BackgroundThe 2002 growing season in the Red River Valley of the North was one of the "wettest years in recent history." From June to August several counties were hit by severe storms causing devastating floods that heavily damaged crops. During this period the area received 150 to 200 % above normal rainfall, with an intensity of 1 to 3 inches an hour.

As an example, Eldred, MN, received 11.36 inches of rain in July, of which 9.53 inches fell on July 9 and 10. The resulting widespread flooding destroyed young sprouts and caused mature crops to turn yellow, suffer from root rot, and die or produce very little. In places where crops were destroyed, replanting was not always possible because the saturated fields took too long to dry-out. In addition, water erosion jeopardized the prospect for late replanting of cash crops.

 Use of DataIn another case during the wet 2002 growing season, a more complex method was developed by an eastern MN farmer where maps outlining water damaged crops were created using imagery in a 4-step process. From the July 29 Landsat image a new layer was created where each pixel's value from the near infrared (NIR) band (fig. a) was represented by a point (fig. b). Each point belongs to one of the four classes defined by the farmer according to the range of the NIR reflectance values.

In the next step, the point layer was transformed in a new raster image (fig. c) with a 10 m pixel resolution. Every pixel value represents a range of NIR reflectance defined from the three first steps. In the last step, the 10 m raster image was classified to create a new water damage map (fig. d). Four classes were defined to depict the level of damage suffered by the crop: drown out, low, medium, and high, corresponding respectively to drown out vegetation, heavy, moderate, and no damage vegetation. The maps were used by the crop insurance agent to help determine the amount of financial compensation for the farmer.

 Economic and Environmental BenefitsThe produced maps not only show the location and surface areas affected by water, they also attempt to quantify the expansion and severity of crop damage. Based on actual information, these products are powerful tools for crop insurance decision makers. By increasing the accuracy of affected acreages and determining the severity of the phenomenon, financial compensations can be refined to better fit the reality of the actual event. It's also helpful in developing an even stronger trust between the producer and the crop insurance agency.