Radiocarbon (¹⁴C) measurements of atmospheric CO₂ and CH₄ enable distinction between fossil and modern biogenic sources, providing critical constraints for emission inventories and mitigation efforts. We present multi-year, multi-site atmospheric ¹⁴CO₂ and ¹⁴CH₄ data from a monitoring network in Switzerland (2019–2023), combining fortnightly sampling at five boundary layer sites (4 rural and 1 urban) on the Swiss Plateau and the alpine background station Jungfraujoch with FLEXPART-COSMO atmospheric transport simulations based on Swiss and European emission inventories. Fossil CO₂ estimates showed expected spatial patterns, with highest values at the urban site and reasonable agreement with inventory-based simulations. Nuclear Δ(¹⁴C, CO₂) contamination was negligible (+0.2 ± 0.4 ‰) and recent observations of Swiss ecosystem respiration Δ(¹⁴C, CO₂) signature indicated that the respiration correction was also quite small. In contrast, ¹⁴CH₄ measurements revealed pervasive nuclear contamination that severely compromised fossil CH₄ source apportionment. Sporadic nuclear ¹⁴CH₄ releases could not be adequately resolved with monthly/quarterly emission data from nuclear power plant operators, leading to large model-observation mismatches and physically implausible fossil CH₄ estimates. This demonstrates a fundamental limitation of ¹⁴CH₄ for quantitative source attribution in nuclear-influenced regions. The contrasting success of ¹⁴CO₂ versus ¹⁴CH₄ underscores the need for tracer-specific strategies and highlights challenges posed by expanding nuclear infrastructure globally.