An Assessment of Ocean Color Model Estimates of Net Primary Productivity (NPP) in the Arctic Ocean

Younjoo Lee, Bigelow Laboratory for Ocean Sciences, ylee@bigelow.org
Patricia Matrai, Bigelow Laboratory for Ocean Science, pmatrai@bigelow.org
Marjorie Friedrichs, Virginia Institute of Marine Science, marjy@vims.edu
Vincent S Saba, NOAA/NMFS, vincent.saba@noaa.gov (Presenter)

The Primary Productivity Algorithm Round Robin activity (PPARR) provides a framework such that the skill and sensitivities of net primary productivity (NPP) estimated by ocean color-based models can be assessed in the Arctic Ocean (AO), based on the model’s ability to reproduce mean and variability of NPP. We present here the first phase results from the 32 models that estimate depth-integrated marine NPP with respect to a unique pan-Arctic data set (1988-2011) that includes in situ, satellite, climatological and/or re-analysis data. Model simulation was mainly furnished with information about surface chlorophyll-a concentration (hereafter chlorophyll), photosynthetically available radiation (PAR), sea surface temperature (SST), mixed-layer depth (MLD), and other satellite-derived properties such as phytoplankton absorption coefficient and remote-sensing reflectance. The models were most sensitive to the source of surface chlorophyll but relatively less sensitive to the source of PAR, SST, and MLD. The models performed better using in situ chlorophyll as opposed to satellite-derived chlorophyll. Regardless of model type or complexity, most models significantly underestimated the variability of NPP, often by more than a factor of two whereas some models exhibited almost no bias. Although the model performance significantly varied seasonally and regionally, the models overestimated NPP in low-productivity regions/seasons and under-estimated NPP in high-productivity regions/seasons. In addition, the models performed better in low NPP regions when no sub-surface chlorophyll-a maximum (SCM) was present. This study suggests that ocean color models need to be tuned seasonally or regionally when applying to the AO since most of the models performed relatively well in estimating NPP were developed or modified for the Arctic environment. Furthermore, future success of ocean color models in estimating NPP rely on improvement of algorithms to derive satellite chlorophyll for the AO.

Presentation Type:  Poster

Session:  General Contributions   (Tue 4:35 PM)

Associated Project(s): 

• Matrai, Paty: Net Primary Productivity Algorithm Round Robin for the Arctic Ocean ...details

Poster Location ID: 220