Published online by Cambridge University Press: 20 January 2017
Weed control strategies based on conserving crop yields rather than preventing weed seed production may result in increased weed densities and management costs over the long-term, particularly in less competitive crops such as tomatoes. The effect of crop, tillage, and duration of weed control on weed seed bank size and composition was examined from spring 2001 to spring 2003 near Lafayette, IN. Main plots in 2001 and 2002 contained soybean or tomato planted in rotation (soybean-tomato, tomato-soybean). Subplots were managed with either conventional or no-till practices. Sub-subplots contrasted threshold strategies in which weeds were either controlled for four to six weeks (period threshold, PT) or throughout the growing season (no-seed-threshold, NST). Seed banks were sampled annually in the spring. Emergent weeds were counted` at four and twelve weeks after planting (WAP) in 2001 and 2002. Weed seed banks did not significantly change in the NST plots in any year. However, seed bank densities increased substantially following tomatoes in PT plots. In contrast, weed seed bank densities decreased following soybeans in PT plots. The difference in seed banks and emergent weeds between soybean and tomatoes could be attributed primarily to greater suppression of giant foxtail by the soybean canopy. Giant foxtail control was greater in PT soybeans than in PT tomatoes in both years and giant foxtail comprised most of the PT tomato seed bank in 2002 and 2003. Tillage did not affect weed seed banks in any year. This study highlights the need to control later emerging weeds in tomatoes to prevent large increases in the weed seed bank.