Once a year we swab restored and wild kelp beds across Tasmania for the provenance program. Two hundred blades, five sites, and what the lab read in them.
Tidewright field notes are our crew's own stories from project sites. Everything in them — sites, numbers, people — is fictional, part of the Tidewright demonstration. The genetics is real.
Once a year, in the last week of winter, I trade the lab coat for a drysuit and go collecting. Genetic Swab Week is the provenance program's health check: roughly two hundred kelp tissue samples, taken from our restored beds and their wild neighbour populations across five Tasmanian sites, sequenced back at the gametophyte bank and compared against the baseline genotypes we banked before a single line was seeded. The question is blunt: is the kelp we planted still the kelp we planted — and is it breeding with the forest next door?
Kelp populations are locally adapted. A Ecklonia radiata blade from a warm, sheltered bay differs in meaningful, heritable ways from one on a cold exposed headland forty kilometres away — the whole premise of our population genetics and provenance work is that matching seedstock to site is what makes restored forests survive their first heatwave. But planting is not the end of the story. Once a restored bed reproduces, its spores mix with wild ones; wild spores settle our lines; selection starts editing whatever we planted. Without annual sampling, we'd be running an experiment with no instruments.
The practical core of the week is unglamorous. Each sample is a 2-cm disc punched from the meristem of a tagged blade — kelp tissue keeps fine for the boat ride in a labelled tube of silica gel — paired with a photo, a depth, and the plant's line-number if it's one of ours. The dive plan for the week looked like this:
Back in the lab, each disc is extracted, PCR-amplified for a panel of a few thousand SNP markers, and sequenced. The method family is the same one described in DNA barcoding seaweed, scaled up from species IDs to population fingerprints.
Three findings, in ascending order of how much they changed our plans.
Restored beds are holding their provenance. Plants grown from Boomer-local broodstock remain genetically Boomer-local: no detectable drift toward hatchery homogeneity, no bottleneck signature. Whatever selection the site applies, it's applying it to the right raw material.
Recruits are overwhelmingly local too. Of the untagged recruits sampled on our lines, about 90% matched the neighbouring wild population's genetic profile rather than the hatchery pedigree — the wild forest is seeding our lines faster than our lines seed it. We take that as a compliment. It means the lines are reading as habitat, not as a plantation — which is precisely how biofilms and settlement cues are supposed to work on clean substrate.
One warm-edge allele is creeping north — in the wild. At our southern reference bed, a marker combination associated with our warm-tolerant Strahan provenance — which we deliberately planted at two northern sites under our assisted-migration framework — turned up in four wild recruits. At 2% prevalence it is not cause for alarm, but it is cause for the next item on this list.
Genetic monitoring can sound like confidence itself — a lab, a sequencer, a number. The truth is that a SNP panel answers the questions you thought to ask. It will tell you about relatedness and provenance, not yet about performance; for that, the freezer stays married to the dive slate and the drone census. The week works because field and lab argue with each other once a year, in August, over two hundred little discs of kelp.
A 2-cm disc from a blade's growing region is lost tissue, but each sampled plant is tagged and revisited; none has failed to keep growing. Total annual take across all sites is trivial next to a single storm's pruning.
Cost and comparability. A few thousand markers across two hundred samples a year is affordable and directly comparable across years; whole genomes are for the archive tissue if a question ever demands them.
Sustained loss of local genetic structure in a restored bed, hatchery-dominated recruitment on wild reefs, or translocated markers crossing our 5% prevalence threshold in wild populations. Each has a pre-written response in the provenance protocol — decision before data, not after.
Yes, and they should. The marker panel, sampling design and archive logic are deliberately boring, standard and shareable — the point of a provenance program is that any operator can adopt it.
A Tidewright field note from the coldest corner of our hatchery: the gametophyte bank, where microscopic kelp cultures wait in the dark for the reefs that will need them.
Dr. Elias Brandt ·
How scientists read the family tree of a kelp forest — from SNP chips to seascape genomics — and what genetic maps reveal about provenance and adaptation.
Dr. Elias Brandt ·
Moving warm-tolerant kelp genotypes ahead of climate heat could buy forests decades. Here's the evidence, the ecological risks and the protocols that make it safer.
Dr. Elias Brandt ·