Evaluation of a site-specific weed management strategy

Studies were conducted in 1997 and 1998 to compare conventional to site-specific weed management techniques. Field experiments were conducted in Knoxville, Tennessee. The study utilized a randomized complete block design with a split-plot treatment arrangement with nesting.

Preliminary research was conducted in 1996 to establish selective weed densities. Monocot and dicot densities were generated by applying selective herbicides. Weeds present included trumpetcreeper (Campsis radicans), common cocklebur (Xanthium strumarium), broadleaf signalgrass (Brachiaria platyphylla), and rhizome johnsongrass (Sorghum halepense). Field perimeters, plots, and weeds were mapped using global positioning systems (GPS).

In 1997 and 1998, weed density maps were used to determine atrazine applications in site-specific areas of the experiment. Site-specific areas determined to contain a high population of common cocklebur in 1996 received a Ix rate of atrazine. Low atrazine rates (0 in 1997 or l/3x in 1998) were applied to remaining site-specific plots. Conventional plots received a 1x rate of atrazine. Broadleaf signalgrass was uniform throughout the field therefore both conventional and site-specific plots received the Ix rate of alachlor.

Early postemergence (POST) herbicides included nicosulfliron and dicamba. Conventional and site-specific plots received a Ix rate of nicosulfuron to control broadleaf signalgrass. Conventional plots received a 1x rate of dicamba to control common cocklebur. Dicamba was applied at 0, l/3x, 2/3x, and Ix (0, 0.093, 0.19, and 0.28 kg ai/ha) to site-specific plots. These rates were determined by the density of common cocklebur at the time of POST applications.

For the 1998 field season, weed densities were counted using a 1 m²area for each sub-plot and recorded. A combine equipped with a yield monitor (AgLeader Yield Monitor 2000) and DGPS receiver (Trimble Ag 132) was used to harvest the corn. Yields were recorded using an AgLeader Yield Monitor 2000. Data were imported into ArcView for storage and generating yield maps.

Corn yields were similar between conventional and site-specific plots in 1997 and 1998 when averaged over treatments. Comparison of POST treatments within the field indicated a significant difference between conventional and site-specific plots. Corn yields were lowest when atrazine was not applied preemergence (PRE) and dicamba rates were 0 or 0.093 kg ai/ha.

These data demonstrated that site-specific weed management techniques may replace conventional techniques while maintaining yields. Site-specific weed management may result in reduced herbicide use, therefore lowering cost of weed control. Data indicates that a PRE herbicide was important in these weed control systems.