Reflectance spectroscopy in the 400-2400 nm to assess settled dust in different urban environments

The aim of this study was to apply a spectral reflectance approach to account for small amounts of sediment dust in occupied homes. We examined the method’s ability to predict the gravimetric weight of sediment dust particles solely from the reflectance data (400-2400 nm). Multivariate data analysis based on Partial Least Squares (PLS) regression was run to predict the dust loads solely from reflectance data. For both experiments use of difference index (when the dust trap background spectra was subtracted from the total measured reflectance) in the PLS analyses was found to demonstrate the best pretreatment in PLS modeling. Maps of dust distribution, based both, on the reference gravimetric and spectrally predicted weight values was generated. Reasonable explanation was provided to the calculated distribution that can be further used by decision makers to improve the indoor life quality. We conclude that this methodology (simple and rapid in-situ spectral measurements with appropriate analyses), can be employed to assess dust in both indoor and outdoor environments (in small and high dust content). This information can be used for initial decision making, improving indoor conditions, and tracking dust contamination following environmental change.