Susan E. Lorenz interned for the Environmental Protection Agency, Region 3.

Working in Region 3í¢â‚¬â„¢s Environmental Protection Agency on a systems-based, iterative environmental decision-making mechanism called Multi-criteria Integrated Resource Assessment (MIRA), a dataset on index of benthic integrity and several indicators for green infrastructure were developed.  Data was collected on benthic integrity from state governments as well as from many volunteer and research driven water-monitoring organizations.  After the initial data query, data was needed for an additional 2000 of 3700 12-digit hydrological units in the Mid-Atlantic region.  A data estimation procedure was developed based on multivariate discriminant function analysis to establish benthic integrity data points for the hydrological units with no data. In an additional project, data was collected on green infrastructure networks from the Mid-Atlantic States and local organizations.  A lengthy case study was developed on green infrastructure projects throughout the United States as means of constructing green infrastructure indicators for MIRA.

Report

During the summer of 2007, I spent 13 weeks working in Region 3 of the Environmental Protection Agency (hereafter EPA) in the Environmental Information and Assessment Division.  My work focused around a project entitled Multi-criteria Integrated Resource Assessment (hereafter MIRA) that informs environmental decision-making.  MIRA has been in development for over 12 years by two environmental specialists in the Air Protection Division of EPA, Cynthia Stahl and Alan Cimorelli. 

Since environmental decision making needs to become informed not just by science but by stakeholder values, MIRA provides a means to organize complex scientific data, expert judgment, and stakeholder values in a transparent, learning-based framework.  MIRA goes beyond traditional approaches to environmental decision-making and í¢â‚¬Å“seeks to facilitate decision analysis through an improved understanding and interconnection between both the scientific data and the societal values that are present in all environmental policy questions.í¢â‚¬ 

MIRA has been used in several different applications including county-based analysis of ozone monitoring networks, a ground level ozone study, a Philadelphia Air Toxics study and most recently a Mid-Atlantic, 12-digit hydrological unit (hereafter HUC12) based analysis of public health and ecosystem risks.  This in-progress study is a demonstration in how public health and ecological criteria can be integrated and linked to EPA program activities.  Among the issues raised were those of scale, missing data, data compatibility, construction of appropriate indicators, and application of indicators to selected ecosystems.  So far, this study has successfully demonstrated the feasibility of using a large number and a large variety of indicators in a holistic systems-based analysis that can be linked to program accountability.  

My role in MIRA involved addressing missing data for the index of benthic integrity (hereafter IBI) in the Mid-Atlantic HUC12 analysis of public health and ecosystem risk.  An IBI is a synthesis of diverse biological information, which numerically depicts associations between human influence and biological attributes, specifically benthic macroinvertebrate populations.  It is composed of several biological attributes or 'metrics' that are sensitive to changes in biological integrity caused by human activities.  The multi-metric (a compilation of metrics) approach compares what is found at a monitoring site to what is expected using a regional baseline condition that reflects little or no human impact.  IBI is considered to be one of the best indicators of water quality by aquatic biologists, thus it was key to have a robust dataset of IBI for MIRA.

States in the Mid-Atlantic Region, Delaware, Maryland, Pennsylvania, Virginia and West Virginia provided data on IBI.  This data was GIS-processed to the HUC12 scale and there was IBI data for 1700 out of the 3700 HUC12s in the Mid-Atlantic region.  To address this issue, I spent several weeks reaching out to volunteer and research driven water-monitoring organizations that take measurements of benthic macroinvertebrates and chemical parameters (pH, dissolved oxygen) in streams and rivers.  Most of the organizations were happy to share their data with the EPA, but there were a few organizations that would not oblige.  After I had collected all of this data it was compiled into one spreadsheet and GIS-processed to the HUC12 scale.  The result was an additional 200 IBI data points to complement the 1700 Mid-Atlantic HUC12í¢â‚¬â„¢s with data.  This was a significant addition to the dataset, as IBI data is difficult to get because it relies on scientists going into the field and spending several hours to collect and identify the macroinvertebrates at each site.  Along with compiling all of this data, I also researched how each organization measured and computed their multimeric indices and their quality assurance plans.

The next steps in addressing the IBI data issues was to find a way to estimate IBI points for the rest of the dataset.  I developed an approach with an EPA aquatic biologist based out of Wheeling, West Virginia.  We identified a type of multivariate statistics called discriminant function analysis (hereafter DFA) as a means of estimating the rest of the data. DFA develops a formula based on several other ecological data parameters and can suggest whether or not a HUC12 has an impaired or unimpaired benthic condition.  Further DFA analysis sub-classifies the data into categories of excellent, good, fair and poor benthic condition.  This analysis was run on the IBI data, and resulted in a dataset with IBI values for every HUC12 in the Mid-Atlantic region.  Along with IBI values, uncertainty parameters were developed from the multivariate model for every HUC12 with an estimated data point.  To complete the IBI dataset development, a PowerPoint presentation and detailed report were generated to complement and enhance the analysis.

As part of my internship, I also spent several weeks researching green infrastructure and suggesting the construction of indicators.  Green infrastructure is an approach to land conservation that facilitates interconnected networks of natural areas and other open spaces to conserve the function of ecosystem services and provide linked habitats for wildlife.  Green infrastructure drives green solutions in terms of smart growth and conservation, low impact development and ecosystem/watershed management.  The key concept behind green infrastructure is linking together large undeveloped tracts of land called í¢â‚¬Å“hubsí¢â‚¬  with smaller thinner land areas called í¢â‚¬Å“linksí¢â‚¬ .

To assess the possibility of indicators for green infrastructure, I put together a lengthy case study that combined the analysis of several different green infrastructure approaches conducted nationwide.  I also contacted state and county level governments and organizations to obtain data on green infrastructure networks that had been developed.  From this analysis, the indicators that could be used in MIRA to represent green infrastructure were determined.

This internship provided me with the opportunity to experience firsthand the time consuming and complicated nature of making environmental decisions.  I learned the difficulties in establishing collaborative partnerships with other organizations, and I also saw the benefits.  In this experience I was able to utilize my knowledge in ecology and environmental sustainability to further develop a decision-making mechanism, MIRA, which truly provides a new way of thinking about environmental decisions.  MIRA has the goal of moving the EPA away from the í¢â‚¬Ëœstove-pipeí¢â‚¬â„¢ approach of environmental management to more of a systems-based, iterative, learning and adaptive management approach.  During my internship, I felt that my skill set embodied many of the concepts embedded in MIRA.  This experience was crucial to my professional development, as I was able to express these skills on many levels.  From this experience, I am considering a career in the Environmental Protection Agency.

Internship Details