Criteria for Levels of Reservoir Nutrient Degradation
An Executive Summary of the report, "Review of the Retrospective Analysis done by the Texas Commission on Environmental Quality for U.S. Environmental Protection Agency Region 6 Regional Technical Advisory Group (RTAG) on January 18, 2006 And A Subsequent Reanalysis Based on Recalculating the Criteria", follows. The report is presented in full in PDF format. If you use assistive technology and the format interferes with your ability to access the information, please contact for assistance Coastal Fisheries staff.
The Environmental Protection Agency (EPA) has tasked the states with developing numeric criteria for nutrients in surface water. Both the Texas Commission on Environmental Quality (TCEQ) and Texas Parks and Wildlife Department (TPWD) have presented proposals for calculating reservoir nutrient criteria based on anti-degradation approaches using different methodologies.
At the annual meeting of the EPA Region 6 Regional Technical Advisory (RTAG) meeting on January 18, 2006, TCEQ presented a comparison between the TCEQ and TPWD approaches. TCEQ conducted a retrospective1 analysis to determine how well nutrient criteria based on the TPWD and TCEQ methods would work using historical data. Because we found there were several minor errors in the TCEQ analysis, we conducted our own retrospective analysis to review their results.
As currently constructed, nutrient criteria based on the TPWD method tended to indicate more reservoirs would have been declared degraded than would nutrient criteria based on the TCEQ method. Primarily, this was because the TCEQ method utilizes the entire historic time series to construct its criterion, whereas the TPWD method uses only the last 10 years. Because many of the reservoirs used in this simulation have experienced changes in their nutrient levels during their historical record, the TCEQ method set criteria that did not accurately reflect the most recent water quality data.
A second reason the TPWD method tended to indicate more reservoirs would have been declared degraded than would the TCEQ method is that the TPWD method uses the 90th percentile of the actual data as the criterion, whereas the TCEQ method uses the 99th percentile of the mean of a subset of the data (outliers have been removed).
When both methods used the last 10 years of data and the 90th percentile, very similar results were obtained for both TPWD and TCEQ methods. However, differences still exist because the TPWD method uses the actual data to set its criterion, whereas the TCEQ method estimates the criterion based on an assumed theoretical distribution for the data.
Both methods can be refined to detect levels of nutrient degradation deemed appropriate. However, the use of a retrospective approach is a poor approach to use for that refinement. A better approach for refining either method would be to use simulated time series that have been strategically altered to test each method's sensitivity and specificity.