The uptake of silica during the spring bloom in the North East Atlantic Ocean

A full understanding of the biogeochemical cycling of silica in the North Atlantic is hampered by a lack of estimates of silica uptake by phytoplankton. We applied the ${}^{32}\text{Si}$ radiotracer incubation technique to determine silica uptake rates at 10 sites during the UK-(Natural Environment Research Council) Faroes-Iceland-Scotland hydrographic and environmental survey (FISHES) cruise in the Northeast Atlantic, May 2001. Column silica uptake rates ranged between 6 and 166 mmol Si $\text{m}^{-2}\ \text{d}^{-1}$; this data set was integrated with concurrent hydrographic, chemical, and primary productivity data to explain these changes in silica uptake in terms of the progress of the spring bloom. In order to interpret data covering a relatively large spatial and temporal scale, we used mean photic zone silica concentration as a proxy time-series measure of diatom bloom progression. Both absolute and specific silica uptake rates were highest at dissolved silica concentrations >2 mmol $\text{L}^{-1}$. Si and C uptake were vertically decoupled at those stations where surface silica was strongly depleted. Absolute primary productivity was not strongly correlated with dissolved silica concentrations, owing to either exhaustion of silica at diatom-dominated stations or to dominance of the community by other phytoplankton. Silica uptake as a function of increased substrate concentration was linear up to 25 $\mu \text{mol}\ \text{L}^{-1}$; we consider some possible reasons for the nonhyperbolic response.

Error sensitive historical GIS: Identifying areal interpolation errors in time series data

Historical GIS has the potential to re-invigorate our use of statistics from historical censuses and related sources. In particular, areal interpolation can be used to create long-run time-series of spatially detailed data that will enable us to enhance significantly our understanding of geographical change over periods of a century or more. The difficulty with areal interpolation, however, is that the data that it generates are estimates which will inevitably contain some error. This paper describes a technique that allows the automated identification of possible errors at the level of the individual data values.