Skip to main content
ARS Home » Midwest Area » Urbana, Illinois » Global Change and Photosynthesis Research » Research » Publications at this Location » Publication #224309

Title: Living boundaries: tracking weed seed movement with non-dormant seeds

Author
item Davis, Adam
item LUSCHEI, EDWARD - UNIVERSITY OF WISCONSIN

Submitted to: Weed Science
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 11/18/2008
Publication Date: 12/22/2008
Citation: Davis, A.S., Luschei, E.C. 2008. Living Boundaries: Tracking Weed Seed Movement with Non-Dormant Seeds. Weed Science. 57(2):163-168.

Interpretive Summary: Synthetic seed banks are a useful tool for increasing precision of counts in studies of weed population dynamics. By sowing a known number of seeds within a spatially well-described area, an investigator can greatly reduce uncertainty in initial population size and seed location. Here we describe a method for providing a living boundary to synthetic seed banks in weed population dynamics experiments managed with real-world farming practices. Study quadrats in East Lansing, MI, and Arlington, WI, were sown with either giant foxtail or velvetleaf at rate of 2000 seeds m-2 .The study quadrats were marked on the perimeter and diagonals using non-dormant seeds of three marker species, kale, radish and rye. The areas were then tilled and planted to corn. Seedling locations for the marker and weed species were obtained through digital image processing. Marker species tracked the weed species very accurately. The results suggest a simple rule for applying the method to field studies of weed population dynamics: after the original quadrat is subjected to agricultural practices and seedlings have emerged, flag the shape containing all marker seedlings to obtain the expanded quadrat containing the study weed population.

Technical Abstract: Synthetic seed banks are a useful tool for increasing precision of counts in weed demography studies. By sowing a known number of seeds of a single accession within a spatially well-described area, an investigator can greatly improve the signal to noise ratio, relative to the ambient weed seed bank, by reducing uncertainty in initial population size and minimizing the biological variability that hinders inferences about population processes. The “pulse” does not, however, eliminate the problems as any field operation moving soil will reintroduce uncertainty in population size and composition unless the area encompassing the “pulsed” seed can be followed in time. Here we describe a method for providing a living boundary to synthetic seed banks in demography experiments subject to field operations. Study quadrats in East Lansing, MI, and Arlington, WI, were sown with either giant foxtail or velvetleaf at rate of 2000 seeds m-2 .The study quadrats were marked on the perimeter and diagonals using non-dormant seeds of three marker species, kale, radish and rye. The areas were then subjected to tillage and planting operations. Spatial coordinates of seedling locations for the marker and weed species were obtained through digital image processing. A non-parametric comparison of the spatial displacement of marker and weed species indicated that their empirical spatial distributions did not differ. Percent containment of weed species by marker species reached 100% within the quadrat described by the 50th, 90th and 99th quantile of movement of the marker species for velvetleaf in WI, velvetleaf in MI, and giant foxtail in MI, respectively. The results suggest a simple rule for applying the method to field demography studies: after the original quadrat is deformed and seedlings have emerged, flag the polygon containing all marker seedlings to obtain the expanded quadrat containing the study weed population.