PhD Research
PhD Thesis
Improved estimates of net community production in the Subarctic Pacific and Canadian Arctic Ocean using ship-based autonomous measurements and computational approaches
My PhD thesis research focuses on improving estimates of marine net community production (NCP), through the use of observational and numerical/statistical tools. By definition, NCP (i.e. the balance between algal photosynthetic production and community-wide aerobic respiration) is a useful metric that informs us of the net metabolic state, or breathing rate, of an ocean region, and on long (e.g. annual) time-scales, it sets upper limits on the ocean’s capacity for biomass production (i.e. the ocean's ability to support other life), and CO2-sequestration from the atmosphere.
As O2 is a main currency of photosynthesis and respiration, in-situ measurements of dissolved O2 can inform us of NCP within an ocean region. My thesis thus revolves around the estimation of NCP from ship-based measurement of O2. To achieve this, I combine autonomous sampling from research vessels, with numerical and statistical models to obtain high-resolution, broad-coverage estimates of NCP.
The oceanographic regions of chief interest to me are the Subarctic Northeast Pacific, and various sub-regions of the Canadian Arctic (Fig. 1.1). My work is based mostly on data obtained from the Line P and La Perouse monitoring / time-series programs (NE Pacific), and annual ArcticNet CCGS Amundsen expeditions (Arctic). Both regions are characterized by seasonally-high NCP and ecosystem productivity, and are experiencing wide-spread alterations as symptoms of climate change. A key goal of my thesis is, thus, to produce a series of tools (including ship-based measurement systems, numerical approaches, and a predictive algorithm; see below for details) for obtaining high-resolution, accurate estimates of NCP in the Subarctic NE Pacific and Canadian Arctic. Subsequently, long-term variability and trends in NCP can be observed and related to changes in the planet’s climate, (commercial) fish production, and overall ocean ecosystem health. This work is important for predicting the ocean's response to future climate change.
My PhD thesis research focuses on improving estimates of marine net community production (NCP), through the use of observational and numerical/statistical tools. By definition, NCP (i.e. the balance between algal photosynthetic production and community-wide aerobic respiration) is a useful metric that informs us of the net metabolic state, or breathing rate, of an ocean region, and on long (e.g. annual) time-scales, it sets upper limits on the ocean’s capacity for biomass production (i.e. the ocean's ability to support other life), and CO2-sequestration from the atmosphere.
As O2 is a main currency of photosynthesis and respiration, in-situ measurements of dissolved O2 can inform us of NCP within an ocean region. My thesis thus revolves around the estimation of NCP from ship-based measurement of O2. To achieve this, I combine autonomous sampling from research vessels, with numerical and statistical models to obtain high-resolution, broad-coverage estimates of NCP.
The oceanographic regions of chief interest to me are the Subarctic Northeast Pacific, and various sub-regions of the Canadian Arctic (Fig. 1.1). My work is based mostly on data obtained from the Line P and La Perouse monitoring / time-series programs (NE Pacific), and annual ArcticNet CCGS Amundsen expeditions (Arctic). Both regions are characterized by seasonally-high NCP and ecosystem productivity, and are experiencing wide-spread alterations as symptoms of climate change. A key goal of my thesis is, thus, to produce a series of tools (including ship-based measurement systems, numerical approaches, and a predictive algorithm; see below for details) for obtaining high-resolution, accurate estimates of NCP in the Subarctic NE Pacific and Canadian Arctic. Subsequently, long-term variability and trends in NCP can be observed and related to changes in the planet’s climate, (commercial) fish production, and overall ocean ecosystem health. This work is important for predicting the ocean's response to future climate change.
1. Refining estimates of NCP for vertical mixing fluxes of O2
A common approach to estimating NCP involves solving the mixed layer O2 mass balance budget, where biological production (i.e. NCP), gas exchange across the air-sea interface (via bubbles or diffusive exchange), and mixing across vertical and horizontal boundaries contribute to O2 variability (Fig. 1.2). To isolate the biological fraction of the O2 pool, a term called the “biological O2 saturation anomaly” (ΔO2/Ar) can be quantified from measurements of O2 and argon (Ar). Because O2 and Ar are gases with similar physical properties, ΔO2/Ar is a tracer of biological activity and is largely unaffected by the physical effects from bubble injection or temperature- and salinity-induced changes in O2 saturation state. By this approach, NCP is proportional to ΔO2/Ar.
ΔO2/Ar measurements can be obtained on very fine scales (<30 second intervals) using ship-based mass spectrometry (Fig. 1.3), and this approach has been used widely. However, it remains a key challenge to quantify the contribution of vertical mixing to a regional O2 (and thus, ΔO2/Ar) budget. In particular, near the coast, where NCP, carbon export and fish production are seasonally high, large mixing fluxes of O2 from the sub-surface (where O2 is typically depleted) can bias surface ΔO2/Ar measurements and NCP estimates. Our understanding of NCP in these ecologically-significant regions may be skewed by the inability to account for mixing fluxes. My thesis research has focused on accurately quantifying these vertical fluxes using both chemical tracers, and numerical models.
One approach to deriving the vertical O2 mixing flux, uses nitrous oxide (N2O) as a chemical tracer. This is based on the observation that O2 and N2O exhibit consistent, linear stoichiometry (Fig. 1.4a) due to nitrification in the the subsurface. This approach was initially shown to be valid in a model framework (Cassar et al., 2014) but I demonstrated its utility for field studies in the subarctic Pacific (Izett et al., 2018). This work suggested that past studies have indeed been significantly underestimating NCP in coastal regions (Fig. 1.4b). Future work on this will involve high-resolution surface N2O sampling to derive a mixing correction at the same resolution as underway ΔO2/Ar observations.
The N2O-based approach has great promise for many ocean regions, but the complexity of the nitrogen cycle in some regions (including the Arctic) precludes its use. N2O surveys are also uncommon on most field campaigns, or measurements are typically made at low spatial resolution. Therefore, numerical models may be used to quantify physical vertical mixing fluxes. This has been the focus of some of my recent work. Specifically, I have paired in-situ observations of ΔO2/Ar (from ship-based mass spectrometry) with estimates of eddy diffusivity coefficients derived from NEMO model output to derive mixing correction estimates along a cruise track in the Arctic (Tortell et al., in preparation).This work is on-going.
Relevant publications: Izett et al., 2018; Izett et al., In preparation
Relevant publications: Izett et al., 2018; Izett et al., In preparation
2. Estimation of NCP from underway O2 and N2 observations
While the spatial and temporal extent of NCP studies is enhanced by the use of ship-based mass spectrometry (i.e. as opposed to discrete gas sampling), obtaining broad coverage estimates nonetheless remains challenging. One difficulty is that mass spectrometers can be cost-prohibitive to many research groups and require considerable expertise and attention to maintain and run at sea. Thanks to new technological developments in autonomous gas sensors, alternative systems may now be feasible. Specifically, a combined O2 Optode and Gas Tension Device (GTD) system could be used as an alternative for NCP estimation, whereby nitrogen (N2), derived from GTD measurements, is used to define ΔO2/N2. The concurrent deployment of an optode and GTD can yield underway estimates of the seawater O2/N2 ratio at high sampling resolution, and like ΔO2/Ar, ΔO2/N2 can be used to isolate the biological fraction of the O2 pool.
The potential to estimate NCP from simultaneous O2 and N2 measurements has led me to develop a ship-based flow-through optode/GTD system (nicknamed the “PIGI” system for the Pressure of In-situ Gases Instrument; see PIGI System). The PIGI system offers many advantages over mass-spectrometry based systems: it is fully autonomous (with automated pumping system, and custom-designed logging software), can be monitored remotely via iridium transmission, is (relatively) inexpensive, can be run for extended deployments with minimal oversight, and is easy to setup and deploy on ships, or land-based field stations. Ultimately, these features mean that multiple optode/GTD units could be simultaneously deployed unsupervised (or in a manner similar to underway TSG systems), and on a number of vessels to obtain unrivalled coverage of surface ocean O2 and N2 measurements, and NCP estimates.
Since N2 is less soluble than O2 and Ar, divergence between ΔO2/Ar and ΔO2/N2 signals is expected under some conditions; in-situ comparisons between ΔO2/Ar and ΔO2/N2 are thus required to validate this approach. This has been the focus of much of my recent work. This work is ongoing, but preliminary results suggest that ΔO2/N2 can be used successfully to obtain accurate NCP estimates.
Relevant publication: In progress.