xe/EPA United States Environmental Protection Agency Office of Water (WH-556F) EPA 503/9-91/008 July 1991 National Estuary Program ?!EADQUAHTERS LIB ,!^ ------- ------- TABLE OF CONTENTS PRACTICAL TOOLS AND TECHNIQUES I. Workshop Objectives 1 n. Opening Roundtable Discussion: NEP Data Management Concerns, Goals and Objectives 1 ffl. Successful Data Management Solutions From Tier I and Tier II Representatives: Goals and Objectives 7 A. Collection of Historical Monitoring Data 7 B. Collection of New Monitoring Data 9 C. Analysis of Data for Characterization , 13 D. Data Management Systems Design 17 IV. Small Group Discussion of Data Management Issues 20 A. Data Collection Methodologies 20 B. QA/QCofData .23 C. Hardware/Software Alternatives 24 D. Tools for Analyzing Monitoring Data 25 V. National Data Management System Resources 27 A. ODES/STORET Bridge 27 B. NEP Inventory 29 C. NOAA Systems (COMPAS) 31 D. ODES/GIS Connection 33 PLANNING AND PUTTING IT ALL TOGETHER I. NEP Data Management Policy: Review 36 A. Need for an NEP Data Management Policy 36 B. Issues Resulting from the Policy 36 C NEP Support 37 D. EPA Data Management System Modernization 38 Workshop Summary ------- II. Costs and Benefits of Data Systems Development ....40 A. ODRM System Life Cycle Development Methodology & EPA Approved Hard ware/Software 40 B. Long-term Systems Maintenance and Operations - Issues 42 HI. Proposed Action Items 45 Appendix A: Small Discussion Groups A-l Appendix B: Workshop Participants B-l Appendix C: Summary of Participants' Comments C-l Workshop Summary ------- PRACTICAL TOOLS AND TECHNIQUES I. Workshop Objectives Mark Curran, Chief of the Estuarine Management Branch of the Oceans and Coastal Protection Division (OCPD), welcomed the participants and thanked them for attending the Workshop. He explained that the Office of Water has been reorganized and that OCPD maintains the responsibility for overseeing the National Estuary Program (NEP). Mr. Curran explained that the issues associated with data management for the NEP have been neglected in the past. He added that there has been a long- standing need for a Workshop to get the "people in the trenches" of data management together to discuss common issues, goals and problems. The importance of data management when developing a CCMP was stressed by Mr. Curran. He noted that the ability to manage data is key in providing understandable information, and that the data provides the ability to measure the effectiveness of management plans. Two objectives for the Data Management Workshop were highlighted by Mr. Curran: 1) To gather together the data management experts from the estuary programs to learn from each others experiences. 2) To identify the types of data management support needed by the programs. Mr. Curran commented that the agenda should accomplish these objectives, but that all of the issues could not possibly be covered during the two day Workshop. However, the contacts the participants made at the Workshop should serve as a source of information to resolve future issues. II. Opening Roundtable Discussion: NEP Data Management Concerns, Goals and Objectives Joel Salter, the data management coordinator for the NEP, led the opening round table discussion which was designed to give a spokesperson from each program the opportunity to describe the program's data management concerns, goals and objectives. Fourteen out of seventeen programs were represented. Workshop Summary ------- A. Individual Estuary Program Comments Albemarle-Pamlico Estuarine Study (Tier I): Tim Johnson stated that the main goal of the APES program is to build an integrated data base that can perform statistical and spatial analyses. APES has spent a considerable amount of effort developing a Geographic Information System (GIS) for spatial data analysis. APES is attempting to reach out to other programs and organizations to gather data and to obtain support for continuing the data management effort after the program ends. An additional data management goal is to develop communication links to other systems including the IBM 9000 computer at the National Computer Center (NCC) in North Carolina. Mr. Johnson also identified several program concerns regarding data management. These concerns include obtaining the resources that will be necessary to maintain the data base, pulling data together from different sources so that it is compatible, making the data base accessible to the public, and developing a method for handling temporal data in a GIS. Barataria-Terrebonne (Tier in); Greg DuCote stated that the program's main goals are to collect as much data as possible, to make the data accessible to researchers and the public, and to be able to successfully manage the data for meaningful analyses. Barataria-Terrebonne has a grant from the Department of Energy to collect spatial data layers which will assist in the program's data management effort. Several concerns were also noted by Mr. DuCote including the problems associated with providing data accessibility for a variety of users. In addition, the program is concerned with coordinating all of the data that is collected, so that it is in a compatible format. Buzzards Bay Project (Tier I): Neil MacGaffey identified several goals for the project. Specifically, the project is striving to develop additional spatial data layers including sub-basin drainage areas and shellfish beds, to compile a descriptive index of data so that available data can be identified, to provide on-line access to users, and to provide data management support for project activities. Mr. MacGaffey also discussed several concerns which stem from the need to prioritize data management objectives within the program. A clear description of data management expectations needs to be established. Establishing expectations early in the program assists data managers in prioritizing objectives and tasks. Casco Bay (Tier ffl): Christopher Kroot relayed to the group that the Casco Bay program sees data management as an essential part of developing a CCMP. As a result, the program started planning for its data management system immediately. Mr. Kroot indicated that the program is in the process of identifying what management decisions need to be made, so that the Workshop Summary ------- necessary data can be collected and coordinated into a useful, compatible, and accessible format. Currently, the data management system developed by Casco Bay addresses at least half of the program's objectives. In order to make the data accessible, Mr. Kroot explained that the program has chosen to use the ODES/GIS Connection developed by EPA. Rather than develop a new system, the program is relying on the existing ODES system for data storage and maintenance. The program is designing a specialized user- interface for the ODES/GIS Connection to specifically address program needs. Delaware Estuary Project (Tier n): Mindy Lemoine stated that currently no one is specifically assigned to focus on data management for the project. The project plans to use existing data management systems for their data storage and analysis needs including ODES, COMPAS, and other NOAA systems. Ms. Lemoine indicated that difficulties arise from conflicting objectives between the three states that are involved in the project (i.e., NJ, DE, and PA). For example, New Jersey has developed a GIS, but Delaware has not. As a result, the project will not develop its own GIS, but will coordinate the data available from the three states and utilize the GIS resources already available. She indicated that the project does not have the financial resources to develop its own GIS. Galveston Bay National Estuary Program (Tier II): Robert McFarlane stated that the Galveston Bay program is not developing a data management system. He indicated that the main goal for the program has been to develop a data inventory which provides an electronic literature search including a bibliography. This data inventory allows a user to query the system to identify where the data can be located or if it is still available. The Galveston Bay Information Center, located at Texas A&M University, maintains all available literature and reports. One of the program's largest problems has been lost due to a lack of maintenance (e.g., data on magnetic tapes, etc.) Archiving data has not been a state priority. As a result, 80% of the data collected for Galveston Bay has been lost. Mr. McFarlane explained that the program has been supporting the use of COMPAS, which has been adopted by many of the state agencies in Texas. The system will be fully functional within the year. The program has also incorporated the Texas National Resources Information System (TNRIS) into its data management strategy. TNRIS coordinates all state and federal data and provides public access to the data. Galveston Bay is not meeting the dates established in its data management schedule for developing its CCMP, according to Mr. McFarlane. As a result, the program is forced to begin developing its CCMP without any data analysis. Mr. McFarlane explained that the program is going to rely on expert opinion, since there is not enough data to perform the necessary analyses. Workshop Summary ------- Indian River Lagoon (Tier III): Bob Day stated that the program is only in the beginning stages of developing its data management plan. The program's main goals include providing accessibility to data for a wide range of users through Florida's environmental learning centers. The environmental learning centers are used by scientists as well as elementary school children, so the program is faced with providing information to a wide audience. The program has stored all of its water quality monitoring data in STORET and is planning on using the ODES/STORET Bridge to access their data. The water management districts have spatial data layers that cover the lagoon, and these data layers will be available to the program. Long Island Sound Study (Tier I); Cynthia Pring-Ham explained that the program is continuing to develop its data management strategy. The program has been extended a year, because it is behind in developing its CCMP. She indicated that data management issues do not receive the attention that they need to help address program objectives. In the short- term, Ms. Pring-Ham indicated that the program is developing an index of data which contains three years worth of data. Ms. Pring-Ham identified several goals for the program which include developing models to determine the monitoring needs of the program, designing a data management system to satisfy the identified monitoring needs, and insuring that the data management system has adequate analytical capabilities. Two concerns were identified by Ms. Pring-Ham. First, the program is attempting to combine data collected by two different states, New York and New Jersey. Second, the program recognizes that it will take considerable time and money to review the historical data for Long Island Sound to see if there is sufficient QA/QC information to make the data useful. In addition, she indicated that the historical data is in inconsistent formats. As a result, it must be converted to a standard format in order for the data to be accessible. Massachusetts Bays Program (Tier ffl): Dillon Scott explained that data management is one of the Massachusetts Bays Program's top priorities. As a result, the program is establishing specific objectives for data management. Ms. Scott indicated that there is only one person responsible for data management, so it is often difficult to determine how to adequately address the variety of data management issues. The program's data management system is linked to a state system which maintains ARC/Info data layers. The program is developing an index of research data collected for the Bay and focusing on making the data compatible with the states GIS. The GIS will provide access to the data for a variety of users including the public. Workshop Summary ------- Narragansett Bay Project (Tier I): Stephen Hale explained that the Narragansett Bay project began a massive data collection effort within the first year of the program. He indicated that the program has set specific goals for characterization. Specifically, the project is striving to collect and verify historical monitoring data, to modify the data management system to be able to sustain long-term monitoring, and to develop a database for managing long-term trends. Mr. Hale explained that the program has been successful in its data management effort. He attributes this success to the development of one central data management system and a computerized bibliography. In addition, the program obtained support and money to further its data management objectives. Although the data management effort for the program has been successful, Mr. Hale identified some concerns. First, the program is trying to consolidate all of its monitoring program objectives, so that the data management system may address as many as possible. Second, the program is beginning to face the problem of obtaining funding for the data management system after the program ends. Santa Monica Bay Restoration Project (Tier n): Robert Smith explained that the program currently has large monitoring programs that are managed by independent sources. One of the program's goals is to move toward data collection by regional programs instead of independent sources. However, Mr. Smith indicated that standardizing all of the data that has been collected in different formats from the independent sources will be a challenge. The program is planning to use a dispersed data management system. The option to develop a centralized system was eliminated because the program feels that this will result in data duplication. In addition, a dispersed system will allow updates to data to be done quickly and will allow immediate access to the data. Mr. Smith also indicated that the program is developing an index of the data in an effort to keep track of where the available data is located. Sarasota Bay National Estuary Program (Tier II): Dave Tomasko indicated that a workshop for all of the Pis collecting data for Sarasota Bay will be held in June. The program has developed a standardized format for collecting data which is based on ODES codes. The workshop is designed to discuss the standard format with the Pis and resolve any questions or concerns. In the long-term, Mr. Tomasko indicated that the program will be moving from the characterization phase of the program to monitoring. In an effort to effectively use the program's financial resources, the number of stations sampled will be reduced, so that monitoring can be concentrated in representative areas. Workshop Summary ------- The data being collected for six out of the eleven studies funded by the NEP will have spatial information. However, the program is not planning to develop a GIS, and will send the data elsewhere for digitization. Mr. Tomasko explained that the program will work with the GIS system developed by the South West Florida Water Management District, but the program currently has not had a strong demand for spatial data. Tampa Bay (Tier HI): Holly Greening explained that the program has not yet developed its data management strategy. The program is faced with coordinating and standardizing the monitoring program data from eleven governmental agencies. A workshop is planned for this summer to facilitate this effort. Ms. Greening indicated that the program is placing an emphasis on GIS to analyze spatial data. In addition, the water quality data for the system is maintained in STORET, and the biological data will be submitted to ODES. Thus, the program is interested in using the ODES/STORET Bridge to access all of the data. B. Common Themes Identified Throughout the Estuary Programs Several common themes were identified throughout many of the comments made by the estuary spokespeople. These themes are listed below and are discussed in more detail in the sections that follow. Data integration issues Communication between systems Data accessibility for principle investigators & the public Available tools to use the data Identification of the intended use of the data Importance of spatial data and GIS applications Data Indexes to identify available data Establishment of priorities and expectations for data management Develop better data coordination between states Identification of perishable data Importance of sound data analysis during characterization, CCMP development and implementation, and long-term monitoring High cost for determining the usefulness of historical data Need to use standardized data collection formats Importance of identifying sources of funding for data management Workshop Summary ------- IE. Successful Data Management Solutions From Tier I and Tier II Representatives: Goals and Objectives A panel of Tier I and II data managers, moderated by Bob King, addressed successful data management solutions. Neil MacGaffey discussed the collection of historical monitoring data. Stephen Hale presented issues concerning the collection of new monitoring data. A discussion on the analysis of data for characterization was presented by Tim Johnson. Tom Gulbransen discussed data management systems design. A question and answer period followed the four presentations. The comments made during the question and answer period are incorporated into the pertinent sections. A. Collection of Historical Monitoring Data Neil MacGaffey - Buzzards Bay Project The use of historical data is strongly affected by changes in data methodologies and available technology. For example, the use of personal computers and commercial data management software has increased due to the reduction in cost. The availability of hardware and software in return has led to a standardization effort to use the same data management tools and data formats. This standardization effort has led to some integration among data management systems. These changes affect the user's expectations of how easily the historical data can be manipulated. As a result, those responsible for the original data collection effort cannot always be faulted for the complexities associated with historical data, because the methods used to manage and collect the data have advanced since the data was collected. The following four major issues were addressed: 1. Documentation Quality control documentation usually does not accompany the data and often is not included in the data set report. For example, information on limits of detection is often missing. As a result the limits must be assumed based on the analytical technique used to collect the data. In addition, data collection references are usually incomplete or missing. Often the agency that collected the data no longer exists. In other cases, the agency no longer has the data report or cannot find it. Additional documentation problems result from missing or incomplete data dictionaries. Without the supporting data element descriptions, it is difficult to identify the information maintained in each field. 2. Variations in Temporal and Spatial Coverage Historical data sets were usually collected for purposes unrelated to current program needs. As a result, the spatial and temporal coverages are often Workshop Summary ------- inadequate for current analyses. However, exceptions do exist. For example, in the one instance in Buzzards Bay where long term data collection has existed, some limited trend analysis is possible even though the data were originally collected for unrelated purposes. 3. File Formats and Standardization Each file usually has a unique format. As a result, each data set poses separate data transfer problems. These problems allow little opportunity for developing efficiencies in the data transfer process. An inordinate amount of time can be spent diagnosing and solving intricate file and field format problems and inconsistencies. The problems associated with massaging data into standard formats exists regardless of the system or application being used. In an effort to facilitate the data standardization process, the Buzzards Bay Project hired an outside contractor to keypunch some data sets, standardize the file formats of digital data, collect quality assurance information, and add important information such as latitude/longitude coordinates for station locations. Locating, obtaining, "scrubbing", and standardizing the format of existing data sets is a task requiring day-to-day attention to details; it is extremely difficult to track these details reliably when other tasks provide distractions. Having a contractor focus exclusively on the task resulted in more consistent data in a standard format. The standard format, the data scrubbing, and some quality assurance performed by the contractor, made moving data into the Project's ORACLE database easier to accomplish. 4. Missing Information Often essential information is missing or available only after considerable collection effort by data management staff. This information must be assembled to determine the potential usefulness of the data set. Missing information may include: sample values (e.g., concentrations), units of measurement, methods for some or all of the parameters being measured, station depths, coordinates for station locations, environmental data (i.e., wind direction, rain, or tide). In addition, the definitions for coded fields are often missing. As a result the codes used in the data set are meaningless to the investigators. If standardized codes were used, such as NODC codes or CAS numbers this problem is reduced. Workshop Summary 8 ------- B. Collection of New Monitoring Data Stephen Hale - Narragansett Bay Project Four major topics concerning the collection of new monitoring data were addressed. These topics include the goals of a long-term monitoring data management effort, general issues related to collection of new monitoring data, information needed by data management personnel in order to effectively collect new monitoring data, and actions that should be taken by data management personnel prior to the start of data collection. 1. Goals of a long-term monitoring data management effort The recommendations developed from the characterization process that are incorporated in the CCMP are tested through a long-term monitoring data management effort. The long-term monitoring data management effort provides a long-term, consistent data base that can be used by managers and scientists to discover and explain trends in the estuary's ecosystem. The long- term monitoring of the estuary should indicate if the recommendations implemented in the estuary's CCMP are having the desired effect. The ultimate goal of the program's data management system is to provide access to quality data so that the data can be assimilated into useful information to provide the support for management decisions. Achieving this goal provides the estuary program managers with accessibility to quality information, so they can do a better job of managing the estuary and its watershed. Although estuary program management decisions are influenced by a variety of factors, it is the responsibility of the program's data managers to provide a platform for decision making based on data and scientific information. An additional goal in the development of the data management system is to make the monitoring data available and easily understandable for a broad range of managers, researchers, and the public. Although these goals are difficult to attain, data managers should strive to achieve the highest quality of data and greatest accessibility when developing the system. Many states are discovering that providing accessibility to the monitoring data at a local level will ensure additional funding for monitoring programs. It is especially important to provide access for the local levels, because local governments, public works departments, and the water districts are the people who are actually making the decisions that affect the environment. Moreover, states are much more inclined to provide funding for monitoring programs that generate data that is accessible to state and local level programs. Workshop Summary ------- The underlying idea behind the estuary program is to build up the state agencies to the point where they can assume the responsibility for the program's data management and analysis. Thus, it is very important to involve the state agencies in developing a monitoring plan for the estuary. 2. General issues related to collection of new monitoring data When planning the data management effort for the program, the program managers must determine if the system will track only monitoring data, or if it will function as a true central data repository for the estuary, Narragansett Bay has developed a central storage center for monitoring data as well as other types of data, including NOAA bathymetry data and USGS water quality information. The program selected this approach because analysts need access to many types of data other than monitoring data. In addition, different analysts need access to the same data. If a central data storage location is not used, two different analysts may store and maintain the same data in their individual systems which results in a duplication of effort. By maintaining all types of data in a central location, users have access to a variety of data and the problems associated with duplication are eliminated. The problem with developing a multi-disciplinary central data base for an estuary is that no agency or program has the responsibility or mandate for developing and maintaining it. Although federal, state and local agencies and universities heartily support the idea, none of these organizations are mandated to provide it. As a result, uncertainty remains among many of the estuary programs regarding how a long-term monitoring data management effort can be funded. Obtaining a firm commitment from the state in support of long-term monitoring plays a key factor in planning the program's data management strategy. One of the key issues in the long-term monitoring program is that many different agencies, contractors and citizens monitoring groups are collecting data. The data managers for the estuary program are trying to pull all of the information collected together to address the key concerns that the estuary program has identified. Assimilating all of the data is a time consuming task, because, the data and quality assurance information are in different formats and must be standardized. Some of the standardization effort can be eliminated by specifying that contractors collect and document the data in a standard format. Since the data is being collected from citizens monitoring groups on a voluntary basis, the data managers can only stress the importance of collecting data in a specific format. In addition, the data files from other agencies will almost certainly need to be massaged into the estuary programs standardized format, unless the individual agencies can agree on a consistent format. Workshop Summary 10 ------- 3. Information needed by data management personnel in order to effectively collect new monitoring data Before a data collection effort is begun, information is needed by the data management personnel. Specifically, a final list of data file types that will be collected must be developed. This list should include the specific parameters and units that will be collected. The number of stations and frequency of sampling at each station must be identified for all data file types and parameters. This information will indicate the amount of data storage space needed to manage the data collected. The precision, significant digits, detection limit and field collection method must be identified and stored in the data base for each parameter. In addition, the data collection, digitization and storage methods must be collected and maintained in the system. For existing data collection efforts, the data management personnel must research this information if it is not part of the data file. In addition, the data management personnel along with the estuary program managers and scientists, must determine exactly what will be done with the long-term monitoring data files once they have been submitted to the central data system. Specifically, the desired tables, graphs, maps, and statistical analyses should be identified. This early discussion also provides the opportunity for the managers and scientists to express their expectations from the system and allows the data managers to realistically explain whether or not their expectations are feasible. 4. Actions that should be taken by data management personnel prior to the start of data collection Prior to the start of long-term monitoring data collection, the data managers should provide historical data to the statisticians and monitoring plan designers so that the most effective sampling design can be developed. The historical data can provide insight into the most useful sampling parameters, the number and location of the stations for sampling, and the frequency and number of samples to be taken at each station. The data management staff should also modify the data base design to incorporate any new data types that were not included in the characterization studies. Please note that this requires the data base structure to provide the flexibility for changes. The data management system must also provide a tracking system for archived samples that records information about each sample taken in the Workshop Summary ------- estuary. The information should include who took the sample, when it was taken, where it is stored, and what methods were used to collect it. After the data base design is modified, the supporting documentation (i.e., the Data Submissions Manual) must be revised to include all data file types in the final long-term monitoring plan. In addition, all data submissions forms and formats must be updated. To facilitate the submission of data, the data management staff should develop procedures with each data collecting agency to transfer the data and documentation. Wherever possible, procedures should be implemented to automate the data transfer, checking, loading, and reporting. Workshop Summary 12 ------- C. Analysis of Data for Characterization Tim Johnson - Albemarle-Pamlico Estuarine Study This discussion defined characterization and presented alternative approaches for analyzing data during characterization. The use of GIS as a data analysis tool was discussed using examples from the Albemarle-Pamlico Estuarine Study. During the open discussion session, several data managers discussed how their programs have dealt with the complexities of using historical data in characterization and CCMP development. In addition, several participants raised questions about the appropriate audience for GIS applications. 1. What is characterization? Characterization is the effort to discover the present state of the estuary. The characterization report should include the status of problems in the estuary and the sources of the problem. In addition, the report should include trends identified by the studies initiated to review the problems. 2. Alternative approaches to data analysis for characterization. Most estuary programs begin the characterization process by using a non- automated approach where the general state of the estuary is assessed by experts by reviewing available trend information. Many programs then move to an automated system where tabulated data may be used to provide reports and conduct statistical analyses. A third alternative is to incorporate a geographic information system in a program's data management system, so that the data may be viewed spatially. Ideally, the estuary program would have the resources to design a specialized monitoring program and collect the required data to address the issues related to the estuary. However, this is not the case, and the estuary programs must use existing data to characterize the estuary and plan the monitoring program. Almost every program is using historical data differently to characterize the estuary. The differences in the data analysis methods stem from the quantity and quality of the historical data available and the gaps in data that must be filled during characterization. All of the programs agree that it is difficult to adequately factor data management into developing the CCMP within five years. However, Congress has imposed the five year time limit, and the NEP developed a program around the time frame. Representatives from several programs explained how their programs are attempting to factor data management into the characterization process and into the CCMP development: Workshop Summary 13 ------- Casco Bay: The Technical Advisory Committee made it a priority during the first year to collect a baseline of data. The program took a data inventory of all state agencies and groups. At the end of the first year, the program will write a "state of data" report. All of the state agencies, local governments, and universities are committed to the initial data collection project and have given their financial support. The program managers understand that a successful CCMP cannot be developed without incorporating sound data analyses. The development of the data management system is going to be the glue that holds all of the agencies and groups together after five years. The whole purpose of developing the system is to build relationships among the groups to help them share data. Delaware Bay: The program is currently developing an inventory of its data, but the inventory will not be complete until after the period of characterization. As a result, the program will base its characterization report on research and expert opinion. The data managers are meeting this summer to identify gaps in the data, so they can focus on collecting the necessary information to complete the CCMP. Albemarle-Pamlico: During the second year of the study, the program completed a data needs assessment. One of the products of the assessment turned out, unintentionally, to be a data inventory. The data managers discovered some purely historical data and identified areas which need to be expanded. As a result, they have set priorities for the data collection. 3. Use of GIS as a data analysis tool A GIS is an automated mapping system that allows users to manipulate data through querying data files and to produce output maps. Data stored in a GIS has two components, a graphic component and a tabulating component. These two components are combined by a unique identifying number. Using GIS provides the advantage of a common frame of reference through spatial data. Analysts are able to display trends in the estuarine environment and can visually identify the available data in order to make recommendations for protecting the estuary. A GIS can be combined with other technologies to create a total system environment. For example, a GIS can be used in conjunction with statistical packages to quantify trends. The ability to communicate the basis for decisions to management and the public is extremely important to the estuary program. The importance is magnified by the drastic changes that are going to be proposed, especially in land use and zoning. The estuary programs are going to have to sell their decisions to the public who are going to be suffering under the resulting ordinances that are going to be recommended. Developing a GIS application provides the tools necessary for displaying the need for radical changes. Workshop Summary 14 ------- The disadvantages associated with developing a GIS are due to the cost. Data development is the most costly aspect of using GIS. It is time-consuming and requires the use of specialized and expensive hardware and software. 4. Case Study: Albemarle-Pamlico Estuarine Study The APES study area covers approximately 10 million acres in North Carolina and 2 million acres in Virginia. The data management team for the program decided to use a GIS as the data management tool. A GIS allows the program to integrate as much data as possible into a system that is easily accessible for researchers as well as the public. The Center for Geographic Information Analysis, a North Carolina state agency, had a considerable amount of spatial data already available, that covered the APES study area. In the initial data needs study, the data management team planned to collect or develop approximately 65 data layers. However, some of these data layers were eliminated for one or more reasons such as incompleteness of source material over a wide geographic area and prohibitive cost of data collection. At this time, almost 50 data layers are under development. Most of the data covers North Carolina, because the state is further along in its GIS effort. The data management team is working with Virginia to ensure that data development specifications are consistent with those used in the North Carolina portion of the study area. The information accessible through the GIS ranges from basic data layers such as hydrography and political boundaries to specialized layers such as shellfish beds. The GIS developed for APES runs on a SUN hardware platform, and uses ARC/Info and ERDAS software. The data management team is stressing the importance of expanding the communications capability of the GIS to reach resource managers and the research community. This type of link eliminates the need for the data to be stored in multiple locations. The GIS currently is being used to study the inventory of resources in the Albemarle-Pamlico estuary. Using the GIS, estuary managers can view the available resources, the location of the resources, and the relationships among them. The GIS allows managers to target critical areas, especially wetlands and shellfish disease areas. The APES program is also using the GIS in water use planning. The managers are focusing on the competing objectives of conservation, preservation, and development activities in the coastal areas. The GIS allows the areas of influence to be viewed spatially with overlays of natural resources (i.e., shellfish beds) to see if these resources are being negatively affected by the development going on in the area. Workshop Summary 15 ------- 5. Audience of GIS Applications Considerable debate exists among the program data managers regarding the appropriate audience for a GIS application. For example, it is arguable that it is unrealistic that a senior manager will turn to a GIS application sitting on his or her desk to readily access the information necessary to make the decision. The senior manager is more likely to request the information in a consolidated form on paper from the senior technical staff. On the other hand, if a GIS application is menu-driven, the senior manager can have all of the information immediately. As senior management becomes increasingly computer literate, they will take advantage of the quick response to the system. Examples of senior management utilizing GIS applications to made decisions currently exist. For example, senior managers used a GIS application to develop Oregon's Clean Water Strategy. A second argument is that a GIS application should be geared towards senior technical staff who are more directly involved in the data analyses than the program managers. In this case, a menu-driven GIS application provides a set of predictable queries which releases the senior technical staff person from performing mundane data queries. In addition, the system can also provide access to the ARC/Info command line, so that the experienced GIS analyst can address more complicated modeling and analyses. In addition, once the scientist has made a technical decision, he or she will have the ability to present the information to senior managers, legislators and the public in a pictorial form to explain why the decision was made. Providing a GIS that is accessible and simple to use for the public is a third alternative. Making the information accessible allows the public to become involved in the decision making process. As a result, strong public opinion can influence the program manager's and legislator's decisions regarding resource allocation for the estuary program. Workshop Summary 16 ------- D. Data Management Systems Design Tom Gulbransen - Battelle Ocean Sciences NEP data managers face unique challenges which make following the standard textbook data management systems design process difficult. For example, the programs must provide accessibility to data for many groups including the public. In addition, the data managers are faced with the difficult task of working with many agencies to define the needs of each group. 1. Factors Influencing a Data Management System Design & Development The specific management decisions and analyses that are to be supported by the system must be written down and prioritized. Each committee should present its objectives, and the Management Committee should prioritize them. The master list of prioritized objectives can then be developed into a data management strategy. From this strategy, an appropriate system design can emerge. The agencies involved in the system's development must commit to a time line and budget. Each task in the system design and development should be identified, so everyone involved understands the level of effort required. The tasks can include needs assessment survey costs, requirements documentation, collecting data, keypunching data, digitizing data, and maintenance. The financial resource planning incorporated into the data management system design cannot be based on single-year workplans. The system will require long-term resources which must be identified from the beginning of the system's design. The program coordinator must confirm the commitment of the states to support the system, because the states will need to take over the responsibility for the system and long-term monitoring after the five year program. It is not necessary or advisable to wait until the five year program is over to start transferring system responsibility to the state. The transfer should begin as soon as the state's commitment has been secured. Several programs commented on the complexity of securing funding for data management and how the program is handling this challenge: Casco Bay: The program recognizes that it would be impossible to develop a data management system specifically for the program, because the funding is not available. The program's entire data management system is designed around existing systems that will be maintained by the EPA. The ODES/GIS application, developed with headquarters funds and contractor support, is simply a front-end to these systems and functions as a data integration tool. Workshop Summary 17 ------- Buzzards Bay: The program's goal is to build up the capacity of the state technical and data management expertise, so that it can take over the system. However, the state is not in a financial position to do that, so realistically the program will only have enough money for long-term monitoring. Even a few of the data collection projects that were established during characterization will not become part of the long-term monitoring program due to lack of funds. The program is currently trying to obtain support from other sources. Sarasota Bay: The program hopes that the data management effort will uncover inefficiencies in the current Bay monitoring program. The data managers need to review the data collection process to determine if the most effective number of stations, times of day, and number of samples are being included in the sampling program. If inefficiencies are discovered, the program feels that the amount of money spent on monitoring will decrease in the long run. 2. Functional Requirements and Documentation Functional requirements are statements of what the data management system needs to do, not how the systems accomplish the processing. The functional requirements of the system are affected by a variety of factors including the need for system flexibility. As priorities in the program shift, the data management system must be able to accommodate those changes. Some programs are designing systems in an "ad hoc" manner. In this situation, a program manager indicates that he or she needs to access a particular data set or to answer a specific question. In response to the request, the data managers formats data sets, or writes a case specific application. Responding to requests on a case by case basis is costly in the long run, because all of the data sets are not formatted consistently and subsequent applications need to manipulate diverse data sets. Documentation created to support the system is an important factor in the success of the system. The documentation can explain the people's roles, system procedures, needs and functional definitions. The documentation serves as a baseline for verifying the system design and should be reviewed periodically. In addition, the system platform should be documented, so that the data can be easily shared among different systems. For example, extensive documentation of the ODES and STORET platforms facilitated the development of a link between the two systems (i.e., the ODES/STORET Bridge). There is detailed data management systems design methodology documentation available from OIRM. However, the NEPs do not have the time or the financial resources to go through the step-by-step life-cycle. The Workshop Summary 18 ------- OIRM methodology must be tailored to meet the NEP needs. In fact, NEP needs will differ for each program. 3. Identifying a Successful Data Management System Even if the system design and documentation are perfect, the system is still not guaranteed to be a success. A system can only be successful if it fits into the way people perform their assessments, characterizations, and analyses. Success can only be achieved fully if these people are involved in the design stages above. Basic question should be asked to examine a system's success, such as: 1) How many CCMP improvements did the system provide? 2) Did any policies or decisions result from the use of the system? 3) How often is the system being used? 4) Were all targeted users served? Workshop Summary 19 ------- IV. Small Group Discussion of Data Management Issues The Workshop participants were divided into three groups to discuss data management issues. The groups were arranged to mix representatives from different tiers and programs. Appendix A includes the breakdown of group members. Four topics were covered by each group: 1) data collection methodologies, 2) QA/QC of data, 3) hardware/software alternatives, and 4) tools for analyzing monitoring data. After the groups met for approximately an hour and a half, a spokesperson presented the main ideas discussed in the group. A. Data Collection Methodologies 1. Group #1 Discussion The data collection methodologies discussed in group #1 were very diverse. Representatives from three programs identified the methodology currently in place to collect data: Albemarle-Pamlico: Data managers from North Carolina and Virginia met after three and a half years to develop an inventory of data. In addition, a data needs survey was developed and 65 data layers were agreed to be collected. The data managers decided on the priorities for collecting the data layers and assigned responsibility for collecting them. Galveston Bay: The program decided that the existing state and federal monitoring data is adequate for its needs, although all data prior to 1980 has been lost. The data managers meet to develop overall goals for data collection, and ad hoc planning is used to better define the data collection methods within the goals. Narragansett Bay: The Science and Technical Committee has identified areas of data that need to be collected. The program has funded studies to collect data to fill in the identified gaps. The program has hired experts to research and assess the available data. The data managers have identified the types of GIS coverages that the program needs, and have started to build them. However, the coverages are being built without the end products in mind. So far, this method has not led to any problems. The group discussed what they felt the perfect methodology would be for collecting data. In general, the group acknowledged the importance in specifically stating the expectations of collecting data in the form of a Workplan. A schedule of deliverables should be established, so that the program receives the data on a timely basis. In addition, the required QA/QC procedures should be stated at the onset of the contract. The QA/QC Workshop Summary 20 ------- information reported to the program should include (at a minimum) the sampling methods, laboratory methods, and QA data (e.g., blanks, spike recovery). In addition, the group decided that there are two options for developing a request for proposal. First, the program can make the RFP very specific. This option should be chosen when the requirements need to be explicitly stated. Second, the RFP can be fairly broad. Developing a general RFP allows the program to assess the contractor's understanding and expertise as related to data collection and program objectives. 2. Group #2 Discussion Every program in the group had different methods and protocols and indicated that achieving consistency among the agencies involved with data collection would be difficult. One example of data consistency among agencies is in Florida where the state legislature requires that all data follow a Quality Assurance Program Plan. In addition, the programs commented that using ODES formats is another example of an effort to achieve consistency among different agencies. Several group members offered comments about the data collection methodologies in use in specific programs: Long Island Sound: The program is following general data collection protocols developed by the State University of New York and the University of Connecticut. Currently, protocols are being developed by the state which the program will follow. In addition, the program is planning to document its data collection strategy. Galveston Bay: The program has devised a bay segmenting scheme. Currently, they are revising the sampling plan by evaluating the stations being sampled. The analytical methods used to collect data are not standardized. The strategy of the program is to coordinate the collection of the data, and to distribute the responsibility for the data. The program is not planning to centrally store the data. Barataria-Terrebonne: Multiple monitoring programs are going on in the estuary. Different methodologies and parameters are being used for each monitoring program. The managers for the monitoring programs are not coordinating the data collection efforts. Standardized protocols need to be developed for the program. At this time, a variety of spatial data is also being collected by the program, but specific conventions are not being followed. Indian River Lagoon: Florida has imposed lagoon-wide Surf Water Improvement and Management (SWIM) legislation which incorporates monitoring at fixed stations throughout the lagoon. Formats for data collection are established. The Indian River Lagoon Science Information Workshop Summary 21 ------- system contains a bibliography and index of the data and QA/QC information for the program. The QA/QC procedures will be reviewed i the near future. Tampa Bay: Tampa Bay is a SWIM region, thus there are legislative requirements to monitor the bay and submit the data to STORET. The program has planned a Workshop for this summer to better coordinate the collection and submission of monitoring data for the bay. Thus, the estuary program is acting as an agent for improved coordination and cooperation. Albemarle-Famlico: QA/QC reports are generally available through the appropriate department for historical monitoring data. New monitoring procedures are specifically stated in proposals prior to collection. Formats and parameters for water quality data have been previously decided. The data management staff visit the Pis to review formats and parameters when necessary. Delaware Estuary: A program study is currently underway to summarize the variety of data types and collection methods being used. The state agencies have consistent long-term data conventions, but the data is not centrally stored. Bibliographic references are available, and the program is currently considering ODES formatting conventions. 3. Group #3 Discussion The group identified several common data collection problems which are related to contractor management. Often contractors do not complete the data collection on time. In addition, the money on the contract runs out before the contractor makes any corrections to the data or provides adequate QA/QC information. The group offered some ideas to solve these problems. Specifically, the contracts should specify data formats, QA/QC requirements, and a schedule for delivery. In addition, the final payment should be held until all data is submitted in the correct format and all QA/QC information is completed. The group added that a program person needs to be appointed to specifically oversee the data collection contract. This contract manager should verify that the data is collected in the correct format with adequate QA/QC information and in a timely manner. Workshop Summary 22 ------- B. QA/QC of Data 1. Group #1 Discussion The group identified the QA/QC procedures in place for the individual programs: Galveston Bay: QA/QC information is available for some historical data. The program is currently screening the QA/QC for all data sources to collect information concerning how and when the data was collected. Before any new data is collected, a QA/QC plan must be established before. The EPA requirements for QA are generally followed, however, they often must be supplemented to provide the level of detail necessary. Albemarle-Pamlico: The three agencies collecting data for the program are coordinating their QA/QC efforts including participating in workshops. Narragansett Bay: New data collection projects must have QA/QC plans that follow EPA guidance. These plans are submitted to the EPA Regional office for review. The program data managers are beginning to develop QA/QC guidance for data collection. During the group discussion, there was considerable concern about how to ensure that the data collected by contractors was submitted in a complete, accurate and timely manner. The group generally agreed that peer reviews of the data work well to detect any problems with the accuracy or data collection methods. In addition, the contract for data collection should require that the data be delivered according to a set schedule and that the complete payment for the work should not be made until the data and QA/QC report are submitted. 2. Group #2 Discussion The group decided that although different data collection methods and protocols are used by individual agencies, information about the methods and protocols must be referenced. In addition, more of an effort needs to be placed on making this information available in electronic form, similar to the on-line QA/QC reports in ODES. For example, Puget Sound uses a grading process to denote the quality of its data. Instead of storing spikes and blanks on-line, a number from 1 to 5 is associated with the data to indicate the relative quality of the data. 3. Group #3 Discussion The group discussed the roles in the QA/QC process and decided that the PI performs the quality control procedures, and the data base manager performs Workshop Summary 23 ------- the quality assurance procedures. Specifically, the PI collects the data and verifies that the procedures are appropriate and accurate. The PI should provide information concerning blanks and spikes associated with the data, as well as a description of how the collection instruments are calibrated. The PI should also provide information on any notable trends in the data. After the PI provides the data and the quality control information, the data manager should then review the data to make certain that it is in the correct format and that all required quality control checks have been performed and documented. The group strongly agreed that documentation is the key to the QA/QC process. Each data set must be documented as completely as possible. Data should never be discarded. Even if no information exists about the data, document that fact. Providing as much documentation as possible allows the user to judge whether or not the data is useful in his or her studies. QA/QC methods used by several programs were discussed during the group session: Narragansett Bay: The data managers perform frequency distributions and range checks on the data. Standard method codes and sampling codes are used across data sets. A text file is associated with each data set which includes contact information for the data. The text file is updated to describe how and when a data set is used. Casco Bay: The program's technical committee is developing standards for all newly collected data. The program is "double checking" its QA/QC process by first performing their own review of the data, then sending the data to the ODES technical staff for its QA/QC review. The program is then comparing their own results with that of the ODES technical staff. Santa Monica Bay: At a minimum multivariate analyses are run on the data set to check QA/QC across the variables. A manual of quality control procedures has been developed, but each data set requires additional review in addition to the standard procedures. C. Hardware/Software Alternatives 1. Group #1 Discussion The hard ware/software in use by the program was discussed: Sarasota Bay: The principle investigators (Pis) use their own hardware and software when collecting the data. The program receives disks in ASCII format and hard copies of the data from the Pis. Workshop Summary 24 ------- Galveston Bay: The Pis collecting data use a variety of hardware and software, and the Pis submit the data in ASCII format. Narragansett Bay: The program decided to use FOCUS, a commercial data base management software package, for its data management system mainly because this software was already in use by the state. The system has worked very well. However, FOCUS and ARC/Info are not on the same hardware platform, so issues have arisen when they have tried to use ARC/Info with the program data. They are currently running their system on a VAX, but stress that there is no one hardware and software solution. 2. Group #2 Discussion The group reviewed Section IX in the notebook which included an overview of the hardware and software in use by each program. In addition, the group discussed the necessity of considerable disk space to manage the quantity of data needed by the programs. Specifically, when programs analyze land use data, multiple gigabytes is required. 3. Group #3 Discussion Each program in the group is using a different combination of hardware and software for its data management effort, as demonstrated by the overview of hardware and software in Section IX of the notebook. Most programs took advantage of what ever was available to them from the state or universities. In some programs, such as Casco Bay, the state is providing funding for all hardware and software the program requires. In general, the programs are satisfied with the hardware and software that is available to them. The group stressed that one of the most important features of the system is to allow data to be stored in a central location so that it can be downloaded to a personal computer. Downloading the data allows an analyst to use his or her favorite software to perform the analyses. Of equal importance to the programs is the need for significant computer power to manage the large data bases being developed. D. Tools for Analyzing Monitoring Data 1. Group #1 Discussion There are a variety of tools in use by the programs for analyzing monitoring data. Similarities exist among the programs relating to the types of analysis being conducted, but there are differences in the actual logistics of how the analyses are accomplished: Workshop Summary 25 ------- Albemarle-Famlico: The program is primarily focusing on using its GIS for analyzing the monitoring data, and the system is working well. Issues have arisen because the GIS applications do not provide the capability to perform statistical routines such as correlation and regression. Sarasota Bay: The program is using its data to perform trend analyses. Narragansett Bay: Contractors are performing statistical analyses such as time-series and cluster analyses. However, it is difficult to correlate causes and effects. The program is now focusing on data modeling, and is drawing on the index of available data to develop the models. Galveston Bay: Highly customized models have been developed for the estuary. Contractors are required to use and contribute to the data base inventory, bibliography and literature search. 2 Group #2 Discussion The group discussed the broad range of models used by individual programs. Some programs have the facility to run a model and capture a specific "time slice" and store the information as a coverage in a GIS. Other programs have models which produce numerical coefficients that are incorporated into reports. 3. Group #3 Discussion The programs in the group, including the Tier I estuaries, have not begun to analyze the data being collected. The data managers are collecting and documenting data, so that studies can begin. Some programs, including Long Island Sound and Santa Monica, have tasked contractors to perform small studies when needed. Among most of the programs in the group, a general lack of objectives for data use exists. The program has not specifically identified why the data is being collected. As a result, thousands of dollars is spent on data collection, but no one is using the data for analysis. On the other hand, Christopher Kroot from Casco Bay explained to the group that the program would not commit to collecting any data until the data managers and program managers decided on what tools would be available. He stressed that this method has worked very well. Workshop Summary 26 ------- V. National Data Management System Resources Three national data management systems were presented. Gary Labovich discussed the ODES/STORET Bridge, Craig McCulloch presented COMPAS, and Christopher Kroot explained the benefits of the ODES/GIS Connection. In addition, Tom Gulbransen briefly discussed the NEP Inventory and requested additional information from the Workshop participants. A ODES/STORET Bridge Gary Labovich - American Management Systems, Inc 1. What is the ODES/STORET Bridge? The ODES/STORET Bridge is a user-friendly link that connects the ODES system and the STORET data base. The Bridge provides access to water quality data stored in STORET through the menu-driven facilities of ODES. The purpose of the bridge was to eliminate the need for a user to leave the ODES system to access STORET data. 2. How has the ODES/STORET Bridge Evolved? The bridge was originally designed to provide user-friendly menus to access STORET data. A Basic Option was added to the ODES system, Basic Option S, which helps users identify STORET agency codes, station codes, and available dates for data. An additional goal for the Bridge was to allow users to access more data for analysis. However, the ODES system creates a geographical boundary for a program, so users can only access data collected within a program's boundary. A geographical boundary may be increased or reduced if necessary. The Bridge was also created to give users access to historical data maintained in STORET when recently collected data is stored in ODES. In addition, the Bridge provides graphical tools that can plot STORET data, and the ability to download STORET data in text or ARC/Info format. Estuary Programs have requested enhancements to the Bridge, so users can access all STORET estuarine water quality parameter codes associated with the data collected for the program. In addition, the enhanced Bridge will access all variables associated with STORET estuarine water quality data. The downloading feature will be expanded to handle all STORET estuarine parameter codes and variables. Workshop Summary 27 ------- 3. How can the Estuary Programs use the ODES/STORET Bridge? Estuary Programs can use the enhanced ODES/STORET Bridge in a variety of ways. For example, user-friendly menus will prompt a user to identify a subset of STORET data. Users can select data by station attributes (e.g., within Delaware or by "ESTURY" station type), and can select data by sample attributes (e.g., by time ranges or depths). In addition, any STORET estuarine water quality data may be downloaded in text or ARC/Info format, and reports of any STORET estuarine water quality data can be created. 4. How does the ODES/STORET Bridge Relate to the NEP Data Management Policy? The enhanced ODES/STORET Bridge provides complete access to estuarine water quality data for OWOW. As a result, OWOW is considering amending the NEP Data Management Policy. The amended policy would allow water quality data to continue to be submitted to STORET instead of ODES, if the program currently submits its water quality data to STORET. If the policy is amended, it will not apply to biological data. The requirement to submit all biological data generated with NEP funds to ODES would remain in effect. OWOW is currently using the state of Florida as a test case to determine if this policy should be amended. Florida has been chosen as the test case because the state requires all water quality monitoring data to be submitted to STORET. Under this test case, a data description form will be required for each water quality data set submitted to STORET. The data description form will require the program's data managers to provide a QA/QC description including the goals of sampling program, station descriptions, and sampling techniques. Each form will be reviewed by ODES Technical staff, and a QA/QC report will be stored on-line in ODES. Workshop Summary 28 ------- B. NEP Inventory Tom Gulbransen - Battelle Ocean Sciences The data management representatives revised the overview of estuary program hardware, software, and documentation located in the Workshop notebook. The following chart indicates the diversity among estuary programs in the hardware and software being used, the number of people dedicated to data management support, and the documentation that has been produced. Program Chesapeake Bay TIER I APES Buzzards Bay LISS Narragansett Puget Sound Hardware PC's Workstations VAX Data General-5+ Sun Workstations (> Gigabyte) PE VAX-State NCC VAX-University State Systems PCs MicroVaxII Prime-Unix VAX-EPA Lab (University) Software ARC/INFO Models DBs Custom SAS Image /Vid. ARC/INFO STORET ERDAS SAS ISM ARC /INFO ORACLE SAS Models Paradox Focus ARC /INFO Custom People Involved 30 1 Part-time 1 Part-time <4 Distributed Documenta- tion Index Data Transfer Doc. Data Dictionary Bibliography Data Mgt. Plans Needs Assessment Data Dictionary Functional Description Data Mgt. Plan Data Reqrmts. System Description System Design Needs Survey Data Index Index of Data Needs Survey User Manual Data Submissions Manual SOPs for System Maintenance Needs Assessment Data Transfer Specs User Manual Bibliography Workshop Summary 29 ------- Program San Francisco TIER II Delaware Bay Delaware Inland Bay Galveston NY/NJ Harbor Santa Monica Sarasota Bay TIER III Barataria- Terrebonne Casco Bay Indian River Mass Bay Tampa Hardware MicroVAX - PC Workstation-St. VAX-St. PC's (University) - - MVIOK DG (St.) Workstation File Server Workstation PCs VAX-State PCs Software Info Custom -ODES ARC/INFO ORACLE Custom STORET -COMPAS -ODES Custom TRAKS (state) -ODES Distributed -ODES ARC/INFO-State MARS MOSS ERDAS Integraph ARC/INFO -ODES -ODES STORET ARC/INFO ORACLE -ODES STORET People Involved <2 Part-time =2 Part-time Distributed Part-time ~2 "1/2 1 Documenta- tion User Manuals Monitoring Plan - Preparing Design/ Dev (User Manual?) Plan Index Bibliography Data Mgt Strategy Data Inventory Data Mgt. Needs Assessment Draft Strategy System Design Data Mgt. Plan - - Workshop Summary 30 ------- C. NOAA Systems (COMPAS) Craig McCulloch - Texas Water Commission 1. What is COMPAS? COMPAS is a desktop information system which integrates maps and graphics. As a desktop information system, COMPAS provides a microcomputer based system which is highly focused for the user. The system allows direct access to the data as well as simple, interactive operations and powerful analytical capabilities. COMPAS runs on an Apple Macintosh n and incorporates several commercial software packages. HyperCard provides the graphical front-end for the system. ORACLE, a relational data base, maintains and manipulates the data behind the front-end. MapMaker displays geographic information on maps. Although MapMaker is not as sophisticated as ARC/Info, it is adequate for the system's mapping requirements. The following modules are currently being developed in COMPAS. 1) Physical & Hydrologic 2) Recreation 3) Non-point Sources 4) Wetlands Permits 5) Land Use 6) Marine Resources 7) Point Sources 8) Facilities 9) Habitats 10) NOAA Monitoring 11) Shellfish Growing Waters 12) Housing & Population 13) WQ Monitoring 14) Coastal Tracts 15) Water Rights 16) Streamflow Gages COMPAS also provides estuarine models. These models including two- dimensional steady-state estuarine models that can allow users to study salinity or conservative and non-conservative water quality pollutants. 2. Who is Developing COMPAS? The first twelve modules listed above primarily contain NOAA data. The last four modules are being developed with data from Texas. Many groups are involved in developing these four modules including: Texas Water Workshop Summary 31 ------- Commission, Department of Health, Parks and Wildlife, General Land Office, Texas Natural Resources Information System, Water Development Board, Railroad Commission, and the University of Texas' Bureau of Economic Geology. The data collected to be maintained in the system is being reformatted to a predefined standard. To eliminate the amount of reformatting, the system developers are trying to identify the specific data that is actually used in the decision making process and provide that data in COMPAS. For example, if summary statistics are used for the analysis, only those statistics are stored in the data base. The entire wealth of source data is maintained on the mainframe. 3. Examples of COMPAS Modules Each module provides the user with a variety of options. For example: The Marine Resources module presents the temporal distribution and relative abundance for a specific taxa in an estuary. Pull down menus easily allow the user to select a different taxa or estuary. The Water Quality Monitoring module allows a user to select a specific monitoring station. COMPAS displays summary statistics for the monitoring data available for the selected station and time period. The user can easily select a different station, time period, parameter group, or statistic using the pull-down menus. A user identifies a study area in the Coastal Tracts module and retrieves a map of the area displaying all of the monitoring stations. The user can then select a specific station in the study area, and COMPAS will display summary statistics for the the monitoring data that has been collected at that station. Workshop Summary 32 ------- D. ODES/GIS Connection Christopher Kroot - Casco Bay Program 1. Why Maine Chose ODES & the ODES/GIS Connection When the Casco Bay program started, managers and scientists evaluated the ODES system as an option for managing the data collected for the Bay. The program decided to use ODES because the system maintains a substantial amount of data, has a flexible data structure, and is integrated with a GIS application, the ODES/GIS Connection. In addition, the managers and scientists agreed that ODES provides the analytical capabilities to meet most of their needs. If the application does not meet their needs, they can download the data and use the file in any commercial software package (e.g., Lotus 1-2-3, SAS). 2. Tailoring the ODES/GIS Connection for Maine The scientists and managers from different agencies in Maine were involved in the design and development of the ODES/GIS Connection application. As a result, the system design specifically meets their needs. The GIS will provide easy access to data for the public, state agencies, environmental groups, universities, and local governments. Approximately 22 agencies are involved in developing Maine's system. By allowing the agencies to be directly involved in the design and development of the system, the agencies have a greater interest in ensuring that the system continues to be supported. The state has developed the majority of the data. In addition, the program has been working with OWOW and AMS to develop the menu-driven interface for the ODES/GIS Connection for about a year. It has been a challenging project because many groups are involved in developing the system. Each group has its own set of expectations from the system, and only a limited amount of money is available to apply towards the contract with AMS to develop the system. The program feels that they have been successful in negotiating compromises among the groups to provide the best system possible to meet each groups needs. 3. How the ODES/GIS Connection Works The key to the ODES/GIS Connection is the menu-driven interface, because it allows someone who does not know any thing about computer programming to easily use the system. The interface consists of a series of general headings. The general headings are pull down menus which allow a user to select monitoring programs, file types, parameters, and date ranges. The "Draw" menu prompts the user with a list of available data coverages such as shellfish beds, marine water quality sampling stations, critical Workshop Summary 33 ------- habitats, and non-point source dischargers. Any additional coverages can be included in the application. The user can review the list of available data layers and select the specific overlays of interest. The ODES/GIS Connection also provides a dynamic legend which displays the specific coverages that user selected and the corresponding color and shape. The system resides on the EPA prime, but it will be ported onto Maine's SUN. The base ODES/GIS Connection application is designed to be hardware independent, thus it can run on any program's hardware. Data from PCS, ODES, STORET, or other systems is downloaded once a week or month (depending on the frequency of data updates) to a floppy disk or tape. The data is then uploaded into the ODES/GIS Connection application. Once the application is completed, Maine will look into developing communication links to these systems similar to the ODES/STORET Bridge. However, at this point developing those communication links is too expensive. Anyone that has access to the EPA computing network will be able to access the application in the future. The user only needs to obtain a modem and a communications software package that emulates a techtronic terminal. The EPA is committed to helping other programs use this software. EPA and Maine's purpose in developing the ODES/GIS Connection was not to give Maine a specialized tool, but to develop an application that can be used by other programs. When the menu-driven interface was designed and developed for the state of Maine, it was particularly challenging to design a user-interface that would be specific enough for Maine, but also general enough to be used by other programs. However, the generic design has been successful, and the application can be tailored to meet the needs of any program. 4. How Maine is Using the ODES/GIS Connection The Department of Environmental Protection will be using the menu-driven system to manage most of its GIS related data. The data base will soon include information from AIRS, PCS, STORET and ODES. As a result, an analyst is not limited to one type of information. The Casco Bay program is fortunate that the senior management in the Department of Environmental Protection are committed to cross-media applications. The managers recognize the importance of making natural resource decisions by evaluating the "big picture". Maine is also planning to use the ODES/GIS Connection application for evaluating the state's watershed. Maine has a considerable amount of non- point source information that is currently being incorporated into the menu- driven system. Workshop Summary 34 ------- The state of Maine is one of three states that has a mandated Growth Management Comprehensive Planning Law where all towns and communities must develop a plan for pollution prevention and growth management. Local officials in the towns in Maine, especially in the water districts, recognize that the ODES/GIS Connection will provide them with access to zoning and growth management information. Workshop Summary 35 ------- PLANNING AND PUTTING IT ALL TOGETHER I. NEP Data Management Policy: Review - Mark Curran A. Need for an NEP Data Management Policy A significant portion of the funds provided by the National Estuary Program are spent on information management. The EPA has the responsibility for maintaining and providing access to the information collected as part of the NEP. As a result, the EPA has established a data management policy which requires all data generated with NEP funds to be submitted to a central data system, the Ocean Data Evaluation System (ODES), so that the data can be maintained on a national level. Please note that the policy does not require a program to use ODES for analysis. The EPA chose ODES as the central data storage system for many reasons. Specifically, the system is agency approved and supported. ODES cost four million dollars to develop, and continues to be supported through EPA funds. In addition, ODES is fully developed and has been in use since the beginning of the 301 (h) program. The system provides data accessibility at a national level. The ODES data submissions process incorporates QA/QC procedures, so that the data maintained is of the highest quality possible. The ODES Technical Staff reviews the data and prepares a QA/QC report which is stored on-line in the system. In addition, the system uses consistent formats for data which will allow EPA, or anyone, to analyze data across programs or study areas. B. Issues Resulting from the Policy The EPA recognizes that this policy raises many concerns for estuary programs. For example, older programs were putting data on the NCC mainframe using NODC codes. To address this problem, the EPA asked the Tier I estuary programs to identify priority data sets. EPA would pay the costs for converting the data sets to ODES, if the program was going to use ODES as its data management system. Historical data presented additional issues related to the policy. Data that was collected prior to the 1990 fiscal year is not required to be submitted to ODES. However, if a program chooses to use ODES as its data management system, this data would likely need to be converted to the same format as the newly collected data. The programs have the responsibility for reformatting the historical data. Workshop Summary 36 ------- EPA recognizes that this policy may require the creation of new ODES file types. A survey was distributed to all estuary programs soliciting their needs for ODES file types to accommodate the program's data. Although the summary results are inconclusive, it seems that a file type related to non- point sources needs to be developed. However, the specific parameters to be stored for non-point source information are difficult to pinpoint. As a result, the specific requirements of the non-point source file type need to be focused. ODES is not a static system. It is flexible and can accommodate new file types as well as changes to existing file structures. One of the largest issues resulting from the policy relates to the cost of using the ODES system. Specifically, the management conference has the responsibility for formatting the data for ODES. EPA recommends that the programs include this stipulation in the contracts established with Pis. Once the data is submitted to ODES, EPA pays the cost for the QA/QC process and loading the data into ODES. When the policy was originally established, the EPA felt the need to disperse the ODES system usage fee among the programs. However, the $20,000 fee per program per year inhibited the use of the system. As a result, the EPA has assumed the cost of the ODES usage fees for the estuary programs. Please note that the funds provided to support ODES usage for the NEP are taken from the operating budget, not program funds. If a representative from a program is interested in using the system, he or she simply needs to call the ODES hotline to obtain a user ID and password for the NCC mainframe. The EPA also recognizes that many programs have stored data in other national data bases, specifically STORET, and need access to this data. As a result, EPA has developed and is enhancing the ODES/STORET Bridge which will provide access to all STORET estuarine water quality data. C. NEP Support The EPA intends to enforce this policy, but is committed to providing support to assist programs in complying with the requirements. Contractor support is available through the NEP from AMS, Tetra Tech and Battelle. General information management support can be obtained from Battelle. ODES support is available through AMS and Tetra Tech and includes, among other things, hotline access and training. The NEP Data Management policy does not apply to GIS data, however the CIS National Program Office (GISNPO) can provide support to the NEP. The GISNPO supports many activities to disseminate information and provide assistance related to GIS applications. The program supports regional GIS teams and an outreach program. The outreach program distributes a newsletter and maintains a bulletin board system to help disseminate Workshop Summary 37 ------- information. The program is in the process of distributing a survey to identify training needs. Training courses should begin this summer. GISNPO supports the GRIDS system which maintains national data layers that have been developed by various agencies. These data layers can be provided to other programs at no cost to the program. National mapping requirements have also been identified by GISNPO. These requirements are sent to the Office of Management and Budget to identify priority data layers that should go on-line in the GRIDS system. GISNPO is coordinating a spatial data management plan and a PC GIS study to establish priorities and identify user needs. In addition, the GISNPO worked with the Office of Policy Planning and Evaluation to conduct an human health and environmental assessment among all 10 regions. For additional information on this project contact Bev Martain. The GISNPO can assist the NEP in its efforts to effectively use GIS and obtain information about recent developments in GIS applications. The names of NEP data managers should be added to the mailing list for the GISNPO newsletter and to the bulletin board system. In addition, the GISNPO can help facilitate GIS information exchange among the programs. For example, sources of GIS support can be identified for a program. In addition, the application of GIS to meet program requirements and objectives can be explored and technical support can be made available. Other participants highlighted several other sources of information that are available to the NEPs: The River Reach System provides hydrography and point-source discharge information. The Information Resources Directory is an index of all Office of Water information systems that was developed by OIRM. COASTNET is a bulletin board system that helps users disperse information D. EPA Data Management System Modernization EPA is beginning a modernization effort to combine the existing water management information systems. The modernization is currently in the planning stages, and EPA is obtaining input from system users. A symposium is planned for September where EPA managers are meeting to identify issues to consider and to develop a preliminary modernization plan. The new system will contain one data base to include all physical, chemical and biological data. The system will be menu-driven and run on the IBM Workshop Summary 38 ------- 9000 with telecommunications and full technical support. The proposed modernization effort will not produce a system that will be ready to test for at least 4-5 years. In the interim, the EPA continues to fully support existing systems, including ODES. Workshop Summary 39 ------- II. Costs and Benefits of Data Systems Development A. OIRM System Life Cycle Development Methodology & EPA Approved Hardware/Software Jean Sammon - Office of Information Resource Management 1. Life Cycle Management Series OIRM has developed the Life Cycle Management Series which contains the following documents: Volume A - Mission Needs Analysis Volume B - Preliminary Design and Options Analysis Volume C - System Design, Development and Implementation Supplemental Guidance to Volumes A and 6 - Guidance for Developing Image Processing Systems in EPA Supplemental Guidance to Volume B - EPA/APD Applications Guidance to Hardware/Software Selection Operations and Maintenance Manual In addition to these documents, additional references are also available through OIRM: EPA Information Security Manual NDPD Operational Policy Manual ADABAS Application Development Procedures Manual Guide to NCC Services 2. Importance of OIRM Documentation These documents were developed to provide criteria for decision-making, and to address project scope, planning and management. In addition, these documents will help managers to clarify expectations and requirements of the system to promote user involvement. Agency-wide standards are also provided in these documents. Workshop Summary 40 ------- These documents are important because millions of dollars are expended each year on developing, operating and maintaining systems. As a result, programs need to invest resources wisely from the beginning of the system life cycle. Agency systems are becoming increasingly complex because more emphasis is being placed on cross-media analysis, data integration and public access. These complexities require improved focus on sound data administration and for a systematic approach to defining needs and developing applications. 3. Selecting EPA Approved Hardware and Software There is a wide range of EPA approved hardware and software options to consider. Selecting EPA approved hardware and software leads to increased compatibility among EPA systems, the elimination of "Islands of Technology", and the facilitation of data sharing. Significant benefits result from selecting EPA approved hardware and software. For example, EPA support is available for training, and hardware or software upgrades. In addition, system-life cycle costs are reduced, because the hardware and software has already been tested, and a pool of experts is available for reference. Workshop Summary 41 ------- B. Long-term Systems Maintenance and Operations - Issues Gary Labovkh - American Management Systems, Inc. 1. Phases of the System Life Cycle The system life cycle consists of four phases: Needs Analysis/Preliminary Options Detailed Design of the System Function Software Development, Testing, Integration, Verification & Validation Operations & Maintenance 2. Percentage Cost Comparison of System Life Cycle Phases Reviewing the percentage cost comparison for System Life Cycle phases indicates the relative importance of maintenance and operations within the system life cycle: Needs Analysis/ Preliminary Options - 5% Detailed Design of the System Function -15% Software Development, Testing, Integration, Verification & Validation -35% Operations & Maintenance - 45% The operations and maintenance phase is the vehicle for preserving the investment in the system. If managers do not adequately plan for operations and maintenance, the system is doomed to failure. 3. Elements of System Maintenance and Operations The first element of system maintenance and operations is system operation. System operation includes activities such as assigning user IDs, updating system dictionaries, and performing system back-ups. In addition, the system and data base usage is continually monitoring, and minor system enhancements are developed, implemented and tested. For a system to operate most efficiently, someone should be appointed system administrator to oversee the above activities. Workshop Summary 42 ------- o Technical user support is the second element of system maintenance and operations. One component of technical user support is user training. For a system to be optimally used, people need hands-on training. Training teaches the user the most efficient ways to use the system. Informed users eliminate the likelihood that he or she will submit useless batch jobs tying up the system resources for others. Access to a user support hotline is an additional component of technical user support. Providing this rapid response service indicates to the users that the system support team is dedicated to making certain they obtain the information needed. Hardware and software support should also be a component of technical user support. If users cannot resolve hardware and communications issues to access the system, they simply will obtain the information they need elsewhere. The third element of system maintenance and operations is developing system documentation. When documentation is developed, it must be geared toward the appropriate audience. For example, the ODES system has four different documents, each developed for a specific purpose: 1) User's Guide, 2) Data Submissions Manual, 3) Data Briefs, and 4) the Tool Manual. After the documentation is developed, it must be updated as the system changes and evolves. In addition, the system support team must develop a plan to efficiently disseminate the documentation. Implementing hardware and software upgrades and improvements is the fourth element of system maintenance and operations. Any affected system features must be modified to appropriately handle the new versions of hardware and software, and the system must be tested thoroughly. In addition, the documentation must be updated to reflect the system changes, and the users must be informed of the system updates. The fifth element of system maintenance and operations is debugging errors. First, the source of the error must be identified. Second, the optimal solution to the problem must be determined. Then, the system support team must identify any effects from implementing the chosen solution, so that another system feature does not crash due to the selected "solution". Finally, the system must be thoroughly tested to ensure that the error is resolved. The sixth and final component of system maintenance and operations is evaluating major system modifications. When someone proposes a major system change such as the ODES/STORET Bridge, the system support team must determine if the modification will effectively solve the information management problem. In addition, the team must determine it is the most efficient way to solve the problem. A realistic evaluation of the extent of the modification must be made to determine how long the modification will take to implement and how much it will cost. Workshop Summary 43 ------- 4. Reducing Operations and Maintenance Costs Several operations and maintenance benefits result from developing a system using EPA approved hardware and software platforms. Specifically, the hardware and software has already been tested, and a pool of experts already exists for reference. In addition, training support is available, and support is also available when hardware and software is upgraded. It is important to note that system maintenance and operations begins after the system is fully functional. Although this is the last segment, 45% of the money spent during the system's life cycle is on system maintenance and operations. As a result, system planners must consider the long-term personnel and financial resources needed to support systems maintenance and operations. Workshop Summary 44 ------- in. Proposed Action Items The Workshop participants were given the opportunity to identify types of data management support that would be useful for the NEPs. The following suggestions were made: Develop a list of documentation that is available through the individual programs, OWOW, and OIRM. Prepare a summary of each program's data management strategy including an assessment of how well the strategy is working. Support the wide spread use of COASTNET to facilitate the communication between programs. Include the names of the Workshop participants on all program newsletters. Prepare a report documenting the QA/QC procedures in use by each program including a technical analysis of those procedures. Support additional opportunities for NEP program data managers to meet to discuss program concerns, goals and objectives. Workshop Summary 45 ------- ------- Appendix A Small Discussion Groups Group Facilitator: Mark Shibata Note Taker: Carl Hanson Spokesperson: Dave Tomasco Name 1) Jenifer Steele 2) Stephen Hale 3) Mindy Lemoine 4) Russell Kiesling 5) Charlotte Schwartz 6) DaveTomasko 7) SueHawes 8) Dillon Scott 9) BUlMuldrow Program/Organization Tier Albemarle-Pamlico Estuarine Study I Narragansett Bay Project I Delaware Estuary Project n Galveston Bay National Estuary Program n Galveston Bay National Estuary Program n Sarasota Bay National Estuary Program n Barataria-Terrebonne HI Massachusetts Bay Program in OIRM Group #2 Facilitator: Bob King Note Taker: Tom Gulbransen Spokesperson: Tom Gulbransen Name 1) Tim Johnson 2) Sue Kenney 3) Frank Shipley 4) Ron Embry 5) Greg Du Cote 6) Holly Greening 7) Bob Day 8) Drew Puffer 9) Jean Sammon Program/Organization Tier Albemarle-Pamlico Estuarine Study I Long Island Sound Study I Galveston Bay National Estuary Program n Galveston Bay National Estuary Program n Barataria-Terrebonne in Tampa Bay ffl Indian River Lagoon HI Gulf of Mexico Program OIRM Workshop Summary A-l ------- Group #3 Facilitator: Gary Labovich Note Taker: Lisa Eunice Spokesperson: Robert Smith Name 1) Neil MacGaffey 2) Susan Smith 3) Cynthia Print-Ham 4) Bob McFarlane 5) Robert Smith 6) Craig McCulloch 7) Chris Kroot 8) John Piper Program/ Organization Buzzards Bay Project Narragansett Bay Project Long Island Sound Study Galveston Bay National Estuary Program Santa Monica Bay Restoration Project Galveston Bay National Estuary Program Casco Bay Great Lakes National Program Office Tier I I I Workshop Summary A-2 ------- Johnson Jennifer Steele Appendix B Appendix D NEP Data Management Workshop Participant List Albermarle-Pamlico Estuarine Study 512 N. Salisbury Street Room 1125 Raleigh NC 27604 Albermarle-Pamlico Estuarine Study P.O. Box 27687 (919) 733-2090 (919) 733-0314 Gary Labovich Lisa Eunice Rosem«try Watt Carl Hcjison Greg DiiCote Sue Hawes Raleigh NC 27611 AMS 1777 North Kent Street 7th Floor Arlington VA 22209 AMS 1777 North Kent Street 7th Floor Arlington VA 22209 AMS 1777 North Kent Street 7th Floor Arlington VA 22209 AMS 1777 North Kent Street 7th Floor Arlington VA 22209 Barataria-Terrebonne Estuary Program Dept. of Natural Resources /Coastal Management P.O. Box 44487 Baton Rouge LA 70804-4487 Barataria-Terrebonne Estuary Program Corps of Engineers P.O. Box 60267 New Orleans LA 70160 (703) 841-6970 (703) 841-6032 (703) 841-5513 (703) 841-6873 504-342-7936 504-862-2518 Tom Gtilbransen Neil MacGaffey Battelle Ocean Sciences 397 Washington Street Duxbury MA Buzzards Bay Project Mass. Coastal Zone Management 100 Cambridge Street, Suite 2006 Boston MA 02332 02202 617-934-0571 (617) 727-9530 Workshop Summary B-l ------- Christopher Kroot Mindy Lemoine Russell Kiesling dixB Appendix is NEP Data Management Workshop Participant List Casco Bay Maine DEP State House Station #17 Agusta ME 04333 Delaware Estuary Project EPA Region III 841 Chestnut Street Philadelphia PA 19144 Galveston Bay National Estuary Program University of Houston-Clearlake 2700 Bay Area Blvd Box 164 Houston TX 77058 (207) 289-4292 (215) 597-3697 (713) 283-3950 Frank Shipley Galveston Bay National Estuary Program University of Houston-Clear lake 2700 Bay Area Blvd Box 164 Houston TX 77058 (713) 283-3950 Craig McCulloch Galveston Bay National Estuary Program Texas Water Commission P.O. Box 13087 Austin TX 78711 (512) 371-6356 Robert McFarlane Galveston Bay National Estuary Program McFarlane & Associates 9503 Sharpview Drive Houston TX 77036 713-772-8294 Charlotte Schwartz Galveston Bay National Estuary Program Texas Natural Resources Information System P.O. Box 12321, Capitol Station Austin TX 78711 (512) 463-8402 RonEmbry Andrew Puffer Bob Day Galveston Bay National Estuary Program Baytown Refinery - Exxon USA P.O. Box 3950 Baytown TX 77522 Gulf of Mexico Program Building 1103, Room 202 John C. Stennis Space Center Stennis Space Center MS Indian River Lagoon 1900 S. Harbor City Blvd. Suite 109 Melbourne FL 39529 32901-4749 (713) 425-3333 (601)688-3726 (407) 984-4950 Workshop Summary B-2 ------- thii Pring-Ham Appendix B Appendix i> NEP Data Management Workshop Participant List Long Island Sound Study U.S. EPA Region I WQE425 JFK Building Boston MA 02203 617-565*4437 Sue Kennedy Long Island Sound Study Connecticut Department of Environmental Protection 122 Washington Street Hartford CT 06106 (203) 566-6690 Dillon :5cott Massachusetts Bays Program 100 Cambridge Street CZM Boston MA 02202 (617)727-9530 ext415 Susan Smith Stephen Hale Naragansett Bay Project EPA Environmental Research Lab 27 Tarzwell Drive Narragansett RI Narragansett Bay Project 291 Promenade Street Providence RI 02882-1198 02908-5767 (401) 782-3044 (401) 782-3044 "ean Sammon Bill Muldrow Mark Curran Joel Saher Bob Kinjj OIRM 401 M Street, S.W. Mail Code: PM-211D Washington D.C. 20460 OIRM 402 M Street, SW Mail Code: PM-218D Washington DC 20460 OWOW U.S. EPA Fairchild Buflding 401 M Street, S.W. Mail Code: WH-556F Washington D.C. 20460 OWOW U.S. EPA Fairchild Building 401 M Street, S.W. Mail Code: WH-585 Washington D.C. 20460 OWOW U.S. EPA Fairchild Building 401 M Street, S.W. Mail Code: WH0556F Washington D.C. 20460 (202) 382-5914 (703) 883-8878 (202) 475-8483 (202) 475-8484 (202) 475-7119 Workshop Summary B-3 ------- Robert Smith Dave Tomasko Appendix B Appendix D NEP Data Management Workshop Participant List Santa Monica Bay Restoration Project EcoAnalysis, Inc. 221 East Matilija Street, Suite A Ojai CA 93023 Sarasota Bay National Estuary Program 1550 Ken Thompson Pkwy Sarasota FL 34236 (805) 646-1461 (813)361-6133 Holly Greening Tampa Bay Estuary Program 111 7th Avenue South (813) 893-2765 Mark Shibata Ken Green St. Petersburg FL Tetra Tech 3746 Mt. Diablo Boulevard Suite 300 Lafayette CA VigYan, Inc. 5203 Leesburg Pike Suite 900 Falls Church VA 33701 94549 22041 (415) 283-3771 (703) 931-1100 Workshop Summary B-4 ------- Appendix C Summary of Participants' Evaluation Sheets This appendix summarizes the comments offered by the participants from the NEP Data Management Workshop held in New Orleans, Louisiana on May 20-21. There were approximately 40 participants from the various estuary programs, the Great Lakes program, the Gulf of Mexico program and OIRM. The participants were generally pleased with the Workshop, as demonstrated by the comments presented below. These comments summarize 18 evaluation forms. (5=excellent) Numerical Ratings: Average 1) Overall format of conference 4.2 2) Relevance of topics discussed 4.2 3) Time allowed for discussion 4.3 4) Opportunity for interaction among 4.4 participants 5) Emphasis on Estuary Program 4.0 concerns 6) Conference facilities 3.8 The following is a summary of the comments made to the following questions: What additional information, relative topics discussed, would you like to have seen presented? How OWOW is interacting with other media program offices Evaluation of how well different programs are working NOAA- Weather/Meteorological data and systems None Which session formats (i.e. panel discussions, small group discussions, etc.) do you think were most effective? The small group discussions were very popular Generally people liked the mixed format of panel discussion and small breakout group discussion Workshop Summary C-l ------- What is the single most important role OWOW should undertake to support the transfer of information among those involved with the National Estuary Program? Keep programs informed on state of the art program designs (i.e. what is successful and provide technical assistance to deal with the challenges). Repeat the data management workshop annually Expansion/improvement of tools which access/analyze/manipulate ODES data. Simply being there Get STORET/ODES modernized Newsletter to help facilitate Data Base Management information exchange "Demand use of COASTNET' and develop an 800 number Do you see a need for other types of conferences? Yes, monitoring program designs and their relationship to Data Base Management. Funding solutions after the NEP Modeling in the NEPs-is it worthwhile? Yes, the data management conference should continue Activity which stresses transfer of relevant, summary info to managers/decision makers. What suggestions do you have for future NEP data management meetings (e.g. format location, participants, etc.)? Same type of format, same participants on a yearly basis More working on-line Data Base Management demonstrations Emphasize more case studies (i.e. programs with completed characterizations) Workshop Summary C-2 ------- 4ft (ft ------- j?i ii MI EI O b § T) I 5 5 DATE DUE SF.P 25 nn HIGHSMITH #45115 ------- |