|
Using Imagery for Fungicide Application Year: 2003 Application Area: Agriculture BackgroundDry edible beans (DEB) are among the major crops grown in North Dakota and Minnesota. North Dakota is the nation's leading DEB producer, producing 31.16% of the nation's total for the years 2000, 2001 and 2002. Minnesota ranks fourth with 8.54% of the national total.1 In both states the largest concentration of DEB acreage is located in the Red River Valley. While optimal yields are sustained by pertinent field practices, adequate crop rotation, and favorable weather conditions, DEB are highly susceptible to many diseases. White mold, a fungal disease that develops in humid and rainy weather, is one of the most serious crop diseases in the Red River Valley, and can cause substantial yield losses. DEB are extremely vulnerable to disease if they become waterlogged at flowering time. The plants first develop watery rot spots on pods, stems, and foliages, then the decaying areas become covered by a cottony white fungal growth. The 2002 growing season was particularly wet in the Red River Valley. From June to August the region was hit by severe storms provoking devastating floods that heavily damaged crops. The standing water destroyed some young sprouts and caused some mature crops to turn yellow, develop root rot, die, or produce very little. Use of DataAfter heavy rain events in late July 2002, an eastern MN farmer had some navy bean fields severely damaged by flooding. The wet conditions also created a favorable environment for white mold infestation. To avoid, or at least try to minimize further crop losses from white mold, the farmer decided to preventively apply fungicide. However, he was not willing to spend money on unnecessary fungicide applications over field areas already irreversibly damaged by drown-out. The farmer created a fungicide spray map by first resampling a Landsat image to a 10 m pixel size (approximately the swath size of his sprayer). He then grouped the 10 m pixels of the resampled Landsat image according to their NIR reflectance values. In a last step he created a two class georeferenced spray map. The "spray" zones were areas with a normal navy bean vegetation development while the "non-spray" zones corresponded to the lost crop due to water damage. Economic and Environmental Benefits The producer was able to save $972.00 of fungicide cost by carefully selecting the areas to be treated. Furthermore, by not applying fungicide in the areas in which it was not necessary, he reduced the amount of chemicals introduced into the environment. 1) Link: Semi-monthly USDA totals (PDF) The farmer was able to save $972.00 of fungicide cost by carefully selecting the areas to be treated. |
Download print version (PDF)
![]() Figure 1) NIR Landsat; the blue-green pixels correspond to drown-out areas, the red to crops not damaged by water. ![]() Figure 2) Spray application map created from the resampled Landsat image. |