People

Hyperspectral radiative measurements spanning from UV to NIR such as will be available with PACE's OCI will improve our ability to estimate radiative products compared to the current ones by constraining atmospheric transmission at many more wavelengths, narrower bands and into a wider range of wavelengths than current satellites. Together with PACE's expanded suite of biogeochemical products, such as absorption by phytoplankton and dissolved organic materials, the improved radiative products will allow for improved constraints of aquatic biogeochemical rate process, for example, near-surface primary production and photo-oxidation.

It has recently become apparent that current radiative products are not optimal for users. In particular, to compute rates of photochemical and photobiological processes the necessary input is the sub-surface scalar irradiance and its attenuation with depth spanning from UV and across the visible. However, current ocean color radiative products are currently limited to above surface downwelling spectral irradiance and PAR.

Here we propose to expand the radiative products suite distributed by NASA's Ocean Biology and biogeochemistry Processing Group (OBPG) to include both daily and instantaneous (at the time of satellite passage) spectral downwelling and scalar irradiance below the surface, ocean albedo and the appropriate attenuation function to propagate them to different depths. We will provide OBPG the algorithms for these novel products and work closely with them on their implementation. In addition, we will generate software tools (subroutines) for users interested in obtaining band integrated PAR and UVA radiative values at a given depth or integrated over a depth range (e.g., mixed- layer). Prior to PACE, we will create and test demonstration algorithms using data from HICO (spectral and PAR) and MOBY (UVA) and validated using radiometry data available on SeaBASS, PANGAEA, profiling floats and from different mooring sites.

We will reach out to the relevant communities (e.g., ecosystem and biogeochemical modelers) early in this project to insure that the algorithms we produce are fit-for-purpose and that these communities are ready to ingest the relevant products as soon as they become available online.

The activity proposed here consists of 1. Creating a consortium of practitioners interested in standardizing collection and processing of high horizontal resolution data collected using flowthrough in-line systems, 2. Sharing associated codes, 3. Providing the community with a plan for deployment, processing and computation of estimated uncertainties associated with in-line data, and 4. Creating a uniformly processed dataset for Cal/Val activities and studies of inherent optical properties (IOP) distributions throughout the world's oceans.

Such a dataset will be unique in its global extent, being ideal for validation of remote sensing product and for algorithm development for a global mission such as PACE. Critical evaluation of the in-line IOP acquisition is necessary to assign realistic uncertainties to those IOPs.

Once processing methodology is agreed upon among the collaborators, UMaine will reprocess historical in-line data collected by the collaborators and provide them to SeaBASS with the processing algorithms and source codes for future use by the ocean color community. Efforts will be made such that data generated will have sufficient details so that alternative processing could be applied without the need to reprocess the raw data.

As part of this proposal we will use the data to answer the following SCIENCE question: What are the characteristics of sub-satellite-pixel variability in IOPs in the ocean?

The utility of the in-line dataset goes well beyond the scope of this proposal and can be used to answer other science questions directly relevant to PACE (a global hyperspectral mission), such as:

• What are the deviations of IOPs from published bio-optical relationships and how do they vary with variables such as temperature, salinity, date, distance from land and ocean depth?
• What information is available in hyperspectral IOPs (and hence hyperspectral ocean color) in addition to that currently obtained with spectral sensors (e.g. added pigments in addition to chlorophyll a (e.g. Chase et al., 2014), size information etc.)?

As part of this proposal we will use the data to answer the following METHODOLOGICAL questions: 1. What should we acquire as discrete samples to increase the utility of the in-line systems? 2. How do in-line data compare to in-water data collected with similar sensors at the same time (e.g. are there noticeable biases in in-line data?)? Answering these methodological questions will improve in-line collection and our estimates of uncertainties for the data collected. Uncertainties are necessary to evaluate the degree of agreement between remote and in-situ estimates of IOPs as well as biogeochemical quantities.

PI Boss also proposes himself to be the IOP Science team lead.