Field experiments were conducted for 3 yr to determine the effect of various biological and physical factors on the operation of the weed-sensing Detectspray system. Plant detection is achieved by sensors measuring differential reflectance of red and near-infrared wavelengths of light from green plants, crop residues, and soil. Weed detection was greatly reduced 70 to 80 min after sunrise and before sunset when operated at lat 50°N because of reduced solar irradiance. Tall, dense-standing crop stubble limited detection of small weeds at the soil surface. Weed detection varied with plant species. Canola with three to four leaves consistently was detected, but wheat or green foxtail usually required five to six leaves to be detected. Small weeds were detected if present at densities greater than 70 plants m−2. Growers and commercial applicators need to be aware of the limitations of the Detectspray system to use it effectively to control weeds with concurrent reductions in herbicide use.

Detectspray™ weed-activated sprayer (WAS) technology enables individual nozzles on a sprayer to turn on and off in response to the presence of green vegetation. WAS operated full season at two fallow sites in North Dakota reduced POST herbicide requirement 47 to 88%. Full-season WAS applied as needed at West Fargo reduced cost to the farmer by only 13% compared with broadcast application with a farmer-owned sprayer and assuming free labor. Winds of 8 to 16 km/h apparently moved spray off target, thereby necessitating retreatment and increasing the relative costs of WAS at Fargo. The percentage of sample quadrats having greater than 3% vegetative cover correlated positively (r = 0.78 to 0.80) with percentage reduction in spray volume achieved by WAS.

Field studies were conducted from 1993 through 1995 to determine potential reductions in herbicide use and associated cost savings by utilizing a weed-sensing sprayer, named Detectspray, to control weeds throughout the fallow season and to control perennial weeds after crop harvest. The Detectspray system gave comparable weed control to conventional broadcast spraying on 80% of the application dates and reduced glyphosate/dicamba use over the fallow season by 19 to 60%. This reduced herbicide use resulted in cost savings of $6 to $32 ha−1. A fallow treatment that combined two herbicide applications with the Detectsprayer plus one to two wide-blade tillage operations was less costly than conventional tillage at two of 11 sites and at all sites retained more surface crop residues to reduce the risk of erosion. Postharvest glyphosate use on quackgrass with the Detectsprayer was reduced 50 to 78% compared to broadcast applications and resulted in cost savings of $16 to $25 ha−1. Clopyralid use on Canada thistle with the Detectsprayer was reduced 71 to 80%, with cost savings of $44 to $50 ha−1. The Detectspray system is a useful tool to effectively manage weeds in conservation fallow and reduced tillage cropping systems.