We still have several gobo (domesticated burdock) roots in the ground. I dug two of them up during a brief thaw about a week ago to see what roots left to overwinter in the ground would be like. They were no more difficult to dig in cold earth than in warm earth, which is to say I still broke a sweat on a cold day.
Turns out I could have simply consulted the Johnny's Seed catalog to answer my questions about whether or not gobo holds in the earth through the winter. Just last night, while surrounded by six open seed catalogs, I came across this: "...roots make a great season extension offering. For fall, winter and spring harvest and storage. Burdock can be overwintered in soil much the same as parsnips." Still, it was an opportune moment to do the digging. Temperatures are barely peaking above freezing once or twice a week.
In the course of answering my question by empirical means I found that the roots have developed a single growth ring, just as any tree does each year. I've seen such rings before in carrots left too long in the ground. I've heard that biennial parsnips do the same, and that the flesh of the parsnip outside the growth ring remains edible in the second year, even though the part within the ring becomes tough and unappetizing. In gobo, the flesh is always somewhat tough, which lends it a meaty texture but also requires more cooking than for other types of root vegetable.
The roots I dug this month were extremely large; more than two inches in diameter. As ever, I was not able to dig deep enough to harvest the entire root. So there's a root fragment of unknown length left in the ground for each plant that I harvested. In wild burdock that I've dug up, the roots had hollowed out to form very large fibrous cavities. I saw just the beginnings of this trait in the Takinagawa gobo I left for winter, though not in the least in those harvested earlier. Where the cavity begins to form, the flesh discolors slightly and becomes sort of spongy. I just trimmed around these sections and composted those parts. I also found some traces of a pink coloration near the top of the root as I trimmed it. I don't know what to make of this as I've never seen any such pigmentation in roots dug earlier in the year. But we ate that root and are none the worse for wear, so it seems there's nothing to worry about.
I've read that burdock root stores only for one week in the refrigerator, but this has not been my experience. I harvest the roots, leaving the very base layers of the leaf stalks intact, only very gently cleaning off the dirt that clings to them. To do so I soak the roots in a big pot of water out on the porch to remove the soil clumps, and then shake off both water and dirt, leaving all the little feeder roots intact. Rinsing the roots is said to hasten spoilage, but with this method all seems well. Then I wrap each individual root as tightly as can be managed in plastic wrap and store it in the vegetable drawer of the refrigerator. So far I've kept them for two weeks in there and they seem fine. (My next experiment will be to store one in the root cellar and see how that does.) I can't say this with absolute certainty, but it seems to me that winter-dug gobo roots are slower to discolor when the flesh is exposed to air than are summer- or fall-dug roots. The discoloration occurs quickly when preparing the roots by scraping them with the back of the knife. (There are more details about preparing gobo on my kinpira gobo recipe page.) Whereas the parts of the root exposed by breaking them off during winter harvest seem very slow to discolor. In any case, the discoloration is purely aesthetic and not in any way dangerous. I now put a generous splash of vinegar in the soaking water during preparation and this prevents most discoloration.
So there's my empirically derived answer as to whether or not gobo will hold in the ground over the winter. They will, though the tendency of the plant to form root cavities may become more and more evident the later the root is harvested. Now the question remains as to whether or not the root fragments left in the ground will rot in place, or manage to grow enough to produce top growth in a second year. The fragments are all at least 10 inches below the surface and usually more than that. If they rot in place, that's fantastic treatment for our clay soil. If they regrow next year, there's a good chance I'll get another harvest from a single planting. In the latter case it would be interesting to see how many years of harvest could derive from the initial planting. It's also possible though that the second year's root growth might be less edible in some way. We shall see...