Application of a probabilistic ammonia model: Identification of important model inputs and critique of a TMDL analysis for an urban lake

Abstract

Modeling analyses are conducted with a probabilistic mass balance ammonia model to demonstrate the important role that specifications of model inputs and toxicity standards by regulators can play in determining the assimilative capacity of polluted Onondaga Lake, NY, and to support a critical review of a related Total Maximum Daily Load (TMDL) analysis. More than 90% of the ammonia received by the lake is from a municipal wastewater treatment plant (Metro). It was found that various decisions to be made by regulators in management applications of the model greatly influence the lake's apparent assimilative capacity (TMDL). In particular, the following issues were critical: (1) anticipated hypolimnetic oxygenation treatment, (2) previously documented in-lake nitrification events, (3) effects of residual industrial pollution on pH, (4) effects of uncertainties and potential bias in pH measurements, (5) anticipated increases in population growth served by Metro, and (6) revisions in national guidance toxicity criteria prepared by the U. S. Environmental Protection Agency. Several limitations in the existing ammonia TMDL analysis are reported, including: (1) arbitrary specification of critical conditions, (2) omission of the important features of duration and allowable number of occurrences in the state standard, (3) identification of the wrong month as critical for determining the lake's assimilative capacity, (4) lack of recognition of the artificial assimilative capacity associated with the effects of residual industrial pollution, (5) the specified “margin of safety” was too low, given the level of modeling and input uncertainties, (6) anticipated increases in discharge from Metro were not considered, and (7) incorrect identification of the critical year for tributary hydrology. Recommendations are made to upgrade the ammonia TMDL analysis, including the supporting model framework and data sets. © 2003 Taylor & Francis Group, LLC.