Weather, soil and microbial interactions determine seed germination and seedling demographic bottlenecks in Chihuahuan desert shrubs

Caroline R. Toth, Plant & Environmental Science, New Mexico State University, Las Cruces, NM, Niall P. Hanan, Plant and Environmental Sciences, New Mexico State University, Las Cruces, NM, Darren James, Jornada Experimental Range, USDA - Agricultural Research Service, Las Cruces, NM and Steven R. Archer, School of Natural Resources and the Environment, University of Arizona, Tucson, AZ

Background/Question/Methods
Shrub encroachment is threatening the productivity of southwestern rangelands by diminishing herbaceous productivity, with potential to negatively impact the sustainability of grazing-supported economies. During the past century native shrubs have expanded in former grasslands to become the dominant species of the Chihuahuan Desert and limiting herbaceous growth and reestablishment through direct competition and indirect positive feedbacks. At the Jornada Basin Long Term Ecological Research (JRN-LTER) site the expansion of three shrub species, creosotebush (Larrea tridentata), honey mesquite (Prosopis glandulosa) and tarbush (Flourensia cernua), has occurred with varying timing (over ~150 years) and distinct spatial patterns associated with soil types. However, while the history of shrub expansion and contemporary distribution patterns are well-documented, the specific mechanisms leading to temporal sequencing and spatial filtering among species are unknown. In this study, we explore the ontogenetic processes relating to soil type, microbial interactions and precipitation that may favor (demographic release) or suppress (demographic bottlenecks) seed germination and seedling survival. A four-stage experiment measured the germination, survival, and growth responses of these three shrub species during the critical and most vulnerable early life stages, from germination to initial establishment and growth (< 3 months post germination).

Results/Conclusions
Bottlenecks to establishment of the three primary shrubs that have proliferated on the JRN-LTER grasslands differ by species. Under current conditions, tarbush establishment is limited by extremely low seed viability (<1%) associated with self-incompatibility during pollination. More research is needed to establish under what conditions tarbush produces viable seed sufficient to explain its historical expansion. Honey mesquite and creosotebush require distinct seed scarification treatments, correlated with dispersal mechanisms (herbivory and wind, respectively), to break dormancy. Following scarification of mesquite and creosotebush, we found little evidence for intraspecific or interspecific allelopathy associated with soils or microbial communities associated with those species. However, we did find evidence that germination and survival of mesquite and creosote are suppressed when exposed to tarbush soil microbial communities. Early seedling establishment (<24 days post-germination) of both honey mesquite and creosotebush is favored by wetter than average conditions. However, in older seedlings (24-75 days), honey mesquite survival is enhanced by wetter than average conditions, whereas creosotebush survival is optimal under average conditions but reduced under wetter than average conditions. Understanding the varying causes of demographic release and demographic bottlenecks is critical for understanding historical shrub encroachment/expansion processes and predicting likely future trends.