5 
Developing a methodology for assessing criteria attainment using these data was 
also critical. Ideally the criteria assessment methodology would prove useful in 
several ways: I) it could be applied consistently for many w r ater quality criteria 
components; 2) it would provide a common framework for assessing data collected 
over multiple scales; 3) it would provide a basis for using as much of the informa¬ 
tion contained in the collected data as possible; 4) it would provide a clear basis for 
making decisions on criteria attainment; and 5) it would provide diagnostic infor¬ 
mation regarding the spatial and temporal patterns of criteria violations. The 
cumulative frequency diagram (CFD) approach, described in the original 2003 
Chesapeake Bay water quality criteria document, was designed with many of these 
objectives in mind (U.S. EPA 2003a). 
OVERVIEW OF THE CFD ASSESSMENT METHODOLOGY 
The original 2003 Chesapeake Bay water-quality criteria document fully describes 
the CFD methodology (Chapter 6, pages 154-178), but is summarized briefly here 
(U.S. EPA 2003a). Criteria assessment using the CFD methodology is based on 
interpolation within a spatially defined grid. Described later in this chapter, this grid- 
based interpolation provides the spatial framework for use of all of the data. It 
weights each data location according to the amount of area (or volume) it represents. 
Water quality parameter levels in all interpolator grid cells are estimated based on 
interpolation algorithms, providing a complete “map” of water quality throughout 
the assessed area (Figure II-1). Water quality parameter levels in each grid cell are 
compared to the applicable criteria levels to establish an estimate of the spatial extent 
of criteria exceedance (non-attainment). Aggregating the total amount of space (area 
or volume) in which the criteria are exceeded provides a basis for estimating the 
percentage of the spatial assessment unit (designated use within a segment) in which 
the criteria were exceeded for that monitoring cruise. These measures of criteria 
exceedance are then compiled over the entire assessment period to develop a cumu¬ 
lative frequency diagram, or CFD. The CFD is a well-known and well-established 
statistical procedure commonly used to describe hydrologic and environmental data 
(Helsel and Hirsch 1992). 
The CFD assessment methodology evolved from the need to allow for variability in 
water quality parameters due to unusual events. For the water quality parameter to 
be assessed, a criterion threshold is established; when the threshold is exceeded, the 
system is considered impaired. All water quality parameters, however, are inherently 
variable in space and time. Because of this variability, it is unlikely that even a 
healthy Chesapeake Bay ecosystem will attain the threshold absolutely in all places 
and at all times. 
Spatially, small regions may persistently exceed the criteria’s threshold due to poor 
flushing or other natural conditions. Such areas should not automatically lead to the 
assumption that the entire assessment unit is impaired. Similar logic applies in the 
chapter ii • 
Refinements to the Chesapeake Bay Water Quality Criteria Assessment Methodology 
