Precision Ag Insight | Optimize Returns with On-Farm Trials

Sponsored by XSInc

Using On-farm Trials to Optimize Your Inputs

Chad Godsey, PhDPrecision Ag Insight WebinarFebruary 17, 2015

Sponsored by XSInc

Webcast Format

• You are muted.

• Please ask questions using the “Questions” box.• Click the red arrow to close/open

controls.• Click the “plus” next to “Questions” if

you don’t see the questions box.

• A recording of today’s webcast will be emailed to you.

Minimized Controls

Sponsored by XSInc

Speakers• Chad Godsey, PhD

Owner, Godsey Precision Ag

• Angela WalkerAgVeritas Product Manager, XSInc

Sponsored by XSInc

On-Farm Research Is Important• Answer questions important to growers

˃ New products˃ Precision ag practices

• Increase grower profitability

• Compliance, stewardship and sustainability

Sponsored by XSInc

Registration Question

14%

29%

29%

21%

7%Rarely, 0-25% of the time Sometimes, 25-50% of the timeUsually, 50-75% of the time Almost always, 75+% of the timeI don't set specific objectives

How often do you/your customers meet your objectives for on-farm trials?

Sponsored by XSInc

3 Keys to Success• Quality over Quantity• Keep It Simple• Reduce Noise

Sponsored by XSInc

3 Keys to Success• Quality over Quantity

˃ Focusing on quality in the short-term will lead to quantity over time.

˃ If we spend a little more time, we can uncover a lot more information with every trial.

• Keep It Simple• Reduce Noise

Sponsored by XSInc

3 Keys to Success• Quality over Quantity• Keep It Simple

˃ Keep it small and simple˃ Improve from year to year

• Reduce Noise

Sponsored by XSInc

3 Keys to Success• Quality over Quantity• Keep It Simple• Reduce Noise

˃ Know and manage sources of “noise”˃ Spatial variability (e.g. soils, nutrients, etc.)˃ Temporal variability (e.g. due to weather)˃ Human error

Sponsored by XSInc

Noise Example: Fungicide Trial

No fungicide appliedFungicide applied

• 7 bu/ac increase BUT• West half has always been 5-7 bu/ac better

Sponsored by XSInc

3 Keys to Success• Quality over Quantity• Keep It Simple• Reduce Noise

˃ You can never be 100% certain that yield differences in a trial are solely due to the treatments being evaluated.

˃ But you can manage sources of noise with good planning and analysis.

Sponsored by XSInc

Registration question

40%

29%

24%

7%Collecting the data Getting a clear answer from the dataGetting the trial put out correctly I don't set specific objectives

What is your biggest challenge in achieving your objectives for on-farm trials?

Sponsored by XSInc

3 Critical Pieces of a Trial Plan

• The Question• The Lay Out• The Data

Sponsored by XSInc

Define the question

• The first step is clearly defining a question that you want to answer.

˃ Is Product A better than Product B?

˃ If I increase or decrease the application rate of this

product by 10% in each management zone, how much

will my yield increase or decrease?

• Consider other factors that may influence trial outcome

˃ Management practices

˃ Hybrid or varietal differences

Sponsored by XSInc

3 Critical Pieces of a Trial Plan

• The Question• The Lay Out• The Data

Sponsored by XSInc

Poll: Trial PlacementWhich factors do you consider when deciding on trial placement?

˃ Whatever is easiest/what the producer is willing to do

˃ Field conditions (e.g. soil types) that might affect response

˃ Variable rate practices/management zones

˃ Data collection plan/equipment widths

Sponsored by XSInc

4 Considerations for Trial Placement

1. Objective

2. Background information

3. Equipment and logistics

4. The “Three Rs” of on-farm trials

1. ObjectivesCheck StripsObjective: test new products or practices

˃ Obtain samples from most areas of the field

˃ Allows you to see how response varies across field

Test PlotsObjective: Refine an existing practice

˃ Take a sample from a previously defined management zone

˃ Test a change on a small area to minimize yield impact

Sponsored by XSInc

2. Background Information

• Avoid it˃ Unstable areas˃ Field edges

• Get a sample from it˃ Soils˃ Topography˃ Yield potential

Sponsored by XSInc

Using Historical Yield Data

• Why?˃ Accounts for temporal variability˃ Provides a baseline to compare treatments against ˃ Useful for problem areas (i.e. Fe deficiency)

• How?˃ Create yield potential zones

˃ Incorporate into management zones˃ Run test strips through all zones

˃ Create yield stability zones˃ Avoid and/or ignore during analysis

Sponsored by XSInc

Multi-year Yield Map Analysis

Relative grain yield Year to year stability

Sponsored by XSInc

3. Equipment & Logistics

• Ease of application• Width of applicator• Width of combine

Compost TrialSeeding Rate Checks

Sponsored by XSInc

4. The “Three Rs”

•Replicate•Randomize•Request help

A poorly planned (statistically-speaking) on-farm trial has a high risk of failure.

Sponsored by XSInc

Replicate

• Repeat treatments to enable the separation of treatment effects from “background noise”.• When possible, design replications to be within

relatively uniform areas (e.g. not up a hill).• Over replication can introduce noise from human or

machine error.

Sponsored by XSInc

Randomize

• Use when testing several things• Randomly select where each test will go • Decreases the odds that spatial variability will

influence treatment effects.˃ For example, areas of the field that are poorly drained.

Sponsored by XSInc

Request Help

• If research is not your vocation, then don’t be shy about requesting help from those who conduct research for a living.

˃ University researchers & Extension specialists˃ Industry researchers & agronomists˃ Crop consultants

Sponsored by XSInc

3 Critical Pieces of a Trial Plan

• The Question• The Lay Out• The Data

Sponsored by XSInc

4 Parts of Data Collection

• Background Information• Trial Data• In-season Observations• Harvest

Sponsored by XSInc

Minimize harvest “noise”

• Faithfully calibrating scales or sensors.˃ This includes grain moisture sensors.

• Triple-check yield monitor settings!˃ Logging of data to card

˃ Swath width

˃ Load identification

Sponsored by XSInc

An Example

The QuestionIs it profitable to apply fungicide to corn at V5?

Background Information

Sponsored by XSInc

An Example

The QuestionIs it profitable to apply fungicide to corn at V5? Lay Out

Sponsored by XSInc

Thank You

Chad Godsey, PhDOwner, Godsey Precision AgPhone: 970-630-7732Email: [email protected]: godseyag.comTwitter: @godseyag

Stick around for an AgVeritas demonstration.

Sponsored by XSInc

AgVeritas Demo

• Web-based yield analysis tool• Uses spatial statistics to analyze yield• Analyzes all the factors that impact yield simultaneously• Planting choice / rate• Treatment choice / rate• Nutrient levels, soil types, elevation• And many more

• Determines which factors impacted yield more and less• Calculates if and where a tested management practice is

profitable

Sponsored by XSInc

AgVeritas DemoFungicide TrialSoil Types Elevation

The Question: Is it profitable to apply fungicide on this field?

Result of Comparison of AveragesTreated areas yield 8 bu/ac higher on average

https://agveritas.xsinc.com/AnalysisResults/Summary/323

Sponsored by XSInc

Thank You

Chad Godsey, PhDOwner, Godsey Precision AgPhone: 970-630-7732Email: [email protected]: godseyag.comTwitter: @godseyag

Stick around for an AgVeritas demonstration.