10 
Table 11-1. Steps for constructing and assessing criteria attainment using cumulative 
frequency diagrams (CFDs). 
1. Collect data from a spatial network of locations on several dates during the 
assessment period. 
2. For each date, interpolate the data spatially over the entire system to obtain esti¬ 
mates of water quality using a two- or three-dimensional grid of interpolation 
cells. 
3. Aggregate interpolations to the appropriate temporal scale (e.g., if evaluating the 
30-day mean, take the average of all interpolations for each date in the month). 
4. For each interpolator cell, assess whether the applicable criterion is exceeded. 
5. For each assessment unit, compute the percentage of interpolator cells that exceed 
the criterion as an estimate of the percent of area (or volume) within the spatial 
assessment unit that exceeds the criterion. 
6. Rank the percent of area estimates for the set of all sample days in the assessment 
period from largest to smallest and sequentially assign to these ranked percents a 
value that estimates percent of time. Add the end points of (100%, 0%) and (0%, 
100 %). 
7. Plot the paired percent of area (or volume) and percent of time data on a graph 
with the percent of area on the x-axis and percent of time on the y-axis. The 
resultant plot is the assessment cumulative frequency diagram or CFD. 
8. Compare the assessment CFD (from step 7) to the appropriate reference CFD. If at 
any point the assessment CFD exceeds the reference CFD (i.e., a given level of 
spatial noncompliance occurs more often than allowed for a given amount of 
time), then the criterion is in non-attainment. Consequently, the segment fails to 
meet that designated use. 
An illustration of the eight steps for computing the CFD for these simplified 
constraints is shown on the facing page. The three columns show the first three steps. 
Column 1 provides fictional data for five dates for five fixed locations in a two- 
dimensional grid. Column 2 shows a fictional interpolation of these data to cover the 
entire grid. Column 3 gives the compliance status of each cell in the grid with 1 indi¬ 
cating non-attainment and 0 signifying attainment. 
In this hypothetical example, the assessment curve is clearly greater than the refer¬ 
ence curve and in non-attainment of the criterion, therefore, the designated use is not 
met. EPA recommends that any exceedance of the attainment CFD above the refer¬ 
ence CFD should be considered non-attainment of the criterion and, consequently, 
the designated use. 
chapter ii 
Refinements to the Chesapeake Bay Water Quality Criteria Assessment Methodology 
