Table of Contents

• Coverage
• Dataset Description
  • Methods & Sampling
  • Data Processing Description
  • BCO-DMO Curation Notes
  • Problem Description
• Data Files
• Related Publications
• Parameters
• Instruments
• Deployments
• Project Information
• Funding

Sediment trap deployments and sample collection

REspiration on Sinking Particles In the subsuRface ocEan (RESPIRE) particle collection and incubation systems (Boyd et al., 2015; Bressac et al., 2024) were deployed on a drifting, surface-tethered trap (STT) array at each station for 2-3 days at depths between 80-225 m. RESPIRE traps were filled with 0.2 µm-filtered seawater collected from a CTD cast 2-3 hours prior to deployment. Upon recovery, overlying seawater in RESPIRE traps was siphoned off immediately after recovery, remaining RESPIRE water was collected for bacterial production, dissolved organic carbon (DOC), and cell count measurements, and remaining sample water was passed through a rotary wet splitter and divided into equal fractions. All trap samples were passed through screens to remove zooplankton “swimmers”. Split fractions of the sample water were then vacuum filtered onto pre-combusted 25mm GF/F (Whatman, Maidstone, Kent, United Kingdom) or quartz (Whatman) filters for CHN analyses. Samples collected for DNA were vacuum-filtered onto 25 mm or 47 mm 0.2 µm pore size polyethersulfone filters (Pall Supor, Port Washington, New York, USA), flash-frozen, then stored at -80°C until further processing in the lab. Filters for cell counts and CHN analyses were stored at -20°C and -80°C respectively until further processing in a shore-based lab.

Dissolved and total organic carbon measurements

Prior to each deployment, the 0.2 µm-filtered seawater used to fill RESPIRE traps prior to deployment was sampled for DOC and used as the DOC_initial timepoint. After recovery, particle-containing water was collected from RESPIRE traps for DOC or total organic carbon (TOC) determination (DOC_final or TOC_final). Combusted 40 mL glass vials with Teflon-lined caps were used for sample collection. For DOC samples, water was filtered into the vial using a 0.2 µm filter to a volume of ~30 mL and was acidified with 50 µL of DOC-free 4 N hydrochloric acid (Stephens et al., 2020). TOC samples were collected following the same protocol but without 0.2 µm filtration. All samples were stored at 4°C until measurements could be made in a shore-based lab. DOC and TOC concentrations were measured via the high temperature combustion method using modified Shimadzu TOC-V or TOC-L analyzers following Halewood et al. (2022).

DAPI staining and cell enumeration on particles

Particle-containing fluid (~15mL) from RESPIRE traps were collected and fixed with 0.405 mL of formalin for the purposes of DAPI staining and cell enumeration. Samples were vacuum filtered onto 25 mm, 0.2 µm poresize polycarbonate filters (Pall Supor). Five mL were pipetted into the filtration tower and filtered using vacuum filtration until ~0.5 mL remained above the filter. 0.5 mL DAPI solution (5 µg/mL) was added to the sample, and the filtration tower was covered with foil and left to stain for 3 min. After 3 min, the remaining sample was filtered. The filtered sample was removed from the filtration tower and mounted on a labeled glass microscope slide using a drop of immersion oil between the slide and filter and another drop between the filter and cover slip. Filters were stored at -20°C until imaging and counting on-shore. Epifluorescence microscopy (63X objective, Zeiss Axio Observer 7 Inverted microscope, Carl Zeiss Microscopy, LLC, White Plains, NY, USA) was used to take 12 digital images across different parts of each sample filter.

RESPIRE trap mechanism and respiration rate calculations

The RESPIRE particle collection and incubation apparatus is designed to non-intrusively intercept sinking particles, deposit them into an incubation chamber containing 0.2 µm-filtered seawater from depth and incubate the particles in situ (Boyd et al., 2015; Bressac et al., 2024). Particle collection phases for the RESPIRE trap deployments in this study ranged from 0.5 to 1 d with the following incubation phases lasting 0.5 to 1.5 d. An oxygen optode (Aanderaa 4831 series) within the RESPIRE incubation chamber measured oxygen concentration every five minutes. Dissolved oxygen concentrations were corrected for pressure, salinity, and temperature (if temperature varied notably over the course of the incubation; AT5010 deployment 3) according to the Aanderaa Operating Manual and Thierry et al. (2018). The change in oxygen over the incubation phase time was determined from these measurements for the volume of the RESPIRE incubation chamber. A respiratory quotient (RQ), the incubation length (t_inc), and the volume of the RESPIRE incubation chamber (V_RES, 1.45 L) were used to convert the oxygen-based remineralization rate (O_2,cons) to units of carbon (Equation 1). Here, we define RQ to be the molar drawdown of dioxygen to carbon (O₂:C). We selected an RQ of 1.1 (L. A. Anderson, 1995) for our calculations. End-point POC measurements (POC_final), the change in DOC over the course of the incubation (delta DOC), and the rate of oxygen consumption (O_2,cons) were used to back-calculate the starting concentration of POC (POC_initial) at the beginning of the RESPIRE incubation phase (Equation 1). For deployments where TOC samples were taken, DOC concentrations were calculated by subtracting the POC concentration (after RESPIRE incubation) from the TOC_final concentration (SR1919, AT5010). The change in dissolved organic carbon (DOC) concentration over the course of RESPIRE incubations was measured from two timepoints—an initial (before particle collection) and final (recovered RESPIRE water minus the measured POC concentration). The calculated amount of POC_initial (mmol C) was divided by the length of the collection period (t_col) and the area of the RESPIRE trap (A_RES, 0.014 m²) to estimate POC fluxes (Equation 2). Carbon remineralization rates were normalized to mean POC concentration ([POC]_avg) of the incubation chamber to achieve carbon-specific (C-specific) remineralization rates (R_norm; Equation 3). Uncertainties related to sample splitting, averaging sample replicates, and linear regressions were propagated through all calculations.
 

Equation 1: POC_initial = POC_final + deltaDOC + (O_2,cons / RQ * t_inc * V_RES)

Equation 2: POC flux = POC_initial / (t_col * A_RES)

Equation 3: R_norm = (O_2,cons / RQ) / [POC]_avg

DAPI staining and cell enumeration on particles

Epifluorescence microscopy images were analyzed using CellProfiler (v4.2.8). Images were converted to grayscale as needed (Relative weight of each channel: Red = 0, Green = 0.8, Blue = 1.0) and images were analyzed sample by sample. If background intensity values were high, an illumination correction was applied to the image using a median filter smoothing method. Cells were identified based on intensity relative to the background intensity which was specific to each sample. The image with the most and least number of cells were removed from analysis and the cell counts for the remaining 10 images were averaged for each sample. Cell concentration in the RESPIRE incubation chamber was back-calculated from cell counts using the area imaged, the area of the filter, the volume of sample filtered, and the volume of the RESPIRE incubation chamber (1.45 L). Cell counts were converted to cell carbon using a carbon conversion factor of 20 fg C cell^-1 (Ducklow, 2000).

This section documents curation actions performed prior to publication review with the submitter, and additional information relevant to understanding and reusing this dataset. It distinguishes changes made to the submitted (meta)data from unresolved issues and/or enhancements that improve future reuse and interoperability.

CURATION ACTIONS PERFORMED ON DATA

- Loaded BCO-DMO_AT5010_SR1919_2026-05-19.csv, using filename as resource name, with header row 1 and row 2 skipped; set missing value markers to "", "nd", and "NA"
- Output file as 1008909_v1_respire_rates_at5010_sr1919.csv

CURATION ACTIONS PERFORMED ON METADATA

- BCO-DMO's standard metadata entry and text formatting steps were performed. See: https://www.bco-dmo.org/how-to/standard-curation-edits

ISSUES POTENTIALLY IMPACTING REUSE

- N/A

NSF abstract:
Though scarce and largely insoluble, trace metals are key components of sophisticated enzymes (protein molecules that speed up biochemical reactions) involved in biogeochemical cycles in the dark ocean (below 1000m). For example, metalloenzymes are involved in nearly every reaction in the nitrogen cycle. Yet, despite direct connections between trace metal and nitrogen cycles, the relationship between trace metal distributions and biological nitrogen cycling processes in the dark ocean have rarely been explored, likely due to the technical challenges associated with their study. Availability of the autonomous underwater vehicle (AUV) Clio, a sampling platform capable of collecting high-resolution vertical profile samples for biochemical and microbial measurements by large volume filtration of microbial particulate material, has overcome this challenge. Thus, this research project plans an interdisciplinary chemistry, biology, and engineering effort to test the hypothesis that certain chemical reactions, such as nitrite oxidation, could become limited by metal availability within the upper mesopelagic and that trace metal demands for nitrite-oxidizing bacteria may be increased under low oxygen conditions. Broader impacts of this study include the continued development and application of the Clio Biogeochemical AUV as a community resource by developing and testing its high-resolution and adaptive sampling capabilities. In addition, metaproteomic data will be deposited into the recently launched Ocean Protein Portal to allow oceanographers and the metals in biology community to examine the distribution of proteins and metalloenzymes in the ocean. Undergraduate students will be supported by this project at all three institutions, with an effort to recruit minority students. The proposed research will also be synergistic with the goals of early community-building efforts for a potential global scale microbial biogeochemistry program modeled after the success of the GEOTRACES program, provisionally called "Biogeoscapes: Ocean metabolism and nutrient cycles on a changing planet".

The proposed research project will test the following three hypotheses: (1) the microbial metalloenzyme distribution of the mesopelagic is spatially dynamic in response to environmental gradients in oxygen and trace metals, (2) nitrite oxidation in the Eastern Tropical Pacific Ocean can be limited by iron availability in the upper mesopelagic through an inability to complete biosynthesis of the microbial protein nitrite oxidoreductase, and (3) nitrite-oxidizing bacteria increase their metalloenzyme requirements at low oxygen, impacting the distribution of both dissolved and particulate metals within oxygen minimum zones. One of the challenges to characterizing the biogeochemistry of the mesopelagic ocean is an inability to effectively sample it. As a sampling platform, we will use the novel biogeochemical AUV Clio that enables high-resolution vertical profile samples for biochemical and microbial measurements by large volume filtration of microbial particulate material on a research expedition in the Eastern Tropical Pacific Ocean. Specific research activities will be orchestrated to test the hypotheses. Hypothesis 1 will be explored by comparison of hydrographic, microbial distributions, dissolved and particulate metal data, and metaproteomic results with profile samples collected by Clio. Hypothesis 2 will be tested by incubation experiments using 15NO2- oxidation rates on Clio-collected incubation samples. Hypothesis 3 will be tested by dividing targeted nitrite oxidoreductase protein copies by qPCR (quantitative polymerase chain reaction)-based nitrite oxidizing bacteria abundance (NOB) to determine if cellular copy number varies with oxygen distributions, and by metalloproteomic analyses of NOB cultures. The demonstration of trace metal limitation of remineralization processes, not just primary production, would transform our understanding of the role of metals in biogeochemical cycling and provide new ways with which to interpret sectional data of dissolved and particulate trace metal distributions in the ocean. The idea that oxygen may play a previously underappreciated role in controlling trace metals due not just to metals' physical chemistry, but also from changing biological demand, will improve our ability to predict trace metal distributions in the face of decreasing ocean oxygen content.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.