Extremely low frequency magnetic fields (ELF-MF) in Switzerland: From exposure monitoring to daily
Abstract
Extremely low frequency magnetic fields (ELF-MF) in Switzerland: From exposure monitoring to daily exposure scenarios Loizeau N, Haas D, Zahner M, Stephan C, Schindler J, Gugler M, Fröhlich J, Ziegler T, Röösli M. Extremely low frequency magnetic fields (ELF-MF) in Switzerland: From exposure monitoring to daily exposure scenarios. Environment International. 2024. doi: 10.1016/j.envint.2024.109181. Abstract Exposure to extremely low frequency magnetic fields (ELF-MF) is ubiquitous in our daily environment. This study aims to provide a comprehensive overview of the ambient ELF-MF exposure in Switzerland and presents a novel environmental exposure matrix for exposure assessment and risk communication. Magnetic flux density levels (μT) were measured using a portable exposimeter carried in a backpack for the main ELF sources: railway power (16.7 Hz), domestic power (50 Hz), and tram ripple current (300 Hz). We collected ELF-MF levels between 2022 and 2024 in various environments representative of the Swiss population: 300 outdoor areas (e.g. city centres, residential areas), 245 public spaces (e.g. train stations, schools), 348 transport journeys (e.g. train, cars), and in 59 homes (e.g. bedrooms, living rooms). Over all environments, the highest ELF-MF exposure levels were measured in train stations (median: 0.48 μT), trains (median: 0.40 μT), and in living rooms near (<200 m) highest voltage lines of 220 kV and 380 kV (median: 0.37 μT). ELF-MF median levels measured two years apart showed high Pearson correlation coefficients in the same 150 outdoor areas (r = 0.88) and 86 public spaces (r = 0.87), without any significant changes. All measurements are well below the Swiss ambient regulatory limit based on the ICNIRP 1998 guidelines (median: 0.2 %). Finally, we derived an environmental exposure matrix and modelled 27 daily time-weighted average ELF-MF exposure scenarios by combining typical time spent at home, work and transport environments. People who do not live near highest voltage lines or work in highly exposed environments are typically exposed to less than 0.3 μT on average, while those who do are likely to exceed this level. This novel environmental exposure matrix is a useful tool for public communication and agent-based exposure modelling for future epidemiological research. : sciencedirect.com
AI evidence extraction
Main findings
Portable exposimeter measurements (2022–2024) found the highest median ELF-MF levels in train stations (0.48 μT), trains (0.40 μT), and living rooms near (<200 m) 220/380 kV lines (0.37 μT). Repeated measurements two years apart in the same locations showed high correlations (outdoor areas r=0.88; public spaces r=0.87) with no significant changes. All measurements were reported as well below the Swiss ambient regulatory limit (median 0.2% of the limit), and modelled daily time-weighted average scenarios suggested typical exposures <0.3 μT unless living near highest voltage lines or working in highly exposed environments.
Outcomes measured
- Magnetic flux density (μT) levels across environments
- Temporal stability of ELF-MF levels (Pearson correlation over two years)
- Comparison to Swiss ambient regulatory limit (ICNIRP 1998-based)
- Modelled daily time-weighted average exposure scenarios
Suggested hubs
- who-icnirp (0.7)Mentions Swiss ambient regulatory limit based on ICNIRP 1998 guidelines.
- occupational-exposure (0.7)Daily exposure scenarios include work in highly exposed environments (no specific occupation detailed).
View raw extracted JSON
{
"study_type": "exposure_assessment",
"exposure": {
"band": "ELF",
"source": "environmental (railway power, domestic power, tram ripple current; proximity to high-voltage lines)",
"frequency_mhz": null,
"sar_wkg": null,
"duration": "Measurements collected between 2022 and 2024; daily time-weighted average scenarios modelled"
},
"population": "Environments representative of the Swiss population (outdoor areas, public spaces, transport journeys, homes)",
"sample_size": 952,
"outcomes": [
"Magnetic flux density (μT) levels across environments",
"Temporal stability of ELF-MF levels (Pearson correlation over two years)",
"Comparison to Swiss ambient regulatory limit (ICNIRP 1998-based)",
"Modelled daily time-weighted average exposure scenarios"
],
"main_findings": "Portable exposimeter measurements (2022–2024) found the highest median ELF-MF levels in train stations (0.48 μT), trains (0.40 μT), and living rooms near (<200 m) 220/380 kV lines (0.37 μT). Repeated measurements two years apart in the same locations showed high correlations (outdoor areas r=0.88; public spaces r=0.87) with no significant changes. All measurements were reported as well below the Swiss ambient regulatory limit (median 0.2% of the limit), and modelled daily time-weighted average scenarios suggested typical exposures <0.3 μT unless living near highest voltage lines or working in highly exposed environments.",
"effect_direction": "unclear",
"limitations": [],
"evidence_strength": "insufficient",
"confidence": 0.78,
"peer_reviewed_likely": "unknown",
"keywords": [
"extremely low frequency",
"ELF-MF",
"magnetic fields",
"exposure monitoring",
"portable exposimeter",
"Switzerland",
"railway power",
"domestic power",
"tram ripple current",
"high-voltage lines",
"exposure matrix",
"time-weighted average",
"ICNIRP 1998",
"regulatory limit"
],
"suggested_hubs": [
{
"slug": "who-icnirp",
"weight": 0.6,
"reason": "Mentions Swiss ambient regulatory limit based on ICNIRP 1998 guidelines."
},
{
"slug": "occupational-exposure",
"weight": 0.35,
"reason": "Daily exposure scenarios include work in highly exposed environments (no specific occupation detailed)."
}
]
}AI can be wrong. Always verify against the paper.
Comments
Log in to comment.
No comments yet.
NNEMF synthesis
Loizeau and colleagues did not test a disease, a cell, or a recovery response. They measured ambient extremely low frequency magnetic fields (ELF-MF) in Switzerland from 2022 to 2024 with a portable exposimeter carried in a backpack. They recorded magnetic flux density in microtesla (μT) for three sources: railway power at 16.7 hertz, domestic power at 50 hertz, and tram ripple current at 300 hertz. The sample was 300 outdoor areas, 245 public spaces, 348 transport journeys, and 59 homes. The highest environment medians were train stations at 0.48 μT, trains at 0.40 μT, and living rooms within 200 metres of 220 kilovolt and 380 kilovolt lines at 0.37 μT. Repeat measurements two years apart in the same 150 outdoor areas and 86 public spaces were stable (Pearson r = 0.88 and r = 0.87) with no significant change. From time spent at home, work, and in transport they modelled 27 daily time-weighted scenarios. People who do not live near the highest-voltage lines and do not work in highly exposed places are typically under 0.3 μT on average; people who do are likely to exceed 0.3 μT. Those figures are modelled scenarios and place medians, not personal biological doses. Dose sits next to the limit the record actually uses. Specific absorption rate (SAR), the radiofrequency energy-absorption metric, is not reported, and this synthesis does not estimate it. The 1.6 watts per kilogram phone limit is a radiofrequency SAR limit and is not a comparator this record supports. What the record does report is magnetic flux density, and it places all measurements well below the Swiss ambient regulatory limit based on the International Commission on Non-Ionizing Radiation Protection (ICNIRP) 1998 guidelines, at a median of 0.2 percent of that limit. Temperature was not measured. On the study's own comparison, these are non-heating exposures. That comparison describes the dose. It does not show that the field is biologically silent, and it does not settle the signaling question a heating-and-stimulation limit was not built to answer. [data_supported] Unit identity, not a new measurement: 0.40 μT is 4.0 milligauss, 0.48 μT is 4.8 milligauss, 0.37 μT is 3.7 milligauss, and the scenario threshold of 0.3 μT is 3.0 milligauss. The highest place medians therefore sit at or above the four-milligauss exposure benchmark. That is an exposure fact. It is not a finding that these fields cause leukemia or any other disease. The paper itself offers the matrix for later epidemiological modelling and public communication. It does not report a health outcome. [data_supported for the flux densities; insufficient_data for any disease] The framework connection is exposure-side, not a causation claim. The framework does not say this field causes a named disease. It says a non-thermal field can reduce the precision of biological regulation: the timing of calcium signals, the behavior of the voltage-sensing S4 segment of ion channels, mitochondrial energy and redox output, and, as background only, field-sensitive spin chemistry. This study supplies the field people actually live in, at non-heating levels, separated by source frequency, stable across two years, and uneven across homes, stations, and trains. It does not supply the biology. What matches a framework prediction, without having been tested: a daily field that is a mix of distinct waveforms (16.7, 50, and 300 hertz) rather than a single absorbed-energy number; a repeating exposure, stable over years, which is the pattern under which extra compensatory effort could draw down reserve; and a split between people typically under 0.3 μT and people likely over 0.3 μT, which is an exposure contrast a later study could use. [framework_prediction] None of those biological predictions was measured here. What was not shown, and must not be inferred: a change in calcium timing, S4 gating, mitochondrial reactive oxygen output, antioxidant or heat-shock compensation, slower recovery after repeated exposure, or a difference by tissue, sex, genotype, or developmental window. Separating three source frequencies shows that the exposure is not energy-only. It does not show that biology depends on waveform. Living near a high-voltage line is a difference in environmental field, not density gating. Density gating is about the receiving cell's channel machinery and reserve. This study measured neither. Schools and bedrooms were measurement sites, not developmental-window tests. [insufficient_data] A null is possible support and is not a limit on the hypothesis. This paper's only null is exposure stability: no significant change in field levels over two years. That is not a biological null, and it is not treated here as support or as a ceiling. A future null on disease, under these non-heating fields, in a low-coupling tissue, on an endpoint downstream of fidelity loss, after a short exposure, or without a recovery challenge, would be possible support for density-gated low-fidelity biology. It would not refute the hypothesis, and the present below-limit finding does not either. No biological null is being forced into support. The mechanistic neighbor in the supplied set is the voltage-gated channel account of animal magnetoreception. That paper argues a magnetic field can drive mobile ions inside voltage-gated ion channels against the voltage sensor. This paper measured the anthropogenic ELF-MF those sensors would encounter in Swiss daily life. The field regimes differ: geomagnetic field plus animal motion, versus oscillating 16.7 to 300 hertz fields at tenths of a microtesla. The sensor step is shared; the quantitative transfer was not tested. Microwave monkey electroencephalogram and electrocardiogram work, and radiofrequency brain-tumor meta-analysis, do not share a mechanism step with this exposure matrix. Removal of those two links is proposed for review and is not described as already applied. Search returned other extremely low frequency calcium and power-line records, but their full summaries were not opened before the lookup limit, so no addition is proposed. What would settle the open prediction is not another map of microtesla. It is a test, at the flux densities and source frequencies this matrix documents, of whether calcium-signal timing or S4-dependent channel behavior shifts while ordinary output is held up by extra redox or stress-protein effort, and whether that shift is larger in a high-reserve-demand window than in a low-coupling tissue.