Using Yield History to Establish a Baseline for OnFarm Research

1
Using Yield History to Establish a Baseline for
On-Farm Research
C.B. Godsey Kansas State University
8th Annual Kansas Precision Agriculture
Technologies Conference Program
2
Overview

• Developing a baseline
• Evaluation of treatments using a baseline

3

• Common problems encountered with on-farm research
• Temporal variability
• Spatial variability

4
Growing season
5
DTPA Fe Test Levels
2 4 ppm
4 - 6 ppm
gt 6 ppm
6

• Using yield data as a tool to minimize impacts of
  temporal and spatial variability
• Accounts for temporal variability
• The greater the number of years the better
• Provides a baseline to compare treatments against
  
• Especially useful for problematic areas in fields

7
Establishing a Baseline

• 1. Utilize previous years yield maps
• Create 60 ft x 60 ft grids
• Normalize yields
• (Cell Avg. Field Avg.)
• Field Avg.
• 2. Average years to determine low, average, and
  high yielding areas

YieldNORM
8

• Example
• Cell 1
• Cell average is 200 bu ac-1
• Field average is 225 bu ac-1
• 200-225
• 225
• Three year normalized yield average is -0.2
• What does this mean?

YieldNORM
-0.11
9
Establishing a Baseline

• Error bars that cross the 0 s line indicate those
  cells are within 1 standard deviation of mean and
  are considered to be average yielding

10
Establishing a Baseline

• Classifications
• Consistently low yielding cells
• Consistently average yields
• Consistently high yielding cells

11
Evaluation of Treatments

• Fe study example
• Soil properties associated with Fe deficiency
• Soil pH 7.4 - 8.5
• Presence of CaCO3
• Elevated CO2 levels in the soil
• Spatial heterogeneity
• Yield loss, sometimes quite severe
• Symptoms - interveinal chlorosis in corn,
  sorghum, and soybeans

12
Field-scale study

• 72 lb acre-1 FeSO4H2O applied in the seed row
• Soil sample transect

SE 14
Fe application strip
Soil sample points
Small plot location
13

• Yield monitor data
• Average yield calculated for 60 x 60 ft grid
  cells
• Observations separated into those within the
  problematic area and those outside

14
1999 Yield and Soil CaCO3 Content
Problematic Area
15
Grain Yield 1998
Grain Yield bu ac-1
16
Normalized Yield for 1998
17
Classification of Cells
Classification
Low and stable
Average
High and stable
18
Determining effect of Fe strip application

• Two methods
• Compare outside Fe application strip to inside
  the Fe application strip (adjacent cells)
• One year of yield data
• Compare the change in yield from treatment based
  on yield potential
• Take into account previous years yield data

19
Paired t-test

• Use paired t-test to evaluate Fe application
• One year of data year of treatment application
• Disadvantage
• Does not account for yield variability inside
  and outside strip
• Does not account for temporal variability

20
Comparing yield potential

• Use multiple years of yield data (yield baseline)
• Grain yield was calculated for 1997, 1998, 1999,
  and 2000
• Compare yield potential inside the application
  strip to outside the application strip based on
  previous yield data
• ? Yield Yield (2000) Avg. Yield (1997-1999)

21
Grain yield in Fe application strip at SE14.
Consistently greater yield outside application
strip
Similar yields in 2000
Fe trt reduced yield less
22
Grain Yield from Application Strip at SE14
23
Results from Fe strip application

• SE14 Fe application increased yield 6.4 in
  problematic area, or 13 bu acre-1

24
Summary

• Historical yield data is a valuable tool in
  treatment evaluation.
• Using yield history may increase probability of
  detecting treatment differences.
• Method works best with well defined or
  problematic areas.

25
Chad Godsey Agronomy Department Kansas State
University Email cgodsey_at_ksu.edu Tel 785 -
532 - 7897
26
(No Transcript)
27
Identifying Problematic Area
28
(No Transcript)