21 
That is, the reference curve defines the degree to which criteria violations can be 
tolerated without resulting in impairment of the designated use. 
Bias and uncertainty are driven in CFD curves by sample densities in time and space. 
Therefore, the STAC panel advised that similar sample densities be used in the deri¬ 
vation of assessment and reference curves. As such densities are not always feasible, 
additional analytical methods are needed to weight sampling densities equally 
between attainment and reference curves. 
APPLICATION OF THE CFD-BASED 
ASSESSMENT METHODOLOGY 
RECOMMENDATIONS FOR APPLICATION OF THE CFD-BASED 
METHODOLOGY 
As stated above, the CFD-based water quality criteria assessment methodology 
offers the potential for significant benefit in accurately assessing Chesapeake Bay 
water quality criteria attainment. As the STAC CFD Review Panel has indicated, 
however, that the methodology is new and additional evaluations and refinements 
should be performed (STAC 2006) (Appendix A). The EPA agrees with the panel’s 
conclusions, strongly supports the findings that further research is needed, and will 
support those efforts in whatever way possible in the coming years. In the meantime, 
the EPA recommends the following approach in undertaking Chesapeake Bay water 
quality criteria assessments. 
As described above, the Chesapeake Bay Program collects data at two different 
scales for water quality criteria attainment assessment. In each case, the design of 
data collection program focuses on assessments at a specific scale. The fixed-station 
data are designed for segment and baywide assessments and the shallow-water moni¬ 
toring data are designed to assess the small tidal tributaries and the Bay’s 
shallow-water habitats. Given the different scales, separate interpolations are likely 
necessary using the most appropriate interpolation algorithm. The STAC CFD 
Review Panel evaluated two possible options for spatial interpolation, recom¬ 
mending kriging as the better of the two alternatives (STAC 2006). Kriging, 
however, has not been fully developed for application in Chesapeake Bay water 
quality criteria attainment assessment. 
Until kriging is fully developed as an option for whole-Bay assessment based on the 
fixed-station data, the EPA recommends that spatial interpolations continue using the 
current Chesapeake Bay Program’s inverse distance weighting (IDW) algorithm- 
based interpolator (Appendix D). Spatial interpolation of the fixed-station data for 
assessment of criteria attainment in the mainstem Bay and major tidal tributaries 
requires several specific capabilities including: 1) the data must be interpolated in 
three-dimensions (i.e., with depth); 2) the data must be interpolated into the tidal 
tributaries and around bends in these tidal rivers; and 3) the interpolation needs to be 
chapter ii 
Refinements to the Chesapeake Bay Water Quality Criteria Assessment Methodology 
