Interesting hypothesis. I've also run across the idea that in arid regions, nutrients were built up over geological timescales, but after one introduces irrigated crops, get depleted over agricultural timescales.
I wonder what effect dust created from fire has on soils over large periods of time.