Slow growth is what makes an ancient grassland durable, and it is also what loses the race after a plow. The perennials that characterise unplowed sites are slow growers, and Ashish Nerlekar, the study's lead author, says that is precisely what stops them competing in post-destruction grasslands [9]. In the pine savannas of the southeastern United States, wiregrass marks the old stands, while the regrowing patches are distinguished by fast-reproducing annuals such as white goosefoot [11]. That is not a queue the original residents are standing in. It is a filter they fail.
It is worth being exact about what the paper measured. The team analysed 742 plant species across six continents, comparing never-plowed grassland with secondary patches at different stages of regrowth [3][4], and what it reports is species composition and growth traits [1]. The often-quoted figure that grasslands hold about a third of Earth's terrestrial carbon is background about grasslands in general [6], not a measured old-growth-minus-secondary gap in this dataset [1]. So nobody gets a discount coefficient out of this paper for a restored-grassland carbon ledger. What they get is the loss of the equivalence assumption: the extensive root systems credited with storing that carbon and slowing runoff belong to the long-lived, tough-leaved perennials [6][8] that the study finds do not come back [9].
The supply arithmetic runs one way. India's savannas fell from 100 million acres in 1880 to 60 million in 2010, according to one study cited in the report [14], a 40 percent loss averaging roughly 308,000 acres a year [2]. Less than half of North America's historical grassland acreage remains [15], and Brazil's Cerrado loses an area the size of London every three months [13]. Against that, millions of acres of farmland are abandoned worldwide, and secondary grassland springs up in their place [16]. Old-growth stock falls, regrowth stock rises, and the two are being counted as the same asset class while the trait evidence says the second is a different community.
There is a density point buried in the figures as well. Grasslands cover nearly a quarter of Earth's land and hold a third of its terrestrial carbon [6][7], so per hectare they carry more than an average share [3]. That makes the plowing decision expensive in a way that a treeless horizon does not advertise, and it makes the recovery timescale the operative number. Nerlekar's earlier work found secondary grasslands taking decades to approach old-growth biodiversity, with some species still absent after centuries of regrowth [12]. Restoration projects do not report on centuries. They report on cover, and cover is the one thing that does come back.
