40 
derivation of the 2003 published deep-water reference curves were identified using 
a measure of benthic community health—the Chesapeake Bay benthic index of 
biological integrity or benthic-IBl (Weisberg et al. 1997). Sessile benthic communi¬ 
ties are good indicators of the water quality of the overlying waters. Although 
relatively tolerant of lower oxygen concentrations, a dissolved oxygen concentration 
of 2 mg liter' 1 is the threshold below which benthic infaunal communities become 
severely stressed (numerous references cited in Chapter 3 of U.S. EPA 2003a). A 
healthy benthic community, therefore, could indicate allowable time and space 
exceedances of the dissolved oxygen criteria that will not impair the biological 
community. 
Benthic infaunal community samples are collected as part of the long-term Chesa¬ 
peake Bay Benthic Monitoring Program at fixed and random locations during the 
summer, usually in August to September. If the benthic-IBl of that sample is “good,” 
(in this case 3 or more on a scale of 1 to 5), dissolved oxygen conditions were likely 
adequate for the previous one to two months (Dauer et al. 2005). 
In order to ensure greater consistency in deriving the open-water and deep-water 
reference curves, factor in the state-adopted designated-use boundaries and take 
advantage of a full two decades on monitoring data, both reference curves were 
updated. To develop updated open-water and deep-water reference curves, the 
monthly fixed and random station locations for the benthic-IBI data from 1985 to 
2005 were matched with the monthly open-water and deep-water designated-use 
boundaries for the same time period. This updated approach differs from the original 
method published by the EPA (2003a), which used a single designated-use boundary 
coverage for the entire data record. An additional difference is that previously this 
method was used to define only the deep-water reference curve. The open-water 
reference curve was based on an analysis in which “good” water quality conditions 
were defined for reference segments by year (see Appendix H in U.S. EPA 2003a). 
Reference locations were identified by sorting the resulting data set by year, 
segment, and designated use. If a designated use in a given segment in a given year 
had only “good” benthic-IBI scores (>3), then the dissolved oxygen data for that 
segment, designated use, and summer period (June-September) can be used to 
compute a reference curve. Appendix F lists these use-segment-year combinations. 
Separate CFDs were generated for open-water and deep-water designated-use habi¬ 
tats from the entire data set of summer dissolved oxygen data from all reference 
locations over the 1985-2005 data record. Figures IV-2 and IV-3 respectively illus¬ 
trate the resultant June-September open-water and deep-water dissolved oxygen 
criteria reference curves. Appendix G documents the equations for the reference 
curves. 
chapter iv 
Refinements to the Chesapeake Bay Dissolved Oxygen Criteria Assessment Procedures 
