Dataset: Nereocystis luetkeana geographic patterns in physiology and microbiome from samples collected from 2019 to 2022 at nine sites spanning more than 200 kilometers in Washington state

These are data on kelp traits, seawater features and 16S rRNA amplicon sequencing. All kelp measurements were made in situ, often via kayak or at the shore using a measuring tape and calipers.

Kelp tissue was collected for elemental analysis by excising a 2 x 2 centimeter (cm) piece of tissue in the meristematic region. Tissue was ground to a fine powder in a GenoGrinder Spex (Metuchen, NJ), weighed in aliquots of 1.2 to 1.5 milligrams (mg) and packed into 3.5 × 5 millimeter (mm) aluminium tins (Costech, Valencia, CA). Packed tins were analysed on an elemental analyser isotope ratio mass spectrometer at Northwestern University Stable Isotope Biogeochemistry Laboratory (NUSIBL).

Carbon fixation was quantified using a section of the youngest, basal frond tissue extending 24 to 30 cm from the bulb. It was weighed (wet mass in grams (g)) with a Pesola scale and placed in a 1 liter (L) Nalgene container; the change in oxygen was measured for 1 hour with a Pyro Science contactless fibre-optic sensors (Oxygen Sensor Spots OXSP5) and an optical oxygen and temperature meter (FireStingO2 FSO2-4), with the sensor spots affixed inside each 1 L chamber.

DOC release and nutrient changes were quantified from seawater filtered through a GF/F filter and immediately frozen for analysis at the University of Washington Marine Chemistry Laboratory (methods from UNESCO), with DOC samples frozen in 40 milliliter (mL) glass vials with Teflon caps (Shimadzu, VOA) and other nutrients in acid-washed 60 mL high density polyethylene (HDPE) Nalgene bottles.

Using pulse amplitude modulated (PAM) fluorometry in situ (Diving Pam, Walz, Germany) we estimated fluorescence over a range of irradiances. PAM is a non-invasive, in situ method that quantifies Photosystem II fluorescence parameters across different light levels. Tissue was dark-adapted for 20 minutes and then rapid light curves were estimated with nine PAR levels from 0 to 1200. We estimated the maximum electron transport rate (ETR_max) and the photosynthetic efficiency at low light, both as the initial linear slope of ETR to irradiance (α) and as the quantum yield of photosystem II (Yield) on nine individual Nereocystis at both Tatoosh Island and Squaxin Island, the populations that span the greatest range of environmental conditions. Tatoosh Island was assayed on 9 and 20 June 2019 and Squaxin Island on 2 July 2019. At both sites, we selected individuals at least 1 meter (m) apart at low tide.

The salinity and seawater temperatures were assessed with a Castaway-CTD (Sontek) from a boat or kayak at all sites except Tatoosh Island where instrumentation is moored (Hach DS5).

Microbial density on individual kelp blades at either end of the geographic distribution, Tatoosh Island versus Squaxin Island, differed using 4,6-diamidino-2-phenylindole (DAPI) fluorescent staining of bacterial cells. We quantified bacterial cells from a section of the blade at approximately 20 cm distal from where the blade articulates with the bulb. We sampled a 3−5 mm slice from 10 individuals at each site (Tatoosh on 4 August 2019, Squaxin on 2 July 2019), preserving each slice immediately in Formalin, transferring to 50 : 50 EtOH : phosphate-buffered saline (PBS) after 1 hour, then freezing. In the laboratory, each piece was again sliced with a razor blade to approximately 0.5 mm, placed on a glass slide, and a solution of 1 microgram (µg) DAPI : 1 mL PBS was dropped on the slide-mounted slice. The slide was kept in dark for 20 minutes, chilled and rinsed with PBS prior to visualizing at 100× with an Olympus BX50 fluorescence microscope with a DAPI filter. We photographed 10 fields of view, selected haphazardly, for each individual slice, resulting in 100 images at each Tatoosh and Squaxin. In samples where bacteria were sparse or non-existent, we repeatedly adjusted the fine focus to verify no bacteria were present. We quantified the bacteria in ImageJ, enumerating the cells in the blue channel, adjusting the background and thresholds identically for all 200 images to eliminate autofluorescence by kelp cells. We calculated bacterial cell density as a percentage of the total imaged area.

Individual kelp was sampled for microbes by rubbing a cotton swab for 20−30 seconds over the mid-blade region. The swab was placed in a 2 mL Eppendorf tube, immediately chilled and transferred to a freezer within 4−6 hours. DNA from the swabs was extracted with a Qiagen DNeasy PowerSoil Kit (Qiagen). DNA was amplified, sequenced and amplicon sequence variants (ASVs) were identified by the Duchossois Family Institute Microbial Metagenomics Facility (DFIMMF) at The University of Chicago. The V4–V5 region within the 16S ribosomal RNA (rRNA) gene was amplified using universal bacterial primers and polymerase chain reaction (PCR) conditions described in Younker, et al. (2024). Approximately 412 bp region amplicons were then purified using a spin column-based method (Minelute, Qiagen), quantified and dual index adapters were ligated. Sequences were generated from the Illumina MiSeq platform using the QIASeq 1-step amplicon kit (Qiagen) for generating libraries and using 2 × 250 paired end reads with 5000 to 10,000 reads per sample. The default pipeline for processing MiSeq 16S rRNA reads was dada2 (v. 1.18.0) with minor modifications in R (v. 4.0.3). Reads were first trimmed at 190 bp for both forward and reverse reads to remove low-quality nucleotides, and chimaeras were detected and removed using the default consensus method in the dada2 pipeline. Then, ASVs with length between 320 and 365 bp were retained. Taxonomy of the resultant ASVs were assigned to the genus level using the RDP classifier (v. 2.13) with a minimum bootstrap confidence score of 80. Species-level classification used blastn (v. 2.13.0) and the refseq_rna database. We analysed ASVs with R, phyloseq and microViz (R version 2023.06.2+561). Chloroplast DNA that was not classified as Cyanobacteria were generally diatoms (Bacillarophyta) and were filtered out, as well as any other chloroplast sequence that did not match to the class Cyanobacteria. Sequences of corn (Zea, in Plantae), in low numbers at some Puget Sound sites, were also removed.

All methods are described in the Pfister, et al. (2025) paper.