROOK 5.47 TOTAL PHOSPHATE VS TIME TRIB #2 CHURCHILL BROOK 5.48 TOTAL BACTERIA VS TIME TRIB #3 DAN CASEY ROAD 5.49 CHLORIDE VS TIME TRIB #3 DAN CASEY ROAD 5.50 TSS VS TIME TRIB #3 DAN CASEY ROAD 5.51 NITRATE VS TIME TRIB #3 DAN CASEY ROAD 5.52 TOTAL PHOSPHATE VS TIME TRIB #3 DAN CASEY ROAD 5.53 TOTAL BACTERIA VS TIME TRIB 14 CAMP WINADU 5.54 CHLORIDE VS TIME TRIB #4 CAMP WINADU 5.55 TSS VS TIME TRIB #4 CAMP WINADU 5.56 NITRATE VS TIME TRIB 14 CAMP WINADU 5.57 TOTAL PHOSPHATE VS TIME TRIB #4 CAMP WINAOU • 5.58 TOTAL BACTERIA VS TIME TRIB #5 PARKER BROOK 5.59 CHLORIDE VS TIME TRIB #5 PARKER BROOK 5.60 TSS VS TIME TRIB #5 PARKER BROOK i 5.61 NITRATE VS TIME TRIB 15 PARKER BROOK • 5.62 TOTAL PHOSPHATE VS TIME TRIB #5 PARKER BROOK i ENV/L22-frt I

I LIST OF FIGURES (CONT'D)

I 5.29 BACTERIA COUNTS - IN-LAKE #1 5.30 BACTERIA COUNTS - IN-LAKE #2 • 5.31 BACTERIA COUNTS TRIB #1 DANIELS BROOK 5.32 BACTERIA COUNTS TRIB #2 CHURCHILL BROOK • 5.33 BACTERIA COUNTS TRIB #3 DAN CASEY ROAD ™ 5.34 BACTERIA COUNTS TRIB #4 CAMP WINADU 5.35 BACTERIA COUNTS TRIB #5 PARKER BROOK I 5.36 BACTERIA COUNTS TRIB #6 BLYTHWOOD DRIVE 5.37 BACTERIA COUNTS ONOTA BROOK | 5.38 TOTAL BACTERIA VS TIME TRIB #1 DANIELS BROOK 5.39 CHLORIDE VS TIME TRIB #1 DANIELS BROOK • 5.40 TSS VS TIME TRIB #1 DANIELS BROOK 5.41 NITRATE VS TIME TRIB #1 DANIELS BROOK • 5.42 TOTAL PHOSPHATE VS TIME TRIB #1 DANIELS BROOK ™ 5.43 TOTAL BACTERIA VS TIME TRIB #2 CHURCHILL BROOK 5.44 CHLORIDE VS TIME TRIB #2 CHURCHILL BROOK 1 5.45 TSS VS TIME TRIB #2 CHURCHILL BROOK _ 5.46 NITRATE VS TIME TRIB #2 CHURCHILL BROOK I 5.47 TOTAL PHOSPHATE VS TIME TRIB #2 CHURCHILL BROOK 5.48 TOTAL BACTERIA VS TIME TRIB #3 DAN CASEY ROAD I 5.49 CHLORIDE VS TIME TRIB #3 DAN CASEY ROAD 5.50 TSS VS TIME TRIB #3 DAN CASEY ROAD • 5.51 NITRATE VS TIME TRIB #3 DAN CASEY ROAD • 5.52 TOTAL PHOSPHATE VS TIME TRIB 13 DAN CASEY ROAD 5.53 TOTAL BACTERIA VS TIME TRIB #4 CAMP WINADU • 5.54 CHLORIDE VS TIME TRIB #4 CAMP WINADU 5.55 TSS VS TIME TRIB #4 CAMP WINADU | 5.56 NITRATE VS TIME TRIB #4 CAMP WINADU 5.57 TOTAL PHOSPHATE VS TIME TRIB #4 CAMP WINADU • 5.58 TOTAL BACTERIA VS TIME TRIB #5 PARKER BROOK 5.59 CHLORIDE VS TIME TRIB #5 PARKER BROOK • 5.60 TSS VS TIME TRIB #5 PARKER BROOK 5.61 NITRATE VS TIME TRIB #5 PARKER BROOK I 5.62 TOTAL PHOSPHATE VS TIME TRIB #5 PARKER BROOK I ENV/L22-frt I i• LIST OF FIGURES (CONT'D) 5.63 TOTAL BACTERIA VS TIME TRIB #6 (BLYTHEWOOD) • 5.64 CHLORIDE VS TIME TRIB #6 (BLYTHEWOOD) 5.65 BACTERIA COUNTS TRIB #6 (BLYTHEWOOD) • 5.66 NITRATE VS TIME TRIB 16 {BLYTHEWOOD) ™ 5.67 TOTAL PHOSPHATE VS TIME TRIB #6 (BLYTHEWOOD) 5.68 PERCENT COMPOSITION OF BENTHOS (NORTH SIDE - SPRING) 1 5.69 PERCENT COMPOSITION OF BENTHOS (SOUTH SIDE - SPRING) 5.70 PERCENT COMPOSITION OF BENTHOS (EAST SIDE - SPRING) | 5.71 PERCENT COMPOSITION OF BENTHOS (NORTH SIDE - SUMMER) 5.72 PERCENT COMPOSITION OF BENTHOS (SOUTH SIDE - SUMMER) • 5.73 PERCENT COMPOSITION OF BENTHOS (EAST SIDE - SUMMER) 5.74 PERCENT COMPOSITION OF BENTHOS (NORTH SIDE - FALL) 5.75 PERCENT COMPOSITION OF BENTHOS (SOUTH SIDE - FALL) 1 5.76 PERCENT COMPOSITION OF BENTHOS (EAST SIDE - FALL) 6.1 MACROPHYTE DISTRIBUTION MAP • 7.1 SEDIMENT REMOVAL DEPTH CONTOUR MAP 7.2 PROPOSED DREDGING AREA | 8.1 POTENTIAL..SEPTIC LEACHATE AREAS 10.1 PREDICTED TROPHIC STATE OF ONOTA LAKE USING THE CRITERIA OF DILLON (1974) • 10.2 PREDICTED TROPHIC STATE OF ONOTA LAKE USING THE CRITERIA OF BAKER, ET AL. (1985) Jj 11.1 PROPOSED FLOW SHORT CIRCUITING PROJECT 11.2 DREDGE.SOIL DISPOSAL SITES • 11.3 WEED BARRIER LOCATIONS PROPOSED IN NORTH BASIN ™ 11.4 PROJECT ORGANIZATION 11.5 TROPHIC STATE AS PREDICTED INDEPENDENTLY FOR ONOTA LAKE'S NORTH AND 1 SOUTH BASINS. 11.6 SCHEDULE OF ACTIVITIES • • • ENV/L22-frt v111 I I I I I EXECUTIVE SUM-WRY i i

i i I

• EXECUTIVE SUMMARY

Lake eutrophlcatlon Is a natural aging process which proceeds over a period of thousands of years. During this process, the lake undergoes a series of successional changes from a clear, clean water body to a marsh and eventually solid land, as a result of erosion and sedimentation of the soils and rocks within the lake's watershed and the associated input of fertilizing nutrients. The combined effect of these processes contributes to the decline i of the lake's water quality and results in the "filling in" of the lake due to the deposition of both silt and organic material (decayed macrophytes, algae, i , etc.) • Anthropogenic activities, such as construction, agriculture, and discharge of sewage effluent accelerates the natural eutrophication process. This • phenomenon, associated with watershed urbanization, is referred to as cultural eutrophication. The eutrophication process, even when subjected to the accelerating effects of cultural eutrophication, still proceeds slowly over • thousands of years. However, the negative effects of nutrient enrichment and _ sedimentation may become apparent within a very short time frame. Initially, | the changes in^.water quality seem minimal, perhaps increased turbidity or a slight greenish hue to the water. Eventually, problems related to the • development of algal scums, the growth of nuisance aquatic macrophytes, unpleasant odors, and fish kills are observed. Under such a scenario, the • lake may degrade from a pristine state to a condition where the aesthetics, recreational potential, and water quality are below acceptable levels within a i few years. _ Nutrients Include Inorganic compounds such as phosphate and nitrate, and | organic, or carbon based materials. As the amount of nutrients entering a lake increases, so does lake productivity. Organic material will stimulate • bacterial activity, and Inorganic material will stimulate the growth and development of primary producers such as phytoplankton, algae and aquatic • weeds. Respiration by bacteria during the decomposition of organic material, and by plants, at night when they are not photosyntheslzlng, may tax the • oxygen content of the lake to a point where more sensitive organisms can no i ENV/L22-frt I

I longer be supported. Although oxygen depletion is Initially observed in the hypolimnion, the deep non-mixed water of the lake, during summer I stratification eutrophication, oxygen depletion (anoxia) may occur even in the I surface waters. There are numerous natural sources of nutrients to a lake. Usually, the • annual contribution from such sources is relatively low and does not promote excessive growth and production of algae and aquatic weeds. Human activities • which disturb the soils and promote erosion, such as improper logging, ™ farming, and construction practices, increase the contribution of fertilizing nutrients to the lake. The discharge of stormwater, municipal or industrial I sewage and septic leakage serve as additional nutrient sources. Increased nutrient contributions stimulate the growth and development of phytoplankton, • algae and aquatic weeds to nuisance levels. At high densities, the aesthetic and recreational attributes of the lake are decreased due to discoloration of • the water and the impairment of swimming and boating. In addition, taste and odor problems may occur. Upon their death, the tissues of the primary • producers settle to the bottom of the lake. Bacterial decomposition of this • material decreases oxygen levels in the lake. Nutrients associated with these tissues initially become chemically bound with the sediments. However, when • environmental conditions are appropr1ate and oxygen 1eve!s are 1ow, these nutrients are released and recycled into the water column and further • stimulate plant growth in subsequent years. I Thus, cultural eutrophication affects a lake as follows: 1) Primary producers increase to nuisance densities and reduce the water I quality, aesthetics, and recreational attributes of the lake. I 2) Death and decomposition of these primary producers contributes to the depletion of oxygen 1n the deep areas of the lake due to bacterial I respiration. The shape of the lake and the volume of the hypolimnion I may exacerbate the oxygen depletion problem. I I ENV/L22-frt 3) The depletion of oxygen shifts the species composition of the fish and other organisms toward that characterized as pollution tolerant. Most of these fish species are of poor harvestability and angling potential.

4) The internal recycling of sediment released nutrients within the lake may contribute additional nutrients further stimulating plant growth.

5) The deposition of sediments transported to the lake as a result of watershed urbanization and organic material associated with 1n-lake productivity contribute to the "filling in" of the lake and its eventual extinction.

By examining key physical, chemical and biological parameters of a water body, the extent of eutrophication, referred to as trophic status, can be measured and quantified. Limnologlsts have historically classified lakes in terms of their productivity. Lakes of low productivity in which no organic matter is sedimented are termed oligotrophic while those of elevated productivity are termed eutrophic. Those lakes which are somewhat productive and are in the stage where organic matter is initially being deposited are referred to as mesotrophic. This classification scheme, although simplistic, provides a means by which, the condition of lakes can be qualified in very general terms. A more detailed assessment of lake trophic status involves measuring nutrient concentrations, water transparency, lake flushing rate and the density of aquatic weeds and phytoplankton.

In order to assess the ecological conditions of Onota Lake, a comprehensive limnologlcal Investigation was conducted. The purpose of the study was to Identify those factors which have contributed to the degradation of the lake, quantify their Impact, and determine what must be done to improve the condition of Onota Lake. In brief:

1) Onota Lake presently receives an annual load of fertilizing nutrients substantial enough to promote the growth of aquatic weeds to nuisance densities throughout the north basin and 1n the shallow backwaters of the south basin.

ENV/L22-frt 2) The majority of nutrients originate from diffuse, nonpoint sources, but improperly functioning septic tanks and a leaking sewer line are responsible for a portion of the annual nutrient load.

3) A large amount of organic matter (dead plant tissue and phytoplankton cells) sinks to the bottom of the lake. Temperature and density differences between the surface and deep water layers physically prevent mixing of the deep water and replenishment of its dissolved oxygen content throughout the summer. As a result, bacterial respiration, associated with the decomposition of the plant tissues, depletes the dissolved oxygen content of the deep water layer. This occurs primarily in the south basin, and to a lesser extent in the north basin.

4) Oxygen depletion in the deep, cool waters affects the fishery of the lake as trout and many other game species require cool, well oxygenated water throughout the year. The deep cool waters of Onota Lake, which amount to a significant percentage of its total volume, are void of oxygen throughout the summer. Valuable habitat needed for the maintenance of a cold water fishery has been lost.

5) During the winter, ice cover coupled with reduced water flow, decreases the opportunity for oxygen exchange in the north basin. The bacterial decomposition of accumulated organic detritus leads to a depletion of oxygen, and could potentially trigger a winter fish kill.

6) Future development in currently forested areas of the lake's watershed, could substantially accelerate the lake's eutrophlcation. Guidelines, ordinances and zoning, which could control or mitigate potential development impacts, must be developed and enacted.

Thus, problems and symptoms associated with accelerated eutrophlcation of Onota Lake have been observed. The Initial effects have been a decrease 1n water quality, the loss of valuable fishery habitat, and Impaired recreational utilization of the lake. The further demise of the recreational attractiveness of the lake must be halted and restoration measures Implemented.

ENV/L22-frt 4 I

I The means by which the accelerated eutrophication of Onota Lake can be mitigated or abated is provided within the context of the I restoration/management plan. The primary goal of this plan is to decrease nutrient and sediment loading. This goal is achievable only through a I comprehensive, well coordinated watershed management program. Such a program must focus on reducing point and non-point source pollution contributions. I Simultaneously, due to the recreational importance of Onota Lake, it is necessary that an intensive in-lake restoration and management program be implemented immediately. The in-lake program must focus on ameliorating I symptoms of accelerated eutrophication such as weed growth, silt accumulation and degradation of the fishery. Through such a program, immediate, user I orientated improvements in lake quality could be realized. Thus, a restoration/management study which encompasses both in-lake and watershed I techniques has been proposed for the long-term improvement of the lake, as I well as short-term symptomatic relief. Such an approach insures that over the long-term, future eutrophication of the water body is better controlled. In addition, it also guarantees that during I the implementation of the long-term program, the aesthetic and recreational attributes of the lake are not compromised any further. The specifics of the I Onota Lake restoration/management plan are discussed within the text of this I report. A summary of the overall scope of the prescribed plan is as follows: I Watershed Management 1) Reduction in loading of septic inputs through sewering of the Blythewood I Drive area and repair of the leaking Pecks Road sewer line. 2) The use of passive treatment devices such as retention basins and catch I basins to decrease the stormwater related contribution of pollutants and I sediments. 3) Enforcemen/ t of existing soil erosion control ordinances in order to I reduce soil transport and the subsequent filling of the lake. This I includes stabilization of existing sources of eroded soils. I ENV/L22-frt I

I 4) Sound land use management to avoid improper development of lake shore properties, environmentally sensitive areas and land adjacent to the I tributaries of Onota Lake. The sub-basins along the lake's southwest shore would benefit from environmentally sensitive ordinances, which I could help avoid improper development of these open areas. I 5) Use of non-phosphorus detergents and low phosphorus fertilizers. 6) Public education through seminars, flyers, etc. to inform the users and I residents of the lake how they themselves can protect the lake. I In-Lake Restoration • 1) The use of mechanical weed harvesters to decrease the density and distribution of aquatic weeds throughout the lake.

2) Short circuiting of flow, through .the creation of a culvert under the Thomas Island causeway. This should help divert nutrient rich waters out • of the north basin of the lake and promote the flushing of certain _ sections which are currently stagnant.

3) Aeration of the hypolimnion of the lake's south basin as a means of • decreasing the annual regeneration of phosphorus from the lake's anoxic sediments during the summer. In the north basin aeration should be • implemented in the winter in order to avoid winter kill.

4) Dredging selected areas of the lake to remove accumulated nutrient rich, I highly organic sediments which presently serve as an internal source of I nutrients for the lake's aquatic macrophytes. 5) Restoration and Improvement of the lake's fishery, particularly In terms I of available habitat. I I I ENV/L22-frt I

I 6) Restriction 1n motor boat speed In certain areas of the lake. This could aid in decreasing the spread of El odea and reduce the opportunity for I nutrient release from shallow littoral sediment currently disturbed by I boat propellers and wakes. The watershed and in-lake management approaches presented above are not prioritized. It would be very effective, however, if the in-lake restoration I tasks begin with the short circuiting of flow. This project would entail the placement of a culvert under Thomas Island Road and sediment removal from I Thomas Island Cove and the small cove north of the causeway. Removal of this material would greatly reduce the organic load and weed densities which I presently pi ague the north basin. With the cu1 vert in pi ace, harvesti ng equipment and winter aeration would not be needed. The remaining macrophyte I problem would probably be best controlled through contract harvesting. Creation of an in-lake sediment catch basin south of Dan Casey Memorial Causeway could also be scheduled so as to coincide with the culvert project I thus making effective use of dredging equipment.

I In conjunction with the restoration/management plan, a public education/ involvement program is highly recommended. In this way, the public will come I to realize that the restoration of Onota Lake will benefit all those who directly and indirectly utilize or rely on the lake. They will be informed of I those practices that can be implemented individually on a voluntary basis to reduce nutrient inputs into the lake. Only through public support and I involvement can the prescribed plan be accomplished. I I I I I I ENV/L22-frt I

I 1.0 ONOTA LAKE - GENERAL DESCRIPTION AND IDENTIFICATION I Onota Lake is located in the northwest sector of the City limits of Pittsfield I and just southwest of Pontoosuc Lake (Figure 1.1). This 250 hectare lake is the largest body of water in the upper Basin. The lake Itself is entirely within the City of Pittsfield but some portions of the I lake's watershed encompass areas of Hancock and Lanesborough. A total of six tributaries flow into the lake, the largest being Parker Brook which drains an I extensive area of the Pittsfield State Forest. The outlet of Onota Lake, Onota Brook, flows southeast into Pecks before entering the West Branch I of the Housatonic River in Pittsfield. I Local climate is characterized as humid with a temperature regime characteristic of the North Temperate Zone. The mean annual temperature is approximately 46°F. Record temperatures are a high of 95°F and a low of -25°F I recorded at the Pittsfield airport. Weather patterns alternate roughly twice- weekly from fair to cloudy or stormy conditions. It is infrequent for any I regular weather pattern to continue for several weeks. Changeability is the most normal characteristic, and a "normal" month, season, or year is indeed I the exception rather than the rule. I Precipitation averages about 117 cm annually, and is rather evenly distributed throughout the year. The average monthly precipitation varies from a low of 6.9 cm in October to a high of 10.9 cm in April. Much of the winter I precipitation falls as snow which averages about 190.5 cm annually. Continuous snow cover lasts on the average from 2 to 3 months depending upon I the elevation. I The topography of the Upper Housatonic River Basin is typical of Berkshire County. The Basin can be divided into a three-part physiographic system I consisting of the valley of the Housatonic River and Its major tributaries, the highland area to the east and the Taconic Range on the west. Elevations range from- a high of 800 meters above sea level at Brodie Mountain 1n Hancock I to a low of approximately 290 meters above sea level near the P1ttsf1eld-Lenox I boundary. The valley area ranges from 290 to 335 meters above sea level. The I ENV/L22-rptl 1-1 FIGURE 1 SITE LOCATION MAP ONOTALAKE PfTTSFIELO. MASSACHUSETTS FT PROJECT NO. 1645 I

I laconic Range and the eastern highland area exhibit relief of about 305 meters I above the valley floor. Residential development around Onota Lake and its immediate watershed is very I sparse. The northeastern section of the watershed is the most densely populated. Specifically, this encompasses the residential areas of Thomas Island and the more northern end of Pecks Road. Some development also exists I along Churchill Street and Blythewood Drive in the western section of the watershed. There are quite a few residences which have been converted from I seasonal to year-round occupancy. The average residence time is 0.895 years or approximately eleven months per year. As is common for the larger lakes in I Berkshire County, there are several summer camps and a city park in the lake's immediate area. Burbank Park, a public access area to Onota Lake, has I undergone extensive improvements as of late including construction of a new boat ramp and fishing pier, designated use areas and planning for a new I pavilion. I I I I I I I I I I ENV/L22-rptl 1-3 2.0 PRESENT AND HISTORICAL USES OF ONOTA LAKE

Onota Lake, once known as West Pond and nicknamed "Lake of the White Deer" has • a long history of recreational utilization. Onota Lake provides for a wide variety of recreational activities including fishing, swimming, boating and • water skiing. Of the numerous lakes and located in Berkshire and Franklin Counties, Onota Lake is one of the few which provides public access in the form of beach, park, boat ramp and fishing pier. Silver Lake, Goodrich • Pond, Morewood Lake, Mud Pond and Pecks Pond, in addition to Onota Lake, are located within Pittsfield City limits. There is no public access to Pecks Jj Pond. Silver Lake, Mud Pond, and Morewood Lake are accessable only to lake residents (MDWPC, 1976). Goodrich Pond does have a public boat ramp, but the • pond is only 13 acres in size. In terms of water quality, all but Onota Lake are classified as eutrophic, and it is the only Pittsfield lake routinely • stocked (MDWPC, 1976). Richmond Pond and Pontoosuc Lake lie partially within ™ Pittsfield's boundaries. Although both water bodies offer public access and are important recreational lakes, they are plagued by aquatic macrophyte • problems and do not offer the recreational potential of Onota Lake. This is particularly true with respect to sport fishing.

The City of Pittsfield has extensively improved public access to Onota Lake. B Since 1968, Burbank Park, located along Lakeway Drive on the lake's eastern shore, has undergone a series of upgrades. Pittsfield officially purchased • the water rights to Onota Lake from the Berkshire Woolen Company in 1971. — Initial improvements consisted of the construction of a fishing pier (1976) and a new boat ramp (1984). To better meet the recreational needs and • interests of lake users, designated use areas have been developed. The objective of these areas 1s to decrease the negative interaction of different | users, such as water skiers, power boaters, sallboarders and swimmers. This has resulted in a more orderly use of the lake and has decreased the potential • for mishap. Burbank Park provides the following user amenities: boat ramp (2 car), fishing pier, on-site parking, swimming beach, picnic areas, sailboard • access, public pavilion, and rest rooms.

_ Lake's and reservoirs are typically Important community focal points. Not

H ENV/L22-rptl 2-1 I I only do they provide aesthetic qualities, but serve as a source of recreational enjoyment. Major recreational waterbodies may even play an I important role in the regional economic base (Souza and Perry, 1977). To establish the recreational use of the lake, a survey was distributed to local I residents. Of 314 questionnaires circulated, 95 were returned (Table 2.1). The survey results indicate that the more popular recreational activities I are: boating, swimming, fishing, and water skiing (Table 2.1). Of those surveyed, 73% have access from their property, while the remainder use the I boat ramp or the public beach. The greatest percentage of boats on the lake are motorized (41%). Other boating includes canoes, rowboats, and I sailboats. Based on the survey, Onota Lake is used primarily from April through October. Of those surveyed, 24% use it weekdays only and 17% use it weekends only. It should be noted though that this survey solicited I information primarily from lake residents. The utilization of Burbank Park, the fishing pier and the boat ramp is probably more greatly skewed toward non- I lake residents. The proximity of the lake to Pittsfield as well as its excellent reputation as a game fishing lake probably results in a larger I number of non-resident and day users than indicated by the survey. Improvements to the park will probably increase the number of transient I users. Informal surveys conducted during water quality monitoring activities, found the lake to be used throughout the year. There were usually individuals I fishing from the pier, even during inclement weather. During the summer, the parking lot near the boat ramp was often near capacity.

I The popularity of the lake is largely a result of the long term stocking and fishery management of the lake. Available data reviewed as part of this study I show that stocking of the lake with game fish dates back to the early 1900's. A 11st of some of the fi.sh historically introduced to the lake are I listed below: I Species Number Date 9,000 1917-1953 Smallmouth-Bass 96,500 1914-1948 I Bullhead 37,500 1916-1945 Pike Perch 4,345,000 1915-1937 I 1,500,000 1915-1945 I ENV/L22-rptl 2-2 I

I Smelt 7,450,000 1903-1937 Brown Trout 17,900 1933-1955 113,000 1902-1951 I White Perch 1,100 1923-1933 Crappie 6,000 1928-1930 Pickerel 1,100 1932-1945 Silver Trout 50,000 1920 I Kokanee Salmon 10,000 1970-1973 I Management techniques have included the application of rotenone in 1957 for the control of non-game fish densities. Creel census data indicate that over 50,000 angler hours per season is not unusual (MDFG, 1972, 1973 and 1978; I Daly, per Comrn, 1987). Interestingly, since 1947 there has been a decline in trout habitat from approximately 40% of lake volume to about 25% (MDFG, 1956 I and 1978). Apparently, this has not adversely affected the sport fishing utilization of the lake. The decrease in aerobic, cold water habitat has I probably influenced the lake's summer holding capacity for trout, but since the lake is aggressively managed, the negative effects appear to have been I minimized in respect to sport fishing pressure and/or attractiveness. I I I I I I I I I I ENV/L22-rptl 2-3 I I I TABLE 2.1 I SUWttRY OF RECREATION ACTIVITIES OH ONOTA LAKE Number Distributed - 314 I Number Returned - 95

I TYPE OF LAKE TYPE OF BOATS DAYS OF LAKE ACCESS TO ONOTA I ACTIVITIES (PERCENT) (PERCENT) USE (NUMBER)* LAKE (PERCENT) Fishing 23% Sail 10% April-Oct. 36 Ramp 6% I Boating 29% In-board 17% Once a Week 1 Home 73% Swimming 27% Row 16% Daily 29 Boat Livery \% Camping 2% Canoe 17% Weekends 17 Causeway 2% I Water Skiing 13% Outboard 24% Year Round 7 Cove 1% Other 2% Other 2% No Response 21 Beach 3% I None 456 None 14% No Response I I I I I I I I I EMV/L22-rptl 2-4 3.0 GEOLOGY AND SOILS OF THE ONOTA LAKE WATERSHED

The Onota Lake watershed lies within the New England Province of the Appalachian Highlands. The watershed encompasses portions of two distinct physiographic regions: the and the Berkshire Valley. The Taconic Mountain Range borders the lake to the west and northwest, comprising most of its drainage area. The Berkshire Valley underlies the immediate vicinity of the lake and nearby Pittsfield. The geology and soils of this region have been described in a publication from the Berkshire County Regional Planning Commission (1978) which forms the basis for much of the following discussion.

The Taconic Range is composed of mostly quartz mica schistose rock with some garnetiferous schist formed 350 million years ago during the Ordovician and perhaps Cambrian Period of the Paleozoic Era. This erosion resistant metamorphic rock contains parallel layers of flakey mica, talc, and possibly iron pyrite and splits easily into thin leaves. This bedrock forms the long valleys which drain several hills of the range down to Onota Lake. These valleys constitute the majority of the lake's drainage area. Elevations along the ridge tops range from 2313 to 2157 feet above mean sea level.

The Berkshire Valley which contains Onota Lake is underlain by carbonate rocks such as limestone, dolomite, and marble. This Stockbridge Limestone is part of the Stockbridge Group of the Ordovician and Cambrian Periods. This relatively level valley with its frequent overlain deposits of glacial material is the product of complex structural relationships and simple scours of the relatively soft limestone. Elevations in the valley range from about 1200 feet at the base of the ridge to the lake surface elevation of 1078 feet.

Each of these types of bedrock contain subsurface water 1n fractures and faults, with the carbonate type typically exhibiting limestone solution cavities. The high porosity of carbonate rocks commonly coupled with overlain porous soils renders the ground water in these wells highly susceptible to contamination. The occurrence of limestone formations in the watershed of Onota Lake may influence the alkalinity and hardness of the lake and its'

ENV/L22-rptl 3-1 I

tributary waters. It may also serve as a natural source of buffering against _ acidic precipitation.

The surficial geology of the Onota Lake watershed can be broadly categorized • into stratified glaciofluvial deposits serving as ground water recharge areas and unstratified glacial till. The lake is largely situated upon and bordered • on the east, west, and south sides by unstratified till. Stratified deposits have accumulated in a broad band at the base of the ridge draining the Taconic Range. The range itself is also overlain with predominantly unstratified • till.

The unstratified till is a poorly-sorted heterogeneous mixture of boulders, stones, sand, silt, and clay in various mixtures formed 25,000 years ago during the Pleistocene epoch of the Cenozoic Era. The limited porosity of • these soils results in slow infiltration and minimal ground water recharge, H with most precipitation lost through surface or lateral downslope subsurface runoff. The lateral subsurface runoff contributes hydrologically to the streams and brooks which drain to the lake.

_ Stratified drift deposits are composed of assorted, relatively homogeneous H materials, predominantly sand with varying amounts of silt and gravel, deposited by waters flowing within or from the glacier. These deposits are I much more conducive to ground water recharge. Precipitation tends to move downward rapidly through the unconsolidated deposits to aquifers. The coarse • nature of these deposits provides minimal filtering action and the aquifers underneath are especially susceptible to soluble pollutants from the surface • (e.g. road salt and septic leachate). • Soils Stratified drift deposits and unstratified till overlay the carbonate rock • surrounding Onota Lake to form a mosaic of soil types (Figure 3.1) which include silty and sandy loams, very stony loams and rock outcrops with steep slopes. Soil depths are also variable and become very shallow in those areas • surrounding the lake that are dominated by rock outcrops. The major soils • surrounding the lake are as follows (SCS, 1983): i ENV/L22-rpti 3-2 I I I I I I I I I I I I I I I

SOURCE: USDA SOL CONSERVATION SERVICE RQURE 3.1 I SOIL ASSOCIATION MAP (1980) SOIL ASSOCIATION MAP FOR ONOTA LAKE I I I "Do NOI Se»»t Trw» Drawing* "^ North - Pittsfield, Hinckley, Warwick series; Taconic-Macomber and Pittstown- • Bernardston Associations Northeast - Groton, Hinckley, Warwick series; Medisaprists, and Gravel Pits. East - Fredon, Hero, Nellis, (and Warwick series;) Rock-Outcrop-Farmington i Complex; Udorthents and Medisaprists. _ Southeast - Amenia, Fredon, Hickley, Nellis, and Pittsfield series; Rock- • Outcrop-Farmington complex; Medisaprists, and Urban Land. South - Amenia, Kendaia, Nellis, and Pittsfield series; Rock-Outcrop- • Farmington complex; Pittstown-Bernardston and Stessing-Mansfield Associations. Southwest: - Hero, Hinckley, Kendaia, and Warwick series; Rock-Outcrop- Farmingdale complex; Pittstown-Bernardston Associations; Gravel Pits, and Medisaprists. • West * Warwick series; Bernardston-Dutchess-Pittstown-Bernardston, and • Taconic-Macomber Associations. Northwest - Warwick series; Pittstown-Bernardston and Taconic-Macomber Associations. AMENIA SERIES (430C, Figure 3.1) - These are moderately well drained soils developed in compact calcareous glacial till that contain many limestone, • schist and phyllite fragments. They have a crumbly fine sandy loam or silt loam surface soil and subsoil. They are underlain at about 24 inches from the • surface by a silt loam hardpan. The permeability is moderately rapid in the surface soil and subsoil and slow in the hardpan. Amenia soils have excess B seepage water or a high water table for 4 to 5 months of the year, generally in late winter and early spring. These soils are very stony or extremely • stony except where they have been cleared of surface stone. They occupy nearly level to steep slopes.

FREDON SERIES (523A, Figure 3.1) - These are poorly drained soils formed in thick deposits of sands and gravel derived largely from limestone, schist, and • phyllite- They have a fine sandy loam surface soil and a fine sandy loam or gravelly fine sandy loam subsoil underlain at a depth of about 2 feet by B layers of sands and gravel. They have rapid or moderately rapid permeability in the surface soil and subsoil and rapid permeability In the substratum. The B water table is at or near the surface for 7 to 9 months of the year. They are generally free of stones and boulders but may contain cobbles. They occupy • depressions and low flat areas. i ENV/L22-rptl 3-4 I I GROTON SERIES (505E, Figure 3.1) - These are excessively drained soils that £ formed in thick deposits of neutral or calcareous sands and gravel. The sands and gravel are derived mainly from limestone, phyllite, schist and gneiss. • They are very gravelly. They commonly have a gravelly sandy loam surface soil and subsoil underlain by stratified sands and gravel. Groton soils are loose • throughout and have a very high permeability. They are usually stone-free but ™ may contain cobbles and a few stones. They occupy level to very steeply M sloping areas.

HERO SERIES (512A, Figure 3.1) - These are moderately well drained soils • formed in thick sand and gravel deposits derived principally from limestone, schist, and phyllite. They have a crumbly fine sandy loam or gravelly fine H sandy loam surface soil and subsoil. They are underlain at a depth of about 24 inches by a substratum consisting of layers of sands and gravel. Hero • soils have a moderately rapid or rapid permeability in the surface soil and ™ subsoil and rapid permeability in the substratum. A seasonal high water table is within 1 % to 2 % feet from the surface for 4 to 5 months of the year, • generally in winter and early spring. These soils are commonly free of surface stones and boulders but may have cobbles. They occupy level areas and j^ gentle slopes.

• HINKLEY-GROTON ASSOCIATION (35-C, Figure 3.1) - These are excessively drained soils developed in thick deposits of sands and gravel derived mainly from • granite and gneiss. They are very sandy and gravelly. They commonly have a ™ sandy loam surface soil and a sandy and gravelly subsoil underlain by stratified sands and gravel. Hinckley soils are loose throughout and water • moves rapidly through them. They are usually stone-free but may contain cobbles and a few stones. They occur on level areas to very steep slopes.

KENDAIA SERIES (432B, Figure 3.1) - These are somewhat poorly drained soils formed in calcareous glacial till derived mainly from limestone, schist, and phyllite. They have a crumbly silt surface soil and subsoil. They have a silt loam, hardpan substratum at a depth of about 18 to 24 inches from the surface. The surface soil and subsoil have a moderately rapid permeability. The hardpan has a slow permeability. The soils are wet at or near the surface

ENV/L22-rptl 3-5 for 6 months of the year due to a high water table or excess seepage water from adjacent higher land. These soils are very stony or extremely stony except where cleared of surface stones. They occupy level areas to moderately sloping areas.

MEDISAPRISTS (ORGANIC) (48, Figure 3.1) - These are very poorly drained bog soils formed in accumulations of organic deposits that are underlain by mineral soil materials. The upper part of the organic material is generally bl ack and has decomposed to such a degree that pi ant remai ns cannot be identified by the unaided eye. Decomposition of the materials in the lower part of the deep organic soils varies from this condition to one of practically no decomposition in which plant remains are rapidly identifiable. Organic soils occupy depressions and potholes. The water table is at or near the surface most of the year. Some organic soils have only 1 to 3 feet of organic deposits over mineral soil materials. In others, the organic deposits are many feet thick.

NELLIS SERIES - These are well drained soils formed in calcareous glacial till derived mainly from dark schist and limestone. They generally have a loam or fine sandy loam surface soil and subsoil and are crumbly to a depth of 30 inches from the surface. Below a depth of 30 inches there is generally a hard or firm loam or fine sandy loam substratum. Nellis soils have moderately rapid permeability in the surface soil and subsoil and moderately slow permeability in the hard, firm substratum. Nellis soils are very stony or extremely stony except where cleared of surface stones. They occupy gentle to very steep slopes.

PITS, GRAVEL (GP, Figure 3.1) - These are areas where the soil material has been removed for fill and other uses. The depth of excavation varies greatly. In most places the excavation consists of an exposure of unweathered sands, gravel and cobblestones.

PITTSFIELD SERIES (4440, Figure 3.1) - These are well drained soils formed in glacial till derived largely from limestone and schist. They have a loam or fine sandy loam surface soil and subsoil that contains moderate amounts of cobbles and stones. They are crumbly to a depth of 30 inches or more from the

ENV/L22-rptl 3-6 surface but may have a hard layer below 30 inches. Pittsfield soils have moderately rapid or rapid permeability in the surface soil and subsoil and moderately slow permeability in the hard layer below 30 inches. Pittsfield soils are very stony or extremely stony except where cleared of surface stone. They occupy gentle to very steep slopes.

ROCK OUTCROP-PITTSFIELD COMPLEX (441C, Figure 3.1) - Shallow to bedrock soils formed in calcareous glacial till. They have a friable, fine sandy loam subsoil. Depth to bedrock is mostly less than 20 inches with many rock outcrops.

UDORTHENTS (520, Figure 3.1) - This land type consists of areas from which soil material has been removed. The depth of excavation is variable. In most areas, all the surface soil and part of the subsoil have been removed. In some areas, all of the surface soil and subsoil have been removed, leaving unweathered soil material exposed. Generally, there are many stones and boulders left scattered throughout the area.

URBAN LAND (52, Figure 3.1) - This land type consists of areas where the soils has been altered or obscured by bui1dings, industrial areas, paved parking lots, sidewalks, roads or railroad yards. These structures cover 75 percent or more of the surface area. Slopes range from nearly level to steep.

WARWICK SERIES (152B-D, Figure 3.1) - These are somewhat excessively drained soils developed in sands and gravel that contain many dark, fine grained, phyllite fragments. The texture of the surface soils and subsoil is fine sandy loam or loam, but in some places the subsoil is sandy loam. The break between subsoil and the underlying sands and gravel is abrupt and occurs at a depth of about 2 feet. The permeability is moderately rapid or rapid in the surface soil and subsoil and rapid in the sandy and gravelly substratum. Warwick soils occupy level areas to very steep slopes.

BERNARDSTON-DUTCHESS ASSOCIATION (BDE, Figure 3.1) - Bernardston and Dutchess soils are deep and well drained. The Bernardston soils are typically on the lower and less steep slopes and the Dutchess soils are on the steeper and higher parts of slopes. Stones and boulders are approximately 5 to 20 feet

ENV/L22-rptl 3-7 apart. This unit consists of about 50% Bernardston soils, 30% Outchess soils M and 20% other soils.

PITTSTOHN-BERNARDSTON ASSOCIATION (PBDC, Figure 3.1) - Deep, moderately well | drained Pittstown soils and deep, well drained Bernardston soils. The Pittstown soils are typically on the lower parts of slopes or in convex • areas. Stones and boulders are approximately 5 to 20 feet apart and are prominent features in the landscape. This unit consists of about 60% • Pittstown soils, 20% Bernardston soils, and 20% other soils.

STISSING-MANSFIELD ASSOCIATION (SWB, Figure 3.1) - This association consists i primarily of deep, poorly drained Stissing soils and deep, very poorly drained Mansfield soils. These nearly level and gently sloping soils are on the foot slopes and drainage ways of hills and mountains. The Stissing soils are typically on the base of slopes or in slightly concave areas with the Mansfield soils in concave areas or depressions. Stones and boulders, approximately 5 to 20 feet apart, are prominent features of the landscape. This unit consists of about 55% Stissing soils, 25% Mansfield soils, and 20% other soils.

TACONIC-MACQMBER ASSOCIATION (NDE, Figure 3.1) - These are somewhat excessively drained, shallow to bedrock soils developed in glacial till derived mainly from phyllite. These soils have a crumbly silt loam surface soil and subsoil and contain many flat, black phyllite fragments. The depth to bedrock is 0-40 inches. They have a moderately rapid or rapid permeability to the bedrock. Bedrock outcrops range from 10 to more than 100 feet apart. These soils contain many stones and rock fragments except where they have been cleared of surface stones. They occupy gentle to very steep slopes.

Most of the land around Onota Lake exhibits shallow slopes, although there are some isolated areas especially along the northern and eastern shores where the slopes exceed 15 percent. These steep slopes are vulnerable to soil erosion especially when the vegetative cover is removed or disturbed. Steep slopes also predominate in the Taconic Range approximately two miles east. Development limitations for the soil groups due to slope, drainage, strength and excessive frost action are given in Table 3.1.

ENV/L22-rptl 3-8 TABLE 3.1 LIMITATIONS OF SOILS FOR DEVELOPMENT

Foundations for Dwel11ngs Local Roads, Streets Septic Tank SOIL MAP UNITS With Basement Without Basement and Parking Lots Absorption Fields

Amenia stony silt loam Severe: wetness Moderate: wetness, Severe: frost action Moderate: wetness; 8-15* slopes slope slope

Fredon Variant fine Severe: wetness Severe: wetness Severe: wetness, frost Severe: wetness, sandy loam action poor filter 0-3% slopes

Groton & Hinekley soils Severe: slope Severe: slope Severe: slope Severe: poor filter, 25-35* slopes slope

Hero fine sandy loam Severe: wetness Moderate: wetness Severe: frost action Severe: wetness, poor 0-3* slopes filter

Hinckley sandy loam Moderate: slope Moderate: slope Moderate: slope Severe: poor filter 8-15* slopes

Kendaia silt loam Severe: wetness Severe: wetness Severe: wetness, frost Severe: wetness, percs 3-8* slopes action slowly

Medisaprists Severe: wetness, Severe: wetness, Severe: wetness, Severe: wetness low strength low strength low strength, frost action

ENV/L22-rptl 3-9 TABLE 3.1 (Continued)

Foundations for Dwellings Local Roads. Streets Septic Tank SOIL MAP UNITS With Basement Without Basement and Parking Lots Absorption Fields

Nell 1s stony loam Moderate: slope Moderate: slope Moderate: slope, frost Severe: percs slowly 8-15% slopes action

Pittsfleld stony fine Severe: slope Severe: slope Severe: slope Severe: slope sandy loam 15-25% slopes

Quonset gravelly sandy Severe: slope Severe: slope Moderate: slope Severe: poor filter loam 8-15% slopes

Rock-outcrop Severe: depth Severe: depth Severe: depth to rock Severe: depth to rock Farmington Complex, to rockto rock 3-15% slopes Rock-outcrop Severe: depth Severe: slope Severe: depth to rock, Severe: depth to rock, Farmington Complex to rock, slope depth to rock slope slope 15-35% slopes Stockbridge stony Moderate: slope Moderate: slope Moderate: low strength, Severe: percs slowly silt loam frost action, slope 8-15% slopes

Warwick gravelly loam Moderate: slope Moderate: slope Moderate: slope Severe: poor filter 8-15% slopes

ENV/L22-rptl 3-10 I

4.0 MORPHOMETRIC DATA •

Onota Lake is approximately 3.4 km long and 1.0 km wide and has a shoreline H length of 16.3 km (Table 4.1). The two basins which comprise the lake have a total surface area of 250 hectares and a combined volume of 15.98 x 10 m3.

The lake's major tributaries are Churchill Brook, Daniels Brook and Parkers Brook. Inflow from these major tributaries total approximately 7.0 x 106 I m /yr. Lulu and Hawthorne Brooks feed into Parkers Brook above Churchill Street and create a small pond prior to flowing into Onota Lake. A series of | small intermittent streams feed the lake along the southwest shore off of Blythewood Drive. One of these flows fairly continuously and was monitored • throughout this study along with Churchill, Daniels and Parker Brooks. Outflow from the lake is to Onota Brook which in turn feeds Pecks Pond.

™ The watershed surrounding Onota Lake . contains ten sub-watersheds which o encompass 25.7 km (Figure 4.1). The sub-watersheds encompassing the greatest • area are VI, VII, and VIII (Table 4.2). These sub-watersheds are predominately forested and undeveloped (Table 4.3). Although the watershed is I not highly developed, sub-watersheds I, II, IV, IX and X have the greatest • percentage of developed versus undeveloped land. The lake's bathymetric profile was established by conducting a survey of the lake using a continuous recording fathometer equipped with a strip chart. • Readings were taken at 20 to 50 foot intervals along 23 transects. These data were integrated using Goldenware Software SURFER program to produce the • bathymetric contours of the lake (Figure 4.2). The lake consists of two basins, the larger of which is the south basin (170 ha, Table 4.1). Water • depths in this section of the lake reach 20.6 m. The northern basin is smaller, (80 ha) generally more shallow (< 3m) and attains a maximum depth of • only 7.3 m. A shallow sand bar bisects the lake and physically separates the two basins. Water depth along the sand bar is approximately 0.5 to 1.5 m. A 2 to 3 meter deep by 30 meter wide access channel has been cut through the • sand bar to allow for unimpeded boat traffic. This channel also allows for some hydrologic exchange between the two basins. No estimate was developed

L22-rptl 4-1 I

I for water exchange between the north and south basin. It is assumed, based on topography, that the net water movement is from the south basin to the north I basin.

• Other prominant physical features of the lake include a backwater area north of the Dan Casey Causeway, Thomas Island, a 5-10 hectare along the • northwest shore , the f i shi ng pi er and boat 1 aunch , and three smal 1 island/peninsulas located in the south basin.

The primary public access point to the lake is via a fishing pier and a public boat ramp located at the base of Lakeway Drive. The ramp and fishing pier are part of the Burbank Park system which also offers bathing beaches, access for sailboarding, and a public pavilion. Public access and common recreational uses are detailed in Section 2.0.

L22-rptl 4-2 I

I TABLE 4.1 I MORPHOMETRIC DATA OF ONOTA LAKE I Official Name Onota Lake Location Pittsfield, Massachusetts I Lake Area (Total) 250 ha = 620 acres North Basin 80 ha = 200 acres I South Basin 170 ha = 420 acres Lake Maximum Depth 20.6 m = 68.0 ft I Lake Mean Depth 6.4 m = 21 ft Lake Volume 15.98 x 106m3 = 572.5 x 10°f6t3 I Watershed Area 2570 ha = 6350 acres Shoreline Length 16.3 km = 10.1 miles I Maximum length 3.4 km = 2.1 miles I Maximum Width 1.0 km = 0.62 miles I I I I I I I I I L22-rptl 4-3 I I I I I I I I I I I I

FIGURE 4.1 WATERSHED AND SUBBASINS OF ONOTA LAKE I SOURCE: USGS TOPOGRAPHIC MAPS - I P1TTSF1ELD AND HANCOCK I I I "Do Not Scale This Drawing" TABLE 4.Z AREA OF ONOTA LAKE SUB-WATERSHED BASINS

Sub-watershed basin Acres . Hectares

I 385 156 II 214 87 III 64 25 IV 66 27 V 189 77 VI 2137 865 VII 766 ' 310 VIII 1568 635 IX 149 60 X 190 , 77 Lake 617 250

TOTAL 6345 2569

L22-rptl 4-5 I I

N

3 ^ I OUTLET

«m

contours are reported in feet. 0 - 4 ft, contour interval is 2 feet in both basins. ///r I ,—*r--U<#/ For depths greater than 4 ft., interval is 4 ft. in the north basin, w^-^J)K o«i 8 ft. in the south basin. slUc^^dr/ FIGURE 4.2 BATHYMETRY MAP

ONOTA LAKE PITTSFIELO, MASS. JUNE 1987 PROJECT *541648 1384 IT CORPORATION ALL COPYRIGHTS RESERVED Do N01 Scale This Drawing" TABLE 4.3 UNO USE PER SUB-BASIN FOR ONOTA LAKE

LAND USE CATEGORIES PRESENTED AS PERCENT TOTAL AREA RESIDENTIAL Sub-Basin Open High Low Forest Covered* Agricultural Density Density

I. 50 50 II. 30 5 50 15 III. 85 15 IV. 50 50 V. 95 5 vi. 94 1 5 VII. 89.5 .5 10 VIII. 89 1 10 IX. 40 60 X. 20 80

* Open areas with vegetative coyer but no appreciable canopy (parks, pasture, etc.)

L22-rptl 4-7 I

I 5.0 LIMNOLOGICAL STUDY I I 5.1 WATER QUALITY - IN-LAKE SAMPLING In order to establish the existing conditions of the lake and its tributaries and to identify the interrelationships among the physical-chemical-biolcgical I components of Onota Lake, a limnological monitoring program was conducted. The goal of this program was to identify the problems of Onota Lake and I develop, from these data, an effective restoration action plan. I Two in-lake stations were monitored on a semi-monthly basis from March through fall circulation and monthly for the balance of the year. Station #1 was I located in the main basin while Station #2 was located in the shallow basin by Thomas Island (Figure 5.1).

I Samples were collected with a 4.1 liter, non-metallic Kemmerer bottle relative to lake strata. Three water samples were collected during periods of I stratification, 0.5 meters below the surface, near the thermocline and 0.5 meters above the bottom. During the periods when Onota Lake was not I stratified, samples were collected at 0.5 meters below the surface and 0.5 meters above the bottom. All in-situ measurements were taken at 1.0 meter I intervals from surface to bottom irrespective of lake stratification. In addition to the regular water samples, surface grab samples were collected I at each station for total coliform bacteria analysis.

I 5.1.1 Temperature I Based on characteristics displayed in the temperature profiles (Figure 5.2 - 5.10), Onota Lake can be classified as a temperature dimictic lake. Lakes are I classified dimictic if they circulate freely twice a year in the spring and fall and are directly stratified in summer and inversely stratified in winter (Wetzel, 1983). The temperature profile of Onota Lake was monitored at two I locations; Station II (Figures 5.2-5.6) and Station #2 (Figures 5.7 - 5.10). I I ENV/L22-rpt2 5-1 I I The thermocline measured at Station #2 was not as defined as that of Station #1. However, the difference in density resulting from the epilimnetic waters I (< 6 m) is significant enough to impede mixing (Kortman et a!., 1982). I The lake remained ice covered until mid-March. Profiles conducted .in the north and south basin from ice breakup through April revealed water temperatures to be uniform from surface to bottom. By mid-May a thermocline I became established (Figures 5.4 and 5.8).

I Station #1 (South Basin) epilimnetic waters (depth < 12 m) were approximately 8°C during the winter sampling event. Over the course of the spring and I summer, the epilemnetic waters of the south basin increased in temperature, but the hypolimnetic waters remained at approximately 8°C (Figures 5.4 and 5.5). During this period of stratification (April 5 to September 23) the I south basin did not experience major mixing.

I A similar pattern of thermal stratification was observed in the shallower Noth Basin. An apparent thermocline became established in mid-May. The North I Basin did not appear however to stratify as strongly as the South Basin. Bottom waters in the North Basin were in the 11-15°C range, whereas surface I waters ranged from 19 to 25°C (Figures 5.8 and 5.9). I The South Basin became destratified sometime between the October 14 and November 14 sampling events (Figure 5.6). Stratification in the South Basin was pronounced until mid-October. In the North Basin, stratification began to I weaken by August 27th and by September 10th water temperatures were relatively uniform from surface to bottom (Figures 5.9 and 5.10). These data indicate I that autumnal overturn occurs twice in Onota Lake, once in late summer in the I North Basin and approximately 45 days later in the South Basin. I 5.1.2 Dissolved Oxygen Annual dissolved oxygen (DO) fluctuations in Onota Lake are influenced by its dimictic nature. As such, Onota Lake exhibits a positive heterograde curve, I or an increase in oxygen in the metalimnion during stratification. I I ENV/L22-rpt2 5-2 I

I Surface concentrations throughout the year were above 6 mg/1 and followed the temperature related solubilities of oxygen in water. That is, dissolved I oxygen concentration increased with decreasing ambient temperatures (Figures 5.11 to 5.18). At times, concentrations in excess of saturation were I measured. This reflects photosynthetic activity of algae or macrophytes. There was a decrease in oxygen concentration of the hypolimnion as stratifi- I cation of Onota Lake intensified. This is due to bacterial respiration during decomposition of organic sediments, detritus, and senecsent algal cells. The I resulting oxygen deficit is sufficient to cause anoxic conditions to persist I towards the bottom of the lake. Near anoxic conditions were observed as early as June 3rd (Figure 5.12). From I June 25th through October 14th the DO concentration was less than 1 mg/1 at depths in excess of 10 meters (Figures 5.13-5.14). The total duration of I anoxic conditions in the South Basin during the 1986 survey was 99 days. In the North Basin, DO concentrations remained high during January, but I decreased during February (Figure 5.15). Dissolved oxygen concentrations dropped to 1-2 mg/1 in depths > 3m. The depletion of oxygen in these lower I depths was the result of the removal of oxygen by bacterial decomposition in the sediments and the slow diffusion of oxygen through the ice cover. This I depletion could create a serious oxygen deficit and result in a winter fish kill beneath the ice.

I No sampling was done during the month of March in the South Basin. Dissolved oxygen concentrations increased throughout the period of April-May (Figure I 5.16). During spring overturn conditions (April 25 - May 6), DO concentrations remained relatively stable between 10.0 - 11.0 mg/1 (Figure I 5.16). With warming water temperatures in June, 00 concentrations decreased to 8-9 mg/1 (Figures 5.16 - 5.17). Because of depth limitations in the North I Basin, the lower waters occurring at depths <6 m remained well oxygenated during the period of July-September (Figure 5.17). Depths > 6m did exhibit oxygen depletion during the summer period. A limited period of stratification I did occur during the later part of August and persisted through early I September (Figure 5.17). Stratification did not persist as long in the South I ENV/L22-rpt2 5-3 compared to the North Basin, nor was it as strongly developed. Although DO concentrat i ons dec!i ned over the summer, near anoxi c condi t ions were not observed until August. As compared to the South Basin, the period of anoxic conditions, 29 days in 1986, was short.

An increase in dissolved oxygen in the metalimnion was observed during summer stratification at Station 1 between 6 and 11 meters. This increase was first measured in early June (Figure 5.12), and is likely the result of oxygen produced by stenothermal algal populations. This phenomenon occurs as a result of settling algae encountering the thermocline. The density difference is sufficient to impede settling. At this depth, light penetration is adequate enough to sustain photosynthesis. The dense accumulation of algae and its photosynthetic activity is sufficient to cause the observed oxygen increase.

5.1.3 Nitrogen

The concentration of ammonia and nitrate are influenced by a variety of bio- logical processes. Ammonia is generated by heterotrophic bacteria as an end product of the decomposition of organic matter. Nitrate is formed as a result of the biological conversion of organic and inorganic nitrogen compounds from a reduced state to a more oxidized state. This process is known as nitrifi- cation. Nitrate and ammonia ions are readily utilized as nutrients by phyto- plankton and aquatic macrophytes (Wetzel, 1983).

The concentration of ammonia in Onota Lake ranged from O.02 to 0.58 mg/1 (Figure 5.19 and 5.21). The observed mean concentrations at 0.5 meters below the surface were 0.098 mg/1 at Station #1 and 0.09 mg/1 at Station 12. Mid- depth concentrations were similar to those observed at the surface. The mean concentration at Station II was 0.08 mg/1 and 0.103 mg/1 at Station #2. Concentrations were generally higher in the lower strata of the lake. At Station #1, the mean concentration at 19 meters was 0.192 mg/1. At 7 meters, near the bottom at Station #2, the concentration was 0.092 mg/1. This appears to be a result of stratification and anoxic conditions in the hypolimnion. Under anoxic conditions, ammonia generated as a result of decomposition will not be oxidized to the nitrate form. This results in an accumulation of

ENV/L22-rpt2 5-4 ammonia in the lower strata which can be recycled into the water column during periods of mixing and subsequent oxidation.

Nitrate-nitrogen exhibited a sharp increase in concentration during the winter season (Figures 5.20 and 5.22). This pattern is similar to that commonly found in temperate lakes and is attributable to the decreased rate of assimilation by plankton and decreased rate of reduction by bacteria. Additional disparity between summer and winter concentrations may also be attributable to the decreased rate of nitrification during periods of stratification as the hypolimnion becomes anaerobic,

5.1.4 Phosphorus

In north temperate waterbodies, phosphorus is usually the limiting nutrient for plant growth (Sawyer, 1970). In essence, the availability of phosphorus, its annual load to the lake and subsequent concentration in the water column will greatly influence productivity. As the supply of phosphorus is increased, productivity is accelerated and/or increases as well (Lee et al., 1976). As a result of this classical relationship between phosphorus and productivity the EPA (USEPA, 1980) and other regulatory agencies have developed trophic "guidelines" based on the in-lake concentration of phosphorus. The Massachusetts Division of Water Pollution Control has identified in-lake total phosphorus (TP) concentrations exceeding 0.01 mg/1 as potentially degrading. This implies that phosphorus concentrations of this magnitude could accelerate the eutrophication process. In Onota Lake, TP concentrations ranged from <0.01 to 0.63 mg/1, but were generally less than 0.05 mg/1 (Figures 5.23 and 5.24).

TP concentrations measured in the lake over the survey period displayed great variation. This variation, under closer inspection, is largely the result of stratification and mixing phenomena, phytoplankton activity and interval regeneration from anaerobic sediments. When viewed in respect to seasonal biophysical processes, the TP dynamics of Onota Lake are fairly typical.

At Station #1, the South Basin, TP concentrations displayed three unique peaks. The first occurred in the spring approximately at the onset of strati-

ENV/L22-rpt2 5-5 fication (Figure 5.23). This peak was probably the result of spring thaw loading and lake mixing which occurred in March and April. As the growing season progressed, the concentration of TP in the surface waters declined, but hypolimnetic concentrations rose. The observed increase in hypolimnetic TP concentrations became most obvious after July 22nd. It is .at approximately this date that the hypolimnion became anoxic. Under anoxic conditions, lake sediment chemistry is altered, and the remineralization of phosphorus occurs (Freedman and Canale, 1977). This can lead to the accumulation of significant quantities of phosphorus in the non-mixed hypolimnion (Souza and Koppen, 1984). When the lake destratifies and undergoes mixing, the soluble phospho- _ rus which was remineralized can be circulated into the epilimnion and may | stimulate algal productivity. This result is an autumnal bloom. As the stratification of the South Basin weakened in the fall, hypolimnetic concen- • trations declined and metalimnetic concentrations increased. When the lake began to turn over in the fall, a dramatic increase in epilimnetic TP was • observed (Figure 5.23). The usual condition is for nutrients to be nearly evenly distributed in the water column after overturn. The reason for the great disparity in surface and bottom total phosphorus concentrations found in • the November South Bas i n samp1e is not known. As such, strati f i cati on,

— mixing, and internal regeneration of sediment bound phosphorus were processes | which appeared to most greatly influence TP dynamics to the South Basin.

• In the North Basin, Station 12, the pattern which developed was somewhat different than observed in the South Basin. In the spring, a peak in TP was • observed in the surface, metalimnion and bottom waters (Figure 5.24). As with the South Basin, these peaks occurred shortly after the North Basin became stratified. Once stratified, epilimnetic TP concentrations declined and • remained low (<0.05 mg/1) for the remainder of the growing season. Bottom „ water TP, however, rose dramatically in July and to a lesser degree in August I (Figure 5.24). It is suspected that the July 8th concentration may be an aberration or faulty data point. Anoxic conditions did not persist in the • North Basin's hypolimnion until August. The second bottom water TP peak coincides with the advent of anoxic conditions, and is probably the result of • internal TP regeneration from anaerobic sediments. The first bottom water TP peak however occurred prior to anoxic conditions. It is possible that • sediment was disturbed during sampling and became entrained 1n the sample.

• ENV/L22-rpt2 5-6 I I The North Basin did not exhibit an increase in epilimnetic TP following de- I stratification as did the South Basin (Figure 5.24). This observation may be the result of two factors. First, if there were sediments entrained in the I July 8 sample, causing a higher TP peak, these results should be discounted. Upon exclusion of the July 8 data point the TP in the North Basin's bottom water never reached the magnitude observed in the South Basin. As such, the I internally regenerated load was probably insignificant (See Section 10). Second, in the North Basin, the volume of the epilimnion is relatively much I greater than that of the hypolimnion, as most of the North Basin is shallow (Figure 4.2). When the North Basin destratifies its volumetric contribution I to the epilimnion is minimal. The combination of the small volume below the thermocline and the small quantity of the internally regenerated TP load results in only a minor increase in epilimnetic TP being realized in the fall I (Figure 5.24).

• 5.1.5 p_H

In typical northern lakes, such as Onota Lake, the pH ranges from 6.0 to 9.0. The pH of Onota Lake was found to range from 6.8 to 8.3 with a marked I decrease evident with increasing depth (Figures 5.25 and 5.26). This vertical distribution is mostly attributable to the (a) photosynthetic utilization of I carbon dioxide (COo) in the trophogenic zone, which tends to reduce (#2 content and to increase pH, and (b) the degradation of organic matter in the I hypolimnion which generates (#2 and tends to decrease pH. Vertical variations in pH became more pronounced during summer stratification I due to increased photosynthetic activity and decreased mixing of the water column. During those periods when Onota Lake exhibited a positive heterograde I oxygen curve at Station #1, a corresponding positive heterograde pH curve was I also evident. 5.1.6 Alkalinity

I Alkalinity, or buffering capacity, is a measure of the ability of a lake to I maintain neutral pH. Onota Lake can be considered to be a waterbody of medium I ENV/L22-rpt2 5-7 I

I buffering capacity. Total alkalinity, measured as CaC03, ranged from 60 to 80 mg/1. This range is typical for most Berkshire Lakes (Mass. Oiv. of Fisheries I and Game, 1979). No substantial fluctuation in the lake's alkalinity was ob- served over the duration of the study. The observed pH and alkalinity values indicate that the lake is, at present, not affected by acid precipitation. I The limestone geology of the watershed contributes to this historical resis- I tance to a decline in pH and a stable buffering capacity. I 5.1.7 Secchi Disc Transparency Secchi disc transparency provides a subjective means of determining trans- I parency and subsequently estimating productivity of a lake. Transparency is affected by a number of factors including density of planktonic algal cells, dissolved and particulate matter, water color, surface conditions, time of I day, and observer bias. When algal biomass is the primary factor influencing transparency, Secchi disc measurements can be used as an indicator of produc- I tivity in the following manner: lakes with transparencies greater than 8 meters are considered oligotrophic, those with transparencies less than 2 I meters are eutrophic and those with transparencies falling in between 2 and 8 meters are mesotrophic. Utilizing these criteria, Onota Lake can be I considered mesotrophic. Readings at Station #1 were generally 5 meters with the maximum transparency I of approximately 7 meters on October 14 and a minimum of 3.5 meters on April I 8. As illustrated in Figure 5.27, these variations relate to algal growth. Readings at Station #2 were generally 3.5 meters with a maximum of I approximately 4.7 meters on June 3, July 22, and October 14. Minimum transparencies of approximately 2 meters were recorded on June 25 and February I 19. The February low is attributable to ice cover whereas the June low is a combination of algal growth and macrophyte shading (Figure 5.28). The I relationship of secchi disc transparency and phytoplankton growth is addressed I further in sections 5.1.8 and 5.2. I I ENV/L22-rpt2 5-8 I I 5.1.8 Chlorophyll I Concentrations of chlorophyll a, b, c, and pheophytin were monitored throughout the study and employed as an indicator of primary production. I These data were utilized in combination with net and whole phytoplankton data (Section 5.3.1) as an indicator of the lake's trophic condition. Changes in chlorophyll a were examined in relation to changes in season, station, and 3 I secchi disc depth. Chlorophyll concentrations ranged from 0.1 to 7.5 mg/m . I These concentrations are generally indicative of mesotrophic water bodies. Generally, chlorophyll concentrations in Station #1 were higher than at • Station #2 (Figures 5.27 and 5.28). Decreased concentrations of chlorophyll a at Station #2 may be attributable to the high density of macrophytes found at • this station. Macrophytes tend to compete with the resident phytoplankton populations primarily for available light but also nutrients. This typically • reduces the establishment of dense phytoplankton communities.

A seasonal increase in the concentration of algae is evident in late May and • September. The peak in May is probably the result of a combination of optimal temperature, light and nutrients. The September peak is probably due to • macrophyte dieback in the north basin, epilimnetic erosion of the hypolimnetic TP rich water (Kortman, et al. 1982) and, eventually, destratification and • turnover of the lake. Chlorophyll concentrations were generally lowest during the winter months when • population densities are reduced as a result of low light and temperatures. • 5.1.9 Bacteria • Coliform bacteria are utilized as indicator organisms of septic or sewage contamination of waters. Although some coliform bacteria occur naturally in • the environment, fecal coliform bacteria are resident in the intestinal tract of birds and mammals. The occurrence of coliform bacteria may signify the potential presence of pathological bacteria or viruses. As such, standard • bacteriological analyses are routinely conducted for any water body which is i used for potable water or contact recreation such as swimming. The • ENV/L22-rpt2 5-9 Massachusetts Surface Water Quality Standards (314 CMR) Identify the water quality criteria required to sustain the designated use of a water body. Class B waters must meet the following criteria in respect to fecal coliform bacteria:

"Shall not exceed a log mean for a set of samples of 200 per 100 ml, nor shall more than 10% of the total sampler exceed 400 per 100 ml during any monthly sampling period, except as provided in 310 CMR 4.01(1)."

Throughout the study, fecal coliform numbers measured in the lake were relatively low and did not exceed state standards. The analysis for non-fecal coliforms is important as well, because their abundance may also result in unsatisfactory water quality (APHA, 1976). In-lake total coliform concentrations were variable (Figure 5.29 and 5.30), and at various times over the sampling program, peaks were observed. These peaks tended to coincide with periods of heavy Take usage (Memorial Day, July 4) or significant rainfall events (6/23; 7/2-7/6; 8/7-8/11; 8/27; 9/23; and 10/15). Total coliform levels were elevated (>400 colonies/100 ml) for most of August in the North Basin. This may be attributable to the unseasonably high rainfall which occurred during July and August 1988. The majority of the lake's watershed drains to the North Basin. Storm activity could have increased soil erosion an'd street runoff which, in turn, exacerbated coliform levels in this section of the lake.

5.2 WATER QUALITY - TRIBUTARY SAMPLING

Six streams are tributaries to Onota Lake. Two are intermittent, seasonal streams. The four major tributaries are: Daniels Brook, Churchill Brook, Parker Brook and a small tributary which drains the Blythewood Drive area along the western shore (Figure 5.1). The lake's primary tributary is Parker Brook. Parker Brook is fed by Hawthorne and Lulu Brook. The three streams converge approximately 1/2 mile upstream of Onota Lake. The three streams drain a total area of 865 hectares (see Section 4). In addition to the intermittent streams, there are seasonal seeps and drainage along the western shore and at the southeast end of the lake near Blythewood Drive.

ENV/L22-rpt2 5-10 I I Surface water samples were collected at mid-channel at the four major tributaries and the outfall on a semi-monthly basis from March through fall I circulation and monthly for the rest of the year. Two additional tributaries (Tributaries 4 and 5) were originally included and sampled on a semi-monthly basis, however after four sampling events it was deemed that neither were I major components of the lake's water or nutrient budgets. Sampling of these two tributaries was terminated at the end of May, and- the effort allocated to I different aspects of the project. I The tributaries were monitored in accordance with the Phase I Scope of Work, under baseline and storm conditions. Storm events were monitored on 9/24/86 I and 3/31/87. Results of the chemical and microbiological analyses of the various I tributaries feeding Onota Lake are presented in Figures 5.31 to 5.67 and Appendix B. In general, these data indicate that the waters which feed Onota I Lake are well oxygenated, and characterized by high nutrient and coliform bacteria concentrations. The high nutrient and coliform bacteria I concentrations are attributable to land use within the drainage areas of these tributaries. Of specific concern is loading resulting from agriculture I activities, residential development, and septic contributions. I 5.2.1 Churchill Brook and Daniels Brook Churchill and Daniels Brooks drain the lakes, two northern-most sub-basins. I Land use within these sub-basins is largely a combination of forest/farmland with sparse residential development. Sections of both sub-watersheds were I quarries at one time and remain erosion prone. I Total phosphorus ranged from 0.008 to 0.07 mg/1 at Churchill and Daniels Brook. The observed mean concentration of TP in Daniels Brook was 0.021 mg/1 and 0.015 mg/1 in Churchill Brook (Appendix B1-B2). The U,S, Environmental I Protection Agency (1976) suggests that to prevent accelerated eutrophication, in-stream total phosphorus concentrations should not exceed 0.05 mg/1, TP I concentrations exceeded this criteria on two occasions, both of which were I associated with storm events (Figures 5.42 and 5.47). I ENV/L22-rpt2 5-11 I I Nitrate-nitrogen concentrations ranged from 0.02 to 1.32 mg/1. The observed mean concentration of N03~N in Daniels Brook was 0.69 mg/1 and 0.45 mg/1 in I Churchill Brook (Appendix B1-B2). The highest concentrations appeared during the late summer months (Figures 5.41 and 5.46). It is suspected that runoff, I caused by the unseasonably high rainfall, leached nitrogen compounds from fertilized fields and contributed to these elevated concentrations. Ammonia- nitrogen ranged from 0.01 - 0.22 mg/1 in Daniels and Churchi"!! Brook. The I observed mean concentration in Daniels Brook was 0.065 mg/1 and 0.083 mg/1 in Churchill Brook. Total Kjeldahl nitrogen ranged from 0.05 to 0.46 in both I streams. The observed mean concentration in Daniel Brook was 0.211 mg/1 and 0.21 mg/1 in Churchill. Elevated concentrations were associated with storm I events. Total suspended solids ranged from <1 mg/1 to 37 mg/1. The observed mean concentration of total suspended solids in Daniels Brook was 5.9 mg/1 and I 5.3 mg/1 in Churchill Brook. Elevated concentrations were generally associated with storm events (Figures 5.39 and 5.44). Chloride concentrations ranged from 1.0 to 19 mg/1 but were generally less than 10 mg/1. The mean I concentration of chloride in Daniels Brook was 6.22 mg/1 and 3.9 mg/1 in Churchill Brook. Highest concentrations were associated with storm events I (Figures 5.39 and 5.44). The high winter concentrations measured were probably due to the application of road salt. Total coliform bacteria ranged I from 2 to 14,360 bacteria/100 ml and fecal coliform bacteria ranged from <2 to 1760 bacteria/100 ml. The mean concentration of total coliform observed in I Daniels Brook was 272 bacteria/100 ml and 1,296 bacteria/100 ml in Churchill Brook. Fecal coliform concentrations were elevated in both tributaries. The mean concentration observed in Daniels Brook was 126 bacteria/100 ml and 132 I bacteria/100 ml in Churchill Brook. Highest concentrations were encountered during the summer months and suggest contamination from animal or human wastes I or runoff from fertilized fields (Figures 5.38 and 5.43). I 5.2.2 Parker Brook I The Parker Brook monitoring data are presented graphically in Figures 5.58 to 5.62. Parker Brook drains Onota Lake's largest watershed. As mentioned previously, Hawthorne Brook and Lulu Brook are tributaries to Parker Brook. I Most of the lands drained by Parker Brook are forested. The topography of the I terrain is steep and the soils somewhat shallow along steep slopes. There is I ENV/L22-rpt2 5-12 a minimum of residential development within this watershed. A major summer camp, Camp Winadu, is located in Parker Brook's watershed. The camp lies along the shoreline of the lake.

Total phosphate-phosphorus ranged from <0.01 mg/1 to 0.14 mg/1. The mean concentration of TP was 0.024 mg/1. The highest concentration was measured on July 22, 1986 (Figure 5.62). At other times, concentrations were generally less than the 0.05 mg/1 maxima suggested by the EPA. Nitrate-nitrogen ranged from 0.12 to 1.1 mg/1 but was less than 0.6 mg/1 on all except one occasion. The mean concentration of nitrate-nitrogen was 0.37 mg/1. Ammonia nitrogen ranged from 0.03 to 0.22 mg/1 and the observed mean concentration was 0.07 mg/1. Animal waste applied to fertilized fields is a likely source for elevated concentrations particularly during storm events. Total suspended solids ranged from <1 mg/1 to 9 mg/1 with the highest concentrations associated with storm events. The mean concentration observed during this study was 3.8 mg/1. Total dissolved solids ranged from 15-350 mg/1. The mean concentration was 93.8 mg/1. Higher concentrations were associated with higher stream discharges. Chloride concentrations ranged from 2 to 12 mg/1 with one peak at 22 mg/1 on July 8, 1986. The reason for this summer chloride peak is not known. The mean chloride concentration in Parker Brook was 5.0 mg/1. Coliform..bacteria counts were extremely variable, with the highest numbers associated with storm events (Figure 5.58). The mean concentrations • of total coliform and fecal coliform bacteria for the study period were 585 and 94 bacteria/100 ml, respectively. Again, human or animal waste, as well • as fertilizers may have caused these elevated concentrations. • 5.2.3 Blythewood Drive - Unnamed Stream There are small seeps, swales and intermittent streams which flow into Onota • Lake. Flow from these waterways is inconsistent and is seasonal or the result of storm events. Most of these waterways are located along the lake's western • shore. None of these are gagable. There does exist a small unnamed stream which flows into the southern basin • from the Blythewood Drive area. This stream drains primarily wooded areas i although a small farm exists 1n the streams upper watershed. Closer to the • ENV/L22-rpt2 5-13 I I lake, residential land use characterizes the stream's watershed. Most of these residences are on 1/4 acre or larger lots and all use septic systems for I the treatment and disposal of wastewater. The terrain draining to the stream ranges from nearly flat to fairly steep. The stream itself is approximately 2 I to 3 meters wide, with a stoney bottom and water depths of 10 to 30 cm. . Water depths however can increase 2 to 3 fold during a storm event.

I The monitoring data for the Blythewood Drive tributary are presented in I Figures 5.63 to 5.67. Nitrate-nitrogen concentrations ranged from 0.05 to 0.95 mg/1, total I phosphate-phosphorus concentrations ranged from 0.01 to 0.11 mg/1 with the highest concentrations associated with storm events. Phosphate concentrations I were generally lower than 0.05 mg/1. TSS concentration ranged from <1 to 48 nig/1 with a significant increase in concentration evidenced during storm events. Total coliform bacteria ranged from 10 to 80,000 bacteria/100 ml and I fecal colifonns ranged from 1 to 2400 bacteria/100 ml. These high coliform concentrations, in addition to the substantial and continuously elevated I chloride and conductivity measurements, appear to be attributable to development in this subbasin, in particular, the use of on-site wastewater I treatment and disposal. The soils, hydrology and steep slopes surrounding Blythewood Drive tend to compromise the efficiency of septic systems, I exacerbate runoff, and contribute to increased nutrient and bacteriological loading to the lake.

I 5.2.4 Storm Sampling

I On two dates, September 23, 1986 and March 31, 1987, the main tributaries to the lake were sampled during storm conditions. The samples collected during I the Fall 1986 storm event were grab samples taken during peak flow from Daniels Brook, Churchill Brook, Parker Brook, the Blythewood Drive Tributary, I and the Pecks Road Tributary (Appendix C). During the Spring, 1987 storm event, ten samples were collected from Churchill I Brook, Daniels Brook, Parker Brook and the unnamed Blythewood Drive I tributary. The sampling for this storm event was begun at first hydraulic I ENV/L22-rpt2 5-14 flush, with samples taken at ten minute intervals for the first 30 minutes. Samples were then taken at 15 minute Intervals for a cumulative period of two hours (ten samples total for each stream). The samples from Churchill and Daniels Brooks were collected simultaneously and composited so that a total of ten samples were analyzed for these two streams. For metals analysis, separate composite samples were taken from the Parker Brook samples, the Blythewood Drive tributary samples, and the combined Churchill and Daniels Brook samples. These data, including summary tables for the sequential storm flow samples, are included in Appendix C. The storm flow data were discussed in previous sections (5.2.1 through 5.2.3) and presented in Figures 5.38 through 5.67.

5.3 FLORA AND FAUNA

Lake biota (excluding macrophytes) sampled during this survey are discussed in the following sections. (The macrophyte discussion may be found in Section 6.0).

5.3.1 Phytoplankton

Phytoplankton is an important component of a lake ecosystem. It may be present in sufficient quantity to color the water but is more often of such low densities that special means must be employed to obtain a sufficient quantity for identification.

Samples of both whole and net plankton were collected from both in-lake sta- tions as part of the regular water quality sampling program. Samples were collected on a semi-monthly basis from March through fall circulation and on a monthly basis for the balance of the year. During periods of stratification, the samples were depth integrated to the metalimnion. During periods on non- stratification, the samples were taken from the surface only.

Net plankton was collected with a 63 micron plankton net, and preserved with a buffered formalin-copper sulfate solution. Whole plankton was collected by immersing one-liter collection bottles into the water. The samples were pre- served immediately with lugols solution.

ENV/L22-rpt2 5-15 Prior to enumeration, the preserved, whole plankton samples were allowed to settle. Upon settling, they were concentrated by decantation and then enumerated using a Sedgwick-Rafter counting cell. Identification of plankton to the genus level was accomplished with a few taxa being identified to species. The algal mass was computed through use of cell volumes derived from direct measurements.

A seasonal succession of phytoplankters typically exists in temperate lakes. This pattern generally involves a winter minimum of flagellates adapted to low light and temperature. In the spring, diatom biomass increases substantially followed by a smaller increase in the green algae population. In the summer, diatom populations increase in mesotrophic and oligotrophic lakes while blue- green algae blooms may occur in eutrophic lakes. As water temperatures cool in the fall, algal densities decrease. However, a resurgence in algal densities, an autumnal bloom, may occur shortly after fall overturn.

Percent composition was chosen as the method .of analysis because it readily shows the seasonal changes in phytoplankton populations. In addition, this method makes obvious significant increases in taxa indicative of unhealthy conditions, such as the blue-green algae (Cyanophyta). The percent composition data are presented in Tables 5.1 through 5.4. In general, the data collected for this study showed that the diatom percent composition reached a spring low at approximately the same time the lake stratified (May 25th to June 3rd). In the early summer, diatom numbers rebounded. Fragilaria was a commonly occurring species in June to early July. As summer progressed, green algae became more numerous. This was best reflected in the net plankton samples and to a lesser degree 1n the whole plankton samples. The percent composition of blue-green algae remained relatively low throughout the summer and never reached densities indicative of unhealthy conditions. The Dinaflagellates Dinobryon and Ceratium continued to occur in the phytoplankton assemblage. The percent composition of both genera was variable and not as great as observed in the spring samples.

ENV/L22-rpt2 5-16 Toward the end of the summer and into early fall (August 27th to October 15), the diatoms once again dominated the phytoplankton assemblage. Green algae displayed a steady decline after the summer. Blue green algal percent composition in the net plankton also declined, but displayed two slight increases in the whole plankton collected from the North Basin. The increases appeared to coincide with the destratification of the North Basin and the seasonal dieback of macrophytes.

5.3,2 Benthic Colonization

The benthic infauna represents an important element in the energy pathways and ecology of a lake system. The benthos (oligochaetes, aquatic insect larvae, snails, amphipods, etc.) represent a direct link between the sediment/detrital component of the lake and upper trophic levels. Organic material produced in the euphotic zone of the lake settles to the bottom. Decomposition of this material releases carbon, nutrients and trace metals needed by primary producers to aid in additional productivity. The benthos are a key factor in the breakdown of this material. In addition, many benthic organisms are important forage for fish. As the majority of benthic forms are sedentary, they serve as good indicator organisms since they are more directly affected by their immediate environment than other members of the biota (phytoplankton, , fish). Their community structure can often times provide insight into the intricacies of a lake ecosystem (Wiederholm, 1980).

Benthic samples were collected on a monthly basis from key sites in the lake, using an Ekman dredge. All samples were preserved with a 10% Rosebengal- Formalin solution with sorting and identification being conducted in the lab. All organisms were identified to the lowest practical taxon and their community structure analysed in terms of species composition, dominant organism, abundance, diversity, and evenness (Pennak, 1978; Merritt and Cummins, 1978; Pielou, 1966).

Benthic colonization was investigated at three in-lake, littoral zone stations. # station was located in the North Basin, and two stations in the South Basin, one at the extreme southern shore and the other near the east shore approximately 20 meters south of the fishing pier. All stations were

ENV/L22-rpt2 5-17 I I TABLE 5.1: WHOLE PLANKTON % COMPOSITION I BASIN 1A I GENUS/DATE 4-8 4-25 5-6 5-20 6-3 6-25 7-8 7-23 8-13 8-27 9-10 9-23 10-15 11-14 CYANOPHYTA {blue green algae) TOTAL <1 5.2 14.2 10.6 9.0 3.6 7.9 1 Anabaena 9.5 5.3 4.5 3.6 7.1 •• Polycystis 5.2 4.5 Microcystis Coelosphaerium <1 4.7 5.3 0.8 CHLOROPHYTA(greens)TOTAL 31.6 28.49 5.1 38.0 27.8 38.0 21.7 61.4 24.5 14.2 24.2 18.0 11.7 Ankistrodesmus 5.4 3.2 1.7 I Characium 8.0 Chlamydomonas 1.7 2.8 19.0 11.1 20.0 10.6 6.1 2.7 2.7 Chlorella 7.9 1.3 <1 19.0 13.9 14.0 11.1 14.2 16.2 7.2 1.8 Gonium 4.0 1.8 Hydrodictyon Protococcus 2.8 36.4 14.3 3.6 Stentor I Volvox 22.7 4.1 8.1 1.8 Ulothrix I Zygnema 16.6 21.19 3.4 <1 2.3 ••• CHRYSOPHYTA (diatoms) TOTAL 8.3 49.2 70.9 20.9 47.2 55.5 44.0 42.3 9.1 71.4 56.8 54.1 65.7 69.3 Asterionella 5.8 26.3 13.2 4.7 12.0 9.1 14.3 14.2 21.7 4.5 47.3 I Closterium 15.2 <1 Coscinodiscus <1 <1 <1 Cyclotella 8.0 15.8 6.1 16.2 0.8 I Cymbella

_ Diatoma Fragilaria 2.9 10.4 3.3 14.2 22.2 4.0 38.8 7.0 61.2 Navicula <1 9.5 5.6 10.6 14.2 8.9 • • Nitzschia Pinnularia I Staurastrum

I ENV/L22-tbls/5 I — - • '

I^H TABLE 5.1: WHOLE PLANKTON % COMPOSITION BASIN 1A (Continued) I GENUS/DATE 4-8 4-25 5-6 5-20 6-3 6-25 7-8 7-23 8-13 8-27 9-10 9-23 10-15 11-1' CHRYSOPHYTA (diatoms) (con't.) Synedra 2.1 16.9 5.5 9.5 8.3 5.3 16.2 6.3 I Synura Tribonema Tabellaria 3.3 11.3 28.7 4.4 9.3 19.4 20.0 10.6 12.2 21. 4 6.3 CRYPTOPHYTA I Cryptomonas 7.9 5.6 3.5 13.8 16.7 6.0 2.3 2.0 7.1 3.6 PYRROPHYTA TOTAL 51.4 15.4 19.0 57.5 4.0 26.4 18.2 2.0 21.4 10.8 5.4 ,5.4 Ceratium 43.9 <1 4.0 5.3 15.9 2.0 10.8 5.4 Dinobryon 7.5 15.4 19.0 57.5 21.1 2.3 21.4 5.4 1 Peridinium PROTOZOANS TOTAL 4.0 1.8 I Acanthocystis Actinophrys 1.8 Pelomyxa I Pronodan Urostyla 4.0 Vorticella ROTLERA TOTAL 4.0 Rellacottia Keratella 4.0 1 Polyarthra >, ""' • ' "X

I ENV/L22-tbls/6 I I

TABLE 5.2: WHOLE PLANKTON % COMPOSITION I BASIN IB I GENUS/DATE 4-8 4-25 5-6 5-20 6-3 6-25 7-8 7-23 8-13 8-27 9-10 9-23 10-15 11-14

CYANOPHYTA (blue green algae) TOTAL 16.9 5.5 18.9 10.0 2.6 17.3 2.4 9,5 6.6 3.1 Anabaena 16.9 5.5 10.8 2.6 13.8 1.2 9.5 6.6 3.1 Polycystis 8.1 5.2 3.4 I Microcystis 4.8 Coelosphaen'um 1.2

CHLOROPHYTA (greens) TOTAL 34.2 32.2 1.2 36.1 32.4 47.4 15.8 25.0 13.8 8.1 14.3 4.6 10.1 I^^^H Ankistrodesmus <1 1.3 <1 1.2 Characium 1.2 Chlamydomonas <1 1.8 11. 1 10.8 9.7 2.3 14.3 2.3 1 CMorella <1 1.3 <1 25.0 21.6 28.0 10.5 25.0 3.4 2.8 Gom'um 0.9 4.7 Hydrodictyon Protococcus 9.7 13.8 1.2 I Stentor Ulothrix Volvox 5.3 0.9 Zygnema 31.6 30.4 3.1

CHRYSOPHYTA (diatoms) 1 TOTAL 82.0 44.3 5.1 1.2 52.8 32.4 24.3 79.0 25.0 51.6 83.8 47.5 138.5 60.8 Asten'onella 1.7 17.2 <1 5.4 4.8 55.3 6.9 81.0 41.4 Closterium <\ 3.4 2.3 Coscinodiscus 13.9 1 Cyclotella 6.5 5.3 3.4 0.9 Cymbella <1 4.7 Diatoma 3.1 Fragilarla 74.6 15.7 <1 13.9 16.2 6.5 9.5 34.0 1.5 Navicula <1 1.3 10.5 23.8 19.8 Nitzschia 73.3 I Pinnularia <1 Staurastrum

I run /i no +• kl r- It TABLE 5.2: WHOLE PLANKTON % COMPOS1T10H BASIN IB (Continued) GENUS/DATE 4-8 4-25 5-6 5-20 6-3 6-25 7-8 7-23 8-13 8-27 9-10 9-23 10-15 11-14 CHRYSOPHYTA (con't.) Synedra 1.4 8.5 8.3 7.9 6.9 2.8 Synura 5.4 0.8 Tabellaria 4.3 1.6 5.1 16.7 5.4 6.5 14.3 31.0 10.5 9.5 11.7 Tribonema 10.7 CRYPTOPHYTA Cryotomonas <1 8.2 5.5 17.0 5.5 10.8 16.1 10.7 5.8 6.6 5.5 PYRROPHYTA TOTAL 55.3 61.8 5.4 39.4 17.3 10.5 29.0 2.8 Ceratium 10.3 10.5 Dinobryon 55.3 61.8 5.4 3.6 29.0 Peridinium 17.9 6.9 2.8 PROTOZOANS TOTAL 2.6 Acanthocystis Actinophrys 2.6 Pelomyxa Pronodan Urostyla Vorticella ROTLERA Rellacottia Keratella Polyarthra

ENV/L22-tbls/4 TABLE 5.3: NET PLANKTON % COMPOSITION BASIN 1A GENUS/DATE 4-8 4-25 5-6 5-25 6-3 6-23 7-8 7-23 8-13 8-27 9-10 10-15 11-14

CYANOPHYTA (blue green algae) TOTAL 2.5 6.4 1.3 16.0 2.4 <1 6.7 2.8 <1 <1 Anabaena 1.3 6.7 2.8 <1 Polycystis 2.5 6.4 0.8 <1 <1 Microcystls 16.0 1.6 Coelosphaerium CHLOROPHYTA (greens) TOTAL - <1 7.3 <1 10.0.. <1 4.6 60.0 14.6 2.1 2.6 18.7 Ankistrodesmus Chiamydomonas . 2.1 1.0 <1 Chlorella 4.6 20.0 6.3 2,1 <1 Euglena 2.0 0.8 Gonium <1 6.7 <1 Hydrodictyon .-. 4.4 Protococcus 7.3 8.0 33.3 4.1 <1 Stentor <1 Ulothrix Vol vox <1 2.1 Zygnema . 1.6 <1 14.3

CHRYSOPHYTA (diatoms) TOTAL 76 35.8 29.0 5.4 4.0- 91.3 88.0 20.0 55.5 35.8 80.5 82.3 66.2 Asterionella 15.2 10.4 6.7 < 2.8 37.6 2.8 4.7 10.9 53.2 Closterium <1 Coscinodiscus Cyclotella 1.3 <1 Cymbella <1 Diatoma x <1 Fragilaria 13.0 5.4 2.0 80.0 12.8 6.3 26.0 71.6 62.9 11.7 Navicula Nitzschia Pinnularia 1.2 Staurastrum Synedra 1.3 4.0 2.0 2.0 _ 6.7 2.1 7.0 4,2 , 6.8 1.3

ENV/L22-tbls/l I I TABLE 5.3: NET PLANKTON % COMPOSITION BASIN 1A (Continued) • GENUS/DATE 4-8 4-25 5-6 5-25 6-3 6-23 7-8 7-23 8-13 8-27 9-10 10-15 11-1' • CHRYSOPHYTA (con't.) Synura

PROTOZOANS TOTAL 5.2 Acanthocystls <1 1.1 « Act inophrys <1 4.3 <1 5.2 H Pelomyxa ™ Pronodan Urostyla • Vorticella 1.6 1.4

ROTLERA TOTAL 5.7 2.9 <1 8.0 1.2 2.1 2.1 2.1 Rellacottia 2.5 6.0 Keratella 1.3 <1 1.2 2.1 2.1 2.1 I Polyarthra 1.9 2.9 2.0 <1

ENV/L22-tbls/2 1 TABLE 5.4: NET PLANKTON % COMPOSITION 1 BASIN IB GENUS/DATE 4-8 4-25 5-6 5-25 6-3 . 6-23 7-8 7-23 8-13 8-27 9-10 10-15 11-14

CYANOPHYTA (blue green 2.1 <1 3.4 11.1 2.4 8.7 4.5 6.4 <1 algae) TOTAL Anabaena <1 4.5 6.4 <1 Polycystis 2,1 <1 6.7 8.7 1 Microcystis 4.4 2.4 Coelosphaen'um 3.4

CHLOROPHYTA (greens) TOTAL 1.7 5.0 11.1 1.2 8.6 13.7 27.2 14.2 6.6 2.5 2.2 I Ankistrodesmus 2.6 Chlamydomonas 9.0 10.3 Chlorella 13.7 4.5 1.9 Gonium 1.2 Hydrodictyon , <1 Protococcus 2.9 13.7 3.3 1 Stentor 2.1 , 4.3 Volvox <1 Ulothrlx 1.7 Zygnema 4.4 4.3 1.3 3.3 0.6 2.2 Euglena 6.7 <1 CHRYSOPHYTA (diatoms) I TOTAL 63.1 76.7 14.6 37.5 22.2 79.3 8.7 3.4 13.5 36.0 76.7 73.9 91,2 Asterionella 8.8 4.0 3.6 * 9.5 9.0 7.8 5.1 7£,9 Closterium <1 1.7 I Coscinodlscus Cyclotella 12.9 Cymbella ; Dlatoma^' 7.2 4.4 <1 4.5 Fragilaria 11.3 10.2 64.5 16.7 62.8 59.2 Navicula 8.7 Nitzschia <1 <1 t 1.1 <} I Pinnulan'a . 1.2 Staurastrum <1 6.7 Synedra <1 <1 <1 2.1 11.1 „ 9.0 9.0 5.0 5.1 11.0 1 Synura t 4.5 1.3 <1 <1 I ENV/L22-tbls/7 TABLE 5.4: NET PLANKTON t COMPOSITION BASIN IB (Continued) GENUS/DATE 4-8 4-25 5-6 5-25 6-3 6-23 7-8 7-23 8-13 8-27 9-10 10-15 1.1-14 CRYSOPHYTA (diatoms) {con't.) Tabellaria 43.0 62.5 24.6 4.1 3.4 1.3 Tribonema <1 CRYPTOPHYTA Cryptomonas 1.4 2.2

PYRROPHYTA TOTAL 31.00 12.44 66.4 49.3 20.0-- 1.88 22.0 3.4 54.22 41.33 12.88 20.3 2.6 Ceratium 20.00 1.88 22.0 3.4 36.00 31.00 6.1 <1 Dinobryon 31.00 12.44 66.4 49.3 4.5 1.3 6.7 20.3 2.6 Peridiaum 13.7 9.0 PROTOZOANS TOTAL . 13.0 Acanthocystis Actinophrys 4.3 Pelomyxa <1 8.7 Pronodan Urostyla <1 Vorticella ROTLERA TOTAL 1.1 9.3 1.4 31.1; 13.0 26,0 4.0 2.6 1.1 . 2.6 Rellacottia <1 1.7 8.9 <1 <1 Keratella 7.6 1.4 13.3; 13.0 13.0 4.0 2.6 1.1 2.6 Polyarthra , 1.1 8.9 <1 13.0 I I I

ENV/L22-tb1s/8 sampled once each in spring, summer and fall. The taxa composition percentages calculated for the samples are presented in Figures 5.68 through 5.76.

Typically, the benthic fauna of the littoral zone is diverse and abundant. Littoral sediments are characteristically rich in organic material and can support a wide variety and high density of organisms due to the availability of food sources. However, benthic colonization is greatly affected by sediment grain size, as grain size can determine stability, influence interstial dissolved oxygen concentrations and reflect the amount of organic matter (i.e. food) present. The benthic samples collected from Onota Lake reflect a community of low diversity and abundance. The sediments from the South Basin were relatively homogeneous sandy materials of low organic content. The sediment from the North Basin, although rich in detritus, was very silty and unstable. These sediment characteristics may be the cause of the relatively low diversity and abundance of benthic organisms found in the Onota Lake samples. The benthos in this lacustrine zone is typically characterized by relatively diverse and abundant population with high oxygen requirements. The homogenous substrate and the low amount of organic matter found in the sampled areas are probably the limiting factors. No substantial seasonal variation in diversity was observed. This is typical of lakes providing there"is a relatively constant food source.

The benthic community was comprised mainly of the scud (Hyallella), midges (Chlronomus), caddlsflies (Trichoptera), and mayflies (Ephemeroptera). Hyalella, the genus most frequently encountered, is a common amphipod found in unpol1uted waters. Hyalella, and amphipods in general, have high oxygen requirements and are usually restricted to waters of high oxygen concentrations. Mayflies and caddlsflies also have high oxygen requirements and are indicative of clean waters. Chlronomid larvae are ubiquitous and may dominate certain benthic communities. No distinct, seasonal or spatial pattern in species composition was observed 1n the benthic community. The community assemblage of benthic organisms 1n Onota Lake indicates a well oxygenated,/unpolluted lake.

ENV/L22-rpt2 5-26 5.4 FISHERY

The fishery of a lake represents its upper trophic level and is a focus of a great deal of recreational attention. A qualitative assessment of Onota Lake's fi shery was conducted usi ng beach sei ne, trap net, and electroshocking. The species captured were examined, measured and released. The fishery analysis included species composition, abundance, dominant species, and proportional, stock density.

Onota Lake supports a diverse fish community. Fourteen different species were collected as part of this study, but a greater number of species are known to occur in the lake (Table 5.5). Many species have been stocked officially and some unofficially (see Section 2.0). The most important game fish present in the lake include and a variety of trout. Currently the Massachusetts Division of Fisheries and Wildlife stocks approximately 16,000 trout annually in Onota Lake.

Of the species collected during the 1986 survey, chain pickeral (Esox niger), pumpkinseed (Lepomis qibbosus), bluegil1 Lepomis macrochirus), rock bass (Ambloplites rupestris) and yellow perch (Perca flavescens) dominated the catch. Also abundant were golden shiner (Notemigonus crysoleucas) and northern pike (Esox lucius). Although not captured, there was evidence of large schools of smelt (Osmerus mordax) being present at mid-water depths. Species distribution conformed generally to that found in a 1978 survey conducted by the Massachusetts Division of Fisheries and Wildlife. However, the 1986 data suggest that dominance has increased.

Proportional Stock Density (PSD) for the more abundant species: the pumpkin- seed, blueglll, yellow perch, rockbass and chain pickerel, was computed. PSD is the proportion of fish of quality size 1n a stock expressed as a percent- age. This analysis demonstrated that the sampled pumpkinseed, bluegill, yellow perch, and rockbass populations Include a high percentage of quality size individuals. Analysis of the 1978 survey indicated that only the yellow perch and rockbass had a high percentage of quality size Individuals.

ENV/L22-rpt2 5-27 TABLE 5.6 FISHERY DATA

1978 PSD Species x 4/24 5/20 6/23 7/22 8/13 8/26 9/23 10/14 11/13 Total Survey PSD* 1978 Golden Shiner 8 2 2 1 18 180 Chain Pickerel 1 1 ' 8 7 4 25 124 9 2 Pumpklnseed 22 6 11 10 2 64 89 85 15 Bluegill 13 12 4 3 1 35 5 61 Rock Bass 7 13 4 8 2 35 26 46 56 Yellow Perch 12 8 1 1 11 35 179 61 44 Brown Bullhead i 2 8 10 Black Grapple 2 2 Smallmouth Bass 6 Carp 6 8 Northern P1ke 10 7 White Sucker Common Shiner 2 Largemouth Bass 1 115 247

*Proportional stock density I

H Onota Lake contains both a warm and cold water fishery. The volume of water available for trout, (the trout layer) was calculated. The trout layer, is • that volume of water less than 71"F but containing at least 5 ppm of dissolved oxygen. The trout layer appears to have declined over the years. In 1947, H this layer amounted to 42% of the lake's volume, but by 1972 it had declined to 18.5$. Anoxic conditions in the hypolirainion act to reduce the volume of this layer. As the duration and extent of hypolimnetic anoxic conditions • increases, the amount of available trout habitat will decrease. Temperature _ and oxygen profiles conducted over the course of the present study indicated • that during the summer stratification, approximately 13% of Onota Lake's volume was suitable for trout. This apparent decline in available trout I waters is likely due to an increased sediment oxygen demand (SOD), which has steadily increased the volume of water with an oxygen deficit. It appears H that a hypolimnetic aeration system which will oxygenate yet maintain the temperature integrity of this layer, is needed. The issue is discussed in • detail as part of the restoration and management plan.

ENV/L22-rpt2 5-29 •TWBUTAHY

\ M//vU! \m&£j i, .OUTLET . TRIBUTARY 3AMPUNG STATION i-<2 ^S-^«/^~-, A*. **> t WQftTHBASM)

NOTE: DEPTHS ARE REPORTED IN FEET

LEGEND:

TRIB 1 DAMEL8 BROOK

3 WAOHAU BROOK* 4 UN-NAMED •• 6 PARKER BROOK * BLYTHEWOjpO OWffi, UN-NAMED

APWL AND MAY, FLOW MCXNMI8TENT

FIGURE 5.1 SAMPUMG LOCATIONS IN-LAKE STATION 1 (SOUTH BASW) STATION 2 (NORTH BASW) PREPARED FOR: ONOTA LAKE PITTSFIELD, MASS. PROJECT No. 541648 SEPTEMBER 1988

^-^h.*^^^***^^J^ INTERNATIONAL « 1984 IT CORPORATION TECHNOLOGY ALL COPYRIGHTS RESERVED CORPORATION "Do Not Scale This Drawing" I

I TEMPERATURE ISOPLETH for In-Lake #1 (Main Basin)

ICE

19.5 JULY AUQ SEPT OCT NOV JAN FEB

MONTH

FIGURE 5.2 FIGURE 5.5 TEMPERRTURE PROFILE - WINTER In-l_al 18 19 20 21 FIGURE 5.4 TEMPERnTURE PROFILE - SPRING In-l_ak© *1 CMaln Basing - ONQTR LRKE TEMPERATURE Cdegrees CD 0 2 6 8 10 12 14 16 18 20 22 24 26 u i i i i i i i i i i i i O £> O 2K 1 """""Oti* P ^ / i T / t <> 2 9 > 9 ¥ + <> 3 9 > 9 ¥ t <> 1 : i ; 4 o t* n ^k ..---+ <> 5 9 > 9 * t"' .. D ^ + .--o' 1 1 7 9 > 9 * .+ .^" 1 : 1 : 8 6 > n x + .0 UJ cg* t> C3 ^K 4- ^ ? / / / ..•••*'"/ 3/26 10 o> n .^'o'"' ..+ i -" /,--•' /,-••" 11 c> j: /': 4/8 12 1 / / / / 13 O n ¥+> n 4/25 14 r- rU>^ nLJ ^K -*-jT•

15 - [> n ?K-t-> •K l"~t M** t S. 5/6 16 > LJ 7K~T/ r> n ^4-> 17 ••i 'i* r. + 18 c> n^f 5/20

19 O / 20 n^+ 6/3

O1 T FIGURE 5.5 l| TEMPERflTURE PROFILE - SUMMER T In-L-ato #1 CMain BasinD - ONOTR LflKE f\ (I TEMPERATURE (degrees C)

-f 0 2 4 6 8 10 12 14 16 18 20 22 24 26 \ 0 i < i i i i i t i i i i • O Q + D>3 4 f L <> 9 t ^ I ; 1 i /i 2 <> 6 + :J>] , : I :' 3 O Q + 3i>3 \ i I 1 ...-•;/ •O O + C> D:

^ *~\ ^s o ~r> n ^ ^J S/ W V^ I—J ?K

£\ F-^ w^\ ! I^ 11lf= • a>. ^o ty g^2£^^° 0 «a a9 1 /f&S' 6/25 i 10 t>iV6 1 11 &;Q^ > .&•• 7/8 - 1O >.pp I if/T lo-h a T1 13 r : I f- 7/22 I 14 . V 1 ^i™j\ i 16 & 8/13 I 17 Qf +

f , 18 Q* 8/27 : 1L 11 19 c?^" o i 20 9/ 1 O O 1 / 21 L

I FIGURE 5.6 TEMPERRTURE PROFILE - FRLL #1 CMain Basin3 - ONOTR LRKE TEMPERATURE (degrees CD 02 6 8 10 12 14 16 18 20 22 24 26 ! i i i i i i -i i i i — r 0 g £> 9 9 > 9 2 9 ^9 3 9 > 9 4 9 > o 5 9 ^ 9

E 6 ni t:> oT 7 9 > 9 Qu 8 9 ^ 9 9 n > ^o ° ' rv^^^ 9/23 10 11 - n cf> > i /.-"• 1 O/ 1 4 12 13 ?n or> n i A* 11/14 14 n o> 15 n cp> 16 n p> 17 n o> ! /> 18 19 n? 4o 20 h 21 FIGURE 5.7 TEMPERRTURE PROFILE - WINTER In-l_ake #2 C Sha 1 low BasInD - ONOTR LRKE TEMPERATURE C degrees C) 02 4 6 8 10 12 14 16 18 20 22 24 26 0 i i i i i i i i i f i i o> 1 - > ''p

2 >p

3 - . >p

a 4 Vf^^+r^*/""^j ^/

£ \ \ O =^ 1 \ - 1/21 i 5

M 2/19 6 0 o

O

7 -

8 -

9 FIGURE 5.8 TEMPERRTURE PROFILE - SPRING In-Lake #2 CShaI low BasinD - ONOTR LRKE TEMPERATURE Cdegrees C) 02 4 6 8 10 12 14 16 18 20 22 24 26 0 n r ? ^ q o > a

o > a

6

B 4 O V—' O 4/8 O

4/25 6 O n 5/6 7 n 5/2O 8 S/3

9 FIGURE 5.9 TEMPERRTURE PROFILE - SUMMER In-L_ake #2 C Sha I low Basin} - ONOTR LRKE TEMPERATURE (degrees C) 0 2 4 6 8 10 12 14 16 18 20 22 24 26 u

<> Q + ;j> 1 1 T ?t /t

2 <> 6+ * J>

3 - o o > p£

'a 4 i i^v

g :' //-V*' / O fc[_tj 5\j " ^P,-« * » • * • 6/25

X * " " " ** k //l-'''V 7/9 ^ k^/* LJ TiSrf' ~r 6 ,^ ,* •" * .-•>'' •-•' /-!"'' D //"/ ,-•// 7/22

7 O U-^l_n^i Ji —i1— xlNv^fc>^^i xijixt / ; / 8/13 > : O

8 1. 8/27

+ 0 9/10 1 FIGURE 5.10 TEMPERRTURE PROFILE - FRLL In-Lake #2 C Sha i low BasirO - ONOTR LRKE TEMPERATURE (degrees C) 02 4 6 8 10 12 14 16 18 20 22 24 26 0 i i i i i i i i i i 1 1 — p c?> 9 6 > p I

2 6 > 6

' 3 - . 6 > p

B 4 - p t> o •~t ; 1 / o | i / / 9/23 Aj 5 p E> Q 3 i / i > j ?| 1O/14 , 6 P > Q

1 O ! :" 7 > 6 I o

! 8 ''' 1 9 - ' FIGURE 5.11 DISSOLVED OXYGEN PROFILE - WINTER In-l_oU:e #1 C Ma 1 n BasinD - ONOTR LRKE DISSOLVED OXYGEN Cmg/1) 0 1 2 3. 4 5 8 7 8 9 10 11 12 13 14 15 16 0 i i i i i i [i i 1 1 1 1 I 1 • 9 cn> o >"" "a.. 2 p j> 'q 3 o >" b 4 o t> ...n /-••' ...-•-'""

'e 6 Or Er> Pr 7 9 > ci fe== •-. / .-•' ^ ou 8 o* a 9 Eb 10 /' 11 .OO 12 > 0 13 E> O O ^ / 1/21 14 > o 15 >' p > 16 p>' o 2/18 // 17 >-•• .-> O ._n, 18 3/4 19 20 - 21 FIGURE 5.12 DISSOLVED OXYGEN PROFILE - SPRING In-L_ake #1 C Ma I n BasInD - ONOTR LRKE DISSOLVED OXYGEN Crag/I) 0 1 2 3 4 5 6 7 8 9 10 II 12 13 14 15 16 0 1 o 2 3 <> + 9 > ; t • f 4 +. g: > 5 b' 6 '•Q; > o 7 8 9 3/26 10 11 ,•-:<* 4/8 12 .....<>•;+ •••" ic^b 13 .0" + ^ n 4/25 14 15 n i 16

+ > D 5/6 17 9> + > D i 18 o + D 5/2O v n 19 O 20 6/3 21 FIGURE 5.15 DISSOLVED OXYGEN PROFILE - SUMMER In-l_ake #1 C Ma 1 n Bas i rO - ONOTR LRKE DISSOLVED OXYGEN Cmg/n 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 i i i i i i i i i i i i i i i O [>-t~C O

1 <>,H> "^ 2 <-.> *!CP 3 - > jjib 4 +> '49 •"•.I \ 5 '* £*9P 6 ^-KOfc, • \ \|\ *•-. + 7 <>,.. c» s "*• : 8 ,+"".... '^ ....^£p to 9 -{-."<>'''"....jK"*" " .t>y o ^ V-AV.'. ...^ / 6/25 10 11 -*•"" +•'' a- -".:>•'""" a^^ > 7/s 12 -i"'^''""""" o-^^"^ 13 - *.D>] o a - * c8X 7/22 14 :- A 15 - •* to ^ 16 - -i ID 8/13 ' '. In 17 - -j ^^ r /^^ 18 -+¥& 8/27

19 *, ^ >N o o 20 r /' 9/10 2O 11 ' FIGURE 5.14 DISSOLVED OXYGEN PROFILE - FRLL In-L-dke #1 C Ma i n Bas i n 2> - ONOTR LRKE ! DJSSDLVED OXYGEN C ing/ ID 0 1 23 4 5 6 7 8 9 10 11 12 13 14 15 16 0 > i i i i i i i i i i i i i i ?f 1 I2 >6 ; 3 i ; /f?f i : 1 ^ ^ .e 6 "* ^^ Ci \~s _. 7 _ f^ ^S(

- ^> C^~l Q*"»" R" w „ ': 1 1 a 9 ^ ^99 10 - f — ^riiri *F> 11 -co r cp 12 «?9 - 9 i • / i 13 c.O P o '• 1 1 9/23 . 14 to n I 15 9" 9 > ' 16 .[0 a 10/14 i , 17 rf- ' n 18 -o a 11/14 i1 1 \ 1 19 -O , D 20 - 21

L. FIGURE 5.15 DISSOLVED OXYGEN PROFILE - WINTER In-l_ake #2 C Shia I low Basin} - ONOTR LRKE DISSOLVED OXYGEN Cmg/I) 0 1 2 3 4 5 6 7 8 9 10 1! 12 13 14 15 16 0

...-O o-"

o

o

B 4A o 1 /21 .aU 2/1 9 > p 6 7

8

9 FIGURE 5.16 DISSOLVED OXYGEN PROFILE - SPRING Jn-I_ake #2 C Sha I low Bas i n D - QNOTF) L.RKE DISSOLVED OXYGEN (mg/n 0 1 2 3 4 5 6 7 8 9 10 II 12 13 14 15 16 0

ot>

O '"5 o 4/8 H-0 o 4/25 6 .-O a 5/6 7 g > ri 5/20 8h i/3

9 FIGURE 5.17 DISSOLVED OXYGEN PROFILE - SUMMER In-l_ake #2 C Sha I low Bas I n 3 - ONOTR LRKE DISSOLVED OXYGEN Cmg/I) 0 1 234 5 6 7 8 9 10 II 12 13 14 15 16 u i i i i 1 I ' 1 1 ! 1 1 I - 1 t 1

O :.Q P ] o *h o a

2 <> *f p p

• y \ 3 . . i^ *t "«

** .* * / / / :- 1 / o : : /s•n ••' r>f^-c " i /^^\XL -•^ ' R/P^u^/ ^_S ^^ e 4 "" V^ i - | \^^ \ V^fN w * «** * *

— ****** " " / ^^> .....••••-"" / X iy " 7/9 QUJ- R& a ,••"'•••''' /-/ a .,•••;>••' ,// 7/22 6 - • »• ***.•* MX - -•* -.^" * ** ^^

iI • " *--'-' .»*' '"--'" T._ S^ / t ^^1^ 3 7 --'^f-O' n o U 8/27 8 o I , . 9/10 "^ x' FIGURE 5.18 DISSOLVED OXYGEN PROFILE - FRLL In-L_ake #2 C Sha I low BasInD - ONOTR LRKE DISSOLVED OXYGEN Cmg/I) 0 1 2 3 4 5 6 7 8 9 10 1! 12 13 14 15 16 Q

>

> a s 4A o > P o 9/23 o

1 O/1 4 o a 11/14 o e> o

8 -

9 FIGURE 5.19 MMONin vs'TI In-Lake #1 - ONOTfl LflKE 0.62 SURFRCE O METflLIMNION A D BOTTOM 0.52-

A

0 " 3-4 4-8 5-6 6-3 7-8 8-13 9-10 10-14 1 DflTE 3-26 4-25 5-20 6-25 7-22 8-27 9-23 11-14 2-18 FIGURE 5.20 ITRflTE vs TIM In-Lake #1 - ONOTfl LflKE 4-8 surface = 1.9 1 4-8 bottom = 1.7 a SURFRCE o METflLIMNION

P A BOTTOM

"3-4 4-8 5-6 6-3 7-8 8-13 9-10 10-14 1-20 DRTE 3-26 4-25 5-20 6-25 7-22 8-27 9-23 11-14 2-18 FIGURE 5.21 MMONin vs'TI In-Lake #2 - ONOTfl LflKE 0.22 i SURFflCE o METflLIMNION BOTTOM 0.20

0.18

0.16

0.14h

X GOO.I2 E v_^ o.io- CC o 0.08 - z: 5 0.06 0.04

' 3-4 4-8 5-6 6-3 7-8 8-13 9-10 10-14 1-20 DflTE 3-26 4-25 5-20 6-25 7-22 8-27 9-23 11-14 FIGURE 5.22 ITRRT vs TIM In-Lake #2 - ONOTR LflKE

] 3-4 surface =1.1

a SURFRCE a

O METflLIMNION

A BOTTOM

3-4 5-6 6-3 7-8 8-13 9-10 10-14 1 -20 DflTE 3-26 4-25 5-20 6-25 7-22 8-27 9-23 11-14 2-18 FIGURE 5.23 TOTflL PHOSPHflTE vs TI In-Lake #1 - ONOTR LRKE 0.23 11-14 surface = 0.41 c D SURFflCE 0.21 O 0.191- o METRLIMNION

BOTTOM

A

II 111 £*-* ••_•• ^^f ^^ •••••» "•! m ^BBB! • Ht •'•'• • •• "^" ••*' ' " 3-4 4-8 5-6 6-3 7-8 8-13 9-10 10-14 1-20 DflTE 3-26 4-25 5-20 6-25 7-22 8-27 9-23 11-14 2-18 FIGURE 5.24 TOTflL PHOSPHflTE vs TI In-Lake #2 - ONOTfl LflKE 0.21 o SURFflCE ] 7-8 bottom = 0.63 0.19- 20.17 [ o METRLIMNION \ A BOTTOM UJ0.13

^0.07 h- °0.05

0.03 AA , A--;6^D— 0.0! -—a ft.— Q—-6 A—- 3-4 4-8 5-6 6-3 7-8 8-13 9-10 10-14 1-20 DflTE 3-26 4-25 5-20 6-25 7-22 8-27 9-23 11-14 2-18 FIGURE 5.25

P vs 1 ^1- In -Lake #1 - ONOTfl LflKE 8.4 A BOTTOM a SURFflCE o METflLIMNION i_i_— i—ni - 8.2 \ cT \\ °l\ 8.0 n /\ n — CL / / CJ \ A 7.8 /\ \/g \P r*\ / co 7.6 r\ • J / v •+^ / .— LAJ / \ i / / ^\ 5 7.4 / * Q / ^ w >°/ \\ // -V - / x D \ / -••' / x U ^ °' f*\ A s \ '• O •°-\ ^ 7.0 - ,A ^-A' \\ s * , X '. .-' A '""O yk_ X \ x^ NO '^'o\ A'''°* *—" .AX V / * \ * (• 6.8 - ''\A'/ b' 6.6 -

6.4 1 1 1 1 1 1 i i i i i i i i i i 3-4 4-8 5-6 6-3 7-8 8-13 9-10 10-14 1-20 DflTE 3-26 4-25 5-20 6-25 7-22 8-27 9-23 11-14 2-18 FIGURE 5.26 pH vs TI In-Lake #2 - ONOTfl LflKE 8.6 SURFRCE o METRLIMNION A BOTTOM

n a n^ p

o

A

&--

A V--2-

3-4 4-8 5-6 6-3 7-8 8-13 9-10 10-14 1-20 DRTE 3-26 4-25 5-20 6-25 7-22 8-27 9-23 11-14 2-18 FIGURE 5.27 SECCHI DEPTH & TOTRL CHLOROPHYLL vs TIME In-Loke #1 (Main Basin) . 9

8 o SECCHI DEPTH A CHLOROPHYLL 10 A 9 o 8 D) E 7

« a_ Q oa: 5

o

3-4 4-8 5-6 6-3 7-8 8-13 o-io 10-14 1-21 3-26 4-25 5-20 6-25 7-22 8-27 9-23 1-14 2-19 SRMPLE DflTE ONOTfl LflKE FIGURE 5.28 SECCHI DEPTH 8 TOTRL CHLOROPHYLL vs TIME In-Loke #2 CShallow Basin) 30 SECCHI DEPTH o CHLOROPHYLL

25

B CD 20

15 a a: o

10 a: h~ o

0 4-8 5-6 6-3 7-8 8-13 9-10 10-14 1-21 4-25 5-20 6-25 7-22 8-27 9-23 11-14 2-19 SHMPLE DflTE QNDTfl LflKE FIGURE 5.29 Bacteria Counts — In Lake 1 600

500 -

400 - o o 300 - o o

200 -

100 -

0 tp tp ^— 3/22 4/8 4/25 5/6 5/21 6/2 6/24 7/9 7/268/148/289/109/2310/1611/141/21 2/19 Dote a FECAL COLJFORMS TOTAL COLJFORMS FIGURE 5.30 Bacteria Counts — In Lake 2

O C *- O o o

3/22 4/8 4/25 5/6 5/21 6/2 6/24 7/9 7/268/148/289/109/2310/1611/141/21 2/19

Date D FECAL COLIFORM TOTAL COLIFORM FIGURE 5.31 Bacteria Counts - DANIELS BROOK 80

70 -

60 -

50 - ^% E 0 S"~o c" i- O >5 40 - o o

30 -

20 -

10 -

[p tp •fi 4/8 4/25 5/6 5/21 6/2 6/24 7/9 7/268/148/289/109/2310/1611/141/21 2/19

Date n hTCAL COL1FORM TOTAL COLIFORM FIGURE 5.32 Bacteria Counts - Churchill Brook 80

70 -

60 -

50 - o c T- O 40 - o o

30 -

20

10 -

V fr ^ 4/8 4/25 5/6 5/21 6/2 6/24 7/9 7/268/148/289/109/2310/1611/141/21 2/19

Date FECAL COLIFORM + TOTAL COLIFORM FIGURE 5.33 Bacteria Counts — Dan Casey Road 1UU -

90 - • -

80 - - 70 -

^ 60 - o o 50 - o* o 40 - N*^

30 -

20 -

10 -

0 -I i i i i i i i i i i i i i i 4/8 4/25 5/6 5/21 6/2 6/24 7/9 7/268/148/289/109/2310/1611/141/21 2/19

Date D FECAL COLIFORM TOTAL COLIFORM FIGURE 5.34 Bacteria Counts — Camp Winadu 100

90

80

70

60 o o 50 o u 40

30 -

20 -

10 -

.- __ 0

4/8 4/25 5/6 5/21 6/2 6/24 7/9 7/268/148/289/109/2310/1611/141/21 2/19

Date D FECAL COLIFORM + TOTAL COLIFORM FIGURE 5.35 Bacteria Counts — Parker Brook 80

70 -

60 -

50 H o c T- O 40 ^ oo

30 -

20 -

10 -H

0 -f* •f 4/8 4/25 5/6 5/21 6/2 6/24 7/9 7/268/148/289/109/2310/1611/141/21 2/19 Date D FECAL COLIFORM + TOTAL COLIFORM FIGURE 5.36 Bacteria Counts — Blythewood Drive 80

70 -

60 -

50 - E /^ o-^ «o O C f- O >5 40 - o o

30 -

20 -

10 -

0 - 4/8 4/25 5/6 5/21 6/2 6/24 7/9 7/268/148/289/109/2310/1611/141/21 2/19

Date FECAL COUFORM + TOTAL COLIFORM FIGURE 5.37 Bacteria Counts — Oncta Brook 80

70 -

60 -i

50 - /-N _E • o9* c° «- o 40 - >o5o

30 -

20 -i

10 -

¥• •f 1 4/8 4/25 5/6 5/21 6/2 6/24 7/9 7/268/148/289/109/2310/1611/141/21 2/19

Date D FECAL COLIFORM -f TOTAL COLIFORM FIGURE 5.38 TQTRL BRCTERI vs TIM Trib. #1 (Daniels Brook) - ONOTfl LRKE 2200

2000 -

1800 - o o 1600 - \ # 1400 - en I—H 1200 - o: LLI 1000 -

£ a

3-4 4-8 5-6 6-3 7-8 8-13 9-9 9-24 11-13 1-20 3-31 DflTE 3-26 4-25 5-20 6-24 7-22 8-26 9-23 10-16 12-16 2-18 m m m FIGURE 5.39 CHLORIDE vs TIM Trib. #1 (Daniels Brook) - ONOTfl LflKE

CD E

LJLI i—Qi cr 0 _i u

3-4 4-8 5-6 6-3 7-8 8-13 9-9 9-24 11-13 1-20 3-31 DflTE 3-26 4-25 5-20 6-24 7-22 8-26 9-23 10-16 12-16 2-18 FIGURE 5.40 TOTflL SUSPENDED SOLIDS vs TIM Trib. #1 CDanie s Brook) - ONOTfl LflKE

3-4 4-8 5-8 6-3 7-8 8-13 9-9 9-24 11-13 1-20 3-31 DflTE 3-26 4-25 5-20 6-24 7-22 8-26 9-23 10-16 12-16 2-18 FIGURE 5.41 NITRRTE vs TI Trib. #1 (Daniels Brook) - ONOTfl LflKE

STORM

3-4 4-8 5-6 6-3 7-8 8-13 9-9 9-24 11-13 1-20 3-31 DflTE 3-26 4-25 5-20 6-24 7-22 8-26 9-23 10-16 12-16 2-18 FIGURE 5.42 TQTRL PHOSP TE vs TIM Trib. #1 CDanie s Brook) - ONOTfl LflKE 0.15 O.Uh

\ 0.12h a STORM E O.11 0.10 111 H0.09 JE 0.08 Wa 0.0n m7 £ 0.06 ^0.05 a 0.04 H 0.03 0.02 0.01 0.00 3-4 4-8 5-6 6-3 7-8 8-13 9-9 9-24 11-13 1-20 3-31 DRTE 3-26 4-25 5-20 6-24 7-22 8-26 9-23 10-16 12-16 2-18 FIGURE 5.43 TOTRL BRCTERIR vs TI Churchilf Brook) - ONQTfl LRKE 2200 9-23 storm = 14,360 ^ 2000h E 1800 o STORM 2 1600

* 1400h 1200

H 1000 m 800 f 600 o h 400 200 0 "3-4 4-8 5-6 6-3 7-8 8-13 9-9 9-24 11-13 1-20 3-31 DflTE 3-26 4-25 5-20 6-24 7-22 8-26 9-23 10-16 12-16 2-18 FIGURE 5.44 CHLDRID VS TIM Trib. #2 (Churchi Brook) ONOTfl LflKE

STORM

CD E v_/

LU a a _j x u

3-4 4-8 5-6 6-3 7-8 8-13 9-9 9-24 11-13 1-20 3-31 DflTE 3-26 4-25 5-20 6-24 7-22 8-26 9-23 10-16 12-16 2-18 FIGURE 5.45 TOTRL SUSPENDED SOLIDS vs Trib. #2 (Church! Brook) - ONOTR LflKE 90

80

70 STORM

60

en 40 CO 30

20

10

0 3-4 4-8 5-6 6-3 7-8 8-13 9-9 9-24 11-13 1-20 3-31 DRTE 3-26 4-25 5-20 6-24 7-22 8-26 9-23 10-16 12-16 2-18 FIGURE 5.46

T J_ TRRT VS TIM Trib. #2 (Churchi Brook) - ONOTfl LflKE

STORM

0.0 "3-4 4-8 5-6 6-3 7-8 8-13 9-9 9-24 11-13 1-20 3-31 DRTE 3-26 4-25 5-20 6-24 7-22 8-26 9-23 10-16 12-16 2-18 FIGURE 5.47 TOTRL PHOSP TE vs TIM Trib. #2 (Churchi Brook) - ONOTfl LflKE 0.15 0.14 ^ OJ3 ^ 0.12 STORM gO.ll ~ 0.10 LLJ £0.09 J 0.08 | 0.07 £ 0.06 ^0.05 a 0.04 ^ 0.03 0.02 0.01 0.00 3-4 4-8 5-6 6-3 7-8 8-13 9-9 9-24 11-13 1-20 3-31 DflTE 3-26 4-25 5-20 6-24 7-22 8-26 9-23 10-16 12-162-18 FIGURE 5.48 TOTflL BflCTERIfl vs TIM Trib. #3 (Dan Casey Rd.) - ONOTfl LflKE 300

250

O O 200

CE (— « 150 cr LU h- u CL GO 100 _J CE h- a 50

0 4-8 4-25 5-6 3-31 DRTE FIGURE 5.49 CHLORIDE vs TIM Trib. #3 (Dan Casey Rd.) - ONOTfl LflKE

3-26 4-8 4-25 5-6 3-31 DRTE FIGURE 5.50 TOTRL SUSPENDED SOLIDS vs TIM Trib. #3 (Dan Casey Rd.) - ONOTfl LflKE 20

18 16 STORM 14

CO ^Ji1o2 ffi 8

6

4

2

0 3-26 4-8 4-25 5-6 3-31 DRTE FIGURE 5.51 NITRRTE vs TIM Trib. #3 (Dan Casey Rd.) - ONOTfl LRKE

STORM

3-26 4-8 4-25 5-6 3-31 DRTE FIGURE 5.52 TOTflL PHOSPHflTE vs TIM Trib. t3 (Dan Casey Rd.) - ONOTfl LflKE 0.10 0.09h

2 0.08 STORM ^O.OTh

0.06 r— CC 0.05 C/) I0.04

0.03

0.02

0.01

0.00 3-26 4-8 4-25 5-6 3-31 DflTE FIGURE 5.53 TOTFL BflCTERIfl vs TIM Trib. #4 (Camp Winadu) - ONOTfl LflKE 350 r

300 o o 250

200 CE i—i (T 150 CJ CE DQ _, 100 cr i- o 50

0L 4-8 4-25 5-6 3-31 DflTE FIGURE 5.54 CHLORIDE vs TIM Trib. #4 (Camp Winadu) - ONOTfl LRKE 20

18- 16- STORM 2 14 \ en v-/ i—ai 0o1 Q° 5 6 4 2 0 3-26 4-8 4-25 5-6 3-31 DRTE FIGURE 5.55 TOTflL SUSPENDED SOLIDS vs TIME #4 (Camp Winadu) - ONOTR LflKE 90

801-

70 STORM

60

|,50 en 40 tn 30

20

10

0 3-26 4-8 4-25 5-6 3-31 DflTE FIGURE 5.56 NITRflTE vs TI Trib. #4 (Camp Winadu) - ONQTfl LflKE 0.9

0.8

0.7H STORM

- 0.6h \ O) E 0.5 LU I- CL 0.4 o: 0.3

0.2

0.1

0.0 3-26 4-8 4-25 5-6 3-31 DRTE FIGURE 5.57 TOTflL PHOSPHflTE vs TI Trib. #4 (Camp Winadu) - ONOTfl LflKE 0.20

0.18 20.16 STORM ^0.14 \^> wO.12 CE £0.10 cn 0.08 o_ _i CE 0.06 h- O 0.04

0.02

0.00 3-26 4-8 4-25 5-6 3-31 DRTE FIGURE 5.58 TOTRL BflCTERIfl vs TIM Trib. #5 (Parker Brook) - ONOTfl LflKE

9-23 storm = 11.210 I

3-4 4-8 5-6 6-3 7-8 8-13 9-9 9-24 11-13 1-20 3-31 DRTE 3-26 4-25 5-20 6-24 7-22 8-26 9-23 10-16 12-16 2-18 FIGURE 5.59 CHLORIDE vs TIM Trib. #5 (Parker Brook) - ONOTfl LflKE

STORM

O) E

LU ia— * a: a -j IE u

3-4 4-8 5-6 6-3 7-8 8-13 9-9 9-24 11-13 1-20 3-31 DRTE 3-26 4-25 5-20 6-24 7-22 8-26 9-23 10-16 12-16 2-18 FIGURE 5.60 TOTflL SUSPENDED SOLIDS vs TIM Trib. #5 (Parker Brook) - ONOTR LflKE 34 321- 30 28- 26- STORM 24

E w 16 tn H H12 10 8 6 4 2 0 3-4 4-8 5-6 6-3 7-8 8-13 9-9 9-24 11-13 1-20 3-31 DflTE 3 26 4-25 5-20 6-24 7-22 8-26 9-23 10-16 12-16 2-18 FIGURE 5.61 NITRRTE vs TIM Trib. #5 (Parker Brook) - ONOTfl LRKE

STORM /

f^ f^ 1 .1 . 1 1 _i_ 1.. I 1 I - --1—. • i._ - - . • 1-. -• 1 • 1 1 1 1 ' 3-4 4-8 5-6 6-3 7-8 8-13 9-9 9-24 11-13 1-20 3-31 DRTE 3-26 4-25 5-20 6-24 7-22 8-26 9-23 10-16 12-16 2-18 FIGURE 5.62 TQTRL PHQSPHRTE vs TIM Trib. #5 (Parker Brook) - ONOTR LRKE 0.16

0.141-

0)0.12 TORM

CE I &)0.08 Io ^ 0.06 CE E0.04

0.02

0.00 3-4 4-8 5-6 6-3 7-8 .8-13 9-9 9-24 11-13 1-20 3-31 DRTE 3-26 4-25 5-20 6-24 7-22 8-26 9-23 10-16 12-16 2-18 FIGURE 5.63 TOTRL BRCTERIR vs TIME Trib. #6 CBIythewood) - ONOTfl LflKE 2200 9-23 storm = 6320 ^ 2000 E 1800 o STORM ° 1600 # 1400 1200 a: in {3 1000 800 fE 600 o 400 200 0 3-4 4-8 5-6 6-3 7-8 8-13 9-9 9-24 11-13 1-20 3-31 DflTE 3-26 4-25 5-20 6-24 7-22 8-26 9-23 10-16 12-16 2-18 FIGURE 5.64 CHLORIDE vs TIM Trib. #6 (Blythewood) - ONOTfl LflKE

STORM

CD E

LU ia— i cr a _i m u

A 4-8 5-6 8-3 7-8 8-13 9-9 9-24 11-13 1-20 3-31 DflTE 3-26 4-25 5-20 6-24 7-22 8-26 9-23 10-16 12-16 2-18 FIGTJKE 5.65 TOTflL SUSPENDED SOLIDS vs TIME Trib. #6 CBIythewood) - ONOTfl LflKE

STORM

3-4 4-8 5-6 6-3 7-8 8-13 9-9 9-24 11-13 1-20 3-31 DflTE 3-26 4-25 5-20 6-24 7-22 8-26 9-23 10-16 12-16 2-18 FIGURE 5.66 NITRRTE vs TIM rib. #6 CBIythewood) - ONOTfl LRKE

STORM

"3-4 4-8 5-6 6-3 7-8 8-13 9-9 9-24 11-13 1-20 3-31 DRTE 3-26 4-25 5-20 6-24 7-22 8-26 9-23 10-16 12-16 2-18 FIGURE 5.67 TQTRL PHOSPHflTE vs TIM Trib. #6 CBIythewood) - ONOTfl LflKE 0.12

STORM

0.00 3-4 4-8 5-6 6-3 7-8 8-13 9-9 9-24 11-13 1-20 3-31 DRTE 3-26 4-25 5-20 6-24 7-22 8-26 9-23 10-16 12-16 2-18 PERCENT COMPOSITION of BENTHOS NORTH SIDE SPRING 1986

Hyalella

Other

Ch i ronom i dae Gastropoda

ONOTR LflKE FIGURE 5.68 PERCENT COMPOSITION of BENTHOS SOUTH SIDE SPRING 1986

Ch i ronom i doe

Trichoptera

Gastropoda

Other Hyalella

01 igochaeta

ONOTfl LflKE FIGURE 5.69 PERCENT COMPOSITION of BENTHOS EflST SIDE SPRING 1986 Hyalella

Other Ch i ronom i dae

H i rud i nea

01 igochaeta ONOTfl LflKE FIGURE 5.70 PERCENT COMPOSITION of BENTHOS NORTH SIDE SUMMER 1986

Hyalella

Other Trichaptera

Ephemeroptera

ONOTR LflKE FIGURE 5.71 PERCENT COMPOSITION of BENTHOS SOUTH SIDE SUMMER 1986

Hyalella

Ch i ronom i dag

Trichoptera

01 igochaeta

Nematoda Ephemeroptera Other

ONOTfl LRKE FIGURE 5.72 PERCENT COMPOSITION of BENTHOS ERST SIDE SUMMER 1986

H i rud i nea Trichoptera

Nematoda Hya!e!i a Other

01 igochaeta TurbelI aria

QNOTR LRKE FIGURE 5.73 PERCENT COMPOSITION of BENTHOS NORTH SIDE FRLL 1986 Hyalel la

Other

Trichoptera Ch i ronom i dae

ONOTR LflKE FIGURE 5.74 PERCENT COMPOSITION of BENTHOS SOUTH SIDE FRLL 1986 Hyalella

Ch i ronom i dae Other

H i rud i nea

Trichoptera 01 igochaeta

ONOTR LflKE FIGURE 5.75 PERCENT COMPOSITION of BENTHOS EflST SIDE FRLL 1986 HyaIe11 a

H i rud i nea

Tr ichoptera

Gastropoda

Ch i ronom i dae Other

ONOTfl LRKE FIGURE 5.76 6.0 MACROPHYTE SURVEY

To establish the distribution of nuisance plant species throughout Onota Lake, multiple qualitative surveys of the submerged and emergent aquatic vegetation were conducted. The species composition, relative abundance, and dominant species of the macrophyte community was determined throughout the growing season.

A quantitative assessment was conducted in August when the macrophytes reached y peak density. Biomass was determined by harvesting a series of 1 nr quadrats, twenty feet apart along transects set perpendicular to the shore line. Each transect extended into the lake to the limit of plant growth. Plants were cut at the sediment-water interface, rinsed, returned to the laboratory, oven dried at 103QC and weighed.

Ten species of aquatic macrophytes were observed throughout the lake (Table 6.1). The distribution and density of macrophytes differed markedly between the North and South Basins. The shallow North Basin is dominated by dense stands of Myriophyllum spicatum, El odea sp.» Vallisneria americana, and Potamogeton perfoliatus. A sizable stand of Nuphar advena exists at the outlet of Wadham Brook. By mid-summer, vegetative cover extends over approximately 70-802S of the North Basin.

Vegetation in the deeper South Basin is confined mainly along the shoreline, and consists almost entirely of Potamoqeton perfoliatus. The excellent clarity of Onota Lake allows this species to exist in dense stands in relatively deep waters off the shoreline (< 2.5m). Areal coverage of Potamogeton increased in the northern portion of the South Basin, particularly along the sandbar (Figure 6.1).

The total biomass and nutrient content of macrophytes which could potentially be harvested were estimated by calculating the total species coverage in the lake and conducting chemical analysis on various samples of harvested plants (Table 6.2). The nutrient analyses conducted on composites of dried plant tissue averaged 15,000 mg/kg TKN and 2000 mg/kg TP. These values show that a

ENV/L22-rpt2 6-1 Thom« PPPUIan

N-NUPHAR ADVENA FIGURE 6.1 M-MYRIOPHYLLUM SPICATUM E-ELODEASP. MACROPHYE DISTRIBUTION MAP P-PpTAMOGETON SUMMER 1986 {PERFOUATUS & CRISPUS) ONOTA LAKE V-VALJJSI«RA AMERICANA PITTSRELD, MASS. C-CHARA (ALGAB JUNE 1987 PROJECT *541648

7.5" TOPOGRAPHIC QUADRANGLE '9Cml9 DC Noi Scaie *"is Oraivng high concentration of nutrients are stored up in the macrophytes. Total potential harvestable biomass amounts to 2.1 x 105kg (dry weight). The equivalent phosphorus content of harvestable biomass is 420 kg.

Welsh, et. al. (1979) reported that macrophyte harvesting can greatly decrease the internally regenerated TP load of a lake. This could, in systems where the internal load accounts for a majority of the total load, actually improve lake trophic state. Nutrient dynamics in the littoral sediments are greatly influenced by aquatic macrophytes (Carignan, 1985). In a lake, such as Onota, where nutrient availability is low, sedimentary organic matter (dead plant tissue) may predicate and influence the density of niacrophyte growth.

Based on a review of existing data and examination of existing macrophyte distributions, it appears that certain dominant plant species in the North Basin may be spreading into the South Basin. Potamogeton perfoliatus, a commonly occurring species in the North Basin, displays a pattern of distribution which suggests that the species is emigrating from the North Basin and becoming progressively established in the South Basin. Shallow areas in the South Basin have, over recent years, supported increasingly denser colonies of this species. This is most evident in the immediate proximity of the sand bar. Elodea sp_. is another nuisance species which occurs in high density in the North Basin and is encountered sporadically in the South Basin. The spread of nuisance species from the North Basin to the South Basin may be accelerated by boat traffic, prop disturbance, and fugitive plants introduced on boat trailers.

ENV/L22-rpt2 6-3 TABLE 6.1 SPECIES LIST AND RELATIVE ABUNDANCE OF AQUATIC MACROPHYTES COLLECTED FROM ONOTA LAKE

MyHophytlum spicatum Abundant/Dominant Chara (Alga) Common Elodea sp. Abundant Vallisneria americana . Abundant Potamogeton cHspus Present Potamogeton perfollatus Abundant/Dominant Potamogeton robinsil Present Nuphar advena Present Heteranthera dubla Sparse i Nltella (Algae) Present i i i i i • • ENV/L22-rpt2 6-4 TABLE 6.2 _ TOTAL HARVESTABLE BIOMASS AND CHEMICAL CONTENT • OF MACROPHYTON IN ONOTA LAKE m Total Harvestable Area (Acres) = 132 Average Harvestable Depth (meter) = 1.5 • Average Biomass/Square Meter (grams) = 265 Total Harvestable Biomass (kg) = 2.1 x 105 U Total Phosphate -P of Total Biomass (kg) = 420 Total Kjeldahl Nitrogen of Total Biomass (kg) = 3,150

ENV/L22-rptZ 6-5 I I 7.0 SEDIMENT CHARACTERISTICS AND DEPOSITION

• Storm water characteristically transports a large amount of suspended sediment. Upon reaching the lake, the sediment particles settle and are deposited at or near their point of entry to the lake. These inflow areas may I subsequently be colonized by aquatic macrophytes or even wetland plants.

™ Historically, Onota Lake has experienced such sediment deposition primarily at _ the northern end. A review of available literature along with bathymetric • data (see Section 4) identified two major areas of sediment deposition to the east and west of Thomas Island. To further quantify the extent of the 8 accumulated, unconsolidated sediment, measurements were conducted using a graduated sediment probe which was hand-driven through the loose material to 8 the point of impenetrability. By conducting numerous measurements along a series of transects, sediment depth contours were established (Figure 7.1).

™ The sediment probes data indicate that appreciable (>0.3 meters) accumulation

M of sediment exists in the North Basin. In the South Basin, limited areas • along the southwest shore were found to have deposits of organic sediments. These sediment deposits, although sporatic, were primarily confined to the • shoreline parallel to Blythewood Drive and ranged from 0.3 to 1.0 meters in depth. The eastern and southern shore are very rocky and overlaid by sandy • sediment. North of the sand bar, sediment deposits were characteristically organic and much less sandy. Deep deposits (1.8-2.0 meters) were recorded along the northwest shoreline. Deposits 1 to 1.2 meters deep were recorded to • the north and west of Thomas Island and in Thomas Island Cove. The sediments _ in the cove included detrital material in the form of macrophyte tissue, 8 deciduous leaves, and woody material.

One potential component of a lake restoration plan is the dredging of shallow, • filled-in areas. Unconsolidated sediment depth measurements provide a good estimate of the volume of material which could be removed. Table 7.1 presents the surface areas and volumes associated with the 3 to 4 foot, the 2 foot, and the 1 foot/sediment removal contours. These data are presented in English as I opposed to metric values as this is the conventional method used to project I spoil volume and dredging costs. The locations associated with this potential I ENV/L22-rpt3 7-1 —'T-rv-i—rrr "^ •vr-°*";—np TTTr—j

\\\XXXXXXXXXXXX X X X X X . X X X X \ \ -v X \ •^\\x\x^xxxxxxxxxx^ sxxx-.

THOMAS ISLAND

• •OAT HOUM

FK3URE 7.1 4 TO 6 FEET DEPTH OF ACCUMUJVTEO 3 TO 4 FEET SECHMENT -

ONOTALAKE PITTSFIELD. MASa PROJECT No. 541648 SEPTEMBER.1988

WTEKNATIONJU TECHN ' 1904 IT CORPORATION CORPO ALL COPYRIGHTS RESERVED dredging activity are depicted in Figure 7.2. Dredging of the North Basin appears potentially feasible from a cost/benefit viewpoint. This is the only area of the lake where considerable deposits of sediment occur. A total of • 4.28 x 105 ydj3 (428,000 cubic yards) could possibly be removed from the Thomas Island section of the lake; 1.87 x 105 yd3 from an area west of Thomas Island • and 1.54 x 105 yd ^ from an area east of Thomas Island (Table 7.1). The feasibility of dredging will be reviewed in Section 11.0.

An area to the northwest of the lake, in the Wadham Brook vicinity (Figure 7.1) with substantial accumulated sediment is not identified as a potential dredging area. This is due to its importance as a fish breeding/nursery area. Currently colonized by Nuphar advena, this area provides excellent fish habitat due to cover, the rich nature of the sediments and benthic invertebrate forage.

Analysis of the chemical composition of Onota Lake sediments was conducted for samples, collected with a modified K-B free-fall coring apparatus, from the two in-lake water quality monitoring stations (Figure 5.1). Samples were retrieved, transferred to plastic bottles and returned to the IT laboratory • for analysis of: Total nitrogen Arsenic Copper Manganese Total phosphate phosphorus Cadmium Iron Zinc i Total organic carbon Chromium Lead

• The results of these analyses are presented in Table 7.2. g| Lead and zinc are among the most prevalent toxic heavy metals found in storm- water runoff (Wllber and Hunter, 1975). Lakes .with urbanized watersheds often • display elevated levels of these metals in their sediment (Koppen and Souza, 1983). The concentrations of these metals in urban runoff have been • quantitatively related to the levels of zinc in automobile tires (an average ™ of 0.73 percent) and lead in gasoline. As a result of that analysis, it has been shown, that the average deposition of these metals on road surfaces are • 0.0030 g zinc/vehicle km and 0.0049 g lead/vehicle km (Christensen and Guinn, 1979). It would, therefore, be expected that surface runoff from roads would

_ ENV/L22-rpt3 7-3 CO ^ CD */

I*D

U* ^ =Z u*j / 3* < 2 d r> G z •/

LEFT INTACT AS FISHERY HABITAT ')

s

<£D CC o

INSTALL BOX CULVERT

kFRANKLIN CONTROY REC. PAVILION

LEGEND FIGURE 7.2

AREA TO BE PROPOSED DREDGING AREA DREDGED PREPARED FOR: ONOTA LAKE PITTSFIELD. MASS PROJECT No. 541648 NOVEMBER 1987 INTERNATIONAL TECHNOLOGY < 1964 IT CORPORATION ALL COPYRIGHTS RESERVED *C«t f««t EB CORPORATION Do Noi Seal* This Drawing' I I have substantial levels of these metals. Both of these metals were detected in the Onota Lake sediment samples at concentrations considered fairly high. I Comparison of the Onota Lake data with sediment data from a representative New Jersey lake with a substantially undeveloped watershed and no motorboat I activity (PAS, 1987) reveals that lead and zinc sediment concentrations are an order of magnitude greater in Onota Lake than in the New Jersey lake.

I Cadmium, like zinc, is a component of automobile tires (Owe, et. al., 1982) which may be transported into the lake with road surface runoff. The I deterioration of galvanized pipe is another source of cadmium in the environment (APHA, 1983). Both cadmium and zinc were detected in low I concentration in the sediments samples. Chromium and manganese, components of automobile bumpers and trim (Sartor and I Boyd, 1972), were detected at elevated concentrations, particularly at Station #2. The proximity of major roads to this shallow basin is likely a major I cause of this metal contamination. I A common material in electrical and plumbing applications, as well as in various pesticides and herbicides, copper was detected in the lake sediment I samples but at low concentrations. The measured values in the North and South Basins (52 and 49 ppm, respectively) are well within DEP limits stated in 310 I CMR 32.00. Arsenic compounds were often utilized as a pesticide in agricultural I applications. Arsenic is very persistent and the historical use of arsenic based compounds may often be reflected in sediments even after a considerable I amount of time. A concentration of 25 ppm arsenic is often used as an arbitrary environmental indicator of compromised condi11ons. The measured I levels in Onota Lake (20 ppm) may require special attention be given to the method of sediment removal and disposal if dredging 1s conducted in the North Basin. At a minimum, additional testing is recommended to better define the I extent of arsenic contamination. Additional sediment analysis should include EP Toxicity'testing to quantify the potential leachability of this material if I distributed. I I ENV/L22-rpt3 7-5 The most abundant metal measured In the sediment core samples was iron. However due to its ubiquitous nature in the environment, high levels of this metal are a natural occurrence.

Due to their role as nutrients that often accelerate the productivity of lake waters, both nitrogen and phosphorus are considered important components of lake sediments (Wetzel, 1983). Phosphorus, and to a lesser extent nitrogen, can be present in lake sediments at concentrations several orders of magnitude greater than in the water column. For the sediment samples analyzed from Onota Lake, the TP concentrations ranged from 2,800 to 3,700 mg/kg and TN concentrations from 9,900 to 16,000 mg/kg. Under anoxic, reducing conditions these nutrients can be liberated from the sediments and recycled to the overlying water. The internal recycling of total phosphorus from the sediments can have a marked influence on the productivity of the lake, particularly if it is a major component of the lake's total phosphorus budget.

Total organic carbon (TOC), a measure of the amount of carbon based material in the sediment, was high for samples from both the deep and shallow basins. TOC concentrations expressed in dry weight were 230,000 mg/kg and 290,000 mg/kg for the South and North Basins, respectively. Concentrations of this magnitude are common for lake sediments. Silt, algae and other organic niaterials settle at the bottom of the lake where they accumulate and lend a highly organic quality to the. sediments. Sediments from the deep profundal area of a lake are rarely disturbed. As a result, over time, these sediments become characteristically organic.

In Section 10.0, detail is provided relative to the annual influx of sediment via each of the lake's main tributaries. Based on unit area loading estimates, (as developed in Section 10.1 - 10.6), the annual sediment influx per tributary 1s as follows:

Churchill Brook - 121,830 kg/yr , Daniels Brook - 255,270 kg/yr Parker Brook - 275,935 kg/yr

Total annual sediment loading to the lake 1s calculated to be 1,080,118 kg.

ENV/L22-rpt3 7-6 I I TABLE 7.1 SURFACE AREA AND VOLUME OF MATERIAL TO BE DREDGED FROM ONOTA LAKE

I Depth of Surface Volume I Sediment Removal Area Dredged I West of Thomas Island 3-4 foot 33.1 acres 1.87 x 105 yd3 2 foot 14.0 acres 4.51 x 104 yd3 I 4 3 1 foot 16.0 acres 2.58 x 10 yd

I Sum 2.58 x 105 yd3

I East of Thomas Island I 3-4 foot 27.3 acres 1.54 x 105 yd3 2 foot 10.3 acres 1.55 x 104 yd3

I 5 3 Sum 1.70 x 10 yd

I TOTAL 4.28 x 105 yd3 I I I I I I I ENV/L22-rpt3 7-7 I TABLE 7.2 ONOTA LAKE SEDIMENT I SAMPLE ANALYSES OCTOBER 16, 1986 Classification of Dredge Material I 314 CMR 9.QQ Category Category Category I Parameter* Station #1 Station #2 One Two Three I Total Nitrogen 9,900 16,000 TOC 230,000 290,000 I Phosphate-P (Total) 2,800 3,700 Arsenic 20 12 10-20 >20 I Cadmium 1.4 0.83 <5 5-10 I Chromium 15 53 <100 100-300 >300 Copper 49 52 <200 200-400 >400 I Iron 43,000 53,000 I Lead 220 160 <100 100-200 >200 Manganese 690 1,600 I Zinc 260 210 <200 200-400 >400 I *Note: All results expressed as mg/kg dry wt. I I I I I I ENV/L22-rpt3 7-8 8.0 WASTEWATER DISPOSAL PRACTICES

Properly operating onslte wastewater disposal systems usually contribute very little to the nutrient budget of lakes (Lee, et. al. 1978). In contrast, faulty systems contribute a significant nutrient load which can stimulate the development of aquatic primary producers to nuisance densities (Kerfoot, 1979). Total phosphorus loads associated with domestic wastewater (kitchen, toilet, bath and laundry) can be as much as 1.5 kg/capita/yr (Ligman, et. al, 1974). Faulty septic waste disposal systems located close to the shoreline could therefore have a serious impact on the trophic status of the lake.

In many lakefront developments, failing septic systems are often attributed to oversaturation of the leach field due to a seasonal elevation of water table height, hydraulic overloading of the system due to overcrowding of vacation homes and conversion of seasonal dwellings to permanent dwellings, or the formation of clogging mats in the leach field which significantly impair wastewater percolation. These problems are often compounded by installation of systems of improper design or capacity and by use of-septic systems in areas with soils of poor absorption quality. All the above conditions reduce the ability of the soils to properly remove nutrients by sedimentation, absorption, filtration or biochemical oxidation (Otis, 1979). The area in immediate proximity to the lake's shoreline will also be more sensitive to these conditions due to water table height and limited soil depth available for bacterial degradation and soil absorption (Kerfoot, 1980). This may result in the discharge of waste-contaminated ground water plumes from these shoreline dwellings and the localized elevation of sediment-nutrient concentrations. In turn, by reducing or eliminating such discharges, noticeable Improvements in water quality and substantial reduction in localized plant growth can be garnered (Otis, 1979).

Septic leachate entering the lake carries with it dissolved nitrate, ammonia, phosphate, and organic substances. These nutrients stimulate the growth of bacteria, algae and aquatic macrophytes, which in turn may effect the recreational and potable use of the water.

In order to determine the importance of septic systems on nutrient loading to

ENV/L22-rpt3 8-1 I I the lake, a portable fluorescence - conductivity meter, commonly called a "septic snooper", was used to detect improper wastewater discharge. The I discharge, referred to as a septic plume, is caused by the faulty operation of onsite disposal systems, and results from the active emergence of septic waste contaminated ground water into the lake (Kerfoot, 1980). Under such I conditions, the septic effluent has not had sufficient time to percolate through the soils, and is usually characterized by elevated organic and I inorganic constituents. I As water is pumped through the "septic snooper", conductance and fluorescence are monitored continuously. In principle, wastewater effluent is partly I comprised of a mixture of near UV fluorescent organics derived from laundry whiteners, surfactants and natural degradation products, and conductive inorganics, such as chloride (Cl~) and sodium (Na+). By monitoring these I parameters in the form of fluorescence and conductance, a leachate plume can be detected as it emanates from the shoreline. This results in three general I conditions: I 1. Elevated fluorescence 2. Elevated conductance I 3. Elevated fluorescence and conductance The third condition is indicative of septic contamination whereas the other I two may indicate "grey water" contamination, ground water intrusion, or discharge from streams, bogs, or marshes. At those sites where both I fluorescence and conductivity were elevated, water quality and bacteriological samples were collected. Analysis of these samples helped verify or refute the I existence of a septic plume. I A septic leachate survey was conducted along the entire shoreline of Onota Lake on August 13 and 14, 1986. A number of septic leachate plumes were encountered in the survey of Onota Lake (Figure 8.1) The highest meter I readings were located primarily along the Blythewood Drive shoreline and in the cove to'the east of Thomas Island. The plume detected in both these areas I was broad and non-distinct. More localized plumes were detected in the cove I adjacent to the peninsula at the south end of the lake, and adjacent to the I ENV/L22-rpt3 8-2 I I Franklin H. Controy Recreation Pavilion. Both the Blythewood Drive and Thomas Island Cove plumes represented broad areas of diverse contamination, the latter condition being exacerbated by the restricted flow in the cove. The other two suspected plumes were more discrete in nature representing a more distinct source of septic leachate infiltration, perhaps indicative of individual septic failures encountered.

A number of water samples were collected to verify whether the observed readings were septic related. A total of 12 water samples were collected during the Septic Snooper Survey and analyzed for total coliform bacteria, fecal coliform bacteria, nitrate-nitrogen, and chloride. The results of these analyses are presented in Table 8.1.

The parameter most substantially elevated in all samples was total coliform bacteria. While bacterial contamination is evident, it can not definitively be attributed to septic leachate as all August in-lake water quality samples displayed elevated total coliform concentrations (see Section 5.1.10). The one exception was the sample collected from Thomas Island Cove. This sample result suggests the presence of sewage contamination. The homes in this area are serviced by city sewer rather than individual septic systems, therefore, the problem was not attributable to septic leachate intrusion. Further investigation of the problem following a discussion with Pittsfield DPW officials disclosed a break in the sewer line along Pecks Road.

The broad plume along Blythewood Drive is indicative of general conditions of malfunctioning or improperly designed septic systems. It underscores the need for widespread system maintenance or sewering in this section of the watershed. Throughout this study, and as documented in previous reports, the septic problem along Blythewood Drive has been identified as a problem i requiring immediate attention. The contribution of septic loading to the total nitrogen and phosphorus budgets is small relative to non-point source inputs (Section 10). However, as this section of the lake becomes i increasingly developed the contribution will become increasingly significant; This area represents one of the few sections of the watershed likely to be developed 1n the future. Immediate action is needed to protect the lake from future septic related perturbations.

ENV/L22-rpt3 8-3 I I I I I I I I 1 I I I I I

I LEGEND: FIGURE 8.1 SAMPLE LOCATION AREAS OF POTENTIAL SEPTIC I LEACHATE EXCURSION INTO AREA OF BROAD CONSISTENTLY ONOTA LAKE HIGH READINGS I PREPARED FOR: I AREA OF MORE CONFINED ONOTA LAKE PITTSFIELD, MASS. HIGH READINGS PROJECT No. 541648 SEPTEMBER 1988 2000 INTERNATIONAL I SOURCE: TECHNOLOGY U.S.aS. 7.5' TOPOGRAPHIC QUADRANGLE f«et GDI CORPORATION I Do Not Scale This Drawing CO O CNJ * * O • O

I r-l q- CSJ un ^H V O O • CO m I i-H CD i—> o o — I in ^yo m ® g O . T— < LJ_ 2 0 o O o^ml ^ ^ i I V "Z. c CO o ID Qj o ^— *r- tif\. +-• •* E ^d CO O to in id o ^3- o (O O I i~H PO • in ^~ o •M C C v41 1 u^^ i— t— i~- r— (J -M E £ T- tO M- -M O O 4- c/1 • o o 3 (J «-l t—1 o> T- **- I 0 O O E E (J 0 • i— i =*= =»: z • =3 CO 0£ o. , J- •— i— Q> O O to u_ i y -M (J t_> I— Csj OJ OO QJ Csl LU QJ I (O (O (O 1 1— 00 J- L) 4-> CO (d (U O o _J •— « 1 Q. U_ h— z U -K ^-* I Sewering of remaining areas which rely on the use of individual on-site septic systems is a major concern of residents and lake users. This is supported by the results of questionnaires and public comment at numerous lake restoration meetings. Individuals who Tive on Blythewood Drive are particularly anxious • to have this area sewered. Although sewering will alleviate the identified ™ problem of septic leachate contamination of the lake, it may encourage additional development. Heavy development of many of these areas could pose • as great an impact to lake water quality as faulty septic systems. Areas such as Blythewood Drive have shallow soils, a high seasonal water table and relatively steep slopes. To combat intense development of such areas following sewering, stricter zoning and new ordinances (steep slope, sensitive environment, etc.) may be warranted. This is discussed further in Section 11, the Restoration/Management Plan.

To further facilitate the examination of current wastewater disposal practices, a questionnaire was distributed to all residents living within 300 • meters of the lakeshore (Table 8.2). Emphasis was placed on canvassing areas of the lake which are non-sewered. Three hundred fourteen (314) questionnaires were distributed and ninety-five (95) were returned. Since the questionnaire also addressed recreational use of the lake, some questionnaires were distributed to residents living in areas of the watershed which are currently sewered.

The only densely populated area of the watershed not currently sewered is the Blythewood Drive section, located along the extreme southwest shoreline of the i lake. There are approximately 30 dwellings along Blythewood Drive, most of which are within 100 feet of the shoreline. Thirty-three of the survey i respondants used a septic system for wastewater treatment and disposal. The survey results suggest that 60% of these dwellings are utilized year round. Average occupancy appears to be 3-5 residents. Only 2Q% of those using a septic system had to alter their systems, but most alterations (63%) were due i to malfunctioning disposal fields and pipes. _ There were .also reported problems with odor and sewage backup. A small number • of residents on septic systems admitted to not having their tanks pumped at all. This cou.ld lead to odor and sewage backup, and Insufficiently treated •I _ ENV/L22-rpt3 8-6 leachate reaching the lake. The majority of respondants however appeared conscious of the problems caused by faulty septic systems and reported their tanks pumped every one to two years.

I I I

ENV/L22-rpt3 8-7 TABLE 8.2

SUW1ARY OF RESPONSES TO ONSITE WASTEWATER DISPOSAL SYSTEM

Number Distributed - 314 i Number Returned - 95 TYPE OF ONSITE WASTEWATER ALTERATION OR IMPROVEMENT DISPOSAL SYSTEM (PERCENT) OF SYSTEM (PERCENT) ALTERATION (PERCENT) • Septic Tank 35% None 32% Enlarge 26% Cess Pool 2% Altered/Improved 20% Malfunction 63% Sewer 61% No Response 48% Both No Response or Did Not Know 2%

TYPICAL PROBLEMS WITH CAUSE OF PROBLEMS SYSTEM (NUMBER)* WITH SYSTEM (NUMBER)* None 40 High Water Table 6 Odor 5 Slowly Permeable Soils 5 Backup Into House 4 Other 2 No Response 46 No Response 83

*Some responded to more than one category

ENV/L22-rpt3 8-8 TABLE 8.2 SUWARY OF RESPONSES TO ONSITE WASTEWATER I DISPOSAL SYSTEM (Continued) Number Distributed 314 I Number Returned 95

LOADS OF LAUNDRY DETERGENTS USED (NUMBER)* I PER WEEK (PERCENT) LAUNDRY DISH 1-5 429S Tide - 14 Dawn 4 I 6-10 12% Bold 1 Joy 2 11-15 6% Cheer 3 Ivory 9 16-20 3% Arm & Hammer 5 Palmolive 12 I Oxydol , 1 Cascade 4 Solo 2 #? 1 Wisk 14 Ajax 4 Dash 1 Dove 2 All 4 Octagon 1 Era 3 Sunlite 5 Fab 1 Calgon 3 I Fresh Start 1 Electrosol 3 Cold Power 1 All 1 Ivory Snow 2 Any Brand 5 I Ajax 2 I Any Brand 6 I *Some responded to more than one category.

I I I I ENV/L22-rpt3 8-9 I

• 9.0 HYDROLOGIC BUDGET

Tributary inflow, surface runoff, precipitation, evaporation, and ground water • infiltration data were used to calculate the volume of water annually entering and leaving the lake. Hydrologic loading due to precipitation falling M directly onto the lake, and normalized stream flows were calculated on the ™ basis of the historical rainfall data. In this way, the hydrologic budget ^ would be more amenable for use in the long term management of the lake. The Ui methodology and results of the hydrologic budget are presented in detail in the following sections.

9.1 PRECIPITATION/EVAPORATION

I The NOAA thirty year average precipitation data, as recorded at the Stockbridge, Massachusetts monitoring station were compared to rain gage • monitoring data recorded during 1986-1987 by LOP A volunteers. The NOAA data _ were disparate with the majority of the 1986-1987 rain gage data for those K months where simultaneous collections were available for comparison (Table 9.1). Noticeable differences occurred in June, July and August, 1986, when • unseasonably greater rainfall than the thirty-year average was measured (approximately 2 times greater than normal). In April and May, 1986 rainfall • was significantly less than the thirty year average. The hydrologic load associated with precipitation onto the lake was calculated using the monthly 30 year average data for the annual contribution. The sum of the monthly § averages was determined to be 1.11 m/yr (Table 9.1). Based on this value and fi ? — a total lake area of 2.5 x 10°m , the annual contribution of precipitation • falling directly on the lake would be calculated as:

fl 1.11 m/yr x 2.5 x 106m2 = 2.78 x 106m3/yr

fl| Evaporative losses from the surface of the lake were determined using the isopleths developed by Hely, et. al. (1961). This method entails the use of USGS pan .evaporation rates. Based on the geographical and limnological • characteristics of Onota Lake, an evaporation coefficient of 75 cm/yr was _ selected. The hydrologic loss from the lake's surface caused by evaporation

_ ENV/L22-rpt3 9-1 I is calculated as:

• 0.75 m/yr x 2.50 x 106 m2

= 1.88 x 106 m3/yr •

Net contribution of precipitation falling directly onto the lake's surface when corrected for evaporative losses would be 9.0 x 10 m/yr. •

9.2 TRIBUTARY INFLOW

• Hydrologic inputs from the ten sub-basins were determined using an empirical precipitation-stream discharge formula (Dunn and Leopold, 1978) and the thirty • year average monthly precipitation data Table 9.2. The empirical stream discharge formula is given as:

i Q = CIA (0.278)

H where C = runoff coefficient I = rainfall intensity (mm/hr) A = area of sub basin (km?) I Q = stream flow (m3 /sec)

~ This approach was necessitated, as much of the surface inflow to the lake is from small streams, swales, and diffuse non-gagable sources. The inflow data • obtained by this method is a reasonable estimate, comparable, in terms of inherent error, with measured discharge methods (Scheider, et al., 1979). ( Inflow to the lake was calculated for each of the subbasins on a monthly basis (Table 9.3). Summation of the monthly inflow calculations was used to • estimate tributary influx. These data were checked against those months when runoff was normal (as based on comparison of measured and historical fl| rainfall). Agreement was deemed acceptable. Tributary inflow was calculated •• to be 5.50 X 106 lAr'1- i • ENV/L22-rpt3 9-2 9.3 OUTFLOW

• Outfow from the lake were monitored routinely by LOPA volunteers. A staff gage was erected at the spillway and was calibrated on 13 dates by IT flj| personnel (Table 9.4). The calibration points were utilized to develop an ™ outflow regression equation (Table 9.5). The resulting regression equation, Y = 78.99 x -12.24, was used to calculate the annual outflow volume. A • Spearman's correlation coefficient (r = 0.88) supported this methodology. The mean gage height as calculated from the data collected by IT personnel was • 0.504 feet or 0.155 m (Table 9.4). Applying this value to the equation results in a discharge of 27.07 ft3/sec or 0.766 m3/sec (Table 9.5). Using • this value, the annual discharge volume was calculated to be: I 27.07 ft3/sec x 3.15 x 107 sec/yr = 8.53 x 108 ft3/yr I or 24.1 x 106 m3/yr.

During the 1986 study year, a short term drawdown was conducted for dock/shoreline repair. The lake was drawn down approximately 3 feet (Sanginelli, personal communication, 1990). This volume reduction was accounted for by calculating the total volume lost during drawdown which would have been lost via the outflow. The volume lost during drawdown would be:

3 feet x 617 acres • = 1851 acre-feet

V or

M 2.28 x 106m3

_ Correcting .for this excess release, the annual outflow volume was calculated • as: I I ENV/L22-rpt3 9-3 I 24.1 x 106m3 - 2.28 x 106m3

£ = 21.8 x I06m3/yr

• Based on this analysis, outflow from Onota Lake totaled 21.8 X 106 m3/yr.

• 9.4 GROUND WATER

There are two components to ground water inputs: that which feeds directly • into the lake and that which seeps into tributaries which in turn feed the lake. Ground water which seeps directly into the lake may occur at discrete • points (springs) and at diffuse sources along the shoreline or in the littoral zone. Ground water which seeps into tributaries constitutes part of the H surficial aquifer. It represents precipitation which has infiltrated the overlying strata and has exceeded soil moisture needs and vegetative evapo- transpiration loss. Such inputs are usually greatest during spring and fall • because evaporation losses and water uptake by plants are low. Winter (1979) •discusses the role of these different ground water flows and their interaction in a lake's hydrologic budget. In areas of considerable seeps and fractured bedrock geology, such as the , an appreciable amount of this • recharge feeds tributaries (Posten, 1982). To account for this potential component of the hydrologic budget, the ground water runoff methodology of • Posten (1982) was applied to those sub-basins which drain to Onota Lake via a stream. For those sub-basins where drainage to the lake is via surface runoff, ground water contributions to the lake were assumed to result from • seepage along the shoreline or nearshore littoral areas. It was also assumed

M that ground water recharge would occur primarily only during those months when • soil moisture needs were low, plant uptake minimal and evaporation inconsequential, typically November through April (Dunne and Leopold, 1978). • This assumption seems to apply to streams in the Onota Lake watershed. In 1986, even with the unseasonably large amount of rain, stream flow 1n three of • the four monitored tributaries was low from June through August. Based on a rate of groundwater seepage of 7.2 X 102 m3/d/km2 (Posten, 1982), applied to sub-baslns/III, IV, V, VI, VII, and VIII for 210 days (November through i April), a recharge value of 5.16 X 106 m3/yr was obtained. This represents the volume of ground water which seeps into the lake's tributaries (Table

ENV/L22-rpt3 9-4 I

I 9.6).

Ground water contributions resulting from diffuse seepage into the littoral I area of the Take and discrete springs were estimated as the difference between • total inflow and total outflow. Based on this approach, direct ground water influx to the lake is calculated to be 3.60 x 10 m . Unconfirmed reports suggest there is substantial spring activity in the lake's South Basin. • Attempts were made to quantify this on two occasions: late September and mid- ^ November, 1986. Seepage meters constructed from 20-gallon metal trash cans J| (Lee, 1977) were driven into the substrate at three locations suggested to IT personnel where spring activity was suspected. The three monitored stations H were along the southeast shoreline, within 50 feet of shore. The substrate was coarse sand to gravel and water depth ranged from 6 to 10 feet. .After 24 • hours the meters were inspected by divers and on both dates very little, if any, influx of water had occurred.

• An additional evaluation of the ground water, loading estimate was developed — through comparison of the Onota Lake estimate and a ground water influx value I mathematically generated for Lake Buel, Montery/New Marlborough, Massachusetts (IEP, 1982). Lake Buel and Onota Lake are in the same physiographic province, • are of the same geological history --and share similar soil types. The shoreline of Lake Buel is approximately 17,400 feet, which is 0.455 the length of Onota. . The ground water favorability study and permeability analysis of • Lake Buel revealed annual ground water inflow to amount to 3.2 to 3.8 X 106m3. As Onota's shoreline is 2.2 times that of Lake Beul, an annual ground i water inflow of 7.04 to 8.4 X 10fi n ^r would be predicted using the same ^ approach as 1n the Lake Beul study. The predicted Onota Lake ground water || contribution from direct seepage (3.6) is 8.7 X 106nryr. This value is similar to that obtained using the Lake Buel methodology. As such, the ground • water influx to Onota Lake as computed in this study was deemed reasonable.

• 9.6 HYDROLOGIC BUDGET

Summing the' various components of the hydrologlc budget yields an annual net 6 3 • water balance of 4.8 X 10 m /yr (Table 9.7). Though this number is negative, i it does not predict annual losses of water from Onota Lake. It is more i ENV/L22-rpt3 9-5 1

H artifactual, based on the assumptions of the model. One reason for this difference is that inflow parameters were based on 30 year averages whereas • outflow was based on 10 months of measurements, during which at least 3 months had a higher than average input via precipitation. Of greater significance is • the linear relationship assumed for lake height and flow rate. The relationship is non-linear, being exponential and asymptotic at low lake i levels. The assumptions made overestimate annual outflow. Given a lake volume of 15.98 X 10fi° nrO, the annual flushing rate is 1.28 • times/year. The inverse of the flushing rate, residence time is 0.78 years. • Areal water load (qs) is an important function in the determination of phosphorus retention. The ratio of annual lake inflow to lake surface area • (Q/SA) is used to compute a qs. For Onota Lake, qs was calculated to be 6.8 I m/yr.

l ENV/L22-rpt3 9-6 I I TABLE 9.1 NOAA AVERAGE AND MEASURED PRECIPITATION I RECORDED FOR ONOTA LAKE I TOTAL MONTHLY PRECIPITATION 30 YEAR MONTHLY AS MEASURED I MONTH AVERAGE (NOAA DATA) DURING STUDY I (INCHES) (INCHES) JAN 3.17 I FE8 2.71 MARCH 3.51 APRIL 3.94 0.25 I 3.77 1.87 MAY JUNE 3.69 7.83 I JULY 3.89 8.28 AUG 4.27 6.33 I SEPT 3.95 2.14 OCT 3.40 3.18 I NOV 3.81 I DEC 3.62 I I I I I I ENV/L22-rpt3 9-7 I TABLE 9.2 TRIBUTARY AND SURFACE RUNOFF I CONTRIBUnONS TO THE ANNUAL HYDRCLOGIC BUDGET 1 GFUKEONOTA* 1 MDN1H.Y RUNOfF WL1ICS IN M3

1^p SUB- RUNOFF BASIN** COEF JAM FEB WRCH APRIL WY JUNE JULY PUG SEPT OCT MOV DEC

I 0.5 7.31 X 105 6.92 X 105 8.1 X 105 7.8 X 104 7.45 X 104 7.3 X 104 7.7 X 104 8.4 X 104 7.8 X 104 6.75 X 104 7,55 X 104 7.15 X 104

II 0.4 3.25 X 10? 3.85 X id5 3.60 X 1C? 3.46 X id* 3.32 X id* 3.24 X id* 3.42 X 104 3.74 X 104 3.46 X id* 2.99 X 104 3.35 X 104 3.18 X id*

I^F 4 4 4 III 0.3 7.29 X 10 6.90 X 10 8.08 X 10 7.77 X 103 7.44 X 103 7.29 X 103 7.68 X 103 8.4 X 103 7.77 X 103 6.72 X 103 7.53 X 103 7.14 X 103

I IV 0.3 7.55 X 104 7.12 X 104 8.33 X 104 8.01 X 103 7.68 X 103 7.50 X 103 7.92 X 103 8.64 X 103 8.01 X 103 6.93 X 103 7.74 X 103 7.35 X 103

I V 0.2 1.43 X 105 1.36 X 105 1.59 X 105 1.53 X 104 1.47 X 104 1.43 X 104 1.51 X 104 1.65 X 104 1.53 X 104 1.32 X 104 1.48 X 104 1.41 X 104

1 VI 0.25 2.02 Xl# 1.92X1# 2.24 X vf 2.16 X 105 2.07 X 105 2.02 X 105 2.14 X 105 2.34 X 105 2.16 X 105 1.87 X 104 2.09 X 1Q5 2.00 X 1Q5

1 VII 0.25 7.27 X 105 6.88 X 105 8.06 X 105 7.75 X 104 7.42 X 104 7.25 X 104 7.65 X 104 8.38 X 104 7.75 X 104 6.7 X 104 7.4 X 104 7.12 X 104

1 VIII 0.25 1.49 X 1# 1.40 X 106 1.42 X 10? 1.59 X 105 1.52 X 10? 1.49 X 105 1.57 X 10? 1.72 X 105 1.59 X 1# 1.37 X 10? 1.54 X 10? 1.46 X 1C? I IX 0.6 3.39 X 105 3.21 X 105 3.23 X 104 3.62 X 104 3.47 X 104 3.39 X 104 3.58 X 104 3.91 X 104 3.62 X 104 3.13 X 104 3.50 X 104 3.32 X 104 1 X 0.6 4.33 X 10? 4.09 X 10? 4.12 X 104 4.61 X 104 4.42 X 104 4.32 X 104 4.56 X 104 4.98 X 104 4.61 X 104 3.98 X 104 4.46 X 104 4.26 X 104 1 TOTAL 6.35 X 106 6.09 X 106 6.06 X 105 6.78 X 105 6.50 X 105 6.35 X 105 6.71 X 105 7.34 X 105 6.78 X 105 4.15 X 105 6.56 X 105 6.25 X 105.

* Calculated using rational formjla (Dunne and Leopold, 1978), based on 30 year ave rainfall. ** Refer to figure 4.1 for location of sub-basins.

1 ENV/L22-rpt3 9-8 TABLE 9.3 VOLUME OF PRECIPITATION AND VOLUME OF EVAPORATIVE WATER LOSS FROM THE SURFACE OF ONOTA LAKE

Surface area of lake 2.5 X 106m2 Precipitation on lake's surface* 2.75 X 106m3 Evaporative loss** 1.875 X 106m3

ANNUAL TOTAL NET GAIN 0.875 X 106m3 DUE TO DIRECT PRECIPITATION

* Based on historical thirty year average rainfall data ** Based on a pan evaporation coefficient of 0.75 m/yr

ENV/L22-rpt3 9-9 I TABLE 9.4 FLOW/STAFF GAGE RELATIONSHIP I FOR THE OUTFALL (SPILLWAY) OF I ONOTA LAKE

Calibration Flow Staff Gage Height (Date) Cftl/s) (ft)

4-8-86 84.1 0.85 I 5-6-86 11.5 0.26 6-3-86 11.9 0.15 I 6-24-86 29.2 0.62 7-22-86 7.4 • 0.20 8-13-86 26.0 0.52 I 9-9-86 2.3 . 0.15 9-24-86 15.6 0.41 10-16-86 18.6 0.50 11-13-86 73.4 1.05 12-16-86 . 23/8 0.60 1-20-87 25.7 0.66 2-18-87 26.2 0.58

N = 13

x 0.155

5D+/- = 0.0834

ENV/L22-rpt3 9-10 I

I TABLE 9.5 RESULT OF LINEAR REGRESSION 1 CONDUCTED ON FLOW/STAFF GAGE DATA RECORDED AT ONOTA LAKE I OUTFALL (SPILLWAY) Y = 78.99 x -12.42 Staff* 1^WP Gage Ht Flow** I X value predicted Y Population 95% C.L.*** Individual 95% C.L. 0.2000 3.3759 -7,8798 14.6316 -25.3379 32.0897 0.3000 11.2750 1.9723 20.5776 -16.7309 39.2809 0.4000 19.1740 11.2870 27.0611 -8.3940 46.7421 I 0.5000 27.0731 19.7459 34.4003 -0.3400 54.4862 0.6000 34.9722 27.1627 42.7817 ' 7.4262 62.5181 0.7000 42.8713 33.7003 52.0422 14.9088 70.8337 1^p 0.8000 50.7703 39.6779 61.8627 22.1202 79.4205 0.9000 58.6694 45.3355 72.0034 29.0791 88.2597 1 1.0000 66.5685 50.8089 82.3281 35.8089 97.3281 1.1000 74.4676 56.1714 92.7638 42.3344 106.6008 1.2000 82.3666 61.4631 103.2701 48.6806 116.0527 1.3000 90.2657 66.7077 113.8237 54.8712 125.6602 I 1.4000 98.1648 71.9194 124.4101 60.9276 135.4020 1.5000 106.0639 77.1074 135.0203 66.8686 145.2591 1.6000 113.9629 82.2778 145.6481 72.7108 155.2151 I^•i 1.7000 121.8620 87.4347 156.2894 78.4681 165.2560 1.8000 129.7611 92.5811 166.9411 84.1525 175.3697 1.9000 137.6602 97.7192 177.6011 89.7742 185.5462 2.0000 145.5593 102.8507 188.2678 95.3416 195.7769 2.1000 153.4583 107.9766 198.9400 100.8620 206.0547 2.2000 161.3574 113.0981 209.6167 106.3415 216.3733 2.3000 169.2565 118.2157 220.2972 111.7852 226.7278 •2.4000 177.1556 123.3302 230.9809 117.1975 237.1136 -'' 1^•1 * Height of water at outfall in feet ** Flow in ft3/s 1 *** Confidence limits of predicted flow 1 ENV/L22-rpt3 9-11 I

TABLE 9.6 GROUND WATER RECHARGE TO TRIBUTARIES OF ONOTA LAKE

Ground water** Sub* Sub-Basin Loading Number*** Recharge Vol 2 3 2 Basin Area (km2) 10 m /d/km Of Days X loV/yr

III 2.59 7.20 210 0.392 IV 2.67 7.20 210 0.404 V 7.65 7.20 210 1.16 VI 8.65 7.20 210 1.31 VII 3.10 7.20 210 0.464 i VIII 6.35 7.20 210 0.960 i TOTAL 4.69 X 104m3/yr * Applied only to those sub-basins with a runoff coefficient (R) < 0.4, in those sub-basins with R >0.4, the % impervious area is probably too great to allow for significant infiltration.

** Posten, 1982

*** November - April (210 days), that time of year when there exists enough soil moisture to allow for recharge. i — ENV/L22-rpt3 9-12 I I TABLE 9.7 I HYDROLOGIC BUDGET OF ONOTA LAKE

I Inputs i 106—m3/yr Tributary and Surface Runoff Inputs 7.36

Precipitation directly on lake surface • corrected for evaporation 0.875

Ground water recharge and subsequent * • influx of surplus to tributaries , 0.0469

| • Direct seepage of ground water to ** lake. 8.70

TOTAL INPUT BUDGET 17.0

Outputs

Total Annual Outflow 21.8 -4.8***

* Represents ground water recharge which occurs only during those periods • when so i1 moi sture demands exceeded. Based on 1oad i ng coeff i c i ent established by Posten (1982).

I ** Difference between outflow and inflow. i• *** Represents the net loss of water volume annually — ENV/L22-rpt3 9-13 10.0 NUTRIENT LOADING AND LAKE TROPHIC STATE ANALYSIS

10.1 SEDIMENT/NUTRIENT BUDGET

An important step in the development of a successful lake restoration action plan is the accurate calculation of the sediment and nutrient budgets. These data provide insight relevant to the trophic dynamics of the lake. The nutrient budget is the quantification of the various sources of nitrogen and phosphorus. The amount of phosphorus and nitrogen annually contributed to the lake is referred to as the annual load. The magnitude of the annual load will greatly influence the productivity of the lake. The nutrient budget therefore is an important determinant of lake trophic state. The sediment budget provides an estimate of the annual influx of eroded soils, occurring naturally and as a result of human activity. Nutrients display a high affinity to soil particles, particularly phosphorus which adsorbs to the surface of particulate material. As well as serving as a source or. vehicle for nutrients, sediment contributions may increase turbidity, decrease flood storage volume, create deltas which are in turn colonized by macrophytes and alter flow and flushing patterns in lakes.

Nutrients and sediments may be of natural sources, anthropogenic sources, or natural sources exacerbated by human activity. In any case, it is important that all major sources be identified and accurately quantified. It is also important that the nature of sediment and nutrient influx, whether point, non- point or internal, be established. Such data are provided through the development and calculation of the nutrient and sediment budgets. In the case of Onota Lake, there are no point sources of pollutants. The major vectors of nutrients and sediments are tributaries and runoff (non-point sources), septic systems, atmospheric contributions and Internal sources. The relative contribution of these sources are discussed below.

10.2 CALCULATION OF THE NON-POINT SOURCE LOADS / // Unit areal loading (U.A.L.) methodology was utilized to calculate annual non- point source phosphorus, nitrogen and sediment loads. An approach similar to

ENV/L22-rpt5 10-1 that in the Clean Lakes Program Guidance Manual (EPA, 1980) was utilized. The EPA provides a range of non-point source loading coefficients for various land uses and development scenarios. Based on site-specific conditions for the Onota Lake watershed (slope, cover, soils, residential density, etc.) coefficients were selected from the EPA values (Table 10.1). Once the coefficients were established, the watershed was delineated into sub-watershed basins, and the area of each watershed sub-basin calculated. U.S.G.S. 7.5 minute topographic contour maps, tax maps, zoning maps and available data (BCRPC, 1978) were reviewed to establish land use activity within each watershed sub-basin. A ground truthing reconnaisance was then conducted to confirm and refine land use patterns. Land use patterns were defined as follows:

High Density Residential - one or more housing units/acre or 2.5 or more units/hectare;

Low Density Residential - less than one housing unit/acre or less than 2.5 units/hectare;

Commercial - business, industry, airports, and parking lots; land use in which the majority of the area is impervious;

Disturbed - Open - landfills and construction sites, areas of barren, undeveloped land use characterized by exposed soils, lacking substantial vegetative cover;

Covered - Open - vacant lots, parks, large lawn areas; land use which has no appreciable canopy and sparse to dense vegetative cover;

Agriculture - active productive farmland;

Forested - areas covered by tree canopy.

For each land use category, an annual load was generated by multiplying the i appropriate loading coefficient, specific for each land use, by the area of i that land use type occurring in each of the watershed sub-basins. This non- ENV/L22-rpt5 10-2 I point source load accounts for loads originating from each land use and conveyed to the lake as a result of storm runoff. Land use within each sub- • basin was detailed in Section 4.0.

• Review of the unit areal, non-point source loading data show that the lake's major nutrient and sediment loads originate from watershed sub-basin's VI, VII and VIII, the lake's three largest sub-watersheds (Table 10.2). However, on a • per unit area basis, the greatest loads are from sub-watershed I. Land use in this sub-watershed is a combination of open-covered and high density | residential land use types.

• 10.3 CALCULATION OF THE SEPTIC LOAD

• Nutrient contributions resulting from septic sources were also accounted for ™ in the development of the nutrient budget. The only major residential area within Onota Lake's watershed still not sewered is located along Blythewood • Drive, sub-watershed II. Phosphorus originating from septic systems can be a significant component of a lake's nutrient budget and greatly affect in-lake productivity (PAS, 1984). In addition, excessive nitrate contributions can pose a health hazard if potable water sources become contaminated by I improperly treated septic effluent. Ligman, et al (1974) reports the phosphorus load of domestic wastewater may be as great as 1.5 kg/capita/yr. Under proper operating conditions, septic systems effectively treat wastewater • and minimize nutrient, as well as bacteriological, contamination of ground water or surfacewater. In a lake front setting, however, conditions such as • high ground water table, sub-optimal soils and high density development, challenge the effectiveness of septic treatment. Unfortunately, many lake community residences originated as seasonal homes with wastewater disposal systems totally Incapable of meeting the needs of year-round, full-time residency. Often, as the summer homes become converted to year-round dwellings, there is no upgrading of the system. The septic snooper survey indicated that the Blythewood Drive area may be contributing nutrients originating from Improperly functioning systems.

There are approximately 50 dwellings on Blythewood Drive. Assuming year-round residency, and a per capita density per house of 3.5, yields 175 residents.

ENV/L22-rpt5 10-3 I I This is possibly a conservative, over-estimate, of the actual Blythewood Drive population, but a reasonable estimate of the potential user population of this I area. Using the National Eutrophication Survey (NES, 1976) loading coeffi- cients for marginally operating septic systems located within 100 meters of a I lake shoreline (TN = 4.39 kg/c/yr, TP = 1.07 kg/c/yr), the estimate of the annual TN and TP septic loads are 768.3 kg/yr and 188.2 kg/yr respectively.

I Based on these data the septic load presently contributed to the lake is low. The phosphorus load is equivalent to that entering via precipitation and I the nitrogen component is equivalent to the dryfall contribution. However, this does not suggest that this nutrient source should be ignored. These I sources, both the Pecks Road sewer line leak and the Blythewood Drive septic systems, require immediate attention. These are much easier nutrient sources to control than the non-point storm runoff and soil erosion sources. In addition, these sources represent a vector by which bacteria and possibly I pathogens enter the lake. I 10.4 INTERNAL REGENERATION OF PHOSPHORUS An often overlooked component of a lake's nutrient budget is the phosphorus I load which is regenerated internally. Sedimentation processes result in particulate phosphorus being deposited in the bottom of the lake. This phosphorus is typically in a form not directly bio-available to phytoplankton I or benthic algae. However, under anoxic conditions, the ferric-phosphate complexes are chemically altered and phosphorus undergoes desorption (Freedman I and Canale, 1977). This can result in a phosphorus load significant enough to stimulate algal bloom even after a significant reduction in the external load I (Weich and Rock, 1980). Most often the 1iberated phosphorus, due to stratification, 1s concentrated in the hypolimnion well below the euphotic I boundary. Occasional mixing effects may erode the metalimnetic/hypolimnetic layer and distribute small concentrated pulses into the euphotic zone (Kortmann, et al., 1982). However, it is usually not until the fall turnover I that the hypolimnetic phosphorus stores are transported in mass into the euphotic la^er. This mixing process and resulting phosphorus load may trigger I an autumnal algae bloom. This appears to be occurring in Onota Lake. I I ENV/L22-rpt5 10-4 Sedimentary phosphorus regeneration has been studied extensively under laboratory and field conditions. The EPA (1980) provides a wide range of phosphorus loading coefficients for anaerobic sediments. Nurnburg (1984) devloped a series of internal loading coefficients based on a large sample of North Ameri can 1akes revi ewed in the 1 i terature. Based on Onota Lake's geographical and limnological characteristics an anoxic sediment TP loading o rate of 6 mg/rrr/day was selected. Souza and Koppen (1984) used a similiar coefficient in estimating internal loading in two large New Jersey lakes, Lake Hopatcong and Greenwood Lake.

Although the lake stratifies, the duration of stratification is much different for the two basins (Section 5). The dissolved oxygen profile data indicate that anoxic conditions exist in the South Basin's hypolimnion (> 10m) for 99 days. The North Basin's bottom water (> 6m) experiences anoxic conditions for only 29 days. The total area of North Basin sediments exposed to anoxic conditions during summer stratification was calculated to be 16.1 ha or 1.6 x c p 10 m. The anoxic bottom waters of the South Basin (> 10m) over lay a total area of 58.1 ha or 5.8 x 105 m2 . Applying the 6 mg/m 2/day loading coefficient, the estimated loads resulting from internal regeneration would be:

South Basin

(6.0 mg/mz/day) (5.8 x 105m2)(99 days)

= 3.45 x 108 mg or 345 kg TP

North Basin

(6.0 mg/ra2/day) (1.6 x 105m2)(29 days)

= 2.8 x 107 mg or 28 kg TP

The estimated loads resulting from internal regeneration are 345 kg and 28 kg i for the South and North Basins, respectively. It should be noted that these estimates are conservative, as some

ENV/L22-rpt5 10-5 I

I regeneration, albeit much less, can be expected from aerobic, littoral I sediments as a result of prop wash, boating activities, and macrophytes. I 10.5 ATMOSPHERIC CONTRIBUTIONS Nutrients, as well as pollutants such as heavy metals, may enter a lake from the atmosphere in a dissolved or particulate state. The loading coefficients I for dryfall and precipitation related nutrients inputs are typically low (Table 10.1), except for TN, which has a precipitation loading coefficient I equivalent to that of agricultural land use. The relative precipitation/dry fall nutrient load tends to be low unless the ratio of watershed to lake I surface area is extremely large (USEPA, 1980). However, it has been calculated that the phosphorus and nitrogen component of the nutrient budget I derived from atmospheric contributions can be as much as 25% and 1096, respectively (Kortmann, 1980).

I For Onota Lake, the potential does exist for precipitation/dryfall contri- butions to be sizable as the watershed-to-lake area ratio is approximately I 10:1 (Cooke, et al. 1980). Using the USEPA loading coefficients, the predicted precipitation and dryfall loads are respectively 62.5 kg/yr and 5.14 I kg/yr for TP, and 2500 kg/yr and 1028 kg/yr for TN (Table 10.3). This amounts to atmospheric contribution which totals 5.42S of the annual TP budget and I 30.5% of the annual TN budget (Table 10.4). I 10.6 CALCULATION OF THE ANNUAL NUTRIENT AND SEDIMENT BUDGETS The annual nitrogen, phosphorus and sediment loads were computed by summing I the above predicted nutrient and sediment loads. The TP load is sizable (1261.7 kg/yr), with the majority (50.236) originating from non-point sources I (Table 10.4). The Internally regenerated TP load accounts for approximately 29.6# of the annual total. In regard to TN, 62.956 of the 11,593.3 kg annual I load originates from non-point sources. A substantial portion (30.5%) of the TN load, however, is from atmospheric sources (Table 10.4). No estimate was developed for internally regenerated TN. Kortmann (1980) demonstrated that I for dissolved forms of nitrogen, a lake functions as a nitrogen sink. I I ENV/L22-rpt5 10-6 I I The septic related phosphorus load is 14.95U of the total annual load. This is a significant component of Onota Lake's nutrient budget. These data indicate I that although only a few dwellings still use septic systems for on-site wastewater treatment and disposal, their TP load is probably of great enough I magnitude to influence lake productivity. I 10.7 LIMITING NUTRIENTS Primary production is affected by the amount of available light, nutrients and I micro-nutrients. The amount of primary production supported by a lake ecosystem will be limited by that essential parameter of least availability. I In the euphotic zone, there is ample light energy to fuel photosynthesis and it is typically a nutrient or micronutrient which is the limiting factor. In most cases, phosphorus is the limiting nutrient, particularly with respect to I north temperate water bodies. However, nitrogen availability has also been found to be of critical importance, particularly with respect to southeastern I lakes. As such, lake managers often determine the nitrogen:phosphorus ratio to establish which of these two essential nutrients is the limiting factor for I a particular lake or pond. Studies have shown that when the N:P ratio is >12, phosphorus is limiting. Ratios <12 usually indicate nitrogen limitation I (USEPA, 1980). However, the 12:1 criteria was originally developed from algal bioassays (Fogg, 1965) designed to examine the response of algal cultures to the addition of dissolved nitrogen and phosphorus compounds. Under in-situ I conditions, the significance of such ratios may be minimal and even misleading, particularly when based on TN and TP data. Although of some I utility, it is probably more informative and meaningful to examine the chemical nature of the nutrients (dissolved vs particulate), bioavailability, I and any temporal pattern of nutrient loading rather than the TN:TP ratios alone. For instance, much of the nitrogen and phosphorus load may be I particulate rather than dissolved. In addition, the introduction of nutrients from various sources is variable over time, thus ratios developed from the total annual loads may not accurately reflect nutrient ratios at discrete I periods over the growing season. i // I The TN:TP ratio, derived from the annual nutrient budget, is 8.4:1 (Table I 10.4), indicative of a nitrogen limited waterbody. This result 1s similar to I ENV/L22-rpt5 10-7 data developed by BCRPB (1978) which were used in that study to formulate the conclusion that Onota Lake is nitrogen limited. However, although the TN:TP ratio data suggests Onota Lake to be nitrogen limited, it is unsupported by data developed 1n this study such as the seasonality of nutrient loading, the consistently low PO^ concentrations, and the temporal variations in the in- lake concentrations of TP and TN. As such, it was determined by IT, following review of all available data, that either TP or TN could be limiting in Onota Lake. Lake trophic state was therefore examined using models developed for both nitrogen and phosphorus limited waterbodies.

10.8 TROPHIC STATE ANALYSIS

The data presented in Sections 10.1 through 10.7 are integrated in this section and utilized to calculate the lake's trophic state. The trophic state of the lake is a quantification of relative potential productivity derived from a regression analysis of nutrient, hydrologic and morphometric data. The trophic state models provide a means by which the impact of nutrient loading to the lake can be assessed. In essence, the utility of the various trophic state analysis models is in the interpretation of hydrologic and nutrient data in a manner which reflects the lake's ecological "health". With only a few key pieces of data it is possible to determine the existing trophic status of a water body with a reasonable degree of accuracy (Dillon, 1974; Reckhow, 1979; USEPA, 1980). In addition, these models can be used to predict the future degradation or improvement of a lake as a result of changes in nutrient loading or implementation of various management practices.

To test the phosphorus limited scenario, the empirical models of Ostrofsky (1978) and Dillon (1974) were utilized to predict TP retention and trophic state, respectively. More sophisticated models exist such as detailed, in Reckhow (1979) and Walker (1977). But these models are very dissimilar from the available TN models. As such, it was determined that the Dillon model would best facilitate comparison of the TN and TP model results. The key input parameters were annual TP load, lake surface area, hydrologic retention time, and mean depth.

The nitrogen limited scenario was evaluated using a model developed by Baker,

ENV/L22-rpt5 10-8 I I et al (1985). Although developed for Florida lakes, the model was utilized for Onota Lake, largely due to the fact that few nitrogen loading criteria I models are available. This model is based on the same concept as the Dillon TP Model. It was anticipated that comparison of predicted lake trophic state I under the TP and TN limited scenarios would be possible if a similar modeling approach was used for both conditions.

I 10.8.1 Phosphorus Retention/Critical Loading I The TP load to the lake is calculated to be 1261.7 kg/yr (Table 10.4). Using this annual load, in conjunction with the trophic state model of Dillon (Table I 10.6), a spring TP concentration of 0.022 g/m 1s predicted. The predicted spring TP concentration is an indication of lake trophic state and is used to predict summer in-lake productivity. The observed mean concentration of TP I measured in the lake's surficial waters was 0.011 mg/1 (whole lake) during spring turnover of 1986. These data compare fairly well with the predicted I spring concentration. However, the Dillon .model also predicts the lake's trophic state to be eutrophic and that the existing TP load exceeds the I minimum eutrophic load (Figure 10.1). This prediction does not agree with observed conditions or monitored in-lake conditions. The observed mean spring I TP concentration characterizes Onota Lake as mesotrophic (Figure 10.1). Also, Secchi disc transparency and in-lake productivity, as expressed in terms of chlorophyll concentrations, cfo not reflect conditions representative of a I eutrophic water body (Section 5). Thus, although the Dillon model accurately predicts the spring TP concentration, it incorrectly characterizes the lake as I eutrophic (Figure 10-1). I 10.8.2 Nitrogen Retention/Critical Loading I An approach similar to that used for TP was used with the TN loading data to predict lake trophic state. The annual TN load is calculated to be 11,593.3 kg (Table 10.4) and nitrogen retention 1s predicted to be 59.6#. Using these I data In conjunction with the Baker, et al. (1980) model, the trophic state analysis predicts the lake to be oligotrophic. The existing annual TN load is I slightly below the minimum mesotrophic load (Figure 10.2). The predicted I I ENV/L22-rpt5 10-9 I to + I (0 MELp MMLp

c _

a: - c. z PREDICTED TROPHIC STATE T I I I I I

I 0.1 I I FIGURE 10-1 PREDICTED TROPHIC STATE OF ONOTA LAKE USING DILLON MElp= Total phosphorus (1974) CRITERIA AND BASED I ^Maximum Eutrophic Load ON TP LOADING / MMlp= Total phosphorus PROJECT NO. 541646 I Maximum Mesotrophic Load

JNTEBNATlONAi TECHNOLOGY '984 IT CORPORATION

Oo Nor PREDICTED TROPHIC STATE

0.1 0.1

FIGURE 10-2 * SEE TABLE 10.4 FOR DETALS ** SEE TABLE 10.3 FOR DETALS PREDICTED TROPHIC STATE OF ONOTA tAKE USWQ THE CRfTERIA OF BAKER et at (1985) MEln= Toal Nitrogen AND BASED ON TN LOADNG Maximum Eutrophic Load PROJECT NO. 541648 / MMln* Total Nitrogen Maximum Mesotrophic Load INTERNATIONAL '984 IT CORPORATION TECHNOLOGY ALL COPYRIGHTS RESERVED CORPORATION Do No* Seal* Tn.j trophic state is not representative of conditions observed in the lake relative to chlorophyll and transparency data. In addition, in the North Basin, macrophyte densities, sediment accumulation and winter dissolved oxygen concentrations reflect conditions associated with a non-oligotrophic waterbody.

Thus, the TP (Dillon) model data indicates the lake to be more eutrophic than measured lake parameters suggest and, in contrast, the TN model data appears to underestimate lake productivity, particularly in respect to conditions observed in the North Basin. As such, neither model satisfactorily predicts lake trophic state.

10.9 ONOTA LAKE, TWO LAKES IN ONE

The widely differing results of the two models leaves some question of their utility for Onota Lake. Classifying the lake as eutrophic, as per the TP model, is inaccurate. Chlorophyll a concentrations, lake transparency and other biological indicators such as fish and benthos, for Onota Lake are characteristically meso to oligotrophic in nature, particularly in the south basin. Likewise, classifying the lake as oligotrophic, as based on the TN model, is also inaccurate. The density of weeds, as well as the magnitude of accumulated silt and the precariously low winter dissolved oxygen levels in the North Basin are more characteristic of a eutrophic waterbody than an oligotrophic lake. As such, the lake is paradoxical. The South Basin is mesotrophic in nature, whereas the North Basin is eutrophic.

It appears that the lake's trophic state may not be properly addressed by dealing with the lake as a whole. The presence of the sand bar which separates the North and South Basins may be acting as a deterent to total lake mixing. Thus, the sand bar in effect may Isolate the bottom waters from being mixed during turnover. This may cause the two basins to function more as independent ecosystems than as one lake. Intuitively, this approach has merit. , First, the sand bar Is a prominent physical feature. It clearly separates ;the two basins, and, if not for the boat access channel, would probably greatly impede hydrologic exchange. Although no measurements were made of inter-basin water exchange, net flow must be from south to north as

ENV/L22-rpt5 10-12 I I the lake's spillway and discharge is in the North Basin. Second, throughout the monitoring program, it was observed that, in general, water quality, I sediments, macrophyte densities, and fish community data were different in the two basins. It is true that the origin and bathymetry of the two basins are dissimilar and these factors in themselves may be causing the observed I physicochemical and biological differences. However, the North Basin proportionally receives a greater percentage of the total nutrient, sediment I and hydrologic loads (Tables 9.2 and 10.2). Its shoreline is also more densely populated and there are a greater number of discrete stormwater I discharges than occur in the South Basin. The sandbar may also limit flushing and dilution of nutrient inputs through partial isolation of the North Basin I from the South Basin. • In an attempt to better characterize the 1ake's trophic state, a dual ecosystem approach was adopted. Lake trophic state was re-calculated using the Dillon (1974) TP model, but the North and South Basins were modeled • separately. The sandbar was designated as a physical boundary separating the >two basins. Hydrologic exchange between the two basins was assumed, for this exercise, to be minimal. Using the data developed in Sections 9 and 10.1 - 10.6, hydrologic and nutrient budgets were calculated separately for each I basin (Table 10.9 and 10.10). Although the North Basin is physically smaller than the South Basin, it is fed by BB% of the lake's total watershed by B area. The non-point source TP load is approximately 3.5 times greater to the North Basin than to the South Basin. However, the overall TP load to the South Basin is greater (Table 10.10) due largely to internal regeneration • processes. I Entering the basin specific data into the phosphorus retention model yielded predicted TP retention coefficients of 0.593 and 0.734 for the North and South B Basins, respectively (Table 10.11). Due to the South Basin's larger volume and lower flushing rate, conditions favor the settling and retention of B particulate phosphorus. In the North Basin, the flushing rate 1s quicker, owing to the large hydrologic load and relatively small volume of the basin, leading to/less settling and retention of particulate TP than occurs in the • South Basin. Aside from trophic state predictions, these data can be particularly useful in selecting lake management techniques. This will be

— ENV/L22-rpt5 10-13 I I elaborated upon further 1n Section 11.0. I The dual ecosystem approach to calculating predicted trophic state yielded Improvement over the total lake approach (Table 10.12). As opposed to the data generated by the Dillon TP model, when the lake was treated as a whole, I the basin specific data do not indicate the lake to be exceedingly eutrophic. Specifically, the south basin is predicted to be slightly I eutrophic while the north basin is predicted to be on the meso-eutrophic I border (Figure 10.3). These data are still not totally in keeping with the perceived condition of I the lake, as based on field observations. However, they appear to be more accurate predictions of trophic state than that estimated under either the total lake TP or TN limited scenarios (Figures 10.1 and 10.2). Calculation of I the south basin's trophic state based only on the external load would no doubt lead to a predicted trophic state in the oligo-mesotrophic range. However, the internal load must be included as part of. the total load. This component of the annual load can significantly affect trophic state (Welsh and Rock, E 1980). I Based on the outcome of this iteration, treating the lake as two lakes rather than one appears correct and justifiable. From a lake manager's viewpoint, it does reveal some interesting" aspects of Onota Lake. Specifically, septic I management, non-point source control and internal load reduction strategies used in the restoration of Onota Lake will differ markedly between the two I basins. These data will be used in determining the feasibility and priority I of various in-1ake and watershed restoration/management techniques. Further refinement of the predicted trophic state is possible if either the Reckhow (1979) anoxic lake model or the Walker (1977) model are used. I However, use of either of these models will make it difficult to compare trophic state under TP versus TN limited conditions. Use of the Dillon model I seems warranted given the uncertainty of whether Onota Lake is TP or TN I limited. / I I ENV/L22-rpt5 10-14 10.10 TROPHIC STATE CRITICAL LOADING BOUNDARIES

As a final check on the validity of the total phosphorus loading calculations, the critical loading boundaries between oligo-mesotrophic and meso-eutrophic conditions were calculated using the method of Vollenweider (1976). These calculations (Table 10.13) indicate that the critical load for the oligo- mesotrophic transition is 404.8 kg TP/yr (whole lake). The meso-eutrophic transition critical load was calculated to be 811.0 kg TP/yr. These TP load boundaries indicate that the TP budget for the lake (Table 10.4) should result in a eutrophic conditon, which, as previously stated, was not demonstrated by the data collected during this study.

The critical loading boundaries were also calculated for each basin, separately (Table 10.13). As with the Dillon TP model, the South Basin is predicted to be eutrophic and the North Basin is predicted to be on the meso- eutrophic border. These data are still not totally in keeping with the perceived condition of the lake.

ENV/L22-rpt5 10-15 1

1 TABLE 10.1 I LOADING COEFFICIENTS FOR VARIOUS LAND USE ACTIVITIES I IN THE ONOTA LAKE WATERSHED

I Land Use ANNUAL LOAD kg/ha/yr I Category TN TP TSS Forest 2.5 0.2 250 I Agriculture 10.0 0.6 1,600 High Density Res1 5.0 0.8 2,000 1 Low Density Res 2.0 0.25 200 Open-covered 5.0 0.30 400 Open-Disturbed 10.0 0.6 1,600 o I Precipitation^ 10.0 0.25 - _ II Dryfall3 0.4 0.002 i 1 - includes commercial properties 2 - applied only to lake surface i 3 - applied to entire watershed

11 Source: USEPA Clean Lake Program Guidance Manual i I (I 11 ii ENV/L22-rpt5 10-16 TABLE 10.2

SUMMARY OF NON-POINT SOURCE NUTRIENT AND SEDIMENT LOADING TO ONOTA LAKE ANNUAL LOAD PER SUB-BASIN

Sub* Area ANNUAL LOAD kg/yr Basin Ha TN TP TSS

I. 156 780.0 85.8 187,200 II. 87 352.4 46.1 103,095 III. 27 78.3 7.8 13,824 IV. 25 56.3 5.6 5,625 V. 77 192.5 15.4 19,019 VI. 865 2,508.5 190.3 275,935 VII. 310 1023.0 74.4 121,830 VIII. 635 1914.0 151.6 255,270 IX. 60 192.0 28.2 55,200 X. 77 200.2 27.7 43,120

Total 2,319 7297.0 632.9 1,080,118

* Refer to Figure 4.1 for sub-basin location.

ENV/L22-rpt5 10-17 TABLE 10.3

RELATIVE ATMOSPHERIC CONTRIBUTIONS OF NITROGEN AND PHOSPHORUS TO ONOTA LAKE

Phosphorus Nitrogen Parameter Coefficient Load Coefficient Load kg/ha/yr kg/yr kg/ha/yr kg/yr

Dryfall 0.002 5.14 0.4 1,028

Precipitation** 0.25 62.5 10.0 2,500

Total atmospheric tt Load 67.64 3,528 ' * Applied to area of lake ** Applied to area of watershed

.it

ENV/L22-rpt5 10-18 TABLE 10.4 NUTRIENT/SEDIMENT BUDGET OF ONOTA LAKE

TP TN TSS SOURCE kg % kg % kg % Direct Runoff and Tributary Loading 632.9 50.2 7297 62.9 1.08 X 106 100 62.5 5.0 2500 21.6 Direct Precipitation — — Dryfall 5.14 0.4 1028 8.9 — — Internal Regeneration 373.0 29.6 — — — — Septic Loading 188.2 14.9 768.3 6.6 — —

TOTAL 1261.7 11593.3 1.08 X 106

ENV/L22-rpt5 10-19 I

I TABLE 10.5

PHOSPHORUS RETENTION ONOTA LAKE

Rp = 0.201e(-°-0425 "s) + 0.5743e(-°-00949

where:

Rp = Phosphorus Retention o qs = Area! water load = annual inflow (m /yr) o surface area of lake (or)

Annual inflow 18.0 X 106m3/yr Surface area 2.497 X 106m2

qs = 18.0 X 1Q6 m3^. o =7.2 m/yr 2.49 X 10fiV

Rp = 0.201e (-0.0425)(7.2) + 0>5743e (-0.00949)(7.2)

• Rp = 0.148 + 0.536

•I Rp = 0.684

_ % TP Retention = 68.4%

— ENV/L22-rpt5 10-20 TABLE 10.6 TROPHIC STATE ANALYSIS OF ONOTA LAKE USING THE CRITERIA OF DILLON (1974)

Parameter Value

Annual Load 1261.7 kg/yr

Areal Load (L) 0.505 g/m2/yr

Phosphorus Retention (R) 0.684

Hydraulic Retention Time (T) . 0.888 yr

Mean Depth (Z) 6.4 m

3 Predicted spring TP concentration (g/m ) = [P.s ] = L(1-R)T Z

2 Ps] = 0.505 g/m /yr (0.316)(0.888 yr) 6.4 m

2 [Ps] = 0.142 g/m 6.4 m

3 [Pc] = 0.022 g/m

ENV/L22-rpt5 10-21 TABLE 10.7

NITROGEN RETENTION ONOTA LAKE

0 00949 = 0.4266 (-0<271qs) + o.574e<- -

where RN = Nitrogen Retention _ qs = Areal water load = annual Inflow (m /yr)r lake surface area (m) Annual discharge 18.0 x 106m3/yr Surface Area 2.497 x 106m2

qs = 18.0 x IQJfoj/yr = 7 2 m/vr 2.49 x 10V /yr 00949 7 2 = 0.426e (-0.271M7.2) + o.574e(-°- >< - )

= 0.06 + 0.536

RN = 0.596

TN retention = 59.63S

ENV/L22-rpt5 10-22 TABLE 10.8 TROPHIC STATE ANALYSIS OF ONOTA LAKE — USING THE TN CRITERIA OF BAKER, ET AL. (1985) Parameter Value Annual load 11593.3 kg/yr • Area! load 4.65 g/m2/yr Nitrogen Retention (R) 0.596 Hydraulic Retention (T) 0.88 yr Mean Depth (z) 6.4 m ITHI . L I'"')7

[TN] = 4.65 g/m2/yr (1-0.596)(0.88 yr) 6.4 m [TN] = 1.65 g/m2 6.4m [TN] = 0.26 g/m3 TN

• — ENV/L22-rpt5 10-23 • s I TABLE 10.9 I PERTINENT HYDROLOGIC AND MORPHOMETRIC CHARACTERISTICS OF ONOTA LAKE'S 1 NORTH AND SOUTH BASINS

Parameter Value North Basin South Basin

B Basin surface area 106m2 0.794 1.7

B Basin volume 106m3 1.5 14.5

B Hydrologic load 106m3/yr 12.85 5.15

H Flushing rate times/yr 8.6 0.36

_ Residence time yr .116 2.82

qs m/yr 16.2 3.02

Mean depth m 1.8 8.5

ENV/L22-rpt5 10-24 TABLE 10.10 ANNUAL TOTAL PHOSPHORUS LOADING (KG/YR) TO THE NORTH AND SOUTH BASINS OF ONOTA LAKE

I^•B Source North Basin South Basin

I Tributary and direct runoff 493.2 139.7

Direct precipitation I on lake surface 19.8 42.6 I Dry fall 4.6 0.5 Internal regeneration 28.0 345.0 I Septic systems 188.2 1 Total load 545.7 716.0

• Area load (g/m2) 0.42 0.687

ENV/L22-rpt5 10-25 TABLE 10.11 PHOSPHORUS RETENTION* AS CALCULATED INDEPENDENTLY FOR ONOTA LAKE'S NORTH AND SOUTH BASINS

SOUTH BASIN

0425 3 02 Rp = o.201e(-°- X - ) + 0.5743e (-0-00949)(3.02)

Rp = 1.76 + 0.558

Rp = 0.734

P retention = 73.4%

NORTH BASIN

0425 16 2 Rp = 0.201e(-°- X - > + 0.57436 (-0.00949) (16.2)

Rp = 0.10 + 0.492

Rp = 0.593

P retention = 59.3*

* Calculated .as per Ostrofsky (1978)

ENV/L22-rpt5 10-26 TABLE 10.12 LAKE TROPHIC STATE AS PER DILLON (1974) CRITERIA CALCULATED INDEPENDENTLY FOR ONOTA LAKE'S NORTH AND SOUTH BASINS

SOUTH BASIN fP 1 - 0.42 g/m^/yr (l-.734)(2.82 yr) I r 2 J 8.5m rp ] - 0.315 g/m£ l • 51 8.5m

3 [Ps] = 0.037 g/m

NORTH BASIN

[p ] = 0.687 g/m^/yr (1-.593H.116 yr) 1.8m

[p ] = 0.032 g/m? 1.8m

3 [Ps] = 0.018 g/m

ENV/L22-rpt5 10-27 I

I TABLE 10.13 CALCULATION OF CRITICAL LOADING BOUNDARIES BETWEEN OLIGO-EUTROPHIC CONDITIONS BASED ON VOLLENWEIDER (1976)

2 l + tw) Lr (mg/m /yr) = TOcn * Z ( Vollenweider (1976) | ^ L bp TU From Vollenweider (1976) • Transitional Boundaries Between Oligotrophy --> Eutrophy (l__c ) 10 mg/md — 20 • APPLICATION TO ONOTA LAKE 1. Oligo-Mesotrophic Transition (10 mg/m3) Boundary • _Tw = 0.73 yrs. , M ^r Z. = 6.4 m j + •j Lc = 10.0 mg/m • (6.4m) / n'^ ^l

3 = 10.0 mg/m • (6.4m) j fe** yrj

• = (10 mg/m3) - (6.4 m) (2.53) 2 LC = 161.9 mg/m /yr | or 404.8 KgTP/yr • 2. Meso-Eutrophic Transition (20 mg/m3) boundary L = 20.0 mg/m3 - (6.4m) (1 + .73 yr) ^_ c ( . / o yr ) • = 20.0 mg/m3 • (6.4m) (2.53 yr) 2 m Lc = 324.4 mg/m,2/yr or 811.0 KgTP/yr

ENV/L22-rpt5 10-28 TABLE 10.13 (Continued) APPLICATION TO NORTH BASIN 1. Oligo-Mesotrophic Transition (10 mg/m3) Boundary

Lc - = 10.0 mg/m3 •• (1.8m) • (11.6/yr = 209 mg/m2/yr or 166 KgTP/yr 2. Meso-Eutrophic Transition boundary

3 1+ - Lc = 20.0 mg/m . (1.8m) j

= 20.0 mg/m3 • (1.8m) (11.6/yr) = 418 or 332 kg Tp/yr.

• ENV/L22-rpt5 10-29 I TABLE 10.13 • (Continued) APPLICATION TO SOUTH BASIN • 1. Oligo-Mesotrophic Transition Basin 3 _ Lc = 10.0 mg/m • (8.5m) . (^ + 2.82)

= 10.0 mg/m3 - (8.5m) • ( 0.950/yr.) • = 80.8 mg/m2/yr 1 or 137 kgTp/yr 2. Meso-Eutrophic Transition boundary 1 3 ^ ™ Lc = 20.0 mg/m • (8.5m) |^—|^|j

• = 20.0 mg/m3 • (8.5m) (0.950/yr) o = 162 mg/m^/yr 1 or 279 KgTP/yr

i i i

i ENV/L22-rpt5 10-30 MELp EUTROPHIC MMLp

< 5 cr z> c z

0.10 -

0,05-

CJ

o II H £ i 0.01-

0.1 0.6 1.0 5.0 10.0 MEAN DEPTH - m

FIGURE 10.3 TROPHIC STATE AS PREDICTED INDEPENDENTLY FOR ONOTA LAKE'S NORTH AND SOUTH BASINS PREPARED FOR: ONOTA LAKE PITTSFIELD, MASS PROJECT NO. 541648 NOVEMBER 1987

INTERNATIONAL 1984 IT CORPORATION TECHNOLOGY ALL COPYRIGHTS RESERVED CORPORATION Do Not Scale This Drawing 11.0 RESTORATION/MANAGEMENT PLAN

11.1 INTRODUCTION

The data base developed in the diagnostic portion of this study identified the problems which affect Onota Lake. Nutrient and sediment loading were quantified and utilized to establish the lake's trophic state. Water quality data were reviewed and the lake's general environmental condition determined. As well as documenting the current status of the lake, these data provided the means by which the factors responsible for the lake's degredation could be identified and prioritized. With this information, a management plan could be developed in a manner insuring that the proper amount of effort and money was allocated to each factor impacting the lake.

11.2 DEVELOPMENT OF THE FEASIBILITY ANALYSIS

In this section, detail is provided concerning the formalization of the management plan. Once the lake's problems were identified and prioritized, a feasibility analysis was conducted to determine which restoration/management techniques are most applicable for the problems of Onota Lake.

In general, the lake's eutrophication can be attributed to watershed urbanization and subsequent increase in sediment and nutrient loading. The lake's morphometry exacerbates the effects of increased sediment and nutrient 1oadi ng. The North Basin's shal1ow depth is favorable to the growth of macrophytes and the South Basin's large anoxic hypolimnion favors the internal regeneration of phosphorus. From the results of the diagnostic evaluation, it appears that the restoration/management plan should address the following:

1. Excessive aquatic macrophyte growth. 2. Accumulation of organic sediment and silting in of.shallow embayments, and 3. Reduction in dissolved oxygen concentrations.

Doing so will require that the following nutrient and/or sediment sources be controlled, corrected or abated:

ENV/L22-rpt4 11-1 I I 1. Septic inputs 2. Non-point source loading of sediments and nutrients caused by watershed I urbanization, and I 3. Internal regeneration of nutrients. I 11.3 MANAGEMENT PLAN The most pervasive cause of the lake's problems stem from excessive sediment I and nutrient loading. Based on data compiled in the diagnostic sections, it is evident that erosional, non-point, and septic sources need to be H controlled. It appears that this is best achieved through stormwater and soil erosion management, land-use management practices, and sewering. In addition, • certain in-lake techniques appear to be needed to aid in the restoration of the lake specifically to improve water quality and aesthetics. Most of the in-lake measures should not be viewed as means of controlling the causes of • eutrophication but as means of reducing the impacts of eutrophication. The •combination of in-lake and watershed management measures will allow for the maintenance of the lake as a viable recreational water body while providing for its long-term management.

I The recommendations presented in the management plan will require the integration of many water resource programs, government agencies and funding sources. The full implementation of the plan will require a number of years to complete. To achieve the goal of restoring and managing the lake, activities will have to be well coordinated. A lead agency, such as the Preservation Association (LOPA), will be required to interface with the various government, funding and planning groups. LOPA will also be needed to ensure that the level of effort and the direction of that effort does not falter.

The following sections provide a detailed evaluation of the general components I which were presented above. Specifically, this includes a feasibility analysis for in-lake and watershed management measures, public education I program, and institutional arrangements. I I ENV/L22-rpt4 11-2 I I 11.3.1 WATERSHED MANAGEMENT I The objective of the watershed management plan is to reduce external nutrient and sediment loading. If done properly, this will effectively slow the rate of the lake's eutrophication. Nutrient and sediment control can be achieved I through a variety of methods, so it is important to thoroughly evaluate the merits, applicability and cost of various options. The following should be considered as part of this evaluation: I Practicality of non-point source control, Cause-effect relationship between existing load, lake response, and • load reduction following management/control, Performance characteristics and cost, • • Required maintenance, and • Ease of implementation.

H Both structural and non-structural methods are available. Structural methods

— rely on the construction of catch basins, silt traps or other engineered | devices to intercept, divert, or control nutrient/sediment inputs. Non- structural methods are conservation techniques, product modification, and • ordinances which reduce loading and regulate activities responsible for nutrient/sediment influx.

In order to objectively evaluate the applicability of various watershed management methods, a feasibility matrix was developed (Table 11.1). An I ordinal ranking, based on a score of 1 to 5, was obtained for each alternative. Those alternatives scoring the highest were given priority consideration. The feasibility of each option was judged on the basis of: I Cause-effect - how effective would this method be in immediately contributing to a noticeable improvement in lake conditions.

• • Degree of load reduction - once implemented how substantial a decrease in nutrient and/or sediment loading could be expected.

• Practicality - can the method be practically implemented.

_ ENV/L22-rpt4 11-3 I • Performance characteristics - based on the scientific literature, how I effective is this measure in abating nutrient/sediment loading. Cost - is the cost of the method, its implementation, or its maintenance I prohibitive relative to the expected returns.

I Although each was given equal weight relative to scoring, attention was focused on the scores associated with load reduction, practicality and cost. I The higher the score for each of these three matrix components, the more I favorable the management option. Review of the matrix scores indicate that the top five watershed management I options are: • Stormwater retention/detention, I • Erosion control, Land use ordinances, I • Stormwater management, and I Product modification. I I I I I

ENV/L22-rpt4 11-4 I I TABLE 11.1 I WATERSHED MANAGEMENT FEASIBILITY MATRIX CAUSE- DEGREE OF PERFORMANCE METHOD EFFECT LOAD REDUCTION PRACTICALITY CHARACTERISTICS COST TOTAL Diversion of inflow 5 4 1 1 1 12

Treatment of inflow 5 2 1 1 1 10

Stormwater Detention - 5 4 4 4 2 19 Retention

Erosion Control 4 4 5 4 3 20

Sewering 3 2 4 5 1 15

Up-Grade of Existing Point Sources 2 2 5 5 1 15

Land Use Ordinances 4 4 5 3 5 21

Stormwater Management 5 4 4 4 2 19

Product Modification 3 3 5 4 5 20

Public Education 2 2 5 4 5 18 I I I ENV/L22-rpt4 11-5 Each of these management options are feasible, cost-effective means by which loading can be reduced and the lake's water quality improved. Implementation • recommendations for the above options are as follows. i 11.3.1.1 Stormwater Retention/Detention As with most established lake communities, Onota Lake is faced with a limited amount of available land for the construction of off-stream detention basins. Of the three main tributaries (Parker, Churchill and Daniels Brooks), the construction of an off-stream basin appears most feasible for Parker Brook at a site located at the base of Cascade Street on property owned by Camp Winadu. The proposed basin should be a dry, grassed basin which evacuates 90% of its volume in 18 hours. The outlet structure should be either a stand pipe device or a dual orifice head wall. If a dual orifice headwall is used, the small outlet orifice should not exceed 4 inches and should be protected by a trash rack. Sizing of the detention basin should be such that it detains a total volume equivalent to 10% of the 1 year storm volume. It is recommended that HECII modelling be conducted as part of the design analysis to ensure that offsite flooding will not occur.

For Churchill and Daniels Brooks a different approach is recommended. Full advantage should be taken of the backwater located north of the causeway. This area is colonized by wetland plants and macrophytes. It serves, in its present state, as a biofilter and retention basin. Storm flow from both brooks empties into this area. Its quiescent nature coupled with the existence of vegetation promotes settling of particulates and an associated decrease in nutrient loading. Additional settling could be promoted by i erecting a gabion wall at the mouth of tributaries. This type of wall, constructed of concrete cylinders or similar material, will aid in dissipating i the energy of storm surges and facilitate settling. At the mouth of each stream, downstream of the gabions, a variety of sedges, rushes and bullrushes, should be planted. These vegetation types can withstand a variety of hydrologic conditions and provide an excellent means of i trapping sediment and uptaking nutrients. If protected from storm surges and i scouring by the gabion energy dissipator, the introduced vegetation should i ENV/L22-rpt4 11-6 This "natural" basin should be augmented by the creation of a small sump on the south side of the causeway. It is recommended that a depression approximately four feet deep and 100 x 100 be excavated. A shallow (3 foot) gabion wall should be erected along the perimeter of the depression. The objective of the sump would be to promote the settling of particulates and sediments. The gabion lip would aid in dissipating flows and facilitate settling of material in the basin. Rather than have sediments accumulate and form a delta, the creation of the sump would promote the settling of materials in an area easily accessible for future maintenance. It would decrease loading to the Take, aid in decreasing the rate of silting in of the north basin, and could even create a new fishery habitat.

A similar approach is recommended for Blythewood Drive. The small tributary which flows off Blythewood into the lake can be addressed in two ways. A small sump, similar to that described for Churchill and Daniels could be excavated at the mouth of the brook. The sump would act to promote settling and trap sediments.

An alternative and more desirable approach would be the creation of a small, confined wetland at the mouth of the brook. This would be accomplished by first erecting a gabion wall and placing a set amount of fill within its confines. Wetland species, such as Typha, would be planted within the confined area. This wetland would serve as a biofilter. Sediment settling would be achieved as well as attenuation of nutrients. Precise bathymetric measurements in the vicinity of the mouth would be necessary to establish the feasible extent of such an area.

11.3.1.2 Erosion Control

Several sites throughout the watershed were observed to be in need of erosion control. Most of these sites are located along the southwest side of the lake. Gully erosion along road beds, building lot excavation and agricultural activities were some of the specific causes of the problems. Three approaches, each of which can be easily implemented, are recommended to

ENV/L22-rpt4 11-7 I I correct erosion. I First, stricter enforcement of the City of Pittsfield's soil erosion control ordinance is required. The existing ordinance is well structured and will protect the lake and its tributaries from excessive sedimentation. This I represents the most effective means of alleviating future soil erosion perturbations. Therefore, it must be stressed that development in the I watershed, whether it be of one or multiple parcels, adhere to existing soil I erosion control guidelines. Second, there 1 s a need to maintai n a buffer of at 1 east 50 feet between I cultivated land and the stream banks. These buffers should be bermed and vegetated. Course vegetation will filter soils eroded from croplands during storm events prior to their transport to the lake's tributaries. Conservation I tillage practices as specified by the Soil Conservation Service should be implemented. If possible, there should be a minimum amount of time when land I lies fallow. Planting of clover or alfalfa is recommended for areas which are being rotated and will not be cultivated over the course of a year.

Improved agricultural practices can greatly decrease sediment, nutrient, and I pesticide losses to a lake (U.S. EPA, 1980). Among the various alternatives, the following Best Management Practices (BMP's) should be implemented for I Onota Lake: (1) Conservation tillage, including no-plow methods, I (2) Winter cover crops, (3) Timing of field operations to avoid long fallow periods, I (4) Contouring, graded rows and terracing, (5) Grassed or vegetated buffer zones along the perimeter of I cropland particularly where they abut roads or streams, (6) Avoiding winter spreading of manure, (7) Proper storage of animal wastes, I (8) Containment and passive treatment of runoff from feed lots, (9) Proper calculation, application, and timing of fertilizer I usage, I (10) Optimization of pesticide formulation, the timing of I ENV/L22-rpt4 11-8 I I placement, and the method of application to insure I effectiveness while minimizing loss. Finally, individual existing sites of erosion must be comprehensively catalogued and the specific corrective actions needed to remediate the sites I must be developed. In many situations, such as roadway washout and unstable stream banks, non-structural corrective actions will be effective. This would I consist essentially of regrading, planting and possibly stormwater diverting. In other cases, structural measures such as retainer walls, will I be needed- I To accomp1i sh thi s catalogi ng, a detai1ed reconnai ssance of the area, concentrating on the general sections of the lake identified above, must be conducted. Due to seasonal nature (road wash out) and unpredictability (new I development) of some of these problems, it is necessary that the reconnaissance immediately precede the development of specific restoration and I erosion control measures. It is highly likely that many of the problem areas observed during the 1986 survey have been corrected as part of the Burbank I Park upgrade and/or comments directed by IT to LOPA or city officials. I 11-3.1.3 Land Use Ordinances In order to protect the lake from future nutrient, erosion, and pollutant I loading, sensitive environment land use ordinances are required. This relates specifically to slopes in excess of 20%, soils of limited stability and septic I suitability, and . These areas are sensitive and must be protected by limiting development. In order to achieve this, development restrictions must I be enacted. This appears most needed in the section of the watershed along Blythewood Drive. This area presently utilizes septic systems. It is I recommended that to decrease future septic related inputs the area be re-zoned to R-43, 1 acre minimum. Even if this area is sewered in the future, the R-43 zoning classification should be implemented. The proximity of this area to I the lake, the presence of small intermittent streams which channel runoff, and the steep to moderate slope, all contribute to making this an environmentally I sensitive area. By designating this area as environmentally sensitive, a I legal basis will exist by which future development can be restricted. I ENV/L22-rpt4 11-9 I I I I 11.3.1.4 Stormwater Management Ordinances Measures must be imposed to control stormwater runoff from new development to the lake. In essence, this requires the creation of stormwater management I ordinances. Such ordinances require that stormwater management provisions be incorporated into sub-division, roadway construction, and commercial I property. These provisions are different from flood control in that they are designed to improve stormwater quality and reduce sediment/nutrient loading. I Attention is focused on small storms and the first flush of major storms. Detention basins, retention basins, catch basins and grassed swales are I commonly employed techniques. When the situation allows, dry wells, roof top detention, pervious soils and created wetlands are other viable stormwater I management techniques. I 11.3.1.5 Product Modification At the grass roots level, there are measures which can be implemented which I have been proven effective in decreasing nutrient loading. These include the use of non-phosphorus detergents and non-phosphorus fertilizers. This also I encompasses public education in septic system maintenance and low flow water devices. It is strongly recommended that non-phosphorus fertilizers be used. A time release nitrogen, non-phosphorus fertilizer is manufactured I specifically for lake front communities. This material can be made available through the Onota Lake Preservation Association or local hardware/garden I supply stores. The use of such fertilizers will be part of the Phase II I public education program. 11.3.1.6 Sewering and Infrastructure Improvements

I The matrix scores for sewering and upgrade of existing infrastructure were lower than for many other management options. This stems primarily from the I high cost of such sewering and the limited achievable reduction in loading. I However, both are highly recommended due to acute factors currently affecting I ENV/L22-rpt4 11-10 I I the lake and recorded improvements in lake quality following sewage diversion (Welch and Rock, 1980). Although, septic loading accounts for only 14.9£ of I the total annual TP load, results of the "septic snooper" study reveal major acute septic problems in Thomas Island Cove and problems along the Blythewood Drive shoreline. The Thomas Island problems stem from a break in the Peck's I Road sewer line. Faulty septic systems, resulting from soils of marginal treatment ability, are responsible for the problems along Blythewood Drive. I Although only a 15% reduction in TP loading will be gained by sewering this area, the health risk problems posed by faulty septics will be corrected. In I addition, if this area is further developed, the septic related TP loading could greatly increase depending on final densities and proximity of new I dwellings to the lake. A bond has recently been approved by the City of Pittsfield which will make I available funds to locate and repair the Peck's Road break. Once this is corrected, sewage infiltration into the storm sewer will be alleviated. The I contamination problem currently affecting Thomas Island Cove will then be I abated. Sewering of the homes along Blythewood Drive is highly recommended. Although I this will conceivably be costly, such action is required. Review of existing data, including preliminary engineering specifications, suggest that the best approach to sewering the area would be to install a line along the lake's I shore line. A pump station would be used to pump the sewage upgrade to a I trunk line which runs along West Street. Although this is a sound approach, other alternatives should be reviewed. A I possible option would be to retain the existing septic tanks for those homes located on the east side of Blythewood Drive. The tanks would be used to I accumulate solids and sludge. Liquid wastes would be piped to a small diameter line located along the lake's shore. Servicing of future development along the west side of the road would be accomplished through the use of I standard lines. There would be two distinct advantages to such an approach. First, it would be less costly. At present the majority of the development is I on the east side of the road along the lake's shoreline. Future development I will occur primarily along the western side of the road. Rather than incur I ENV/L22-rpt4 11-11 I I costs for a single large system designed to service both existing and future densities, it would be possible to correct the existing septic problems at a I much lower cost to the individual home owner. New development will inevitably require the up-grade of Blythewood Drive in order to better meet the transportation and servicing needs of this area. Sewering of new development I could be timed to coincide with this. Also, the smaller diameter mains would require a smaller pump station. This would translate to a savings once again I in cost and also in operations and maintenance. I Second, the installation of a small diameter pipe would require less excavation. This would reduce potential impacts to the lake resulting from I construction activity. In order to protect that section of the lake from excessive nutrient and I sediment loading resulting from future development, re-zoning of this area is warranted. Due to the environmentally sensitive nature of this area, re- I zoning to R-43 is highly recommended. Limits should also be placed on the use of this area for cluster development. In this way, sewering of the area will I not lead to environmental repercussions resulting from increased nutrient and I sediment loading. 11.3.1.7 Diversion and Pretreatment

I Neither diversion of inflow or the chemical treatment of stormwater is recommended. It is felt that both would be very impractical and too costly. I Diversion would also disrupt the lake's hydrologic budget. Pretreatment would have a high O&M cost associated with it. For these reasons, neither of these I watershed management options are recommended. I 11.3.2 In lake Restoration Measures Given Onota Lake's recreational and aesthetic attributes, the development of I an in-lake restoration program is necessary. In-lake programs offer, in some cases, an immediate improvement in lake quality. Unlike watershed management I techniques, in-lake measures are usually easier to implement, somewhat less I expensive, and provide much quicker improvements in lake quality and I ENV/L22-rpt4 11-12 I I utilization. Although some In-lake management measures are primarily cosmetic, they can lead to substantial reductions in nutrient loading as well I as to an improvement in the appearance and recreational opportunities of the lake.

I As was performed with the watershed management measures, a feasibility matrix of the various in-lake options was conducted. Matrix components and scoring I were similar to that utilized for watershed management. The results of the analysis are presented in Table 11.2. Of the numerous techniques assessed, I the following proved to have the highest scores. • Short circuiting of inflow I • Aquatic macrophyte harvesting • Aeration • Use of macrophyte barriers I • Spot dredging Each of these promises to yield the greatest benefits to the lake. I Interestingly, lake conditions are somewhat dissimilar between the two basins. As such, different in-lake management options are prescribed for each I basin. In the South Basin, summer aeration and limited macrophyte harvesting are needed. In the North Basin extensive harvesting, flow short-circuiting, spot dredging, weed barriers and winter aeration are recommended. The utility I of each in-lake technique is detailed in the following sections. I 11.3.2.1 Short Circuiting of Flow I The presence of the sand bar effectively isolates the North and South Basins. Although water exchange occurs between the two, the sand bar physically impedes mixing. This has been documented in Section 10 and is I supported by the observed conditions and problems of the two basins. The majority of tributary inflow is to the north basin. Thus, the majority of I tributary related influxes of nutrients and sediments is to the North Basin I (refer to Table 10.10). Review of the lake's morphometry and the location of the tributaries suggest I that it would be possible to increase flushing and short circuit flow by I construction of a culvert under Thomas Island Road. Presently, there exists a I ENV/L22-rpt4 11-13 narrow causeway (-50 feet wide). Two box culverts, each 4-8 feet wide, could be placed under the roadway. Opening of this area, coupled with dredging, could promote flushing of Thomas Island Cove and the small cove northwest of the island. In addition, inflow from Daniels and Churchill Brooks would be partly diverted as a result of the construction, and induced to flow through the culverts as opposed to around the island. As the outlet of the lake is located in Thomas Island Cove, this would, in effect, short-circuit nutrient rich water out of the lake and the altered circulation pattern would promote better flushing of the North Basin (Figure 11.1).

There are a number of ways to engineer this project. The first would be to hydraulically dredge the sediments on either side of the causeway, possibly disposing of the spoils in the abandoned quarry pits in the Daniels Brook watershed (Figure 11.2) Once dredging had been completed the lake could be lowered and the box culverts installed. The optimum level of the culverts would be determined during the engineering survey, but would be such that flow would not be interrupted by normal low lake levels.

An alternate approach would be to lower the lake. Once the sediments had been allowed to dewater, conventional equipment would be used to remove accumulated sediment. This would probably include the use of muddozers, a pond excavator such as the Smalley Excavator, and a dragline. Spoils would have to be trucked off site. Once again, the quarry pits could be used for spoil disposal. Other approaches including the use of coffer dams, and a combination of dry and hydraulic dredging could be pursued.

Based on our review of the site and discussions with LOPA and Pittsfield officials, it appears the most feasible approach would be dry excavation. First, dry excavation will minimize the spoil handling/disposal problems. Hydraulic dredging will require pumping of the slurry. Due to the distance and topography of the potential disposal site, it will be necessary to traverse one of two roads and pump the material upgrade (approximate 30' change in elevation). Sophisticated sediment containment/dewatering basins will need to be constructed to accept the slurry, all of which increases the cost of the project. Dry excavation will alleviate much of these problems and

ENV/L22-rpt4 11-14 LEFT INTACT AS FISHERY HABITAT

CREATE CHANNEL

THOMAS ISLAND CREATE SEDIMENT CATCH SUMP INSTALL BOX CULVERT

FRANKLIN CONTROY REC. PAVILION

• BOAT HOUSE

FIGURE 11.1

AREA TO BE PROPOSED FLOW SHORT DREDGED CIRCUITING PROJECT PREPARED FOR: ONOTA LAKE PITTSFIELO. MASS PROJECT No. 541648

INTERNATIONAL 1984 IT CORPORATION TECHNOLOGY ALL COPYRIGHTS RESERVED CORPORATION Do Not Scale This Drawing DISPOSAL SITE

FIGURE 11.2 DREDGE MATERIAL DISPOSAL SITE

PREPARED FOR: ONOTA LAKE PITTSFtELD, MASS PROJECT No. 541648 APRIL 1988 additional costs. By allowing the sediments to dewater In place, for 1-2 months, before excavation, an appreciable amount of water will drain from the spoils. (Dewatering characteristics would be determined from core samples taken prior to the project.) If needed, as the spoils are being excavated, localized containment/dewatering areas or spoil stockpiling areas could be created within the boundaries of the excavation area to allow for final de- watering.

Second, during the lowering, the lake additional restoration activities could £>e conducted. This could include a general cleanup of refuse and debris, placement of benthic barriers, and excavation activities associated with in- lake sumps or biofilters (created wetlands).

Third, the total operation could be completed much quicker using dry excavation techniques.

The short circuiting project should coincide with the repair of the dam. It will be necessary to lower the lake to accomplish both projects. As such, it would be prudent to schedule both projects to occur simultaneously.

The nutrient budget data indicate that approximately 58.3% (517.7 kg/yr) of the total annual external TP load enters the North Basin (Table 10.10). Approximately 71.456 of the lake's annual hydrologic load discharges into the North Basin. In addition, the TP load generated by the Churchill Brook and Daniels Brook watersheds amounts to 226 kg (35.7%) of the total annual load. As such, the North Basin, although only 10% the volume of the South Basin, receives a significant proportion of the annual pollutant load. In addition, this basin, particularly 1n the vicinity of Thomas Island, is characterized by deep, organic sediments (Section 7.0), low dissolved oxygen (Section 5.0), and a substantial weed problem (Section 5.0). Implementation of the proposed project would Improve conditions of the North Basin and the entire lake as follows:

1. Remove/'a significant (2-4 x 105yd3) amount of accumulated organic sediment.

ENV/L22-rpt4 11-17 2. Decrease the density of weed growth in approximately 25 to 40 ha of lake surface.

3. Help alleviate the water D,0. problem by removing the organic silt and macrophyte layer.

4. Improve flushing of the two backwater coves which lie on either side of the Thomas Island Road causeway.

5. Decrease the hydraulic residence time in the North Basin thereby decreasing the retention of sediment and associated particulate forms of nutrients.

11.3.2.2 Aquatic Macrophyte Harvesting

Dense beds of Elodea and mixed beds of Myriophyllum, Vallisneria and El odea occur throughout the North Basin. These plants impede recreational use of the area, primarily in relation to boating and fishing. The seasonal dieback of the weeds also contributes to the lake's nutrient load. The bacterial decomposition of the dead macrophytes during the winter appears to be causing a severe reduction in dissolved oxygen (D.O.). This depletion in D.O. is substantial enough to potentially cause winter fish kills in the North Basin.

ENV/L22-rpt4 11-18 I I TABLE 11.2 IN-LAKE RESTORATION FEASIBILITY MATRIX I DEGREE CAUSE/ OF LOAD PERFORMANCE I METHOD EFFECT REDUCTION PRACTICALITY CHARACTERISTICS COST TOTAL Macrophyte I harvesting 3 3 5 5 3 19 Short Circuit I in-flow 5 5 5 5 3 23 Spot I Dredging 3 3 3 4 1 14

Aeration 5 5 3 4 2 19

Use of Macrophyte Barriers 1 1 4 • 4 5 15

Algaecide/ I Herbicide 1 1 1 1 4 8 Lake I Lowering 1 1 4 2 5 13 I I i

ENV/L22-rpt4 11-19 I Potamoqeton is the primary species of concern in the South Basin, although • Myriophyllum. and Elodea have been observed. Densities reach nuisance proportions primarily in the shallow littoral areas along the lake's south I shore. The implementation of an intensive weed harvesting program would greatly aid I in the control of macrophyte densities (Nichols, 1974). In addition to « increasing the total weed free surface area of the lake, harvesting could I greatly influence the internal regeneration of nutrients in the North Basin (Souza, et al 1987, Carignan, 1985). The objective of the harvesting • operation should be management of macrophytes to acceptable levels and not their elimination. For this reason, even in the North Basin, areas of • appreciable standing crop of macrophytes should be allowed.

Specifically, intensive harvesting (two cuts per summer) is required in the • following locations:

• • Thomas Island Cove (between Lakeway Drive and Thomas Island Road), — • The majority of the North Basin from Dan Casey Memorial Drive, • Along the west shore in the vicinity of Camp Winadu, and • To a lesser extent in the vicinity of Apple Tree Point in the • Main Basin. The majority of the macrophyte harvesting will occur in the North Basin. • Harvesting could be effectively accomplished using a mid-size Harvester such as the UMI HP7-400 or a full size unit such as the UMI HP8-700. Although i appreciable densities of macrophytes exist in the main basin, water depth and obstructions may inhibit harvesting. In addition, relative to cost/benefits, harvesting 1n the North Basin will yield the greatest returns.

If harvesting 1s Implemented independent of any of the other recommended in- Take restoration measures (particularly short circuiting of flow), it will be necessary to purchase equipment for Onota Lake. The proposed program will be effective only if done intensively. Based on existing weed growth patterns and given weed management objectives, it is estimated that 75 to 100 ha would require harvesting. Over 8036 of the total area 1s 1n the North Basin. The limited truck access to much of the North Basin will influence harvesting

ENV/L22-rpt4 11-20 efficiency. It is estimated based on review of design specifications of various harvesters, that it will take 30 to 40 working days, to conduct a • single harvest of the lake. Since the objective is to remove as much biomass as possible, multiple harvests are recommended. A double harvesting of the • lake, and possibly triple harvesting in particularly nuisance areas such as — Thomas Island Cove, will require 60 to 80 working days to accomplish. As such it does not appear feasible to borrow equipment used at Lake Pontoosuc or • Richmond Pond for use at Onota Lake. To accomplish the macrophyte management objectives, a single unit must be dedicated full time to operation at Onota lake.

However, if the short circuiting of flow project is implemented, a different weed control scenario will develop. The sediment excavation required as part of the short circuiting channel will eliminate most of the weeds in Thomas Island Cove and to the north of Thomas Island Drive. If this project is combined with the creation of a sump on the south side of the Dan Casey • Causeway (11.3.1.1) and use of benthic barriers the total amount of harvestable lake surface will probably be reduced to approximately 25 ha. The purchase of machinery to harvest such a small amount of lake surface is not justified. Given such a scenario, it would be more cost effective to rent equipment from Pontoosuc Lake or Richmond Pond. Based on a price of $150 to • $200/acre (much less than commercial harvesting costs), Onota Lake could be maintained in a managed state for approximately $7500 to $20,000/season (depending on the number of harvests). Since the county owns the equipment, and harvesting could be done by Pittsfleld DPW employees, it is highly likely that the harvesting costs could be less.

11.3.2.3 Spot Dredging

The sediment accumulation study (Section 7.0) indicates that the majority of accumulated sediment 1s 1n the North Basin. Considerable quantities (4.29 x 105 yd3) of organic sediments have been deposited 1n the Thomas Island Cove and at,the base of Dan Casey Memorial Drive. If previous identified projects were implemented, such as the creation of sumps (11.3.1.1) and the short- circuiting of flow discussed above, the amount of ancillary dredging required 1n the lake would be minimal. If the short-circuiting recommendation is not

ENV/L22-rpt4 11-21 I Implemented, however, dredging of the coves on either side of Thomas Island should still proceed. As much as 6 feet of silty, organic material was • measured in these areas. Its removal would benefit the lake as follows:

• Decrease the sediment organic load, thereby decreasing oxygen depletion, • Decrease the availability of in-lake pools of nitrogen and phosphorus, • • Decrease dense stands of nuisance macrophytes, • Improve recreational utilization, and • • Improve aesthetic attributes. Either dry or hydraulic dredging could be implemented. The use of diver- • operated suction dredge equipment is promising only if an adequate and ecologically sound means of handling the spoils and dredge slurry is developed. As discussed above, IT recommends that as much dredging as • possible be done using dry excavation techniques. • As with any dredging operation, some negative impacts will result. These are namely destruction of the benthic community, temporary increase in turbidity (hydraulic), possible regeneration of nutrients, possible algal bloom resulting from removal of shading macrophytes and regeneration of nutrients. All of these problems are temporary and short-term (USACOE, 1977). The benefits and improvements to the lake definitely outweigh these negative impacts.

11.3.2.4 Aeration

A sizable percentage of the lake's total phosphorus load is generated from anoxic sediment. The internal load is 345 kg TP/yr from the main (deep) basin and 28 kg TP/yr from the shallow basin. The majority of sediment regeneration of phosphorus occurs in the main basin. Hypolimnetic aeration could effectively reduce the internal load by oxygenating the deep level of the lake. An additional benefit which will be accrued through aeration will be the addition of available trout habitat. Based on an anoxic volume of 5.85 x i 106 m, one large or two small aerators would be required. Installation of the units would require the fabrication of a pump house and O&M associated with the compressor system. It will be necessary to shut the system down in the winter, first because 1t is not needed and second to avoid

ENV/L22-rpt4 11-22 interference with ice fishing.

Conversely, a smaller aeration unit is required for the North Basin. Its purpose would not be associated with aerating the hypolimnion in the summer but in contravening winter oxygen depletion. The DO curves show a substantial decline in winter 00 concentrations. February concentrations reach potentially critical levels (Figure 5.16). The cause for this depletion is most likely bacterial decomposition of detrital plant material and subsequent respiratory oxygen demands. The aeration unit would disrupt ice fishing in a section of the North Basin. However, it would prevent fish kills from occurring. In addition, it would most likely improve the lake's fishing as it would increase the volume of the North Basin suitable for cold water game fish (DO > 7 mg/1).

An alternate approach to controlling the internal regeneration of phosphorus is through sediment inactivation. This typically involves the subsurface application of sodium aluminate or aluminum sulfate. The alum inactivates the phosphorus, significantly decreasing Its availability to primary producers (Cooke and Kennedy, 1980).

Alum, in the form of aluminum sulfate, could be added to the hypolimnion using a submergible manifold. The correct concentration of alum would be determined on the basis of analyses conducted just prior to the application. The concentrate on of phosphorus, the pH, and the alkalinity of the hypo"!imnion will determine the dosage rate and concentration of the alum. Application should occur in mid-August after hypolimnetic phosphorus release has commenced, but before lake turnover. In this manner, adequate precautions could be taken to help insure that the sediment-liberated phosphorus is inactivated before it is circulated into the trophogenic zone. The specifics concerning cost and manpower allocation are yet to be fully developed. However, based on previous studies, the cost for an operation of this magnitude may exceed $25,000 in the cost of alum alone. Conversely, a single treatment could yield as much as 5 or 6 years of control of the internal phosphorus/load (Cooke, et al., 1982).

However, the utility of alum inactivation is limited at Onota Lake. First, in

ENV/L22-rpt4 11-23 lakes of low buffering capacity, problems arising from aluminum toxicity are possible. Trout are particularly sensitive to aluminum. Second, TP inactivation does nothing to improve hypolimnetic dissolved oxygen. As it is as equally desirable to increase available fish habitat as it is to decrease internal TP loading, aeration appears to be more feasible than nutrient inactivation.

11.3.2.5 Macrophyte Barriers

An extensive weed bed occurs along the northwest shore of the lake adjacent to a 1arge wet1 and (Figure 11.3). A combination of emergent and subtnergent species dominate the area. Fishing access cou1d be improved in this area through the development of "fishing lanes". These lanes can be created by using mechanical barriers which impede weed growth as opposed to harvesting. Material such as Texel or Dartec could be laid out. Lanes 50 - 75 feet long and approximately 12 feet wide would be created. This would also cause the creation of an edge ecotone which could also improve diversity and utilization of the area by large predator fish. The approach is relatively inexpensive.

11.3.2.6 Algacide/Herbicide Application

The present concentrations of algae do not require control via application of algacides. This will be better achieved through decreasing the nutrient load to the lake. Harvesting is greatly preferred in respect to herbicide application as the means of macrophyte control. For these reasons neither is recommended.

11.3.2.7 Lake Lowering

Manipulation of a lake's water level is used as a means of controlling macrophyte growth. The practice basically relies on the exposure of the lake bottom and the subsequent dessication or freezing of the weeds. When done properly, the roots and rhizomes of the plants, are adversely affected. In turn, growth in the subsequent season is impeded. The literature (Cooke, 1980) shows this practice to be successful for a number of weed species. Myrophillufi) spicatum, Elodea sp., Nuphar sp. and Chara sp. would be expected

ENV/L22-rpt4 11-24 I I I

I LEFT INTACT AS I FISHERY HABITAT, I I I I IL FRANKLIN CONTROY I REG. PAVILION

I FIGURE 11.3 PROPOSED WEED BARRIERS I PREPARED FOR: ONOTA LAKE LEGEND / V U DAM PITTSFIELD, MASS PROJECT No. 541648 I APRL 1988 PROPOSED WEED BARKERS I 600 1984 IT CORPORATION ALL COPYRIGHTS RESERVED scale feet I 'Do Not Scale This Drawing" I

I to show a decrease in abundance after lake level drawdown. Other weed species such as Potamoqeton sp. and Vallisneria sp. may not be affected by drawdown, • and Najas sp. may increase in abundance after drawdown (Cooke, 1980). The following three case studies indicate the varying degree of macrophyte control • that may be obtained via lake level' drawdown.

• In Mondeaux Flowage, Wisconsin, Potamoqeton robbinsii (Robbin's pondweed) • was the dominant vascular plant. Over-winter drawdown was performed during _ the winters of 1971 through 1972 and 1972 through 1973 (Nichols, 1975). B Forty percent (66 ha) of the 166 ha was relieved of nuisance plant growth after one drawdown. The second drawdown provided little additional control • except for a further reduction in the abundance of Nuphar variaqatum. In 1974, the abundance of plants returned to predrawn levels and Ceratophyllum • demersum became the dominant macrophyte. In addition, dissolved oxygen levels during the winter drawdown became low but no fish kills were reported. Based on these results, Nichols (1975) suggested that a drawdown • every 2 to 3 years would have been more, effective than an annual water

— withdrawal since resistant plants might not be able to become established.

• • Beard (1973) reported that 42 percent (303 hectares) of Murphy Flowage, • Wisconsin was obstructed to the extent that fishing was not possible. Robbin's pondweed (P. rofabinsii) was the dominant nuisance plant, and the • sub-dominants were Nuphar sp. (water lily), Ceratophyllum demersum (coontail), Potamogeton natans, (floating-leaf pondweed), and Myriophyllum sp. (water milfoil). In 1967 and 1968 the water level was lowered 1.5 • meters (5 feet) from November through March, and restored in April. There _ was a 92 percent reduction in area covered by plants after drawdown, and I all plant species were controlled or eliminated. Fishing became possible in 87 percent of the area previously covered by the plants. In 1969, • Potamoqeton natans, Meqalondonta beckii (bur marigold), Najas flexilis (naiad) and Potamoqeton diversifolius became abundant but did not impair • fishing to the extent of the previous plants. In this case, nuisance rooted plants were controlled by a winter exposure of sediments, and areas previously closed to fishing were opened. However, an algal bloom occurred • in August of the year after drawdown.

ENV/L22-rpt4 11-26 I

• • Kahle Lake (102 hectares/12 m average depth) in Clarion and Venango Counties 1n Pennsylvania was created in 1974 by impounding Mill Creek. Few | aquatic vascular plants were present until 1976. The dominant species, bushy pondweed (Najas flexilis) comprised approximately 90 percent of the H plant community (Tazik et. al., 1982). N. flexilis was abundant in the shallow waters of Lake Kahle. Drawdown was performed in September 1977 and refilling was completed in April of the following year. Results indicate • that drawdown was ineffective in controlling N. flexilis.

B Lake level manipulation has been used extensively at other lakes neighboring Onota Lake, also with varying degrees of success. William Enser reports • (Person. Comm., 1987) that drawdown has had successful results in other Berkshire County Lakes. Drawdown at Richmond Pond produced mixed results. At H Pontoosuc Lake, drawdown successfully reduced the densities of both Myriophyllum spicatum and Potamoqeton crispus.

In 1988, IT Corporation performed an evaluation of lake level control as a _ macrophyte management strategy in Pontoosuc Lake. The general conclusions • drawn from this study were as follows:

• » The selection of drawdown for macrophyte control should be made using site specific data which must include data on macrophyte species and • densities. The two dominant nuisance species in the lake, Myriophyllum spicatum and Potamogeton crispus can be effectively control 1ed through M drawdown procedures.

^ • Fish and benthos populations can be impacted by drawdown. Obviously, an • extensive drawdown that significantly reduced fish habitat would result in increased competition and predation, and therefore a decrease in fish • populations. Benthos in the exposed sediments would be impacted, thereby reducing a food source for fish species which feed on invertebrates. Also, • reduction or elimination of near-shore macrophytes via drawdown would reduce cover utilized by fish for spawning and/or protection. These potential impacts would be minimized by drawing down to a moderate, • carefully chosen level. Refill of the lake should be complete before the _ spawning period of the first species to spawn which 1n Pontoosuc and Onota | Lake is the Northern Pike.

— ENV/L22-rpt4 11-27 I I • Small numbers of herptiles hibernating in mud in the shallows may be • impacted by the removal of the water thermolayer during drawdown. • • Little adverse impact would be expected on migratory waterfowl, mammal populations, and emergent vegetation. i These conclusions were based on a six foot drawdown, which was judged to provide the greatest cost/benefit for conditions existing in Pontoosuc Lake • (IT Corp. 1989). M The depth of drawdown for macrophyte control must often be a compromise between a depth that will affect the maximum acreage of nuisance vegetation, H and a lesser depth which will ensure that the lake can be re-filled prior to spring spawning of fish. In Onota Lake, a drawdown of four feet would expose large areas of the North Basin, where Myriophyllum spicatum is the dominant • nuisance plant species (Figure 6-1). Myriophyllum spicatum has been shown to •be susceptible to dewatering (Smith, 1971), and would be expected to show reduced growth if sediments were dessicated for at least 21 days. Potamogeton crispus, which is abundant in the southern end of the North Basin and along • the shoreline of the South Basin, may not respond to drawdown (Cooke, 1980), although, as previously stated, this species has been reduced by drawdown in H nearby Pontoosuc Lake. Assuming the calculated hydrologic input to Onota Lake (17 x 10 m /yr, Table 9-7) which gives an average monthly input of 1.4 x 10°mC ",J and interpolating for reduced area at lower lake levels, less than two | months (1.87 months) would be required to refill the lake subsequent to a four — foot drawdown.

The potential for lake level drawdown as a macrophyte control strategy may be fl evaluated by considering the following factors:

• • Winter climate of sufficient severity to freeze exposed sediments, and adversely affect macrophyte roots and rhizomes

• Hydrologic input sufficient to refill lake prior to fish spawning, after lowering to a level which exposes significant macrophyte infested areas of i lake bed. i ENV/L22-rpt4 11-28 • Presence of macrophyte species susceptable to control by exposure.

Since all three factors are judged to be positive for Onota Lake, and given that negative impacts are expected to be minimal, it is recommended that drawdown be incorporated into the lake management program. Drawdown offers the potential for cost effective macrophyte control, and should be particularly effective in controlling Myriophyllum and Elodea in the northern basin.

Given the occurrence of species (Najas sp.) whose growth may increase after exposure of short duration (Cooke, 1980), it is best to ensure that the drawdown is of sufficient duration to avoid this possibility. The lake should, of course, be drawn down during the winter months, and its success as a macrophyte control measure may be directly related to the duration and intensity of sub-freezing temperatures.

11.3.2.8. Power Boat Limitations

As indicated in Section 6.0 (Macrophyte Survey), there exists some evidence to support the contention that nuisance plants are being introduced to the South Basin as a result of prop disturbance, fragmentation and boat-related distribution. In order to curtail the spread of weeds by boats a speed restriction (no wake, <5 mph) should be applied to the entire area northwest of Thomas Island. Although water skiing is popular in the North Basin, most activity is confined to the south side of Thomas Island and the beach area adjacent to Camp Winadu.

Low speeds 1n the northwest end, which is heavily weed-choked, will decrease the fragmentation of Elodea and help reduce the boat related spread of nuisance weeds.

In addition, through a public education effort, boaters should be encouraged to clear their props when travelling from the North to South Basin. The shallow water depths in the vicinity of the sand bar already require boats to slow down during passage to the south. Removal of weeds fouling props would take but a few additional minutes. ENV/L22-rpt4 11-29 I I 11.3.2.9 Low Priority Options

Other in-lake management options were reviewed, including;

• Bottom sealing, • • Selective discharge, ™ • Dilution flushing, i • Biological controls. None of these appear to be reasonably feasible management options. Bottom • sealing and selective discharge are means of decreasing internal regeneration of nutrients. The use of the aeration system will accomplish the same, while • increasing hypolimnetic DO, increasing cold water fish habitat and only slightly disrupting the thermocline.

™ As a large proportion of the lake's hydrologic budget is comprised of ground- water seepage, dilution/flushing will not provide a great benefit. In addition, a substantial quantity of ground water would be required to decrease the lake's flushing rate (more frequent flushing). As such, it is not recommended.

At present, biological controls have had only a limited success in lake management. Of the various techniques, management of macrophytes by herbivorous fish such as grass carp is the most promising. However, at .this time, the Commonwealth of Massachusetts prohibits the introduction of grass carp. Until proper stocking ratios are better established, use of grass carp should not be pursued.

11.3.3 Recommended Plan

Following the review of the diagnostic data, the feasibility assessment and d i alogue wi th LOPA and c1 ty off1ci als, a Restoration Management PI an consisting/of a combination of in-lake and watershed techniques was developed. In order to achieve improved water quality, recreational potential, and aesthetic attributes in a cost-effective manner which 1s responsive to local

ENV/L22-rpt4 11-30 needs, the schedule of implementation should be as follows:

1. Lower lake level in fall a minimum of four feet. Allow sediments to de- water in situ and conduct the necessary excavation and construction required to facilitate short circuiting of flow in the North Basin.

2. While the lake is lowered, conduct any repairs needed to maintain the structural integrity of the lake's spillway.

3. Concurrent with the short circuiting project, create a sump at the south side of Dan Casey Causeway and supplement the wetland area north of the causeway with additional planting to further facilitate nutrient and sediment retention,

4. Install benthic barriers along the northwest shore of the North Basin, placement of which is designed to improve fishing access and enhance existing fish habitat by creating edge ecotone,

5. During the first spring/summer following these activities, observe weed growth and determine the necessity for harvesting. Given low remaining acreage of nuisance beds following implementation of Tasks 1, 3, and 4, contract with the county to share a harvester from either Pontoosuc Lake or Richmond Pond.

6. Conduct reconnaisance of immediate shoreline, identify areas of existing soil erosion and erosion prone areas. Develop the required-approach to address each site and provide for future monitoring to identify new areas as they arise.

7. During spring, install hypolimnetic aeration unit(s) 1n South Basin,

8. Initiate required bonding provisions for the sewering of Blythewood Drive,

9. Initiate public education programs,

ENV/L22-rpt4 11-31 10. Develop final engineering drawings and bid specifications for the Parker Brook detention basin and the Blythewood Drive biofliter, once land availability and easement issues are rectified, and

11. During course of any restoration, Implement a monitoring program, designed specifically to examine the objectives of each project. These project specific monitoring programs should be part of a continuing lake-wide monitoring program designed to examine the status of the lake.

11.3.4 Public Education and Involvement Program

An integral component of the Restoration/Management plan is Public Education. In order to maintain public interest and support, it is important that the public be kept informed of interim findings and proposed solutions. Equally important is the fact that they must know what is being done, why it is being done, and how they, the lake users, will benefit. In addition, they must become educated as to what they can do as individuals to improve the lake. This can all be accomplished through several avenues, including:

1. News releases to local newspapers, 2. Special quarterly public meetings or seminars to discuss the issues that affect the lake, progress being made, and proposed activities, 3. Monthly public meetings of LOPA, 4. Presentation of papers at conferences and Technology Transfer Seminars, 5. For all projects which may impact the lake or its tributaries, involvement of LOPA in the decision-making processes of local government, particularly the Environmental Commission, Board of Health and Planning Board, 6. Special presentations to the various officials and professionals of the communities 1n the watershed, 7. Active participation in the data-gathering activities of the study, and 8. Dissemination of information to watershed residents concerning Best Management Practices and grass roots activities which help reduce pollutan/ t loading to the lake, LOPA and the City of Pittsfield recognize that, as well as being informed, the public has to be involved. The residents of the watershed will be requested

ENV/L22-rpt4 11-32 ^ to support the plan. To date, volunteers have provided assistance in the monitoring of rain and staff gages and the collection of storm samples. • Public involvement of this type identifies the community as being sincerely concerned and interested in the lake's restoration. It is important that the • public considers the plan and the lake as their's and that they are obligated to implement it and provide for the lake's future management.

• Local citizens can be especially helpful in contacting their elected

— representatives and gaining support for the project. The public, and public | officials at all levels of government, must become more fully aware that Onota Lake is the city's most important natural resource. The economic welfare of • the area also depends in part upon the quality of Onota Lake as a recreational resource. This effort has to be continued and intensified if the plan M developed herein is to be successful.

In addition to the above, the following are specific recommendations which • will aid in the acceptance and implementation.of the lake restoration plan:

|| 1. The City of Pittsfield should examine, strengthen, enforce, and/or adopt storm runoff control and erosion control ordinances and septic system ordinances. 2. The master planning process and zoning ordinances should channel development toward areas which will have no impact on lake water quality 'and away from environmentally sensitive areas such as Blythewood Drive. 3. Once new zoning and development ordinances are adopted, the communities should insist on rigid compliance with the plan. The temptation to zone by parcel through variances and zoning adjustments should be resisted. A municipality can have an excellent master plan and ordinances, but, fail to achieve their goals because of lack of enforcement. 4. Tighten up ordinances to include Environmental Impact Statements for all majpr and minor subdivision projects. 11.3.5 Institutional Arrangements

The proposed lake restoration and watershed management plan will be successful only 1f local, state and federal institutions see them through. The institutional arrangements for implementing the plan presented in this report will necessitate approval by the city, the county and the Commonwealth (Figure 11.4). The number of agencies, laws, regulations and levels of government

ENV/L22-rpt4 11-33 I I FIGURE 11.4 PROJECT ORGANIZATION I iFUNOlNfii | PROJECT TFE1ERAU II 3EPr Term ItaUNtYl ADMINISTRATION L j ' 1 • fity of putsfleld I QVQC 1 IDEP 1 PROJEtt MANAGER LOPA I PROJECT 1 CONSULTANT) i 1~~ -, | I SHORT CIRCUITr ; Stormwater Public I PROGRAM Management Education ] 1 —} Project I n-TVi i -r i. m — i— i— 1— i- it ' 1 1 1 1 Consultant ROAD • i SOIL EROSION EWIRONMENT AERATION CONTRACT FISHERY ~ '~ '' DREDGE - RECONSTRUC. MONITORING PERMITS ' CONTROL HONITOR SO. BASIN HARVEST IMPROVEMENT OESIGK PERMIT CONSTRUCTION MOHITORIHG MAINTAIN Contractor Contractor Project Project City or Project contractor Contractor Engineer Project Contractor Project City and City and City Consultant Consultant County Consultant Project Consultant Consultant Consultant County I In-Kind Consul tant C1 ty

ISEHEl "IMS HABITAT VTinriwp I I IMPROVE 1 DEP - 1 DESIGN CONSTRUCTION MONITORING MAINTENANCE lonsultant

Engineer Contractor Project City I LDEPi city City Consultant Couniv County Engineer I Proj. Consult. Review I I I I I I I I • that are involved with implementing the program could become complicated. For instance, construction of stormwater management basins (as proposed at the f mouth of Parker Brook) could involve DEP, COE, and County Environmental Commission approval.

Therefore, it is essential that one local organization be made responsible for •| coordinating and implementing the plan. The organization serving this ™ coordinating function at this time is the Onota Lake Preservation Association. It is recommended that the OEQE and City of Pittsfield recognize • the LOPA to implement and oversee the management/restoration plan. LOPA should act on behalf of the city in the management of Phase II activities.

To provide for the long-term management of Onota Lake, the Commonwealth of • Massachusetts should examine the feasibility of creating an Onota Lake Management District. Among other powers, the district could, assess taxes throughout the watershed, apply for, receive and administer grants, and assess • user charges to develop and implement lake management and restoration and «M watershed management programs.

LOPA should continue coordinating with the various political entities to • maintain and improve Onota Lake as a viable recreational resource. It should be LOPA's responsibility to implement the management and restoration plan in • cooperation with the various governmental and private institutions. Also, as an independent organization concerned with the lake's improvement, they should • review and comment on all zoning changes, master plans, development and ™ regulations that affect Onota Lake.

• 11.3.6 Permit Requirements

• Permits must be secured from the required state, federal and local conservation agencies before implementation of the restoration measures can • begin. i Activity / Permit Required Permitting Agency 1. Lake lowering Notice of Intent (NOI) Pittsfield Conservation Commission

ENV/L22-rpt4 11-34 Ch. 91 DEP-Dlv. of Wetlands and Waterways Notification ifState Stocked Mass. Div. of Fisheries and Wildlife (MDFW) ENF MEPA (if >5000 ft2 of wetlands involved)

2. Dredging Notice of Intent (NOI) Pittsfield Conservation Commission Ch. 91 DEP-Div. of Wetlands and Waterways Sec. 404 Army Corps of Engineers (ACOE) Water Quality Cert. DEP - Div. of Water Pollution Control ENF MEPA

3. Construction N.O.I. Conservation Commission Ch. 91 DEP Sec. 404 ACOE ENF MEPA Water Quality Cert, OEP-Oiv. of Water Pollution Control

4. Wetland N.O.I. Conservation Commission ENF MEPA Sec. 404 ACOE

The Soil Conservation Service does not-have a required permit. The Phase II consultant would file Section 404 and Chapter 91 permits. The City of Pittsfield would file Notice of Intent and the Environmental Notification Form (ENF). The city must write a letter to the Director of Massachusetts Oivison of Fisheries and Wildlife ten days prior to lowering the lake specifying depth of drawdown, duration and expected refill date.

11.3.7 Monitoring Program

A two-stage monitoring program should be developed as part of the Phase II Program. The first stage would be designed for the continued assessment of lake quality. This should involve quarterly monitoring at each of the in-lake and tributary stations monitored during the 1986 study. Monitored parameters should include:

ENV/L22-rpt4 11-35 I

• TP Fecal coliforms Soluble reactive phosphorus (SRP) Transparency

| N03 D.O.*

NH3 Temperature* • TKN Conductivity* TSS pH* Chlorophyll a,b,c phaeophytin • Alkalinity i Those parameters followed by an asterisk should be conducted in-situ i throughout the entire water column at 1.0 m intervals. Concurrent with this "background" monitoring program, individual programs need H to be developed as part of each restoration task. These project-specific programs should be designed to assess environmental impact during program • implementation and assess the effectiveness of the program. Lake lowering and projects which include construction (in-lake sumps, • biofilters, short circuiting of flow, dredging) should have a monitoring program which requires weekly sampling during drawdown or construction. • During drawdown an in-situ station in each basin should be monitored for D.O., NH3, N03, TP, and SRP. Samples should be collected 1/2 meter below the • surface and 1/2 meter above the bottom. D.O. should be monitored in-situ throughout the water column at 1-meter intervals. i For any construction related projects, sampling should occur up-stream, within, and downstream of the project boundaries. If the lake is lowered to • facilitate operations, a station should be located within 50' of the project boundary, or where water depths are at least one meter. A second station • should be located approximately 50' downstream of the first. Regardless of the scenario, the following parameters should be monitored:

D.O. / NH3 pH N03 TP TKN

ENV/L22-rpt4 11-36 SRP TSS Transparency

Following the completion of each stormwater management project, sampling emphasis should be placed on quantification of storm load reduction. This requires wet weather sampling. At a minimum, samples should be collected during three storms every year for a total of 3 years.

Stations should be established above, within and below the stormwater management structure. Storm sampling should be consistent with DEQE Phase II requirements and include analysis for:

TP TSS SRP As, Cd, Cr, Pb, Hg, Fe, Al, Ni, Zn, Cu

N03 Oil and Grease (Petroleum Hydrocarbons)

NH3 Chlorides TKN Fecal coliform TDS

The monitoring program associated with macrophyte control should utilize the in-lake "background" water quality data. This will reduce sampling redundancy. The water quality should be augmented by plant surveys done in June, August and September. 7\ 10 x 10 meter test plot (harvested quadrats) and control plot (non-harvested quadrats) located in each basin should be monitored on each date for:

density species composition percent coverage total biomass

Seine or electroshocking techniques should be used to assess the fish community associated with each plot. These data could be used to assess the success of the harvesting program relative to changes in macrophyte species composition, re-growth time, biomass harvested, and effects on fish utilization.

ENV/L22-rpt4 11-37 11.4 ANTICIPATED IMPROVEMENTS

Implementation of the recommended restoration projects are anticipated to improve the water quality, recreational use and aesthetic attributes of the lake. Improvements in water quality are based on literature cited performance capabilities of the proposed stormwater management measures and the loading data developed in Chapter 10. The load reduction and improvements in lake use assume the following restoration plan will be implemented:

South Basin

1. Sewer Blythewood Drive

2. Aerate hypolimnion during summer.

North Basin

1. Repair Peck Road sewer break,

2. Implement short circuiting project:

a. lower lake, expose north basin sediments, b. allow sediments to dewater, c. install box culvert under Thomas Island Road, d. excavate accumulated sediment, e. create sump at the base of Dan Casey Causeway f. Refill Lake

3. Conduct selective weed harvesting.

4. Improve fishery habitat.

Watershed Management

1. Implement stormwater management measures.

ENV/L22-rpt4 11-38 I I I 2. Conduct public education programs. I 3. Correct soil erosion problems. It is anticipated that the South Basin projects should reduce the total I phosphorus load associated with septic sources by 188.2 kg and from internal sources by 173 kg, assuming a 50% improvement due to aeration. The total South Basin TP reduction would, therefore, be 361.2 kg (based on Table 10.10 I values). The North Basin project will result in an estimated reduction in loading of 197.3 kg. This is a conservative estimate and assumes that the I combination of short circuiting, the Dan Casey sump, and the biofilters at the mouth of Churchill and Daniel Brooks will provide only a 40% reduction in TP I loading. The excavation associated with the short circuiting program should substantially reduce the North Basin's volume of organic sediments. This should effectively eliminate the TP regeneration associated with the anoxic I hypolimnion, resulting in the elimination of 28.0 kg/yr of total phosphorus.

I The sum of the load reductions associated with the various restoration measures totals 361.2 kg for the South Basin and 225.3 kg for the North I Basin. This amounts to a 50% and 41% TP load reduction for the South and North Basins, respectively, and results in a total TP load of 354.8 kg for the South Basin and 320.4 in the North Basin. The resulting areal load will be I 2 2 0.212 g/m /yr and 0.403 g/ni /yr for the South and North Basins, I respectively. Using these annual loads, in conjunction with the trophic state model of Dillon, the predicted spring TP concentrations (Ps) for each basin will be reduced. The trophic state for each basin will improve given these I reductions (Table 11.3, Figure 11.5). I As equally Important, the implementation of the proposed Restoration/ Management Plan will increase the recreational potential of the lake. This I will be achieved through macrophyte control, deepening of shallow areas, improvements in fishing access, and improvement in cold water fish habitat. These improvements in the lake will greatly benefit lake users. In addition, I the public education program will help users become attuned to the overall I scope of the Restoration Management Plan. Through seminars and other planned I ENV/L22-rpt4 11-39 I

I TABLE 11.3 PREDICTED POST-RESTORATION TROPHIC STATE FOR THE NORTH AND SOUTH BASINS OF ONOTA LAKE

Predicted spring TP concentration (g/m ) = [Ps] = L(I-R)T

SOUTH BASIN

2 [Ps] = 0.212 q/m /yr (1-0.734) (2.82 yr) 8.5 m

2 • [Ps] = 0.159 g/m 8.5m

H r n i _ n 1 n « / o Ps] = .019 g/m-m

NORTH BASIN

2 Ps] = 0.403 g/m /yr (1 - 0.593) (0.116 yr) 1.8 m

2 Ps] = 0.019 g/m 1.8 m

3 [Ps] = 0.011 g/m

ENV/L23-rpt4 11-39 I I activities, individuals can also become more aware of how they can help I improve or maintain the quality of the lake at the grass roots level. I 11.5 SCHEDULE OF ACTIVITIES The proposed plan will take at least five years to implement. A projected time line has been prepared (Figure 11.6). The time line shows the preferred I scheduling of activities. If this schedule is followed, the restoration of Onota Lake should proceed in a manner consistent with the prioritization I analysis conducted in previous sections of this report. This will help insure I cost effectiveness. 11.6 BUDGET

I The budget has been prepared using engineering estimates (Means, 1985), review of literature, and communication with restoration contractors and/or I suppliers. The budget has been prepared assuming the Thomas Island short circuiting project is implemented and the schedule of activities as presented I in Figure 11.6. I The budget breakdown is presented in Table 11.4. It should be noted that an approximate 5 year completion period is projected for the proposed plan._ The elements of the plan have been scheduled not only to meet restoration needs I but to avoid an excessive monetary burden being placed on the city or I Commonwealth. I I I I I I ENV/L32-rpt4 11-40 I

TABLE 11.3 • PREDICTED POST-RESTORATION TROPHIC STATE FOR THE NORTH AND SOUTH BASINS OF ONOTA LAKE

3 Predicted spring TP concentration (g/m ) = [Ps] = L(I-R)T

— SOUTH BASIN

2 [Ps] = 0.212 g/m /yr (1-Q.734)(2.82 yr) • 8.5 m

2 • [Psl = 0.159 g/m 8.5m

3 ™ [Psl = .019 g/m

• NORTH BASIN

2 I [Ps] = 0.403 q/m /yr (1 - 0.593) (0.116 yr) 1.8 m

2 [Ps] = 0.019 g/m • 1.8 m

— [PJ = 0.011 g/m3

ENV/L22-rpt4 11-41 TABLE 11.4 ANTICIPATED COSTS OF THE ONOTA LAKE RESTORATION PROGRAM

ACTIVITY COST

Weed harvesting (contract) $19,500/yr x 4 yrs Aeration South Basin 150,000 Watershed Management 1,500,000 Pecks Road Sewer"!ine 1,200,000 Public Education 7,500/yr x 5 yrs Annual Monitoring 30,000/yr x 5 yrs Short Circuiting Project Dredging 730,000 Construction 750,000 Miscellaneous 75,000

PROJECTED TOTAL* $4,670,500

* Excludes sewering of Blythewood Drive

ENV/L22-rpt4 11-42 TROPHIC STATE A5FIGURS PREDOEDFOR ONOTEA lAKE' 11.S NORT5H AMA£) SOUTH BASINS 10

(N

II 0.1 -

NORTtTBASIN ORTH BASIN

0.01 -

0.001 0.1 DEPTH (m) 10 100 BEFORE RESTORATION v AFTER RESTORATION I 1 FIGURE 11.6 I SCHEDULE OF ACTIVITIES AND MILESTONES FOR THE RESTORATION OF I ONOTA LAKE

Fft j ACTIVITY YEAR1 YEAR 2 YEARS YEAR4 YEARS I 01J08|D3|04|D5|06|D7|03|09 10 11|12 13IHI15 16 17118119120 SiEeB3E4 55e6C7EeE9130 31Q2p3p4B5t36 37p8|39|40 4l|42|43|44i45l46|47|48 4960l51p8p3p4| SEWERING 01 REPAIR PECK ROAD BREAK I02 SEVER BLYTHEWDDD DRIVE SHORT CIRCUIT DF FLOW IN NORTH BASIN 03 LOVER LAKE I04 SEDIMENT DEWATETR 05 BOX CULVERT INSTALLATION 06 EXCAVATION OF SEDIMENT TD CREATE CHANNEL Io; CREATE SUMP AT DAN CASEY CAUSEWAY 08 REFILL LAKE WEED HARVESTING I09 MONITOR NEED AFTER COMPLETION OF NORTH BASIN EXCAVATION 10 FISHERY IMPROVEMENTS n IF NEEDED, CONTRACT WEED HARVEST I12 INSTALL AERATION UNIT CONSTRUCTION OF STORM WATER MGMT. STRUCTURES 13 CHURCHHILL/DANIELS I 14 PARKER BRDDK I15 BLYTHEWDDD SOIL EROSION' CONTROL ft CONDUCT INTENSIVE SURVEY I7 PREPARE RESTORATION MEASURES R IMPLEMENT „ 9 CONTINUE SURVEILLANCE I PUBLIC EDUCATION ~ 20 SEMINARS ?1 MONTHLY LUPA MEETINGS f ' *> \r—i I?? SCIENTIFIC PRESENTATIONS ' * P3 NEWS PAPER ARTICLES 4 INTERIM REPORTS -I y— -H 25 ^PIMTTnPTMn TM t Ai/r 126 STORM MANAGEMENT I ~" * — • " . i . . I

I The budget associated with the restoration and management of Onota Lake is difficult to actually project (Table 11.4). This is largely due to the costs I associated with the sewering of Blythewood Drive. Basically, there were a few different engineering approaches to this project. These alternatives were I presented previously. Depending on which is deemed most feasible, the cost could vary by as much as an order of magnitude.

I Weed harvesting costs were estimated on a contract cost/unit area of $325/acre. If the total acreage in need of harvesting, following the Thomas I Island short circuiting project, is 60 acres, the annual contract harvesting cost will be $19,500/yr. This is admittedly an over estimate. IT experience I with other lake harvesting projects show that as much as 175 acres can be harvested for $20,000/yr, particularly if the county owned equipment and City I DPW employees are used to do the harvesting. The contract cost should be anticipated to increase by IQ% annually.

I The estimated cost of aeration includes the cost of equipment, its deployment and set up and its annual cost of operation (Atlas COPCO, pers. comm., I 1987). In terms of annual cost, aside from routine maintenance which requires scuba divers, cost of operation is only $2,000-$4,000 according to the I manufacturer, with most of that associated with the electricity needed to I power the compressor. Land use management and the reduction in nutrient/sediment loading to Onota Lake will require implementation of a number of small to medium independently I executed projects. The specifics of these projects were presented previously in this section. The creation of wetlands and construction of in-lake sumps I are best accomplished by first drawing down the lake and utilizing standard construction equipment. There is, a substantial amount of design work I associated with these programs and environmental monitoring is also required. In addition, permits will be needed. These costs are ancillary I expenditures associated with the actual construction but need to be addressed. It is expected that the total cost of watershed management projects w,m be $1,500,000.00. This will cover the design and construction I of the Blythewood Drive Biofilter ($75,000.00), the Daniels/Churchill wetland I enhancement ($85,000.00), and the Parker Brook Detention Basin I ENV/L22-rpt4 11-45 ($750,000.00). The remaining $590,000.00 would be used to correct existing soil erosion problems.

The Pecks Road sewer line break requires immediate attention. The City of Pittsfield has recognized the need to correct this problem rapidly and as such has authorized a $1.2 million construction bond. A city council resolution authorizing this expenditure was enacted in May of 1987.

A continuing public education program is strongly recommended. At present, the lake community is essentially represented by a few independent lake associations. To do so will require a moderate budget to prepare presentations, distribute information and conduct seminars on current lake issues. A consultant will most likely be required to provide technical guidance and support. An annual operating budget of $7,500 should cover such activities.

In addition, further lake monitoring is recommended. This will generate the data base required to quantify the success of restoration practices listed below and enable identification of improvements observed in lake quality, aesthetics or use: 1. lower lake 2. excavate sediment 3. re-build Thomas Island Road culvert.

A comprehensive monitoring program would cost in the vicinity of $30,000/year.

As with any dredging project, costs associated with lab analysis, permitting and monitoring must be accounted for. Mobilization and demobilization of major construction equipment and provisions for a temporary access road for Thomas Island residents are other factors which must be considered.

It is anticipated that a total of 4.5 x 105 yd3 of sediment will be excavated as part of this project. This includes dredging to the north and south of Thomas Island Road, creation of a sump at the base of Dan Casey Memorial Causeway, and excavation of a "canal" from the sump toward the new culverts to promote short circuiting of flow. Preliminary cost estimates developed

ENV/L22-rpt4 11-46 I through Means (1985) projections set the sediment excavation component of the I project at $730,000. Reconstruction of the road at current engineering rates will cost approximately $750,000. This includes the demolition of the old road, I installation of box culverts and provisions made for temporary access and I relocation of water, sewer and power lines. Permit requirements, including analytical costs, monitoring and consulting I could add as much as $75,000 to the project cost. I In total, the short circuiting project is anticipated to cost $1,555,000. The cost associated with most of these projects could be decreased substantially if in-kind services were provided by the city or county. These services could be in the form of labor and equipment. This could include the preparation of design spec drawings and calculation of engineering estimates.

11.7 ADMINISTRATION OF FUNDING I The proposed management plan was developed with the assumption that funding for implementation would be available through the Massachusetts Clean Lakes Program. This program is no longer in existence and state funding is not I presently available for lake restoration projects. Therefore, it will be necessary to develop innovative sources of funding if the elements of the plan I are to be carried out. I The successful implementation of the proposed plan will require substantial cooperation and integration of effort at the local, county and state levels. The City of Pittsfield will officially be the recipient of restoration I funds. However, LOPA should act as the City's grant administrator. LOPA should remain prominent in the restoration of the lake by assisting the City I in the .selection of contractors, overseeing project activities, etc.

• If sources of funding become available, it may be required that matching funds I be provided by the City of Pittsfield and Berkshire County. It is recommended I ENV/L22-rpt4 11-47 that in-kind services be provided to help defray the cash contribution that may be needed for local match. Potential in-kind services include:

1. LOPA Project Administration 2. LOPA volunteer monitoring 3. City and county donated equipment and labor for use in weed disposal, sediment excavation, dredge spoil disposal, soil erosion control 4. Cost of sewer construction and repair 5. Improvements to Burbank Park related directly to increased public access or use of Onota Lake.

To properly implement the plan, allocate funds, and provide in-kind services, the project organization should be as identified in Figure 11.4.

11.8 ENVIRONMENTAL EVALUATION

1. Will the project displace people?

No.

2. Will the project deface existing residences or residential areas?

The construction of detention basins and the retrofitting of existing stormwater collection systems with sediment traps may temporarily disrupt residential areas during the construction phase. The installation and repair of sanitary sewers may have similar effects on residential areas. Residential land use will not be impaired and existing residences will not be defaced.

3. Will the project be likely to lead to changes 1n land use patterns or an Increase 1n development pressure?

There will be increased pressure to develop along Blythewood Drive once this area is sewered. Stricter zoning and identification of areas as environmentally sensitive will be required to protect this section of the lake from future impacts. Other sub-basins of the

ENV/L22-rpt4 11-48 I lake's watershed are either established residential areas or forests and parks protected from development. Implementation of sound • watershed management should mitigate or greatly reduce the impacts of changes in land use or development induced by the improvement of the • lake's trophic state.

4. Will the project adversely affect prime agricultural land or 1 activities? • No. Recommendations include the implementation of Best Management Practices to decrease nutrient and sediment loading from agricultural • sources (Section 11,3.1.2). However, the BMPs do not adversely affect ™ farming practices or prime agricultural land.

5, Will the project adversely affect parkland, public land or scenic land?

No. In fact, the project may actually lead to the preservation of • additional open spaces through the implementation of environmentally sensitive zoning ordinances.

™ 6. Will the project adversely affect lands or structures of historic, architectural, archaeological, or cultural value?

No.

7. Will the project lead to a significant long-range increase in energy i demands? The most significant energy demands will be associated with aquatic • macrophyte harvesting and hypolimnetic aeration. As both are recommended annual restoration/management practices, there will be • long term O&M costs. Part of this will be the energy demands required ™ , to operate both.

• Previous analyses conducted by the manufacturer of both the harvesting equipment and the aerator indicate the energy demands to be minor.

_ ENV/L22-rpt4 11-49 First, both will be in operation only 5 months per year (mid-May through mid-October). Based on fuel consumption guidelines provided by the manufacturers, the harvester should require only 900 gallons of diesel fuel/season. This is based on an average operating time of 200 hours/month and fuel usage of 0.90 gal/hour. The requirement for proper aeration of the hypolimnion will be 50 ft •5 /minute. The compressor(s) will most likely be electrically powered. The draw will be comparable to the residential use of electricity for a one year period.

Dredging and the construction of sedimentation traps/detentions basins and the Thomas Island culvert will have associated energy demands.

8. Will the project adversely affect short-term or long-term ambient air quality?

Due to construction of detention basins and repair and/installation of sanitary sewer lines, and the construction of the Thomas Island culvert, there may be a short-term decrease in air quality relative to the generation of dust from construction sites. Proper sediment erosion and dust control practices will minimize such impacts. Increased air emissions through activities such as dredging and weed harvesting, may result. However, as these equipment are fitted with emission control devices, their impact should be negligible.

9. Will the project adversely affect short-term or long-term noise levels?

Yes. A temporary increase in noise associated with construction or maintenance activity related with the restoration of the lake may result. All construction and maintenance vehicles will be equipped with the proper noise pollution control devices, thus an increase 1n noise related to the restoration of the lake should be minimal. There will be some increase in noise associated with weed harvesting.

10. If the project involves the use of 1n-lake chemical treatment, will it cause any short-term or long-term effects?

ENV/L22-rpt4 11-50 The use of chemicals to control weed growth are not advocated at this time 1n Onota Lake. Thus, the potential for toxicity problems associated with the use of herbicides will be avoided under this plan.

11. Will the project be located in a floodplain?

Yes, in part. All activities which occur along stream corridors or in the near shore, littoral areas of the lake will be considered in the i floodplain. • 12. Will structures be constructed in the floodplain? H It is conceivable that some of the passive stormwater treatment structures (i.e. detention basins, catch basins, sediment traps etc.) • will be constructed in floodplains. However, the construction of such ™ structures will be in compliance with existing laws and regulations of mm the Commonwealth of Massachusetts.

^ 13. If the project involves physically modifying the lake shore, its bed, H or its watershed, will the project cause any short or long-term ™ adverse- effects?

• Dredging will be limited to only a few areas of .the lake. Recolonization of these areas by benthic organisms should not be a H problem due to the limited nature of the dredging operation. Destruction of aquatic macrophytes will occur; however, this is • desirable and will enhance the quality of the lake. The dredging and macrophyte removal will actually improve fish habitat.

™ 14. Will the project have a significant adverse impact on fish and wildlife, wetlands or other wildlife habitat?

• Yes/No. Activities proposed in the plan will improve fish habitat and • will actually benefit fish and waterfowl use. Specifically, this will ™ be achieved in that the plan calls for reestablishing and protecting fishery habitat, reducing weed densities, minimizing siltation, influx

ENV/L22-rpt4 11-51 of pollutants, and improving water quality. The full impact of lake level drawdown on adjacent wetlands must be further researched to fully determine potential impact upon wildlife, habitat and wetland resource via the E/R.

15. Have all feasible alternatives to the project been considered in terms of environmental impacts, resource commitment, public interest and cost?

Yes. All feasible alternatives have been thoroughly examined and analyzed. The proposed plan is the most cost effective restoration program for Onota Lake. It is also characterized by a minimal potential negative environmental impact. The efficiency of the proposed plan will be greatly diminished if only part of the plan is implemented. In order to preserve the integrated nature of this plan it is important that both the watershed and in-lake restoration measures be implemented in full.

16. Are there other measures not previously discussed which are necessary to mitigate adverse effects from the project?

Offsite disposal of the dredge spoils through the use of trucks will increase vehicular traffic and may cause some damage to roads. It wi11 also be necessary to conduct comprehensive testing of the sediments to determine its suitability for upland disposal. The dewatering of the spoils, when done hydraulically, will create a supernatant. This water must be of a quality equal to that of the lake or receiving stream. These potential impacts associated with dredging can be mitigated using environmentally sound engineering practices and adhering to soil erosion and water quality regulations.

ENV/L22-rpt4 11-52 I I I I I I

I I • m 11-53 EHV/L22-rpt4 12.0 LITERATURE CITED American Public Health Association. 1976. Standard Methods for the Examination of Water and Wastewater, 14th ed., A.P.H.A., Washington DC. 1193 pp. Baker, L. A., P. L. Brezonik and C. R. Kratzer. 1985. Nutrient loading models for Florida lakes. In Lake and Reservoir Management - Practical Applications, 253-258. Proceedings of the 4th Annual Conference of the North American Lake Management Society, McAfee, New Jersey. Berkshire County Regional Planning Commission. 1978. Water Quality Management Plan for the Upper Housatonic River, Final Plan/Environmental Impact Statement. BCRPC, Pittsfield, Massachusetts, 385 pp. Carignan, R. 1985. Nutrient dynamics in a littoral sediment colonized by the submergec macrophyte Myrlophilum spicatum. Can J Fisheries and Aquatic Sci. 42(7). 1303-1311. Cooke, G.D., 1980. Lake level drawdown as a macrophyte control technique. Water Res. Bull. 16(2): 317-322. Cooke, G.D. and R.H. Kennedy. 1980. Phosphorus inactivation: A Summary of Knowledge and Research Needs. In, Restoration of Lakes and Inland Waters EPA 440/5-81-010, 395-404. Cooke, G.D., R.T. Heath, R.H. Kennedy, and M.R. McConas. 1982. Change in lake trophic state and internal phosphorus release after aluminum sulfate application. Water Res. Bull. 18(4):699-705. Cooke, G.O., E.B. Welch, S.A. Peterson and P.R. Newroth. 1986. Lake and Reservoir Restoration. Butterworth's Publishing, Boston, Mass. Dillon, P.J. 1974. A critical review of Vollenweider's nutrient budget model and other related models. Water Res. Bull. 10:969-705. Dunne, T. and L.B. Leopold. 1978. Water in Environmental Planning. W.H. Freeman and Co., San Francisco, CA., 68p. Fogg, G.E. 1965. Algal Cultures and Phytoplankton Ecology, The University of Wisconsin Press, Madison Wisonsin, 126p. Freedman, P.L. and R.P. Canale. 1977. Nutrient release from anaerobic sediments. J. Env. Engineering Div., ASCE 103 (EE2):223-244. Geldrelch, E. E. and B. A. Remner, 1969. Concentrations of fecal streptococci in stream pollution. J. Water Pollution Control Federation, 41:336-352. Hely, A.G., T.J. Nordenson, et al. 1961. Precipitation, Water Loss, and Runoff in the Delaware River Basin and New Jersey. U.S. Geological Survey, Hydrologic Investigations Areas HA-11. IT Corporation, 1989. Technical and Environmental Evaluation of lake level Control for Management 1n Pontoosuc Lake, Berkshire County, Massachusetts. Berkshire County Commissioners, Pittsfield, MA.

ENV/L22-ref 12-1 I I Kerfoot, W. 1979. Septic system leachate surveys for rural lake communities: A winter survey of Otter Tail Lake, Minnesota. N.S.F. I Sixth National Conference, Proceedings, 435-469. Kerfoot, W. 1980. Operation Manual for ENDECO Type 2100 Septic Leachate I Detector System. Environmental Devices Corp., Marion, Md. Kirchner, W. B. and P. J. Dillon. 1975. An empirical model of estimating the I retention of phosphorus. Water Res Research. 2(1):182-183. Kortmann, R,W., D.D. Henry, A. Kuether, and S. Kaufman. 1982. Epilmnetic nutrient loading by metalimnetic erosion and resultant algal responses in I Lake Waramug, . Hydrobiologia 91-2:501-510. Lee, G.F., W. Rast, and R.A, Jones. 1976. Eutrophication of water bodes; • Insights for an age-old problem. Env Sci Tech 12(8): 900-908. Ligman, K., N. Hutzler, and W.C. Boyle. 1974. Household wastewater « characterization. J. Environ. Eng. Division, ASCE 100 (EE1): 201-213. Merritt, R.W. and K.W. Cummins. 1978. An Introduction to the Aquatic Insects •of North America. Kendall/Hunt Publishing Co., Dubuque, Iowa. 441 p. • Means. 1987. Means Building Construction Cost Data. 45th Annual Edition, 466 pp. | Nichols, S. A. 1974. Mechanical and Habitat Manipulation for Aquatic Plant Management, a Review of Techniques. Tech. Bull. No. 77, Dept. Natural _ Resources, Madison, Wisconsin. " Nurnberg, G.K. 1984. The prediction of internal phosphorus load in lakes with anoxic hypolimnia. Limnol. Oceanogr. 29(1):111-124, • Ostrofsky, M.L. 1978. Modification of phosphorus retention models for use with lakes and low areal water loading. J. Fish Res. Board Can. M 35(12):1532-1536. Otis, R.J. 1979. Treatment of domestic waste in lakeshore developments. In, _ Lake Restoration. EPA 440/5 79-001, 95-102.

™ Owe, M.s P, J. Craul, and H. G. Halverson. 1982. Contaminant levels in precipitation and urban surface runoff. Water Res. Bull. 18(5):863-868. H Pennak, R.W. 1978. Freshwater Invertebrates of the United States, 2nd ed. John Wiley & Sons, Inc. Hew York. 803 p. Pielou, E.G. 1966. The measurement of diversity in different types of biological collections. J. Theoret. B1ol. 13:131-144. Posten, S.'1982. Estimation of mean ground water runoff and safe yield using hydrograph analysis in selected New Jersey hard rock aquifers. 1n, Pollution and Water Resources: Columbia University Seminar Series, Volume XV, Pergamon Press, Elmsford, N.Y.

EMV/L22-ref 12-2 Princeton Aqua Science. 1982. New Jersey Lakes Management Program, Lakes Classification Study. Reckhow, K. H. 1979. Quantitative Techniques for the Assessment of Lake Quality. EPA Report No. EPA 4405-79-015. p. 146. Sartor, J.D. and G. B. Boyd. 1972. Water Pollution Aspects of Street Surface Contaminants. Prepared for, Office of Research and Monitoring, USEPA.

Scheider, W. A., J. J. Moss4 and P. J. Dillon. 1979. Measurement and uses of hydraulic and nutrient budgets. In, Lake Restoration, 77-84. E{A 44-5- 79-001. Soil Conservation Service. 1983. General Soils Report, Berkshire County, Massachusetts, Southern Part. Prepared by U.S. Dept. of Agriculture, in cooperation with Berkshire-Franklin Resource Conservation and Development Project. Souza, S. J. and J. D. Koppen. 1984. The role of internal phosphorus loading on the trophic status of New Jersey's largest lakes. Lake and Reservoir Management, 111-128. EPA-440/5/84-001. Souza, S.J.' and P.A. Perry. 1977. Land and Recreational Development at New Jersey Reservoirs. Water Resources Research Institute technical publication A-043-N.J., Rutgers University, N.J. U.S.E.P.A. 1976. National Eutrophication Survey Methods 1973-1976. U.S. Environmental Protection Agency, National Eutrophication Survey. Working Paper No. 175. U.S.E.P.A. 1980. Clean Lakes Program Guidance Manual, U.S. Environmental Protection Agency EPA-440/5-81-003. Vollenweider, R. A. 1975. Input-output models with Special -Reference to the phosphorus loading concept in limnology. Schweiz. Z. Hydrol. 37(i):53-84. Walker, W. W. 1977. Some analytical methods applied to lake water quality problems. Ph.D. dissertation, Harvard University, 1977. Welch, E.B. and C.A. Rock. 1980. Lake Sammamish response to waste water diversion and increasing urban runoff. Water Res. 14:821-828. Wetzel, R. G. 1983. Limnology, 2nd ed., W. B. Saunders Co., Publishing, New York, N.Y. 767 pp. Wiederholm, T. 1980. Use of benthos in lake monitoring. JWPCF 52(3):537- 547. Wilber^ W. G. and J. V. Hunter. 1975. Contributions of metals resulting from stormwater runoff and precipitation in Lodi, New Jersey. Urbanization and Water Quality Control Conf., American Water Resources Assoc. Proc. No. 20: 45-54. Winter, T.C. 1978. Groundwater Component of 1 ake Water and Nutrient Budgets. Verk. Internat. Terein. Limnol. 20:438-444.

ENV/L22-ref 12-3 APPENDIX Al WATER CHEMISTRY DATA COLLECTED AT 0.5 METERS AT STATION il

Alkalinity N-NH3 N-N03 TKN TP cr TDS TSS Chlorophyll (mg/nr)

Date (mgCaCOn/11^ ) mg/1 a b c Pheophyton

3/4/86 77 0.02 0.13 0.58 0.02 5.0 70 11 1.2 0.17 ND 3/22/86 40 0.12 0.28 0.23 0.01 4.0 15 <1 0.39 0.19 0.19 4/8/86 72 0.57 1.9 0.76 O.01 8.0 120 2 0.93 0.14 0.07 4/25/86 76 0,13 0.10 0.35 0.02 4.4 82 1 3.3 0..03 0.47 5/6/86 76 0.07 0.18 0.33 <0.01 8.0 93 1 -- — 5/21/86 74 0.09 <0.02 0.44 0.07 7.0 99 2 1.9 0.34 0.8—5 6/2/86 70 0.07 <0.02 0.32 0.02 6.0 58 4 1.3 0.99 2.7 6.24/86 68 <0.08 0.32 <0.08 0.01 4.5 97 1 2.8 0.13 0.60 7/8/86 67 <0.05 0.03 0.08 0.02 6.0 100 3 2.1 0.40 1.1 7/26/86 67 <0.08 0.03 0.23 <0.01 6.0 89 <1 1.52 1.13 0.514 8/14/86 64 <0.05 0.07 0.22 O.01 5.5 84 2 2.0 0.10 ND 8/28/86 66 <0.05 0.08 <0.05< 0.01 5.0 100 6 1.5 0.28 ND 9/10/86 69 <0.08 0.04 0.32 <0.01 7.5 230 2 2.7 0.59 ND 9/23/86 73 <0.05 <0.02 0.25 0.01 6.5 86 2 3.6 NO ND 10/16/86 74 <0.08 0.02 0.08 <0.01 6.0 120 <1 3.2 ND ND 11/14/86 78 0.07 0.07 0.30 0.41 10.0 55 2 1.9 0.12 ND 1/21/87 80 <0.05 0.29 0.23 <0.01 5.5 83 2.4 3.7 ND 3.2 2/19/87 84 <0.05 0.08 0.39 0.05 8.9 68 4 1.1 0.14

N 18 18 18 18 18 18 18 18 17 17 17 X 70.8 0.098 0.204 0.57 0.04 6.3 91.6 2.69 2.07 0.28 0.75 SD+/- 9.4 0.121 0.435 1.11 0.094 1.63 42.4 2.47 0.99 0.33 1.13

ENV/L22-appA/l APPENDIX A2 WATER CHEMISTRY DATA COLLECTED AT MID DEPTH (9.0M) AT STATION #1

Alkalinity N-NH3 N-N03 TKN TP cr TDS TSS Chlorophyll (mg/nr) Date (mgCaOVl) mg/1 a b c Pheophyton "-

3/4/86 ------3/22/86 _ 4/8/86 ------4/25/86 64 0.17 0.11 0.38 0.01 5.2 7.8 1.0 3.0 0.18 0.72 5/6/86 76 0.05 0.14 0.29 0.01 8.0 83 2.0 2.0 0.15 0.36 5/21/86 73 0.13 <0.02 0.65 0.04 7.0 20 4.0 6/2/86 70 0.10 <0.02 0.40 0,03 6.0 38 7.0 - - 6/24/86 70 <0.08 0.32 0.12 0.01 5.0 96

N 12 12 12 12 12 12 12 12 2 2 2 X 71 0.08 0.078 0.273 0.033 6.27 78.7 3.0 2.5 0.165 0.54 SD+/- 3.74 0.049 0.094 0.176 0.057 0.967 38.0 1.75 0.707 0.021 0.25

ENV/L22-appA/2 APPENDIX A3 WATER CHEMISTRY DATA COLLECTED AT 0.5M OFF THE BOTTOM (19.0 M) AT STATION II

Alkalinity N-NH3 N-N03 TKN TP cr IDS TSS Chlorophyll (mg/nr) mg/1 Date (mgCaCO^/1w ) a b c Pheophyton

3/4/86 88 0.02 0.21 0.50 0.01 4 100 _ <1.0 0.79 0.18 ND 3/22/86 ------4/8/86 91 0.12 1.7 0.43 <0.01 11 100 3,0 4/25/86 82 0.15 0.13 0.49 0.01 5.7 76 6.3 - - 5/6/86 76 0.12 0.10 0.39 0.03 8.0 100 3.0 - - 5/21/86 74 0.15 0.04 0.51 0.04 8.0 no 6.0 - - 6/2/86 72 0.13 <0.02 . 0.46 0.02 6.0 100 2.0 - - 6/24/86 77 0.09 0.34 0.09 0.02 4.7 99 6.0 7/8/86 71 <0.05 <0.02 <0.05 0.04 5.5 100 4.0 7/26/86 74 <0.10 0.04 0.20 0.02 6.0 110 5.0 8/14/86 76 0.47 0.10 0.65 0.06 6.0 99 8.0 8/28/86 80 0.40 . 0.07 0.40 0.08 5.5 120 15.0 - - - 9/10/86 87 0.26 0.19 0.70 0.10 8.0 113 11.0 9/23/86 86 0.34 <0.02 0.57 0.09 7.5 104 21 10/16/86 88 0.70 0.11 1.0 0.05 . 6.0 120 12 11/14/86 76 0.06 0.04 0.26 0.03 11.0 64 <1 1/21/87 108 <0.05 0.10 0.23 <0.01 8 130 4.8 - - 2/19/87 88 0.06 0.20 0.16 0.01 9 69 5.0 - -

N 17 17 17 17 17 17 17 17 NA NA NA X 82 0.192 0.202 0.417 0.037 7.05 100.8 6.71 SO+/- 9.32 0.185 0.396 0.242 0.029 2.01 17.5 5.33 -

ENV/L22-appA/3 APPENDIX A4 WATER CHEMISTRY DATA COLLECTED AT 0.5 METERS BELOW SURFACE AT STATION il

Alkalinity N-KH'3 N-N03 TKN TP cr TDS TSS Chlorophyl 1 (mg/m ) Date (mgCaOh/1) mg/1 a b c Pheophyton v« ' 3/4/86 _ — _ . 3/22/86 86 0.21 0.28 0.40 <0.02 5.0 94 <1 - - - 4/8/86 62 0.15 0.21 0.37 O.01 4.0 84 2.0 1.7 ND ND 4/25/86 73 0.16 0.11 0.49 ,. <0.01 2.2 39 4.0 3.0 0.46 0.81 5/6/86 73 0.07 0.17 0.38 0.02 7.0 86 3.0 2.1 ND ND 5/21/86 70 0.09 0.05 0.47 0.04 9.0 110 3.0 2.6 0.16 0.63 6/2/86 70 <0.05 <0.02 0.32 0.14 9.5 56 5.0 3.1 3.4 8.8 6/24/86 58 <0.08 0.33 <0.08 0.01 4.7 92 1.0 3.3 0.51 1.1 7/8/86 58 <0.05 O.02 0.13 0.02 5.5 180 <1 1.7 1.0 1.9 7/26/86 60 0.18 <0.02 0.47 0.01 6.0 79 <1 0.932 1.01 5.63 8/14/86 62 O.05 0.05 0.24 O.01 6.0 88 2 2.1 0.2 ND 8/28/86 64 <0.05 0.05 , <0.05 0.02 5.5 110 2 3.0 0.14 0.71 9/10/86 150 <0.05 0.04 0.46 0.02 6.5 96 8 - - - 9/23/86 71 <0.08 0.02 0.24 0.01 6.2 84 4 2.8 0.75 0.15 10/16/86 74 0.13 0.02 0.46 <0.01 6.0 110 2 3.3 0.82 0.45 11/14/86 70 <0.05 0.09 0.19 0.02 12 69 <2 1.5 0.26 0.16 1/21/87 80 <0.05 0.58 0.13 0.01 4 no 1.2 3.8 0.14 2.0 2/19/87 94 <0.05 0.42 0.23 0.03 6.5 81 1.0 2.3 0.6 1.6

N 17 17 17 17 17 17 17 17 15 15 15 X 75.3 0.09 0.145 0.30 0.024 6.2 92.2 2.5 2.5 0.69 1.60 SD+/- 11.2 0.054 0.167 0.15 0.031 2.3 30.0 1.9 0.80 0.84 2.50

ENV/L22-appA/4 APPENDIX A5 WATER CHEMISTRY DATA COLLECTED AT 4.0 METERS (MIDDEPTH) AT STATION i2

Alkalinity N-NH3 N-N03 TKN TP CI IDS TSS Chlorophyll (mg/nr)

Date (mg'OaCO,/!•j ) mg/1 a b c Pheophyton

3/4/86 72.0 0.21 1.1 0.65 <0.02 9.0 80 4 0.93 0.21 ND 3/22/86 - - - - - . 4/8/86 - - - - - ______, 4/25/86 - - - - - 5/6/86 70.0_ 0.09 0.05 0.53 0.04 9.0 75 <1 5/21/86 - - - - _ 6/2/86 63 0.07 <0.02 0.41 0.02 5.5 39 9 6/24/86 59 <0.08 0.29 <0.08 <0.01 5.0 110 3 7/8/86 57 <0.08 <0.02 <0.08 0.02 5.0 79 <1 7/26/86 59 <0.13 0.03 0.14 <0.01 6.0 97 <1 8/14/86 63 <0.08 0.04 0.29 0.02 5.5 97 5 8/28/86 65 <0.08 0.06 <0.08 <0.01 6.0 99 2 9/10/86 - . - - - ______9/23/86 - - - - - _ 10/16/86 - - - - - _____ 11/14/86 - - - - -_ _ 1/21/87 ------2/19/87 - —

N 8 8 8 8 8 8 8.8 1 1 X 63.5 0.103 0.20 0.28 0.018 6.4 84.5 3.25 NA NA NA SD+/- 5.34 0.047 0.37 0.23 0.010 1.67 21.9 2.76 NA NA NA

ENV/L22-appA/5 APPENDIX A6 WATER CHEMISTRY DATA COLLECTED 0.5 M OFF THE BOTTOM (7.0 M) AT STATION

Alkalinity N-NH3 N-N03 TKN TP cr TDS TSS Chlorophyll (mg/m )

Date (mgCaCOi/1*j ) mg/i a b c Pheophyton i 3/4/86 . _ 3/22/86 • ------_ 4/8/86 110 0.10 0,27 0.43 0.01 4 140 5 - - 4/25/86 72 0.08 0.12 0.46 0.02 5.5 52 2 5/6/86 70 0.08 0.23 0.35 0.02 7.0 85 1 5/21/86 73 0.11 0.04 0.45 0.06 11 34 7 6/2/86 70 0.07 <0.02 0.32 0.02 6.0 58 4 1.3 0.99 2.7 6/24/86 71 0.19 0.30 0.19 0.04 2.0 110 5 7/8/86 74 0.13 <0.02 0.31 0.63 6.0 63 13 7/26/86 75 0.20 <0.02 0.44 0.02 6.0 99 22 8/14/86 64 <0.05 0.05 0.22 0.02 4.5 105 3 - - 8/28/86 84 O.05 0.09 0.08' 0.09 6.0 110 26 9/10/86 NA <0.05 0.07 0.23 0.01 6.0 110 5 9/23/86 74 <0.08 0.05 0.24 0.01 6.5 82 3 - - 10/16/86 76 0.13 0.07 0.25 0.03 6.0 120 9 11/14/86 71 O.05 0.13 0.20 0.02 12.0 86 <2_ 1/21/87 90 <0.05 0.51 0.11 0.01 7.0 2.8 2/19/87 111 <0.05 0.29 0.22 0.01 10 53 5.0

N 15 16 16 16 16 16 16 16 NA NA NA X 79 0.092 0.143 0.28 0.064 6.6 89.8 5.9 SD+/- 14.2 0.049 0.140 0.12 0.153 2.5 31 6.2

ENV/L22-appA/6 I APPENDIX Bl I ONOTA LAKE TRIBUTARY #6 - BLYTHEWOOD DRIVE

Date of Alkalinity N-NH-, N-N0 P-TP TKN TSS Total Fecal O 3 cr TDS Sample {mgCaC03/l) Col iform Col iform I •* 02/19/87 260 .05 .31 .02 20 .29 240 3 N/A <10 09/24/86 280 <.05 .13 .02 23 .50 317 7 1600 30 07/26/86 76 <.25 .10 .02 32 <.25 336 2 320 112 07/08/86 280 <.05 .24 .02 1.5 <.05 88 4 920 30 05/21/86 270 .66 .11 .02 22 <.08 310 9 90 <2 06/24/86 266 <.08 .34 .02 16 <.08 320 7 300 <2 I 06/02/86 270 <.05 .06 .05 24 .43 300 10 100 2 08/14/86 260 .06 .19 .01 14 .35 320 1 1200 44 09/10/86 320 .11 .10 22 .47 370 1 200 1 08/28/86 300 .16 .12 7= 250,6 7= 0.118 7= 0.27 7= O.Q21 7 = 21.7 7 = 0.40 7 = 305 7 = 6.GI 7 = 440 7= 17.5 I SD = 55 SD = 0.141 SD = 0.21 SD = 0.016 SD = 12.3 SD = 0.23 SD = 92.3 SD = 6.7 SD = 503 SD = 28 I 1ENV/L22-appB/4 APPENDIX Bl ONOTA LAKE TRIBUTARY #5 - PARKER BROOK

Date of Alkal inity N-NH3 N-N03 P-TP cr TKN TDS TSS Total Fecal Sample Col iform Col iform

02/19/87 66 <.05 0.34 .05 3.5 .08 63 2 N/A N/A 09/23/86 95 <.05 0.45 .05 7 0.25 125 3 610 610 07/26/86 N/A .12 .20 - .14 N/A .34 N/A N/A 220 84 07/08/86 59 <.05 0.26 .01 22 <.05 350 4 850 190 05/21/86 62 .10 .18 .03 2 .40 76 9 TNTC <2 06/02/86 56 <.05 .14 .03 2.5 .12 36 5 60 <1 06/24/86 46 <.08 .35 <.01 2.5 <.08 72 <1 1300 <2 08/14/86 73 <.05 .43 .01 3 .12 110 3 920 18 08/28/86 79 <.D5 .39 .01 2.5 .10 110 4 TNTC TNTC 09/10/86 74 <.05 0.30 ' .02 . 2.5 .13 65 2 2400 160 11/14/86 62 <.05 0.37 <.01 6.5 .10 55 <2 60 <10 03/22/86 40 .22 .33 .01 4 .28 15 5 N/A N/A 03/04/86 56 .05 1.1 .01 3 .18 60 7 N/A N/A 04/08/86 130 .08 .51 <.01 7 1.5 120 4 14 <2 05/06/86 55 .03 .12 .02 4 .22 70 5 180 4 04/25/86 58 .06 .32 <.01 7 .28 79 3 24 <1 11/14/86 62 <.05 .37 <.01 6.5 .10 55 <2 60 <10 10/16/86 60 .14 .35 .01 2 .21 160 <1 1000 150 01/21/87 68 <.05 .59 <.01 2.5 .17 68 i.6 500 170 TR8" fi^IT WW • TFT5" .. WTZ N=19 N=l8 ""1=18 . N=14 N=15 ' '•• > ; 1 (*-''•„•. 1 . • v -.^.' ••- ^i ~x. X 7 = 66.7 7 = 0.073 7 = 0.37 7= 0.024 7 = 5.0 7 = 0.25 7= 93.8 7= 3.8 7 = 585 7= 94.2 SD = 20.0 SD = 0.045 SD = 0.21 SD = 0.030 SD = 4.6 SD = 0.32 SD = 72.7 S'J = 2.1 SD = 675 SD = 160.0

ENV/L22-appB/3 APPENDIX Bl ONOTA LAKE TRIBUTARY *2 - CHURCHILL BROOK

Date of A 1 k a 1 i n i ty N-NH3 N-N03 P-TP cr TKN TDS TSS . Total . Fecal Sample (mgCaC03/l) Coliform Coliform

02/19/87 60 <.05 0.51 0.01 3.5 0.42 84 12 N/A <10 09/23/86 N/A <.05 0.71 0.05 3.0 0.13 92 11 12600 1760 07/26/86 58 .25 .57 .01 1.5 .25 73 1 360 88 07/08/86 59 <.05 .27 .01 3 <-05 72 <1 330 44 05/21/86 44 .08 .36 .02 4 .34 25 11 16 <2 06/02/86 46 .06 .22 .008 1.5 .29 18 30 16 <1 06/24/86 29 .11 .33 <.01 1.5 .11 68 <1 180 <2 08/14/86 68 .05 .73 .02 4 .05 86 2 TNTC <1 08/28/86 77 .10 1.0 <.01 4 .10 95 1 TNTC 292 09/10/86 80 .16 .57 <.01 2.5 .27 83 1 2000 <1 11/14/86 36 <.05 .27 <.01 10 .11 49 <2 260 <10 03/22/86 26 .17 .27 .01 4 .28 19 11 N/A N/A 03/04/86 50 .05 .77 .01 6 .46 50 6 N/A N/A 04/08/86 31 .03 .43 .01 1 .30 62 3 20 2 05/06/86 44 .01 ,38 .03 7 .14 54 1 8 <2 04/25/86 76 .13 .10 .02 4.4 .35 82 1 No Growth <1 11/14/86 36 <.05 .27 <.01 10 .11 49 <2 60 <10 10/16/86 41 <.08 .33 <-01 <1 .14 47 2 600 10 01/21/86 57 <.05 .56 .03' 2 .19 35 <1 400 <10 N=18 N=19 N=19 N=19 N=19 N=19 N=19 N=19 N-13 N=17

7 = 51 7 = 0.083 7 = 0.45 7 = 0,015 7 = 3.9 7 = 0.21 7 = 60.1 7= 5.3 7 = 1296 7= 132 SD = 16.8 SD = 0.059 SD = 0.23 SD = 0.011 SD = 2.7 SD = 0.12 SD = 24.4 SD = 7.3 SD = 3437 SD = 425

ENV/L22-appB/2 APPENDIX Bl ONOTA LAKE TRIBUTARY #1 - DANIELS BROOK

Date of Alkalinity N-NH3 N-N03 P-TP cr TKN TDS TSS Total Fecal Sample (mgCaC03/l) Col iform Col iform

01/21/87 75 <0.05 0.85 <0.01 4 0.13 90 7.6 100 <10 10/16/86 60 '• <0.08 0.51 <0.01 3 0.26 83 1 500 40 11/14/86 40 <0.05 0.29 . 0.03 10 0.13 36 <2 260 80 04/25/86 60 0.02 0.71 <0.01 2.7 0.17 76 4 24 <1 05/06/86 70 0.06 0.60 <0.01 7 0.23 69 2 88 <4 04/08/86 42 0.07 0.77 <0.01 2 0.30 69 3 2 2 03/04/86 65 0.09 1.0 0.01 19 0.30 80 4 N/A N/A 03/22/86 28 0.07 0.36 . 0.04 4 0.44 23 15 N/A N/A 12/17/86 56 0.04 0.69 <0.01 4 0.22 97 7 80 60 09/10/86 120 <0.05 1.32 0.01 6.5 0.06 150 1 1100 40 08/28/86 86 0.07 1.10 0.01 4.5 0.13 100 .3 TNTC 716 08/14/86 72 <0.05 0.68 <0.01 5 0.08 88 ? 340 8 06/24/86 42 <0.08 0.29 <0.01 5.2 <0.08 81 1 150 2 06/02/86 82 <0.08 . 0.88 0.02 6.0 0.25 55 5 180 <1 05/21/86 74 0.09 <0.02 0.07 7.0 0.44 99 2 <4 <2 07/08/86 75 0.05 0.33 0.01 5.5 0.05 110 3 620 170 07/26/86 75 0.12 0.67 <0.01 5 0.12 126 <1 360 72 09/23/86 N/A 0.07 1.15 0.07 6.5 0.41 141 37 TNTC ; 930 02/19/87 86 <0.05 0.96 0.03 9.5 0.08' 120 <1 N/A <10 N=18 N=19 N=19 1 N=19 N=19 N=18 N=19 N=A TFTT N=17 1 , i i,k , 1' ' •-, :'! . '' ' 7* 67.1 7 = 0.065 7=0.69 7= 0.021 7= 6.12 7= 0.211> 7 = 89.1 7- 5.9 7= 272 7= 126.4 SD * 21.3 SD = 0.022 SD=0.33 SD =0.019 SD = 3.75 SD = 0.127 SD = 32.4 SD - 8.5 SD = 303.9 SD = 268.5

ENV/L22-appB/l APPENDIX Bl ONOTA LAKE LAKE OUTFALL

Date of Alkal nity N-NH3X N-N03 P-TP cr TKN IDS TSS Total Fecal Sample (mgCa(:03/l) Col iform Col iform

01/21/87 8*. <.05 .22 <.01 5.5 .496 150 <1 20 <10 10/16/86 7f <.08 <.02 .01 6 .29 160 1 700 <10 11/14/86 9< <-05 0.09 <.01 8.5 .19 85 6 60 <10 04/25/86 7< .29 .09 .01 6 .68 75 5 N/A N/A 05/06/86 7r .05 .21 .02 4 .36 95 1 20 4 04/08/86 6' .07 .27 <.01 5 .41 67 4 6 <2 03/04/86 7 .09 .34 .02 8 .29 100 1 N/A N/A 11/14/86 9~ <.05 .09 <.01 8.5 .19 85 6 60 <10 03/22/86 5 .10 .24 .01 6 .27 62 10 N/A N/A 08/28/86 6 .08 .04 .01 6 .29 96 1 TNTC 12 09/10/86 6£ <.15 .06 .01 7 .33 92 3 400 <1 08/14/86 61 <.05 .10 .02 4 .23 74 2 80 8 06/03/86 52 .06 <.02 .09 5.3 .58 42 7 16 <1 06/24/86 59 <.08 .33 .01 4.5 <.08 110 6 30 <2 05/21/86 ?n .15 .02 .04 6 .45 70 7 4 <2 07/08/86 62 <.Q5 0.25 .02 6 <.05 93 3 TNTC 24 07/26/86 N/" <.25 .05 .03 6 .44 75 23 20 8 09/23/86 7-r .05 .05 .01 7.5 .29 85 2 1,100 50 02/19/86 91 <-05 .33 .01 9.5 .09 105 3 <10 <10 FT5' N=19 N=19 N=19 N=19 N=19 "FIT N=19 N=14 N=16

7 = 71.9 7 = 0.09 7 = 0.148 7 = 0.018 7 = 6.3 7 = 0.32 7 = 90.5 x~= 4.8 7= 180 7= 10.25 SD = 13.3 SD = 0.069 SD = 0.117 SD - 0.019 SD = 1.5 SD = 0.16 SD = 28 SD = 5.1 SD = 330 SD = 12.1

ENV/L22-appB/5 APPENDIX C TRIBUTARY *6 - BLYTHEWOOD DRIVE STORM EVENT MARCH 31, 1987

CoHform N- N- N- P- cr TDS TSS PH Sample # Fecal Total NH3 N03 TKN TP

#1 200 200 .08 .97 .81 .11 17 150 27 7.3 2 50 <50 .09 .41 .86 .08 14 120 20 7.5 3 300 <50 .13 .26 .88 .05 18 150 22 7.5 4 50 50 .08 .41 .77 .06 15 140 23 7.5 5 100 <50 .08 .22 .87 .09 15 130 41 7.4 6 100 50 .07 .66 .95 .10 14 170 40 7.4 7 100 50 .09 1.06 .93 .09 14 180 21 7.5 8 50 <50 .09 1.0 1.0 .11 14 200 27 7.4 9 <50 <50 <.05 1.0 .83 .096 13 210 35 7.4 10 <50 100 <.05 .68 .73 .09 15 200 24 7.5 N=10 PTo Mo N=10 N=10 N=10 N=10 N=10 N=10 N=10 PTO

x _ .081 .67 .86 .087 14.9 165 28 7.4 SD = .022 .33 .08 .019 1.5 31.7 7.4 0.069

ENV/L22-appC/l APPENDIX C TRIBUTARY il & 2 - CHURCHILL & DANIELS BROOK STORM EVENT MARCH 31, 1987

Coliform N- N- N- P- CI" TDS TSS PH Sample # Fecal Total NH3 N03 TKN TP

#1 50 100 0.1 .70 .50 .08

ENV/L22-appC/2 APPENDIX C TRIBUTARY #5 - PARKER BROOK STORM EVENT MARCH 31, 1987

Conform M- , H- N- P- CI TOS TSS pH Sample # Fecal Total NH3 N03 TKN TP

#1 <50 50 <.05 .33 .49 .05 6.7 220 23 6.7 2 50 100 <.05 .39 .46 .05 7.0 64 23 7.0 3 50 150 <.05 .34 .24 .04 4.0 76 18 6.9 4 <50 150 <.05 .36 .42 .08 15 37 15 6.7 5 100 100 <.05 .32 .44 .06 6 49 19 6.6 6 <50 100 <.05 .35 .36 .07 10 54 20 7.0 7 50 <50 <.05 .48 .41 .06 3 13 26 6.7 8 50 <50 <.06 .33 .33 .08 5.7 37 31 6.8 9 50 100 .05 , .34 .49 .099 3.2 35 30 6.4 10 50 50 <.06 .31 .45 .09 3 33 120 6.2 JFI5 FR3 N^TO IFTO PTO N^TO N^TO PTO N^TO NFTO PTO x _ .052 .355 .41 .068 6.4 61.8 32.5 6.7 SO = .0042 .049 .078 .02 3.8 58.3 31.2 .25

ENV/L22-appC/3 APPENDIX C TRIBUTARY DATA STORM EVENT MARCH 31, 1987

Cd Cr i Cu Fe Pb Mn Zn Tributary (mg/L) (mg/L) (mg/L) (mg/L) (mg/L) (mg/L) (mg/L)

Church111 & Daniels Brook <0.005 <0.01 0.055 1.0 O.005 0.099 0.028

Parker Brook <0.005 <0.01 0.04 2.8 0.019 0.22 0.047

Blythewood Drive O.005 <0.01 0.032 0.69 0.008 0.052 0.026

ENV/L22-appC/4 APPENDIX C TRIBUTARY DATA STORM EVENT SEPTEMBER 23, 1986

Cd Cr • Cu Fe Pb Mn Zn Tributary (mg/L) (mg/L) (mg/L) (mg/L) (mg/L) (mg/L) (mg/L)

Daniels Brook <0.0007 <0.01 <0.002 0.32 <0.005 0.15 0.007 Church 111 Brook <0.0007 <0.01 <0.002 0.14 <0.007 0.024 0.007 Parker Brook <0.0007 <0'.01 O.002 0.28 <0.005 0.048 0.007 Blythewood Drive <0.0007 <0.01 0.004 0.26 <0.005 0.13 0.01 Pecks Road <0.0007 <0.01 0.007 0.8 0.01 0.1 0.016

ENV/L22-appC/5 APPENDIX C TRIBUTARY DATA STORM EVENT SEPTEMBER 23, 1986

Col i form N- N- N- P- cr TDS TSS PH Sample # Fecal Total NH3 N03 TKN TP

Daniels Brook 930 TNTC 0.07 1.15 0.41 0.07 6.5 141 37 7.6 Churchill Brook 1760 12600 <0.05 0.71 0.13 0.05 3.0 92 11 7.7 Parker Brook 2110 9100 0.05 0.40 0.19 0.05 4.5 96 11 7.8 Blythewood Drive 520 5800 0.09 0.26 1.1 0.11 26 309 48 8.1 Pecks Road 610 * 0.14 0.71 0.67 0.18 3.5 42 27 7.4

TNTC - Too numerous to count * - Confluent with Total Coliform Colony Growth

ENV/L22-appC/6 .

USDA SOIL CONSERVATION SERVICE FIGURE 3.1 SOIL ASSOCIATION MAP (1980) SOIL ASSOCIATION MAP

4200

s •cat*

Do Not Sc§»« Tnis Drawing'