Publications

Export 2 results:
Sort by: [ Author  (Asc)] Title Type Year
A B C D E F G H I J K L M N O P Q R S T U V W X Y Z   [Show ALL]
K
Kahru, M, Lee ZP, Mitchell BG, Nevison CD.  2016.  Effects of sea ice cover on satellite-detected primary production in the Arctic Ocean. Biology Letters. 12   10.1098/rsbl.2016.0223   AbstractWebsite

The influence of decreasing Arctic sea ice on net primary production (NPP) in the Arctic Ocean has been considered in multiple publications but is not well constrained owing to the potentially large errors in satellite algorithms. In particular, the Arctic Ocean is rich in coloured dissolved organic matter (CDOM) that interferes in the detection of chlorophyll a concentration of the standard algorithm, which is the primary input to NPP models. We used the quasianalytic algorithm (Lee et al. 2002 Appl. Opti. 41, 575525772. (doi: 10.1364/AO.41.005755)) that separates absorption by phytoplankton from absorption by CDOMand detrital matter. We merged satellite data from multiple satellite sensors and created a 19 year time series (1997-2015) of NPP. During this period, both the estimated annual total and the summer monthly maximum pan-Arctic NPP increased by about 47%. Positive monthly anomalies in NPP are highly correlated with positive anomalies in open water area during the summer months. Following the earlier ice retreat, the start of the high-productivity season has become earlier, e.g. at a mean rate of -3.0 d yr(-1) in the northern Barents Sea, and the length of the high-productivity period has increased from15 days in 1998 to 62 days in 2015. While in some areas, the termination of the productive season has been extended, owing to delayed ice formation, the termination has also become earlier in other areas, likely owing to limited nutrients.

S
Schwarz, JN, Kowalczuk P, Kaczmarek S, Cota GF, Mitchell BG, Kahru M, Chavez FP, Cunningham A, McKee D, Gege P, Kishino T, Phinney DA, Raine R.  2002.  Two models for absorption by coloured dissolved organic matter (CDOM). Oceanologia. 44:209-241. AbstractWebsite

The standard exponential model for CDOM absorption has been applied to data from diverse waters. Absorption at 440 nm (a(g)440) ranged between close to zero and 10 m(-1), and the slope of the semilogarithmic absorption spectrum over a minimum range of 400 to 440 nm (s440) ranged between < 0.01 and 0.04 nm(-1). NO relationship was found between a(g)440 or s440 and salinity. Except in the southern Baltic, s440 was found to have abroad distribution (0.0165 +/- 0.0035), suggesting that it should be introduced as an additional variable in bio-optical models when a(g)440 is large. An alternative model for CDOM absorption was applied to available high quality UV-visible absorption spectra from the Wisla river (Poland). This model assumes that, the CDOM absorption spectrum comprises distinct Gaussian absorption bands in the UV, similar to those of benzene. Five bands were fit to the data. The mean central energy of all hands was higher in early summer (E &SIM;7.2, 6.6, 6.4, 6.2 and 5.5 eV or 172, 188, 194, 200 and 226 nm)) than in winter. The higher energy bands were found to decay in both height and width with increasing salinity, while lower energy bands broadened with increasing salinity. 8440 was found to be correlated with shape parameters of the bands centred at 6.4 and 5.5 eV. While the exponential model is convenient for optical modelling and remote sensing applications, these results suggest that the Gaussian model offers a deeper understanding of chemical interactions affecting CDOM molecular structure.