Just by looking at the Choropleth map, one can begin to see the disparity in population trends by looking at the Parallel Coordinate Plot (PCP) below along with the Lab 5 information. When looking at the Lab 5 choropleth map for instance, it is readily apparent that the central part of the island has historically been the residence of the majority of the island of Martinique. However, what this map reveals is that there is large population growth in the Central and Southern parts of the island while the communes in the North have a static or slowly decreasing population. These larger patterns could be attributed to some common causes but what is a little more unexplainable is that small fishing villages such as Case Pilote have in some cases more than doubled in size.
Clearly what these patterns actually mean and what there numerous causes might be is not an easily answerable question. There are many factors which are likely playing a part in these patterns which more detailed statistical and census data would help to see. From visiting these areas in person a few things become clear in regards to the larger patterns: firstly, the areas of low population are often at the far north and south of the island as noted in the Lab 5 Map. These are areas which contain high levels of either tropical forest or mangroves which make large scale habitation difficult due to remoteness, lack of supplies, and the harsh qualities of the climate. The north in particular is home to the island's volcano Mt. Pelee which has erupted several times throughout history, decimating the living space around it and so it is also likely that people are moving away from it as an eruption becomes more and more likely for the near future and has been predicted to be due soon. In addition to this, almost all of the land on the island has remained within families for generations and is a source of pride for some which helps explain in part why so few people are moving around, particularly in the more remote areas such as the North. Some of the growth can likely be attributed to common causes such as people moving out of overpopulated cities (St. Joseph) or visitors or businesses relocating to the more active tourist regions (Ducos, Riviere-Salee, Trois-Ilets, and Le Diamant).The large growth in Lamentin specifically could be explained by the fact that it was a mangrove which was "filled" thus recently become habitable and is today an area where industry is growing rapidly. These of course are not the only reasons why these changes might be happening, but they are plausible potential explanations one could think of without more readily available information.
This evidence is of course anecdotal and small scale. To really discover what is happening on the Island of Martinique, quantitative and qualitative data should be collected over long periods of time. Unfortunately, population counts are some of the only data that is available by commune currently with all other data from the French Overseas departments being on an island by island basis which is certainly good for comparing islands but this makes understanding intra-isle economy, crime, and other sociological phenomena very difficult to study.