Remote sensing based in-season N recommendations

David Clay and Cheryl Reese

Dryland wheat systems

Copyright 2010, Cheryl Reese SDSU Plant Science Dept.

2

N and P recommendationsMycorrhizae, carbon footprints, salinity

• The northern Great Plains has some of the highest climate variability in the United States.

• Develop a sustainable system that increases soil and crop resilience.– Cover crops, were estimated by NRCS to be

implemented in over 140,000 acres in South Dakota last year.

– In-season N rates based on remote sensing may be a tool that can be used to help manage this variability.

Changes in our cultural practices and soil quality.

Is soil carbon increasing the need for sensors.

What has happened to soil carbon

Year

1990 2000 2010

SO

C (

lbs/

acre

)

25000

30000

35000

40000

45000

50000

NC and C regions, dSOC/dt = 328 lbs/(a x year), r= 0.6** NE, EC, and SE region, dSOC/dt = 316 lbs/(a x year), r=0.55**All region, dsoc/dt = 326 klbs(a x year), r= 0.5**

Increasing yields and conservation tillage adoption

Rapid no-tillageadoption

Clean or conventional tillage

1990 2000 2010

# cl

ean

till/#

of s

urve

ys

0.0

0.2

0.4

0.6

0.8

1.0No-tillage

1990 2000 2010

# no

-till

age/

# su

rvey

s

0.0

0.2

0.4

0.6

0.8

1.0

1.2

North centralCentralNortheastEastcentralSoutheast

Reduced tillage

Year1990 2000 2010

# re

duce

d til

lge/

# of

sur

veys

0.0

0.2

0.4

0.6

0.8

1.0

1.2Grain yield

Year1990 2000 2010

Cor

n yi

eld

(bu/

acre

)

30

60

90

120

150

180

Carbon footprint (Corn)

Carbon footprintsProduct gCO2 equivalent

Hamburger (burger) 3,600-6,100

Light bulb (4 hour day for 1 year) 11,000

Nebraska (bu corn) (Liska et al.) 7,640

North Dakota (bu corn) 6,630

South Dakota 5,840

Minnesota 5,968

Gal of gasoline 12,600

Carbon footprints (g CO2 eq/MJ)(Greet model)

SD Region Footprint w/o soil Soil (g CO2 eq/MJ) Adjusted footprint

Gasoline 95.86

NC 39.61 -23.6 16.4

C 39.67 -15.9 23.0

NE 39.63 -6.6 33.0

EC 39.67 -13.1 26.6

SE 39.6 -17.1 22.4

Relative to gasoline there is a 65 to 83% reduction in the footprint

Soil carbon• Changes in our soil carbon values may be producing

changes in our fertility requirements,

• Data from our soil testing laboratories can provide important information,

• A good way to account for OM differences is to include organic matter content in the N recommendation.

• Remote sensing-based recommendation may help account for differences as well.

Sensor use in wheat grown in a semi-arid environment

N rate impact on yield and wheat quality

• Research conducted in 2007 and 2008• 5 N rates (0, 25%, 50%, 100%, and 150% of

recommended rate,• 2 water rates (adequate and deficit), • N and water budgets developed,• YLNS and YLWS determined using 13C isotopic

discrimination,• Protein and dough stability measured.

Farinograph

Poor quality bread

N rate influenced water use and N use efficiency (Dakota Lakes, Overly 2007)

N rate Yield bu

WUEBu/in

NUE% fert

Protein%

Stabilitymin

0 55 8.52 12.4 5.2

¼ 61.6 9.55 42 13.5 8.0

½ 63.9 9.83 33 14.6 9.9

1 66.2 10.3 22 15.4 10.0

1.5 64.2 9.93 14 15.8 11.3

YieldsTreatment 2007 (Mg/ha) 2008 (mg/ha)

0 3.69 4.4825% 4.13 5.0550% 4.29 5.04100% 4.44 5.54150% 4.31 5.73p 0.001 0.001LSD 0.2 0.28WaterNot 4.52 5.32Stressed 3.82 5.02p 0.005 0.1

Protein vs fert + min N

N fert plus mineralized N

100 150 200 250 300 350 400

Pro

tein

(g

/kg)

100

110

120

130

140

150

160

170

2007 2008

Protein = 102 + 0.131(available N), r2 = 0.81**

Protein (g/kg)100 110 120 130 140 150 160 170

Sta

bilit

y (m

inut

e)

2

4

6

8

10

12

14

16

18

Overley,2007Alice, 2008

Overley:Stability = -14+0.16 (protein), r2 =0.62**

Aluce: Stability = -18+0.22(protein) r2 =0.53**

We may see cultivar differences in dough quality

Reflectance calculations

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

SI-NDVIwf = NDVI / NDVIwf

SI-NDVImz = NDVI / NDVImz

Stress impacts on reflectanceTreat Blue Green Red Red nir NDVI SI-NDVI

WfSI-NDVIMZ

0 3.24 5.68 4.62 9.95 36.3 0.81 0.91 0.940.25 3.01 5.35 4.20 9.43 38.7 0.85 0.96 0.980.5 2.96 5.25 4.08 9.28 40.3 0.86 0.97 0.991.0 2.95 5.19 4.03 9.16 40 0.87 0.97 11.5 2.92 5.06 3.91 8.9 41.3 0.87 0.97 1P 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.01lsd 0.16 0.23 0.26 0.40 1.87 0.01 0.01 0.02WaterNot 2.64 4.91 3.84 9.13 39.9 0.88 0.99 0.99Stressed 3.39 5.71 4.49 9.56 38.8 0.83 0.93 0.98P 0.001 0.001 0.001 0.03 0.09 0.002 0.002 0.36

N rate (kg/ha)0 50 100 150 200 250 300

SI-

ND

VI w

f

0.65

0.70

0.75

0.80

0.85

0.90

0.95

1.00

1.05

2007 well watered2007 water stress2008 well watered2008 water stressed

N rate (kg N/ha)

0 50 100 150 200 250 300

SI-

ND

VI m

z

0.65

0.70

0.75

0.80

0.85

0.90

0.95

1.00

1.05

SI-NDVIwf0.75 0.90 1.05

N f

ertil

izer

yie

ld in

crea

se

-400

0

400

800

1200

160020072008

Stem extension, r= -0.29

SI-NDVImz0.75 0.90 1.05

N f

ertil

izer

yie

ld in

crea

se

-400

0

400

800

1200

1600Management zoneWhole field

Stem extensionr= 0.49*

SI-NDVIwf

0.75 0.90 1.05

N f

ertil

izer

yie

ld in

crea

se

-400

0

400

800

1200

1600

SI-NDVImz

0.75 0.90 1.05N f

ertil

izer

yie

ld in

crea

se

-400

0

400

800

1200

1600

Flag leaf, r= -0.33Flag leafr= -0.51**

Summary• Reference areas can be used to reduce variety

and water stress impacts on sufficiency index values.

• Reference areas can be placed in a strip or within different management zones.

• Using a reference area in high yielding areas can result in diagnosing water stress as N stress.

• In low yielding areas N stress can be diagnosed as water stress.