Ecosystems Studies & Management
Monitoring Stream Health
Undercut stream banks and sedimentation often occur in urban streams.
The altered composition and configuration of land use, such as expansion of impervious surface areas within a watershed, disrupt the hydrology and ecology of stream ecosystems. The inhibited infiltration of rainwater and snow melt in impervious areas results in reduced base flows and flashier stream hydro graphs that exhibit a reduced lag time between storm events and peak discharge. Stream channels are modified by these changes, quickening bank and stream bed erosion and increasing sediment loads. WHRC scientists have demonstrated the association of these land use changes with the degradation of biological, chemical and physical properties of streams within the Chesapeake Bay watershed. Stream health impacts have been carefully documented in Maryland by the Department of Natural Resources as part of the Maryland Biological Stream Survey. These metrics include, among others, macro invertebrate and fish Indices of Biological Integrity (IBI), Benthic Index of Biotic Integrity (BIBI) and number of Ephemeroptera, Plecoptera, and Trichoptera (EPT) species.
Maryland
WHRC researchers documented the statistical association between our mapped land cover variables (impervious cover, tree cover, crop and pasture land) and BIBI/EPT across small watersheds in Maryland. These spanned a wide range of land uses, from predominantly agricultural to mostly residential. The BIBI and EPT rankings were based on data from a number of MBSS sampling stations collected over a five year time period.
Map of Maryland showing MBSS (Maryland Biological Stream Survey) watershed locations and year sampled, and land cover variables depicted for a sample watershed in the Coastal Plain province. ISA is a per-pixel impervious surface area.
Center methodology included geolocating each of the MBSS sample points within the stream network, defining a catchment area representative of each sampling point, and overlaying these areas with the land cover maps. Inverse Distance Weighting schemes were used to further define land cover contributions to stream measurements.
WHRC scientists constructed linear models that include the predictor variables; physiographic region, stream length, stream order, watershed area, percent impervious surfaces, percent crop area, percent tree cover, and percent grassland. The response variables were BIBI and number of EPT. The results of this analysis show that impervious area is the primary predictor of stream health, followed by percent tree cover. The results vary by physiographic region and catchment size. The models allow predictions (Figure 2) stream water quality based on land cover metrics for the Piedmont and Highland regions of Maryland. Watersheds in the Coastal Plain were more difficult to predict due to the challenge of identifying basin catchments in this topographically invariant area.
Figure 2. A map showing the predicted vs. the observed BIBI and EPT for selected watersheds in MD.
Montgomery County, Maryland
WHRC scientists also worked using finer scale resolution in Montgomery county. Stream health was ranked as excellent, good, fair, or poor by the Montgomery county Department of the Environment, based on a combination of the IBI scores and physical stream properties such as dissolved oxygen, pH, and temperature measured between 1996 and 2001 (Figure 3).
Figure 3. Stream health ratings and land cover variables for watersheds in Montgomery County, MD.
The suite of land cover variables were incorporated as independent predictor variables, as were landscape configuration metrics such as mean distance from impervious areas to the stream channel along a topographically defined flow path, and lumpiness and contagion indices, which define the dispersion or aggregation of land cover within the watershed.
WHRC results, based on statistical models, demonstrated that the primary indicator of stream health was, once again, the amount of impervious surface within a watershed, followed by the amount of tree cover within the stream buffer zone (30m either side of the stream channel). These observations, summarized in Table 1 and Figure 4, support anecdotal evidence that reducing impervious cover in new residential and commercial development, or reducing the impacts of impervious areas through mitigation measures such as retention ponds, is beneficial to stream water quality and associated biotic health. The results also indicate that despite the importance of tree cover in the stream buffer zone, the overall proportion of impervious cover throughout the watershed was the overriding factor in predicting the health of streams within small watersheds.
Table 1: Small watershed sample size and average statistics by stream health rating category.
| Stream health rating | Number of data points | Area (km2) | Impervious (%) | Tree cover (%) | Buffered (%) |
| Excellent | 38 | 272 | 3.6 | 50.6 | 76.8 |
| Good | 81 | 658 | 4.9 | 44.6 | 71.3 |
| Fair | 76 | 451 | 13.9 | 37.0 | 63.2 |
| Poor | 50 | 356 | 19.5 | 29.6 | 56.3 |
Figure 4: Stream health ratings and average statistics.
Based on these fine scale results, guidelines for achieving a rating of excellent stream health would be to restrict watershed impervious surfaces to no more than 6% of the total area, and ensure that at least 65% of the riparian buffer zones were occupied by vegetation, in this case tree cover. To achieve an overall rating of good watershed health required no more than 10% impervious area, and at least 60% buffer zone vegetation cover. Using these criteria to assess watersheds across the state of Maryland (Figure 2) those that should be targeted for protection (excellent and good rankings) versus those that require restoration (fair or poor rankings) can be identified.
Management practices such as storm water retention ponds and/or stream restoration efforts may influence these relationships. Other factors, such as point source pollution (like sewage treatment plants) not easily identified from imagery, could dominate disturbance within the stream. These relatioships with stream reach scale measurements, where they exist, are being explored and the efficacy of increasingly relevant policies that fall under the umbrella of Smart Growth, Low Impact Development, Green Infrastructure, and Best Management Practices is being assessed.
Links to relevant publications:
