Reference Strips and Precision Sensors for Nitrogen Management

Precision Agriculture Seminar

February 24, 2015

Boise, ID

Reference Strips

and Precision Sensors

Olga WalshAssistant Professor, Cropping Systems Specialist

Parma Research & Extension Center

University of Idaho

Presentation Outline:

1. Precision Ag (PA) benefits to

producers (specific example)

2. Level of PA adoption

3. Future of PA

4. Yield goal vs Yield potential

5. How sensors work

6. Reference strips

7. Misconceptions

8. Idaho research

Precision ag: Specific benefits

to producers

next few slides from Mr. Robert

Blair (2009 Precision Ag Farmer of the Year

2011 Eisenhower Fellow in Ag

2012 McCloy Fellow in Ag

Fourth-generation farmer Kendrick, Idaho)

Variable Rate Technology (VRT) VRT includes computer controllers and

associated hardware to vary output of fertilizer

and chemicals.

Utilize application map and GPS info to control

the hardware that varies the application rate.

Yield-goal/map based systems

Yield-potential/”on-the-go” systemsRT200, 6 GreenSeeker

sensor system

Sense and spray on the go

Benefits of VRT

• Sensor data + Algorithm = Fertilizer Recommendation

• Manage areas instead of whole field

• Precise input placement – As needed where needed

• Environmentally friendly

• 20-25% Cost savings

2009 NITROGEN HIT A RECORD HIGH

N cost was $.80 - $.90 per pound

100 pounds of N cost between $80 to $90 per acre

N costs on 500 acres of winter wheat at $.85/lb

at a 100 lb rate totaled:

$42,500

USING THE $42,500 TOTAL, COST

SAVINGS ON 500acres:

5% - $2,125

10% - $4,250

15% - $6,375

20% - $8,500

APPLICATION

Guidance

/Steering

VR Fertilizer

Seed Control

Auto Boom

NET

SAVINGS

2.5%

21%

5%

5%

PERCENT

SAVINGS

$10,137

$17,850

$2,716

$4,600

TOTAL PRECISION AG SAVINGS = $35,303

Current Idaho projects

Systems for Improving Water and

Nitrogen Use Efficiency in Spring

Wheat (WUE and NUE)

Precision Sensing for Improved

Wheat Production (N, wheat varietal

nurseries, diseases, pests)

Level of Precision Ag Adoption

PRECISION AG ADOPTION IN THE

U.S.

20% Adoption

amongst farmers

80% Adoption

amongst service

providers

WHY ONLY A 20% ADOPTION

RATE FOR FARMERS

Average Age of

Farmers

Understanding the

Technology

Capital Outlay

Lender’s Position

Landlords/Others

Involved

Future of Precision Ag

Unmanned Aerial Systems -

Drones From military/security forces to precision agriculture.

The U.S. Congress has mandated the Federal

Aviation Administration (FAA) incorporate drones into

national airspace by Sept. 30, 2015

Next 10 years: annual spending on drones will

increase by 73%; worldwide spending - $89 billion;

U.S. will account for 62% of the research and

development

Multirotor Ready to Fly Kit

X4 - Aerial Precision Ag.

Ready to Fly Skyjib X4

Titanium Film Kit – Aerial

Systems International

Robotics

Chemical Applications in Orchards

Mechanical Weeding

Autonomous TractorsBlue River Technology - $3.1

million funding to develop

agricultural robots to kill

weeds and thin out plants

Vision Robotics Corporation

– harvesting robots

Yield goal vs Yield potential

Nitrogen Use Efficiency

N is a key nutrient limiting crop production

N use efficiency (NUE) is only about 40%

About 60% of applied N is lost via volatilization and plant loss, run off and leaching, immobilization and denitrification =>

60% of funds growers invest in N fertilizer is lost

Reference strips and crop sensors help to accurately estimate crop yield potential and crop’s responsiveness to N mid-season.

Yield Goal vs Yield Potential

• Yield Goal:

Average yield for past 5 years + 30% (just in case we have a good year)

Based on past (historical data)

Uses average N rates

• Yield Potential:

Estimated using in-season data

Based on current crop nutrient status

Precise N rate (crop- and site-specific)

Yield Potential Varies Year to Year

0

20

40

60

80

100

120

140

160

180

200

1940 1950 1960 1970 1980 1990 2000 2010 2020

“Maximum Attainable Yield” (Yield Goal)

ActualHarvested

Yield

Taylor, 2009

5 times in 70 years, harvested wheat yield = yield goalShould we fertilize for maximum yield every year?

Alternative to Yield Goal - Yield Potential

How Sensors Work

Crop Sensors

hollandscientific.com; topconpositioning.com; nue.okstate.edu; agleader.com

Why use crop sensors?

Feed the crop, maintain the soil

Nitrogen = fuel for plant growth and development

To achieve highest efficiency – need to provide the crop with exactly how much it needs (no more and no less)

Why use crop sensors? Sensors = plant fuel gages, tell us:

How much the crop needs to reach yield potential

How much the crop received already from the soil (residual N, mineralization, rainfall …)

Do we always add the same amount of fuel no matter what?

Should we apply the same amount of N every year to every field?

E F

1/2

E F

1/2

E F

1/2

Potential and Response

Research showed that both crop yield potentialAND crop response to applied N changes:

year to year (temporal variability) and

field to field (spatial variability)

To get good estimates of N fertilizer demand, both the crops ability to respond to additional N and the grain yield potential must be known.

Sensor-Based N Rate

1. We needa lot

2. We don’t need much

3. We needa little1. High Yield

Potential, plants some-what deficient in N2. Very high Yield Potential, almost adequate N nutrition3. Low Yield Potential, plants are very deficient

Reference Strips

Nitrogen Reference Strips

Reference Strip Conrad, MT

Everything is relative and understood in comparison

The easiest way to assess nitrogen status –establish a non-limiting N strip and compareit to the rest of the field

Help, My Strips Did Not Work!

I can’t see my Reference Strip in my field, everything looks the same

It worked! – Enough N was delivered to the crop “for free”

Your crop probably will not benefit from addition of N fertilizer

Aren’t you glad you did not apply that high N rate to the whole field/farm?

www.blog.iastatebk.com, 2013

N Reference Strips

STEP-BY-STEP:

Establishing N Reference Strips every year

Apply starter fertilizer at seeding

Evaluate the Strip vs the rest of the field

Make N fertilizer decisions

www.noble.org, 2013

CIMMYT, Mexico

300 lb N/ac 0 lb N/ac

Large-scale on-farm studies

Sensor Basics

• Emits light and measures reflectance from plants

• Red light is used for photosynthesis (absorbed)

• NearInfrared light – not enough energy, not used (reflected)

• Sensor reading - Similar to a plant physical examination

Lightgeneration

Light signal

Lightdetection

? Calculate NDVI

“Sensor”

www.nue.okstate.edu, 2014

Sensor Basics

• Sensor can detect: Plant Biomass

Plant Chlorophyll

Crop Yield

Water Stress

Plant diseases, and

Insect damage

Sensors are used by agronomists, breeders, plant pathologists, weed scientists, crop consultants, growers

Lightgeneration

Light signal

Lightdetection

?Calculate NDVI

“Sensor”

www.nue.okstate.edu, 2014

red

redNIR

NIR

30%50%

60% 8%

NDVI = (NIR-red)/(NIR+red)

Sensor detects the amount of light reflected from the crop and calculates NDVI Tubana, 2007

NDVI = 0.76

NDVI = 0.25

What the GreenSeeker sensor “Sees”

The vigor of the leaves and

the ratio of plant to soil

affect NDVI values

N-Tech Ind., 2009

N-Tech Ind., 2009

What the GreenSeeker sensor “Sees”

Most Common Misconceptions About Sensors

Misconceptions

Misconception 1: GreenSeeker is a Nitrogen Sensor

Facts:

Nitrogen leaf content is not a good predictor of yield potential

GreenSeeker is a biomass sensor

Biomass/color is highly correlated with yield potential

Gerhardt, 2009

Misconceptions

Misconception 2: “I can see variability with my eyes—I don’t need a sensor”

Facts:

We can see macro variation in a field, but not subtle changes

We cannot remember where the variations are and to what degree

A flat rate is typically not the optimal answer to variability

Gerhardt, 2009

Misconceptions

Misconception 3: “I can do variable rate N with historic data—I don’t have the time nor the need for an in-season device.”

Facts:

In certain seasons, 1 year of quality in-season data can be more valuable than 10 years of historic data

Averaging information has bias that will limit the high end and over estimate the low end

In wet years hills do best, in dry years the lower areas do best Gerhardt, 2009

Misconceptions

Misconception 4: “If you give me your yield goal, I’ll tell you how much nitrogen to apply”

Facts:

“Yield Goal” fertility is a not the best approach economically and environmentally

You can’t estimate mineralization rates, residual N, or lost N without in-season information(ex. Nitrogen Rich Strip vs. Farmer Practice)

Gerhardt, 2009

Idaho research

Sensor-Based Work in Idaho

Pre-Season Variable Rate Nitrogen in Potatoes

Cook, Hopkins, Ellsworth, Bowen, and Funk (University of Idaho, Idaho Falls and Twin Falls)

2 growing seasons (2003-04), 5 fields, Eastern ID

Objective: To compare traditional and sensor-based variable rate fertilization

Results:

Support the concept of variable rate N application in potato production

The recognized increase in yield and quality more than compensated for the increased cost of this method of variable rate N fertilization

Sensor-Based Work in Idaho

In-Season Variable Rare N in Potato and Barley Production Using Optical Sensing Instrumentation (2004)

Bowen, Hopkins, Ellsworth, Cook, and Funk (University of Idaho at Idaho Falls and Twin Falls)

Objective: To evaluate the use of optical sensing instrumentation to help manage in-season N for potato and malt barley

1 growing season, 4 potato fields, 5 barley fields

Results: Sensors can be used to prescribe variable N rates to malt barley at jointing and to potatoes prior to row closure

A research assistant for Dr. Bryan Hopkins (currently a soil scientist at Brigham Young University, Utah)

evaluates sugarbeet canopy health using a GreenSeeker, Idaho

Sugar Producer, 2013

Thank you!

Olga Walsh

Assistant Professor, Cropping Systems Specialist

Parma Research & Extension Center

University of Idaho

(208) 722-6701 (ext 218)

[email protected]

Blog: Idaho Crops & Soils –

www.idcrops.blogspot.com

Follow us on Twitter: https://twitter.com/IDCrops

From sensor to N Rate

Established Reference Strip (at seeding)

Compared to the rest of the field (sensed at tillering)

Now we know: yield potential and crop responsiveness to N

How do we determine the needed N rate?

Need a formula (algorithm) to translate sensor readings into N recommendation

Algorithms

•Accurate mid-season fertilizer Nitrogen recommendations based on NDVI

•26 algorithms•> 10 crops and•> 20 regions (US and worldwide)

•15% increase in NUE•Savings: $10 - $20/ac

online since 2002

N-Rate Calculator on-line

www.nue.okstate.edu

N-Rate Calculator on-line

N-Rate Calculator on-line

Sensor-Based Approach Recognition

Named “the most revolutionary approach in a century to fertilizing crops” - the U.S. Department of Agriculture

Voted "the best and the brightest developed throughout the world for the agricultural, food, and biological systems industries" -American Society of Agricultural Engineering