Interesting study. When I look at the numbers, though, it seems to me that this is sort of the natural baseline of the biosphere under which conditions humans have evolved. Rather like natural background radiation. It's the concentrated man-made sources that pose a much greater hazard to health.
A couple of interesting quotes from the paper:
In the extant literature, we found reports of long-lived organic free radicals in other plant structures formed at certain conditions: in the woody tissue of a sycamore tree when infected by a fungus, (13) in coffee beans when roasted, (14,15) in seeds and pollen during artificial aging (16) and dehydration, (17) and in wheat leaf tissue when exposed to high ozone levels.
... The corrected (see Materials and Methods) calculated concentration was ∼1015–1016, which is 100× less than the [EPFR] in ambient PM2.5. The [PFR] in leaves was orders of magnitude lower than in other matrices such as those in PM2.5 and tobacco smoke (∼1016–1018 spins/g), (4−6) biochar (∼1018 spins/g), (20,21) wildfire charcoal (∼1019–1020 spins/g), (22) and PM2.5 from extreme haze event (∼1020 spins/g). (23) Also, the [PFR] values from all samples stored as dried powder for 1 year remained nearly the same.