Can population structure be a useful proxy for projected population growth? A meta-analysis of herbaceous plant populations

Background/Question/Methods

Determining whether a plant population is expected to grow, remain stable or decline is often the first step in designing management actions for species of conservation concern, such as threatened and endangered species or species harvested from the wild. While modeling population dynamics is the preferred method for evaluating population status, developing population models can be resource-intensive, requiring both time and expertise. In practice, population structure – which can be assessed more easily and rapidly – is often used as a proxy for population status, with the assumption that a greater proportion of small individuals relative to reproductive adults is indicative of a growing population. However, previous studies of the relationship between population structure and growth rates of tropical tree species have found only weak relationships between the two, concluding that size distributions are not a good predictor of future population trends. The relationship between structure and growth remains to be evaluated in other plant life forms. Here, we ask if population structure is predictive of projected future population growth for herbaceous plant species. We use a meta-analysis of data from 78 herbaceous plant species to evaluate the relationship between observed population size distributions and projected population dynamics from matrix projection models.

Results/Conclusions

We found moderate correlation between short-term projected population dynamics (1-year transient growth rates) and population structure (r = 0.32, p<0.01). However, populations with a greater proportion of large, and likely reproductive, individuals were expected to grow more in the next year, which is the opposite of the generally assumed relationship between structure and growth. Across longer time periods (5- and 10-year transient dynamics), the strength of the correlation between structure and projected growth rates declined.  We found only a very weak correlation between population size distributions and long-term, asymptotic projected population growth (r = 0.09, p = 0.047). At this timescale, populations with a greater proportion of small individuals did tend to have higher projected growth rates. While this pattern is consistent with the assumed relationship between structure and growth, the low degree of correlation shows that population structure is only a poor indicator of projected dynamics. Our results suggest that caution is needed in using population structure as a proxy for population dynamics in guiding management decisions for herbaceous plant species.