Water Quality Monitoring in the Source Water Areas for New York City: In

Water Quality Monitoring in the Source Water Areas for New York City: An Integrative Watershed Approach

A Report on Year 5 (2004) Monitoring Activities

Submitted by: Stroud Water Research Center Contribution No. 2005007 970 Spencer Road Avondale, PA 19311

27 October 2005 NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Report Prepared by:

David B. Arscott, Ph.D. Anthony K. Aufdenkampe, Ph.D. Thomas L. Bott, Ph.D. Charles L. Dow, Ph.D. John K. Jackson, Ph.D. Louis A. Kaplan, Ph.D. J. Denis Newbold, Ph.D. Bernard W. Sweeney, Ph.D.

- II - NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Table of Contents Chapter 1 - Introduction...... - 1 - Literature cited...... - 3 -

Chapter 2 - Technical Design ...... - 5 - Overview...... - 5 - Phase II Study Site Descriptions...... - 7 - Literature Cited...... - 13 -

Chapter 3 - Nutrients, Major Ions and Suspended Particles in transport...... - 31 - Research Task...... - 31 - Methods ...... - 31 - Results and Discussion ...... - 34 - Literature Cited...... - 39 -

Chapter 4 – Molecular Tracers ...... - 53 - Introduction...... - 53 - Research Task...... - 54 - Methods ...... - 54 - Results ...... - 56 - Discussion...... - 59 - Appendix Discussion: QA/QC...... - 61 - Literature Cited...... - 63 -

Chapter 5 - Macroinvertebrate Community...... - 91 - Introduction...... - 91 - Methods ...... - 92 - Results and Discussion ...... - 97 - Literature Cited...... - 105 -

Chapter 6 - DOC and BDOC Dynamics...... - 123 - Research Task...... - 123 - Methods ...... - 123 - Results ...... - 124 - Discussion...... - 126 - Literature Cited...... - 127 -

Chapter 7 -Nitrogen (N), Phosphorus (P), and Dissolved Organic Carbon (DOC) Spiraling - 133 - Introduction...... - 133 - Methods ...... - 134 - Results and Discussion ...... - 137 - Literature Cited...... - 141 -

Chapter 8 - Stream Metabolism ...... - 151 - Purpose and Significance...... - 151 - Methods ...... - 151 - QAQC Results ...... - 155 - Results, Discussion and Conclusions...... - 156 - Literature Cited...... - 160 -

Chapter 9 - Reservoir Primary Productivity ...... - 171 - Purpose and Significance...... - 171 - Methods ...... - 171 - QA/QC Results ...... - 173 - Results and Discussion ...... - 175 - Literature Cited...... - 178 -

- III - NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

APPENDICES to Year 5 New York Watersheds Report ...... - 189 -

Appendix to Chapter 3 - Nutrients, Major Ions and Suspended Particles in Transport...... - 191 -

Appendix to Chapter 4 – Molecular Tracers...... - 199 -

Appendix to Chapter 5 - Macroinvertebrate Community...... - 215 -

Appendix to Chapter 6 - DOC and BDOC Dynamics ...... - 217 -

Appendix to Chapter 7 – Nitrogen (N), Phosphorus (P), and Dissolved Organic Carbon (DOC) Spiraling...... - 219 -

Appendix to Chapter 8 - Stream Metabolism ...... - 225 -

Appendix to Chapter 9 - Reservoir Primary Productivity ...... - 241 -

- IV - NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Table of common abbreviations used in this report.

Abbreviation Description 1MP 1-methyl phenanthrene 2MP 2-methyl phenanthrene aCOP Cholestanol AFDM Ash Free Dry Mass AHTN Tonalide ALKL Alkalinity ANOVA Analysis of Variance ANT Anthracene aONE Cholestanone BAA Benzo(a)anthracene BAP Benzo(a)pyrene BBF Benzo(b)fluoranthrene bCOP coprostanol BDOC Biodegradable Dissolved Organic Carbon BKF Benzo(k)fluoranthene BMP Best Management Practice BOD Biological Oxygen Demand BOM Benthic Organic Matter Ca Calcium CAF Caffeine CCS Continuing Calibration Standards CHOL Cholesterol CHR Chrysene Cl Chloride cm centimeter CO2 Carbon Dioxide CoIA Co-Inertia Analysis COND Specific Conductance CR24 Community Respiration CV Coefficient of Variation DO Dissolved Oxygen DOC Dissolved Organic Carbon DON Dissolved Organic Nitrogen eCOP 24-ethyl-coprostanol EOH East of Hudson Watersheds (a.k.a., Croton-Kensico System) EPI Epicoprastanol EPT Ephemeroptera, Plecoptera, and Trichoptera FB Field Blank FD Field Duplicate FLR Fluoranthene FLU Fluorene FM Fragrance Materials FS Fecal Steroids GF/F Glass Fiber Filters GIS Geographic Information System GPP Gross Primary Productivity GPS Global Positioning System h hour H/LPAH High-to-Low Molecular Ration for PAHs HBI Hilsenhoff Biotic Index HDPE High Density Polyethylene HEV Hydraulic Uptake Velocity HHCB Galaxolide HMW High Molecular Weight ID inner diameter ISCO Automated field water sampling device K Potassium L Liter

- V - NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Abbreviation Description LB Lab Blank LCS Laboratory Control Standard LMW Low Molecular Weight MDL Method Detection Limit Mg Magnesium min minute mL milliliter MLR Multiple Linear Regression MS Matrix Spike Na Sodium NDM Net Daily Metabolism NH4-N Ammonium Nitrogen NO3-N Nitrate Nitrogen NUTW Whole water (unfiltered) nutrient sample NY DEC New York State Department of Environmental Conservation NYC New York City NYCDEP New York City Department of Environmental Protection NYS DOH New York State Department of Health NYSDEC New York State Department of Environmental Conservation OD Optical Density OM Organic Matter OTIS One Dimensional Transport In Streams PAH Polycyclic Aromatic Hydrocarbons PAR Photosynthetically Active Radiation PCA Principal Components Analysis PD Percent Difference PHE Phenanthrene PMA Percent Model Affinity PYR Pyrene QA/QC Quality Assurance Quality Control QAPP Quality Assurance Project Plan RPD Relative Percent Difference SAS Statistical Software Package in use at Stroud SD Standard Deviation sec second SKN Soluble Kjeldahl Nitrogen SNOL 24-ethylcholestanol SO4 Sulfate SOP Standard Operating Procedure SPDES State Pollutant Discharge Elimination System SRM Surface Reaeration Model SRP Soluble Reactive Phosphorus SWRC Stroud Water Research Center TDP Total Dissolved Phosphorus TKN Total Kjeldahl Nitrogen TM Thematic Mapper TP Total Phosphorus TSS Total Suspended Solids US EPA United States Environmental Protection Agency USGS United States Geological Survey vf Uptake Velocity VSS Volatile Suspended Solids WOH West of Hudson Watersheds (a.k.a., Catskill-Delaware) WQS Water Quality Score (Macroinvertebrate based) WWTP Waste Water Treatment Plant

- VI - NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Chapter 1 - Introduction

The drinking water industry in the United States and abroad now recognizes that protecting the sources of fresh water is a critical component of any long-term plan for a drinking water system. With this recognition has come a new understanding of the central role that watersheds – and their aquatic ecosystems – play in the filtration/treatment process that is necessary to provide clean, safe drinking water to the public in the most cost-effective way. In the recent past, providing safe drinking water has meant chemical treatment or filtration for finished water and better regulation of point source discharges to improve surface water quality. Today, it is also recognized that, in addition to other strategies, protection of source water areas and management of human activities in source water areas are paramount for providing high quality water for consumption. Forested watersheds provide higher quality stream water than non-forested (Dudley and Stolton 2003) and having high quality water can greatly reduce treatment costs, even fitration costs (Ernst 2005). Source water protection requires managing these water supply watersheds and ecosystems. Consequently, a successful management plan for New York City’s drinking water must be based on a solid understanding of the streams and the watersheds they drain in order to make source watershed protection a reality.

Watersheds and their ecosystems have three critical functions: (i) they are the ultimate sources of water; (ii) they are major sources of naturally occurring and anthropogenic constituents (physical, chemical, and biological) in water; and (iii) they are primary natural processors of water-borne constituents. Because past, present, and future land-use activities in source water areas affect each of these functions, successful source water protection requires an "Integrated Watershed Approach" to assess sources, impacts, and processes relevant to the streams and reservoirs of the source area.

A monitoring program for drinking water source areas should focus primarily on constituents of natural and anthropogenic origin (hereafter contaminants) that can, at certain concentrations, contaminate water and render it unsuitable for human consumption and/or unable to support wildlife. An integrated watershed approach to contaminant dynamics in the NYC source area needs to recognize four basic elements: Source, Transport, Ecosystem Impairment, and Symptom. The existing monitoring programs, like most other source water programs, include strong elements of Transport (levels of contaminants in the source water and distribution system, consisting of streams, rivers, reservoirs, and distribution pipes) and Symptom (turbidity, oxygen deficits, taste and odor, disinfection byproduct formation potential, etc.). These elements are driven by local, state, and federal regulations and by operational needs (understanding ambient quality of water for treatment purposes). This program, which focuses on elements of Ecosystem Impairment and Source, is intended to enhance on-going efforts by introducing both new study variables and a different scale (spatial and temporal) for certain study variables.

Each contaminant in the NYC system can have multiple sources that can be diffuse (non-point source) or concentrated at a point. Identifying the historical and current sources of the principal contaminants in the various watersheds and sub-watersheds is critical to developing long-term plans for current remediation and future protection and development. This requires an intensive and coordinated spatial and temporal sampling program as well as sophisticated analytical

- 1 - CHAPTER 1 - INTRODUCTION NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005 techniques that can distinguish among the various possible sources of contaminants within each of the NYC source watersheds.

Each contaminant is capable of causing some impairment to streams, rivers, and reservoirs of the NYC water supply system. This impairment can cause a change in the structural and/or functional properties of the ecosystem which renders it unable to effectively or efficiently utilize, process, metabolize, or otherwise sequester materials, including contaminants entering from the watershed.

Careful assessment of contaminant sources in watersheds supplying NYC drinking water and of key structural and functional properties of streams, rivers, and reservoirs in the NYC distribution system will provide: (i) a basis for measuring spatial variation in the source of contaminants and their impacts on ecosystem functioning and biological communities; (ii) a basis for measuring temporal/spatial change in both the source of contaminants and their impacts on stream, river, and reservoir functionality; and (iii) a stronger scientific basis for the overall management plan for the NYC source water area.

The principal objectives of this monitoring program have been:

1. To provide dependent variables for statistical analyses relating aquatic ecosystem structure and function to land use, best management practice (BMP) implementation, and other watershed inputs or factors.

2. To provide chemical and biological indicators for evaluating the occurrence and source of selected aquatic contaminants.

3. To provide a baseline data set of population, community and ecosystem-level variables and molecular indicators of contaminants to assess changes in water quality and aquatic ecosystem structure and function in response to on-going and/or future shifts in land use/cover. For example, (i) quantitative measures of stream ecosystem structure/function can be used in a before-after analytical framework; or (ii) measurements made across sites help define the true range of conditions throughout these watersheds. This range can then be compared to future changes to understand improvements or degradations at specific points in a watershed.

Data from Source and Ecosystem Impairment monitoring will put into perspective: (i) the magnitude and complexity of contaminant/source issues throughout the NYC source water area and (ii) the current status of ecosystem health within the NYC source water system (i.e., where the ability to process excess nutrients in watersheds is in good to excellent condition and where that ability has been compromised). In addition, these data will provide a baseline for measuring success of on-going remediation efforts in NYC source water areas and will be helpful in designing/implementing future remediation or conservation efforts (e.g., BMP, stream restoration, zoning) as part of an overall NYC management plan.

This monitoring program was designed to complement existing programs of the New York State Department of Environmental Conservation (NYS DEC), New York City Department of Environmental Protection (NYC DEP), United States Environmental Protection Agency (US

- 2 - CHAPTER 1 - INTRODUCTION NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

EPA), and the New York State Department of Health (NYS DOH), as well as programs under the direction of -- and/or in cooperation with -- the various counties in the study area. Several of the principal study elements in this program were not monitored by any of the above groups at the outset of this endeavor. While one or more groups are monitoring some elements (or parts of an element) they are doing so with lower spatial intensity and, in some cases, less accuracy or precision. Although the Stroud Center's program was designed to have some overlap of study site locations with NYS DEC and NYC DEP programs, to allow data generated from each program to supplement and add perspective to one another, this program is an independent effort designed to enhance overall monitoring in these source areas.

This report covers year 5 of this 6 year study (a.k.a., Phase II, Year 2) and is intended to be an interim report documenting year 5 activities. A technical overview of study design and study site descriptions (Chapter 2) is followed by separate chapters on the specific tasks of this program: Nutrients, Major Ions, and Particulates in Transport (Chapter 3); Molecular Tracers Analysis (Chapter 4); Macroinvertebrate Community Structure and Function (Chapter 5); Dissolved Organic Carbon (DOC) and Biodegradable Dissolved Organic Carbon (BDOC) Dynamics (Chapter 6); Nitrogen (N), Phosphorus (P), and DOC Spiraling (Chapter 7); Net Stream Metabolism (Chapter 8); and Reservoir Productivity (Chapter 9).

Each chapter contains an overview of field and laboratory methods along with a results and discussion of data from year five research/monitoring activities. A summary of quality assurance/quality control (QA/QC) efforts for the fith year of monitoring is also included in the Appendix. Data discussion focuses on year five and describes how data within each task characterized individual study sites, subwatersheds, and the two regions (East and West of the Hudson River) that comprise the NYC source water areas. Integration across monitoring tasks and comparison to Phase I results, to the extent possible, are also discussed. However, thorough synthesis across phases and across study tasks is not part of this project status report, as this type of analysis will be conducted upon completion of the 6-year project.

Literature cited

Dudley, N., and S. Stolton. 2003. Running Pure: The importance of forest protected areas to drinking water. World Bank/WWF Alliance for Forest Conservation and Sustainable Use.

Ernst, C. 2004. Protecting the source. Land conservation and the furute of America's drinking water. Trust for Public Land and America Water Works Association.

- 3 - CHAPTER 1 - INTRODUCTION NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

------Intentionally Blank------

- 4 - CHAPTER 1 - INTRODUCTION NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Chapter 2 - Technical Design

Overview

This project was designed as a six-year study divided into two discrete three-year phases; Phase I from 2000 to 2002 and Phase II from 2003-2005. New York City drinking water source areas are located in primary locations referred to as East of Hudson (EOH; a.k.a., Croton/Kensico System) watersheds and West of Hudson (WOH; a.k.a., Catskill/Delaware System) watersheds (Figs. 2.1 and 2.2). Reports on Phase I activities (3-yr summary) and Phase II year 4 activities can be accessed on-line at http://www.stroudcenter.org/research/nyproject/

During Phase I, annual studies of various project elements at 60 stream and 8 reservoir stations were performed. Each element was studied at a specified portion of study sites. These activities were replicated in Phase II but occurred at new locations throughout the source watersheds. The scientific strength of this program is a result of the kinds and number of elements measured, spatial scope of the study (108 stream and 14 reservoir stations visited over 6 years) and its replication over time.

In Phase I, 60 stream (30 EOH, 30 WOH) and eight reservoir (2 EOH, 6 WOH) sampling stations were established (Stroud Report 2004) to provide complete spatial coverage of source watersheds. For Phase II, 48 new (differing from Phase I) stream stations were established, and monitoring at 12 of the Phase I stations has continued (Tables 2.1 through 2.4; 27 EOH, 33 WOH stations). Continued monitoring at the 12 Phase-I stations during Phase II tie these phases together for a continuous temporal perspective (i.e., 6 years of measures at 12 sites). Reservoir monitoring stations in Phase II occured in seven (7) reservoirs (Figs. 2.5 and 2.6). Four stations were located in new reservoirs, two within reservoirs studied in Phase I but with new substation locations, and one within a Phase I reservoir at the same Phase I substations (Figs. 2.3 and 2.4, Table 2.5 and 2.6). All of theses stations will be subject to annual monitoring reported herein for three years (through 2005). Additionally, two stream monitoring sites in the WOH region were added in 2004. These sites (159 and 160) were located at headwater sites in the Esopus Creek and East Branch Delaware River watersheds.

Selection of Phase II sites was based on a combination of (i) “areas of concern” revealed during Phase I, (ii) desire to broaden spatial coverage within source watersheds, and (iii) the need to measure, quantify, and determine more sources and effects of contaminants in watersheds and the present condition and ability of existing ecosystems to process both natural and unnatural (contaminants) watershed inputs. In general, Phase II stream sites were located on other important tributaries to primary reservoirs in the system or further upstream in the watershed from Phase I stations.

Phase I stream sampling stations were distributed among the major sub-basins of the principal source watersheds (designated as 50 "targeted" and 10 "integrative" sampling stations depending on the project elements being measured at each station). Criteria for selecting Phase I study sites were as follows: (1) land use (forested — upland and riparian; agriculture — row crop, dairy, and beef; suburban — septic and sewage treatment, road runoff, fertilization, pesticides; urban — waste water treatment plants, urban runoff); (2) Gauged stream flow (USGS records); (3) NYC DEP / NY DEC / EPA study or demonstration sites; (4) NYC DEP / NY DEC / EPA /

- 5 - CHAPTER 2 – TECHNICAL DESIGN NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

County background data; (5) BMP’s in progress, BMP implementation, or BMP pending implementation; (6) Feasibility in studying various elements of our monitoring program.

Selection of Phase II study sites was a combination of the above but relied more heavily on providing information that would supplement Phase I results. For example, Phase I integrative sites in EOH watersheds lacked a “least-impaired” site, while integrative sites in the WOH watersheds lacked an impacted site. Integrative stations occurred sufficiently downstream in a watershed to integrate effects of land use and other factors on a given project element or task under study over a large portion of the watershed. Further, at “Integrative” stations monitoring activities included detailed study of nutrient spiraling and stream metabolism (see Chapters 7 and 8, respectively). In some instances ‘downstream’ distance was constrained by feasibility of one or more of the study elements. For Phase II, sites were selected to potentially “fill” gaps along measured “impact gradients”. For “targeted” stations, site selection included new sites in major tributaries not sampled in Phase I. "Targeted" stations occurred on streams of varying size and monitoring activities were limited to measurement of nutrients, major ions, and organic particles in transport, molecular tracers, dissolved organic carbon dynamics, and macroinvertebrate communities (see Chapters 3-6). Also, in a few instances new “targeted” sites were selected to be upstream of potential negative impacts to water quality in “degraded” Phase I sites to help identify “areas of concern”. Finally, several “targeted” sites were selected to broaden the spatial extent of this program (i.e., moving even further upstream).

Overall project design was intended to: (i) expand understanding of sources of principal contaminants in "source water" watersheds of NYC; (ii) provide new information about present structure and function of the aquatic ecosystems comprising the "system"; and (iii) use that information as (a) a measure of anthropogenic stress, (b) an estimate of "functional capacity" of these ecosystems to absorb, sequester, or otherwise process natural inputs and contaminants, and (c) a baseline to determine future improvement and/or deterioration in watershed conditions. This work plan was designed to have elements complement and build on one another. Some project elements (e.g., grab samples of water chemistry) were instantaneous with regard to condition over time. Some elements (e.g., macroinvertebrates) contain information about water quality/habitat condition over time. Some (e.g., N, P, DOC spiraling, and stream metabolism) were integrative in a spatial sense. This programs strength lies in its breadth of study elements and its high degree of integration (same sites/timing/personnel).

This project is a spatially intense, broad synoptic survey repeated annually, rather than a highly targeted survey with limited spatial scope and high repetition within a site. All major watersheds throughout the study area were subjected to this monitoring regime, rather than one or two watersheds representing a small portion of the study area. This broad synoptic approach avoids two serious problems associated with a spatially limited, temporally intense approach: (1) pseudoreplication - where multiple samples taken from a given stream throughout the year still only represent one stream and one watershed; and (2) serial autocorrelation - where repeated measures of the same variable during the year tend to be correlated with one another or are non- independent (e.g., baseline chemistry, macroinvertebrates). This applies to both baseflow and stormflow sampling. For example, the project focuses on between-stream variability rather than between-storm variability for a given stream.

- 6 - CHAPTER 2 – TECHNICAL DESIGN NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Phase II Study Site Descriptions

Study sites were separated into two groups: 52 "targeted" and 10 "integrative" sampling stations (see above for definitions; Figs. 2.1 and 2.2). Several of the specific task components involved all 62 sites, while a few tasks were conducted only at integrative sites. Tweny eight (Tables 2.1 and 2.2) of the 62 Phase II stream sites were visited during winter baseflow to collect water samples for molecular tracer analyses (see Chapter 4). Autosamplers were deployed at 3 (Tables 2.1 and 2.2) of the 62 Phase II stream sites to collect water samples for stormflow chemistry (see Chapter 3). Seven reservoirs were also studied for certain project elements (Figs. 2.3 and 2.4). Stream and reservoir stations were located using a Trimble GPS Pathfinder ™ ProXR receiver unit (Tables 2.3, 2.4, and 2.6). Figs. 2.1 and 2.2 illustrate Phase II stream stations and Figs. 2.3 and 2.4 illustrate Phase II reservoir stations and substations (Phase I sites were also included).

All geographic data were manipulated using ArcMapTM 9.0 (Copyright 1999-2004 ESRI Inc.). For each stream sampling site, basin boundaries were derived from an existing NYC DEP coverage of subbasin delineations for the region. For most sites, NYC DEP subbasin delineations did not exactly match our study sites. Therefore, where necessary, boundaries were modified by hand digitizing using USGS 1:24,000 topographic maps (20-ft contours). Basin boundaries were used to quantify and summarize land use/cover, population density, road density, and discharge facilities permitted by the State Pollution Discharge Elimination System (SPDES) (Table 2.1, 2.2, and 2.5).

Bedrock and surficial geology of the Catskill/Delaware and Croton/Kensico basins was summarized from the 1:250,000 Bedrock Geology or Surficial Geology Maps of New York State (NYS Geological Survey Map & Chart Series Number 15 and Number 40). Map interpretation was guided by a publication on NY State geology (Isachsen et al. 2000) and the NY State Museum web site and metadata for these GIS data layers (http://www.nysm.nysed.gov/). Geographic data for bedrock and surficial geology were intersected with basin delineations of our study sites and percent of each geologic formation within each basin was summarized for descriptive purposes. These data are only presented here in a descriptive format. The final 6-year report will include a tabular summary of these data.

Soil data layers were provided by NYC DEP Bureau of Water Supply and were originally prepared by the Soil Information Systems Laboratory at Cornell University from air photos (1:24k, 1:20k, and 1:15,840). Soil series delineations were related to the SSURGO database (Soil Survey Geographic database) which is useful for quantifying organic matter content, clay content, pH, and soil erodibility (Kf), among other factors. Geographic soil data were intersected with basin delineations of our study sites and percentage of each soil series with each basin was summarized for descriptive purposes. Organic matter content, clay content, erodibility, and soil pH were summarized following procedures outlined in the STATSGO manual (NRCS 1994). These data are only presented here in a descriptive format. The final 6-year report will include a tabular summary of these data.

Rasterized land use/land cover data were obtained from the New York City Department of Environmental Protection (DEP). The land use/cover data were derived from 2001 Landsat Enhanced Thematic Mapper Plus (ETM+) satellite imagery (5 April, 8 June, 10 July, and 12 September 2001). An Anderson (Anderson et al. 1976) level 4 based classification scheme was

- 7 - CHAPTER 2 – TECHNICAL DESIGN NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

developed by DEP to classify the ETM+ images after overlays with the National Wetland Inventory polygon data (mid 1980’s), New York State Office of Real Property Services Tax Parcel data, and other ancillary data from the USDA-Farm Security Agency and Watershed Agriculture programs, Central Business District locations, farmstead locations, and some forest disturbance and strip mine locations. Resolution of this composite land use/land cover data layer was 10-meter cell size. A complete report on this classification process is available from DEP (Adair et al. 2004).

Prior to summarizing land use/cover data at the best available land use/cover classification resolution possible, a re-classification routine had to be run for those cells that were only classified at the Level-1 category of ‘urban’ (see Anderson et al. 1976 for classification). The re- classification scheme determined the majority value of all Level-2 categorized cells surrounding the cell requiring re-classification based on a given ‘neighborhood’ about that cell. The Level-2 category determined as the majority value in the assigned neighborhood was then assigned to the cell requiring re-classification. This scheme was run iteratively, with the neighborhood of cells increasing by one cell-width each run, until less than 5% of all cells requiring re-classification in the input raster remained. There were cells that were only classified at the Level-1 categories of ‘agriculture’ and ‘brushland’, however, these cells were either too few within a watershed, or too isolated from other similarly-categorized cells, to allow the re-classification scheme to work. Only 13 of the 60 study sites had greater than 1% of their watershed area classified only to the Level-1 ‘urban’, ‘agriculture’, or ‘brushland’ categories and of those only 2 had greater than 5% of their total area in any of these Level-1 categories. Following this re-classification scheme, land use/cover at the Level-2 Anderson classification was summarized, as percentages for basin delineations.

After regressing road density (see below) against % transporation for WOH sites, it was apparent that the land use/cover layer had errors associated with the Pepacton (East Branch Delware) watershed. Upon further review it was noted that even the land use/cover data layer prior to our manipulations did not adequately represent roads in Pepacton watersheds compared to all other watersheds (i.e., regressions coefficients for road density versus transportation for sites within other subcatchments were significant and high compared to within the Pepacton). Therefore, the transporation land cover category for WOH sites was eliminated from any subsequent analyses.

Population density data were compiled from total population counts from the 2000 census using census blocks within each county (Tables 2.1, 2.2, and 2.5). Census blocks are the smallest unit for which census data are available (Census 2000 Geographic terms and concepts; http://www.census.gov/geo/www/census2k.html). Census 2000 Geographic Census data, including population counts, were retrieved as Census 2000 TIGER/Line data through the Environmental Systems Research Institute, Inc. (ESRI) web page at http://www.esri.com/data/download/census2000_tigerline/index.html. Watershed boundaries were used to determine what proportion of each census block fell within a given study site watershed. The fraction of the census block area falling within a given watershed was multiplied by the total population count for that census block. This product of fractional census block area and corresponding population count was summed for all census blocks falling within a watershed and then divided by the watershed area to arrive at a watershed population density estimate.

- 8 - CHAPTER 2 – TECHNICAL DESIGN NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

A GIS point coverage of SPDES discharge points with 2002 permits was supplied by DEP. The coverage was used to determine the number of active (i.e., discharging) plants located upstream of each stream and reservoir station (Tables 2.1, 2.2, and 2.5). The average annual discharge (2000-2002) was calculated for each point and the total flow within a sub-basin was divided by basin area to normalize flows to watershed area (m3/m2).

Road density was quantified from data layers provided by the Mapping and Modelling Unit of the NYC-DEP Bureau of Water Supply. East of Hudson roads were digitized from 1996 NY- DOT Planimetric Images and categorized as Primary (Interstate, Route, or Parkway) roads, Secondary (all others) roads, and Total roads (sum of primary and secondary). West of Hudson roads were digitized in 1993 from USGS DLGs. Road categorization was different between EOH and WOH data sets but only total roads (sum of Primary and Secondary in EOH sites and sum of road classes 1-4 in WOH sties) was used for analyses. Road data layers were intersected with basin delineations for our 110 sites and road length in each category was summed and divided by basin area to derive road densities (m/km2).

Geology and Soils

East and WOH regions are underlain by distinctly different bedrock geology. Geology of the entire WOH region has roots in the Late Devonian Period (ca. 375 million yrs. b.p.) and all of the formations in the region are sedimentary (quartz-dominated shale, sandstone, siltstone, and conglomerates). Although there are different formations from north-to-south and east-to-west (Oneonta-to-Lower Walton-to-Upper Walton Formations) within the WOH region, rock composition remains relatively similar throughout the region.

Bedrock geology in the EOH region is a complex mosaic of formations associated with two distinct geologic regions (Isachsen et al. 2000), the Hudson Highlands (Middle Proterozoic; ~1,100 million yrs. b.p.) and the Manhattan Prong (~ 500 million yrs. b.p.). The Hudson Highlands, composed of layered and unlayered metamorphic units highly resistant to erosion, crosses the northwestern portion of the EOH region. These rock layers can contain large deposits of magnetite, a kind of iron ore mined in the 18th and 19th centuries (Isachsen et al. 2000). Biotite, mica, quartz, and feldspar gneiss are other common constituents in the Hudson Highlands region. The Manhattan Prong dominates the southern portion of the EOH region and also covers the northeastern tip of the region. The metamorphic rocks of Manhattan Prong include Fordham, Yonkers, Pundridge, and Bedford Gneiss, and Inwood (to the south) and Stockbridge (to the north) Marble. A pocket of limestone, dolostone, and siltstone occurs near the northern tip of the EOH region.

Surficial geology in both regions is the result of past glaciation by the Hudson-Champlain Lobe of the Laurentide Ice Sheet. In the Catskills, glacial history is further complicated by the occurrences of both the Laurentide Ice Sheet and local mountain glaciers (Isachsen et al. 2000). Surficial geology in the EOH region is predominately glacial till (unsorted mixture of clay, sand, gravel, and boulders deposited directly by a glacier) riddled with kame deposits throughout (steep-sided mounds of sand and gravel deposited by meltwater from a glacier) and bedrock outcrops and swamp deposits more prevalent towards the north. The WOH region can be separated to 2 distinct regions based on surficial geology: (1) the southeast (Neversink, Esopus, and upper Schoharie basins) is primarily bedrock outcrop mountain tops with till, kame, and

- 9 - CHAPTER 2 – TECHNICAL DESIGN NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

outwash sand and gravel deposits in narrow valleys; and (2) the northern and western areas have more till and deeper soils on the ridges and side slopes and valleys with recent alluvium, outwash sand and gravel, and kame deposits.

Coupled with surficial geology and bedrock, soils help define distinct regions within the EOH and WOH source water areas. At a course scale, soils are primarily Udept Inceptisols (suborder/order) that are moderate to highly acidic. Inceptisols usually occur on relatively active landscapes, e.g., mountain slopes and river valleys, where erosional processes actively expose and deposit relatively unweathered material (Olson et al. 1999). Udepts (a.k.a., Ochrepts) are freely drained Inceptisols and often have only thin, light-colored surface horizons. Udepts extend from southern New York through central and western Pennsylvania, West Virginia, and eastern Ohio. Some Udepts in southern New York and northern Pennsylvania are not naturally productive due to low organic matter content and have thus been associated with silviculture and pasture/grazing activities following earlier periods of crop production (Brady and Weil 1999). Aquepts (fluvially deposited wet Inceptisols) and Fluvents (fluvially deposited wet Entisols [younger and less developed than Inceptisols), organically rich soils but too wet without artificial drainage for crop production, are also present in E and WOH valleys. Saprists, wet Histosols of well decomposed plant material usually associated with wetlands, are limited in extent in the WOH region but occur throughout the EOH region and account for up to ~5% of soil surface area in some sub-basins.

In the EOH, several soil series helped to define 4 groups along a north-south gradient. Important to the northern portion of the EOH are the Hollis-Chatfield and Chatfield-Hollis complexes which are well-to-excessively well drained, have depth to bedrock of 10 to 20 inches, are strongly acidic, and have low erodability (Kf). Important to the southern portion of the EOH are the Paxton, Sutton, and Woodbridge series which are well drained, have depth to bedrock commonly more than 6 feet, are moderately acidic, and have higher erosion potential (i.e., Kf ~ 0.3). Several sub-basins of the Kensico reservoir have a high proportion of disturbed soils (Udorthents) which are typically the result of earth-moving activities.

Soil series in the WOH region clearly distinguished 4 groups: the East and West Branch of the Delaware River basin; the Neversink River basin; the Esopus Creek basin; and the Schoharie Creek basin. Rondout Creek basin has characteristics similar to the Neversink and Esopus basins. Within the WOH region the E. and W. Br. of the Delaware River basins have the richest soils with the highest percent organic material, lowest acidity, and greatest depth to bedrock. The Lewbeach, Onteora, and Willowemoc series, dominant in the E. and W. Br. Delaware River basins (30-40%), all have depth to bedrock at or greater than 5 ft and pH ~6 - 6.5. However, these soils have higher erodability (Kf) than other dominant soils in the WOH region. The Halcot-Mongaup-Vly complex is more common in the other 3 basin groupings (30-50%). These soils occupy ridge tops and are shallow due to outcropping of bedrock. Of the basins with higher percent area of the Halcot complex, the Schoharie basin has appreciable coverage of Lewbeach and Willowemoc series and the Esopus/Rondout basins have Oquaga-Lordstown-Arnot, Wellsboro-Wurtsboro complexs, and Elka series that are associated with fairly steep slopes and are very stoney. There are two Aquept series (wet Inceptisols; hydric soils) that are more common in some valley areas of the E. and W. Br. Delaware River (Lackawanna-Bath complex) and the Schoharie River basin (Onteora-Volusia complex).

- 10 - CHAPTER 2 – TECHNICAL DESIGN NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Phase II versus Phase I landscape comparison

Correlation PCA (Fig. 2.5) of 19 watershed-specific, GIS derived variables, and 110 sites (48 Phase I, 50 Phase II, and 12 Phase I and II sites) illustrated the landscape differences among study sites and similarities between Phase I and II site characteristics. Land cover variables included 15 land use/cover variables (e.g., % residential, % commercial, % cropland), watershed area (WtsdArea), population density (ind./m2; 2000 census), road density (m/km2), and annual average SPDES permitted discharge per unit area (cm3/cm2). The first two principal components accounted for 50.6 % of the variance in the data-matrix (factor 1 = 32.9 %, F2 = 17.7 %). The first factor (Fig. 2.5a) was primarily defined by population density (+) (13.8 % contribution to F1 definition), road density (+) (14.0 %), residential (+) (11.3%), wetland (+) (9.5%), and mixed forest (-) (10.8%). The second factor was defined by farmstead area (+) (20.4%), grass cover (+) (17.8%), cropland (+) (15.9%), and mixed brush/grass cover (+) (15.2%). Watershed area, % commercial, and % industrial contributed most to the 3rd axis (not shown; 16.6, 16.9, and 13.2%, respectively).

East of Hudson sites (circles and triangles) clustered separately from WOH sites (squares and diamonds) and revealed distinct land use/cover gradients in these two regions. East of Hudson sites ranged from ~85% forest (Table 2.2) with ~4% urban/suburban cover, no permitted dischargers, low population and road density (relative to other EOH watersheds; site 125) to sites with low forest cover (<40%), >25% urban/suburban cover, several SPDES discharges, and high population and road density (e.g., site 58). Agriculture played a minor role in EOH site character (except at site 32). West of Hudson sites were typically located in larger watersheds compared to EOH sites (Table 2.1 and 2.2). West of Hudson sites ranged from > 95% forested with very low population density and zero SPDES permitted dischargers to sites dominated by agricultural lands (Table 2.2). Urban/suburban (not including transporation cover), wetland area, population and road density, and the number of SPDES permitted dischargers in WOH watersheds were all considerably lower than EOH watersheds.

One objective of this analysis was to understand land use/cover character of Phase II study sites relative to Phase I study sites. The resulting PCA site map coded for Phase I, II, and sites sampled in both phases (Fig. 2.5c) illustrated that (a) Phase II sites were similar in land use/cover composition to Phase I study sites and (b) objectives in site selection were met with regard to landscape variables. For example, a primary objective for Phase II site selection in WOH watersheds was to increase the number of stream sites (particularly “integrative” sites) that had potential for greater anthropogenic impact and sites in EOH watersheds should be selected to help round out the “least” impacted sites. By comparing Figs. 2.5b and c, it is evident that several Phase II Integrative sites in the WOH occurred towards the “agricultural” end and at least one EOH sited (34) occurred towards the “forest” end of the landscape gradient. Additionally, several Phase II EOH sites helped to better define EOH forested dominated streams. Finally, by selecting Phase II sites further upstream in the WOH end points in the agricultural-to-forest gradient were extended (i.e., Phase II sites with greater forest and agricultural land cover than Phase I sites).

- 11 - CHAPTER 2 – TECHNICAL DESIGN NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Stormflow

Stormflow sampling occurred at three of the 60 baseflow monitoring sites: W. Br. Delaware River at Hawleys (6); Neversink River near Claryville (29); and the Kisco River near Stanwood (55). A USGS gaging station is co-located with the monitoring site on the Kisco River (USGS ID 01374987), and a USGS station is located approximately 1.5 km downstream of the site on the Neversink River (USGS ID 01435000). Since there were no significant tributaries entering the Neversink between the monitoring site and the gaging station, this USGS gaging station was considered to be representative of discharge at site 29. For the third monitoring site, the W. Br. Delaware River at Hawleys, a USGS gaging station was located several miles downstream in Walton (USGS ID 01423000). Storm discharge was estimated for this site using the USGS gaging station located in Walton and watershed area relationships (see text below).

Due to equipment failures (e.g., dead batteries; trigger mechanism dislodged by debris), attempts to collect two samples during at least one storm at each monitoring site in the fall of 2003 yielded only one successfully sampled event (Fig. 2.6a). However, efforts during the 2004 field season yielded successful sampling of two events at each of the three storm sampling sites (Figs. 2.6-2.8). Instantaneous discharge data were considered provisional data by the USGS and were not subjected to final review or approval by the USGS. High turbidity and high flow samples for analysis of major ions and nutrients (Chapter 3), particles in transport (Chapter 3), dissolved organic carbon (Chapter 3), and molecular tracers (Chapter 4) were collected from each storm event observed to compare to baseflow chemistry.

Discharge at our WBD site for all storms was estimated using the measured discharge at the USGS Walton gage multiplied by the ratio of the watershed area at the monitoring site to the watershed area for the USGS gauging station at Walton. A MiniTroll stage recorder (pressure transducer), which records relative stage height, was located at the monitoring site on the W.Br. Delaware and was used to adjust the timing of hydrograph for the peak stage at the monitoring site. The offset in peak flow time was also applied to all estimates of discharge during automated sampling of storms. Because instantaneous flow data were collected at 15-minute intervals at each of the gaging stations, discharge taken at the interval closest to the actual sampling time of each sample was used.

West Branch Delaware storms.- High turbidity and high flow samples were collected from 3 storms (Fig. 2.6) occurring between 23 September 2003 and 26 April 2004 at the W. Br. Delaware River (site 6). The 23 Sept. 2003 storm event had the greatest magnitude, resulting in peak discharge of 2119 ft3 s-1 (~60 m3 s-1). This storm amounted to a 10-fold increase in flow within 12 hours. The two samples during stormflow represented high turbidity at a discharge of 2016 ft3 s-1 (~57.1 m3 s-1) and high flow at 2119 ft3 s-1 (60 m3 s-1). High turbidity and flow samples from the 13-14 April 2004 storm occurred at discharges of 756 ft3 s-1 (21.4 m3 s-1) and 886 ft3 s-1 (25.1 m3 s-1), respectively. The 26 April 2004 storm was intermediate between the two other storms and high turbidity and flow samples occurred at discharges of 1181 ft3 s-1 (33.5 m3 s-1) and 1169 ft3 s-1 (33.1 m3 s-1), respectively.

Neversink storms.- High turbidity and high flow samples were collected from 2 storms (Fig. 2.7) occurring on 13-14 April 2004 and 27 May 2004 at the Neversink River (site 29). The 27 May storm event had the greatest magnitude, resulting in peak discharge of 1119 ft3 s-1 (31.7 m3 s-1).

- 12 - CHAPTER 2 – TECHNICAL DESIGN NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

This storm amounted to a 7-fold increase in flow within 17 hours. The two samples during stormflow represented high turbidity at a discharge of 1119 ft3 s-1 (31.7 m3 s-1) and high flow at 1109 ft3 s-1 (31.4 m3 s-1). High turbidity and flow samples from the 13-14 April 2004 storm occurred at discharges of 693 ft3 s-1 (19.6 m3 s-1) and 707 ft3 s-1 (20.0 m3 s-1), respectively.

Kisco storms.- High turbidity and high flow samples were collected from 2 storms (Fig. 2.8) occurring on 13-14 April 2004 and 8 September 2004 at the Kisco River (site 55). The 8 Sept. storm event had the greatest magnitude, resulting in peak discharge of 535 ft3 s-1 (15.2 m3 s-1). This storm amounted to an 84-fold increase in flow within 14.7 hours (6.6 ft3 s-1to 535 ft3 s-1). Since the Kisco hydrograph consists of two different peaks (first minor peak associated with more local runoff), an attempt is made to collect 2 sets of high turbitidy and high flow samples from each peak in the storm event hydrograph. For the September storm, samples from the first peak occurred at flows of 149 ft3 s-1 (4.2 m3 s-1) and 209 ft3 s-1 (5.9 m3 s-1), HT and HF respectively, and samples from the second peak occurred at flows of 420 ft3 s-1 (11.9 m3 s-1) and 535 ft3 s-1 (15.2 m3 s-1), HT and HF respectively. Only a high flow sample was collected during the first peak of the April storm (135 ft3 s-1; 3.8 m3 s-1) and the high turbidity and flow samples from the second peak occurred at flows of 125 ft3 s-1 (3.5 m3 s-1) and 205 ft3 s-1 (5.8 m3 s-1), respectively.

Literature Cited

- 13 - CHAPTER 2 – TECHNICAL DESIGN NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

nition of site of nition Agricultural Forest Water Wetland Suburban Urban and ) 2 Road Density Density (m/km West of Hudson region . P I and II Type columns West of Hudson region . P I and II Type ) 2 /cm 3 SPDES (cm Annual Flow ) 2 (ind./km Pop. Density Pop. Density % transporation land cover. ) r baseflow” (WB) and/or “stormflow” (S) site (see text for defi for (S) site (see text “stormflow” (WB) and/or r baseflow” 2 5.9 10.0 0 820 10.0 20.1 66.8 0.5 1.1 41.4 9.534.853.3 0 9.1 1046 10.990.7 077.8 0.00993.1 4.4 1171 984 8.874.4 33.4 8.4 8.1 60.4 8.5 0 4.676.7 0 0.015 0.1 3.683.5 21.0 26.864.5 909 0.5 865 1204 67.4 65.8 3.2 087.171.1 0.3 8.0 4.7 0.2 11.0 2.9 4.8 0 332 1.2 38.1 1.3 1.4 0 14.512.5 0 18.5 2.9 4.739.7 364 80.2 1.936.5 75.0 89.7 061.5 631 3.9 0 0.024.8 637 0.3 3.9 0.7 0.4 10.7 2.6 0.5 361 7.9 0 1.0 1.3 94.5 1.5 1.8 509 7.0 0 0 0.5 6.0 0.0 0 1.1 98.0 0.5 604 0 1.0 1.5 644 0.6 1026 0 97.5 0.0 95.4 1074 0.8 3.4 1041 0.1 1.1 0.6 2.7 0.0 5.1 97.4 803 97.1 5.7 0.6 20.1 0.2 0.9 5.2 10.5 30.5 0.0 75.3 4.7 32.1 85.9 0.5 61.8 29.4 0.5 0.0 59.4 0.0 0.2 63.4 8.0 0.4 0.1 85.6 0.2 0.2 1.1 1.5 0.1 0.8 0.6 Area Wtsh Wtsh 122.6135.4372.3663.9 22.5878.6 7.0 21.4 0.506181.6 16.2121.1 0.171 0211.9 18.3 1277 0.176448.2 11.9 1135 0.386 1094 10.4 1127 5.7 0.007 7.7133.2 12.1 1147 5.7 0.004250.1 5.2 5.6 32.3 0.002 1113 0.056189.4 1019 58.8 5.8 28.4589.0 14.2 19.6 25.8 806 10.7 62.7 5.9 0.4 983163.4 72.7 65.6 5.5 0.328 25.3 13.3 0.2 0.180 1.3 10.5 65.8 3.8 0.4 0.3 13.6 5.1438.8 1001 0.025 1.3 7.8 79.5 0.2 0.085 6.3 936 0.7 1.2 85.8 6.0165.9 9.5 4.0 0.2 888 0.104 1.2 100.3 89.4 8.6 831 3.6 0.1 84.4 0.7 0.1 4.9 0.042 571 3.5 0.1 0.6 2.3 3.8 4.3 4.0 89.3 0.6 0.7 539 89.0 1.6 7.4 0 0.3 5.8 0 84.8 0.4 87.6 1.5 1.9 1.2 0.2 453 1.7 0.3 584 95.9 1.7 1.0 1.1 0.0 1.5 96.2 0.9 0.6 0.1 1.0 1.3 97.0 0.8 97.2 0.1 0.0 0.7 0.4 (km over categories for stream sampling sites in the sites in the sampling for stream over categories Urban/suburban did not include or “Integrative” (I) and a “winte : Phase I and II percent land c land : Phase I and II percent W. Br. Delaware R. nr Stamford Delaware Br. W. nr Hobart Brk. Town Kortright South at R Br. Delaware W. R. nr Delhi Delaware Little R. nr Delhi Delaware Br. W. Hawleys at R. Br. Delaware W. Walton nr Brk. West T WB T R. nr Walton Delaware Br. W. Cr. Trout nr Cr. Trout WB I Arkville nr R. E. Br. Delaware nr Arkville Kill Bush Brk. nr ArkvilleDry WB T T S WB T I R. nr Dunraven Delaware E. Br. S WB T Dunraven at Kill Platt Andes Kill nr Tremper WB T Center Jewett nr Kill East T WB Center nr Jewett Cr. Schoharie T WB I nr Lexington Cr. Schoharie T nr Lexington Kill West WB T Kill nr Batavia Prattsville WB I Cr. nr Prattsville WB Schoharie I Indian Big nr Cr. Esopus WB T Allaben nr Cr. Esopus T Phoenicia at Cr. Stony T WB T WB T Mount Tremper at Kill Beaver Tremper Mount nr Cr. Esopus WB I WB I nr Claryville Neversink Br. W. nr Claryville Neversink E. Br. T WB T Claryville nr R. Neversink WB T Corner nr Lowes Cr. Rondout WB T Rose Brk. nr South KortrightCoulter Brk. nr Bovina Center WB I T T Delhi East at Cr. Elk T WB WB T Fraser at Brk. Planter WB T T WB Walton nr Brk. East T Beerston nr Brk. Dryden WB S I WB I WB S I T T WB WB T T T T WB 2 4 5 6 7 9 1 3 8 # Name Site Type PI - Type PII - 11 12 13 14 18 19 21 22 24 27 30 10 15 16 17 20 23 25 26 28 29 102 103 104 105 106 101 types). Phase II sites are bolded. define sites as “Targeted” (T) Table 2.1 - 14 - CHAPTER 2 – TECHNICAL DESIGN NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005 Agricultural Forest Water Wetland Suburban Urban and Urban ) 2 Road Density Density (m/km ) 2 /cm 3 SPDES (cm Annual Flow ) 2 (ind./km Pop. Density Pop. Density ) 2 2.4 16.9 0.537 1159 4.1 5.9 89.1 0.1 0.4 m 34.249.655.144.0 6.864.535.9 9.512.1 3.9 035.5 1.039.0 2.0 0 3.4 066.9 738 2.0 032.2 4.6 028.8 1079 4.3 023.4 763 0 3.1 0.01322.1 14.8 494 0.03045.6 13.6 6.3 51623.6 700 0.219 5.4 3.9 11.131.7 563 751 0.528 5.3 0.811.6 567 13.2 84.0 2.1 1.016.5 1023 1.7 0 3.8 79.8 1143 2.0 7.1 9.4 0 3.8 0.2 11.4 1.4 2.8 0 88.0 10.2 6.7 0.0 5.2 0 17.3 95.7 11.8 1.5 381 4.0 0 2.3 91.9 0.0 0 6.4 546 77.9 0.6 1.9 0.1 0 328 93.1 19.5 89.1 0.0 0 520 94.7 0.6 1.4 0.1 3.4 1045 73.1 0.4 1046 0.3 0.1 0.1 0.4 1478 91.1 0.6 0.7 0.8 0.3 1102 0.1 2.0 0.8 4.8 0.5 6.8 0.0 1.0 0.4 4.3 97.3 1.5 0.2 5.9 98.3 0.1 0.2 22.9 98.9 20.6 0.0 99.5 67.6 0.0 3.1 69.9 0.0 6.9 0.1 91.2 0.0 0.7 0.7 0.1 86.5 0.2 0.0 0.1 2.5 0.0 0.8 0.1 0.4 Area Area Wtsh Wtsh 132.4 14.1 0.036 884 5.0 7.1 84.7 0.3 1.9 (k :Continued E. Br. Delaware at Roxbury at E. Br. Delaware Scudders Run nr Roxbury Run Kellys Corner Kill nr Batavia Fleischmanns nr Cr. Vly Mapledale nr Brk. Dry Mills Grant nr Mill Brk. Downsville nr Clove Coles Downsville nr Brk. Holiday Park Elka near Cr. Schoharie Jewett East KillEast near Batavia Kill near Windham Gorge Grand Kill nr Bear WB T T WB Big Indian at Cr. Birch T Shandaken at Cr. Bushnellsville Warner Cr. nr Chichester Valley Frost above R. Br. Neversink W. T Rondout Cr. near Peekamoose T Betty Brk. nr South Kortright T T Cr.. Trout nr Brk. Loomis T T Hill Pine above Cr. Birch Emory Brk. at Fleischmanns T T WB T T T T WB T T WB T T T T # Name Site Type PI - PIIType - Table 2.1 108 110 111 112 113 114 115 116 117 119 120 121 122 123 151 153 159 160 118 109 107 - 15 - CHAPTER 2 – TECHNICAL DESIGN NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005 define sites as ypes). Phase II sites are Agricultural Forest Water Wetland Suburban Urban and Urban ) 2 Road Density Density (m/km ) 2 /cm 3 SPDES (cm Annual Flow ) 2 sites in the East of Hudson region . P I and II Type columns sites in the East of Hudson region (ind./km Pop. Density Pop. Density ) 2 nd/or “stormflow” (S) site (see text for definition of t 8.4 123.91.5 0 646.06.8 2262 0 15.1 462.75.3 3.075 6134 6.17.7 70.0 4854 40.6 33.0 0.9 163.23.6 33.7 0 5.4 7.5 3.1 0 53.15.4 60.7 5.6 28220.2 0.0 101.6 45.00.4 25644.0 0 465.0 4.2 0.6 2.1 465.9 0 15.89.5 6.102 11.7 86.5 21166.0 68.5 2.4 0 5123 15.0 5273 87.2 0 20.6 61.9 45.1 0 110.2 22.2 5826 2.0 27.4 0.1 3751 9.5 0 2740 3.6 0 57.1 13.7 6.3 67.9 11.7 54.4 57.4 54.1 12.5 1641 0.0 2164 7.0 2.3 16.6 0.1 32.8 1.8 6.6 8.4 14.8 26.2 4.1 5.1 0.0 73.6 0.0 5.6 6.3 1.3 3.0 2.7 69.3 75.9 3.7 5.9 0.1 12.3 2.9 m 11.017.725.126.9 85.528.4 22.410.358.2 56.4 0 104.7 035.5 35.1 108.8 0.100 0 213211.9 63.8 1971 0 259.6 3077 0 12.0 1515 0 7.935.1 0.100 87.2 1720 16.363.9 2240 5.4 4.9 0.077 2066 448229.5 34.8 76.8 391.9 5.4 13.2 4.2 54.8 137.015.6 2864 6.1 22.4 70.8 8.6 0.744.5 0.683 516.0 0.134.6 83.4 4.3 3.8 0 16.9 1.0 5.1 444.4 428845.5 75.2 6.941 4.4 76.0 0.3 1.9 3.7 130.6 7.3 63.638.9 73.9 0.060 90.6 11.7 3020 3.2 1.2 29.3 4.5 5339 0.003 380.4 62.7 2.8 4.6 11.6 5131 5.4 17.2 0 33.1 371.1 3545 0.2 0.068 5.8 6.4 8.9 17.9 27.8 45.312.9 11.7 6.8 13.7 3431 4901 13.6 0 56.7 8.3 42.631.2 28.6 9.4 17.2 26.7 4.6 92.2 5.9 4678 9.7 1.7 58.5 65.8 0 0.021 8.4 206.1 6.4 26.4 0.1 8.0 6.5 3.4 65.5 60.4 3066 3.9 1511 0 9.8 6.5 0.6 0.2 16.6 59.9 3.8 3420 9.4 5.9 0.2 4.7 21.4 4.6 6.6 69.6 86.8 14.2 0.7 0.4 51.8 8.2 4.0 1.5 9.5 Area Area Wtsh Wtsh 117.1238.3209.1 96.9 0.003 147.3 163.1 0.277 2192 0.612 2680 10.9 2931 15.2 3.6 16.1 71.0 11.4 4.4 61.6 7.1 65.3 2.8 7.1 7.2 8.6 6.4 (k % transporation land cover. land transporation % : Phase I and II percent land cover categories for stream sampling W. Patterson nr PattersonCr. Patterson W. nr Pawling Brk. Brady Mills nr Farmers Stream Leetown Haviland Hollow Br. at Haviland Hollow R. Croton Br. to Middle Trib Corners Allen nr R. Croton Br. W. Carmel Lake nr Pound Brk. Horse T Cliffs Kent nr R. Croton Br. W. T toTrib E. Br. Croton R. nr Brewster WB T nr Carmel R. Croton Br. Middle I WB Br. CrotonWest R. nr Crafts WB T Holly Stream nr Deans Corner T T Mahopac at West Brk. Secor T Falls nr Croton R. Croton E. Br. T T Br. CrotonW. R. nr Butterville Baldwin Place R. nr Muscoot I WB Station Purdys R. nr Titicus CornerCrook Brk. nr Grant T Amawalk nr Mill Brk. Hallocks T Angle Fly Brk. nr Whitehall Corners T WB T Park State nr Mohansic Brk. Hunter Resv Ridge Pound Cross R.Ward in T Hills Bedford nr R. Hill Stone I WB R. Croton Tributray Unnamed I WB Stanwood nr R. Kisco T T WB T Kitchawan nr Brk. Gedney T T I WB Mount Kisco at R. Kisco Purdy Lake nr R. Croton of Trib I WB nr Hawthorn Resv Kensico of Trib nr WCA Resv Kensico of Trib T Pawling nr R. E. Br. Croton trib. Unnamed T Park County Merrit W.G. at Brk. Quaker Stump Pond Stream nr PawlingBlack Pond Brk. at Meads Corner T S I WB WB T R. E. Br. Croton trib. Unnamed T S WB T Center T Salem R. near Titicus WB T T T WB T T T I WB # Name Site Type PI - PIIType - 42 45 46 34 40 44 49 51 55 57 59 60 32 33 35 36 37 38 39 41 43 47 48 50 52 53 54 56 58 31 126 127 124 125 129 130 bolded. Urban/suburban did include did bolded. Urban/suburban “Targeted” (T) or “Integrative” (I) and a “winter baseflow” (WB) Table 2.2

- 16 - CHAPTER 2 – TECHNICAL DESIGN NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005 Agricultural Forest Water Wetland Suburban Urban and Urban ) 2 Road m/km Density Density ( ) 2 /cm 3 SPDES cm ( Annual Flow ) 2 (ind./km Pop. Density Pop. Density ) 2 5.65.95.4 125.8 563.17.1 2.256 211.93.2 1.027 29955.7 0.056 528.8 200.1 5073 20.6 44234.0 130.9 01.0 34.3 0 25.3 0.087 8.5 228.9 5670 439.4 62.2 3.9 3865 2186 3.1 55.9 0 32.9 0.0 62.9 0 16.1 18.0 1.1 8.3 1.4 5329 11.2 6776 2.7 5.8 53.3 8.5 6.0 17.1 76.8 60.5 53.7 0.0 0.0 0.0 2.4 6.2 10.0 0.9 11.6 71.2 35.9 0.9 0.0 4.2 0.4 m 24.016.1 245.4 332.096.9 0 0.04315.0 344.1 3565 406813.419.4 2.771 20.9 95.5 25.8 204.0 426527.8 98.1 11.3 0 3.602 3.4 28.6 53.3 158.9 46.2 0 3289 5370 2.0 16.6 0.005 7.5 11.0 16.2 3148 33.2 45.3 6.9 3972 8.3 12.0 9.2 8.9 16.5 8.7 67.1 52.4 3.0 0.5 0.2 6.8 71.9 61.1 6.7 4.1 1.0 4.3 11.0 9.6 k Area Area Wtsh Wtsh 116.7 107.9 0.118 3295248.2 14.1 123.8 5.9 64.4 0.099 5.3 2433 8.9 12.0 12.9 64.9 1.9 7.9 ( : Continued : Continued Titicus R. near North Salem North near R. Titicus Corner Sears nr Brk. Bog Unnamed trib. Muscoot R. at Mahopac Falls Brk.Plum at ShenorockUnnamed trib. Muscoot Res. nr Goldens BridgeCross R. nr KatonahMuscoot R. Whitehallnr Corners Yorktown. nr Brk. Hunter Unnamed trib. Croton Res. nr Croton Heights T WB Bedford nr R. Kisco T trib. R. to Cross nr Cross R.Unnamed Broad Brk. nr T Bedford Hills Bedford nr R. Hill Stone T Unnamed trib. Kensico Res. at Mt Pleasant T Unnamed trib. Kensico Res. nr ThornwoodWaccabuc R. at Boutonville T I WB Brewster at R. E. Br. Croton T T T T T WB T WB WB T T T T WB Table 2.1 # Name Site Type PI - PIIType - 132 134 140 143 146 147 150 139 131 133 137 138 141 142 145 148 149

- 17 - CHAPTER 2 – TECHNICAL DESIGN NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Table 2.3: Location information for the 27 East of Hudson Phase II stream sites. Each site (except 150; from map) was located using a Trimble GPS Pathfinder TM ProXR receiver unit, with real-time correction (Datum = WGS 84). See Table 2.1 for site names and descriptive information. Phase I site locations can be found in Stroud (2004).

Maximum Horizontal Position Precision Site Latitude Longitude Decimal Degrees Dilution of (95% CI) Precision m 34 41.49438077 -73.54641599 3.5 1.217 46 41.33265904 -73.76496965 4.5 1.037 52 41.26028843 -73.60198649 4.4 0.860 55 41.22898049 -73.74356273 2.5 1.281 124 41.54005559 -73.61557599 5.3 1.443 125 41.49874968 -73.53383643 4.7 1.122 126 41.50820788 -73.68247079 2.9 0.881 127 41.48360034 -73.76890208 3.1 0.953 129 41.42346001 -73.55755546 5.2 0.963 130 41.32768298 -73.58078559 5.5 0.887 131 41.33487622 -73.55814227 4.1 1.152 132 41.42927864 -73.58463644 6.0 1.202 133 41.37483505 -73.76203475 2.3 0.723 134 41.33612911 -73.73477869 4.5 1.655 137 41.28965343 -73.65908981 4.2 0.915 138 41.26682982 -73.66836286 3.1 1.022 139 41.27257487 -73.74575572 5.8 2.170 140 41.29094958 -73.83465386 2.3 0.602 141 41.24316019 -73.81795596 4.8 1.203 142 41.19248383 -73.72695417 4.8 0.929 143 41.27460174 -73.61832933 3.1 0.932 145 41.24776891 -73.67044354 5.3 1.093 146 41.21572580 -73.63194603 5.1 1.161 147 41.12490180 -73.74346656 24.8 3.505 148 41.10273616 -73.75709573 4.3 1.376 149 41.25844110 -73.56610994 2.3 0.804 150 41.40277778 -73.59305556 NA NA

- 18 - CHAPTER 2 – TECHNICAL DESIGN NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Table 2.4: Location information for the 35 West of Hudson Phase II stream sites. Each site was located using a Trimble GPS Pathfinder TM ProXR receiver unit, with real-time correction (Datum = WGS 84). See Table 2.2 for site names and descriptive information. Phase I site locations can be found in Stroud (2004).

Horizontal Latitude Longitude Maximum Position Precision Site Dilution of Precision (95% CI) Decimal Degrees m 3 42.34367436 -74.71979975 2.2 0.656 6 42.17548414 -75.01828999 4.2 0.924 9 42.17376864 -75.27943302 2.2 1.082 10 42.16987985 -74.61151354 2.4 0.651 15 42.12610104 -74.81170240 2.6 0.688 23 42.11731029 -74.37679339 2.4 0.572 26 42.03961869 -74.28169149 5.2 1.512 29 41.90174954 -74.58072348 3.0 0.752 101 42.33553867 -74.73917211 2.9 0.768 102 42.25703301 -74.77161393 3.5 0.947 103 42.29949546 -74.89223927 4.3 1.122 104 42.24288046 -74.96426207 7.3 1.911 105 42.18096091 -75.10621802 8.3 1.510 106 42.11789180 -75.24962674 12.2 0.861 107 42.29375686 -74.55917861 3.6 0.876 108 42.23393721 -74.59036309 2.8 0.643 109 42.18139839 -74.59126882 5.0 1.239 110 42.17068580 -74.51546992 4.7 1.185 111 42.10767973 -74.56120015 5.8 1.172 112 42.10617040 -74.73026838 5.7 0.879 113 42.12910492 -74.89832297 5.0 1.066 114 42.06546841 -74.87722641 3.3 0.769 115 42.17198754 -74.14987465 2.4 0.605 116 42.24242108 -74.17846641 3.4 0.834 117 42.29337212 -74.30532935 3.3 0.738 118 42.33792438 -74.45102525 4.0 0.927 119 42.10938420 -74.45181128 6.1 0.938 120 42.12109310 -74.39868066 3.1 0.837 121 42.10009907 -74.29540223 8.8 0.985 122 41.99047876 -74.49263600 5.6 1.066 123 41.91630759 -74.43564844 4.4 1.013 151 42.34512936 -74.73337798 3.3 0.919 153 42.15946134 -75.27729896 11.6 1.235 159 42.14392778 -74.48091667 NA NA 160 42.15120278 -74.51986944 NA NA

- 19 - CHAPTER 2 – TECHNICAL DESIGN NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

7.4 7.4 8.1 8.4 8.6 1.2 0.8 1.2 Wetland

4.7 4.7 9.9 4.6 6.1 2.6 2.4 1.8 Water

60.3 60.3 43.6 56.7 63.3 92.7 82.3 68.8 Forest

7.9 7.9 8.7 6.1 1.7 11.7 11.7 10.2 21.4 Agriculture Agriculture

e seven reservoir sites. “Active reservoir sites. “Active e seven Land Cover Percentages (%) Percentages Cover Land

1.6 1.6 4.1 5.2 18.3 18.3 28.3 17.2 14.6

Urban/ Urban/ Suburban

land cover methodology. WOH reservoir ) 2

575 575 952 3213 4214 3020 3170 1084 Road Density Density (m/km

) 2

4 8 15 186 186 392 137 130 Pop. Density Density (ind./km

, and descriptive information for th information descriptive , and 9 0 4 0 7 7 23 # of # of Active SPDES SPDES

) 2

censed dischargers. See text for 51 64 78 rban does not include transporation. 238 238 961 820 820 Area Area 1178 Wtsh Wtsh (km

295 295 224 276 363 599 (ha) (ha) Area Area 2096 1849 Surface

Titicus Reservoir : Station names, substation locations Muscoot Reservoir Muscoot Reservoir Pepacton Reservoir Reservoir Pepacton Amawalk Reservoir Reservoir Amawalk Neversink Reservoir Reservoir Neversink Cross River Reservoir Cannonsville Reservoir Station Name – Name Station East of HudsonEast of Reservoirs 65 66 67 West of Hudson Reservoirs Reservoirs West of Hudson 155 155 156 157 158 Site SPDES” = number of active SPDES li watershed land covers for urban/subu Table 2.5

- 20 - CHAPTER 2 – TECHNICAL DESIGN NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Table 2.6: Location data for reservoir substations. Sites were located using a Trimble GPS Pathfinder TM ProXR receiver unit, with realtime correction (Datum = WGS 84).

Maximum Horizontal Substation Latitude Longitude Position Site Reservoir Precision Number Dilution of (95% CI) Precision degrees, minutes, seconds 155 Muscoot 1 41° 16' 41" N 73° 41' 27" W 2.0 0.4 2 41° 16' 05" N 73° 41' 30" W 3.1 0.7 3 41° 16' 18" N 73° 42' 48" W 2.7 0.5 156 Amawalk 1 41° 19' 03" N 73° 44' 31" W 3.0 0.5 2 41° 18' 45" N 73° 44' 23" W 4.8 0.8 3 41° 17' 47" N 73° 44' 51" W 6.0 0.9 157 Titicus 1 41° 19' 52" N 73° 36' 47" W 4.8 0.8 2 41° 19' 53" N 73° 37' 40" W 3.6 0.6 3 41° 19' 39" N 73° 38' 32" W 3.9 0.6 158 Cross River 1 41° 15' 10" N 73° 37' 18" W 2.4 0.3 2 41° 15' 14" N 73° 37' 55" W 4.1 0.9 3 41° 15' 45" N 73° 39' 17" W 4.3 0.5 65 Neversink 1 41° 50' 19" N 74° 38' 53" W 2.3 0.3 2 41° 49' 53" N 74° 39' 52" W 1.8 0.3 3 41° 50' 52" N 74° 40' 09" W 1.7 0.3 66 Pepacton 1 42° 05' 13" N 74° 48' 05" W 2.6 0.4 3 42° 04' 54" N 74° 52' 30" W 2.1 0.4 4 42° 06' 04" N 74° 49' 39" W 2.8 0.3 5 42° 04' 27" N 74° 50' 14" W 2.2 0.3 6 42° 04' 30" N 74° 53' 15" W 2.4 0.3 7 42° 06' 07" N 74° 54' 05" W 2.2 0.3 67 Cannonsville 3 42° 06' 07" N 75° 17' 51" W 3.1 0.4 4 42° 05' 52" N 75° 19' 07" W 2.2 0.4 5 42° 07' 25" N 75° 18' 25" W 4.3 0

- 21 - CHAPTER 2 – TECHNICAL DESIGN NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Figure 2.1: Location of Phase II sampling sites in the East of Hudson Watersheds (a.k.a., Croton/Kensico System). Study sites names and descriptive information is found in Tables 2.1 and 2.3, by site number.

- 22 - CHAPTER 2 – TECHNICAL DESIGN NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005 em). Study site names and .k.a. Delaware/Catskills Syst of Hudson watersheds (a bles 2.2 and 2.4, by site number. Location of Phase II sampling sites in West Figure 2.2: descriptive information are found in Ta

- 23 - CHAPTER 2 – TECHNICAL DESIGN NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Figure 2.3: Location of reservoir sampling stations (each reservoir) and substations (actual sample locations) within each reservoir for the East of Hudson Reservoirs. Solid circles represent Phase II substations and Phase I substations are represented by open circles in in-lay map.

- 24 - CHAPTER 2 – TECHNICAL DESIGN NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Figure 2.4: Location of reservoir sampling stations (each reservoir) and substations (actual sample locations) within each reservoir for the West of Hudson Reservoirs. Solid circles represent Phase II substations and white crosses represent Phase I substations.

- 25 - CHAPTER 2 – TECHNICAL DESIGN NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

A Brush

MxdBrushCroplandGrass WtsdArea Farmstead

OtherUrban

MxdForest Orchard IndustrialCommercial SPDEScm Water Conifer RoadDens ResidentialWetlandPopDens00 Deciduous

WOH EOH EBD B EMC WBD KSC ESP MNC 3 NRD TCS SCH WBC

-3

Phase I C Phase II Phase I and II Integrative 3 PCA Factor 2 - Inertia = 17.8% PCA

-3 -30 3 PCA Factor 1 - Inertia = 32.9 % Figure 2.5: A correlation Principal Components Analysis (PCA) for (A) land cover/use variables (percent agriculture, forest, urban, water and wetland categories, 2000 population density, road density, average annual SPDES permitted discharge per unit area, and watershed area at each stream study site (110 sites from Phase I and II). Site relationships with the F1 and F2 axes are shown separated by (b) major catchment or (c) phase and integrative versus targeted.

- 26 - CHAPTER 2 – TECHNICAL DESIGN NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Site 6 - West Branch Delaware River - Storm on 23 September 2003

3000 Discharge - Delhi Discharge - Walton Transducer x1350 2500 Discharge - Estimate Sampling Points High Turbidity Sample High Flow Sample

2000 ) -1 s 3

1500 Discharge (ft 1000

500

0 0:00 2:00 4:00 6:00 8:00 10:00 12:00 14:00 16:00 18:00 20:00 22:00 0:00 Time

Site 6 - West Branch Delaware River - Storm on 13-14 April, 2004

1200 Discharge - Walton Discharge - Delhi Sampling Points 1000 High Turbidity Sample Transducer x1000 Discharge -Estimate

800 ) -1 s 3

600 High Flow Sample Discharge (ft 400

200

0 0 2 4 6 8 1 1 1 1 18:00 2 22:0 0:00 2:00 4:00 6:00 8:00 10:00 12:00 1 1 1 2 2 0 :0 :0 :0 :0 :0 0: 2: 4: 6 0 4:00 6: 8: 0: 2: :0 0 0 0 0 0 0 0 0 :0 :0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

Site 6 - West Branch Delaware River - Storm on 27 April, 2004

1800 Discharge - Walton

1600 Discharge - Delhi Sampling Points

1400 Transducer x1000 High Flow Sample Discharge -Estimate

1200 ) -1 s 3 1000

800 High Turbidity Discharge (ft 600

400

200

0 0:00 2:00 4:00 6:00 8:00 10:00 12:00 14:00 16:00 18:00 20:00 22:00 0:00 2:00 4:00 6:00 8:00 10:00 12:00 14:00 16:00 18:00 20:00 22:00 0:00 Figure 2.6: Storm hydrographs showing sampling times for high turbidity and flow samples collected from the West Branch of the Delaware River near Hawleys (site 6) for storms on 23 September 2003, 13- 14 April 2004, and 27 April 2004. Discharge estimated from USGS instantaneous (provisional) discharge data for the W. Br. Delaware at Walton (USGS ID 01423000) gaging station by correcting to estimate flow at site 6 using watershed area to discharge relationships (see text for details).

- 27 - CHAPTER 2 – TECHNICAL DESIGN NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Site 29 - Neversink River - Storm on April 13-14, 2004

800

Discharge High Turbidity Sample 700 Sample Points

600 High Flow Sample

) 500 -1 s 3

400

Discharge (ft 300

200 Discharge Data: USGS

100

0 0:00 2:00 4:00 6:00 8:00 10:00 12:00 14:00 16:00 18:00 20:00 22:00 0:00 2:00 4:00 6:00 8:00 10:00 12:00 14:00 16:00 18:00 20:00 22:00 Site 29 - Neversink River - Storm on May 27, 2004

1200 High Turbidity Sample High Flow Sample Discharge Sample Points 1000

800 ) -1 s 3

600 Discharge (ft 400

Discharge Data: USGS

200

0 0:00 2:00 4:00 6:00 8:00 10:00 12:00 14:00 16:00 18:00 20:00 22:00 0:00 2:00 4:00 6:00 8:00 10:00 12:00 14:00 16:00 18:00 20:00 22:00

Figure 2.7: Storm hydrographs showing sampling times for high turbidity and flow samples collected from the Neversink River for storms on 13-14 April 2004 and 27 May 2004. Discharge estimated from USGS instantaneous (provisional) discharge data for the Neversink River near Claryville, NY (USGS ID 01435000).

- 28 - CHAPTER 2 – TECHNICAL DESIGN NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Site 55 - Kisco River nr Stanwood - Storm on April 13-14, 2004

3 250 Rain Storm Total (in) Discharge Sampling Points 2nd High Flow 2.5 200

2 st 1 High Flow Sample ( (ftDischarge 150

1.5 3 s -1

High Turbidity Sample 100 ) 1

Precipitation Storm Total (inches) Total Storm Precipitation Rain Data: Byram Lake Weather Station 50 0.5

0 0 0 0 0 0 0 0 0 0 0 0 :00 :0 :00 :0 :0 :0 :00 :0 :00 0:0 4 6 8:0 2 4 2:0 4 6 8:0 2 4 0 0 02:00 0 0 0 10:00 1 1 16:00 18:00 20:00 2 00:00 02:00 0 0 0 10:00 1 1 16:00 18:00 2 22:0 00:00 Site 55 - Kisco River nr Stanwood - Storm on Sept 8-9, 2004

6 600 Rain Storm Total (in) Discharge

Sampling Points nd 2 High Turbidity Sample 5 2nd High Flow 500

4 Rain Data: Byram Lake Weather Station 400 Discharge (ft Discharge

3 300 st 3

1 High Flow s -1 )

2 200 PrecipitationStorm Total (inches)

1 1st High Turbidity Sample 100

0 0 6:00 7:00 8:00 9:00 10:00 11:00 12:00 13:00 14:00 15:00 16:00 17:00 18:00 19:00 20:00 21:00 22:00 23:00 0:00 Figure 2.8: Storm hydrographs showing sampling times for high turbidity and flow samples collected from the Kisco River (EOH) for storms on 13-14 April 2004 and 8-9 September 2004. Discharge from USGS instantaneous (provisional) discharge data for the Kisco River below Mt. Kisco, NY (USGS ID 01374987).

- 29 - CHAPTER 2 – TECHNICAL DESIGN NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

------Intentionally Blank------

- 30 - CHAPTER 2 – TECHNICAL DESIGN NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Chapter 3 - Nutrients, Major Ions and Suspended Particles in transport

Research Task

Concentrations of nutrients and major ions transported in streams can be sensitive indicators of ecosystem impairment, particularly when monitored over a several year period of time and across landscapes of complex land-use patterns. They also provide important supplementary data for the other aspects of this monitoring project. These stream constituents provide an assessment of inorganic and nutrient water quality relative to differences in existing watershed characteristics and can be used to quantify and predict changes in water quality in response to changes in land use. We monitored nutrients and major ions, as well as suspended particles, in year five of the project, under baseflow conditions at 62 sites located throughout the NYC drinking-water source watersheds, including 48 Phase II sites, 12 existing Phase I sites, and 2 new sites in the Highmount/Fleischmanns area. Sampling was also carried out during storms at three stormflow sampling sites.

The suspended particles monitored included both organic and inorganic particles. Organic particle dynamics are indicative of upstream processing of organic matter, stream linkage (i.e., the upstream to downstream transfer of organic energy), and carbon loading to downstream reservoirs. The objectives for the suspended solids portion of this task are to characterize the concentrations and transport of inorganic and organic particles under baseflow conditions at both the targeted and integrated sampling stations, relate particle concentrations to land use and land cover, and to describe, to a first approximation, the response of organic particle transport to runoff events at three storm-sampling sites. Here we present a summary of the year five investigation, including observations concerning the spatial and temporal variation of suspended particles throughout the study region.

Methods

Baseflow

Samples analyzed for nutrients, major ions, and suspended particles for each of the 62 study sites east and west of the Hudson River (EOH and WOH, respectively) were collected between June and August. Nutrient and major ion baseflow sampling was coordinated with the molecular tracer and dissolved organic carbon (DOC)/biodegradable DOC (BDOC) tasks when possible. If a river appeared unusually turbid, or otherwise displayed signs of high flow (based on available USGS real-time gauging stations at or within the vicinity of the sampling sites), at the time scheduled for sampling, the site was (re)sampled at a later date.

Upon arrival at each sampling site, pH, specific conductance, and temperature were measured in situ. Nutrients, major ions, and suspended particles were collected from below the water surface in regions of significant current using polyethylene bottles which were rinsed thoroughly with deionized water followed by stream water (3 rinses with stream water).

Specifically at each site, two 125-mL polyethylene bottles were filled with stream water for subsequent analysis of Total Kjeldahl Nitrogen (TKN) and Total Phosphorus (TP) (bottle 1) and alkalinity (bottle 2). A 450-mL polyethylene bottle was then filled with stream water used for

- 31 - CHAPTER 3 – NUTRIENTS IN TRANSPORT NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

immediate filtering. This water was filtered on-site through a cellulose nitrate membrane (0.45µm) filter and split among three 100-mL samples into three 125-mL polyethylene bottles for dissolved nutrients (total dissolved phosphate [TDP], ortho-phosphate [SRP], soluble kjeldahl nitrogen [SKN], nitrate [NO3-N], and ammonium [NH4-N]; bottle 3), anions (bottle 4), and cations (bottle 5), respectively. The remaining water in the 450-mL bottle was discarded. Alkalinity, anion, and cation samples (i.e., 3 separate 125-mL bottles) were chilled in a cooler filled with ice until returned to Stroud where they were stored in a refrigerator at 4°C until analysis. Two 125-mL bottles (1 filtered for dissolved nutrients, 1 not filtered for TKN and TP) were frozen for subsequent analyses. Anions [Chloride (Cl) and Sulfate (SO4)], cations [calcium (Ca), potassium (K), sodium (Na), and magnesium (Mg)], and total and dissolved nutrients were analyzed by a contract laboratory (the Patrick Center for Environmental Research laboratory) and alkalinity was measured at Stroud.

Samples for analyses of total (TSS) and volatile suspended (VSS) solids (seston) were also collected as part of the stream water sampling described above. Three replicate samples were collected at each site in three 5-L plastic jugs with the replicates taken in the same location, one after another. Samples were chilled at ~4°C using ice or freezer packs in coolers and transferred to a refrigerator until they could be processed (within 7 days from the time of collection). Suspended solids were filtered onto an organic-free, tared (0.5 µm) GF/F filter, filtering approximately 1 to 3 L of sample water. Filters were analyzed for suspended and volatile solids by drying at 60°C for ~48 hours and muffling at 500°C for ~2 hours and obtaining dry and ash weights at the respective stages.

Stormflow

The concentrations of nutrients, major ions, and suspended solids in transport (as well as molecular tracers [Chapter 4], and DOC [Chapter 6]) were quantified for two storms at each of our three storm-sampling sites (Neversink River – 29, W. Br. Delaware River – 6, Kisco River – 55). Storm events were sampled on April 13,14 (all three sites), April 26,27 (site 6), May 27, (site 29), and September 7,8 (site 55). Additional baseflow sampling, as described above, was conducted at each of these sites in order to have representative baseflow chemistry near the time of the separate storm events. These additional baseflow samples were only used in conjunction with analysis of the storm samples and were not used in any baseflow-specific analyses.

Each sampling site was instrumented with at least two ISCO automated samplers, the first of which was set to trigger following a 10-15-cm rise in stage height. The second ISCO was triggered by the completion of sampling by the first ISCO. Once triggered, the ISCOs sampled hourly in duplicate for a total of 6 h. If a third or fourth ISCO was used at a particular site, the triggering sequence was established in a similar fashion but with a greater delay after onset of sampling to capture later portions of the storm hydrograph. When a run-off event was imminent, changes in stream discharge and stage height at or near each site were monitored using real-time updates of stream-specific gauging stations on the USGS website. There was no USGS gauging station in close proximity to the storm sampling site at the W. Branch of the Delaware River, so we equipped this site with a single water-level recorder (pressure transducer) to record actual changes in stage height.

- 32 - CHAPTER 3 – NUTRIENTS IN TRANSPORT NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

The first sample from each duplicate pair of hourly samples (n=12) was used for the analyses in this task, and the second in each pair (n=12) was used for molecular tracer analyses. Two sets of duplicate samples were analyzed for each storm; one sample corresponded to peak flow (±1 h), as determined by provisional hydrograph data provided by the U.S. Geological Survey (or pressure transducer data) and the other corresponded to peak TSS transport (±1 h) selected by visual comparison of sample turbidity among collected sample jars. Each sample was filtered, with ~50mL of the filtrate from each sample used for DOC analysis. After thorough shaking of the original sample, the unfiltered nutrient and alkalinity sample splits were collected, as described above. The remaining unfiltered sample volume was processed for total and volatile suspended solids also as described above (following thorough shaking at each step). The filtrate from the DOC processing was subsequently re-filtered through a cellulose nitrate membrane (0.45µm) filter and split among (3) 125-mL HDPE bottles for dissolved nutrients and major ions, as described above. The resulting (5-6) 100-mL sample splits were filtered, fixed, and stored as necessary for subsequent analysis of whole and dissolved nutrients (SKN, TKN, TP, NO3-N, NH4-N, SRP, TDP), anions (Cl, SO4), cations (Ca, K, Na, Mg), and alkalinity. Samples were collected and stored on ice within 6-12 hours of sample collection by the ISCO automated sampler.

QA/QC

The QA/QC procedures of the Patrick Center for Environmental Research laboratory for all sample analyses for this project included analysis of lab blanks, duplicated samples, matrix spikes, reference or lab control standards (LCS), and continuing calibration standards (CCS). Laboratory quality control for suspended particles was evaluated using lab blanks (LB) of deionized water and a laboratory control sample (LCS) of resuspended stock particles collected from the benthos of White Clay Creek (certified to 16.7% OM by Lancaster Laboratory, PA). Each day of sample processing included the filtration of 1 to 2 LBs (~1 L) and 1 to 2 LCS (~90- 100 mL of freshly resuspended particles) onto organic-free, tared (~0.7-µm) GF/F filters and processed as above.

Baseflow. A field duplicate (FD) and field blank (FB) sample were collected at one predetermined site during four of the baseflow sampling weeks. QA/QC sites were randomly selected prior to the onset of the sampling season. Major ion/nutrient field-duplicated samples were collected simultaneously and in close proximity to each other in the water column and analyzed as discrete samples. Conductivity, pH, and temperature, which were measured in situ, were field-duplicated (each measurement was made twice) at only one site. For suspended particles, field duplicates were collected simultaneously by filling three additional 5-L bottles per duplicate while field blanks involved taking 4 L of deionized water and following the same protocol used to filter and analyze stream samples. Ion/nutrient field blanks were taken by transferring deionized water into the 1-L grab sample bottles at each of the selected QA/QC sites. Conductivity, temperature, and pH of the field blank samples were measured at the time of collection, although pH of the deionized water had little significance. For seston, the replicate sampling as previously described for field samples, including FB and FD samples, added another measure of precision for field sampling and laboratory analyses. The replicate sampling was also conducted for a few LB samples.

- 33 - CHAPTER 3 – NUTRIENTS IN TRANSPORT NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Stormflow. Field blanks (or "equipment rinsate blanks") were collected during baseflow by collecting a volume of deionized water through the ISCO sampling apparatus immediately following routine equipment maintenance. Field duplicates were collected from a sampled storm event by selecting a pair of duplicate samples other than those samples selected for actual stormflow sample analyses. A total of three FB samples were taken; two preceding a storm collection and one following a storm collection. Two sets of field duplicate samples were taken.

Ion balance. A cation/anion balance and the difference between measured and calculated conductivity were used as additional consistency checks on the baseflow and stormflow inorganic chemistry data. All concentrations were converted from mg/L to µeq/L for the ion balance calculation. DOC concentrations, which are often used to estimate the organic anion contribution to the ion balance, were not included. The equivalent concentrations were summed separately for the cations and anions, with the final ion balance expressed as a percent difference (PD):

(PD[ion]) = ((cations – anions)/(cations + anions)) * 100

Conductivity check. An additional data consistency check was performed using measured and calculated conductivity. Conductivity was calculated by summing the product of ion concentration (converted to µM) and associated equivalent conductivities. This sum was then adjusted for ionic strength at finite concentrations (APHA, 1992). As with the ion balance calculations, the conductivity calculation did not include the organic anion contribution (estimated from DOC concentrations). Agreement between measured and calculated conductivity was assessed through a percent difference:

PD[cond] = ((meas. conductivity – calc. conductivity)/meas. conductivity)*100

Results and Discussion

QA/QC

A full field and laboratory QAQC summary for major ions and nutrients is included in Appendix A3 of this report. It indicates that there were no QAQC issues within the laboratory effort and that no serious issues arose from the field QAQC. There were several continuing calibration sample exceedances for N and P species and one for Na, however, nearly all of these exceedances were for low-concentration samples (i.e., near stated detection limits) where relative (e.g., percent) recoveries can be inflated. Four field QAQC exceedances occurred within the field duplicate data. One exceedance occurred within a baseflow field duplicate pair for Organic N, with one for TKN and two for TP within the stormflow field duplicates. The Organic N exceedance within the baseflow field duplicate was not considered an issue given the low concentration of the duplicates (mean concentration = 0.04 mg/L), the fact that this analyte was a derived value (as the difference between SKN and NH4-N), and the fact that no other exceedances occurred within that particular duplicate set. The exceedances of TKN and TP within the stormflow duplicates does warrant further consideration as will be discussed below for the seston field duplicate results.

- 34 - CHAPTER 3 – NUTRIENTS IN TRANSPORT NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

A total of 87 laboratory blanks (LB) were analyzed for whole-sample TSS and 83 for VSS. TSS and VSS concentrations in these blanks averaged 0.37 and 0.32 mg/L, respectively. All but one of the TSS blanks met our acceptance criterion of a 1.2 mg/L detection limit, and 65 of the 83 VSS blanks met the acceptance criterion of the 0.22 mg/L detection limit. A total of 44 TSS and 42 VSS laboratory control standards were analyzed with a mean recovery of 98.9% for TSS and 116.6% for VSS. None of the TSS LCS samples exceeded the acceptance criterion of 80 to 120% accuracy, while 16 of the 42 VSS LCS samples did exceed the acceptance criterion. The mean coefficient of variation (CV; standard deviation divided by the mean multiplied by 100) for all field TSS replicates (triplicate sampling) was 7.9% with a range of 0.9% to 25.7%. For VSS, the mean CV was 8.2% with a range of 0.5% to 23.5%. The total number of replicated field samples (including field duplicates) was 55. The mean CV across the four FB was 36.1% (range of 7.6% to 83.8%) for TSS and 26.6% (range of 7.8% to 66.7%). For LB, a total of five samples were collected in triplicate, with a resulting mean CV of 5.0% (range of -177% to 72%) for TSS and 98.7% (range of -68.1% to 476%) for VSS. These results for blank sample precision, along with the number of blank exceedances noted above indicate potential problems with sampling the very small amounts of seston associated with both field and lab blanks. The field sample precision results though, do suggest that the problems associated with taking and analyzing blank samples do not extend to the field samples analyzed for either TSS or VSS. Further, we also believe that the LCS exceedances are associated with improper preparation of the LCS samples and not an indication of any deficiencies in the laboratory method. We will evaluate and rectify problems with both the blank and LCS sampling issues prior to the start of field sampling in year 6 (2005) of the project.

All of the baseflow field blanks (FB) for suspended particles processed during the year 5 field season met the QC criterion of less than 2x detection limit, with a mean TSS of 0.52 and a range of 0.38 to 0.72 mg/L and VSS mean of 0.26 mg/L (range 0.17 to 0.34 mg/L). The one exceedance for a FB was the VSS associated with storm sampling. However, the FB value of 0.52 mg /L was considerably below the lowest stormflow value for VSS of 7.65 mg/L. Only one set of field duplicates, collected as stormflow duplicates, did not meet the defined precision criterion of 40% (as a relative percent difference). The second sample in this particular duplicate pair had much less seston (both TSS and VSS) relative to the first sample in the pair. This result was mirrored by the TKN and TP results for this same duplicate pair as previously mentioned. Taken together, this storm-related field duplicate result suggests potential problems with the paired sampling of suspended solids during storms. However, another set of samples were analyzed as storm field duplicates and no issues were found. Both sets of field duplicates were sampled at the same site, using the same ISCO sampling set up. At this point, we will reserve any further judgment on our ability to collect simultaneous samples with a reasonable expectation of precision until further storm field duplicates are analyzed (which will occur in year 6 of the project). .

The ion balance and conductivity checks demonstrated strong consistency between the cation and anion sums and the measured versus-calculated conductivity, respectively. The mean PD[ion] (±1 standard deviation) for baseflow samples was -0.02% (±8.1%) while for stormflow samples the mean was -2.7%(±3.8%). Two sites had baseflow sample PD[ion] values that exceeded 10%; a difference of 30% was found for a Kisco River near Stanwood (55) sample and a -15% difference was determined for an E.Br. Croton River near Brewster (150) sample. There were 2 sets of samples taken from the Kisco River because it happened to be a site where a field

- 35 - CHAPTER 3 – NUTRIENTS IN TRANSPORT NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

duplicate was collected. The alkalinity sample was lost (bottle damaged in transit) from the first duplicate set and this is the set that results in the 30% PD exceedance (i.e., because carbonate alkalinity was missing from the balance). The ion balance for the second duplicate was valid. The PD[ion] value for the E.Br. Croton River sample was not very large, and the conductivity balance was reasonable (PD[cond] = -13%). No further action was deemed necessary in regard to either of these two samples. Only one stormflow sample had a PD[ion] value that exceeded 10%; the high flow sample on May 27, 2004 at the Neversink River (29) with a PD[ion] = -15%. This difference was not very large, and the conductivity balance was reasonable (PD[cond] = -5%) so no further action was deemed necessary in regard to this sample as well.

The mean PD[cond] (±1 standard deviation) across baseflow samples was -1.7% (±19%) and for stormflow samples the mean was 2.1% (±6.5%). Seven baseflow samples and no stormflow samples exceeded a PD[cond] value of 20%. Of the seven baseflow samples, six were for sites with relatively low conductivity values (measured conductivity ranges from 22 µS/cm to 75 µS/cm, but only one site had a measured value > 45 µS/cm) which leads to magnified percent differences; no further action was deemed necessary in regard to these samples. The seventh sample, a storm-associated baseflow sample at site 55 measured on September 14, 2004 had a PD[cond] = -116%, but a PD[ion] value of only 0.1%. Given the lack of any issues with the ion balance, the decision was made to substitute the calculated conductivity value of 326 µS/cm for the measured value of 151 µS/cm. The calculated conductivity value is much more in line with the two other baseflow-measured conductivity values for site 55; 343 µS/cm measured April 20, 2004 and 450 µS/cm measured on July 6, 2004.

Baseflow ionic, nutrient and suspended particle water quality

General relationships and patterns. The general patterns in cation and anion sums observed for the monitoring sites in 2003 continue with the 2004 data (Fig. 3.1). In general, the ionic composition, which indicates the amount of dissolved solids in stream water, continues to be two to four times greater for EOH sites relative to WOH sites. A majority of sites in both regions had higher ionic sums in 2004 relative to 2003 (Fig. 3.2). These inter-annual differences may be explained in part by differences in hydrology. In fact, 2004 discharge values, measured or estimated for the day of sampling at a site, were lower than corresponding 2003 values for 42 of the 60 sampling sites having two years of data.

The relationship between stream chemistry and discharge under baseflow conditions can be explored further using the 12 Phase II sites (WOH site numbers 3, 6, 9, 10, 15, 23, 26, and 29; EOH site numbers 34, 46, 52, and 55) that were also sampled during Phase I of the project. Conductivity versus baseflow discharge relationships (Figs. 3.3a, 3.3b) provide evidence that a significant controlling factor behind any inter-annual variability in stream chemistry in the study area is hydrology. Of the 12 regressions shown in Figs. 3.3a and 3.3b, 5 are significant at an α = 0.05, and 9 have R2 values greater than 0.5. All but one relationship is negative; as baseflow discharge increases, conductivity and more generally ionic chemistry concentrations, decrease. The one site with a slight (though very weak and not significant) positive relationship between baseflow discharge and conductivity is site 26 (Esopus Cr. near Mt. Tremper). Site 26 experiences strong flow modification from the Shandanken Tunnel inflow upstream of the sampling site which is likely the driving force in the stream chemistry signal observed at this sampling site. For the other 11 sites though, similar source-area contributions to baseflow may be

- 36 - CHAPTER 3 – NUTRIENTS IN TRANSPORT NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

occurring where lower baseflow chemistry reflects a greater contribution from deeper groundwater sources containing higher ionic concentrations and higher baseflow constituting a dilution of this groundwater signal by shallower groundwater/soil water contributions to stream flow. Nine of the 12 sites do have some type of contribution from WWTP effluent. Therefore, these relationships between conductivity and discharge at these sites may simply reflect a dilution of WWTP effluent during wetter periods (i.e., greater baseflow discharge).

There were more pronounced differences between the WOH and EOH sites for suspended solids relative to 2003 values. TSS averaged 4.6 mg/L for the EOH sites (range 1.0 to 25.5 mg/L) and 3.0 mg/L (range 0.6 to 7.2 mg/L) for the WOH sites (Fig. 3.4). VSS averaged 1.6 mg/L (range 0.3 to 9.9 mg/L) for the EOH and 0.9 mg/L (range 0.3 to 3.0 mg/L) for the WOH sites (Fig. 3.4). The organic matter content was 36.5% for the EOH (range 22.7 to 60.6%) and 33.4% for the WOH (range 15.2 to 53.0%) (Fig. 3.5). The mean values for TSS and VSS are greater than 2003 values for both EOH and WOH sites, which is evident in individual site comparison between years for both fractions (Fig. 3.6). Based on regression analysis of TSS versus discharge, there is a positive, though not statistically significant relationship between TSS and discharge for 6 of the 12 sites (site numbers 3, 6, 9, 10, 15, and 55) having 5 years of data (data are not provided). None of the regressions were significant at an α = 0.05; R2 values for these 6 regressions ranged from 0.34 to 0.71. Positive relationships between VSS and stream discharge were only found for 3 of the 12 sites (site numbers 9, 15, 55): R2 values were 0.93 (significant at α = 0.05), 0.39, and 0.47, respectively. The organic matter content values are lower than 2003 values for both regions.

Following observations from previous years, the primary factor in separating the stream sites based on relative baseflow chemistry is total ionic composition as determined by separate Principal Components Analyses (PCA) on baseflow major ion, nutrient, and seston concentrations for each region (Figs. 3.7 and 3.8) including both Phase I and Phase II sites. This is especially the case for WOH sites (Fig. 3.7) with the first axis of the PCA explaining 54% of the variability in the data and having all the analytes essentially loading along this first axis. Along the second axis of the WOH PCA, sites are somewhat separated along an acidity gradient defined by increasing pH in the positive direction. Separation between watershed groupings is also evident in the WOH PCA with West Branch Delaware sites plotting in line along the first axis defining higher ionic concentrations and Neversink/Rondout and Esopus sites plotting in the opposite direction. The pH gradient separates Schoharie sites having relatively higher pH values from the lower pH values of the Neversink/Rondout sites. The EOH PCA (Fig. 3.8) dramatically points to the difference in stream chemistry for the three sites (all Phase I sites) that have large effluent volume inputs upstream of the sampling sites relative to the other EOH study sites. These three sites, Hallocks Mill Brook (49), Secor Brook (43), and an unnamed tributary to the New Croton Reservoir near Lake Purdy (58) are being separated from the rest of the sites by high NH4-N, Organic N, and Total Dissolved N concentrations. EOH sites along the second PCA axis are being separated along an acidity gradient in one direction and particulates in the other direction. The acidity gradient is being defined by increasing pH and alkalinity, while the particulate gradient is being defined by increasing TSS/VSS and also increasing particulate N and P. Unlike the WOH PCA result, there is no clear watershed separation evident in the EOH PCA.

- 37 - CHAPTER 3 – NUTRIENTS IN TRANSPORT NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Watershed condition and baseflow chemistry. With two years of data for the Phase II sites, we have begun to examine relationships between baseflow chemistry and watershed condition using data collected across the entire project. The Phase II sites were not specifically selected to reflect a particular type of watershed condition. Yet, when combined with data collected in Phase I of the project and related to watershed conditions, definite shifts occurred in the set of significant predictors of baseflow chemistry for certain analytes (Table 3.1). Results shown in Table 3.1 reflect separate multiple linear regressions (MLR) run using just Phase I data versus Phase I and II data for selected analytes (conductivity, particulate and dissolved P, total and volatile suspend solids) against a suite of land use/cover percentages and permitted discharger annual effluent volumes (WWTP).

The conductivity MLR models for WOH sites are an example of the difference in land use/cover predictors that result depending upon which set of data are used in the models. Both models are strong, having similar R2 values (0.78 using Phase I data, 0.76 using Phase I and II data). When using just Phase I sites though, %grass was the dominant predictor based on the partial R2 values, while %residential area was the dominant predictor when using Phase I and II data. Likewise, the particulate P MLR results using EOH sites, show WWTP as the dominant predictor of particulate P concentrations using Phase I data, with %wetland the dominant predictor when using all data. Caution should be exercised in interpreting these results for a number of reasons: only 2 of the 3 years of Phase II data have been collected; the difference in number of observations used to generate mean chemistry values (n = 1, 2, 3, or 5); potential spatial autocorrelation issues within the land use/cover data (King et al. 2005) and, no screening for the existence of potential outlier values exerting undue influence on the MLR results was carried out at this time. Phase II site selection was in part governed by a desire to move further upstream in the watershed from our existing Phase I sites. Therefore, the differences in MLR results presented here may also reflect the issue of distance between a particular watershed land use/cover and a stream sampling point (King et al. 2005). It should also be pointed out that certain MLR results, for example dissolved P, show consistent results regardless of what data period is used. Within both EOH and WOH results, the dominant predictor of dissolved P is the same.

The subset of analyte MLR models presented, suggest that the primary watershed characteristics influencing ionic and nutrient baseflow chemistry were often anthropogenic in nature as noted in earlier annual project reports. Conductivity (as a surrogate of ionic chemistry) along with particulate and dissolved P have as their primary predictors either %residential, WWTP, or %grass. Note that %grass is thought to comprise agricultural activity such as hay fields and pasture areas especially in the WOH region. Notable exceptions occur within the particulate P results where %wetland is the dominant predictor for EOH, Phase I and II data, and %brush the dominant predictor for the WOH, Phase I and II data. The suspended and volatile solids MLR models deviate even further from the notion of anthropogenic influences governing baseflow concentrations, with %mixed forest (MXDFR in Table 3.1) the primary predictor in 5 of the 8 models shown in Table 3.1, regardless of dataset used or region. The caveats noted in the last paragraph regarding interpretation of the MLR results also apply to these observations of the dominant watershed influences on baseflow chemistry.

Stormflow inorganic and nutrient water quality

Two storm samples, one high turbidity (HT) and one high flow (HF), were collected for each storm sampled at the Neversink and W. Br. Delaware River sites. For the Kisco River site, three

- 38 - CHAPTER 3 – NUTRIENTS IN TRANSPORT NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

samples (1 HT and 2 HF) were collected during the April storm and four samples (2 HT, 2HF) were collected during the September storm. The typical Kisco River hydrograph (see Fig. 2.8 in Chapter 2) has two peaks; one that occurs early in the hydrograph representing a flashier response, and a secondary, more drawn-out peak response later in the hydrograph. Given these dual-peak hydrographs, we though it important to collect at least two peak flow samples, one for each peak, and if possible, two high turbidity samples to coincide with the two peak-flow samples.

Instantaneous baseflow and stormflow fluxes, as the product of measured concentration and instantaneous discharge, are provided for selected ions (Fig. 3.9), particulate and dissolved N and P (Fig. 3.10) and seston (both TSS and VSS) along with total and particulate organic carbon (OC and POC, respectively; Fig. 3.11). Particulate and dissolved N, particulate P, POC and OC are all derived values. POC is calculated as being 45% of the measured VSS concentration and OC as the sum of POC and dissolved organic carbon (DOC) concentrations. Particulate N (PN) is the difference between total kjeldahl N (TKN) and soluble kjeldahl N (SKN); total dissolved N (TDN) is the sum of SKN and NO3-N. Particulate P (PP) is the difference between total P and total dissolved P. Results in these figures are provided for all storm samples collected to date.

Almost no overlap exists between the ion flux values under baseflow conditions relative to stormflow for the forested watershed (Neversink River, 29) while overlap occurs for both the agricultural (W. Br. Delaware River, 6) and urbanized (Kisco River, 55) watersheds (Fig. 3.9). This suggests that even under baseflow, appreciable transport of dissolved solids can occur on these human-influenced watersheds relative to storm transport. The generally higher ion fluxes during storms on the Kisco River suggest that even with taking into account the generally higher dissolved solid concentrations for EOH sites versus WOH sites (see Fig. 3.1), this relatively small, yet urbanized, watershed is transporting considerable amount of dissolved material downstream, relative to the other two larger watersheds.

Unlike the ion fluxes, much less overlap exists between the baseflow and storm fluxes for N, P, seston, and organic carbon (Figs. 3.10, 3.11). However, certain baseflow fluxes, namely TDN, TDP, PP, and OC measured at the W. Br. Delaware River site fell within the range of values measured during storms more often than the fluxes measured at the other two sites. Baseflow transport of nutrients and organic material during baseflow on the W. Br. Delaware River, and by extension to other agriculturally-influenced watersheds in the study region, can not be considered insignificant relative to storm transport. The storm fluxes for nutrients, seston, and organic carbon also demonstrate that appreciable downstream transport can occur in forested watersheds, based on the Neversink River results, relative to agricultural and urbanized watersheds.

Literature Cited

American Public Health Association (APHA). 1992. Standard methods for the examination of water and wastewater. 18th edition. American Water Works Association, Water Environment Federation, Washington, DC. pp. 2-44. King, R. S., M. E. Baker, D. F. Whigham, D. E. Weller, T. E. Jordan, P. F. Kazyak, and M. K. Hurd. 2005. Spatial considerations for linking watershed land cover to ecological indicators in streams. Ecological Applications 15:137-153.

- 39 - CHAPTER 3 – NUTRIENTS IN TRANSPORT NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

2 R 0.33 0.64 0.63 0.15 0.74

0.25 (+)0.25 0.34 (+)0.34 yte and a

0.10 (+) 0.36 (+) 0.10 0.13 0.28 0.25 0.46 (+) 0.11 0.41 (+) 0.16

0.08 (-) (-) 0.08 0.94 0.15 (-) 0.13 (-) 0.35 (+)0.35 (+)0.14 0.25 (+)0.25 MXDFR

oncentration values (n=1, 2, 3, values provided for those variables 2 DECFR CONFR cover variables and permitted and permitted cover variables

(-) 0.15 (+) 0.06 0.54 MXDBR

value for a given model). Separate models were 0.76 2 0.14 (+) 0.21 (+) 0.78 (+) 0.21 (+) 0.14 (+) 0.17 0.12 (+) (+)0.07 (-) 0.17 0.80

0.19 (+) 0.19 GRASS BRUSH 0.52 (+)0.52 (+)0.69 (+)0.51 0.42 (+)0.42

FARMD

ed analytes against land use/ ed analytes ocess was used with partial R OH n=53, WOH n=57) using mean c OH indicate the largest partial R 0.11 (-) 0.53 (-) 0.11

gns indicate increasing or decreasing, respectively, relationships between an anal between relationships increasing or decreasing, respectively, gns indicate (WWTP). A stepwise selection pr 0.09 (-) 0.14 (-) 0.57 0.52 0.39 (-) (-) 0.14 0.05 (-) (+) 0.09 0.08

0.13 (+) 0.17 (+) 0.17 (+) 0.13

0.03 (+) 0.17 (-) (-) 0.17 (+) 0.03 0.21 (+) 0.12 (+) 0.12 (+) (-) 0.21 (+) 0.10 0.11 = 0.05) predictors (bolded values α 0.34 (+) 0.29 (+) 0.47 (+) 0.35 (+) 0.30 (+)

0.30 (+) 0.30 (+) 0.16 (+) 0.17 redictor variable. Results of multiple linear regressions (MLR) run for select run for (MLR) regressions linear multiple Results of p articular Particulate Phosphorus (+) Phosphorus 0.11 EOH EOH Region Period (+) (+)I Conductivity I&II Phase I %Urban/Suburban Phase0.13 I&II 0.05 WOH EOH EOH I Phase 0.10 (+) 0.08 (+) Phase I&II WOH WOH (+) (+) WOH WOH Particulate I WWTP RESID COMML %Agriculture I Phase INDST TRANS OTHERORCHD I&II CRPLD I %Herbaceous/Brush Phase0.04 I&II 0.06Phosphorus EOH EOH Phase %Wetland 0.09 (+) %Forest Phase WOH WOH Solids %Water Dissolved I I Phase I&II Phase I&II Suspended (+) EOH EOH Phase 0.09 (+) Phase Solids Total 0.08 Phase Phase EOH SuspendedEOH I Phase I Phase I&II I&II WOH WOH Volatile Phase Phase Phase Phase I&II WOH or 5 depending upon site). The ‘+’ and ‘-‘ si or 5 depending upon site). The ‘+’ and Table 3.1. p discharger annual effluent volumes run for Phase I data (EOH n=30, WOH n=30) and I&II (E found to be significant (

- 40 - CHAPTER 3 – NUTRIENTS IN TRANSPORT NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

8000 TCS Cations WBC EMC MUS

CA KS C 6000 MG NA K

4000 Concentration (ueq/L) 2000 EBD SCH WBD ESP

NRD

0 3 6 9 10 15 23 26 29 34 46 52 55 101 102 103 104 105 106 151 153 107 108 109 110 111 112 113 114 160 115 116 117 118 119 120 121 159 122 123 124 125 126 129 132 150 127 133 134 139 130 131 137 138 143 145 146 149 140 141 142 147 148

8000 Anions WBC TCS MUS CL KS C 6000 SO4 ALKL EMC NO3N NH3N SRP 4000 Concentration (ueq/L) Concentration 2000 WBD SCH EBD ESP NRD

Figure 3.1. Major cation and anion summaries for all 62 stream monitoring sites for the 2004 sampling year. Subwatershed designations are: Neversink River and Rondout Creek (NRD); Esopus Creek (ESP); Schoharie Creek (SCH); E. Br. Delaware River (EBD); and W. Br. Delaware River (WBD). E. & M. Br. Croton R. (EMC); W. Br.Croton R. (WBC); Muscoot R. (MUC); Titicus, Cross, and Stone Hill Rivers (TCS); and Kensico Resv. & Lower New Croton Resv. (KSC).

- 41 - CHAPTER 3 – NUTRIENTS IN TRANSPORT NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

3500 WBD SCH 2004 3000 2003

EBD 2500 ESP

2000

1500 Ionic sum (ueq/L) sum Ionic 1000 NRD

500 WOH 0 3 6 9 10 15 23 26 29 101 102 103 104 105 106 151 153 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123

18000 TCS 2004 MUS 2003 15000 WBC KSC EMC 12000

9000

Ionic sum (ueq/L) 6000

3000

EOH 0 34 46 52 55 124 125 126 129 132 150 127 133 134 139 130 131 137 138 143 145 146 149 140 141 142 147 148

Figure 3.2. Comparison of 2003 (open circles) to 2004 (closed circles) ionic sums for sites WOH (top) and EOH (bottom). Ionic sums include all ions shown in Figure 3.1 plus H+ and OH- . See Fig. 3.1 for subwatershed designations.

- 42 - CHAPTER 3 – NUTRIENTS IN TRANSPORT NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005 Conductivity (µS/cm)

3 2 Discharge (m /s/km ) Figure 3.3a. Conductivity versus stream discharge relationships for 6 WOH sites having 5 years of data (2000-2004). Rsq values > 0.79 significant at an α = 0.05. See Tables 2.1 and 2.2 for stream names corresponding to station number.

- 43 - CHAPTER 3 – NUTRIENTS IN TRANSPORT NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005 Conductivity (µS/cm)

3 2 Discharge (m /s/km ) Figure 3.3b. Conductivity versus stream discharge relationships for 2 WOH and 4 EOH sites having 5 years of data (2000-2004). Rsq values > 0.79 significant at an α = 0.05. See Tables 2.1 and 2.2 for stream names corresponding to station number.

- 44 - CHAPTER 3 – NUTRIENTS IN TRANSPORT NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

12 West of Hudson 10

4 Suspended Solids (mg/L) 2

3 6 9 1 2 5 3 6 9 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 1 2 0 1 2 3 1 3 9 0 - - 1 2 2 2 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 - - 2 2 2 2 5 5 5 6 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 9 9 1 1 1 1 1 1 1 1 1 1 1 1 1 1

30 East of Hudson

10 Suspended Solids(mg/L) 5

4 6 2 1 2 4 5 6 7 9 0 1 2 3 4 7 8 9 0 1 2 3 5 6 7 8 9 0 3 4 5 - - 2 2 2 2 2 3 3 3 3 3 3 3 3 4 4 4 4 4 4 4 4 4 5 5 5 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 5 5

TSS Site Number VSS

Figure 3.4. Whole sample Total Suspended Solids (TSS) and Volatile Suspended Solids (VSS) from samples collected at EOH and WOH sites in 2004. Error bars (standard deviation) reflect variability among three replicate samples collected per site.

- 45 - CHAPTER 3 – NUTRIENTS IN TRANSPORT NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

70 West of Hudson 60

20 Percent Organic Matter

3 6 9 1 2 5 3 6 9 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 1 2 0 1 2 3 1 3 9 0 - - 1 2 2 2 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 - - 2 2 2 2 5 5 5 6 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 9 9 1 1 1 1 1 1 1 1 1 1 1 1 1 1 70 Eas t of Huds on 60

20 Percent Organic Matter Organic Percent

4 6 2 1 2 4 5 6 7 9 0 1 2 3 4 7 8 9 0 1 2 3 5 6 7 8 9 0 3 4 5 - - 2 2 2 2 2 3 3 3 3 3 3 3 3 4 4 4 4 4 4 4 4 4 5 5 5 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 5 5

Site Number Figure 3.5. Percent organic matter (as VSS/TSS*100) for the 2004 sampling effort at all EOH and WOH sites. Error bars (standard deviation) reflect variability among three replicate samples collected per site.

- 46 - CHAPTER 3 – NUTRIENTS IN TRANSPORT NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005 10 SCH WOH 8 WBD EBD ESP

TSS (mg/L)TSS NRD 2

0 3 6 9 10 15 23 26 29 101 102 103 104 105 106 151 153 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 30 25 WBC TCS EOH 20 15 MUS EMC KSC 10 TSS (mg/L) 5 0 34 46 52 55 124 125 126 129 132 150 127 133 134 139 130 131 137 138 143 145 146 149 140 141 142 147 148

4 SCH WBD ESP WOH 3 EBD

VSS (mg/L) 1 NRD

0 3 6 9 10 15 23 26 29 101 102 103 104 105 106 151 153 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 12 10 WBC TCS EOH 8 6 MUS EMC KSC 4 VSS (mg/L) 2 0 34 46 52 55 124 125 126 129 132 150 127 133 134 139 130 131 137 138 143 145 146 149 140 141 142 147 148

2004 2003

Figure 3.6. Comparison of 2003 (open circles) to 2004 (closed circles) TSS (top two graphs) and VSS (bottom two graphs) for sites WOH and EOH. See Fig. 3.1 for subwatershed designations.

- 47 - CHAPTER 3 – NUTRIENTS IN TRANSPORT NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

pH CondNa Cl Ca PN

PP TSS Mg K DON NH4 SO4 TDPVSS Alk TDN 3 NO B 5 20 21

18 19 17

4 25 14 13 16 23 22 11 117 109 24 10 119 1 26 12 108 15 120 116 118 0 101 27 153 159 121 30 28 8 56107 9 7 110 104 102 2 29 105 160 112114 PCA Factor 2 - 12.3 % 106 115 122 3 151 123 113 111

103

-4 -8 0 7 PCA Factor 1 - 53.8 %

West Branch Delaware River East Branch Delaware River Schoharie Creek Esopus Creek Neversink River & Rondout Creek Figure 3.7. Loadings (A) and the first and second principal component scores (results separate by watershed) [B] from a PCA of inorganic chemistry for the 57 WOH sites. All input variables were log –transformed (except for pH). Site variability explained by the first two PC scores is provided in the axis labels. Subwatershed designations are: Neversink River and Rondout Creek; Esopus Creek; Schoharie Creek; E. Br. Delaware River; and W. Br. Delaware River.

- 48 - CHAPTER 3 – NUTRIENTS IN TRANSPORT NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

VSS PP A TSS NH PN 4 DON

TDN

TDP K Na SOCl4 NO3

Cond Alk Mg

Ca pH

5 49

43 38

36 41

44 127 150 143 47 149 125 139 57 34 133 142 138 51 42 146 45 40 132 58 0 147 131 52 46 126 33 54 50 55 130 31 35 53 134 129 137 60 145 59 39 148 56 48 37 PCA Factor 2 - 18.0 % 32

124 140 141

-4 -8 0 7 PCA Factor 1 - 44.7 %

East & Middle Branch Croton River West Branch Croton River Titicus Cross Stone Hill Kensico and South of Croton Res. Muscoot and North of Croton Res. Figure 3.8. Loadings (A) and the first and second principal component scores (results separate by watershed) [B] from a PCA of inorganic chemistry for the 53 EOH sites. All input variables were log –transformed (except for pH). Site variability explained by the first two PC scores is provided in the axis labels. Subwatershed designations are: Kenisco Resv and Lower New Croton Resv; Titicus , Cross, and Stone Hill Rivers; Muscoot River; E. and M. Br. Croton Rivers; and W. Br. Croton River.

- 49 - CHAPTER 3 – NUTRIENTS IN TRANSPORT NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Figure 3.9. Instantaneous fluxes, as the product of concentration and stream discharge, of selected ions sampled during stormflows and under baseflow (BF) at three sites in the NY watershed from 2000 through 2004. At least two stormflow samples collected for each of 5 (sites 29 and 55) or 6 (site 6) total storms representing the peak turbidity (HT) and peak flow (HF) points on the storm hydrograph. The vertical lines connect the minimum and maximum values and the horizontal dash is the mean.

- 50 - CHAPTER 3 – NUTRIENTS IN TRANSPORT NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Figure 3.10. Instantaneous fluxes, as the product of concentration and stream discharge, of total dissolved and particulate N and P sampled during stormflows and under baseflow (BF) at three sites in the NY watershed from 2000 through 2004. At least two stormflow samples collected for each of 5 (sites 29 and 55) or 6 (site 6) total storms representing the peak turbidity (HT) and peak flow (HF) points on the storm hydrograph. The vertical lines connect the minimum and maximum values and the horizontal dash is the mean.

- 51 - CHAPTER 3 – NUTRIENTS IN TRANSPORT NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Figure 3.11. Instantaneous fluxes, as the product of concentration and stream discharge, of seston (TSS and VSS), total organic carbon (OC), and particulate organic carbon (POC) sampled during stormflows and under baseflow (BF) at three sites in the NY watershed from 2000 through 2004. At least two stormflow samples collected for each of 5 (sites 29 and 55) or 6 (site 6) total storms representing the peak turbidity (HT) and peak flow (HF) points on the storm hydrograph. The vertical lines connect the minimum and maximum values and the horizontal dash is the mean.

- 52 - CHAPTER 3 – NUTRIENTS IN TRANSPORT NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Chapter 4 – Molecular Tracers

Introduction

Degradation of water quality can occur from a variety of point and non-point sources of natural and anthropogenic contamination, such as sewage (from septic leakage or waste water treatment plants (WWTP) effluent), atmospheric deposition, road and agricultural runoff, and wildlife. The range of contaminants includes nutrients, heavy metals, pesticides, other toxic organic compounds, and pathogens. In order to best maintain the quality of drinking water resources, targeted efforts to reduce or eliminate primary contamination sources first require the accurate identification and quantification of all contaminant sources that contribute to water quality impairment. The use of molecular tracers to identify sources of contaminants is an emerging technique that qualitatively links the presence of components unique to these sources with contaminants of concern (Leeming et al. 1996, Simonich et al. 2000, Standley et al. 2000, Kolpin et al. 2002, Buerge et al. 2003).

We have chosen a suite of 25 organic compounds that act as robust proxies for a variety of contamination sources (Table 4.1). These compounds include twelve polycyclic aromatic hydrocarbons (PAH), two fragrance materials (FM), caffeine (CAF) and seven fecal steroids (FS). Polycyclic aromatic hydrocarbons are found in raw and refined petroleum and coal products and are also formed during the combustion of vegetation, wood, waste, coal and petroleum. Thus PAHs have both natural and anthropogenic sources. The compounds that we quantify here are fluorene (FLU), phenanthrene (PHE), anthracene (ANT), 2-methyl phenanthrene (2MP), 1-methyl phenanthrene (1MP), fluoranthene (FLR), pyrene (PYR), benzo(a)anthracene (BAA), chrysene (CHR), benzo(b)fluoranthene (BBF), benzo(k)fluoranthene (BKF), and benzo(a)pyrene (BAP). Fragrance materials are anthropogenic compounds used in a variety of consumer products such as soaps, detergents and lotions. Thus, FMs enter the environment primarily through greywater sewage (Simonich et al. 2000). The compounds that we quantify here are tonalide (7-acetyl-1,1,3,4,4,6,-hexamethyl-1,2,3,4-tetrahydronaphthalene, AHTN) and galaxolide (1,3,4,6,7,8-hexahydro-4,6,6,7,8,8-hexamethylcyclopenta[γ]-2- benzopyran, HHCB). Both AHTN and HHCB are non-biodegradable, making them particularly suited for tracers studies (Simonich et al. 2002). Caffeine is a natural compound that occurs in certain tropical plants, including tea and coffee, and is added to numerous food products and pharmaceuticals. In temperate climates, the primary source of caffeine to watersheds is via the urine of those who consume caffeine-containing products (Buerge et al. 2003). Fecal steroids are natural compounds that are produced in the intestines of birds and mammals. Ratios of certain steroids to others allow for the discrimination between human fecal material and that of other animals (Leeming et al. 1996). The steroids we quantify for this study are coprostanol (5β- cholestan-3β-ol, bCOP), epicoprostanol (5β-cholestan-3α-ol, EPI), cholesterol (cholest-5-en-3β- ol, CHOL), cholestanol (5α-cholestan-3β-ol, aCOP), ), coprostanone (5β-cholestan-3-one, bONE), cholestanone (5α-cholestan-3-one, aONE), 24-ethyl-coprostanol (24-ethyl-5β-cholestan- 3β-ol, eCOP), 24-ethyl-epicoprostanol (24-ethyl-5β-cholestan-3α-ol, eEPI), 24-ethyl-cholesterol (24-ethyl-cholest-5-en-3β-ol, eCHO), and 24-ethylcholestanol (24-ethyl-5α-cholestan-3β-ol, SNOL).

- 53 - CHAPTER 4 – MOLECULAR TRACERS NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Research Task

For Phase II of this project, molecular tracer concentrations were analyzed at each of 60 stream sampling sites (10 continuing from Phase I, for a total of 110 sites) within New York City drinking water source watersheds (see Chapt. 2 and Tables 2.1-2.4 and Figs. 2.1-2.2) during summer baseflow conditions to determine the relative influence of contaminant sources on water quality. In addition, molecular tracers were analyzed in samples collected during winter baseflow conditions at 28 sites targeted to provide information on background levels, winter recreation area influences, and the effect of low temperatures on sewage treatment efficiency. However, as we have done in previous years (with the exception of the Phase I report), in this year 5 report we present winter baseflow data from project year 4 and this winter’s data will be presented in next year’s report. Storm flow samples were collected at three "integrative" sampling stations to determine changes in source composition with storm runoff.

The molecular tracer investigation targets two of the projects primary objectives, as listed in the Quality Assurance Project Plan (QAPP). First, molecular tracers are designed to act as indicators for evaluating the occurrence and source of selected aquatic chemical contaminants. Second, the three-year data sets will be utilized as a baseline for assessing changes in water quality in response to changes in land use and best management practice (BMP) implementations.

Methods

Detailed descriptions of our field and laboratory methods are provided in the Quality Assurance Project Plan (QAPP) for Project Year 5 and the Standard Operating Procedures attached therein. What follows is a brief description of these methods.

Field

During summer and winter baseflow collections, 8 L water samples were collected for tracer analysis in pre-cleaned glass jars. At the same time, samples were collected for all other baseflow analyses (i.e. nutrients, major ions, TSS, DOC, etc. see Chapt. 3 & 6). Storm flow samples were collected using ISCO samplers fitted with pre-cleaned glass receiving bottles. The ISCO samplers were set to begin sampling with a small rise (approximately 10 to 15 cm) of stream water and took two 1-liter samples every hour for up to 12 hours. A subset of two paired 1-L samples for each stream – representing high flow (HF) and high turbidity (HT) – were chosen for analysis. Selection of the two samples were based on examining the storm hydrograph available from a nearby USGS gauging station collecting near real-time data or from a co-located stage recorder (In-Situ, Inc., Mini-Troll). Peak particulate concentrations were determined visually. High flow samples were removed from the ISCO apparatus within 18 hours of collection and then handled in the same manner as baseflow water samples. Water samples were stored in a cool and dark place and extracted within 7 days.

All glass sampling equipment and sample jars were precleaned of organic contaminants baking in a kiln at 480oC for 4 hours. Metal and Teflon sampling equipment was cleaned with solvent rinses, as was any field equipment that needed to be reused between sites. The probe and collection tubing on ISCO samplers were cleaned weekly with 0.1 N HCl, 0.1 N NaOH, and deionized water.

- 54 - CHAPTER 4 – MOLECULAR TRACERS NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Field blanks and duplicates were each collected at three sites during summer baseflow sampling (5% of sites) and two sites during winter baseflow sampling. One set of field blanks and duplicates was collected through the ISCO sampler during stormflow collections.

Laboratory

Molecular tracers were were extracted from all samples by liquid-solid extraction onto an Empore™ disk, using protocols similar to EPA approved alternate test method 608 ATM 3M0222 or to EPA Method 3535. In 2003, we modified our protocol slightly over that used previously in order to increase reproducibility, recovery and sensitivity, as was described in an addendum to our Year 4 QAPP. As a whole, our updated set of SOPs for extraction and GC-MS analysis describe a method very similar to EPA method 8270..

In brief, sample water was filtered through a glass fiber filter stacked on top of an Empore™ C- 18 disk. Particulate tracer compounds were extracted from the filter by sonic extraction and dissolved tracers were eluted from the Empore disk with solvents. Surrogate recovery standards – perdeuterated phenanthrene (PHE-D10), perdeuterated chrysene (CHR-D12), perdeuterated perylene (PER-D12), perdeuterated caffeine (CAF-D9) and perdeuterated cholesterol (CHO-D6) – were added to the surface of both the filter and the disk, after they were separated but prior to extraction. Dissolved and particulate extracts were then back-extracted in a separatory funnel with an aqueous salt solution to remove impurities, mixed with anhydrous sodium sulfate to remove moisture, rotoevaporated, and transferred to auto-injector vials. Samples were gently dried under a stream of nitrogen, redissolved in 15 µL pyridine and spiked with 5 µL internal standard solution (25-ng/µL in each of p-terphenyl-d14 and 5α-cholestane in pyridine). In order to analyze fecal sterols, which contain alcohol groups, samples were then derivatized by purging sealed vials with N2, adding 15 µL of BSTFA (N,O-bis(trimethylsilyl)trifluoroacetamide) with 1% TMCS (Trimethylchlorosilane), and heating to 70°C for 30 minutes in an aluminum heating block. These derivitized sample extracts were analyzed for each of the molecular tracers compounds by capillary gas chromatography – mass spectrometry (GC/MS) in selective ion monitoring (SIM) mode, using a J&W DB1701 column (30 m, 0.25 mm i.d., 250 µm coating) on an Agilent 6890 series GC interfaced with a 5973n series MSD.

Laboratory blanks and duplicates, and matrix spike samples, were prepared in conjunction with all sites having field blanks and duplicates (3 sites during summer baseflow sampling and two sites during winter baseflow sampling).

Quantification

As described in our Phase I report, we now quantify molecular tracer data with an automated data quantification system. In brief, after confirmation by the analyst, compound peak areas for standards and samples are exported from the Agilent GC-MS “ChemStation” chromatography software directly into our central server. We then manipulate this raw data within the server with SAS-based scripts to produce the final data. Thus, decisions – regarding how to fit the calibration curve, when to drop outlying standards, whether or not peak identity is adequately confirmed, etc. – were all made uniformly for 2004 data using the same objective criteria used in the previous four years.

- 55 - CHAPTER 4 – MOLECULAR TRACERS NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

All data presented here are surrogate-corrected with the extraction recoveries measured within each sample for each surrogate standard, which were associated with tracer compounds as follows (see Table 4.1): perdeuterated phenanthrene (FLU, PHE, ANT), perdeuterated chrysene (FLR, PYR, BAA, CHR, HHCB, AHTN), perdeuterated perylene (BBF, BKF, BAP), perdeuterated caffeine (CAF) and perdeuterated cholesterol (fecal steroids). The average recovery of perdeuterated phenanthrene and perdeuterated chrysene is used for 2MP and 1MP. These associations between target and surrogate compounds have been slightly modified from previous years, based on analysis of our matrix spike data. However, these assignments have now all been retroactively assigned to all project data, from 2000 to present.

Results

Summer Baseflow

Total concentrations of PAHs in 2004 summer base flow samples varied by up to three orders of magnitude between sites (Fig 4.1, 4.2, 4.3), whereas interannual variability within a site was generally less than one order of magnitude . For the ten sites that were also sampled during the first three years of this project (Phase I), concentrations measured in 2003 and 2004 (Phase II) appear generally lower but only slightly and not outside the range of inter-annual variability within each phase. We thus consider means from the two Phases to be comparable. In general, total PAH concentrations were higher at east of Hudson (EOH) sites relative to west of Hudson (WOH) sites, such that 41 of the 53 EOH sites were in top 50% of most contaminated sites (Fig. 4.1).

Molecular tracers do not necessarily need to be toxic compounds. However, ten of the twelve PAHs analyzed for this project are listed by the EPA as Priority Toxic Pollutants and five of these are known human carcinogens (EPA 2002a, EPA 2002b). These five most toxic PAHs (BAA, CHR, BAP, BBF, BKF) have been given exceptionally low “National Recommended Water Quality Criteria for Human Health” of 0.0038 µg/L for the consumption of the water or 0.018 µg/L for the consumption of organisms living in the water (EPA 2002b). Similarly, NY State Department of Environmental Conservation (NYSDEC) has set water quality guidance values of 0.002 µg/L for these same compounds (BAA, CHR, BAP, BBF, BKF) for ambient waters directly feeding water supplies, 0.0012 µg/L for BAP in waters used for fish consumption, and 0.03 µg/L for BAA as a flag of chronic toxicity to aquatic life [NYSDEC, 1998 #2724]. Although these are non-regulatory guidance values that are not enforceable, and none of the sites are near water supply intakes, these guidance values are useful to place measured PAH concentrations in the context of potential human and ecosystem toxicity. When compared to the more conservative EPA guidance values, measured concentrations of one or more of these five compounds exceeded the lower EPA guidance limit of 0.0038 µg/L in 58 summer baseflow samples out of total of 298 collected, and 10 out of 298 samples for the upper guidance limit of 0.018 µg/L. At on average 10 sites per compound, geometric mean concentrations exceeded the lower guidance limit. Figures 4.2 and 4.3 show data for BAA and BKF; concentrations of the other three compounds show similar trends. For example, the average number of compounds to exceed a limit in a sample was 3.7 out of 5 (i.e. If one compound exceeded a limit, so did a majority of the other five).

- 56 - CHAPTER 4 – MOLECULAR TRACERS NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

The ratios of certain PAH compounds to others have been used to identify both petroleum sources, such as spills of kerosene, diesel oil, lubricating oil and crude oil (Yunker et al. 2002, Zakaria et al. 2002), and combustion sources, such as automotive exhaust, smelter emissions, coal burning emissions and wood smoke (Dickhut et al. 2000, Yunker et al. 2002). Two of the most useful of these ratios are that of ANT/(ANT+PHE) and FLR/(FLR+PYR), where low values suggest petroleum sources and high values combustion sources (Fig. 4.4, 4.5). The petroleum/combustion transition point for ANT/(ANT+PHE) is considered to be 0.1. For FLR/(FLR+PYR) the transition is less clear, and values between 0.4 and 0.5 are considered to indicate mixed sources (Yunker et al. 2002). Another useful source indicator is the ratio of high molecular weight (HMW) PAH compounds to low molecular weight (LMW) PAH compounds (H/LPAH)(Fig. 4.3c). In general, LMW, volatile PAHs strongly predominate over HMW PAHs in crude oil and most refined petroleum products (with the exception of asphalt) (Zakaria et al. 2002), whereas HMW PAHs are the primary constituents of soot (Countway et al. 2003). Ratios above approximately 0.5 appear to indicate combustion sources.

In general, PAH source indicator ratios were high. For our 2004 data, only five sites exhibited ANT/(ANT+PHE) ratios less than 0.1 (Fig. 4.4) and only three sites exhibited FLR/(FLR+PYR) ratios less than 0.4 (Fig. 4.5). H/LPAH ratios showed a similar pattern, with 4 of 61 sites having values below 0.5 (Fig. 4.6). For EOH sites, H/LPAH ratios were generally higher in 2004 than in 2003.

Caffeine concentrations spanned almost four orders of magnitude between sites (Fig. 4.7). Concentrations were generally higher in 2004 than in 2003, especially EOH. Fragrance materials showed generally uniform concentrations between sites, with the exception of some sites with notably high concentrations, and interannual variability within a site was minimal (Fig. 4.8).

Total fecal steroid (FS) concentrations in 2004 summer baseflow were in the range of Phase I values, unlike 2003 concentrations which were substantially lower (Fig. 4.9). Coprostanol concentrations showed a similar pattern to that of total fecal steroids (Fig. 4.10). The primary exception is that concentrations of bCOP ranged over five orders of magnitude, whereas total fecal steroids only ranged two orders of magnitude. Because bCOP is the dominant FS found for humans and is a minor FS component for all other animals (Leeming et al. 1996), concentrations of bCOP in surface waters tend to directly correlate with human sewage inputs (Leeming and Nichols 1996). Thus, linear relationships between bCOP and bacterial indicators of human sewage (fecal streptococci and thermotolerant coliforms), allow for the translation of bCOP concentrations into fecal bacterial counts. Using the relationships in Leeming and Nichols (1996), four EOH sites (139, 143, 138 and 55) exceeded 0.1 µg/L of bCOP corresponding to 35 enterococci (a subset of fecal streptococci) and 300 thermotolerant coliforms per 100 mL of water.

Similar to PAHs, ratios of fecal steroids can help distinguish potential sources (Fig. 4.11). The ratio bCOP/(bCOP+aCOP) has been used to identify sites where fecal material from humans relative to that from livestock and wildlife dominates (Grimalt et al. 1990, O'Leary et al. 1999). O'Leary et al. (1999) suggested that values of this ratio >0.3 were a clear indication of human fecal material in greater relative abundance compared to non-human fecal material and values between 0.2 and 0.3 suggest mixed sources. The ratio bCOP/(bCOP+EPI) has also been used to

- 57 - CHAPTER 4 – MOLECULAR TRACERS NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

distinguish human sewage from other fecal sources, with values > 0.5 attributed only to humans (Leeming et al. 1998). Last, because cholesterol is widely found in all organisms, bCOP/CHO has also been used to trace human sewage (Mudge and Seguel 1999).

For our 2004 data, values of fecal steroid ratios were substantially lower than in 2003 for many sites but not all sites. Values of bCOP/(bCOP+aCOP) in particular (Fig. 4.11) show that some sites exhibited substantial interannual variability while others exhibited very stable ratios from year to year. 28 sites have mean bCOP/(bCOP+aCOP) ratios over 0.2. Ratios of bCOP/CHO mirror these patterns.

Winter Baseflow

The concentrations and ratios of PAHs measured during winter baseflow exhibit similar levels of interannual variability at each site as was seen for summer baseflow (Fig. 4.13-4.14). As observed in Phase I, no general seasonal trend was observed for all sites – the majority of sites show no difference between winter and summer baseflow. There are however, a few notable exceptions in either direction. Sites 2, 20, 58 and 46 tended to have higher total PAH concentrations in winter whereas sites 3, 10 and 117 had higher PAHs during the summer.

Caffeine and fragrance materials also appeared to show no seasonal trend within observed interannual variability (Fig. 4.15), with the notable exceptions of sites 58, 43 and 49 which all had higher summer concentrations of fragrances.

Fecal steroids and their source indicator ratios, on the other hand, did show distinct seasonal trends (Fig. 4.16). In general, the total fecal steroids were less concentrated in winter months, especially west of the Hudson (Fig. 4.16a). The ratio of coprostanol to conprostanol plus cholestanol, an indicator of human fecal material, tended to be higher during the winter than summer, again predominantly at West-of-Hudson sites (Fig. 4.16b).

Stormflow

Concentrations of PAHs increased with increasing discharge during storm events at all three sites where storms were sampled (Fig. 4.17). Also, PAH concentration tended to be higher during the peak in turbidity during each storm, rather than during the maximum discharge. Ratios of Anthracene to Anthracene plus Fluoranthene tended to decrease with increasing storm discharge (Fig. 4.18), whereas the ratio of high to low molecular weight PAHs showed the opposite trend of increasing with increasing storm discharge (Fig. 4.19).

Caffeine and fragrance material concentrations all increased substantially during high storm flows, but only site 29 showed a consistent relationship between discharge and concentration when the data from all years are pooled (Fig. 4.20 and 4.21).

The sum of fecal steroids increase with increasing storm flow at all sites (Fig. 4.22), but the ratio of coprostanol to coprostanol plus cholestanol, bCOP/(bCOP+aCOP), shows different patterns at each site (Fig. 4.23). At site 6, the West Branch of the Delaware, bCOP/(bCOP+aCOP) tends to increase during a storm, whereas tend to decrease with storm flow at site 55, the Kisco (Fig.

- 58 - CHAPTER 4 – MOLECULAR TRACERS NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

4.23). The Neversink, site 29, generally shows little change in bCOP/(bCOP+aCOP) during storms.

Discussion

Sources

Polycyclic aromatic hydrocarbons have three groups of sources that can be distinguished from one another based on compound distributions and ratios. These are: 1) petroleum products – such as kerosene, diesel oil, lubricating oil and crude oil – which are characterized by lower ratios of less stable to more stable isomers (i.e. ANT/(ANT+PHE) or FLR/(FLR+PYR)) and by lower ratios of high to low molecular weight PAHs (H/LPAH) (Yunker et al. 2002, Zakaria et al. 2002); 2) combustion byproducts – such as automotive exhaust, smelter emissions, coal burning emissions and wood smoke – which are characterized by higher ratios of less stable to more stable isomers and by higher H/LPAH ratios (Dickhut et al. 2000, Yunker et al. 2002); and 3) asphalt, which is characterized by low ratios of less stable to more stable isomers (similar to petroleum products) and by higher H/LPAH ratios (similar to combustion byproducts) (Yunker et al. 2002).

At Phase I and Phase II sites, high ratios of ANT/(ANT+PHE), FLR/(FLR+PYR) and H/LPAH all suggest that combustion emissions appear to dominate over petroleum spills as the primary source of PAHs to most of the stream sites (Fig. 4.4, 4.5, 4.6). However, interannual variability in these ratios suggest that the relative proportions of these sources could change with time. For instance, H/LPAH ratios in particular might suggest that petroleum products could have been an important source of PAHs in 2003 at nearly a third of all sites. However, in 2003 total concentrations of PAHs from any source were substantially reduced relative to Phase I findings (Fig. 4.1). These observations can possibly be explained by the high levels of precipitation throughout the summer of 2003. Higher water flows would preferentially dilute PAH components from sources that have a constant flux (i. e. soot from automobiles and coal-fired power plants), whereas flushing of pavement and sewage systems by heavy rains would tend increase the fluxes (per unit watershed area) of petroleum products into streams and rivers. This flushing would increase the relative proportions of petroleum sources while at the same time concentrations might still decrease due to dilution. Caution should be taken, however, when interpreting subtle differences in H/LPAH ratios because of the substantial enrichment of high molecular weight PAHs in particles. Clearly, in-stream variations in flow that change the concentrations of suspended particulates can in turn have a strong affect on these ratios.

Therefore, measured PAH concentrations and ratios at NY sites suggest that PAHs are introduced to stream waters via a combination of combustion sources (via soot deposition from local or distant locations) along with contributions from asphalt from road runoff. Although spills of petroleum products appear to be a neglible source in general, their relative contributions may increase during exceedling wet years. However, preferential evaporative losses of the more volatile and bioavailable low molecular weight PAHs during transport and storage in the environment could transform petroleum products to give a low ANT/(ANT+PHE) and high H/LPAH signature similar to that of asphalt (Countway et al. 2003).

- 59 - CHAPTER 4 – MOLECULAR TRACERS NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Fragrance materials, HHCB and AHTN, and caffeine are introduced to streams and rivers relatively unambiguous sources. Fragrance materials are anthropogenic compounds introduced to the environment primarily in domestic greywater sewage. Because of the low biodegradability of these polycyclic compounds, they are transported relatively conservatively though sewage treatment and down streams (Simonich et al. 2002, Artola-Garicano et al. 2003). However, because of the hydrophobicity of HHCB and AHTN (LOG10 of their octanol-water partition coefficients are 5.9 and 5.7 respectively), it has been suggested that their concentrations are often a function of total suspended solid concentrations in sewage treatment plant effluent (Simonich et al. 2002, Artola-Garicano et al. 2003). Our 2003 data, showing that most of these fragrances are found in the dissolved phase, suggest the opposite and explain why the concentrations of fragrance materials are closely correlated with caffeine, which is very hydrophilic (SWRC 2003).

The only source of caffeine to streams and rivers in temperate climates is the urine of humans (and sometimes domestic animals). Although removed more effectively than HHCB and AHTN by waste water treatment processes, caffeine still displays relatively low rates of biodegradation in the environment and is transported though waterways relatively efficiently (Buerge et al. 2003). However, caffeine has much lower particle affinity (LOG10 of octanol-water partition coefficients are -0.7, 5.7, 5.9 for caffeine, AHTN and HHCB respectively) and is thus much less affected by the dynamics of the particulate phase. In addition, these low particle affinities suggest that caffeine is much more likely to enter streams from leaking septic systems via ground water inputs.

Fecal steroids have three primary potential sources to streams and rivers; human sewage, agricultural wastes from domestic animals, and wildlife (mammals and birds). These sources can generally be differentiated because animal species excrete fecal steroids in characteristic patterns (Leeming et al. 1996). The most striking of these patterns is that human fecal material has extremely high concentrations of coprostanol (bCOP) relative to other steroids, whereas coprostanol is a minor component of the fecal steroids of other animals (Leeming et al. 1996). Although fecal steroids are known have high particle affinity, little is known about the biodegradation rates or residence times of fecal steroids in the environment.

Temporal Variability

As can be seen in Figures 4.1 to 4.16, temporal variability is appreciable within sites. Understanding the underlying causes of this variability is useful not only for better intersite comparisions, but also for exploring source processes at given sites. As a whole, concentrations of molecular tracers could be partially explained by hydrology. Most tracers (log transformed) showed a negative correlation to basin-area-normalized discharge (log transformed) (Fig. 4.24), with correlation coefficients ranging from -0.11 for caffeine (p = 0.05) to -0.28 for total PAHs (p = 1 x 10-6) to -0.38 for cholesterol (p = 5 x 10-12). Similarly, most tracers (with the exception of caffeine) showed a positive correlation to concentrations of total suspended solids (TSS), with correlation coefficients ranging from 0.16 for total PAHs (p = 0.005) to 0.34 for the sum of seven fecal steroids (p = 7 x 10-8). When combined using multiple linear regression (MLR), discharge and TSS together explain 5% to 21% of the variability in tracer concentrations.

- 60 - CHAPTER 4 – MOLECULAR TRACERS NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Although most of the interannual variability within site is statistically unexplainable, it is nonetheless real and offers valuable information for understanding source dynamics. For instance, the highest bCOP/(bCOP+aCOP) ratios at many sites with no upstream waste water treatment plant discharge (such as sites 103, 105, 107, 113, 109, 110, 112) occurred during the wettest summer baseflow conditions sampled (Fig. 4.16 and 4.24), opposite to the overall trend of lower concentrations during high flows and suggesting septic field failures.

Appendix Discussion: QA/QC

Data quality objectives for molecular tracers data are described in the Year 5 Quality Assurance Project Plan (2004), and summarized below. Evaluation of these QA/QC data needs to consider that all of these objective criteria are being evaluated at concentrations typical of those measured in streams, which are generally in the range of 0.001 to 0.1 µg/L. These concentrations are 3-6 orders of magnitude lower than the quantitation limits listed in Table 2 of EPA 8270 (generally 10-200 µg/L) and 1-3 orders of magnitude lower than the reporting levels listed by the recent USGS-NWQA methods (ranging from 0.005 to 2.5 µg/L) (Koplin et. al. 2002). As such, our blanks and absolute differences between duplicates are much lower than any other published dataset when considered in absolute concentration units. However, we evaluate our QA/QC data with respect to Method Detection Limits (MDL) that we measured for our high sensitivity method (ranging from 0.00009 to 0.01 µg/L) or with respect to relative differences between replicates. As a result, simply counting QA/QC flags might lead to the impression that data quality using our method is no better than other methods. However, we feel that if viewed in terms of absolute concentration units, our method provides the highest quality data available by any method.

Method Detection Limits (MDL). As an ongoing reference point for many of our data quality objectives, we again determined MDL values for each tracer compound using the procedures described by the Code of Federal Regulations (40 CFR Part 136). In brief, we analyzed replicates (n = 8) of nanopure water spiked with standards at 0.001 µg/L. This spike concentration was lower than used for previous MDL studies, but appropriate given slight modifications in our method which increased sensitivity and decreased instrument detection limits by nearly an order of magnitude. We calculated MDLs by multiplying the standard deviation of concentrations by the appropriate n - 1 Student t distribution value for 99% confidence. MDL results are presented in Appendix A.4. The first MDL study results were applied to all 2004 data.

Field blanks (FB). In order to evaluate potential contamination issues during baseflow sample collection, we collected field blanks by transferring nanopure water into sample jugs on the banks of a number of stream sites. For stormflow collections, nanopure water was pumped through the ISCO sampling device into sample bottles. Our objective was to collect field blanks for 5% of samples. Our QC criterion is that the field blank should be less than twice the method detection limit (MDL). Data are presented in the Appendix A.4. For project year 5 (2004), we collected 4 FBs during summer baseflow, 2 during winter baseflow and 3 during stormflow, totaling 9.6% of collected samples. FBs exceeded 2*MDL for 23% of baseflow and 58% of stormflow measurements.

- 61 - CHAPTER 4 – MOLECULAR TRACERS NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Lab blanks (LB). In order to better track down sources of contamination in field blanks (which include any potential contamination in the lab), lab blanks were also analyzed with the same frequency and same criteria as field blanks. Data are presented in the Appendix A.4. For project year 5 (2004), we collected 6 LBs during summer baseflow, 2 during winter baseflow, totaling 8.5% of collected samples. LBs exceeded 2*MDL for 17% of baseflow measurements.

Field duplicates (FD). In order to evaluate precision in our collection, preparation and analysis procedures, we collected duplicate samples in the field. Our objective was to collect FDs for 5% of samples. Our QC criterion is that the relative percent difference (RPD) of FDs should be less than the precision values specified in the Appendix A.4. However, if the measured concentration is less than the MDL divided by the stated precision limit (Appendix A.4), then duplicates are evaluated by their absolute difference, which should be less than the MDL. Summer baseflow data are presented in Appendix A.4. For project year 5 (2004), we collected 4 FDs during summer baseflow, 2 during winter baseflow and 2 during stormflow, totaling 8.9% of collected samples. FDs exceeded limits for 36% of summer baseflow measurements, 22% of winter baseflow measurements, and 42% of stormflow measurements.

Lab duplicates (LD). Lab duplicates were phased out for phase II because they offered no additional information over field duplicates and matrix spike duplicates.

Matrix spikes (MS). In order to evaluate our ability to recover known additions of tracer compounds using our extraction and analytical methods, we spiked a set of field duplicate samples with a mixture of 0.01 ug/L of each standard. We performed the full laboratory extraction and analysis procedure on matrix spike duplicates, to assess the reproducibility of our method at moderate concentrations (many compounds in most field and laboratory duplicates are near our detection limits). These matrix spike duplicates are thus different from those in Phase I, where we simply injected and analyzed one matrix spike extract twice on the GC/MS to assess the instrumental precision. Our objective was to analyze one MS for every 20 samples (5%). Percent recoveries were evaluated against our accuracy criteria presented in the Appendix A.4. Recovery data are presented in Appendix A.4. Replication data are presented Appendix A.4. For project year 5 (2004), we analyzed 4 MSs during summer baseflow, zero during winter baseflow (due to lost samples in the lab), and 1 during storm sampling, totaling 5.3% of collected samples. Percent recoveries exceeded limits for 34% of measurements during summer baseflow. Large exceedences only occurred for fecal steroids in 2 samples. These exceedences occurred because for two of the four the stream samples chosen as the matrix (sites 10 & 55), in situ concentrations of fecal steroids were very high (CHO of 0.3 and 0.4 µg/L respectively). Since our matrix spike was designed to increase ambient concentrations by 0.01 µg/L, for these samples, we were looking for very small differences in two large numbers, which inherently has substantial error. MS duplicates exceeded limits for 4 % of summer baseflow measurements and 9 % of storm measurements. Furthermore, with the exception of some fecal steroids from two samples, RPDs of MS duplicates were generally under 12%.

Standard reference material (SRM). Similar to matrix spiked samples, we analyzed SRMs each year to evaluate our ability to quantify known concentrations of tracer compounds using our extraction and analytical methods. However, the analyst was blind to certified concentrations while conducting the analyses. Our concentration data were evaluated relative to the acceptance limits provided by the supplier. SRM data are presented in Appendix A.4. For project year 5

- 62 - CHAPTER 4 – MOLECULAR TRACERS NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

(2004), we analyzed a total of 6 SRM samples (4 in summer and 2 in winter). We passed criteria for all compounds and in all SRM samples, with the exception FLU where measured values were 12% and 50% below criteria in our two SRMs analyzed during winter baseflow.

Literature Cited

Adair, E. C., D. Binkley, and D. C. Andersen. 2004. Patterns of nitrogen accumulation and cycling in riparian floodplain ecosystems along the Green and Yampa rivers. Oecologia 139:108-116. Anderson, J. R., E. E. Hardy, J. T. Roach, and R. E. Witmer. 1976. A land use and land cover classification system for use with remote sensor data. Professional paper 964, U.S. Geological Survey, Washington, D.C. Artola-Garicano, E., I. Borkent, J. L. M. Hermens, and W. H. J. Vaes. 2003. Removal of Two Polycyclic Musks in Sewage Treatment Plants: Freely Dissolved and Total Concentrations. Environmental Science and Technology 37:3111 -3116. Brady, N. C., and R. R. Weil. 1999. The nature and properties of soils, 12 edition. Prentice Hall, Upper Saddle River, NJ. Buerge, I. J., T. Poiger, M. D. Müller, and H.-R. Buser. 2003. Caffeine, an Anthropogenic Marker for Wastewater Contamination of Surface Waters. Environmental Science and Technology 37:691 - 700. Countway, R. E., R. M. Dickhut, and E. A. Canuel. 2003. Polycyclic aromatic hydrocarbon (PAH) distributions and associations with organic matter in surface waters of the York River, VA Estuary. Organic Geochemistry 34:209-224. Dickhut, R. M., E. A. Canuel, K. E. Gustafson, K. Liu, K. M. Arzayus, S. E. Walker, G. Edgecombe, M. O. Gaylor, and E. H. MacDonald. 2000. Automotive Sources of Carcinogenic Polycyclic Aromatic Hydrocarbons Associated with Particulate Matter in the Chesapeake Bay Region. Environmental Science and Technology 34:4635 -4640. EPA. 2002. 2002 Edition of the Drinking Water Standards and Health Advisories. EPA 822-R-02-038, US EPA Office of Water, Office of Science and Technology. Grimalt, J. O., P. Fernandez, J. M. Bayona, and J. Albaiges. 1990. Assessment of fecal sterols and ketones as indicators of urban sewage inputs to coastal waters. Environmental Science and Technology 24:357 - 363. Isachsen, Y. W., E. Landing, J. M. Lauber, L. V. Rickard, and W. B. Rogers, editors. 2000. Geology of New York. A simplified account, 2nd edition. The New York State Geological Survey, Albany, NY. King, R. S., M. E. Baker, D. F. Whigham, D. E. Weller, T. E. Jordan, P. F. Kazyak, and M. K. Hurd. 2005. Spatial considerations for linking watershed land cover to ecological indicators in streams. Ecological Applications 15:137-153. Kolpin, D. W., E. T. Furlong, M. T. Meyer, E. M. Thurman, S. D. Zaugg, L. B. Barber, and H. T. Buxton. 2002. Pharmaceuticals, Hormones, and Other Organic Wastewater Contaminants in U.S. Streams, 1999-2000: A National Reconnaissance. Environmental Science and Technology 36:1202 -1211. Leeming, R., A. Ball, N. Ashbolt, and P. Nichols. 1996. Using faecal sterols from humans and animals to distinguish faecal pollution in receiving waters. Water Research 30:2893-2900. Leeming, R., and P. Nichols. 1996. Concentrations of coprostanol that correspond to existing bacterial indicator guideline limits. Water Research 30:2997-3006. Leeming, R., P. D. Nichols, and N. J. Ashbolt. 1998. Distinguishing sources of faecal pollution in Australian inland and coastal waters using sterol biomarkers and microbial faecal indicators. Research Report No 204., Water Services Association of Australia. Mudge, S. M., and C. G. Seguel. 1999. Organic contamination of San Vicente Bay, Chile. Marine Pollution Bulletin 38:1011-1021. NRCS. 1994. State soil geographic (STATSGO) data base. Data use information. Miscellaneous

- 63 - CHAPTER 4 – MOLECULAR TRACERS NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Publication 1492, USDA Natural Resources Conservation Service, Fort Worth, TX. O'Leary, T., R. Leeming, P. D. Nichols, and J. K. Volkman. 1999. Assessment of the sources, transport and fate of sewage-derived organic matter in Port Phillip Bay, Australia, using the signature lipid coprostanol. Marine and Freshwater Research 50:547-556. Olson, R. J., J. M. Briggs, J. H. Porter, G. R. Mah, and S. G. Stafford. 1999. Managing data from multiple disciplines, scales, and sites to support synthesis and modeling. Remote Sensing of Environment 70:99-107. Simonich, S. L., W. M. Begley, G. Debaere, and W. S. Eckhoff. 2000. Trace Analysis of Fragrance Materials in Wastewater and Treated Wastewater. Environmental Science and Technology 34:959 -965. Simonich, S. L., T. W. Federle, W. S. Eckhoff, A. Rottiers, S. Webb, D. Sabaliunas, and W. de Wolf. 2002. Removal of Fragrance Materials during U.S. and European Wastewater Treatment. Environmental Science and Technology 36:2839 -2847. Standley, L. J., L. A. Kaplan, and D. Smith. 2000. Molecular tracers of organic matter sources to surface water resources. Environmental Science and Technology 34:3124-3130. SWRC. 2003. Water quality monitoring in the source water areas for New York City: An integrative approach. A report on the first phase of monitoring. Stroud Water Research Center, Avondale, PA. Yunker, M. B., R. W. Macdonald, R. Vingarzan, R. H. Mitchell, D. Goyette, and S. Sylvestre. 2002. PAHs in the Fraser River basin: a critical appraisal of PAH ratios as indicators of PAH source and composition. Organic Geochemistry 33:489-515. Zakaria, M. P., H. Takada, S. Tsutsumi, K. Ohno, J. Yamada, E. Kouno, and H. Kumata. 2002. Distribution of Polycyclic Aromatic Hydrocarbons (PAHs) in Rivers and Estuaries in Malaysia: A Widespread Input of Petrogenic PAHs. Environmental Science and Technology 36:1907 -1918.

- 64 - CHAPTER 4 – MOLECULAR TRACERS NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Table 4.1. Compounds chosen as molecular tracers, abbreviations used in this report, and ions (mass-to-charge ratios) selected for quantitation and confirmation of each compound using Selected Ion Monitoring (SIM) mode with our Gas Chromatography-Mass Spectrometry (GC- MS) system.

Abbrevia Quant. 1st Confirm. 2nd Confirm. 3rd Confirm. Compound tion Ion Ion Ion Ion Internal Standards p-terphenyl-D14 TERd14 244 212 160 122 5α-cholestane aCHO 217 357 372 149 PAH fluorene FLU 166 82 139 phenanthrene PHE 178 89 152 76 anthracene ANT 178 89 152 76 2-methyl phenanthrene 2MP 192 165 94 1-methyl phenanthrene 1MP 192 165 94 fluoranthene FLR 202 101 88 174 pyrene PYR 202 101 88 174 benz(a)anthracene BAA 228 114 101 200 chrysene CHR 228 113 101 200 benzo(b)fluoranthene BBF 252 126 113 224 benzo(k)fluoranthene BKF 252 126 113 224 benzo(a)pyrene BAP 252 126 113 224 phenanthrene-D10 (surrogate) PHEd10 188 94 160 80 chrysene-D12 (surrogate) CHRd12 240 120 106 208 perylene-D12 (surrogate) PERd12 264 132 118 232 Fragrances tonalide (1,3,4,6,7,8-hexahydro- 4,6,6,7,8,8-hexamethylcyclopenta[γ]- 2-benzopyran) HHCB 243 258 213 galaxolide (7-acetyl-1,1,3,4,4,6,- hexamethyl-1,2,3,4- tetrahydronaphthalene) AHTN 243 258 213 159 Caffeine caffeine (surrogate) CAF 194 109 82 67 caffeine-D9 CAFd9 203 115 88 70 Steroids coprostanol (5β-cholestan-3β-ol) bCOP 370 355 215 257 epi-coprostanol (5β-cholestan-3α-ol) EPI 370 215 355 257 cholesterol (cholest-5-en-3β-ol) CHOL 368 129 329 458 cholestanol (5α-cholestan-3β-ol) aCOP 215 460 445 335 24-ethyl-coprostanol (24-ethyl-5β- cholestan-3β-ol) eCOP 398 383 215 257

- 65 - CHAPTER 4 – MOLECULAR TRACERS NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Abbrevia Quant. 1st Confirm. 2nd Confirm. 3rd Confirm. Compound tion Ion Ion Ion Ion 24-ethyl-epicoprostanol (24-ethyl-5β- cholestan-3α-ol) eEPI 398 383 215 257 cholestanone (5α-cholestan-3-one) aONE 231 386 371 coprostanone (5β-cholestan-3-one bONE 231 386 371 316 24-ethyl-cholesterol (24-ethyl-cholest-5- en-3β-ol) eCHO 129 357 486 396 24-ethyl-cholestanol (24-ethyl-5α- cholestan-3β-ol) SNOL 215 488 473 383 cholesterol-D6 (surrogate) CHOL13 374 131 333 464

- 66 - CHAPTER 4 – MOLECULAR TRACERS NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

10 Mean 2000 Increasing Drainage Area 2001 1 2002 2003 2004 0.1

0.01 Sum Total PAHs (µg/L) 0.001

WBD EBD SCH ESP NRD 0.0001 1 7 2 9 3 4 5 6 8 15 14 11 10 12 13 19 16 17 20 18 21 25 22 24 23 26 28 27 30 29 102 151 106 153 104 101 103 105 108 114 160 107 113 111 109 110 112 115 116 118 117 159 121 120 119 122 123 Station ID 10 Increasing Drainage Area 1

0.1

0.01 Sum Total PAHs (µg/L) 0.001

EMC WBC MNC TCS KSC 0.0001 39 31 42 32 34 35 40 44 33 37 36 38 41 45 54 58 43 50 51 49 46 48 53 52 47 59 60 56 57 55 124 132 129 125 150 127 126 141 134 140 133 139 137 143 145 146 131 149 130 138 148 147 142 Station ID

Figure 4.1: Summer baseflow stream water concentrations of the sum of all measured PAH compounds (1MP, 2MP, ANT, BAA, BAP, BBF, BKF, CHR, FLR, FLU, PHE, PYR) for all Phase I and Phase II sites both (A) West of Hudson and (B) East of Hudson. Geomentric means of all measured values at each site are given by the solid grey line. Sites are arranged by subwatershed in order of increasing basin area. Subwatershed designations are identical to those used in Chapter 3: Neversink River and Rondout Creek (NRD); Esopus Creek (ESP); Schoharie Creek (SCH); E. Br. Delaware River (EBD); and W. Br. Delaware River (WBD). E. & M. Br. Croton R. (EMC); W. Br.Croton R. (WBC); Muscoot R. (MUC); Titicus, Cross, and Stone Hill Rivers (TCS); and Kensico Resv. & Lower New Croton Resv. (KSC).

- 67 - CHAPTER 4 – MOLECULAR TRACERS NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

1 Mean 2000 Increasing Drainage Area WOH 2001 2002 0.1 2003 2004 Criteria

0.01

0.001

0.0001 Benzo(a)anthracene (BAA) (µg/L) Benzo(a)anthracene (BAA)

WBD EBD SCH ESP NRD 0.00001 1 7 2 9 3 4 5 6 8 15 14 11 10 12 13 19 16 17 20 18 21 25 22 24 23 26 28 27 30 29 102 151 106 153 104 101 103 105 108 114 160 107 113 111 109 110 112 115 116 118 117 159 121 120 119 122 123 Station ID 1 Increasing EOH Drainage Area 0.1

0.01

0.001

0.0001 Benzo(a)anthracene (BAA) (µg/L) (BAA) Benzo(a)anthracene

EMC WBC MNC TCS KSC 0.00001 39 31 42 32 34 35 40 44 33 37 36 38 41 45 54 58 43 50 51 49 46 48 53 52 47 59 60 56 57 55 124 132 129 125 150 127 126 141 134 140 133 139 137 143 145 146 131 149 130 138 148 147 142 Station ID

Figure 4.2: Summer baseflow stream water concentrations of Benzo(a)anthracene for all Phase I and Phase II sites sites both (A) West of Hudson and (B) East of Hudson. Geomentric means of all measured values at each site are given by the solid grey line. Dotted lines show EPA water quality criteria levels for human health risks for the consumption of water (lower line) and for the consumption of organisms living in the water (upper line) (EPA 2002). Sites are arranged in order of increasing basin area by subwatershed, as designated in Chapter 3 and in Fig. 4.1.

- 68 - CHAPTER 4 – MOLECULAR TRACERS NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

1 Mean 2000 Increasing Drainage Area WOH 2001 2002 0.1 2003 2004 Criteria

0.01

0.001

0.0001 Benzo(k)fluoranthene (BKF) (µg/L) WBD EBD SCH ESP NRD 0.00001 1 7 2 9 3 4 5 6 8 15 14 11 10 12 13 19 16 17 20 18 21 25 22 24 23 26 28 27 30 29 102 151 106 153 104 101 103 105 108 114 160 107 113 111 109 110 112 115 116 118 117 159 121 120 119 122 123 Station ID 1 Increasing EOH Drainage Area 0.1

0.01

0.001

0.0001 Benzo(k)fluoranthene(µg/L) (BKF) EMC WBC MNC TCS KSC 0.00001 39 31 42 32 34 35 40 44 33 37 36 38 41 45 54 58 43 50 51 49 46 48 53 52 47 59 60 56 57 55 124 132 129 125 150 127 126 141 134 140 133 139 137 143 145 146 131 149 130 138 148 147 142 Station ID

Figure 4.3: Summer baseflow stream water concentrations of Benzo(k)fluoranthene for all Phase I and Phase II sites sites both (A) West of Hudson and (B) East of Hudson. Geomentric means of all measured values at each site are given by the solid grey line. Dotted lines show EPA water quality criteria levels for human health risks for the consumption of water (lower line) and for the consumption of organisms living in the water (upper line) (EPA 2002). Sites are arranged in order of increasing basin area by subwatershed, as designated in Chapter 3 and in Fig. 4.1.

- 69 - CHAPTER 4 – MOLECULAR TRACERS NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

1.0 Mean WOH 2000 Increasing Drainage Area 2001 0.8 2002 2003 2004 0.6 Limit

0.4 ANT / (ANT+PHE) / ANT

0.2

WBD EBD SCH ESP NRD 0.0 1 7 2 9 3 4 5 6 8 15 14 11 10 12 13 19 16 17 20 18 21 25 22 24 23 26 28 27 30 29 102 151 106 153 104 101 103 105 108 114 160 107 113 111 109 110 112 115 116 118 117 159 121 120 119 122 123 Station ID 1.0 Increasing EOH Drainage Area 0.8

0.6

0.4 ANT / (ANT+PHE) ANT

0.2

EMC WBC MNC TCS KSC 0.0 39 31 42 32 34 35 40 44 33 37 36 38 41 45 54 58 43 50 51 49 46 48 53 52 47 59 60 56 57 55 124 132 129 125 150 127 126 141 134 140 133 139 137 143 145 146 131 149 130 138 148 147 142 Station ID

Figure 4.4: Summer baseflow ratios of anthracene/(anthracene+phenanthrene), a PAH source indicator, for all Phase I and Phase II sites both (A) West of Hudson and (B) East of Hudson. High values of ANT/(ANT+PHE) indicate combustion sources dominate over petroleum sources. Dotted lines show a suggested limit between sources of ANT/(ANT+PHE) = 0.1 (Yunker et al. 2002). Arithmetic means of all measured values at each site are given by the solid grey line.

- 70 - CHAPTER 4 – MOLECULAR TRACERS NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

1.0 Mean 2000 Increasing Drainage Area WOH 2001 2002 0.8 2003 2004 Limit

0.6

0.4 FLR / (FLR+PYR)FLR /

0.2

0.6

0.4 FLR / (FLR+PYR)

0.2

Figure 4.5: Summer baseflow ratios of fluoranthene/(fluoranthene+pyrene), a PAH source indicator, for all Phase I and Phase II sites both (A) West of Hudson and (B) East of Hudson. High values of FLR/(FLR+PYR) indicate combustion sources dominate over petroleum sources. Values between dotted lines (0.4 to 0.5) suggest mixed sources (Yunker et al. 2002). Arithmetic means of all measured values at each site are given by the solid grey line.

- 71 - CHAPTER 4 – MOLECULAR TRACERS NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

PAH 100 Mean WOH 2000 Increasing Drainage Area 2001 2002 10 2003 2004 Limit

0.1

WBD EBD SCH ESP NRD 0.01 1 7 2 9 3 4 5 6 8 High Mol. Weight / Low Mol. Weight PAHs (H/L) High Mol. Weight / Low Mol. PAHs Weight 15 14 11 10 12 13 19 16 17 20 18 21 25 22 24 23 26 28 27 30 29 102 151 106 153 104 101 103 105 108 114 160 107 113 111 109 110 112 115 116 118 117 159 121 120 119 122 123 Station ID

PAH 100 Increasing EOH Drainage Area

0.1

EMC WBC MNC TCS KSC 0.01 High Mol. Weight / Low Mol. Weight PAHs (H/L) High Mol. Weight / Low Mol. PAHs Weight 39 31 42 32 34 35 40 44 33 37 36 38 41 45 54 58 43 50 51 49 46 48 53 52 47 59 60 56 57 55 124 132 129 125 150 127 126 141 134 140 133 139 137 143 145 146 131 149 130 138 148 147 142 Station ID Figure 4.6: Summer baseflow ratios of high versus low molecular weight PAHs, a PAH source indicator, for all Phase I and Phase II sites both (A) West of Hudson and (B) East of Hudson. High values of H/LPAH indicate combustion sources or asphalt versus fresh petroleum. Dotted lines show a suggested limit between sources of H/LPAH = 0.5 (Zakaria et al. 2002). Geomentric means of all measured values at each site are given by the solid grey line.

- 72 - CHAPTER 4 – MOLECULAR TRACERS NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

10 Mean WOH 2000 Increasing Drainage Area 2001 1 2002 2003 2004 g/L) µ 0.1

0.01 Caffeine (CAF) Caffeine( (CAF) 0.001

0.01 Caffeine (CAF) ( Caffeine (CAF) 0.001

Figure 4.7: Summer baseflow stream water concentrations of Caffeine (CAF) for all Phase I and Phase II sites sites both (A) West of Hudson and (B) East of Hudson. Geomentric means of all measured values at each site are given by the solid grey line.

- 73 - CHAPTER 4 – MOLECULAR TRACERS NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

10 Mean WOH 2000 Increasing Drainage Area

g/L) 2001 µ 1 2002 2003 2004 0.1

0.01

0.001 Sum Fragrance Materials (FM) ( (FM) Materials Sum Fragrance WBD EBD SCH ESP NRD 0.0001 1 7 2 9 3 4 5 6 8 15 14 11 10 12 13 19 16 17 20 18 21 25 22 24 23 26 28 27 30 29 102 151 106 153 104 101 103 105 108 114 160 107 113 111 109 110 112 115 116 118 117 159 121 120 119 122 123 Station ID 10 Increasing EOH Drainage Area g/L)

µ 1

0.1

0.01

0.001 Sum Fragrance Materials (FM) ( Sum Fragrance Materials (FM) EMC WBC MNC TCS KSC 0.0001 39 31 42 32 34 35 40 44 33 37 36 38 41 45 54 58 43 50 51 49 46 48 53 52 47 59 60 56 57 55 124 132 129 125 150 127 126 141 134 140 133 139 137 143 145 146 131 149 130 138 148 147 142 Station ID

Figure 4.8: Summer baseflow stream water concentrations of the sum of fragrance materials (FM) tonalide (AHTN) and galaxolide (HHCB) for all Phase I and Phase II sites sites both (A) West of Hudson and (B) East of Hudson. Geomentric means of all measured values at each site are given by the solid grey line.

- 74 - CHAPTER 4 – MOLECULAR TRACERS NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Mean WOH 100 2000 Increasing Drainage Area 2001

g/L) 2002 µ 2003 10 2004

0.1 Sum Fecal ( Sum Steroids (FS)

WBD EBD SCH ESP NRD 0.01 1 7 2 9 3 4 5 6 8 15 14 11 10 12 13 19 16 17 20 18 21 25 22 24 23 26 28 27 30 29 102 151 106 153 104 101 103 105 108 114 160 107 113 111 109 110 112 115 116 118 117 159 121 120 119 122 123 Station ID

Increasing EOH 100 Drainage Area g/L) µ 10

0.1 Sum Fecal Steroids (FS) (

EMC WBC MNC TCS KSC 0.01 39 31 42 32 34 35 40 44 33 37 36 38 41 45 54 58 43 50 51 49 46 48 53 52 47 59 60 56 57 55 124 132 129 125 150 127 126 141 134 140 133 139 137 143 145 146 131 149 130 138 148 147 142 Station ID

Figure 4.9: Summer baseflow stream water concentrations of the sum of seven fecal steroids (FS) – aCOP, aONE, bCOP, bONE, CHO, EPI and SNOL (see table 4.1) – for all Phase I and Phase II sites sites both (A) West of Hudson and (B) East of Hudson. Geomentric means of all measured values at each site are given by the solid grey line.

- 75 - CHAPTER 4 – MOLECULAR TRACERS NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

100 Mean WOH 2000 Increasing Drainage Area 10 2001 2002

g/L) 2003 µ 1 2004

0.1

0.01 Coprostanol (bCOP) ( (bCOP) Coprostanol 0.001

0.1

0.01 Coprostanol (bCOP) ( (bCOP) Coprostanol 0.001

Figure 4.10: Summer baseflow stream water concentrations of coprostanol (bCOP) for all Phase I and Phase II sites sites both (A) West of Hudson and (B) East of Hudson. Geomentric means of all measured values at each site are given by the solid grey line. Dotted lines show concentrations of this specific indicator of human fecal material that that have been demonstrated to correspond to fecal bacteria concentrations of: 1) 35 enterococci per 100 mL and 300 thermotolerant coliforms per 100 mL for the lower line, and 2) 200 enterococci per 100 mL and 1100 thermotolerant coliforms per 100 mL for the upper line (Leeming and Nichols 1996).

- 76 - CHAPTER 4 – MOLECULAR TRACERS NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

1.0 Mean 2000 Increasing Drainage Area WOH 2001 2002 0.8 2003 2004 Limit

0.6

0.4 bCOP/(bCOP+aCOP) 0.2

0.6

0.4 bCOP/(bCOP+aCOP) 0.2

Figure 4.11: Summer baseflow ratios of coprostanol/(coprostanol+cholestanol) (bCOP/(bCOP+aCOP)), a fecal steroid source indicator, for all Phase I and Phase II sites both (A) West of Hudson and (B) East of Hudson. High values of bCOP/(bCOP+aCOP) indicate human fecal sources dominate over livestock and wildlife sources. Dotted lines show a suggested limit between sources of bCOP/(bCOP+aCOP) = 0.2 [Grimalt, 1990 #3849][O'Leary, 1999 #4989]. Arithmetic means of all measured values at each site are given by the solid grey line.

- 77 - CHAPTER 4 – MOLECULAR TRACERS NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

100 Mean WOH 2000 Increasing Drainage Area 2001 10 2002 2003 2004 1

bCOP/CHO 0.1

0.01

WBD EBD SCH ESP NRD 0.001 1 7 2 9 3 4 5 6 8 15 14 11 10 12 13 19 16 17 20 18 21 25 22 24 23 26 28 27 30 29 102 151 106 153 104 101 103 105 108 114 160 107 113 111 109 110 112 115 116 118 117 159 121 120 119 122 123 Station ID 100 Increasing EOH Drainage Area 10

bCOP/CHO 0.1

0.01

EMC WBC MNC TCS KSC 0.001 39 31 42 32 34 35 40 44 33 37 36 38 41 45 54 58 43 50 51 49 46 48 53 52 47 59 60 56 57 55 124 132 129 125 150 127 126 141 134 140 133 139 137 143 145 146 131 149 130 138 148 147 142 Station ID

Figure 4.12: Summer baseflow ratios of coprostanol/cholesterol (bCOP/CHO), a fecal steroid source indicator, for all Phase I and Phase II sites both (A) West of Hudson and (B) East of Hudson. High values of bCOP/CHO indicate human fecal sources dominate over livestock and wildlife sources [Mudge, 1999 #4878]. Geometric means of all measured values at each site are given by the solid grey line.

- 78 - CHAPTER 4 – MOLECULAR TRACERS NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

10 2000 2001 WOH EOH 2002 Increasing 2003 Drainage Area 1 2004 g/L) µ winter

0.1 summer

Sum Total PAH ( PAH Total Sum 0.01

WBD EBD SCH ESP NRD EMC WBC MNC TCS KSC 0.001 1 2 9 3 6 8 15 11 10 12 13 16 17 20 18 22 24 23 26 27 30 29 32 34 40 33 58 43 49 46 52 60 55 102 103 105 108 107 117 119 123 125 150 133 139 145 130 148 142 Station ID

1.0 2000 2001 WOH EOH 2002 Increasing 0.8 2003 Drainage Area 2004

0.6 winter summer

0.4 ANT / (ANT+PHE) / ANT

0.2

WBD EBD SCH ESP NRD EMC WBC MNC TCS KSC 0.0 1 2 9 3 6 8 15 11 10 12 13 16 17 20 18 22 24 23 26 27 30 29 32 34 40 33 58 43 49 46 52 60 55 102 103 105 108 107 117 119 123 125 150 133 139 145 130 148 142 Station ID

Figure 4.13: A comparison of winter baseflow with summer baseflow values of (A) the sum of all PAH compounds (as in Fig. 4.1), and (B) the anthracene/(anthracene+phenanthrene) ratio (as in Fig. 4.4), for Phase I and Phase II sites having winterbaseflow data. Sites are arranged in order of increasing basin area by subwatershed within both West of Hudson (WOH) and East of Hudson (EOH) regions as in Figures 4.1 to 4.12. Winter baseflow data are shown as grey symbols, and summer data as unfilled symbols. Symbol shapes as in Figures 4.1 to 4.12.

- 79 - CHAPTER 4 – MOLECULAR TRACERS NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

100 2000 2001 WOH EOH 2002 Increasing 2003 Drainage Area 10 2004 winter summer PAH

H/L 1

0.1

WBD EBD SCH ESP NRD EMC WBC MNC TCS KSC 1 2 9 3 6 8 15 11 10 12 13 16 17 20 18 22 24 23 26 27 30 29 32 34 40 33 58 43 49 46 52 60 55 102 103 105 108 107 117 119 123 125 150 133 139 145 130 148 142 Station ID

Figure 4.14: A comparison of winter baseflow with summer baseflow values of the high to low molecular weight PAH ratio (H/LPAH) (as in Fig. 4.6), for Phase I and Phase II sites having winterbaseflow data. Sites are arranged in order of increasing basin area by subwatershed within both West of Hudson (WOH) and East of Hudson (EOH) regions as in Figures 4.1 to 4.12. Winter baseflow data are shown as grey symbols, and summer data as unfilled symbols. Symbol shapes as in Figures 4.1 to 4.12.

- 80 - CHAPTER 4 – MOLECULAR TRACERS NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

10 2000 2001 WOH EOH 2002 Increasing 1 2003 Drainage Area 2004 g/L) µ 0.1 winter summer

0.01 Caffeine ( (CAF) 0.001

WBD EBD SCH ESP NRD EMC WBC MNC TCS KSC 0.0001 1 2 9 3 6 8 15 11 10 12 13 16 17 20 18 22 24 23 26 27 30 29 32 34 40 33 58 43 49 46 52 60 55 102 103 105 108 107 117 119 123 125 150 133 139 145 130 148 142 Station ID

10 2000 2001 WOH EOH Increasing

g/L) 2002 µ 1 2003 Drainage Area 2004

0.1 winter summer

0.01

0.001 Sum Fragrance Materials (FM) ( Sum Fragrance Materials (FM) WBD EBD SCH ESP NRD EMC WBC MNC TCS KSC 0.0001 1 2 9 3 6 8 15 11 10 12 13 16 17 20 18 22 24 23 26 27 30 29 32 34 40 33 58 43 49 46 52 60 55 102 103 105 108 107 117 119 123 125 150 133 139 145 130 148 142 Station ID

Figure 4.15: A comparison of winter baseflow with summer baseflow concentrations of (A) caffeine (as in Fig. 4.7), and (B) the sum of fragrance materials (FM) tonalide (AHTN) and galaxolide (HHCB) (as in Fig. 4.8), for Phase I and Phase II sites having winterbaseflow data. Sites are arranged in order of increasing basin area by subwatershed within both West of Hudson (WOH) and East of Hudson (EOH) regions as in Figures 4.1 to 4.12. Winter baseflow data are shown as grey symbols, and summer data as unfilled symbols. Symbol shapes as in Figures 4.1 to 4.12.

- 81 - CHAPTER 4 – MOLECULAR TRACERS NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

100 2000 2001 WOH EOH 2002 Increasing 2003 Drainage Area g/L)

µ 10 2004 winter

1 summer

0.1 Sum Fecal Steroids (FS) ( (FS) Steroids Fecal Sum

WBD EBD SCH ESP NRD EMC WBC MNC TCS KSC 0.01 1 2 9 3 6 8 15 11 10 12 13 16 17 20 18 22 24 23 26 27 30 29 32 34 40 33 58 43 49 46 52 60 55 102 103 105 108 107 117 119 123 125 150 133 139 145 130 148 142 Station ID

1.0 2000 2001 WOH EOH 2002 Increasing 0.8 2003 Drainage Area 2004

0.6 winter summer

0.4 bCOP / (bCOP+aCOP) bCOP 0.2

Figure 4.16: A comparison of winter baseflow with summer baseflow concentrations of (A) the sum of seven fecal steroids (FS) – aCOP, aONE, bCOP, bONE, CHO, EPI and SNOL (as in Fig. 4.9), and (B) the coprostanol/(coprostanol+cholestanol) ratio (bCOP/(bCOP+aCOP)) (as in Fig. 4.11), for Phase I and Phase II sites having winterbaseflow data. Sites are arranged in order of increasing basin area by subwatershed within both West of Hudson (WOH) and East of Hudson (EOH) regions as in Figures 4.1 to 4.12. Winter baseflow data are shown as grey symbols, and summer data as unfilled symbols. Symbol shapes as in Figures 4.1 to 4.12.

- 82 - CHAPTER 4 – MOLECULAR TRACERS NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Figure 4.17: Comparison of the sum of PAH concentrations versus discharge in high-tubidity (HT) and high-flow (HF) storm samples and summer baseflow (BF) samples at each of the three storm sites.

- 83 - CHAPTER 4 – MOLECULAR TRACERS NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Figure 4.18: Comparison of ANT/(ANT+PHE) ratios versus discharge in high-tubidity (HT) and high-flow (HF) storm samples and summer baseflow (BF) samples at each of the three storm sites.

- 84 - CHAPTER 4 – MOLECULAR TRACERS NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Figure 4.19: Comparison of (H/L)PAH ratios versus discharge in high-tubidity (HT) and high- flow (HF) storm samples and summer baseflow (BF) samples at each of the three storm sites.

- 85 - CHAPTER 4 – MOLECULAR TRACERS NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Figure 4.20: Comparison of caffeine concentrations versus discharge in high-tubidity (HT) and high-flow (HF) storm samples and summer baseflow (BF) samples at each of the three storm sites.

- 86 - CHAPTER 4 – MOLECULAR TRACERS NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Figure 4.21: Comparison of fragrance material concentrations versus discharge in high-tubidity (HT) and high-flow (HF) storm samples and summer baseflow (BF) samples at each of the three storm sites.

- 87 - CHAPTER 4 – MOLECULAR TRACERS NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Figure 4.22: Comparison of the sum of seven fecal steroids (see Fig. 4.9) versus discharge in high-tubidity (HT) & high-flow (HF) storm samples and summer baseflow (BF) samples at each of the three storm sites.

- 88 - CHAPTER 4 – MOLECULAR TRACERS NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Figure 4.23: Comparison of bCOP/(bCOP+aCOP) ratios versus discharge in high-tubidity (HT) and high-flow (HF) storm samples and summer baseflow (BF) samples at each of the three storm sites.

- 89 - CHAPTER 4 – MOLECULAR TRACERS NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005 ) 2 Mean WOH km- 2000 Increasing Drainage Area -1 2001

s 0.1

2 2002 2003 2004

0.01

0.001

WBD EBD SCH ESP NRD Basin Area normalized Discharge (m Discharge normalized Basin Area 1 7 2 9 3 4 5 6 8 15 14 11 10 12 13 19 16 17 20 18 21 25 22 24 23 26 28 27 30 29 102 151 106 153 104 101 103 105 108 114 160 107 113 111 109 110 112 115 116 118 117 159 121 120 119 122 123 Station ID ) 2 Increasing EOH km-

-1 Drainage Area

s 0.1 2

0.01

0.001

EMC WBC MNC TCS KSC Basin Area normalized Discharge (m Area Basin 39 31 42 32 34 35 40 44 33 37 36 38 41 45 54 58 43 50 51 49 46 48 53 52 47 59 60 56 57 55 124 132 129 125 150 127 126 141 134 140 133 139 137 143 145 146 131 149 130 138 148 147 142 Station ID

Figure 4.24: Discharge, normalized to upstream basin area, on the days that summer baseflow tracer samples were collected for all Phase I and Phase II sites sites both (A) West of Hudson and (B) East of Hudson. Values are calculated from dailly discharge at the nearest USGS gauging station that was scaled to each site by basin area. Geomentric means of all measured values at each site are given by the solid grey line.

- 90 - CHAPTER 4 – MOLECULAR TRACERS NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Chapter 5 - Macroinvertebrate Community

Introduction

This portion of the NY Watersheds study uses naturally occurring benthic (i.e., bottom-dwelling) macroinvertebrate populations in the streams and rivers of the NY Watersheds to assess whether statistically significant and ecologically meaningful differences in environmental quality occur. Benthic macroinvertebrates such as insects, worms, and molluscs are the preferred group of aquatic organisms monitored in water quality assessment programs (Hellawell 1986) because: (1) they provide an extended temporal perspective (relative to traditional water samples that are collected periodically) because they have limited mobility and relatively long life spans (e.g., a few months for some chironomid midges to a year or more for some insects and molluscs); (2) the group has measurable responses to a wide variety of environmental changes and stresses; (3) they are an important link in the aquatic food web, converting plant and microbial matter into animal tissue that is then available to fish; and (4) they are abundant and their responses can be easily analyzed statistically (Weber 1973). Thus, the presence or conspicuous absence of certain macroinvertebrate species at a site is a meaningful record of environmental conditions during the recent past, including ephemeral events that might be missed by assessment programs that rely only on periodic sampling of water chemistry. Most stream ecosystems have relatively diverse macroinvertebrate assemblages with species from a number of different orders [e.g., mayflies (Ephemeroptera), caddisflies (Trichoptera), stoneflies (Plecoptera), beetles (Coleoptera), true flies (Diptera)]. Likewise, the common trophic groups (i.e., herbivores, detritivores, predators) are represented by a number of different species. Various abiotic factors (e.g., hydrology, substrate, temperature, oxygen, pH) and biotic factors (e.g., food quality and quantity, interactions with competitors or predators) have molded, through natural selection, a unique set of optimum environmental requirements for each species. These environmental requirements contribute significantly to the distribution and abundance of these organisms within and among natural stream ecosystems, and influence their response to environmental perturbation.

Aquatic macroinvertebrate species characteristic of the streams and rivers of New York can be typically divided into three subsets based on their period of major growth and activity: (1) species with their principal larval growth during fall - winter - spring and whose adults (in the case of aquatic insects) emerge during spring or early summer (e.g., Ephemerella, Eurylophella, Ameletus, many stoneflies, some Hydropsyche and Cheumatopsyche, Prosimulium); (2) species with their principal larval growth and adult emergence during summer (e.g., Tricorythodes, some Simulium); and (3) species with one or more cohorts per year that include significant larval growth during fall - winter - spring as well as during summer (Baetis, Centroptilum, some Hydropsyche and Cheumatopsyche, Chimarra, many chironomids). Early spring sampling in this program focuses on collecting species near the end of the fall-winter-spring growth cycle, which is when individuals for many species are largest and often easier to identify. In addition, because they have been actively growing and developing the streams since at least the previous September, the presence/absence, absolute abundance, physiological state, etc. of larvae collected in spring integrates both habitat and water quality conditions in a given stream or river over the previous 6-9 months. Thus, the macroinvertebrates collected in spring provide a strong "temporal perspective" during an important and significant portion of the year.

- 91 - CHAPTER 5 – MACROINVERTEBRATES NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Methods

Field Collection of Macroinvertebrate Samples

Macroinvertebrates were collected at 62 locations distributed throughout the watersheds (see Figs. 2.1 and 2.2) between 3 May and 13 May 2004. The sampling protocol was designed to characterize riffle-inhabiting macroinvertebrates in a reach that included several riffles (i.e., for additional habitat and biotic diversity) rather than the approach of characterizing macroinvertebrates from a single riffle or part of a riffle. Reach length varied among streams and rivers, but generally included 20-50 m of riffle. Random sampling locations were chosen based on their longitudinal (e.g., along the length of the study reach) and lateral positions. For example, a sampling location in a stream might be designated as 17-25, which would represent 17 m upstream and 25% across the stream from the bank. The sampling protocol called for a total of four composite samples representing 16 samples to be collected at each site. At 7 of 62 sites (Appendix Table A.5.1), riffle habitat was limited and we modified the sampling design by collecting fewer samples (i.e., 4 or 8 versus 16).

Benthic macroinvertebrates were collected in riffle habitats with a Surber sampler (1 ft2 or 0.093 m2; 0.250-mm mesh) using a quantitative composite sampling regime that was modified from Stroud SOP S-04-09. Sampling started at the downstream end of the sampling area and proceeded in an upstream direction. The operator identified the location of each sampling area based on the longitudinal and lateral position. If boulders or large woody debris interfered with sampling at the designated sampling location, the location was moved slightly until there was no obstruction. If it was impossible to obtain a good sample from this location, an alternative sampling site that was also randomly chosen was used for this sample.

To collect the macroinvertebrate sample, the back edge of the Surber sampler is set on the stream bottom so that there is a tight seal across the substrate to prevent animals from escaping under the sampler. The square bottom frame is then laid out on the stream bottom to delimit the 1 ft2 sample area. Rocks that were under the frame were included in the sample if more than half of the rock was inside the frame; if more than half of the rock was outside of the frame it was not included in the sample. Larger rocks (> 65 mm in longest dimension) are removed individually, and scrubbed with a soft bristled brush under the water in front of the net. Scrubbing removes most attached organisms while the water current moving through the sampler carries these dislodged organisms into the sample net. Each scrubbed rock was placed in a plastic bucket (held by a second person) for subsequent counting. The minimum rock counted and/or measured is > 65 mm on the longest axis. Large rocks that could not be moved were scrubbed in place. After all rocks were scrubbed and removed, the enclosed benthic area was rapidly stirred and agitated for at least 20 seconds to suspend any residual organisms in the water column and subsequently into the sample net. The sampler was then removed from the bottom and stream water splashed onto the outside of the net in order to wash clinging animals into the bottom of the net. Each sample was randomly assigned to one of four composite samples so the net for a sample was inverted and the contents washed into a plastic bucket designated for that composite sample.

Composite samples resulted from combining four 1ft2 samples (if possible) into one composite sample (i.e., containing macroinvertebrates from 4 ft2) and then subsampling the combined samples in the field such that a subsample equaled one sample (i.e., macroinvertebrates

- 92 - CHAPTER 5 – MACROINVERTEBRATES NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

representative of 1 ft2). After all samples (usually 16) had been collected and combined into four composite samples, each composite sample was split into subsamples (each representing 1 ft2), with one of the subsamples being preserved and brought back to the laboratory for analysis. Each composite sample was washed into a large sample splitter that was placed in a large plastic trash can half filled with water. The mixture of macroinvertebrates, detritus, and sediments was homogenized and resuspended by stirring, agitating, and pushing water into the subsampler. The material then resettled across the bottom of the subsampler while slowly drawing the subsampler out of the barrel. If the material did not appear evenly distributed, the resuspension and settling process was repeated. The net (0.250-mm mesh)-covered bottom was separated from the rest of the subsampler, and the + shaped plastic separator was pushed into the sample material, dividing the material into four equal parts. A spatula and scissors was used to separate subsamples and transfer a subsample to a labeled sample jar filled with 5% buffered formalin, which was then transported to the laboratory. If the composite sample contained four samples, then 1/4th of the composite material represented macroinvertebrates from 1 ft2. If only eight samples were collected, then each composite sample contained the contents of two samples (i.e., macroinvertebrates from 2 ft2), and the composite sample was split into two subsamples (each representing 1 ft2).

Sample compositing has advantages because it can increase both accuracy and precision relative to standard (non-compositing) macroinvertebrate sampling. For example, compositing increases the accuracy of the desired description by increasing the number of samples collected and therefore the area sampled in these riffles without increasing the number of samples processed. At the same time, compositing homogenizes spatial variation when these samples are combined, which reduces variance among samples in statistical analyses.

Associated with each sample, water depth was measured to the nearest cm and current velocity was estimated with a current meter set at a point 0.6 of the distance from the bottom to the water surface. The number of large rocks (> 65 mm in longest dimension) that had been in that sample was also recorded. Periphyton biomass (as chlorophyll a and ash free dry mass) was measured for each composite sample by collecting a small algae-covered stone (3-5 cm in diameter) near where each sample were collected and placing it in labeled plastic Tupperware containers associated with each composite sample (i.e., 2 or 4 rocks per composite sample). The plastic Tupperware containers were stored on dry ice (in field) or in a freezer (in laboratory) until chlorophyll a and ash free dry mass analyses were completed in the laboratory (< 30 d for chlorophyll a).

Laboratory Processing of Macroinvertebrate Samples

Benthic materials (i.e., macroinvertebrates and detritus) were transferred from the sample jar into a 0.250-mm mesh sieve and rinsed thoroughly with water to remove fine particles. Because macroinvertebrates were abundant (hundreds to thousands per sample), each sample was split into four subsamples, and then one of those subsamples was split into four subsamples (i.e., 1/16th of a sample). Actual subsample size processed varied among samples (1/16, 1/8, 3/16, 1/4) and reflected the number of macroinvertebrate per sample. Our target was to identify 100-300 macroinvertebrates per subsample. Macroinvertebrates were separated from detritus by taking a small portion from the subsample and placing it in a plastic sorting tray partially filled with 80% ethanol. This material was then carefully examined with the aid of a dissecting microscope (12 X

- 93 - CHAPTER 5 – MACROINVERTEBRATES NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

magnification). All macroinvertebrates were removed from the detrital material collected in the subsample, and the detrital material was transferred to an aluminum weigh boat (see Benthic Organic Matter below).

Aquatic insects were generally identified to genus or species; other macroinvertebrates (e.g., crustaceans, mites, flatworms, oligochaetes, and nematodes) were commonly left at higher taxonomic levels (e.g., order, family). Specimens that were damaged or extremely small were identified to the taxonomic levels possible, but these were higher than species and even genus. Chironomids were subsampled before identification, and the number examined represented the percentage of chironomids in that sample. For example, if a sample contained 300 macroinvertebrates and 40% of them were chironomids, then 40 chironomids were identified to genus/species and these identifications were applied proportionally to the remaining 80 chironomids. Identified macroinvertebrates were placed in vials containing 80% ethanol and a permanent label containing the appropriate information (project name, project number, study site, sampling device, sample number sample date, name of individual who sorted and identified sample). Macroinvertebrate specimens (sorted and unsorted material) are being archived by the Stroud Center for at least 10 years after the collection date. After verification, selected voucher specimens may be incorporated into the permanent macroinvertebrate collection at the Stroud Center.

Periphyton chlorophyll a and biomass were estimated for rocks collected in association with each composite sample. For chlorophyll a analyses, rocks were extracted overnight in alkaline acetone and optical densities determined at 665 nm and 750 nm (for turbidity) before and after acidification with a drop of 1 N HCL. Optical densities were used to determine chlorophyll a concentrations with correction for phaeophytin (Lorenzen 1967). These rocks were then scrubbed with small brushes to remove attached organic material (i.e., the biofilm of algae, fungi, and bacteria). This organic material was captured on a pre-ashed GF/F filter, dried at 60 °C for >48 h, weighed (dry mass of organic and inorganic matter on rock surfaces), ashed at 550°C for 5 h, and then weighed again (dry mass of inorganic materials). Weight loss during ashing represents the organic content of the periphyton expressed as mg or g AFDM/m2. Periphyton chlorophyll a and biomass are measures of the biofilm that represents macroinvertebrate food attached to rocks

Benthic Organic Matter (BOM) is also a measure of macroinvertebrate food, but in the form of medium and coarse organic particles (i.e., captured by a 0.250 mm mesh sieve) intermixed among rocks and finer substrates in the stream bed. BOM was estimated as the detrital material associated with each processed subsample. After the macroinvertebrates were removed, the wet detritus (organic and inorganic material) was transferred to an aluminum weigh boat and dried at 60 °C for > 48 h. The sample was weighed (dry mass of organic and inorganic materials), ashed at 550°C for 5 h, and then weighed again (dry mass of inorganic materials). Weight loss during ashing represents BOM expressed as mg or g AFDM/m2.

QA/QC of Macroinvertebrate Data

Errors for macroinvertebrate data were measured three ways: sorting errors, identification/count errors, and identification accuracy. Sorting error (or efficiency) was measured on 12 samples (the number of samples required by the QA/QC Plan) by resorting through the processed detrital

- 94 - CHAPTER 5 – MACROINVERTEBRATES NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

material looking for macroinvertebrates that were not found in the first sort. Sorting error was reported as the number of individuals found expressed as a percentage of the total number of macroinvertebrates found for a sample). Error in macroinvertebrates identifications and counts was estimated by reexamining the specimens identified in 12 samples (the number of samples required by the QA/QC Plan). Errors arose due to incorrect identifications or counts or placing an individual in the wrong vial. Error in macroinvertebrate identification or count was reported as the number of mistakes expressed as a percentage of the total number of macroinvertebrates identified. Finally, identification accuracy was assessed by sending voucher specimens for each genus and/or species to be verified at the Aquatic Resource Center, Inc, 545 Cathy Jo Circle, Nashville, TN 37211.

All macroinvertebrate and associated data was compiled into SAS data sets. The contents of these data sets were then compared with original laboratory or field data sheets, with 100% of the data being proofread and any discrepancies being corrected.

Status of Macroinvertebrate Samples

All 248 macroinvertebrate samples collected in 2004 have been processed, and identifications have been completed.

QA/QC review of the macroinvertebrate data was carried out as the samples were processed and thus has been completed. Sorting error ranged from 1.57% and 17.49% (Appendix Table A.5.2), with only one sample exceeding the 15% limit defined in the QA/QC Plan. The overall sorting error rate across the 12 samples was 8.20%, which was below the 15% limit defined in the QA/QC Plan. There was a single sample that had a sorting error rate >15% but this was the only sample processed by this individual. Error in macroinvertebrate identification or count ranged from 0 to 1.0% (Appendix Table A.5.3), with none of the samples exceeding the 10% limit defined in the QA/QC Plan. The overall error in identifications or counts across the 12 samples was 0.2%, which was well below the 10% limit defined in the QA/QC Plan. Finally, the voucher specimens have been sent to the Aquatic Resource Center, and we expect their assessment within a month. Any serious discrepancies (i.e., species-level differences) will be resolved by our senior entomologist (25 years of experience).

All macroinvertebrate and associated data was compiled into SAS datasets, with 100% of the data being proofread and corrected. Statistical analyses programs that compare macroinvertebrate assemblages among stations have been developed and some initial results are provided below. The results presented in this progress report address the metrics that are used to calculate the NY Water Quality Score (Bode et al. 2002b). This includes:

Total Richness Total Richness summarizes species responses (as presence/absence but not abundance) of all taxa, including pollution-sensitive and pollution-tolerant taxa. It is reported as the mean number of aquatic macroinvertebrate species found in each subsample. Total Richness generally decreases in response to moderate to severe pollution. Total Richness is often split into EPT Richness and Chironomid Richness.

EPT Richness EPT Richness is reported as the mean number of Ephemeroptera, Plecoptera, and Trichoptera species found in each subsample. These three insect orders contain many pollution-

- 95 - CHAPTER 5 – MACROINVERTEBRATES NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005 sensitive taxa; thus, this metric summarizes responses of mostly pollution-sensitive taxa. EPT Richness generally decreases in response to moderate to severe pollution.

HBI Hilsenhoff Biotic Index (HBI) - Analyses involving abundance (i.e., density) or presence/absence (richness) are only able to incorporate pollution tolerance information indirectly, through the interpretation of results for individual taxa or groups of taxa. Biotic indexes combine abundance data and pollution tolerance values for each taxon to form a weighted average for the aquatic macroinvertebrates at that site. A biotic index is estimated with data from each sample, and summarized as a mean per sample. Tolerance values (values range from 0 to 10, with 10 being most tolerant and 0 being least tolerant of pollution) for the Hilsenhoff Biotic Index were obtained from two sources: NY QA/QC 2002, and unpublished data obtained from US EPA

PMA Percent Model Affinity - PMA compares the observed distribution of individuals among seven orders with a hypothetical macroinvertebrate community representing an unimpacted macroinvertebrate assemblage. The model community consists of 40% Ephemeroptera, 5% Plecoptera, 10% Trichoptera, 10% Coleoptera, 20% Chironomidae, 5% Oligochaeta, 10% Other Taxa. The PMA is calculated by comparing values for each taxonomic group from the model and observed communities, and taking the sum of the smaller of the two values from each taxonomic group.

WQS Water Quality Score - The values for each of the four metrics (Total Richness, EPT Richness, HBI, and PMA are converted to a WQS (range = 0-10) using the Biological Assessment Profile in Bode et al. (2002b). The WQS for the site is the mean of the WQSs for the four individual indexes. Based on data collected with a kick sampler (0.8 x 0.9 mm-mesh) between July and September, a WQS of 7.5-10 indicates no impact, 5.0-7.5 indicates slight impact, 2.5-5.0 indicates moderate impact, and 0.0-2.5 indicates severe impact. The applicability of this system to other sampling designs (e.g., different sampling efforts or different seasons) remains unknown.

Total Richness, EPT Richness, and HBI all have a long history in water quality monitoring. PMA is less commonly used. While numerous multimetric indexes have been developed for stream macroinvertebrate assemblages and are widely used in water quality monitoring, the Biological Assessment Profile used to calculate a Water Quality Score and to assess water quality impact have been developed specifically for New York streams by New York State Department Environmental Conservation (NYS DEC) in Bode et al. (2002b).

Because the total number of individuals identified differed greatly among our samples and always exceeded the 100 individuals that are standard in the NYS DEC protocol, we used a rarefaction process to produce standardized samples. Standardized samples were created by randomly resampling (without replacement) 100 individuals from each raw sample, and individual measures of community structure (i.e., Total Species Richness, EPT Richness, HBI, PMA) were calculated from this standardized sample. The resampling process was repeated 1000 times for each sample, and the means of the 1000 replicates were used to calculate the WQS for that sample.

- 96 - CHAPTER 5 – MACROINVERTEBRATES NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Modifications to the QA/QC Plan for 2005

Overall, the field and laboratory protocols outlined in the 2004 QA/QC Plan worked well, and we do not suggest any changes to the QA/QC Plan for 2005 at this time. Other modifications (e.g., the addition of sites) may become apparent after the sites have been sampled and data have been analyzed.

Results and Discussion

Range of conditions across sites in 2004

Macroinvertebrate communities sampled in 2004 varied greatly across the New York City drinking water watersheds. The range of conditions was similar to that observed in Phase I (years 1-3) of this study. Biological metrics (i.e., Species Richness, EPT Richness, HBI, PMA, and WQS) indicated that the macroinvertebrate communities at the 62 sites represented a continuum of conditions from relatively high water and habitat quality to relatively low water and habitat quality (see Table 5.1 for values for each site).

Water Quality Score ranged from a high of 9.1 at Site 121 (Warner Creek nr Chichester) to a low of 3.6 at Site 138 (Cross R. nr Katonah) (Table 5.1, Fig. 5.1). Over half of the sites (33 of 62) had a WQS characteristic of no impact, 19 sites had a WQS characteristic of slight impact, and 10 sites had a WQS characteristic of moderate impact. The 9 of the 10 sites with the highest WQS were WOH sites, while the 10 sites with the lowest WQS were all EOH sites (Table 5.2). A similar pattern was observed for each of the four components of the WQS [i.e., Species Richness (Fig. 5.2); EPT Richness (Fig. 5.3); HBI (Fig. 5.4); PMA (Fig. 5.5)], where the majority of the sites with the highest scores were WOH, while the sites with the lowest scores were generally EOH.

The degree and frequency of impairment was less among sites in the WOH region. Twenty-six of the 35 WOH sites had WQSs of 7.5 or higher, indicating no impact. Nine WOH sites fell in the slightly impaired range (7.55). No WOH sites had WQSs in the moderate or severely impaired ranges (WQS<5). The five sites in the WOH with the highest WQS were (in descending order): Site 121 - Warner Creek nr Chichester, Site 10 - E. Br. Delaware nr Arkville, Site 106 - Dryden Brook nr Beerston, Site 160 - Emory Brook at Fleischmens, and Site 105 - East Brook nr Walton (Table 5.3). Sites with lowest WQS in the WOH were (in ascending order): Site 117 - Batavia Kill nr Windham, Site 26 - Esopus Creek nr Mount Tremper, Site 122 - W. Br. Neversink nr Frost Valley, Site 29 - Neversink R. nr Claryville, and Site 116 - East Kill nr East Jewett (Table 5.3). Seven sites in the EOH region were classified as non-impaired with WQSs at or above 7.5, ten EOH sites were in the slightly impaired range (7.5 < WQS > 5), and ten in the moderately impaired range (5 < WQS > 2.5). No sites were classified as severely degraded in the EOH. The five EOH sites with the highest WQS were (in descending order): Site 125 - Quaker Brook at Merrit County Park, Site 52 - Cross R. in W.P.R. Reservation, Site 34 - Haviland Hollow Br. at Haviland Hollow, Site 124 - trib. E. Br. Croton nr Pawling, and Site 149 - Waccabuc R. at Boutonville (Table 5.4). The five EOH sites with the lowest WQS were (in ascending order): Site 138 - Cross R. nr Katonah, Site 148 - trib. of Kensico Res. nr Thornwood, Site 140 - Hunter Brook nr Yorktown, Site 133 - trib. of Muscoot R. at Mahopac Falls, and Site 145 Broad Brook nr Bedford Falls (Table 5.4).

- 97 - CHAPTER 5 – MACROINVERTEBRATES NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Total macroinvertebrate densities in 2004 ranged from 4545/m2 at Site 108 (Scudders Run nr Roxbury Run) to 61,699/m2 at Site 138 (Cross R. nr Katonah) (Table 5.5). Total densities ranged from 9556/m2 to 61,699/m2 among moderately impacted sites, 8642/m2 to 43,957/m2 among sites classified as slightly impacted and 4545/m2 to 57,462/m2 among sites classified as non-impacted. Density for several major taxonomic groups correlated significantly with WQS (Table 5.6). Density of EPT (r = 0.59), Ephemeroptera (r = 0.73), Plecoptera (r = 0.56), Trichoptera (r = 0.40), and Coleoptera (r = 0.34) had significant positive relationships with WQS. Total density of Oligochaeta (r = -0.31), Noninsect (which were primarily Oligochaeta; r = -0.21), and Diptera (r = -0.48) had significant negative relationships with WQS. We did not find significant relationships between Total Macroinvertebrate density and WQS or Total Insect density and WQS.

Integrative Sites in 2004

The integrative sites in Phase II included eight sites continued from Phase I (i.e., Sites 3, 9, 10, 15, 29, 34, 46, 52) and two new sites (i.e., Sites 130 and 139). The integrative sites were representative of the range of conditions among all 62 Phase II sites (Fig. 5.6): six sites were classified as non-impacted (WQS at Site 10 = 9.0, Site 15 = 8.7, Site 52 = 8.5, Site 34 = 8.4, Site 9 = 8.3, and Site 3 = 8.2), three were classified as slightly impacted (WQS at Site 29 = 6.8, Site 46 = 6.3, and Site 139 = 5.7), and one was classified as moderately impacted (WQS at Site 130 = 4.4).

Phase II Sites compared to Phase I sites

Fig. 5.7 shows the range of WQSs for WOH sites in Phase II 2004 combined with average WQSs for WOH sites in Phase I 2000-2002. Phase II in 2004 included a few WOH sites toward the more impacted end of the range of WQSs (i.e., Sites 123, 116, 122, and 117). However, none of these sites were classified in the moderately or severely impacted categories. Thus, the WOH sites represent only part of the conditions represented in the EOH. EOH Phase II sites also spanned the range of the average WQSs from Phase I (2000-2002) (Fig. 5.8). Several Phase II EOH sites in 2004 were non-impacted sites (i.e., Sites 125, 124, 127, 149, and 129).

The gray bars in Figs. 5.7 and 5.8 are WQSs from the 12 sites (Sites 3, 6, 9, 10, 15, 23, 26, 29, 34, 46, 52, 55) that were common to both Phase I and II of this 6-year project. These sites represent the range of conditions seen in both the EOH and WOH regions in 2004. Overall, these sites address three of the four water quality classification categories: non-impact, slight impact, and moderate impact (no sites scored in the severe impact category in 2004). WQSs from 2004 overlapped or were within the range of scores found at most of the 12 sites in 2000 to 2002 (Fig. 5.9). The 2003 and 2004 WQSs were lower than 2000-2002 years at Site 9 (Trout Creek nr Trout Creek) and Site 29 (Neversink R. nr Claryville). The Trout Creek site was classified as non- impacted for all five years. The Neversink R. site was classified as non-impacted from 2000 to 2003 (WQS ≤7.5), but as slightly impacted in 2004 (WQS = 6.8). At Site 15 (Tremper Kill nr Andes) and at Site 52 (Cross R. in Ward Pound Ridge Resv), the 2003 and 2004 WQSs were higher than in 2000-2002. Interannual variability in WQS was greatest at Site 46 (Muscoot R. nr Baldwin Place) and Site 55 (Kisco R. nr Stanwood). For example, WQSs ranged from 3.3 to 6.7 at Site 46 and from 3.5 to 6.3 at Site 55 during the five years. Overall, for the five years of data

- 98 - CHAPTER 5 – MACROINVERTEBRATES NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

there were two sites (Sites 9 and 34) classified as non-impacted, eight sites (Sites 3, 6, 10, 15, 23, 26, 29, and 52) classified as non- to slightly impacted, and two sites (Sites 46 and 55) classified as slightly to moderately impacted (Fig. 5.9).

Comparisons to reference conditions and within watershed comparisons in 2004

Reference conditions were defined as the three sites with the highest WQS in WOH and the EOH (top 10% of sites in each region). The three WOH reference sites were: Site 121 - Warner Creek nr Chichester (WQS = 9.1), Site 10 - E. Br. Delaware R. nr Arkville (WQS = 9.0), and Site 106 - Dryden Brook nr Beerston (WQS = 9.0). The three EOH reference sites were: Site 125 - Quaker Brook at Merrit County Park (WQS = 9.1), Site 52 - Cross R. in Ward Pound Ridge Resv (WQS = 8.5), and Site 34 - Haviland Hollow Brook at Haviland Hollow (WQS = 8.4). These sites were compared with sites in each watershed (ANOVA with Tukey’s means comparison test, see Fig. 5.10 for Tukey’s results by watershed). These reference sites and analyses were specific for the 2004 analyses.

The a priori hypothesis in these analyses was that the macroinvertebrate assemblages at potentially impacted sites would not differ from the macroinvertebrate assemblages at Sites 121, 10, and 106 in the WOH and Sites 125, 52, and 34 in the EOH. Evidence of a negative impact was defined as a difference in the macroinvertebrate assemblage that resulted in a significantly lower WQS at a potentially impacted site relative to all three reference sites. This difference in the macroinvertebrate assemblage may reflect lower Total Richness, lower EPT Richness (i.e., primarily pollution-sensitive species), higher HBI (i.e., lower relative abundance of pollution- sensitive groups such as mayflies), and/or a reduction in the Percent Model Affinity (i.e., the similarity of the macroinvertebrate community structure relative to the model community). In our interpretation of the quantitative data, differences between a potentially impacted site and the reference sites must be parallel. Differences that were significant for only one or two of the reference sites were not considered evidence of environmental change potentially because these differences did not exceed natural variation observed among three reference sites.

In the WOH, 13 of 35 sites (37%) had WQSs that were statistically lower than at all three reference sites. And, in the EOH, 20 of 27 sites (74%) had WQSs that were statistically lower than at the three reference sites. The larger proportion of EOH sites with WQSs less than reference sites illustrates the greater range in WQSs in the EOH region. WQSs in the slightly and moderately impacted categories were usually significantly less than all three reference sites. For five cases, sites classified in the non-impacted category had a WQS that was statistically lower than the reference sites (i.e., Sites 101 and 151 in Cannonsville watershed, Site 107 in Pepacton watershed, Site 118 in Schoharie watershed, and Site 23 in the Ashokan watershed).

WOH Watersheds in 2004

Half of the WOH watersheds had sites with WQSs that were significantly lower than at the three reference sites, and with the exception of the Cannonsville and Rondout watersheds, all watersheds had sites within that watershed that differed from each other. Occasionally, one site within each watershed had a WQS that was significantly lower than at all other sites in the watershed.

- 99 - CHAPTER 5 – MACROINVERTEBRATES NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Cannonsville Watershed – Reference Site 106 is in the Cannonsville watershed. In the Cannonsville watershed, Site 151 - Betty Brook nr South Kortright (WQS = 7.7) was the only site that had a lower WQS than the WOH reference sites. However, all 11 sites in the Cannonsville watershed had statistically similar WQSs and were all classified as non-impacted.

Pepacton Watershed – Reference Site 10 is in the Pepacton watershed. WQSs at the Pepacton sites ranged from 9.0 at Site 10 to 7.3 at Site 111 (Dry Brook nr Mapledale). Relative to the three WOH reference sites, WQSs were significantly lower at Site 111, Site 112 (Mill Brook nr Grant Mills) and Site 107 (E. Br. Delaware at Roxbury). Sites 112 and 111 were the only sites in the Pepacton classified as slightly impacted, all other sites were considered non-impacted.

Schoharie Watershed – All four sites in the Schoharie watershed had lower WQSs than the WOH reference sites. Site 117 (= 5.1) had the lowest WQS in the WOH and was considered slightly impacted. The remaining three sites (Sites 118, 115 and 116) were all statistically similar but Site 118 was considered non-impacted, whereas Sites 115 and 116 were classified as slightly impacted.

Ashokan Watershed – Reference Site 121 is in the Ashokan watershed. Site 23 (Esopus Creek nr Allaben) and Site 26 (Esopus Creek nr Mount Tremper) had WQSs that were statistically lower than at the WOH reference sites. Site 26 had a lower WQS than all other Ashokan sites and was classified as slightly impacted (WQS = 6.3). All other Ashokan sites were considered non- impacted.

Neversink Watershed – Two sites were sampled in the Neversink watershed: Site 122 (W. Br. Neversink above Frost Valley) and Site 29 (Neversink R. nr Claryville). Both of these sites were classified as slightly impacted (WQS = 6.8) and WQSs were statistically lower than the WOH reference sites.

Rondout Watershed – Only one site (Site 123 - Rondout Creek nr Peekamoose) was sampled in the Rondout watershed. The WQS at Site 123 (= 7.2) was significantly lower than the three WOH reference sites, which classified the site as slightly impacted.

EOH Watersheds in 2004

Five of the six EOH watersheds had sites with WQSs that were lower than at the EOH reference sites. In four of these watersheds, WQSs at all sites were lower than at the reference sites.

Middle & West Br. Croton Watershed – One site (Site 126 - Stump Pond Stream nr Pawling) in the Middle Br. Croton watershed was sampled in 2004 and it was statistically lower than the reference sites. One site (Site 127 - Black Pond Brook at Meads Corner) was sampled in the West Br. Croton watershed in 2004 and it was not different from the reference sites. Site 126 was regarded as slightly impacted and Site 127 was labeled non-impacted.

East Branch Croton Watershed – A total of five sites were sampled in the East Branch of the Croton watershed, and this included all two EOH reference sites (i.e., Sites 34 and 125). In the East Br. Croton watershed, WQSs at Site 129 (Unnamed trib. of E. Br. Croton R.), Site 132 (Bog Brook nr Sears Corner), and Site 150 (E. Br. Croton R. at Brewster) were lower than the

- 100 - CHAPTER 5 – MACROINVERTEBRATES NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

reference sites. Site 150 had the lowest WQS (= 5.1), followed by Site 132 (WQS = 6.3), then by Site 129 (WQS = 7.5). Site 124 (Unnamed trib. of E. Br. Croton R. nr Pawling) had a WQS similar to the reference sites. Sites 150 and 132 were classified as slightly impacted and the remaining E. Br. Croton sites were classified as non-impacted.

East & South of the Croton Watershed – A total of 12 sites were sampled in the East and South of the Croton watershed, and this included one EOH reference site (i.e., Site 52). Only Site 149 (Waccabuc R. at Boutonville) and Site 146 (Stone Hill R. nr Bedford) had WQSs similar to the reference sites. Site 149 was rated non-impacted and Site 146 was considered slightly impacted. The remaining nine sites (Site 141 - Unnamed trib. of Croton Res. nr Croton Heights, Site 143 - Unnamed trib. to Cross R. nr Cross R., Site 55 - Kisco R. nr Stanwood, Site 131 - Titicus R. nr North Salem, Site 137 - Unnamed trib. of Muscoot Res. nr Goldens Br., Site 130 - Titicus R. nr Salem Center, Site 142 - Kisco R. nr Bedford, Site 145 - Broad Brook nr Bedford Hills, and Site 138 - Cross R. nr Katonah) had WQSs that were lower than the reference sites. These sites had WQSs that ranged from a high of 5.7 at Site 141 to a low of 3.6 at Site 138, which was also the lowest score observed among EOH sites in 2004. This indicates that the majority of sites in the East and South of the Croton watershed were categorized as slightly (e.g., Sites 146, 141, 143, and 55) to moderately impacted (e.g., Sites 131, 137, 130, 142, 145, and 138).

North of the Croton Watershed – All five sites in the North of the Croton watershed had WQSs that were lower than the EOH reference sites (i.e., Site 46 - Muscoot R. nr Baldwin Place, Site 139 - Muscoot R. nr Whitehall Corners, Site 134 - Plum Brook at Shenorock, Site 133 - Unnamed trib. of Muscoot R. at Mahopac Falls, and Site 140 - Hunter Brook nr Yorktown). Sites 46 and 139 were classified as slightly impacted, whereas Sites 134, 133, and 140 were classified as moderately impacted.

Kensico R. Watershed – WQSs at the two Kensico R. watershed sites (Site 147 - Unnamed trib. of Kensico Res. at Mt Pleasant and Site 148 - Unnamed trib. of Kensico Res. nr Thornwood) were lower than at the EOH reference sites. In addition, Site 148 was lower than Site 147. Site 147 (WQS = 6.2) was classified as slightly impacted and Site 148 (WQS = 3.7) was classified as moderately impacted.

WQS related to landscape variables

Principal Components Analysis (Fig. 2.5) separated WOH and EOH sites based on a suite of landscape variables (i.e., watershed area, percent forest cover, number of active SPDES permits, etc.), thus revealing markedly different anthropogenic impact gradients. EOH sites fell out along a high population density/percent impervious cover to high percent forest gradient. WOH sites oriented vertically with Factor 2, explained largely by percent agriculture and percent forest cover. To examine whether macroinvertebrate assemblages might be responding to these gradients, we ran simple linear regressions between WQS in 2004 and the landscape variables describing these gradients. Among EOH sites, there was a significant negative relationship between WQS and population density (r2 = 0.34, p = 0.001; Fig. 5.11) and between WQS and percent impervious (r2 = 0.51, p < 0.0001; Fig. 5.12). Among WOH sites, WQS related to percent agriculture (r2 = 0.21, p = 0.007; Fig. 5.13) and percent forest (r2 = 0.20, p = 0.008; Fig. 5.14).

- 101 - CHAPTER 5 – MACROINVERTEBRATES NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Biological Metrics in 2003 & 2004

Individual indexes as well as the WQSs from 2003 and 2004 were compared across the 60 sites (Sites 159 and 160 excluded since they were not samples in 2003). Overall metrics suggest that 2003 and 2004 data were similar (Fig 5.15). Specifically, Total Richness was similar in 2003 and 2004 (r2 = 0.61). Average Total Richness was 27 taxa/100 individuals in 2003 and 24 taxa/100 individuals in 2004. Total Richness ranged among sites from a low of 12 taxa/100 individuals to a high of 40/100 individuals. EPT Richness was also similar in 2003 and 2004 (r2 = 0.78), for both years the average EPT Richness was 9 taxa/100 individuals. HBI also varied little between years (r2 = 0.67): HBI was 4.6 in 2003 and 4.5 in 2004. Average PMA was 60% in 2003 and 64% in 2004 (r2 = 0.64). This slight difference may suggest a general improvement in stream quality or natural variation. As expected from the previous metrics, WQS was very similar between 2003 and 2004 (r2 = 0.84) and averaged 7.0 in 2003 and 7.1 in 2004 (Fig. 5.15).

Comparisons to reference conditions and within watershed comparisons in 2003 & 2004

Data from 2003 and 2004 were combined to give an “average” (i.e., based on two years) assessment of conditions at each site and differences among sites. Use of multi-year data sets to derive site classifications reduces the errors in site classification that result from natural or anthropogenic contributions to annual variation. Based on the combined 2003 and 2004 data, the three (i.e., top 10%) WOH sites with the highest average WQS were: Site 121 - Warner Creek nr Chichester (WQS = 9.2), Site 106 - Dryden Brook nr Beerston (WQS = 9.1), and Site 15 - Tremper Kill nr Andes (WQS = 8.9). The three reference sites for the analysis of the EOH watersheds based on the combined 2003 and 2004 data were: Site 125 - Quaker Brook at Merrit County Park (WQS = 8.6), Site 52 - Cross R. in Ward Pound Ridge Resv (WQS = 8.4), and Site 34 - Haviland Hollow Brook at Haviland Hollow (WQS = 8.2).

WOH Watersheds in 2003 & 2004

In 2003, 15 of 33 WOH sites exhibited statistically significant differences relative to the WOH reference sites (Stroud 2004) and in 2004, 14 of 35 WOH sites exhibited statistically significant differences relative to the WOH reference sites. But when 2003 and 2004 were averaged only 5 of 35 WOH sites (e.g., Sites 3, 117, 23, 26, 123) exhibited statistically significant differences relative to the WOH reference sites (Fig. 5.16). The number of significant differences between WOH sites and reference sites was reduced using both years of data. In 2003-2004, no differences were observed in the Pepacton and Neversink watersheds, even though differences relative to reference sites were observed in 2003 and 2004 when the years were analyzed separately. Differences for 2003-2004 combined data relative to reference sites are listed below:

Cannonsville Watershed – Reference Site 106 is in this watershed. Only one site (e.g., Site 3) of the 11 sites in the Cannonsville watershed differed from the reference sites for 2003-2004. Site 3 (W. Br. Delaware R. at South Knotright) was labeled non-impacted (WQS = 7.5).

Pepacton Watershed – Reference Site 15 is in this watershed. No differences relative to reference sites for 2003-2004.

- 102 - CHAPTER 5 – MACROINVERTEBRATES NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Schoharie Watershed – Only one site (e.g., Site 117) of the four sites in the Cannonsville watershed differed from the reference sites for 2003-2004. Site 117 (Batavia Kill nr Windham) was considered slightly impacted (WQS = 5.7).

Ashokan Watershed – Reference Site 15 is in this watershed. Two sites (e.g., Sites 23 and 26) in the Asokan watershed differed from the reference sites. Site 23 (Esopus Creek nr Allaben) and Site 26 (Esopus Creek nr Mount Tremper) were classified as slightly impacted (WQS = 7.4 and 6.1, respectively)

Neversink Watershed – No differences relative to reference sites for 2003-2004.

Rondout Watershed – The only site in Rondout watershed, Site 123, was different from reference sites for 2003-2004. Site 123 (Rondout Creek nr Peekamoose) was considered slightly impacted (WQS = 7.2).

EOH Watersheds in 2003 & 2004

In 2003, 18 of 27 EOH sites exhibited statistically significant differences relative to the EOH reference sites (Stroud 2004) and in 2004, 20 of 27 EOH sites exhibited statistically significant differences relative to the EOH reference sites. When 2003 and 2004 were averaged 17 of 27 EOH sites exhibited statistically significant differences relative to the EOH reference sites (Fig. 5.16). Averaging 2003 and 2004 data did little to reduce the number of significant differences between EOH sites and reference sites. No differences were observed in the Middle Branch and West Branch of the Croton watersheds. Differences relative to reference sites are listed below:

Middle Br. Croton Watershed – No differences relative to reference sites for 2003-2004.

East Branch Croton Watershed – Reference Sites 34 and 125 are in this watershed. Only one site (e.g., Site 150) in the East Branch Croton watershed differed from the reference sites for 2003- 2004. Site 150 (E. Br. Croton R. at Brewster) was considered moderately impacted (WQS = 4.9).

West Br. Croton Watershed – No differences relative to reference sites for 2003-2004.

East & South of the Croton Watershed – Reference Site 52 is in this watershed. Nine of the 12 sites in East & South of the Croton watershed (e.g., Sites 143, 141, 131, 137, 142, 145, 55, 130, and 138) differed from the reference sites for 2003-2004. These sites had WQSs that ranged from 3.8 at Site 138 (Cross R. nr Katonah) to 5.9 at Site 143 (Unnamed trib. to Cross R. nr Cross R.). Of the nine sites, two were classified as slightly impacted (e.g., Sites 143 and 141) and seven (e.g., Sites 131, 137, 142, 145, 55, 130, and 138) were considered moderately impacted.

North Croton Watershed – All sites in the North Croton watershed (e.g., Sites 139, 46, 134, 140, and 133) differed from the reference sites for 2003-2004. These sites had WQSs that ranged from 4.0 at Site 133 (Unnamed trib. of Muscoot R. at Mahopac Falls) to 5.9 at Site 139 (Muscoot R. nr Whitehall Corners). In the North Croton watershed two sites (e.g., Sites 139 and 46) were considered slightly impacted and three sites (e.g., Sites 134, 140 and 133) were classified as moderately impacted.

- 103 - CHAPTER 5 – MACROINVERTEBRATES NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Kensico R. Watershed – Both sites in the Kensico R. watershed differed from the reference sites. Site 147 (Unnamed trib. of Kensico Res. at Mt Pleasant) was classified as slightly impacted (WQS = 5.9) and Site 148 (Unnamed trib. of Kensico Res. nr Thornwood) was considered moderately impacted.

Overall, the 2003-2004 data classified 32 of the 56 non-reference sites as similar to the reference sites and 22 as different. Of those different sites, one site was considered non-impacted, nine sites were considered slightly impacted, and 12 as moderately impacted. No sites in 2003-2004 were considered severely impacted. Overall, the results from the analysis of the combined 2003 and 2004 data indicate that greater differences relative to reference conditions were statistically significant with only two years of data, although the degree of difference (i.e., the degree of impairment) can vary greatly between years. Most of the significantly impaired sites were EOH sites. In contrast, the degree of differences at most WOH sites (and some EOH sites) relative to reference conditions were smaller and not statistically significant in analyses with only two years of data. Additional data (i.e., more years and/or greater replication within a year) are needed to confidently describe the degree of difference and to resolve some of these differences statistically.

Comparison with Recent NYS DEC Data

The WQS protocol developed by NYS DEC summarizes data from 100 macroinvertebrates collected from riffles in July-September with a kick net (0.8 x 0.9-mm mesh) into a single water quality score, and then classifies the site primarily based on that score (Bode et al. 2002b). In our study, WQSs were calculated based on approximately 200 macroinvertebrates from each of four quantitative samples collected from riffles in April/May with a Surber sampler (0.250-mm mesh). The applicability of the NYS DEC summarization and classification protocol to other sampling designs remains unknown. A standard semi-quantitative method, like NYS DEC protocol, have been incorporated into numerous state and federal monitoring programs, especially since USEPA’s rapid bioassessment manual was first published in 1989 (Barbour et al. 1999, Carter and Resh 2001). The NYS DEC program predates the rapid bioassessment manual and has a much longer history than most of these other monitoring programs. Our sampling method is a standard quantitative approach that also has a long history in stream ecology (Surber 1937). We expected that WQSs calculated using NYS DEC and our methods might vary because of the different collecting devices, sampling areas, numbers of individuals identified, and seasons. However, comparison of recent NYS DEC data (i.e., 1998-2001) to our results from a limited number of sites suggests that the two techniques produce comparable WQSs (Fig 5.17) and that the variation among years was similar. Overall, NYS DEC estimates on WQSs were approximately 7% higher than those determined by Stroud (Fig. 5.18). These conclusions are preliminary and will benefit from an additional year of data from Stroud and any additional data obtained from NYS DEC.

- 104 - CHAPTER 5 – MACROINVERTEBRATES NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Literature Cited

Barbour, M. T., J. Gerritsen, B. D. Snyder, and J. B. Stribling. 1999. Rapid Bioassessment Protocols for use in streams and wadeable rivers: Periphyton, benthic macroinvertebrates and fish. Second Edition. EPA 841-F-99-002. U.S. Environmental Protection Agency; Office of Water; Washington, D.C. Bode, R.W., M.A. Novak, L.E. Abele and D.L. Heitzman. 2000a. Assessment of water quality of streams in the New York City watershed based on analysis of invertebrate tissues and invertebrate communities. New York State Department of Environmental Conservation Report, Albany, New York. Bode, R.W., M.A. Novak, L.E. Abele, D.L. Heitzman and S. Passy. 2000b. Assessment of water quality of streams in the New York City watershed based on analysis of invertebrate tissues and invertebrate communities. Part II: 1999 sampling results. New York State Department of Environmental Conservation Report, Albany, New York. Bode, R.W., M.A. Novak, L.E. Abele, D.L. Heitzman and A. J. Smith. 2002a. Pesticides in streams of the Croton System, and their biological impacts. Part I. Targeted land uses. New York State Department of Environmental Conservation Report, Albany, New York. Bode, R. W., M. A. Novak. L. E. Abele, D. L. Heitzman, and A. J. Smith. 2002b. Quality assurance work plan for biological stream monitoring in New York State. New York State Department of Environmental Conservation Report, Albany, New York. Bode, R. W., M. A. Novak. L. E. Abele, D. L. Heitzman, and A. J. Smith. 2003. Pesticides in streams of the Croton System, and their biological impacts. Part II. Longitudinal trend studies. New York State Department of Environmental Conservation Report, Albany, New York. Carter, J. L., and V. H. Resh. 2001. After site selection and before data analysis: sampling, sorting, and laboratory procedures used in stream benthic macroinvertebrate monitoring programs by USA state agencies. Journal of the North American Benthological Society 20: 658-682. Hellawell, J. M. 1986. Biological Indicators of Freshwater Pollution and Environmental Management. Elsevier Applied Science, London, England. Lorenzen, C.J. 1967. Determinations of chlorophyll and phaeo-pigments: specrophotometric equations. Limnol. Oceanogr. 12: 343-346. Riva-Murray, K., R. W. Bode, P. J. Phillips, and G. L. Wall. 2002. Impact source determination with biomonitoring data in New York state: concordance with environmental data. Northeastern Naturalist 9:127-162. Stroud Water Research Center (Stroud). 2004. Water Quality Monitoring in the Source Water Areas for New York City: An Integrative Watershed Approach. A Report on Year 4 (2003) Monitoring Activities. Contribution No. 2004009 of the Stroud Water Research Center. Stroud Water Research Center, Avondale PA. Surber, E. W. 1937. Rainbow trout and bottom fauna production in one mile of stream. Transactions of the American Fisheries Society 66:193-202. Weber, C. I. 1973. Biological field and laboratory methods for measuring the quality of surface waters and effluents. EPA-670/4-73-001.

- 105 - CHAPTER 5 – MACROINVERTEBRATES NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Table 5.1: Aquatic macroinvertebrate assemblages from 2004 described with mean (1 ± SE) values for four individual biometrics [Total Richness, EPT Richness, Hilsenhoff Biotic Index (HBI), and Percent Model Affinity (PMA)], which were combined in the multimetric index [Water Quality Score (WQS)].

Site Total EPT HBI PMA WQS Richness Richness 3 35.4+1.2 11.7+0.9 4.82+0.11 67+4 8.2+0.3 6 32.7+2.7 12.2+0.9 4.39+0.17 78+2 8.5+0.3 9 32.5+1.2 11.9+0.4 4.55+0.08 69+2 8.3+0.1 10 31.9+1.6 13.8+0.6 4.07+0.07 87+1 9.0+0.2 15 36.4+1.0 13.7+0.9 4.36+0.05 69+2 8.7+0.1 23 23.6+1.0 9.6+0.6 3.71+0.10 78+1 7.8+0.2 26 21.9+0.6 6.4+0.4 5.07+0.13 60+3 6.3+0.2 29 31.7+2.9 8.5+1.3 5.06+0.17 48+1 6.8+0.4 34 32.3+2.1 12.2+0.8 4.38+0.21 72+2 8.4+0.2 46 23.5+1.3 8.5+0.7 5.66+0.22 55+2 6.3+0.3 52 28.1+1.8 13.6+0.7 3.19+0.11 70+3 8.5+0.2 55 20.8+1.2 3.8+0.4 5.88+0.12 47+1 5.0+0.1 101 26.3+0.7 9.6+0.5 4.10+0.27 73+6 7.7+0.3 102 31.7+1.2 12.1+0.7 4.36+0.12 76+1 8.5+0.1 103 30.7+1.1 12.2+1.1 3.93+0.32 81+5 8.6+0.3 104 30.8+0.7 13.2+0.6 3.79+0.13 81+2 8.8+0.2 105 30.8+0.4 14.2+0.8 3.82+0.45 78+5 8.8+0.3 106 29.8+1.1 14.3+0.9 3.42+0.15 82+2 9.0+0.2 107 27.9+0.7 10.0+0.6 4.57+0.25 69+2 7.7+0.2 108 28.5+1.1 12.6+1.2 3.91+0.21 71+3 8.3+0.2 109 28.6+0.5 12.4+0.8 3.76+0.05 86+2 8.7+0.1 110 27.1+1.7 11.5+0.2 3.42+0.17 81+2 8.4+0.2 111 20.6+1.5 9.1+0.6 3.09+0.25 67+5 7.3+0.4 112 22.7+1.2 9.5+0.1 3.79+0.06 70+6 7.4+0.2 113 28.5+1.7 11.1+0.9 4.51+0.11 68+2 7.9+0.3 114 26.6+0.3 12.9+0.7 3.84+0.14 85+4 8.5+0.2 115 31.6+2.6 9.9+0.9 4.95+0.22 60+1 7.4+0.3 116 25.8+1.7 10.3+1.6 4.99+0.15 58+4 7.0+0.5 117 19.4+1.2 4.6+0.5 5.03+0.10 70+1 6.2+0.1 118 28.1+1.1 10.6+0.7 5.13+0.07 70+5 7.6+0.1 119 25.4+1.9 10.8+0.5 3.55+0.11 82+2 8.2+0.2 120 25.5+0.2 10.9+0.4 3.59+0.08 83+1 8.2+0.1 121 31.1+1.5 14.4+1.1 3.38+0.13 85+2 9.1+0.2 122 25.4+1.5 7.5+0.1 3.99+0.20 54+1 6.8+0.2 123 26.8+0.5 8.5+0.7 3.86+0.07 58+2 7.2+0.1 124 28.3+1.6 12.1+0.6 3.22+0.14 68+2 8.3+0.1 125 30.9+1.7 14.2+0.7 3.04+0.05 79+2 9.1+0.1 126 21.9+1.0 9.2+1.1 4.04+0.09 68+2 7.2+0.2 127 26.4+1.6 9.9+0.6 3.08+0.03 65+4 7.8+0.3

- 106 - CHAPTER 5 – MACROINVERTEBRATES NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Table 5.1: Continued.

Site Total EPT HBI PMA WQS Richness Richness 129 24.1+0.3 7.9+0.5 4.27+0.08 83+2 7.5+0.1 130 15.5+0.8 4.0+0.4 6.02+0.15 42+1 4.4+0.2 131 20.9+0.4 4.1+0.5 5.72+0.10 42+2 5.0+0.2 132 23.5+1.5 4.7+0.1 5.13+0.20 64+2 6.3+0.1 133 14.4+0.5 1.9+0.7 5.86+0.15 39+1 3.9+0.3 134 18.2+1.2 4.7+0.6 5.80+0.09 42+2 4.8+0.2 137 22.2+0.5 4.2+0.3 6.40+0.17 39+1 4.8+0.1 138 14.8+1.8 1.3+0.4 5.70+0.13 34+2 3.6+0.4 139 16.7+0.8 4.6+0.6 4.88+0.15 59+3 5.7+0.3 140 16.1+1.3 2.7+1.1 6.06+0.22 34+2 3.8+0.4 141 16.3+0.8 7.2+0.4 3.72+0.51 44+2 5.7+0.2 142 24.7+1.3 1.8+0.5 6.46+0.06 35+1 4.4+0.2 143 23.6+4.1 6.0+1.0 5.94+0.52 47+5 5.5+0.8 145 17.8+1.2 3.4+1.0 5.83+0.16 39+3 4.3+0.5 146 24.0+2.2 9.8+0.3 4.20+0.14 61+1 7.2+0.2 147 23.9+2.0 6.9+0.5 5.66+0.41 57+3 6.2+0.4 148 14.6+2.1 2.3+0.6 5.50+0.08 29+1 3.7+0.3 149 30.2+0.7 10.5+0.8 4.65+0.09 66+2 7.8+0.2 150 17.7+0.8 3.4+0.3 5.18+0.05 51+2 5.1+0.1 151 32.1+1.5 10.8+1.2 5.25+0.15 63+2 7.7+0.3 153 27.4+1.2 14.2+1.2 2.83+0.05 79+2 8.8+0.2 159 26.6+2.0 13.1+0.9 3.27+0.16 80+1 8.6+0.3 160 33.3+0.7 14.3+0.7 4.02+0.06 72+2 8.8+0.2

Table 5.2: Ten sites where macroinvertebrates indicated the highest and lowest stream quality based on Water Quality Score (WQS) in 2004.

Site Site Description WQS Site Site Description WQS Highest Quality (descending order) Lowest Quality (ascending order) 121 Warner Creek nr Chichester 9.1 138 Cross R. nr Katonah 3.6 125 Quaker Br.at Merrit Cnty Park 9.0 148 trib. Kensico Res. nr Thorn. 3.7 10 E. Br. Delaware nr Arkville 9.0 140 Hunter Brook nr Yorktown 3.8 106 Dryden Brook nr Beerston 9.0 133 trib. Muscoot at Mahopac Fls 3.9 160 Emory Brook at Fleischmens 8.8 145 Broad Brook nr Bedford Falls 4.3 105 East Brook nr Walton 8.8 142 Kisco R. nr Bedford 4.4 104 Planter Brook at Fraser 8.8 130 Titicus R. nr Salem Center 4.4 153 Loomis Brook nr Trout Creek 8.8 137 trib. Muscoot R. nr Goldens Br. 4.7 109 Batavia Kill nr Kellys Corner 8.7 134 Plum Brook at Shenorock 4.8 15 Tremper Kill nr Andes 8.7 131 Titicus R. nr North Salem 5.0

- 107 - CHAPTER 5 – MACROINVERTEBRATES NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Table 5.3: Five WOH sites where macroinvertebrates indicated the highest and lowest stream quality based on Water Quality Score (WQS) in 2004.

Site Site Description WQS Site Site Description WQS Highest Quality (descending order) Lowest Quality (ascending order) 121 Warner Creek nr Chichester 9.1 117 Batavia Kill nr Windham 6.2 10 E. Br. Delaware nr Arkville 9.0 26 Esopus Creek nr Mt. Tremper 6.3 106 Dryden Brook nr Beerston 9.0 122 W. Br. Neversink nr Frost Val 6.8 160 Emory Brook at Fleischmens 8.8 29 Neversink R. nr Claryville 6.8 105 East Brook nr Walton 8.8 116 East Kill nr East Jewett 7.0

Table 5.4: Five EOH sites where macroinvertebrates indicated the highest and lowest stream quality based on Water Quality Score (WQS) in 2004.

Site Site Description WQS Site Site Description WQS Highest Quality (descending order) Lowest Quality (ascending order) 125 Quaker Br. at Merrit Cnty Park 9.0 138 Cross R. nr Katonah 3.6 52 Cross R. in W. P. R. Resv 8.5 148 trib. Kensico Res. nr Thorn. 3.7 34 Haviland Hollow Brook 8.4 140 Hunter Brook nr Yorktown 3.8 124 trib. E. Br. Croton nr Pawling 8.3 133 trib. Muscoot at Mahopac Fls 3.9 149 Waccabuc R. at Boutonville 7.8 145 Broad Brook nr Bedford Falls 4.3

- 108 - CHAPTER 5 – MACROINVERTEBRATES NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

density, density, density, EPT om 62 sites in the NYC drinking oleoptera density, Total Noninsect Macroinvertebrate density, Total Insect iptera density, C of aquatic macroinvertebrates fr richoptera density, D labels are underlined: Total ± 1 SE) of selected groups 2 lecoptera density, T Density (individuals/m ligochaeta density. phemeroptera density, P Site Total Insects EPT E P T D C O Noninsects

6 39226±4221 33333±4244 13978±2336 9075±2122 989±82 3914±446 15054±2018 4301±539 5892±1355 2323±738 5892±1355 989±82 3914±446 15054±2018 4301±539 6108±826 3376±911 86±35 1720±274 9075±2122 2645±234 9226±566 16194±1360 2753±328 13978±2336 3 860±246 18946±1496 33333±4244 6 39226±4221 9 7591±2324 305±132 2989±1142 5007±1337 16731±4743 11312±3427 2624±559 2796±820 10885±3558 649±180 43828±10537 10 16685±4858 792±169 1957±367 24172±5039 624±162 50581±12708 18642±5002 6753±2174 15 21677±5047 12946±3190 1591±858 7140±1365 2774±1014 50409±10688 27656±5337 946±388 1247±961 6731±1225 6065±1142 398±92 2943±895 2735±842 158±50 50±22 1419±452 57462±12433 21591±2109 86±86 23 10753±2136 269±87 9358±2492 1075±332 7054±1759 5161±666 2667±802 1548±140 3097±937 11677±2061 26 6409±1800 11720±3115 26839±2729 4000±925 7±7 556±152 29 29935±2461 22±22 2362±627 34 652±338 925±146 376±133 4868±1166 2475±594 6235±1419 1148±277 880±174 1319±219 46 3510±826 1474±410 887±345 1182±425 869±301 287±109 5695±1744 28172±3030 14495±1839 2925±600 42280±5108 52 1297±831 6877±2167 1896±519 4946±751 6624±949 55 537±83 5900±1637 5039±1406 667±212 194±22 479±156 1367±272 13247±1488 12208±2005 22430±4059 3699±987 30688±6054 101 12559±2231 14394±2140 7444±1524 5781±1202 738±280 925±213 387±191 5656±410 14158±2722 817±177 14108±2482 102 6588±1199 14771±2690 201±86 652±136 2186±138 125±55 103 1892±727 989±412 613±44 24373±6296 13670±4106 9297±2865 29728±7155 122±28 2602±708 1771±547 104 18237±2977 20882±3578 13032±2468 9828±2044 23011±3970 8530±1623 1763±419 1441±441 105 495±194 6989±1601 25634±958 36746±3769 8258±2032 7484±1148 14509±1169 8961±1163 106 2172±896 5355±863 2072±224 1950±327 3599±920 14280±2895 8896±1292 15713±3451 6029±687 107 366±113 8946±950 2129±396 452±177 21785±3902 10968±1661 5892±1711 33806±9236 255±152 2072±553 796±140 366±166 108 4186±2541 4280±1246 2555±736 4545±1338 307±82 1240±296 10292±4249 30±20 4710±591 1064±319 237±82 6434±2275 1295±236 2050±510 11111±3264 1254±377 7981±2014 109 5032±1534 7376±1163 7806±4810 21656±3948 15470±2926 11799±2704 1470±342 2201±276 25434±4407 1333±303 1299±462 18304±6213 5484±723 4294±614 401±52 789±124 110 12022±5469 1333±229 11785±1342 86±61 330±108 5892±1073 7065±1200 330±36 3398±1170 12748±3415 8826±2542 14796±4023 2452±181 19599±6358 1434±623 545±313 12635±1641 65±65 6692±1836 6817±786 4387±692 1097±146 111 414±217 264±122 127±57 1498±313 1049±416 1084±378 294±30 14133±1928 18688±1366 112 3785±1225 3778±573 1247±340 789±122 21140±1401 2968±326 1513±236 667±100 6918±910 767±158 113 6817±1585 5133±1507 792±126 892±81 138±58 11208±2247 10007±1093 115±52 2194±340 114 13556±2639 2047±748 327±120 3867±621 12201±1383 2110±249 1520±160 117±61 473±108 140±40 115 7412±1562 5282±1438 29±29 4337±519 4068±437 129±102 20±14 8642±1929 523±91 1437±260 9477±1291 5140±944 116 1230±373 2348±434 0±0 10530±1653 2649±252 2036±140 204±40 409±159 117 82±59 6928±1589 4047±1409 301±109 233±33 455±151 7384±1570 1054±430 118 119 7846±1416 6416±1243 254±118 1176±129 21462±3679 14129±1868 11097±1576 1591±422 1441±387 25806±4748 12265±2075 4419±711 120 13713±2139 6753±1648 0±0 581±313 882±142 4344±1183 871±183 1448±313 water watersheds during 2004. Column Table 5.5: O E

- 109 - CHAPTER 5 – MACROINVERTEBRATES NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Continued.

121 8234±1688 5917±1313 4504±1025 684±136 729±159 9413±1956 2853±920 9774±2145 2183±412 2250±423 1720±692 122 710±90 10602±2280 123 4839±1263 68±42 946±249 527±153 23477±3482 5928±1445 27677±4027 566±164 743±121 2968±1223 172±105 124 1179±269 0±0 4817±575 7094±2037 3792±1177 2031±761 618±178 8083±2354 8645±4010 6871±1570 1635±242 2079±116 882±218 308±194 14287±3956 1142±342 6603±776 4616±561 2586±450 1287±239 126±33 7390±867 1212±357 125 710±249 828±159 22±22 332±33 787±163 394±138 609±127 326±53 15140±2034 1312±252 55699±20536 989±374 151±74 1176±145 9075±4074 1566±281 126 7669±2340 2067±677 208±92 5667±989 4262±673 1079±338 3258±667 2409±403 4201±652 394±76 10369±1372 5470±672 12337±1647 810±238 380±122 222±88 9061±963 1419±43 258±103 9480±1674 1857±285 15771±4168 65±41 0±0 1118±568 127 337±95 258±113 11634±4411 556±195 3047±397 29075±3817 18323±2182 8344±1402 989±163 37806±4556 11337±1907 8442±2574 8989±1091 10222±1207 129 14280±2337 1498±285 925±69 20609±5372 409±95 1183±600 2968±883 969±516 221±109 29±29 720±409 19437±3826 8043±1582 882±349 4108±1726 130 11642±1182 1247±275 5247±2106 903±82 1090±157 20814±4332 131 2366±354 8731±1271 19398±446 86±50 15183±2535 3233±933 645±147 774±228 56882±20470 17161±3021 3670±1493 132 2323±149 208±86 29256±6642 17804±3514 5121±1149 20728±4629 158±79 8057±1085 3785±647 1968±276 1057±149 43±25 975±253 8151±2125 0±0 2469±541 21892±446 133 26237±4309 2918±418 136±70 100±82 3935±560 5168±1694 1890±557 763±353 394±217 134 37075±2848 61699±21004 12086±2545 137 10257±1978 8659±1027 1204±142 1150±443 1344±578 7161±631 2377±572 2925±1225 138 237±95 17254±3936 720±173 581±95 387±102 11892±1482 7183±2155 5118±1525 139 208±54 72±27 1484±149 0±0 4344±2098 14925±2306 8072±653 136±52 7753±626 111±42 140 15312±2381 9556±614 2419±781 269±37 141 1194±505 7075±1839 11194±3174 10086±2471 957±330 548±234 142 1755±927 407±187 2965±837 21280±3970 18047±868 201±72 1147±848 15983±602 2982±1942 143 310±93 3556±957 21603±1240 731±392 108±81 0±0 145 93±36 530±290 16065±2893 3111±2011 16796±3192 146 13312±1829 7068±1470 17319±2188 7739±1518 2152±470 1434±391 147 242±29 12581±2955 19907±4819 3627±902 0±0 3140±909 2495±355 6939±1174 2810±768 148 783±176 1127±408 5763±580 4252±744 4119±1756 37892±7889 3312±541 30910±6874 43±43 149 6065±1720 8403±3376 4842±1843 43957±9249 7136±2919 150 1313±403 4720±1124 2430±562 444±283 3392±1537 14892±2920 441±153 1849±472 4007±513 1061±236 814±384 151 9710±1848 16860±3230 1290±287 441±117 1968±387 153 1073±493 1505±674 14509±4898 11301±3684 9029±2825 15731±5211 2201±713 756±269 159 6555±2489 4570±2522 3274±2003 862±374 435±210 7202±2817 1516±708 1859±696 160 33527±9388 15484±4720 9462±2852 3018±1151 2645±907 3376±1253 42473±12005 1464±666 1267±458 787±348 17376±4591 125±84 147±38 647±342 435±251 667±229 3892±1727 8946±2661 1222±316 649±244 Table 5.5: Site Total Insects EPT E P T D O C Noninsects

- 110 - CHAPTER 5 – MACROINVERTEBRATES NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Table 5.6: Significance (p) of correlation coefficients (r) describing the relationship between the WQS and 10 density measures (log x+1 transformation) for 2004. Column labels are underlined: Total Macroinvertebrate density, Total Insect density, EPT density, Ephemeroptera density, Plecoptera density, Trichoptera density, Diptera density, Coleoptera density, Total Noninsect density, Oligochaeta density.

Total Insect EPT E P T D C Noninsect O

r -0.24 -0.18 0.59 0.73 0.56 0.40 -0.48 0.34 -0.21 -0.31 p 0.06 0.15 <0.0001 <0.0001 <0.0001 0.001 <0.0001 0.008 0.10 0.013

- 111 - CHAPTER 5 – MACROINVERTEBRATES NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

10 EOH

WOH 7.5

5 Water Quality Score 2.5

0 6 9 3 10 15 52 34 23 29 26 46 55 121 125 106 160 105 104 153 109 103 114 159 102 110 124 108 120 119 113 149 127 101 151 107 118 129 115 112 111 126 123 146 116 122 132 117 147 139 141 143 150 131 134 137 130 142 145 133 140 148 138 Site Figure 5.1: WQSs at all 62 Sites in 2004. White bars represent WOH sites and black bars represent EOH sites. Sites are arranged from highest to lowest WQS.

EOH

WOH 30

20 Number of Species 10

0 6 9 3 34 10 29 52 23 46 26 55 15 160 151 102 115 121 104 105 125 103 149 106 108 109 113 124 118 107 153 110 123 114 159 101 127 116 120 119 122 142 129 146 147 143 132 112 137 126 131 111 117 134 145 150 139 141 140 130 138 148 133 Site Figure 5.2: Total species richness for all 62 sites in 2004. White bars represent WOH sites and black bars represent EOH sites. Sites are arranged from highest to lowest species richness.

- 112 - CHAPTER 5 – MACROINVERTEBRATES NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

15 EOH

WOH

5 Number of EPT Species

0 6 9 3 10 15 52 34 23 29 46 26 55 106 121 160 125 153 105 104 159 114 108 109 103 124 102 110 113 120 119 118 151 149 116 115 107 127 146 101 112 126 111 123 129 122 141 147 143 117 132 134 139 137 131 130 145 150 140 148 133 142 138 Site Figure 5.3: Number of EPT species at all 62 sites in 2004. White bars represent WOH sites and black bars represent EOH sites. Sites are arranged from highest to lowest EPT richness.

6 EOH WOH

4 HBI

0 6 9 3 52 23 10 15 34 29 26 46 55 153 125 127 111 124 159 121 110 106 119 120 141 109 104 112 105 114 123 108 103 122 160 126 101 146 129 102 113 107 149 139 115 116 117 118 132 150 151 148 147 138 131 134 145 133 143 130 140 137 142 Site Figure 5.4: Hilsenhoff Biotic Index scores for all 62 sites in 2004. White bars represent WOH sites and black bars represent EOH sites. Sites are arranged from lowest HBI score to highest HBI score.

- 113 - CHAPTER 5 – MACROINVERTEBRATES NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

100

EOH

80 WOH

40 Percent Model Affinity 20

0 6 9 3 10 23 34 52 15 26 46 29 55 109 114 121 120 129 106 119 103 104 110 159 153 105 125 102 101 160 108 112 117 118 107 113 124 126 111 127 149 132 151 146 115 139 116 123 147 122 150 143 141 130 131 134 133 137 145 142 138 140 148 Site Figure 5.5: Percent Model Affinity values for all 62 sites in 2004. White bars represent WOH sites and black bars represent EOH sites.

Integrative Sites

Targeted Sites 7.5

5 Water Quality Score

2.5

0 6 9 3 10 15 52 34 23 29 26 46 55 121 125 106 160 104 105 153 109 103 102 114 159 110 108 124 119 120 113 127 149 101 107 151 118 129 112 115 111 123 126 146 116 122 132 117 147 139 141 143 150 131 134 137 130 142 145 133 140 148 138 Site Figure 5.6: 2004 WQSs for Integrative (white) and targeted (black) sites.

- 114 - CHAPTER 5 – MACROINVERTEBRATES NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

10 2000-2002

2004 only

Phase I in 2004 7.5

5 Water Quality Score

2.5

0 9 6 7 9 3 1 4 2 6 8 5 3 10 15 25 11 27 13 14 29 10 16 23 23 28 18 22 12 19 24 15 17 21 30 29 20 26 26 121 106 104 105 153 160 109 103 102 114 159 110 108 119 120 113 101 107 151 118 112 115 111 123 116 122 117 Sites Figure 5.7: Average WQS for 2000 to 2002 (in black) and 2004 (in white) WOH sites. Gray bars represent Phase I site WQS in 2004. 10 2000-2002

2004 only

Phase I in 2004 7.5

5 Water Quality Score

2.5

0 52 34 34 33 52 36 50 41 37 54 31 35 48 42 40 45 53 56 46 38 32 44 55 46 55 51 39 47 57 43 58 60 59 49 125 124 127 149 129 126 146 132 147 139 141 143 150 131 134 137 130 142 145 133 140 148 138 Site Figure 5.8: Average WQS for 2000 to 2002 (in black) and 2004 (in white) for EOH sites. Gray bars indicate Phase I site WQS in 2004.

- 115 - CHAPTER 5 – MACROINVERTEBRATES NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

10 WQS 2000 WQS 2001

7.5 WQS 2002 WQS 2003 WQS 2004 5 Water Quality Score 2.5

0 3 6 9 101523262934465255 Site Figure 5.9: 2000 to 2004 WQS at the 12 sites in common to Phases I and II.

- 116 - CHAPTER 5 – MACROINVERTEBRATES NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Cannonsville Middle Br. Croton R.

121 10 106 105 104 153 103 6 102 9 3 101 151 125 52 34 126 Pepacton East Br. Croton R.

121 10 106 160 109 15 114 110 108 113 107 112 111 125 52 34 124 129 132 150 Schoharie West Br. Croton R.

121 10 106 118 115 116 117 125 52 34 127

Ashokan East and South of Croton R.

121 10 106 159 120 119 23 26 125 52 34 149 146 141 143 55 131 137 130 142 145 138

Neversink North of Croton R.

121 10 106 29 122 125 52 34 46 139 134 133 140

Roundout Kensico R.

121 10 106 123 125 52 34 147 148

Figure 5.10: Results of Tukey's test comparing WQS in reference sites to sites within the respective watershed in 2004. Stations under the same line were not significantly different. WOH reference sites are 121, 10, and 106. EOH reference sites are 125, 52, and 34.

- 117 - CHAPTER 5 – MACROINVERTEBRATES NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

r2=0.34 p=0.001 7.5

Water Quality Score 2.5

0 0 100 200 300 400 500 600 2 Population Density 2000 (individuals/km ) Figure 5.11: Relationship between WQS and population density in the EOH.

7.5 r2=0.51 p<0.0001

Water Quality Score 2.5

0 0 5 10 15 20 25 30 Percent Impervious

Figure 5.12: Relationship between WQS and percent impervious cover in the EOH.

- 118 - CHAPTER 5 – MACROINVERTEBRATES NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

7.5

r2=0.21 p=0.007 5

Water Quality Score 2.5

0 0 20406080100 Percent Agriculture

Figure 5.13: Relationship between WQS and percent agriculture cover in the WOH.

7.5

r2=0.20 p=0.008 5

Water Quality Score 2.5

0 0 20406080100 Percent Forest

Figure 5.14: Relationship between WQS and percent forest cover in the WOH.

- 119 - CHAPTER 5 – MACROINVERTEBRATES NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

40 35 r2 = 0.61 30 25 20 15

2003 Richness 10 90 80 5 r2 = 0.64 70 0 0 5 10 15 20 25 30 35 40 60 2004 Richness 50 40 7

2003 PMA 30 2 6 r = 0.78 20 5 10 4 0 0 102030405060708090 3 2004 PMA

2 10

2003 EPT Richness 9 1 2 8 r = 0.84 0 7 01234567 2004 EPT Richness 6 5 18 4

16 2003 WQS 2 3 r = 0.67 14 2 12 1 10 0 8 012345678910

2003 HBI 2004 WQS 6 4 2 0 0 2 4 6 8 1012141618 2004 HBI

Figure 5.15: Biological metrics (Total Richness, EPT Richness, HBI, PMA, and WQS) for Phase II sites in 2003 and 2004.

- 120 - CHAPTER 5 – MACROINVERTEBRATES NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Cannonsville Middle Br. Croton R.

121 106 15 153 105 104 102 103 9 6 101 151 3 125 52 34 126 Pepacton East Br. Croton R.

121 106 15 160 10 110 109 108 111 114 112 113 107 125 52 34 124 129 132 150 Schoharie West Br. Croton R.

121 106 15 115 118 116 117 125 52 34 127

Ashokan East and South of Croton R.

121 106 15 159 119 120 23 26 125 52 34 149 146 143 141 131 137 142 145 55 130 138

Neversink North of Croton R.

121 106 15 122 29 125 52 34 139 46 134 140 133

Roundout Kensico R.

121 106 15 123 125 52 34 147 148

Figure 5.16: Results of Tukey's test comparing WQS in reference sites to sites within the respective watershed for Phase II sites in 2003 and 2004. Stations under the same line were not significantly different. WOH reference sites are 121, 106, and 15. EOH reference sites are 125, 52, and 34.

- 121 - CHAPTER 5 – MACROINVERTEBRATES NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

1998

7.5 1999

2000 5 2001

2003 Water Quality Scores 2.5 2004

0 3 6 9 29 34 46 52 115 119 122 123 153 125 130 134 139 140 142 146 WOH EOH Site

Figure 5.17: Water Quality Scores from 1998 to 2001 determined by NYS DEC (Bode et al. 2000a, 2000b, 2002a, 2003) and 2003-2004 determined by Stroud (2004 and this study).

7.5

2.5 Average Water Quality Score - Stroud 0 0 2.5 5 7.5 10 Average Water Quality Score - NYS DEC

Figure 5.18: Mean Water Quality Score reported by NYS DEC (Bode et al. 2000a, 2000b, 2002a, 2003) and by Stroud (2004 and this study). Not all site means represent the same years or the same number of years: NYS DEC n = 1-4, and this study n = 2.

- 122 - CHAPTER 5 – MACROINVERTEBRATES NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Chapter 6 - DOC and BDOC Dynamics

Research Task

Dissolved organic carbon (DOC) is the sum of all reduced carbon molecules dissolved in water and represents the largest pool of detrital carbon in stream ecosystems (Wetzel 2001). Heterotrophic bacteria utilize DOC as a source of energy and C, and DOC in transport provides a metabolic link between upstream and downstream segments. DOC concentrations provide a bulk indicator of organic pollution and are also indicative of terrestrial processing of organic matter. In the absence of extensive wetlands, bogs or swamps, baseflow concentrations of DOC in undisturbed watersheds generally range from approximately 1 to 3 mg C/L (1000 to 3000 µg/L) (Thurman 1984). Higher concentrations suggest sources of organic pollution such as point sources from sewage treatment plant discharges and eutrophic, algal-rich farm ponds, or non- point source runoff from urban or rural landscapes. The biodegradable DOC fraction (BDOC) consists of organic molecules that heterotrophic bacteria can utilize as a source of energy and carbon (Servais and Ventresque 1989). Within the context of drinking water quality, DOC is of interest because molecules within a subset of the DOC pool constitute the precursors of disinfection byproducts, DOC constituents, at very low concentrations, can generate taste and odor problems, and BDOC constitutes the nutritional resources that can contribute to biological regrowth within water distribution systems (Escobar et al. 2001) when carbon is the limiting nutrient.

Data generated in this research task were particularly relevant to research objectives 1 and 3, indicating how well best management practice (BMP) implementation controls sources of DOC and BDOC and how well the ecosystem conserves and processes organic matter. These data provide baseline targets for BMPs and insights into potential land uses that contribute DOC and BDOC, including natural sources such as wetlands. Additionally, these DOC and BDOC data provide supporting information to help interpret the tracer, spiraling, and metabolism studies.

Methods

Samples for DOC were collected and processed with particular attention to avoiding contamination (Kaplan 1994) and analyzed by Pt-catalyzed persulfate oxidation (Kaplan 1992) or UV-promoted persulfate oxidation (Kaplan 2000). Briefly, all glassware used for water collection was rendered organic-carbon (C) free by combustion at 500°C for six hours, and samples were protected from the atmosphere by sealing the collection vessels with persulfate- cleaned Teflon-backed silicone-septa. Baseflow stream samples were collected in 500-mL borosilicate bottles that were rinsed twice with site water, filled, capped, and placed on ice in the dark. Within 36 hours, the samples were filtered into 40-9mL borosilicate vials. Filtration to remove particles > 0.7 µm was performed with precombusted glass fiber filters (Whatman GF/F), an acetal-resin syringe type filter holder, and a peristaltic pump.

Analysis was performed with either an OI 1010 analyzer or a Sievers 800 analyzer. The OI 1010 acidifies the sample to convert all inorganic carbon to CO2, sparges the sample with ultrapure N2 (taken from the headspace in a liquid nitrogen tank) and the sparge gases are continuously fed through a non-dispersive infrared detector (NDIR) to measure the concentration of inorganic C.

- 123 - CHAPTER 6 – DOC/BDOC DYNAMICS NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Next sodium persulfate is added and the sample is heated to 95°C to convert the DOC to CO2. A second sparging sends the CO2 generated to the NDIR to quantify the amount of DOC.

The Sievers 800 uses a different method based on continuous flow analysis of total dissolved C (TC) and dissolved inorganic carbon (DIC), with DOC determined by difference. The oxidation and detection protocols are different as well. The analyzer injects acid and persulfate into the sample as it flows into the analyzer, and then the reagent-amended sample flows through a vacuum-degassing module that removes > 99% of the DIC. The sample is split into two streams, one for DIC analysis and one for total C analysis. The TC analytical stream flows through a quartz coil surrounding an UV lamp for DOC oxidation, while the DIC analytical stream bypasses this oxidation step. Both sample streams flow though temperature compensated conductivity detectors and an algorithm is used to convert conductivity to C-concentration. An extensive comparison of the Sievers 800 and OI 1010 analyzers, including groundwaters and surface waters from 102 geographically dispersed watersheds showed no differences between the concentrations measured with the two instruments (Kaplan 2000).

The BDOC method relies on the measurement of DOC in water samples before and after incubation for 28 days at room temperature in the dark (Kaplan et al. 1994). In the BDOC method, 10 organic-C free 40-mL vials are filled with subsamples of the filtered water from each site. DOC concentrations in five of the vials were measured immediately and the other five vials were incubated to allow the bacterial inoculum contained in the filtered water to grow and metabolize the BDOC. After 28 days, the samples in the five vials were refiltered using a syringe and syringe type filter holder, and analyzed for DOC. BDOC was calculated as the difference between the initial and final DOC concentrations.

Results

QA/QC

QA/QC for year 5 include laboratory measurements of standards and blanks, and baseflow field sampling of duplicates and blanks. QA/QC summary data are presented in Appendix A.6.

Laboratory QA/QC procedures involved the analyses of standards prepared from potassium hydrogen phthalate (KHP), blanks of deionized water, analytical replicates, and spikes of samples with concentrated KHP. Periodic checks of our laboratory standards were performed with a Demand standard purchased from QC SPEX. The Demand standard is an organic-C molecule (glucose) that can be used for assessment of DOC analyses or biochemical oxygen demand (BOD) analyses.

All analyses of DOC, including field samples, standards, and blanks were performed in duplicate. A total of 296 KHP standards were measured with an average recovery of 100.7% (range 90.95 to 108.8%). Duplicate analyses of the KHP standards had a relative percent difference (RPD) that averaged 0.66% (range 0 to 9.51%). When lab blanks were analyzed for RPD, the values ranged from 0 to 200% (41.4% ± 47.4%, mean ± standard deviation). The concentration of the lab blanks ranged from 0 to 216 µg C/L, well below the lowest DOC concentration for a field sample (1005 µg C/L).

- 124 - CHAPTER 6 – DOC/BDOC DYNAMICS NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Matrix spikes consisted of amending a field sample with a concentrated stock solution of KHP. Typically 380 µL of a 200 mg C/L standard were added to approximately 38 mL of sample. We used one of the five field replicates for the matrix spike and compared that concentration to the mean of the four remaining, unspiked field replicates. Overall, recovery of the three matrix spikes for the year five sampling ranged from 94.7 to 100.4% (97.5% ± 2.8%, mean ± standard deviation).

Field QA/QC procedures involved field duplicates and field blanks. For the 60 stream sites and the reservoir sites sampled in the fifth year, we performed fourteen field duplicates and fourteen field blanks. A field duplicate consisted of taking a second 500 mL water sample in a separate borosilicate bottle and treating the duplicate in the same manner as all other samples. The field blank for a reservoir or baseflow stream sample consisted of filling a 500 mL borosilicate bottle with deionized water and then processing this sample along with all other samples. Between each sample, the filtration apparatus was rinsed with deionized water, and the field blank typically was processed in the middle of the field samples collected on a given day. A field blank for a stormflow sample involved processing deionized water through the automated ISCO sampler, placing the inlet probe into a bottle of deionzed water and pumping the water into the collection vessel. The DOC field duplicates for baseflow and reservoir samples had a relative percent difference of 2.3% ± 1.7 % (mean ± standard deviation) and a range of 0.3% to 6.2%. For stormflow samples, the two values were 0.6% and 2.3% relative percent difference. The field blanks ranged from 15 to 33 µg C/L for reservoir samples, 18 to 53 µg /L for baseflow samples, and 234 µg /L to 387 µg/L for stormflow samples.

Field Data

Overview of entire study region –Baseflow sampling.- The year 5 data set of baseflow concentrations substantiates clear differences in DOC concentrations that were observed between the East of Hudson (EOH) sites and the West of Hudson (WOH) sites. The average DOC concentration WOH was 1923 µg C/L, and the range was 780 to 3852 µg C/L. Seven of the WOH sites had average concentrations above 2500 µg C/L, and 14 of the 35 sites had DOC concentrations that were less than 1500 µg C/L (Fig. 6.1). The average DOC concentration EOH was 3312 µg C/L, the range was 1862 to 5863 µg C/L, and 7 sites had concentrations below 2500 µg C/L (Fig. 6.1). The average BDOC concentrations for EOH and WOH sites also differed, with mean concentrations at EOH sites 1.7-fold higher (207 µg C/L WOH, 355 µg C/L EOH) (Fig. 6.2). BDOC as a percent of DOC in the WOH sites averaged 10.7% (range 1.8 to 19.9%) and in the EOH sites averaged 10.5% (range 3.5 to 17.4%) (Fig. 6.2).

The extremes within the DOC data set show that the five sites with the lowest concentrations were all from the WOH and the five sites with the highest concentrations were all from the EOH. The five lowest DOC concentrations include site 120, Bushnellsville Creek at Shandaken (780 µg C/L), site 119, Birch Creek at Big Indian (916 µg C/L), site 112, Mill Brook near Grant Mills (1005 µg C/L), site 159, Birch Creek above Pine Hill (1032 µg C/L), and site 23, Esopus Creek, near Allaben (1050 µg C/L). The five highest DOC concentrations include site 55, Kisco River near Stanwood (4097 µg C/L), site 126, Stump Pond Stream near Pawling (4204 µg C/L), site 132, Bog Brook near Sears Corner (4425 µg C/L), site 139, Muscoot River near Whitehall Corners (4718 µg C/L), and site 149, Waccabuc River at Boutonville (5863 µg C/L).

- 125 - CHAPTER 6 – DOC/BDOC DYNAMICS NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Similarities between year-4 and year-5 data were apparent (Fig. 6.1), although for several EOH sites, especially 46, 127, 143, and 150, year-5 DOC concentrations were as much as 2-fold lower than in year-4. For each of these sites, as well as sites 52, 130, 131, 132, 134, and 146, where DOC concentrations were lower in year-5, the estimated baseflow discharge in year-5 was also 3-fold or more lower than in year-4. When considered on a watershed basis, and separated into Phase I and Phase II sites, there was general overlap in concentrations of DOC and BDOC between the phases (Fig. 6.3 and 6.4). The exceptions were some elevated DOC concentrations in the Schoharie Creek watershed (Fig. 6.3 B), and lower BDOC concentrations in the West Branch Delaware watershed (Fig. 6.4 B)

Step-wise multiple linear regressions were used to identify land-use variables that could predict DOC and BDOC concentrations. All concentration data were log transformed to improve linearity and equality of variance. This analysis was performed with data from Phase I, and then with data from Phase I and II to see how the Phase II sites impacted our evaluation of the watersheds. The percentage of watersheds in wetlands was a significant predictor of DOC concentrations in both the WOH (r2 = 0.58) and the EOH (r2 = 0.75) sites (Table 6.1). For the EOH region, waste water treatment plant effluents and wetlands explained most of the variance in DOC and BDOC data in both Phase I and Phase II (Table 6.1). For the WOH region, a positive correlation with grasslands as a predictor of DOC and BDOC concentrations was replaced by a negative correlation with deciduous forests (DOC) or a positive correlation with wetlands (BDOC) (Table 6.1).

Discussion

The year-5 data set complements and reinforces the information obtained during the other 4 years of this investigation, and the DOC and BDOC concentrations for the 12 sites carried over from Phase I generally fell within the range of the prior 4-year data set. The five WOH sites with the lowest DOC concentrations (23, 112, 119, 120, and 159) provide a good baseline target for BMP implementation. All the sites were located in watersheds with low population densities (2- to-14 people/km2), extensive forest cover (91 to 97%), little impervious cover, and little, if any, sewage treatment discharge (reported in monthly State Pollutant Discharge Elimination System reports; SPDES). Even these high quality watersheds had measurable concentrations of BDOC (66 to 198 µg C/L range; approximately 7% to 17% of the DOC), as BDOC production is a natural function of healthy ecosystems. This helps place possible limits or boundaries on BMP expectations.

Sites in the EOH region with low (< 2500 µg C/L) DOC concentrations (sites 34, 124, 125, 129, 140, 141, and 143) were smaller watersheds, and with the exception of 143, had no reported SPDES discharges and less than 5% wetlands. The five EOH sites with the highest DOC concentrations (55, 126, 132, 139, and 149) reflect the impacts of wetland processes, as the watersheds had 6 to 10 % wetland cover. All of these sites had active SPDES discharges, however the range was broad (0.005 to 2.771 cm3/cm2)

Wetlands have been identified as an important source of DOC (Mulholland 1979; Hemond 1990), and while this DOC has a high humic content which is generally considered refractory to biological decomposition, others have shown that humic-C in stream is biologically labile (Volk et al. 1997). Impervious surfaces have also been identified as a source of DOC (Jordan et al.

- 126 - CHAPTER 6 – DOC/BDOC DYNAMICS NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

2000; Wallace et al. 2002), though these studies were not peer reviewed and do not address the issues of DOC quality (i.e., BDOC concentrations). There are relatively few published reports of DOC and BDOC concentrations in watersheds that supply drinking water. One study that measured concentrations in 80 watersheds across the continental United States where drinking water treatment plants are supplied by surface waters reported DOC concentrations of 800 to 5000 µg C/L (mean 2300 µg C/L) and BDOC concentrations of 100 to 1000 µg C/L (mean 300 µg C/L) (Kaplan et al. 1994). For the most part, the data reported for the 60 study sites fell within these ranges.

Literature Cited

Hemond, H. F. 1990. Wetlands as the source of dissolved organic carbon to surface waters. in Organic Acids in Aquatic Ecosystems, E. M. Perdue and E. T. Gjessing, eds., pp. 301-313. John Wiley and Sons, Ltd. Hynes, H. B. N. 1960. The Biology of Polluted Waters. Liverpool University Press. Jordan, T. E., D. L. Correll, D. E. Weller. 2000. Mattawoman Creek Watershed: Nutrient and Sediment Dynamics. Final Report, Smithsonian Environmental Research Center, Edgewater, MD. Kaplan, L. A. 1992. Comparison of high-temperature and persulfate oxidation methods for determination of dissolved organic carbon in freshwaters. Limnology and Oceanography 37:1119-1125. Kaplan, L. A. 1994.A field and laboratory procedure to collect, process, and preserve freshwater samples for dissolved organic carbon analysis. Limnology and Oceanography 39:1470-1476. Kaplan, L. A., D. J. Reasoner, and E. W. Rice. 1994. A survey of BOM in US drinking waters. J. AWWA 86:121-132. Kaplan, L. A. 2000. Comparison of three TOC methodologies. J. AWWA 92: 149-156. Mulholland, P. J., and E. J. Kuenzler. 1979. Organic carbon export from upland and forested wetland watersheds. Limnology and Oceanography 24:960-966. Servais, P. A. Anzil, and C. Ventresque. 1989. Simple method for determination of biodegradable dissolved organic carbon in water. Applied and Environmental Microbiology 55:2732-2734. Volk, C. J., C. B. Volk, and L. A. Kaplan. 1997. Chemical composition of biodegradable organic matter in streamwater. Limnology and Oceanography 42:39-44. Wallace, T.A., Ganf, G.G. and Brookes, J.D. (2002). Sources of Organic Carbon in the Torrens River Catchment: Report to Torrens Catchment Water Management Board, Adelaide. 23pp.

- 127 - CHAPTER 6 – DOC/BDOC DYNAMICS NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

n 2 R 0.63 0.63 0.59 0.59 0.76 0.54 0.54 0.60 0.60 0.76 0.76

0.37 (+) (+) 0.37 0.39 (+) (+) 0.39 0.30 (+) (+) 0.30 0.47 (+) (+) 0.47

predictor variable.

0.11 (-) (-) 0.11 (+) 0.05 0.71 lues (n=1,2,3, or 5 depending upo 0.46 (-) l). Separate models were run for Phase I an analyte and a particular values provided for those variables found to be 2 0.18 (+) 0.09 (+) 0.08 (+) (+) 0.08 0.09 0.77 0.04 (+)

0.53 (+) (+) 0.53 0.60 (+) (+) 0.60 using mean concentration va value for a given mode 2 BDOC against land use/cover variables and permitted discharger discharger and permitted variables BDOC against land use/cover ely, relationships between ss was used with partial R

indicate the largest partial R easing or decreasing, respectiv . A stepwise selection proce

0.06 (+) (+) 0.06 (-) 0.18

0.06 (+) (+) 0.06 (+) 0.26

0.44 (+) = 0.05) predictors (bolded values α Results of multiple linear regressions (MLR) run for DOC and run for (MLR) regressions linear multiple Results of DOC Region Period DOC Phase I %Urban/Suburban Phase I&II Phase I EOH EOH Phase I&II 0.14 0.20 (+) (+) MXDFR DECFR CONFR MXDBR BDOC WWTP BRUSH GRASS FARMD INDST COMML RESID WOH WOH ORCHD OTHER TRANS CRPLD %Agriculture Phase I Phase I&II %Herbaceous/Brush Phase I %Wetland EOH EOH 0.04 (-) Phase I&II 0.26 (+) %Forest %Water WOH WOH 0.05 (+)

annual effluent volumes (WWTP) Table 6.1. data (EOH n=30, WOH n=30) and Phase I&II data (EOH n=53, WOH n=57) data (EOH n=30, WOH n=30) and Phase site). The ‘+’ and ‘-‘ signs indicate incr significant (

- 128 - CHAPTER 6 – DOC/BDOC DYNAMICS NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

12000 2003 DOC 2003 BDOC

East of Hudson 2004 DOC 2004 BDOC 10000

8000

6000

4000 Organic CarbonOrganic (ug/L)

2000

0 4 0 8 34 46 52 55 29 37 43 50 12 125 126 127 1 13 131 132 133 134 1 13 139 140 141 142 1 145 146 147 148 149 1 Site Number

4000 West of Hudson

3000

2000 Organic Carbon (ug/L) Carbon Organic 1000

0 3 6 9 1 3 10 15 23 26 29 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 5 5 101102 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 159160 Site Number

Figure 6.1 Baseflow DOC and BDOC concentrations measured at the 60 EOH and WOH stream sampling sites for years 4 and 5 (Phase II).

- 129 - CHAPTER 6 – DOC/BDOC DYNAMICS NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

20 2003 East of Hudson 2004

10 Percent BDOC Percent

0 4 6 2 4 1 2 0 1 8 9 3 4 5 55 2 25 26 27 3 3 33 34 37 4 4 42 43 45 4 4 50 1 1 1 1 129 130 1 1 1 1 1 138 139 1 1 1 1 1 146 147 1 1 1 Site Number

West of Hudson 30

20 Percent BDOC Percent 10

0 3 6 9 3 4 9 3 7 8 2 0 10 15 2 26 29 03 07 08 12 16 21 53 59 1011021 10 1051061 1 10 1101111 11 1141151 11 11 1191201 12 1231511 1 16 Site Number

Figure 6.2 Baseflow BDOC as a percentage of DOC concentrations measured at the 60 EOH and WOH stream sampling sites for years 4 and 5 (Phase II).

- 130 - CHAPTER 6 – DOC/BDOC DYNAMICS NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

a MNCWBC EMC TCS KSC

b WBD EBD SCH ESP NRD

Figure 6.3. DOC concentrations for baseflow samples in Phase I and Phase II in the watersheds East of Hudson (a) and West of Hudson (b).

- 131 - CHAPTER 6 – DOC/BDOC DYNAMICS NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

a MNCWBC EMC TCS KSC

b WBD EBD SCH ESP NRD

Figure 6.4. BDOC concentrations for baseflow samples in Phase I and Phase II in the watersheds East of Hudson (a) and West of Hudson (b).

- 132 - CHAPTER 6 – DOC/BDOC DYNAMICS NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Chapter 7 - Nitrogen (N), Phosphorus (P), and Dissolved Organic Carbon (DOC) Spiraling

Introduction

Phosphorus and nitrogen entering the streams that feed the NYC reservoirs are likely to be taken up and recycled at least once and probably several times within the stream ecosystem prior to reaching the reservoirs. Because this cycling occurs simultaneously with downstream transport it is sometimes referred to as "spiraling". The "spiraling length" represents the distance over which the average nutrient atom travels as it completes one cycle of utilization from a dissolved available form, passes through one or more metabolic transformations and is returned to a dissolved available form. Quantitatively, it is the ratio of the downstream flux of nutrient to the uptake of nutrient per unit length of stream. More intense utilization of the nutrient, along with more effective retention of particulate forms, shortens the spiraling length so that an individual nutrient atom completes more cycles in its passage through a stream-river network. Dissolved organic carbon (DOC) undergoes similar spiraling except that its utilization eventually results in oxidation, and the spiraling length in this case refers to the distance traveled until oxidation.

The significance of spiraling to the NYC watersheds relates both to the function of the stream ecosystem itself, as well as implications for downstream ecosystems (the reservoirs) and resulting water quality. For the stream ecosystem, spiraling reflects the degree of metabolic activity within the system, the ability of the system to retain nutrients, and the relative utilization rates (hence degree of nutrient limitation) among different nutrients. Spiraling length also describes the scale on which upstream processes are linked to downstream processes. Thus spiraling represents a fundamental measure of stream ecosystem function. Ecosystem impairment is likely to increase spiraling length (reduce the cycling intensity), through reduced uptake, excessive loading, or decreased retentive ability of the ecosystem. An exception to this rule would occur when the increased loading of a single nutrient stimulates uptake of a second nutrient, whose spiraling length would shorten.

The processing or spiraling of nutrients may have a variety of implications to downstream ecosystems. Uptake may sequester nutrients for long periods resulting in seasonal alterations of downstream nutrient loads. Processing may also alter the partitioning of the nutrient forms (inorganic/organic, dissolved/particulate), which has implications for the availability of the nutrient moving downstream. In the case of nitrogen, significant in-stream removal may occur through denitrification. In the case of DOC, more intense utilization within the stream ecosystem directly reduces the downstream loading.

The measurement of uptake length will provide a first step in addressing the role of spiraling as an indicator of ecosystem function and as a potential influence on downstream water quality. A complete evaluation of spiraling length requires use of isotopic tracers, but previous studies have shown that the uptake that can be observed by incrementing the background nutrient concentrations by small amounts provides a reasonable first approximation to the uptake length (or distance traveled in the available form). Past studies have also shown that uptake length is normally >90% of the total spiraling length (Newbold 1992), a result that can be evaluated from the fractions of dissolved and particulate nutrient in the water column.

- 133 - CHAPTER 7 – NUTRIENT SPIRALING NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

+ - This section reports the uptake length of inorganic nitrogen (NH4 ), inorganic phosphorus (PO4 3), and organic carbon (glucose and arabinose) in streams derived from whole-stream injections of standard solutions of N, P, and C along with a conservative tracer (bromide). Peak concentrations of the added nutrients were in the µM range, and the carbohydrates were in the nM range. Concentrations of each constituent were measured at five stations downstream and the data fitted to a one-dimensional advection-dispersion model augmented to include transient storage. The task was performed at each of the 10 integrative stations between June and October 2004.

In phase II of the project (2003 and 2004) we sampled three sites that were continued from Phase I, and seven sites that had not previously been sampled. The three continuing sites were station #52 (Cross River in Ward Pound Ridge Reserve), #46 (Muscoot River, near Baldwin), and #29 (Neversink River near Claryville).

Methods

Uptake lengths for dissolved phosphate, ammonium, glucose, and arabinose were determined by whole-stream solute injections, following the general approach described by the Stream Solute Workshop (1990).

One injection was made at or near each of the 10 "integrative" sampling stations. Each injection -3 + involved simultaneous addition of a conservative tracer (sodium bromide), PO4 , NH4 , glucose, and arabinose at rates designed to achieve maximum concentration (after mixing) in the stream -3 + of 30 µg/L PO4 , 30 µg/L NH4 , and 0.20 µM for the carbohydrates. Amendments were metered in at a constant rate, using a peristaltic pump for time periods ranging from 70 to 120 minutes, depending on flow and channel characteristics. These injections were conducted simultaneously with propane injections made for the purpose of assessing gas exchange rates (Chapter 8).

On the day prior to an injection, preliminary measurements were made of streamflow and travel times. Streamflow was measured by wading cross sections (or from a bridge), and measuring depth and velocity (Swoffer current meter) at 10-20 intervals. Time-of-travel was estimated from the introduction of a pulse of rhodamine dye, which was tracked visually. From these measurements the following design parameters for the actual injection were determined: (i) quantity of conservative tracer, phosphorus, nitrogen, and carbon to be injected; (ii) duration of injection; (iii) concentrations and metering rates for the injection solutions; (iv) longitudinal locations of five sampling stations downstream from injections; (v) schedule for collection of water samples from each station. The design objective was to achieve target concentrations (with thorough lateral mixing) of all constituents by the upstream most sampling station, to minimize longitudinal variation in the peak concentration of the conservative tracer, and to observe approximately 60% uptake of each nutrient within the study reach. In locating downstream stations, an uptake mass transfer coefficient of 5x10-5 m/s was assumed. A spreadsheet-based model was used to calculate the design parameters.

Immediately prior to the beginning of an injection, background water samples for nutrient concentrations were taken at each downstream sampling station. Subsequent samples were taken

- 134 - CHAPTER 7 – NUTRIENT SPIRALING NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

according to the sampling schedule relative to initiation of the injection. Five water samples for assay of N, P, glucose, and arabinose, in addition to the conservative tracer, were taken from each station within the period of plateau concentration, or in the period of maximal concentrations. Samples for N and P assay were field-filtered through a rinsed, Whatman® 0.45 µm cellulose nitrate membrane filter and frozen within 24 h of collection for analysis within 60 days. Samples for glucose and arabinose assay were sterile-filtered (0.2 µm HT Tuffryn Acrodisc®) and frozen within 24 h for analysis within 2 months. At the upstream-most and downstream-most sampling stations additional water samples for the conservative tracer were collected to describe the complete passage of the injection pulse. Additionally, samples were collected at the injection site and at the downstream-most sampling station before, during, and after the injection for supplemental water chemistry analyses (NO3-N, SKN, TKN, TDP, and TP) to monitor changes in other N and P species throughout the injections.

Uptake length for ammonium, phosphate, and carbohydrates was estimated from the average concentration elevation, ∆c(x)=c(x)-cb(x) where c(x) is the average concentration of the added nutrient sampled during the plateau at a distance x meters downstream into the reach; and cb(x) is the background (i.e., un-amended) concentration at distance x of the nutrient measured immediately prior to injection. To adjust for longitudinal dilution and dispersion, we calculated the ratio, r, of the concentration elevation to that of the bromide (the conservative tracer): i.e., r=∆c/∆ [Br-]. The uptake length, S, for the respective nutrient was calculated as the inverse of the longitudinal loss-rate, kl, which, in turn was estimated by non-linear regression from the

relation r(x)=r0 exp(-kl x), where r0 is the concentration ratio elevation at x=0. The mass transfer coefficient for uptake, vf, or “uptake velocity” was then calculated as vf =kl d, where vw is the reach-average water velocity, and d is the reach-average depth.

This approach to estimating the uptake velocity assumes that uptake increases linearly with the concentration, c(x), of the nutrient throughout the varying concentration exposures of the injection, i.e., that the vf measured at elevated c(x) is the same as vf at background concentration. It is likely, however, that some decline in vf occurs at higher concentrations as biological uptake approaches saturation, so that, in this respect, the estimates of vf should be considered as lower- bound approximations. The potential error introduced by possible non-linearity in uptake was held to a minimum by limiting the injection concentrations to ~30 µg/L above background, for ammonium and phosphate, and to 0.2 µM above background for carbohydrates.

In 2004 (as in 2003) the ammonium concentrations measured at Station 139 (Muscoot R. near Whitehall Corners) were anomalous. Background ammonium concentrations measured prior to the injection ranged from 0.22 to 0.35 mg/L, with a mean of 0.28 mg/L (Table 7.3). These backgrounds were similar to those from the same station in 2003 (mean of 0.33 mg/L), but were far higher than all other backgrounds measured during the total of 50 injections conducted for this project. Background ammonium concentration typically has been near the detection limit of 0.01 mg/L and in no case has exceeded 0.030 mg/L. The background at Station 139 also greatly exceeded the design concentration elevation, ∆c(x=0) of 0.030 mg/L for the injection. The injection was conducted without modification because the results of these analyses were not available until several weeks later. The plateau concentrations for the injection ranged from 0.09 to 0.21 mg/L, which was lower than the backgrounds but considerably higher than the design elevation. It is suspected that there is a varying upstream input of ammonium that declined

- 135 - CHAPTER 7 – NUTRIENT SPIRALING NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

between the time the background concentrations were sampled and the plateau measurements were taken. The data were analyzed as though the observed ammonium had been experimentally injected, but using an assumed background of 0.010 mg/L. The non-linear regression explained 98% of the variance. Note that although the suspected input may have been declining, the experimental sampling was precisely timed to follow a downstream pulse, so that the temporal variation of the input would not have compromised the estimate. As discussed above, the relatively high concentrations would be expected to reduce the estimated uptake velocity. The resulting estimate of 0.037 mm/s was the second lowest uptake velocity measured in 2004. We therefore report this estimate for the ammonium uptake, but note that it was made under unusual circumstances.

Conservative tracer (bromide) data were analyzed by fitting the concentration-time curves to a one-dimensional advection-dispersion model that includes a transient storage component (OTIS- P, Runkel 1998). Streamflow, a fixed parameter, was calculated from corrected injection pump rate and duration and area under the empirically derived bromide curves at the downstream station. Where sufficient empirical data were not available to complete the tail of the curves, the tail was extrapolated and the estimated area under this portion of the curve was included in the flow calculation. The model yields estimates of the longitudinal dispersion (D), cross-sectional stream area (A), the hydraulic exchange velocity (the rate of vertical mixing into the stream bed), and the transient storage volume (expressed as the non-advecting cross-section of channel As). The parameters were estimated iteratively by the OTIS-P model, which uses a non-linear least squares estimation procedure.

The carbohydrates, glucose and arabinose, were analyzed by HPLC with pulsed amperometric detection (Dionex 500) using a protocol recently improved for resolution, sensitivity, and precision (Cheng and Kaplan 2001). These modifications have resulted in improved baseline stability, allowing detection limits of less than or equal to 0.4 nM for monosaccharides, average coefficient of variation of 1.3% for a 100 nM standard, and recovery between 92 and 109% for individual monosaccharides in stream water amended with standards.

Bromide was analyzed by ion chromatography with conductivity detection (Dionex 500).

Data analysis—Reported correlations coefficients, r, refer to the Pearson product-moment correlation coefficient. Unless otherwise stated, a reported correlation was significant at P<0.05. For comparisons involving data only from 2004, for which the sample size, n=10, the non- parametric Spearman rank-correlation coefficient was also computed in order to eliminate the potential for excessive leverage from a single data point. Correlation coefficients for the 2004 data are reported as significant only if the P<0.05 for both the Pearson and Spearman coefficients.

For analysis-of-variance (ANOVA) and analysis-of-covariance (ANCOVA) among the 5-year dataset, we used the General Linear Model procedure of SAS Institute to accommodate the lack of balance in design (most sites were sampled in only 2 or 3 of the 5 years).

- 136 - CHAPTER 7 – NUTRIENT SPIRALING NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

The uptake of nutrients was described by a rectangular hyperbola in analogy with Michaelis- Menten enzyme kinetics. In a form applicable to uptake onto streambeds, the uptake flux U (M L-2 T-1) of nutrient is be described as:

U = Umax C / (Ks + C) Eq. 7.1

in which Umax is the maximum uptake flux that would occur under high (saturating) water -3 -1 column nutrient concentration, C [M L T ], and Ks is the “half-saturation” concentration, i.e., at which U = Umax /2. The uptake velocity reported here is related to the flux by vf = U/C which, substituting into Eq. 7.1, yields, vf = Umax / (Ks + C) Eq. 7.2

The parameters Umax and Ks were estimated from the measurements of phosphate vf and background nutrient concentrations (as C) by non-linear regression.

Results and Discussion

QA/QC

Quality Assurance/Quality Control results are summarized in Appendix A.7, Tables 7.1-7.4.

Field blanks—Ammonium (NH4-N), and soluble reactive phosphorus (SRP) each had only one field blank exceedance (concentration greater than twice the method detection limit, or MDL), constituting 3% of all field blanks. There were 4 exceedance for Br (7%) and 16 for glucose. Three of the bromide exceedances were near the acceptable limit (all < 3 X MDL), whereas the fourth, had a concentration of 0.47 mg/L or nearly 50 X the MDL. The latter sample was erroneously sampled as a duplicate (the concentration of the corresponding sample was 0.46 mg/L) rather than as a field blank. The field-blank exceedances for glucose are a concern, but it is unlikely that field contamination compromised the estimates of glucose uptake because the mean and maximum concentrations of blanks (4.1 nM and 17 nM) were far lower than the target concentration (200 nM above background) of the stream injections. Both the frequency and magnitude of high glucose field blanks was substantially lower than in 2003, presumably as a result of improvements in field methods that were instituted for 2004.

Laboratory blanks—For bromide there was one laboratory blank exceedance which, at 3 × MDL, is not of concern. Eleven (17%) of the glucose laboratory blanks exceeded twice the MDL. However, the average and maximum concentrations of the exceeding blanks (7.5 nM, and 13 nM, respectively) were far below the target concentration (200 nM above background) of the nutrient injection, so that these exceedances should not have influenced the uptake estimates.

Field duplicates—Exceedances in field duplicates occurred for bromide (2 of 106), SRP (6 of 78), arabinose (4 of 77), and glucose (22 of 77). The two bromide exceedances were less than twice the detection limit and are not a concern (0.017 mg/L and 0.013 mg/L at a MDL 0.01mg/L). The 6 SRP exceedances consisted of five relative percent difference (RPD) exceedances and one absolute difference (ABSD) exceedance. The absolute difference

- 137 - CHAPTER 7 – NUTRIENT SPIRALING NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

exceedance was less than 2 X MDL and is not a cause for concern (0.004 mg/L at a MDL of 0.003 mg/L). The RPD’s for the five failures (RPD>20%) were all between 20 and 25% except one, which was 32%. All of these RPD failures occurred at three sites (52, 139, and 130) where we also observed that variation in SRP among plateau samples was high. The corresponding laboratory duplicates were all satisfactory. Field contamination seems unlikely because all sampling locations at each site were uniformly affected and because the ammonium field duplicates, taken from the same sample split as the SRP, were satisfactory. All three sites were sampled early in the field season (June) under conditions of high baseflow and slightly elevated concentrations of suspended solids (TSS: 3.3-8.6 mg/L) and particulate phosphorus (PP: 0.010- 0.018 mg/L) (Table 7.4). It appears that the large RPD’s in the field duplicates from these stations involved interference from the particulate phosphorus, probably adsorption and desorption of the added phosphate either in the water column prior to sampling or in the few minutes between sampling in the field and field filtration. The effect of the high particulates was to increase the uncertainty of the estimates of phosphate uptake length, as presented in Table 7.1. There was, in fact, a fourth site (#46, Muscoot River near Baldwin Place) sampled in June with elevated PP (0.036 mg/L) and TSS (4.0 mg/L), where the uncertainty of the uptake length estimate was high. Despite the high uncertainties, the resulting estimates were reasonable and consistent with previous years’ data, as discussed below. In future field work, additional care will be taken to avoid injections when suspended particulates are elevated.

The carbohydrate field duplicates had 26 total exceedances (out of 154 duplicates, 77 each for arabinose, and glucose). Of these, only four involved arabinose and they were all ABSD exceedances, ranging from 2 to 14 nM. The other 22 exceedances involved glucose and were evenly split between RPD exceedances (10 total) and ABSD (12 total). The ABSD exceedances ranged up to 12 nM. For both arabinose and glucose the ABSD exceedances were low in comparison to the injected concentration of 200 nM and as such do not pose a significant concern towards interpreting carbohydrate uptake dynamics.

The ten glucose field-duplicate exceedances ranged in RPD from 22% to 67%. The RPD exceedances were distributed among seven different injections, most in the first part of the field season. The problem was not detected until near the end of the field season when analyses were completed. At that point all technicians involved in field sampling were re-trained and more rigorous procedures (such as the use of latex gloves during sampling) were adopted. After these measures, only two field duplicate carbohydrate exceedances occurred over the course of the last two injections, as compared to 24 exceedances during the first eight. Although the rate of RPD exceedances is a concern, it is unlikely that they appreciably affected the uptake estimates because each is based on a non-linear regression through 25 sample points.

Lab duplicates—Among lab duplicates, exceedances occurred for glucose (4 of 45), and arabinose (1 of 34). The glucose exceedances were ABSD exceedances of 3.2 nM and 7.2 nM (the MDL is 2 nM), and RPD’s of 22% and 37%. The arabinose exceedance was 4.3 nM.

Matrix spikes were satisfactory for all analytes with the exception of bromide (8 exceedances out of 63 total) and arabinose (one exceedance out of 31 total). The arabinose recovery was 72% relative accuracy limits of 75-125% and is not a cause for concern. The bromide exceedances involved recoveries between 86 and 90%, or just below the accuracy limits of 90-110%.

- 138 - CHAPTER 7 – NUTRIENT SPIRALING NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Laboratory control standards were acceptable for all analytes but bromide which had 9 exceedances out of 164 total observations. The exceedances were all confined to two consecutive runs and the recoveries ranged from 117-119%, suggesting that an improper LCS dilution might have caused the exceedances.

Field Data Summary

Significant uptake was measured for each of the nutrients at each of the ten stations, except at Station 130 (Titicus near Salem Center), where the uptake of SRP was not significant (Table 7.1). The uptake length for phosphate ranged from 739 to 9,276 m. Ammonium uptake length ranged from 353 m to 4902 m and was in all cases shorter than the respective phosphate uptake length. Glucose uptake length ranged from 224 m to 4338 m, and arabinose from 466 to 18,051 m. The uptake lengths for all four nutrients were significantly correlated among each other (r= 0.88 to 0.98, P<0.01).

Uptake lengths (S) for all nutrients (NH4-N, SRP, glucose, or arabinose) were positively correlated with streamflow (r = 0.90, 0.70, 0.76, or 0.85, respectively), water depth (r = 0.85, 0.72, 0.87, or 0.98, respectively), and water velocity (r = 0.94, 0.76, 0.98, or 0.88, respectively). This strong dependence of uptake length on stream size has been widely observed (e.g., Peterson et al. 2001) and is expected from scaling considerations. In general, uptake length is shorter and the likelihood that a nutrient atom will be utilized within a reach of a given length is larger in smaller streams.

Uptake velocity, vf, (Table 7.2) expresses the nutrient utilization rate as a mass transfer coefficient from which the scale-effects of stream size (Table 7.3), specifically depth and velocity, on uptake length have been removed. The transformation from S to vf removed most but not all of the correlation with stream size. Among the measurements made in 2004, vf for phosphate correlated with streamflow (r=0.67), and among all measurements made between 2000 and 2004 (the “5-year dataset,” n=49-50), vf for both phosphorus and ammonium correlated with streamflow (r=0.36 and 0.29, respectively), velocity (r=0.30 and 0.29), and depth (r=0.27 and 0.30). These correlations, which are much smaller than those with uptake length, may arise from secondary influences of stream size, such as a stimulatory effect of current velocity on nutrient uptake (Borchardt et al. 1994) or of reduced riparian shading on primary production (see Chapter 8). However, they may be the co-incidental result of a potentially confounding correlation of nutrient concentration with stream size. In the 5-year dataset background SRP, total dissolved phosphorus (TDP), and total dissolved nitrogen (TDN) were all negatively correlated with streamflow (r=-0.35, -0.38, and -0.28, respectively) and these correlations, in turn, reflect the regional pattern of larger streams with lower nutrient concentrations west of the Hudson, relative to streams located east of the Hudson (see Chapter 3). The influence of nutrient concentration on uptake velocity is presented below.

Phosphate uptake velocity in 2004 ranged from 0.007 to 0.045 mm/s (Table 7.2), with a mean of 0.021 (±0.010, standard deviation) mm/s, somewhat higher than the mean for all streams (0.018 mm/s) measured in the years (2000-2003). In 2004, there was a weak negative correlation between the vf of phosphate and the background concentration of SRP (r=-0.59, P=0.08). In the

- 139 - CHAPTER 7 – NUTRIENT SPIRALING NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

previous analysis of Phase 1 data (2000-2002), SRP was found to be the variable most strongly correlated with phosphate uptake velocity, and this continues to be the case when the data from all 5 years (2000-2004) are considered together. For the 5-year dataset, vf correlated with SRP at r=0.59, the same as for the 2004 data, but significant at P<0.0001 because of the larger sample size (n=50). For both datasets SRP explained 35% (r2) of the variation in uptake. In contrast, the Michaelis-Menten uptake kinetics model (Eqs. 7.1 and 7.2) applied to the 5-year dataset explained 60% of the variance in uptake velocity (Fig. 7.1). The corresponding parameter -2 -1 estimates were: Ks=0.0096 mg/L and Umax=0.33 µg m s , which are comparable to those -2 -1 previously obtained for the 2000-2002 data: Ks =0.007 mg/L and Umax =0.26 µg m s .

Uptake velocity, vf, of ammonium ranged from 0.035 to 0.079 mm/s, with a mean of 0.048 (±0.013) mm/s, which is below the 2000-2003 mean of 0.060 mm/s. Unlike previous years, the ammonium uptake velocity in 2004 did not correlate significantly with background concentrations of dissolved nitrogen (ammonium, nitrate, or TDN). However, the ammonium uptake velocities measured in 2004 were generally consistent with the previous years’ data (Fig. 7.2), and the absence of a correlation in 2004 may be in part attributable to the somewhat higher values of background TDN measured in 2004 in comparison with previous years. For the 5-year dataset, ammonium vf correlated with background TDN and nitrate (r=-0.36 for both analytes), and less strongly with background ammonium (r=-0.20). For this reason, TDN, rather than ammonium was used to fit the Michaelis-Menten uptake kinetics model. This substitution assumes that dissolved nitrogen in forms other than ammonium are utilized by benthic microbes and, when present in abundance, can reduce the demand for ammonium. The Michaelis-Menten model (Fig. 7.2) explained 20% of the variance in ammonium uptake, which is more than the linear correlation (r2=0.13, or 13%), but considerably less than the 60% obtained for the phosphorus model. The resulting parameter estimate of the half-saturation constant, Ks , was 1.9 mg/L of TDN. The parameter Umax could not be estimated because of the substitution of TDN for ammonium. The previous estimate for Ks, based on the 2000-2002 data was 1.0 mg/L, and for those years the model explained 38% of the variance.

The mean glucose uptake velocity was 0.063±0.028 mm/s, nearly the same as that of the previous 4 years (0.061±0.032 mm/s). The mean uptake velocity for arabinose was 0.021±0.006 mm/s, also similar to the mean of the previous four years (0.022±0.010). Glucose and arabinose uptake velocities were not significantly correlated with each other (r=0.42, P=0.22), in contrast to the strong correlation (r=0.85) in the 5-year dataset. There were no significant correlations of uptake velocity for either glucose or arabinose with their respective background concentrations (Table 7.11), indicating these carbohydrates did not occur in sufficient concentrations to saturate uptake.

The hydraulic exchange velocity (HEV) ranged from 0.01 to 0.25 mm/s with a mean of 0.09 mm/s, considerably higher than the mean of 0.028 mm/s from 2000-2003. There were no significant correlations between HEV and the uptake velocity for any of the four nutrients. The standardized transient storage volume (As/A) ranged from 0.01 to 0.31, with a mean of 0.09, somewhat lower than the 2000-2003 mean of 15%.

Interannual Variability

- 140 - CHAPTER 7 – NUTRIENT SPIRALING NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Among the 17 sites sampled since 2000, seven were sampled three times (Phase 1), another seven twice (Phase 2 to date), and three were sample five times, i.e., once each year. For both phosphorus (Fig. 7.3 ) and ammonium (Fig. 7.4), the range of inter-annual variation in uptake velocities generally increased with increasing mean uptake and the variation was large in relation to site-to-site differences. Yet, for both analytes, there was clearly some consistency within sites, particularly those with the lowest mean annual uptake velocities. A two-way analysis of variance (SAS, general linear model) on log-transformed uptake velocities confirmed a significant (P=0.03) site effect for both phosphate (P=0.002), and ammonium (P=0.04) uptake, but no significant year effects (P>0.05). To examine the sources of the inter-annual variation an analysis of covariance was conducted using site as a class variable and background nutrient concentrations and streamflow as covariates. For uptake of both ammonium and phosphorus, the covariates failed to explain significant variation beyond that accounted for by site alone. That is, nearly all of the variation in uptake that is attributable to nutrient concentration and flow (as discussed above) occurs between sites, and none can be unequivocally ascribed to annual variation within sites.

Associations with Ecosystem Metabolism and Water Quality Scores

In Phase I analyses (years 2000-2002), associations of nutrient uptake were observed with community respiration (Chapter 8), and the macroinvertebrate Water Quality Score (Chapter 5). Using the 5-year dataset, community respiration correlated with uptake velocities of phosphate (r = 0.53, Fig. 7.5), glucose (r = 0.32, Fig. 7.6), and arabinose (r = 0.33, not shown), but only marginally with ammonium uptake (r = 0.27, P = 0.07, Fig. 7.7). The benthic macroinvertebrate Water Quality Score correlated with the uptake velocity of phosphate (r = 0.30, 7.8), but not with the uptake of other nutrients.

Literature Cited

Borchardt, M. A., J. P. Hoffmann, and P. W. Cook. 1994. Phosphorus uptake kinetics of Spirogyra fluviatilis (Charophyceae) in flowing water. Journal of Phycology 30: 403-417. Cheng, X. and L. A. Kaplan. 2001. Improved analysis of dissolved carbohydrates in stream water with HPLC-PAD. Analytical Chemistry 73: 458-461. Peterson, B. J., W. M. Wollheim, P. J. Mulholland, J. R. Webster, J. L. Meyer, J. L. Tank, E. Martí, W. B. Bowden, H. M. Valett, A. E. Hershey, W. H. McDowell, W. K. Dodds, S. K. Hamilton, S. V. Gregory, and D. D. Morrall. 2001. Controls of nitrogen export from watersheds by headwater streams. Science 292:86-90. Runkel, R. L. 1998. One-dimensional transport with inflow and storage (OTIS): A solute transport model for streams and rivers. U. S. Geological Survey Water-Resources Investigations Report 98-4018. Stream Solute Workshop. 1990. Concepts and methods for assessing solute dynamics in stream ecosystems. Journal of the North American Benthological Society 9(2): 95-119.

- 141 - CHAPTER 7 – NUTRIENT SPIRALING NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

314 688 224 753 4338 1271 1464 2202 1585 1435 Glucose (291, 342) (608, 792) (191, 270) (642, 909) (3826, 5013) (1171, 1390) (1397, 1539) (2063, 2362) (1459, 1733) (1363, 1515) 466 1201 1753 4796 5051 6676 1229 5444 6882 18051 Arabinose (414, 533) ce intervals are listed in (1102, 1319) (1614, 1916) (4090, 5800) (4486, 5780) (5935, 7622) (1163, 1303) (4892, 6131) (6094, 7905) (14451, 23981) )* ∞ 739 SRP 1901 1900 9276 3237 5081 3490 4105 6112 4662 (1732, (445, 2178) (1226, 4232) (1662, 2218) (3093, 3396) (3115, 3968) (5708, 6583) (4120, 5368) (7806, 11416) (3102, 14025) -N 4 532 353 784 1394 4902 1406 2291 1976 2042 1818 jections in 2004. 95% Confiden NH (446, 658) (297, 433) (695, 899) (1107, 1881) (2166, 2431) (1714, 2332) (1866, 2256) (1683, 1976) (3643, 7496 ) (1271, 1572 )

Date 9-Jun-04 6-Oct-04 1-Sep-04 16-Jun-04 23-Jun-04 30-Jun-04 13-Oct-04 11-Aug-04 18-Aug-04 19-May-04

3 9 mated from nutrient in 52 34 10 15 46 29 139 130 was not significant (P>0.05). Station # Uptake lengths (m) esti parentheses. Stream Cross R., Ward Pound Ridge Reserve Haviland Hollow E. Br. Delaware R., Arkville Tremper Kill Muscoot R., Baldwin Whitehall Muscoot R., Neversink R. Salem Center R. nr Titicus W. Br. Delaware R., Kortright Trout Creek *Uptake of SRP at Station 130 Table 7.1:

- 142 - CHAPTER 7 – NUTRIENT SPIRALING NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Table 7.2 Uptake velocity, vf (mm/s), estimated from nutrient injections in 2004. Stream Station # Date Ammonium Phosphate Arabinose Glucose Cross R. 52 16-Jun-04 0.048 0.014 0.021 0.082

Haviland Hollow 34 19-May-04 0.041 0.030 0.032 0.082

E. Br. Delaware R., 10 6-Oct-04 0.043 0.023 0.012 0.048 Arkville. Tremperkill 15 1-Sep-04 0.050 0.022 0.015 0.055

Muscoot R., B. 46 23-Jun-04 0.035 0.017 0.026 0.055

Muscoot R., W.C.* 139 30-Jun-04 0.037 0.016 0.017 0.057

Neversink R. 29 11-Aug-04 0.079 0.045 0.023 0.071

Titicus 130 9-Jun-04 0.038 0.007** 0.024 0.039

W. Br. Del. R., K. 3 13-Oct-04 0.058 0.020 0.022 0.075

Trout Creek 9 18-Aug-04 0.049 0.019 0.013 0.063

Mean 0.048 0.021 0.021 0.063

Std. Deviation 0.013 0.010 0.006 0.014

* - Ammonium uptake estimate at Muscoot R. Whitehall was inferred from apparent upstream ammonium source that exceeded experimental input. ** - Not significant

- 143 - CHAPTER 7 – NUTRIENT SPIRALING NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

0.022 0.012 0.028 0.045 0.015 0.253 0.146 0.068 0.240 0.040 0.087 0.042 0.012 0.253 (mm/s) Velocity Exchange Hydraulic /A)

S 0.08 0.01 0.08 0.11 0.13 0.13 0.23 0.16 0.31 0.10 0.13 0.12 0.01 0.31 (A Transient Standardized Storage Volume /s)

2 0.40 1.60 8.72 2.57 0.32 0.83 2.95 0.58 1.15 3.03 2.21 1.38 0.32 8.72 (m Dispersion Longitudinal

0.20 0.23 0.46 0.28 0.11 0.24 0.46 0.15 0.31 0.35 0.28 0.26 0.11 0.46 Average Depth (m) injections in 2004.

7.2 7.3 6.7 8.5 7.4 9.6 6.7 16.1 11.0 10.7 22.2 11.5 10.9 22.2 Average Width (m) /s)

3 ured during nutrient 0.18 0.41 3.37 0.78 0.08 0.89 3.47 0.25 1.38 0.67 1.15 0.72 0.08 3.47 (m Average Streamflow Streamflow

0.127 0.242 0.458 0.248 0.107 0.347 0.337 0.202 0.391 0.256 0.271 0.252 0.107 0.458 Water Average Velocity (m/s)

# 3 9 52 34 10 15 46 29 139 130 Station Channel and flow characteristics meas Table 7.3: Stream Cross R. Haviland Hollow E. Br. Del. R., A. Tremperkill Muscoot R., B. Muscoot R., W.C. Neversink R. Titicus W. Br. Del. R., K. Trout Creek mean median min max

- 144 - CHAPTER 7 – NUTRIENT SPIRALING NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

1.45 1.45 1.81 1.09 0.55 1.43 1.31 0.50 2.71 0.31 0.87 VSS VSS (mg/L) (mg/L) TSS 3.40 3.40 5.04 1.86 1.21 3.96 3.26 1.37 8.55 1.84 <1.2 (mg/L) (mg/L) 8.32 8.32 8.69 8.86 5.52 9.36 4.69 (nM) (nM) 15.05 15.05 11.24 20.22 13.92 Glucose Glucose <2 <2 <2 <2 <2 <2 <2 <2 <2 2.57 2.57 (nM) (nM) Arabinose Arabinose PP 0.014 0.014 0.007 0.005 0.003 0.036 0.010 0.000 0.018 0.002 0.005 (mg/L) (mg/L) TP 0.044 0.044 0.017 0.017 0.011 0.074 0.017 0.006 0.048 0.014 0.016 (mg/L) (mg/L) TDP 0.030 0.030 0.010 0.012 0.008 0.038 0.007 0.011 0.030 0.012 0.012 (mg/L) (mg/L) SRP 0.027 0.027 0.003 0.011 0.006 0.039 0.002 0.002 0.024 0.013 0.010 (mg/L) (mg/L) PN 0.068 0.068 0.042 0.000 0.006 0.078 0.099 0.000 0.045 0.000 0.027 (mg/L) (mg/L) TKN 0.314 0.314 0.205 0.130 0.133 0.378 0.695 0.127 0.338 0.177 0.146 (mg/L) (mg/L) SKN SKN 0.247 0.247 0.163 0.133 0.128 0.301 0.596 0.406 0.293 0.182 0.119 (mg/L) (mg/L) -N -N 3 0.357 0.357 0.262 0.208 0.264 1.409 1.484 0.192 0.629 1.004 0.261 NO (mg/L) (mg/L) -N -N 4 0.014 0.014 0.021 0.007 0.009 0.010 0.277 0.008 0.023 0.007 0.008 NH (mg/L) (mg/L) Date 9-Jun-04 9-Jun-04 6-Oct-04 6-Oct-04 1-Sep-04 16-Jun-04 16-Jun-04 23-Jun-04 23-Jun-04 30-Jun-04 13-Oct-04 13-Oct-04 18-Aug-04 18-Aug-04 11-Aug-04 11-Aug-04 19-May-04 19-May-04 # 3 9 52 34 10 15 46 29 139 139 130 Site Background nutrient concentrationsduringnutrient measured injections 2004. in Table 7.4. Table Stream Cross R. Haviland Hollow E.B. Del., A. Tremperkill Muscoot R., B. MuscootW.C. R., R. Neversink Titicus Del., K. W. B. Trout Creek

- 145 - CHAPTER 7 – NUTRIENT SPIRALING NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Figure 7.1: Uptake velocity, vf , of phosphate measured in 2004 (black dots) plotted with measurements from 2000-2003 (circles). The curve represents the fitted uptake kinetics model (Eq. 7.2).

Figure 7.2: Uptake velocity, vf , of ammonium measured in 2004 (black dots) plotted with measurements from 2000-2003 (circles). The curve represents the fitted uptake kinetics model (Eq. 7.2).

- 146 - CHAPTER 7 – NUTRIENT SPIRALING NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Figure 7.3. Inter-annual within-site variability in phosphate uptake velocity. The dots represent measurements from a single year, the line connects the means within sites.

Figure 7.4. Inter-annual within-site variability in ammonium uptake velocity. The dots represent measurements from a single year, the line connects the means within sites.

- 147 - CHAPTER 7 – NUTRIENT SPIRALING NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Figure 7.5: Uptake velocity of phosphate versus community respiration. Symbols as in Fig. 7.1

Figure 7.6: Uptake velocity of glucose versus community respiration. Symbols as in Fig. 7.1

- 148 - CHAPTER 7 – NUTRIENT SPIRALING NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Figure 7.7: Uptake velocity of ammonium versus community respiration. Symbols as in Fig. 7.1

Figure 7.8: Uptake velocity of phosphate versus benthic macroinvertebrate-based Water Quality Score. Symbols as in Fig. 7.1

- 149 - CHAPTER 7 – NUTRIENT SPIRALING NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

------Intentionally Blank ------

- 150 - CHAPTER 7 – NUTRIENT SPIRALING NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Chapter 8 - Stream Metabolism

Purpose and Significance

Metabolism measurements, conducted at the 10 integrated study sites, provide data on the ecosystem processes of primary productivity and community respiration. Primary productivity measures the rate of synthesis of plant biomass, and respiration, the rate of utilization of reduced chemical energy, including the metabolic costs of photosynthesis. Algal productivity predominates primary productivity in the streams studied here. The goals of these studies are to rank sites according to the intensity of these metabolic processes and to relate our findings to environmental correlates. The expectation is that these ecosystem functions will be related principally to the biomasses of algae, heterotrophic microorganisms, and (to a lesser extent) macrophytes and macroinvertebrates and also to the environmental variables of light, temperature, and dissolved and particulate nutrients. Some of those environmental variables, in turn, are related to watershed uses and sources of contaminants. Changes in process rates or in the balance of these functions over time would indicate changes in watershed activities and signal that investigative work on upstream tributaries and the watershed is needed to address causative factors. These measures add a functional dimension to this research program that complements descriptive variables (e.g., nutrient concentrations, invertebrate densities).

Three streams studied as integrative sites in Phase I (Neversink [29], Muscoot at Baldwin [site 46], and Cross [site 52]) were retained as study sites for Phase II. Seven new streams were added: West Br. Delaware at South Kortright (site 3), Trout Creek (site 9), East Br. Delaware (site 10), Tremper Kill near Andes (site 15), Haviland Hollow (site 34), Muscoot near Whitehall Corners (site 139), and Titicus River near Salem Center (site 130) (see Chapter 2).

Methods

Field procedures

Community metabolism was determined using open-system measurements of dissolved O2 change. Pairs of sondes (YSI model 600XLM, Yellow Springs Inc., Yellow Springs, OH) were deployed at the upstream and downstream ends of each study reach to measure and log to memory dissolved O2 and temperature for three-day periods. The EPA-approved 600XLM sonde coupled with a rapid pulse dissolved O2 probe has a manufacturer-certified precision of 0.01 mg/L and an accuracy of ± 0.2 mg/L. Temperature was monitored with a manufacturer-specified precision of 0.01°C and an accuracy of ± 0.15°C. The study reaches were those referred to in the nutrient spiraling studies. Spatially, they included a top (injection) substation, an upstream sonde substation approximately mid-way through the reach, and a downstream sonde substation. Conditions affecting reaeration were similar above the upstream and the downstream sondes. Reach lengths in 2004 were nearly identical to those in 2003 at Haviland Hollow (610 m both years), the Titicus (350 m and 327 m, respectively), and the Neversink (1591 m and 1754 m). Reach lengths in 2004 were less than 2003 for the Cross (492 m and 693 m, respectively), Tremper Kill (885 m and 1202 m), Muscoot at Baldwin (214 m and 708 m), Muscoot at Whitehall (1046 m and 1529 m), but were greater on the East Br. Delaware (1445 m and 1175 m, respectively), West Br. Delaware at South Kortright (1446 m and 745 m), and Trout Creek (865 m and 385 m).

- 151 - CHAPTER 8 – STREAM METABOLISM NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

The sondes (including one designated as QAQC) were placed in water-saturated Turkish towels (M. Lizzote, YSI, personal communication) at the field site and calibrated according to the manufacturer’s instructions. The sondes then were placed at a single location in the thalweg of the study stream for an overnight (7-12 h) period prior to deployment. Differences between sondes were used to calculate offsets that were applied when analyzing data according to the upstream-downstream approach. Two sondes were transferred to the upstream and downstream substations, with pairings based on the similarities of dissolved O2 concentrations toward the end of the field calibration period and probe characteristics (e.g., DO charge and voltage). Dissolved O2 concentration and water temperature were measured and logged at 15-min. intervals for 3 diel periods (diel = 25 hr period). QAQC checks were made each day by securing the QAQC sonde to the stake holding the data sondes. After a 0.5 h equilibration period, instantaneous measures of dissolved O2, % saturation, temperature, conductivity, and DO charge were taken for each data sonde and compared to the QAQC sonde using a YSI 650MDS meter.

Above-water photosynthetically active radiation (PAR) at the upstream and downstream locations was measured at 15 sec intervals using four quantum sensors (Li-Cor, Lincoln, NE) secured to stakes (two per location). Each 15-min. average was logged on a Li-COR 1400 data logger.

Reaeration coefficients were determined empirically from a propane evasion experiment (Marzolf et al. 1994, Young and Huryn 1998, Marzolf et al. 1998) performed once during each measurement period, in conjunction with nutrient spiraling studies. On the day prior to the experiment, the time of travel of water through the study reach was estimated using rhodamine WT. Data were used to assign sampling times for the propane evasion experiment. For that experiment propane was bubbled from two barbecue tanks into the stream using gas diffuser tubes at the injection site (a point ~ 50 - 175 m upstream from the uppermost sampling substation). A bromide conservative tracer solution was injected simultaneously a few cm upstream of the propane using a peristaltic pump. Sources were mixed by the bubbling propane and turbulence during transit from the injection point to the upstream sampling substation. Five sampling substations were set over the length of the study reach. The entire injection was monitored for bromide at the top substation and at either the fourth or fifth downstream substation (with 5 propane samples taken at 2-10-min. intervals when concentrations were at the plateau). Only plateau samples were collected for both propane and bromide at the remaining substations. Field blanks were collected at each substation prior to the start of the injection. For each field injection, a standard curve of propane concentration was prepared by diluting water from the plateau (maximum propane concentration) at the uppermost sampling substation to three lower percentages (50%, 10%, 1%) in site water collected prior to the injection. Bromide samples were collected in 500-ml plastic bottles. Subsamples were drawn into a syringe and filtered through a 0.22µm pore size Millipore Express PES membrane filter into 25 ml plastic bottles. Propane samples were collected in heavy-walled 75-ml glass serum bottles that were capped with rubber septa and crimp-sealed in the field. Water samples were collected by immersing a bucket into the stream in an upstream direction and filling sample bottles from the bucket. This approach reduced turbulence during sampling. Propane bottles were completely filled (no head space) and were stored under refrigeration.

Open-system metabolism measures include both benthic and water column activity. Water column metabolism was measured separately at each site as follows. Ten BOD bottles (six light and four dark) were filled with stream water. Initial dissolved O2 concentration, temperature and percent saturation were measured in each bottle using a YSI Model 58 DO meter and probe with - 152 - CHAPTER 8 – STREAM METABOLISM NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

stirrer suitable for use with BOD bottles. Water used for incubation in the bottles was bubbled with N2 gas to lower the percent oxygen saturation to 77 – 91 %. The bottles were then incubated in the stream for a 4 - 6 h period. PAR was monitored during the incubation. At the end of the incubation period the dissolved O2 concentrations were again determined.

Types of streambed substrata and periphyton in each reach were characterized by a mapping effort. Twenty transects were set at intervals between the top and bottom sondes. At each transect the width of the river was measured, and 10 - 12 equidistant lateral sampling points were set. At each point, river depth was measured and the appearance of both the bed (substratum) and biomass attached to the substratum (referred to as “cover type”) were characterized using a viewing bucket. Our substrata categories follow those of Hynes (1970), except that our pebble category included material he classified as gravel. Width and depth of the reach was measured at another 8-15 transects between the top sonde and the injection point.

Benthic samples for analysis of chlorophyll a and organic mass were collected from substrata constituting 10% or more of the cover types encountered during the mapping effort. Samples of soft substrata were collected by inserting a plastic tube (100 cm2 id) into the river bed to isolate a portion of sediment and removing surface sediments with a meat baster. Samples of periphyton on rocks were scraped, brushed and washed into a jar. Samples were held on ice until return to the laboratory. The planar surface area of the rock was traced onto a piece of paper.

Laboratory analyses

Bromide was analyzed by ion chromatography (see Chapter 7). For propane analyses, two needles were inserted through the rubber septum and 10 ml of water were displaced through one by injecting air through the other into the serum bottle to produce a head-space. Bottles were shaken for 3 h at room temperature to equilibrate propane between the water and head-space. Samples (50 µl) of head-space gases were analyzed using a capillary gas chromatograph with flame ionization detector and helium carrier gas. Propane concentrations (determined from the standard curve) were ~100% at each top substation and ranged between <10 % and ~60% of that at the most downstream substation. Absolute concentrations are not critical to assessing reaeration; proportional loss over distance is used to compute the coefficient.

Periphyton samples for chlorophyll analyses were centrifuged at the field laboratory and frozen until extraction. Samples were extracted overnight in acetone (made basic by adding a pinch of MgCO3 to the reagent bottle). The next day samples were centrifuged for 10 - 20 min at 10,000 x g at 4°C and the optical densities of the supernatant fluids were determined at 665 nm and 750 nm (for turbidity) before and after acidification with 2 drops of 1 N HCl. Acetone was removed and samples were re-extracted with additional acetone until the extract yielded a chlorophyll a absorbance that was either 10 - 15% of the OD for the first extract of that sample or <0.1 absorbance units at 665 nm. Chlorophyll samples were iced and handled under low light during analyses. Concentrations were determined using the equations of Lorenzen (1967, APHA 1992) with correction for pheophytin. A subset of the scraped rocks were frozen and brought to the laboratory for chlorophyll analyses in order to determine how completely the rock was scraped. Prior to analysis, the bottom side of the rock (marked in the field) was flamed with a blowtorch (while the top was immersed in water). Since rocks may tumble in the field this step reduced the chance of including chlorophyll from the bottom of the rock in the estimate of rock-associated unscraped chlorophyll. The ratio of scraped to total chlorophyll (scraped and unscraped) is an

- 153 - CHAPTER 8 – STREAM METABOLISM NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

estimate of how completely we scraped the rocks in the field. Both the spectrophotometer and Mettler balance were recalibrated at the beginning of the field season.

Matrix spikes for chlorophyll were performed as follows. In the laboratory, samples from a subset of rocks were divided into three aliquots of equal wet weight. The chlorophyll in two subsamples was determined and used as lab duplicates. The mean concentration was used as a guide for the appropriate addition of chlorophyll standard to generate a matrix spike for the third subsample.

Following extraction the sample was dried at 60 °C, weighed, ashed (450°C for 6 h), cooled, and reweighed for an analysis of organic matter content (ash free dry mass, AFDM).

Rock outlines were digitized and planar surface area was determined using the public domain NIH Image software (developed at the U.S. National Institutes of Health and available on the Internet at hhtp://rsb.info.nih.gov/nih-image/.

Data analysis

Chlorophyll concentrations were measured for the most important algal cover types, which in 2004 accrued to 89 - 100% of the cover types in all streams. Chlorophyll concentrations were matched with the percentage of total reach area of that cover type to generate a weighted chlorophyll concentration.m-2. Values for organic matter content were treated similarly to generate a weighted estimated for the reach.

Propane data from each sampling station were normalized for bromide concentration and regressed against downstream distance using non-linear regression (SAS, Institute, Cary, NC) (after Wanninkhof et al. 1990). The derivative represents the proportion of propane (corrected for dilution) lost/m, which when multiplied by water velocity, 1.39 (to correct for molecular size) and 60 (sec/min) yields a KO2. Water velocity through the reach and mean depth of the reach were derived from the computer modeling of bromide concentrations using the OTIS-P model as described in Chapter 7. As a back-up to the propane technique, reaeration was also computed from geomorphic parameters entered into a surface reaeration model (Owens 1974) and an energy dissipation model (Tsivoglou and Neal 1976, APHA et al. 1992).

Oxygen data for metabolism estimates were analyzed using the two-station (upstream - downstream) approach (Owens 1974) using SAS software. The rate of change of dissolved O2 concentration (Odum 1956) corrected for reaeration was computed at each 15-min interval over a 24 h diel period. The average hourly rate of community respiration during darkness (corrected for reaeration) was extrapolated to 24 h (CR24). Gross Primary Productivity (GPP) was computed by adding daytime respiration to net oxygen change during the photoperiod. Net daily metabolism was computed as the difference between GPP and CR24 (NDM = GPP - CR24). In addition data were analyzed using the single-station approach applied to the individual upstream and downstream data sets

The separate measurements of water column metabolism were analyzed as follows. Dissolved O2 changes in dark bottles were averaged and that value was added to oxygen changes in the light bottles to yield, after averaging, an estimate of GPP in the water column. Whole system metabolism for the corresponding time period was determined by integration of the area under

- 154 - CHAPTER 8 – STREAM METABOLISM NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

the diel rate of change curve using the overnight respiration rate extrapolated through the daylight hours as a baseline.

Differences between streams were determined using ANOVA and the Scheffe Multiple Range test (p = 0.05) on log10-transformed data with a constant added before the transformation when appropriate. Correlations were performed using log10 transformed variables.

QAQC Results

Chlorophyll a

Rock scrapings were analyzed for chlorophyll content within 21 days except for the samples from the West Br. Delaware (S. Kortright) and Tremper Kill, which were analyzed at 22 and 24 days, respectively. The common unit for comparing samples, blanks, and laboratory control standards is µg chlorophyll a.sample-1. Blanks had negligible absorbance and OD at 665 nm for them averaged 0.000 ± 0.002 (x ± SD, n=267) whereas the OD665 for the first extract of river samples was 0.500 ± 0.523 (x ± SD, n=595). The OD665 of 95% of the most dilute extracts (Ext. 6) was > 0.067 whereas the absorbance of 95% of the blanks was < 0.003, approximately a 22- fold difference in cutoff between blanks and samples after 6 extractions (22 samples), and 13 samples required 7 extracts, 9 eight extracts, 4 nine extracts, and 1 ten extracts. The consistently low blanks confirm that there is little between-sample contamination and that the spectrophotometer was working properly. However, blanks do not enter into the calculation of chlorophyll concentrations in samples, which are based on turbidity corrected changes in OD at 665 nm.

Measured chlorophyll concentrations in laboratory control standards analyzed with river samples averaged 100.3 ± 2.0 % of the expected concentration (34.5 µg/sample). Standards do not enter into our calculation of chlorophyll concentrations in samples.

The accuracy of field procedures was tested by determining how completely a rock was scraped. For this, 42 rocks were analyzed for the amount of chlorophyll remaining on them after the periphyton was scraped. Thirty-eight of the 42 rocks were assayed within 30 days of collection. Sixty-nine percent of the rocks were scraped with 70% or better efficiency. Only 12% of the rocks (5) were scraped with less than 50% efficiency and the cover type on those rocks was “green algae”, some species of which, e.g., Cladophora, are difficult to scrape completely, or “bare”, which have low chlorophyll concentrations.

The precision of chlorophyll a determinations in the laboratory was based on the analysis of laboratory duplicates. The average RPD for the laboratory dups was 48.7%. Only 7 out of 28 pairs were within a 20% target RPD, although another 6 samples were in the range of 20 – 30.13%. Matrix spikes were added to 37 samples with an overall recovery of 93.14 ± 14.25% of the added chlorophyll. Most recoveries were between 80 and 100%, and only three samples had recoveries < 67%. Given the poor reproducibility of the laboratory duplicate samples, it is somewhat surprising that the recovery of matrix spikes was as good as it was. Note, however, that in computing environmental concentrations the sequential extraction procedure we employ assures nearly complete recovery of chlorophyll and improves the accuracy of data from field samples.

- 155 - CHAPTER 8 – STREAM METABOLISM NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Propane

All samples were analyzed within the 21-day holding period. Field duplicates were collected at two or three substations on each river during the propane injections. The RPD of field duplicates averaged 4.6 ± 3.9 (x ± SD, n=29) and all were within 17.2%. In addition to this precision, our propane data are in a time series that makes it easier to identify outliers at any sampling location. Absolute propane concentrations are not required to estimate reaeration, but propane values for samples collected at the top (Substation 1), middle (Substation 3), and bottom (Substation 5) of the reach were compared with data for propane blanks and standards. Values at the most downstream station were usually from 3 to 5 fold higher than the blanks, and were in the range of the blanks only on the Neversink and East Br. Delaware. Values at the most downstream station were close to (or twice) the 10% concentration measured in the standard curve for that stream.

Dissolved oxygen

The calibration of the data sondes was checked daily against the QAQC sonde yielding 134 such QAQC checks. Overall the mean percent difference between data sondes and the QAQC sonde was 1.005 ± 0.667 % (x ± SD, n=134). In contrast, the overall mean percent difference between data sondes and Winkler titrations of dissolved O2 was 9.67 ± 1.73 % (x ± SD, n= 66). We calibrated the sondes against air but made Winkler determinations of dissolved oxygen to allow documentation of our data against historical records or Winkler-based data from other sources.

Results, Discussion and Conclusions

Benthic substrata and cover types

The proportions of benthic substrata in each study stream are shown in Fig. 8.2. Cobble was the predominant substrate type encountered in all WOH rivers, comprising approximately 40 - 50 % of bed material, but this figure was down from the proportions reported last year. The percentage of cobble encountered in EOH streams was lower, ranging from 24% to 39%, except for a value of 46% at Muscoot (Baldwin). Of the WOH streams, Trout Creek had the greatest proportion of soft substrata (sand, silt), amounting to 14.6 %, which was roughly twice that at the other WOH rivers. The EOH streams had greater percentages of soft substrata (10 – 20 %) and an exceptionally high value of nearly 40% in the Cross. Three streams, West Br. Delaware (South Kortright), East Br. Delaware, and Titicus were visited twice during the season because heavy rains prevented completion of work on the first visit. This allowed computation of the RPD for different substrate categories between visits. For substrate categories of boulder, cobble, and pebble the RPDs were < 34.7%, but for the silt and sand categories they were between 50 and 180 %. Those categories would be expected to have a higher RPD given their capacity for resuspension and deposition between visits.

Major categories of plant cover types (macroscopic appearance through the viewing bucket) were bare, black cover (a slime scraped from black colored rocks), blue green algae (Cyanobacteria with blue green or black velvet appearance), diatoms (brown velvet appearing growths or filamentous, e.g., Melosira spp.), green filamentous algae, green cover (green slime), leaf pack, moss, macrophytes, silt, and tufts (filamentous algae or moss either in an immature state or following scour often overlain with silt). Microscopic examination has documented the presence of diatoms in the black covers and silt cover types. - 156 - CHAPTER 8 – STREAM METABOLISM NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

The percentage encounters of different cover types in each stream are shown in Fig. 8.3. Diatoms were the predominant algal cover type in the East Br. Delaware, Tremper Kill, and West Br. Delaware, comprising between 42 and 58% of the encounters. Green filamentous algae made up ~17% of the encounters in the West Br. Delaware (S. Kortright) and 16 % of encounters in the Neversink. A green cover occurred in the Neversink (54% of encounters), and made up between 14.5 and 23 % of encounters in the Tremper Kill, Titicus, East Br. Delaware, Cross, and Trout Creek. Bare substrata were encountered at a significant number of points in Trout Creek, East Br. Delaware, and the Cross River. Black cover was noted most often in the Titicus and both stations on the Muscoot. Tufts were significant component of cover types only in the Muscoot at Whitehall Corners. Mosses made up ~ 15 % of the encounters at Haviland Hollow, Muscoot (Whitehall Corners), and the Neversink. Macrophytes never accounted for significant biomass, except in the Cross, where they made up ~ 10 % of the encounters. Silt with associated diatoms was a significant cover type only in Trout Creek out of the WOH rivers, where the stream gradient was low, but was important in several EOH streams, viz., Haviland Hollow, Cross, both stations on the Muscoot, and the Titicus. As for substrate categories, the two visits to the West Br. Delaware (S. Kortright), East Br. Delaware, and Titicus allowed for the computation of the RPD for different cover types between visits. About half of the RPDs were < 50 % and the other half > 50%. There was little consistency in RPD within types of cover, suggesting that seasonal differences in temperature, light, nutrient status, and time since last storm had greater influence on cover type than substratum type.

Benthic algal chlorophyll a data are presented in Fig. 8.4. Chlorophyll data are based on samples from > 94 % of cover types for all streams but Titicus (85%), Muscoot (Baldwin) (85%), and Cross (87%). Total chlorophyll values ranged from 17.4 mg/m2 (Neversink) to 121.3 mg/m2 (West Br. Delaware). Diatoms made the greatest contribution to total chlorophyll a in the East and West Branches of the Delaware, Tremper Kill, Muscoot (Baldwin), and Trout Creek. Of the WOH streams, Trout Creek had the lowest chlorophyll concentration and West Br. Delaware the highest. Of the EOH streams, Haviland Hollow the highest concentration and Muscoot (Baldwin) and Titicus the lowest. Except for the West Br. Delaware (S. Kortright), East Br. Delaware, Tremper Kill, and Haviland Hollow, concentrations were below 35 mg/m2. Moss made the greatest contribution to total chlorophyll in the Neversink (~ 48 %), a minor contribution in the East Br. Delaware (7 %), and modest contributions in the West Br. Delaware (18 %), Haviland Hollow (27 %), and Muscoot (Whitehall) (32%). The RPDs in chlorophyll a per cover type were usually ≤ 50% but when weighted for the type of cover in the reach the RPD values increased greatly, reflecting the variability in cover type in the reach between visits. Although streams ranked differently in chlorophyll a between 2003 and 2004, concentrations were in the same range each year.

Weights of benthic organic matter associated with periphyton were greatest at both stations in the Muscoot and Haviland Hollow (> 31 g/m2). Elsewhere concentrations were < 20 g/m2, and ranged down to 6.5 – 7.0 g/m2 in the Tremper Kill and East Br. Delaware. Streams ranked quite differently than for chlorophyll a (Fig. 8.5). Organic matter concentrations were higher in 2004 than in 2003, probably the result of less intensive or frequent scouring prior to our sampling. For example, gauging records for study sites at USGS gauging stations indicated that total discharge for the 90 days prior to our 2003 and 2004 field visits were greater in 2003 (Tremper Kill, Neversink, Muscoot nr Baldwin, and Cross River) and peak instantaneous discharge 30 days prior to our sampling was also greatest in 2003 for these same sites.

- 157 - CHAPTER 8 – STREAM METABOLISM NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Periphyton chlorophyll a showed no statistically significant correlations with any environmental variable. Periphyton organic matter (AFDM) correlated positively with NH4-N and total Kjeldahl N (r = 0.693, p = 0.04 and r = 0.770, p = 0.007, respectively).

Stream Ecosystem Metabolism

Metabolism was measured between May 18, 2004 (Haviland Hollow) and October 24, 2004 (West Br. Delaware, South Kortright). Except for one day on the West Br. Delaware PAR values were in the range of 13 – 35 mol quanta per day during WOH measures, whereas PAR values were in the range of ~ 1 – 21 mol quanta per day during EOH studies, primarily the result of greater shade especially at Haviland Hollow, Titicus, and Muscoot (Baldwin).

All metabolism estimates for 2004 were based on the two-station method with reaeration determined from propane evasion, with data collected for three days on all streams but the . -2. -1 Neversink, which had only two. Mean values for GPP ranged from ~0.67 – 4.98 g O2 m day (Fig. 8.6, Table 8.2). As in prior years, the EOH streams tended to rank lower than the WOH streams, with the exception of East Br. Delaware, presumably a function of lower light levels at many of the EOH sites. The low activity in East Br. Delaware is not a function of light levels which were moderately high, but rather of biomass. In 2003 the East Br. Delaware had greatest periphyton biomass, but in 2004, there were major storms ~ 1 and 3 weeks prior to the study and measurements were conducted at a time the stream was on the receding limb of the hydrograph and recovering from scour. The Cross ranked highest of the EOH sites, and Muscoot (Baldwin) the lowest. The Neversink ranked highest of the WOH sites in GPP normalized for area. As in 2003, concentrations of NH4 were high in the Muscoot at Whitehall (0.277 mg/L vs. 0.007 – 0.023 mg/L elsewhere). GPP per area in the Muscoot (Baldwin) and East Br. Delaware was significantly lower than in the Cross, Tremper Kill, Trout Creek, West Br. Delaware, and Neversink and GPP per area in the Titicus was lower than in Trout Creek, West Br. Delaware, and the Neversink (ANOVA, Scheffé MRT, p ≤ 0.05 on log transformed data).

GPP was normalized for the incident PAR received each day. Photosynthesis has been reported to saturate at PAR intensities between 200 and 400 µmol quanta.m-2.s-1 (Hill 1996). We are in the processes of analyzing photosynthesis-irradiation curves for each daily measurement, but for now the upper end of that range (400 µmol quanta.m-2.s-1) was extrapolated to a total daily PAR saturation value assuming a 12-h photoperiod, yielding a saturation value of 17.28 mol quanta. m-2.d-1. Analysis of Phase I data indicated that photosynthesis saturated at 17.95 mol quanta. m-2.d-1 at the WOH sites, which is very close to the computed value but at a much lower intensity (5.13 mol quanta m-2.d-1) at the EOH sites. We used 17.28 as the saturation intensity for all sites in the analysis described below.

Sites were ranked according to GPP normalized for total daily PAR (GPP/PARtot, Fig. 8.7, top panel) and for GPP normalized for saturating PAR (GPP/PARsat Fig. 8.7, bottom panel), which was computed by substituting 17.28 mol quanta.m-2.d-1 for any PAR value greater than it. Substitution of the saturation intensity changed the rankings only of the Cross River and Trout Creek. GPP normalized for PARtot in the East Br. Delaware was significantly lower than in the Titicus, West Br. Delaware, Haviland Hollow, and the Neversink, and that in Muscoot (Whitehall) was significantly lower than in the Titicus (ANOVA, Scheffé MRT, p ≤ 0.05 on log transformed data). Substitution of saturating intensities (GPP/PARsat) resulted in the inclusion of Trout Creek, Cross River, and Tremper Kill among those streams that differed from the East Br. Delaware (ANOVA, Scheffé MRT, p ≤ 0.05 on log transformed data). - 158 - CHAPTER 8 – STREAM METABOLISM NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Data for water column metabolism, measured using light and dark bottle incubations, are summarized in Table 8.3. Water column metabolism was a negligible proportion of total ecosystem metabolism in all streams and thus system activity was related primarily to benthic periphyton. Oxygen changes even in light bottles were negative in the Cross, East Br. Delaware, and Tremper Kill, indicating a predominance of respiration at the time of our measurements. Water column GPP amounted to a maximum percentage of total system activity (0.72%) in the Muscoot (Baldwin) and West Br. Delaware.

Streams ranked by mean daily respiration (CR24) as shown in Fig. 8.8. Values ranged from~ 3 to . -2. -1 14 g O2 m d , similar to most values for 2003. Highest respiration occurred in the Neversink (as in 2003), and lowest respiration in the Muscoot (Baldwin). Respiration in the Tremper Kill and East Br. Delaware were at the lower end of the WOH streams, and respiration in the Muscoot (Whitehall) and Titicus were at the high end of the EOH streams. Average CR24 in the both Muscoot (Baldwin) and Haviland Hollow was significantly lower than in Titicus, Muscoot (Whitehall), West Br. Delaware, Trout Creek, and Neversink River (ANOVA, Scheffé MRT, p ≤ 0.05 on log transformed absolute values). Respiration in East Br. Delaware was also significantly lower than in Muscoot (Whitehall), West Br. Delaware, Trout Creek, and Neversink River. Respiration in the Tremper Kill was significant lower than in the Titicus, Muscoot (Whitehall), and Neversink and that in the Cross was significantly lower than in the West Br. Delaware, Trout Creek, and Neversink. It is potentially noteworthy that water column respiration (measured during the water column productivity studies) was greatest in Trout Creek followed by West Br. Delaware. Those streams ranked high in total system respiration.

All of the study streams were heterotrophic, i.e., respiration exceeded photosynthesis and net daily metabolism (NDM) was negative (Fig 8.9). This implied a net consumption of energy at the times measurements were made. Reflecting the high respiration rate in the Neversink, NDM was most negative there and the value was significantly lower than in all the other streams (ANOVA, Scheffé MRT, p ≤ 0.05 on log transformed NDM values + a constant). Additionally, Tremper Kill was significantly less heterotrophic than West Br. Delaware, Titicus, Muscoot (Whitehall) and Trout Creek and Haviland hollow was less negative than Muscoot (Whitehall) and Trout Creek.

The rates of GPP and CR24 were examined on a relative basis using the P/R ratio [GPP/CR24], which produced the rankings shown in Fig. 8.10. The highest mean P/R ratio, was 0.70 (Tremper Kill), but the second highest, despite relatively low absolute rates of activity was Haviland Hollow. The lowest P/R ratio occurred in the East Br. Delaware (0.20), and data for the Muscoot (Baldwin) and Titicus were only a little higher (0.22). The remaining streams had ratios between 0.30 (Muscoot Whitehall) and 0.61 (Haviland Hollow). Differences were not statistically significant, although the P/R ratio for the Tremper Kill was greater than the value for the Titicus and East Br. Delaware (Scheffé MRT, both at p = 0.07). The Titicus, Muscoot (Baldwin), and Haviland Hollow had the lowest light levels and the Titicus and Muscoot (Baldwin) some of the lowest chlorophyll concentrations, resulting in low GPP and PR ratios.

GPP was negatively correlated with total P (r = -0.664, p < 0.03) and volatile suspended solids (VSS, r= -0.666, p = 0.03) and positively correlated with the uptake of NH4 (VfNH4, r = 0.787, p = 0.005). Some streams with high total P and VSS , e.g., Cross, Titicus, and Muscoot (Baldwin) had low GPP. The absolute value of CR24 also was positively correlated with the uptake of NH4 (VfNH4, r = 0.744, p = 0.01). The correlation of ammonium uptake with respiration had been - 159 - CHAPTER 8 – STREAM METABOLISM NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005 observed before, but the correlation of ammonium uptake with GPP had not. NDM (a negative number) thus correlated negatively both with VfNH4 (r = -0.669, p = 0.03) and with width (r = - 0.631, p = 0.05). That relationship was probably driven by the value for the Neversink. The P/R ratio correlated positively with glucose uptake (Vf Glucose, r = 0.673, p = 0.03) and total chlorophyll a (r = 0.655, p = 0.04), and was nearly significantly correlated with periphyton chlorophyll a (p = 0.057). GPP/PARtot correlated only with Vf arabinose (r = 0.728, p = 0.01) in this year’s data set.

Literature Cited

American Public Health Association (APHA). 1992. Standard methods for the examination of water and wastewater. American Public Health Association, Washington. Hill, W. 1996. Effects of light, pp. 121 - 148. In: R. J. Stevenson, M. L. Bothwell, R. L. Lowe, eds. Algal Ecology, Freshwater Benthic Ecosystems. Academic Press, San Diego. Hynes, H. B. N. The Ecology of Running Waters. University of Toronto Press, Toronto. Lorenzen, C. J. 1967. Determination of chlorophyll and pheo-pigments: Spectrophotometric equations. Limnology and Oceanography 12: 343-346. Marzolf, E. R., P. J. Mulholland and A. D. Steinman. 1994. Improvements to the diurnal upstream- downstream dissolved oxygen change technique for determining whole-stream metabolism in small streams. Canadian Journal of Fisheries and Aquatic Science 51: 1591-1599. Marzolf, E. R., P. J. Mulholland and A. D. Steinman. 1998. Reply: Improvements to the diurnal upstream-downstream dissolved oxygen change technique for determining whole-stream metabolism in small streams. Canadian Journal of Fisheries and Aquatic Science. 55: 1786-1787. Odum, H. T. 1956. Primary production in flowing waters. Limnology and Oceanography 1: 102 - 117. Owens, M. 1974. Measurements on non-isolated natural communities in running waters, pp. 111 - 119. In: R.A. Vollenweider (ed). A manual on methods for measuring primary production in aquatic environments. IBP Handbook No. 12, Blackwell Scientific, Oxford. Tsivoglou, H.C. and L. A. Neal. 1976. Tracer measurement of reaeration: III. Predicting the reaeration capacity of inland streams. Journal of the Water Pollution Control Federation 489: 2669-2689. Wanninkhof, R., P. J. Mulholland and J. W. Elwood. 1990. Gas exchange rates for a first-order stream determined with deliberate and natural tracers. Water Resources Research 26: 1621-1630. Young, R. G. and A. D. Huryn. 1998. Comment: Improvements to the diurnal upstream-downstream dissolved oxygen change technique for determining whole-stream metabolism in small streams. Canadian Journal of Fisheries and Aquatic Science 55: 1784-1785.

- 160 - CHAPTER 8 – STREAM METABOLISM NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Table 8.1: Relative percent difference (RPD) in Substratum, Cover types and chlorophyll a between two visits to three study sites, West Branch Delaware at South Kortright, East Branch Delaware near Arkville, and the Titicus.

W. Br. Delaware (S. Kortright) E. Br. Delaware (Arkville) Titicus River Category 27-Jul-04 12-Oct-04 RPD 14-Jul-04 5-Oct-04 RPD 25-May-04 8-Jun-04 RPD

Substratum (as %) Cobble 55 45.4 19.1 69 50.6 30.8 48.3 38.5 22.6 Boulder 18.3 16.7 9.1 18.8 22.1 16.1 19.6 13.8 34.7 Pebble 21 25.6 19.7 10.2 14 31.4 19.6 19.6 0.0 Sand 2.3 6.5 95.5 0.4 7.5 179.7 4.6 20 125.2 Wood 2.9 3.8 26.9 0.8 0.4 Silt 0.4 1.3 105.9 0.8 2.1 89.7 2.5 1.5 50.0 Bedrock 0.8 0.4 66.7 5.4 6.3 15.4 Grass 0.8 2.5

Cover type (as %) Bare 1 1.5 40.0 0.5 16 187.9 14.5 8 57.8 Black cover 0.5 4.5 21.5 34.5 46.4 Blue Green Algae 10.75 12.25 Diatoms 10 57.25 140.5 37 42.25 13.2 Exposed 0.5 1 66.7 1 0.5 66.7 5 5 0.0 Filamentous algae 17.25 0.75 2.75 114.3 3.5 0.75 129.4 Green cover 42.5 5.25 156.0 31.75 20.5 43.1 19.75 14.5 30.7 Leaf Pack 2 1.5 Moss 5.5 7.5 30.8 2 Macropytes 110.0 Silt 29.25 7.75 116.2 16.75 10 50.5 19 31.25 48.8 Tufts 0.5 15.75 5 103.6

mean Chlorophyll a (as mg/m2) Bare 8.77 12.82 Black cover 20.16 16.52 34.05 69.3 Blue Green Algae 82.34 85.1 Diatoms 85.17 126.21 38.8 67.95 42.63 50.5 Exposed Filamentous algae 254.33 156.81 200.22 50.5 40.05 Green cover 58.27 83.77 35.9 44.9 41.34 50.5 21.1 23.2 9.5 Moss 214.77 353.66 48.9 135.28 Macropytes Silt 15.35 10.24 39.9 47.23 2.4 50.5 140.39 14.39 162.8 Tufts 19.51 50.17 88.0

Reach weighted Chlorophyll a (as mg/m2) Bare 1.4 1.86 Black cover 0.91 3.55 11.75 107.2 Blue Green Algae 8.85 10.42 Diatoms 8.52 72.25 157.8 25.14 18.01 33.0 Exposed Filamentous algae 43.87 1.18 5.51 129.4 1.4 Green cover 24.77 4.4 139.7 14.26 8.47 50.9 4.17 3.36 21.5 Moss 11.81 26.52 76.8 2.71 Macropytes Silt 4.49 0.79 140.2 7.91 0.24 188.2 26.67 4.5 142.3 Tufts 3.07 2.51 20.1 Total weighted 58.44 147.83 86.7 58.91 37.25 45.0 40.72 22.12 59.2

- 161 - CHAPTER 8 – STREAM METABOLISM NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Table 8.2: Ecosystem metabolism at integrated stations in study streams determined by the 2- station (upstream-downstream) method. Reaeration determined by the propane evasion method.

-2 -1 PAR Mean g O2•m •day (mole q•m-2•day-1) temperature 0 River Date GPP CR24 NDM( C) GPP/PAR Haviland Hollow 18-May-04 1.7108 4.5105 -2.7997 3.82 17.09 0.4479 19-May-04 1.7042 3.6153 -1.9111 4.67 17.33 0.3649 20-May-04 1.9738 2.0365 -0.0627 11.35 16.80 0.1739 Mean 1.7963 3.3874 -1.5912 6.61 17.07 0.3289 SD 0.1538 1.2526 1.3963 4.12 0.27 0.1405 Titicus 8-Jun-04 1.9519 5.3230 -3.3711 4.08 18.62 0.4784 9-Jun-04 1.3300 5.9009 -4.5708 4.28 20.05 0.3107 10-Jun-04 0.4729 5.6625 -5.1897 0.96 20.37 0.4926 Mean 1.2516 5.6288 -4.3772 3.11 19.68 0.4273 SD 0.7426 0.2904 0.9246 1.86 0.93 0.1011 Cross River 15-Jun-04 2.1918 4.5041 -2.3122 21.87 19.81 0.1002 16-Jun-04 2.9084 5.2163 -2.3079 20.62 21.26 0.1410 17-Jun-04 2.0281 5.3535 -3.3254 8.15 20.79 0.2488 Mean 2.3761 5.0246 -2.6485 16.88 20.62 0.1634 SD 0.4682 0.4560 0.5862 7.59 0.74 0.0768 Muscoot (Baldwin) 22-Jun-04 0.5231 3.0021 -2.4790 2.43 18.06 0.2153 23-Jun-04 0.6874 2.9883 -2.3009 7.94 19.24 0.0866 24-Jun-04 0.7955 3.0085 -2.2130 7.27 19.19 0.1094 Mean 0.6687 2.9996 -2.3310 5.88 18.83 0.1371 SD 0.1372 0.0103 0.1355 3.01 0.67 0.0687 Muscoot (Whitehall) 29-Jun-04 1.2087 7.1101 -5.9014 21.16 14.72 0.0571 30-Jun-04 2.2613 6.8507 -4.5894 18.57 14.18 0.1218 1-Jul-04 2.4823 6.3019 -3.8196 18.39 14.47 0.1350 Mean 1.9841 6.7542 -4.7701 19.37 14.46 0.1046 SD 0.6805 0.4126 1.0526 1.55 0.27 0.0417 Neversink 10-Aug-04 5.5842 14.4616 -8.8744 24.66 15.49 0.2264 11-Aug-04 4.3737 14.2446 -9.8709 16.61 15.59 0.2633 Mean 4.9790 14.3531 -9.3727 20.64 15.54 0.2449 SD 0.8560 0.1534 0.7046 5.69 0.07 0.0261 Trout Creek 17-Aug-04 4.0479 8.7337 -4.6858 34.84 16.52 0.1162 18-Aug-04 4.2221 8.9779 -4.7558 32.83 16.94 0.1286 19-Aug-04 3.5174 9.5120 -5.9946 15.68 16.86 0.2243 Mean 3.9291 9.0745 -5.1454 27.78 16.77 0.1564 SD 0.3671 0.3980 0.7363 10.53 0.22 0.0592 Tremper Kill 31-Aug-04 2.6140 4.0899 -1.4759 19.43 18.32 0.1345 1-Sep-04 2.8953 4.0922 -1.1969 20.03 17.25 0.1445 2-Sep-04 3.1651 4.1667 -1.0016 21.79 16.78 0.1453 Mean 2.8915 4.1163 -1.2248 20.42 17.45 0.1414 SD 0.2756 0.0437 0.2384 1.23 0.79 0.0060 East Br. Delaware 5-Oct-04 0.4648 3.3518 -2.8870 18.63 10.66 0.0249 (Arkville) 6-Oct-04 0.7206 3.4447 -2.7241 18.08 9.93 0.0399 7-Oct-04 0.9071 3.7685 -2.8614 17.50 10.67 0.0518 Mean 0.6975 3.5217 -2.8242 18.07 10.42 0.0389 SD 0.2221 0.2188 0.0876 0.57 0.42 0.0135 West Br. Delaware 12-Oct-04 4.0477 8.6535 -4.6058 13.08 9.78 0.3095 (S. Kortright) 13-Oct-04 4.5409 8.7561 -4.2151 14.06 10.18 0.3230 14-Oct-04 4.8351 8.8969 -4.0619 8.67 10.31 0.5577 Mean 4.4746 8.7688 -4.2943 11.94 10.09 0.3967 SD 0.3979 0.1222 0.2805 2.87 0.28 0.1396

- 162 - CHAPTER 8 – STREAM METABOLISM NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

(as %) Whole stream GPP GPP Whole stream Water column GPP / GPP column Water

Whole Stream Whole Stream GPP (g•m-2•h-1) Water Column Water Column GPP (g•m-2•h-1) stem metabolism in study streams, Year 5, 2004 7 -0.0009 0.4342 -0.21 .07 -0.0032 0.0470 -6.81 Surface PAR Surface (mol quanta/h) (mol nn 0.82 4.47 0.0006 0.0012 0.0832 0.2696 0.72 0.45 Date of Measure Water column metabolism as a fraction of total ecosy Table 8.3:

Stream CrossEast Br. DelawareMuscoot (Baldwin) 6-OctMuscoot (Whitehall) 23-Ju 16-JunNeversink 30-Ju 3 KillTremper 2.97 Trout Creek 11-Aug DelawareWest Br. -0.0004 1-Sep 18-Aug 13-Oct 1.52 0.2763 4.9 4.18 2.78 0.002 -0.14 0.0039 0 0.6181 0.5446 0.3963 0.32 0.72 0.00 - 163 - CHAPTER 8 – STREAM METABOLISM NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Number of occurrences 4

0 0-10 10.1-20 20.1-30 30.1-40 40.1-50 50.1-60 60.1-70 70.1-80 Percent Chlorophyll a left on rock Figure 8.1: Frequency distribution of percent chlorophyll a left on rocks after being scraped, 2004.

100 Wood 90 Bedrock

e 80 Boulder Cobble 70 Pebble 60 Sand Silt 50 Clay 40

% encounters of substrat indicated 20

0 West Br. Tremper Haviland East Br. Muscoot Neversink Cross Trout Muscoot Titicus Delaware Kill Hollow Delaware (Whitehall) Creek (Baldwin)

Figure 8.2: Substrate types encountered in study reaches, 2004. Site Locations: West Branch Delaware – S. Kortright (3); Trout Creek – Trout Creek (9); East Br. Delaware – Arkville (10); Tremper Kill – Andes (15); Neversink – Claryville (29); Haviland Hollow – Haviland Hollow (34); Muscoot – Baldwin Place (46); Cross – Ward Pound (52); Titicus – Salem Center (130); Muscoot – Whitehall Corners (139).

- 164 - CHAPTER 8 – STREAM METABOLISM NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

100

Other 80

e Moss 70 Filamentous algae Tufts 60 Blue green algae 50 Silt Diatoms 40 Green 30 Black % encountered by individual typ % encountered by individual

20 Bare Exposed 10

0 W. Br. Tremper Haviland E. Br. Muscoot Neversink Cross Trout Muscoot Titicus Delaware Kill Hollow Delaware (Whitehall) Creek (Baldwin)

Figure 8.3: Cover types encountered in study reaches, 2004. Specific site locations are noted in Fig. 8.2.

150

125 Moss Filamentous algae 100 Tufts 2 Blue green algae Silt 75 Diatoms Green Black mgChlorophyll a / m 50 Bare

0 West Br. Tremper Haviland East Br. Muscoot Neversink Cross Trout Muscoot Titicus Delaware Kill Hollow Delaware (Whitehall) Creek (Baldwin)

Figure 8.4: Benthic chlorophyll a concentrations weighted by cover type, 2004. Specific site locations are noted in Fig. 8.2.

- 165 - CHAPTER 8 – STREAM METABOLISM NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

35 ) 2 30

15 Organic Weight Per Area (g/m Area Per Weight Organic

0 Muscoot Haviland Muscoot Cross West Br. Titicus Trout Neversink Tremper East Br. (Whitehall) Hollow (Baldwin) Delaware Creek Kill Delaware

Figure 8.5: Benthic organic matter concentrations (as ash-free dry mass) associated with periphyton in study reaches, 2004. Specific locations are noted in Fig. 8.2.

10.00 ) -1 .day -2 .m

2 1.00 GPP (g O

0.10 Neversink West Br. Trout Creek Tremper Kill Cross River Muscoot Haviland Titicus East Br. Muscoot Delaware (Whitehall) Hollow Delaware (Baldwin)

Figure 8.6: Gross Primary Productivity (GPP) in study streams, 2004. Note the log transformation of the y-axis. Specific sites locations are noted in Fig. 8.2.

- 166 - CHAPTER 8 – STREAM METABOLISM NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

1.00

0.10 (g O2/mol quanta) O2/mol (g tot tot GPP/PAR

0.01 Titicus West Br. Haviland Neversink Cross River Trout Creek Tremper Kill Muscoot Muscoot East Br. Delaware Hollow (Baldwin) (Whitehall) Delaware

1.00

0.10 (g O2/mol quanta) sat GPP/PAR

0.01 Titicus West Br. Haviland Neversink Trout Creek Cross River Tremper Kill Muscoot Muscoot East Br. Delaware Hollow (Baldwin) (Whitehall) Delaware

Figure 8.7: Gross Primary Productivity (GPP) per unit light (gO2/mol quanta) using field data (top panel) or field data with the substitution of 17.28 mol quanta as a saturation intensity for values greater than that value (bottom panel), 2004. Note log transformation of the y-axis. Specific site locations are noted in Fig. 8.2.

- 167 - CHAPTER 8 – STREAM METABOLISM NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

100.00 ) -1 .day -2 .m

2 10.00 (g O 24 CR

1.00 Neversink Trout Creek West Br. Muscoot Titicus Cross River Tremper Kill East Br. Haviland Muscoot Delaware (Whitehall) Delaware Hollow (Baldwin)

Figure 8.8: Community Respiration (CR24) in study streams, 2004. Note the log transformation of the y-axis. Specific site locations are noted in Fig. 8.2.

Haviland Muscoot East Br. West Br. Muscoot Tremper Kill Hollow (Baldwin) Cross River Delaware Delaware Titicus (Whitehall) Trout Creek Neversink 0

-1

-2

-3

-4 ) _1 -5 .day -2 .m 2 -6

-7 NDM (g O (g NDM -8

-9

-10

-11

-12

Figure 8.9: Net Daily Metabolism (NDM) in study streams, 2004. Specific site locations are noted in Fig. 8.2.

- 168 - CHAPTER 8 – STREAM METABOLISM NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

1.0

0.9

0.8

0.7

0.6

0.5 P/R ratio P/R 0.4

0.3

0.2

0.1

0.0 Tremper Kill Haviland West Br. Cross Trout Creek Neversink Muscoot Titicus Muscoot East Br. Hollow Delaware (Whitehall) (Baldwin) Delaware

Figure 8.10: P/R (GPP/CR24) ratios in study streams. Specific site locations are noted in Fig. 8.2.

- 169 - CHAPTER 8 – STREAM METABOLISM NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

------Intentionally Blank ------

- 170 - CHAPTER 8 – STREAM METABOLISM NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Chapter 9 - Reservoir Primary Productivity

Purpose and Significance

The goals in these studies were to (1) quantify algal biomass and primary productivity in selected reservoirs, (2) rank reservoirs according to levels of algal biomass and productivity, and (3) assess potential links between algal responses, the physical chemical characteristics of reservoirs and watershed sources of nutrients. Reservoir condition was assessed using primary productivity and chlorophyll a concentration because these variables are directly related to the amount of particulate matter in the reservoirs. At the start of Phase II (1) spatial coverage was expanded to include four new reservoirs (Amawalk, Titicus, Muscoot, and Cross), (2) studies on the Ashokan, Rondout, Schoharie, Kensico and New Croton were discontinued, (3) measurements on the Pepacton were expanded to six stations, and (4) studies on the Neversink and Cannonsville (extremes of low and high productivity, respectively, during Phase I) were continued, but with new stations on the Cannonsville. Data from both Phases I and II provide a baseline of data for biomass, productivity and related environmental variables and changes in response parameters over time can be compared to the baseline and related to changes in watershed use.

Methods

Field Procedures

On the day prior to productivity measurements, an anchored buoy was placed at each of three sampling locations (substations, subst.) on the reservoir. The location of each reservoir substation was fixed using GPS. Locations are shown in Fig. 2.3 and Fig. 2.4 and GPS coordinates are recorded in Table 2.6.

The depth of the photic zone was determined at each substation by measuring photosynthetically active radiation (PAR) at successive 0.5 m depths through the water column using a spherical underwater quantum sensor and LI-Model 1400 light meter (Li-Cor, Lincoln, NB) coupled with simultaneous measures of above-water PAR made with a quantum sensor for use in air. Measurements were made as close to mid-day as possible. Depths at which underwater PAR was 50%, 25%, 10% and 1% of above-water PAR were determined. At the same time, data were collected to determine dissolved O2 and temperature profiles from surface to the bottom, using a YSI model 5739 probe coupled with a model 58 meter (Yellow Springs Instruments, Yellow Springs, OH).

The next day reservoir primary productivity was measured by conducting light and dark bottle experiments of dissolved O2 change at multiple depths at each substation. Incubations were conducted during the 4.5 -6 h around solar noon on days when objects cast a distinct shadow. Water was collected using Van Dorn samplers from just under the water surface and at depths of 50%, 25%, 10% and 1% incident light. Water (15 - 18 L) from a given depth was pooled in an 18 L bucket and bubbled with N2 gas for approximately 2 – 4 min. to lower the dissolved O2 saturation (often >95%) to a value usually between 70 and 80%. Because there was no measurable change in pH before and after the bubbling, it was assumed that the concentration of dissolved CO2 (which is highly water-soluble) was not affected. Water from the lowest depths

- 171 - RESERVOIR PRODUCTIVITY NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

usually did not require bubbling. Three BOD bottles (2 light and 1 dark) were used at each depth. Bottles were rinsed 3 times with bucket water and then immersed in the bucket in order to cool them to ambient water temperature and avoid the formation of gas bubble on the walls of bottles. The bottles were filled through a hose located approximately 5 cm up from the bottom of the bucket with care to avoid introducing bubbles. Bottles were allowed to overflow for ~ 3-5 bottle volumes, filled, stoppered, and transferred to a holding bath of surface water in a shaded location on the boat. Water temperature, dissolved O2 concentration and percent O2 saturation were measured on each bottle using a YSI Model 58 meter and model 5905 probe with stirrer suitable for use with BOD bottles. Each bottle then was topped off with water from the bucket (0.5 – 1 ml at most) and replaced in the holding bath. The process was repeated with water collected from each depth. After preparing the bottles from all depths at a substation, they were placed horizontally in Plexiglas holders (Wetzel and Likens 1991) suspended in the reservoir at the depth from which the water had been collected. The entire process was repeated at each substation. After incubation, the bottles were retrieved and dissolved O2 concentration and saturation and water temperature were measured again. During each incubation period PAR was measured from the boat deck.

A 2 L sample of water from each depth was collected for chlorophyll a analysis. Those samples were immediately placed on ice, filtered onto GF/F filters within 24 h, and filters were stored frozen until extraction. Field blanks (2 L of nanopure water) were filtered through the filtering apparatus. The filter was treated as a chlorophyll sample.

Water samples for nutrient determinations were taken from the bucket of surface water and filtered through precombusted Gelman GF/F filters (250 ml for inorganic analyses and 40 ml for dissolved organic carbon, DOC). Inorganic samples were placed on ice until they could be frozen in the laboratory; DOC samples were iced and later refrigerated. Other samples of bucket water were collected directly into 125 ml bottles (leaving no head space), placed on ice and refrigerated for total alkalinity determinations.

BOD bottles were treated to prevent microbial wall growth between each reservoir study. Bottles were filled with a 30% bleach solution for 15 min., rinsed with copious amounts of water and allowed to air dry.

The PAR sensors were re-calibrated by the manufacturer prior to the start of the field season. They carry a stated accuracy of ± 3-5% traceable to the U. S. National Institute of Standards and Technology with calibration guaranteed for 2 years. The balances used for weighing were recalibrated prior to the field season

Laboratory analyses

Chlorophyll was analyzed fluorometrically (after Arar and Collins 1997). The frozen filters were snipped and macerated for 30 sec. in 9 ml of a 90% acetone/10% saturated MgCO3 solution (1 g MgCO3/100 ml H2O; APHA 1997) at 4°C. The samples were returned to the freezer for 16 - 24 h in darkness for extraction of chlorophyll. After extraction, the filters were compressed using a Teflon pestle and the supernatant fluid was transferred to a centrifuge tube. The samples were centrifuged at 8,000 x g for 20 min. at 4°C. The supernatant fluid was then transferred to a test

- 172 - RESERVOIR PRODUCTIVITY NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

tube in an ice bath and covered with aluminum foil; the pellet along with the filter was held in the freezer for re-extraction if necessary. All manipulations were performed in subdued light to avoid photobleaching of pigments. An aliquot of the supernatant fluid (4 ml) was analyzed fluorometrically by making an appropriate dilution (between 1:2 and 1:32) in a 9 ml cuvette and measuring fluorescence intensity before and after acidification in a Turner Model 10-AU fluorometer. This year, there were problems on some reservoirs (Pepacton, Titicus and Cross) with fluorescence readings having higher values after acidification. Thus, the chlorophyll concentrations for all reservoirs were computed based on initial fluorescence values. Note, however, that data for most reservoirs were close to values for last year, suggesting that the pheophytin contribution is small. Samples were extracted a second and sometimes a third time to insure complete extraction of each sample. Throughout the field season laboratory control standards were prepared to generate ~ 40 µg of chlorophyll a (Sigma, St. Louis, MO) in a sample. The initial concentration of each bottle was determined and values for individual days were compared either to that concentration or to the mean of all samples from that bottle. A solid standard calibrated against a fluorometrically-determined chlorophyll standard was used for fluorometric assay on a daily basis.

Water chemistry determinations were performed according to procedures documented in Chapters 3 and 6.

Data analysis

Oxygen changes in the light bottles at each depth were converted to an estimate of Gross Primary Productivity (GPP) by adding the O2 change in the dark bottle to the net O2 change in each light bottle for that depth. Data were integrated over the depth of the photic zone to generate an area- specific productivity. Similarly, chlorophyll concentrations determined volumetrically were integrated similarly to generate a value per-unit-surface-area.

Statistical procedures were performed using log-transformed data. Differences between sites were determined using ANOVA and Scheffe Multiple Range tests (p = 0.05).

QA/QC Results

Pertinent QA/QC data are presented in Appendices 9.1 through 9.5.

Chlorophyll a

Chlorophyll samples for the Amawalk, Cross and Muscoot reservoirs were analyzed within 21 days, for the Neversink and Cannonsville within 22 days, the Titicus within 24 days, and the Pepacton within either 24 or 26 days.

Mean chlorophyll a concentrations from fluorometric analyses of the 121 first extractions and 106 second extractions were 2,510-fold and 1,005-fold greater, respectively, than the mean chlorophyll concentrations in 31 lab blanks. The mean fluorescence emission of the lab blanks was 0.51, while the mean for the second extracts of samples was 183 and of 15 third extracts was 88.5. Ninety-five percent of blanks had emissions <1.3, whereas 95% of the second extracts had

- 173 - RESERVOIR PRODUCTIVITY NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

emissions >57.3, a 43-fold difference. The blanks assure that between-sample contamination was negligible and (along with lab control standards) that the fluorometer was operating correctly. Field blanks, which serve as a check for cross-contamination at the filtering step, had a mean chlorophyll concentration of 3.5% of the mean concentration in samples. Further examination of data for field blanks indicated that the mean for EOH reservoirs was nearly 4-fold greater that the mean for WOH reservoirs (0.76 µg/sample vs. 0.20 µg/sample). Thus, while data are not seriously compromised by cross-contamination between samples, greater attention to cleaning the filtration apparatus between samples, especially for EOH reservoirs, is warranted.

Laboratory control standards were analyzed with each set of extractions. This year the seasonal mean concentration of the primary standard was 177 ± 6.73 µg/L. The average relative percent difference (RPD) of 16 daily values from the seasonal mean was 2.54 ± 2.68 % (mean ± SD). The stability of the fluorometer was checked before and after each use using solid standards calibrated to the chlorophyll standard. The RPDs of daily determinations from the seasonal means for the high and low range solid secondary standards averaged 0.73% and 0.61%, respectively (n = 31 determinations each).

The RPD between field duplicate chlorophyll samples analyzed fluorometrically averaged 20.94 ± 19.20 (n=16 paired samples), with a median RPD of approximately 13.48. Five RPD determinations exceeded 20%, and 4 of those exceeded 30%.

Dissolved oxygen

All the changes in dissolved O2 are used in computing metabolism data for the reservoirs. However, the manufacturer’s specification of an accuracy of ± 0.1 mg/L for the Model 58 meter/probe combination provides one means of control of data quality. The substations and depths on each reservoir where changes in dissolved oxygen were equal to or exceeded 0.2 mg/L are noted in the data tables.

Duplicate incubations were conducted with separate water samples collected either from the surface or the 50% depth at two substations on each reservoir. GPP and respiration were computed as oxygen change m-3.h-1 (equivalent to mg.L-1.h-1). The RPD in GPP/h was < 20% in 11 of the 16 determinations and between 20% and 29% in another 3 instances. However, the RPDs were > 50% for one pair of incubations in the Neversink and Pepacton. This occurred at the same substation on the Pepacton (no. 1) last year. If those two comparisons are excluded the RPD for GPP/h for the remaining 14 comparisons averaged 11.74%. The RPD in respiration between duplicates was high for one subst. in the Muscoot (46.5%), at subst. 1, 3, and 5 on the Pepacton (63.9 ± 30.3 %, x ± SD, n=3) and at subst. 3 and 5 on the Cannonsville (200%, where the respiration rates were especially low). However, the average RPD for CR/h for the remaining 10 comparisons averaged 11.8 ± 7.2 % (mean ± SD, n=10).

The grand mean GPP/PAR of surface water incubations across reservoirs had a CV of 18.2% in this years work. Only two reservoirs had a lower within-reservoir CV for that parameter, viz., the Cross (13.1%) and Amawalk (1.2%), although the Pepacton and Titicus had CVs of 27%. The CVs for the Cannonsville, Muscoot and Neversink were higher, ranging between 32 and 55% (Table 9.1). None of these CVs were as high as those generated last year, presumably because

- 174 - RESERVOIR PRODUCTIVITY NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

lighting was more consistent between reservoir studies this year. Within-reservoir gradients in productivity and chlorophyll have been demonstrated in these systems and study sites were selected to maximize the detection of such variability and potential localized influences. However, the presence of the high variability within the Neversink was surprising.

Nutrient Chemistry

Concentrations of NH4-N, NO3-N, SRP and TDP, and total alkalinity were determined for each reservoir. Blanks never exceeded twice the detection limit. Laboratory duplicates were always within the 20% limit for quality control. The mean % recovery of matrix spikes was within the range of 98.7% to 100.3% for all analytes. Quality control data for DOC are presented in Chapter 7. Field duplicates never exceeded the 20% QC standard.

Results and Discussion

Physical – Chemical Profiles

Light, temperature, and dissolved O2 profiles for each reservoir are shown in Figs. 9.1 - 9.8. The depth of the photic zone in the EOH reservoirs ranged from minimum depths of 4 or 4.75 m (Titicus subst. 1, 2, Muscoot subst. 2) to maximum depths of 7.25 m (Cross subst. 3) and 7.5 m (Amawalk, subst. 3). In WOH reservoirs, the photic zone ranged from 8 m (Neversink, all subst.) to 10.75 m (Cannonsville, subst. 4 and 5) deep. A light attenuation coefficient (Kd) was computed for each reservoir substation and a reservoir average determined (Table 9.2). The Kd values for all EOH reservoirs were greater than for all WOH reservoirs. Surprisingly, the Cannonsville had the lowest average Kd value of all the reservoirs, and the Neversink the highest Kd of the WOH reservoirs.

The WOH reservoirs exhibited gradual temperature changes and broad, deep thermoclines. In contrast, the thermocline in each EOH reservoir began between 2 (Muscoot) and 5 m (Cross and Titicus) and was well-developed in all but the Muscoot, with the hypolimnion beginning between 7 and 10 m.

Oxygen saturation profiles in all of the EOH reservoirs were demonstrably clinograde, with saturation values and dissolved O2 concentrations approaching zero in the hypolimnion at more than one substation (Lampert and Sommer 1997). In the Cannonsville, oxygen concentrations approached 1 mg/L and 10% saturation at subst. 5 and 2 mg/L (20% saturation) at the other substations. Concentrations in the Pepacton approached 1 mg/L at subst. 1 (near the mouth of the East Br. Delaware river) but profiles at the other substations and in the Neversink were orthograde, with higher hypolimnetic dissolved O2 concentrations and saturation values. As noted in other years, it is possible that organic matter settled to the thermocline in the Pepacton reservoir where it underwent decomposition as reflected by lower dissolved O2 concentrations at that depth.

Field notes indicated the presence of visible amounts of fine suspended particles at all substations in all EOH reservoirs. Floating particles were also noted at Amawalk subst. 1 and Titicus subst. 1 and 2. Both floating and suspended particles were especially abundant at

- 175 - RESERVOIR PRODUCTIVITY NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Muscoot subst. 1. At Muscoot subst. 2 the density of both types was lower, and the abundance at subst. 3 was between that subst. 1 and 2. Fine suspended material was present at Cannonsville subst. 3 and present to a lesser extent at subst. 4 and 5 (both located in the Trout Creek arm of the reservoir). Very fine suspended particles were noted at all substations in the Pepacton. They were most numerous at subst. 1 at the mouth of the East Br. Delaware River, and densities were lowest at subst. 7. Water clarity was good in the Neversink.

Algal Biomass and Primary Productivity

Chlorophyll a.- Reservoirs were sampled between July 8 (Amawalk) and October 1(Cross). Reservoir average photic zone chlorophyll concentrations ranged from 16.7 mg/m2 in the Neversink to 81.3 mg/m2 in the Titicus and reservoirs ranked according to the mean chlorophyll concentration as shown in Fig. 9.9. The mean concentration in the Neversink was significantly lower than in all of the other reservoirs and the concentration in the Pepacton was significantly lower than in the Titicus (ANOVA and Scheffe MRT on log transformed data, p ≤ 0.05).

Some within-reservoir gradients in chlorophyll concentration were noted. Concentrations decreased from subst. 1 to subst. 3 in both the Titicus and Cross, just as they did in 2003. In the Cannonsville, the highest concentration occurred at subst. 3 in the main arm of the reservoir, but in the Trout Creek arm the concentration was greater at subst. 5 (closer to Trout Creek) than subst. 4 suggesting a tributary influence. The chlorophyll concentration at Muscoot subst. 1 was substantially higher then at subst. 2 and 3, which were similar to each other. This pattern also was observed in 2003 and it is in keeping with field observations of suspended particle densities. Concentrations tended to increase through the Pepacton, and were from 21% to 33% greater at substs. 3, 6 and 7 than at subst. 1. The concentration at subst. 4, located at the mouth of the Tremper Kill, was less than at subst. 1. As in other years, chlorophyll concentrations were low at all stations in the Neversink.

Chlorophyll concentration was positively correlated with GPP/m2 (r = 0.744), GPP/PAR (r = 0.615), both significant at p < 0.001, and also with SRP (r = 0.434, p= 0.03), DOC (r = 0.614, p = 0.001) and total alkalinity (r = 0.765, p < 0.001), and negatively with NO3 (r = -542, p = 0.005); all correlations done with log10 transformed variables). The average nitrate concentration was greater in WOH reservoirs, whereas SRP and DOC were greater in EOH reservoirs, which is where greater chlorophyll concentrations usually occurred.

Primary productivity.- Reservoir mean Gross Primary Productivity (GPP) ranged from 0.199 g . -1. -1 . -1. -1 O2 m h in the Neversink to 0.710 O2 m h in the Titicus. Greatest productivity occurred in two of the EOH reservoirs, the Titicus and Amawalk, which were followed by the Cannonsville (Fig. 9.11). GPP per area in the Neversink was significantly lower than in all other reservoirs but the Muscoot (ANOVA and Scheffe MRT on log transformed data, p ≤ 0.05), but even that difference was nearly significant (p = 0.055). GPP per area was correlated positively not only with chlorophyll but also with SRP (r = 0.419, p= 0.05), DOC (r = 0.483, p = 0.05) and total alkalinity (r = 0.702, p < 0.001), and negatively with NO3 (r = -562, p = 0.01; all correlations done with log10 transformed variables).

- 176 - RESERVOIR PRODUCTIVITY NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

When GPP was normalized for PAR the ordering of the reservoirs remained quite similar to the ordering according to GPP per area, except that all substations on the Pepacton grouped together and were higher than the Cannonsville (Fig. 9.12). The Neversink ranked lowest in GPP/PAR and activity there was significantly lower than in the Titicus, Amawalk, Pepacton and Cannonsville (ANOVA and Scheffe MRT on log transformed data, p ≤ 0.05).

Primary productivity data for each reservoir are summarized in Table 9.3 along with chlorophyll concentrations and pertinent environmental data. Data are presented for each substation, along with the mean for each reservoir, and the reservoirs are ordered according to GPP/mol quanta PAR. The depths at which changes in dissolved O2 in the light bottles or in both light and dark bottles exceeded the error bounds of the probe are indicated in the table. This was the situation for most substations and depths on all reservoirs but the Neversink.

GPP/PAR in the Titicus was higher at subst. 1 and 3 than at subst. 2. In both the Amawalk and the Cross, there was GPP/PAR increased going from subst. 1 to subst. 3, presumably a response to greater water clarity and a deeper photic zone approaching subst. 3. In contrast to the pattern of increasing chlorophyll through the Pepacton, GPP/PAR showed no striking increase. In the Cannonsville, GPP/PAR, as for chlorophyll, was highest at subst. 3 but the value for subst. 5 was higher than for subst. 4. This pattern was observed last year, and while it is in contrast to that for chlorophyll a, it suggests the possibility of tributary influences of the West Br. Delaware (at subst. 3), and perhaps Trout Creek (at subst. 5). Although the chlorophyll concentration was highest at subst. 1 in the Muscoot, GPP/PAR was roughly a third lower there than at the other substations. GPP/PAR was lowest in the Neversink.

Concentrations of NO3 were highest in the Neversink, followed by the Cannonsville and Muscoot (both about 50% lower), with much lower concentrations elsewhere (Table 9.3). Within the Muscoot there was an increasing gradient of NO3 from subst. 1 to 3. NH4 concentrations were highest in the Neversink, Muscoot, and Amawalk reservoirs but patterns within reservoirs were not easily related to patterns in productivity or chlorophyll concentrations. TDP and SRP concentrations were low in all reservoirs with few distinct gradients. DOC concentrations were nearly 2-fold higher in the Amawalk, Titicus, Cross and Muscoot than in the WOH reservoirs.

GPP/PAR was positively correlated not only with chlorophyll a as noted above, but also with SRP (r = 0.473, p = 0.02), and total alkalinity (r = 0.518, p = 0.01) and negatively with nitrate (r = -0.664, p = <0.001) (all correlations done with log10 transformed variables). Thus primary productivity normalized for light was related to algal biomass, and was greater in EOH reservoirs where total alkalinity was greater and nitrate lower. The correlation of GPP/PAR with temperature was negative and nearly significant (r = -0.403, p = 0.0503) as some of the reservoirs with highest temperatures had the lowest productivity and reservoirs with highest productivity had temperatures at the lower end of the range.

- 177 - RESERVOIR PRODUCTIVITY NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Literature Cited

APHA. 1997. Standard methods for the evaluation of water and wastewater, 20th ed. American Public Health Association, American Water Works Association, Water Environment Federation, Washington, DC. Arar, E. J. and G. B. Collins. 1997. In vitro determination of chlorophyll a and pheophytin a in marine and freshwater algae by fluorescence. Method 445.0-1. U. S. Environmental Protection Agency, Cincinnati, OH. Wetzel, R. G. and G. E. Likens. 1991. Limnological Analyses, 2nd ed. Springer, New York. 391 pp.

- 178 - RESERVOIR PRODUCTIVITY NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Table 9.1: Comparison of surface water incubations within each reservoir and between reservoirs. Volumetric Reservoir Data GPP/h Surface PAR Volumetric Date Location Quality (g.m-3.h-1) Hr. ave. GPP/PAR Muscoot Reservoir 1 a 0.0463 5.21 0.0089 21-Jul-04 2 c 0.1072 4.56 0.0235 3 0.0594 4.26 0.0139 Mean 0.0710 4.68 0.0154 St. dev. 0.0321 0.49 0.0074 CV 45.21 10.38 48.21 Cannonsville Reservoir 3 b 0.0657 3.72 0.0177 26-Aug-04 4 b 0.0370 3.96 0.0093 5 b 0.0495 4.12 0.0120 Mean 0.0507 3.94 0.0130 St. dev. 0.0144 0.20 0.0042 CV 28.35 5.15 32.64 Cross River Reservoir 1 0.0433 4.73 0.0092 1-Oct-04 2 0.0359 4.22 0.0085 3 0.0394 3.62 0.0109 Mean 0.0395 4.19 0.0095 St. dev. 0.0037 0.56 0.0012 CV 9.41 13.27 13.06 Titicus Reservoir 1 b 0.1496 4.84 0.0309 22-Sep-04 2 b 0.0938 4.79 0.0196 3 b 0.0829 4.12 0.0201 Mean 0.1088 4.58 0.0235 St. dev. 0.0358 0.40 0.0064 CV 32.90 8.82 27.09 Pepacton Reservoir 1 0.0419 3.05 0.0138 15-16-Sep-04 3 0.0353 2.97 0.0119 4 0.0322 3.20 0.0101 5 0.0254 3.16 0.0080 6 0.0234 3.25 0.0072 7 0.0455 3.14 0.0145 Mean 0.0339 3.13 0.0109 St. dev. 0.0088 0.10 0.0030 CV 25.96 3.22 27.47 Amawalk Reservoir 1 c 0.0914 4.13 0.0221 8-Jul-04 2 b 0.1002 4.49 0.0223 3 b 0.0789 3.62 0.0218 Mean 0.0902 4.08 0.0221 St. dev. 0.0107 0.43 0.0003 CV 11.88 10.63 1.26 Neversink Reservoir 1 0.0134 4.46 0.0030 4-Aug-04 2 0.0185 4.65 0.0040 3 0.0335 4.02 0.0083 Mean 0.0218 4.38 0.0051 St. dev. 0.0104 0.32 0.0028 CV 47.87 7.36 55.49 Grand Mean 0.0594 4.14 0.0142 St. Dev. 0.0123 0.16 0.0026 CV 20.78 3.92 18.23 a: change ≥ 0.2 mg/L for light bottle b: change ≥ 0.2 mg/L for dark bottle c: change ≥ 0.2 mg/L for both light and dark bottles

- 179 - RESERVOIR PRODUCTIVITY NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Table 9.2: Light attenuation coefficients for each reservoir, Year 5, 2004.

Reservoir

Reservoir Substation Kd Mean SD

Titicus 1 1.11 0.993 0.107 2 0.967 3 0.901

Muscoot 1 0.863 0.865 0.015 2 0.881 3 0.852

Amawalk 1 0.776 0.703 0.086 2 0.724 3 0.608

Cross 1 0.702 0.658 0.039 2 0.642 3 0.629

Neversink 1 0.571 0.570 0.004 2 0.574 3 0.566

Pepacton 1 0.444 0.470 0.020 3 0.464 4 0.498 5 0.461 6 0.462 7 0.488

Cannonsville 3 0.485 0.445 0.035 4 0.426 5 0.424

- 180 - RESERVOIR PRODUCTIVITY NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

/L) 3 Total mean and standard mean and standard -P DOC Alkalinity 4 oir location with the oir location -N TDP o-PO 3 -N NH one, GPP normalized for incident PAR, one, GPP normalized for incident PAR, 3 (mg/L) (mg/L) (mg/L) (mg/L) (mg/L) CaCO (mg NO a ) 2 (mg/m Chlorophyll Chlorophyll PAR ) -1 h -2 m ) -1 h -2 data. Data shown are those for each reserv for data. Data shown are those m 2 (GPP) per unit area over the photic z 0.2 mg/L zone Photic PAR Surface GPP/ oirs are ordered using GPP/PAR. ≥ Data change Depth of of Depth 13 8 8 26.4-20.9 0.183 25.6-20.1 0.272 4.430 3.930 0.041 0.069 13.92 17.16 0.174 0.171 0.016 0.010 0.004 0.003 0.001 0.001 2.134 2.079 2.608 2.776 13 5 5.25 4 3 28.2-20.0 28.5-20.5 0.338 0.394 5.170 4.200 0.066 0.094 82.99 31.36 0.036 0.133 0.024 0.022 0.004 0.003 0.002 0.001 3.770 4.107 76.631 75.162 13 6.5 7.25 2,3,4 2,3 21.2-18.8 0.409 21.7-18.1 0.407 4.690 3.550 0.087 0.114 57.96 51.42 0.014 0.013 0.009 0.003 0.007 0.003 0.000 4.357 0.001 4.138 45.238 43.942 35 9.5 10.75 1,2,3,4 1,2,3 24.1-19.6 22.8-19.8 0.476 0.507 3.720 4.100 0.128 0.123 68.14 50.80 0.060 0.097 0.005 0.008 0.003 0.002 0.001 0.001 2.136 2.133 18.482 17.800 15 10.2567 10 2,3 10 9.5 22.1-19.5 2,3 0.425 3 22.7-18.6 2,3 2.970 0.449 22.6-19.8 22.2-20.1 0.143 0.312 0.475 3.180 36.40 3.280 0.141 3.120 0.095 0.002 0.152 30.52 0.007 48.66 0.005 44.21 0.000 0.008 0.003 0.002 0.003 0.000 0.006 0.005 2.079 0.003 0.000 0.000 13.997 2.031 0.003 0.003 13.435 2.026 2.151 13.248 13 6 7.5 1,2,3,4 1,2,3,4 28.0-20.5 27.0-16.5 0.326 0.750 4.130 3.540 0.079 0.211 76.80 78.30 0.005 0.006 0.011 0.024 0.004 0.005 0.003 0.002 5.596 4.651 35.934 33.826 13 4 5 1,2,3,4 22.8-19.9 1,2,3,4 0.800 23.5-20.6 0.716 4.830 4.050 0.166 0.177 96.51 64.98 0.002 0.002 0.005 0.004 0.005 0.004 0.003 0.003 4.558 4.422 65.337 65.901 Mean 8.00 0.199 4.320 0.047 16.657 0.172 0.012 0.003 0.001 2.092 2.692 Mean 5.00 0.391 4.637 0.086 49.017 0.081 0.018 0.004 0.002 3.903 75.897 Mean 6.83 0.409 4.133 0.100 54.627 0.012 0.008 0.001 0.002 4.173 44.590 Mean 10.33 0.457 3.907 0.117 56.067 0.087 0.007 0.003 0.001 2.124 18.141 Mean 9.83 0.401 3.123 0.129 40.223 0.003 0.006 0.000 0.003 2.079 13.520 Mean 6.58 0.547 4.040 0.139 76.723 0.005 0.014 0.005 0.002 4.971 34.880 Mean 4.58 0.710 4.537 0.158 81.273 0.002 0.006 0.004 0.003 4.555 65.619 St. Dev.St. 0.00 0.066 0.348 0.020 2.523 0.002 0.003 0.001 0.000 0.038 0.119 St. Dev.St. 0.25 0.052 0.492 0.017 29.429 0.049 0.008 0.001 0.001 0.179 1.038 St. Dev.St. 0.38 0.002 0.570 0.014 3.272 0.003 0.004 0.002 0.001 0.170 0.916 St. Dev.St. 0.72 0.062 0.190 0.015 10.484 0.024 0.002 0.001 0.000 0.019 0.482 St. Dev.St. 0.38 0.061 0.130 0.022 7.498 0.001 0.001 0.000 0.000 0.048 0.329 St. Dev.St. 0.80 0.213 0.462 0.067 1.616 0.001 0.009 0.001 0.001 0.541 1.491 St. Dev.St. 0.52 0.094 0.424 0.025 15.792 0.000 0.003 0.001 0.000 0.132 0.399 and nutrients for each reservoir, Year 5 Year and nutrients for each reservoir, a Summary of gross primary productivity Date Station (m) depths: at (°C) O (g Titicus 8-Jul-04 2 6.25 1,2,3,4 27.8-21.1 0.566 4.450 0.127 75.07 0.004 0.008 0.005 0.002 4.667 Reservoir zone photic or both LB for range temp. GPP quanta (mol quanta mol 1-Oct-04 2 6.75 2,3 21.1-18.6 0.411 4.160 0.099 54.50 0.012 0.005 0.001 0.002 4.023 Muscoot 21-Jul-04 2 4.75 1,2,3,4 29.3-21.1 0.441 4.540 0.097 32.70 0.075 0.009 0.004 0.002 3.832 4-Aug-04 2 8 25.8-20.5 0.142 4.600 0.031 18.89 0.171 0.010 0.003 0.001 2.062 Pepacton Amawalk 15-Sep-0416-Sep-04 4 3 9.25 10 2 2,3,4 22.0-19.5 22.7-20.1 0.354 0.392 3.220 2.970 0.110 0.132 34.27 47.28 0.002 0.004 0.006 0.005 0.001 0.000 0.003 0.003 2.112 2.072 13.399 22-Sep-04 2 4.75 1,2,3,4 22.5-20.7 0.613 4.730 0.130 82.33 0.002 0.010 0.004 0.003 4.685 Neversink 26-Aug-04 4 10.75 1,2,3 22.8-19.1 0.387 3.900 0.099 49.26 0.105 0.009 0.003 0.001 2.102 Cross River Cannonsville Table 9.3: deviation for the reservoir. Reserv the reservoir. deviation for chlorophyll

- 181 - RESERVOIR PRODUCTIVITY NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Amawalk Reservoir 7 July 04

0 0

-1 -2

-2 -4

-3 -6

-4 -8

-5 -10

-6 -12 Depth (m)

-7 -14

-8 -16

-9 -18 1 - 1266 1 1 1 -10 2 - 1268 -20 2 2 2 3 - 1350 3 3 3 -11 -22 0 20406080100120 5 1015202530 0 2 4 6 8 10 12 14 16 0 20 40 60 80 100 120 Temperature (C) DO (mg/L) % DO saturation Percent of above-water PAR Figure 9.1: Profiles of light, temperature, and dissolved oxygen in Amawalk Reservoir, 7 July 2004.

Muscoot Reservoir 20 July 04 0 0

-1 -2

-2 -4

-3 -6

-4 -8

-5 -10

-6 -12 Depth (m)

-7 -14

-8 -16

-9 -18 1 - 1498 1 1 1 2 - 1674 2 -10 -20 2 2 3 - 1940 3 3 3 -11 -22 0 20 40 60 80 100 120 5 10152025300246810121416 0 25 50 75 100 125 150 175

Percent of above-water PAR Temperature (C) DO (mg/L) % DO saturation

Figure 9.2: Profiles of light, temperature, and dissolved oxygen in Muscoot Reservoir, 20 July 2004.

- 182 - RESERVOIR PRODUCTIVITY NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Neversink Reservoir 3 August 04

0 0

-1 -2

-2 -4

-3 -6

-4 -8

-5 -10

-6 -12 Depth (m)

-7 -14

-8 -16

-9 1 - 1555 -18 1 1 1 2 -1736 2 2 -10 -20 2 3 - 1814 3 3 3 -11 -22 0 20 40 60 80 100 120 5 10152025300 2 4 6 8 10121416 0 20406080100

Percent of above-water PAR Temperature (C) DO (mg/L) % DO saturation

Figure 9.3: Profiles of light, temperature, and dissolved oxygen in Neversink Reservoir, 3 August 2004.

Cannonsville Reservoir 24 August 04

0 0

3 -1 -2 4 -4 -2 5 -6 -3 -8 -4 -10

-5 -12

-6 -14 Depth (m) -7 -16

-18 -8 -20 -9 -22 -10 3 - 1626 -24 3 3 4 - 1309 4 4 -11 -26 5 - 1718 5 5 -12 -28 0 20 40 60 80 100 120 5 10152025300 2 4 6 8 10 12 14 16 0 20 40 60 80 100 120 Percent of above-water PAR Temperature (C) DO (mg/L) % DO saturation

Figure 9.4: Profiles of light, temperature, and dissolved oxygen in Cannonsville Reservoir, 24 August 2004.

- 183 - RESERVOIR PRODUCTIVITY NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Pepacton Reservoir, Stations 1, 4, 5 14 September 04

0 0

-1 -2

-2 -4

-3 -6

-4 -8

-5 -10

-6 -12 Depth (m)

-7 -14

-8 -16

-9 -18 1 1 1 - 597 1 4 4 -10 4 - 625 -20 4 5 - 906 5 5 5 -11 -22 0 20 40 60 80 100 120 5 10152025300 2 4 6 8 10 12 14 16 0 20406080100 Percent of above-water PAR Temperature (C) DO (mg/L) % DO saturation

Figure 9.5: Profiles of light, temperature, and dissolved oxygen in Pepacton Reservoir, Substations1,4,5 14 September 2004. Pepacton Reservoir, Stations 3, 6, 7 14September 04

0 0

-1 -2

-2 -4

-3 -6

-4 -8

-5 -10

Depth (m) -6 -12

-7 -14

-8 -16

-9 -18 3 - 1375 3 3 3 6 - 1639 6 -10 -20 6 6 7 - 524 7 7 7 -11 -22 0 20 40 60 80 100 120 5 10152025300246810121416 0 20406080100

Temperature (C) DO (mg/L) % DO saturation Percent of above-water PAR

Figure 9.6: Profiles of light, temperature, and dissolved oxygen in Pepacton Reservoir, Substations 3,6,7 14 September 2004.

- 184 - RESERVOIR PRODUCTIVITY NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Cross River Reservoir 29 September 04

0 0

-1 -2

-2 -4

-3 -6

-4 -8

-5 -10

Depth (m) -6 -12

-7 -14

-8 -16

1 - 1658 -9 -18 1 1 1 2 - 2142 2 2 2 -10 -20 3 - 1367 3 3 3

-11 -22 0 20 40 60 80 100 120 5 10152025300 2 4 6 8 10 12 14 16 0 20 40 60 80 100 Percent of above-water PAR Temperature (C) DO (mg/L) % DO saturation

Figure 9.7: Profiles of light, temperature, and dissolved oxygen in Cross River Reservoir, 29 September 2004.

Titicus Reservoir 21 October 04

0.0 0

-1.0 -2

-2.0 -4

-3.0 -6

-4.0 -8

-5.0 -10

Depth (m) -6.0 -12

-7.0 -14

-8.0 -16

-9.0 -18 1 - 986 1 1 1 -10.0 2 - 1244 -20 2 2 2 3 - 1386 3 3 3 -11.0 -22 0 20 40 60 80 100 120 5 10152025300 2 4 6 8 10 12 14 16 0 20406080100 Temperature (C) DO (mg/L) % DO saturation Percent of above-water PAR

Figure 9.8: Profiles of light, temperature, and dissolved oxygen in Titicus Reservoir, 21 October 2004.

- 185 - RESERVOIR PRODUCTIVITY NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

100

70 2 60

40 mg Chlorophyll a/m Chlorophyll mg

0 Titicus Amawalk Cannonsville Cross Muscoot Pepacton Neversink Reservoir

Figure 9.9: Reservoir chlorophyll a concentrations (mean ± SD), 2004 field season.

100

70 2 /m

a 60

40 mg Chlorophyll mg Chlorophyll 30

0 Cross 1 Cross 2 Cross 3 Titicus 1 Titicus 2 Titicus Titicus 3 Titicus Muscoot 2 Muscoot 3 Muscoot Muscoot 1 Muscoot Amawalk 3 Amawalk 1 Amawalk 2 Amawalk Pepacton 6 Pepacton 3 Pepacton 7 Pepacton 1 Pepacton 4 Pepacton 5 Pepacton Neversink 2 Neversink 3 Neversink 1 Neversink Cannonsville 5 Cannonsville 4 Cannonsville 3 Reservoir and Substation

Figure 9.10: Chlorophyll a concentrations of each reservoir substation, 2004 field season.

- 186 - RESERVOIR PRODUCTIVITY NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

0.9

0.8

0.7

0.6

0.5

0.4

0.3 GPP (g O2·m-2·h-1) (g GPP 0.2

0.1

0 Titicus Muscoot Amawalk Neversink Cross River Cannonsville Pepacton 1,4,5 Pepacton 3,6,7 Reservoir

Figure 9.11: Gross primary productivity in each reservoir, 2004 field season. Data shown are mean ± SD, n=3. The incubations on the Pepacton were conducted on different days and are shown separately.

0.25

0.20

0.15 /mol quanta) 2

0.10

0.05 GPP/PAR (g O

0.00 Titicus Muscoot Amawalk Neversink Cross River Cross Cannonsville Pepacton 1,4,5Pepacton 3,6,7Pepacton Reservoir

Figure 9.12: GPP normalized for PAR during the incubation period, 2004 field season.

- 187 - RESERVOIR PRODUCTIVITY NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

------Intentionally Blank ------

- 188 - RESERVOIR PRODUCTIVITY NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

APPENDICES to Year 5 New York Watersheds Report

Stroud Water Research Center Avondale, PA 19311

Note: No Appendices to Chapters 1 or 2

- 189 - APPENDICIES NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

------Intentionally Blank ------

- 190 - APPENDICIES NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Appendix to Chapter 3 - Nutrients, Major Ions and Suspended Particles in Transport

______

2004 QUALITY ASSURANCE/QUALITY CONTROL DATA FOR MAJOR IONS/NUTRIENTS DATA SUMMARIES ------

METHOD DETECTION LIMITS (MDL), QUALITY CONTROL CRITERIA (AS EXCEEDANCE LIMITS): (Units = mg/L; Alkalinity (ALKL) reported as ueq/L CaCO3)

CMPD MDL (2000-2004) Precision (RPD specific) Accuracy (%) PH - 20 75-125 ALKL (ueq/L) - 20 75-125 COND (uS/cm) - 20 75-125 CA (mg/L) 0.1 20 75-125 MG (mg/L) 0.1 20 75-125 NA (mg/L) 1.5 20 75-125 K (mg/L) 0.4 20 75-125 CL (mg/L) 1 20 75-125 SO4 (mg/L) 2 20 75-125 NH4N (mg/L) 0.011 20 75-125 NO3N (mg/L) 0.02 20 75-125 SKN (mg/L) 0.1 20 75-125 TKN (mg/L) 0.1 20 75-125 SRP (mg/L) 0.003 20 75-125 TDP (mg/L) 0.01 20 75-125 TP (mg/L) 0.01 20 75-125

ADDITIONAL DEFINITIONS AND NOTES:

______Appendix source: \\StroudSAS\research\nywatershed\nywatershed2004\spiraling2004\NUTR_MAJORIONS_04_QAQC_APPENDIX_C.LST Produced on 07APR05:09:11

- 191 - APPENDIX TO CHAPTER 3 (A.3) NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

______

2004 QUALITY ASSURANCE/QUALITY CONTROL DATA FOR MAJOR IONS/NUTRIENTS Summary data for: Field (FB) & Lab (LB) Blanks Year: 2004

- Sample Type: SB=Summer Baseflow; ST=Stormflow - Laboratory QA/QC results include baseflow and stormflow sample analyses. ______

------SB------ST------QAQC Cmpd Mean Stdev Obs #flag Mean Stdev Obs #flag

FB PH . . . . 7.08 1.16 3 0 FB ALKL (ueq/L) 63.0 125.7 4 0 7.94 2.96 3 0 FB ALKL (ueq/L) 63.0 125.7 4 0 -0.5 . 1 0 FB COND (uS/cm) . . . . 3.50 1.47 3 0 FB CA (mg/L) 0.0080 0.010 4 0 0.040 0.069 3 0 FB MG (mg/L) 0 0 4 0 0 0 3 0 FB NA (mg/L) 0.72 1.26 4 0 0.32 0.16 3 0 FB K (mg/L) 0.0050 0.010 4 0 0.23 0.25 3 0 FB CL (mg/L) 0 0 4 0 0.20 0.35 3 0 FB SO4 (mg/L) 1.13 0.49 4 0 1.23 0.12 3 0 FB NH4N (mg/L) 0.0030 0.0010 4 0 0.0080 0.0010 3 0 FB NO3N (mg/L) 0.0050 0.0040 4 0 0.014 0.0010 3 0 FB ORGN (mg/L) 0.0020 0.0050 4 0 0.022 0.0090 3 0 FB SKN (mg/L) 0.0030 0.0060 4 0 0.030 0.0080 3 0 FB TKN (mg/L) 0.0020 0.0030 3 0 0.024 0.014 3 0 FB SRP (mg/L) 0.0010 0.0010 4 0 0.0030 0.0010 3 0 FB TDP (mg/L) 0.0020 0.0010 4 0 0.0010 0.0010 3 0 FB TP (mg/L) 0.0020 0.0010 4 0 0.0010 0.0020 3 0

LB CA (mg/L) 0.00043 0.0021 23 0 . . . . LB MG (mg/L) 0 0 23 0 . . . . LB NA (mg/L) 0.034 0.030 19 0 . . . . LB K (mg/L) 0.0053 0.0092 15 0 . . . . LB CL (mg/L) 0.014 0.065 21 0 . . . . LB SO4 (mg/L) 1.01 0.52 31 0 . . . . LB NH4N (mg/L) 0.0038 0.0024 68 0 . . . . LB NO3N (mg/L) 0.0041 0.0027 61 0 . . . . LB SKN&TKN (mg/L) 0.012 0.019 39 0 . . . . LB SRP (mg/L) 0.00078 0.0011 73 0 . . . . LB TDP&TP (mg/L) 0.0012 0.00095 49 0 . . . .

Notes: (1) #flag indicates the number of samples whose concentration exceeded the defined QC criterion of 2*MDL (Method Detection Limit - see page A1.1 for values) ______Appendix source: \\StroudSAS\research\nywatershed\nywatershed2004\spiraling2004\NUTR_MAJORIONS_04_QAQC_APPENDIX_C.LST Produced on 07APR05:09:11

- 192 - APPENDIX TO CHAPTER 3 (A.3) NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

______

2004 QUALITY ASSURANCE/QUALITY CONTROL DATA FOR MAJOR IONS/NUTRIENTS Summary data for: Field (FD) & Lab (LD) Duplicates Year: 2004

- Sample Type: SB=Summer Baseflow; ST=Stormflow - Precision Evaluation Type: RPD=Relative Percent Difference; ABD DIFF=Absolute value of the difference - Laboratory QA/QC results include baseflow and stormflow sample analyses. ______------RPD (%)------ABS DIFF (conc.)----- Type QAQC Cmpd Mean Stdev Obs #flag Mean Stdev Obs #flag

SB FD PH . . 0 0 0.010 . 1 0 SB FD ALKL (ueq/L) 0.16 0.038 3 0 . . 0 0 SB FD COND (uS/cm) 0 . 1 0 . . 0 0 SB FD CA (mg/L) 0.77 0.68 4 0 . . 0 0 SB FD MG (mg/L) 0.68 0.83 4 0 . . 0 0 SB FD NA (mg/L) 0.17 0.30 3 0 0 . 1 0 SB FD K (mg/L) 0.29 0.41 2 0 0.020 0.028 2 0 SB FD CL (mg/L) 2.86 2.88 4 0 . . 0 0 SB FD SO4 (mg/L) 4.91 3.49 2 0 0.25 0.071 2 0 SB FD NH4N (mg/L) 1.68 . 1 0 0 0.0010 3 0 SB FD NO3N (mg/L) 1.83 0.0090 4 0 . . 0 0 SB FD ORGN (mg/L) 17.6 23.1 4 1 . . 0 0 SB FD SKN (mg/L) . . 0 0 0.015 0.014 4 0 SB FD TKN (mg/L) . . 0 0 0.015 0.0090 3 0 SB FD SRP (mg/L) 12.4 10.2 2 0 0.0010 0 2 0 SB FD TDP (mg/L) . . 0 0 0.0020 0.0020 4 0 SB FD TP (mg/L) . . 0 0 0.0010 0 3 0

SB/ST LD CA (mg/L) 0.50 0.60 12 0 . . 0 0 SB/ST LD MG (mg/L) 0.19 0.45 12 0 . . 0 0 SB/ST LD NA (mg/L) 0.32 0.28 6 0 0.013 0.021 6 0 SB/ST LD K (mg/L) 0.35 0.29 6 0 0.0033 0.0082 6 0 SB/ST LD CL (mg/L) 1.81 2.23 7 0 0.30 0.14 2 0 SB/ST LD SO4 (mg/L) 4.42 6.07 7 0 0.60 0.49 7 0 SB/ST LD NH4N (mg/L) 1.63 0.98 6 0 0.0011 0.0019 33 0 SB/ST LD NO3N (mg/L) 0.55 0.37 16 0 0.0010 0.00082 4 0 SB/ST LD SKN&TKN (mg/L) 7.62 4.95 2 0 0.013 0.010 22 0 SB/ST LD SRP (mg/L) 2.49 3.90 19 0 0.00014 0.00036 21 0 SB/ST LD TDP&TP (mg/L) 6.74 5.66 4 0 0.00076 0.00094 21 0

ST FD PH . . 0 0 0 . 1 0 ST FD ALKL (ueq/L) 0.56 . 1 0 . . 0 0 ST FD COND (uS/cm) 0 . 1 0 . . 0 0 ST FD CA (mg/L) 5.63 6.54 2 0 . . 0 0 ST FD MG (mg/L) 4.51 6.38 2 0 . . 0 0 ST FD NA (mg/L) 0.74 0.056 2 0 . . 0 0 ST FD K (mg/L) 0.65 0.91 2 0 . . 0 0 ST FD CL (mg/L) 0.96 1.35 2 0 . . 0 0 ST FD SO4 (mg/L) 0.77 . 1 0 1.60 . 1 0 ST FD NH4N (mg/L) 4.38 . 1 0 0.0010 . 1 0 ST FD NO3N (mg/L) 1.36 1.08 2 0 . . 0 0 ST FD ORGN (mg/L) 8.69 6.13 2 0 . . 0 0 ST FD SKN (mg/L) . . 0 0 0.025 0.013 2 0 ST FD TKN (mg/L) 33.3 25.6 2 1 . . 0 0 ST FD SRP (mg/L) 0.90 . 1 0 0.0010 . 1 0 ST FD TDP (mg/L) 0 . 1 0 0.0020 . 1 0 ST FD TP (mg/L) 59.2 28.3 2 2 . . 0 0

Notes: (1) #flag indicates the number of samples whose concentration exceeded the defined QC criterion (see page A1.1 for values). ______Appendix source: \\StroudSAS\research\nywatershed\nywatershed2004\spiraling2004\NUTR_MAJORIONS_04_QAQC_APPENDIX_C.LST Produced on 07APR05:09:11

- 193 - APPENDIX TO CHAPTER 3 (A.3) NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

______

2004 QUALITY ASSURANCE/QUALITY CONTROL DATA FOR MAJOR IONS/NUTRIENTS Summary data for: Matrix Spikes (as percent recovery)

- Sample Type: SB=Summer Baseflow; ST=Stormflow - Laboratory QA/QC results include baseflow and stormflow sample analyses. ______

Year Cmpd Mean Stdev Obs #flag

2004 CA (mg/L) 97.7 5.84 12 0 2004 MG (mg/L) 100.1 2.43 12 0 2004 NA (mg/L) 99.9 3.70 12 0 2004 K (mg/L) 102.5 3.55 12 0 2004 CL (mg/L) 104.3 6.73 7 0 2004 SO4 (mg/L) 93.2 13.4 15 0 2004 NH4N (mg/L) 100.1 2.87 38 0 2004 NO3N (mg/L) 100.0 2.25 20 0 2004 SKN&TKN (mg/L) 102.6 8.31 24 0 2004 SRP (mg/L) 101.8 3.10 39 0 2004 TDP&TP (mg/L) 97.6 3.08 25 0

Notes: (1) #flag indicates the number of samples whose percent recovery exceeded the defined QC criterion (see page A1.1 for values). ______Appendix source: \\StroudSAS\research\nywatershed\nywatershed2004\spiralings2004\NUTR_MAJORIONS_04_QAQC_APPENDIX_C.LST Produced on 07APR05:09:11

- 194 - APPENDIX TO CHAPTER 3 (A.3) NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

______

2004 QUALITY ASSURANCE/QUALITY CONTROL DATA FOR MAJOR IONS/NUTRIENTS Summary data for: Laboratory Control Standards (as percent recovery)

- Sample Type: SB=Summer Baseflow; ST=Stormflow - Laboratory QA/QC results include baseflow and stormflow sample analyses. ______

Year Cmpd Mean Stdev Obs #flag

2004 CA (mg/L) 99.4 3.81 8 0 2004 MG (mg/L) 102.0 1.07 8 0 2004 NA (mg/L) 102.0 2.08 7 0 2004 K (mg/L) 102.3 2.07 6 0 2004 CL (mg/L) 102.0 3.67 31 0 2004 SO4 (mg/L) 102.2 3.69 30 0 2004 NH4N (mg/L) 105.4 1.81 32 0 2004 NO3N (mg/L) 101.8 1.46 36 0 2004 SKN&TKN (mg/L) 99.9 4.72 28 0 2004 SRP (mg/L) 101.1 2.67 32 0 2004 TDP&TP (mg/L) 100.7 5.22 28 0

- 195 - APPENDIX TO CHAPTER 3 (A.3) NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

______

2004 QUALITY ASSURANCE/QUALITY CONTROL DATA FOR MAJOR IONS/NUTRIENTS Summary data for: Continuing Calibration Standards (as percent recovery)

- Sample Type: SB=Summer Baseflow; ST=Stormflow - Laboratory QA/QC results include baseflow and stormflow sample analyses. ______

Year Cmpd Mean Stdev Obs #flag

2004 CA (mg/L) 97.7 4.75 23 0 2004 MG (mg/L) 99.8 2.59 24 0 2004 NA (mg/L) 100.7 15.1 19 1 2004 K (mg/L) 103.3 4.17 16 0 2004 SO4 (mg/L) 106.8 9.94 37 0 2004 NH4N (mg/L) 96.6 10.6 111 6 2004 NO3N (mg/L) 103.7 9.73 97 6 2004 SKN&TKN (mg/L) 103.1 11.5 72 3 2004 SRP (mg/L) 102.8 14.3 119 5 2004 TDP&TP (mg/L) 99.8 12.3 87 5

- 196 - APPENDIX TO CHAPTER 3 (A.3) NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

______

2004 QUALITY ASSURANCE/QUALITY CONTROL DATA FOR MAJOR IONS/NUTRIENTS Summary data for: Conductivity Checks and Cation/Anion Balance (as percent differences)

- Sample Type: SB=Summer Baseflow; ST=Stormflow - Conductivity Check as the difference between measured and calculating conductivity divided by measured conductivity - Cation/Anion Balance as the sum of anions subtracted from the sum of cations divided by the sum of all ions (units = ueq/L) ______

Year Task Mean Stdev Min Max N obs

CONDUCTIVITY CHECK (uS/cm)

2004 SB -1.7 19.0 -115.6 15.5 65 2004 ST 2.08 6.51 -10.8 10.8 16

CATION/ANION BALANCE (ueq/L)

2004 SB -0.8 4.51 -15.2 29.6 70 2004 ST -2.7 3.75 -15.0 3.96 19

Appendix source: \\StroudSAS\research\nywatershed\nywatershed2004\spiralings2004\NUTR_MAJORIONS_04_QAQC_APPENDIX_C.LST ______Produced on 07APR05:09:11

- 197 - APPENDIX TO CHAPTER 3 (A.3) NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

------Intentionally Blank ------

- 198 - APPENDIX TO CHAPTER 3 (A.3) NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Appendix to Chapter 4 – Molecular Tracers

______

QUALITY ASSURANCE/QUALITY CONTROL DATA FOR MOLECULAR TRACERS - 2004. DATA SUMMARY BY SAMPLING SEASON ------

METHOD DETECTION LIMITS (MDL), QUALITY CONTROL CRITERIA (AS EXCEEDANCE LIMITS):

CMPD MDL (2002-2004) Precision (RPD specific) Accuracy (MS specific) (ug/L) (%) (%)

FLU 0.00061 30 30-150 PHE 0.00082 30 30-150 ANT 0.00029 30 30-150 2MP 0.00027 30 30-150 1MP 0.00021 30 30-150 FLR 0.00012 30 30-150 PYR 0.00009 30 30-150 BAA 0.00040 30 30-150 CHR 0.00062 30 30-150 BBF 0.00006 30 30-150 BKF 0.0074 30 30-150 BAP 0.011 30 30-150 2MN 0.0013 30 30-150 HHCB 0.0039 50 25-175 AHTN 0.0012 50 25-175 CAF 0.0076 30 30-150 bCOP 0.00040 50 25-175 EPI 0.0039 50 25-175 CHOL 0.020 50 25-175 aCOP 0.0016 50 25-175 eCOP 0.00031 50 25-175 eEPI . 50 25-175 bONE 0.0015 75 25-175 aONE 0.0011 75 25-175 eCHO 0.19 50 25-175 SNOL 0.0022 75 25-175

ADDITIONAL DEFINITIONS AND NOTES:

1. Sample blank (field and laboratory) exceedances are assessed at concentrations > 2*MDL value 2. Precision assessed using either the Relative Percent Difference (RPD) value or the Absolute Difference between duplicate values - When the mean concentration of a duplicate set of samples is < the quantity of (1/stated precision limit * MDL) then the appropriate precision assessment is the absolute difference with a corresponding exceedance criterion equal to the MDL. - Precision summaries presented in the Molecular Tracer QAQC appendix include all duplicate pairs for the absolute difference summaries, while only those duplicate pairs whose mean concentration is greater than the criterion defined above are included in the RPD summaries. 3. Accuracy assessed from percent recovery of matrix spike (MS) and standard reference material (SRM) samples

______Appendix source: \\StroudSAS\research\nywatershed\nywatershed2004\tracers2004\NYC_TRCRS04_QAQC_APPENDIX_C.LST Produced on 02MAY05:11:39

- 199 - APPENDIX TO CHAPTER 4 (A.4) NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

______

QUALITY ASSURANCE/QUALITY CONTROL DATA FOR MOLECULAR TRACERS - 2004. Summary data for: Field (FB) & Lab (LB) Blanks Year: 2004 Compound Group: PAH/Fragrances/Caffeine - Data are sorted by compound retention times - SB=Summer Baseflow; ST=Stormflow - Lab blanks cover both baseflow and stormflow sampling ______

------SB (ug/L)------WB (ug/L)------ST (ug/L)------QAQC Cmpd Mean Stdev Obs #flag Mean Stdev Obs #flag Mean Stdev Obs #flag

FB FLU 0.12 0.23 4 1 0.00037 0.00010 2 0 0.0057 0.0022 3 3 FB PHE 0.23 0.45 4 2 0.0029 0.0013 2 0 0.015 0.0010 3 3 FB ANT 0.11 0.21 4 2 0.0020 0.0028 2 0 0.0013 0.00082 3 3 FB 2MP 0.15 0.31 4 3 0.00066 0.00004 2 0 0.0046 0.0021 3 3 FB 1MP 0.050 0.099 4 2 0.00049 0.00016 2 0 0.0033 0.0021 3 3 FB FLR 0.033 0.065 4 3 0.0015 0.0011 2 0 0.0052 0.0011 3 3 FB PYR 0.042 0.082 4 3 0.0013 0.00091 2 0 0.0023 0.00027 3 3 FB BAA 0.020 0.039 4 1 0.00054 0.00029 2 0 0.043 0.071 3 3 FB CHR 0.0069 0.011 4 2 0.00098 0.00019 2 0 0.037 0.062 3 2 FB BBF 0.00008 0.00007 4 1 0.015 0.011 2 0 0.0018 0.0010 3 3 FB BKF 0.015 0.029 4 1 0.013 0.0046 2 0 0.00085 0.00050 3 0 FB BAP 0.059 0.088 4 2 0.0019 0.00077 2 0 0.00043 0.00041 3 0 FB 2MN 0.063 0.13 4 0 0.00017 . 1 0 0.0086 0.0041 3 0

Notes: (1) #flag indicates the number of samples whose concentration exceeded the defined QC criterion of 2*MDL (Method Detection Limit - see first page for values) ______Appendix source: \\StroudSAS\research\nywatershed\nywatershed2004\tracers2004\NYC_TRCRS04_QAQC_APPENDIX_C.LST Produced on 02MAY05:11:39

- 200 - APPENDIX TO CHAPTER 4 (A.4) NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

______

------SB/ST (ug/L)------WB (ug/L)------QAQC Cmpd Mean Stdev Obs #flag Mean Stdev Obs #flag

LB FLU 0.0010 0.00048 6 2 0.00024 0.00027 2 0 LB PHE 0.018 0.026 6 4 0.0034 0.0027 2 0 LB ANT 0.00043 0.00022 6 1 0.00031 0.00039 2 0 LB 2MP 0.00055 0.00034 6 2 0.00084 0.00054 2 0 LB 1MP 0.00030 0.00017 6 2 0.00051 0.00019 2 0 LB FLR 0.0011 0.0010 6 4 0.0010 0.0012 2 0 LB PYR 0.0011 0.0012 6 6 0.00075 0.00066 2 0 LB BAA 0.0015 0.0024 6 2 0.00083 0.00056 2 0 LB CHR 0.0025 0.0035 6 2 0.0033 0.0045 2 0 LB BBF 0.00010 0.00006 6 2 0.0012 0.0017 2 0 LB BKF 0.00016 0.00014 6 0 0.00057 0.00081 2 0 LB BAP 0.00010 0.00010 6 0 0.00052 0.00034 2 0 LB 2MN 0.00046 0.00035 6 0 0.00011 . 1 0

- 201 - APPENDIX TO CHAPTER 4 (A.4) NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

______

QUALITY ASSURANCE/QUALITY CONTROL DATA FOR MOLECULAR TRACERS - 2004. Summary data for: Field (FB) & Lab (LB) Blanks Year: 2004 Compound Group: Fecal Sterols - Data are sorted by compound retention times - SB=Summer Baseflow; ST=Stormflow - Lab blanks cover both baseflow and stormflow sampling ______

------SB (ug/L)------WB (ug/L)------ST (ug/L)------QAQC Cmpd Mean Stdev Obs #flag Mean Stdev Obs #flag Mean Stdev Obs #flag

FB HHCB 0.85 1.70 4 1 0.0027 0.0019 2 0 0.014 0.011 3 2 FB AHTN 0.25 0.49 4 1 0.0014 0.0011 2 0 0.0050 0.0010 3 3 FB CAF 1.00 1.99 4 1 0.0023 0.00091 2 0 0.011 0.0080 3 1 FB bCOP 0.041 0.082 4 1 0.019 0.027 2 0 0.0044 0.0039 3 2 FB EPI 0.67 1.31 4 2 0.0020 0.00034 2 0 0.018 0.016 3 2 FB CHOL 3.60 7.14 4 0 0.11 0.14 2 0 0.16 0.16 3 0 FB aCOP 0.094 0.19 4 2 0.013 0.017 2 0 0.0034 0.0036 3 1 FB eCOP 0.00001 0.00001 4 0 0 0 2 0 0.00014 0.00025 3 0 FB eEPI 0.00002 0.00004 4 0 0 0 2 0 0.00027 0.00036 3 0 FB bONE 0.11 0.22 4 2 0.0092 0.012 2 0 0.011 0.012 3 2 FB aONE 0.0012 0.0023 4 1 0.0064 0.0076 2 0 0.0011 0.0010 3 0 FB eCHO 37.4 74.5 4 0 0.10 0.11 2 0 0.38 0.53 3 0 FB SNOL 0.35 0.68 4 2 0.0083 0.011 2 0 0.020 0.017 3 3

- 202 - APPENDIX TO CHAPTER 4 (A.4) NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

______

------SB/ST (ug/L)------WB (ug/L)------QAQC Cmpd Mean Stdev Obs #flag Mean Stdev Obs #flag

LB HHCB 0.0079 0.0061 6 3 0.0045 0.0031 2 0 LB AHTN 0.0015 0.00087 6 2 0.0013 0.00074 2 0 LB CAF 0.0031 0.0038 6 0 0.013 0.015 2 0 LB bCOP 0.00005 0.00007 6 0 0.00024 0.00010 2 0 LB EPI 0.0029 0.0028 6 0 0.0012 0.00077 2 0 LB CHOL 0.015 0.0064 6 0 0.023 0.0092 2 0 LB aCOP 0.00056 0.00040 6 0 0.00081 0.00018 2 0 LB eCOP 0.00001 0.00001 6 0 0 0 2 0 LB eEPI 0 0 6 0 0 0 2 0 LB bONE 0.0021 0.0032 6 1 0.00077 0.00051 2 0 LB aONE 0.0017 0.0027 6 1 0.00011 0.00015 2 0 LB eCHO 0.26 0.22 6 0 0.16 0.042 2 0 LB SNOL 0.0062 0.0085 6 2 0.0040 0.00039 2 0

- 203 - APPENDIX TO CHAPTER 4 (A.4) NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

______

QUALITY ASSURANCE/QUALITY CONTROL DATA FOR MOLECULAR TRACERS - 2004. Summary data for: Field (FD) & Lab (LD) Duplicates Year: 2004 Compound Group: PAH/Fragrances/Caffeine Type: SB

- Data are sorted by compound retention times - SB=Summer Baseflow; ST=Stormflow - Precision Evaluation Type: RPD=Relative Percent Difference; ABD DIFF=Absolute value of the difference ______------RPD (%)------ABS DIFF (ug/L)----- QAQC Cmpd Mean Stdev Obs #flag Mean Stdev Obs #flag

FD FLU . . 0 0 0.00039 0.00030 4 1 FD PHE 17.9 7.71 3 0 0.00062 . 1 0 FD ANT 46.9 36.3 2 1 0.00043 0.00050 2 1 FD 2MP 38.7 26.8 3 2 0.00060 . 1 1 FD 1MP 30.1 27.8 4 1 . . 0 0 FD FLR 26.7 19.3 4 2 . . 0 0 FD PYR 25.8 16.7 4 2 . . 0 0 FD BAA 102.9 80.7 3 3 0.00001 . 1 0 FD CHR 45.4 0.087 2 2 0.00041 0.00050 2 1 FD BBF 54.1 40.9 4 2 . . 0 0 FD BKF . . 0 0 0.0015 0.0014 4 0 FD BAP . . 0 0 0.0014 0.0015 4 0 FD 2MN . . 0 0 0.00019 0.00011 4 0 FD PHEd10 0.69 0.45 4 0 . . 0 0 FD CHR-D 0.69 0.45 4 0 . . 0 0 FD PERd12 0.69 0.45 4 0 . . 0 0

Notes: (1) #flag indicates the number of samples whose concentration exceeded the defined QC criterion (see first page for values). ______Appendix source: \\StroudSAS\research\nywatershed\nywatershed2004\tracers2004\NYC_TRCRS04_QAQC_APPENDIX_C.LST Produced on 02MAY05:11:39

- 204 - APPENDIX TO CHAPTER 4 (A.4) NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

______

QUALITY ASSURANCE/QUALITY CONTROL DATA FOR MOLECULAR TRACERS - 2004. Summary data for: Field (FD) & Lab (LD) Duplicates Year: 2004 Compound Group: PAH/Fragrances/Caffeine Type: ST

FD FLU 11.9 13.3 2 0 . . 0 0 FD PHE 20.0 12.4 2 0 . . 0 0 FD ANT 40.4 33.8 2 1 . . 0 0 FD 2MP 6.45 8.12 2 0 . . 0 0 FD 1MP 11.3 3.51 2 0 . . 0 0 FD FLR 26.7 16.7 2 1 . . 0 0 FD PYR 20.0 12.6 2 0 . . 0 0 FD BAA 32.1 10.8 2 1 . . 0 0 FD CHR 30.1 11.2 2 1 . . 0 0 FD BBF 33.7 7.97 2 1 . . 0 0 FD BKF 37.5 3.21 2 2 . . 0 0 FD BAP 37.6 4.04 2 2 . . 0 0 FD 2MN 24.5 29.3 2 1 . . 0 0 FD PHEd10 0.73 0.26 2 0 . . 0 0 FD CHR-D 0.73 0.26 2 0 . . 0 0 FD PERd12 0.73 0.26 2 0 . . 0 0

- 205 - APPENDIX TO CHAPTER 4 (A.4) NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

______

QUALITY ASSURANCE/QUALITY CONTROL DATA FOR MOLECULAR TRACERS - 2004. Summary data for: Field (FD) & Lab (LD) Duplicates Year: 2004 Compound Group: PAH/Fragrances/Caffeine Type: WB

FD FLU . . 0 0 0.00011 0.00005 2 0 FD PHE 11.2 5.73 2 0 . . 0 0 FD ANT . . 0 0 0.00001 0.00000 2 0 FD 2MP 3.20 . 1 0 0.00001 . 1 0 FD 1MP 90.4 . 1 1 0.00026 . 1 1 FD FLR 22.2 5.01 2 0 . . 0 0 FD PYR 26.6 0.86 2 0 . . 0 0 FD BAA 30.7 . 1 1 0.00049 . 1 1 FD CHR 26.9 . 1 0 0.00047 . 1 0 FD BBF 41.8 7.36 2 2 . . 0 0 FD BKF . . 0 0 0.0031 0.00098 2 0 FD BAP . . 0 0 0.0024 0.0029 2 0 FD 2MN . . 0 0 0.00019 . 1 0 FD PHEd10 1.15 1.32 2 0 . . 0 0 FD CHR-D 1.15 1.32 2 0 . . 0 0 FD PERd12 1.15 1.32 2 0 . . 0 0

- 206 - APPENDIX TO CHAPTER 4 (A.4) NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

______

QUALITY ASSURANCE/QUALITY CONTROL DATA FOR MOLECULAR TRACERS - 2004. Summary data for: Field (FD) & Lab (LD) Duplicates Year: 2004 Compound Group: Fecal Sterols Type: SB

FD HHCB 54.2 . 1 1 0.0027 0.0022 3 1 FD AHTN 61.0 . 1 1 0.00062 0.00078 3 1 FD CAF 22.2 . 1 0 0.0016 0.0010 3 0 FD CAFd9 0.69 0.45 4 0 . . 0 0 FD bCOP 60.2 62.5 4 1 . . 0 0 FD EPI 190.8 . 1 1 0.0011 0.00032 3 0 FD CHOL 54.1 69.6 4 1 . . 0 0 FD aCOP 50.1 66.2 4 1 . . 0 0 FD eCOP 108.7 83.7 3 2 0.00098 . 1 1 FD eEPI 162.4 75.2 4 3 . . 0 0 FD bONE 65.4 80.4 4 1 . . 0 0 FD aONE 26.0 16.8 4 0 . . 0 0 FD eCHO 81.7 66.5 3 2 0.0031 . 1 0 FD SNOL 45.7 73.3 4 1 . . 0 0 FD CHOLd6 0.69 0.45 4 0 . . 0 0

- 207 - APPENDIX TO CHAPTER 4 (A.4) NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

______

QUALITY ASSURANCE/QUALITY CONTROL DATA FOR MOLECULAR TRACERS - 2004. Summary data for: Field (FD) & Lab (LD) Duplicates Year: 2004 Compound Group: Fecal Sterols Type: ST

FD HHCB 32.2 8.75 2 0 . . 0 0 FD AHTN 25.5 15.9 2 0 . . 0 0 FD CAF 11.1 11.5 2 0 . . 0 0 FD CAFd9 0.73 0.26 2 0 . . 0 0 FD bCOP 96.2 106.2 2 1 . . 0 0 FD EPI 118.3 15.4 2 2 . . 0 0 FD CHOL 87.0 113.8 2 1 . . 0 0 FD aCOP 80.9 112.5 2 1 . . 0 0 FD eCOP 92.9 110.4 2 1 . . 0 0 FD eEPI 187.8 17.3 2 2 . . 0 0 FD bONE 90.7 111.3 2 1 . . 0 0 FD aONE 85.5 112.5 2 1 . . 0 0 FD eCHO 86.9 120.4 2 1 . . 0 0 FD SNOL 84.1 118.7 2 1 . . 0 0 FD CHOLd6 0.73 0.26 2 0 . . 0 0

- 208 - APPENDIX TO CHAPTER 4 (A.4) NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

______

QUALITY ASSURANCE/QUALITY CONTROL DATA FOR MOLECULAR TRACERS - 2004. Summary data for: Field (FD) & Lab (LD) Duplicates Year: 2004 Compound Group: Fecal Sterols Type: WB

FD HHCB . . 0 0 0.0011 0.00026 2 0 FD AHTN 169.0 . 1 1 0.00065 . 1 0 FD CAF . . 0 0 0.0031 0.0039 2 0 FD CAFd9 1.15 1.32 2 0 . . 0 0 FD bCOP 49.1 27.1 2 1 . . 0 0 FD EPI . . 0 0 0.00056 0.00028 2 0 FD CHOL 45.3 7.57 2 1 . . 0 0 FD aCOP 21.0 5.80 2 0 . . 0 0 FD eCOP 16.9 . 1 0 0 . 1 0 FD eEPI 43.6 . 1 0 . . 0 0 FD bONE 46.5 18.0 2 0 . . 0 0 FD aONE 61.0 7.25 2 0 . . 0 0 FD eCHO 71.3 29.5 2 2 . . 0 0 FD SNOL 43.4 7.89 2 0 . . 0 0 FD CHOLd6 1.15 1.32 2 0 . . 0 0

- 209 - APPENDIX TO CHAPTER 4 (A.4) NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

______

QUALITY ASSURANCE/QUALITY CONTROL DATA FOR MOLECULAR TRACERS - 2004. Summary data for: Matrix Spikes (as percent recovery) Year: 2004

- Data are sorted by compound retention times - SB=Summer Baseflow; ST=Stormflow - Matrix spikes cover both baseflow and stormflow sampling ______

------SB/ST (%)------WB (%)------Tracer Group Cmpd Mean Stdev Obs#flag Mean Stdev Obs #flag

PAH/Fragrances/Caffeine FLU 81.5 12.6 4 0 16.5 31.1 2 1 PAH/Fragrances/Caffeine PHE 70.9 23.2 4 0 34.1 64.0 2 1 PAH/Fragrances/Caffeine ANT 78.2 12.3 4 0 8.18 6.85 2 2 PAH/Fragrances/Caffeine 2MP 120.1 43.5 4 1 54.3 74.4 2 1 PAH/Fragrances/Caffeine 1MP 108.2 13.2 4 0 34.9 51.9 2 1 PAH/Fragrances/Caffeine FLR 80.4 45.9 4 1 28.9 43.8 2 1 PAH/Fragrances/Caffeine PYR 75.1 38.9 4 1 30.8 47.5 2 1 PAH/Fragrances/Caffeine BAA 13.9 144.5 4 1 29.9 47.7 2 1 PAH/Fragrances/Caffeine CHR 73.0 16.0 4 0 29.1 47.7 2 1 PAH/Fragrances/Caffeine BBF 98.4 20.4 4 0 78.7 130.7 2 2 PAH/Fragrances/Caffeine BKF 83.7 25.2 4 0 44.1 67.1 2 1 PAH/Fragrances/Caffeine BAP 74.5 11.2 4 0 32.3 68.8 2 1 PAH/Fragrances/Caffeine 2MN 42.5 11.3 4 1 -1.6 . 1 1

Notes: (1) #flag indicates the number of samples whose percent recovery exceeded the defined QC criterion (see first page for values). ______Appendix source: \\StroudSAS\research\nywatershed\nywatershed2004\tracers2004\NYC_TRCRS04_QAQC_APPENDIX_C.LST Produced on 02MAY05:11:39

- 210 - APPENDIX TO CHAPTER 4 (A.4) NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

______

QUALITY ASSURANCE/QUALITY CONTROL DATA FOR MOLECULAR TRACERS - 2004. Summary data for: Matrix Spikes (as percent recovery) Year: 2004

- Data are sorted by compound retention times - SB=Summer Baseflow; ST=Stormflow - Matrix spikes cover both baseflow and stormflow sampling ______

------SB/ST (%)------WB (%)------Tracer Group Cmpd Mean Stdev Obs#flag Mean Stdev Obs #flag

Fecal Sterols HHCB 97.9 47.2 4 0 13.9 33.8 2 1 Fecal Sterols AHTN 153.8 91.2 4 1 12.0 68.4 2 1 Fecal Sterols CAF 184.7 198.7 4 2 17.1 14.1 2 2 Fecal Sterols bCOP -56.3 585.1 4 2 14.0 15.7 2 1 Fecal Sterols EPI 59.5 192.6 4 2 30.5 21.9 2 1 Fecal Sterols CHOL -1284 8781.7 4 4 14.5 1479.0 2 2 Fecal Sterols aCOP -89.2 953.7 4 2 -30.6 4.16 2 2 Fecal Sterols eCOP -22.9 124.7 4 3 -1.5 2.17 2 2 Fecal Sterols eEPI 19.5 49.8 4 3 -1.2 1.73 2 2 Fecal Sterols bONE 26.0 205.7 4 2 29.0 8.33 2 1 Fecal Sterols aONE 96.6 110.6 4 1 40.4 31.3 2 1 Fecal Sterols eCHO -1096 7621.4 4 4 -222.6 2812.3 2 2 Fecal Sterols SNOL -107.8 1054.1 4 3 4.60 96.8 2 1

- 211 - APPENDIX TO CHAPTER 4 (A.4) NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

______

QUALITY ASSURANCE/QUALITY CONTROL DATA FOR MOLECULAR TRACERS - 2004. Summary data for: Matrix Spike Duplicates Year: 2004 Compound Group: PAH/Fragrances/Caffeine

- Data are sorted by compound retention times - SB=Summer Baseflow; ST=Stormflow - Precision Evaluation Type: RPD=Relative Percent Difference; ABD DIFF=Absolute value of the difference ______------RPD (%)------ABS DIFF (ug/L)------Type Cmpd Mean Stdev Obs #flag Mean Stdev Obs #flag

SB FLU 2.87 3.32 4 0 . . 0 0 SB PHE 0.95 0.54 4 0 . . 0 0 SB ANT 5.19 4.86 4 0 . . 0 0 SB 2MP 2.45 0.84 4 0 . . 0 0 SB 1MP 4.55 6.70 4 0 . . 0 0 SB FLR 3.59 1.92 4 0 . . 0 0 SB PYR 4.42 3.57 4 0 . . 0 0 SB BAA 4.86 5.46 4 0 . . 0 0 SB CHR 2.87 4.36 4 0 . . 0 0 SB BBF 8.14 7.00 4 0 . . 0 0 SB BKF . . 0 0 0.00068 0.00043 4 0 SB BAP . . 0 0 0.0011 0.00070 4 0 ST FLU 1.27 . 1 0 . . 0 0 ST PHE 1.14 . 1 0 . . 0 0 ST ANT 2.54 . 1 0 . . 0 0 ST 2MP 0.97 . 1 0 . . 0 0 ST 1MP 3.79 . 1 0 . . 0 0 ST FLR 0.0036 . 1 0 . . 0 0 ST PYR 0.37 . 1 0 . . 0 0 ST BAA 0.99 . 1 0 . . 0 0 ST CHR 0.080 . 1 0 . . 0 0 ST BBF 2.76 . 1 0 . . 0 0 ST BKF 8.00 . 1 0 . . 0 0 ST BAP 0.79 . 1 0 . . 0 0 WB FLU 2.00 . 1 0 0.00004 . 1 0 WB PHE 4.31 1.53 2 0 . . 0 0 WB ANT 26.6 17.7 2 1 . . 0 0 WB 2MP 2.25 0.46 2 0 . . 0 0 WB 1MP 0.43 0.12 2 0 . . 0 0 WB FLR 0.72 0.71 2 0 . . 0 0 WB PYR 6.07 1.49 2 0 . . 0 0 WB BAA 0.52 0.44 2 0 . . 0 0 WB CHR 1.35 1.17 2 0 . . 0 0 WB BBF 8.03 1.01 2 0 . . 0 0 WB BKF . . 0 0 0.0015 0.00042 2 0 WB BAP . . 0 0 0.0011 0.0013 2 0

______Appendix source: \\StroudSAS\research\nywatershed\nywatershed2004\tracers2004\NYC_TRCRS04_QAQC_APPENDIX_C.LST Produced on 02MAY05:11:39

- 212 - APPENDIX TO CHAPTER 4 (A.4) NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

______

QUALITY ASSURANCE/QUALITY CONTROL DATA FOR MOLECULAR TRACERS - 2004. Summary data for: Matrix Spike Duplicates Year: 2004 Compound Group: Fecal Sterols

- Data are sorted by compound retention times - SB=Summer Baseflow; ST=Stormflow - Precision Evaluation Type: RPD=Relative Percent Difference; ABD DIFF=Absolute value of the difference ______------RPD (%)------ABS DIFF (ug/L)------Type Cmpd Mean Stdev Obs #flag Mean Stdev Obs #flag

SB HHCB 7.31 3.74 4 0 . . 0 0 SB AHTN 4.59 1.44 4 0 . . 0 0 SB CAF 19.0 . 1 0 0.0029 0.0018 3 0 SB bCOP 35.2 28.6 4 1 . . 0 0 SB EPI 19.5 15.9 3 0 0.0026 . 1 0 SB CHOL 33.9 20.1 4 1 . . 0 0 SB aCOP 29.1 17.0 4 0 . . 0 0 SB bONE 21.8 23.1 4 0 . . 0 0 SB aONE 19.9 16.7 4 0 . . 0 0 SB eCHO 52.0 32.2 2 1 0.043 0.014 2 0 SB SNOL 20.7 26.0 4 0 . . 0 0 ST HHCB 1.29 . 1 0 . . 0 0 ST AHTN 0.024 . 1 0 . . 0 0 ST CAF 5.63 . 1 0 . . 0 0 ST bCOP 19.7 . 1 0 . . 0 0 ST EPI 1.36 . 1 0 . . 0 0 ST CHOL 66.7 . 1 1 . . 0 0 ST aCOP 36.3 . 1 0 . . 0 0 ST bONE 18.2 . 1 0 . . 0 0 ST aONE 3.22 . 1 0 . . 0 0 ST eCHO 52.9 . 1 1 . . 0 0 ST SNOL 6.19 . 1 0 . . 0 0 WB HHCB . . 0 0 0.00019 0.00002 2 0 WB AHTN 3.48 . 1 0 0.00000 . 1 0 WB CAF . . 0 0 0.00067 0.00000 2 0 WB bCOP 9.33 7.94 2 0 . . 0 0 WB EPI . . 0 0 0.00037 0.00036 2 0 WB CHOL 2.18 2.57 2 0 . . 0 0 WB aCOP 0.56 0.34 2 0 . . 0 0 WB bONE 48.0 24.5 2 0 . . 0 0 WB aONE 7.62 5.05 2 0 . . 0 0 WB eCHO 6.27 . 1 0 0.015 . 1 0 WB SNOL 8.48 1.37 2 0 . . 0 0

______Appendix source: \\StroudSAS\research\nywatershed\nywatershed2004\tracers2004\NYC_TRCRS04_QAQC_APPENDIX_C.LST Produced on 02MAY05:11:39

- 213 - APPENDIX TO CHAPTER 4 (A.4) NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

______

QUALITY ASSURANCE/QUALITY CONTROL DATA FOR MOLECULAR TRACERS - 2004. Summary data for: Standard Reference Material

- Data are sorted by compound retention times - Sample Type: SB=Summer Baseflow; WB=Winter Baseflow; ST=Stormflow ______

Type SRM lot# Year Cmpd Cert value (ug/L) Mean Stdev Obs #flag

SB QCO121_057 2004 FLU 139 (68.5- 209) 122.6 25.5 4 0 SB QCO121_057 2004 PHE 142 (55.6- 228) 144.1 19.6 4 0 SB QCO121_057 2004 ANT 0.0 ( 0.0-10.0) 1.45 0.77 4 0 SB QCO121_057 2004 2MP 0.0 ( 0.0-10.0) 0.31 0.16 4 0 SB QCO121_057 2004 1MP 0.0 ( 0.0-10.0) 0.21 0.13 4 0 SB QCO121_057 2004 FLR 133 (65.4- 201) 125.8 21.5 4 0 SB QCO121_057 2004 PYR 134 (53.4- 215) 121.5 20.2 4 0 SB QCO121_057 2004 BAA 126 (71.7- 180) 105.4 9.85 4 0 SB QCO121_057 2004 CHR 0.0 ( 0.0-10.0) 1.09 0.77 4 0 SB QCO121_057 2004 BBF 0.0 ( 0.0-10.0) 0.12 0.15 4 0 SB QCO121_057 2004 BKF 0.0 ( 0.0-10.0) 0.099 0.12 4 0 SB QCO121_057 2004 BAP 0.0 ( 0.0-10.0) 0.75 1.12 4 0 WB QCO121_057 2004 FLU 139 (68.5- 209) 47.1 18.3 2 2 WB QCO121_057 2004 PHE 142 (55.6- 228) 123.4 17.6 2 0 WB QCO121_057 2004 ANT 0.0 ( 0.0-10.0) 1.14 1.62 2 0 WB QCO121_057 2004 2MP 0.0 ( 0.0-10.0) 0.36 0.12 2 0 WB QCO121_057 2004 1MP 0.0 ( 0.0-10.0) 0.52 0.52 2 0 WB QCO121_057 2004 FLR 133 (65.4- 201) 68.2 0.47 2 0 WB QCO121_057 2004 PYR 134 (53.4- 215) 72.6 1.16 2 0 WB QCO121_057 2004 BAA 126 (71.7- 180) 97.5 1.81 2 0 WB QCO121_057 2004 CHR 0.0 ( 0.0-10.0) 1.34 1.90 2 0 WB QCO121_057 2004 BBF 0.0 ( 0.0-10.0) 0.25 0.069 2 0 WB QCO121_057 2004 BKF 0.0 ( 0.0-10.0) 0.14 0.052 2 0 WB QCO121_057 2004 BAP 0.0 ( 0.0-10.0) 0.18 0.069 2 0

Notes: (1) #flag indicates the number of SRM samples outside of acceptance limits (as defined by range given in the 'Cert value (ug/L)' field). ______Appendix source: \\StroudSAS\research\nywatershed\nywatershed2004\tracers2004\NYC_TRCRS04_QAQC_APPENDIX_C.LST Produced on 02MAY05:11:39

- 214 - APPENDIX TO CHAPTER 4 (A.4) NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Appendix to Chapter 5 - Macroinvertebrate Community

Appendix Table A.5.1: Macroinvertebrate sampling sites where the sampling protocol was modified in response to field conditions during 2004.

Site Site Description Samples Collected 6 W. Br. Delaware R. at Hawleys 8 random samples 105 East Brook nr Walton 8 random samples 126 Stump Pond Stream nr Pawling 8 random samples 132 Bog Brook nr Sears Corner 8 random samples 143 Unnamed trib. to Cross River 4 random samples 148 Unnamed trib. to Kensico Res. 8 random samples 149 Waccabuc River at Boutonville 8 random samples

Table A.5.2: Macroinvertebrate sorting errors found by resorting a processed sample from QA/QC 2004.

Site Sample # ID # Number of macroinvertebrates % Missed Initial sort Resort 52 3 40023 171 20 10.47 124 4 40040 176 24 12.00 125 4 40048 151 13 7.93 127 1 40061 279 9 3.13 127 2 40062 210 13 5.83 132 2 40102 250 4 1.57 132 3 40103 231 25 9.77 137 2 40126 179 10 5.29 140 4 40152 157 15 8.72 141 1 40157 205 18 8.07 142 1 40165 151 32 17.49a 145 1 40181 204 18 8.11

Overall 8.20 a Only sample with >15% sorting error and the only sample processed by this individual.

- 215 - APPENDIX TO CHAPTER 5 (A.5) NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Table A.5.3: Errors in non-midge macroinvertebrate identifications from QA/QC 2004.

Site Sample # ID # % Incorrect 52 3 40023 0.00 124 4 40040 0.00 125 4 40048 0.00 127 1 40061 0.40 127 2 40062 1.00 132 2 40102 0.00 132 3 40103 0.90 137 2 40126 0.00 140 4 40152 0.00 141 1 40157 0.00 142 1 40165 0.00 145 1 40181 0.50

Overall 0.23

- 216 - APPENDIX TO CHAPTER 5 (A.5) NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Appendix to Chapter 6 - DOC and BDOC Dynamics ______

2004 QUALITY ASSURANCE/QUALITY CONTROL DATA FOR DOC/BDOC. Summary data for all FIELD QA/QC data

- Sample Type: SB=Summer Baseflow; ST=Stormflow - QAQC Sample Type: FB = Field Blank (ug/L), FD - Field Duplicate (Relative Percent Difference) - Method Detection Limit (MDL) = 50 ug/L; Precision QC Limit = 15% (applicable to RPD); Accuracy = 85 to 115%. ______

QAQC_Type Sample_Type Year DOC_Sample_Type Mean Stdev #obs #flag (ug/L)

FB SB 2004 DOC 24.5 12.7 10 0 FB SB 2004 DOC(day28) 41.6 17.9 10 0 FB ST 2004 DOC 295 81.3 3 3

FD SB 2004 DOC 2.26 1.70 12 0 FD SB 2004 DOC(day28) 1.97 2.93 12 0 FD ST 2004 DOC 1.45 1.18 2 0

ADDITIONAL DEFINITIONS AND NOTES: 1. Sample blank (field and laboratory) exceedances are assessed at concentrations > 2*MDL value 2. Precision assessed using the Relative Percent Difference (RPD) value. 3. #flag indicates the number of samples whose concentration exceeded the defined QC criterion of 2*MDL (Method Detection Limit - see above for values) ______Appendix source: \\StroudSAS\research\nywatershed\nywatershed2004\doc2004\DOC04_QAQC_APPENDIX_C.LST Produced on 26APR05:10:22

- 217 - APPENDIX TO CHAPTER 6 (A.6) NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

______

2004 QUALITY ASSURANCE/QUALITY CONTROL DATA FOR DOC/BDOC. Summary data for all LAB QA/QC data

- Lab QAQC samples include baseflow and stormflow sampling efforts - QAQC Sample Type: LB = LAB Blank (ug/L), LCS - Lab Control Standard (ug/L & %recovery) MS = Matrix Spike (ug/L), LD = Lab Duplicate (Relative Percent Diff. & Absolute Diff.) - Method Detection Limit (MDL) = 50 ug/L; Precision QC Limit = 15% (applicable to RPD); Accuracy = 85 to 115%. ______

--% Recovery-- --(LCS only)--

QAQC_Type Year Mean Stdev #obs #flag Mean Stdev (ug/L)

LB 2004 38.1 34.9 218 12

LCS 2004 2054 46.1 296 2 101 2.41

------RPD (%)------Abs. Diff. (ug/L)------

QAQC_Type Year Type* Mean Stdev #obs #flag Mean Stdev #obs #flag

LD 2004 KHP 0.66 0.87 296 0 13.6 18.4 296 11 LD 2004 NANO 41.4 47.4 215 136 13.0 19.7 217 16

ADDITIONAL DEFINITIONS AND NOTES: 1. Sample blank (field and laboratory) exceedances are assessed at concentrations > 2*MDL value 2. Precision assessed using the Relative Percent Difference (RPD) value. 3. #flag indicates the number of samples whose concentration exceeded the defined QC criterion of 2*MDL (Method Detection Limit - see above for values) * 'Type' for lab duplicates applies to the specific group of samples summarized

______Appendix source: \\StroudSAS\research\nywatershed\nywatershed2004\doc2004\DOC04_QAQC_APPENDIX_C.LST Produced on 26APR05:10:22

- 218 - APPENDIX TO CHAPTER 6 (A.6) NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Appendix to Chapter 7 – Nitrogen (N), Phosphorus (P), and Dissolved Organic Carbon (DOC) Spiraling

______

QUALITY ASSURANCE/QUALITY CONTROL DATA FOR NUTRIENT & CARBOHYDRATE SPIRALING - 2004.

------

METHOD DETECTION LIMITS (MDL), QUALITY CONTROL CRITERIA (AS EXCEEDANCE LIMITS):

CMPD MDL (2004) Precision (% RPD specific) Accuracy (%)

BR (mg/L) 0.01 10 90-110 ARAB (nM) 2 20 75-125 GLUC (nM) 2 20 75-125 NH4N (mg/L) 0.011 20 75-125 NO3N (mg/L) 0.02 20 75-125 SKN (mg/L) 0.1 20 75-125 TKN (mg/L) 0.1 20 75-125 SRP (mg/L) 0.003 20 75-125 TDP (mg/L) 0.01 20 75-125 TP (mg/L) 0.01 20 75-125

ADDITIONAL DEFINITIONS AND NOTES:

______Appendix source: \\StroudSAS\research\nywatershed\nywatershed2004\spiraling2004\INJECT_04_QAQC_APPENDIX_C.LST Produced on 27APR05:11:51

- 219 - APPENDIX TO CHAPTER 7 (A.7) NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

______

QUALITY ASSURANCE/QUALITY CONTROL DATA FOR NUTRIENT & CARBOHYDRATE SPIRALING - 2004. Summary data for: Field (FB) & Lab (LB) Blanks

______

QAQC Year Cmpd Mean Stdev Obs #flag

FB 2004 BR (mg/L) 0.01 0.06 57 4 FB 2004 ARAB (nM) 0.3 0.9 39 0 FB 2004 GLUC (nM) 4.1 5.6 39 16 FB 2004 NH4N (mg/L) 0.005 0.006 40 1 FB 2004 SRP (mg/L) 0.002 0.007 40 1 LB 2004 BR (mg/L) 0.001 0.005 84 1 LB 2004 ARAB (nM) 0.3 1.9 63 2 LB 2004 GLUC (nM) 1.4 3.1 63 11 LB 2004 NH4N (mg/L) 0.004 0.002 84 0 LB 2004 NO3N (mg/L) 0.005 0.003 51 0 LB 2004 SKN&TKN (mg/L) 0.01 0.02 38 0 LB 2004 SRP (mg/L) 0.0007 0.0010 90 0 LB 2004 TDP&TP (mg/L) 0.001 0.0010 45 0

Notes: (1) #flag indicates the number of samples whose concentration exceeded the defined QC criterion of 2*MDL (Method Detection Limit - see page A7.1 for values)

______Appendix source: \\StroudSAS\research\nywatershed\nywatershed2004\spiraling2004\INJECT_04_QAQC_APPENDIX_C.LST Produced on 27APR05:11:51

- 220 - APPENDIX TO CHAPTER 7 (A.7) NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

______

QUALITY ASSURANCE/QUALITY CONTROL DATA FOR NUTRIENT & CARBOHYDRATE SPIRALING - 2004. Summary data for: Field (FD) & Lab (LD) Duplicates

- Precision Evaluation Type: RPD=Relative Percent Difference; ABD DIFF=Absolute value of the difference

______

------RPD (%)------ABS DIFF------

QAQC Year Cmpd Mean Stdev Obs #flag Mean Stdev Obs #flag

FD 2004 BR (mg/L) 0.7 0.9 35 0 0.002 0.003 71 2 FD 2004 ARAB (nM) 3.6 4.2 31 0 0.7 2.5 46 4 FD 2004 GLUC (nM) 13 15 50 10 2.9 3.0 27 12 FD 2004 NH4N (mg/L) 4.7 3.3 7 0 0.001 0.001 71 0 FD 2004 SRP (mg/L) 7.4 8.5 40 5 0.0006 0.0009 38 1 LD 2004 BR (mg/L) 0.6 0.6 38 0 0.0007 0.0007 25 0 LD 2004 ARAB (nM) 2.3 2.0 33 0 4.3 . 1 1 LD 2004 GLUC (nM) 5.3 7.4 38 2 2.2 2.4 7 2 LD 2004 NH4N (mg/L) 1.6 1.0 6 0 0.001 0.002 50 0 LD 2004 NO3N (mg/L) 0.5 0.3 11 0 0.001 . 1 0 LD 2004 SKN&TKN (mg/L) 5.9 5.2 4 0 0.01 0.02 21 0 LD 2004 SRP (mg/L) 1.4 2.6 32 0 0.0002 0.0004 25 0 LD 2004 TDP&TP (mg/L) 2.0 1.8 4 0 0.0009 0.0009 21 0

Notes: (1) #flag indicates the number of samples whose concentration exceeded the defined QC criterion (see page A7.1 for values).

______Appendix source: \\StroudSAS\research\nywatershed\nywatershed2004\spiraling2004\INJECT_04_QAQC_APPENDIX_C.LST Produced on 27APR05:11:51

- 221 - APPENDIX TO CHAPTER 7 (A.7) NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

______

QUALITY ASSURANCE/QUALITY CONTROL DATA FOR NUTRIENT & CARBOHYDRATE SPIRALING - 2004. Summary data for: Matrix Spikes (as percent recovery)

______

Year Cmpd Mean Stdev Obs #flag

2004 BR (mg/L) 96 4.9 63 8 2004 ARAB (nM) 91 7.9 31 1 2004 GLUC (nM) 97 6.1 41 0 2004 NH4N (mg/L) 100 2.6 55 0 2004 NO3N (mg/L) 99 2.5 12 0 2004 SKN&TKN (mg/L) 101 6.1 25 0 2004 SRP (mg/L) 102 3.4 56 0 2004 TDP&TP (mg/L) 98 2.5 25 0

Notes: (1) #flag indicates the number of samples whose percent recovery exceeded the defined QC criterion (see page A7.1 for values).

______Appendix source: \\StroudSAS\research\nywatershed\nywatershed2004\spiralings2004\INJECT_04_QAQC_APPENDIX_C.LST

Produced on 27APR05:11:51

- 222 - APPENDIX TO CHAPTER 7 (A.7) NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

______

QUALITY ASSURANCE/QUALITY CONTROL DATA FOR NUTRIENT & CARBOHYDRATE SPIRALING - 2004. Summary data for: Laboratory Control Standards (as percent recovery)

______

Year Cmpd Mean Stdev Obs #flag

2004 BR (mg/L) 102 5.2 164 9 2004 ARAB (nM) 94 5.5 44 0 2004 GLUC (nM) 94 4.8 44 0 2004 NH4N (mg/L) 105 1.3 32 0 2004 NO3N (mg/L) 102 1.5 24 0 2004 SKN&TKN (mg/L) 99 3.4 23 0 2004 SRP (mg/L) 101 2.6 33 0 2004 TDP&TP (mg/L) 98 5.6 28 0

Notes: (1) #flag indicates the number of samples whose percent recovery exceeded the defined QC criterion (see page A7.1 for values).

______Appendix source: \\StroudSAS\research\nywatershed\nywatershed2004\spiralings2004\INJECT_04_QAQC_APPENDIX_C.LST

Produced on 27APR05:11:51

- 223 - APPENDIX TO CHAPTER 7 (A.7) NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

------Intentionally Blank ------

- 224 - APPENDIX TO CHAPTER 7 (A.7) NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Appendix to Chapter 8 - Stream Metabolism

Appendix Table 8.1: Summary of chlorophyll a concentrations and initial (before acidification) absorbances at 665 nm in samples of river periphyton, blanks and standards, Year 5 data, 2004.

Chlorophyll a (µg/sample) OD665B 5th 95th 5th 95th nx SDPercentile Percentile n x SD Percentile Percentile

Sample extracts 595 122.46 231.00 1.20 563.74 1320 0.345 0.454 0.019 1.430 Ext. 1 595 84.46 140.81 1.20 333.11 595 0.500 0.523 0.022 1.712 Ext. 2 438 33.86 114.34 1.24 146.15 438 0.207 0.356 0.016 1.105 Ext. 3 136 34.01 69.45 1.88 221.24 136 0.256 0.370 0.025 1.186 Ext. 4 66 27.47 43.20 1.92 149.96 66 0.261 0.342 0.037 1.156 Ext. 5 36 19.33 20.38 2.61 71.09 36 0.209 0.170 0.057 0.595 Ext. 6 22 15.21 10.80 5.01 33.52 22 0.170 0.091 0.067 0.309 Ext. 7 13 12.68 6.44 3.77 26.34 13 0.181 0.154 0.074 0.677 Ext. 8 9 11.24 6.83 4.17 25.50 9 0.130 0.051 0.070 0.243 Ext. 9 4 7.89 4.61 3.01 13.79 4 0.119 0.054 0.064 0.189 Ext. 10 1 11.35 11.35 11.35 1 0.092 0.092 0.092

Blanks 267 0.07 0.724 0.000 0.084 267 0.000 0.002 -0.002 0.003

Standards 147 34.49 3.63 28.85 39.86 147 0.878 0.089 0.734 0.983

All samples processed within 28 day holding period.

- 225 - APPENDIX TO CHAPTER 8 (A.8) NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Appendix Table 8.2: Reproducibility of preparing and quantifying chlorophyll in standards run with river periphyton chlorophyll analyses, Year 5 data, 2004.

Chlorophyll a Measured Relative % Percent Date of added Chlorophyll a difference measured / analysis LCS No. Vial # (µg/sample) (µg/sample) (Lab Dups 1&2) added

25-May-04 1 1 35.39 34.57 3.51 97.68 25-May-04 2 1 35.39 35.80 101.17 26-May-04 1 1 35.39 * * 26-May-04 2 1 35.39 35.80 101.17 27-May-04 1 1 35.39 34.75 2.00 98.18 27-May-04 2 1 35.39 35.45 100.16 28-May-04 1 1 35.39 * * 28-May-04 2 1 35.39 * * 1-Jun-04 1 2 39.36 39.42 2.07 100.16 1-Jun-04 2 2 39.36 40.25 102.25 2-Jun-04 1 2 39.36 39.28 99.81 2-Jun-04 2 2 39.36 * * 2-Jun-04 3 2 39.36 38.74 0.96 98.43 2-Jun-04 4 2 39.36 39.12 99.38 3-Jun-04 1 2 39.36 38.72 1.36 98.38 3-Jun-04 2 2 39.36 38.20 97.05 4-Jun-04 1 2 39.36 38.39 1.07 97.52 4-Jun-04 2 2 39.36 38.80 98.57 7-Jun-04 1 2 39.36 38.22 0.51 97.10 7-Jun-04 2 2 39.36 38.41 97.60 8-Jun-04 1 2 39.36 38.29 0.68 97.29 8-Jun-04 2 2 39.36 38.55 97.95 8-Jun-04 3 2 39.36 38.82 0.27 98.62 8-Jun-04 4 2 39.36 38.71 98.36 9-Jun-04 1 2 39.36 39.26 2.24 99.74 9-Jun-04 2 2 39.36 38.39 97.52 9-Jun-04 3 2 39.36 38.31 1.62 97.33 9-Jun-04 4 2 39.36 38.94 98.93 10-Jun-04 1 2 39.36 38.70 2.15 98.33 10-Jun-04 2 2 39.36 37.88 96.24 11-Jun-04 1 2 39.36 38.96 0.14 98.97 11-Jun-04 2 2 39.36 38.90 98.83 14-Jun-04 1 2 39.36 38.29 0.10 97.29 14-Jun-04 2 2 39.36 38.33 97.38 15-Jun-04 1 2 39.36 39.20 1.68 99.59 15-Jun-04 2 2 39.36 39.86 101.28 15-Jun-04 3 2 39.36 39.34 0.57 99.95 15-Jun-04 4 2 39.36 39.56 100.52 16-Jun-04 1 3 41.73 41.59 2.20 99.66 16-Jun-04 2 3 41.73 42.51 101.88

- 226 - APPENDIX TO CHAPTER 8 (A.8) NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Appendix Table 8.2: Continued. Chlorophyll a Measured Relative % Percent Date of added Chlorophyll a difference measured / analysis LCS No. Vial # (µg/sample) (µg/sample) (Lab Dups 1&2) added 17-Jun-04 1 3 41.73 41.37 0.16 99.15 17-Jun-04 2 3 41.73 41.44 99.31 18-Jun-04 1 3 41.73 41.40 1.08 99.22 18-Jun-04 2 3 41.73 41.85 100.29 22-Jun-04 1 4 28.91 * * 22-Jun-04 2 4 28.91 * * 23-Jun-04 1 4 28.91 * * 23-Jun-04 2 4 28.91 29.35 101.52 24-Jun-04 1 4 28.91 * * 24-Jun-04 2 4 28.91 * * 25-Jun-04 1 4 28.91 28.57 0.82 98.83 25-Jun-04 2 4 28.91 28.81 99.64 28-Jun-04 1 4 28.91 28.61 2.52 98.96 28-Jun-04 2 4 28.91 29.34 101.48 29-Jun-04 1 4 28.91 28.56 1.98 98.80 29-Jun-04 2 4 28.91 29.13 100.77 30-Jun-04 1 4 28.91 28.64 2.23 99.06 30-Jun-04 2 4 28.91 29.28 101.29 1-Jul-04 1 4 28.91 29.08 0.13 100.58 1-Jul-04 2 4 28.91 29.11 100.71 2-Jul-04 1 4 28.91 29.11 0.90 100.71 2-Jul-04 2 4 28.91 29.38 101.61 6-Jul-04 1 4 28.91 29.05 0.10 100.48 6-Jul-04 2 4 28.91 29.08 100.58 7-Jul-04 1 4 28.91 28.77 2.03 99.51 7-Jul-04 2 4 28.91 29.36 101.55 8-Jul-04 1 4 28.91 28.90 0.55 99.96 8-Jul-04 2 4 28.91 29.06 100.51 9-Jul-04 1 4 28.91 28.85 0.26 99.80 9-Jul-04 2 4 28.91 28.93 100.06 12-Jul-04 1 4 28.91 28.85 1.83 99.80 12-Jul-04 2 4 28.91 29.39 101.65 13-Jul-04 1 4 28.91 * * 13-Jul-04 2 4 28.91 29.21 101.03 14-Jul-04 1 4 28.91 29.11 100.71 14-Jul-04 2 4 28.91 * * 15-Jul-04 1 4 28.91 29.00 1.03 100.32 15-Jul-04 2 4 28.91 29.30 101.35 20-Jul-04 1 4 28.91 28.94 0.65 100.09 20-Jul-04 2 4 28.91 28.75 99.44 21-Jul-04 1 4 28.91 28.97 0.32 100.22 21-Jul-04 2 4 28.91 29.07 100.54

- 227 - APPENDIX TO CHAPTER 8 (A.8) NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Appendix Table 8.2: Continued. Chlorophyll a Measured Relative % Percent Date of added Chlorophyll a difference measured / analysis LCS No. Vial # (µg/sample) (µg/sample) (Lab Dups 1&2) added 22-Jul-04 1 4 28.91 31.50 0.24 108.96 22-Jul-04 2 4 28.91 31.57 109.22 26-Jul-04 1 5 33.63 33.79 0.55 100.48 26-Jul-04 2 5 33.63 33.98 101.04 27-Jul-04 1 5 33.63 33.05 1.91 98.28 27-Jul-04 2 5 33.63 33.69 100.18 3-Aug-04 1 5 33.63 32.79 0.54 97.51 3-Aug-04 2 5 33.63 32.97 98.03 4-Aug-04 1 5 33.63 34.00 1.12 101.09 4-Aug-04 2 5 33.63 34.38 102.23 6-Aug-04 1 6 34.09 33.51 3.13 98.30 6-Aug-04 2 6 34.09 34.58 101.43 10-Aug-04 1 6 34.09 33.91 1.34 99.48 10-Aug-04 2 6 34.09 34.37 100.83 17-Aug-04 1 6 34.09 34.13 0.16 100.11 17-Aug-04 2 6 34.09 34.19 100.28 18-Aug-04 1 6 34.09 34.05 0.03 99.89 18-Aug-04 2 6 34.09 34.06 99.92 19-Aug-04 1 6 34.09 34.13 4.08 100.11 19-Aug-04 2 6 34.09 35.55 104.29 20-Aug-04 1 6 34.09 32.61 2.04 95.67 20-Aug-04 2 6 34.09 33.29 97.64 21-Aug-04 1 6 34.09 34.30 1.32 100.61 21-Aug-04 2 6 34.09 33.85 99.29 23-Aug-04 1 6 34.09 34.65 101.65 23-Aug-04 2 6 34.09 * * 24-Aug-04 1 7 35.07 * * 24-Aug-04 2 7 35.07 36.19 103.19 25-Aug-04 1 7 35.07 34.10 2.36 97.24 25-Aug-04 2 7 35.07 34.91 99.56 26-Aug-04 1 7 35.07 35.27 0.69 100.57 26-Aug-04 2 7 35.07 35.51 101.26 3-Sep-04 1 7 35.07 36.01 0.42 102.68 3-Sep-04 2 7 35.07 35.86 102.25 9-Sep-04 1 7 35.07 35.90 1.36 102.36 9-Sep-04 2 7 35.07 35.41 100.97 10-Sep-04 1 7 35.07 35.76 0.13 101.96 10-Sep-04 2 7 35.07 35.71 101.82 11-Sep-04 1 7 35.07 35.60 0.34 101.50 11-Sep-04 2 7 35.07 35.72 101.85 12-Sep-04 1 7 35.07 35.65 1.07 101.66 12-Sep-04 2 7 35.07 36.04 102.76

- 228 - APPENDIX TO CHAPTER 8 (A.8) NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Appendix Table 8.2: Continued. Chlorophyll a Measured Relative % Percent Date of added Chlorophyll a difference measured / analysis LCS No. Vial # (µg/sample) (µg/sample) (Lab Dups 1&2) added 13-Sep-04 1 7 35.07 36.83 1.02 105.03 13-Sep-04 2 7 35.07 36.46 103.96 16-Sep-04 1 7 35.07 35.59 0.60 101.48 16-Sep-04 2 7 35.07 35.80 102.09 24-Sep-04 1 7 35.07 35.89 0.13 102.33 24-Sep-04 2 7 35.07 35.84 102.20 25-Sep-04 1 7 35.07 35.27 0.35 100.57 25-Sep-04 2 7 35.07 35.15 100.22 26-Sep-04 1 7 35.07 35.88 1.87 102.30 26-Sep-04 2 7 35.07 35.21 100.41 27-Sep-04 1 7 35.07 35.21 2.60 100.41 27-Sep-04 2 7 35.07 36.14 103.05 30-Sep-04 1 7 35.07 35.30 0.50 100.65 30-Sep-04 2 7 35.07 35.12 100.14 7-Oct-04 1 7 35.07 35.65 0.42 101.66 7-Oct-04 2 7 35.07 35.80 102.09 26-Oct-04 1 7 35.07 37.11 3.75 105.83 26-Oct-04 2 7 35.07 35.75 101.93 27-Oct-04 1 8 34.06 * * 27-Oct-04 2 8 34.06 * * 28-Oct-04 1 8 34.06 * * 28-Oct-04 2 8 34.06 * * 29-Oct-04 1 8 34.06 * * 29-Oct-04 2 8 34.06 33.76 99.13 1-Nov-04 1 8 34.06 34.36 100.89 1-Nov-04 2 8 34.06 * * 2-Nov-04 1 8 34.06 * * 2-Nov-04 2 8 34.06 * * 3-Nov-04 1 8 34.06 33.08 97.13 3-Nov-04 2 9 33.42 33.66 100.72 4-Nov-04 1 9 33.42 33.53 1.40 100.33 4-Nov-04 2 9 33.42 33.06 98.93 5-Nov-04 1 9 33.42 33.59 0.11 100.50 5-Nov-04 2 9 33.42 33.55 100.39 8-Nov-04 1 9 33.42 33.61 100.58 8-Nov-04 2 10 32.77 32.01 97.69 9-Nov-04 1 10 32.77 33.02 0.85 100.75 9-Nov-04 2 10 32.77 33.30 101.61 10-Nov-04 1 10 32.77 32.44 1.09 99.01 10-Nov-04 2 10 32.77 32.80 100.09 11-Nov-04 1 10 32.77 32.40 1.49 98.87 11-Nov-04 2 10 32.77 32.88 100.35 15-Nov-04 1 10 32.77 33.10 0.42 101.01 15-Nov-04 2 10 32.77 33.24 101.43 Grand Mean 34.49 1.18 100.30 SD 3.63 0.96 2.00 n 147 67 147 * Sample values rejected due to condensation on cuvette at time of reading.

- 229 - APPENDIX TO CHAPTER 8 (A.8) NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Appendix Table 8.3: Precision of chlorophyll a determinations in samples of river periphyton assessed from the relative % difference of lab duplicates, Year 5, 2004.

Relative % Chlorophyll a Average Lab difference Stream Tag No. Lab Dup. (µg/sample) Dups Lab Dups

Haviland Hollow 46008 1 383.47 359.23 13.49 5/18/2004 46008 2 334.99 46015 1 64.07 56.27 27.72 * 46015 2 48.47

Titicus River 46041 1 230.67 188.25 45.07 * 6/8/2004 46041 2 145.83 46046 1 45.83 61.67 51.37 * 46046 2 77.50 46047 1 50.64 59.62 30.13 * 46047 2 68.60

Cross River 46055 1 57.70 53.93 13.98 6/15/2004 46055 2 50.16 46059 1 76.10 50.08 103.92 * 46059 2 24.06 46064 1 7.10 5.05 80.95 * 46064 2 3.01

Muscoot near 46067 1 33.88 22.25 104.50 * Baldwin Place 46067 2 10.63 6/22/2004 46072 1 72.61 164.45 111.69 * 46072 2 256.29 46077 1 41.42 37.97 18.16 46077 2 34.52

Muscoot below 46080 1 14.07 12.01 34.39 * Amawalk Res. 46080 2 9.94 6/29/2004 46083 1 7.66 8.36 16.79 46083 2 9.06 46090 1 40.50 61.57 68.45 * 46090 2 82.64

Neversink River 46133 1 43.26 42.06 5.72 near Claryville 46133 2 40.86 8/10/2004 46136 1 99.36 85.14 33.39 * 46136 2 70.93 46139 1 45.59 52.42 26.08 * 46139 2 59.26

- 230 - APPENDIX TO CHAPTER 8 (A.8) NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Appendix Table 8.3: Continued.

Relative % Chlorophyll a Average Lab difference Stream Tag No. Lab Dup. (µg/sample) Dups Lab Dups

West Br. Delaware 46154 1 117.00 137.06 29.28 * Trout Creek 46154 2 157.13 8/17/2004 46155 1 68.88 48.80 82.33 * 46155 2 28.71 46158 1 21.73 14.59 97.80 * 46158 2 7.46

East Br. Delaware 46165 1 22.57 37.53 79.70 * Tremper Kill 46165 2 52.48 8/31/2004 46167 1 139.29 113.49 45.47 * 46167 2 87.69 46173 1 282.07 314.12 20.41 * 46173 2 346.18

East Br. Delaware 46182 1 22.53 14.82 104.20 * near Arkville 46182 2 7.10 10/5/2004 46181 1 22.01 32.88 66.10 * 46181 2 43.74

West Br. Delaware 46204 1 178.10 156.75 27.24 * at South Kortright 46204 2 135.40 10/12/2004 46206 1 109.02 112.53 6.24 46206 2 116.03

46207 1 53.45 59.10 19.13 46207 2 64.75

Mean 48.70 SD 34.30

* Exceeds 20% limit for RPD.

- 231 - APPENDIX TO CHAPTER 8 (A.8) NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Days held Cover type rock % Left on in scrapings in % Recovered % Recovered Total in study streams, Year 5 data, 2004 (µg/cm2) Chlorophyll (cm2) Rock area Rock 77.52 49.13 23.97 98.04 1.96 Filamentous Algae FA 28 (µg) Total Total Chlorophyll Chlorophyll Chlorophyll (µg on rock) on (µg Recovery of chlorophyll from rocks Chlorophyll Chlorophyll (µg in scrapings) (µg Appendix Table 8.4: Cross Cross Cross Neversink Muscoot (Whitehall) Muscoot (Whitehall) Titicus Neversink Haviland Hollow Titicus 210.88 Trout Creek 160.25 Neversink 17.61 94.44 140.43 Cross 174.28 Trout Creek 128.57 E.Br.Delaware (Arkville) 167.63 48.46 (Whitehall) Muscoot 385.16 234.69 101.31 155.62 43.49 Neversink 288.82 Tremper Kill 46.02 66.07 375.12 195.75 122.32 101.82 199.09 Trout Creek 55.60 43.73 120.95 Muscoot (Baldwin) 33.41 8.37 386.24 59.82 67.71 24.05 289.95 W.Br.Delaware (S.Kortright) 5.19 57.67 352.31 Neversink 276.57 33.15 90.88 676.68 6.27 2.75 2.89 76.90 49.11 Tremper Kill 20.05 54.75 375.99 157.71 E.Br.Delaware (Arkville) 59.74 55.48 134.96 289.97 Muscoot (Baldwin) 60.51 14.93 136.11 5.90 358.61 316.72 685.34 63.78 E.Br.Delaware 26.65 37.44(Arkville) 48.25 543.95 45.25 268.06 4.63 50.50 344.47 50.61 Haviland Hollow Green 44.52 61.14 1.27 488.42 72.60 324.46 W.Br.Delaware (S.Kortright) 57.11 Tufts 27.24 944.75 73.35 57.81 62.56 51.75 90.61 E.Br.Delaware (Arkville) 66.45 118.36 2.67 5.73 Green Green 61.05 Green 56.27 437.13 100.34 Tremper Kill 47.91 70.57 451.93 53.70 9.52 2.34 56.55 275.93 337.10 157.41 Tremper Kill 42.19 35.82 803.70 449.22 383.17 37.29 8.00Trout Creek Diatoms 43.73 1.52 75.44 68.66 G 17.59 209.20 415.04 W.Br.Delaware (S.Kortright) 43.45 Black 549.07 65.50 66.55 8.34 68.57 71.01 56.41 360.28 11.72 64.74 Green 37.69 25.13 16.32 G 35.69 24.56 31.34 G G 72.13 1154.42 79.43 71.63 12.27 51.31 6.75 Black Green 6.79 31.43 28.99 108.68 14.97 516.67 313.62 362.23 44.72 173.73 86.01 Green Diatoms 11.63 27.87 491.62 20.57 28.37 63.96 7.02 35.47 Diatoms 565.36 55.05 85.27 30.23 23.09 224.17 48.99 Black 79.83 Tufts 82.66 593.79 13.99 G 302.74 83.00 Green 1.70 14.57 53.51 11.98 5.70 11 29.14 14.73 3.55 D 40.01 90.06 G 20.17 17.34 27.26 Diatoms T Black G 17.00 Green 4.19 12.03 14.84 Diatoms 520.76 35 87.47 92.33 93.06 9.94 87.98 90.61 592.63 35 Diatoms 48.80 BK G 314.77 24 24 24 12.53 D 93.32 7.67 92.47 6.94 49.96 12.02 9.39 Moss Green Diatoms Filamentous Algae 10.67 D 34.85 7 G Diatoms BK 11.86 6.68 7.53 9 FA 9.03 Diatoms Blue Green Algae 24 94.41 D BK 95.40 28 27 D 24 G 96.18 T 5.59 9 24 4.60 D Diatoms 35 BK Diatoms 24 3.82 M 28 D Filamentous Algae D BG 23 15 30 27 D FA 28 28 7 30 31 D 23 28 D 28 27 30 30 Stream

- 232 - APPENDIX TO CHAPTER 8 (A.8) NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Days held Cover type Cover rock % Left on % Left on in scrapings % Recovered % Recovered Total Total (µg/cm2) Chlorophyll Chlorophyll (cm2) Rock area Rock (µg) Total Total Chlorophyll Chlorophyll Chlorophyll (µg on rock) on (µg Chlorophyll Chlorophyll (µg in(µg scrapings) Appendix Table 8.4: Continued. First Visit Titicus Titicus W.Br.Delaware (S.Kortright) W.Br.Delaware (S.Kortright) W.Br.Delaware (S.Kortright) 132.87 (Arkville) E.Br.Delaware 543.74 (Arkville) E.Br.Delaware 212.39 (S.Kortright) W.Br.Delaware (Arkville) E.Br.Delaware 130.87 445.80 (Arkville) E.Br.Delaware 151.29 446.87 41.93 434.64 44.91 263.74 152.10 695.04 285.28 91.42 43.53 257.30 253.28 81.79 81.79 58.80 90.31 209.47 44.45 537.22 6.06 50.25 528.67 516.43 7.70 100.73 361.57 20.85 67.00 5.79 54.73 335.54 52.66 32.13 50.38 48.28 274.13 78.23 8.02 79.80 9.66 9.81 82.55 3.13 49.62 7.49 34.92 Blue 21.77 Green Algae 4.20 82.98 Diatoms 17.45 84.53 7.85 84.16 41.63 42.07 Diatoms 17.02 85.02 15.47 Green 15.84 58.37 57.93 92.39 Green Diatoms Bare Filamentous Algae 14.98 Diatoms BG 7.61 FA Green G D G D 21 B G D D 28 16 21 23 23 28 16 23 23 Stream

- 233 - APPENDIX TO CHAPTER 8 (A.8) NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Appendix Table 8.5: Accuracy of determining chlorophyll a in samples of river periphyton assessed from recovery of chlorophyll from spiked samples, Year 5 data, 2004.

Matrix Chlorophyll a Average Matrix spike Stream Tag No. Lab Dup. Spike (µg/sample) Lab Dups (µg/sample) % Recovery

Haviland Hollow 46008 1 383.47 359.23 18-May-04 2 334.99 3 1 665.82 354.26 86.54 46015 1 64.07 56.27 248.47 3 2 192.70 166.13 82.12 Titicus 46025 1 41.46 37.89 25-May-04 2 34.32 3 3 120.37 82.15 100.40 46027 1 59.54 39.57 219.61 3 4 131.11 85.80 106.69 46030 1 208.61 191.65 2174.69 3 5 574.72 415.53 92.19 Titicus 46041 1 230.67 188.25 8-Jun-04 2 145.83 3 6 590.24 408.14 98.49 46046 1 45.83 61.67 277.50 3 7 208.01 133.70 109.46 46047 1 50.64 59.62 268.60 3 8 231.47 129.26 132.94 Cross 46055 1 57.70 53.93 15-Jun-04 2 50.16 3 9 188.33 119.79 112.20 46059 1 76.10 50.08 224.06 3 10 138.01 108.58 80.98 46064 1 7.10 5.05 23.01 3 11 13.91 10.95 80.90 Muscoot (Baldwin) 46067 1 33.88 22.25 22-Jun-04 2 10.63 3 12 63.95 48.25 86.43 46072 1 72.61 164.45 2256.29 3 13 476.69 356.54 87.57 46077 1 41.42 37.97 234.52 3 14 109.06 82.32 86.35 Muscoot (Whitehall) 46080 1 14.07 12.01 29-Jun-04 2 9.94 3 15 93.98 84.79 96.68 46083 1 7.66 8.36 29.06 3 16 28.19 21.08 94.04 46090 1 40.50 61.57 282.64 3 17 181.43 155.28 77.19

- 234 - APPENDIX TO CHAPTER 8 (A.8) NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Appendix Table 8.5: Continued.

Matrix Chlorophyll a Average Matrix spike Stream Tag No. Lab Dup. Spike (µg/sample) Lab Dups (µg/sample) % Recovery

E. Branch Delaware 46097 1 334.91 337.16 (Arkville) 2 339.40 14-Jul-04 3 18 830.13 850.35 57.97 46100 1 247.11 176.94 2 106.77 319617.66 446.26 98.76 46110 1 23.42 17.70 2 11.99 320 62.75 44.65 100.90

W. Branch Delaware 46119 1 263.62 188.91 (South Kortright) 2 114.19 27-Jul-04 3 21 633.94 476.44 93.41 46127 1 72.41 46.29 2 20.17 322159.14 116.75 96.66 46131 1 283.23 238.63 2 194.02 323824.51 601.84 97.35 Neversink 46133 1 43.26 42.06 10-Aug-04 2 40.86 324154.93 110.62 102.03 46136 1 99.36 85.14 2 70.93 3 25 302.60 223.93 97.11 46139 1 45.59 52.42 2 59.26 3 26 201.00 137.88 107.75 Trout Creek 46154 1 117.00 137.06 17-Aug-04 2 157.13 3 27 507.52 360.49 102.76 46155 1 68.88 48.80 2 28.71 3 28 135.24 128.34 67.36 46158 1 21.73 14.59 27.46 3 29 37.17 38.39 58.81 Tremper Kill 46165 1 22.57 37.53 31-Aug-04 2 52.48 3 30 125.66 95.88 91.91 46167 1 139.29 113.49 2 87.69 3 31 401.95 289.96 99.48 46173 1 282.07 314.12 2 346.18 3 32 965.09 802.57 81.11 E. Branch Delaware 46182 1 22.53 14.82 (Arkville) 27.10 5-Sep-04 3 33 144.74 140.86 92.24 46181 1 22.01 32.88 2 43.74 3 34 361.66 324.82 101.22

- 235 - APPENDIX TO CHAPTER 8 (A.8) NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Appendix Table 8.5: Continued.

Matrix Chlorophyll a Average Matrix spike Stream Tag No. Lab Dup. Spike (µg/sample) Lab Dups (µg/sample) % Recovery W. Branch Delaware 46204 1 178.10 156.75 (South Kortright) 2 135.40 12-Oct-04 3 35 506.12 392.87 88.93 46206 1 109.02 112.53 2 116.03 3 36 384.87 282.03 96.57 46207 1 53.45 59.10 2 64.75 3 37 211.14 148.13 102.64

Mean 93.14 SD 14.25

- 236 - APPENDIX TO CHAPTER 8 (A.8) NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Appendix Table 8.6: Relative percent difference in propane concentrations on each stream, Year 5, 2004. The measure on field duplicate samples provides an estimate of precision of the measures.

Propane (% ) Field Dup Field Dup Relative percent Stream Date Substation 1 2 difference (%) Haviland Hollow 19-May-04 2 63.82 62.48 2.1 4 11.15 11.35 1.8 5 6.58 5.83 12.0 Titicus 9-Jun-04 2 39.41 38.83 1.5 4 10.07 10.49 4.1 5 6.48 6.48 0.0 Cross 16-Jun-04 2 69.34 66.97 3.5 4 24.19 22.87 5.6 5 16.52 16.29 1.4 Muscoot (Baldwin) 23-Jun-04 2 75.25 70.22 6.9 4 39.54 37.17 6.2 5 25.54 23.38 8.8 Muscoot (Whitehall) 30-Jun-04 2 66.52 64.27 3.4 4 30.64 30.87 0.7 5 12.26 12.69 3.5 Neversink 11-Aug-04 2 48.01 45.91 4.5 4 15.78 13.27 17.2 5 3.95 4.00 1.2 Trout Creek 18-Aug-04 2 74.51 69.41 7.1 4RDFRDF- 5 33.18 33.44 0.8 Tremper Kill 1-Sep-04 2 69.15 72.42 4.6 4 41.18 41.70 1.3 5 26.94 29.31 8.4 E. Br. Delaware 6-Oct-04 2 70.27 78.03 10.5 (Arkville) 4 67.77 69.35 2.3 5 51.72 54.06 4.4 W. Br. Delaware 13-Oct-04 2 76.54 77.02 0.6 (S. Kortright) 4 35.17 36.17 2.8 5 24.13 22.78 5.8 Mean 4.6 SD 3.9 Median 3.5 RDF - Reject, damaged in field, sample not analyzed.

- 237 - APPENDIX TO CHAPTER 8 (A.8) NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Reg. R2 Lin. ±SD 100% x 2004. , ±SD 50% x Year 5 , Standards ±SD 10% x and for standards , ) 3 ±SD 1% x for blanks , ±SD x Blanks Propane peak area (x10 peak Propane ±SD x Subst. 5 les at different substations p ±SD x Subst. 3 Samples eak areas for sam p ±SD x Subst. 1 ane p Pro endix Table 8.7: pp A Haviland HollowHaviland Titicus Cross 53.1 (Baldwin)Muscoot ± 1.4 (Whitehall)Muscoot 16.3 ± 7.3Neversink 1.1 ± 63.2 54.9 0.3 ± ± 3.5 Trout Creek 2.3 6.1 22.1 ± ± 59.0 2.7 11.5 0.5 1.2 KillTremper ± ± 32.0 ± 0.7 0.1 ± 0.3 1.0 1.3 13.3 DelawareE. Br. (Arkville) ± ± 21.2 0.9 0.5 ± 1.6 5.7 15.7 0.7 ± Br. Delaware (S. Kortright)W. ± ± 0.1 0.2 0.6 3.6 15.1 8.7 5.3 19.3 reject areaintegration below peak value. # ± ± 11.4 ± ± ± 0.7 1.2 1.2 ± 1.5 1.0 0.4 3.4 16.3 0.4 ± # ± ± 0.3 ± 0.5 0.6 7.2 11.1 # 13.3 1.8 6.2 ± ± 2.6 ± ± 0.5 ± 1.0 1.3 ± 0.2 0.5 4.5 10.0 0.4 0.5 ± ± ± 0.2 2.9 0.4 0.3 27.0 ± 2.8 6.6 ± 0.6 3.3 6.8 4.7 ± ± 0.5 2.1 ± ± 0.8 ± 0.1 7.8 0.2 ± 0.2 5.8 0.3 ± 0.1 7.4 ± 1.1 53.0 ± 0.4 4.6 1.8 0.1 ± 7.4 1.8 2.5 ± ± 0.9 ± ± 0.0 ± 0.5 4.6 27.1 0.4 0.5 1.9 0.4 ± ± 0.999 31.8 ± 0.3 0.2 ± 0.6 12.7 2.3 1.1 29.9 2.9 2.0 ± ± ± ± ± 0.5 0.7 54.0 0.8 0.1 7.6 2.1 0.2 ± ± 63.1 ± 0.6 0.3 0.2 ± 3.7 22.7 0.5 58.6 4.6 2.9 0.998 ± ± 0.998 ± ± ± 0.3 0.5 0.9 1.000 0.4 3.4 0.3 ± 1.000 1.000 0.1 10.6 8.5 4.6 ± ± ± 0.7 0.2 9.5 0.7 ± 0.4 20.6 14.8 ± ± 7.6 0.6 0.3 ± 17.3 0.3 ± 0.995 1.000 1.0 0.992 1.000 Stream

- 238 - APPENDIX TO CHAPTER 8 (A.8) NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

for each % Difference ainst winklers mgDO/L mgDO/L Abs. Difference, Difference, Abs. QAQC sondes and data ag % Difference n n Deployed - QAQC SondeDeployed - Winkler Deployed - mgDO/L mgDO/L Abs. Difference, Summary of daily QAQC checks of data sondes against StreamHaviland HollowTiticus 18-21May04 DatesCross 0.078 0.042Muscoot (Baldwin)Muscoot (Whitehall) 0.837 Mean 22-25Jun04 8-11Jun04 0.445 29Jun-2Jul04 SDNeversink 0.073 0.058 12 15-18Jun04 0.163 0.046 0.030Trout Creek 0.073 Mean 0.065 0.983 0.816 0.052 SD 0.646Tremper Kill 0.046 0.526 1.661 0.341 10-12Aug04 0.674 0.843Delaware E. Br. 10.456 12 12 0.127 17-20Aug04 0.585 0.441 16 31Aug-3Sep04W. Br. Delaware 0.043 0.087 Mean 12 Kortright) (S. 0.142 0.740 6 0.962 0.025 0.071 1.133 1.329 0.139 5-8Oct04 0.057 SDGrand Means 0.222 12-15Oct04 0.900 0.897 0.443 0.110SD 10.222 Mean 8.214 0.127 0.110 1.375 0.256 11.534 6 1.552 0.062 0.732 0.525 1.778 0.084 10.170 16 SD 6 6 1.293 16 0.991 8 0.535 0.942 0.549 6 0.035 1.093 0.705 1.195 0.057 16 16 0.093 5.700 10.866 0.355 0.099 1.058 11.086 0.509 0.809 0.129 0.758 2 0.024 8 10.050 8 0.068 0.801 8.170 1.005 0.216 8 8 0.667 0.935 0.215 9.674 1.731 Appendix Table 8.8: stream, Year 5, 2004.

- 239 - APPENDIX TO CHAPTER 8 (A.8) NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

------Intentionally Blank ------

- 240 - APPENDIX TO CHAPTER 8 (A.8) NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Appendix to Chapter 9 - Reservoir Primary Productivity th 95 Percentile th 5 Percentile b R ) fluorescence in for a represents a fore acidification nx SD Fluorometer for Chlorophyll n th 95 Percentile th 5 Percentile concentrations and initial (i.e., be (µg/sample) a Chlorophyll mples and blanks, Year 5 data. nx SD 121 13.76 11.99 3.13 34.45 242 160.00 65.96 63.5 261 Summary of chlorophyll a in extractsinextract 3 addedtotal2 andthe 1 to generate to the amount chlorophyll a a represents each initial absorbance/fluorescenceThe reading. b for R for n The The Sample extracts Sample Ext. 1Ext. 2Ext. 3Lab Blanks 121Field Blanks 106 10.04 15 31 4.02 chlorophyll the 10.22 8 1.59 0.004 2.81 2.25 0.482 0.008 0.37 1.26 28.58 0.398 0.000 1.03 9.36 0.095 0.013 2.12 121 1.254 148.60 106 31 70.15 15 183.13 16 0.512 53.73 88.53 51.3 25.786 1.115 105.0 20.58 29.259 254 0.040 4.943 280 57.3 1.338 78.500 118 that sample. Therefore, thatfinal sample. the n cannot exceedthefor extract n 1. a Appendix Table 9.1: fluorometrically-read reservoir sa reservoir fluorometrically-read

- 241 - APPENDIX TO CHAPTER 9 (A.9) NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Appendix Table 9.2: Means and relative percent differences from the mean fluorometric readings for lab quality standards run with reservoir chlorophyll analyses, Year 5 data, 2004.

Primary Standard Solid Secondary Standards Analysis Chlorophyll a RPD from High Range RPD from Low Range RPD from Date u g/L mean u g/L mean u g/L mean 29-Jul-04 180 1.61 55.1 1.94 10.10 0.50 54.9 1.58 10.10 0.50 30-Jul-04 171 3.52 54.1 0.11 10.10 0.50 54.5 0.85 10.10 0.50 11-Aug-04 175 1.21 54.0 0.07 10.10 0.50 54.0 0.07 10.10 0.50 12-Aug-04 173 2.36 55.0 1.76 10.10 0.50 55.2 2.12 10.10 0.50 26-Aug-04 175 1.21 53.1 1.75 9.82 2.32 54.1 0.11 9.90 1.50 17-Sep-04 174 1.78 54.0 0.07 10.10 0.50 54.2 0.30 10.10 0.50 20-Sep-04 176 0.64 54.3 0.48 10.10 0.50 9-Oct-04 164 7.70 53.6 0.82 9.99 0.60 53.8 0.45 10.10 0.50 11-Oct-04 175 1.21 54.1 0.11 10.10 0.50 53.8 0.45 10.00 0.50 12-Oct-04 194 9.09 54.8 1.40 10.10 0.50 54.1 0.11 10.10 0.50 14-Oct-04 175 1.21 54.1 0.11 10.10 0.50 54.0 0.07 9.99 0.60 16-Oct-04 178 0.49 53.6 0.82 10.00 0.50 53.8 0.45 9.98 0.70 18-Oct-04 180 1.61 53.9 0.26 10.00 0.50 53.8 0.45 9.98 0.70 19-Oct-04 179 1.05 53.8 0.45 10.00 0.50 54.1 0.11 9.99 0.60 22-Oct-04 188 5.96 52.8 2.32 10.10 0.50 53.1 1.75 10.00 0.50 25-Oct-04 177 0.07 54.1 0.11 10.10 0.50 53.4 1.19 10.00 0.50 Mean 177.125 2.544 54.04 0.73 10.05 0.61 SD 6.73 2.68 0.56 0.73 0.07 0.37

- 242 - APPENDIX TO CHAPTER 9 (A.9) NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Appendix Table 9.3: Precision of chlorophyll a determinations in reservoir samples assessed from the relative % difference of field duplicates by fluorometer.

Relative % Reservoir Tag Field Chlorophyll a difference Reservoir Date Station No. Dup. (µg/L) Field Dups Amawalk 8-Jul-04 1 46400 1 14.37 13.84 46401 2 16.51 3 46413 1 8.12 8.13 46414 2 7.49 Muscoot 21-Jul-04 1 46435 1 5.41 50.49 * 46436 2 3.23 3 46448 1 4.76 49.53 * 46449 2 2.87 Neversink 4-Aug-04 1 46470 1 1.62 11.02 46471 2 1.81 3 46483 1 2.06 5.65 46484 2 2.18 Cannonsville 26-Aug-04 3 46505 1 7.13 2.43 46506 2 6.96 5 46518 1 5.87 23.10 * 46519 2 4.65 Pepacton 15-Sep-04 1 46541 1 2.83 5.43 46542 2 2.99 5 46554 1 3.89 6.06 46555 2 3.66 16-Sep-04 3 46577 1 4.80 11.10 46578 2 5.36 7 46590 1 5.27 19.77 46591 2 4.32 Titicus 22-Sep-04 1 46613 1 25.13 14.40 46614 2 29.03 3 46626 1 13.29 34.22 * 46627 2 9.41 Cross River 1-Oct-04 1 46649 1 15.59 66.79 * 46650 2 7.78 3 46662 1 7.96 13.11 46663 2 9.07 Mean 20.94 SD 19.20 Median 13.48 * Exceeds 20% limit for RPD.

- 243 - APPENDIX TO CHAPTER 9 (A.9) NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

ter from the same location. ) quality* GPP/h CR/h -1 h -3. m . SD DO (g licate bottles filled with wa p n 0679 0.0012 -0.0038 b 0.89 200.00 Volumetric GPP/hVolumetric % Difference Relative )Mea -1 h -3. e associated with re e associated m . g 0596 0.0586 0.0014 0.0065 b 21.61 200.00 0.06880.0711 b b 0.05760.0735 b b 0.01180.02210.03990.03430.0528 0.0258 0.06460.05920.0405 0.0355 0.0678 0.0567 b b b chan 2 0 dup 0.0721 0.0728 0.0009 0 b 0 dup 0.0120 0.0170 0.00710 dup 0.0334 -0.014 0.0296 0.0054 0 dup 0.0356 -0.0077 0.0356 0.0001 -0.0168 2.0 dup 0.0636 0.0673 0.0053 0 b 1.5 dup 0.0366 0.0354 0.00161.5 dup 0.0557 -0.017 0.0574 0.0024 -0.0147 b 1.25 dup 0.0638 0.0603 0.0050 -0.0171 in dissolved O in dissolved y Variabilit Reservoir Depth GPP/h CR/h Volumetric Data and Subsite Date (m) (g DO Pepacton (1)Pepacton 15-Sep-04 (5)Pepacton 0 (3)Pepacton 0.0556 16-Sep-04 (7)Pepacton 0.0542 1.5 0 0.0020 0.0661 0.0295 -0.0226 0.0653 0.0350 1.25 0.0011 0.0078 0.0705 -0.0249 58.71 -0.011 0.0692 0.0019 b 98.35 12.88 -0.0209 1.70 51.52 b 41.73 13.75 20.00 Neversink (1)Neversink 4-Aug-04 (3)Neversink 0 0.0078 0.0098 1.5 0.0028 0.0374 -0.0137 0.0387 0.0018 -0.0187 53.73 2.17 8.91 9.52 Cannonsville (5)Cannonsville 2.0 0.0671 0. Cannonsville (3)Cannonsville 26-Aug-04 0 0. endix Table 9.4: pp A

- 244 - APPENDIX TO CHAPTER 9 (A.9) NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

) quality* GPP/h CR/h -1 ·h -3 m SD DO· (g n 91 0.0071 -0.0545 a 3.70 19.98 Volumetric GPP/hVolumetric % Difference Relative 0.1548 0.0091 -0.0283 b 6.92 1.75 )Mea -1 ·h -3 m 0.04640.04500.07330.0552 0.10260.09030.12970.13750.16120.1353 c c c a c b a b b 0.17410.17960.0465 0.0392 0.08900.0883 a b b b 0 dup 0.0558 0.0504 0.00760 dup 0.1065 -0.0471 0.0984 c 0.01140 dup 0.1535 -0.051 0.1444 0.0129 a -0.0288 b 0 dup 0.0349 0.0370 0.0030 -0.0205 1.0 dup 0.1656 0.1726 0.00991.0 dup 0.0813 -0.0446 0.0848 b 0.0049 -0.0293 b 1.25 dup 0.0475 0.0514 0.00551.25 dup 0.1357 -0.0314 0.1366 0.0012 -0.0528 a > 0.2 mg/L for NPP & CR. b: Change in dissolved O2 > 0.2 mg/L for NPP c: Change in dissolved O2 > 0.2 mg/L for CR. for CR. O2 > 0.2 mg/L dissolved in NPP for c: Change O2 > 0.2 mg/L dissolved in b: Change NPP for CR. & > 0.2 mg/L 2 Continued. Reservoir Depth GPP/h CR/h Volumetric Data Titicus (1) 22-Sep-04Titicus (3) 0 0.1483 1.0 0.1841 0.17 and Subsite Date (m) (g DO· Muscoot (3)Muscoot 1.25 0.0581 0.0657 0.0108 -0.0504 c 24.52 46.45 Muscoot (1)Muscoot 21-Jul-04 0 0.0380 0.0422 0.0060 -0.0528 c 17.71 11.41 Amawalk (1)Amawalk 8-Jul-04 (3)Amawalk 0 0.0663 1.25 0.0845 0.0257 0.1337 0.1317 -0.0582 0.0028 c -0.0437 15.26 b 13.19 3.65 18.86 Cross River (1) River Cross 1-Oct-04 (3) River Cross 0 0.0527 0.0496 1.0 0.0044 0.0871 -0.0242 0.0880 0.0013 -0.028 29.10 b 16.55 3.70 4.54 endix Table 9.4: pp A * a: Change in dissolved O dissolved in a: Change *

- 245 - APPENDIX TO CHAPTER 9 (A.9) NY WATERSHEDS PROJECT – YEAR 5 REPORT – 27 OCTOBER 2005

Appendix Table 9.5: QC summary of alkalinity field duplicates and field blanks, and nutrient laboratory blanks, duplicates and spikes performed with reservoir samples, Year 5 data, 2004.

Alkalinity NH3NNO3N SRP TDP

mg CaCO3/L mg/L Field duplicates* # of samples 118888 # of duplicate pairs for RPD calc. 8 0200 Mean RPD 2.019 . 0.585 . . Max. RPD 11.105 . 1.170 . . # of samples >20% RPD 0 0000

# of duplicate pairs for Abs Diff calc. 2 8686 Mean Abs Diff 0.2165 0.0020 0.0005 0.0001 0.0007 Max. Abs Diff 0.422 0.004 0.001 0.001 0.001 # of samples > DL 0 0000

Blanks # of samples 8 40413340 Mean 0.08 0.003 0.004 0.001 0.001 SD 0.16 0.003 0.002 0.001 0.001 QC detection limit (DL) 2 0.011 0.02 0.003 0.01 # exceeding 2x(DL) 0 0000

Laboratory duplicates* # of samples 14 15 11 23 16 # of duplicate pairs for RPD calc. 7 1 8113 Mean RPD 1.39 0.930 0.435 1.918 5.061 Max. RPD 3.58 0.930 0.939 11.765 11.512 # of samples >20% RPD 0 0000

# of duplicate pairs for Abs Diff calc. 1431213 Mean Abs Diff 0.0007 0.0010 0.0003 0.0003 Max. Abs Diff 0.001 0.002 0.001 0.001 # of samples > DL 0000

Matrix spike recovery # of samples 14 11 22 16 Mean % Recovery 100.3 99.7 98.7 99.1 SD 4.32 1.79 2.27 2.91

* If the mean of a pair of duplicate samples was < 5 times the DL then the absolute difference was used to evaluate the duplicate precision. Values are flagged if the absolute difference of a duplicate pair was > the DL.

- 246 - APPENDIX TO CHAPTER 9 (A.9)