(Open Management Zones folder in Store to access more background information.)
Landsat Yield Quantity & Variability (Productivity) Management Zones
(based on Method 4 Productivity Zones from Kleinjan et al., 2006; [pdf])
Yield Quantity & Variability Zones based on Landsat by GIS Ag Maps (the example below corresponds to corn and soybeans, but zones can be developed from many different crops). GIS Ag Maps carefully looks through the archive of Landsat imagery from the most recent season to about 10 years in the past (and also looks at other historic imagery to see if the field area has changed) and, based on years of experience, only uses proper imagery to make zones. Preferably, there is imagery for two normal, one wet, and one dry weather season (though this ideal combination is not always available). Imagery is combined with proprietary methods for different seasons and crops and a zone map is produced to the field extent. (See the Recommended Zones page for maps and information from Kleinjan et al. [2006] describing the zones).
Above graphics from left to right (1-4, respectively) are: 1) Landsat-based high (green) and low (red) yield amount (based on many seasons); 2) Landsat-based high (red) and low (green) yield variability; 3) Pixels are divided into a combination of relatively high and low areas; symbology is as follows (this method is based on Method 4 from Kleinjan et al. [2006]; see below for reference); High yield and low variability (dark green); High yield and high variability (light green); Low yield and high variability (orange); Low yield and low variability (red); 4) Zone map to field extent based on pixels. (Methods to remove outlier effects are applied in process.)
Text from Kleinjan et al. of (in order): Method 4 zones, Summary and Conclusion
Method 4 (recommended zones): "In method 4, both yield and standard deviation data are used to identify four different productivity zones. Both yield and standard deviation were split into two categories, above and below the average value. Combining these categories resulted in four productivity zones with the following characteristics: 1) high yield, high deviation, 2) low yield, high deviation, 3) high yield, low deviation, and 4) low yield and low deviation are created." Kleinjan et al. (2006)
(In Kleinjan et al. the publication, there are three other types of zone researched, which are: Method 1, which are created from average yields separated the field into areas with low, medium, and high yields; Method 2, which are created from yield standard deviation separated the field into areas with low, medium, and high yields; Method 3, which are created from yield coefficient of variation (CV) separated the field into areas with low, medium, and high yields. Access the full article above for more details)
Summary
"Method 4: A better understanding of how the average yields and standard deviations in a field interact may be attained if both factors are used to develop zones. In the example field, areas with both high yields and high standard deviations were found in footslope areas that have occasional problems with excess moisture. These areas produce high yields in dry years and low to average yields in wet years. These areas are good candidates for installing tile drainage…provided drainage is allowed by the NRCS or other authorities. Selecting yield goals for these areas is difficult due to the large variability. Footslopes areas in dry years produce corn yields greater than 200 bu/A in this field, whereas yields can be as low as zero/A in wet years. Areas with low yield and high deviation are typically toeslopes that are extremely wet or in compacted regions surrounding wet field areas. High variability observed in these areas may be controlled by limiting compaction, installing drainage, or planting extremely low areas to grass waterways. Field areas with high yields and low deviations were typically backslope areas that had adequate soil moisture and do not flood. Nutrient inputs may need to be increased in these areas due to crop removal from high yields that occur on a regular basis. Field areas that have low yields and low deviations are typically summit or shoulder areas with limited soil moisture holding capacity. These areas may be eroded with low soil organic matter content and limited production capabilities. Inputs may be decreased in the summit or shoulder areas due to low stable yields and corresponding low nutrient removal rates." Kleinjan et al. (2006)
Conclusion
"There are several ways to use yield data to analyze yield trends both spatially and temporally. Production zones will become more accurate with time as more and more growing seasons are added to the database. Different maps can be used for different purposes. Average yield maps can be used to define yield goals. However, field areas that have high standard deviation in yield will have yield goals that are year-dependent. Standard deviation maps are very useful in identifying areas requiring corrective treatments. By comparing yield and standard deviation maps, the potential yield losses associated with not implementing a corrective treatment can be determined. For example, a 10-acre footslope area with excess water in 1 out of 10 years that results in a 90% yield loss for that season. The monetary loss from that year would be $4,500 (10 acres @ 180 bu/A loss @ $2.50/bu). This situation occurred in this field in 1996.
In this case study, the preferred method for explaining yield variability used a combination of average yields and standard deviation to delineate productivity zones. Due to the yield variability present in most agricultural production systems, producers need to analyze the different zone creation methods for each field and determine for themselves how management decisions may be altered to increase production efficiency." Kleinjan et al. (2006)
Reference
Kleinjan, D.E. Clay, C.G. Carlson, and S.A. Clay. 2006. Developing Productivity Zones from Multiple Years of Yield Monitor Data. Site Specific Management Guidelines; SSMG-45, 10/06.