Publication Abstract

Spatial distribution functions and abundances inferred from sparse noisy plankton data: an application of constrained thin plate splines

Spatial distribution functions andabundances inferred from sparse noisy plankton data: an application of constrained thinplate splines

S. Wood, S. and J.W.H. Horwood

Asplankton biologists ask more detailed questions of necessarily sparse and noisy spatialdata, the need for well founded methods for statistical analysis of such data grows. Thisnote examines the utility of constrained thin-plate smoothing splines as a tool forinferring underlying spatial distribution functions from sparse noisy data. Constrainedthin-plate splines are described in a straightforward manner. An economical method ofcalculation is suggested, which sacrifices mathematical optimality for ease ofcomputation. Using simulated data several methods for choosing the complexity of theinferred distribution function are compared and robustness to large amplitude noise isexamined. Confidence intervals are calculated and tested. The method is applied to eggdata from Dover sole (Solea solea) in the Bristol Channel.

Reference:

S. Wood and J.W. Horwood, 1995. Spatial distribution functions and abundances inferredfrom sparse noisy plankton data: an application of constrained thin plate splines. Journalof Plankton Research, 17(6): 1189-1208.

S. Wood and J.W. Horwood*

Journal of Plankton Research, 17(6): 1189-1208