1.

Net N

min

m

g kg

−1so

il=

𝐼𝑛𝑜

𝑟𝑔𝑎

𝑛𝑖𝑐 𝑁

−𝐼𝑛

𝑜𝑟𝑔

𝑎𝑛

𝑖𝑐 𝑁𝑡=

0

Estimating Mineralizable Nitrogen from Organic Materials

The u

se of n

ear in

frared

reflectance sp

ectrosco

py

allow

ed fo

r rapid

estimates

of th

e in

itial organ

ic N o

f the

material (r2

=0.9

7) an

d rate

con

stant o

f min

eralization

(r2

=0.9

2) w

hen

net N

m

ineralizatio

n w

as fit to first

ord

er kin

etics.

Introduction

Objectives

Conclusions

1. Determine mineralizable N and rate of mineralization from 42 different organic-certified materials (24 commercially available organic fertilizers, 8 composts, and 10 poultry litters).

ACKNOWLEDGEMENTS

PMN INCUBATION 1. Organic materials were characterized for initial water content (drying at 65oC for 48 h),

inorganic N (KCl), total C and N (dry combustion) and total elements (acid digestion). 2. Materials were applied at a rate to supply an estimated 75 mg N kg-1 of PAN to Cecil Sandy

Loam soil at 50% water holding capacity and incubated at 30oC for 99 days. PAN was estimated to be 10%, 40% and 50% of the total N for composts, poultry litters, and fertilizers respectively.

3. Subsamples were taken on d 1, 3, 7, 14, 35, 56, 78, 99 to determine inorganic N (1M KCl, 1:5) 4. Net N mineralization (Net Nmin) was determined for each d by subtracting the d.1 initial

inorganic N and fitting the data to a first order exponential model. N mineralization measured in the soil control was then subtracted from the model.

5. Net Nmin (g kg-1 material) was fit to first-order kinetics using SAS PROC NLIN:

Fig.4. Simplified model structure.

Kate Cassity-Duffey, M. Cabrera, D. Franklin, J. Gaskin, D.E. Kissel, and U. Saha Department of Crop & Soil Sciences - University of Georgia, Athens, GA;katecass@uga.edu

2. Evaluate NIRS for predicting initial characteristics and rate constants of mineralization of organic-certified materials.

Results: Net N Mineralized

Fig. 2. Stella ® Model Software Input Parameters

Fig.5.Model in Stella® Modeling Software

Fig.6. Litter WC sub-model driven by air water potential.

Results: NIRS

Materials and Methods

NIRS ANALYSIS 1. Samples were oven dried (65oC), ground, and packed in quartz ring cups. 2. Samples were then scanned using FOSS NIRSystem 6500; 400-2500 nm at 2-nm intervals 3. WinSCAN v 1.50 software (FOSS North America, Eden Prairie, Minnesota) reported

reflectance (R) as log (1/R) and was used for statistical analysis. 4. Software used principle component regression for analysis of the spectra to develop

calibration and cross validation with initial characteristics, and the rate constant of mineralization.

In the southeastern United States, compost, poultry litter, and commercial organic fertilizers (made from plant and animal byproducts) are commonly used in vegetable production to supply plant-available nitrogen (PAN). However, due to the high variability in the nitrogen (N) content and composition of these materials, predicting the amount of PAN and the rate of N mineralization from these products is difficult. The ability to accurately predict N mineralization through the use of near infrared spectroscopy (NIRS) would allow for rapid estimations of PAN immediately prior to application, leading to increased precision of application of these products and reduced farmer costs.

Fig.

1. T

he

mat

eria

ls u

sed

fo

r P

MN

an

d N

IRS

anal

ysis

Net Nmin = 𝑖OrgN 1 − e−kt where iOrgN is the initial organic N (g kg material), k is the rate constant of mineralization, and t is time (d)

PL2

PL3

PL5

PL8

PL9

PL10

PL11

PL12

PL13

PL14

CO

MP

1C

OM

P2

Blo

od M

Feath

er M

Sym

phony

Fis

h M

Cra

b S

hell

Harm

ony

Bio

Fis

h M

Must

ard

MP

G V

eg M

ixS

oyb

ean M

Veggie

Mix

Fis

h B

one M

Bone M

Cotton M

Vegan m

ixN

S 7

-12-0

NS

5-6

-6N

S 1

0-2

-8P

MP

M P

lus

MG

3-3

-3

Ne

t N

Min

era

lize

d %

O

rga

nic

Nitro

ge

n A

pp

lied

0

20

40

60

80

100

Ra

te c

on

sta

nt

(k x

10

2 d

-1)

-1

0

1

2

3

4

5

6

Net N Min

Rate Constant

PL2

PL3

PL5

PL8

PL9

PL10

PL11

PL12

PL13

PL14

CO

MP

1C

OM

P2

Blo

od M

Feath

er M

Sym

phony

Fis

h M

Cra

b S

hell

Harm

ony

Bio

Fis

h M

Must

ard

MP

G V

eg M

ixS

oyb

ean M

Veggie

Mix

Fis

h B

one M

Bone M

Cotton M

Vegan m

ixN

S 7

-12-0

NS

5-6

-6N

S 1

0-2

-8P

MP

M P

lus

MG

3-3

-3

Ne

t N

Min

era

lize

d %

O

rga

nic

Nitro

ge

n A

pp

lie

d

0

20

40

60

80

100

Ra

te c

on

sta

nt

(k x

10

2 d

-1)

-1

0

1

2

3

4

5

6

Net N Min

Rate Constant Fig. 2. Net N min from materials (PL=poultry litter, COMP=compost, M=meal) that mineralized and respective rate constants of mineralization determined through first-order kinetics.

Characteristic Mean SD RSQ SECV 1-VR

Water Content g g-1 0.33 0.35 0.86 0.17 0.76 Total C g 100 g-1 35.82 9.67 0.74 7.13 0.49 Total N g 100g-1 4.64 2.50 0.79 1.37 0.70 Organic N g kg-1 40.38 22.66 0.97 10.17 0.79 k x 102 d-1 0.51 0.48 0.92 0.42 0.30

The use of near NIRS allowed for rapid estimates of the initial organic N of the materials (r2=0.97) and the rate constant of mineralization (r2=0.92) when net N mineralization was fit to first-order kinetics for a range of organic materials.

Acknowledgement This work was supported by USDA-SARE Grant LS16-269

• NIRS accurately estimated initial material water content, total N and C, organic N as well as the rate constant of mineralization allowing for prediction of mineralizable N using first-order kinetics.

• Net N min calculated using NIRS predicted variables led to a r2 =0.92 with measured values

• Blood and feather meal values were not estimated well using NIRS and were removed

PL2

PL3

PL5

PL8

PL9

PL10

PL11

PL12

PL13

PL14

CO

MP

1C

OM

P2

Blo

od M

Feath

er M

Sym

phony

Fis

h M

Cra

b S

hell

Harm

ony

Bio

Fis

h M

Must

ard

MP

G V

eg M

ixS

oyb

ean M

Veggie

Mix

Fis

h B

one M

Bone M

Cotton M

Vegan m

ixN

S 7

-12-0

NS

5-6

-6N

S 1

0-2

-8P

MP

M P

lus

MG

3-3

-3

Ne

t N

Min

era

lize

d %

O

rga

nic

Nitro

ge

n A

pp

lied

0

20

40

60

80

100

Rate

con

sta

nt

(k x

10

2 d

-1)

-1

0

1

2

3

4

5

6

Rate Constant

Poultry LItter

Compost

Fertilizer

• Avg Net Nmin % of Total N applied: • Composts: -3.5% • Poultry litter: 17.0% • Fertilizers: 30.8%

• Rate constants (k) varied among materials.

• Very high “k” in blood and feather meal may be due to the high solubility of these products.

NIRS Predicted Net Nmin 100 d (g kg-1 material)

0 20 40 60 80

Mea

sure

d N

et N

min

100

d (

g k

g-1

mate

ria

l)

0

20

40

60

80

Poultry Litter

Compost

Fertilizer

Measured=0.81*Predicted+ 0

r2=0.92

Fig. 3. Predicted Nmin using organic N values and k constants determined through NIRS versus measured Nmin at 99 d

Table. 1. NIRS determined mean, standard deviation (SD), r-square (RSQ), standard error of cross validation (SECV), and 1- validation ratio (VR) of selected material characteristics.