@article{kamykowski_zentara_2005, title={Changes in world ocean nitrate availability through the 20th century}, volume={52}, ISSN={["1879-0119"]}, DOI={10.1016/j.dsr.2005.04.007}, abstractNote={Conceptual models linking climate change with fluctuations in fish population abundances are based on how cyclic patterns in air–sea interactions influence pelagic food web dynamics. The effect of changing mixed layer dynamics on phytoplankton light and nutrient exposure is a prominent focal point in the overall mechanism. The Extended Reconstruction (ER) of Sea Surface Temperature (SST) version one (ERSSTv1) and version two (ERSSTv2) monthly time series from 1854 to 2003, interpreted with the aid of a historically based global Nitrate Depletion Temperature (NDT) climatology, provide a qualitative tool for examining relative temporal and spatial patterns in nitrate availability in normal salinity areas of the world ocean. After an analysis of local NDT variability at four time series stations demonstrated temporal stability compared to SST, [SST–NDT] as a Nitrate Availability Index (NAI) was calculated for April (boreal spring or austral fall) and October (boreal fall and austral spring) for the whole ERSSTv1 data set and for selected years of the ERSSTv2 data set using the global NDT climatology. The more negative the NAI difference, the greater the expected surface nitrate. The more positive the NAI difference, the greater the intensity of temperature stratification between the surface and the nitracline and thus the less likely that nitrate mixed to the surface. The records from April and October both showed that decreased nitrate availability, defined by both smaller negative NAI differences and larger positive NAI differences, generally though not universally occurred throughout the 20th century in association with global warming. The greatest decreases in nitrate availability occurred in two warming events in the time periods 1909–1937 and 1977–present in the Northern Hemisphere and 1926–1937 and 1950–1990 in the Southern Hemisphere. Different areas of the world ocean were affected in each warming event. Prominent exceptions in the ERSSTv1 analysis where 20th century nitrate availability actually increased in at least one season were in western parts of the South Indian, the North Pacific, the equatorial Pacific, the South Pacific, the North Atlantic, and the South Atlantic and in eastern parts of the South Pacific and South Atlantic. The ERSSTv2 analysis also showed increased 20th century nitrate availability in the eastern subarctic Pacific. The nitrate availability trends resulting from the present analysis agree with various literature reports on regional changes in plant nutrient availability, on increased occurrences of harmful algal blooms often associated with dinoflagellates that are better able to access subsurface nitrate pools, and on decadal changes in marine fisheries. They also raise concerns about the resilience of historical patterns of pelagic community structure and function in response to a warming trend continuing into the 21st century.}, number={9}, journal={DEEP-SEA RESEARCH PART I-OCEANOGRAPHIC RESEARCH PAPERS}, author={Kamykowski, D and Zentara, SJ}, year={2005}, month={Sep}, pages={1719–1744} }
 @article{kamykowski_zentara_2003, title={Can phytoplankton community structure be inferred from satellite-derived sea surface temperature anomalies calculated relative to nitrate depletion temperatures?}, volume={86}, ISSN={["0034-4257"]}, DOI={10.1016/S0034-4257(03)00123-8}, abstractNote={Hydrographic data collected in the upper 50 m off La Jolla, CA, USA (31°N, 117°W) between 1970 and 1972 were reanalyzed to examine temporal variability in the local temperature–nitrate relationship and to document how chlorophyll a concentration and phytoplankton community structure covary with the temperature–nitrate relationship. Based on the linear expression y=mx+b, the y-intercepts (b), slopes (m), and x-intercepts (−b/m or nitrate depletion temperature, NDT) of four seasonal (January–March, April–June, July–September, and October–December) temperature–nitrate relationships, obtained from the combined multiyear data set, were statistically different from each other and varied around overall multiyear values of b=72.73 μM, m=−5.33 μM °C−1, and NDT=13.65 °C. Three interannual temperature–nitrate relationships from February to April 1970, 1971, and 1972 also had y-intercepts, slopes, and x-intercepts that were statistically different from each other. Nevertheless, limited variability in direct comparisons among seasonal or interannual regression lines and a 1 °C La Jolla NDT range compared to a 25 °C global NDT range supported the general utility of NDT-based comparisons. A nitrate-normalized temperature axis (T−NDT) was created for the La Jolla data set by subtracting NDT from the recorded water column temperatures (T). Chlorophyll a reached a maximum between 0 and 2 °C on this T–NDT axis that ranged from −4 to 10 °C. Microscope-based determinations of La Jolla centric diatom, pennate diatom and dinoflagellate abundances, and La Jolla chlorophyll a, partitioned in proportion to the numerical abundance of the three groups, both peaked in logical progression along the T–NDT axis. In a separate analysis of high-performance liquid chromatography (HPLC) data from three Atlantic Meridional Transect (AMT) cruises (50°N to 52°S), chlorophyll a peaked below 0 °C and three different phytoplankton classes, nanoflagellates, large eukaryotes and prokaryotes, distributed in logical progression along a sea surface temperature (SST) minus NDT axis. To further generalize these results, a previously reported 1° latitude×1° longitude grid of NDTs for the world ocean was applied to satellite-derived grids of SST for March 1999 through June 2000. The SST−NDT calculation provided a standard nitrate-normalized axis simultaneously applicable to all locations in the world ocean. Sixteen plots of satellite-derived chlorophyll a versus SST−NDT for March 1999 through June 2000 demonstrated the opposing seasonal movements of northern and southern hemisphere chlorophyll a along the SST–NDT axis. Based on the phytoplankton community patterns along the temperature minus NDT in the La Jolla and AMT data sets, this chlorophyll a movement along the SST–NDT axis can be associated with phytoplankton community changes related to location around SST−NDT=0 °C. The SST−NDT index appears to provide a useful tool for interpreting the character of the phytoplankton community structure contributing to satellite-derived chlorophyll a in the world ocean.}, number={4}, journal={REMOTE SENSING OF ENVIRONMENT}, author={Kamykowski, D and Zentara, SJ}, year={2003}, month={Aug}, pages={444–457} }
 @article{switzer_kamykowski_zentara_2003, title={Mapping nitrate in the global ocean using remotely sensed sea surface temperature}, volume={108}, number={C8}, journal={Journal of Geophysical Research. Oceans}, author={Switzer, A. C. and Kamykowski, D. and Zentara, S. J.}, year={2003} }
 @article{kamykowski_zentara_morrison_switzer_2002, title={Dynamic global patterns of nitrate, phosphate, silicate, and iron availability and phytoplankton community composition from remote sensing data}, volume={16}, number={4}, journal={Global Biogeochemical Cycles}, author={Kamykowski, D. and Zentara, S. J. and Morrison, J. M. and Switzer, A. C.}, year={2002}, pages={1077–1} }