Fig. 2

The forest plots showed the 95% confidence intervals (CI) of six datasets of tidal volume (TV), obtained from short sections comprising several breaths from the initial, middle, or late periods of the recording. The plots depicted data from all sites (left) and 4 individual sites (right). At the top of each forest plot, the conventional data averaging breaths over a 5-minute period (TV-All) was displayed. The vertical line and gray area represented the median and the 95% CI range of TV-All, respectively. (Abbreviations: Initial20, Mid20, and Late20 represent data calculated from 20 breaths of the initial, middle, and late period of the recording; Initial5, Mid5, and Late5 represent data from 5 breaths of each period; Initial30s, Mid30s, and Late30s represent data from 30 s of recording of the initial, middle, and late period; Initial10s, Mid10s, and Late10s represent data from the average of 10 s recording of each period)