Science Serving Maryland's Coasts


From Genes to Ecosystems: Integrating Measures of Aquatic Biodiversity and Ecosystem Health Within Urbanizing Bay Watersheds

Principal Investigator: 

Robert Hilderbrand

Start/End Year: 

2014 to 2016


Appalachian Laboratory, University of Maryland Center for Environmental Science

Co-Principal investigator: 

Stephen R. Keller, Appalachian Laboratory, University of Maryland Center for Environmental Science; Alyson Santoro, Horn Point Laboratory, University of Maryland Center for Environmental Science

Strategic focus area: 

Resilient ecosystem processes and responses


Land-use changes create numerous adverse impacts to stream ecosystems within the Chesapeake Bay watershed, including degrading water quality for both human and non-human use. Fish and benthic macroinvertebrates are traditionally used as indicators of biotic response to watershed disturbance.  However, these indicators do not necessarily reflect the status of key ecosystem processes such as the metabolism of nutrients and other pollutants that otherwise can flow to coastal waters. Conversely, microbial communities drive many ecosystem processes, including nutrient cycling, and thus their diversity and composition have great potential to assess and possibly mitigate impacts on aquatic ecosystems. Microbial pathogens are also important to human health, and identifying their prevalence and landscape associations may have profound implications for resource management and planning. However, almost nothing is known about the biogeography of microbial community diversity in stream ecosystems and how this varies with watershed alterations.

Robust methods to characterize the diversity of microbes, many of which cannot be cultured in the laboratory, have arisen recently due to high-throughput DNA sequencing of whole microbial communities. Our objectives for this study are to characterize microbial community diversity in headwater streams throughout the Chesapeake Bay watershed of Maryland to test the following predictions: (1) Microbial diversity declines along gradients of land use and urbanization; (2) Microbial diversity is a more sensitive and reliable indicator of stream impairment than traditional bioindicators such as fish and benthic macroinvertebrates; and (3) The taxonomic composition of microbial communities reflects specific impacts of anthropogenic land use such as high loadings of nutrients or sediments into aquatic ecosystems.

The Maryland Biological Stream Survey (MBSS) performs annual assessment of stream condition across the state's watersheds. In spring and summer of 2014, we will team with MBSS to sample microbial DNA from water and sediments of 96 headwater streams. An overlapping set of 24 "Sentinel Sites" will be sampled again in 2015. In addition, 96 sites will be sampled by 4 citizen watershed associations, who will help design studies to characterize microbial response to a local environmental stressor. Microbial DNA will be used for high throughput 16S amplicon sequencing to characterize the composition and diversity of the microbial metagenomic community at each site and season. Microbial diversity will then be related to water quality, traditional bioindicators, and measurements of land use using multivariate statistical models. Environmental predictors include a suite of physic-chemical and land use variables that define an environmental gradient within Maryland ranging from pristine forested, to primarily agricultural, to highly urbanized watersheds.

Our study will robustly characterize the relationships between microbial diversity and land use change within the Chesapeake Bay watershed. Our results will be accessible and useful to diverse stakeholders with interests in water quality, aquatic resource management, and the health of stream ecosystems.


This section describes how this project has advanced scientific knowledge and/or made a difference for 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.

Summary: Researchers developed a method that uses the taxonomic diversity and abundance of microbial populations to measure effects of increasing urbanization and other environmental stressors on streams in the Chesapeake Bay watershed. The method offers an alternative approach to quantify the resulting impairments in water quality and may be more flexible and cost effective than existing measures.

Relevance: Scientists traditionally have measured diversity of fish and benthic macroinvertebrate species (e.g. mayflies) as indicators of stream health. However, these indicators do not necessarily provide useful information about key ecosystem processes, such as nutrient cycling, that are driven by microbial communities and can mitigate the impacts of disturbances on aquatic ecosystems. Microbial pathogens are also important to human health, and identifying their prevalence and location may have important implications for resource management and planning. However, little is known about how microbial community composition and diversity vary within a watershed and respond to changes in land use. 

Response: Robert Hilderbrand, Stephen Keller, and Alyson Santoro at the University of Maryland Center for Environmental Science performed genetic analyses of microbe populations collected from 96 stream sites across the Chesapeake Bay watershed in Maryland. Using recent advances in DNA sequencing, the researchers sequenced entire microbial communities at once. This allowed scientists to determine taxonomic richness and diversity. A graduate fellow funded by Maryland Sea Grant, Sarah Laperriere, organized work by three local groups of citizen scientists to collect stream water samples for analysis for this project. 

Results:  The researchers found evidence of the utility of this approach. They identified, for example, correlations between microbial taxonomic richness and specific ecosystem factors of interest, such as dissolved organic carbon concentrations, quality of stream-bottom substrate, and the area of forest and agricultural land in the surrounding watershed.  The researchers did not find statistically significant correlations between microbial richness and the standard, commonly used indices that are based on macroinvertebrate and fish richness. However, the researchers suggest that the microbiome data complement those indices usefully; analyses indicated that the microbes can faithfully repeat stream health assessments as designated by macroinvertebrates. The findings were presented to the Maryland Department of Natural Resources’ Monitoring and Non-Tidal Assessment Division and at an annual statewide meeting on stream monitoring. While the approach requires further testing, it has the potential to greatly expand stream monitoring cost effectively.