Addressing a Critical Gap in the Development of Sediment TMDLs
Erie County, NY

Grantee: SUNY College at Buffalo
Basin Program Funds: $29,873
Non-federal Funds: $13,970
Project Duration: 07/2003 - 08/2005
Status: complete

Problem Statement
There are no protocols available for models that link the water quality with the biological condition of the waters. It is virtually impossible to track how the reduction in pollution is leading to the recovery of biological conditions and attainment of the "designated use" for a water body. An example of such a model could be the relationship between sediment concentrations and the benthic macroinvetebrate assemblage in the stream.

There is a critical gap in the Total maximum Daily Load (TMDL) process which was highlighted in the recent National Research Council report of 2001. The report recommended that - "EPA should promote the development of models that can effectively link environmental stressors to biological response". The report also emphasized that simple models based on process understanding and observed data should be preferred over complex mechanistic models with numerous parameters. The report further recommended that models be developed for "benchmark" watersheds so that they could be extrapolated to similar watersheds in the same eco-region. Although the National Research Council (NRC) recommended development of simple models, it also underscored the need for rigorous statistical analysis when developing relationships between stressors and the biological response. The report states that "all chemical and biological criteria should be defined in terms of magnitude, frequency, and duration."

Our interest in this project was to address these critical needs associated with the linkage of stressors and biological conditions. We are currently in the process of developing a pilot sediment-TMDL plan for a 135 square mile subwatershed (Cazenovia Creek) of the Buffalo River basin. In addition to sediment concentrations, we conducted weekly measurements of water quality parameters such as turbidity, dissolved oxygen, temperature, and pH at six sites in the watershed. We were also able to perform a preliminary survey of the benthic community at these six sites for three months (July - September) this year. This data highlighted that it was not only important to know the magnitude of the stressor, but also its frequency and duration. Although these results are promising, the data was limited and did not allow for development of significant relationships that could be used for TMDL analysis.

This study assessed the use of water quality - benthic habitat relationships in the development of the sediment-TMDL plan for the Buffalo River watershed. Specific objectives included:

  • Identify and develop relationships between sediment concentrations and other stressors and the benthic habitat indices.
  • Evaluate the impact of frequency and duration of stressors on benthic habitat indices. Develop models for stressor concentration-duration and frequency of exceedances.
  • Determine the "threshold" stressor concentrations that correspond to "desirable" or "undisturbed" benthic habitat conditions.
  • Use the stressor-benthos models to determine the sediment reductions required for "impaired" stream reaches to attain "desirable" benthic quality status. Determine if the targeted reductions are feasible given the choice of available management practices (this will determined via "scenario analysis" simulations in the calibrated BASINS-SWAT model).

Benthic metrics and physiochemical water quality parameters significantly (alpha = 0.05) indicated that the urban/residential site (C1A) was the most impaired site. The agricultural site (C3D) was assessed as impaired on a majority of metrics under NYSDEC and USEPA biological impairment criteria but, both biological and physiochemical impairment was greatest at the urban/residential site (C1A). Of the parameters assessed, dissolved oxygen and temperature are the primary water quality variables affecting the benthic assemblage. Analysis of summer versus winter benthic metrics showed dramatic seasonal variation in water quality and benthic assemblages.

Results comparing the reference site (C3C) to the urban/residential site (C1A) suggest that changes as small as a 14% increase in mean DO levels and a 14% decrease in mean temperature levels at the urban/residential site would allow this site to improve its assessment. This result suggests that these benthic metric/water quality models may be effectively utilized for TMDL development in the Cazenovia Creek watershed. (See final report below.)

Contact: Dr. Shreeram Inamdar, 716-878-6229


Great Lakes Commission des Grands Lacs.  2805 S. Industrial Highway, Suite 100.  Ann Arbor, MI  48104-6791.  phone: 734/971.9135.  fax: 734/971-9150. Join the Friends of the Great Lakes GLIN Partner