ESCANABA — The amount of DNA from the virus responsible for COVID-19 that is detectable in Escanaba’s sewer water has risen, ending a downward trend that has lasted since early January.

According to the Sentinel Wastewater Epidemiology Evaluation Project (SWEEP), a state-run program that monitors select sewersheds for viral DNA, the two most recent samples from the Escanaba Wastewater Treatment Plant have had contained more viral DNA than more than 30% of all samples submitted to the program from the site. Specifically, the samples dated Feb. 12 and Feb. 14 contained 30.31% and 30.71% more DNA than all other samples collected at the site, respectively.

Historically, the increase in viral DNA has corresponded with an increase in cases of the virus. However, the data for confirmed infections reported by SWEEP is provided by the Michigan Department of Health and Human Services and hinges on infected individuals being tested in specific environments, like hospitals or nursing homes. Home testing, which is the predominant way COVID-19 infections are now identified, is not included in the data.

MDHHS also does not report infection rates if the number of infected people would not equate to 10 or more people if the population of the area were 100,000. For Escanaba, that means more than two people must have tested positive in these specific settings. According to the SWEEP data, that threshold has either not been met or — like the wastewater data itself — there is a delay in reporting, as no case information is currently available.

Trend data from SWEEP was not reporting the increase in cases as of Thursday afternoon, instead showing the data from the last 15-day trend period. However, the sample collected immediately prior to the Feb. 12 sample, dated Feb. 7, contained only 18.5% more DNA than all other samples.

The most recent samples also signify a shift in the way data is reported by SWEEP. Until recently, the project only reported the percentage comparing individual samples in whole numbers. Now, these percentages are available down to hundredths of a percent, allowing for a more fine-tuned comparison of historical sample data.

While the change does allow any specific analysis to be more precise, it does mean that historical values may shift more dramatically based on current viral loads. For example, if a significant and prolonged spike in viral activity hits the area, many older samples may be reclassified as having less viral load because the average baseline of viral activity has shifted. This was the case with the old reporting system as well, but more precise reporting means these swings will likely be more obvious to anyone diving into the data themselves.