At Huntington, nowcasts were provided to the public for 399 days during the recreational season of 2006-09.

• The nowcast provided a correct response 83.7 percent of time.
• The sensitivity of the nowcast was 58.5 percent. That means that the nowcast correctly predicted that the standard was exceeded on 38 out of 65 days that the standard was actually exceeded.
• The specificity of the nowcast was 88.6 percent. That means that the nowcast correctly predicted that the standard was NOT exceeded on 296 out of 334 days that the standard was actually NOT exceeded.

The Huntington nowcast provided more accurate information and better estimates of public health risk than the use of the previous day’s E. coli (the current method used by most beach managers.), especially in regard to sensitivity.

• The previous day’s E. coli provided a correct response 75.8 percent of time.
• The sensitivity of using the previous days E. coli was 25.8 percent. That means that the current method correctly predicted that the standard was exceeded on only 16 out of 62 days that the standard was actually exceeded.
• The specificity of using the previous days E. coli was 85.3 percent, slightly lower than using the nowcast.

At Edgewater, nowcasts were provided to the public for 235 days during the recreational season of 2008-09.

• The nowcast provided a correct response 69.4 percent of time.
• The sensitivity of the nowcast was 46.5 percent. That means that the nowcast correctly predicted that the standard was exceeded on 33 out of 71 days that the standard was actually exceeded.
• The specificity of the nowcast was 79.3 percent. That means that the nowcast correctly predicted that the standard was NOT exceeded on 130 out of 164 days that the standard was actually NOT exceeded.

Although the Edgewater nowcast provided more accurate information and better estimates of public health risk than the use of the previous day’s E. coli (the current method used by most beach managers.), more work needs to be done to improve the accuracy of the Edgewater nowcast.

• Many of the false negatives in the nowcast were due to significantly higher lake levels in 2008 than those found in the data used to develop the models (2004−07). Because of the negative relation between E. coli concentrations and lake levels for 2004−07, the model was underpredicting the probability that the standard was exceeded during 2008. To find out more about lake levels, see the full report.
• The previous day’s E. coli provided a correct response 60.2 percent of time.
• The sensitivity of using the previous days E. coli was 34.3 percent. That means that the current method correctly predicted that the standard was exceeded on only 24 out of 70 days that the standard was actually exceeded.
• The specificity of using the previous days E. coli was 71.4 percent, lower than using the nowcast.
• Work will continue in 2010 to measure additional variables that may be used in the Edgewater nowcast. These include foreshore head, photosynthetically-active radiation, and measures of sand scouring.