The advent of missions comprised of phased arrays of spacecraft, with separ
ation distances ranging down to at least mesoscales, provides the scientifi
c community with an opportunity to accurately analyse the spatial and tempo
ral dependencies of structures in space plasmas. Exploitation of the multi-
point data sets, giving vastly more information than in previous missions,
thereby allows unique study of their small-scale physics. It remains an out
standing problem, however, to understand in what way comparative informatio
n across spacecraft is best built into any analysis of the combined data. D
ifferent investigations appear to demand different methods of data co-ordin
ation. Of the various multi-spacecraft data analysis techniques developed t
o affect this exploitation, the discontinuity analyser has been designed to
investigate the macroscopic properties (topology and motion) of boundaries
, revealed by multi-spacecraft magnetometer data, where the possibility of
at least mesoscale structure is considered. It has been found that the anal
ysis of planar structures is more straightforward than the analysis of non-
planar boundaries, where the effects of topology and motion become interwov
en in the data, and we argue here that it becomes necessary to customise th
e analysis for non-planar events to the type of structure at hand. One issu
e central to the discontinuity analyser, for instance, is the calculation o
f normal vectors to the structure. In the case of planar and 'thin' non-pla
nar structures, the method of normal determination is well-defined, althoug
h subject to uncertainties arising from unwanted signatures. In the case of
'thick', non-planar structures, however, the method of determination becom
es particularly sensitive to the type of physical sampling that is present.
It is the purpose of this article to firstly review the discontinuity anal
yser technique and secondly, to discuss the analysis of the normals to thic
k non-planar structures detected in magnetometer data.