A final technique I have developed for analyzing population movement over time and space is the population density change map. Examples are Figures 5.6, 6.11, and 7.8. In section 4.2.3, I described how a population density dataset is generated. It consists essentially of a grid within which each square has a total population index value. A population density change dataset is generated by subtracting one of these population density datasets from another. That is, a new grid is generated within which the value of each square is equal to the difference between the values of the same square in the two population density datasets.
The resulting dataset is then used to produce a surface in the same manner as was done with the population density datasets.
![]()
where
= grid square value in population density change dataset
= total population index of same grid square in phase
= total population index of same grid square in previous phase
The population density change map is actually a very useful tool for visual inspection of changes in population across space and through time. For example, Figure 5.6 is a settlement density change map for the Early Formative 2. The value of the z axis is the grid square population index value for the EF2 phase minus the population index value of the same grid square in the EF1 phase. It is immediately apparent that most sites in this period experienced healthy growth, but that sometime late in the EF1 phase (before the EF2 phase began) three sites experienced population declines. Two of these turn out to have been abandoned (the ones near the tip of the peninsula), while one fissioned into at least two separate sites (the one that lost the least population, roughly in the middle of the peninsula). This example is discussed more thoroughly in Chapter 5 . My intention here is only to show the usefulness of the population density change map. Another very striking example is Figure 9.5, which eloquently captures the wholesale depopulation of the peninsula in the Early Pacajes phase.