Grantee Research Project Results
2002 Progress Report: Real-Time Water Quality Monitoring and Modeling for Equitable Recreation on the Mystic River
EPA Grant Number: R829338Title: Real-Time Water Quality Monitoring and Modeling for Equitable Recreation on the Mystic River
Investigators: Deshpande, Vithal V. , Kirshen, Paul , Perez, Grace , Durant, John , Baise, Laurie , Minardi, Lee , Brukilacchio, Lisa , Chapra, Steve
Current Investigators: Deshpande, Vithal V. , Kirshen, Paul , Perez, Grace , Durant, John , Baise, Laurie , Minardi, Lee , Brukilacchio, Lisa , Hammett, Nancy , Chapra, Steve
Institution: City of Somerville , Mystic River Watershed Association , Harvard Graduate School of Arts and Science , Tufts University
Current Institution: City of Somerville , Mystic River Watershed Association , Tufts University
EPA Project Officer: Packard, Benjamin H
Project Period: October 1, 2001 through September 30, 2003
Project Period Covered by this Report: October 1, 2001 through September 30, 2002
Project Amount: $427,969
RFA: Environmental Monitoring for Public Access and Community Tracking (EMPACT) (2001) RFA Text | Recipients Lists
Research Category: Environmental Statistics , Water , Aquatic Ecosystems , Air , Ecological Indicators/Assessment/Restoration
Objective:
The objectives of this research project are to: (1) collect data on water quality indicators-fecal coliform, enterococcus, dissolved oxygen (DO), turbidity-along with data on depth, temperature, pH, conductivity, and meteorological conditions in real-time to develop an "early warning" water quality forecasting model; and to present real-time and timely water quality data and information to the public, using the Internet, color-coded flags, a variety of other printed and visual materials in several languages, and regular advisories in local print and broadcast media. The presentations will interpret the data into indices useful for everyday decisions about contact with the water. The presentations also will make recreational users and watershed residents aware of the real-time and timely data, indices, and the model available through this project. The focus is on reaching a large segment of the public, including those without easy access to the Internet, to stimulate interest in finding and using the information from the project, and to inform watershed residents about the connections between human behavior and water quality through educational efforts that emphasize the value of this information for personal decisions and community health.
Progress Summary:
During the past project year, we accomplished the following milestones: (1) equipment installation; (2) experimentation planning and execution; (3) biological and physical parameters data collection at five locations across the lower Mystic River Basin and Alewife Brook; (4) Web site design and real-time data collection on the Web; (5) planning and initial development for a predictive model based on logistic regression and multiple linear regression and statistical time-series modeling; and (6) public participation and input in developing the Web site design.
All above-mentioned milestones are required to pursue the state's goal to make the river swimmable and fishable by the year 2010. The data collection and analysis are helping to identify where biological parameters are affecting the stream. They also are helpful in understanding the impact of human activities on this vital natural resource in the region.
Equipment Installation and Data Collection. During the summer and fall of 2002, five real-time water quality monitoring stations were established along the Mystic River and Upper Mystic Lake. The five sites were chosen to provide information about water quality at locations where recreational contact is prevalent. Each station was equipped with a multiparameter water quality sensor, and a data acquisition and telemetry system. The sensors measured dissolved oxygen, temperature, pH, conductivity, and water depth at 15-minute intervals. The data were sent in real-time at radio frequency to a server on the Tufts campus in Medford and immediately uploaded onto a dedicated Web site. Biological parameter data was collected for (Monday-Saturday) 5 days a week and analyzed over a 24-hour period. The total number of data points collected during the first season is about 222,000. Data collected and confirmed through the quality assurance procedure are available online at http://www.mysticriveronline.org Exit .
Partnerships and Permits. Four of the stations were located at facilities owned by the Metropolitan District Commission (MDC), and the fifth station was set up in a facility owned and operated by the Massachusetts Water Resources Authority. Two Notice of Intent applications were filed¾one with the town of Winchester, the other with the city of Somerville¾to obtain permission to dig shallow trenches in wetland areas.
Changes to the System. We originally had intended to use cell-phone transmission to send the real-time monitoring data to Tufts; however, it was discovered by experimentation that radio-frequency transmission was more effective, and thus the decision was made to use radio transmission. We also had originally planned to operate a station at the Henderson Street bridge on the Alewife Brook between Cambridge and Arlington. The station was installed successfully; however, it was badly vandalized shortly after. To better secure the equipment, the station was relocated at approximately 1-km downstream to a Massachusetts Water Resources Authority facility. The Authority generously allowed us to install the equipment in one of its buildings and run the cables down to the Brook through a buried pipe.
Web Development. The specific objective for this part of the project is to: develop and implement effective methods to present real-time and timely water quality data and information to the public through a dedicated Web site showing the data. L. Minardi initiated an investigation into the software structure necessary to automatically post information from a data file onto the Web in September 2001. It was decided to use "ASP" Web pages with embedded VB Scripts (rather than Java Scripts) to control the changing data of the custom html Web pages. The Tufts Arts, Sciences, and Engineering server was chosen to host the Web page. C. Perez (Graduate Student-EMPACT Research Assistant) joined the effort in January 2002, and was given responsibility to assist in the design and implementation of the Web pages. M. Heberger (Graduate Student-EMPACT Research Assistant) was responsible for the installation of field sensors and the radio receiver that was mounted on a building on campus (Miller Hall).
When we learned that the PC receiving the data must be within 100 feet of the receiver, some design changes were made for the Web posting of data. The data collecting software and routines for automatically transferring the data to the Web server would be installed on the PC (named Aqua) near the receiver. Both daily created data files and archived data reside on Aqua. Custom routines were created to strip the most recent data from incoming field data and to transfer files via FTP to the Web server every 15 minutes. Bacterial data, which are manually gathered from the field sites, also are posted on the Web page. In addition, monthly data for all five sites are combined into a single file and posted as part of a download page available to the public.
We sought to make the design of the Web pages consistent with the Mystic River Watershed Association (MyRWA) Web Site. Moreover, it was decided after meeting with association representatives that two versions of data display be provided, one for "novice" users and another for more technically inclined viewers. As a result, the Web page includes a tab for "Summary" information and a tab for "Detail" information. Much effort was devoted to producing an attractive and user-friendly Web page with a formal, professional look to support the importance of water quality.
The Web site went online on May 3 with one field unit at the Amelia Earhart Dam. A second Web site was added on May 20, 2002. Real-time data posted every 15 minutes were available in June 2002, and in September a fourth Web site was added. Because of vandalism of instrumentation at one site, the fifth site was only available for a short time. The field sensors were removed from the water in early November for the winter. Data gathering is scheduled to resume in mid-April 2003. The Web site is active and available for downloading archived data over the winter months.
Data Modeling "Early Warning" Water-Quality Forecasts. The water-quality forecast model is in the initial stages of development. The water-quality and meteorological data collected from May to November 2002 are the data sets for initial model calibration and model validation. The current modeling effort is following a two-prong approach. The first effort uses logistic regression to develop an indicator model for water-quality in terms of the exceedence of standards for daily bacterial concentrations. This simple model will result in a safe or not safe prediction for each of the monitoring stations paired with probabilities of exceedence. The initial explanatory variables under investigation are: precipitation, air temperature, water temperature, stream flow, and combined sewer overflow (CSO) activation. We have gathered CSO activation data (from other sources) for the watershed to help understand the relationship between CSO activation and bacterial concentrations, although it will not be a part of the real-time model. The second effort uses a combination of multiple linear regression and statistical time-series modeling to predict a numerical value for bacterial concentration at each of the monitoring stations. In addition to the explanatory variables used in the simple model, this second effort will explore using dissolved oxygen, conductivity, pH, and turbidity as explanatory variables. Both models currently are under development.
Public Outreach. Public outreach has focused primarily on gathering a range of input from community members on the Web site, including issues of usability, flexibility, and universal access. Informal presentations were made to selected community members, including: Mystic Watershed Steering Committee (composed of various Tufts Department representatives and Mystic community members), the MyRWA staff and board members, and the City of Somerville MIS Director. Input also was gathered informally during meetings of local municipal officials and community organizations. A formal presentation was made last summer at the Environmental Educator's Conference, with useful feedback provided by audience members from across the country. The Massachusetts Executive Office of Environmental Affairs (EOEA) Watershed Initiative-Boston Harbor/Mystic Basin Team Meeting was the site of another formal presentation for outreach and input purposes, reaching local community members, as well as federal, state and regional officials. Feedback and input from these sources have been delivered to the EMPACT team at regular meetings and incorporated into the planning and implementation of the Web site design to the degree feasible given technical considerations.
Our next steps include expanding our Advisory Board beyond the limits of the Massachusetts Water Coalition (MWC). A revised board is anticipated to meet at least bimonthly, starting in late 2002, to address the Spring launch, which this team has negotiated with the scientists to be moved back to mid-April to address public information needs based on our observation of past recreational usage patterns of local waters. The revised team is to include representatives from the private boating community, the birding community, a Sandy Beach swimmer, an educator from one of the public schools located on the river, the MDC, a designee from the Blessing of the Bay Boat Club, a fisherman, the EOEA Basin Team Leader, the MWC, and local community river groups.
Future Activities:
In activities for Year 2 of the project, include: (1) data modeling; (2) data collection; (3) formal public display of the real-time data monitoring Web site; and (4) public outreach. A special public outreach program for the Web site is in the planning stage at this time and will be held on May 18, 2003.
Journal Articles:
No journal articles submitted with this report: View all 2 publications for this projectSupplemental Keywords:
real-time data monitoring, public education, environmental education, public awareness, natural resources, predictive modeling for river quality, Mystic River, Massachusetts, MA., RFA, Scientific Discipline, Water, ECOSYSTEMS, Ecosystem Protection/Environmental Exposure & Risk, Water & Watershed, Monitoring/Modeling, Civil/Environmental Engineering, Terrestrial Ecosystems, Environmental Monitoring, Watersheds, aquatic ecosystem, remote sensing, water resources, health effects, wetlands, community-based approach, bacteria, fecal contamination, sediment transport, stream ecosystems, community water quality information system, recreational water , community outreach, community tracking, water quality, aquatic ecosystems, ecological models, ecology assessment models, water management options, watershed assessment, land management, stream ecosystem, river ecosystemsRelevant Websites:
http://www.mysticriveronline.org Exit
Progress and Final Reports:
Original AbstractThe perspectives, information and conclusions conveyed in research project abstracts, progress reports, final reports, journal abstracts and journal publications convey the viewpoints of the principal investigator and may not represent the views and policies of ORD and EPA. Conclusions drawn by the principal investigators have not been reviewed by the Agency.