A new procedure to optimize the selection of groups in a classification tree

Agglomerative cluster analyses encompass many techniques, which have been widely used in various fields of science. In biology, and specifically ecology, datasets are generally highly variable and may contain outliers, which increase the difficulty to identify the number of clusters. Here we present a new criterion to determine statistically the optimal level of partition in a classification tree. The criterion robustness is tested against perturbated data (outliers) using an observation or variable with values randomly generated. The technique, called Random Simulation Test (RST), is tested on (1) the well-known Iris dataset [Fisher, R.A., 1936. The use of multiple measurements in taxonomic problems. Ann. Eugenic. 7, 179–188], (2) simulated data with predetermined numbers of clusters following Milligan and Cooper [Milligan, G.W., Cooper, M.C., 1985. An examination of procedures for determining the number of clusters in a data set. Psychometrika 50, 159–179] and finally (3) is applied on real copepod communities data previously analyzed in Beaugrand et al. [Beaugrand, G., Ibanez, F., Lindley, J.A., Reid, P.C., 2002. Diversity of calanoid copepods in the North Atlantic and adjacent seas: species associations and biogeography. Mar. Ecol. Prog. Ser. 232, 179–195]. The technique is compared to several standard techniques. RST performed generally better than existing algorithms on simulated data and proved to be especially efficient with highly variable datasets.

Analytical phytoplankton carbon measurements spanning diverse ecosystems

The measurement of phytoplankton carbon (Cphyto) in the field has been a long-sought but elusive goal in oceanography. Proxy measurements of Cphyto have been employed in the past, but are subject to many confounding influences that undermine their accuracy. Here we report the first directly measured Cphyto values from the open ocean. The Cphyto samples were collected from a diversity of environments, ranging from Pacific and Atlantic oligotrophic gyres to equatorial upwelling systems to temperate spring conditions. When compared to earlier proxies, direct measurements of Cphyto exhibit the strongest relationship with particulate backscattering coefficients (bbp) (R2=0.69). Chlorophyll concentration and total particulate organic carbon (POC) concentration accounted for ~20% less variability in Cphyto than bbp. Ratios of Cphyto to Chl a span an order of magnitude moving across and within distinct ecosystems. Similarly, Cphyto:POC ratios were variable with the lowest values coming from productive temperate waters and the highest from oligotrophic gyres. A strong relationship between Cphyto and bbp is particularly significant because bbp is a property retrievable from satellite ocean color measurements. Our results, therefore, are highly encouraging for the global monitoring of phytoplankton biomass from space. The continued application of our Cphyto measurement approach will enable validation of satellite retrievals and contribute to an improved understanding of environmental controls on phytoplankton biomass and physiology.