Geoscience Reference
In-Depth Information
question may then arise about the evolution of these parameters over time, and thus
about the reference frame associated with the method upon which the interpretation
will depend. It is also typically the case of multivariate factorial analyses made at
several dates: while the shares of the variance explained by the first factor at each
date can, for example, be compared, the positions of an entity on the first factor
computed for different dates cannot be directly compared. Indeed, these analyses are
describing differentiations according to a specific reference frame at each date.
The rest of this chapter will be dedicated to applications of spatio-temporal
analyses. The distinction between an input directly centered upon the evolution of
the spatial organization
versus
an input centered on the evolution of entities, is used
as a support for the adopted plan. Section 3.2 is dedicated to the first of these inputs
and to the steps to understand the evolution of a spatial system's structure. We will
distinguish here between two types of approach:
- those consisting of calculating a time series of indicators summarizing the
spatial organization (for example a center-periphery gradient or the spatial
autocorrelation of a phenomenon) (section 3.2.1);
- those based on the follow-up in time of spatial associations (correlation
between variables describing spatial entities, for example, or factorial analyzes in the
multidimensional case) (section 3.2.2).
Section 3.3 presents the approaches centered on the evolutions of the spatial
entities themselves. The point is to describe and explain (in the statistical sense)
change at the level of these entities (by integrating a number of explanatory factors
and various temporalities) and to represent the trajectories corresponding to their
evolution (section 3.3).
3.2. Following the evolution of the structure of spatial systems
The examples presented in this section give an account of the evolution of the
structure of a spatial system, whether concerning its spatial configuration, another
kind of structure, hierarchical, for example, or even the relations' structure between
the variables characterizing the entities of the system considered.
3.2.1.
Describing the evolution of space and hierarchical organizations
3.2.1.1.
The indicators measuring spatial organization
The objective is to implement methods to give account of the different “states”
of the spatial organization of the system studied in order to identify and measure the
different aspects of change. In some cases, very simple indicators allow us to show
