Past research

A portion of Figure 1 in Godsoe et al. 2021.	Estimating biodiversity change: Current estimates suggest strikingly slow change in traditional measures of biodiversity, but these estimates seem to contradict rapid shifts in the abundance of individual species observed in nature. Conceptual models have emphasized the role of competition among species, but it's to quantify how these mechanisms contribute to biodiversity change. For instance, in some cases strong competition or selection on species identity leads to no biodiversity change. In view of this disconnect, we develop a new approach to studying biodiversity change using the Price equation. We show that biodiversity change responds to selection on species’ rarity, rather than to either competition or selection on species identity. We then show how this insight can be used to quantify the effects of the mechanisms previously thought to influence biodiversity: (1) selection, (2) (ecological) drift, (3) immigration and (4) speciation. Our results suggest the connection between species’ fates and their rarity is fundamental to understanding biodiversity change. doi
Sampling pollen presentation on Lobelia siphilitica. Photo: K. Eisen.	Pollen presentation: Plants have evolved many mechanisms that promote pollen export and transfer via animal pollinators. One such mechanism is pollen dispensing, where plants regulate the amount of pollen available to a pollinator during a single visit. But in order for pollen dispensing to evolve, there must be variation in dispensing schedules within populations, and this variation must have a heritable component. We measured variation in pollen dispensing schedules for ten open-pollinated maternal families within a population of Lobelia siphilitica (Campanulaceae). We found that L. siphilitica flowers dispensed an average of 8.76 % ± 1.37 % of available pollen during any visit, which is low relative to other measured species. We found that the repeatability, which places an upper limit on the heritability of a trait, was 0.29, which suggests there could be constraints on the evolution of constant or increasing proportional pollination schedules. doi
A piece of the phylogenetic distribution of gynodioecy on the cover of IJPS.	Mating system evolution: Gynodioecy is a dimorphic sexual system where individual plants are either female or hermaphroditic. Because gynodioecy has been documented in ≪1% of plant species, it is possible that gynodioecy can evolve and persist only under a restrictive set of conditions. To identify the phenotypic traits and ecological factors that are associated with gynodioecy, we assembled an angiosperm-wide database of gynodioecious species and used this database to test whether gynodioecy was associated with two phenotypic traits/ecological factors: an herbaceous growth form and a temperate geographic distribution. We found that gynodioecy was associated with an herbaceous growth form and a temperate distribution, suggesting that herbaceousness and temperateness can facilitate the evolution and/or persistence of gynodioecy. In addition, our results suggest that gynodioecy is associated with different phenotypic traits/ecological factors than dioecy, a closely related sexual system. doi
The Lyford Grid permanent plot at Harvard Forest. Photo: A. Barker Plotkin.	Forest ecology: The majority of forested ecosystems in the northeastern U.S. are second-growth stands that established following the abandonment of agricultural fields in the mid-to-late 1800s. Repeated measurements of these forests provide an opportunity to assess how change in northeastern forests impacts the globally significant carbon sinks documented in the region; the capacity for storage may change over decades to centuries with stand-level changes in age, species composition, and structure, and these long-term processes are not well represented in current models of carbon sequestration. Using 42 years of census data for > 6,000 individual trees in a 2.9-ha permanent plot at the Harvard Forest (Petersham, MA, USA), we found that the biomass of live trees increased linearly, confirming that this ca. 110-year-old stand is still in the aggradation phase of stand development. Quercus rubra L. (red oak) accounts for 80 percent of the increase in aboveground biomass due to the rapid growth of dominant stems and low canopy mortality rates, and we predict this dominance will continue over the next century, in the absence of major canopy disturbance. doi