Do primary sources confirm the figures for the post-1963 decline in atmospheric background radiation that make low-background steel increasingly unnecessary?

claim-low-background-steel-need-is-fading-as-atmospheric-fallout-decays carries specific quantitative claims sourced only to Tier-4 Wikipedia:

• Anthropogenic background peaked at roughly 4% above natural levels.
• It began falling in 1963, following the Partial Nuclear Test Ban Treaty.
• By 2008 it had decreased to only ~0.005 mSv/yr above natural levels.
• Consequently special low-background steel is "no longer necessary for most radiation-sensitive uses."

Numbers are exactly the kind of claim the sourcing floor says must rest on Tier 1–2. Each figure almost certainly has a primary home in radiological-survey data.

What would answer it.

• UNSCEAR reports (UN Scientific Committee on the Effects of Atomic Radiation) on fallout dose trends since atmospheric testing — the canonical primary for global average dose figures.
• National dose-reconstruction datasets or a health-physics review tracing the post-1963 decline curve.
• A detector-programme or steel-industry primary stating that modern steel now meets low-background specifications without pre-atomic salvage — the load-bearing "no longer necessary" claim.

The interesting structural point (a lost capability quietly un-losing itself as the contaminant decays) survives even if the exact figures move, but the note stays seedling with its [unverified-quant] flag until the numbers are grounded in a primary.