FINAL REPORT for A Scientific Assessment of Nutrient Concentrations, Loads, and Biological Response in the Northern Gulf of Mexico EPA Contract No. 68-C-03-041 Work Assignment No. 1-01 Task 5 September 13, 2004 U.S. Environmental Protection Agency Oceans and Coastal Protection Division Prepared by BATTELLE 397 Washington Street Duxbury, MA 02332 (781) 934-0571 ------- Final Report September 2004 Work Assignment 1-01 Task 7 Page i TABLE OF CONTENTS 1.0 EXECUTIVE SUMMARY 1 2.0 PROJECT DESCRIPTION 1 2.1 Problem Definition/Background 1 2.2 Proj ect Obj ective 4 2.3 Project/Task Description 4 3.0 DATA COLLECTION, DATABASE DEVELOPMENT, FORMATTING, PROCESSING, AND QUALITY REVIEW METHODS 4 3.1 Data Collection 4 3.2 Database Development 6 3.3 Data Loading 9 3.4 Data Screening in Database 9 3.5 Format Standardization 9 3.5.1 Parameter Names 9 3.5.2 Result Values and Result Units 12 3.5.3 Locations 12 3.6 Sampling Locations Linked to Spatial Segments 12 3.7 Evaluation for Duplicate Records 13 3.8 Flag and Qualifier Consideration 13 3.9 Quality Assurance and Quality Control of Data Processing 13 3.10 Statistical Data Quality Assessment 13 4.0 DATA SUMMARIES 14 4.1 Study Area 14 4.2 Alabama 18 4.3 Louisiana 21 4.4 Mississippi 24 5.0 TOWARD THE DEVELOPMENT OF NUTRIENT CRITERIA 27 5.1 Scale 27 5.2 Characterization/typology 27 5.3 Statistical Power 28 5.4 Nutrient Criteria Technical Guidance Manual: Estuarine and Coastal Marine Waters (EPA 2001) 28 5.5 NOAA National Estuarine Eutrophication Assessment 30 5.6 Louisiana Coast-wide Reference Monitoring System (CRMS) 31 5.7 Australian Waterway Classifications and Deviation Index 32 5.7.1 U.S EPA Aquatic Stressors Framework 33 5.8 Discussion (change to section 5.6) 34 6.0 RECOMMENDATIONS 36 6.1 Data Gaps Analysis 36 6.2 Identify Appropriate Classification Process 36 6.3 Further Develop Database and Information Access Program 36 7.0 REFERENCES CITED 37 Baiteiie The Business of Innovation ------- Final Report Work Assignment 1-01 Task 7 September 2004 Page ii LIST OF TABLES Table 3-1. Data Required to Support Nutrient Characterization Assessment 7 Table 3-2. Priority Parameters and Standardized Names in the Nutrient Characterization Database 11 Table 4-1. Qualifiers and Flags within the Database and Actions Taken when Observed 15 Table 4-2. Total number of observed data associated with each primary parameter (n = 49) within the entire study area 16 Table 4-3. Method of aggregation associated with parameter sampling station spatial distribution shown in Attachments F and G 17 Table 4-4. Ranking of Alabama 12-digit and 10-digit coastal HUCs by total records of primary causal/response parameters 19 Table 4-5. Ranking of Louisiana coastal HUCs by total records of primary causal/response parameters 22 Table 4-6. Ranking of Mississippi coastal HUCs by total records of primary causal/response parameters 25 LIST OF FIGURES Figure 2-1. Study Area 3 Figure 3-1. Entity Relationship Diagram of the Nutrient Characterization Database 8 Figure 3-2. Locations of all sampling sites also indicating those outside of the project study area 10 Figure 5-1. Ternary classification of coastal systems divided into seven classes (after Dalrymple et al. [1992] and taken from Ryan et al. [2003]) 33 ATTACHMENTS Attachment A: Current inventory of collected datasets from sources in Louisiana, Mississippi, Alabama, and other regional sources. Attachment B: Bibliography of publications and reports that are relevant to some datasets reviewed for incorporation in the database Attachment C: Status matrix of identified sources of relevant project data indicating individual names, affiliations, and type of response to data request Attachment D: Data Dictionary Attachment E: Statistical summaries of 12 representative primary standardized parameters. Attachment F: Maps showing discrete sampling locations for each primary parameter (aggregated categories) within the study area. Attachment G: Maps showing relative magnitudes of observed data for each primary parameter (aggregated categories) within the study area. Attachment H: Temporal distribution of observed data for each primary parameter (n = 49) for each selected hydrologic unit (HUC) within the study area. Attachment I: Total observation counts for all primary parameter (n = 49) in Alabama. Attachment J: Total observation counts for all primary parameter (n = 49) in Louisiana. Attachment K: Total observation counts for all primary parameter (n = 49) in Mississippi. Baiteiie The Business of Innovation ------- Final Report Work Assignment 1-01 Task 7 September 2004 Page 1 1.0 EXECUTIVE SUMMARY A Scientific Assessment of Nutrient Concentrations, Loads, and Biological Responses in the Northern Gulf of Mexico is centered on compiling information to support nutrient criteria development for coastal ecosystems of the northern Gulf of Mexico. This project was conducted to identify and access the best available data for establishing nutrient criteria for the project study area: the estuaries and nearshore waters associated with Lake Borgne, Chandeleur and Breton Sounds, Mississippi Sound, and Mobile Bay. This project will support the ultimate goal of establishing nutrient criteria, identifying data gaps, and providing recommendations to the Gulf of Mexico Program (GMP) related to future monitoring and assessment activities. Approximately 50 water quality parameters, physical parameters, and biological attributes necessary to characterize the nutrient conditions of the study area were selected as guidance for data collection. These parameters were provided to approximately 60 academic investigators, state and federal agency scientists and managers, and private industry researchers to request whether they had or knew of existing datasets that may contain information pertaining to the parameters. Several datasets were identified and 36 datasets were collected. These datasets contain a total of 2.1 million records from 3,124 sampling stations within the study area. The temporal timeframe of the data spans the time period between 1905 and 2003. When datasets were received, they were pre-screened for several criteria (e.g., spatially and temporally referenced locations, within study area, contained parameters of interest). Those datasets that met the criteria were loaded into the uniform relational database established for this project. Within the database for this project, 678 unique parameters have been loaded, For the purpose of this report, several parameters of interest were prioritized, standardized and summarized. This report provides a characterization of the database. It summarizes key parameters that will be useful in the development of nutrient criteria (i.e., total phosphorus, ammonia, chlorophyll a, etc). It provides a look into the data gaps for key parameters on an individual HUC spatial scale and an annual temporal scale. The report does not, however, generate nutrient criteria for the individual states. It includes an assessment of potential approaches to the development of reference conditions that will support nutrient criteria development. A series of approaches that are relevant to coastal system classification and reference condition are reviewed, and compared, to provide guidance, additional knowledge and increased efficiency. It is suggested that scientists and managers carefully review the data for the spatial regions included in this database and build upon this information to design monitoring programs that would address any data gaps that need more detailed information prior to development of nutrient criteria. 2.0 PROJECT DESCRIPTION 2.1 Problem Definition/Background Section 304(a) of the Clean Water Act (CWA) directs the U.S. Environmental Protection Agency (EPA) to develop and publish criteria guidance to assist states and authorized tribes in developing water quality standards for all waterbodies (i.e., lakes, rivers, streams, estuaries) that are protective of the designated uses of those waterbodies. Among these standards, nutrient criteria recommendations are intended to protect the integrity of coastal ecosystems from the adverse effects of cultural eutrophication. Cultural eutrophication manifests itself in a variety of interrelated ecosystem responses, both primary and secondary, which include phytoplankton and macroalgal blooms, increased production and decay of organic matter, decreased light transmission, and several additional responses that are considered detrimental to ecosystem health. It has been determined that the primary causal stressor associated with cultural eutrophication in coastal ecosystems is increased nutrient enrichment, primarily nitrogen, Baiteiie The Business of Innovation ------- Final Report Work Assignment 1-01 Task 7 September 2004 Page 2 although phosphorus also contributes to the process. Therefore, water quality standards for nutrients (nitrogen and phosphorus) must be established to support Section 304(a) guidance. Estuaries and nearshore coastal waters naturally vary in the type, physical and biological characteristics, as well as geographic coverage of the ecological communities at risk from nutrient overenrichment. The factors associated with this variability are numerous, but include qualities such as water residence time, depth, water column stratification, salinity, and temperature. Therefore, it is important to note that one single criterion is not applicable to national or regional systems, and nutrient criteria must be developed in a manner that is consistent with the uniqueness of system types. The EPA guidance document, Nutrient Criteria Technical Guidance Manual: Estuarine and Coastal Marine Waters (EPA-822-B-01-003) suggests a reference approach be applied to nutrient and algal criteria development. Reference conditions are defined as the best existing estuarine or marine water within a watershed or coastal area. In some cases, historical data can be compared to more recent observations to assess ecological departures from a reference state. A thorough review of water quality data and supporting information must occur in order to identify and characterize reference conditions. This review should identify current and historical data that are necessary for the development of nutrient criteria. A Scientific Assessment of Nutrient Concentrations, Loads, and Biological Responses in the Northern Gulf of Mexico, previously Task 7 of Work Assignment 0-01 and currently Task 5 of Work Assignment 1-01, is centered on the first of two phases of compiling information to support nutrient criteria development for coastal ecosystems of the northern Gulf of Mexico. This project was conducted to identify and access the best available data for establishing nutrient criteria for the project study area. The project study area includes the estuaries and nearshore waters associated with Lake Borgne, Chandeleur and Breton Sounds, Mississippi Sound, and Mobile Bay (Figure 1). As part of this project Battelle is providing assistance to the states of Louisiana, Mississippi, and Alabama in an analysis of data compiled during this current effort toward the development nutrient criteria throughout the study area. The analyses associated with this phase of work is focused on illustrating the type of data acquired and their spatial and temporal distribution within the study area. Therefore, the analyses are considered preliminary in nature and do not result in proposed site-specific nutrient criteria. Rather, the database is described through a series of spatial and temporal assessments, including descriptions of data quality and database development. A review of nutrient criteria development methods and potential applications within the study area is provided. The remainder of this document summarizes the progress associated with the project tasks. Section 2.2 defines the objectives of the overall project. Section 2.3 describes each of the tasks conducted. Section 3.0 summarizes data collection and the activities associated with the database development, including formatting, processing and quality review methods. Data summaries for each state are provided in Section 4.0 for the primary causal and response parameters that have been chosen by the project Steering Committee. In Section 5.0, a review of methods and activities associated with the development of nutrient criteria is provided. Finally, Section 6.0 presents lessons learned, key findings and further recommendations with respect to developing nutrient criteria within the project study area. Battelle The Business of Innovation ------- Final Report Work Assignment 1-01 Task 7 September 2004 Page 3 MJ0,0,16 N.,0,0,0E Figure 2-1. Study Area. Batreiie The Business of Innovation ------- Final Report Work Assignment 1-01 Task 7 September 2004 Page 4 2.2 Project Objective The objective of the overall "Scientific Assessment of Nutrient Concentrations, Loads, and Biological Responses in the Northern Gulf of Mexico" project is to provide data and information that can be used by states to develop nutrient criteria and associated management responses. 2.3 Project/Task Description This project has focused on the identification, collection, and compilation of available water quality data associated with the study area into a comprehensive project database. Analyses conducted will support the ultimate goal of establishing nutrient criteria, identifying data gaps, and providing recommendations to the Gulf of Mexico Program (GMP) related to future monitoring and assessment activities. Project activities include: 1. Developing a detailed Combined Work/Quality Assurance Project Plan (CW/QAPP) for the data collection, QA/QC, compilation, and analysis, and gain work assignment manager (WAM) approval; 2. Selecting attributes that best characterize the nutrient condition of the study area, and describe rationale behind these selections and rationale for omitting other attributes that were considered; 3. Identifying and acquiring existing data sets from agencies and universities; 4. Performing QA/QC for all data sets; 5. Compiling data sets into a uniform, electronic relational database; 6. Preparing a draft and final Interim Report detailing an analysis strategy for the data, with specific information about each data set including: • Collection method • Data gaps • GIS layers representing temporal and spatial data density • Preliminary evaluation of waterbodies and corresponding nutrient data • Preliminary evaluation of reference conditions within the study area • Recommendations for procedures for analysis 3.0 DATA COLLECTION, DATABASE DEVELOPMENT, FORMATTING, PROCESSING, AND QUALITY REVIEW METHODS This section summarizes the methods used to acquire the historical data, develop the database, and process the data (i.e., load, organize, format, standardize, and evaluate). Working with multiple formats, different presentations, and unique nomenclature, data processing required frequent checks to ensure data were loaded correctly and to compare records for duplication and correctness. 3.1 Data Collection The identification and acquisition of data requires significant effort and efficiency. The GMP initiated the process by developing an inventory of data sources and had also begun the process of acquiring data. To minimize duplication of effort, Battelle initially searched existing federal and state government sources (e.g., STORET, NEPs, DEP/DEQs) and developed a "living" database inventory (Attachment A). This Battelle The Business of Innovation ------- Final Report Work Assignment 1-01 Task 7 September 2004 Page 5 inventory has been continually updated as additional sources were identified. The inventory was also distributed along with a list of the targeted parameters (Table 3.1) to scientists for review and response. This database inventory was critical for minimizing duplication of effort as many data sets being sought from particular sources may already have been available from existing sources. For example, many of the state monitoring programs upload their data into STORET. If the information was being sought from that state, but was already included in the STORET acquisition, duplicate data would exist in the database. Loading duplicate data was avoided if possible and checks were necessary once all data were loaded. Battelle used a variety of means to identify potential sources of relevant data. These included monthly conference calls, internet searches, literature searches, personal inquiries to scientists and contacts with individuals from conferences and/or scientific meetings. Monthly telephone conference calls. Monthly conference calls with the nutrient pilot study Steering Committee members were held throughout the project period. Attendance on the calls varied, but participants typically included Battelle scientists, GMP staff, and representatives from Louisiana, Mississippi, and Alabama state agencies. Additional scientists and managers from EPA contributed to some of the calls. Meeting agendas consisted of briefing steering committee members on progress and problems associated with this task. The calls also provided Battelle and the GMP the opportunity to inquire about specific data sources within the study area. Hence, several data sets and potential sources (individual scientists and managers) were identified throughout this process. When data or sources were identified during these calls, Battelle followed up with the appropriate points of contact to acquire this data. Data were collected in various formats (e.g., Excel, ASCI, hardcopy), screened for appropriateness and relevance, inventoried, and loaded into the Oracle database for this project. Internet searches. Battelle conducted numerous data searches through a variety of internet search methods. Several large data sets were available and documented online. Examples of internet-based data sources include EPA's STORET, USGS's NWIS, and a variety of state agencies that had data search engines on, and/or linked, to their respective homepages. Additional agencies such as NOAA and programs such as the National Estuarine Research Reserve (NERR) also provided data search engines from internet websites. Often these searches resulted in identification of technical documents, peer reviewed literature or points of contacts for electronic data sets. Data were collected in various formats (e.g., Excel, ASCI, hardcopy), screened for appropriateness and relevance, inventoried, and prepared for loaded into the Oracle database for this project. Literature searches. Scientific literature was searched through a variety of methods, including key word searches using various search engines including Current Contents, Ingenta, and DIALOG. Many journal articles were collected and analyzed for relevant data. In some cases the authors of relevant journal articles were contacted directly in order to inquire about the availability of the data that were presented in the paper. A bibliography of literature associated with this component of the project is listed in Attachment B. A contact list of research scientists and organizations was developed and maintained throughout this project (Attachment C). This list identifies points of contact for potential information, whether contact was attempted, and the status of inquiry (e.g., response). Those scientists that authored papers containing potentially relevant data to the project are included in the list presented in Attachment B. Personal inquiries (existing relationships). Several scientists with Gulf of Mexico research experience are colleagues of Battelle staff. These individuals were contacted and asked if they knew of data sets that may be available. Individuals contacted included faculty at Dauphin Island, Louisiana State University, University of Southern Mississippi, LUMCON, and the University of New Orleans. Those researchers that were contacted are also listed in Attachment C. Battelle The Business of Innovation ------- Final Report Work Assignment 1-01 Task 7 September 2004 Page 6 Miscellaneous contacts (conferences, meetings). Battelle scientists are active members of many professional and scientific societies and often attend meetings and conferences related to nutrient eutrophication issues in coastal areas. Battelle staff attending recent conferences (e.g., Estuarine Research Federation, Northwest Atlantic Indicators Workshop, etc) made contact with scientists from other parts of the country that are engaged in similar efforts. Those contacts are also listed in Attachment C. In some cases, these efforts were fruitful and some significant datasets were attained (e.g., J. Pennock, Mobile Bay, AL). Environmental data were screened prior to loading them into the database. In order for a dataset to be considered for loading, it had to meet the following three screening criteria: • data were in the project study area • data were geo-referenced • contained parameters that would support the nutrient criteria development. The datasets meeting these criteria were prioritized for loading into the relational database. A Nutrient Characterization Database was developed to support the data review process for nutrient criteria development. The database was developed in Oracle 8i and later upgraded to Oracle 9i. The relational database was designed to reflect the requirements of the end users and data systems into which project data may be exported. The Nutrient Characterization Database is a short-term repository for specific data sets and is not meant to be a complete environmental data system. The database was developed to generate the best available data not the necessarily the best interface to other databases. The Nutrient Characterization Database allows for the recording of laboratory analyses and field measurements. It also contains location information for all sampling and field measuring events. Waterbody and Ecoregion tables support the linkages of locations to geospatial boundaries. Figure 3.1 depicts the entity relationship diagram (ERD) for the Nutrient Characterization Database. The ERD shows the tables, data fields, and linkages of the database. The data dictionary for the Nutrient Characterization Database (Attachment D) defines the use of the fields used. Sample-based analyses and field data can be linked by location and time period to provide an overall description of a site or region. Trend analysis can also be performed by parameter in similar fashion. 3.2 Database Development Battelle The Business of Innovation ------- Final Report Work Assignment 1-01 Task 7 September 2004 Page 7 Table 3-1. Data Required to Support Nutrient Characterization Assessment. I'liv sic:il Alkalinity Benthic Community Air Temperature Ammonia (NH3) Biochemical Oxygen Demand (BOD) Color (CDOM) Ammonium (NH4) Chlorophyll a, b, c Conductivity Benthic Flux Measurements (nutrients, Dissolved Oxygen (DO) Depth DO, salinity) Dissolved Oxygen (% Saturation) Depth Distribution (Hypsography) Chloride Fecals (bateria, pathogens) Flushing Time Dissolved Inorganic Nitrogen (DIN) Fish Community Groundwater Discharge Dissolved Inorganic Phosphorus (DIP) Macroalgae Biomass and Production Light Extinction Coefficient (Kd) Dissolved Organic Nitrogen (DON) Macro invertebrate and Fish Community Meteorological Conditions Hardness Phytoplankton Production Precipitation Nutrients (all available) Primary Production (02 or 14C) River Discharge Oil Sheen Salt Marsh Distribution Secchi Depth PH Submerged Aquatic Vegetation (SAV) Surface Irradiance Salinity Total System Metabolism (TSM) Tidal Amplitude Solids Zooplankton Biomass and Growth Velocities Sulfate Water Residence Time Total Kjeldahl Nitrogen (TKN) Water Surface Elevation Total Nitrogen (TN) Water Temperature Total Organic Carbon (TOC) Wind Speed/Direction Total Phosphorus (TP) Total Organic Phosphorus Total Suspended Sediments (TSS) Turbidity Baiteiie The Business of Innovation ------- Final Report Work Assignment 1-01 Task 7 September 2004 Page 8 t*K STUPY 10 STUDY NAME STARTJ3ATE END DATE ORGANIZATION CONTACT_NAME T..PHONE viCE..FlLE uriN ID PK STAT ON jD STATION NAME STATION TYPE LOCATION TYPE Kt> l-*m!UDE GITUDE TH FFFLULNT RELEASE LATITUDE EPF LI LNT RELEASE. LONGITUDE LOGIN 10 F*l ENAME LOAD. DATE UPDATE SCRIPT UPDATE DATE bT^ND " h h h stand' latitude ¦vrmc !nhi-«ruoE Mi.JG PASit ^YAIERBODY NAME HUC. 8 HUC jo HUG II HUGH- NAME JL" 1 NAME HUC USE ¦ i ¦¦ PK SAMP ID FIELD QC CODE PA TEMP MARKER PLANNED ANALYSIS LIST CLIENT SAMP ID FK2 STUDY ID FK1 STATION ID SURVEY ID ECTOR NAVIGATION S 3TEM DL DL UNIT COMMENTS TEST TYPE UPDATE SCRIPT UPDATE DATE DEPTH UNIT CODE ft f PK : K1 STATION ID PARFfjT STATIC") KM MION^HIP TYPE LOCATION DESGR xgfkcy n I OCJ* T|r?j rspf Fil bf.Ak'f- I A/vn fUTF UPDATE SCRIPT update date PK,FK1 PK STATION ID ECOREGION ID WATERBODY ID# EPA ECOREGION ID ECOREGION DESCR FILENAME LOGIN ID LOAD DATE UPDATE SCRIPT UPDATE DATE PK.FK1 STATION ID '_TYPE_COOE [LAKE AREA SOURCE I LAKE_OUTL£T.TYPE | USEJMPAIRED I'IPAIKN I N > CONDITION I REFERENCE .COMMENTS \>h i ( Pf L NTS I filename I Oc-- N Id I LOAD DATE RlPT TE SAMP PARAM CODE HK>Ki iSAfgiUa PK STUDY ID PK,FK1 STATION ID PK ME AS DATE TIME PK DEPTH PK PARAM CODE SURVEY ID DEPTH UNIT CODE RESULT UNIT INSTR CODE LOGIN ID FILENAME LOAD DATE COMMENTS UPDATE SCRIPT UPDATE... DATE b i -NL> PAKmM NAME ;RiPTION LOGIN ID I 'I i Ml L LOAD DATE UPDATE SCRIPT UPDATE,DATE LAB * Th PUN ID LAB SAMP To I £ I 'H 'H SAMP., WGT SAMP WGT DL DL UNIT ! 3iN t U t. I4IL LOAD DATE ! NT ! STURE ; D REV NUM PT UPDATE DATE PK,FK1 ANALYSIS IP PK PARAM CODE RESULT LAB OUAL I NJAw OUAL I1W1 us uu i n. if DETECT. LIM. CODE ANALrbiS UAH SDG VALIDATION LEVEL LOGIN ID FILENAME LOAD.DATE DILUTION f NMU». avu h< ' NUM UPDATE SCRIPT UPDATE.DATE i (AM Pm *!„ r mME MANP ! Nh «¦ i*ND F LM !i_I REPORTABLE _YN COMMENTS >M I Lh'-, REPORT RESULT PK.F K1 AMLY§JJLJD ' SDG cP%fTION I RESULT | LAB _QUAL FINAL_QUAL. I UNIT j CE^Ei T LIMIT | IFTCT LIT,' CODE ANA L Y S i S_D AT E | W.LH^Tuh LEVEL LOGIN ID FILENAME jLOAD_DATE I DILUTION |minimum_level | REV NUM [ UPDATE.. SCRIPT |UPDATE_DATE | PARAM..NAME i ST f D P \P 1..NAME STAND_UNIT I STAND RESULT : PK siudvjd PK SURVEY ID PK.FK1 STATION ID PK OBS DATE TIME PK PARAM CODE OBS DESCRIPTION LOGIN ID FILENAME LOAD DATE UPDATE SCRIPT UPDATE DATE RESULT LAB QUAL UNIT LOGIN ID FILENAME L0AD...DATE COMMENTS Figure 3-1. Entity Relationship Diagram of the Nutrient Characterization Database. Baiteiie The Business of Innovation ------- Final Report Work Assignment 1-01 Task 7 September 2004 Page 9 3.3 Data Loading A unique identifier or 'Login_ID' was given to each dataset (i.e., filename) loaded into the database. Every record in every database table contains a 'LoginID', the original filename, the date it was loaded, the filename of the last script used to process or update it, and the date the last update took place. If any changes were made previously, the output script will show the date and script name of that update. This combination of information allows for rapid and complete auditing of the data loading and processing. Thirty six data sets met the screening criteria for loading and were loaded into the database. The loaded data comprise over 2.1 million results from 3,124 stations over the temporal span of 1905 to 2003. Of the 3,124 locations included in the datasets, 2,441 are within the study area. Figure 3.2 illustrates the spatial distribution of these stations. A total of 678 original parameter names are associated with the stations within the study area. Because of time constraints and the multitude of parameter names that were included in the database, parameters were prioritized as to whether they were key to nutrient criteria development. Over 200 of these original parameter names were prioritized and standardized to reduce the number of ways to indicate the same parameter. The Format Standardization section further describes the process used to normalize the parameter names as well as other fields in the database. The data for the priority parameters were screened in the database for reasonableness with respect to high and low outliers. A database field was created called "reportable_yn" to allow values to be "turned off' and not used in subsequent analysis - (reportable_yn = N) or left "on" for future use (reportable _yn = Y). Any negative values observed in the database were turned "off' (i.e., reportable_yn='N') and the yn_flag 'NN' was assigned. Potential outliers on the upper end (i.e., very high values for a given parameter) were not addressed at this time. It will be left to the discretion of the scientists and managers using this information to determine if these high values are true outliers and whether to exclude them (reportable_yn = N) from subsequent analysis. Because data originated from multiple agencies and organizations, they arrived in disparate formats. To be able to compare the data to check for duplicates and subsequent summarization, the data had to be normalized for parameter name, location, result value, and result unit. 3.5.1 Parameter Names Different sources often reported the same parameter differently. For example, the parameter dissolved oxygen can be reported as DO, Dissolved O, Diss O, etc. To facilitate future queries of a specific parameter, the priority parameter names were standardized in a new parameter name field (i.e., STAND_PARAM_NAME) of the database. This new parameter name field was populated with the standardized name for the prioritized parameters and the original parameter name remained in a separate (original) database field. Table 3.2 depicts the priority parameters (bold) and the standardized naming convention used in the database. Source documentation was reviewed to establish the meaning and new naming convention for original parameter names having the same meaning. Even what would seem to be a simple parameter to express like 'Water Temperature' was provided under multiple naming conventions: 'Bottom Water Temperature', 'Surface Water Temperature', 'TEMP', 'Temperature', 'Water Temp', 'Water temp', 'TEMP (C)', 'TEMPERATURE WATER (DEGREES CENTIGRADE)', 3.4 Data Screening in Database 3.5 Format Standardization Baiteiie The Business of Innovation ------- Final Report Work Assignment 1-01 Task 7 September 2004 Page 10 Figure 3-2. Locations of all sampling sites also indicating those outside of the project study area. Batreiie The Business of Innovation ------- Final Report Work Assignment 1-01 Task 7 September 2004 Page 11 Table 3-2. Priority Parameters and Standardized Names in the Nutrient Characterization Database. AIR TEMPERATURE ORTHOPHOSPHATE AS P - FILTERED AMMONIA NITROGEN AS N - FILTERED ORTHOPHOSPHATE AS P - UNFILTERED AMMONIA NITROGEN AS N - UNFILTERED OXYGEN TEMPERATURE AMMONIA NITROGEN BOTTOM DEPOSIT DRY WEIGHT PARTICULATE NITROGEN AS N CHLOROPHYLL A CHROMA-FLUORO PERIPHYTON CHLOROPHYLL A CHROMA-FLUORO CHLOROPHYLL A CHROMA-SPECTRO CHLOROPHYLL A FLUORO PERIPHYTON CHLOROPHYLL A FLUORO PERIPHYTON CHLOROPHYLL A SPECTRO UNCORRECTED CHLOROPHYLL A HPLC PH CHLOROPHYLL A SPECTRO ACID CORRECTED PHOSPHATE AS P - UNFILTERED CHLOROPHYLL A SPECTRO UNCORRECTED PHOSPHATE AS P BOTTOM DEPOSIT DRY WEIGHT CHLOROPHYLL A TRICHROM UNCORRECTED POLYPHOSPHATE AS P DISSOLVED ORGANIC PHOSPHORUS AS P - FILTERED POTENTIAL WATER TEMPERATURE DISSOLVED OXYGEN CONCENTRATION POTENTIAL WATER TEMPERATURE DISSOLVED OXYGEN PERCENT SATURATION PRODUCTIVITY AS C DISSOLVED PHOSPHORUS AS P - FILTERED SALINITY INORGANIC NITROGEN AS N - FILTERED SALINITY - DAILY MEAN KJELDAHL NITROGEN AS N - FILTERED SECCHIDEPTH KJELDAHL NITROGEN AS N BOTTOM DEPOSIT DRY WEIGHT SUSPENDED KJELDAHL NITROGEN AS N MEAN ANNUAL AIR TEMPERATURE TOTAL KJELDAHL NITROGEN AS N MEAN MONTHLY AIR TEMPERATURE TOTAL NITROGEN AS N - FILTERED NITRATE + NITRITE NITROGEN AS N - FILTERED TOTAL NITROGEN AS N - UNFILTERED NITRATE + NITRITE NITROGEN AS N - UNFILTERED NITRATE + NITRITE NITROGEN AS N BOTTOM DEPOSIT DRY WEIGHT TOTAL ORGANIC PHOSPHORUS AS P TOTAL PHOSPHORUS AS P - UNFILTERED NITRATE NITROGEN AS N - FILTERED TOTAL PHOSPHORUS BOTTOM DEPOSIT DRY WEIGHT NITRATE NITROGEN AS N - UNFILTERED TURBIDITY - LAB NITRITE NITROGEN AS N - FILTERED TURBIDITY - UNFILTERED NITRITE NITROGEN AS N - UNFILTERED TURBIDITY AS SI02 - UNFILTERED ORGANIC NITROGEN AS N - FILTERED UNIONIZED AMMONIA AS N ORGANIC NITROGEN AS N - UNFILTERED WATER TEMPERATURE Note - BOLD indicates the priority parameters that are further summarized in this report. Baiteiie The Business of Innovation ------- Final Report Work Assignment 1-01 Task 7 September 2004 Page 12 'TEMPERATURE WATER (DEGREES FAHRENHEIT)', 'Temperature water', 'Temperature, water, degrees Celsius', 'pOOOlO - Temperature, water, degrees Celsius', and 'pOOOl 1 - Temperature, water, degrees Fahrenheit'. The differences in units were also normalized into a standard unit. See the Result Values and Result Units section below. 3.5.2 Result Values and Result Units Units can also prevent data from being directly compared. Syntax used to express units can also vary. Examples of syntax variations seen for the unit milligrams for liter, included: 'MG/L', 'PPM', 'mg/L', 'mg/1', 'mg/1', 'mg/1 ', and 'ppm'. Some units appear to be the same but extra spaces in the unit make the unit expressions different when comparing records. Additionally the unit was not always in a separate field when received but instead part of the original parameter name. In these cases, the unit, as indicated in the original parameter name, was copied to the unit field. The units for milligram per liter were standardized in the database in a new result unit field (i.e., STAND_UNIT) as 'MG/L' and the original unit syntax were maintained in the UNIT database field. To achieve a single result unit for a given standardized parameter name, result values with different units were converted. These converted or standardized values were placed in a new result value field (i.e., STAND RESULT) while maintaining the original value and original unit in the database (database fields RESULT and UNIT, respectively). When original result units were equivalent (e.g., MG/L = PPM and MG/M3 = UG/L), one unit was chosen as the standardized unit and the other was translated. In the cases of equivalent units, the original result value was copied to the standardized result value field. When parameters were expressed in different but related units, the corresponding conversion was used to generate the standardized result and update the standardized unit field (e.g., G/L x 1000 = MG/L). By maintaining the original values and units, translations or conversions may be made if another unit is preferred. 3.5.3 Locations Each station's latitude and longitude, expressed as decimal degrees were rounded to three decimal places to generate standardized latitude (STAND LATITUDE) and standardized longitude (STAND LONGITUDE) fields in the database. By rounding to three decimal places, stations within approximately 100 feet of each other would have the same standardized latitude and standardized longitude. Those stations with the same standardized latitude and standardized longitude were given a unique identifier or standardized station name (STAND STATION). The new standardized location identifier allowed records to be grouped according to location and later compared to identify duplicate records. The 1,972 original stations in the project area were grouped according to the method described above and converted into 1,553 standardized stations. For example, the three original station identifiers, 'MP_237E_01_N1', 'MP_237E_01_N2', and 'MP_237E _01', describe a similar location (i.e., are within 100 feet of each other) and were given the same standardized station name. 3.6 Sampling Locations Linked to Spatial Segments The Gulf of Mexico project area covers portions of Alabama, Louisiana, and Mississippi. These states are segmented into geospatial watersheds or basins. Each watershed has an eight digit hydrologic unit code or HUC 8. These watersheds are further segmented into drainage basins, subsegments, or HUC 12 areas. The location data in the Nutrient Characterization Database are linked to these geospatial regions to allow for queries according to state, project area, watershed (i.e., HUC 8), or drainage basin (i.e., HUC 12). Baiteiie The Business of Innovation ------- Final Report Work Assignment 1-01 Task 7 September 2004 Page 13 3.7 Evaluation for Duplicate Records Once the data were normalized by location identifier, parameter name, result value, and unit, records could be compared to identify multiple occurrences of the same measurements in the database. Identifying duplicate records prevents their use in future characterizations and analyses. Characterization results may be skewed if duplicates remain unchecked. Duplicate data were identified as measurements or results having the same standardized parameter name, standardized station location, date/time of sample or measurement, sample depth, standardized result (rounded to two decimal places), and standardized unit. Once identified the duplicate records that had qualifying information (i.e., flags, qualifiers, method detection limit, or comments) were maintained as reportable "yes" and the related record was turned "off (i.e., reportable_yn = 'N'). Those duplicate records that were turned "off were also given a flag of 'ND' in the 'YN_FLAG' field. This flag indicates that the record was turned "off because it was a duplicate record. Counts of record changes were reviewed to ensure that the duplicate record evaluation procedure ran correctly. Of the priority parameter records evaluated for duplication, 2,793 records were turned "off as duplicates. 3.8 Flag and Qualifier Consideration Data qualifiers and flags provided with the original data sets were loaded into the database and taken into consideration. Actions were taken to turn "off or adjust the result value based on particular flags or qualifiers. Table 3.3 shows the original flags and qualifiers used by the data sources and the actions taken in the database. The fatal qualifiers associated with the data led to 15,080 records being turned "off and flagged as 'NQ' in the YN_FLAG field. 3.9 Quality Assurance and Quality Control of Data Processing Quality assurance and quality control (QA/QC) was performed on a randomly sampled portion of the database contents. Approximately 10% of the records in the database were pulled and reviewed by Battelle's QA staff. This group checked that the data sets were correctly mapped to the database structure, checked the loading scripts, and verified completeness and accuracy of the data in the database compared to the data sets received for loading. The standardization process was reviewed and the conversion factors checked. Findings were reported and corrections were made accordingly in the database. 3.10 Statistical Data Quality Assessment Following data loading, formatting, standardization, duplication and qualifier checks, the priority parameters were queried. Summary statistics of these priority parameters were run using SAS Release 8.02. Each priority parameter was summarized in tabular form by state, HUC12 name, and year. The statistical summaries include the number of observations, mean value, minimum value, maximum value, standard deviation, standard error, coefficient of variation, median and various percentiles that show the distribution of data by HUC12 and year. The statistical summaries are presented in detail in Attachment E. Battene The Business of Innovation ------- Final Report Work Assignment 1-01 Task 7 September 2004 Page 14 4.0 DATA SUMMARIES This section summarizes the spatial and temporal distribution of data that have been identified to support the development of nutrient criteria in each of the states within the study area. Two approaches to this summary have been chosen. First, the full set of 49 causal and response variables (that have been selected by the project Steering Committee to be of primary interest) are summarized within the study area and each state. This summary describes the number of observations that exist for each parameter in each state in tabular form. For effective visualization of the spatial distribution of these data, several inter- associated constituents have been aggregated and mapped. This was done for chlorophyll, nitrogen and phosphorus because each of these general parameters consists of either additional constituents (i.e., sub- species of analytes) or varied analytical methods. For example, the response variable "chlorophyll" is associated with 10 varieties of component variables which are based on the application of varying analytical methods and sub-species (e.g., periphyton chlorophyll a). Further examples reside in the nutrient data. Total nitrogen was identified as a primary causal variable of interest, however the individual constituents of TN include inorganic (nitrate, nitrite, and ammonium) and organic (organic nitrogen and total kjeldahl nitrogen) forms. Further, these nitrogen constituents were reported in different ways due to analytical methods (e.g., filtered vs. unfiltered). The following subsections focus on all 49 parameters/variables within the study area and each individual state. In order to facilitate the understanding of how data are distributed throughout space and time, a series of key questions that support the ultimate goal of identifying which regions, subregions and waterbodies possess adequate data and which do not. • Which waterbodies are richest in causal (nutrient) and response (chl, DO, Secchi) data? • Which waterbodies have earliest records of causal and response data? • Which waterbodies have riches physical data (salinity, temperature)? • Which waterbodies possess most significant data gaps? The spatial and temporal distribution of the 49 standardized parameters are summarized for the entire study area (Figure 1) in this section. These parameters, and their total counts (data counts) are shown in Table 4-1. The process by which these parameters were standardized within the database is discussed in Section 3.5. In addition, the statistical analyses conducted on the most popularly sampled parameters within the study area (shown as shaded rows in Table 4-1) are discussed in Section 3.10 and illustrated in Attachment E. 4.1 Study Area Baiteiie The Business of Innovation ------- Final Report Work Assignment 1-01 Task 7 September 2004 Page 15 Table 4-1. Qualifiers and Flags within the Database and Actions Taken when Observed. ORGANIZATION LAB QUAL QUAL DESC ACTION Texas A and M -9 not sampled or bad data Turn record "off Texas A and M -1 not detected Keep, use % the reported value USEPA $ Calculated by retrieval software. Numerical value was neither measured nor reported to the database, but was calculated from other data available during generation of the retrieval report Keep AL-DEM, Field Ops < Value Is Less Than Method Detection Limit Keep, use % the reported value Smith and Hitchcock < Value Is Less Than Method Detection Limit Keep, use % the reported value USGS < Value Is Less Than Method Detection Limit Keep, use % the reported value USGS > Actual value is known to be greater than the value shown Keep USEPA A Value reported is the mean of two or more determinations Keep - leave as is USEPA B Results based on colony counts outside acceptable range Turn record "off USEPA C Calculated. Value stored was not measured directly, but was calculated from other data Keep - leave as is USEPA D Field measurement. Keep, leave as is USGS E Value Is Estimated, Daily Detection Limit > Value > Method Detection Limit Turn record "off USEPA G Value reported is maximum of two or more determinations Turn record "off USEPA I Value reported is less than the practical quantification limit and greater than or equal to the method detection limit Keep, use % the reported value USEPA J Estimated. Value shown is not a result of analytical measurement Turn record "off USEPA K Off-scale low. Actual value not known, but known to be less than value shown Keep, use % the reported value USEPA L Off-scale high. Actual value not known, but known to be greater than value shown Keep, use the reported value USEPA M Presence of material verified, but not quantified. Indicates a positive detection at a level too low to permit accurate quantification. In the case of species, M indicates male sex Keep, use % the reported value USGS M Presence of material verified, but not quantified. Indicates a positive detection at a level too low to permit accurate quantification. In the case of species, M indicates male sex Keep, use % the reported value AL-DEM-BMP nd not detected Keep, use % the reported value USGS nd not detected Keep, use % the reported value USEPA 0 Sampled for, but analysis lost. Accompanying value is not meaningful for analysis Turn record "off USEPA P Too numerous to count Turn record "off USEPA Q Sample held beyond normal holding time Turn record "off USEPA R Significant rain in the past 48 hours Keep - leave as is USEPA U Material was analyzed for but not detected. Value stored is the limit of detection for the process in use. In the case of species, U indicated undetermined sex Keep, use % the reported value USGS U Analyzed for, not detected Keep, use % the reported value USEPA Z Too many colonies to count Turn record "off * denotes fatal qualifier - record turned "off' and yn_flag = 'NQ' Battene The Business of Innovation ------- Final Report Work Assignment 1-01 Task 7 September 2004 Page 16 Table 4-2. Total number of observed data associated with each primary parameter (n = 49) within the entire study area. I'lii'iimclcr ( ) III III AMMONIA NITROGEN AS N - FILTERED 4,552 AMMONIA NITROGEN AS N - UNFILTERED 2,333 AMMONIA NITROGEN BOTTOM DEPOSIT DRY WEIGHT 12 CHLOROPHYLL A CHROMA-FLUORO 340 CHLOROPHYLL A CHROMA-SPECTRO 15 CHLOROPHYLL A FLUORO 3,560 CHLOROPHYLL A SPECTRO ACID CORRECTED 111 CHLOROPHYLL A SPECTRO UNCORRECTED 9 CHLOROPHYLL A TRICHROM UNCORRECTED 42 DISSOLVED ORGANIC PHOSPHORUS AS P - FILTERED 2 DISSOLVED OXYGEN CONCENTRATION 101,748 DISSOLVED OXYGEN PERCENT SATURATION 93,027 DISSOLVED PHOSPHORUS AS P - FILTERED 3,194 KJELDAHL NITROGEN AS N - FILTERED 937 KJELDAHL NITROGEN AS N BOTTOM DEPOSIT DRY WEIGHT 128 NITRATE + NITRITE NITROGEN AS N - FILTERED 7,222 NITRATE + NITRITE NITROGEN AS N - UNFILTERED 2,852 NITRATE NITROGEN AS N - FILTERED 1,582 NITRATE NITROGEN AS N - UNFILTERED 1,155 NITRITE NITROGEN AS N - FILTERED 4,476 NITRITE NITROGEN AS N - UNFILTERED 1,287 ORGANIC NITROGEN AS N - FILTERED 88 ORGANIC NITROGEN AS N - UNFILTERED 556 ORTHOPHOSPHATE AS P - FILTERED 561 ORTHOPHOSPHATE AS P - UNFILTERED 532 PARTICULATE NITROGEN AS N 1,921 PERIPHYTON CHLOROPHYLL A CHROMA-FLUORO 1 PERIPHYTON CHLOROPHYLL A FLUORO 2 PERIPHYTON CHLOROPHYLL A SPECTRO UNCORRECTED 1 PH 92,703 PHOSPHATE AS P - UNFILTERED 5,174 PHOSPHATE AS P BOTTOM DEPOSIT DRY WEIGHT 2 POLYPHOSPHATE AS P 2 PRODUCTIVITY AS C 1,152 SALINITY 94,183 SALINITY - DAILY MEAN 7,960 SECCHI DEPTH 3,234 SUSPENDED KJELDAHL NITROGEN AS N 94 TOTAL KJELDAHL NITROGEN AS N 3,556 TOTAL NITROGEN AS N - FILTERED 1,516 TOTAL NITROGEN AS N - UNFILTERED 1,016 TOTAL ORGANIC PHOSPHORUS AS P 8 TOTAL PHOSPHORUS AS P - UNFILTERED 7,998 TOTAL PHOSPHORUS BOTTOM DEPOSIT DRY WEIGHT 45 TURBIDITY - LAB 509 TURBIDITY - UNFILTERED 92,576 TURBIDITY AS SI02 - UNFILTERED 216 UNIONIZED AMMONIA AS N 760 WATER TEMPERATURE 107,834 Baiteiie The Business of Innovation ------- Final Report Work Assignment 1-01 Task 7 September 2004 Page 17 An analysis of the spatial distribution of these primary parameters was conducted by mapping individual sampling locations and relative magnitudes of sampling sites within each 12-digit HUC. Nitrogen, phosphorus and chlorophyll parameters were aggregated to provide an easier visual depiction of the spatial distribution of these parameters. These aggregations are illustrated in Table 4-2. Table 4-3. Method of aggregation associated with parameter sampling station spatial distribution shown in Attachments F and G. I'lii'iimclcr NITROGEN AMMONIA NITROGEN AS N - FILTERED AMMONIA NITROGEN AS N - UNFILTERED AMMONIA NITROGEN BOTTOM DEPOSIT DRY WEIGHT KJELDAHL NITROGEN AS N - FILTERED KJELDAHL NITROGEN AS N BOTTOM DEPOSIT DRY WEIGHT NITRATE + NITRITE NITROGEN AS N - FILTERED NITRATE + NITRITE NITROGEN AS N - UNFILTERED NITRATE NITROGEN AS N - FILTERED NITRATE NITROGEN AS N - UNFILTERED NITRITE NITROGEN AS N - FILTERED NITRITE NITROGEN AS N - UNFILTERED ORGANIC NITROGEN AS N - FILTERED ORGANIC NITROGEN AS N - UNFILTERED SUSPENDED KJELDAHL NITROGEN AS N TOTAL KJELDAHL NITROGEN AS N TOTAL NITROGEN AS N - FILTERED TOTAL NITROGEN AS N - UNFILTERED UNIONIZED AMMONIA AS N PARTICULATE NITROGEN AS N PHOSPHORUS CHLOROPHYLL ORTHOPHOSPHATE AS P - FILTERED ORTHOPHOSPHATE AS P - UNFILTERED PHOSPHATE AS P - UNFILTERED PHOSPHATE AS P BOTTOM DEPOSIT DRY WEIGHT POLYPHOSPHATE AS P TOTAL ORGANIC PHOSPHORUS AS P TOTAL PHOSPHORUS AS P - UNFILTERED TOTAL PHOSPHORUS BOTTOM DEPOSIT DRY WEIGHT CHLOROPHYLL A CHROMA-FLUORO CHLOROPHYLL A CHROMA-SPECTRO CHLOROPHYLL A FLUORO CHLOROPHYLL A SPECTRO ACID CORRECTED CHLOROPHYLL A SPECTRO UNCORRECTED CHLOROPHYLL A TRICHROM UNCORRECTED DISSOLVED ORGANIC PHOSPHORUS AS P - FILTERED DISSOLVED PHOSPHORUS AS P - FILTERED PERIPHYTON CHLOROPHYLL A CHROMA-FLUORO PERIPHYTON CHLOROPHYLL A FLUORO PERIPHYTON CHLOROPHYLL A SPECTRO UNCORRECTED DISSOLVED OXYGEN DISSOLVED OXYGEN CONCENTRATION DISSOLVED OXYGEN PERCENT SATURATION PRODUCTIVITY PRODUCTIVITY AS C SALIINITY SALINITY SALINITY - DAILY MEAN SECCHI SECCHI DEPTH TURBIDITY TURBIDITY - LAB TURBIDITY - UNFILTERED TURBIDITY AS SI02 - UNFILTERED TEMPERATURE WATER TEMPERATURE The spatial distributions of these standardized parameters are shown in Attachments F and G. Battene The Business of Innovation ------- Final Report Work Assignment 1-01 Task 7 September 2004 Page 18 The following sections summarize data richness and distribution within Alabama, Louisiana, and Mississippi. 4.2 Alabama Data distribution within Alabama coastal waters is illustrated in several summaries. Table 4.3 (below) contains an aggregated summary of total observations in Alabama waterbodies. Analytical species of nitrogen, phosphorus and chlorophyll were aggregated as shown in Table 4.2. The total observations are indicated for each Alabama HUC and these have been placed in descending order; from the greatest to least total number of observations. Attachments F and G provide visual distribution of total stations within the Alabama portion of the project study area (Attachment F) and relative magnitudes among the coastal HUCs (magnitudinal centroids shown in Attachment G). Temporal distributions are indicated in Attachment H where information on total annual observations are provided for each HUC. Attachment I summarizes the total counts for all primary standardized parameters (non-aggregated). Table 4.3 indicates a significantly high overall percentage of DO, salinity, temperature, and turbidity observations. However, these are the result of a continuous monitoring program which utilized an in situ data logger (e.g., CTD). This is a good example of how summaries of water quality data can be biased based on the application of different monitoring practices. Clearly the thousands of continuously monitored data are valuable, but not necessarily comparable to other historical and current discrete data in terms of identifying data gaps. If we disregard this first row of Table 4.3, we see that there are approximately 7 to 10 HUCs that possess relatively high total numbers of all primary parameters. The remaining HUCs possess various combinations of observations that are shown in descending order. One interesting aspect of the Alabama data set is that phytoplankton productivity was measured quite rigorously. The Pennock et al. Mobile Bay and Weeks Bay cruises in the late 1980s and 1990s resulted in a very good spatial and temporal survey of productivity, standing stock, DO, nutrients, and physical factors. This type of surveying, which encompasses many key nutrient response characteristics, are not as common in other regions of the study area. Baiteiie The Business of Innovation ------- Table 4-4. Ranking of Alabama 12-digit and 10-digit coastal HUCs by total records of primary causal/response parameters. Also shown are total records for all nitrogen (N), phosphorus (P) and chlorophyll (Chi) analytical values, and phytoplankton productivity (Prod), dissolved oxygen concentration and percent saturation (DO), Secchi depth, turbidity, salinity and water temperature (Temp). ST HUC HUC Name Basin Name Totals N p Chi Prod DO Secchi Turbidity Salinity Temp AL 3160205060 Magnolia River Mobile 425,730 0 0 0 0 170,292 0 85,146 85,146 85,146 AL 031602050307 Lower Fish River MOBILE BAY 15,514 5,416 2,645 1,362 424 1,767 1,121 32 1,344 1,403 AL 031602050104 Bon Secour Bay MOBILE BAY 8,210 2,844 1,121 669 210 1,242 413 23 841 847 AL 031602050102 Middle Mobile Bay MOBILE BAY 6,483 2,068 807 434 175 1,152 248 145 706 748 AL 031602050101 Upper Mobile Bay MOBILE BAY 6,444 2,196 803 485 144 1,174 298 21 618 705 AL 031602050103 Lower Mobile Bay MOBILE BAY 4,889 1,688 632 382 109 844 185 28 510 511 AL 031700090201 Mississippi Sound MISSISSIPPI 2,715 807 269 142 40 624 99 62 329 343 COASTAL AL 031602040201 Mittlin Lake MOBILE-TENSAW 2,084 17 15 0 0 1,349 0 14 0 689 AL 031602050309 Skunk Bayou MOBILE BAY 1,984 691 352 168 50 241 142 0 170 170 AL 031602050310 Bon Secour River MOBILE BAY 1,854 14 6 0 0 975 0 2 366 491 AL 031602040505 Tensaw River- MOBILE-TENSAW 1,139 152 50 0 0 580 0 42 16 299 Apalachee River AL 031602040303 Grand Bay MOBILE-TENSAW 832 101 35 0 0 419 0 40 14 223 AL 031602040404 Lower Chasaw Creek MOBILE-TENSAW 664 124 41 0 0 278 0 28 54 139 AL 031602050308 Magnolia River MOBILE BAY 540 220 84 35 0 68 9 15 36 73 AL 031602050204 Lower Dog River MOBILE BAY 513 2 1 1 0 145 1 115 117 131 AL 031602040105 Big Chippewa Lake MOBILE-TENSAW 497 98 41 0 0 201 0 4 0 153 AL 031602050301 Yancey Branch MOBILE BAY 484 0 0 0 0 121 0 121 121 121 AL 031602040402 Seabury Creek MOBILE-TENSAW 477 66 3 0 0 87 0 0 0 321 AL 031602050304 Upper Fish River MOBILE BAY 432 159 67 0 0 27 0 13 0 166 AL 031602040504 Three Mile Creek MOBILE-TENSAW 297 65 20 0 0 128 0 20 0 64 AL 031401070205 Little Lagoon PERDIDO BAY 249 35 20 0 0 37 0 5 115 37 AL 031602040501 Upper Bay Minette MOBILE-TENSAW 236 120 50 0 0 28 0 14 0 24 Creek AL 031502040705 Majors Creek LOWER ALABAMA 189 101 45 0 0 13 0 14 0 16 AL 031602050302 Fly Creek MOBILE BAY 174 14 6 0 0 40 0 38 36 40 AL 031602040101 Barrow Creek MOBILE-TENSAW 159 0 0 0 0 106 0 0 0 53 AL 031602040203 The Basin MOBILE-TENSAW 96 7 3 0 0 54 0 1 0 31 AL 031602050311 Oyster Bay MOBILE BAY 86 0 0 0 0 1 0 0 84 1 AL 031602040103 Upper Cedar Creek MOBILE-TENSAW 79 0 0 0 0 0 0 0 0 79 AL 3160205010 Mobile Bay Mobile 77 0 0 0 0 22 11 0 22 22 ------- ST HUC HUC Name Basin Name Totals AL 031602040104 Lower Cedar Creek MOBILE-TENSAW 62 AL 031602050206 Fowl River MOBILE BAY 60 AL 031602050201 Garrows Bend- MOBILE BAY 60 Mobile Bay AL 031602040401 Meekers Creek MOBILE-TENSAW 60 AL 031602040302 Bayou Sara MOBILE-TENSAW 51 AL 031602050205 Deer River MOBILE BAY 44 AL 031700090103 Grand Bay Swamp MISSISSIPPI 42 COASTAL AL 031602050303 Gum Swamp MOBILE BAY 40 AL 031602050202 Upper Dog River MOBILE BAY 36 AL 3170009030 Mississippi Sound Escatawpa 35 AL 031700080302 Powell Creek Escatawpa 25 AL 031700090202 Dauphin Island MISSISSIPPI 18 COASTAL AL 031602050306 Middle Fish River MOBILE BAY 15 AL 031602040202 Rains Creek MOBILE-TENSAW 15 AL 3170009020 Dauphin Island Escatawpa 14 AL 031700090203 Pelican Bay MISSISSIPPI 13 COASTAL AL 031602040102 Halls Creek MOBILE-TENSAW 13 AL 3170009040 West Fowl River Escatawpa 7 AL 3160205050 Fish River Mobile 7 AL 3160204060 Three Mile Creek Mobile 7 AL 3160204040 Lower Tensaw River Mobile 7 AL 3140107040 Wolf Creek Perdido-Escambia 7 AL 031602040403 Eightmile Creek MOBILE-TENSAW 2 Total Counts 483,767 % Chi Prod DO Secchi Turbidity Salinity Temp 2 0 8 0 3 6 0 0 16 0 7 0 7 6 0 6 7 0 0 0 0 0 0 17 15 12 0 29 9 28 10 8 10 0 15 4 0 1 3 0 10 4 0 0 0 1 1 0 0 1 0 0 0 0 0 0 0 15 12 0 0 9 0 10 0 10 0 18 0 0 4 2 0 2 2 2 2 2 0 43 15 12 60 15 9 14 10 4 10 4 0 2 3 4 2 1 2 2 2 2 2 2 17,073 7,147 3,686 1,152 182,180 2,545 85,983 90,735 93,266 4% 1% 1% 0% 38% 1% 18% 19% 19% ------- Final Report Work Assignment 1-01 Task 7 September 2004 Page 21 4.3 Louisiana Data distribution within Louisiana coastal waters is illustrated in several summaries. Table 4.4 (below) contains an aggregated summary of total observations in Louisiana waterbodies. Analytical species of nitrogen, phosphorus and chlorophyll were aggregated as shown in Table 4.2. The total observations are indicated for each Louisiana HUC and these have been placed in descending order; from the greatest to least total number of observations. Attachments F and G provide visual distribution of total stations within the Louisiana portion of the project study area (Attachment F) and relative magnitudes among the coastal HUCs (magnitudinal centroids shown in Attachment G). Temporal distributions are indicated in Attachment H where information on total annual observations are provided for each HUC. Attachment J summarizes the total counts for all primary standardized parameters (non-aggregated). The Louisiana HUCs indicate the highest concentration of nutrient criteria data in nitrogen (27%), DO (9%), turbidity (18%), salinity (26%) and temperature (18%). Currently the project database does not posses any phosphorus or productivity data in Louisiana. Also, relatively few records associated with chlorophyll (1%) are present. Baiteiie The Business of Innovation ------- Table 4-5. Ranking of Louisiana coastal HUCs by total records of primary causal/response parameters. Also shown are total records for all nitrogen (N), phosphorus (P) and chlorophyll (Chi) analytical values, and phytoplankton productivity (Prod), dissolved oxygen concentration and percent saturation (DO), Secchi depth, turbidity, salinity and water temperature (Temp). ST HUC HUC Name Totals N P Chi Prod DO Secchi Turbidity Salinity Temp LA LA070301 LOWER MISSISSIPPI-BATON ROUGE 11,699 4,836 0 20 0 1,239 0 1,452 18 4,134 LA LA042202 LAKE PONTCHARTRAIN 3,982 906 0 0 0 0 0 2,050 653 373 LA LA042001 LAKE PONTCHARTRAIN 2,984 893 0 0 0 464 22 1,295 22 288 LA LA042202 PONTCHARTRAIN 2,697 0 0 0 0 23 14 0 2,607 53 LA LA041702 LAKE PONTCHARTRAIN 2,177 696 0 70 0 514 212 291 167 227 LA LA041701 PONTCHARTRAIN 2,168 0 0 0 0 2 1 0 2,159 6 LA LA042208 PONTCHARTRAIN 2,057 0 0 0 0 2 1 0 2,025 29 LA LA070401 MISSISSIPPI 1,348 680 0 100 0 189 11 182 24 162 LA LA041702 PONTCHARTRAIN 960 0 0 0 0 0 0 0 953 7 LA LA041601 LAKE PONTCHARTRAIN 839 469 0 65 0 123 0 126 1 55 LA LA042208 LAKE PONTCHARTRAIN 671 151 0 35 0 34 0 215 158 78 LA LA042003 LAKE PONTCHARTRAIN 523 224 0 0 0 0 0 299 0 0 LA LA041805 LAKE PONTCHARTRAIN 489 70 0 0 0 246 0 28 0 145 LA LA042101 PONTCHARTRAIN 488 0 0 0 0 0 0 0 238 250 LA LA042102 LAKE PONTCHARTRAIN 444 7 0 0 0 0 0 167 169 101 LA LA041401 LAKE PONTCHARTRAIN 335 159 0 0 0 8 0 167 0 1 LA LA041901 LAKE PONTCHARTRAIN 278 171 0 38 0 35 0 34 0 0 LA LA042207 LAKE PONTCHARTRAIN 264 160 0 33 0 36 0 35 0 0 LA LA041601 PONTCHARTRAIN 244 0 0 0 0 0 0 0 242 2 LA LA042201 PONTCHARTRAIN 219 0 0 0 0 50 27 0 50 92 LA LA070301 MISSISSIPPI 181 0 0 0 0 46 23 0 45 67 LA LA042001 PONTCHARTRAIN 156 0 0 0 0 30 15 0 30 81 LA LA041501 LAKE PONTCHARTRAIN 142 106 0 3 0 14 0 17 0 2 LA LA042201 LAKE PONTCHARTRAIN 121 83 0 0 0 18 0 19 0 1 LA LA041701 LAKE PONTCHARTRAIN 94 51 0 1 0 18 0 14 0 10 LA LA042206 LAKE PONTCHARTRAIN 73 59 0 0 0 9 0 5 0 0 LA LA090202 LOWER PEARL 66 57 0 0 0 6 0 0 0 3 LA LA070404 LOWER MISSISSIPPI-NEW ORLEANS 65 57 0 0 0 4 0 4 0 0 LA LA042003 PONTCHARTRAIN 63 0 0 0 0 0 0 0 0 63 LA LA090102 LOWER PEARL 46 0 0 0 0 28 0 4 0 14 LA LA042102 PONTCHARTRAIN 44 0 0 0 0 0 0 0 0 44 LA LA070404 MISSISSIPPI 39 0 0 0 0 0 0 0 0 39 ------- ST HUC HUC Name Totals LA LA042203 PONTCHARTRAIN 35 LA LA042205 PONTCHARTRAIN 31 LA LA041703 PONTCHARTRAIN 30 LA LA070403 MISSISSIPPI 30 LA LA042207 PONTCHARTRAIN 27 LA LA070402 MISSISSIPPI 26 LA LA041801 LAKE PONTCHARTRAIN 21 LA LA042203 LAKE PONTCHARTRAIN 21 LA LA042206 PONTCHARTRAIN 20 LA LA042104 PONTCHARTRAIN 18 LA LA041704 PONTCHARTRAIN 15 LA LA090208 PEARL 15 LA LA042004 PONTCHARTRAIN 13 LA LA040910 PONTCHARTRAIN 12 LA LA042204 PONTCHARTRAIN 10 LA LA090103 PEARL 10 LA LA041901 PONTCHARTRAIN 9 LA LA020802 BARATARIA 7 LA LA041703 LAKE PONTCHARTRAIN 7 LA LA041801 PONTCHARTRAIN 4 LA LA070403 LOWER MISSISSIPPI-NEW ORLEANS 4 LA LA042103 PONTCHARTRAIN 3 LA LA090202 PEARL 3 LA LA042004 LAKE PONTCHARTRAIN 2 LA LA070601 MISSISSIPPI 2 LA LA040910 LAKE PONTCHARTRAIN 1 LA LA041401 PONTCHARTRAIN 1 LA LA041704 LAKE PONTCHARTRAIN 1 LA LA042101 LAKE PONTCHARTRAIN 1 0 0 0 0 0 0 9 0 0 0 0 0 0 0 0 0 0 0 7 0 0 0 0 2 0 1 0 1 1 Total % 36,335 9,876 27% P Chi Prod DO Secchi Turbidity Salinity Temp 0 0 0 4 2 0 4 25 0 0 0 0 0 0 0 31 0 0 0 0 0 0 0 30 0 0 0 0 0 0 0 30 0 0 0 4 1 0 4 18 0 0 0 0 0 0 0 26 0 1 0 1 0 0 0 0 0 0 0 0 0 11 0 0 0 0 0 2 1 0 2 15 0 0 0 0 0 0 0 18 0 0 0 2 1 0 2 10 0 0 0 2 1 0 2 10 0 0 0 0 0 0 0 13 0 0 0 0 0 0 0 12 0 0 0 0 0 0 0 10 0 0 0 2 1 0 2 5 0 0 0 2 1 0 2 4 0 0 0 2 1 0 2 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 4 0 0 0 0 0 4 0 0 0 0 0 0 0 0 0 3 0 0 0 0 0 0 0 3 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 366 0 3,159 335 6,419 9,581 6,599 0% 1% 0% 9% 1% 18% 26% 18% ------- Final Report Work Assignment 1-01 Task 7 September 2004 Page 24 4.4 Mississippi Data distribution within Mississippi coastal waters is illustrated in several summaries. Table 4.5 (below) contains an aggregated summary of total observations in Mississippi waterbodies. Analytical species of nitrogen, phosphorus and chlorophyll were aggregated as shown in Table 4.2. The total observations are indicated for each Mississippi HUC and these have been placed in descending order; from the greatest to least total number of observations. Attachments F and G provide visual distribution of total stations within the Mississippi portion of the project study area (Attachment F) and relative magnitudes among the coastal HUCs (magnitudinal centroids shown in Attachment G). Temporal distributions are indicated in Attachment H where information on total annual observations are provided for each HUC. Attachment K summarizes the total counts for all primary standardized parameters (non-aggregated). The data distribution in the Mississippi HUCs is concentrated within nitrogen (26%), phosphorus (14%), DO (28%) and temperature (23%). The other parameters are relatively underrepresented. However, there is a relatively good coverage of the database, i.e., some nutrient and DO data available for many of the HUCs. Biological response information such as chlorophyll and productivity are significantly low, in addition to information on water column light transparency (Secchi and turbidity). Baiteiie The Business of Innovation ------- Table 4-6. Ranking of Mississippi coastal HUCs by total records of primary causal/response parameters. Shown are total records for all nitrogen (N), phosphorus (P) and chlorophyll (Chi) analytical values, and phytoplankton productivity (Prod), dissolved oxygen concentration and percent saturation (DO), Secchi depth, turbidity, salinity and water temperature (Temp). ^ !3 o as' & S. ST HUC HUC Name Basin Name Totals N P Chi Prod DO Secchi Turbitity Salinity Temp MS 3170009 West Mississippi Sound MISSISSIPPI COASTAL 6,148 2,074 1,539 0 0 1,067 43 71 531 823 MS 031700060303 N/A PASCAGOULA 4,965 1,072 511 0 0 1,663 38 114 142 1,425 MS 031700090309 Bayou Talla- Jourdan River MISSISSIPPI COASTAL 3,259 476 213 1 0 1,253 13 88 174 1,041 MS 031700090704 N/A MISSISSIPPI COASTAL 3,159 557 247 0 0 1,128 34 68 100 1,025 MS 031700090703 Flat Branch- Turkey Creek MISSISSIPPI COASTAL 2,394 703 274 0 0 651 55 102 86 523 MS 031700099999 N/A MISSISSIPPI COASTAL 2,204 531 281 0 0 632 80 98 205 377 MS 031700060302 Big Swamp- Pascagoula River PASCAGOULA 1,712 136 218 0 0 625 0 0 101 632 MS 3170009 East Mississippi Sound MISSISSIPPI COASTAL 1,659 455 340 1 0 444 1 41 100 277 MS 031700090203 Polar Branch-Wolf River MISSISSIPPI COASTAL 1,348 764 275 0 0 126 0 62 0 121 MS 031700090403 Cutoff Bayou- Jourdan River MISSISSIPPI COASTAL 1,168 267 91 0 0 387 19 38 63 303 MS 031700090702 Bayou Portage MISSISSIPPI COASTAL 1,166 472 148 0 0 262 19 14 40 211 MS 031700090706 Old Fort Bayou- Biloxi Bay MISSISSIPPI COASTAL 905 420 165 0 0 138 0 35 17 130 MS 031700090801 Bayou Cumbest MISSISSIPPI COASTAL 784 247 112 0 0 219 0 30 26 150 MS 031700090605 Parker Creek- Tchoutacabouffa River MISSISSIPPI COASTAL 697 148 93 6 0 222 0 1 26 201 MS 031800040704 Mulatto Bayou-Pearl River LOWER PEARL 588 142 55 1 0 174 16 58 70 72 MS 031700090701 Flat Branch- Bayou Bernard MISSISSIPPI COASTAL 529 187 50 1 0 146 1 2 0 142 MS 031700080706 Cunningham Branch- Escatawpa River ESCATAWPA 413 113 81 0 0 106 0 21 26 66 MS 031700090602 Bayou Costapia- Tchoutacabouffa River MISSISSIPPI COASTAL 339 115 41 15 0 96 0 24 0 48 MS 031700090506 Mill Creek- Biloxi River MISSISSIPPI COASTAL 266 26 52 0 0 91 0 0 26 71 MS 031700090308 Rotten Bayou MISSISSIPPI COASTAL 181 6 2 0 0 86 0 2 4 81 MS LA92LR14 East Mississippi Mississippi Coastal 163 12 24 0 0 38 13 0 38 38 ------- ST HUC HUC Name Basin Name Totals N P Chi Prod DO Secchi Turbitity Salinity Temp Sound MS 031700090402 Lower Bayou la Croix MISSISSIPPI COASTAL 120 24 4 0 0 46 0 4 6 36 MS 031700060301 Black Creek Swamp- Pascagoula River PASCAGOULA 114 30 6 0 0 34 0 5 0 39 MS 03170009 N/A MISSISSIPPI COASTAL 91 0 0 0 0 26 13 0 26 26 MS 031700090604 Choctaw Creek- Tuxachanie Creek MISSISSIPPI COASTAL 80 32 8 4 0 16 4 0 8 8 MS 031800040703 Upper Mulatto Bayou LOWER PEARL 57 0 0 0 0 38 0 0 0 19 MS 031800040702 West Pearl River- Pearl River LOWER PEARL 43 1 0 0 0 26 0 2 0 14 MS 031700090705 Old Fort Bayou MISSISSIPPI COASTAL 34 14 3 0 0 6 1 3 0 7 MS 031700090501 Horse Creek- Biloxi River MISSISSIPPI COASTAL 30 0 0 0 0 19 0 1 0 10 MS 031700090505 Lower Little Biloxi River MISSISSIPPI COASTAL 30 0 0 0 0 20 0 0 0 10 MS 031700090202 Bell Creek-Wolf River MISSISSIPPI COASTAL 26 20 0 0 0 0 0 0 0 6 MS 031700090307 Bayou La Terre MISSISSIPPI COASTAL 25 7 2 0 0 8 0 2 4 2 MS 031800040701 Mikes River LOWER PEARL 22 0 0 0 0 12 0 4 0 6 MS 031700060107 Indian Creek- Pascagoula River PASCAGOULA 18 6 0 0 0 1 0 0 0 11 MS 031700090201 Crane Creek MISSISSIPPI COASTAL 18 8 2 0 0 4 0 2 0 2 MS 031700090302 Catahoula Creek MISSISSIPPI COASTAL 18 8 2 0 0 4 0 2 0 2 MS 031700090303 White Cypress Creek- Hickory Creek MISSISSIPPI COASTAL 17 7 2 0 0 4 0 2 0 2 MS 031700060203 Fourmile Branch- Moungers Creek PASCAGOULA 16 7 1 0 0 4 0 2 0 2 MS 031700060204 Paige Bayou- Moungers Creek PASCAGOULA 15 7 1 0 0 4 0 1 0 2 MS 031700060303 West Pascagoula River-East Pascagoula River PASCAGOULA 7 0 0 0 0 2 1 0 2 2 MS 031700090704 Back Bay of Biloxi MISSISSIPPI COASTAL 7 0 0 0 0 2 1 0 2 2 MS 031700099999 N/A MISSISSIPPI COASTAL 7 0 0 0 0 2 1 0 2 2 Total Counts 34,842 9,094 4,843 29 0 9,832 353 899 1,825 7,967 % 26% 14% 0% 0% 28% 1% 3% 5% 23% o K> O 4-- ------- Final Report Work Assignment 1-01 Task 7 September 2004 Page 27 5.0 TOWARD l lll DEVELOPMENT OF NUTRIENT CRITERIA The ultimate goal of this project is to provide the best available means to support the states of Louisiana, Mississippi and Alabama in the development of nutrient criteria for coastal waters. The Steering Committee has determined that one essential step toward developing nutrient criteria is to focus on the identification of reference conditions for estuarine and other coastal waters of interest within the study area. Reference conditions support comparative analyses among groups of similarly functioning ecosystems such that responses to agents of change (e.g., nutrient delivery) can be equated. Therefore, relative ecological states can be assessed and management sectors can be well-informed of likely responses to management decisions. The term reference condition can have multiple meanings. EPA (2001) defines reference conditions to be "...the comprehensive representation of data from several similar, minimally impacted, 'natural' sites on a waterbody or from within a similar class of waterbodies..." The use of reference conditions in some ecosystem restoration activities (e.g., salt marsh restoration) are often used to track relative changes in ecosystem function from some baseline state which is not necessarily pristine or uninfluenced by human activities. This is often referred to as functional equivalency. Reference sites can be paired, or multiple sites can be combined to create functional equivalency assessments over large areas (Steyer et al. in press). For estuarine and coastal waterbodies, the definition for reference conditions is focused on the assessment of a desired, or target condition. These desired conditions may be based on existing unimpacted (or pristine) sites, or some knowledge of historical conditions at individual sites. The determination of reference conditions is key to the development of nutrient criteria. The process is dependent upon the classification of estuarine and coastal waters into similar groups. The approach of "one size fits all" has proven futile in most ecological characterizations. This is particularly true for estuarine systems where biophysical variability is relatively high (Costanza et al. 1993). Therefore, a series of steps aimed at normalizing whole systems to a set of controlling variables is required to enhance comparability across space and time. There are several important factors associated with establishing reference conditions in ecological studies. These include (but are not limited to) scale, characterization (system type), and statistical power: 5.1 Scale Temporal and spatial scales play a significant role in ecological comparative analysis. Temporal scale is important in determining previous, or pristine, conditions, as well as forward-moving trends. Temporal scale is also critical for understanding responses to management activities (i.e., nutrient reduction practices). Spatial scale is extremely important in defining and understanding coastal ecosystem processes. For example, local vs. regional conditions vary but can be strongly or weakly associated based on their level of "communication." 5.2 Characterization/typology Physical characteristics such as geomorphologic properties and sediment type are increasingly being used to normalize ecosystem functional behavior in coastal systems (Boyd et al. 1992, Dalrymple et al. 1992, EPA 2001, Bricker et al. 2003). For instance, shallow lagoon systems that exhibit strong benthic-pelagic coupling behave dramatically different than deeper, river-dominated systems. Grouping similar types of coastal systems is required to perform reasonable comparisons over space and time. There are several efforts currently underway that are associated with the process of system characterization and typing that are highlighted in Section 3.1. Baiteiie The Business of Innovation ------- Final Report Work Assignment 1-01 Task 7 September 2004 Page 28 5.3 Statistical Power Comparative analyses are only as good as their ability to interpret significant differences among causal and responsive factors. In order to assess trends or differences to some quantifiable degree of certainty, minimum numbers and distributions of observations are necessary. Historical data applied to reference condition assessment may not always represent conditions in a statistically sound manner (i.e., inappropriate sampling frequency, replication in time and space, etc.). Future monitoring activities should be planned such that their results fit into an appropriate sound statistical design. Issues associated with scale, characterization/typology and statistical power are inherent in several documents that provide guidance for developing reference conditions in coastal systems. The following subsections highlight and summarize four examples of methodological review and guidance: 1. The EPA Nutrient Criteria Technical Guidance Manual (EPA 2001); 2. NOAA's National Estuarine Eutrophication Assessment (Bricker et al. 1999, Bricker et al. 2003, Bricker et al. in press); 3. The Louisiana Coast-wide Reference Monitoring System (CRMS) (Steyer et al. in press), and; 4. Australia's National Estuaries Assessment and Management (NE) Project (Heap et al. 2001, Ryan et al. 2003). 5.4 Nutrient Criteria Technical Guidance Manual: Estuarine and Coastal Marine Waters (EPA 2001) This project is consistent with much of the direction contained within the Nutrient Criteria Technical Guidance Manual for Estuarine and Coastal Marine waters (EPA, 2001). There are several chapters devoted to background and historical research, monitoring, and modeling of nutrient enrichment to coastal water bodies. The key chapters associated with the development of reference conditions for estuaries and coastal waters in Northern Gulf of Mexico region include Chapter 3 (Classification of Estuarine and Coastal Waters), Chapter 6 (Determining the Reference Condition), and Chapter 7 (Nutrient and Algal Criteria Development). Chapter 3 deals specifically with classification of estuarine and coastal waters. According to the manual, the purpose of classification is stated: "Where possible, similar estuaries, tributaries, or coastal reaches should be equated through physical classification to reduce the magnitude of the criteria development problem and to enhance predictability of management responses. To be useful, classification should reduce variability of ecosystem-related measures (e.g., water quality factors) within identified classes and maximize interclass variability. This is important because managers need to understand how different types of estuaries and coastal waters, as well as important habitat differences within these systems, respond to nutrient over-enrichment in or to plan effective management strategies." Several classification schemes are reviewed in this manual. These classification schemes are introduced in conjunction with a series of factors associated with determining the susceptibility of estuarine systems to nutrient loading. These factors, modified from the NRC (2000), include the following: 1. System dilution and water residence time, or flushing rate 2. Ratio of nutrient load per unit area of estuary Baiteiie The Business of Innovation ------- Final Report Work Assignment 1-01 Task 7 September 2004 Page 29 3. Vertical mixing and stratification 4. Algal biomass (phytoplankton and macroalgae) 5. Wave exposure 6. Depth distribution (hypsography) 7. Ratio of side embayment volume to open estuary (local vs. regional communication) With the exception of nutrient load (2) and algal biomass (4), all of these factors are physical influences. These physical factors, in conjunction with nutrient loading rates, are significant determinants in how primary (e.g., algal production) and secondary (e.g., consumers, system metabolism, light attenuation) factors respond to nutrient loads. Other physical factors include temperature, surface irradiance, and meteorological events. These are assumed to remain relatively consistent at the regional or subregional scale. Although the guidance manual does not endorse any single method of classification, it does offer several examples including geomorphic classification, man-made estuaries, physical and/or hydrodynamic factor- based classifications, and other considerations such as habitat type or theoretical issues. These classification approaches vary from the relatively simple (geomorphologic) to complex (hydrodynamic). Geomorphic classification provides a first-order estimate of physical characteristics. The manual identifies the four main groups including: coastal plain estuaries, lagoons, fjords and fjord-like estuaries, and tectonically caused estuaries (Pritchard 1955, 1967; Dyer 1973). Man-made estuaries are estuarine, or near-estuarine systems such as drainage canals, boat basins, and other types of waterbodies resulting directly from anthropogenic activities. They are noted because they have different physical evolutionary histories and different response characteristics than natural coastal features. Physical and/or hydrodynamic classifications are based on physical or hydrodynamic factors. Approaches for classifying estuaries using this approach employs parameters such as stratification, mixing, circulation, water residence time, river flow, tides, waves, and tidal amplitude. NOAA also has a scheme for determining estuarine susceptibility which is summarized below (Section 3.2) and one other approach for classifying estuaries based on physical/hydrodynamic parameters uses comparative systems empirical modeling. Classification using other types of estuarine categories involves evaluations of habitat type or theoretical considerations. Since system response to nutrient enrichment varies considerably, the National Nutrient Criteria Program (EPA 2001) suggests measuring the nutrient characteristics of minimally impacted sites as a reliable goal. Chapter 6 (Determining the Reference Condition) of the guidance manual compares and summarizes the identification of "pristine" vs. minimally impacted waters. The determination of reference conditions requires some level of appropriate comparison; that is, a minimum number of similar systems to compare. Five concentration-based approaches for establishing reference conditions are presented along with a series of relative levels of estuarine degradation (see Table 6-1 in the manual). The guidance manual describes potential approaches for determining reference conditions by applying historical data, paired or multisystem comparisons, and dose-response models. In all cases, the point of developing the reference condition in any system is analogous to establishing a target or minimum water quality metric for either TN, TP, Chlorophyll-a, Secchi depth, or some combination of these parameters based on site-specific characteristics. Chapter 7 (Nutrient and Algal Criteria Development) of the guidance manual suggests a generalized sequence of milestones in the development of nutrient criteria. Several aspects of this sequence may run simultaneously; however, they should progress as follows: Baiteiie The Business of Innovation ------- Final Report Work Assignment 1-01 Task 7 September 2004 Page 30 1. Historical Information. The identification, collection and review of historical, or antecedent, conditions may provide immediate and useful information related to historically-driven reference conditions. 2. Development of Reference Condition. Resource managers should identify, if possible, a set of reference sites or historical reference conditions. 3. Models. Estimates of dose-response relationships, historical nutrient loads, and the management requirements associated with nutrient reduction can be made through the development of models. Models may range in complexity from simple, empirical regressions to time-dependent, spatially-discrete ecosystem response models. 4. Antidegradation Policy and Allowable Downstream Effects. Nutrient criteria must not change due to the degradation of reference conditions, and downstream (e.g., shelf) water quality should be considered in the development of nutrient criteria for estuarine and near-shore systems. 5. Regional Technical Assistance Group (RTAG). A group of regional experts should be assembled to work with the States in the review of water quality data, development of monitoring plans, and the development of nutrient criteria. 5.5 NOAA National Estuarine Eutrophication Assessment The National Estuarine Eutrophication Assessment (NEEA) is an approach that surveyed the characteristics of 16 nutrient related water quality parameters within 82 U.S. estuaries (Bricker et al. 1999). The results of the assessment provided a series of qualitative indicators for these estuarine systems, and these were compiled into regional condition assessments. NOAA sent a survey to approximately 400 coastal scientists and managers in the U.S. This survey asked that the scientists assess regional coastal conditions based on the set of water quality parameters. Based on the responses from the survey, a series of qualitative indicators were developed that illustrated relative levels (e.g., high, medium, low, no expression, improving, and insufficient data) of the following primary (Chlorophyll a, epiphytes, and macroalgae) and secondary (low dissolved oxygen, SAV loss, and nuisance/toxic algal blooms) symptoms of eutrophication. An overall eutrophic condition was then assessed based on the average of the magnitudes of these symptoms. This work also produced a eutrophication survey of the Gulf of Mexico (NOAA 1997). The eutrophication survey contains summaries of water quality (e.g., nutrients, dissolved oxygen, algae), physical processes, and physical characteristics (e.g., salinity zones, tidal zones) of the region. This summary also expresses the relative degree of primary and secondary symptoms in a manner consistent with the overall assessment report (Bricker et al. 1999). NOAA is currently planning the next steps of this National Assessment (S. Bricker, personal communication 1/04). These are currently in draft condition, and include the following: 1. Identification of a Steering Group to guide further development of the monitoring and assessment program. 2. Determination of members for several groups including "Typology" and "Monitoring, assessment and classification" (including reference conditions and classification issues, such as variable thresholds), "Modeling and Management" (for development of operational framework and analytical tools). 3. Drafting of the scope-of-work for each of the groups, remembering that overlap among groups is key. Baiteiie The Business of Innovation ------- Final Report Work Assignment 1-01 Task 7 September 2004 Page 31 4. Identification and contact of appropriate economists and social scientists for participation and guidance on development of the economic impacts/society's uses component. 5. Compilation of a summary of data in preparation for selection of candidate estuaries for a pilot study (includes searches for models of susceptibility and input/response, such as that for Tampa Bay). 6. Begin designing pilot studies. 5.6 Louisiana Coast-wide Reference Monitoring System (CRMS) The Louisiana Coast-wide Reference Monitoring System (CRMS) is currently a proposed project focusing on an integrated, statistically-designed, 4-5 year coastal wetland monitoring program. There are two primary objectives associated with this proposed work. The first objective is to determine, collectively, the effectiveness of the projects carried out under the Breaux Act restoration plan. This would be done by providing a network or "pool" of reference sites by which to evaluate project effectiveness, thereby eliminating the need for one-on-one pairs of reference and project areas. The second objective of CRMS is to determine the ecological condition of the coastal wetlands based on the variables measured to ensure that the strategic coastal plan for Louisiana (Coast 2050) is effective in recreating a sustainable coastal ecosystem (Steyer et al. in press). A series of questions are related to these two objectives: 1. Did the projects carried out under the Breaux Act contribute to reducing coastal wetland loss in Louisiana? 2. Which project types are the most effective in creating, restoring, protecting and enhancing wetlands? 3. Is the Breaux Act effective in reducing the major stressors on wetlands (i.e., flooding regime, salinity, marsh elevation change)? 4. Are Louisiana's coastal wetlands accumulating vertically at a sufficient rate to achieve sustainability? 5. Are estuarine salinity gradients being maintained to sustain a diversity of vegetation types in the hydrologic basins? 6. Has the coastal landscape been maintained with sufficient hydrologic linkages to support efficient exchange and interface between the marshes and estuaries? The strategy formed to answer these questions and to achieve the two primary objectives is based on a state-wide system of reference sites. These would be characterized into ecologically similar groups and statistically evaluated for changes in function over time in comparison to ongoing restoration activities (restoration sites). Steyer et al. (in press) state that pristine conditions are rare or perhaps non-existent in Louisiana and that there are very few sites, if any, that possess pre-project conditions (controls). Therefore, the project proposes a design that is a composite of several wetland functional assessment approaches including: 1. Hydrogeomorphic (HGM) 2. Functional Assessment of European Wetland Ecosystems (FAEWE) 3. EPA Ecological Monitoring and Assessment Program (EMAP) 4. Cumulative Impact Assessment (Lee and Gosselink 1988, Gosselink and Lee 1989) Baiteiie The Business of Innovation ------- Final Report Work Assignment 1-01 Task 7 September 2004 Page 32 5. Before-Afiter-Control-Impact (BACI) 6. Other miscellaneous sources The approach is based on classifying the coast into a discrete number of similarly functioning regions and basins, and into project and non-project areas. Further classification is performed based on dominant vegetation types within these areas. The monitoring design for the project includes six variables to be measured within 9 basins (project and non-project results in 18 strata). Based on the requirements implied by the Central Limit Theorem, 30 samples per stratum are being proposed; thus a total of 540 reference sites will be evaluated in the study. Monitoring information will provide an instantaneous "snapshot" that places project sites within the functional continuum of all reference sites (current conditions). Monitoring will also track the trajectory of change of project sites through time with respect to changes in functionally equivalent reference sites (accounting for regional/sub-regional natural variability). Finally, monitoring information will allow for comparisons to a group of reference standards (relative distance from desired conditions). A statistically sound, synoptic monitoring plan is being developed. Remaining design activities involve identifying an indicator aimed at evaluating responses rapidly, and testing the design in terms of its power to characterize temporal and spatial variability. Ecological functional models following HGM methodologies also need to be developed. Due to the inherent degree of coupling between Louisiana's coastal marshes and estuaries (Childers et al. 1999, Turner 2001), Steyer et al. (in press) suggest that this cumulative effort of restoration projects and parallel evaluation will "collectively lead to the general restoration of Louisiana's estuaries." 5.7 Australian Waterway Classifications and Deviation Index The National Estuaries Assessment and Management (NE) Project of Australia has made significant progress in the development of an inventory of estuarine (or waterbody) type (classification) and condition. This assessment includes 974 discrete systems throughout Australia and Tasmania. Their efforts are illustrated on their web site, the "OzEstuaries Database" (www.ozestuaries.org). The Australian NE Project is based on a geomorphic classification scheme which was developed by Harris et al. (2002). The conceptual model of this classification scheme (Figure 3) includes the general geomorphic and sedimentary environments (or habitats) of seven classes: 1. Embayments and Drowned River Valley (EMB) 2. Wave-dominated Estuaries (WDE) 3. Wave-dominated Deltas (WDD) 4. Coastal Lagoons and Strandplain-associated Creeks (CL/SP) 5. Tide-dominated Estuaries (TDE) 6. Tide-dominated Deltas (TDD) 7. Tidal Creeks (TC) Battene The Business of Innovation ------- Final Report Work Assignment 1-01 Task 7 September 2004 Page 33 River Wave Tide Figure 5-1. Ternary classification of coastal systems divided into seven classes (after Dalryinple et al. [1992] and taken from Ryan et al. [2003]). Conceptual models, examples of systems associated with these classifications, and a full classification of all 974 water bodies are documented in Ryan et al. (2003). A report on the quantitative assessment of the health of Australia's estuaries and coastal waterways (Heap et al. 2001) includes a review of the evolution and geomorphologic types of Australian coastal systems, conceptual models of dose-response of nutrients in four classes of coastal systems, an overview of biogeochemical and physical factors affecting the nature of dose-response relationships, a series of statistical analyses of existing data, and the development of a "deviation index'' which is used to determine relative ecological states and to rank all systems for comparative analysis (reference conditions). Technical descriptions of data collection and mapping activities are also included in this assessment. The approach employed by the Australian NE Project focuses on habitat type based on sediment qualities (facies). Much emphasis is put on sediment organic matter and sediment nutrient concentrations and their use as indicators of system denitrification rates and habitat quality. The deviation index is aimed at developing a reference condition-based tool to assess target or desired conditions with the consideration of natural variability as well as ranking coastal systems in terms of their level of relative degradation. The Estuarine Geoscience Database (OzEstuaries) is an online warehouse of publications, estuarine GIS mapping, estuarine conditions assessments, estuarine water quality data, conceptual models and descriptive information, an indicators program, and other sources of information. The site allows the user to create and view site maps and Landsat images, review condition assessment reports, and review water quality data for most of the 974 coastal systems throughout Australia and Tasmania. 5.7.1 U.S EPA Aquatic Stressors Framework The EPA is currently developing a framework by which to assess coastal ecosystem responses to multiple stressors. The process is driven by the EPA Aquatic Stressors Framework (USEPA 2002) which has formed a Diagnostics Committee that oversees research and development associated with this work. The Batreiie The Business of Innovation ------- Final Report Work Assignment 1-01 Task 7 September 2004 Page 34 goal of the research is to develop tools and methods that will promote management decisions related to major aquatic stressors. At this time there are four primary stressors under evaluation: (1) nutrients, (2) suspended and bedded sediments, (3) toxics, and (4) altered habitat. A series of conceptual models are being developed that are intended to provide guidance in how coastal ecosystems respond to these multiple stressors. The ultimate goal of this work includes an improved ability to support State and Tribe 303(d) impaired waters assessments and the implementation of necessary management activities to improve water and ecosystem quality. The approach to this study involves the development of four distinct stressor-response conceptual models that possess physical and modifying factors. A series of primary physical factors tend to control, or govern, the extent of ecosystem response to agents of change. For example, water residence time, depth, hypsography, salinity, and other physical controls influence the type of response an ecosystem has to individual, or multiple, agents of change. Therefore, this framework is based on a coastal classification system whereby classifying similar biogeophysical attributes of systems will increase the ability to assess response characteristics. The classification framework (in development) is currently at the scale of 8-digit HUCs which are modifications of NOAAs estuarine and coastal drainage area (EDA & CDA) and USGS HUC classifications. Further refinement to smaller spatial units (e.g., 12-digit HUC scale) may increase the ability to assess ecological response across many systems. Therefore, this research is quite relevant to studies of nutrient criteria in coastal waters. The approaches being developed are similar to the Australian classification and indices work, but are more relevant because they are covering water bodies and contributing watersheds within the project study area and at similar spatial and temporal scales. The efforts associated with this program will likely be valuable to nutrient criteria development, and vice versa. 5.8 Discussion (change to section 5.6) In a recent paper by R.E. Turner (Turner 2001) the fundamental question is posed: are Gulf of Mexico estuaries unique? In attempting to answer this question, Turner (2001) produced a series of descriptive comparisons between Gulf estuaries and others across the U.S. He also reviewed historical and existing research on Gulf estuaries to synthesize information that would support an overall classification and comparative functional assessment associated with estuarine systems that are susceptible to eutrophication. His findings and conclusions can be summarized as follows: 1. Although the microtidal regime of Gulf estuaries sets them apart from most other U.S. regions, there is no apparent regional signature to the estuarine landscape as measured by geomorphic criteria (e.g., per unit watershed drainage, estuarine surface area, freshwater turnover time, and wetland area/water surface area). 2. Wetland biogeochemical responses to watershed and open water coupling can determine important sources, losses, and transformations of dissolved and particulate organic matter in Gulf systems (see also Childers et al. 1999). 3. It would be useful for estuarine managers to know if wetland restoration rates vary among estuaries in order to plan, allocate resources, and effectively evaluate projects. 4. One primary goal associated with the understanding of variability imposed by physics, chemistry, geology, and biology is to develop an efficient mechanism that makes comparable measurements of key parameters in many estuarine systems. 5. A field-based, cross-system comparison of estuaries that examines different landscape scales, ages, and vegetation types, etc., will further reveal the strength of these bio-geo-physico- chemical interactions. Baiteiie The Business of Innovation ------- Final Report Work Assignment 1-01 Task 7 September 2004 Page 35 6. Important physical parameters to consider are wave energy, freshwater turnover times, size, watershed morphology, and water column stratification. 7. Important biological parameters include wetland:water ratios, landscape mosaics (fragmentation, source materials, and composition), nutrient loading rates, light regime, and couplings with upstream and continental shelf systems. 8. In summary, a comparative analysis emphasizing forcing functions, rather than geographic uniqueness, will lead to significant progress in understanding how all estuaries function, are perturbed, and even how they can be restored. Thus, Turner (2001) suggests that the estuarine systems within the study area are no less variable in nature than others throughout the U.S. However, they are significantly influenced by extensive riverine discharge (e.g., Mississippi River Delta and tributaries to Mobile Bay) and relatively high ratios of wetland:estuarine surface area. He further suggests that a comparative analysis, based on focused field monitoring, will contribute positively to the understanding of system behavior, and therefore ecological responses to nutrient loads. This view is consistent with that offered by EPA (2001) and NOAA (Bricker etal. 1999). The five approaches and one review paper summarized in this section all share similar fundamental properties; however, they differ in specific methods. A synthesis and comparison of particular methods, results, and lessons learned in these and other examples would likely benefit this project. The EPA guidance manual (EPA 2001) offers a broad overview of methods and approaches available to state and tribal managers for the development of nutrient criteria. The emphasis, however, seems to be on using nutrient concentrations and Secchi depth as the primary response indicators. Additional investigations into the use of other supporting biogeochemical and biological response indicators may be necessary, particularly in cases where water column nutrient concentrations are too variable or appropriate data do not exist. The NOAA NEEA (Bricker et al. 1999) provides a broad comparison of relatively large estuarine systems in the Gulf (i.e., on the scale of Mississippi Sound, Mobile Bay) using a significant number of causal and response variables. Although this assessment includes several Gulf of Mexico coastal systems, it may be too general (large-scale) and qualitative to be directly applicable to many of the water bodies of interest within this project's study area. However, according to NOAA, further refinement of this assessment is planned and collaboration between EPA, NOAA, Louisiana, Mississippi, and Alabama would be prudent in order to maximize efficiency. This is particularly true if the NOAA NEEA work will include a national "typology" effort. Although the methods for conducting the CRMS work in Louisiana proposed by Steyer et al. (in press) are associated primarily with wetlands assessments, there is significant overlap in relevance to water quality assessments in many of the water bodies within the study area. The close coupling between coastal wetland and estuarine habitats and processes within the study area justify a close collaborative effort between the CRMS work and nutrient criteria assessments. The approach for determining reference conditions will require significant resources and effort; however, it will likely be one of the most comprehensive comparative studies to date. Similar classification and referencing methods may be considered for developing nutrient criteria in the estuarine systems of the northern Gulf of Mexico. The Australian National Estuaries Assessment and Management (NE) Project applies a geomorphologic and sedimentary type habitat classification (facies) approach to developing a cross-system comparison of ecological status. The approach to classification and comparisons is well-documented (Heap et al. 2001, Ryan et al. 2003). Condition assessments are being completed for each system based on available data. Baiteiie The Business of Innovation ------- Final Report Work Assignment 1-01 Task 7 September 2004 Page 36 Water quality data are available online through a GIS-centered, user-friendly website. This overall approach may provide a decent model for further development of the database and criteria development associated with this project. In addition, important "lessons learned" may provide added value and efficiency to further regional comparative eutrophication studies. 6.0 RECOMMENDATIONS This section is a summary of recommendations that are associated with further progress toward development of nutrient criteria in the coastal waters within the study area. 6.1 Data Gaps Analysis A thorough data gap analysis should be performed on priority and representative water bodies within the study area. A data gap analysis will result in the prioritization of candidate reference sites for comparative studies and also can be used to evaluate the need to augment and/or revise current and future monitoring plans. This will be based on at least two methods of analysis: (1) identify locations in priority coastal waters where temporal monitoring has been ceased or interrupted and (2) identify locations within priority coastal waters where monitoring should be implemented. An overall, statistically robust plan would be beneficial to the desired outcome of creating nutrient-related criteria (water quality or ecosystem functional targets) and reference sites used to quantify improvement or degradation of particular systems over time. Future monitoring may overlap or coincide with parallel efforts of other investigators and agencies. Collaboration in planning and implementation would reduce unwanted redundancy and augment efficiency. 6.2 Identify Appropriate Classification Process Coastal and estuarine waterbodies are physically and biologically complex systems. System classification is necessary in order to limit the problems of interpretation caused by natural variability among system types. There are several approaches being applied to system classification and most are directly associated with physical or developmental characteristics. The project's steering committee should review the various methods of classification and determine the most appropriate one(s). Section 5 of this report summarizes some recently developed methods related to system classification, biogeochemical states, and responses to stressors. They are all focused on determining reference conditions and methods to quantify system change attributed to management activities and independent of natural variability. Perhaps the most relevant, regionally-specific method is associated with the EPA Multiple Stressor research. Battelle has been communicating with members of the Diagnostics Committee to find relevance between projects to increase efficiency. This research has already accomplished a classification analysis within the Northern Gulf study area, but at a spatial resolution that may need to be further optimized. Their conceptual model on ecosystem responses to nutrient enrichment can be further developed and modified in combination with further site-specific system classification. Ongoing research in Europe, Australia, and other regions of the US should still be reviewed on a regular basis to capture innovative approaches and discoveries. 6.3 Further Develop Database and Information Access Program The database for this project is being developed with no specific application design, i.e., for a variety of end-users. Plans to further develop a user-friendly web interface or front-end where specific information can be easily reviewed and/or retrieved is strongly suggested. Examples of such a design exists in individual state websites, STORET, the newly created EPA National Nutrient Criteria database and the Australian NE Project's web site. An interactive database as such would provide a tool for assessing Battelle The Business of Innovation ------- Final Report Work Assignment 1-01 Task 7 September 2004 Page 37 quantitative changes in ecosystem function, or state, over time. It would also be an important resource in communicating project scope and results to critical stakeholders and similar programmatic entities. Modification of this database to make it compatible with the EPA's National Nutrient Criteria database is a high priority. Recent dialog between project steering committee members, Battelle, and EPA Headquarters has resulted in an initial assessment of good compatibility between databases. Further work on this is necessary to effectively transfer data from the Northern Gulf database to the emerging National Nutrient Criteria database. 7.0 REFERENCES CITED Boyd, R., Dalrymple, R., and Zaitlin, B. A., (1992) Classification of clastic coastal depositional environments. Sedimentary Geology. 80:139-150. Bricker, S.B., C.G. Clement, D.E. Pirhalla, S.P. Orlando, and D.R.G. Farrow. 1999. National Estuarine Eutrophication Assessment: Effects of Nutrient Enrichment in the Nation's Estuaries. NOAA, National Ocean Service, Special Projects Office and the National Centers for Coastal Ocean Science. Silver Spring, MD: 71 pp. Bricker, S.B., G. Matlock, J. Snider, and A. Mason. In Press. National Eutrophication Assessment Update: Workshop Summary and Recommendations for Development of a Long-term Monitoring and Assessment Program (Pre-Distribution Copy). September 4-5, 2002. Patuxent National Wildlife Research Refuge, Laurel, MD. Bricker, S.B., J.G. Ferreira, and T. Simas. 2003. An integrated methodology for assessment of estuarine trophic status. Ecological Modelling. 169:39-60. Childers, D.L., S.E. Davis III, R. Twilley, and V. Rivera-Monroy. 1999. Wetland-water column interactions and the biogeochemistry of estuary-watershed coupling around the Gulf of Mexico, p. 211-235. In T.S. Bianchi, J.R. Pennock, and R.W. Twilley (eds), Biogeochemistry of Gulf of Mexico Estuaries. J. Wiley & Sons, Inc., New York. Costanza, R., Kemp, W.M. & Boynton, W. 1993. Predictability, scale, and biodiversity in coastal and estuarine ecosystems : implications for management. Ambio. 22(2):88-96. Dalrymple, R. W., Zaitlin, B. A., and Boyd, R., (1992) Estuarine facies models; conceptual basis and stratigraphic implications. Journal of Sedimentary Petrology. 62:1130-1146. Dyer, K. 1973. Estuaries: A Physical Introduction. New York: Wiley-Interscience. Gosselink, J.G. and L.C. Lee. 1989. Cumulative impact assessment in bottomland hardwood forests. Wetlands. 9:83-174. Harris, P.T., A.D. Heap, S.M. Bryce, R. Porter-Smith, D.A. Ryan, and D.T. Heggie. 2002. Classification of Australian clastic coastal depositional environments based on a quantitative analysis of wave, tide and river power. Journal of Sedimentary Research. 72(6):858-870. Battelle The Business of Innovation ------- Final Report Work Assignment 1-01 Task 7 September 2004 Page 38 Heap, A., S. Bryce, D. Ryan, L. Radke, C. Smith, R. Smith, P. Harris, and D. Heggie. 2001. Australian Estuaries and Coastal Waterways: A Geoscience Perspective for Improved and Integrated Resource Management. AGSO Record 2001/07. http://www.ozestuaries.org/document/documentation.html Lee, L.C. and J.G. Gosselink. 1988. Cumulative impacts in wetlands: linking scientific assessments and regulatory alternatives. Environmental Management. 12:591-602 Monsen, N.E, J.E. Cloern, and L.V. Lucas. 2002. A comment on the use of flushing time, residence time, and age as transport time scales. Limnology & Oceanography. 47(5): 1545-1553. National Oceanographic and Atmospheric Administration (NOAA). 1997. NOAA's Estuarine Eutrophication Survey, Volume 4: Gulf of Mexico Region. Silver Spring, MD: Office of Ocean Resources Conservation and Assessment. 77 pp. NRC (National Research Council). 2000. Clean Coastal Waters: Understanding and Reducing the Effects of Nutrient Pollution. Washington, DC: National Academy Press. Pritchard, D. 1955. Estuarine circulation patterns. Proc. Am. Soc. Civil Eng. 81:717-1 - 717-11. Pritchard. D. 1967. What is an estuary: physical viewpoint. In: G. Lauff, ed. Estuaries. Washington, DC: American Association for the Advancement of Science. Publication 83. Ryan, D.A., A.D. Heap, L. Radke, and D.T. Heggie. 2003. Conceptual models of Australia's estuaries and coastal waterways: application for coastal resource management. Geoscience Australia Record 2003/09. http://www.ozestuaries .org/document/documentation.html Steyer, G.D., C.E. Sasser, J.M. Visser, E.M. Swenson, J.A. Nyman, and R.C. Raynie. In Press. A proposed coast-wide reference monitoring system for evaluating wetland restoration trajectories. Turner, R.E. 2001. Of manatees, mangroves, and the Mississippi River: Is there an estuarine signature for the Gulf of Mexico? Estuaries. 24(2): 139-150. U.S. EPA. 2001. Nutrient Criteria Technical Guidance Manual: Estuarine and Coastal Marine Waters. EPA-822-B-01-003, October 2001, U.S. Environmental Protection Agency, Office of Water, Washington DC USEPA. (2002). Aquatic Stressors: Framework and Implementation Plan for Effects Research (EPA 600/R-02/074). Research Triangle Park, NC: U. S. Environmental Protection Agency, Office of Research and Development. Baiteiie The Business of Innovation ------- |