A Biologically-Optimized Environmental Classification of Maryland Streams: Assessing Impacts of Stream Burial and Responses to Climate Change

Principal Investigator:

Matthew C. Fitzpatrick

Start/End Year:

2012 - 2014


Appalachian Laboratory, University of Maryland Center for Environmental Science

Co-Principal Investigator:

Andrew J. Elmore, Appalachian Laboratory, University of Maryland Center for Environmental Science

Strategic focus area:

Resilient ecosystem processes and responses


OBJECTIVES: Our primary objective is to develop and make readily available to managers a biologically-optimized environmental classification of freshwater stream reaches in Maryland west of the Chesapeake Bay. In addition to considering environmental and land use influences on stream biota, we will investigate relationships between landscape connectivity and stream community composition and how these relationships vary by fish and benthic macroinvertebrates. We will assess the proportion of stream biodiversity lost as a result of 40 years of urbanization in the state and will identify stream reaches that may be particularity sensitive to land use change. Working with Maryland Department of Environment and Maryland Department of Natural Resources, our models will inform strategies to prioritize the protection and restoration of streams.

METHODOLOGY: We will use new maps of streams and stream burial and methodological advances in landscape ecology to develop environmental parameters that describe both local and landscape-scale characteristics of stream reaches. In addition to environmental predictors, we will use circuit theory, a new technique that has not previously been applied to aquatic networks, to quantify connectivity of streams. These variables, along with survey data for fish and macroinvertebrates from the Maryland Biological Stream Survey, will be analyzed using a new statistical approach for modeling community composition, Generalized Dissimilarity Modeling (GDM). GDM can transform and weight environmental variables such that they most closely match biological patterns and can accommodate nonlinearities commonly encountered in ecological datasets. Our models will be used to quantify impacts of stream burial on biodiversity and will identify stream reaches that have experienced the greatest impacts due to land use change.

RATIONALE: In 2010, the Environmental Protection Agency established a Total Maximum Daily Load (TMDL) for the Chesapeake Bay. Meeting the demands of the TMDL will require significant improvements to water quality throughout the Bay's watershed, including the restoration and protection of freshwater streams. Our work will provide comprehensive and high-resolution information on the spatial distribution of aquatic community types in Maryland streams that can be used to inform management actions. By classifying stream reaches into biological communities and by quantifying their relative similarity, our research will provide a means to prioritize stream reaches for restoration or protection. We will host our geospatial datasets on a web server and will hold a workshop to inform stakeholders of our findings and to solicit feedback to ensure the resulting products are of greatest applied benefit as possible.


This section describes how this project has advanced scientific knowledge and/or made a difference in the lives of coastal residents, communities, and environments. Maryland Sea Grant has reported these details to the National Oceanic and Atmospheric Administration (NOAA), one of our funding sponsors.

RECAP: Researchers created a computer model that predicts the distribution and composition of biodiversity in freshwater streams across a large region of Maryland. The scientists shared the tool with natural-resource managers to inform ongoing efforts to protect and improve these streams, which in turn could help improve water quality in Chesapeake Bay.

RELEVANCE: In 2010, the Environmental Protection Agency established a Total Maximum Daily Load (TMDL) to reduce excess amounts of nutrients and sediment flowing into the Chesapeake Bay. Meeting the demands of the TMDL requires significant improvements to water quality throughout the Bay’s watershed. One tool to accomplish this goal is to restore and protect freshwater streams. This research project has provided comprehensive and high-resolution information on the spatial distribution of aquatic species in Maryland streams. Categorizing stream reaches by species composition provides an empirical basis for identifying degraded and vulnerable stream reaches and prioritizing them for restoration and protection.

RESPONSE: Maryland Sea Grant supported research by principal investigators Matthew Fitzpatrick and Andrew Elmore of the University of Maryland Center for Environmental Science’s Appalachian Laboratory. They developed a biological classification of freshwater stream reaches in Maryland west of the Chesapeake Bay. A challenge in creating this model was to accurately predict species composition in stream locations that have not been monitored directly by the Maryland Biological Stream Survey (MBSS). The new model incorporates statewide data about physical parameters (e.g. slope and soil type) and biological parameters (e.g. fish and benthic macroinvertebrate species) to predict species composition in streams across the study area. The researchers created the model using statistical approaches, such as Gradient Forests, not previously used in this context. Through a competitive process, Miriam Johnston was selected as a Maryland Sea Grant Research Fellow and conducted research on this project.

RESULTS: Completed in 2014, the model classifies Maryland streams into 29 unique groupings of macroinvertebrate species and 22 of fish. The model indicates that the biodiversity of fish and macroinvertebrate assemblages can decline in a rapid, non-linear fashion in response to changes in physical parameters, such as declining forest cover caused by construction projects and other development. The researchers created an online website (http://streammapper.al.umces.edu/streamsbiodiv.html) that maps these community types in streams across Maryland. The website provides information regarding the physical and biological attributes of each 10-meter stream reach in Maryland west of Chesapeake Bay.

The researchers shared the model in a meeting with the Maryland Department of Natural Resources’ Monitoring and Non-Tidal Assessment Division. The scientists also described the model during a day-long workshop sponsored by the Maryland Water Quality Monitoring Council. Attendees included prospective users from management agencies, industry, non-profits, and academia.

Related Publications:

Johnston, MR; Elmore, AJ; Mokany, K; Lisk, M; Fitzpatrick, MC. 2017. Field-measured variables outperform derived alternatives in Maryland stream biodiversity models. Diversity and Distributions23(9):1054 -1066. doi:10.1111/ddi.12598. UM-SG-RS-2017-06.

Blois, JL; Zarnetske, PL; Fitzpatrick, MC; Finnegan, S. 2013. Climate change and the past, present, and future of biotic interactions. Science341(6145):499 -504. doi:10.1126/science.1237184. UM-SG-RS-2013-09.

Elmore, AJ; Julian, JP; Guinn, SM; Fitzpatrick, MC. 2013. Potential stream density in mid-Atlantic US watersheds. PLOS ONE8(8):1 -15. doi:10.1371/journal.pone.0074819. UM-SG-RS-2013-08.

Fitzpatrick, MC; Sanders, NJ; Normand, S; Svenning, JC; Ferrier, S; Gove, AD; Dunn, RR. 2013. Environmental and historical imprints on beta diversity: insights from variation in rates of species turnover along gradients. Proceedings of the Royal Society B-Biological Sciences280(1768) . doi:10.1098/rspb.2013.1201. UM-SG-RS-2013-11.

The Blue Crab: Callinectes Sapidus

An essential resource for researchers, students, and managers.  Get your copy today!

pile of cooked crabs