A-52 
determinations in the Chesapeake Bay. Using criteria for Best Science and Best 
Available Science developed by the American Fisheries Society and the Estuarine 
Research Federation (Sullivan et al. 2006), we list relevant attributes of the CFD 
approach (Table 6.1). 
The CFD builds on important statistical theory related to the cumulative distribution 
function and as such, its statistical properties can be simulated and deduced. We have 
also shown that it is feasible to construct confidence ellipses that support inferences 
related to threshold curves or other tests of spatial and temporal compliance. Work 
remains to be done in understanding fundamental properties of how the CFD repre¬ 
sents likely covariances of attainment in time and space and how temporal and 
spatial correlations interact with sample size effects. Further, more work is needed 
in analyzing biases across regions and designated use segments. The panel expects 
that a two-three year time frame of directed research and development will be 
required to identify and measure these sources of bias and imprecision in support of 
attainment determinations. 
Through simulations of the CFD approach, it is feasible to analyze bias and error for 
both temporal and spatial sources of attainment variability. In particular, conditional 
simulations merit additional investigation as a relatively unbiased approach for 
supporting statistical comparisons among CFD curves. Much work remains to be 
done in understanding fundamental properties of how the CFD represents likely 
covariances of attainment in time and space. Still, the panel finds the approach 
feasible: one which merits additional development, testing, and application. Indeed, 
the CFD approach is beginning to attract scientific and management attention 
outside the Chesapeake Bay community. 
As shown by analyses in previous sections, the approach can efficiently combine 
spatial and temporal data to support inferences on whether regions within the Chesa¬ 
peake Bay attain or exceed water quality standards. On the other hand, we recognize 
substantial bias and imprecision can occur due to small sample size, non-independ¬ 
ence in temporal trends, and inadequate spatial interpolations. More work is needed 
in analyzing these biases across regions and designated use segments. Further, the 
old saw of needing more samples cannot be ignored. In particular, the panel is opti¬ 
mistic in the application of continuous spatial data streams made available through 
the cruise-track monitoring program, and the promise of continuous temporal data 
through further deployment of remote sensing platforms in the Chesapeake Bay 
(CBOS web site, etc). These data sets will support greater precision and accuracy in 
both threshold and attainment determinations made through the CFD approach. 
In classifying the CFD approach as best available science, we seek to make several 
important distinctions (Table 6.1). First, the CFD approach is a scientifically based 
approach based upon its clear purpose, conceptual and design framework, empirical 
procedures, documentation, and intent to develop rigorous statistical and review 
procedures (Sullivan et al. 2006, Daubert v. Merrell Dow Pharmaceuticals, Inc., 
1993). That the approach permits evaluation of uncertainty also supports its classi¬ 
fication as best available science (Christman 2006). On the other hand, we do not 
believe that the CFD approach yet constitutes best science. Here, further analyses of 
underlying statistical properties of the approach (including sampling design and 
appendix a 
The Cumulative Frequency Diagram Method for Determining Water Quality Attainment 
