Figure 1.
Airborne mass and bacterial genome concentrations.
Box and whisker plots of (A) total particle mass and (B) bacterial genome copy number (GCN) measured in indoor air, ventilation duct supply air, and outside air during occupied and vacant periods. The box frames the upper quartile and lower quartile, the line represents the median, and whiskers denote range.
Table 1.
Airborne particulate matter, filter dust, and floor dust samples acquired and analyzed in this study.
Figure 2.
The influence of floor dust resuspension and particle shedding on particle number concentrations of varying optical diameter.
Plotted are the ratio of occupied indoor to simultaneous outdoor particle number concentrations for five size ranges from 0.3 µm to 10 µm under the following three conditions. Black bars represent the case of 30 people sitting on a carpeted floor that is covered with plastic sheeting (to prevent resuspension of floor dust). White bars represent one person walking on a carpeted floor covered with plastic sheeting. Gray bars represent one person walking on a carpeted floor (without plastic sheeting). Error bars indicate one standard error of the mean for replicate experiments. The experiment in which 30 people were sitting on a carpeted floor covered with plastic sheeting was conducted only once.
Figure 3.
Enrichment of bacteria in airborne particulate matter and floor dust.
Bacterial mass percentage (100×bacterial mass divided by total particle mass) in indoor air, outdoor air, and duct supply air samples and in the PM2.5 and PM10 size fraction of resuspended floor dust samples. Mass fractions were estimated assuming an average mass of 655 fg per bacterium [25]. Box and whisker plots have the same interpretation as in Figure 1.
Figure 4.
Relative abundances of bacteria in the indoor air, ventilation duct air, floor dust, and HVAC filter dust samples.
Relative abundances of the 20 most common bacterial taxa in indoor air, ventilation duct air, HVAC filter dust, and floor dust. Indoor and ventilation duct air include PM10 samples from indoor air when the room was occupied. Floor dust samples were sieved PM37 floor dust and resuspended PM10 floor dust taken after occupancy. HVAC filter dust represents samples from the filter of the building HVAC system that handled a variable mixture of outdoor air and indoor return air. Taxa are classified to the highest taxonomic level to which they could be confidently assigned. Error bars represent one standard error of the mean for nine indoor air PM10 samples, four floor dust samples, and three HVAC duct samples. Groups shown represent 55% of floor dust, 83% of HVAC filter dust, 51% of indoor air taxa, and 46% of ventilation duct air taxa.
Figure 5.
Comparison of indoor bacterial populations.
(A) Weighted UniFrac-based bacterial diversity principal coordinate analysis of indoor air (yellow outlined squares), ventilation duct supply air (orange outlined squares), floor dust (outlined circles) samples, and HVAC filter dust samples (outlined triangles) from this study. (B) UniFrac-based bacterial diversity principal coordinate analysis displaying the two coordinates that explain most of the variation between samples from this study (open squares, circles, and triangles) and the bacterial ecology of human skin samples (filled diamonds) from Costello et al. [31], outdoor air samples (filled triangles) from Bowers et al. [33], and floor dust samples from Täubel et al. [20] and Rintala et al. [27] (filled circles).