Generally, social scientists look at migration as a combination of both "push" and "pull" factors. Push factors are negative inducements compelling someone to leave their place of origin (natural disasters for instance), while pull factors are positive factors of a destination enticing someone to move there (greater job opportunities, for instance). When studying migration, a researcher needs to understand what combination of pull and push factors are at work compelling movement of human populations.

For example, take the mass outmigration of Ancestral Puebloan people from the Four Corners region of the U.S. between about A.D. 1270 and 1300. Over the course of these 30 or so years, the Four Corners went from the highest density concentration of people in the Southwest (other than southern Arizona) to being almost completely uninhabited.

One of the primary push factors was in fact environmental degradation. I've written about this before here, but to summarize, several environmental factors contributed to making corn agriculture in the Four Corners region nonviable in supporting the large populations that existed there.

These are summarized from Ahlstrom and Dean 1995:

• The "Great Drought" between 1276-1299

• Lower average temperatures

• Changing patterns of precipitation (so unpredictable wet seasons)

• Dropping water tables and river channel degradation (extreme cutting of river embankments) reducing groundwater availability

• Increased temporal and spatial variability in wild resources that supplemented farming activities and guarded against poor harvests

All these factors combined around 1270 to make continued subsistence extremely difficult. Not only would the environmental situation work as a push factor, but all the attendant social changes. Resource stress in a densely inhabited area (close to the carrying capacity of the environment) means increased conflict between groups (which we have some archaeological evidence for). It also likely means decreasing faith in the ability of existing institutions (social and religious) to maintain social order. All these things in the case of the four corners are significant push factors.

As for pull factors, the people in the four corners were not isolated, and were in fact part of an extensive web of social relationships that crossed much of the U.S. Southwest. Because of this, many likely had significant social relationships that tied them to groups outside the four corners in areas that had more stable environmental conditions, like in the areas of modern day Zuni and Hopi Pueblos along the upper Little Colorado river in Arizona and New Mexico, or along the Rio Grande river in New Mexico. The combination of more stable water regimes and existing social connections (relatives or trading partners, for example) meant these places provided an attractive alternative to continued living in the Four Corners region.

In other words, by A.D. 1270, a number of common push factors (environmental instability, violence, and failure of social institutions) combined with the pull factors of greater environmental and social stability along the Rio Grande and the Mogollon Rim (including the Little Colorado) to create a massive outmigration from the Colorado Plateau. Importantly, we also have to consider the impact of existing social relations. A significant pull factor that makes a spot attractive to migrate to is existing social connections with the locals, whether that is familial relationships, trade partnerships, or just shared religious beliefs.

Finally it is important to consider the social aspect of these kinds of migrations beyond just environmental concerns (even if those are the root cause). For example, in the Four Corners example, we see a nearly complete outmigration in that 30-year timeframe, yet we have good paleoenvironmental data to suggest that a significant portion of the population could have remained and done well for themselves as farmers. Why the discrepancy? Clearly there is a social factor, on top of the environmental. Consider the impact on a society of a large portion of people migrating away. A lot of social connections would be severed, and so to maintain those connections, you need to follow the group (or find some other method for maintaining that connection). You could consider this as a push factor - the ghost-town effect, if you will - or as a pull factor - having a significant number of your relations living elsewhere.

Source

Ahlstrom, Richard V.N., Carla R. Van West, Jeffrey S. Dean. 1995. Environmental and Chronological Factors in the Mesa Verde-Northern Rio Grande Migration. Journal of Anthropological Archaeology, 14(2): 125-142.