Neubauer G, Cecil S, Giczi W, Petric B, Preiner P, Fröhlich J, Röösli M. The association between exposure determined by radiofrequency personal exposimeters and human exposure: A simulation study. Bioelectromagnetics. Ahead of print. Jun 1, 2010.
We have seen in recent year’s higher environmental exposure to radiofrequency fields coming from wireless telecommunication devices such as mobile phones but also other new technologies (Wifi). The public concern of RF exposure and adverse health effects is growing and the problems with exposure assessment methodology for day-to-day measurements of RF emitted from different sources needs to be addressed. Exposure patterns in the general population have changed and it can be assumed that future exposure patterns will change resulting in a less dominant exposure from mobile phones and an increasing exposure from other sources.
The objective of the study is to systematically investigate the relationship between real exposure by sources not operated on, or in close proximity to the body such as mobile phones and exposimeter measurements. Numerical simulations of various scenarios (indoor and outdoor environments) are carried out.
The visible human phantom is used as an anatomical phantom in the simulation. The relation between the field levels at the location of the exposimeter near the anatomical phantom, and the field levels averaged over the volume occupied by the phantom, without the phantom present, is calculated for different exposure conditions (indoor and outdoor) and frequency bands (100MHz (FM), 946MHz (GSM 900), and 2140MHz (UMTS)), using the finite-difference time-domain (FDTD) method in combination with a ray-tracing method. The readings of two exposimeters were calculated for all 10 positions on the phantom using the French system, Satimo EME SPY 120 and the German Maschek ESM 140. The applied anatomical body model consisted of 113 different tissues. To examine the relationship between whole-body exposure and exposimeter reading, two different scenarios were simulated by the authors. In the case of GSM, UMTS, and frequency modulation (FM), an urban outdoor environmental scenario was set-up, and for WLAN exposure, an indoor environmental scenario was used.
The results of the simulations between the exposimeter and phantom are provided for all 4 studied frequencies. At the frequency of 946MHz (GSM), results suggest a trend of underestimation of exposure by the exposimeter (reference exposure obtained by averaging the field strengths over the volume of the body). The underestimation is between 0.47 and 0.97 and varies depending on the location of the exposimeter on the phantom. For all locations other than the hips, the level of underestimation at the exposimeter location is the lowest at FM (100MHz).
On average, exposimeter quotients are higher at the location of the arms (0.93) compared to the hips (0.8) and back (0.78). Results indicate that between 56% and 81% of the field levels at the exposimeter location are below 1 (60% at 100 MHz, 70% at 946 MHz, 56% at 2140 MHz, and 81% at 2450 MHz). A good indication of an underestimation by exposimeters was found in this study.
Interpretation and Limitations
The average field levels at the location of the exposimeter were found to be mostly lower than the electric field strengths at the position of the phantom (true exposure) although some overestimation was found in some cases. Underestimation of exposure by body-worn exposimeters has also been reported by other authors. The authors asked if correction factors should be applied to adjust exposimeter readings to obtain a better estimate of personal exposure. Studies have several limitations such as all studies are restricted to very few exposure scenarios and it’s unsure how these scenarios are representative of typical exposure of the general population. Also, other morphologies and anatomies of the exposed phantoms or individuals can lead to other exposure conditions.
It was concluded that the simulation study suggested that personal exposimeter measurements can underestimate true exposure because of the impact of the human body on instrument’s reading. This underestimation should be accounted for in future exposure assessment studies when assessing exposure by body-worn devices such as personal exposimeters. This study indicates that a thorough evaluation of correction factors for different scenarios is critical prior to the definition of a study protocol.