Understanding the relative importance of density-dependent and density-independent factors in driving population dynamics is one of the oldest challenges in ecology, and may play a critical role in predicting the effects of climate change on populations. We used long-term observational data to describe patterns in plant population regulation for 57 forb and grass species from three different ecosystems (arid desert grassland, semiarid sagebrush steppe, and subhumid mixed-grass prairie). Using a hierarchical partitioning approach, we (i) quantified the relative influence of conspecific density, heterospecific composition, and climate on temporal variation in population growth rates, and (ii) asked how the relative importance of these drivers depends on site aridity, species growth form and life expectancy, and abundance and spatial patterns. The data from one of the sites in this analysis are presented in one of the chapters of this thesis. We found that density-dependence had the strongest effect on species. Climate often had a significant effect, but its strength depended on growth form. Community composition rarely explained significant variation in growth rates. The relative importance of density, composition, and climate did not vary among sites, but was related to species' life histories: compared to forbs, grasses were more sensitive to climate drivers. Abundance and spatial clustering were negatively correlated with the importance of density dependence, suggesting that local rarity is a consequence of self-limitation. Our results show that interspecific interactions play a weaker role than intraspecific interactions and climate variability in regulating plant populations. Forecasting the impacts of climate change on populations may require understanding how changes in climate variables will affect the strength of density-dependence, especially for rare species.