Because the monitoring data necessary for retrospective exposure assessment are lacking in most cases, assessors typically must rely on less exact methods to estimate historical exposure, such as interviews, questionnaires, documentation of occupational histories, centralized monitoring data for air and water pollutants, or construction of exposure scenarios (NRC 1991; Sexton et al. 1995b, 2004a).
The latter is important because using the monitoring data alone resulted in too much predictive uncertainty.
Because the highly unbalanced structure of the monitoring data means that LTAs at home sites are computed from a handful of observations of a few weeks' duration, whereas LTAs at fixed sites are computed from long time series, we performed cross-validation separately for each site type.
O3 has a longer length scale (8 km) because it is a regional pollutant and because estimations are derived from spatial interpolation of the monitoring data.
However, the utility of SDWIS/FED for environmental public health surveillance is limited because actual monitoring data are available at the federal level only when results exceed the maximum contaminant levels, and the consistency of data elements over space and time can vary (Niskar 2003).
The estimation of PM2.5 was similar; however, because EPA AQS monitoring data for PM2.5 was not available prior to 1999, separate models were created for pre- and post-1999 PM2.5. post-1999 PM2.5
We also tried to calculate the index but unfortunately we found the results were imprecise because our obtained monitoring data of air concentration of Mn in various workplaces were not enough, and not all of the working places were included in monitoring spots.
Because routine monitoring data were available only for NO2, and PM10 in all study areas, back-extrapolation was limited to these pollutants.
Because monitoring data on PM2.5 are limited before 1999, PM2.5 in the period before 1999 was modeled using data on PM10 (Yanosky et al. 2008).
Because monitoring data on PM2.5 are limited before 1999, PM2.5 in the period before 1999 was modeled using data on PM10 and visibility data at airports (Yanosky et al. 2009, 2014).
However, because monitoring data had shown that known carcinogens were present in the atmosphere, with a support of the Region of the Marches, a further study was initiated to investigate the risk of death from leukemia, non-Hodgkin lymphoma, and other HMs [ 23– 25] (an incidence study was difficult if not impossible as the area was not covered by a cancer registry).
