|Liu, Jianyang - UNIVERSITY OF ILLINOIS|
|Harrison, Kent - THE OHIO ST UNIVERSITY|
|Regnier, Emilie - THE OHIO ST UNIVERSITY|
Submitted to: Weed Science Society of America Meeting Abstracts
Publication Type: Abstract Only
Publication Acceptance Date: January 19, 2009
Publication Date: July 20, 2009
Citation: Schutte, B.J., Liu, J., Davis, A.S., Harrison, K., Regnier, E. 2009. The Association Between the Earthworm Lumbricus terrestris and Giant Ragweed (Ambrosia trifida) in Agricultural Fields Across the Eastern U.S. Corn Belt. Weed Science Society of America Meeting Abstracts. 48:225. Technical Abstract: Previous research indicated that secondary seed dispersal by the earthworm Lumbricus terrestris can improve giant ragweed seed survival and influence seedling spatial structure at the quadrat (m2) scale. Here, we examine the association between L. terrestris and giant ragweed at plant neighborhood, transect and regional scales. Our specific objectives were: 1) compare nearest neighbor distances and neighborhood densities between midden and nonmidden seedlings, 2) determine the L. terrestris-ragweed association’s strength at the quadrat and transect scales in agricultural fields across the eastern U.S. Corn Belt, and 3) conduct exploratory analyses to identify potential causes of strength differentials among sites. The L. terrestris-ragweed association was studied in no-till soybean fields across Illinois, Indiana, and Ohio during the spring of 2008. Seedling neighborhoods were quantified using two-dimensional images of 1-m2 quadrats, which were placed in areas of L. terrestris-giant ragweed co-occurrence. "Midden seedlings" were located within 10-cm diameter circles centered on earthworm burrow openings. Seedlings outside of burrow-centered circles were classified as "nonmidden" seedlings. Except for seedlings 15 cm from quadrat edges, neighborhoods were described by first and third nearest neighbor distances, and neighborhood densities at 5, 10, and 15 cm. Quadrat images were also used to calculate quadrat level strengths of association by dividing percent midden emergence by percent of area occupied by burrow-centered circles. For transect level analyses, two 23-m transects were positioned in areas abound with giant ragweed. At 0.9-m intervals, the presence or absence L. terrestris middens and giant ragweed seedlings in 0.04-m2 quadrats was noted. The proportion of quadrats in which middens and seedlings co-occurred relative to the number of quadrats containing seedlings provided a measure of strength at the transect scale. First neighbor distances were similar between seedling types, but third nearest neighbor distances for midden seedlings were less than those for nonmidden seedlings. Midden seedling neighborhoods at 5 and 10 cm were denser than nonmidden seedling neighborhoods. These results indicate that midden and nonmidden seedlings experience different competitive environments, which may impact subsequent plant growth and development. Quadrat level strengths of association ranged from 0 to 30 and varied among sites. Regional variation in quadrat level strength of association was negatively associated with longitude, latitude, and air temperatures the previous October. Transect level strengths of association ranged from 27 to 100 and varied among sites. Regional variation in transect-level strength of association was positively associated with total precipitation during the previous autumn (October to December). Because the L. terrestris-ragweed association represents an antagonistic force to post-dispersal seed predation, influential factors at the regional scale may indicate where seed predation-based integrated weed management tactics will be most effective.