Agricultural crops can sustain extensive damage caused by Canada geese (Branta canadensis) when these crops are planted near wetlands or brood-rearing sites. From 2000 to 2015, South Dakota Game, Fish and Parks spent >$5.6 million to manage damages caused by Canada geese to agricultural crops (primarily soybeans) in South Dakota, USA. For the purpose of developing a repellent application strategy for nonlethal goose damage management, we comparatively evaluated the width of anthraquinone applications (i.e., 9.4 L Flight Control^® Plus goose repellent/ha [active ingredient: 50% 9,10-anthraquinone] at 0–36 m versus 0–73 m perpendicular to the edge of wetlands in 2014), the timing of the first repellent application (i.e., 9.4 L Flight Control Plus goose repellent/ha at 50% versus 75% seedling emergence in 2015), the yield of soybeans (Glycine max) within repellent-treated and untreated subplots, and anthraquinone chemical residues in Day County, South Dakota. Soybean yield was greater in subplots 73 m from the water’s edge than that in the 36-m subplots (P < 0.02). Among subplots first sprayed at 50% seedling emergence, soybean yield was greater at 73 m and 82 m than that at 36 m (P < 0.005). In contrast, we observed no difference in yield at 36 m, 73 m, or 82 m in the subplots first sprayed at 72% seedling emergence (P > 0.09). We therefore conclude that goose damages were effectively managed in subplots first sprayed at 72% seedling emergence. Anthraquinone residues averaged 674 and 629 ppm anthraquinone upon the first application of the repellent (June to July), 22 and 35 ppm anthraquinone in the mid-season hay (August to September), and 36 and 28 ppb anthraquinone in the harvested seed (October to November) in 2014 and 2015, respectively. Our results suggest that a 73-m bandwidth of anthraquinone-based repellents first applied at approximately 72% or 65–85% seedling emergence can protect soybeans from Canada goose depredation.