23 
any criterion exceedance greater than that of the appropriate reference curve indi¬ 
cates non-attainment of that criterion and, therefore, the designated use. 
RECOMMENDATIONS FOR FUTURE REFINEMENT OF 
THE CFD-BASED ASSESSMENT METHODOLOGY 
As part of its conclusions, the STAC CFD Review Panel identified several critical 
remaining issues requiring resolution in the near future (STAC 2006). The EPA 
agrees with the recommendations for future development and advises that the Chesa¬ 
peake Bay Program partners ensure that the work is completed in a timely, 
appropriate manner. 
The following list identifies some of the critical aspects requiring further research as 
recommended by STAC (2006). See Appendix A for additional details. 
1. Effects of Sampling Density on CFD Results. The CFD is a special case of 
an unbiased estimator for a cumulative distribution function of a population. 
Like the cumulative distribution function, the CFD is a function of the mean 
and the variance of the population under assessment. The better the mean and 
variance are characterized with sample data, the more accurate the shape of the 
CFD. As the sampling density increases, the estimated CFD begins to approach 
the true CFD. If the sampling density is low, however, then sampling error 
could become important with the potential to affect the shape of the CFD and 
ultimately the accuracy of the compliance assessment. Furthermore, the poten¬ 
tial for sample size to affect the shape could create an attainment assessment 
bias if the reference curve and assessment curve are based on different 
sampling densities. Conditional simulation methods developed by the STAC 
panel show promise in resolving these issues and mitigating potential biases 
caused by sample size differences. 
2. Choice of Interpolation Method, The STAC panel’s research considered 
several interpolation methods and outlined the features of each (Table C-l in 
Appendix C). These features illustrated tradeoffs between ease of implementa¬ 
tion and maximizing information garnered from the data. Further work is 
needed to compare the features to the requirements of wide-scale implementa¬ 
tion of Chesapeake Bay criteria assessment procedures and to formulate a plan 
for tractable implementation that results in credible assessments. One strategy 
is to implement easily performed analysis (e.g., IDW) as a screening tool to 
identify cases for which attainment/non-attainment is clear, and then imple¬ 
ment more labor-intensive methods (e.g., kriging) for cases in which 
attainment is more difficult to resolve. 
3. Three-Dimensional Interpolation. Assessments of the dissolved oxygen 
criteria attainment requires three-dimensional interpolation. The field of three- 
dimensional interpolation, however, is not as highly developed as that of 
two-dimensional interpolation. Efforts are needed to evaluate research in three- 
dimensional interpolation further and to seek additional outside scientific input 
chapter ii 
Refinements to the Chesapeake Bay Water Quality Criteria Assessment Methodology 
