I spent this morning in Los Angeles city parks, pulling up clover. This attracted less attention than you might expect. Angelenos are, as a group, not inclined to bother people who aren’t doing anyone else any obvious harm, and honestly the range of activity in a typical LA park includes behaviors that are objectively odder (group yoga) and more damaging to the turf (pickup soccer games). The single fellow who shouted good morning to me followed up only with “happy hunting!” He may have thought I was looking for a lucky charm.
In fact I wasn’t counting leaflets, but trying to make sure I got to twenty more-or-less distinct patches of Trifolium repens, white clover. White clover is native to Europe, but it loves lawns, and it’s been introduced all over the world. Recently Ken Thompson, Marie Renaudin, and Marc Johnson reported something interesting about white clover populations in Toronto, New York, and Boston: close to the centers of those cities, clover samples were less likely to respond to injury by producing toxic hydrogen cyanide, HCN. HCN production has a simple genetic basis, so parallel loss of the phenotype is likely an evolved response. The HCN reaction defends against herbivores — but it’s also triggered by freezing damage, so HCN-producing plants in cold climates risk poisoning themselves. Thompson et al. hypothesized that winter road plowing in built-up urban areas exposes roadside clover to sharper cold than it would cope with in the countryside, insulated under snow cover. And it turns out that in a fourth city, Montreal, where there were below-freezing days with no snow cover at the sampling sites, clover populations didn’t show the pattern of reduced HCN production.
That’s an interesting story from northeast North America, but white clover grows worldwide. The Global Urban Evolution (GLUE) Project aims build on Thompson et al.‘s original four-city dataset with replicate transects in cities on every continent. I’m one of three biologists at institutions in greater Los Angeles (it’s a big city!) that have pitched in to take samples and test them for HCN content.
Collecting my samples has taken a little more thought than I’d initially figured on. The GLUE sampling protocol envisions relatively contained, convex urban areas that shade smoothly into suburbs and then countryside, with clover available pretty much anywhere along the way. The first problem with this, for me, is that the development of greater Los Angeles hasn’t so much sprawled outward from a central built-up area as it’s poured around and occasionally over the complex of coastal mountains that surround downtown LA on three sides. The second is that white clover doesn’t grown in truly natural rural areas beyond the limits of that development — it seems the only suitable habitat for T. repens in greater LA really is a regularly watered lawn.
Not just any lawn, either! I went through a couple of draft sampling plans, picking likely looking lawns out from Google Earth aerial imagery, only to find, upon showing up in person, that they’d be too well weeded, or closely mowed, for clover. Or just too small — many a median strip of grass has spreading patches of Medicago or Oxalis where white clover might otherwise move in. A first attempted collecting trip with a couple of my students turned up patch after patch of beautiful, carpet-like grass without a single trifoliate leaf to be seen.
That is, until we got to our first city park. It seems that the LA Department of Recreation and Parks provides just the right degree of lawn care — enough water, not too much re-seeding or mowing — to make white clover happy. So I went back to Google Earth, loaded the map layer with public park locations, and stitched together a transect from downtown north and west across the San Fernando Valley. It’ll take a few more days working with the students and on my own, but by the end of it I’ll have seen a pretty large sample of the public green spaces in the city.
Thompson KA, M Renaudin, MTJ Johnson. 2016. Urbanization drives the evolution of parallel clines in plant populations. Proc. Royal Soc. B 283, 20162180. doi: 10.1098/rspb.2016.2180