The Effect of Model Uncertainty on Water Quality Trading in the Non-Tidal Passaic River Basin (New Jersey)EPA Grant Number: F07A10316
Title: The Effect of Model Uncertainty on Water Quality Trading in the Non-Tidal Passaic River Basin (New Jersey)
Investigators: Kardos, Josef
Institution: State University of New Jersey
EPA Project Officer: Michaud, Jayne
Project Period: September 1, 2007 through December 20, 2009
RFA: STAR Graduate Fellowships (2007) RFA Text | Recipients Lists
Research Category: Economics and Decision Sciences , Academic Fellowships , Fellowship - Environmental Engineering , Fellowship - Model Uncertainty , Fellowship - Water Quality Trading
The objective of this research is to analyze the effect of water quality model uncertainty on the phosphorus Total Maximum Daily Load (TMDL) and trading program in the Non-Tidal Passaic River Basin of New Jersey. The following hypotheses will be tested by model uncertainty analysis: a) The TMDL for total phosphorus will result in attainment of the site-specific chlorophyll-a criteria at Dundee Lake with less than 10% exceedance probability; b) The TMDL will result in attainment of the targeted 70% reduction in total phosphorus loads diverted to the Wanaque Reservoir from the Wanaque South pump intake, with less than 10% exceedance probability; and c) Trades will protect water quality (in terms of total phosphorus, chlorophyll-a and dissolved oxygen) at each of five potential hot spots (e.g., localized areas vulnerable to degraded water quality due to excessive phosphorus) in the watershed, with less than 10% expected exceedance.
The study will analyze the uncertainty of the water quality model used to develop the Passaic TMDL and trade scenario outcomes. The study will apply Latin Hypercube Sampling with restricted pairing to account for parameter covariance and provide a robust and efficient output. Selected kinetic parameters and boundary conditions are the focus of the proposed uncertainty analysis, which will be executed in three stages. In the first stage, the performance of the uncertainty analysis will be tested with the binomial model against observed data. In this way, the credibility of the uncertainty analysis and the validity of its assumptions will be assessed before applying it to alternate scenarios. In the second stage, the uncertainty analysis developed in the previous stage will be applied to the TMDL scenario. In the third stage, the uncertainty yielded by different trading ratios across a range of trade scenarios will be examined.
The uncertainty analysis will demonstrate the likelihood that the TMDL and potential trades will protect water quality at each of the five most vulnerable locations in the watershed. Potential impacts could lead to reevaluation of the TMDL margin of safety or restriction of certain trades to reflect model uncertainty; conversely, potential impacts could lead to greater credibility in the TMDL margin of safety and greater assurance that proposed trades will protect water quality.