The Journal of Ecology Editors are delighted to announce that Helen Alexander is our Eminent Ecologist award winner for 2019. In recognition of her work, we have asked Helen to put together a special Virtual Issue of some of her excellent contributions to the journal. Helen has written a summary of her career, research and personal experiences. If you missed part 1, you can find it here. Below is the rest of Helen’s story. Happy reading!
Theme 2 – Seed and seed bank ecology
Seeds are, of course, an important stage in the plant life cycle and their patterns of dormancy and germination have important ecological and evolutionary implications (Baskin & Baskin 2014). My interest in seeds started while I was still in graduate school. Renata Wulff, a talented Venezuelan scientist, had the idea to explore effects of maternal environments on seed properties. This led to the seventh chosen paper, Alexander & Wulff (1985). We found that maternal temperatures altered seed germination in Plantago lanceolata, and that half-sib families differed in their response. Intriguingly, effects of maternal treatments affected seed germination into a second generation. In Kansas, my research has focused on the well-known phenomenon that many seeds persist in the soil (i.e., seed bank) for years. The contribution of the seed bank to the dynamics of above-ground plant populations is, however, poorly understood. In my eighth chosen paper, I addressed this question with an undergraduate student, Anne Schrag, using the annual sunflower, Helianthus annuus (Alexander & Schrag 2003) (Fig. 5). We took data on seedling production under large isolated sunflower plants where we had removed all flowers (and thus any seedlings must have been derived from the seed bank) or under control plants without manipulation (seedlings could come from either recently produced seeds or the seed bank). From 10-23% of seedlings under individual plants came from the seed bank. We also found that seedlings under the manipulated plants occurred at lower density and at maturity produced ~4 times more inflorescences than control plants. To our knowledge, few have examined these kinds of density-dependent processes that could enhance the effective contribution of dormant seed to the plant population.
Most seed bank studies are done on spatial scales of centimeters to meters. In Alexander, Pilson et al. (2009) my ninth chosen paper, we addressed seed banks on a landscape spatial scale. Following approaches developed with Silene latifolia (Antonovics 2004), we set up roadside surveys in northeast Kansas and southeast Nebraska. Each fall, when flowering sunflowers are very visible (Fig. 6), we drove 24 km (Kansas) and 19 km (Nebraska) routes and counted the number of sunflowers on both sides of the road in 80 m units. Despite flat tires and the occasional curious dog, we repeated these surveys for 6 years in Kansas and 4 years in Nebraska. These data allowed us to document geographical and temporal variation in plant occupancy, colonization, and extinction. Seed banks were clearly important: the strongest evidence comes from a year following a spring drought in Nebraska, when 100s to 1000s of plants were found in units that lacked plants the year before. Besides their ecological interest, seed dormancy and seed banks in H. annuus are important in evolutionary questions as well, such as questions about the degree to which crop sunflower alleles persist in wild sunflower populations (Alexander, Emry et al. 2014).
My tenth chosen paper, Alexander, Foster et al. (2012), was part of Journal of Ecology’s Centenary Symposium series. I was invited to explore the challenging topic of “metapopulations and metacommunities,” or more generally the importance of spatial processes in plant ecology. My coauthors and I tackled this problem from multiple approaches. Ford Ballantyne and I used the sunflower roadside survey data described above to build spatial models of plant populations with and without seed banks. We found local population processes and seed banks were crucial for landscape level patterns of plant abundance. Janis Antonovics used 13 years of data on roadside populations of Silene latifolia to estimate rates of long distance colonization by seed, a challenging phenomenon to study. Finally, my community ecologist colleagues (Cathy Collins, Bryan Foster, and Bob Holt) analyzed data from a long term experiment on landscape fragmentation and succession. They found that habitat connectivity and constraints on seed dispersal affected how the species composition of plant communities changed over both space and time. Overall, we concluded that ignoring spatial properties of plant individuals, populations, and communities risks discounting important processes; consideration of “space” is needed at a wide variety of spatial scales, from local (Mack & Bever 2014) to landscape (Collins, Banks-Leite et al. 2017).
Theme 3 – Ecology of long-lived herbaceous plants and the challenge of “detectability”
Like most plant population biologists, my experimental organisms are often short-lived: annual plants are far easier for studies of population dynamics or for experimentation than species that live for decades. However, long-lived perennial plants dominate most landscapes, including the tallgrass prairies of my home state of Kansas. My last chosen paper, Alexander, Slade et al. (2009), focused on Mead’s milkweed (Asclepias meadii), a long-lived perennial that is an indicator of high quality remnant prairie (Fig. 7). Given its rare status, Dean Kettle, Research Programs Director of the University of Kansas Field Station, asked me to examine multi-year survey data to address what appeared to be a simple question: is the plant population increasing, decreasing, or staying constant in numbers over time? These data were, however, confusing. Plants present in some years seemed to disappear in other years, and then reappear. With the important help of mammologists Norm Slade and Aaron Reed, we realized that our ability to find plants (i.e., probability of detection) was less than one — the same challenge typically encountered in studies of mobile small mammals. Low plant detectability can occur for many reasons, including observer error (vegetative plants are hard to find in thick prairie vegetation, Fig. 7). Further, vegetative dormancy and herbivory can mean that no above-ground plant parts are visible. To address this problem, we estimated plant survival and population size by analyzing 14 years of survey data with mark-recapture approaches. Additionally, repeated surveys within a single year allowed us to estimate the degree to which ‘lack of detection” is due to observer error, herbivory, or dormancy.
Our long-term data on Mead’s milkweed was included in a recent synthesis on ecological and evolutionary correlates of vegetative dormancy in plants (Shefferson, Kull et al. 2018). Furthermore, our studies of incomplete detection have contributed to a (slowly) increasing recognition that plant ecologists need to address “low detectability” (Kellner & Swihart 2014). Alas, detection problems with plants occur frequently because plants of different sizes and species are more or less likely to be found by human observers, and certain habitats with dense vegetation are challenging for surveys. Using mark-recapture approaches can correct for otherwise serious errors that incomplete detection introduces into estimates of population size, survival rates, and biodiversity.
A personal byproduct of this award was the opportunity to reread papers that I had written long ago and hadn’t looked at for decades. Thanks again to Journal of Ecology for not only choosing me for this award but also their hard day-to-day work of producing a quality scientific journal. As I reflect on my research, it is clear that many projects have shared features: my papers largely address studies of plants in field environments, many report on field experiments, and several studies spanned multiple years. Given these characteristics, I close by emphasizing that much of my work has been done at field stations, specifically the University of Virginia’s Mountain Lake Biological Station (MLBS) and the University of Kansas Field Station (KUFS). Field stations such as MLBS and KUFS provide diverse environments, places where one can securely carry out long term studies, and an excellent community of researchers and staff. I’m glad that I’ve had the opportunity to work at these high quality centers of research, education, and stewardship of natural lands.
Helen Alexander, The University of Kansas, Kansas, USA.