Nitrification, denitrification, and sorption-desorption of NH4-N in sands during water movement to subsurface drains

MISSING IMAGE

Material Information

Title:
Nitrification, denitrification, and sorption-desorption of NH4-N in sands during water movement to subsurface drains
Series Title:
Florida Water Resources Research Center Publication Number 65
Physical Description:
Book
Creator:
Mansell, R. S.
Fiskell, J. G. A.
Calvert, D. V.
Publisher:
University of Florida
Place of Publication:
Gainesville, Fla.
Publication Date:

Notes

Abstract:
Isotopically labelled (15N-depleted (NH4)2S04) N fertilizer was applied at a rate of 115 kg N/ha to a Florida citrus grove. located in a subsurface drained Spodosol in order to determine the fate of applied N. The fertilizer was applied to one 0.22-ha field plot each of shallow-tilled (ST), deep-tilled (DT) , and deep-limed/tilled (DTL) soil. Analyses of soil, soil solution, drainage water, and citrus leaf samples revealed that much of the NH4-N was nitrified to form NO3-N during the first 42 days after application of fertilizer. Partially because of a local drought, most of the applied N was absorbed by plant roots during the first 134 days and leaching losses in drainage water were less than 4% of the amount applied. Rates of nitrification and N uptake by roots were highest for DTL and lowest for ST treatments. Selective deep tillage with lime incorporation of Spodosols appears to be an effective means to manage fertilizer and water resources for citrus and possibly other agricultural crops without contaminating groundwater supplies.

Record Information

Source Institution:
University of Florida Institutional Repository
Holding Location:
University of Florida
Rights Management:
All rights reserved by the source institution and holding location.
System ID:
AA00001542:00001


This item is only available as the following downloads:


Full Text

















WATER RESOURCES RESEARCH CENTER


Publication No. 65


NITRIFICATION, DENITRIFICATION, AND SORPTION-
DESORPTION OF NH -N IN SANDS DURING WATER
MOVEMENT T6 SUBSURFACE DRAINS


By


R. S. Mansell
J. G. A. Fiskell
D. V. Calvert


Soil Science Department
University of Florida
Gainesville















NITRIFICATION, DENITRIFICATION, AND SORPTION-
DESORPTION OF NH -N IN SANDS DURING WATER
MOVEMENT T6 SUBSURFACE DRAINS



By


R. S. Mansell
J. G. A. Fiskell
D. V. Calvert



Publication No. 65

Florida Water Resources Research Center

Research Project Technical Completion Report

Project Number A-042-FLA

Annual Allotment Agreement Numbers
14-34-0001-0110
14-34-0001-1110
Report Submitted August 23, 1982


The work upon which this report is based was supported in
part by funds provided by the United States Department of
the Interior.











TABLE OF CONTENTS

Section Page

Title . . . . . . i

Table of Contents . . . . ii

Acknowledgments . . . . iii

Abstract . . . . . v

Chapter 1: Introduction . . . 1

Chapter 2: Isotopic Tracer Methods for Nitrogen in

Soil-Plant Systems . . 5

Chapter 3: Experimental Methods and Procedure 9

Chapter 4: Results and Discussion . . 18

Chapter 5: Summary and Conclusions . . 70

Literature Cited . . . . 76

Appendix: Synopsis of Master of Science Thesis 78










ACKNOWLEDGEMENTS

The Soils and Fertilizer Research Branch Division of

Agricultural Development, Tennessee Valley Authority (TVA)

located at the National Fertilizer Development Center near

Muscle Shoals, Alabama supplied the 15N-depleted (NH4)2SO4

used in this research. The project investigators are appre-

ciative to TVA and to Dr. Roland D. Hauck, Research Soil

Scientist with TVA. We are also indebted to Dr. Hauck for

making us aware of the automatic mass spectrometer facility

at the Los Alamos National Laboratory for analyses of
15N/ 14N ratios in water samples.

We wish to acknowledge the University of California,

Los Alamos National Laboratory (LANL) in Los Alamos, New

Mexico for performing nitrogen isotope analyses for the 1300

water and soil samples used in this investigation. We are

particularly appreciative to Dr. B. B. McInteer for permit-

ting us to utilize the analytical service of the Los Alamos

National Laboratory.

The authors are indebted to Dr. J. S. Rogers, Agricul-

tural Engineer with the United States Department of Agricul-

ture and University of Florida, for providing water drainage

data from the Soil, Water, Atmosphere, and Plant (SWAP)

research citrus grove where the field experiment for this

grant was performed.

Acknowledgement is also given to Marc E. Hall, former

Laboratory Technologist II in the Soil Science Department

and Mariano Delacasa, student assistant, for performing com-






iv



puter analysis of the data. The investigators also acknowl-

edge assistance from Mrs. Jennifer Johnson, former Labora-

tory Technologist for preparation of camera-ready figures

presented in this report.










ABSTRACT

Isotopically labelled (15N-depleted (NH4)2SO4) N fer-

tilizer was applied at a rate of 115 kg N/ha to a Florida

citrus grove located in a subsurface drained Spodosol in

order to determine the fate of applied N. The fertilizer

was applied to one 0.22-ha field plot each of shallow-tilled

(ST), deep-tilled (DT), and deep-limed/tilled (DTL) soil.

Analyses of soil, soil solution, drainage water, and citrus

leaf samples revealed that much of the NH4-N was nitrified

to form NO3-N during the first 42 days after application of

fertilizer. Partially because of a local drought, most of

the applied N was absorbed by plant roots during the first

134 days and leaching losses in drainage water were less

than 4% of the amount applied. Rates of nitrification and N

uptake by roots were highest for DTL and lowest for ST

treatments. Selective deep tillage with lime incorporation

of Spodosols appears to be an effective means to manage fer-

tilizer and water resources for citrus and possibly other

agricultural crops without contaminating groundwater

supplies.










Chapter 1: Introduction

Acid, sandy soils with shallow water tables are common-

ly utilized in Florida for intensive production of high-

value vegetable, fruit, and horticultural crops. Productiv-

ity of these soils using the definition by Hillel (1981) may

be severely restricted by limiting levels of both "chemical

fertility" and "physical fertility". High contents of

silica sand and low contents of colloidal material provide

porous soil matrices that primarily have only short-term

capacity to supply water and nutrients to active plant root

systems. Relatively large pores and low ion exchange capac-

ities of the solid particles also result in potential

leaching losses of fertilizer applied to such soils. During

rain storms or when excess irrigation is applied, soluble

nutrients from applied fertilizer may easily be displaced

from the rooting zone of the soil profile and be transported

into underlying groundwater. Obviously, careful management

of fertilizer and soil water for crops growing in these

soils is needed to simultaneously minimize chemical contami-

nation of groundwater supplies, maximize the use-efficiency

of crop roots for nutrients applied in chemical fertilizer,

and optimize the water use-efficiency of crop roots.

Although the average annual rainfall in Florida is high

(1.3 to 1.5 meters) the distribution during the year is

highly nonuniform giving a relatively dry period during the

winter and spring months and a relatively wet period during

the summer and fall months. For crops growing in sandy











soils with an impermeable layer located in the profile at

some depth less than 2 m, ditch or subsurface drainage sys-

tems are frequently installed so as to prevent development

of anaerobic soil conditions in the rooting zone during the

wet season. These systems are particularly needed to remove

excess water during intense thunderstorms from soil areas

with nearly flat topography. However, during the dry period

the accelerated removal of drainage water may well decrease

the effectiveness of rainfall events and the water use-

efficiency of the crop. Leaching loss of soluble fertilizer

nutrients may also be increased as a result of accelerated

water discharge from ditches or subsurface drain tubes

(Mansell, Wheeler and Calvert 1980; Mansell et al. 1977).

Such enhanced leaching of applied fertilizer should be most

evident during the wet period but also would be expected for

large storms even during the dry period. On the one hand

artificial drainage tends to enhance use-efficiency of crops

for water and fertilizer by permitting active growth of the

root system during the wet season, but accelerated leaching

loss of nutrients due to the drainage system tends to

restrict the use-efficiency of applied fertilizer. During

the dry period, the limited water-holding capacity of these

coarse-textured soils requires that irrigation be used to

maintain economically profitable crop yields. Insufficient

irrigation will limit yields whereas excess irrigation

enhances leaching loss of fertilizer. Thus effective soil

water management for crops growing in shallow sands must of










necessity include both irrigation and drainage practices.

The effectiveness of soil water management practices for

these soils will ultimately be reflected in the magnitude of

crop yields and in the quality of groundwater.

Results from a 3-year investigation (Mansell et al.

1977, Rogers et al. 1977, Calvert et al. 1981) of fertilizer

leaching losses from an experimental citrus grove located

on a Spodosol at the University of Florida Agricultural Re-

search Center near Ft. Pierce show that significant quan-

tities of NO3-N and PO4-N were removed from the soil through

discharge of drainage water. These losses occurred despite

a management practice of applying small quantities of fer-

tilizer frequently (every 3 months) to minimize leaching

losses. As much as 22% of 169.52 kg/ha of N (55% of fertil-

izer N was NH 4-N and 45% was NO3-N) applied in 4 split ap-

plications during a 12-month period was lost as NO3-N in the

drainage water. However, for treatments which had been

deep-tilled to a depth of 105 cm prior to planting of citrus

trees the average annual loss of NO3-N in the drainage water

was equivalent to only 3% of the total N applied as fertil-

izer. The effectiveness of the deep tillage plots in mini-

mizing the leaching loss of N was attributed to increased

retention of applied NH4-N in the top soil due to upward

transport of colloidal material from subsurface horizons and

to increased potential for denitrification of NO3-N in the

soil profile due to slower drainage rates which resulted in

generally higher water contents following rainfall events.











A major limitation of the interpretation of some of the

results from this study however was that the applied ferti-

lizer N was not labelled with a tracer such as 15N or 14N.

Thus a field investigation was performed to quantita-

tively evaluate the fate of 15N-depleted NH4 applied as fer-

tilizer to a subsurface-drained sandy soil in an experi-

mental citrus grove in South Florida. Isotopically labelled

fertilizer N was applied to plots of a Spodosol which had

previously received shallow- and deep-tillage treatments.

The primary objective of this investigation was to de-

termine the influence of movement (leaching), ion exchange,

and transformations (nitrification, denitrification, etc)

upon N- depleted NH4 applied as a single fertilizer appli-

cation to a tube-drained sandy soil in a citrus grove. A

secondary objective was to evaluate the influence of 3 soil

modifications --- shallow tillage ST, deep tillage DT, and

deep tillage with initial incorporation of limestone into

the profile DTL --- upon the fate of applied 15N- depleted

NH4. Both objectives require evaluation under field

conditions.

A brief review of the use of 15N and 14N isotopes as a

tracer for fertilizer nitrogen in soil-plant systems will be

given before presenting a description of experimental me-

thods and results.









Chapter 2: Isotopic Tracer Methods for

Nitrogen in Soil-Plant Systems

The recovery of fertilizer N applied to a given soil is

commonly determined by both difference (non-tracer) and

isotope tracer methods. Results from these two methods

(Kowalenko 1980) may however lead to different conclusions.

The difference method gives a measure of the net recovery of

fertilizer N after several processes (denitrification,

nitrification, ion exchange, leaching, immobilization, etc.)

have exchanged with and transformed the fertilizer N.

Broadbent and Carlton (1980) have shown that the assumption

in difference method calculations that plant uptake of unla-

beled N is the same in fertilized and unfertilized plots is
15 14
not valid. In contrast, N and N tracer techniques give

a measure of the actual fate of the applied N (Kowalenko

1980) as well as provide more accurate results than the dif-

ference method (Broadbent and Carlton 1980). Tracer methods

permit actual measurement of fertilizer recovery as well as

the distribution of recovered N in various chemical

fractions.

Nitrogen tracer techniques (Hauck and Bremner 1976) are

based upon the observation that naturally occurring N com-

pounds contain about 0.366 atom % 15N and 99.634 atom %
14
N. Addition of a fertilizer with an unusually high

(15N-enriched) or low ( 15N-depleted) concentration of 15N to

a soil-plant system can thus be used as a tracer. Measured

changes of 15N/ 14N ratios in samples removed from the system











provide a means to investigate transformations of the added

fertilizer. The magnitude of the change in the isotope

ratio R from the background level R can be used to calcu-

late the extent to which the tracer N has interacted with

and become part of the system. The percentage of N present

in a sample of soil or soil water initially from either

15N-depleted or 15N-enriched material can be calculated

using the relationship
R R
R R. 1)
o 1

where R. is the isotopic ratio for the tracer material.

Hauck and Bremner (1976) state that isotope tracer

methods offer much potential for studying ways to maximize

the efficiency of fertilizer nitrogen in crop production.

They state that movement of N into, within, and from soil

can accurately be obtained only by the use of 15N-depleted

or 5N-enriched fertilizers. Isotopically labelled ferti-

lizer is particularly needed in soil-plant systems

(Broadbent and Carlton 1980) which contain a background of

large amounts of indigeneous N. Although 15N-enriched ma-

terials have been used in most nitrogen tracer studies, use

of 15N-depleted material has gained in popularity due to the

lower cost of those materials. However, the use of 15N-

depleted materials (Hauck and Bremner 1976) is restricted to

experiments where excessive dilution does not occur in the

soil-plant system. Thus studies of plant uptake or movement











of applied nitrogen should be performed for single- rather

than multiple-seasons.

Kowalenko (1980) used N-enriched (NH4)2SO4 to inves-

tigate transport and transformation of fertilizer NH in

eight micro-size (20-cm diameter) fallow field plots of a

sandy soil in Canada. The 15N-enriched (5.5% enrichment)

fertilizer was applied to the soil at a rate of 184 kg N/ha

in June 1977. Net fertilizer recoveries (as determined by

differences between extracted N in fertilized and unfer-

tilized plots) from the upper 75 cm of the soil profile were

determined to be 117 and 19% of that originally applied,

respectively, for 35 and 102 days after application of the

fertilizer. Measurements of total 15N concentrations

however revealed that recoveries of fertilizer N were ac-

tually 73 and 25% for 35 and 103 days after fertilization.

Almost all (93%) of the fertilizer NH4 initially applied was

nitrified within 35 days. If the nitrification was assumed

to have occurred at a constant average rate, then that rate

for this sandy soil would have been 4.89 kg N/ha/day which

was considerably higher than a rate of 2.30 kg N/ha/day ob-

tained previously by the same author for a clay loam soil.

Both soils had been in fallow prior to the experiments such

that neither soil had crop residues to influence microbial

processes. Although relatively low organic matter content

of the sandy soil suggested limited microbial activity,

tracer data showed the microbial process for nitrification

to be rapid. Leaching, denitrification, clay fixation of







8


+
NH4, mineralization, and immobilization all were important
+
in the transport and transformation of fertilizer NH4 ap-

plied to a sandy soil.










Chapter 3: Experimental Methods

and Procedure

An experimental citrus grove originally developed

(Knipling and Hammond 1971) in 1970 on 9 hectares of

flatwood land at the University of Florida Agricultural

Research Center (A.R.C.) near Fort Pierce was selected as

the location for this investigation. The location of the

experimental site as well as the distribution of Spodosols

over the land area of Florida is shown in Fig. 1. The grove

was established as a cooperative venture between the Univer-

sity of Florida Institute of Food and Agricultural Sciences

(IFAS) and the Agricultural Research Service (ARS) of the

United States Department of Agriculture. The IFAS-ARS

effort was designated as the Soil, Water, Atmosphere, and

Plant Relationships Project (SWAP). The original purpose

for the SWAP grow was to evaluate the influence of deep

tillage, with and without incorporation of. limestone into

the profile, upon subsurface drainage and growth of citrus

in a Spodosol. The SWAP grove was chosen as the site for

this work partially because of the background information on

chemical and physical characteristics that has been accu-

mulated (Mansell et al. 1977; Mansell, Wheeler and Calvert

1980; Calvert et al. 1981; Rogers et al. 1977) for the acid,

sandy soils in the grove. Another reason for that choice

was the fact that it is one of the best designed Coastal

Plains experimental drainage sites in the Southeastern

United States.










Figures
Fig. 1: Map of Florida showing major land areas of
Spodosols and the location of the experimental
site.


Location of Spodosols in Florida


Spodosols eD


Pierce










Three 0.5-hectare plots of citrus were selected for

this research. One plot was selected for each of the three

soil profile modification treatments: (i) shallow-tilled

(ST) to 15-cm depth, (ii) deep-tilled (DT) to 105-cm depth,

and (iii) deep-tilled (DTL) to 105 cm depth with an initial

incorporation of 56 metric tons per ha of dolomitic lime-

stone into the profile. All treatments received annual

applications of 2.24 metric tons per ha of limestone to the

soil surface. During the initial deep tillage operation, a

trenching machine incorporated and mixed spodic and underly-

ing sandy clay loam material with sandy soil material from

the A horizons. The primary soil type at the site is

Oldsmar fine sand (a member of the sandy siliceous, hyper-

thermic family of Alfic Arenic Haplaquods). In undisturbed

profiles, the A horizon of the acid sand has an average

(Calvert et al. 1981) depth of 82 cm and contains about 1%

organic matter. Underneath the A horizon a nearly imperme-

able spodic layer ranging in thickness from 10 to 20-cm con-

tains about 3.5% organic matter. A layer of sandy clay

loam, also with low permeability to water, occurs beneath

the spodic horizon.

Surface drainage in each plot was provided by a system

of elevated beds (38 cm height of bed.crown above the bottom

of water furrows) separated by parallel water furrows. The

width of the beds as measured from centers of adjacent water

furrows was 15.2 m. During very intense rain storms any

surface drainage water from each plot was removed by











collection ditches at the end of the water furrows.

Subsurface drainage was provided by a system of 10-cm

corrugated plastic tubes buried beneath the soil surface at

an average depth of 107 cm and spaced 18.3 m apart. The

drain tubes were located perpendicular to the elevated soil

beds. Two rows of citrus 7.6-m apart with a spacing of

4.6 m between trees were planted along the top of each bed

in 1970. Water from the center tube in each plot discharged

into a concrete manhole where flow was measured with 30.5

cm, 30 degree, V-notch weirs and Stevens type F, Model 68

waterstage recorders. Pensacola bahiagrass was seeded to

each plot in 1970. A strip of surface along each tree row

was maintained bare of bahiagrass and weeds.

In order to investigate the fate of N in fertilizer ap-

plied to these coarse-textured acid soils, 15N-depleted

NH4-N was applied as (NH4)2S04 in an otherwise complete fer-

tilizer to shallow-tilled (ST) and deep-tilled (DT and DTL)

plots. On June 5 (Julian Day 157) of 1980, 1N-depleted

8-2-8 (% N % P205 % K20) fertilizer was applied at a

rate of 115 kg N/ha to selected areas (The shaded area in

the plot diagram in Fig. 2 indicates the fertilized area.)

for each of ST, DT, and DTL plots. A batch of the isotopic-

ally labelled fertilizer (625 kg) was prepared by mixing 232

kg of 15N-depleted (NH4 2SO4 (courtesy of Dr. R. D. Hauck,

Tennessee Valley Authority, Muscle Shoals, Alabama) with 63

kg of ordinary superphosphate, 84 kg of muriate of potash,

and 246 kg of powdered dolomite. Isotopically labelled










fertilizer was broadcast by hand to the elevated beds (but

not to the water furrows) in the shaded area shown in the

plot diagram (Fig. 2) centered about the drain field for the

center drain in each of the three plots. Dots in Fig. 2

designate the locations for citrus trees. Approximately

0.22 ha in each plot received labelled fertilizer. The re-

mainder of each plot was fertilized with the same rate of

non-labelled 8-2-8 fertilizer.

Irrigation was applied immediately after the fertilizer

application to prevent volatilization of NH 4-N due to previ-

ous annual applications of limestone to the surface soil.

One gallon cans containing open bottles of.sulfuric acid

were placed at selected locations in the treatment plots in

order to trap any gaseous NH3 released to the atmosphere.

The acid solutions were later analyzed for N concentration.

During an 8-month period following the application of

15N-depleted fertilizer, concentrations of NO3-N and NH4-N

were determined in subsurface drainage water, samples of

soil solution, in soil cores, and in selected leaf samples

from citrus trees. When nitrogen concentrations in water,

soil and tissue samples were sufficiently high, ratios of

15N-to- 14N concentrations were determined. The central

drain in each of the selected ST, DT and DTL plots was con-

tinually monitored for water flow rates (Data provided by

Dr. J. S. Rogers, Agricultural Engineer, USDA, University of

Florida) and intermittently monitored for water quality sam-

ples. Automated water samplers (ISCO Model 391) were used








Fig. 2: Schematic diagram (scale: 1-to-695 cm) of experi-
mental citrus plots for ST, DT, and DTI, treat-
ments.





PLOT DIAGRAM


1 -1a* e -
.T 11l


0 0
0 0 0 0
jf 7111 L7RIL#A~ ..&~.fr
uh11i:~~.. ~.:H0r~
~
S
I ,.
6 Oil
0 0 0 0
: :1. :~ : -
.1 0 0
0 *1* ole SI. 010 0.1


DRAIN
-TUBE
OUTLETS


m


54.9
m


T
183

I
4.6
T


15.2 water / -f76
m furrow m
/^ A-










to take samples from weirs at the outflow of each drain.

Samples were taken most frequently during periods of drain

discharge after rainfall events. All water samples were

frozen and stored for later analysis.

Soil solution samplers were constructed of medium po-

rosity Pyrex glass discs permanently glued to the bottom end

of various lengths of 3/4 inch PVC pipe. Rubber stoppers

were placed in the top end of the pipes to provide an air

pressure seal. The fritted disc samplers were placed at 4

soil depths 60, 75, 90, and 105 cm and at 6 horizontal

distances from the central drain 5, 50, 100, 350, 500, and

900 cm along one row of citrus in each plot. Hand-

operated air pumps were used to apply approximately 80 cm of

water to each solution sampler during sampling periods. The

samplers were stoppered for 4 to 6 hours to enable extrac-
3
tion of 25 to 100 cm of water depending upon the water con-

tent of the soil. Solution samples were taken 57 (August

1), 69 (August 13), and 76 (August 20) days after fertilizer

application. Samples were frozen and stored before

analysis.

Soil water suction was determined from mercury

manometer-type tensiometers located at 4 soil depths 30,

60, 90, and 105 cm and 6 horizontal distances 5, 50,

100, 350, 500, and 900 cm from the central drain in each

treatment. The tensiometers were located along one row of

trees in each plot. Water inside the tensiometers made hy-

draulic contact with water in soil pores through a porous











ceramic cup at the bottom of each tensiometer. Mercury

manometers permitted measurements of water suction of soil

water.

A hydraulic coring machine was used to take continuous

soil cores down to a depth of 70 cm in the ST plot and to a

depth of 100 cm in each of DT and DTL plots. The holes in

the soil profile were later refilled with inert builders

sand to prevent preferred water flow to drain tubes. Each

core was divided into subsamples corresponding to depths of

0-8, 9-23, 24-38, 39-53, 54-70, 71-84, and 85-100cm. The

subsamples of soil were frozen and stored until analysis

could be perform. Twelve cores were removed from each of

the tree plots on dates corresponding to 12 (June 15-16), 42

(July 16-17), 75 (August 19-20), and 134 days (October

16-17) after the initial surface application of labelled

fertilizer. Six of the twelve cores from each plot were

taken near each drain and the remaining cores were taken

from 450 cm lateral distance from each drain. A total of

912 soil samples were collected and analyzed for this inves-

tigation.

Leaf samples were collected from six rootstock vari-

eties for both grapefruit and orange trees once during July

and again in October. The samples were ground in a Wiley

mill and frozen until analysis could be performed.

All water samples were thawed and analyzed for pH using

a glass electrode, NO3-N concentration using a nitrate spe-

cific ion electrode, NH4-N concentration using the phenolate











(EPA 1976) method, and soluble salts using an electrical

conductivity meter. The remaining volume of each sample was

measured and analyzed for NH4-N and NO3-N using a

macrokjeldahl procedure (Bremner 1965). Soil samples were

also thawed, extracted with water and 1 N KC1 and both

extracts analyzed similarly to that for the water samples.

Leaf samples were dissolved in concentrated H2SO4 for total

analyses and the solutions analyzed for NO3-N and NH4-N.

After titration, samples containing more than 1.0 mg N were

redistilled by microkjeldahl apparatus into 5 ml of 4% boric

acid and 1 ml of 0.08 N sulfuric acid. Reagent grade

ammonium sulfate was added to all other samples in order to

increase the N content to 1.0 mg per sample. These solu-

tions were analyzed in duplicate for isotopical ratios
15N/ 14N by an automated mass spectrometer at the Los Alamos

National Laboratory near Los Alamos, New Mexico. Isotopic

ratios for reagent grade (NH4)2SO4, 15N-depleted (NH4)2SO4,

and for soil samples taken from soil that had only been fer-

tilized with unlabelled N materials were used along with the

ratios from the samples to calculate the percentages of N

derived from the N-depleted fertilizer. Approximately

1300 samples of water, soil extracts, and leaf extracts were

analyzed for isotopic N ratios. Concentrations of labelled

and total N were expressed as mg N/liter of effluent.











Chapter 4: Results and Discussion

The shallow tillage (ST) treatment of the Spodosol

located at the SWAP citrus grove is representative of agri-

cultural management practiced on vast areas of similar acid,

sandy soils in Florida. Water and nutrient retention capac-

ities of soil in the ST plot (Mansell et al. 1977) have been

demonstrated to be very limited. Recommended cultural prac-

tices for agricultural crops growing on this and similar

soils include split application of fertilizer throughout the

growth season and frequent application of relatively small

quantities of water during periods of drought or limited

rainfall. Because of subsurface horizons with low

permeability for water and relatively flat terrain, artifi-

cial drainage is often needed to maintain an aerobic root

zone during periods of high rainfall. Because of the high

values of hydraulic conductivity when these soils are

water-saturated (Mansell, et al. 1980) rapid rates of drain-

age tend to enhance leaching losses of fertilizer nutrients

such as nitrogen.

Deep tillage (DT and DTL treatments) of this soil has

been observed (Mansell et al. 1977) to increase the nutrient

retention capacity by increasing the cation.exchange capaci-

ty, to increase the water retention capacity by incorporat-

ing colloids from an underlying Spodic horizon, and to

decrease rates of nutrient leaching loss by decreasing

drainage discharge rates. Although this treatment is expen-

sive and requires the presence of a subsurface layer high in










colloid content, selective deep tillage over limited soil

zones only in the vicinity of crop rows offers a potential

conservation practice for applied irrigation water and fer-

tilizer.

As expected, the subsoil pH in the ST plot was observed

during this study (Fig. 3) to be considerably lower than the

DTL plot but higher than the DT plot. The pH of the top

9 cm of the DTL soil profile was relatively uniform with a

value of about 6.1. The higher pH of the DTL soil is attri-

butable to the original incorporation of limestone prior to

the deep tillage operation. Annual applications of lime-

stone however were shown to result in pH values near 6.0 for

the surface soil of all three treatments. These acid, sandy

soils have a predominance (Fiskell and Calvert, 1975) of

variable-charge colloids such as organic matter, iron

oxides, and aluminum oxides; therefore application of lime-

stone tends to raise the cation exchange capacity. The

higher pH of the surface soil of ST and DT plots and the

higher pH of both surface and subsurface soil zones in the

DTL plot should therefore preferentially favor microbial

nitrification of applied fertilizer NH -N as opposed to the

more acid subsoil conditions of ST and DTL plots.

During the period of this study soil water content

generally was higher between 60 to 90 cm depths (Fig. 4)

than in the uppermost 30 cm of the profiles for all three

tillage treatments. During periods of drainage discharge

this was particularly true and the subsoil of the deep-











Fig. 3: Distributions of soil pH with depth measured on

June 18, 1980 for ST, DT, and DTL tillage

treatments.


0

10


20

30-


U 40-

- 50-
60
6O-

70
S70-


80-

90


SOIL pH
O 5 I __6,0


100-


S



0


DTL-









Fig. 4: Distributions of soil water content (% by weight)
with depth on June 15, 1980 for ST, DT, and DTL
tillage treatments.


SOIL WATER
2 4 6 8 10 12


00

10

20

30

E40


0-
LC
060
_ 7


CONTENT(/o by wt)
14 16 18 20 22 24 26 28


JULIAN DAY167
(15 JUNE 1980)













\ \4- DTL


DT


100











tilled plots had generally higher water contents than did

the shallow-tilled plot. The water-unsaturated condition of

the surface soil in all of the treatments however should

tend to enhance nitrification of applied fertilizer. Previ-

ous research (Mansell et al. 1977) has also shown that the

actual water retention capacity of the shallow tilled soil

in the ST plot is generally much less than that in deep-

tilled soil in DT and DTL plots.

Soil water pressure head values for a 10-day period in

October also indicate that the soil at 90 cm depth was drier

in the ST (Tables 1, 2 and 3) plot versus the DTL plot.

Pressure head values were generally slightly negative in ST

whereas the values were positive in DTL. Negative pressure

heads in the DT soil however indicated that DT soil was

drier than the DTL soil.

Volatilization losses from applied fertilizer (Table 4)

averaged approximately 0.31 kg/ha of NH4-N for all 3 plots

during the first two months after application of fertilizer.

Losses were 0.24, 0.25, and 0.42 kg/ha of NH4-N respectively

for ST, DT and DTL plots. Based upon this information we

conclude that volatilization losses of N were sufficiently

small to be insignificant.

The original intent of this investigation was to evalu-

ate leaching loss of fertilizer N during a typical summer

rainy period when large quantities of water moved through

the soil to subsurface drains. However, an unexpected

drought occurred during 1980 such that the summer and fall












Table 1: Soil water pressure head (cm of water) at 0.9 m

depth and lateral distances of 0, 0.5, 1.0, 3.5,

5.0, and 9.0 m from the center drain tube in the

ST plot for 10 days in October 1980.


Lateral
Date Distances:


October

It


if


II

I"





"I

I"


0



-6

-9

-8

-3

-6

+14

-10

-14

-13

-16


0.5 1.0

- (cm of

-8 0

-8 0

-10 -1

-8 0

-8 0

+24 +2

-12 -4

-14 -6

-15 -6

-17 -8


3.5

water)

0

0

+1

-1

-1

+9

-2

-4

-4

-6


5.0 9.0m



S- -12

- -12

- -12

- -1

- -12



- -12

-12

-13

-16













Table 2: Soil water pressure head (cm of water) at 0.9 m

depth and lateral distances of 0, 0.5, 1.0, 3.5,

5.0, and 9.0 m from the center drain tube in the

DT plot for 10 days in October 1980.


Lateral
Date Distance:


0


0.5 1.0 3.5 5.0 9.0m


October 20

21

22

23

24

27

28

29

30

31


- - (cm of

-5 -15 -14

-5 -15 -15

-3 -15 -14

-3 -14 -14

-3 -15 -14

-2 -16 -14

-2 -15 -16

-2 -16 -18

-4 -16

-4 -20


water) -

-10 -6

-10 -10

--8

-- -9

--8

-10

-10

-12

-12

-14


-5

-5

-5

-5

-4

-7

-8

-10

-10

-10











Table 3: Soil water pressure head (cm of water) at 0.9 m

depth and lateral distances of 0, 0.5, 1.0, 3.5,

5.0, and 9.0 m from the center drain tube in the

DTL plot for 10 days in October 1980.



Lateral
Date Distance: 0 0.5 1.0 3.5 5.0 9.0m

October 20 +16 +8 +9 +16 +14 +16

21 +26 +16 +9 +16 +14 +16

22 +16 +8 +10 +14 +11 +16

23 +17 +8 +10 +16 +14 +17

24 +17 +6 +10 +14 +14 -

27 -9 -12 +9 0 -

28 +16 +3 +2 +8 +8 -

29 +14 +4 +10 +8 +6 -

30 +14 +12 +2 +8 +5 -

31 +14 0 0 +8 +4 -












Table 4.


Volatilization of NH3 from surface-applied

fertilizer during 1980 as measured by sorption of

NH4-N in containers (cross-sectional surface area
2
of 178 cm per container). Effective volatiliza-

tion losses of NH4-N are given as averages for

each plot.


Starting
Date


June

June

June

June

July

July


6

10

18

24

1

21

Total


Sorption
Period

(days)

4

8

6

7

20

7


ST



0.0455

0.1219

0.0197

0.0191

0.0185

0.0185

0.2432


Treatment Plots
DT DTL

- (kg NH4-N/ha) -

0.0371 0.0680

0.1191 0.2427

0.0230 0.0331

0.0169 0.0208

0.0157 0.0197

0.0416 0.0404

0.2534 0.4247


These values were calculated by dividing measured

of NH4-N sorption per container ( g NH4-N/178 cm2)

conversion factor 1780.


Mean



0.0500

0.1612

0.0253

0.0185

0.0180

0.0337

0.3067


amounts

by the











periods received less rainfall than during normal years.

Total rainfall plus irrigation was only 84 cm during the 196

day (Fig. 5) period between June 18 and December 3. The

rainfall distribution (Figs. 6-11) was such that periods of

net drainage discharge were intermittent during the time of

this study. For the ST, DT, and DTL plots, discharge

occurred during the periods from June 21 to July 3 (Julian

Days 173 to 185), July 13 to August 2 (Julian Days 195-215),

November 10 to December 5 (Julian Days 315 to 340), and

December 20 to 30 (Julian Days 355 to 365). Water discharge

occurred from the drains for approximately 67 days of the

196 day period. Drainage rates were obviously highest for

the shallow tillage plot. Total drainage from ST was 30.7

cm (Fig. 6) which amounts to only 37% of the total amount of

rainfall plus irrigation. Thus roughly 63% of the input

water for the 196 day period was assumed to be used during

evapotranspiration. Total drainage amounts for DT and DTL

deep-tilled plots were however much less than for ST being

only 5.6 (7%) and 8.5 (10%) cm, respectively. Evapotrans-

piration was therefore assumed to account for roughly 93 and

90% of the input water for DT and DTL plots. The evapo-

transpiration estimates assume no vertical deep seepage

water loss and only small changes in soil water storage.

Water fluxes or drainage rates were much greater for

the shallow-tilled plot than the deep-tilled plots. For all

drains the maximum water fluxes occurred during November

(Figs. 9, 10, and 11) with values of 3.0, 0.6, and 1.5









Fig. 5: Accumulative amounts (cm) of rain and irrigation
between June 18 (Julian Day 170) and December 31
(Julian Day 366).







JULIAN DAYS


Z 84

< 72

06
c 60

Z
< 48

LJ 36


I 24

D 12

Sr)










Fig. 6: Accumulative drainage (m 3/ha) from the ST tillage

treatment over the period from June 18 to December

31, 1980.







JULIAN DAYS


2800

2400

r2O000

j 1600

1200

S800

S400

U o
(9










Fig. 7: Accumulative drainage (m3 /ha) from the DT
treatment over the period from June 18 to December
31, 1980.


E 560

-480
Z
< 400

> 320

240

> 160

Z) 80

I o
D-


JULIAN DAYS
180 200 220 240 260 280 300 320 340 360


DT SOIL









Fig. 8: Accumulative drainage (m /ha) from the DTL

treatment over the period from June 18 to December

31, 1980.


-
E840

S720
Z
<600

>- 480

(360
w
< 240

120

U
U. 0
<


JULIAN DAYS
180 200 220 240 260 280 300 320 340 360











Fig. 9: Drainage flux (m /ha/day) from the ST treatment

over the period from June 18 to December 31, 1980.








JULIAN DAYS


>& 28,
-o
284

\' 24
c2

O 20
E










Fig. 10: Drainage flux (m 3/ha/day) from the DT treatment

over the period from June 18 to December 31, 1980.






JULIAN DAYS
180 200 220 240 260 280 300 320 340 360


I -


DT SOIL











Fig. 11: Drainage flux (m3/ha/day) from the DTL treatment

over the period from June 18 to December 31, 1980.


140


120


I00


80


60


40


20


JULIAN DAYS

I 80 200 220 240 260 280 300 320 340 360








DTL SOIL







-ri~ s~. ^j ^


I -










cm/day respectively for ST, DT, and DTL plots. Drainage was

very small during the summer period for the deep tillage

plots.

Drainage outflow through subsurface tubes has been con-

sistently greater for all shallow tillage (ST) treatments at

the SWAP citrus grove than from both deep tillage (DT and

DTL) treatments. Mean annual drainage discharges over the

10 year period from 1971 through 1980 (Dr. J. S. Rogers.

1982. private communication.) for ST, DT, and DTL treat-

ments were 55.4, 33.3, and 28.9 cm.

As expected, the highest concentrations of N (the sum

of NH -N plus NO3-N) in the drainage water occurred during

the first 2 months after application (Figs. 12, 13, and 14)

of fertilizer. For ST drainage water, maximum N concen-

trations of 25 and 11.7 mg/l occurred during 25 and 48 days

after fertilization. For DT soil, maximum N concentrations

of 8.2 and 10.3 mg/l occurred during 20 and 27 days after

fertilization. For DTL soil, maximum N concentrations of

6.0 and 13.4 mg/l occurred 20 and 26 days after fertiliza-

tion. Thus N concentrations in drainage water were general-

ly higher in ST (Mansell, et al. 1980) than for the deep-

tilled plots (DT and DTL). The higher N concentrations in

the drainage is largely attributable to the low cation ex-

change capacity of this soil. Concentrations of N were

lower during the fall versus the summer for DT and DTL

drainage waters. This observation was true except for an

unexplainable peak that occurred during the latter part of

December.












Isotopic N analysis indicated (Figs. 12, 13, and 14)

that N from the 15N-depleted fertilizer accounted for as

much as 50% of the N in drainage water that occurred during

the summer. This was true for all three plots. The

percentage of N attributable to the labelled fertilizer

decreased with time during the season. For example on

November 15 (163 days after application of the fertilizer)

percentages of N due to the N-depleted fertilizer were 20,

17, and 25% respectively for ST, DT and DTL soils. This

data implies that much of the N being discharged through the

drains was due to residual N from mineralization of soil

organic matter and from previous applications of fertilizer.

Values of nitrogen flux through the drains were cal-

culated by multiplying water flux and N concentrations for

each day that drainage occurred. This data (Figs. 15, 16,

and 17) indicate that rates of N discharge were greater in

the summer than in the fall for the shallow tillage treat-

ment, but were greater in the fall than in the summer for

the deep tillage plots. For the ST plot, rates of nitrogen

leaching losses as high as 0.850 kg/day were observed for 2

drainage events during the summer. Much smaller rates of

0.012 and 0.015 kg/day were observed for single summer

drainage events in each of DT and DTL plots. This informa-

tion further substantiates (Mansell et al. 1977) the effec-

tiveness of deep tillage in decreasing N leaching loss in

these acid sandy soils. Measurements of the area beneath

the curves in Fig. 15 shows that total leaching losses of






37


Fig. 12: Concentration (mg/1) of extracted and labelled N
(NH4-N plus NO3-N) in drainage water from the ST

plot over the period from June 18, 1980 to

February 14, 1981.





Julian Days










Fig. 13: Concentration (mg/1) of extracted and labelled
N(NH4-N plus NO3-N) in drainage water from the DT
plot over the period from June 18, 1980 to
February 14, 1981.





Julian Days
180 2QO 2?0 240 20O 210 390 3?0 340 3C0 3~Q 4QO

24
20O
13
12
E
10
DT Soil
0 NO3-N NH4-N
S.-- Labelled N
c6






0. ".... -----.
2-










Fig. 14:


Concentration (mg/1) of extracted and labelled

N(NH4-N plus NO3-N) in drainage water from the DTL

plot over the period from June 18, 1980 to

February 14, 1981.


Julian Days
180 200 220 240 260 20 3Q00 3O 3 40 300 3$0 400


24
20
^3-
-12
E

10.
O
0n
L-


U
C 4-
0-
U
2

0


DTL Soil
S-- N03-N + NH4-N
.-- Labelled N











Fig. 15: Flux (g/ha/day) of extracted and labelled N(NH4-N

plus NO3-N) in drainage water from the ST

treatment over the period from June 18 to December

31, 1980.






JULIAN DAYS

180 200 220 240 260 280 300 .320 340 360


> 840
-o |
c( 720
N ST SOIL
600 N03--N+NH4-N
X *LABELLED N
..I
480 1
LL

LI
Z 360
(D It
O240




0 ---- --. ...-










Fig. 16: Flux (g/ha/day) of extracted and labelled N(NH4-N
plus NO3-N) in drainage water from the DT

treatment over the period from June 18 to December

31, 1980.






JULIAN DAYS
180 200 220 240 260 280 300 320 340 360


>, 28

0 24

Co 20 DT SOIL
X eNO3-N+NH4-N
16 *LABELLED N
U-.
Z 12

O
IZ 8
I-l f











Fig. 17: Flux (g/ha/day) of extracted and labelled N(NH 4-N

plus NO3-N) in drainage water from the DTL

treatment over the period from June 18 to December

31, 1980.






JULIAN DAYS
180 200 220 240 260 280 300 320 340 360


> 12


c 96
-c
C 80 DTL SOIL
*N03-N+NH4-N
S64* LABELLED N
b-
Z 48
C!
0
32
I 6L
Z 16 ..


M -.










total and labelled N were 9.10 and 4.34 kg/ha, respectively,

for the first 196 days after application of the fertilizer.

Thus 48% of the N discharged from the ST drain was attribut-

able to the 15N-depleted (NH4)2SO4. Total leaching losses

of N in drainage from the deep-tilled plots were extremely

small.

Distributions of total and labelled NH4-N with soil

depth are shown for each of the three plots in Figs. 18, 19,

and 20 for dates corresponding to 12, 42, 75, and 134 days

after application of the fertilizer. For the deep-tilled

plots, NH4-N contents in the soil profile were low in com-

parison to those for the shallow-tilled plot. Total NH4-N

quantities in the soil 12 days after fertilization (Table 5)

were 142.0, 25.9, and 71.4 kg/ha, respectively, for ST, DT,

and DTL plots. After 42 days from fertilization quantities

of NH4-N had decreased to 57.3, 8.5, and 19.3 kg/ha for ST,

DT and DTL plots. This data suggests that much of the NH -N

applied to the deep-tilled plots was either converted to

NO3-N during nitrification and/or absorbed by plant roots

from citrus or bahiagrass since volatilization and leaching

losses of N were extremely small. Distributions of total

and labelled NO3-N in the soil (Figs 21, 22, and 23) do in

fact suggest that nitrification of applied NH -N occurred.

A sandy soil in British Columbia, Canada with relatively low

organic carbon content was shown by Kowalenko (1980) to sup-

port surprisingly high microbial activity for N transfor-

mations. Between 12 and 42 days after fertilization, quan-















Fig. 18: Distributions of extracted and isotopically

labelled NH4-N in the ST soil profile during 1980

for dates corresponding to 12, 42, 75 and 134 days

after fertilization.










NH4-N (pg/g) NH4-N (pg/g)


NH4-N (pg/g)
5 10 15 20










19 AUGUST
ST

X EXTRACTED
e LABELLED


1 20-
o /

S30
I-
W 40 16 JULY
a ST
-50
50 X EXTRACTED
o0 LABELLED
60

70
NH4-N (pg/g)
00 5 10 15 2


I0E
E
220


30

40 16 OCTOBER
SST
50 X EXTRACTED
e LABELLED
60

70


"' 20


a:30
I-
LW 40
0
-J
550
CO
60

70

O


I0
E
220

-30
ai.

40
-J
o
S50

60

70















Fig. 19: Distributions of extracted and isotopically


labelled NH4-N in the DT soil profile during 1980


for dates corresponding to 12, 42, 75, and 134


days after fertilization.


NH4-N (pg/g)

0O 5 10 15 20


10

20

30

40 16 JUNE
DT
50 X EXTRACTED
LABELLED
5-J
60

70

80

90

100
NH4-N (pg/g)
0 5 10 15 20
0
10




"2 30
U !

40 19 AUGUST ;
t- DT

50 X EXTRACTED
LABELLED
560

70

80


90

100 .


NH4-N (pg/g)
0 5 10 15 20

10


20

30

40 16 JULY
DT
a-
W 50 X EXTRACTED
LABELLED
-60
70
TO

80

90

100
NH4-N (pg/g)
0 5 10 15 20


10

20 /

30

o 40 16 OCTOBER
C x DT
L 50X EXTRACTED
LI LABELLED
j 60

70

80

90

100L
















Fig. 20: Distributions of extracted and isotopically


labelled NH 4-N in the DTL soil profile during


1980 for dates corresponding to 12, 42, 75, and


134 days after fertilization.


NH4-N (pg/g)
0 5 10 15 20
0

10 /
/ X
20 /




40 / 16 JUNE
I- ODTL
S0-
S50 X EXTRACTED
e LABELLED
560

70

80

90

100
NH4-N (pg/g)
0 5 10 15 20

SIO
10

20

30

S40 19 AUGUST
I- DTL
. 50 X EXTRACTED
a LABELLED U
.-j
Z 60
om
70 U

80

90

100


NH4-N (pg/g)
0 5 10 15 20


10

20

30
u

40 16 JULY
I'- DTL
S50 X EXTRACTED
e LABELLED
..J
60

70

80

90

100
NH4-N (pg/g)
0 5 10 15 20
0

10

20

30

40 16 OCTOBER
DTL

50 X EXTRACTED
S LABELLED
j 60

70

80

90

I00






47



Fig. 21: Distributions of extracted and isotopically

labelled NO3-N in the ST soil profile during 19S0

for dates corresponding to 12, 42, 75, and 134

days after fertilization.






NO3-N (pg/g) N03-N (pg/g)
0 5 10 15 20 0O 5 10 15 20 25


10. // 10

20/ E 20\

30 XI 30 x

W 40 \ 6 JUNE J 40- 6 JULY
Smx ST ST
I
S501 I X EXTRACTED O 50 X EXTRACTED
C LABELLED LABELLED
60- 60*

70 70
N03-N (pg/g) NO-N (pg/g)
0 5 10 15 20 00 5 10 15 20


10 10

20 E 20

S30 30

L 40 19 AUGUST W 40 16 OCTOBER
0 ST ST
o 50 X EXTRACTED 0 50 X EXTRACTED
0 LABELLED C LABELLED
60 60

7n















Fig. 22:


Distributions of extracted and isotopically


labelled NO3-N in the DT soil profile during 1980


for dates corresponding to 12, 42, 75 and 134 days


after fertilization.


NO3-N (pig/g)
.5 10 15


NO3-N (pg/g)
) 00 5 10 15. 20


10

20

E 30

S40 16 JULY
DT
W 50 X EXTRACTED
LABELLED
5 60

70

x
80

90

100
N03-N (pg/g)
) 0 5 10 15 20


10

20

E 30

I40 I 6 OCTOBER
I-" DT
0-
U 50 X EXTRACTED
e LABELLED
.J
S60

70

80

90

100


E 30

S40
0-
S50

O 60


10

20

E 30

I40
I-
0-
J50
-J
S60

70

80








49




Fig. 23: Distributions of extracted and isotopically


labelled NO3-N in the DTL soil profile during 1980

for dates corresponding to 12, 42, 75, and 134

days after fertilization.






NO-N (pg/g) N03-N (pg/g)
0 5 10 15 20 00 5 10 15 20 25
0 X
go. / o ll

20 / 20

E 30 E 30 x

I 40 16 JUNE 3 40 16 JULY
I DTL D OTL
W 50 X EXTRACTED W 50 X EXTRACTED
LABELLED LABELLED
-jI



X
so o I

90 90/

100 100
N03-N (pg/g) NO3-N (pg/g)
O0 5 10 15 20 O0 5 10 15 20


10 10
X
20 20

E 30 30

: 40 1 9 AUGUST 40 16 OCTOBER
OTL DTL
SX 0-
S50 EXTRACTED W 50 EXTRACTED
LABELLED LABELLED
-j
60 5 60

70 70











tities (Table 5) of total NO3-N in the soil increased con-

siderably for all treatment plots in this study but less

dramatically for the DT soil. After 42 days, the NO3-N

quantities decreased with time for all treatments probably

due to uptake by plant roots.

Quantities of the soil N due to the applied 15N-

depleted fertilizer are given in Table 6. Twelve days after

fertilization the sum of the labelled NH4-N and NO3-N in the

soil represented 105, 28, and 54% of the fertilizer, respec-

tively, for ST, DT and DTL plots. The low recoveries of la-

belled N in the deep-tilled soils imply that root uptake of

N was much faster during the first 12 days after fertilizer

application than in the ST soil. This deduction assumes the

extraction procedure was 100% effective for DT and DTL

soils. The ratios of NO3-N to NH4-N for the labelled N in

the soil were 0.2, 1.5, and 0.5 respectively for ST, DT, and

DTL plots. The larger ratios for the DT and DTL plots

suggest that nitrification occurred more rapidly in the

deep-tilled soils than in the shallow-tilled soil during the

first 12 days after fertilization. After 42 days of elapsed

time these ratios were 1.0, 7.8, and 7.2 for ST, DT, and DTL

plots. These dramatic increases of the NO3-N/NH4-N ratios

over a 30-day period indicate that nitrification occurred in

all soils. After 134 days of elapsed time, quantities of

labelled N in all soils were less than 3% of the applied

fertilizer, and the ratios of NO3-N/NH4-N for ST, DT and DTL

plots were 1.9, 0.6, and 0.9 respectively.











Table 5. Mean quantities of extracted NH4-N and NO3-N in

soil profiles from ST, DT, and DTL tillage treat-

ments on dates corresponding to 12, 42, 75, and

134 days after application of 115 kg/ha of

15N-depleted NH4-N on June 5, 1980.


NH -N

Time ST DT DTL
(elapsed
days)----- (kg/ha)-----

12 142.0 25.9 71.4

42 57.3 8.5 19.3

75 12.2 5.6 14.4

134 10.5 20.5 17.4


NO3-N

ST DT DTL

----(kg/ha)-----

40.8 50.6 33.4

96.3 58.1 77.3

23.9 26.0 30.7

19.7 12.1 9.5


NH4-N + NO3-N

ST DT DTL

------(kg/ha)----

182.8 76.5 104.8

153.6 66.6 96.6

36.1 31.6 45.1

30.2 32.6 26.9











Table 6. Mean quantities of isotopically labelled NH4-N and

NO3-N in soil profiles from ST, DT, and DTL

tillage treatments on dates corresponding to 12,

42, 75, and 134 days after application of 115

kg/ha of 15N-depleted NH4-N on June 5, 1980.


NH4-N

Time ST DT DTL
(elapsed
days) ----- (kg/ha) -----

12 102.6 13.2 42.7

42 11.9 0.8 2.0

75 1.8 1.0 1.1

134 1.1 1.9 0.9


NO -N

ST DT DTL

-----(kg/ha)---

17.6 19.2 19.7

11.5 6.2 14.3

3.7 5.6 3.5

2.1 1.1 0.8


NH -N + NO -N

ST DT DTL

-----(kg/ha)-----

120.2 32.4 62.4

23.4 7.0 16.3

5.5 6.6 4.6

3.2 3.0 1.7











The distributions of total NO3-N content in the soil

(Figs. 21, 22 and 23) imply that some downward movement oc-

curred through the profile during leaching but that a sink

possibility uptake by plant roots had extracted fertilizer N

from the soil of each of the tillage treatments. Dis-

tributions of the sum of NO3-N and NH4-N (Figs. 24, 25 and

26) also support that hypothesis. Results from the drainage

discharge of N however show that the leaching loss was quite

small during the 6-month period of this study. Previous in-

vestigators (Calvert et al. 1977) have shown that citrus

root systems are generally shallower and less extensive in

the profile for shallow-tilled versus deep-tilled DTL soil.

Thus one should expect more extensive N uptake from citrus

trees growing in DTL soil profiles compared to ST.

Due to the lack of rain and the corresponding lower

water content of the shallow-tilled soil compared to the

deep-tilled soil, only a very limited number of soil

solution samples were collected during August. Despite the

very small number of samples reported in Tables 7-9, NO3-N

was definitely the predominant form of N in the ST soil so-

lution. The decrease of the geometric mean concentration of

labelled N with time over this 19-day period implies that

nitrification occurred. The decrease of the geometric mean

concentration of labelled N implies that plant uptake of N

actively occurred during this period. Samples were more

easily extracted from the generally wetter DT and DTL soils

as shown by the N concentrations in Tables 10-15. Geometric














Fig. 24: Distributions of extracted and labelled NH 4-N plus

NO -N in the ST soil profile during 1980 for 12,

42, 75, and 134 days after fertilization.


NO3-N NH4-N (pg/g)


-N (pg/g)


NO3-N* NH4-N (pg/g)
10 20 30 40


NO3-N NH4-N (pg/g)
10 20 30 40











16 OCTOBER-ST

X EXTRACTED
LABELLED
















Distributions of extracted and labelled NH4-N plus


NO3-N in the DT soil profile during 1980 for 12,


42, 75, and 134 days after fertilization.


NO3-N NH4-N (pg/g)
0 10 20 30 40


10
/



20 /


30


40 16 JUNE-OT

X EXTRACTED
50 LABELLED


60


70


80


90

N03-N NH4-N (pg/g)
0 0 20 30 40


10


20


30


40 19 AUGUST- DT

X EXTRACTED
50 *LABELLED


60


70


80


90


NO3-N. NH4-N (pg/g)


Fig. 25:
















Fig. 26: Distributions of extracted and labelled NH -N plus


NO3-N in the DTL soil profile during 1980 for 12,


42, 75, and 134 days after fertilization.


N03-N NH4-N (pg/g)


N03-N. NH4-N (pg/g)
/10 20 30 4









x

S9 AUGUST DTL

X EXTRACTED
LABELLED


N03-N NH-N (pg/g)
10 20 30 4











16 OCTOBER-DTL

X EXTRACTED
LABELLED


30
E

7 40
I.-

, 50
-.J
0
60


70


80


90



0


10


20


30
E
40
I-
S40

Q 50

0
a) 60


70


80


90











Table 7. Concentrations (mg/1) of total and labelled N in

soil solution samples from 4 depths and 6 lateral

distances from the central drain in the ST plot on

August 1, 1980 (Julian Day 214). Geometric means

are shown in parentheses.


soil Lateral Total NH4-N (0.15 mg/1)
Soil Lateral
Depth Distance(cm): 5 50 100 350 500 900
(cm)

60 -

75

90 -

105 0.38 0.06




Soil Lateral Total NO3-N (12.90 mg/1)
Depth Distance(cm): 5 50 100 350 500 900
(cm)

60 -

75

90 -

105 10.60 15.70



Soil Lateral Labelled NH -N + NO 3-N (7.89 mg/1)
Soil Lateral
Depth Distance(cm) 5 50 100 350 500 900


(cm)

60

75

90

105


0- 13.26


4.70










Table 8. Concentrations (mg/1) of total and labelled N in

soil solution samples from 4 depths and 6 lateral

distances from the central drain in the ST plot on

August 13, 1980 (Julian Day 226). Geometric means

shown in parentheses.


_Soil Lateral Total NH4-N (0.32 mg/1)
Soil Lateral
Depth Distance(cm): 5 50 100 350 500 900
(cm)

60 0.12 7.97 0.95

75 0.01 0.33

90 -

105 0.34 -



Total NO3-N (12.78 mg/1)
Soil Lateral
Depth Distance(cm): 5 50 100 350 500 900
(cm)

60 11.00 21.0 3.80

75 84.0 10.00

90 -

105 5.90 -



Labelled NH4-N + NO3-N (0.52 mg/1)
Soil Lateral
Depth Distance(cm) 5 50 100 350 500 900


(cm)

60

75

90

105


0.65 0.16

- 0.35


0.52










Table 9. Concentrations (mg/1) of total and labelled N in

soil solution samples from 4 depths and 6 lateral

distances from the central drain in the ST plot on

August 20, 1980 (Julian Day 233). Geometric means

shown in parentheses.


soil Lateral Total NH4-N (0.11 mg/1)
Soil Lateral
Depth Distance(cm): 5 50 100 350 500 900
(cm)

60 0.98 0.25

75 0.01

90 -

105 0.07 -




Soil Lateral Total NO3-N (1.19 mg/1)
Depth Distance(cm): 5 50 100 350 500 900
(cm)

60 1.2 2.70

75 3.30

90 -

105 0.19 -



soil Lateral Labelled NH -N + NO3-N (0.38 mg/1)
Soil Lateral
Depth Distance (cm) : 5 50 100 350 500 900


(cm)

60

75

90

105


0.54 0.74

- 0.83


0.06










Table 10. Concentrations (mg/1) of total and labelled N in

soil solution samples from 4 depths and 6 lateral

distances from the central drain in the DT plot on

August 1, 1980 (Julian Day 214). Geometric means

are shown in parentheses.


Total NH -N


(0.50 mg/1)


Soil Lateral
Depth Distance(cm): 5 50 100 350 500 900
(cm)

60 -

75 -

90 0.60 0.49 0.12 0.33

105 0.83 0.93 0.48 0.91



soil Lateral Total NO3-N (8.79 mg/1)
Soil Lateral
Depth Distance(cm): 5 50 100 350 500 900
(cm)

60 -

75 -

90 14.40 8.70 5.50 9.10

105 9.60 9.60 8.70 7.10



Labelled NH.-N + NO3-N (4.80 mg/1)
Soil Lateral
Depth Distance(cm): 5 50 100 350 500 900


(cm)

60

75

90 8.30

105 5.77


-

7.25

8.31


- 3.28

- 5.36


- 5.51

- 4.68










Table 11. Concentrations (mg/1) of total and labelled N in

soil solution samples from 4 depths and 6 lateral

distances from the central drain in the DT plot on

August 13, 1980 (Julian Day 226). Geometric means

are shown in parentheses.



_Soil Lateral Total NH4-N (0.12 mg/1)
Soil Lateral
Depth Distance(cm): 5 50 100 350 500 900
(cm)

60 0.30

75 0.01 0.09 0.01 0.01 0.01 0.01

90 1.34 0.01 2.49 0.22

105 6.17 0.01 0.60 1.63 0.14 1.74



_Soil Lateral Total NO3-N (9.62 mg/1)
Soil Lateral
Depth Distance(cm): 5 50 100 350 500 900
(cm)

60 6.20

75 14.50 44.50 68.00 38.00 9.40 8.00

90 55.00 29.00 6.20 3.80

105 8.50 39.00 12.50 8.10 5.25 3.65



soil Lateral Labelled NH -N + NO 3-N (1.71 mg/1)
Soil Lateral
Depth Distance(cm) 5 50 100 350 500 900


3.47



3.51


17.78

22.46

15.55


29.56

12.61

5.69


1.87 0.79

0.43 -

0.48 0.45


0.16

0.20

0.10

0.13


(cm)

60

75

90

105











Table 12. Concentrations (mg/1) of total and labelled N in

soil solution samples from 4 depths and 6 lateral

distances from the central drain in the DT plot on

August 20, 1980 (Julian Day 233). Geometric means

are shown in parentheses.


soil Lateral Total NH4-N (0.02 mg/1)
Soil Lateral
Depth Distance(cm): 5 50 100 350 500 900
(cm)

60 0.01 0.01

75 0.01 0.01 0.01 0.01 -

90 0.01 0.01 0.01 0.01 0.13

105 0.01 0.01 0.54 0.74



soil Lateral Total NO 3-N (4.03 mg/1)
Soil Lateral
Depth Distance(cm): 5 50 100 350 500 900
(cm)

60 1.05 1.90

75 18.00 40.00 1.60 1.60 -

90 23.50 52.50 5.75 2.00 0.61

105 7.00 10.00 0.81 0.67



_Soil Lateral Labelled NH -N + NO -N (0.51 mg/1)
Soil Lateral
Depth Distance(cm) : 5 50 100 350 500 900


(cm)

60

75

90

105


9.31

12.16

3.63


0.57

21.56

28.30

5.39


0.08

0.29

0.07


0.17

0.21

-


0.01

-

0.01

0.01











Table 13. Concentrations (mg/1) of total and labelled N in

soil solution samples from 4 depths and 6 lateral

distances from the central drain in the DTL plot

on August 1, 1980 (Julian Day 214). Geometric

means are shown in parentheses.


Total


NHA-N (0.18 mg/1)


Soil Lateral
Depth Distance(cm): 5 50 100 350 500 900
(cm)

60 0.05 0.07 0.06 0.29 0.38 0.28

75 0.22 0.04 0.24 0.36 0.41 0.23

90 0.04 0.08 0.43 0.39 0.33 0.34

105 0.06 0.09 0.27 0.38 0.32 0.33



soil Lateral Total NO3-N (6.87 mg/1)
Soil Lateral
Depth Distance(cm): 5 50 100 350 500 900
(cm)

60 11.2 18.0 7.1 5.0 4.3 4.3

75 14.3 15.7 6.7 5.0 3.5 4.5

90 13.9 15.0 7.4 6.1 4.1 4.1

105 12.8 13.6 5.2 4.1 4.0 3.3



Soil L lateral Labelled NH 4-N + NO3-N (3.10 mg/1)
Soil Lateral
Depth Distance(cm) : 5 50 100 350 500 900


(cm)

60

75

90

105


7.2

9.3

8.9

8.2


11.5

10.1

9.6

8.7


4.6

4.4

5.0

3.5


2.2

2.2

2.7

1.9


1.7

1.4

1.6

1.6


0.7

0.8

0.7

0.6











Table 14. Concentrations (mg/1) of total and labelled N in

soil solution samples from 4 depths and 6 lateral

distances from the central drain in the DTL plot

on August 13, 1980 (Julian Day 226). Geometric

means are shown in parentheses.



_Soil Lateral Total NH4-N (0.18 mg/1)
Soil Lateral
Depth Distance(cm): 5 50 100 350 500 900
(cm)

60 0.01 0.01 0.12 0.62

75 0.01 0.24 1.08 3.04 2.90 5.51

90 0.10 0.20 0.01 0.01 0.49 3.63

105 0.55 0.01 0.10 0.01 1.53 2.20



Total NO -N (14.76 mg/1)
Soil Lateral
Depth Distance(cm): 5 50 100 350 500 900
(cm)

60 10.50 10.00 32.50 10.50

75 10.00 12.50 42.00 29.00 16.00 34.00

90 10.50 9.40 17.00 17.00 27.50 21.00

105 8.10 20.00 13.50 14.50 21.00 16.00



Labelled NH4-N + NO3-N (0.97 mg/l)


Soil Lateral
Depth Distance(cm): 5 50 100
(cm)

60 1.19 0.85

75 1.13 1.11 3.66

90 1.20 0.84 1.45

105 0.98 1.75 1.16


350 500 900


2.67

2.62

1.39

1.19


3.65

5.41

4.35


0.29

0.95

0.64

0.48










Table 15. Concentrations (mg/1) of total and labelled N in

soil solution samples from 4 depths and 6 lateral

distances from the central drain in the DTL plot

on August 20, 1980 (Julian Day 233). Geometric

means shown in parentheses.


_Soil Lateral Total NH4-N (0.13 mg/1)
Soil Lateral
Depth Distance(cm): 5 50 100 350 500 900
(cm)

60 0.01 0.01 0.01 0.28

75 0.82 0.96 -

90 0.01 0.06 0.23 0.58 0.27 0.26

105 0.43 0.04 0.14 0.72 0.73



soil Lateral Total NO3-N (4.14 mg/1)
Soil Lateral
Depth Distance(cm): 5 50 100 350 500 900
(cm)

60 2.10 2.60 5.30 18.00

75 1.85 29.00 -

90 4.00 1.40 3.70 7.50 11.00 8.00

105 1.00 3.20 1.50 5.30 3.20



Labelled NH -N + NO3-N (0.09 mg/1)
Soil Lateral
Depth Distance(cm): 5 50 100 350 500 900


(cm)

60

75

90

105


0.06

0.07

0.11

0.04


0.04


0.07



0.10

0.08


0.03

0.19

0.05

0.01


0.71

0.38


0.42



0.19

0.09











means for DT soil solution were only 0.50, 0.12, and 0.02

mg/1 of NH4-N respectively for 57, 69, and 76 days after

fertilizer application; however, geometric means for NO3-N

were 8.79, 9.62, and 4.03 mg/l on these same dates. These

concentrations were much higher however than those observed

in drainage water (Fig. 13) after those dates. This obser-

vation was probably due to plant uptake combined with a lim-

ited number of subsurface drainage periods. This data also

shows that NO 3-N constitutes the predominant form for N in

the soil solution. Geometric means for labelled N for 57,

69, and 76 days after fertilization were 4.80, 1.71, and

0.51 mg/l respectively. Thus the concentration of solution

N attributable to the 15N-depleted (NH4)2SO4 was observed to

decrease more than 9 times within the 19-day period. This

information is also consistent with the reasoning that N up-

take by plant roots was very significant in this deep-tilled

DT soil. The slight increase in NO3-N in soil solution be-

tween 57 and 69 days after fertilization further suggests

that some nitrification occurred, but was probably limited

in rate due to acid subsoil conditions.

Geometric mean concentrations of NH -N in DTL soil so-

lution were 0.18, 0.18, and 0.13 mg/l respectively for 57,

69, and 76 days after fertilization; however geometric means

for NO3-N were 6.87, 14.76, and 4.14 mg/l on these same

dates. As with DT soil, the N concentrations in soil so-

lution were considerably higher than in the drainage

(Fig. 14) water essentially for the same reasons. Nitrate











was the dominant form of N in solution and the large in-

crease in NO 3-N concentration between 57 and 69 days after

fertilization implies nitrification had occurred. Geometric

means for labelled N for 57, 69, and 76 days after fer-

tilization were 3.10, 0.97, and 0.09 mg/1 respectively. For

this deep-tilled treatment, the concentration of solution N

attributable to the 15N-depleted (NH4)2SO4 was observed to

decrease more than 34 times during the 19-day period. This

observation suggests that rates of N uptake by plant roots

were higher in DTL soil in comparison to DT soil, probably

due to higher pH values in the subsoil. Calvert et al.

(1977) have shown that citrus yields were higher and that

root systems were deeper in DTL soil compared to DT soil.

Concentrations for selected cations in the DTL soil

solution are presented in Table 16 for 57 days after

application of fertilizer. Geometric mean concentrations

for K Ca, and Mg+2 were 9.2 x 104, 2.15 x 10, and

2.47 x 10-3 m.e./l, respectively. The presence of these

relatively high cation concentrations in the soil solution

implies that the ion exchange sites in this soil are

predominantly occupied by divalent Ca+2 and Mg+2 cations

which are more competitive than NH ions for the exchange

si les.

Citrus leaves (Table 17) sampled during the summer and

fa 1l had 2.28 to 2.69% N and 12 to 15% of the tissue N was

attributable to the 15N-depleted fertilizer. Essentially no

difference were observed between the 3 plots.











Table 16. Concentrations (mg/1) of K Ca+2, and Mg+2 n

soil solution samples from 4 depths anc' 6 la eraL

distances from the central drain in the DTL -plot

on August 1, 1980 (Julian Day 214). Geometr c

means shown in parentheses.


(36


mg/1)


Soil Lateral
Depth Distance(cm): 5 50 100 350 500 900
(cm)

60 41 43 26 33 35 31

75 29 25 60 46 41 47

90 52 27 27 41 29 55

105 21 27 48 53 28 32



aCa+2 (43 mg/l)
Soil Lateral
Depth Distance(cm): 5 50 100 350 500 900
(cm)

60 44 41 44 42 56 46

75 39 37 44 39 61 31

90 62 28 44 40 44 29

105 46 43 29 63 46 59



Mg+2 (30 mg/l)
Soil Lateral
Depth Distance(cm): 5 50 100 350 500 900
(cm)

60 29 26 29 27 48 30

75 27 24 29 28 56 22

90 51 19 29 28 31 21

105 26 28 20 55 30 50











Table 17.


Mean percentages (%) of N in leaf tissue and per-

centages (%) of leaf N that was labelled for

foliage samples taken in July and October of 1980.

Mean values represent 12 rootstock-scion combin-

ations of citrus.


Percentage N in

ST DT

2,69 2.50

2.30 2.28


Leaves

DTL

2,50

2.30


Percentage Labelled N

ST DT DTL

13.33 15.07 14.73

13.68 13.06 12.67


Date

July

October











Chapter 5: Summary and Conclusions

During the summer of 1980, 115 kg/ha of N in the form

of 15N-depleted (NH4)2SO4 was applied to portions of a

10-year old experimental citrus grove in South Florida for

the purpose of determining the fate of fertilizer N applied

to an acid, sandy soil (Spodosol). The influence of deep

tillage upon the disposition of the applied N was determined

by applying the labelled fertilizer to shallow-tilled (15 cm

depth) ST, deep-tilled (105 cm depth) DT, and limed (56 met-

ric tons/ha) plus deep-tilled (105 cm depth) field plots of

a Spodosol with a 10-20 cm thick spodic layer located at ap-

proximately 80 cm depth. All plots have received annual ap-

plications of 2.24 metric tons per ha of limestone. The ST

plot is typical of tillage and liming practices for agricul-

tural crops growing in Florida Spodosols. The citrus grove

was surface-drained by elevated beds and subsurfaced-

drained by 10-cm diameter plastic tubes placed 107 cm deep

and 18.3 m apart. Elevated beds with 2 rows of citrus trees

were oriented perpendicular to the drain tubes, and the area

between tree rows supported a thick Bahiagrass sod. Soil,

soil solution, drainage water, and citrus leaf samples were

taken at selected times during the first six months follow-

ing the application of fertilizer. Chemical analyses were

performed for NH4-N and NO3-N concentrations and isotopic

15 N14 ratios were determined. A local drought during the

summer and fall restricted the quantities of drainage as

well as the sampling frequency for soil solution. During










typical years as much as 90% of the total annual rainfall

often occurs during the summer and fall seasons.

Accumulative leaching losses of NH4-N and NO3-N in

drainage water discharge from the deep-tilled DT and DTL

plots were extremely small during the first 196 days after

the application of fertilizer. The small leaching losses

for DT and DTL plots were due to low N concentrations in the

water but more importantly to small water discharge rates.

Drainage water from the shallow-tilled ST plot resulted in a

slightly larger loss of 4.76 and 4.34 kg/ha respectively of

unlabelled (residual) and labelled nitrogen. Thus the la-

belled N leached from the ST soil represented only 4% of the

quantity 15N-depleted fertilizer nitrogen. Concentrations

of residual plus labelled N in ST drainage water were as

high as 25 mg/l during a drainage event which occurred with-

in the first month after fertilization but decreased to

about 1 mg/l within 4 months after fertilization. The rela-

tive influence of N in the drainage water from ST upon water

quality of the downstream canal was small because of the

small volumes of drainage that occurred during 1980.

Nitrogen discharged in drainage water from all plots was

predominantly in the form of NO3-N with lesser amounts as

NH4-N.

Analyses of soil cores to a depth of 70 cm in the ST

plot and 100 cm in the deep-tilled plots revealed that the

15N-depleted NH4-N was largely nitrified to the NO 3-N form

within the first 30 days. During the first 12 days











nitrification appeared to proceed at faster rates in the

deep-tilled plots as compared to the shallow-tilled plot

possibly due to slightly higher water retention capacity and

higher pH of the DTL subsoil.

Initially, larger quantities of residual and labelled N

(NO3-N plus NH -N) were present in the soil profile for the

ST plot (182.8 kg/ha) than for DT (76.5 kg/ha) and DTL

(104.8 kg/ha) plots. The quantities of labelled N in the

soil at that date were equivalent to 105, 28, and 54% of the

applied 15N-depleted (NH4)2SO4 for ST, DT, and DTL plots re-

spectively. These data indicate that within 12 days after

fertilizer application, 72 and 46% of the labelled fertiliz-

er N, respectively, had been removed from the profiles for

DT and DTL plots. Virtually all of the labelled N was pre-

sent in the shallow-tilled ST soil, however. Since leaching

losses for the deep-tilled soils were extremely small and

denitrification losses of NO -N were most unlikely due to

the infrequent rainfall during that period, the unaccounted

for N appears to have been absorbed by root systems of

bahiagrass and citrus. Volatilization losses of the applied

(NH4)2SO4 were less than 0.5% for all three plots. Earlier

results by Calvert et al. (1977) showed that fruit yields

were significantly higher and root development more exten-

sive in the soil profile for the DTL treatment than for ST

treatment. Yields and root development for the DT treat-

ments was intermediate. Shallower root systems in the

shallow-tilled plot would thus be expected to have lower










rates of absorption of soil N by actively growing roots than

either of the deep-tilled plots.

Total quantities of N measured in ST, DT, and DTL soil

profiles for 12, 42, 75, and 134 days after fertilization

showed dramatic decreases in labelled as well as residual N

from previous applications of fertilizer. After the first

month, much of the soil N appeared to be coming from

non-fertilizer sources of N, presumably from mineralization

of soil organic matter. By 134 days after fertilization,

less than 3% of the 15N-depleted fertilizer N was still

present in the soil. These data also imply that most of the

applied fertilizer N was utilized by citrus trees and

bahiagrass within 4.5 months after application. Previous

investigations with non-labelled N (Mansell et al. 1977) and

with higher rainfall amounts during summer months revealed

that an average of 20% of the fertilizer applied to the

shallow-tilled plot of this same citrus grove was leached

over a three-year period. Average leaching losses for the

deep-tilled plots however were much less ( 5%) As seen

from the current investigation where high residual

concentrations of soil N occur from previous fertilizer ap-

pl cations, the use of 15N-depleted fertilizer N provides a

more effective means to study the fate of applied N under

field conditions. Thus results from this study show that

fertilizer use efficiency for labelled N was higher in deep-











tilled plots than in the ST plot. Even the N use efficiency

for citrus growing in ST soil however was relatively h.gh

due to the low frequency of drainage events during 198i.

As expected, soil solution samples taken 57, 69 aid 76

days after fertilizer application showed that NO3-N wa. the

predominant N form in solution in all plots. Concentrations

of N in solution decreased during this 19-day period indica-

ting that plant roots were actively absorbing N. Mean con-

centrations of solution N attributable to the 15N-

depleted (NH4)2SO4 in the DTL soil were observed to decrease

more than 34-fold during this 19-day period. Compared to

the DT plot, rates of N uptake by plant roots appeared to be

higher for the DTL plot during that period probably due to

higher pH values observed for the subsoil.

Isotopic analyses of citrus leaves during the summer

and fall showed that 12 to 15% of N in the tissue was attri-

butable to the 15N-depleted fertilizer. Concentrations of N

in the tissue did not vary greatly between plots.

Results from this study using labelled N confirm earli-

er investigations (Mansell et al. 1977) that deep tillage

plus lime incorporation into the profile of a Spodosol is an

effective means to enhance root absorption of fertilizer N,

minimize N leaching losses, and thus minimize contamination

of water resources with NO3-N. Shallow development of root

systems, low cation and water retention characteristics, and

high values of saturated hydraulic conductivity of shallow-

tilled Spodosols provide these acid, sandy soils with po-











tential for leaching loss of fertilizer N during periods of

soil drainage. Establishment of narrow (1.0 m wide) zones

of deep tillage with lime only in the vicinity of the soil

where citrus or other agricultural crops are to be located

offers an effective means to increase fertilizer- and water-

use efficiencies for newly developed Spodosols. Restriction

of the deep tillage operation to only the soil near the crop

rows decreases the cost to a fraction of that required to

deep till entire fields.











Literature Cited

1. Bremner, J. M. 1965. Inorganic forms of nitrogen:. In

Methods of Soil Analysis, C. A. Black (Ed). Part 2.

Chapter 84, pp. 1179-1206.

2. Broadbent, F. E. and A. B. Carlton. 1980. Methodology

for field trials with nitrogen-15-depleted nitrogen.

J. Environ. Qual. 9:236-242.

3. Calvert, D. V., E. H. Stewart, R. S. Mansell, J. G. A.

Fiskell, J. S. Rogers, L. S. Allen, and D. A. Graetz.

1981. Leaching losses of nitrate and phosphate from a

Spodosol as influenced by tillage and irrigation level.

Soil and Crop Sci. Soc. Florida Proc. 40:62-71.

4. Calvert, D. V., H. W. Ford, E. H. Stewart and F. G.

Martin. 1977. Growth response of twelve citrus

rootstock combinations on a Spodosol modified by deep

tillage and profile drainage. Proc. of Int. Soc.

Citriculture 1:79-84.

5. Fiskell, J. G. A. and D. V. Calvert. 1975. Effects of

deep tillage, lime in corporation, and drainage on

chemical properties of Spodosol profiles. Soil Sci.

120:132-139.

6. Hauck, R. D. and J. M. Bremner. 1976. Use of tracers

for soil nitrogen research. Advances in Agronomy

28:219-266.

7. Hillel, D. 1980. Fundamentals of Soil Physics.

Academic Press, New York, page 8.











8. Kowalenko, C. G. 1980. Transport and transformation

of fertilizer nitrogen in a sandy field plot using

tracer techniques. Soil Sci. 129:218-221.

9. Knipling, E. B. and L. C. Hammond. 1971. The SWAP

study of soil water management for citrus on flatwood

soils in Florida. Soil and Crop Sci. Soc. Florida

Proc. 31:208-210.

10. Mansell, R. S., D. V. Calvert, E. H. Stewart, W. B.

Wheeler, J. S. Rogers, D. A. Graetz, L. H. Allen, A. R.

Overman, and E. B. Knipling. 1977. Fertilizer and

Pesticide Movement from Citrus Groves in Florida

Flatwood Soils. Completion Report for Project

R-800517, Environmental Protection Agency, Environ-

mental Research Laboratory, Athens, GA.

11. Mansell, R. S., W. B. Wheeler, and D. V. Calvert.

1980. Leaching losses of two nutrients and an

herbicide from two sandy soils during transient drain-

age. Soil Sci. 130:140-150.

12. Rogers, J. S., E. H. Stewart, D. V. Calvert, and R. S.

Mansell. 1977. Water quality from a subsurface

drained citrus grove. Third National Drainage Sympo-

sium Proceedings, American Society of Agricultural

Engineers, pp. 99-103.

13. U.S. Environmental Protection Agency. 1976. Nitrogen,

ammonia. In Methods for Chemical Analyses of Water and

Water, U.S. E.P.A. Technical Transfer Manual.

pp. 169-172.











Appendix: Synopsis from Master of Science Thesis (August

1.982) for Miss Ko Hui Liu in the Soil Science

Department, University of Florida.

Title: Miscible Displacement of Aqueous Solutions of

NH4-N through Columns of Sandy Soil.

A. Summary

Ion exchange for K+ and NH4 during miscible displace-

ment was studied using water-saturated soil columns of ST

and DTL soils. The soil in each column was initially

"saturated" with 0.03 N KC1 prior to each displacement

experiments. During the displacement experiments steady

pore water velocities of 7.8 and 30.1 cm/hr were maintained

in the columns. Potassium in the soil was exchanged with

either 0.03 N NH Cl or 0.01 N NH C1 + 0.02 N KC1, and NH+ in

the soil was exchanged with either 0.03 N KC1 or

0.01 N NH4 C + 0.02 N KC1.

Adsorption isotherms for NH were determined and retar-

dation R values were computed for each soil These R values

were slightly less than those calculated from BTC's solutee

breakthrough curves). Because of the presence of lime in
2+ 2+
the soils, Ca and Mg were continually leached out of the

soil columns. The presence of these divalent cations in the

soil solution suppressed the sorption of NH4 and K The

zero point of charge ZPC of both soils were determined and

found to be 4.8 and 6.1 for ST and DTL soils respectively.

A time study of NH4 exchange revealed that ion exchange was

instantaneous.











B. Conclusions

Based on the results in this study, the following con-

clusions were made for ion exchange during transport through

ST and DTL soil under water-saturated conditions: (1) The

ZPC of DTL soil was higher than that for ST soil due to a

lower content of organic matter and higher contents of Fe

and Al oxides in the mixed soil. Thus ion exchange in these

soils is dependent both on pH and ionic strength of the so-

lution. (2) Ion exchange was shown to be instantaneous and

thus the use of a kinetic model for ion exchange in these

soils is not justified. (3) Because of the dissolution of

lime during displacement experiments, the presence of Ca2+
2+
and Mg in the system and changes in pH affected the sorp-

tion of K and NH (4) For these soils, which have been

receiving lime for the last 10 years, it is difficult to

establish a homo-ionic soil. During these laboratory ex-

periments, the soils were apparently never completely satu-

rated with either K + or NH Thus the ion exchange ob-

tained was not only for K+ and NH+ but included other ions
2+ 2+
such as Mg and Ca2. Due to these considerations, adsorp-

tion isotherms rather than ion exchange isotherms were mea-

sured and used to calculate the retardation factors.

(5) Since ion exchange was shown to be instantaneous the

BTC's in these soils should not have been affected by the

flcw velocity imposed. A small apparent velocity effect on

the translation of BTC's for both K+ and NH+ was attributed
4
to a change in the chemical environment in the system due to











changes in pH, ionic strength, and ionic composition all of

which affect the retardation of ionic species in these

soils. (6) The ST soil showed no selective adsorption for

either K or NH4. This was anticipated from the similar

properties of these ions such as ionic radii and charge.

(7) Due to the presence of vermiculite in the DTL soil how-

ever, K was fixed during ion exchange and transport. This

study shows that even if an ion is fixed or consumed in the

system it is still possible to predict its front during

transport if it is not completely removed from solution.

(8) For the ST soil the displacement studies showed that ion

exchange was reversible for ST soil since all the input

amount of either K or NH were recovered. (9) although the

application of lime to these soils has raised the pH, the

presence of Ca2+ and Mg2+ in the soils will inhibit the

leaching of either NH4 or K when applied to these soils as

fertilizer.




Full Text

PAGE 1

WATER RESOURCES RESEARCH CENTER Publication No. 65 NITRIFICATION, DENITRIFICATION, AND SORPTION DESORPTION OF NH4-N IN SANDS DURING WATER MOVEMENT TO SUBSURFACE DRAINS By R. S. H.ansell J. G. A. Fiskell D. V. Calvert Soil Science Department University of Florida Gainesville

PAGE 3

i NITRIFICATION, DENITRIFICATION, AND SORPTION DESORPTION OF NH4-N IN SANDS DURING WATER TO SUBSURFACE DRAINS By R. S. Mansell J. G. A. Fiskell D. V. Calvert Publication No. 65 Florida Water Resources Research Center Research Project Technical Completion Report project Number A-042-FLA Annual Allotment Agreement Numbers 14-34-0001-0110 14-34-0001-1110 Report Submitted August 23, 1982 The work upon which this report is based was supported in part by funds provided by the United States Department of the Interior.

PAGE 4

ii TABLE OF CONTENTS Section Title Table of Contents Acknowledgments Abstract Chapter 1: Ihtroduction Chapter 2: Isotopic Tracer Methods for Nitrogen in Soil-Plant Systems Chapter 3: Experimental Methods and Procedure Chapter 4: Results and Discussion Chapter 5: Summary and Conclusions. Literature Cited Appendix: Synopsis of Master of Science Thesis Page i ii iii v 1 5 9 18 70 76 78

PAGE 5

iii ACKNOWLEDGEMENTS The Soils and Fertilizer Research Branch Division of Agricultural Development, Tennessee Valley Authority (TVA) located at the National Fertilizer Development Center near Muscle Shoals, Alabama supplied the 15N-depleted (NH4)2S04 used in this research. The project are appre-ciative/to TVA and to Dr. Roland D. Hauck, Research Soil Scientist with TVA. We are also indebted to Dr. Hauck for making us aware of the automatic mass spectrometer facility at the Los Alamos National Laboratory for analyses of 15N /14N ratios in water samples. We wish to acknowledge the University of California, Los Alamos National Laboratory (LANL) in Los Alamos, New Mexico for performing nitrogen isotope analyses for the 1300 water and soil samples used in this investigation. We are particularly appreciative to Dr. B. B. McInteer for pe!mit ting us to utilize the analytical service of the Los Alamos National Laboratory. The authors are indebted to Dr. J. S. Rogers, Agricul-tural Engineer with the United States Department of Agricul-ture and University of Florida, for providing water drainage data from the Soil, Water, Atmosphere, and Plant (SWAP) research citrus grove ,.,here the field experiment for this grant was performed. Acknowledgement is also given to Marc E. Hall, former Laboratory Technologist II in the Soil Science Department and Mariano Delacasa, student assistant, for performing COID-

PAGE 6

iv puteranalysis of the data. The investigators also acknowledge assistance from Mrs. Jennifer Johnson, former Labora tory Technologist for preparation of camera-reaqy figures presented in this

PAGE 7

v ABSTRACT Isotopically labelled (15N-depleted (NH4)2S04) N tilizer was applied at a rate of 115 kg N/ha to a Florida citrus grove. located in a subsurface drained Spodosol in order to determine the fate of applied N. The fertilizer was applied to one O.22-ha field plot each of shallow-tilled (ST), deep-tilled (DT) and deep-limed/tilled (DTL) soil. Analyses of soil, soil solution, drainage water, and citrus leaf samples revealed that much of the NH4-N was nitrified to form N0 3-N during the first 42 days after application of fertilizer. Partially because of a .local drought, most of the applied N was absorbed by plant roots during the first 134 days and leaching losses in drainage water were less than 4% of the amount applied. Rates of and N uptake by roots were highest for DTL and lowest for ST treatments. Selective deep tillage with lime incorporation of Spodosols appears to be an effective means to manage ferti1izerand water resources for citrus and possibly other agricultural crops without contaminating groundwater supplies.

PAGE 8

1 Chapter 1: Introduction Acid, sahdy soils with shallow water tables are commonly utilized in Florida for intensive production of highvalue vegetable, fruit, and horticultural crops. Producti vity of these soils using the definition by Hillel (1981) may be severely restricted by limiting levels of both "chemical fertility" and "physical fertility". High contents of silica sand and low contents of colloidal material provide porous soil matrices that primarily have only short-term capacity to supply water and nutrients to active plant root systems. Relatively large pores and low ion exchange capacities of the solid particles also result in potential leaching losses of fertilizer applied to such soils. During rain storms or when excess irrigation is applied, soluble nutrients from applied fertilizer may easily be displaced from the rooting zone of the soil profile and be transported into underlying groundwater. Obviously, careful management of fertilizer and soil water for crops growing in these soils is needed to simultaneously minimize chemical contamination of groundwater supplies, maximize the use-efficiency of crop roots for nutrients applied in chemical fertilizer, and optimize the water use-efficiency of crop roots. Although the average annual rainfall in Florida is high (1.3 to 1.5 meters) the distribution the year is highly nonuniform giving a relat:i,vely dry period during the winter and spring months and a relatively wet period during the summer and fall months. For crops growing in sandy

PAGE 9

2 soils with an impermeable layer located in the profile at' some depth less than 2 m, ditch or subsurface drainage systems are frequently installed so as to prevent development of anaerobic soil conditions in the rooting zone during the wet season. These systems are particularly needed to remove excess water during' intense thunderstorms from soil areas with nearly flat topography. However, during the dry, period the accelerated removal of drainage water may well decrease the effectiveness of rainfall events and the water useefficiency of the crop. Leaching loss of soluble fertilizer nutrients may also be increased as a result of accelerated water discharge from ditches or subsurface drain tubes (Mansell, Wheeler and Calvert 19801 Mansell et ale 1977). Such enhanced leaching of applied fertilizer should be most evident during the wet period but also would be expected for large storms even during the dry period. On the one hand artificial drainage tends to enhance use-efficiency of crops for'water and fertilizer by permit:t;:ing active growth of the root system during the wet season, but accelerated leaching loss of nutrients due to the drainage system tends to restrict the use-efficiency of applied fertilizer. During the dry period, the limited water-holding capacity of these coarse-textured soils requires that irrigation be used to maintain economically profitable crop yields. Insufficient irrigation will limit yields whereas excess irrigation enhances' leaching'loss of fertilizer. Thus effective soil water management for crops growing in shallow sands must of

PAGE 10

3 necessity include both irrigation and drainage practices The effectiveness of soil water management practices for these soils will ultimately be reflected in the magnitude of crop yields and in the quality of groundwater. Results from a 3-year investigation (Mansell et ale 1977, Rogers et ale 1977, Calvert et al. 1981) of fertilizer leaching losses from an experimental citrus grove located on a Spodosol at the University at Florida Agricultural Re-search Center near Ft. Pierce show that significant quan-tities of N0 3-N and P04-N were removed from the soil through discharge of drainage water. These losses occurred despite a management practice of applying small quantities of fer-tilizer frequently' (every 3 months) to minimize leaching losses. As much as 22% of 169.52 kg/ha of N (55% of fertil-izer N was NH4-N and 45% was N0 3-N) applied in 4 split ap-' plications during a 12-month period was lost as N0 3-N in the drainage water. However, for treatments which had been deep-tilled to a depth of 105 cm prior to planting of citrus trees the average annual loss of N0 3-N in the drainage water was equivalent to only 3% of the total N applied as fertil-izer. The effectiveness of the deep tillage plots in mizing the leaching loss of N was attributed to increased retention of applied NH4-N in the top soil due to upward transport of colloidal material from subsurface horizons and \ to increased potential for denitrification of N0 3-N in the soil profile due to slower drainage rates which resulted in generally higher water contents following rainfall events.

PAGE 11

4 A major limitation of the interpretation of some of the results from this study however was that the applied fertilizer N was not labelled with a tracer such as l5N or 14N Thus a field investigation was performed to quantitatively evaluate the fate of lSN-depleted NH: applied as fertilizer to a subsurface-drained sandy soil in an experi-mental citrus grove in South Florida. Isotopically labelled fertilizer N was applied to plots of a Spodosol which had previously received shallow-and deep-tillage treatments. The primary objective of this investigation was to determine the influence of movement (leaching), ion exchange, and transformations (nitrification, denitrification, etc) upon lSN depleted NH: applied as a single fertilizer application to a tube-drained sandy soil in a citrus grove. A secondary objective was to evaluate the influence of 3 soil modifications ---shallow tillage ST, deep tillage DT, and deep tillage with initial incorporation of limestone into the profile DTL ---upon the fate of applied lSN depleted NH:. Both objectives require evaluation under field conditions. A brief review of the use of lSN and 14N isotopes as a tracer for fertilizer nitrogen in soil-plant systems will be given before presenting a description of experimental me-thods and results.

PAGE 12

5 Chapter 2: Isotopic Tracer Methods for Nitrogen in Soil-Plant Systems The recovery of fertilizer N applied to a given soil is commonly determined by both difference (non-tracer) and isotope tracer methods. Results from these two methods (Kowalenko 1980) may however lead to different conclusions. The difference method gives a measure of the net recovery of fertilizer N after several processes (denitrification, nitrification, ion exchange, leaching, immobilization, etc.) have exchanged with and transformed the fertilizer N. Broadbent and Carlton (1980) have shown that the assumption in difference method calculations that plant uptake of unla-beled N is the same in fertilized and unfertilized plots is not valid. In contrast, 15N and 14N tracer techniques give a measure of the actual fate of the applied N (Kowalenko 1980) as well as provide more accurate results than the dif-ference method (Broadbent and Carlton 1980) Tracer methods permit actual measurement of fertilizer recovery as well as the distribution of recovered N in various chemical fractions. Nitrogen tracer techniques (Hauck and Bremner 1976) are based upon the observation that naturally occurring N compounds contain about 0.366 atom % 15N and 99.634 atom % 14N Addition of a fertilizer with an unusually high (15N-enriched) or low (15N-depleted) concentration of 15N to a soil-plant system can thus be used as a tracer. Measured changes of 15N/14N ratios in samples removed from the system

PAGE 13

6 provide a means to investigate transformations of tl;1e,added fertilizer. The magnitude of the change in the isotope ratio R from the background level R can be used to calcu-o late the extent to which the tracer N ,has interac:ted with and become part of the system. The percentage of N present in a sample of soil or soil water initially from either 15N-depleted or 15N-enriched material can be calculated using the relationship R R o B = R -R. o 1 (1 )' where R. is the isotopic ratio for the tracer material. 1 Hauck and Bremner (1976) state that isotope tracer methods offer much potential for studying ways tJmaximize the efficiency of fertilizer nitrogen in crop production. They state that movement of N into, within, and from soil b b 15 can accurately e 0 talned only by the use of N-depleted or 15N-enriched fertilizers. .Isotopically labelled fertilizer is particularly needed in systems (Broadbent and Carlton 1980) which contain a background of large amounts of indigeneous N. Although 15N-enriched ma-terials have been used in most nitrogen tracer studies, use of 15N-depleted material has gained in popularity due to the lower cost of those materials. 15 However, the use of N-depleted materials (Hauck and Bremner 1976) is' re'stricted to experiments where excessive dilution does not'occur in the soil-plant system. Thus studies of plant uptake or movement

PAGE 14

7 of applied nitrogen should be performed for single-rather than multiple-seasons. Kowalenko (1980) used 15N-enriched (NH4)2S04 to investigate transport and transformation of fertilizer NH: in eight micro-size (20-cm diameter) fallow field plots of a sandy soil in Canada. The 15N-enriched (5.5% enrichment). fertilizer was applied to the soil at a rate of 184 kg N/ha in June 1977. Net fertilizer recoveries (as determined by differences between extracted N in fertilized and unfer-tilized plots) from the upper 75 cm of the soil profile were determined to be 117 and 19% of that originally applied, respectively, for 35 and 102 days after application of the f t'l' f 1 15 er 1 lzer. Measurements 0 tota N concentratlons however revealed that recoveries of fertilizer N were ac-tually 73 and 25% for 35 and 103 days after fertilization. +'" Almost all (93%) of the fertlllzer NH4 lnltlally applled was nitrified within 35 days. If the nitrification was assumed to have occurred at a constant average rate, then that rate for this sandy soil would have been 4.89 kg N/ha/day which was considerably higher than a rate.of 2.30 kg N/ha/day obtaihed previously by the same author for a clay loam soil. Both soils had been in fallow prior to the experiments such that neither soil had crop residues to influence microbial processes. Although relatively low organic matter content of the sandy soil suggested limited microbial activity, tracer data $howed the microbial process for nitrification to be rapid. Leaching, denitrification, clay fixation of

PAGE 15

8 mineralization, and immobilization all were important. in the transport and transformation of fertilizer applied to a sandy soil.

PAGE 16

9 Chapter 3: Experimental Methods and Procedure An experimE7ntal citrus grove originally developed (Knipling and Hammond 1971) in 1970 on 9 hectares of flatwood land at the University of Florida Agricultural Research Center (A.R.C.) near Fort Pierce was selected as the location for this investigation. The location of the experimental site as well as the distribution of Spodosols over the land area of Florida is shown in Fig. 1. The grove was established as a cooperative venture between the Univer sity of Florida Institute of Food and Agricultural Sciences (IFAS) and the Agricultural Research Service (ARS) of the United States Department of Agriculture. The IFAS-ARS effort was designated as the Soil, Water, Atmosphere, and Plapt Relationships Project (SWAP). The original purpose for the SWAP grow was to evaluate the influence of deep tillage, with and without incorporation of.limestone into the profile, upon subsurface drainage and growth of citrus in a Spodosol. The SWAP grove was chosen as the site for this work partially because of the background information on chemical and physical characteristics that has been accumulated (Mansell et ale 1977; Mansell, Wheeler and Calvert 1980; Calvert et ale 1981; Rogers et ale 1977) for the acid, sandy soils in the grove. Another reason for that choice was the fact that it is one of the best designed Coastal Plains experimental drainage sites in the Southeastern United States.

PAGE 17

Fig. 1: 1{) Figures Map of Florida showing major land areas of Spodosols and the location of the .experimental site. Location of Spodosols in Florida N Spodosols ([[!) Fort Pierce

PAGE 18

11 Three O.S-hectare plots of citrus were .selected for this research. One plot was selected for each of the three soil profile modification treatments: (i) shallow-tilled (ST) to lS-cm depth, (ii) deep-tilled (DT) to 10S-cm depth, and (iii) deep-tilled (DTL) to 105 cm depth with an initial incorporation of 56 metric tons per ha of dolomitic limestone into the profile. All treatments received annual applications of 2.24 metric tons per ha of limestone to the soil surface. During the initial deep tillage operation, a trenching machine incorporated and mixed spodic and underlying sandy clay loam material with sandy soil material from the A horizons. The primary soil type at the site is Oldsmar fine sand (a member of the sandy siliceous, hyperthermic family of Alfic Arenic Haplaquods). In undisturbed profiles, the A horizon of the acid sand has an average (Calvert et a1. 1981) depth of 82 em and contains about 1% organic matter. Underneath the A horizon a nearly impermeable spodic layer ranging in thickness from 10 to 20':"cm contains about 3.5% organic matter. A layer of sandy clay loam, also with low permeability to water, occurs peneath the spodic horizon. Surface drainage in each plot was provided by a system of elevated beds (38 cm height of bed.crown above t.he bottom of water furrows) separated by parallel water furrows. The width of the beds as measured from centers of adjacent water furrows was 15.2 m. During very intense rain storms any surface drainage water from each plot was removed by

PAGE 19

12 collection ditches at the end of the water furrows. Subsurface drainage was provided by a system of 10-cm corrugated plastic tubes buried beneath the soil surface at an average depth of 107cm and spaced 18.3 m apart. The drain tubes were located perpendicular to the elevated soil beds. Two rows6f citrus 7.6-m apart with a spacing of 4.6 m between trees were planted along the top of each bed in 1970. Water from the center tube in each plot discharged into a concrete manhole where flow was measured with 30.5 cm, 30 degree, V-notch weirs and Stevens typeF, Model 68 water stage recorders. Pensacola bahiagrass was seeded to each plot in 1970. A strip of surface along each tree row was maintained bare of bahiagrass and weeds. In order to investigate the fate of N in fertilizer applied to these coarse-textured acid soils, NH4-N was applied as (NH4)2S04 in an otherwise complete fertiliier to shallow-tilled (ST) and deep-tilled (DT and DTL) plots. On June 5 (Julian Day 157) of 1980, 8-2-8 (% N -% P 2 0 S -% K 20) fertilizer was applied at a rate of 115 kg N/ha to selected areas (The shaded area in the plot diagram in Fig. 2 indicates the fertilized area.) for each of ST, DT, and DTL plots. A batch of the isotopic-ally labelled fertilizer (625 kg) was prepared by mixing 232 15 kg of N-depleted (NH4'2S04 (courtesy of Dr. R. D. Hauck, I Tennessee Valley Authority, Muscle Shoals, Alabama) with 63 kg of ordinary superphosphate, 84 kg of muriate of potash, and 246 kg of powdered dolomite. Isotopically labelled

PAGE 20

13 fertilizer was broadcast by hand to the elevated beds (but not to the water furrows) in the shaded area shown in the plot diagram (Fig. 2) centered about the drain field for the center drain in each of the three plots. Dots in Fig. 2 designate the locations for citrus trees. Approximately 0.22 ha in each plot received labelled fertilizer. The re-mainder of each plot was fertilized with the same rate of non-labelled 8-2-8 fertilizer. Irrigation was applied immediately after the fertilizer application to prevent volatilization of NH4-N due to previous annual applications of limestone to the surface soil. One gallon cans containing open bottles of.su1furic acid were placed at selected locations in the treatment plots in order to any gaseous NH3 released to the atmosphere. The acid solutions were later analyzed for N concentration. During an 8-month period following the application of 15N-dep1eted fertilizer, concentrations of N0 3-N NH4-N were determined in subsurface drainage water, samples of soil solution, in soil cores, and in selected leaf samples from citrus trees. When nitrogen concentrations in water, soil and tissue samples were sufficiently high, ratios of 15 14 t' d' d N-to-N concentra lons were etermlne. The central drain in each of the selected ST, DT and DTL plots was con-tinua11y monitored for water flow rates (Data provided by Dr. J. S. Rogers, Agricultural Engineer, USDA, University of Florida) and intermittently monitored for water quality sam-p1es. Automated water samplers (ISCO Model 391) were used

PAGE 21

Fig. 2: r 54.9 m II1II 14 Schematic diagram (scale: 1-to-695 ern) of experimental citrus plots for ST, DT, and DTL treatments. PLOT DIAGRAM .. .. to .. .. .. .. I .. .. .. .. I .. .. .. .. I .. .. OIl .. 91.4 m N 1 DRAIN USE OUTLETS

PAGE 22

15 to take samples from weirs at the outflow of each drain. Samples were taken most frequently during periods of drain discharge after rainfall events. All water samples were frozen and stored for later analysis. Soil solution samplers were constructed of medium porosity Pyrex glass discs permanently glued to the bottom end of various lengths of 3/4 inch PVC pipe. Rubber stoppers were placed in the top end of the pipes to provide an air pressure seal. The fritted disc samplers were placed at 4 soil depths 60,75, 90, and 105 cm -and at 6 horizontal distances from the central drain -5, 50, 100, 350, 500, and 900 em -along one row of citrus in each plot. Handoperated air pumps were used to apply approximately 80 cm of water to each solution sampler during sampling periods. The samplers were stoppered for 4 to 6 hours to enable extraction of 25 to 100 cm 3 of water depending upon the water content of the soil. Solution samples were taken 57 (August 1), 69 (August 13), and 76 (August 20) days after fertilizer application. Samples were frozen and stored before analysis. Soil water suction was determined from mercury manometer-type tensiometers located at 4 soil depths 30, 60, 90, and 105 cm -and 6 horizontal distances -5, 50, 100, 350, 500, and 900 cm -from the central drain in each treatment. The tensiometers were located along one row of trees in each plot. Water inside the tensiometers made hydraulic contact with water in soil pores through a porous

PAGE 23

16 ceramic cup at the bottom of each tensiometer. Mercury manometers permitted measurements of water suction of soil water. A hydraulic coring machine was used to take continuous soil cores down to a depth of 70 cm in the ST plot and to a depth of 100 cm in each of DT and DTL plots. The holes in the soil profile were later refilled with inert builders sand to prevent preferred water flow to drain tubes. Each core was divided into subsamples corresponding to depths of 0-8, 9-23, 24-38, 39-53, 54-70, 71-84, and 85-100cm. The. subsamples of soil were frozen and stored until analysis could be perform. Twelve cores were removed from each of the tree plots on dates corresponding to 12 (June 15-16), 42 (July 16-17), 75 (August 19-20), and 134 days (October 16-17) after the initial surface application.of labelled fertilizer. Six of the twelve cores from each plot were taken near each drain and the remaining cores were taken from 450 cm lateral distance from each drain. A total of 912 soil samples were collected and analyzed for this investigation. Leaf samples were collected from six r.ootstock varieties for both grapefruit and orange trees once during July and again in October. Thesamples were ground ina Wiley mill and frozen until analysis could be performed. All water samples were thawed and analyzed for pH using a glass electrode, N0 3-N concentration using a nitrate specific ion electrode, concentration using the phenolate

PAGE 24

17 (EPA 1976) method, and soluble salts using an electrical conductivity meter. The remaining volume of each sample was measured. and analyzed for NH4-N and N0 3-N using a .macrokjeldahl procedure (Bremner 1965). Soil" samples were also thawed, extracted with water and 1 N KCl and both extracts analyzed similarly to that for the water samples. Leaf samples were dissolved in concentrated H 2S04 for total analyses and the solutions analyzed for N03-N and NH4-N. After titration, samples containing more than 1.0 mg N were redistilled by microkjeldahl apparatus into 5 ml of 4% boric acid and 1 ml of 0.08 N sulfuric acid. Reagent grade ammonium sulfate was added to all other samples in "order to increase the N content to 1.0 mg per sample". These solu-tions were analyzed in duplicate for isotopi.cal ratios 15 14 Nl N by an automated mass spectrometer at the Los Alamos National Laboratory near Los Alamos, New Mexico. Isotopic () 15 ( rat lOS for reagent grade NH4 2S04' N-depleted NH4)2S04' and for soil samples taken from soil that had only been fer-tilized with unlabelled N materials were used along with the ratios from the samples to calculate the percentages of" N derived from the 15N-depleted fertilizer. Approximately 1300 samples of water, soil "extracts, and leaf extracts were analyzed for isotopic N ratios. Concentrations of labelled and total N were expressed as mg N/liter of effluent.

PAGE 25

18 Chapter 4: Results and Discussion The shallow tillage (ST) treatment of the Spodosol located at the SWAP citrus grove is representative of agri-cultural management practiced on vast areas of similar acid, sandy soils in Florida. Water and nutrient retention capacities of soil in the ST plot (Mansell et ale 1977) have been demonstrated to be very limited. Recommended cultural practices for agricultural crops growing on this and similar soils include split application of fertilizer throughout the growth season and frequent application of relatively small quantities of water during periods of drought or limited rainfall. Because of subsurface horizons with low permeability for water and relatively flat terrain, artificial drainage is often needed to maintain an aerobic root zone during periods of high rainfall. Because of the high values of hydraulic conducti vi ty when these s.oils are water-saturated (Mansell, et ale 1980) rapid rates of drainage tend to enhance leaching losses of fertilizer nutrients such as nitrogen. Deep tillage (DT and DTL treatments) of this soil has been observed (Mansell et ale 1977) to increase the nutrient retention capacity by increasing the cation. exchange capacity, to increase the water retention capacity by incorporating colloids from an underlying Spodic horizon, and to decrease rates of nutrient leaching loss by decreasing drainage discharge rates. Although this treatment is expensive and requires the presence of a subsurface layer high in

PAGE 26

19 colloid content, selective deep tillage over limited. soil zones only in the vicinity of crop rows offers a potential conservation practice for applied irrigation water and fertilizer. As expected, the subsoil pH in the ST plot was observed this study (Fig. 3) to be considerably lower than the DTL plot but higher than the DT plot.. The pH of the top 9 cm of the DTL soil profile was relatively uniform with a value of about 6.i. The higher pH of the DTL soil is attributable to the original incorporation of limestone prior to tillage operation. Annual applications of limestone however were shown to result in pH values near 6.0 for the surface soil of all three treatments. These acid, sandy soils have a predominance (Fiskell and Calvert, 1975) of variable-charge colloids such as organic matter, iron oxides, and aluminum oxides; therefore application of limestone tends to raise the cation exchange capacity. The higher pH of the surface soil of ST and DT plots and the higher pH of both surface and subsurface soil zones in the DTL plot should therefore preferentially favor microbial nitrification of applied fertilizer NH4-N as opposed to the more acid subsoil conditions of ST and DTL plots. During the period of this study soil water content generally was higher between 60 to 90 cm depths (Fig. 4) than in the uppermost 30 crn of the profi.les for all three tillage treatments. During periods of drainage discharge this was particularly true and the subsoil of the deep-

PAGE 27

20 Fig. 3: Distributions of soil pH with depth measured on June 18, 1980 for ST, DT, and DTL tillage treatments. SOIL pH 7.0 I 6,0 5P. 4P. 01 /' 101 20-1 /\ \ 30 I ......, I tt \ \ \ ) 80i \ / 90" 100

PAGE 28

21 Fig. 4: Distributions of soil water content (% by weiqhtl with depth on June 15, 1980 for ST, DT, and DTL tillage treatments. SOIL WATER by wt) 00 ?' 4 6 8 1P 1,2 14 1p 18 2,0 2,2 26 2,8 I fa ... JULIAN DAY167 / (15 JUNE 1980) 20 30 .,..... E40 0.. W 01 .-I. O If)70 '\

PAGE 29

22 tilled plots had generally higher water contents than the shallow-tilled plot. The water-unsaturated condition of the surface soil in all of the treatments however should tend to enhance nitrification of applied fertilizer. Previous research (Mansell et ale 1977) has also shown that the actual water retention capacity of the shallow tilled soil, in'the ST plot is generally much less than that in deep tilled soil in DT and DTL plots. Soil water pressure head values for a 10-day period in October also indicate that the soil at 90 cm depth was drier in the ST (Tables 1, 2 and 3) plot versus the DTL plot. Pressure head values were generally slightly negative in ST whereas the values were positive in DTL. Negative pressure heads in the DT, soil however indicated that DT soil was drier than the DTL soil. Volatilization losses from applied fertilizer (Table 4) averaged approximately 0.31 kg/ha of NH4-N for all 3 plots during the first two months after application of fertilizer. Losses were 0.24, 0.25, and 0.42 kg/ha of NH4-N respectively for ST, DT and DTL plots. Based upon this information we conclude that volatilization losses of N were sufficiently small to be insignificant. The original intent of this investigation was to evaluate leaching loss of fertilizer N during a typical summer rainy period when large quantities of water moved through the soil to subsurface drains. However, an unexpected drought occurred during 1980 such that the summer and fall

PAGE 30

23 Table 1: Soil water pressure head(cm of water) at 0.9 m depih and lateral distances of 0, .0.5, 1.0, 3.5, 5.0, and 9.0 m from the center drain tube in the ST plot for 10 days in October 1980. Lateral Date Distances: 0 0.5 1.0 3.5 5.0 9.0m ----(cm of water) _. --. October 20 -6 -8 0 0 -12 II 21 -9 -8 0 0 -12 22 -8 -10 -1 +1 -12 II 23 -3 -8 0 -1 -1 24 -6 -8 0 -1 -12 II 27 +14 +24 +2 +9 28 -10 -12 -4 -2 -12 II 29 -14 -14 -6 -4 -12 II 30 -13 -15 -6 -4 -13 II 31 -16 -17 .... 8 -6 -16

PAGE 31

24 Table 2: Soil water pressure head (em of water) at 0.9 m depth and lateral distances of 0, 0.5, 1.0, 3.5, 5.0, and 9.0 m from the center drain tube in the DT plot for 10 days in October 1980. Lateral Date Distance: a 0.5 1.0 3.5 5.0 9.0m ---(em of water) --...... -October 20 -5 -15 -14 -10 -6 -5 21 -5 -15 -15 -10 -10 '-5 II 22 -3 -15 -14 -8 -5 23 -3 -14 -14 -9 -5 24 -3 -15 -14 -8 -4 ,II 27 -2 -16 -14 -10 -7 II 28 -2 -15 -16 -10 -8 29 -2 -16 --18 -12 -10 II 30 -4 -16 -12 -10 31 -4 -20 -14 -10

PAGE 32

25 Table 3: Soil water pressure head (cm of water) at 0.9 m <, depth and lateral distances of 0, 0.5, 1.0, 3.5, 5.0, and 9.0 m from the center drain tube in the DTL plot for 10 days in October 1980. Lateral Date Distance: 0 0.5 1. 3.5 5.0 9.0m October 20 +16 +8 +9 +16 +14 +16 21 +26 +16 +9 +16 +14 +16 22 +16 +8 +10 +14 +11 +16 23 +17 +8 +10 +16 +14 +17 24 +17 +6 +10 +14 +14 27 -9 -12 +9 0 28 +16 +3 +2 +8 +8 29 +14 +4 +10 +8 +6 30 +14 +12 +2 +8 +5 31 +14 0 0 +8 +4

PAGE 33

26 Table 4. Volatilization of NH3 from surface-applied fertilizer during 1980 as measured by sorption of Starting Date June 6 June 10 June 18 June 24 July 1 July 21 Total NH4-N in containers (cross-sectional surface area. of 178 cm 2 per container). Effective volatiliza-tion losses of NH4-N are given as averages for each plot. Sorption Treatment Plots Period ST DT DTL Mean (days) ------(kg NH4-N/ha) -----4 0.0455 0.0371 0.0680 0.0500 8 0.1219 0.1191 O. i427 0.1612 6 0.0197 0.0230 0.0331 0.0253 7 0.0191 0.0169 0.0208 0.0185 20 0.0185 0.0157 0.0197 0.0180, 7 0.0185 0.0416 0.0404 0.0337 0.2432 0.2534 0.4247 0.3067 These values were calculated by dividing measured amounts of NH4-N sorption per'container 9 NH4-N/178 cm 2 ) by the conversion factor 1780. ".

PAGE 34

t. 27 periods received less rainfall than during normal years. Total rainfall plus irrigation was only 84 cm during the 196 day (Fig. 5) period between June 18 and December 3. The rainfall distribution (Figs. 6-11) was such that periods of net drainage discharge were intermittent during the time of this study. For the ST, DT, and DTL plots, discharge occurred during the periods from 21 to July 3 (Julian Days 173 to 185), July 13 to August 2 (Julian Days 195-215), November 10 to December 5 (Julian Days 315 to 340), and December 20 to 30 (Julian Days 355 to 365). Water discharge occurred from the drains for approximately 67 days of the .196 'day period. Drainage rates were obviously highest for the shallow tillage plot. Total drainaqe from ST was 30.7 cm (Fig. 6) which amounts to only 37% of the total amount of rainfall plus irrigation. Thus roughly 63% of the input water for the 196 day period was assumed to be used during evcq)otranspiration. Total drainage amounts for DT and DTL deep-tilled plots were however much less than for ST being only 5.6 (7%) and 8.5 (10%) cm, respectively. Evapotranspiration was therefore assumed to account for roughly 93 and 90% of the input water for DT and DTL plots. The evapotranspiration estimates assume no vertical deep seepage water loss and only small changes in soil water storage. Water fluxes or drainage rates were much greater for the shallow-tilled plot than the deep-tilled plots. For all drains the maximum water fluxes occurred during November (Figs. 9, 10, and 11) with values of 3.0, 0.6, and 1.5

PAGE 35

Fig. 5: ,.-... E z o <.9 0::: 0::: c6 Z W > ....J => 2: => u 84 72 60 48 36 7.8 Accumulative amounts (em) of rain ilnd between June 18 (Julian Day 170) and December 31 .. (Julian Day 366). JULIAN DAYS

PAGE 36

29 Fig. 6: Accumulative Ctrainage (m3/ha) from the ST tillage treatment over the period from June 18 to December 31, 1980. ,.-..... CU ...c ('I') E2800 '-'" W Z o 1600 <{ o 1200 W > 800 t( ..:.J ::J 400 2: .::J JULIAN DAYS 180 200 220 240 260 280 300 320 340 360 i' ,.. ST SOIL

PAGE 37

Fig. 7: ..--.... ro ..c:: M-30 Accumulative drain<1.qe (m3/ha) from D'l' treatment over the period from ,-Tune 18 to December 31, 1980. JULIAN DAYS 180 200 220 240 260 280 300 320 340 360 i i' i' E 560r "--' W 480 Z <{ 400 0:: o 320 <{ 0240 W > 160 -1 DT SOIL -A U

PAGE 38

31 Fig. 8: Accumulative drainage (m3/ha) from the DTL treatment over the period from June 18 to December 31, 1980. JULIAN DAYS ,....... 180 200 220 240 260 280 300 320 340 360 CU ..c ("') E 840 ........... W 720 Z 240 ....J ::::> 2120 ::::> U U
PAGE 39

Fig. 9: >, 280 tU -0 ....... tU 240 .s:::. ....... E 20 16011 .-.J LL 120 0:: w I 3: 8 32 3 Drainage flux (m/ha/dRY) from the ST treatment over the period from June 18 to 31, 1980. JULIAN DAYS 180 200 220 240 60 ST SOIL

PAGE 40

33 Fig. 10: Drainage flux (m3/ha/day) from the DT treatment over the period from June 18 to December 31, 1980. ---.. _._--_. __ .. _----.---> ........ ----.. ---.--.-------------.-.--------. -----_. __ ... JULIAN DAYS t80 200 220 240260 280 300 320 340 360 56 as -,::J 48 as ..c 40l DT SOIL (") E 32 .x :::> ...J 24 LL. a:: W 16 3: 8

PAGE 41

Fig. 11: 140 >, cU' -,::J I 20 loot C") .. E I I sol .X : ::::> .....J so IJ... :0:: W 40 tI 34 Drainage flux (m3/ha/day) from the DTL treCltment over the period from 18 to Dece"!tlber 3], 1980. JULIAN DAYS I SO, ,200 220 240 2S0 2S0 300 320 340 3S0 ; i U, i cUi DTL SOIL 0

PAGE 42

35 cm/day respectively for ST, DT, and DTL plots. Drainage was very sma.ll during the summer period for the deep tillage plots. Drainage outflow through subsurface tubes has been consistently greater for all shallow tillage (ST) treatments at the SWAP citrus grove than from both deep tillage (DT and DTL) treatments. Mean annual drainage discharges over the period from 1971 through 1980 (Dr. J. S. Rogers. 1982. private communication.) for ST, DT, and DTL treatments were 55.4, 33.3, and 28.9 cm. As expected, the highest of N (the sum of NH4-N.pluSN03-N) in the drainage water occurred during the first 2 months after application (Figs. 12, 13, and 14) of fertilizer. For ST drainage water, maximum N concentrations of 25 and 11.7 mg/l occurred during 25 and 48 days after fer.tilization.. For DT soil, maximum N concentrations of 8.2 and 10.3 mg/1 occurred during 20 and 27 days after ferti1ization. For DTL soil, maximum N concentrations of 6.0 and 13.4 mg/l odcurred 20 and 26 days after fertiliza tion. Thus N concentrations in drainage water were generally higher inST (Mansell, et ale 1980) than for the deeptilled plots (DT and DTL). The higher N concentrations in the drainage is largely attributable to the low cation exchange capacity of this soil. Concentrations of N were lower during the fall versus the summer for DT and DTL drainage waters. This observation was true except for an unexplainable peak that occurred during the latter part of December.

PAGE 43

36 Isotopic N analysis indicated (Figs. 12, 13, and 14) 15 that N from the N-depleted fertilizer accounted for as much as 50% of the N in drainage water that occurred during .the summer. This was true for all three plots. The percentage of N attributable to the labelled fertilizer decreased .with time during the season. For example on November 15 (163 days after application of the fertilizer) 15 percentages of N due to the N-depleted fertilizer were 20, 17, and 25% respectively for ST, DT and DTL soils. This data implies that much of the N being discharged through the drains was due to residual N from mineralization of soil matter. and from previous applications of fertilizer. Values of nitrogen flux through the drains were cal-culated by multiplying water flux and N concentrations for each day that drainage occurred This data (Figs. 15, 16, and 17) indicate that rates of N discharge were greater in the summer than in the fall for .the shallow tillage treat-ment, but were greater in the fall than in the summer for the deep tillage plots. For the ST plot, rates of nitrogen leaching losses as high as 0.850 kg/day were observed for 2 drainage events during the summer. Much smaller rates of 0.012 and 0.015 kg/day were observed for single summer drainage events in each of DT and DTL plots. This informa-tion further substantiates (Mansell et al. 1977) the effec-tiveness of deep tillage in qecreasing N leaching loss in these acid sandy soils. Measurements of the area beneath the curves in Fig. 15 shows total leaching losses of

PAGE 44

37 Fig. 12: Concentration (mg/l) of extracted and labelled N c o .-8 +-' ro L OJ u C4 o U 2 o (NH4-N plus N0 3-N) in drainage water from the ST plot over the period from June 18, 1980 to February 14, 1981. Julian Days 180 200 220 240 260 280 3 320 340 360 ST Soil .-. N03-:N + NH4-N .---. Labelled N _--A---_____ \-----"'1--. ----_ .. 1------1-. -.. -------.. ----------.

PAGE 45

38 Fig. 13: Concentration (mg/l) of extracted and labelled 24 20 -13 .-.J --.12 0) E --10 c o . u t::4 o U 2 o N(NH 4-N plus N0 3':"N) in drainage wa.ter from the DT plot over the period from June 18, 1980 to Februa.ry 14, 1981. Julian 240 Dr Soil ._. N03 -N + ._--. Labelled N ---------------

PAGE 46

24 '-" 10 c o .-8 +-' CU L +-'6 c Q) u c4 o U 2 o 39 Fig. 14: Concentration (mg/l) of extracted and .labelled N(NH 4-N plus N0 3-N) in drainage water from the DTL plot over the period from June 18, 1980 to February 14, 1981. Julian 180 200 2 0 240 I .'1 I DTL Soil .-. N03-N + NH4-N ._-_. Labelled. N ___ lvL J-----vL -...... ---''''----'"------, ......

PAGE 47

40 Fig. 15: Flux (g/ha/day) of extracted and labelled N(NH 4-N plus N0 3-N) in drainage water from the ST treatment over the period from June 18 to December 840 CU -g CU 720 ..&:; "-0) 600 X ::> 480 lJ.. Z 360 W (!) 0 240 0::: Z 120 0 31, 1980. JULIAN DAYS 180 200 220 240 260 280 300 .320 340 360 II /I II II I, II I I I I I II I \ ST SOIL eN03-N+NH4-N LABELLED N ----...:---' -.--,-_._--------------.-"----

PAGE 48

41 Fig. 16: Flux (g/ha/day) of extracted and labelled N(NH4 -N plus N03 -N) in drainagewat.er from the DT treatment over the period from June 18 to December 31, 198"0. JULIAN DAYS ISO 200 220 40 260 280 300 320 340 360 2S tU "'0 .24 tU .c Q 20 DT SOIL X eNOa-N+NH .. -N. ::::> 16 -LABELLED N ;..J I.L. Z 1 LLJ. (!) 0 S 0:: Z __ -e

PAGE 49

42 Fig. 17: Flux (g/ha/day) of extracted and labelled N(NH 4-N plus N0 3-N) iIi drainage water from the DTL treatment over the period from June .. 18 to December 31, 1980. JULIAN DAYS 180 200 220 240 260 280 300 '320 340 360 112 as as 96 .&:. DTL.$OIL IC) 80 X ::> 64 LABELLED N ....J LL Z 48 W (!) 0 32 c:: t-Z

PAGE 50

43 total and N were 9.10 and 4.34 kg/ha, respectively, for the first 196 days after application of the fertilizer. Thus 48% of the N discharged from the ST drain was attribut-15 able to the N-depleted (NH4)2S04. Total leaching losses of N in drainage from the deep-tilled plots were extremely small. Distributions of total and labelledNH4-N with soil depth are shown for each of the three plots in Figs. 18, and 20 for dates corresponding to 12, 42, 75, and 134 days after application of the fertilizer. For the deep-tilled plots, NH4-Ncontents in the soil profile were low in comparison .to those for the shallow-tilled plot. Total NH4-N quantities in the soil 12 days after fertilization (Table 5) were 142.0, 25.9, and 71.4 kg/ha, respectively, for ST, DT,. and DTL plots. After 42 days from fertilization quantities of NH4-N had decreased to 57.3, 8.5, and 19.3 kg/ha for ST, D'l' and DTL plots.. This. data suggests that much of the NH4-N applied to the deep-tilled plots was either converted to N0 3-N during ni"t7rification and/or absorbed by plant roots from citrus orbahiagrass since volatilization and leaching losses of N extremely small. Distributions of total and labelled the soil {Figs 21, 22, and 23) do in fact suggest that nitrification of applied NH4-N occurred. A sandy soil in British Columbia, Canada with relatively low organic carbon content was shown by Kowalenko (1980) to support suprisingly high microbial activity for N transformations. Between 12 and 42 days after fertilization, quan-

PAGE 51

44 Fig. 18: Distributions of extracted and isotopLcally labelled NH4-N in the ST soil profile rluring 1980. for dates corresponding to 12, 42, 75 and 134 days after fertilization. NH4,"N (/l9 /9) 00 5 10 15 10 / / ;. / / x :r: 30 I. t-/ a.. / w 40 I o / ... J / x 050 / CJ) / 60 I / :r: W 040 ...J 6 CJ) 50 60 16 JUKE ST X EXTRACTE-D LABELLED 19 AUGUST ST 20 X EXTRACTED: LABELLED NH -N (/lg/g) 4 00. 5 10 15 10 \ \ ., E 20 I 30 I. a.. 16 JULY ST 20 ...J 1 050, CJ) X EXTRACTED LABELLED 60 NH4-N (/lg/g) 00 5 10 15 20 10 ,...,. E :r: t30 a.. w o ...J 40 o CJ) 50 60 16 OCTOBER ST X EXTRACTED LABELLED

PAGE 52

Fig. 19: 45 Distributions of extracted .and isotopically labelled NH4-N in the DT soil profile during 1980 for dates corresponding to 12, 42, 75, and 134 days after fertilization. NH4-N (lIg / 9) NH4 -N (jlg/g) 00 5 10 15 20 00 5 10 15 '20 10 10 20 ,po ... ::I: 40 16 JUNE ::I: 40 16 JULY DT OT no no w 50 x EXTRACTED W 50 X EXTRACTED 0 LABELLED 0 LABELLED :! 60 0 :! 60 0 C/) C/) 10 10 80 80 90 90 100 100 NH4-N (1I9/ 9) NH4-N (1I9/ 9) 00 5 10 15 20 5 10 1 20 10 I 20 e 30 ,... .., '-" e ::I: 40 19 AUGUST ... 40 16 OCTOBER '-" IOT ::I: x OT no w 50 50 0 )( EXTRACTED no )( EXTRACTED LABELLED W LABELLED -l 0 060 -l 60 C/) 0 70 C/) 70 80 80 90 90 100 --100 ----

PAGE 53

Fig. 20: 46 Distributions of extracted and isotopically labelled NH4-N in the DTL soil profile during 1980 for dates corresponding to 12, 42, 75, and 134 days after fertilization. NH4-N <1I9/9) NH4-N (llg/g) 00 5 10 15 00 5 10 15 20 "/ 10 I 10 I I / x 20 / 20 / I E 30 I x u I v :r:: 40 / 16 JUNE :r:: 40 16 JULY I- OTl I-OTl a.. a.. W 50 x EXTRACTED W 50 X EXTRACTED 0 LABELLED 0 I LABELLED ..J ..J (560 <560 en en 70 I 70 I 80 80 t 90 90 I 100 100 NH4-N
PAGE 54

47 rig. 21: Distributions of extracted and isotopically labelled N0 3-N in the ST soil profile during 1980 for dates corresponding to 12, 42, 75, and 134 days after fertilization. g) O 5 10 15 00 5 10 15 20 25 0 10 E 20 u \oJ :I: 30 I-0. \ 16 JUNE IJ.J 40 16 JULY ST C I ST ..J x EXTRACTED 050 I X EXTRACTED LABELLED en I LABELLED I .60. 70 N0 3-N
PAGE 55

Fig. 22: 48 Distributions of nnd isotopically labelled N0 3-N in the DT soil profile during 1980 for dates corresponding to 12, 42, 75 and 134 days after fertilization. N0:3-N (1I9/ 9) 00 5 10 .15 20 10 20 (560 (/) ::t: 40 l-0. W 50 0 (560 (/) 70 I 80 I 90. 100 x 16 JUNE DT X EXTRACTED LABELLED N0:3-N (1I91 g) 5 10 15 20 19 AUGI.lST DT X f:XTRA.CTEO LABLLED 20 E 30 ... '-l ::t: 40 Q.. W 50 o (560 (/) 70 80 90 N03-N gl 5 10 15. 20 16 JULY DT X ExTRACTED LABELLED N03-N (1I9/ 9) 5 10 15 20 160CT08ER DT-X EXTRACTED LABELLED

PAGE 56

Fig. 23: 49 Distributions of extracted and isotopically labelled N0 3-N in the DTL soil profile during 1980 for dates corresponding to 12, 42, 75, and 134 days after fertilization. 10 ...J (560 en 70 I x 80 90 100 N03-N <1I91 g) 5 10 15 20 16 JUNE OTL X EXTRACTED LABELLED 19 AUGUST OTL x EXTRACTED LABELLED 20 ...J (560 en 70 N03-N (lIg l g) 5 10 15 20 16 JULY OTL X EXTRACTED LABELLED 25 N03-N (1I9/ 9) 5 10 15 20 10 20 ,.., E 30 u '"" ::I: 40 16 OCTOBER t-OlL CL. 50 X EXTRACTED LABELLED ...J (560 en 70 80 90 100

PAGE 57

50 tities (TableS) of total N03-N in the soil increased considerably for all treatment p16ts this study but less for the DT soil. After 42 days, the N03-N decreased with time for all treatments probably due to uptake by plant roots. Quantities of the soil N due to the applied lSN depleted fertilizer are given in Table 6. Twelve days after fertilization the sum of the labelled NH4-N and N03-N in the soil represented 105, 28, and 54% of the fertilizer, respec-tively, for ST, DT and DTL plots. The low recoveries of la-belled N in the deep-tilled soils imply that uptake of N was much faster during the first 12 days after fertilizer application than in the ST soil. This deduction assumes the extraction procedure was 100% effective for DT and DTL soils. The ratios cif N03-N to NH4-N for the labelled N in the soil were 0.2, 1.5, and 0.5 respectively for ST, DT, and DTL plots. The larger ratios for the DT and DTL plots suggest that nitrification occurred more rapidly in the deep-tilled soils than in the shallow-tilled soil during the first 12 days after fertilization. After 42 days of elapsed time these ratios were 1.0, 7.8, arid 7.2 for ST, DT, and DTL plots. These dramatic increases of the N03-N/NH4-N ratios over a 30-day period indicate that nitrfication occurred in all soils. After 134 days of elapsed time, quantities of labelled N in all soils were less than 3% of the applied fertilizer, and the ratios of N03-N/NH4-N for ST, DT and DTL plots were 1. 9, 0.6, and 0.9 respectively. ..

PAGE 58

51 Table 5. Mean quantities of extracted NH4-N and N0 3-N in soil profiles from ST, OT, and DTL tillage Time ments on dates corresponding to 12, 42, 75, and ST 134 days after application of 115 kg/ha of 15 N-depleted NH4-N on June 5, 1980. NH -N 4 NO -N 3 --B.!!4-N + N03::L. DT DTL ST DT OTL ST DT DTL (elapsed days)-----(kg/ha)---------(kg/ha)-----------(kg/ha)----12 142.0 25.9 71.4 40:8 50.6 33.4 182.8 76.5 104.8 42 57.3 8.5 19.3 96.3 58.1 77.3 153.6 66.6 96.6 75 12.2 5.6 14.4 23.9 26.0 30.7 36.1 31.6 45.1 134 10.5 20.5 17.4 19.7 12.1 9.5 30.2 32'.6 26.9

PAGE 59

52 Table 6. Mean quantities of isotopically labelled NH4-N and in soil profiles from ST, DT, and DTL tillage treatments on dates corresponding to 12, 42, 75, and 134 days after application of 115 15 kg/ha of N-depleted NH4-N on June 5, 1980. NH -N 4 NO -N 3 NH -N --4 + Time ST DT DTL ST DT DTL ST DT DTL (elapsed-----(kg/ha)-----12 102.6 13.2 42.7 17.6 19.2 19.7 120.2 32.4 62.4 42 .11. 9. 0.8 2.0 11.5 6.2 14.3 23.4 7.0 16.3 75 1.8 1.0 1.1 3.7 5.6, 3.5 5.5 6.6 4.6 134 1.1 1.9 0.9 2.1 1.1 0.8 3.2 3.0 1.7

PAGE 60

53 The distributions of total N03-N content in the soil (Figs. 21, 22 and 23) imply that some downward movement occurred through the profile during leaching but that a sink possibility uptake by plant roots had extracted fertilizer N from the soil of each of the tillage treatments. Distributions of the sum of N03-N and NH4-N (Figs. 24, 25 and. 26) also support that hypothesis. Results from the drainage discharge of N however show that the leaching loss was quite small during the 6-month period of this study. Previous investigators (Calvert et ale 1977) have shown that citrus root systems are generally shallower and less extensive in the profile for shallow-tilled versus deep-tilled DTL soil. Thus one should expect more extensive N uptake from citrus trees growing in DTL soil profiles compared to ST. Due to the lack of rain and the c.orresponding lower water content of the shallow-tilled soil compared to the deep-tilled soil, only a very limited number of soil solution samples were collected during August. Despite the very small number of samples reported in Tables 7-9, N03-N was definitely the predominant form of N in the ST soil so lution. The decrease of the geometric mean concentration of labelled N with time over this period implies that occurred. The decrease of the geometric mean concentration of labelled N implies that plant uptake of N actively occurred during this period. Samples were more eaf!ily extracted from the generally wetter DT and DTL soils as shown by the N concentrations in Tables 10-15. Geometric

PAGE 61

Fig. 24: 54 Distributions of extracted and labelled NH4-N plus N0 3-N in the ST soil profile during 1980 for 12, 42, 75, and 134 days after fertilization. N03,N + NH4-N (lIgl g) N03-N + NH4-N (lIgl g) 0 10 0 20 30 40 10 20 30 40 /. 10 / / .f x 20 I ....... / I E / E u u I '-' t. '-' ::r:: 30 ::r:: 30 1-. I l-I a.. f. a.. I UJ 16 JUNE-ST UJ 16 JULY-ST 0 40 f 0 40 I ...J .f x ...J 6 X EXTRACTED 6 X EXTRACTED (J) 50 I LABELLED (J) 50 I LABELLED I I / 60 60 I 70 -70 N03-N. NH4-N (lIgl g) N03-N+ NH4-N (1I9/ 9) 20 30 40 10 20 30 40. 20 '"' '"" E E u u '-' '-' ::r:: 30 ::r:: 30 l-Ia.. a.. IJ.J 19 AUGUST-ST IJ.J 40 16 OCTOBER-ST 0 40 0 ...J ...J 6 X EXTRACTED 6 X EXTRACTED (J) 50 LABELLED (J) 50 LABELLED 60 70 70 _._---

PAGE 62

Fig. 25: 55 Distributions of extracted and labelled NH4-N plus N03-N in the DT soil profile during 1980 for 12, 42, 75, and 134 days after fertilization. N03-N+ NH4-N + NH4-N g) 00 10 20 30 40 "1(" 30 40 10 x 20 30 '" 'E E v v 16 JULY-DT 40 16 JUNE-DT 40 :c :c l-I-0.. X EXTRACTED 0.. X EXTRACTED W LABELLED W LABELLED 0 50 0 50 ..J ..J 6 6 (J) 60 (J) 60 70 70 I x BO 80 I 90 90 r 1 + NH4-N g) N03-N N'14-N g) 10 20 30 40 '11' 20 30 40 10 x 20 20 30 30 '" '" E e .. .... I 8 OCT08ER -DT 40 19 AUOUBTDT 40 :c :c l-I-0.. )C, EXTRACTED 0.. X EXTRACTED W LABELLED W 50 LAI[LLtD 0 50 0 d I 0 (J) 80 (J) 80 I 70 I 70 eo eo I 90 I 90 .. -------.-.. -... _._-

PAGE 63

Fig. 26: 56 Distributions of extracted and labelled NH4-N plus N0 3-N in the DTL soil profile during 1980 for 12, 42, 75, and 134 days after fertilization. N03-N NH4-N g) NO)-N' NH4-N g) 00 10 20 30 40 40 10 20 30 'E 'E 16 JUNE-DTL 40 :x: :x: l-I-a.. x EXTRACTED a.. l.IJ LABEllED l.IJ C C ...J 0 CJ) 60 CJ) 70 I 80 80 90 N03-N. NH4-N g) (pg/g) 0 0 10 20 30 40 0 10 0 20 30 40 10 20 30 'E '"' E -3 I 9 AUGUST -DTL 40 I 6 OCTOBERDTl :x: :x: l-I-a.. x EXTRACTED a.. x EXTRACTED l.IJ LABELLED l.IJ LABELLED C 50 C 50 ...J ...J 0 0 CJ) 60 CJ) 60 70 70 x 80 80 90 90

PAGE 64

57 Table 7. Concentrations (mg/l) of total and labelled N in soil solution samples from 4 depths and 6 lateral distances from the central drain in the ST plot on August 1, 1980 (Julian Day 214). Geometric means are shown in parentheses. Total NH4-N (0.15 mg:/ 1) Soil Lateral DeEth (em) Distance (cm) : 5 50 100 350 500 900 60 75 90 105 0.38 0.06 Total N03-N (12.90 mg:/l) Soil Lateral DeEth Distance (cm) : 5 50 100 350 500 900 ( cm) 60 75 90 105 10.60 15.70 Labelled NH -N + NO -N (7.89 mg/l) Soil Lateral 4 3 Depth Distance (cm) : 5 50 100 350 500 900 (cm) 60 75 90 105 4.70 13.26

PAGE 65

58 Table 8. Concentrations (mg/l) of total and labelled N in Soil DeEth (em) 60 75 90 105 Soil DeEth (em) 60 75 90 105 soil solution samples from 4 depths and 6 lateral distances from the central drain in the ST plot on August 13, 1980 (Julian Day 226). Geometric means shown in parentheses. Total NH -N (0.32 mg/l ) Lateral 4 Distance (em) : 5 50 100 350 500 900 0.12 7.97 0.95 0.01 0.33 0.34 Total NO -N (12.78 mgt 1) Lateral 3 Distance (em) : 5 !;O 100 350 500 900 11.00 21.0 3.80 84.0 10.00 ,... 5.90 __ ...,...."..--.,...---"...----...;;;L;.;.a...;.b...;.e...;;;lo.;.;.l;..;;.e-"-d----N ...... H 4 -N + N03 .... N (0.52 mg/1) Soil Lateral DeEth Distance (em) : 5 100 350 500 900 (em) 60 75 90 105 0.65 0.52 0.16 0.35

PAGE 66

59 Table 9. Concentrations (mg/l) of total and labelled N in Soil DeEth (cm) 60 75 90 105 Soil DeEth (cm) 60 75 90 105 Soil DeEth ( cm) 60 75 90 105 soil solution samples from 4 depths and 6 lateral distances from the central drain in the ST plot on August 20, 1980 (Julian Day 233). Geometric means shown in parentheses. Total NH -N (0.11 mg/l) Lateral 4 Distance (cm) : 5 50 100 350 500 900 0.98 0.25 0.01 0.07 Total NO -N (1.19 mg/l) Lateral 3 Distance (cm) : 5 50 100 350 500 900 1.2 2.70 3.30 0.19 Labelled NH4-N + NO -N (0.38 mg/l) Lateral 3 Distance (cm) : .5 50 100 350 500 900 0.54 0.74 0.83 0.06

PAGE 67

60 Table 10. Concentrations (mg/l) of total and labelled N in soil solution samples from 4 depths and 6 lateral distances from the central drain in the DT plot on August 1, 1980 (Julian Day 214). Geometric means are shown in parentheses. Total NH -N (0.50 mg/ l) Soil Lateral 4 DeEth Distance (cm) : 5 50 100 350 500 900 (cm) 60 75 90 0.60 0.49 0.12 0.33 105 0.83 0.93 0.48 0.91 Total NO -N (8.79 mg/l) Soil Lateral "'" 3 DeEth Distance (cm) : 5 50 100 350 500 900 (em) 60 ... 75 90 14.40 8.70 5.50 9.10 105 9.60 9.60 8.70 7.10 Labelled + N0 3...;N (4.80 Soil Lateral DeEth Distance (cm) : 5 50 100 350 500 900 (cm) 60 75 .,.. 90 8.30 7.25 3.28 5.51 105 5.77 8.31 5.36 4.68

PAGE 68

61 Table 11. Concentrations (mg/l) of total and labelled N in soil solution samples from 4 depths and 6 lateral distances from the central drain in the DT plot on August 13, 1980 (Julian Day 226). Geometric means are shown in parentheses. Total NH -N (0.12 mg/l) Soil Lateral 4 DeEth Distance (cm) : 5 50 100 350 500 900 (cm) 60 0.30 75 0.01 0.09 0.01 0.01 0.01 0.01 90 1.34 0.01 2.49 0.22 105 6.17 0.01 0.60 1.63 0.14 1.74 Total NO -N (9.62 mg/l) Soil Lateral 3 DeEth Distance (cm) : 5 50 100 350 500 900 (cm) 60 6.20 75 14.50 44.50 68.00 38.00 9.40 8.00 90 55.00 29.00 6.20 3.80 105 8.50 39.00 12.50 8.10 5.25 3.65 Soil Lateral Labelled NH4-N + N0 3-N ( 1. 71 mg/ l) DeEth (cm) Distance (cm) : 5 50 100 350 500 900 60 0.16 75 3.47 17.78 29.56 1. 87 0.79 0.20 90 22.46 12.61 0.43 0.10 105 3.51 15.55 5.69 0.48 0.45 0.13

PAGE 69

62 Table 12. Concentrations (mg/l) of total and labelled N in soil solution samples from 4 depths and 6 lateral distances from the central drain in the DT plot on August 20, 1980 (Julian Day 233). Geometric means are shown in parentheses. Total NH -N (0.02 mg/l) Soil Lateral 4 DeEth Distance (cm) : 5 50 100 350 500 900 (cm) 60 0.01 0.01 75 0.01 0.01 0.01 0.01 90 0.01 0.01 0.01 0.01 0.13 105 0.01 0.01 0.54 0.74 Total NO -N (4.03 mg/l) Soil Lateral 3 DeEth Distance (cm) : 5 50 100 350 500 900 (cm) 60 1.05 1.90 75 18.00 40.00 1. 60 1. 60 90 23.50 52.50 5.75 2.00 0.61 105 7.00 10.00 0.81 0.67 Labelled NH4-N + NO -N (0.51 mg/l) Soil Lateral 3 DeEth Distance (cm) : 5 50 100 350 500 900 (cm) 60 0.57 0.01 75 9.31 21.56 0.08 0.17 90 12.16 28.30 0.29 0.21 0.01 105 3.63 5.39 0.07 0.01

PAGE 70

63 Table 13. Concentrations (mg/l) of total and labelled N in soil solution samples from 4 depths and 6 lateral distances from the central drain in the DTL plot on August 1, 1980 (Julian Day 214). Geometric means are shown in parentheses. Total NH -N (0 18 mg/l) Soil Lateral 4 DeEth Distance (cm) : 5 50 100 350 500 900 (cm) 60 0.05 0.07 0.06 0.29 0.38 0.28 75 0.22 0.04 0.24 0.36 0.41 0.23 90 0.04 0.08 0.43 0.39 0.33 0.34 105 0.06 0.09 0.27 0.38 0.32 0.33 Total NO -N ( 6 8 7 mg / 1 ) Soil Lateral 3 DeEth Distance (cm) : 5 50 100 350 500 900 (cm) 60 11.2 18.0 7.1 5.0 4.3 4.3 75 14.3 15.7 6.7 5.0 3.5 4.5 90 13.9 15.0 7.4 6.1 4.1 4.1 105 12.8 13.6 5.2 4.1 4.0 3.3 Labelled NH4-N + NO -N ( 3 10 mg:ll) Soil Lateral 3 DeEth Distance (cm) : 5 50 100 350 500 900 (em) 60 7.2 11.5 4.6 2.2 1.7 0.7 75 9.3 10.1 4.4 2.2 1.4 0.8 90 8.9 9.6 5.0 2.7 1.6 0.7 105 8.2 8.7 3.5 1.9 1.6 0.6

PAGE 71

64 Table 14. Concentrations (mg/l) of total and.labelled N in soil solution samples from 4 depths and 6 lateral distances from the central drain in the DTL plot on August 13, 1980 (Julian Day 226). Geometric means are shown in parentheses. Total NH -N (0.18 mg/l) Soil Lateral 4 DeEth Distance (cm) : 5 50 100 350 500 900 (cm) 60 0.01 0.01 0.12 0.62 75 0.01 0.24 1. 08 3.04 2.90 5.51 90 0.10 0.20 0.01 0.01 0.49 3.63 105 0.55 0.01 0.10 0.01 1.53 2.20 Total NO -N (14.76 mg/ 1) Soil Lateral 3 DeEth Distance (cm) : 5 50 100 350 500 900 (em) 60 10.50 10.00 32.50 10.50 75 10.00 12.50 42.00 29.00 16.00 34.00 90 10.50 9.40 17.00 17.00 27.50 21.00 105 8.10 20.00 13.50 14.50 21.00 16.00 Labelled NH4-N + NO-N (0.97 mg/l) Soil Lateral 3 DeEth Distance (cm) : 5 50 100 350 500 900 (cm) 60 1.19 0.85 0.29 75 1.13 1.11 3.66 2.62 3.65 0.95 90 1.20 0.84 1.45 1.39 5.41 0.64 105 0.98 1. 75 1.16 1.19 4.35 0.48

PAGE 72

65 Table 15. Concentrations (mg/l) of total and labelled N in soil solution samples from 4 depths and 6 lateral distances from the central drain in the DTL plot on August 20, 1980 (Julian Day 233). Geometric means shown in parentheses. Total NH -N (0.13 mg/l) Soil Lateral 4 DeEth Distance (cm) : 5 50 100 350 500 900 (cm) 60 0.01 0.01 0.01 0.28 75 0.82 0.96 90 0.01 0.06 0.23 0.58 0.27 0.26 105 0.43 0.04 0.14 0.72 0.73 Total NO -N (4.14 mg/l) Soil Lateral 3 DeEth Distance (cm) : 5 50 100 350 500 900 ( cm) 60 2.10 2.60 5.30 18.00 75 1.85 29.00 90 4.00 1. 40 3.70 7.50 11.00 8.00 105 1.00 3.20 1. 50 5.30 3.20 Labelled NH4-N + NO -N (0.09 mg/l) Soil Lateral 3 DeEth Distance (cm) : 5 50 100 350 500 900 (cm) 60 0.06 0.07 0.03 0.42 75 0.07 0.19 90 0.11 0.04 0.10 0.05 0.71 0.19 105 0.04 0.08 0.01 0.38 0.09

PAGE 73

66 means for DT soil solution were only 0.50, 0.12, and 0.02 mg/l of NH4-N respectively for 57, 69, and 76 days after fertilizer application; however, geometric means for N0 3-N were 8.79, 9.62, and 4.03 mg/l on these same dates. These concentrations were much higher however than those observed in drainage water (Fig. 13) after those dates. This vation was probably due to plant uptake combined with a limited number of subsurface drainage periods. This data also shows that N0 3-N constitutes the predominant form for N in the soil solution. Geometric means fOr labelled N for 57, 69, and 76 days after fertili.ation were 4.80, 1.71, and 0.51 mg/l respectively. Thus the concentration of solution N attributable to the 15N-depleted (NH4)2S04 was observed to decrease more than 9 times within the 19-day period. This information is also consistent with the reasoning that N uptake by plant roots was very significant in this deep-tilled DT soil. The slight increase in N0 3-N in soil solution between 57 and 69 days after fertilization further suggests that some nitrification occurred, but was probably limited in rate due to acid subsoil conditions. Geometric mean concentrations of NH4-N in bTL soil solution were 0.18, 0.18, and 0.13 mg/l for 57, 69, and 76 days after fertilization; however geometric means for N0 3-N were 6.87, 14.76, and 4.14 mg/l on these same dates. As with DT soil, the N concentrations in soil solution were considerably higher than in the drainage (Fig. 14) water essentially for the same reasons. Nitrate

PAGE 74

67 was the dominant form of N in solution and the large in-crease in concentration between 57 and 69 days after fertilization implies nitrification had occurred. Geometric means for labelled N for 57, 69, and 76 days after fer-tilization were 3.10, 0.97, and 0.09 mg/l respectively. For this deep-tilled treatment, the concentration of solution N attributable to the 15N-depleted (NH4)2S04 was observed to decrease more than 34 times during the 19-day period. This observation suggests that rates of N uptake by plant roots were higher in DTL soil in comparison to DT soil, probably due to higher pH values in the subsoil .. Calvert et ale (1977) have shown that citrus yields were higher and that root systems were deeper in DTL soil compared to DT soil. Concentrations for selected cations in the DTL soil solution are presented in Table 16 for 57 days after application of fertilizer. Geometric mean concentrations + +2 +2 -4 -3 for K ,Ca ,and Mg were 9.2 x 10 2.15 x 10 and -3 2.47 x 10 m.e./l, respectively. The presence of these relatively high cation concentrations in the soil solution implies that the ion exchange sites in this soil are d 1 d b d' 1 +2 d +2 pre omlnant y occuple y lva ent Ca an Mg catlons which are more competitive than NH: ions for the exchange sii-es. Citrus leaves (Table 17) sampled during the summer and had 2.28 to 2.69% Nand 12 to 15% of the tissue N was attributnble to the 15N-depleted fertilizer. Essentially no difference were observed between the 3 plots.

PAGE 75

68 Table 16. Concentrations (mg/l) of K+, ca+2 and Mg+2 soil solution samples from 4 depths ane; 6 la
PAGE 76

69 Table 17. Mean percentages (%) of N in leaf tissue and percentages (%) of leaf N that was labelled for foliage samples taken in July and October of 1980. Mean values represent 12 rootstock-scion comb inations of citrus. Date July October Percentage N in Leaves ST 2.69 2.30 DT 2.50 2.28 DTL 2.50 2.30 Percentage Labelled N ST DT DTL 13.33 15.07 14.73 13.68 13.06 12.67

PAGE 77

70 Chapter 5: Summary and Conclusions During the summer of 1980, 115 kg/ha of N in the form 15 of N-depleted (NH4'2S04 was applied to portions of a 10-year old experimental citrus grove in South Florida for the purpose of determining the fate of fertilizer N applied to an acid, sandy soil (Spodosol). The influence of deep tillage upon the disposition of the applied N was determined by applying the labelled fertilizer to shallow-tilled (15 cm depth) ST, deep-tilled (105 cm depth) DT, and limed (56 metric tons/ha) plus deep-tilled (105 cm depth) field plots of a Spodosol with a 10-20 cm thick spodic layer located at ap-proximately 80 cm depth. All plots have received annual applications of 2.24 metric tons per ha of limestone. The ST plot is typical of tillage and liming practices for agricul-tural crops growing in Florida Spodosols. The citrus grove was surface-drained by elevated beds and subsurfaced-drained by 10-cm diameter plastic tubes placed 107 cm deep and 18.3 m apart. Elevated beds with 2 rows of citrus trees were oriented perpendicular to the drain tubes, and the area between tree rows supported a thick Bahiagrass sod. Soil, soil solution, drainage water, and citrus leaf samples were taken at selected times during the first six months follow-ing the application of fertilizer. Chemical analyses were performed for NH4-N and N03-N concentrations and isotopic 15N/14N ratios were determined. A local drought during the summer and fall restricted the quantities of drainage as well as the sampling frequency for soil solution. During

PAGE 78

71 typical years as much as 90% of the total annual rainfall often occurs during the summer and fall seasons. Accumulative leaching losses of NH4-N and N0 3-N in drainage water discharge from the deep-tilled DT and DTL p19ts were extremely small during the first 196 days after the application of fertilizer. The small leaching losses for DT and DTL plots were due to low N concentrations in the water but more importantly to small water discharge rates. Drainage water from the shallow-tilled ST ptot resulted in a slightly larger loss of 4.76 and 4.34 kg/ha respectively of unlabelled (residual) and labelled nitrogen. Thus the la-belled N leached from the ST soil represented only 4% of the 15 quantity fertilizer nitrogen. Concentrations of residual plus labelled N in ST drainage water were as high as 25 mg/l during a drainage event which occurred with-in the first month after fertilization but decreased to about 1 mg/l within 4 months after fertilization. The rela-tive influence of N in the drainage water from ST upon water quality of the downstream canal was small because of the small volumes of drainage that occurred during 1980. Nitrogen discharged in drainage water from all plots was predominantly in the form of N0 3-N with lesser amounts as NH4-N. Analyses of soil cores to a depth of 70 cm in the ST plot and 100 cm in the deep-tilled plots revealed that the 15 N-depleted NH4-N was largely nitrified to the N0 3-N form within the first 30 days. During the first 12 days

PAGE 79

72 nitrification appeared to proceed at faster rates in the deep-tilled plots as compared to the shallow-tilled plot possibly due to slightly higher water retention capacity and higher pH of the DTL subsoil. Initially, larger quantities of residual and labelled N (N0 3-N plus NH4-N) were present in the soil profile for the ST plot (182.8 kg/hal than for DT (76.5 kg/ha) and DTL (104.8 kg/ha) plots. The quantities of labelled N in the soil at that date were equivalent to 105, 28, and 54% of the 15 applied N-depleted (NH4}2S04 for ST, DT, and DTL plots re-spectively. These data indicate that within 12 days after fertilizer application, 72 and 46% of the labelled fertiliz-er N, respectively, had been removed from the profiles for DT and DTL plots. Virtually all of the labelled N was pre-sent in the shallow-tilled ST soil, however. Since leaching losses for the deep-tilled soils were extremely small and denitrification losses of N0 3-N were most unlikely due to the infrequent rainfall during that period, the unaccounted for N appears to have been absorbed by root systems of bahiagrass and citrus. Volatilization losses of the applied (NH4)2S04 were less than 0.5% for all three plots. Earlier results by.Calvert et ale (1977) showed that fruit yields were significantly higher and root development more exten-sive in the.soil profile for the DTL treatment than for ST treatment. Yields and root development for the DT treat-ments was intermediate. Shallower root systems in the shallow-tilled plot would thus be expected to have lower

PAGE 80

73 rates of absorption of soil N by actively growing roots than either of the deep-tilled plots. Total quantities of N measured in ST, DT, and DTL soil profiles for 12, 42, 75, and 134 days after fertilization showed dramatic decreases in labelled as well as residual N from previous applications of fertilizer. After the first month, much of the soil N appeared to be coming from non-fertilizer sources of N, presumably from mineralization of soil organic matter. By 134 days after fertilization, 15 less than 3% of the N-depleted fertilizer N was still present in the soil. These data also imply that most of the applied fertilizer N was utilized by citrus trees and bahiagrass within 4.5 months after application. Previous investigations with non-labelled N (Mansell et ale 1977) and with higher rainfall amounts during summer months revealed that an average of 20% of the fertilizer applied to the shallow-tilled plot of this same citrus grove was leached over a three-year period. Average leaching losses for the deep-tilled plots however were much less ( 5%). As seen from the current investigation where high residual concentrations of soil N occur from previous fertilizer ap-15 the use N-depleted fertilizer N provides a more effective means to study the fate of applied N under field conditions. Thus results from this study show that fertilizer use efficiency for labelled N was higher in deep-

PAGE 81

74 tilled plots than in the ST plot. Even the N use efficiency for citrus growing in ST soil however was relatively h i .. gh due to the low frequency of drainage events during I98u. As expected, soil solution samples taken 57, 69 aId 7(>, days after fertilizer application showed that N0 3-N waf the predominant N form in solution in all plots. Concentrations of N in solution decreased during this I9-day period indicating that plant roots were actively absorbing N. Mean concentrations of solution N attributable to the I5N depleted (NH4)2S04 in the DTL soil were observed to decrease more than 34-fold during this 19-day period. Compared to the DT plot, rates of N uptake by plant roots appeared to be higher for the DTL plot during that period probably due to higher pH values observed for the subsoil. Isotopic analyses of citrus leaves during the summer and fall showed that 12 to 15% of N in the tissue was attributable to the I5N-depleted fertilizer. Concentrations of N in the tissue did riot vary greatly between plots. Results from this study using labelled N confirm earlier investigations (Mansell et ale 1977) that deep tillage plus lime incorporation into the profile of a Spodosol is an effective means to enhance root absorption of fertilizer N, minimize N leaching losses, and thus minimize contamination of water resources with N03-N. Shallow development of root systems, low cation and water retention characteristics, and high values of saturate4 hydraulic conductivity of shallowtilled Spodosols provide these acid, sandy soils with po-

PAGE 82

75 tential for leaching loss of fertilizer N during periods of soil drainage. Establishment of narrow (1.0 in wide) zones of deep tillage with lime only in the vicinity of the soil where citrus or other agricultural crops are to be located offers em effective means to increase fertilizer-and wateruse effi.ciencies for newly developed Spodosols. Restriction of the deep tillage operation to only the soil near the crop rows decreases the cost to a fraction of that required to deep till entire fields.

PAGE 83

76 Literature Cited 1. Bremner, J. M. 1965. Inorganic forms of ni trogerl. In Methods of Soil Analysis, C. A. Black (Ed). Part 2. Chapter 84, pp. 1179-1206. 2. Broadbent, F. E. and A. B. Carlton. 1980. Methodology for field trials with nitrogen-IS-depleted nitrogen. J. Environ. Qual. 9:236-242. 3. Calvert, D. V., E. H. stewart, R. S. Mansell, J. G. A. Fiskell, J. S. Rogers, L. S. Allen, and D. A. Graetz. 1981. Leaching losses of nitrate and phosphate from a Spodosol as influenced by tillage and irrigation level. Soil and Crop Sci. Soc. Florida Proc. 40:62-71. 4. Calvert, D. V., H. W. Ford, E. H. Stewart and F. G. Martin. 1977. Growth response of twelve citrus rootstock combinations on a Spodosol modified by deep tillage and profile drainage. Proc. of Int. Soc. Citriculture 1:79-84. 5. Fiskell, J. G. A. and O. V. Calvert. 1975. Effects of deep tillage, lime in corporation, and drainage on chemical properties of Spodosol profiles. Soil Sci. 120:132-139. 6. Hauck, R. D. and J. M. Bremner. 1976. Use of tracers for soil nitrogen research. Advances in Agronomy 28:219-266. 7. Hillel, D. 1980. Fundamentals of Soil Physics. Academic Press, New York, page 8.

PAGE 84

77 8. Kowalenko, C. G. 1980. Transport and transformation of fertilizer nitrogen in a sandy field plot using tracer techniques. Soil Sci. 129:218-221. 9. Knipling, E. B. and L. C. Hammond. 1971. The SWAP study of soil water management for citrus on flatwood soils in Florida. Soil and Crop Sci. Soc. Florida Proc. 31:208-210. 10. Mansell, R. S., D. V. Calvert, E. H. Stewart, W. B. Wheeler, J. S. Rogers, D. A. Graetz, L. H. Allen, A. R. Overman, and E. B. Knipling. 1977. Fertilizer and Pesticide Movement from Citrus Groves in Florida Flatwood Soils. Completion Report for Project R-800517, Environmental Protection Agency, Environmental Research Laboratory, Athens, GA. 11. Mansell, R. S., W. B. Wheeler, and D. V. Calvert. 1980. Leaching losses of two nutrients and an herbicide from two sandy soils during transient drainage. Soil Sci. 130:140-150. 12. Rogers, J. So, E. H. Stewart, D. V. Calvert, and R. S. Mansell. 1977. Water quality from a subsurface drained citrus grove. Third National Drainage Sympo sium Proceedings, American Society of Agricultural Engineers, pp. 99-103. 13. u.S. Environmental Protection Agency. 1976. Nitrogen, ammonia. In Methods for Chemical Analyses of Water and Water, U.S. E.P.A. Technical Transfer Manual. pp 1 6 9 -1 72

PAGE 85

78 Appendix: Synopsis from Master of Science Thesis (August 1982) for Miss Ko Hui Liu in the Soil Department, University of Florida. Title: Miscible Displacement of Aqueous Solutions of NH4-N through ColUmns of Sandy Soil. A. Summary Ion exchange for K+ and NH: during miscible displacement was studied using water-saturated soil columns of ST and DTLsoils. The soil in each ca.lumn was initially "saturated" with 0.03 N KCl prior to each displacement experiments. During the displacement experiments steady pore water velocities of 7.8 and 30.1 cm/hr were maintained in the columns. Potassium in the soil was exchanged with either 0.03 N NH4Cl or 0.01 N NH4Cl + 0.02 N kCl, and NH: in the soil was exchanged with either 0.03 NKCl or 0.01 N NH4Cl + 0.02 N KC1. Adsorption isotherms for NH: were determined and retardation R values were computed for each soil These R values were slightly less than those calculated from BTC's (solute breakthrough curves). Because of the presence of lime in the soils, ca2 + and Mg2 + were continually leached out of the soil columns. The presence of these divalent cations in the soil solution suppressed the sorption of NH: and K+. The zero point of charge ZPC of both soils were determined and found to be 4.8 and 6.1 for ST and DTL soils respectively. A time study of NH: exchange revealed that ion exchange was instantaneous.

PAGE 86

79 B. Conclusions Based on the results in this study, the following con-clusions were made for ion exchange during transport through ST and DTL soil under water-saturated conditions: (1) The ZPC of DTL soil was higher than that for ST soil due to a lower content of organic matter and higher contents of Fe and Al oxides in the mixed soil. Thus ion exchange in these soils is dependent both on pH and ionic strength of the so-lution. (2) Ion exchange was shown to be instantaneous and thus the use of a kinetic model for ion exchange in these soils is not justified. (3) Because of the dissolution of lime during displacement experiments, the presence of ca2 + and Mg2+ in the system and changes in pH affected the sorp-+ + tion of K and NH4 (4) For these soils, which have been receiving lime for the last 10 years, it is difficult to establish a homo-ionic soil. During these laboratory ex':' periments, the soils were apparently never completely saturated with either K+ or NH4+. Thus the ion exchange obtained was not only for K+ and NH: but included other ions 2+ 2+ such as Mg and Ca Due to these considerations, adsorp-tion isotherms rather than ion exchange isotherms were mea-sured and used to calculate the retardation factors. (5) Since ion exchange was shown to be instantaneous the BTC's in these soils should not have been affected by the flew velocity imposed. A small apparent velocity effect on thE translation of BTC's for both K+ and NH: was attributed to a change in the chemical environment in the system due to

PAGE 87

80 changes in pH, ionic strength, and ionic compositi.>!'l all of which affect.the retardation of ionic species in these soils. (6) The ST soil showed no selective adsorption for + .+ either K or NH4 This was anticipated from the similar properties of these ions such as ionic radii and charge. (7) Due to the presence of vermiculite in the DTL soil however, K+ was fixed during ion exchange and transport. This study shows that even if an ion is fixed or consumed in the system it is still possible to predict its front during transport if it is not completely removed from solution. (8) For the ST soil the displacement studies showed that ion exchange was reversible for ST soil since all the input amount of either K+ or NH: were recovered. (9) although the application of lime to these soils has raised the pH, the presence of Ca 2+ and Mg2+ in the soils will inhibit the leaching of either NH+ 4 or K+ when applied to these soils as fertilizer.