On the importance of landscape history for assessing extinction risk, Schrott (2005), Ecological Applications

Pretty interesting paper detailing the risk of extinction in various landscape situations over time, and showing the potential for lagged responses in population size. If there's a lagged response, then any population viability assessment could be quite erroneous if based on only the current situtation, and not looking at the past history. Makes sense. It's pretty theoretical, so there are several arbitrary deliniations for a viable population, threshold values, etc, but that's fine. Of course, there is a caveat that extinction thresholds are so different for individual species that a cookbook approach (i.e. how much habitat is "enough") is to be avoided- but then the whole paper is "strategic," in that it's meant to be applicable broadly... So when calculating suitable and unsuitable habitat, aren't you more or less calculating what is "enough?" Perhaps splitting hairs and being unfair to a theoretical argument, I suppose.

One problem is the way they calculate spread. Both the persistence and the "invasion" calculations are the same, i.e. the species has the same chance at persistence in suitable habitat as it does spread into neighboring suitable habitat. Work by Kendi has shown that in many cases, it's more difficult to spread then persist, and some sort of favorable conditions need to exist before a species spreads into neighboring habitats, even if that habitat would be perfectly suitable once the species arrives (like an activation energy barrier from chemistry or something), mainly due to competition, low initial population sizes, stochastic issues, etc.

While this is based entirely from the conservation viewpoint, there's no reason the same ideas wouldn't be directly applicable to invasions... for an invader to invade, it's got to avoid going extinct when colonizing. Makes sense for the bark beetle too- they are effectivly destroying their own habitat as they go along, changing the landscape structure and therefore altering the environment in which the next generation has to make a living. Finally, how do you do this assessment on a species intitially? If you haven't made repeated population and landscape assessments over the last 20 years, then this analysis doesn't seem to help, unless we can find some clever way to circumvent that problem...