Citation
Evaluation of lime requirement methods for Florida's sandy soils

Material Information

Title:
Evaluation of lime requirement methods for Florida's sandy soils
Added title page title:
Lime requirement methods for Florida's sandy soils
Creator:
Dierolf, Thomas S., 1957-
Publication Date:
Language:
English
Physical Description:
xii, 113 leaves : ill. ; 28 cm.

Subjects

Subjects / Keywords:
Soils -- Florida ( lcsh )
Liming of soils -- Florida ( lcsh )
Sandy soils -- Florida ( lcsh )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (M.S.)--University of Florida, 1986.
Bibliography:
Includes bibliographical references (leaves 103-111).
General Note:
Typescript.
General Note:
Vita.
Funding:
This collection includes items related to Florida’s environments, ecosystems, and species. It includes the subcollections of Florida Cooperative Fish and Wildlife Research Unit project documents, the Sea Grant technical series, the Florida Geological Survey series, the Coastal Engineering Department series, the Howard T. Odum Center for Wetland technical reports, and other entities devoted to the study and preservation of Florida's natural resources.
Statement of Responsibility:
by Thomas S. Dierolf.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. This item may be protected by copyright but is made available here under a claim of fair use (17 U.S.C. §107) for non-profit research and educational purposes. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact Digital Services (UFDC@uflib.ufl.edu) with any additional information they can provide.
Resource Identifier:
13887331 ( OCLC )

Downloads

This item has the following downloads:


Full Text













EVALUATION OF LIME REQUIREMENT METHODS
FOR FLORIDA'S SANDY SOILS







By

THOMAS S. DIEROLF


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN
PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF MASTER OF SCIENCE



UNIVERSITY OF FLORIDA


1986


















EVALUATION OF LIME REQUIREMENT METHODS
FOR FLORIDA'S SANDY SOILS








By

THOMAS S. DIEROLF


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN
PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF MASTER OF SCIENCE



UNIVERSITY OF FLORIDA


1986




























In dedication to Helga, Curt, Barb,
and Sylvia Dierolf

















ACKNOWLEDGMENTS


As with most accomplishments during one's lifetime, a

work, such as this one, is possible only through the

cooperation of many others besides the one whose sole name

appears on the cover. It would not be feasible to thank

everyone individually as the length of this thesis may then be

doubled.

Many thanks are extended to my major advisor Dr. G.

Kidder, for always finding the time to work out any problems I

encountered while I was at the University of Florida. He was

especially helpful during the dreaded thesis writing stage. I

also wish to thank the rest of the members of my supervisory

committee, Dr. Dean Rhue and Dr. Peter Hildebrand, for

reviewing the thesis and offering their helpful suggestions on

improving the manuscript.

I am grateful to Dr. Dean Rhue, Dr. Luther Hammond, the

Soil Characterization Lab, and the Extension Soil Testing Lab

for allowing me to make use of their already limited labora-

tory space and facilities. Thanks are also extended to all of

the lab technicians who lent me a hand and put up with my

usurpation of portions of the labs. Special thanks are

offered to Bill Reve and Ed Hopwood for the extra considera-

tion they gave me.












For helping me initiate and conduct the field trials in

Suwannee County I am thoroughly indebted to Dr. Mickey Swisher

who really went out of her way to accommodate me. I am thank-

ful to Dr. Tito French and the rest of the people involved

with the North Florida FSR/E program for providing logistical

support. I am also grateful to the Andrews, Chamberlain, and

Barr families for allowing me to conduct the trials on their

farms.

Friends are essential in helping us make our everyday life

bearable. The friends I've made here are no exception, and

they all will be sorely missed. I hope that I will be able to

keep in touch with all of them in the future. I owe Steve

Grant a lot of thanks for allowing me to type this thesis on

his PC, although at times I wondered if it might of have been

easier to hire a typist.

Finally, there is a very special friend, Kate Gieger,

whose love and friendship have helped make it all worthwhile.



















TABLE OF CONTENTS


Page

ACKNOWLEDGMENTS.......................................... iii

LIST OF TABLES........................................... vii

LIST OF FIGURES.......................................... ix

ABSTRACT ................................................. xi

CHAPTERS

I INTRODUCTION........................................ 1

II REVIEW OF THE LITERATURE ............................ 4

Introduction........................................ 4
History of Lime Requirement Methods................. 5
Liming Practices................................ 5
Litmus Test ..................................... 6
Soil-lime Titrations............................ 7
Soil-lime Potentiometric Titrations............ 8
Buffer Methods ................................. 10
Double-buffer Methods........................... 13
Soil Acidity........................................ 14
Forms of Soil Acidity........................... 15
Buffering Capacity.............................. 16
Role of Al in Soil pH Buffering................ 16
Base Saturation................................ 18
pH-BU Relationship............................. 18
Measurement of Base Saturation................. 20
Characteristics of Florida Soils .............. 21
Methods of Testing Lime Requirement Methods.......... 22
Reference Methods............................... 22
Salt Accumulation. ............................ 23
Comparison of Reference Methods................ 25
Effectiveness of Buffers in Measuring Total
Soil Acidity.................................. 26
Field Calibration.............................. 27
Calculation of Adams-Evans Lime Requirement.... 28
Previous Evaluations of Lime Requirement Methods.... 29
Adams-Evans Buffer Method....................... 30
Yuan Double Buffer Method....................... 31












Page

SMP Single Buffer Method........................ 34
SMP Double Buffer Method........................ 34

III INCUBATION STUDY.................................... 36

Introduction........................................ 36
Materials and Methods................................ 37
Results and Discussion............................... 46
pHw and Base Unsaturation..... .................. 46
pHw and pH ..................................... 46
Titration Curves................................ 49
AE Method and CaCO3 Incubation.................. 57
AEmod Method and CaCO3 Incubation............... 59
Yuan DB Method and CaCO3 Incubation............. 61
All Three Methods and BaC12-TEA................. 63
AE Total Acidity and BaCl2-TEA.................. 66
BaCl2-TEA and CaCO3 Incubation................... 66
Suggested Modifications......................... 68
Summary and Conclusions ............................. 71

IV FIELD STUDY ........................ ............... 74

Introduction............................. ........... 74
Materials and Methods................................. 76
Description of Sites and Soils.................. 76
Experimental Design............................. 80
Land Preparation, Use, and Analyses............. 80
Site 1.................................... 80
Site 2 .................................... 81
Site 3.................................... 81
Results and Discussion................................ 82
Observations of Field LR Over the Length of
the Study ....................... .......... 82
Calculation of Field Lime Requirement........... 87
Comparison of Field and Laboratory Data......... 87
Summary ............................................. 91

V CORRELATION STUDY.................................... 92

Introduction........................................ 92
Materials and Methods ............................... 93
Results and Discussion............................... 93

VI SUMMARY AND CONCLUSIONS ............................. 99

REFERENCES....................... ......................... 103

BIOGRAPHICAL SKETCH...................................... .112



















LIST OF TABLES


Page


Table 3-1.



Table 3-2.


Table 3-3.





Table 3-4.





Table 3-5.




Table 3-6.






Table 3-7.




Table 4-1.


Classification of 34 soils from which the
top 15 cm was used in the incubation study.
(Soil Survey Staff, 1985)....................

Chemical and physical characteristics of the
34 soils used in the incubation study........

Mean weights and standard deviations of soil
contained in an 11 ml scoop. An excess amount
of soil was scooped into the container, the
side of the scoop was gently tapped three
times, the soil was leveled off, and weighed.

Individual lime rates applied to the 34
incubated soils. ESTL measured pHw and AE
buffer pH. LR (Ibs acre-1) was determined
from published tables and converted to g
CaCO3 100 g-1 soil to attain pHw 6.5 ..........

Periodic pHw (1:2) measurements taken from 14
of the experimental units over the length of
the incubation study to determine when
equilibrium pHw was reached ..................

Regression equations of final pHs (y) versus
g CaC03 kg-1 soil added (x) used to compute
incubation LR. Also shown are the comparison
of r2 for linear and curvilinear
relationships and the standard deviation
(s) for the linear equations................

Regression statistics of various lime
requirement determinations versus the
BaCl2-TEA (pH 8.2) extractable acidity
reference method (g CaCO3 kg-1 soil)..........

Selected chemical and physical
characteristics of the top 15 cm of the
three soils used in the field study..........














Table 4-2.


Table 4-3.


Table 5-1.


Table 5-2.


Average pHw by treatment for the three field
site s ...... ............ .................. .....

Comparison of various methods of predicting
soil LR to pHw 6.0 and 6.5....................

Selected characteristics of 98 soils used in
the correlation study.........................

Regression statistics between AE LR and the
BaCl2-TEA reference method....................


viii


Page



















LIST OF FIGURES


Page


Figure 3-1.


Figure 3-2.



Figure 3-3.


Figure 3-4.



Figure 3-5.



Figure 3-6.



Figure 3-7.



Figure 3-8.



Figure 3-9.



Figure 3-10.


Location of Alachua (A) and Suwannee (S)
Counties, Florida ...........................

Relationship between pH, and base
unsaturation for 567 Florida topsoils
(0 15 cm ) .................................

Plot of pHs versus pHw for the 34 soils used
in the incubation study.....................

Titration curves of final pH versus amount
of CaCO3 added for soils 7, g, 17, 18,
and 20 ......................................

Titration curves of final pH versus the
amount of CaCO3 added for soils 15, 16, 23,
24 and 27 ...................................

Titration curves of final pHs versus the
amount of CaCO3 added for soils 3, 5, 11,
12, and 14 ..................................

Titration curves of final pH versus the
amount of CaC03 added for soils 10, 13, 19,
20, and 21 ..................................

Titration curves of final pH versus the
amount of CaCO3 added for soils 1, 2, 4, 6,
and 9 .......................................

Titration curves of final pH versus the
amount of CaCO3 added for soils 22, 26, 29,
and 31 ......................................

Titration curves of final pHs versus the
amount of CaCO3 added for soils 25, 28, 32,
33, and 34 ..................................














Figure 3-11.



Figure 3-12.



Figure 3-13.



Figure 3-14.




Figure 3-15.



Figure 3-16.


Figure 4-1.


Figure 4-2.




Figure 4-3.




Figure 4-4.




Figure 5-1.


Regression curves and statistics of AE LR
versus incubation LR. a) to pH 5.5; b) to
pH 6.0; c) to pH 6.5 .......................

Regression curves and statistics of AEmod
versus incubation LR. a) to pH 5.5; b) to
pH 6.0; c) to pH 6.5.......................

Regression curves and statistics of Yuan DB
versus incubation LR. a) to pH 5.5; b) to
pH 6.0; c) to pH 6.5........................

The regression equation and statistics
between the total amount of acidity
measured by the AE buffer (TA) and
BaC12-TEA extractable acidity...............

The regression equation and statistics
between the AE soil-buffer equilibrium pH
and the incubation LR......................

The regression equation and statistics
between AE-F and incubation LR to pH 6.0...

The location of the three field trials in
Suwannee County, Florida ...................

The effect of lime treatments on soil pH,,
over the time period of the study, at site
1. Each point represents the average of
four replications ..........................

The effect of lime treatments on soil pH,,
over the time period of the study, at site
2. Each point represents the average of
four replications...........................

The effect of lime treatments on soil pH,,
over the time period of the study, at site
3. Each point represents the average of
four replications...........................

The location of the 98 soils used in the
correlation study. Each number represents
the amount of samples originating from the
respective county...........................


Pare

















Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science

EVALUATION OF LIME REQUIREMENT METHODS
FOR FLORIDA'S SANDY SOILS

By

THOMAS S. DIEROLF

May 1986

Chairman: Gerald Kidder

Major Department: Soil Science

Laboratory and field studies were conducted to evaluate

two soil-test methods used to determine soil lime requirement

(LR). The Adams-Evans (AE) LR method which is currently used

by the Extension Soil Testing Laboratory (ESTL) to make

liming recommendations in Florida and the Yuan double buffer

(DB) method which was specifically designed to measure the LR

of Florida's sandy soils were compared to a standard CaCO3

incubation procedure.

A laboratory study of 34 Florida soils, with low organic

matter and clay contents, indicated that the Yuan DB method

was a poor predictor of the incubation LR. The AE method was

highly correlated with, but underestimated incubation LR.

An alternative method of calculating the AE LR (AE-F) was

developed from the study data. The AE-F method eliminated the












non-specific pH-base unsaturation (BU) relationship and cor-

rected for the underestimation of incubation LR by the AE me-

thod. The alternative method used the same laboratory deter-

minations as the AE method but used different calculations.

The AE method was also evaluated using a group of 98

soils sampled from various areas throughout Florida. The

soils represented a wider range of physical and chemical

characteristics than did the soils used in the incubation

study. Results indicated that the AE method provided an

accurate index of the soil LR on the 98 soils studied.

Field studies revealed that current lime recommendations

made by the ESTL underestimated actual field LR. The AE-F

method also underestimated field LR which implied that the

liming factor of 1.5 used by the ESTL was too low. The field

study indicated liming factors of 3 and 5 for soils with

relatively low and high initial pHw, respectively. More field

studies are needed to obtain better liming factors for

Florida's soils.

Results indicated that although the AE LR method was

appropriate for use on Florida's sandy soils, two adjustments

in the calculation of AE LR would improve the method. First,

the equation developed from the incubation study should be

employed instead of the one from the original AE LR method.

Second, the current liming factor used by the ESTL should be

increased to at least 3.

Chairman



















CHAPTER I
INTRODUCTION

Prior to the adoption of the Adams-Evans (AE) lime re-

quirement (LR) method by the University of Florida Extension

Soil Testing Laboratory (ESTL), soil lime requirement was

estimated by the county extension agents. They used soil pH,

texture, extractable Ca and Mg, and organic matter content to

estimate the amount of lime needed to accomplish the desired

pH change (Rhue and Kidder, 1984).

The desire for a quantitative method for determining soil

LR had been discussed since at least the early 1970's (Yuan,

1970). Unpublished work at that time had found that the

existing LR buffer methods were not suitable for use on

Florida soils. Yuan expressed a need for modification of the

existing methods, or the development of a new one.

In response to the need for a new LR method in Florida,

Yuan (1974) created a new buffer employing the double buffer

concept. Then Yuan (1975) compared his buffer with three other

methods, including among them the AE method. Using the BaC12-

TEA method as a reference method, he found the Yuan double

buffer (DB) method to be preferable over the AE method which

greatly underestimated the reference method. Yuan followed up












this work with further field studies (Yuan et al., 1977; Yuan

et al., 1978).

When the decision to adopt a new LR method for routine

use by the ESTL was made, the AE method was chosen over the

Yuan DB method. Reasons included were that the AE method had

been developed for low cation exchange capacity (CEC) soils

such as those.which are commonly found in Florida. The

neighboring states of Alabama and Georgia had also adopted the

method thus providing for some interstate collaboration.

Finally the AE method used less buffer and required less

determinations than the Yuan DB method, making the AE method

less expensive and more rapid then the Yuan DB method.

Although Yuan (1975) examined the AE method he did not

give the method the full attention needed if a method is to be

adopted. Since the AE method was adopted by the Extension

State Soil Testing Laboratory a critical examination of the

buffer has not been performed.

Prior to the adoption of a particular soil test method,

in this case the LR method, several methods are usually

evaluated in laboratory and/or greenhouse studies. These are

also known as correlation studies. A representative and large

number of soils for a region can be evaluated. This process

allows for the selection of the method that provides the best

index of the soil LR.

The next step is calibration. Here field trials are

conducted to relate the laboratory values to actual field












values. Field recommendations can then be made from these

results.

The overall objective of this thesis was to evaluate the

AE and Yuan DB methods on Florida's sandy soils. These are the

only two methods being evaluated because of the following

reasons:

1/ There is a desire for collaboration between state soil

testing laboratories, thus the desire to keep the

tests as similar as possible, possibly sacrificing

some accuracy.

2/ There may be a push in the future for DB methodology,

so these methods should continue to be evaluated.

The thesis is divided into five parts. Chapter II reviews

the literature pertinent to the study. Chapter III presents

the laboratory incubation study designed to evaluate the AE

and Yuan DB methods. Chapter IV relates the laboratory data to

actual field responses. Chapter V relates the results from the

incubation study to a wider range of Florida soils to

determine the wider applicability of the AE method. Finally

Chapter VI summarizes the study and provides recommendations

for further work.



















CHAPTER II
REVIEW OF THE LITERATURE

Introduction

The lime requirement of a soil can be defined as the

amount of lime or other base required to neutralize the

undissociated and dissociated acidity in the range from the

initial acid condition to a selected neutral or less acid

condition (McLean, 1980). The key word in this definition is

'selected' such that the reason for selecting the desired pH

is unrelated to the definition. This distinction is necessary

because some definitions may include the reason for attaining

a certain pH within the definition. This is true of the one

given by Hesse (1971), where he relates lime requirement to

the amount of lime needed for maximum economic return from a

particular soil.

The term pH is also left out of the definition because

other workers such as Kamprath (1970) have reported that on

highly weathered leached soils exchangeable Al was a valid

criterion on which to base lime rates.

The remainder of the literature review is divided into

four sections with the purpose of presenting a general

understanding of soil lime requirement (LR) methods used to

predict soil LR. The first section covers a general history












of the development and improvement of soil LR methods.

Section two presents a general description of soil acidity

and related soil characteristics that influence soil LR.

Section three describes techniques of correlating and

calibrating soil LR methods. Finally section four reviews

findings of previous work evaluating the precision and

accuracy of the Adams-Evans (AE), Yuan double buffer (Yuan

DB), and SMP methods.

History of Lime Requirement Methods

The methods of lime requirement determination used today

are the result of an evolution of methodology and theory. A

historical review of the pertinent works can lead to a better

understanding of why the current lime requirement methods are

being employed.

Limina Practices

Recently, several reviews summarizing the early work

pertaining to liming, have been published (Adams, 1984;

Barber, 1984; Lathwell and Reid, 1984; and McLean and Brown,

1984). References to the use of lime date back to the first

and second century B.C.. Ruffin, through his writings in the

early and mid 19th century, promoted the use of lime in the

United States by reporting that marl applications improved

crop yields on his farms.

With the advent of agricultural research stations,

several states in the late 19th century began research on

liming using burned lime, gas lime, or marl. Lime had been













used for centuries in Europe and knowledge of its benefits

were carried over to the U.S from previous experience and

also from writings of Europeans on the principles of plant

growth and nutrition. This knowledge was rarely put to use,

however, since it was easier for the farmers to move to more

fertile grounds than to haul lime the long distances usually

required.

Extensive investigations on crop responses to lime

appeared to originate in the Northeastern region of the U.S..

Near the end of the 19th century Wheeler and co-workers

related the need for lime on acid upland soils in Rhode

Island. Prior to this time the practice of liming had become

well established in certain localities based primarily on

farmer experience. It was not until the early 20th century

that extensive liming trials were begun in the Southern and

Midwestern U.S..

Litmus Test

Although litmus was first used as an indicator of acidity

in 1865, it had no general application for recommending lime

until many years later. Wheeler and Tucker (1896), reported

that soils testing more acid on litmus paper benefited more

from liming than did soils testing neutral or nearly neutral.

They called for a chemical test that, by determining the

relative acidity of a soil, would be able to prescribe the

correct quantity of lime needed.












Soil-lime Titrations

The years 1897 to 1920 were the pioneering era of the

quantitative determination of lime requirement. The earliest

methods of determining soil acidity were based upon the

reaction of soil with excess CaCO3. The oldest of these

methods is that of Tacke (1897) which consisted of suspending

an excess of CaCO3 with the soil and aspirating the evolved

CO2 three hours at room temperature. Shaw (1953) reported

that, in 1900, Wheeler and his co-workers investigated the

possibility of utilizing the evolved CO2, from the reaction

of soil with a suspension of CaCO3 at boiling temperature, as

a measure of lime requirement. They were unable, though, to

determine a reasonable time limit within which the

elimination of CO2 would be ended and observed.

Veitch (1902) developed a lime-water method based on the

Tacke procedure. The Veitch method consisted of a series of

CaO equilibrations with the soil followed by boiling. The

smallest amount of lime-water that gave the characteristic

pink color, in the presence of phenolphthalein, was taken as

the acidity equivalent of the soil. This method was designed

to bring the soil pH up to neutrality because Veitch felt

that an alkaline soil pH was necessary for optimum plant

performance. The Veitch method was considered the most

reliable for the first two decades of the 20th century, but

at the same time was plagued by poor reproducibility and

considered as too laborious (Shaw, 1953).












Hopkins et al. (1903) proposed a method in which a

neutral salt solution of NaCl was mixed with a soil sample.

Their theory held that the mineral acids in the soil would

unite with the mineral bases. A standard fixed alkali was

then used to titrate the liberated mineral acid. In the

laboratory they added quantitative amounts of lime to soil

samples in accordance with the method and found that

practically all of the acidity was neutralized.

Jones (1913) recommended a method in which calcium

acetate was used to extract the acidity. The resultant acetic

acid was then titrated with 0.1M NaOH in the presence of

phenolphthalein. This method was later shown to underestimate

the lime requirement because the extractant was most strongly

buffered at pH 4.76 which was too low to effect complete

replacement of exchangeable hydrogen.

MacIntire et al. (1917) reacted soils with CaCO3 and

MgCO3 in the field and in the laboratory. They concluded that

soils have capacities to decompose CaCO3 in the field greatly

beyond the Veitch lime requirement. MacIntire (1915)

developed a method where they evaporated calcium

bicarbonate-soil suspensions to a thin paste on a steam bath

and determined the soil-CaCO3 reaction from analysis of the

residual CaCO3.

Soil-lime Potentiometric Titrations

Sharp and Hoagland (1915) were the first ones to use

potentiometric titration with Ca(OH)2, using the hydrogen












electrode. They added Ca(OH)2 to soil suspensions until a

definite alkaline reaction was obtained. They also

acknowledged that there was incomplete time for complete

neutralization of the soils acids.

MacIntire (1920) reported at a 1917 meeting on a

collaborative effort that compared all of the aforementioned

methods as well some others. Calcium carbonate-soil incuba-

tions were used to assess the predicted lime requirements by

the various methods. Although the lab assigned to evaluate the

Jones method did not provide any data, MacIntire still

reported that the Jones method offered the best possibilities

for obtaining the coefficient of lime determination.

In 1919, two Danes, Bjerrum and Gjaldbaek, made an

epochal contribution to the field when they developed

titration or buffer curves as determined potentiometrically,

to study the acidic and basic properties of soils. They also

established the relationship between partial pressure of C02

and pH values of saturated solutions of CaCO3 (Shaw, 1953).

Jensen (1924) and Christensen and Jensen (1926), also of

Denmark, used Ca(OH)2 and CaCO3, respectively, to obtain

soil-buffer pH curves. Pierre and Worley (1928), in Alabama,

incubated soil with increments of Ba(OH)2 for three days.

Determinations were made on the clear diffusate by the

colorimetric pH method.

The residual carbonate procedure, developed by Bradfield

and Allison in 1933, is the best known titration procedure












(Thomas-and Hargrove, 1984). They defined a 100% base

saturated soil as one which had reached equilibrium with a

surplus of CaCO3 at the partial pressure of CO2 existing in

the. atmosphere and at a temperature of 25 C. They determined

that .the pH of a base saturated soil would be approximately

8.2.

Buffer Methods

Schofield (1933) proposed a soil-buffer method which

required two titrations. A lime and para-nitrophenol solution

was added to a soil sample and allowed to equilibrate for at

least 16 hours. Both the original solution and soil-solution

were titrated. The difference in cubic centimeters between

the two titrations was equal to the milligram equivalents of

lime taken up per 100 g of soil.

Mehlich (1938) reported that rapid and accurate analyses

were needed in order to measure exchangeable H+, base-

exchange reactions, and lime requirement on large numbers of

soil samples. He proposed the triethanolamine acetate-barium

hydroxide buffer at pH 8.15 for this. The Ba2+ would exchange

the H+ on the soil colloids and the hydroxide would neutra-

lize the resultant acidity. The base exchange could be

measured by titration of the original buffer and soil-buffer

solution and subtracting the two. Titratable acidity was

determined electrometrically by titration with 0.2M HC1 to

approximately pH 6.0. The difference between the titration












values of a like aliquot of the original extracting solution

and the soil extract was equivalent to the neutralization of

the barium hydroxide by soil acids.

Mehlich (1942a) proposed an improved buffer consisting of

barium chloride-triethanolamine at pH 8.2. He substituted

barium chloride for barium hydroxide because the CO2 of the

air did not interfere with the titrations. The base-exchange

capacity and the exchangeable H+ were determined by titrating

one aliquot of extract.

Dunn (1943) incubated soil, Ca(OH)2, and distilled water

for four days with thorough shaking twice daily. He compared

this titration curve to field and laboratory experiments,

where he had added increments of CaCO3 or Ca(OH)2 to soil for

several months. The titration curve underestimated the field

lime requirement but was accurate enough to bring the field

soils to within 0.5 pH of the desired pH.

Brown (1943) first developed the idea of combining the

use of the glass electrode and the concept of depression of

the buffer pH as a measure of exchangeable H+ present in the

soil. Soil was added to either neutral normal acetate or

normal acetic acid. The pH was determined in the mixture. The

depression in pH, in the case of the NH40Ac, or the increase

in pH, in the case of the acetic acid was then read off of a

titration curve as millequivalents of acid or base,

respectively.












Woodruff (1948) developed a buffer method, suitable for

routine testing, that required only one pH reading. The

buffer solution consisted of calcium acetate, para-nitro-

phenol, and magnesium oxide. The titration curve of the

buffer was linear from pH 7.0 to 6.0.

McLean et al. (1958) observed that the lime requirements

of Ohio soils were not met as determined by the Woodruff

method. They felt that the Woodruff buffer method for

determining lime requirement generally did not take into

account the extractable Al.

Shoemaker et al. (1961) developed a buffer method (SMP)

which consisted of a more dilute mixture of triethanolamine,

para-nitrophenol, potassium chromate, and calcium acetate

than the Woodruff procedure. They found that the Woodruff and

Mehlich buffers were too strong to indicate by pH change the

relatively weaker acidity of soils high in extractable Al.

The titration curve of this buffer was linear from pH 7.5 to

4.8.

Adams and Evans (1962) developed a buffer designed pri-

marily for Red-Yellow Podzolic soils with only small amounts

of 2:1 type clays. It had a large buffering capacity relative

to the soil but at the same time it was sensitive to the

acidity of soils of low exchange capacity. They included

para-nitrophenol in a greater concentration than the Woodruff

method, and used K-borate instead of Ca-acetate. They

employed a pH-base unsaturation (BU) relationship in the












calculation of field lime requirement. The titration curve of

the buffer was linear from pH 8.0 to 7.0.

Mehlich (1976) developed a buffer for the rapid estima-

tion of unbuffered salt-exchangeable acidity and lime

requirement. This method was primarily calibrated against

exchangeable acidity. The lime recommendations made are based

on the neutralization of a portion of the exchangeable

acidity rather than the amount needed to attain a target pH.

This buffer exhibited linearity from pH 6.6 to 4.0.

Nommik (1983) developed a buffer to measure titratable

acidity and LR on Swedish soils. The buffer had a linear pH

range from pH 7.00 to 5.20. He found the new buffer method

was satisfactorily correlated with both the Yuan DB and the

SMP methods.

About twenty years after Woodruff modified his original

buffer method, work supporting the change was finally pub-

lished (Brown and Cisco, 1984). The new buffer was intended

to more accurately reflect the contribution of aluminum. They

cautioned that the method should not be adopted on Ultisols

and Oxisols until it had been properly calibrated for those

soils. The buffer method was designed for acid soils with a

lime requirement <10 cmol (+) kg1.

Double buffer Methods

Yuan (1974) first proposed the double buffer concept. The

method was based on two separate measurements in buffer

solutions of the same composition but initially adjusted to












pH 6.0 and 7.0. This allowed the individual buffering

capacity of each soil to be determined. The soil acidity to

be neutralized was determined by the buffering property of a

soil which, in turn, was defined by dividing the difference

in soil acidity neutralized in the two buffer systems by the

difference of the two equilibrium pH values.

McLean et al. (1978) borrowed Yuan's double buffer

concept and applied it to the original SMP buffer method.

They adjusted the two buffer solutions to initial pH values

of 6.0 and 7.5. They found the double buffer procedure (SMP

DB) to be more promising- as an improved method, especially on

soils of low lime requirement where the SMP SB method was

known to lack accuracy.

Soil Acidity

The forms and sources of soil acidity influence the

buffering capacity of a soil. Through an increased under-

standing of soil acidity, scientists have been able to

improve the rapid lime requirement methods.

Soil scientists spent the first half of the Twentieth

Century debating the nature of soil acidity. Jenny (1961)

described this debate as a merry-go-round, that began with

Al-clay theories initiated by Veitch in 1904. In 1922, Van

der Spek promoted the concept of H-clays. This theory

dominated for thirty years. In the early and mid fifties, N.

T. Coleman and co-workers as well as P. L. Low, convinced












Americans once again to accept the Al-clay theory first

advanced by Veitch fifty years before.

Forms of Soil Acidity

Soil acidity can be divided into three forms, active

acidity, exchangeable acidity, and total acidity. Active

acidity is expressed as the pH of a soil. If the soil pH is

<7.0 the soil is considered acidic, conversely if the soil pH

is >7.0 the soil is considered alkaline. Soil pH governs,

among others, nutrient availability to plants (Corey, 1973),

as well as microbial activity in the soil (Stotzky, 1972).

Exchangeable acidity is defined as that which is

replaceable by the cation of a neutral unbuffered salt such

as KC1, CaC12, or NaCl (Lin and Coleman, 1960). Theoretically

this value measures the amount of acidity present in the soil

at the pH of the soil or at least at the pH of the soil-salt

solution. Exchangeable acidity is due almost entirely to

monomeric A13+ (Coleman et al., 1959; Dewan and Rich, 1970)

Yuan (1959) observed that H+ dominated over A13+ in the

exchangeable acidity for some Florida soils, where organic

matter was an important contributor to the cation exchange

capacity (CEC). Thomas and Hargrove (1984) felt this might

have been more apparent than real because some of the H Yuan

observed was probably from the hydrolysis of Al3+ held in

nonexchangeable form by the organic matter.

Total acidity is that which is neutralized at a

designated pH. Opinion has shifted toward the use of












BaC12-TEA at pH 8.2 (Thomas, 1982). Total acidity gives an

indication of the amount of acidity that must be neutralized

to attain a pH somewhere between the original soil pH and pH

8.2. This value would give an indication of the lime

requirement of a soil.

Buffering Capacity

The buffering capacity of a soil is its ability to resist

pH change. The greater the buffering capacity of a soil the

greater its lime requirement.

There are several ways a soil can exhibit buffering

capabilities. Most important in highly weathered soils are

crystalline and noncrystalline oxides and hydrous oxides of

Al and Fe, kaolinite, and allophane. The minerals have

surface hydroxyls that protonate or deprotonate in response

to pH fluctuations (Keng and Uehara, 1974). Acidic groups in

organic matter, mostly carboxyls, also ionize at higher pH

levels releasing H+ ions (Stevenson, 1982). The H+ ions

released from both the mineral and organic matter can

neutralize any added bases.

Role of Al in Soil DH Buffering

Aluminum hydrolysis is associated with both the mineral

and organic fractions of the soil and can be a significant

contributor to the buffering capacity of a soil. Hydrogen

ions are subsequently released according to the following

sequence of reactions occurring in acid soils (Tisdale and

Nelson, 1975):












1/ Al3+ H20 <--> Al(OH)2+ + H

2/ Al(OH)2+ + H20 <--> Al(OH)2+ + H+

3/ Al(OH)2+ H20 <--> A1(OH)3 + H

The Al3+ ion is predominant below pH 4.7, Al(OH)2+

between pH 4.7 and pH 6.5, and Al(OH)3 between pH 6.5 and pH

8.0 (Bohn et al., 1979). The reaction products of Al hydroly-

sis may remain in soil solution, be adsorbed as monomers to

CEC sites of the soil, be adsorbed and then polymerized on

surfaces of clay minerals, or be adsorbed and then completed

by organic matter (McLean, 1976). The H+ ions resulting from

hydrolysis lower the pH of the soil solution and react with

soil minerals to further break them down.

The presence of Al and Fe on organic matter exchange

sites causes the organic matter to exhibit a greater weakness

as an acid. This results in less contribution to the CEC

determined by extraction with a neutral salt solution, or

effective cation exchange capacity (ECEC), of the soil,

especially at low pH values (Hargrove and Thomas, 1984). Al

forms rather stable complexes with the soil organic matter by

reaction primarily with carboxyl groups and to a lesser

extent with phenolic hydroxyl groups (Hargrove and Thomas,

1984). The amount of complex formed is dependent on the pH

and the A13+ concentration in the soil solution (Hargrove and

Thomas, 1984). Schnitzer and Skinner (1963) reported that Al

was predominantly hydroxylated in organic matter as Al(OH)2+












Base Saturation

The base saturation (BS) of a soil can be expressed in

several ways. There are several factors which can influence

the reported BS of a soil. Base saturation and its relation-

ship to soil pH plays an integral part in the determination

of lime requirement by the Adams-Evans lime requirement

method. There are several factors that influence the measure

of base saturation itself as well as the use of the soil

pH-base saturation relationship that detract from its

usefulness in lime requirement methods.

Percent base saturation is a measure of the amount of

exchangeable bases, mostly Ca, Mg, K, and Na, that occupy the

cation exchange sites of a soil. The other sites are occupied

by acidic cations, mostly Al and H. Base unsaturation (BU) is

simply the complement of BS.

pH-BU Relationship

Adams and Evans (1962) used a BU (exchangeable acidity

divided by exchangeable bases plus exchangeable acidity)

versus pH in water relationship from 348 red-yellow Podzolic

soils as a basis of their method. The relationship was used

to describe the general buffering capacities of a group of

soils. Even though they employed this generalized relation-

ship, they also stated that since no constant relationship

existed between soil pH and BU for all soils, pH was

considered to be a measure of BS only for a particular soil.












The major reason that soils exhibit different buffering

capacities is that the type of base exchange material

influences the pH-BS relationship. Mehlich (1942b) using

BaCl2-TEA as the extractant, found kaolinite, which acts as a

weak acid, to be only 65% BS at pH 7.0, whereas montmoril-

lonite was 95% BS at pH 7.0. The BS values for organic matter

were between the values for kaolinite and montmorillonite for

almost the entire pH range.

Base saturation generally increases with an increase in

soil pH for a group of soils. Peech (1939) reported on the

pH-BS relationship for Florida citrus soils. He determined

CEC by pH 7.0 1M NH40Ac. Base saturation was determined by

subtracting exchangeable bases from the CEC and dividing that

by the CEC. He used a nonlinear relationship to show that

soils were well buffered between pH 5.0 to 6.5. Base

saturation was 25% at pH 5.0 and 90% at pH 6.5.

Recent literature has shown that the pH-BS relationship

is a linear one (Loynachan, 1981; Magdoff and Bartlett,

1985). In all of the pH-BS relationships shown, there is a

wide scatter of points about the regression line, regardless

of the shape of the regression line. Difficulties have been

encountered when applying this relationship to very poorly

buffered soils such as those commonly found in Florida (Yuan,

1974).

Use of the pH-BS curve implies that all of the soils used

have a titration curve similar to each other as well as to












the general pH-BS curve. Soils actually exhibit widely

differing titration curves (McLean et al., 1960; Magdoff and

Bartlett, 1985).

Measurement of Base Saturation

Debate also occurs as to what criteria are to be used to

determine base saturation. One way is to use a buffered ex-

tractant such as BaC12-TEA at pH 8.2 or NH4OAc at pH 7.0 to

measure total exchangeable cations (Peech, 1939; Mehlich,

1942b).

Coleman et al. (1959) proposed determining percent base

saturation based on the ECEC, where a neutral salt such as

KC1 is used to extract the exchangeable cations at the pH of

the soil under field conditions.

Sanchez (1976) reported that calculating base saturation

based on BaC12-TEA at pH 8.2 or pH 7.0 1M NH40Ac makes a soil

seem more acid than it is if the field pH is lower than the

pH of the extractant. He cited work by Buol in 1973, where

Buol compared 88 soils from the Midwest, and the Southeast

U.S., and Puerto Rico and found 35% base saturation at pH 8.2

was equal to 55% base saturation at effective CEC. For sandy

soils or soils with >1% organic matter, the relationship

would be different.

Methods measuring CEC, which are determined at pH values

appreciably higher than the soil pH, overestimate the

ability of variable charge soil to retain cations in the

field (Horn et al., 1982; Gillman et al., 1983). Methods more












consistent with field conditions should be used for agronomic

evaluation (Gillman et al., 1983).

Characteristics of Florida Soils

A general knowledge of some of the properties of

Florida's soils can give an indication of their buffering

characteristics. Most of Florida's topsoils are sandy. The

particle size fraction is often composed of greater than 95%

sand. Clay contents of the topsoils are resultantly also low.

Using published data accumulated by the Soil Character-

ization Lab at the University of Florida (Calhoun et al.,

1974; Carlisle et al., 1978, 1981, and 1985) several

generalizations of the properties of Florida's surface soils

according to soil order can be made. Most of the Entisols and

Spodosols have a clay content of less than 4% and a CEC (pH

7.0 NH4OAc) of less than 10 meq 100 g'1 soil. The Ultisols

tend to have a slightly higher clay content but the CEC is

still relatively low. Organic matter content for the Ultisols

and Entisols tends to range from 1 to 2%. The Spodosols tend

to have a higher organic matter with most soils ranging from

2 to 3% organic matter. The clay fraction is usually

dominated by kaolinite, halloysite, gibbsite, quartz, and

vermiculite (Fiskell and Carlisle, 1963).

The organic matter is the most important contributor to

the CEC of acid, sandy Florida virgin topsoils (Yuan et al.,

1967). They found that organic matter contributed from 66 to

96% of the CEC depending on the soil order.













Zelazny et al. (1974) found clay content to give a higher

correlation coefficient with total acidity of fifteen surface

soils of Florida Paleudults than organic matter whereas

organic matter gave a higher correlation coefficient with

exchangeable acidity.

Methods of Testina ,Lime Requirement Methods

Due to the diversity of forms of soil acidity along with

the influence of soil solution pH on their availability,

changes in pH should be gradual so that all available acidity

at a particular pH is neutralized (McLean, 1982a). This is an

important consideration when attempting to use a quick-test

buffer method, where the soil is subjected to sudden pH

changes, as a measure of the effect of lime in the field

where the changes are much more gradual.

In correlation studies, the researcher is attempting to

discern a lime requirement method that provides an accurate

index of the lime needs of soil samples representative of the

area of interest. Calibration studies provide field results

that are used as the basis for recommendations (Hanway,

1973).

Reference Methods

McLean et al. (1966) listed the order.for lime require-

ment test methods,, with respect to increasing amount of time

needed for completion as: pH measurement < titration < incu-

bation < field studies. Early researchers relied primarily on

field studies to base lime recommendations. Besides requiring












a long time for completion, field trials also limit the re-

searcher as to the amount and variety of soils that can be

studied. As they began to develop laboratory tests they still

used field trials as well as.pot and laboratory incubations

as a basis of comparison. With.the advent of more accurate

and rapid laboratory methods, some of the older more tedious

laboratory methods were used as the basis of comparison or

"actual lime requirement."

Examples of commonly used reference methods today include

the CaCO3-soil incubation method (Webber et al., 1977;

Shoemaker et al., 1961) which can last from one month up to

17 months or more. A second reference method used is the

BaC12-TEA method by Peech, which is a modification of

Mehlich's earlier method, and takes one to two days to

complete. The Ca(OH)2 titration method is also commonly

employed as reference method. The most recent modification of

the method by McLean et al. (1978) takes several days to

complete. Although they are not used as frequently, field

studies are still very important for correlating laboratory

test data with actual field response (Yuan et al., 1977,

1978; Baker and Chae, 1977).

Salt Accumulation

Although a long term lime-soil incubation appears to be

the best method short of field trials, there are some

problems associated with it. Microbial activity is intense

under incubation conditions where the soil is at or near












field capacity and at room temperature. Salts (particularly

nitrates of Ca, Mg, and K) accumulate in the soil (McLean,

1982a). These should be leached out or otherwise taken into

account lest they depress the soil pH.

Dumford (1965) reported that on 38 acid soils of the

U.S., with organic matter contents ranging from 0.2 to 6.0%,

less salt was present on the average at the end of the

incubation period than the equivalent of 0.02M CaCI2. Baker

and Chae (1977) found that the pH in CaC12 of their unlimed

soils dropped during incubation. They speculated that this

was due to organic matter decomposition.

Besides leaching or measuring pH in a salt solution some

other methods have been tried to overcome the salt problem.

Some workers have used greenhouse pot trials planted with a

crop (Brown and Cisco, 1984; Loynachan, 1981). The effect of

water leaching through the soil as well as plant uptake of

nutrients would serve to prevent accumulation of salts. Dunn

(1943) used three drops of chloroform, in 100 cc distilled

water to 10 g soil suspension, in a four day incubation study

to prevent microbial activity. Nommik (1983) added 10 ppm

dicyandiamide to a ten week incubation study to suppress

nitrification. The organic matter content of the soils ranged

from <1.0% up to >30%.

An opposite effect of salt accumulation results if the

limestone does not completely react with the soil, where pH

values can be artificially high. Baker and Chae (1977) found












that less lime was required when soils and lime were not well

mixed due to the presence of significant amounts of unreacted

CaCO3 at all but the lowest lime levels.

Comparison of Reference Methods

As. researchers have moved to using relatively short term

analyses as a measure of a soils lime requirement, the

question remains as to what the true lime requirement of a

soil is. Some workers have compared some of the reference

methods or modifications of them. Yuan (1974) compared the

BaC12-TEA method of Peech (1965) with a Ca(OH)2-CaCl2 one

week incubation on twenty Florida soils. The mean IR

determined by the BaCl2-TEA method was 4.54 T/A CaCO3 as

compared to 4.47 T/A CaCO3 for Ca(OH)2-CaC12. McLean et al.

(1978) reported that Ca(OH)2 titration to pH 7.2 with 72

hours of intermittent shaking gave values an average of 5%

lower than those for CaCO3 incubation to pH 6.8 for 17 months

(r = 0.99).

Fox (1980) compared the BaC12-TEA method of Peech (1965),

the Ca(OH)2 titration of McLean et al. (1978), and a six

month CaCO3 incubation. The BaC12-TEA method overestimated

the CaCO3 incubation LR below 9.28 meq CaCO3 100 g-1 soil and

underestimated the LR above 9.28 meq CaCO3 100 g-1 soil. The

Ca(OH)2 titration underestimated the CaCO3 incubation LR for

all determinations. At 2 and 4 meq CaCO3 100 g-1 soil, the

Ca(OH)2 method estimated only 29% and 39% of the CaCO3 LR,

respectively.












Brown and Cisco (1984) compared the Ca(OH)2-CaC12

titration of Benham (1970) to a CaCO3 incubation-cropping

greenhouse method. The LR by the CaCO3 incubation-cropping

method overestimated Ca(OH)2-CaC12 LR values of 2, 4, and 8

cmol (+) kg-1, by 194%, 136%, and 119%, respectively.

Effectiveness of Buffers in Measuring Total Soil Acidity

Total soil acidity is usually defined as the amount of

acidity that must be neutralized to attain a pH at or near

8.2. This is the maximum pH attainable with CaCO3 in the

presence of air having a CO2 content of 0.03% (Bradfield and

Allison, 1933).

The BaCl2-TEA method has been regarded as the rapid

method that most closely approximates the titration method of

Bradfield and Allison (Thomas, 1982). However, Shoemaker et

al. (1961) found that the BaCl2-TEA buffer did not react with

all of the extractable Al. Since Al is an important component

of soil buffering, through Al hydrolysis, this means that the

BaC12-TEA buffer does not react with all of the acidity

present in the soil.

Adams and Evans (1962) reported that their buffer

measured slightly more acidity than was measured by 1M NH40Ac

(pH 7.0) extraction. Acidity, by the latter method, was

calculated by subtracting total bases from the NH40Ac

measured CEC.

The AE buffer measured less acidity than the BaCl2-TEA

(pH 8.2) method for Florida soils (Yuan, 1974). This may have












been due to the ability of the stronger BaCl2-TEA buffer's

ability to react with more soil acidity than the AE buffer.

Shoemaker et al. (1961) reasoned that buffers weaker than

BaCl2-TEA would be expected to extract less soil acidity.

Field Calibration

Recommendations based on laboratory analyses are usually

multiplied by a limingg factor' when making field liming

recommendations. The liming factor accounts for the decreased

effectiveness of agricultural limestone applied in the field

versus the finely ground CaCO3 used in laboratory and pot

studies. The reduced effectiveness is due to the larger

particle size of agricultural limestone and incomplete mixing

of lime in the field. The Adams and Evans method incorporates

a liming factor of 1.5 based on previous data of Pierre and

Worley (1928) and Schollenberger and Salter (1943).

Thomas and Hargrove (1984) stated that since in practice,

lime applications cannot be made that precisely, inaccuracies

in the method are not likely to cause major problems. Adams

(1984) supported the previous point of view, feeling that a

high degree of precision in methods is usually wasted because

of problems encountered in the field application of lime.

These problems include soil variation, lime spreading

irregularities, lime quality, and incomplete mixing of the

lime with soil. McLean (1982b) countered with his argument

that the option of taking some additional simple steps with

the double buffer method after taking the reading for the












single buffer method should be weighed against the increased

accuracy of measurement, especially for soils of low lime

requirement.

Calculation of Adams-Evans Lime Reauirement

A brief explanation of the calculation of the AE LR will

be presented. A more detailed example of the calculation is

given by Rhue and Kidder (1984).

Determination of the AE LR employs two major steps, one a

laboratory measurement and the other a set of calculations.

In the laboratory step a known amount of buffer is added to a

known quantity of soil. The buffer is formulated with a

beginning pH of 8.00 and has the property of a linear

decrease in pH between 8.00 and 7.00. Thus the amount of-

acidity neutralized by the buffer is determined by measuring

the soil-buffer equilibrium pH. The amount of acidity

neutralized by the buffer will be herein referred to as the

AE total acidity (TA).

The calculation step of AE LR involves determining the

percent of AE TA that must be neutralized to attain a desired

soil-water pH (pH,). This step is necessary because the soil

buffer equilibrium pH will be higher than the pHw normally

desired for crops. The buffer theoretically neutralizes more

pH-dependent soil acidity at the higher soil-buffer pH

(between 8.0 and 7.0) than would need to be neutralized at

the lower pHw value desired for plants. This is especially












important for Florida's sandy soils where much of the

buffering capacity is pH-dependent.

The calculation employs the pH-BU relationship found for

348 Alabama Ultisols (Adams and Evans, 1962). The BU value

for a particular pH can be determined from the curvilinear

regression equation that was determined to give the best fit

for the pH-BU relationship.

Thus, knowing the initial pHw of the soil, the desired

pH, of the soil, and the AE TA, the following equation of

Adams and Evans (1962) is used to determine the AE LR:


AE LR = AE TA x (Desired change in BU)
Initial BU
The AE TA is calculated from the soil-buffer equilibrium

pH. The initial BU is calculated from the initial pHw of the

soil and the regression equation for the pH-BU relationship.

The desired change in BU is calculated by subtracting the BU

value corresponding to the desired pH, from the BU value

corresponding to the initial pHw of the soil. The resulting

AE LR is the amount of CaCO3 that will neutralize the portion

of AE TA necessary to attain the desired pHw. The LR may be

presented in terms of agricultural limestone if a limingg

factor' is applied to the CaCO3 LR.

Previous Evaluations of Lime Requirement Methods

The various rapid soil test lime requirement methods have

been evaluated by many workers and on a wide assortment of

soils. This section will detail the strengths and weaknesses












of the various methods. The Adams-Evans method and the SMP SB

method are included because they are currently being used by

certain organizations in Florida. The Yuan DB is included

because it was developed specifically for Florida soils.

Since the SMP DB is reportedly an improvement of the SMP SB

method it was also included in the review. The reference

methods, target pH levels, and the soils used by the workers

differed from experiment to experiment, making it difficult

to compare results between experiments. However, general

observations can be made and these are presented next.

Adams-Evans Buffer Method

Adams and Evans (1962) data showed that the AE method

tended to underestimate the LR of three soils when compared

to CaCO3-soil incubation. The difference between AE estimated

and incubation predicted LR decreased as the LR decreased.

Although McLean et al. (1966) did not publish data for

the AE method, they did evaluate it. They mentioned that the

AE method predicted less lime than the SMP SB method when

CaCO3-soil incubation was used as the reference method. It

can then be inferred from the published data that the AE LR

underestimated the LR as predicted by the CaCO3-soil

incubation.

Fox (1980) reported a high correlation (r = 0.919) for

the AE versus CaCO3-soil incubation. The AE overestimated the

soil LR for soils requiring <4.87 meq CaCO3 100 g'1 soil and












underestimated the soil LR for soils requiring >4.87 meq 100

g-1 soil.

Yuan (1975) correlated the AE to BaC12-TEA method at pH

8.1 (r = 0.78) for 31 Florida soils. The AE only predicted

1/3 of the LR of the reference method. Yuan felt that only

part of this difference was because the AE LR gives the

amount of lime required to raise the soil pH to 6.5, while

the BaCl2-TEA method estimates the total acidity. Most of the

difference was apparently due to discrepancies in the AE LR

method itself.

Contrary to the findings of the previous papers, Tran and

van Lierop (1981) found the average AE LR was 145% of that

determined by the CaCO3-soil incubation. Using a wider

soil:buffer ratio (1:4) expanded the buffer range, improved

the correlation, and was only 114% of the CaCO3-soil

incubation determined LR.

Yuan Double Buffer Method

Yuan (1974) compared the Yuan DB method to both the

BaCl2-TEA method and the Ca(OH)2 titration-incubation using

20 Florida soils. The Yuan DB was highly and equally corre-

lated to both reference methods (r = 0.97). Yuan DB method

estimates of 1.0, 2.0, and 4.0 T/A compared to 1.38, 2.36,

and 4.32 T/A as determined by the BaCl2-TEA method. The

Ca(OH)2 method gave results comparable to the BaCl2-TEA

method.












Yuan (1975) found a very high correlation for 44 Florida

soils between the Yuan DB method and the BaC12-TEA method

(r = 0.99). Yuan DB method LR estimates of 0.37, 0.85, and

1.80 T/A compared to 0.5, 1.0, and 2.0 T/A for the BaCl2-TEA

method, respectively.

'Fox (1980) evaluated the Yuan DB method on 20 Pennsyl-

vania soils. A better correlation was determined when CaCO3-

soil incubation (r = 0.967) was used as the reference method

than when Ca(OH)2 titration (r = 0.914) was used as the

reference method. Using the CaCO3-soil incubation as a

reference method, the Yuan DB method overestimated the LR for

soils with a LR <7.44 meq CaCO3 100 g-1 soil and underes-

timated the LR for soils with a LR >7.44 meq CaCO3 100 g-1

soil.

Tran and van Lierop (1981) worked with 70 Quebec soils

and an eight week CaCO3-soil incubation. The Yuan DB method

had a high correlation coefficient (r = 0.959) for soils

ranging in LR from 1.4 to 40.0 meq CaCO3 100 g-1 soil to pH

6.5. A poorer correlation was found for soils with a LR <10

meq 100 g-1 soil (r = 0.877). The Yuan DB method, on the

average, predicted 66% of the CaCO3 LR(6.5).

McLean et al. (1978) reported the Yuan DB method

underestimated the-LR for 28 low LR soils (<4.0 meq 100 g-1

soil). The reference method predicted LR was 1.0, 2.0, and

4.0 meq 100 g-1 soil versus 0.54, 1.65, and 3.87 meq 100 g-1

soil as determined by the Yuan DB method. The Yuan DB method












was more accurate for low lime requiring soils than for high

lime requiring soils. It predicted 89% and 73% of the LR,

respectively for 54 soils from the U.S.

Tran and van Lierop (1982) used 37 acid, coarse-textured

soils. They evaluated the Yuan buffer both as a double buffer

and as a single buffer (Yuan SB). Linearity of the Yuan SB

method increased as its. initial pH approached the desired

soil pH. The precision was lower when the initial pH was 6.5

versus 6.0 or 7.0. They felt this could be due to a break in

the titration curve at pH 6.2 (Yuan, 1976). For a target pH

of 5.5 (LR5.5) or a target pH of 6.0 (LR6.0) the precision of

the relationship was good for either the initial buffer pH of

6.0 or 7.0. The DB for LR(6.0) had a high correlation

coefficient (r = 0.97) versus incubation with CaCO3 but only

measured an average of 67% of the incubation LR. They

proposed using a regression equation to adjust for the

difference. The DB was less precise for LR(5.5) (r = 0.76).

Yet the Yuan SB was as precise for LR(6.0) and more precise

for LR(5.5) versus the Yuan DB.

Van Lierop (1983) found that the Yuan DB performed poorly

on organic soils. He felt this might be because many of the

soil-buffer pH values ranged from pH 5.0 to 6.0 which fell on

the curvilinear portion of the buffer curve. Altering the

soil-buffer ratio to 1:5 vol/vol to raise the soil-buffer pH

to the linear portion of the curve did not the improve the












precision of the method as it resulted in a lowering of the

correlation coefficient.

SMP Single Buffer Method

The SMP SB LR method has been evaluated by numerous

researchers. Most of the researchers report that the SMP SB LR

method underestimates the LR for low lime requiring soils

(Yuan, 1975; Fox, 1980; Brown and Cisco, 1984).

Most of the workers found the SMP SB LR method to be

significantly correlated (r > 0.90) with their respective

reference methods (Webber et al., 1977; Loynachan, 1981). The

buffer exhibits relatively little buffering capacity from an

initial buffer pH of 7.5 down to a pH of 6.9 (McLean, 1978).

The vertical change in buffer pH in the pH range of'7.5 to 6.9

would be too great to indicate adequate lime for acid soils

very low in cation exchange capacity (McLean et al., 1966).

McLean et al. (1966) reported that the SMP SB was reasonably

accurate for soils requiring >4000 lb of lime per acre.

SMP Double Buffer Method

McLean et al. (1978) found that through the use of the

SMP DB method and regression equations the error of estimate

for low LR soils could be significantly improved over the SMP

SB method. However, they still recommended that the SMP SB was

the most satisfactory compromise between simplicity of

determination and reasonable accuracy for soils of a wide

range of lime requirement.












Tran and van Lierop (1982) reported that the SMP DB was

not substantially more accurate than the SMP SB for LR 5.5 and

6.0. In a previous paper, Tran and van Lierop (1981) found

that the SMP DB did not significantly improve the correlation

as compared to the SMP SB.

Fox (1980) found the SMP DB overestimated the LR for

soils with a LR of <4.12 meq CaCO3 100 g-1 soil, and under-

estimated the LR for soils with >4.12 meq CaCO3 100 g-1 soil.

The SMP SB method performed much better on these soils than

the SMP DB method.

Ssali and Nuwamanya (1981), reporting on two separate

experiments, found the SMP DB method to better predict the

mean LR to pH 6.5. For soils with a LR <4 meq 100 g'1 both SMP

methods underestimated the predicted LR. The SMP DB provided a

better estimate than the SMP SB. The second trial used a

CaCO3-soil incubation'to determine the LR to pH 6.0. This time

the two SMP methods both overestimated the predicted LR for

soils requiring <4 meq 100 g-1. The SMP DB provided a better

estimate than did the SMP SB method.


















CHAPTER III
INCUBATION STUDY
Introduction
Rapid soil test methods for the determination of lime

requirement provide indexes of the actual soil lime

requirements that are found from field studies. Several

factors influence the incomplete measurement of soil lime

requirement not the least of which is the short time of

soil-buffer contact. The first step in evaluating the

suitability of a rapid soil test method to a particular group

of soils is through lab correlation studies.

In lab correlation studies the various rapid soil-test

methods being evaluated are compared to one or more reference

methods. The reference methods, which usually take a

relatively long time to complete, are assumed to give a more

accurate indication of the total amount of lime needed to

attain a particular pH level. The ability of a soil-test

method to be an accurate index of the reference method is

then determined by regressing the results of the two methods.

In this study, a commonly used reference method,

soil-CaCO3 incubation, was used to evaluate both the AE and

Yuan DB LR methods. The following objectives were sought to

be determined from this study:












1/ To determine the precision and accuracy of the AE and

Yuan DB LR methods using a soil-CaCO3 incubation as a

reference method.

2/ To determine if a modification of the AE method could

improve, if needed, the AE method for Florida sandy

soils.

3/ To determine if a relatively faster reference method,

BaCl2-TEA extractable acidity, could serve as a

reference method for evaluating the AE method on a

larger amount and wider range of soils.

Materials and Methods

Published data (Calhoun et al., 1974; Carlisle et al.,

1978, 1981, and 1985) were used to develop a relationship

between pHw and BU for Florida soils. Base unsaturation was

calculated by dividing the BaCl2-TEA (pH 8.2) extractable

acidity by the sum of the extractable acidity and 1M NH40Ac

exchangeable base values, and multiplying by 100. With the

exception of Histosols, all soils described in the cited

reports were included in the analysis, resulting in 567 pairs

of data. The regression equation for this relationship was

substituted for the curvilinear regression equation that

Adams and Evans (1962) found for Alabama Ultisols.

From the same source of data, a relationship between pH,

and pHs for Florida soils was also developed. Target pH

values for Florida soils are only given in terms of pHw. The

regression equation determined from the pHw-pHs relationship












would allow the expression of desired pH levels in terms

of pHs values that correspond to commonly used target pH,

levels of 5.5, 6.0, and 6.5. This calculation was done to

facilitate interpretation of the incubation data.

For the incubation study, thirty-four surface horizon (0

to 15 cm) soils (13 Entisols, 6 Spodosols, and 15 Ultisols)

were sampled from Alachua and Suwannee Counties (Fig. 3-1).

The soils were characterized by the use of the Alachua and

Suwannee County soil surveys (Soil Survey Staff, 1985, and

1965) (Table 3-1). Sampling areas were chosen to represent

cultivated, previously cultivated, and native vegetation

sites. All soils were relatively low in organic matter and

clay contents and thus were representative of many

agricultural soils of Florida.

The soils were air dried and passed through a 2 mm sieve.

Soil pHw was determined in triplicate using both 10 ml soil

to 10 ml water (1:1 ratio) and 10 ml soil to 20 ml water (1:2

ratio). Soil pHs was also determined on the same sample in

0.01M CaC12 using 10 ml soil to 20 ml 0.01M CaC12 (1:2 ratio)

(McLean, 1982a). Particle size analysis by the pipette

method, organic carbon by acid dichromate digestion, 1M KC1

exchangeable Al, and BaCl2-TEA (pH 8.2) extractable acidity

were also determined (Soil Survey Staff, 1972). Percent

organic matter was calculated by multiplying percent organic

carbon by 1.724. Mehlich I extractable P, K, Ca, Mg, and Al





































FLORIDA


Figure 3-1. Location of Alachua (A) and Suwannee (S)
Counties, Florida.












Table 3-1. Classification of 34 soils from which the top 15 cm was used in the incubation
stuidv. (Soil Sirvey Staff. 1985 .


Soil
1
2
3
4
5
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
14


County
Alachua
Alachua
Alachua
Alachua
Alachua
Alachua
Alachua
Alachua
Alachua
Alachua
Alachua
Suwannee
Suwannee
Suwannee
Suwannee
Suwannee
Suwannee
Alachua
Alachua
Alachua
Alachua
Alachua
Alachua
Alachua
Alachua
Alachua
Alachua
Alachua
Alachua
Alachua
Alachua
Al a hnII


Series
Tavares
Chipley
Arredondo
Chipley
Wauchula
Kendrick
Kendrick
Chipley
Millhopper
Plummer
Gainesville
Lakeland
Lakeland
Lakeland
Blanton
Blanton
Blanton
Newnan
Sparr
Blichton
Sparr
Kanapaha
Newnan
Tavares
Pomona
Wauchula
Pomona
Kendrick
Arredondo
Gainesville
Gainesville
Ca inpivi 1 IP


Family or higher taxonomic class
Hyperthermic, uncoated Typic Quartzipsamment
Thermic, coated Aquic Quartzipsamment
Loamy, siliceous, Grossarenic Paleudult
Thermic, coated Aquic Quartzipsamment
Sandy, siliceous, hyperthermic Ultic Haplaquod
Loamy, siliceous, hyperthermic Arenic Paleudult
Loamy, siliceous, hyperthermic Arenic Paleudult
Thermic, coated Aquic Quartzipsamment
Loamy, siliceous, hyperthermic Grossarenic Paleudult
Loamy, siliceous, thermic Grossarenic Paleaqult
Hyperthermic, coated, Typic Quartzipsamment
Thermic, coated, Typic Quartzipsamment
Thermic, coated, Typic Quartzipsamment
Thermic, coated, Typic Quartzipsamment
Loamy, siliceous, thermic Grossarenic Paleudult
Loamy, siliceous, thermic Grossarenic Paleudult
Loamy, siliceous, thermic Grossarenic Paleudult
Sandy, siliceous, hyperthermic Ultic Haplohumod
Loamy, siliceous, hyperthermic Grossarenic Paleudult
Loamy, siliceous, hyperthermic Arenic Plinthic Paleaqult
Loamy, siliceous, hyperthermic Grossarenic Paleudult
Loamy, siliceous, hyperthermic Grossarenic Paleaqult
Sandy, siliceous, hyperthermic Ultic Haplohumod
Hyperthermic, uncoated Typic Quartzipsamment
Sandy, siliceous, hyperthermic Ultic Haplaquod
Sandy, siliceous, hyperthermic Ultic Haplaquod
Sandy, siliceous, hyperthermic Ultic Haplaquod
Loamy, siliceous, hyperthermic Arenic Paleudult
Loamy, siliceous, hyperthermic Grossarenic Paleudult
Hyperthermic, coated Typic Quartzipsamment
Hyperthermic, coated Typic Quartzipsamment
Hvnerthermic. rnat-rP T\nic- (minart-7insamment


_1 flr\_____\F __ii_l_____ 1 It_~LIC tIV YI___Y flr_~V V\____C___I__ _












were determined by the University of Florida Extension Soil

Testing Lab (ESTL) (Rhue and Kidder, 1983) (Table 3-2).

The Adams and Evans buffer pH was determined in duplicate

by the ESTL (Rhue and Kidder, 1983). As mentioned previously,

the AE method theoretically measures more of the soil acidity

than needs to be neutralized to attain a pHw of 6.5. The

total amount of acidity measured by the AE buffer solution

will be termed AE total acidity. For comparison purposes this

value can be expressed either as the pH of the buffer-soil

solution or as g CaCO3 kg-1 soil needed to neutralize all of

the measured acidity.

Calculations of AE LR were made on a weight basis to

allow direct comparison to incubation LR data, which were

also computed on a weight basis. Each of the 34 soils were

scooped with an 11 cc container, tapped 3 times, leveled with

a straight edge, and weighed. This was repeated four times

for each soil (Table 3-3). The mean value was found to be

1.32 g cm"3 and was used as an average for all conversions of

volume to weight.

The Yuan double buffer (Yuan DB) method was also employed

to determine soil LR on all 34 soils (Yuan, 1974, 1976).

Separate 50 ml portions, in duplicate, of the Yuan DB with pH

values adjusted to 6.0 and 7.0 were added to separate 5 g

soil samples. The equation given by Yuan allowed for

calculation of the LR to various pH levels.

















Table 3-2. Chemical and physical characteristics of the 34 soils used
in the incubation study.

1M KC1 Mehlich I Extractable
Soil pHw pHw pHs exch.
No. (1:1) (1:2) (1:2) O.C.O.M. Clay Silt Sand Al P K Ca Mg Al

--------g 100 g-1 soil--- ---------------mg kg-l-----------------

1 5.66 5.61 4.88 1.1 1.9 3 3 94 0.11 120 16 397 197 393
2 4.82 4.86 4.04 0.8 1.4 5 2 93 0.51 7 6 36 15 128
3 5.56 5.61 4.78 0.8 1.7 5 4 92 0.28 22 13 125 72 389
4 5.89 5.89 5.06 1.0 1.8 6 1 93 0.02 90 20 440 235 267
5 4.88 4.82 4.05 1.2 2.0 3 1 96 0.54 13 10 71 35 162
6 4.01 4.17 3.15 2.2 3.8 1 2 97 0.61 3 12 92 61 84
7 5.24 5.27 4.70 1.9 3.2 5 3 92 0.24 55 46 396 236 710
8 6.00 5.92 5.11 1.2 2.1 3 4 92 0.00 27 12 570 331 176
9 6.32 6.32 5.69 1.3 2.2 5 4 91 0.00 97 48 1000 555 246
10 6.06 6.05 5.12 1.5 2.6 4 2 94 0.08 49 42 265 156 398
11 4.40 4.46 3.58 1.5 2.6 2 2 96 0.27 7 12 114 66 46
12 4.32 4.35 3.51 1.8 3.1 5 2 93 0.26 6 14 201 116 63
13 5.99 6.06 5.09 0.9 1.5 4 3 93 0.06 125 145 312 182 328
14 5.96 5.89 5.19 1.2 2.0 2 3 95 0.00 126 44 580 308 226
15 5.69 5.70 4.96 1.2 2.1 2 4 94 0.06 56 40 292 174 234
16 5.94 5.85 5.11 1.1 1.9 3 3 94 0.02 56 36 293 171 238
17 5.23 5.19 4.35 0.5 0.9 4 6 90 0.58 3 12 106 74 126
18 4.98 4.99 4.22 0.7 1.2 4 10 86 1.62 3 21 190 119 158
19 5.01 4.96 4.12 1.1 2.0 4 2 93 0.26 8 12 178 100 87
20 5.25 5.26 4.40 1.0 1.7 4 1 95 0.00 5 15 226 125 19
21 4.95 4.97 4.12 0.4 0.8 3 1 96 0.06 2 7 69 37 19
22 5.32 5.23 4.51 1.3 2.2 4 9 86 0.32 81 36 400 237 209
23 5.99 5.90 5.04 0.5 1.0 4 2 95 0.07 18 6 142 63 116
24 4.77 4.74 3.95 0.7 1.2 6 1 93 0.14 8 8 115 65 41
25 4.57 4.57 3.67 1.1 1.9 3 1 96 0.09 3 16 178 106 26
26 4.51 4.52 3.82 0.7 1.2 3 1 96 0.25 3 8 75 41 45
27 5.04 4.98 4.23 1.3 2.2 5 3 92 0.57 92 11 258 136 253
28 4.33 4.31 3.45 3.7 6.3 8 2 91 1.14 4 30 295 165 160
29 4.39 4.50 3.46 1.0 1.7 5 1 94 0.12 2 9 78 43 24
30 6.21 6.14 5.34 0.6 1.0 4 3 93 0.01 52 25 323 169 178
31 4.44 4.41 3.98 1.2 2.1 2 4 94 1.14 54 5 35 18 436
32 6.01 5.98 5.26 1.2 2.0 6 4 90 0.00 201 151 820 458 330
33 6.12 6.05 5.14 1.1 1.9 4 6 90 0.00 209 122 570 309 558
34 5.87 5.80 4.90 1.1 2.0 2 6 91 0.06 162 91 397 232 482









Table 3-3. Mean weights and standard deviations of soil contained in an 11 ml
scoop. An excess amount of soil was scooped into the container, the
side of the scoop was gently tapped three times, the soil was
leveled off, and weighed.

Mean of standard Mean of standard
Soil 4 weighing deviation Soil 4 weighing deviation

------ g 11 cm-3 -----------g 11 cm-3-----
1 14.77 0.43 18 12.88 0.16
2 15.14 0.17 19 13.94 0.13
3 14.96 0.26 20 14.67 0.12
4 15.97 0.14 21 15.23 0.04
5 14.83 0.12 22 13.20 0.13
6 14.24 0.19 23 15.80 0.09
7 14.43 0.17 24 14.60 0.27
8 13.97 0.15 25 14.65 0.17
9 14.55 0.16 26 14.28 0.23
10 15.38 0.14 27 13.96 0.20
11 13.23 0.63 28 12.70 0.19
12 13.32 0.10 29 15.33 0.12
13 14.86 0.23 30 15.68 0.17
14 15.36 0.13 31 14.23 0.24
15 14.89 0.09 32 13.64 0.12
16 15.54 0.12 33 13.81 0.09
17 13.92 0.22 34 14.16 0.12


Overall mean of 34 soils = 14.47 g

standard deviation = 0.83 g


range of means = 12.70 15.97 g












Extractable acidity by the BaCl2-TEA (pH 8.2) method

(Soil Survey Staff, 1972) was determined in duplicate on all

of the soils. The acidity measured by this method was

expressed as the amount of base required to neutralize the

acidity in g CaCO3 kg-1 soil. The BaCl2-TEA method was also

used as a reference method.

Each of the 34 soils was incubated in duplicate with six

increments of carefully weighed 320 mesh reagent grade CaCO3.

The CaCO3 additions were 0, 1/3, 2/3, 1, 4/3, and 5/3 times

the AE LR to attain pHw 6.5 (Table 3-4). The AE LR was

determined separately for each soil. One hundred grams of

each soil were placed in an unsealed polyethylene bag. This

bag was placed inside another polyethylene bag to protect

against accidental soil loss. The CaCO3 was added to the dry

soil and the bag was vigorously hand-mixed for three minutes.

Eight ml of deionized water were added to each bag to

approximate field capacity. The soil was again mixed until

all of the soil appeared moistened. The bags were stored on a

lab bench at room temperature for the duration of the

incubation period. After six weeks the bags were vigorously

shaken to maximize soil-CaCO3 contact. To prevent large

moisture losses while also providing for some gas exchange,

outer bags were sealed but had a 5 mm hole punched in the

side. Some bags were periodically weighed over the 12-week

incubation period and minimal moisture loss of about 1 ml per

sample was confirmed.












Table 3-4. Individual lime rates applied to the 34 incubated
soils. ESTL measured pHw and AE buffer pH. LR (lbs
acre-1) was determined from published tables and
converted to g CaCO3 100 g-1 soil to attain pHw6.5.

pHw AE Buffer
Soil (1:2) pH LR 1/3LR 2/3LR LR 4/3LR 5/3LR

100 lbs A-1 ----mg CaCO3 100 g-1 soil----


73 110
58 87
60 90
33 50
58 87
133 200
93 140
31 47
20 30
33 50
67 100
13 170
42 63
35 53
51 77
33 50
51 77
82 123
55 83
31 47
25 37
89 133
29 43
51 77
65 97
38 57
82 123
167 250
55 83
18 27
127 190
47 70
47 70
53 80


147 183
116 145
120 150
67 83
116 145
267 333
187 233
63 78
40 50
67 83
133 167
227 283
84 105
71 88
103 128
67 83
103 128
164 205
111 138
63 78
49 62
177 222
57 72
103 128
129 162
76 95
164 205
333 417
111 138
36 45
253 317
93 117
93 117
107 133


5.3
5.0
5.5
5.7
5.0
4.3
5.2
5.8
6.1
6.2
4.6
4.4
5.9
5.8
5.7
5.9
5.3
5.0
5.1
5.2
4.9
5.3
5.7
4.9
4.7
4.5
5.0
4.5
4.6
5.9
4.5
5.7
6.0
5.9


7.50
7.65
7.55
7.70
7 .65
7.35
7.40
7 .70
7.70
7.50
7 .65
7.45
7.55
7.65
7 .55
7.65
7.65
7.50
7.65
7.80
7.85
7 .40
7.75
7.70
7.65
7.80
7.50
7.15
7.70
7.80
7.35
7.60
7.45
7.45


3.3
2.6
2.7
1.5
2.6
6.0
4.2
1.4
0.9
1.5
3.0
5.1
1.9
1.6
2.3
1.5
2.3
3.7
2.5
1.4
1.1
4.0
1.3
2.3
2.9
1.7
3.7
7.5
2.5
0.8
5.7
2.1
2.1
2.4













To determine when the soils had reached equilibrium pHw

measurements were taken on 14 samples after 6, 10, and 12

weeks of incubation (Table 3-5). At the end of 12 weeks,

soils were transferred into plastic cups and allowed to air

dry. Duplicate soil pH, and pHs measurements were then taken.

Lime requirements for different target pH levels were

determined using simple regression and correlation tech-

niques. Final pH, or pHs were regressed versus additions of

CaCO3 for each soil. The amount of CaCO3 needed to attain a

desired pH for each soil was calculated from the appropriate

regression equation.

Results and Discussion

PHF and Base Unsaturation

The best relationship between pHw and base unsaturation

(BU) for the 567 soils in the cited soil characterization

reports was found to be: pHw = 7.34 0.03 (%BU) (r = -0.752,

n = 567) (Fig. 3-2). This regression equation will be

referred to as AEmod in the following discussion. The equa-

tion of the original AE method for 347 Ultisols was: pH, =

7.79 5.55 (BU) + 2.27 (BU2) (Adams and Evans ,1962). The

authors did not include the r2 or C.V. values in their paper.

pH- and pHs

Commonly used target pHw values are 5.5, 6.0, and 6.5. To

allow for interpretation of the incubation pHs measurements,

target pHs values were calculated that correspond to the

aforementioned pH, values.











Table 3-5. Periodic pHw (1:2) measurements taken from 14 of
the experimental units over the length of the
incubation study to determine when equilibrium pHw
was reached.

pHw (1:2)
Lab Sample After After After
Number 6 weeks 10 weeks 12 weeks


35 6.80 6.58 6.57

51 5.32 5.27 5.21

81 5.85 5.75 5.73

96 6.72 6.42 6.38

104 6.17 6.10 6.12

113 6.47 6.30 6.25

226 5.82 5.75 5.74

256 6.30 6.12 6.10

327 6.02 5.70 5.64

360 7.10 6.83 6.77

365 7.02 6.72 6.74

380 5.90- 5.80 5.86

384 6.97 6.83 6.81

396 6.80 6.60 6.57


Mean pHw 6.38 6.20 6.18

a There were a total of 408 experimental units. 34 soil x 6
treatments x 2 replications.



































* N..............




N.
N.


5.0 *


pHw = 7.34 -- 0.03(%BU).

(r = -0.752, n = 567)


20 40 60 80 100
BASE UNSATURATION (%)
-_I I' --- -- 0
80 60 40 20 0
BASE SATURATION (%)



Figure 3-2. Relationship between pHw and base unsaturation for 567
Florida topsoils (0 15 cm).


N.. .-r. .


N


''



'












The following relationship was calculated for data

reported from 567 Florida soils: pHs = -0.488 + 0.954(pH,)

(r = 0.90, n = 567). A very similar relationship was found

between pHs and pHw for the 34 soils used in the study: pHs =

-0.676 + 0.974(pH,) (Fig. 3-3). These two equations predict

an average difference of 0.8 pH between pHs and pHw, in the

pH range of 4.0 to 6.5. An assumption is then made that most

of these sandy soils under normal field conditions will have

a pHs that is about 0.8 pH less than pHw. By subtracting 0.8

pH from target pHw values of 5.5, 6.0, and 6.5, the

corresponding target pHs values of 4.7, 5.2, and 5.7 were

obtained.

Titration Curves

Twenty-eight of the soils experienced a drop in pHw of

the blank samples (no lime added) during incubation. Fifteen

soils showed a decrease of at least 0.40 pH. Only one soil

actually attained pHw 7.0. Evaluation of pHs values for the

blank samples showed much less deviation from the original

pHs. The mean pHw for blank samples for the 34 soils was 5.29

before incubation and 5.03 after incubation. The mean pHs for

the blank samples was 4.47 prior to incubation and 4.49 after

incubation.

Dumford (1965) found that the average pH of 38 acid soils

rose from 5.84 to 6.33 when they were leached with pure

water, apparently due to the leaching out of salts, and then

fell to 5.76 with the additions of 0.01M CaC12. The lower






















y = -0.676 + 0.974x

(r = 0.99, n = 34)


0 *
4.0-










3.0 I
4.0 5.0

Figure 3-3. Plot of pHs versus
incubation study.


pH,
pHw for the 34 soils used in the












average 0.01M CaC12 pH than for the unleached soils in pHw

seems to indicate that the amount of salts accumulated was

less than the concentration of 0.01M CaCl2. Thus by using pHs

the salt effect can be accounted for. For this reason only

results for incubation LR determined from pHs versus CaCO3

added curves will be reported as incubation LR.

To facilitate the practical application of the following

results, target pH values will be reported in terms of pH,,

but will have been calculated from pHs values. Figures 3-4

thru 3-10 show the titration curves for all 34 soils. The

curves for pHw and pHs were very similar except that the pHs

values were lower. For the reason explained below, only pHs

titration curves are shown.

The titration curves were analyzed for both linearity and

curvilinearity. All of the relationships were best described

as linear. Table 3-6 lists the equations that were used to

determine the LR values to the respective target pH values.

Also shown are the comparison of r2 for linear and

curvilinear equations. Due to the small differences in r2

between the linear and curvilinear equations as well as the

small number of coordinate pairs (6) the linear equations

were used in all of the cases. Linear increases in pH to

CaCO3 additions have been reported by several other

researchers. McLean et al. (1960) observed linear increases

in pHw to CaCO3 additions to incubated soils. Van Lierop

(1983) reported a linear increase in pH, to CaCO3 additions

















K/ /
6.0- 8i





5.0





4.0 1
0.0 1.0 2.0
-1
g CaCO3 kg soil
Figure 3-4. Titration curves of final pHs versus
amount of CaCO3 added for soils 7, 8,
17, 18, and 20.


S23 t24

/7 /16 -15
6.0 27





5.0





4.0.
0.0 1.0 2.0
g CaCO3 kg soil
Figure 3-5. Titration curves of final pHs versus
the amount of CaCO added for soils
15, 16, 23, 24, and 27.

























S /





4.0



0.0


Figure 3-6.


-1
g CaCO3 kg soil

Titration curves of final pHs versus the
amount of CaCO added for soils 3, 5, 11,
12, and 14.


-1 .
g CaCO3 kg soil
Figure 3-7. Titration curves of final
amount of CaCO added for
19, 20, and 21.


pHs versus the
soils 10, 13,




























4.0 -






0.0


Figure 3-8.


Titration
amount of
6, and 9.


-. -1 J.V
g CaCO3 kg soil

curves of final pHs versus the
CaCO3 added for soils 1, 2, 4,


g CaCO kg soil
Figure 3-9. Titration curves of final pHs versus the
amount of CaCO3 added for soils 22, 26,
29, and 31.


















































Figure 3-10. Titration curves of final pHs versus the amount of CaCO.
added for soils 25, 28, 32, 33, and 34.












Table 3-6.Regression equations of final pHs (y) versus g
CaCO3 kg-1 soil added (x) used to compute
incubation LR. Also shown are the comparison of r2
for linear and curvilinear relationships and the
standard deviation (s) for the linear equations.

Soil r2
No. Regression Equation Linear Curvilinear s


pHs=4.87+0.81x
pHs=3.98+1.51x
pHs=4.69+1.06x
pHs=5.00+1.22x
pHs=3.98+1.30x
pHs=3.13+0.93x
pHs=4.63+0.66x
pHs=4.96+1.25x
pHs=5.53+0.95x
pHs=5.03+0.83x
pHs=3.65+1.51x
pHs=3.59+0.98x
pHs=5.07+0.92x
pHs=5.16+1.16x
pHs=4.87+0.97x
pHs=5.32+1.06x
pHs=4.25+1.70x
pHs=4.07+1.21x
pHs=4.21+1.16x
pHs=4.21+1.84x
pHs=4.30+2.74x
pHs=4.44+0.88x
pHs=4.92+2.27x
pHs=4.22+1.79x
pHs=3.92+1.47x
pHs=4.29+1.76x
pHs=4.22+0.83x
pHs=3.55+0.56x
pHs=3.47+2.01x
pHs=5.32+2.01x
pHs=4.27+0.74x
pHs=5.19+0.98x
pHs=5.13+0.71x
pHs=4.90+0.85x


99.3 99.5


99.6
99.1
99.0
99.9
99.4
99.5
99.6
99.1
99.1
94.8
98.7
98.4
99.0
99.7
99.4
98.2
98.6
97.9
96.3
90.5
99.5
99.4
98.4
98.3
99.1
99.1
96.3
99.6
99.2
99.4
99.5
98.1
99.5


99.5
99.4
99.6
99.0
99.8
99.8
99.6
98.9
99.2
95.3
98.7
99.3
99.0
99.7
99.7
98.0
99.8
97.7
99.1
97.6
99.8
99.8
99.2
98.2
99.2
99.4
96.0
99.6
99.1
99.9
99.7
98.2
99.5


0.046
0.052
0.054
0.036
0.023
0.087
0.038
0.020
0.017
0.024
0.210
0.116
0.044
0.036
0.024
0.024
0.105
0.106
0.088
0.101
0.197
0.051
0.045
0.103
0.113
0.056
0.057
0.165
0.065
0.028
0.064
0.027
0.039
0.028


1












in organic soils. Brown and Cisco (1984) found that for 14

soils, all but one showed a linear response up to pHw 7.0 to

CaCO3 additions in a greenhouse pot trial. The AEmod LR

values, using the pHW-BU relationship for Florida soils, were

calculated for all 34 soils, six of which had pHw values

<4.5. The original AE method regression of BU and pH did not

include soils with such low pH values and thus was only used

for 28 of the soils studied.

To facilitate discussion of the data, the AE, AEmod, and

Yuan DB tests will each be covered separately, first using

the incubation as the reference method. The second section

will cover the soil buffer methods using BaC12-TEA as the

reference method. Finally the last section will present the

findings of the two reference methods and how they compare to

each other.

AE Method and CaCO3 Incubation

The lowest correlation coefficient for any of the three

methods versus the three target pH levels was for the AE

method to LR5.5 (r = 0.68) (Fig.3-11). The AE method

estimated an average of only about 0.56 g CaCO3 kg-1 soil for

every 1.0 g CaCO3 kg-1 soil (equivalent to 2.24 Mg CaCO3 ha-1

or 1 ton CaCO3 acre-1; also equivalent to 3.36 Mg ag lime ha-1

or 1.5 tons ag lime acre-1) predicted by incubation LR5.5.

Although the AE method was highly correlated with

incubation LR6.0 and LR6.5 (r = 0.96, r = 0.96), the AE

method underestimated incubation LR at all practical levels

















a)


y = 0.10 + 0.49x
(r=0.68, n=13, C1.=29%)


O
'1-




01
C-)

10
O 1J


0
0 0


3
INCUBATION LR 5.5


1.0
(g CaC kg soil)
(g CaCO3 kg soil)


b)


y = -0.04 + 0.70x
(r=0.96, n=23, CV.=21%)



,


u0.
INCUBATION LR 6.0


10
(g CaCO3 kgl soil)
(g CaCO3 kg soil)


Sc)


y = 0.02 + 0.57x
8 (r=0.96, n=28, C.V.=12%)
a 1D-







0.0 10 2.0
INCUBATION LR 6.5 (g CaCO3 k-1 soil)
Figure 3-11. Regression curves and statistics of AE LR versus incubation LR.
a) to pH 5.5; b) to pH 6.0; c) to pH 6.5.


1.0-


0.o
Q(


uv, --L I












of lime addition. Figure 3-11 reveals that for soils that are

to be limed to pH 6.0, the AE method will only estimate an

average of about 0.67 g CaCO3 kg-1 soil for every 1.0 g CaCO3

kg-1 soil predicted by incubation. For soils that are to be

limed to a higher pH of 6.5, the AE method estimated only

about 0.59 g CaCO3 kg-1 soil for every 1.0 g CaCO3 kg-1 soil

predicted by incubation. Figure 3-11 also shows that as the

incubation LR increased for a particular target pH the AE

method estimated even less of the incubation LR. This was

also reported by Adams and Evans (1962).

The AE method seems to be more precise when soils are to

be limed to target pH 6.5. There was not a significant change

in r values at the two highest target pH levels. However, the

C.V. values decreased as the target pH values increased from

5.5 to 6.5. The decrease in C.V. values reflects a decrease

in dispersion of points about the regression line.

Fox (1980) found the AE method overestimated CaCO3

incubation LR to pH 6.5 when the CaC03 LR was below 2.4 g

CaCO3 kg-1 soil, and underestimated above that amount (r =

0.92). The findings of Fox were similar to those of Tran and

van Lierop (1981) who found that the AE method overestimated

CaCO3 predicted LR to pH 6.5.

AEmod Method and CaCO3 Incubation

The AEmod method was much better correlated to LR5.5 (r =

0.94) than the AE method '(Fig. 3-12). This may be due to the

lack of range on the x-axis. The AEmod method estimated on















30r


a)
y = 0.07 + 0.53x
(r=0.94, n=19, CV.=25%)


-1
INCUBATION LR 5.5 (g CaCO3 kg soil)
c) y = 0.09 + 0.54x
(r=0.98, n=34, CV.=12%)


b) y = 0.01 + 0.58x
(r=0.97, n=29, C.V.=19%)


20-






1D-1

INCUBATION LR 6.0 (g CaCO kg soil)
INCUBATION LR 6.0 (g CaCO3 kg-I soil)


1" 0 1.0 2D0 3.0
INCUBATION LR 6.5 (g CaCO3 kg soil)
Figure 3-12. Regression curves and statistics of AEmod versus incubation LR. a) to pH 5.5;
b) to pH 6.0; c) to pH 6.5.












the average about 0.57 g CaCO3 kg-I soil for every 1.0 g CaCO3

kg-1 soil predicted by incubation LR5.5.

The AEmod method was also highly correlated with

incubation LR6.0 and LR6.5 (r = 0.97, r = 0.98). However, the

AEmod method also underestimated the incubation LR. For soils

that were to be limed to pH 6.0 the AEmod method estimated an

average of 0.59 g CaCO3 kg-1 soil for every 1.0 g CaCO3 kg-1

soil predicted by incubation. For soils that were to be limed

to pH 6.5 the AEmod method estimated an average of 0.60 g

CaCO3 kg-1 soil for every 1.0 g CaCO3 kg-1 soil predicted by

incubation. Thus there appears to be no practical advantage

in using the AEmod method versus the AE method.

Yuan DB Method and CaCO3 Incubation

The Yuan DB method did not perform as well as had been

expected. Yuan (1974, 1975, 1976) showed that the method

performed well when compared to BaCl2-TEA (pH = 8.2) acidity,

Ca(OH)2 incubation to pH 7.0, and CaC03 incubation to various

pH levels, respectively as reference methods.

The Yuan DB overestimated incubation LR at all three

target pH levels (Fig. 3-13). Although the Yuan DB method had

a high r value (r = 0.91) for LR5.5, the method also had a

relatively high coefficient of variation (C.V. = 46%).

Comparing Fig. 3-13 and 3-11 it is obvious that the spread of

points about the regression line for the Yuan DB is much

greater than for the AE method. The Yuan DB method estimated













a)
y = -0.18 + 1.33x
(r=0.91, n=15, CV.=46%)


0
(-4
o






1.0
On








>-1


** "" -1
INCUBATION LR 5.5 (g CaCO3 kg soil)

c) = -0.02 + 1.28x
(r=0.96, n=34, CV.=21%)


3.0
0
U1
o



"2.0
Cr
oo

S1.0



CIO
p n


INCUBATION


1.0
LR 6.0


o0 2.0 /.
o/%


0 -
,y--


*



0.0 10 2.0 3.0
INCUBATION LR 6.5 (g CaCO3 kg soil)
Figure 3-13. Regression curves and statistics of Yuan DB versus incubation LR. a) to pH 5.5; b) to pHl
6.0; c) to pH 6.5.


y = -0.03 + 1.22x
(r 0.93, n=31, C.V.=35%)





S *


s *
*
0* *
i* **
0S ..
1",. .


2.0
(g CaCO3 kg-


soil)


-~ -L I


I












an average of 1.21 g CaCO3 kg-1 soil for every 1.0 g CaCO3

kg-1 soil predicted by incubation.

Fox (1980) also found that the Yuan DB method greatly

overestimated the CaCO3 LR to pHw 7.0 up to a LR of 3.72 g

CaCO3 kg-1 soil. The regression equation found was: Yuan LR =

0.64(incubation LR to pHw 7.0) + 2.68, (r = 0.93).

However Tran and Lierop (1981) reported the Yuan DB

method only estimated 77% of the incubation LR for soils with

a LR <5 g CaCO3 kg-1 soil. They also found that the Yuan

method as published by Yuan (1974) only estimated 91% of the

LR as shown by his data. McLean (1978) determined that the

Yuan DB actually underestimated the soil LR for soils with a

LR <2 g CaCO3 kg-1 soil by 11% (r = 0.81, C.V. = 24%).

The r values increased and C.V. values decreased for the

Yuan DB as the target pH increased to 6.0 (r = 0.93, C.V. =

35%) and 6.5 (r = 0.96, C.V. = 21%), respectively. The Yuan

DB estimated an average of 1.21 g CaCO3 for every 1.0 g CaCO3

kg-1 soil found by incubation to LR6.0. For incubation LR6.5,

the Yuan DB estimated on the average 1.27 g CaCO3 for every

1.0 g CaCO3 kg-1 soil found by incubation LR6.5.

All Three Methods and BaCl2-TEA

The AE method was very poorly correlated with the

BaC12-TEA method to all three target pH levels of pH 5.5,

6.0, and 6.5 (Table 3-7). As the target pH increased, the AE

method was increasingly better correlated with the reference

method. The AE method severely underestimated the reference













Table 3-7. Regression statistics of various lime requirement determinations versus the
BaCl2-TEA (pH 8.2) extractable acidity reference method (g CaCO3 kg-1 soil).

Dependent Intercept Slope Independent
Variable Variable n r CV

--%--
AE to pH 5.5 0.22 0.02 BaC12-TEA 13 0.00 41

AE to pH 6.0 0.14 0.07 BaC12-TEA 24 0.19 78

AE to pH 6.5 0.20 0.12 BaC12-TEA 28 0.57 41

AEmod to pH 5.5 -0.02 0.14 BaC12-TEA 19 0.78 46

AEmod to pH 6.0 -0.12 0.20 BaCI2-TEA 30 0.70 62

AEmod to pH 6.5 0.01 0.24 BaCl2-TEA 34 0.80 36

Yuan to pH 5.5 -0.61 0.38 BaCl2-TEA 16 0.77 78

Yuan to pH 6.0 -0.54 0.46 BaC12-TEA 34 0.77 67

Yuan to pH 6.5 -0.33 0.60 BaCl2-TEA 34 0.84 39

Incub to pH 5.5 -0.08 0.22 BaCl2-TEA 20 0.67 66

Incub to pH 6.0 -0.11 0.28 BaCl2-TEA 31 0.60 72

Incub to pH 6.5 -0,08 0.41 BaC1R-TEA 34 0.76 44












method. For example, for LR6.5 the AE method estimated only

about 0.2 g CaCO3 for every 1.0 g CaCO3 kg-1 soil estimated by

the reference method.

The AE LR to lower pH levels would be expected to

underestimate BaCl2-TEA LR since the BaCl2-TEA extractant is

buffered at pH 8.2. More acidity would theoretically be

extracted at the higher pH than would be neutralized by the

incubation. As the incubation target pH LR increases it would

begin to measure the same amount of acidity as the BaCl2-TEA

method.

The AEmod method was much better correlated with the

reference method than the AE method (Table 3-7). For LR6.0

the correlation coefficient was slightly lower and the C.V.

value higher than for LR5.5 and LR6.5. The AEmod method also

severely underestimated the reference method. For LR6.5 the

AEmod method estimated on the average 0.24 g CaCO3 for every

1.0 g CaCO3 kg-1 soil estimated by the reference method. The

Yuan DB method was correlated as well with the reference

method as was the AEmod method (Table 3-7). Yet the Yuan DB

method also severely underestimated the reference method LR.

For LR6.5 the Yuan DB method measured 0.5 g CaCO3 for every

1.0 g CaCO3 kg-1 soil predicted by the reference method.

When the total amount of acidity measured by the

BaCl2-TEA (pH 8.2) method was used as a reference method,

lower r values and higher C.V. values were found for each of

the three soil-test methods than when CaCO3 incubation to pH












5.5, 6.0, and 6.5 was used as the reference method. This is

most probably because the BaC12-TEA method measures more

pH-dependent acidity above pH 7.0.

AE Total Acidity and BaC12-TEA

AE total acidity (TA), was highly correlated with

BaC12-TEA acidity (Fig. 3-14). AE TA underestimated that of

the reference method by about half. When.the reference method

predicted that the amount of base required to neutralize the

measured acidity was 1.0, 2.0, and 3.0 g CaCO3 kg-1 soil, the

AE TA estimated only 0.59, 0.46, and 0.42 g CaCO3 kg-1 soil,

respectively.

This seems to indicate that although the BaCl2-TEA method

is not a good measure of the AE LR to pH 5.5, 6.0, and 6.5,

the total amount of acidity measured by the AE buffer was

highly correlated to the total amount of soil acidity

measured by the BaC12-TEA method. Savant and Kibe (1971)

reported that the AE buffer measured about 1/3 of the total

acidity of the soil determined by BaC12-TEA + 1M KC1

exchangeable acidity on lateritic rice soils (r = 0.77). The

exchangeable acidity was almost negligible, constituting less

than 3% of the total acidity.

BaC12-TEA and CaCO3 Incubation

The two reference methods were not as well correlated

(Table 3-7) as other researchers have found (McLean et al.,

1966; Webber et al., 1977; Fox, 1980). However, as the target

pH increased the precision of the relationship increased.






















-l
o AE TA = 0.25 + 0.34(BaC2 -TEA)
2

S20 (r = 0.97, n = 34, C. =10%)

C-
O n J I*
8 *




CH


p I
0.0 1.0 20 30 4D 5.0

BaC12-TEA EXTRACTABLE ACIDITY (g CaCO3 kg-1 soil)
Figure 3-14. The regression equation and statistics between the total amount
of acidity measured by the AE buffer (TA) and BaCl -TEA
extractable acidity.












The BaCl2-TEA acidity overestimated the incubation LR at

all three target pH levels. This difference was most

pronounced at the lower target pH level but even for LR6.5

the BaCl2-TEA method estimated about twice as much acidity as

the incubation method.

The overestimation of CaCO3 incubation LR to pH 6.5 by

the BaCl2-TEA method was expected. Similar results have been

reported by several researchers comparing lime-incubation

methods to BaCl2-TEA extractable acidity. Savant and Kibe

(1971) in the study mentioned earlier, found that Ca(OH)2

titration to pH 6.5 to 7.0 (Puri, 1963), estimated about 57%

of the total acidity determined mostly from BaCl2-TEA

extractable acidity. Loynachan (1981) used BaCl2-TEA adjusted

to 80% of the theoretical LR as proposed by Peech (1965).

Loynachan found that the BaCl2-TEA method method

overestimated the CaCO3 incubation LR to pH 6.5 by over 4

times for soils with high amounts of organic matter. The

slope of 0.68 determined for this relationship was similar to

the slope found in this study.

Suggested Modifications

The underestimation by the AE LR method can be corrected

for in one of three ways. The first one would be to calculate

the AE LR the same as is currently done and then use the

regression equations (Fig. 3-11) to adjust for the

underestimation of the AE LR method. The use of regression












equations for this type of adjustment has already been

proposed (McLean, 1982).

The drawback to this approach is that the pH-BU

relationship is still employed in the calculation. As

recently as 1985, Magdoff and Bartlett questioned the use of

the pH-BU relationship for use in predicting LR. The wide

spread of data about the regression line found for Florida

soils (Fig. 3-2) confirms the notion that the pH-BU

relationship is a very general one.

The second approach would eliminate the use of the pH-BU

relationship, using only the soil-buffer equilibrium pH.

Figure 3-15 shows the soil-buffer equilibrium pH values

regressed against incubation determined LR. This regression

equation can then be used to obtain the amount of lime needed

according to the final soil-buffer equilibrium pH. It is

evident, however that the variability using this approach is

greater than if the first one were used. Tran and van Lierop

(1981) also tried using the soil-buffer equilibrium pH

reading to predict incubation LR. They had expanded the soil-

buffer ratio from 1:2 to 1:4 however. At the 1:4 soil-buffer

ratio they found no significant difference between using

buffer pH or converting to AE LR by means of the pH-BU

relationship. At the 1:2 soil-buffer ratio they found that

using the pH-BU relationship improved the r value (r =

0.78**) over using the soil-buffer pH (r = -0.63**).





















y = 7.71 0.19x

(r = 0.72, n = 31, CV. = 27%)




L4
S5- o


0

W C

8.0







1.0 20 3D
-i
INCUBATION LR (g CaCO3 k" soil)
Figure 3-15. The regression equation and statistics between the AE
soil-buffer equilibrium pH and the incubation LR.












The most satisfactory method of prediction of the

incubation LR found in this study was one which used the AE

soil-buffer equilibrium pH value and the difference. between

initial pH and target soil pH (AE-F). When incubation LR, to

pH 5.5, 6.0, and 6.5 (n = 85), was regressed on AE buffer pH

and the difference between target and initial pH, the

following equation (AE-F) resulted:

Incubation LR = 17.4 2.27 (AE buffer pH) +

0.68 (target pH initial pH)

(r = 0.94, C.V. = 27%, n = 85)

Employing the change in pH and AE buffer pH (AE-F)

appeared to be as accurate as using the AE method (Fig 3-16).

Use of the latter correction factor eliminated the use of the

pH-BU relationship that has been shown to be a very

generalized relationship for Florida soils.

The ESTL assumes that 15 ml soil is equal to 15 g soil

and the lime recommendations are based on this. However, 15

ml soil was found to be equal to 20 g soil in this study.

Hence, the recommendations given by the ESTL are actually 33%

greater than the AE test predicts. Thus current ESTL

recommendations already compensate for some of the

underestimation of the AE method.

Summary and Conclusions

In summary the following conclusions are reached from

this study of a group of surface soils representative of

sand-textured Florida agricultural soils:
























y = 0.07 + 0.95x

(r = 0.96, n = 30, C. = 24%)


O 20-


1

0
^d yJ
u
u


o y

3 *
1.0 *



00


0
f .*


0



Q 1.0 2.0 3.0

INCUBATION LR 6.0 (g CaCO3 kg- soil)
Figure 3-16. The regression equation and statistics between AE-F
and incubation LR to pH 6.0.












1/ The AE LR method estimated 0.56, 0.67, and 0.59 of

the incubation LR to pH 5.5, 6.0, and 6.5,

respectively.

2/ Use of the AEmod method did not show a practical

increase in accuracy of LR prediction.

3/ The BaC12-TEA method for determining extractable

acidity was highly correlated with the total acidity

measured by the AE buffer.

4/ The Yuan DB method overestimated the incubation LR

and was not as precise as the AE method.

5/ A regression equation, shown below, determined from

the data found in this study gave a very satisfactory

estimate of LR by using both the AE soil-buffer

equilibrium pH and the difference between initial

soil pH and target pH.

Incubation LR = 17.4 2.27 (AE buffer pH) +

0.68 (target pH initial pH).



















CHAPTER IV
FIELD STUDY

Introduction

Studies have shown that lime-soil reactions behave

differently under field conditions than in laboratory incuba-

tion studies. In the field, liming materials are normally

applied as agricultural limestone and are partially mixed with

the soil by disking or plowing. In the laboratory, it is

customary to use 320 mesh reagent grade CaCO3 and thoroughly

mix it with the soil. The agricultural limestone (ag lime) is

usually less effective and slower in neutralizing the soil

acidity due both to its larger particle size and to nonuni-

formity of mixing with the field soil. Thus reaction under

field conditions usually takes longer and is less complete

than in laboratory incubation studies. So the amount of ag

lime needed to accomplish a desired pH change is greater than

the laboratory determined LR.

The term limingg factor' refers to the multiplication

factor used to account for the difference in neutralization

efficiency in the laboratory and in the field. This concept

can be expressed by the following simplified equation:

LR X LIMING FACTOR = FIELD LR












Thus by knowing the LR, which can be determined by a buffer

method, and the actual field LR, the liming factor can be

determined.

The AE method uses a liming factor of 1.5, which was

obtained from work done by Pierre and Worley (1928),

Schollenberger and Salter (1943), and research station data.

Yuan (1976) recommended that a liming factor of 1.2 to 1.8 be

employed for his method, on Florida soils. The exact value

would be dependent on the limestone specifications, desired

soil pH, and the amount of lime to be applied. The range of

liming factors found by researchers varies widely. Walker

(1952) found that 2 to 3 times as much lime was needed in the

field as was calculated from complete mixing in the

laboratory.

Incubation studies are valuable in that they allow for

the correlation of a soil test method for a large number of

soils with a broad range of different characteristics. Final

adoption of any soil test method should, however, be preceded

by field calibration studies to determine what the actual

liming factor should be for Florida soils.

The field trials discussed in this chapter will attempt

to give some indication of the applicability of a liming

factor of 1.5, as is assumed by the AE method. Field

experiments were conducted on two farms and a research station

to determine the long-term effects of increasing rates of lime

on soil pH. The main objective of this study was to obtain












field data showing the approximate lime levels needed to

adjust soil pH to predetermined target pHw values. These

results could then be compared to what was found under

laboratory conditions and thus give an indication of the

liming factor.

Materials and Methods

Description of Sites and Soils

Of the four experiments initially laid out in the fall of

1983, three were located in Suwannee County (Fig. 4-1). The

fourth experiment will not be reported on since only six

months of data were recorded prior to the loss of the site due

to experimental difficulties. Site 1 was on a farm located in

Section 21 T.3S.-R.15E.. The soil on which the trial was

performed belongs to the Blanton series (loamy, siliceous,

coated Typic Quartzipsamment). Blanton soils are described as

deep, light-colored, sandy soils principally on gentle

slopes, that developed from thick beds of marine sand very low

in silt and clay (Soil Survey Staff, 1965). Virgin sites of

these soils are usually strongly acid, low in natural

fertility, and low in organic matter content. As shown by

Table 4-1, under cultivation this soil can have relatively

high pH values and extractable base contents.

Site 2 was located on another farm in Section 35

T.3S.-R.12E.. The soil belongs to the Blanton-Susquehenna

complex. The Susquehenna series is described as a fine,

montmorillonitic, thermic, vertic Paleudalf. This complex





















SITE 3
SITE 3


SITE 2


SITE 1


Location of


Florida J


Figure 4-1. The location of the three field trials in
Suwannee County, Florida.













Table 4-1.Selected chemical and physical characteristics of the top 15 cm
of the three soils used in the field study.


pHw pHw pHs 1M KC1.
Site (1:1) (1:2) (1:2) OC OM Clay Silt Sand Mehlich I extractable exch.
Al P K Ca Mg Al

------g 100 g-------- ------------mg kg-1------------
1 6.0 5.9 5.2 1.2 2.0 2 3 95 226 126 44 580 308 0.00

2 5.0 5.0 4.1 0.5 0.8 5 2 93 84 27 20 46 27 0.19

3 5.9 5.8 5.1 1.1 1.9 2 3 94 238 56 36 293 171 0.02












often has a clayey substrata that underlies a fine sand

surface layer which was found at this site (Soil Survey Staff,

1965). This complex is also low in organic matter and in

natural fertility, with a low available moisture capacity.

Table 4-1 shows that even though this soil is under cultiva-

tion, soil pH and extractable base measures are relatively

low. It was noted that the owner of the land where site 2 is

located did not provide for a large amount of inputs. The

particular field on which the experiment was located was used

mostly for pasture and apparently few inputs were applied to

the field.

Site 3 was located on the Live Oak Agricultural Research

Center (ARC) at Section 22 T.2S.-R.14E.. The soil was a

Lakeland series described as a thermic, coated, Typic

Quartzipsamment. Virgin sites of these soils are usually deep,

well drained to excessively drained, strongly acid, coarse

textured, and generally low in natural fertility and organic

matter content (Soil Survey Staff, 1965,). Once again, soil pH

and extractable bases were relatively high, in the trial area

(Table 4-1), probably due to continued cultivation and the

addition of proper soil amendments.

During the process of locating suitable sites for the

field experiments, great difficulty was encountered in

obtaining fields with low pHw values. Most of the farmers in

Suwannee County already had their soil pHw at or above 6.0.












Experimental Design

All of the trials were set up as a randomized complete

block design (RCBD). Individual plot size was 6 x 10 meters.

There were five treatment levels consisting of no lime, no

lime + 15 kg ha-1 Mg, 1.12 Mg ha-1 lime, 2.24 Mg ha-1 lime, and

4.48 Mg ha-1 lime.

The limestone applied was a bagged dolomitic limestone

with the following analyses: Effective neutralizing power,

89.6%; CaC03 equivalent, 104%; 100% passed through 10 mesh;

80% passed through 50 mesh; 75% passed through 100 mesh; 1730

lbs = 1 ton standard liming material. There were four replica-

tions resulting in a total of 20 plots per experiment. The no

lime + 15 kg ha-1 Mg treatment was included for a yield study

and will not be discussed in the observation of soil pH, over

the duration of the trials.

Land Preparation. Use, and Analyses

Site 1. The site was established on November 16, 1983.

Soil cores were taken by walking a zigzag pattern up one half

of the experimental plot and down the other half. Twenty cores

to a depth of 15 cm were taken in all. The soil was mixed and

bagged for future analyses. This sample would be representa-

tive of the entire experimental field prior to the initiation

of the experiment.

The agricultural lime was hand applied to the respective

plots. A blanket application of NH4NO3, (NH4)2SO4, and KCl

fertilizer was also applied. The lime and fertilizer were












disked in and wheat was planted a few days later by the

landowner. The wheat was harvested on May 10, 1984.

Soybeans were planted in the first week of July 1984 and

harvested on November 9, 1984. The field was left fallow until

April 1985 when corn was planted and harvested in July 1985.

Periodic soil samples were taken to monitor the soil pH,.

Sampling consisted of taking four soil cores to a 15 cm depth

from a 3 x 6 meter area in the middle of each plot. The four

cores were mixed together so that only one pH, value was

obtained per plot. Sampling dates were 11/83, 1/84, 3/84,

5/84, 11/84, and 8/85.

Site 2. Site 2 was established on November 26, 1983. Soil

samples were taken in accordance with the method mentioned for

the previous site. The lime and fertilizer were hand applied

and then disked in by the landowner. Wheat was then broadcast

by a hand-held spreader and subsequently disked in.

The wheat was harvested in early June. At that time the

landowner expressed his desire to discontinue with the

double-cropping trials. He wished to utilize the land for

grazing his livestock on rye. He did allow continued sampling

and monitoring of the pH in the field. Samples dates were

11/83, 1/84, 3/84, 5/84, 2/85, and 8/85.

Site 3. The plots on this site were established on

December 12, 1983. Soil samples were taken according to

previously mentioned procedures. The limestone and fertilizer

were hand-applied and immediately disked in.













The wheat was harvested in early June and on July 1, 1984

soybeans were planted. The soybeans were harvested on October

30. Wheat was once again planted in December with the same

fertilizer additions as the previous season. After the wheat

was harvested in May 1985, soybeans were planted in June 1985.

As of the last sampling date the soybeans were still in the

field. The field was sampled on the following dates; 11/83,

3/84, 5/84, 7/84, 11/84, and 8/85.

The soils from sites 1 and 3 were also included in the

incubation study reported on in Chapter III. Soil numbers 16

and 18 in the incubation study were taken from sites 1 and 3,

respectively.

Results and Discussion

Observations of Field LR Over the Length of the Study

Commonly used target pHw levels in Florida are 5.5, 6.0,

and 6.5. For purposes of discussing the field results these

target pHw levels will be used as reference points.

Evaluation of the field data was complicated by the dif-

ficulty in determining actual field LR. The problems associ-

ated with determining the actual field LR and the decisions

made to deal with these problems are discussed next.

First, a decision had to be reached on when the field pH,

had reacted sufficiently with the lime so that it could be

considered to be at equilibrium. Observations indicated that

at all of the sites pHw increased linearly up to six months

after liming (Fig. 4-2, 4-3, 4-4). The initial decrease in


















4.48 Mg ha-1
SITE 1

65 -


2.24 Mg ha-1



ll.12 tMg ha-1

,60 Mg ha-1








55-







5 0C 1 1I.... I I I
11/83 1/84 3/84 5/84 11/84 8/85
SAMPLING DATE
Figure 4-2. The effect of lime treatments on soil pHw over the time period of
the study, at site 1. Each point represents the average of four
replications.






























CL --02.24 Mg ha-"

00

55 1.12 Mg ha-1







1 I I 0 Mg ha-
11/83 1/84 3/84 5/84 2/84 8/85
SAMPLING DATE
Figure 4-3. The effect of lime treatments on soil pHw, over the time period
of the study, at site 2. Each point represents the average of
four replications.































4.48 Mg ha-1



1.12 Mg ha-1



2.24 Mg ha-1


0 Mg ha-1


F.-L U .------_- ------L--


3/84 5/84 7/84


11/84


SAMPLING DATE


Figure 4-4.


The effect of lime treatments on soil pHw, over the time period of
the study, at site 3. Each point represents the average of four
replications.


SITE 3


65








O.


11/83


I I


--


8/8b












soil pHw at site 1 (Fig. 4-2) may have been a result of the

application of ammonium fertilizer.

Although Yuan et al. (1978) observed continued pH,

increases 2 years after liming, it was decided to include all

pHw measurements taken after the six month equilibration

period in the determination of a representative field pHw. The

major reason that this decision was made is that a farmer is

more concerned with the pHw values attained one or two years

after liming and not the pHw that is theoretically obtainable.

Another reason is that the further increases in pHw after the

initial six month period are probably relatively minor

compared to the increases observed during the initial period.

Once a decision was reached on what the actual time

period and respective pHw measurements were to be used,

another problem had to be dealt with. More specifically, the

noticeable fluctuation in pHw values after the initial six

month period made determining actual pH, levels difficult. For

example, Fig. 4-3 shows a variation of pHw for the no lime

added treatment of about 0.5 pH. The wide pHw fluctuations

observed for the above three trials is a well documented

occurrence. Shuman et al. (1983) cite several references that

have reported these observations. Hester and Shelton (1933)

concluded that these seasonal fluctuations are attributed to a

combination of rainfall, fertilization, and nitrification.












The wide pHw fluctuations were dealt with by averaging and

the application of regression equations. The actual calcula-

tions are discussed in more detail in the following section.

Calculation of Field Lime Requirement

To provide an indication of the field lime'requirement

the following calculations were made:

1/ The average pH, of four replications for each

treatment level were calculated for each of the

sampling dates (Fig. 4-2, 4-3, 4-4).

2/ For each treatment level, the average pHw of all

measurements taken at least six months after the

lime was added to the soil was calculated.

3/ The treatment levels were regressed with the average

pHw values shown in Table 4-2. The following

regression equations were calculated, with (x) being

limestone in Mg ha-1, and (y) being the predicted

pHw of the field soil:

Site 1 ( y = 6.18 + 0.09x ) [1]

Site 2 ( y = 5.35 + 0.23x ) [2]

Site 3 ( y = 6.15 + 0.08x ) [3]

These regression equations were then used to calculate

the field LR.

Comparison of Field and Laboratory Data

In Table 4-3 are compared the various lime requirement

determinations for the three soils. All LRs were converted to

g CaCO3 kg-1 soil to allow for direct comparison.




Full Text

PAGE 1

EVALUATION OF LIME REQUIREMENT METHODS FOR FLORIDA'S SANDY SOILS By THOMAS S. DIEROLF A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 1986



PAGE 1

8 Hopkins et al. (1903) proposed a method in which a neutral salt solution of NaCl was mixed with a soil sample. Their theory held that the mineral acids in the soil would unite with the mineral bases. A standard fixed alkali was then used to titrate the liberated mineral acid. In the laboratory they added quantitative amounts of lime to soil samples in accordance with the method and found that practically all of the acidity was neutralized. Jones (1913) recommended a method in which calcium acetate was used to extract the acidity. The resultant acetic acid was then titrated with 0.1M NaOH in the presence of phenolphthalein. This method was later shown to underestimate the lime requirement because the extractant was most strongly buffered at pH 4.76 which was too low to effect complete replacement of exchangeable hydrogen. MacIntire et al. (1917) reacted soils with CaCO3 and MgCO3 in the field and in the laboratory. They concluded that soils have capacities to decompose CaCO3 in the field greatly beyond the Veitch lime requirement. MacIntire (1915) developed a method where they evaporated calcium bicarbonate-soil suspensions to a thin paste on a steam bath and determined the soil-CaCO3 reaction from analysis of the residual CaCO3. Soil-lime Potentiometric Titrations Sharp and Hoagland (1915) were the first ones to use potentiometric titration with Ca(OH)2, using the hydrogen



PAGE 1

56 Table 3-6.Regression equations of final pHs (y) versus g CaCO3 kg-1 soil added (x) used to compute incubation LR. Also shown are the comparison of r2 for linear and curvilinear relationships and the standard deviation (s) for the linear equations. Soil r2 No. Regression Equation Linear Curvilinear s 1 pHs=4.87+0.81x 99.3 99.5 0.046 2 pHs=3.98+1.51x 99.6 99.5 0.052 3 pHs=4.69+1.06x 99.1 99.4 0.054 4 pHs=5.00+1.22x 99.0 99.6 0.036 5 pHs=3.98+1.30x 99.9 99.0 0.023 6 pHs=3.13+0.93x 99.4 99.8 0.087 7 pHs=4.63+0.66x 99.5 99.8 0.038 8 pHs=4.96+1.25x 99.6 99.6 0.020 9 pHs=5.53+0.95x 99.1 98.9 0.017 10 pHs=5.03+0.83x 99.1 99.2 0.024 11 pHs=3.65+1.51x 94.8 95.3 0.210 12 pHs=3.59+0.98x 98.7 98.7 0.116 13 pHs=5.07+0.92x 98.4 99.3 0.044 14 pHs=5.16+1.16x 99.0 99.0 0.036 15 pHs=4.87+0.97x 99.7 99.7 0.024 16 pHs=5.32+1.06x 99.4 99.7 0.024 17 pHs=4.25+1.70x 98.2 98.0 0.105 18 pHs=4.07+1.21x 98.6 99.8 0.106 19 pHs=4.21+1.16x 97.9 97.7 0.088 20 pHs=4.21+1.84x 96.3 99.1 0.101 21 pHs=4.30+2.74x 90.5 97.6 0.197 22 pHs=4.44+0.88x 99.5 99.8 0.051 23 pHs=4.92+2.27x 99.4 99.8 0.045 24 pHs=4.22+1.79x 98.4 99.2 0.103 25 pHs=3.92+1.47x 98.3 98.2 0.113 26 pHs=4.29+1.76x 99.1 99.2 0.056 27 pHs=4.22+0.83x 99.1 99.4 0.057 28 pHs=3.55+0.56x 96.3 96.0 0.165 29 pHs=3.47+2.01x 99.6 99.6 0.065 30 pHs=5.32+2.01x 99.2 99.1 0.028 31 pHs=4.27+0.74x 99.4 99.9 0.064 32 pHs=5.19+0.98x 99.5 99.7 0.027 33 pHs=5.13+0.71x 98.1 98.2 0.039 34 pHs=4.90+0.85x 99.5 99.5 0.028



PAGE 1

9 electrode. They added Ca(OH)2 to soil suspensions until a definite alkaline reaction was obtained. They also acknowledged that there was incomplete time for complete neutralization of the soils acids. MacIntire (1920) reported at a 1917 meeting on a collaborative effort that compared all of the aforementioned methods as well some others. Calcium carbonate-soil incubations were used to assess the predicted lime requirements by the various methods. Although the lab assigned to evaluate the Jones method did not provide any data, MacIntire still reported that the Jones method offered the best possibilities for obtaining the coefficient of lime determination. In 1919, two Danes, Bjerrum and Gjaldbaek, made an epochal contribution to the field when they developed titration or buffer curves as determined potentiometrically, to study the acidic and basic properties of soils. They also established the relationship between partial pressure of C02 and pH values of saturated solutions of CaCO3 (Shaw, 1953). Jensen (1924) and Christensen and Jensen (1926), also of Denmark, used Ca(OH)2 and CaCO3, respectively, to obtain soil-buffer pH curves. Pierre and Worley (1928), in Alabama, incubated soil with increments of Ba(OH)2 for three days. Determinations were made on the clear diffusate by the colorimetric pH method. The residual carbonate procedure, developed by Bradfield and Allison in 1933, is the best known titration procedure



PAGE 1

52 718 ,20 /17 / 4.00.0 1.0 2.0 -1 g CaCO3 kg soil Figure 3-4. Titration curves of final pHs versus amount of CaCO3 added for soils 7, 8, 17, 18, and 20. S /23 /2 / /16 -15 6.0 / 27 5.0 4.0 0.0 1.0 2.0 g CaCO3 kg soil Figure 3-5. Titration curves of final pHs versus the amount of CaCO added for soils 15, 16, 23, 24, an2 27.



PAGE 1

6 used for centuries in Europe and knowledge of its benefits were carried over to the U.S from previous experience and also from writings of Europeans on the principles of plant growth and nutrition. This knowledge was rarely put to use, however, since it was easier for the farmers to move to more fertile grounds than to haul lime the long distances usually required. Extensive investigations on crop responses to lime appeared to originate in the Northeastern region of the U.S.. Near the end of the 19th century Wheeler and co-workers related the need for lime on acid upland soils in Rhode Island. Prior to this time the practice of liming had become well established in certain localities based primarily on farmer experience. It was not until the early 20th century that extensive liming trials were begun in the Southern and Midwestern U.S.. Litmus Test Although litmus was first used as an indicator of acidity in 1865, it had no general application for recommending lime until many years later. Wheeler and Tucker (1896), reported that soils testing more acid on litmus paper benefited more from liming than did soils testing neutral or nearly neutral. They called for a chemical test that, by determining the relative acidity of a soil, would be able to prescribe the correct quantity of lime needed.



PAGE 1

11 values of a like aliquot of the original extracting solution and the soil extract was equivalent to the neutralization of the barium hydroxide by soil acids. Mehlich (1942a) proposed an improved buffer consisting of barium chloride-triethanolamine at pH 8.2. He substituted barium chloride for barium hydroxide because the CO2 of the air did not interfere with the titrations. The base-exchange capacity and the exchangeable H+ were determined by titrating one aliquot of extract. Dunn (1943) incubated soil, Ca(OH)2, and distilled water for four days with thorough shaking twice daily. He compared this titration curve to field and laboratory experiments, where he had added increments of CaCO3 or Ca(OH)2 to soil for several months. The titration curve underestimated the field lime requirement but was accurate enough to bring the field soils to within 0.5 pH of the desired pH. Brown (1943) first developed the idea of combining the use of the glass electrode and the concept of depression of the buffer pH as a measure of exchangeable H+ present in the soil. Soil was added to either neutral normal acetate or normal acetic acid. The pH was determined in the mixture. The depression in pH, in the case of the NH40Ac, or the increase in pH, in the case of the acetic acid was then read off of a titration curve as millequivalents of acid or base, respectively.



PAGE 1

Table 3-1. Classification of 34 soils from which the top 15 cm was used in the incubation study. (Soil Survey Staff, 1985). oil County Series Family or higher taxonomic class 1 Alachua Tavares Hyperthermic, uncoated Typic Quartzipsamment 2 Alachua Chipley Thermic, coated Aquic Quartzipsamment 3 Alachua Arredondo Loamy, siliceous, Grossarenic Paleudult 4 Alachua Chipley Thermic, coated Aquic Quartzipsamment 5 Alachua Wauchula Sandy, siliceous, hyperthermic Ultic Haplaquod 8 Alachua Kendrick Loamy, siliceous, hyperthermic Arenic Paleudult 9 Alachua Kendrick Loamy, siliceous, hyperthermic Arenic Paleudult 10 Alachua Chipley Thermic, coated Aquic Quartzipsamment 11 Alachua Millhopper Loamy, siliceous, hyperthermic Grossarenic Paleudult 12 Alachua Plummer Loamy, siliceous, thermic Grossarenic Paleaqult 13 Alachua Gainesville Hyperthermic, coated, Typic Quartzipsamment 14 Suwannee Lakeland Thermic, coated, Typic Quartzipsamment 15 Suwannee Lakeland Thermic, coated, Typic Quartzipsamment 16 Suwannee Lakeland Thermic, coated, Typic Quartzipsamment 17 Suwannee Blanton Loamy, siliceous, thermic Grossarenic Paleudult 18 Suwannee Blanton Loamy, siliceous, thermic Grossarenic Paleudult 19 Suwannee Blanton Loamy, siliceous, thermic Grossarenic Paleudult 20 Alachua Newnan Sandy, siliceous, hyperthermic Ultic Haplohumod 21 Alachua Sparr Loamy, siliceous, hyperthermic Grossarenic Paleudult 22 Alachua Blichton Loamy, siliceous, hyperthermic Arenic Plinthic Paleaqult 23 Alachua Sparr Loamy, siliceous, hyperthermic Grossarenic Paleudult 24 Alachua Kanapaha Loamy, siliceous, hyperthermic Grossarenic Paleaqult 25 Alachua Newnan Sandy, siliceous, hyperthermic Ultic Haplohumod 26 Alachua Tavares Hyperthermic, uncoated Typic Quartzipsamment 27 Alachua Pomona Sandy, siliceous, hyperthermic Ultic Haplaquod 28 Alachua Wauchula Sandy, siliceous, hyperthermic Ultic Haplaquod 29 Alachua Pomona Sandy, siliceous, hyperthermic Ultic Haplaquod 30 Alachua Kendrick Loamy, siliceous, hyperthermic Arenic Paleudult 31 Alachua Arredondo Loamy, siliceous, hyperthermic Grossarenic Paleudult 32 Alachua Gainesville Hyperthermic, coated Typic Quartzipsamment 33 Alachua Gainesville Hyperthermic, coated Typic Quartzipsamment 34 Alachua Gainesville Hyperthermic, coated Typic Ouartzipsamment



PAGE 1

Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science EVALUATION OF LIME REQUIREMENT METHODS FOR FLORIDA'S SANDY SOILS By THOMAS S. DIEROLF May 1986 Chairman: Gerald Kidder Major Department: Soil Science Laboratory and field studies were conducted to evaluate two soil-test methods used to determine soil lime requirement (LR). The Adams-Evans (AE) LR method which is currently used by the Extension Soil Testing Laboratory (ESTL) to make liming recommendations in Florida and the Yuan double buffer (DB) method which was specifically designed to measure the LR of Florida's sandy soils were compared to a standard CaCO3 incubation procedure. A laboratory study of 34 Florida soils, with low organic matter and clay contents, indicated that the Yuan DB method was a poor predictor of the incubation LR. The AE method was highly correlated with, but underestimated incubation LR. An alternative method of calculating the AE LR (AE-F) was developed from the study data. The AE-F method eliminated the xi



PAGE 1

LIST OF TABLES Page Table 3-1. Classification of 34 soils from which the top 15 cm was used in the incubation study. (Soil Survey Staff, 1985) .................... 40 Table 3-2. Chemical and physical characteristics of the 34 soils used in the incubation study........ 42 Table 3-3. -Mean weights and standard deviations of soil contained in an 11 ml scoop. An excess amount of soil was scooped into the container, the side of the scoop was gently tapped three times, the soil was leveled off, and weighed. 43 Table 3-4. Individual lime rates applied to the 34 incubated soils. ESTL measured pHw and AE buffer pH. LR (Ibs acre-1) was determined from published tables and converted to g CaCO3 100 g-1 soil to attain pHw 6.5.......... 45 Table 3-5. Periodic pHw (1:2) measurements taken from 14 of the experimental units over the length of the incubation study to determine when equilibrium pHw was reached................... 47 Table 3-6. Regression equations of final pHs (y) versus g CaC03 kg-1 soil added (x) used to compute incubation LR. Also shown are the comparison of r2 for linear and curvilinear relationships and the standard deviation (s) for the linear equations ................ 56 Table 3-7. Regression statistics of various lime requirement determinations versus the BaCl2-TEA (pH 8.2) extractable acidity reference method (g CaCO3 kg-1 soil).......... 64 Table 4-1. Selected chemical and physical characteristics of the top 15 cm of the three soils used in the field study........... 78 vii



PAGE 1

5 of the development and improvement of soil LR methods. Section two presents a general description of soil acidity and related soil characteristics that influence soil LR. Section three describes techniques of correlating and calibrating soil LR methods. Finally section four reviews findings of previous work evaluating the precision and accuracy of the Adams-Evans (AE), Yuan double buffer (Yuan DB), and SMP methods. History of Lime Requirement Methods The methods of lime requirement determination used today are the result of an evolution of methodology and theory. A historical review of the pertinent works can lead to a better understanding of why the current lime requirement methods are being employed. Limina Practices Recently, several reviews summarizing the early work pertaining to liming, have been published (Adams, 1984; Barber, 1984; Lathwell and Reid, 1984; and McLean and Brown, 1984). References to the use of lime date back to the first and second century B.C.. Ruffin, through his writings in the early and mid 19th century, promoted the use of lime in the United States by reporting that marl applications improved crop yields on his farms. With the advent of agricultural research stations, several states in the late 19th century began research on liming using burned lime, gas lime, or marl. Lime had been



PAGE 1

20 the general pH-BS curve. Soils actually exhibit widely differing titration curves (McLean et al., 1960; Magdoff and Bartlett, 1985). Measurement of Base Saturation Debate also occurs as to what criteria are to be used to determine base saturation. One way is to use a buffered extractant such as BaCl2-TEA at pH 8.2 or NH4OAc at pH 7.0 to measure total exchangeable cations (Peech, 1939; Mehlich, 1942b). Coleman et al. (1959) proposed determining percent base saturation based on the ECEC, where a neutral salt such as KC1 is used to extract the exchangeable cations at the pH of the soil under field conditions. Sanchez (1976) reported that calculating base saturation based on BaCl2-TEA at pH 8.2 or pH 7.0 1M NH40Ac makes a soil seem more acid than it is if the field pH is lower than the pH of the extractant. He cited work by Buol in 1973, where Buol compared 88 soils from the Midwest, and the Southeast U.S., and Puerto Rico and found 35% base saturation at pH 8.2 was equal to 55% base saturation at effective CEC. For sandy soils or soils with >1% organic matter, the relationship would be different. Methods measuring CEC, which are determined at pH values appreciably higher than the soil pH, overestimate the ability of variable charge soil to retain cations in the field (Horn et al., 1982; Gillman et al., 1983). Methods more



PAGE 1

59 of lime addition. Figure 3-11 reveals that for soils that are to be limed to pH 6.0, the AE method will only estimate an average of about 0.67 g CaCO3 kg-1 soil for every 1.0 g CaCO3 kg-1 soil predicted by incubation. For soils that are to be limed to a higher pH of 6.5, the AE method estimated only about 0.59 g CaCO3 kg-1 soil for every 1.0 g CaCO3 kg-1 soil predicted by incubation. Figure 3-11 also shows that as the incubation LR increased for a particular target pH the AE method estimated even less of the incubation LR. This was also reported by Adams and Evans (1962). The AE method seems to be more precise when soils are to be limed to target pH 6.5. There was not a significant change in r values at the two highest target pH levels. However, the C.V. values decreased as the target pH values increased from 5.5 to 6.5. The decrease in C.V. values reflects a decrease in dispersion of points about the regression line. Fox (1980) found the AE method overestimated CaCO3 incubation LR to pH 6.5 when the CaC03 LR was below 2.4 g CaCO3 kg-1 soil, and underestimated above that amount (r = 0.92). The findings of Fox were similar to those of Tran and van Lierop (1981) who found that the AE method overestimated CaCO3 predicted LR to pH 6.5. AEmod Method and CaCO3 Incubation The AEmod method was much better correlated to LR5.5 (r = 0.94) than the AE method '(Fig. 3-12). This may be due to the lack of range on the x-axis. The AEmod method estimated on



PAGE 1

12 Woodruff (1948) developed a buffer method, suitable for routine testing, that required only one pH reading. The buffer solution consisted of calcium acetate, para-nitrophenol, and magnesium oxide. The titration curve of the buffer was linear from pH 7.0 to 6.0. McLean et al. (1958) observed that the lime requirements of Ohio soils were not met as determined by the Woodruff method. They felt that the Woodruff buffer method for determining lime requirement generally did not take into account the extractable Al. Shoemaker et al. (1961) developed a buffer method (SMP) which consisted of a more dilute mixture of triethanolamine, para-nitrophenol, potassium chromate, and calcium acetate than the Woodruff procedure. They found that the Woodruff and Mehlich buffers were too strong to indicate by pH change the relatively weaker acidity of soils high in extractable Al. The titration curve of this buffer was linear from pH 7.5 to 4.8. Adams and Evans (1962) developed a buffer designed primarily for Red-Yellow Podzolic soils with only small amounts of 2:1 type clays. It had a large buffering capacity relative to the soil but at the same time it was sensitive to the acidity of soils of low exchange capacity. They included para-nitrophenol in a greater concentration than the Woodruff method, and used K-borate instead of Ca-acetate. They employed a pH-base unsaturation (BU) relationship in the



PAGE 1

35 Tran and van Lierop (1982) reported that the SMP DB was not substantially more accurate than the SMP SB for LR 5.5 and 6.0. In a previous paper, Tran and van Lierop (1981) found that the SMP DB did not significantly improve the correlation as compared to the SMP SB. Fox (1980) found the SMP DB overestimated the LR for soils with a LR of <4.12 meq CaCO3 100 g-1 soil, and underestimated the LR for soils with >4.12 meq CaCO3 100 g-1 soil. The SMP SB method performed much better on these soils than the SMP DB method. Ssali and Nuwamanya (1981), reporting on two separate experiments, found the SMP DB method to better predict the mean LR to pH 6.5. For soils with a LR <4 meq 100 g'1 both SMP methods underestimated the predicted LR. The SMP DB provided a better estimate than the SMP SB. The second trial used a CaCO3-soil incubation'to determine the LR to pH 6.0. This time the two SMP methods both overestimated the predicted LR for soils requiring <4 meq 100 g-1. The SMP DB provided a better estimate than did the SMP SB method.



PAGE 1

63 an average of 1.21 g CaCO3 kg-1 soil for every 1.0 g CaCO3 kg-1 soil predicted by incubation. Fox (1980) also found that the Yuan DB method greatly overestimated the CaCO3 LR to pHw 7.0 up to a LR of 3.72 g CaCO3 kg-1 soil. The regression equation found was: Yuan LR = 0.64(incubation LR to pHw 7.0) + 2.68, (r = 0.93). However Tran and Lierop (1981) reported the Yuan DB method only estimated 77% of the incubation LR for soils with a LR <5 g CaCO3 kg-1 soil. They also found that the Yuan method as published by Yuan (1974) only estimated 91% of the LR as shown by his data. McLean (1978) determined that the Yuan DB actually underestimated the soil LR for soils with a LR <2 g CaCO3 kg-1 soil by 11% (r = 0.81, C.V. = 24%). The r values increased and C.V. values decreased for the Yuan DB as the target pH increased to 6.0 (r = 0.93, C.V. = 35%) and 6.5 (r = 0.96, C.V. = 21%), respectively. The Yuan DB estimated an average of 1.21 g CaCO3 for every 1.0 g CaCO3 kg-1 soil found by incubation to LR6.0. For incubation LR6.5, the Yuan DB estimated on the average 1.27 g CaCO3 for every 1.0 g CaCO3 kg-1 soil found by incubation LR6.5. All Three Methods and BaCl2-TEA The AE method was very poorly correlated with the BaC12-TEA method to all three target pH levels of pH 5.5, 6.0, and 6.5 (Table 3-7). As the target pH increased, the AE method was increasingly better correlated with the reference method. The AE method severely underestimated the reference



PAGE 1

CHAPTER I INTRODUCTION Prior to the adoption of the Adams-Evans (AE) lime requirement (LR) method by the University of Florida Extension Soil Testing Laboratory (ESTL), soil lime requirement was estimated by the county extension agents. They used soil pH, texture, extractable Ca and Mg, and organic matter content to estimate the amount of lime needed to accomplish the desired pH change (Rhue and Kidder, 1984). The desire for a quantitative method for determining soil LR had been discussed since at least the early 1970's (Yuan, 1970). Unpublished work at that time had found that the existing LR buffer methods were not suitable for use on Florida soils. Yuan expressed a need for modification of the existing methods, or the development of a new one. In response to the need for a new LR method in Florida, Yuan (1974) created a new buffer employing the double buffer concept. Then Yuan (1975) compared his buffer with three other methods, including among them the AE method. Using the BaC12TEA method as a reference method, he found the Yuan double buffer (DB) method to be preferable over the AE method which greatly underestimated the reference method. Yuan followed up 1



PAGE 1

Page SMP Single Buffer Method........................ 34 SMP Double Buffer Method........................ 34 III INCUBATION STUDY.................................... 36 Introduction........................................ 36 Materials and Methods................................ 37 Results and Discussion............................... 46 pHw and Base Unsaturation..... .................. 46 pHw and pH ..................................... 46 Titration Curves................................ 49 AE Method and CaCO3 Incubation.................. 57 AEmod Method and CaCO3 Incubation............... 59 Yuan DB Method and CaCO3 Incubation............. 61 All Three Methods and BaC12-TEA ................. 63 AE Total Acidity and BaCl2-TEA.................. 66 BaCl2-TEA and CaCO3 Incubation................... 66 Suggested Modifications......................... 68 Summary and Conclusions ............................. 71 IV FIELD STUDY ........................ ................ .74 Introduction............................. ........... 74 Materials and Methods................................. 76 Description of Sites and Soils.................. 76 Experimental Design............................. 80 Land Preparation, Use, and Analyses............. 80 Site 1.................................... 80 Site 2 .................................... 81 Site 3.................................... 81 Results and Discussion................................ 82 Observations of Field LR Over the Length of the Study ....................... ........... .82 Calculation of Field Lime Requirement........... 87 Comparison of Field and Laboratory Data......... 87 Summary ............................................. 91 V CORRELATION STUDY.................................... 92 Introduction........................................ 92 Materials and Methods ............................... 93 Results and Discussion............................... 93 VI SUMMARY AND CONCLUSIONS ............................. 99 REFERENCES....................... ......................... 103 BIOGRAPHICAL SKETCH...................................... .112 vi



PAGE 1

102 incubation study should be employed instead of using the pH-BU relationship of the original AE method. Second, the liming factor should be increased from 1.5 to at least 3. Evidence from the various studies indicated that the liming factor is not constant, but is in fact inversely related to pHw increases in response to liming. Further work is needed to determine the appropriate liming factors that would be applicable under a wide range of conditions. It is suggested that several more liming trials be conducted on farms throughout Florida, especially on soils with relatively low pH values.



PAGE 1

23 a long time for completion, field trials also limit the researcher as to the amount and variety of soils that can be studied. As they began to develop laboratory tests they still used field trials as well as.pot and laboratory incubations as a basis of comparison. With.the advent of more accurate and rapid laboratory methods, some of the older more tedious laboratory methods were used as the basis of comparison or "actual lime requirement." Examples of commonly used reference methods today include the CaCO3-soil incubation method (Webber et al., 1977; Shoemaker et al., 1961) which can last from one month up to 17 months or more. A second reference method used is the BaC12-TEA method by Peech, which is a modification of Mehlich's earlier method, and takes one to two days to complete. The Ca(OH)2 titration method is also commonly employed as reference method. The most recent modification of the method by McLean et al. (1978) takes several days to complete. Although they are not used as frequently, field studies are still very important for correlating laboratory test data with actual field response (Yuan et al., 1977, 1978; Baker and Chae, 1977). Salt Accumulation Although a long term lime-soil incubation appears to be the best method short of field trials, there are some problems associated with it. Microbial activity is intense under incubation conditions where the soil is at or near



PAGE 1

Page Table 4-2. Average pHw by treatment for the three field sites.......................................... 88 Table 4-3. Comparison of various methods of predicting soil LR to pHw 6.0 and 6.5.................... 89 Table 5-1. Selected characteristics of 98 soils used in the correlation study......................... 95 Table 5-2. Regression statistics between AE LR and the BaCl2-TEA reference method.................. .. .96 viii



PAGE 1

LIST OF FIGURES Page Figure 3-1. Location of Alachua (A) and Suwannee (S) Counties, Florida ........................... 39 Figure 3-2. Relationship between pH, and base unsaturation for 567 Florida topsoils (0 -15 cm ) ................................. 48 Figure 3-3. Plot of pHs versus pHw for the 34 soils used in the incubation study..................... 50 Figure 3-4. Titration curves of final pH versus amount of CaCO3 added for soils 7, g, 17, 18, and 20 ....................................... 52 Figure 3-5. Titration curves of final pH versus the amount of CaCO3 added for soils 15, 16, 23, 24 and 27 .................................... 52 Figure 3-6. Titration curves of final pHs versus the amount of CaCO3 added for soils 3, 5, 11, 12, and 14 ................................... 53 Figure 3-7. Titration curves of final pHP versus the amount of CaC03 added for soils 10, 13, 19, 20, and 21................................... 53 Figure 3-8. Titration curves of final pH versus the amount of CaCO3 added for soils 1, 2, 4, 6, and 9 ........................................ 54 Figure 3-9. Titration curves of final pH versus the amount of CaCO3 added for soils 22, 26, 29, and 31 ....................................... 54 Figure 3-10. Titration curves of final pHs versus the amount of CaCO3 added for soils 25, 28, 32, 33, and 34.................................. 55 ix



PAGE 1

24 field capacity and at room temperature. Salts (particularly nitrates of Ca, Mg, and K) accumulate in the soil (McLean, 1982a). These should be leached out or otherwise taken into account lest they depress the soil pH. Dumford (1965) reported that on 38 acid soils of the U.S., with organic matter contents ranging from 0.2 to 6.0%, less salt was present on the average at the end of the incubation period than the equivalent of 0.02M CaCI2.Baker and Chae (1977) found that the pH in CaC12 of their unlimed soils dropped during incubation. They speculated that this was due to organic matter decomposition. Besides leaching or measuring pH in a salt solution some other methods have been tried to overcome the salt problem. Some workers have used greenhouse pot trials planted with a crop (Brown and Cisco, 1984; Loynachan, 1981). The effect of water leaching through the soil as well as plant uptake of nutrients would serve to prevent accumulation of salts. Dunn (1943) used three drops of chloroform, in 100 cc distilled water to 10 g soil suspension, in a four day incubation study to prevent microbial activity. Nommik (1983) added 10 ppm dicyandiamide to a ten week incubation study to suppress nitrification. The organic matter content of the soils ranged from <1.0% up to >30%. An opposite effect of salt accumulation results if the limestone does not completely react with the soil, where pH values can be artificially high. Baker and Chae (1977) found



PAGE 1

55 32 25 6.028 5.04.0 0.0 1.0 2.0 3.0 4.0 5.0 -1 g CaCO3 kg soil Figure 3-10. Titration curves of final pHs versus the amount of CaCO added for soils 25, 28, 32, 33, and 34.



PAGE 1

32 Yuan (1975) found a very high correlation for 44 Florida soils between the Yuan DB method and the BaCl2-TEA method (r = 0.99). Yuan DB method LR estimates of 0.37, 0.85, and 1.80 T/A compared to 0.5, 1.0, and 2.0 T/A for the BaCl2-TEA method, respectively. Fox (1980) evaluated the Yuan DB method on 20 Pennsylvania soils. A better correlation was determined when CaCO3soil incubation (r = 0.967) was used as the reference method than when Ca(OH)2 titration (r = 0.914) was used as the reference method. Using the CaCO3-soil incubation as a reference method, the Yuan DB method overestimated the LR for soils with a LR <7.44 meq CaCO3 100 g-1 soil and underestimated the LR for soils with a LR >7.44 meq CaCO3 100 g-1 soil. Tran and van Lierop (1981) worked with 70 Quebec soils and an eight week CaCO3-soil incubation. The Yuan DB method had a high correlation coefficient (r = 0.959) for soils ranging in LR from 1.4 to 40.0 meq CaCO3 100 g-1 soil to pH 6.5. A poorer correlation was found for soils with a LR <10 meq 100 g-1 soil (r = 0.877). The Yuan DB method, on the average, predicted 66% of the CaCO3 LR(6.5). McLean et al. (1978) reported the Yuan DB method underestimated the-LR for 28 low LR soils (<4.0 meq 100 g-1 soil). The reference method predicted LR was 1.0, 2.0, and 4.0 meq 100 g-1 soil versus 0.54, 1.65, and 3.87 meq 100 g-1 soil as determined by the Yuan DB method. The Yuan DB method



PAGE 1

91 more lime would be required for soils with an already high pH, as compared to soils with a low pHw, to obtain the same increase in pHw.This helped to explain the higher liming factors needed for the soils from sites 1 and 3 as compared to the site 2 soil. Summary The field studies provided evidence that the current AE LR recommendations of the ESTL underestimate the actual field LR. The AE LR method underestimated the field LR even when the arbitrary liming factor of 1.5 was applied. A liming factor of 3 was appropriate for the site 2 soil where the initial pHw was relatively low. A larger liming factor of about 5 was found for site 1 and 3 soils. These soils had a high initial pHw and did not respond as much to liming as the site 2 soil. Incubation LR did underestimate field LR as had been widely reported elsewhere. The use of pHs measurements in the field should minimize seasonal pH fluctuations and allow for more accurate determinations of the effect of liming on field pH.



PAGE 1

66 5.5, 6.0, and 6.5 was used as the reference method. This is most probably because the BaC12-TEA method measures more pH-dependent acidity above pH 7.0. AE Total Acidity and BaC12-TEA AE total acidity (TA), was highly correlated with BaC12-TEA acidity (Fig. 3-14). AE TA underestimated that of the reference method by about half. When.the reference method predicted that the amount of base required to neutralize the measured acidity was 1.0, 2.0, and 3.0 g CaCO3 kg-1 soil, the AE TA estimated only 0.59, 0.46, and 0.42 g CaCO3 kg-1 soil, respectively. This seems to indicate that although the BaCl2-TEA method is not a good measure of the AE LR to pH 5.5, 6.0, and 6.5, the total amount of acidity measured by the AE buffer was highly correlated to the total amount of soil acidity measured by the BaC12-TEA method. Savant and Kibe (1971) reported that the AE buffer measured about 1/3 of the total acidity of the soil determined by BaC12-TEA + 1M KC1 exchangeable acidity on lateritic rice soils (r = 0.77). The exchangeable acidity was almost negligible, constituting less than 3% of the total acidity. BaC12-TEA and CaCO3 Incubation The two reference methods were not as well correlated (Table 3-7) as other researchers have found (McLean et al., 1966; Webber et al., 1977; Fox, 1980). However, as the target pH increased the precision of the relationship increased.



PAGE 1

81 disked in and wheat was planted a few days later by the landowner. The wheat was harvested on May 10, 1984. Soybeans were planted in the first week of July 1984 and harvested on November 9, 1984. The field was left fallow until April 1985 when corn was planted and harvested in July 1985. Periodic soil samples were taken to monitor the soil pH,. Sampling consisted of taking four soil cores to a 15 cm depth from a 3 x 6 meter area in the middle of each plot. The four cores were mixed together so that only one pH, value was obtained per plot. Sampling dates were 11/83, 1/84, 3/84, 5/84, 11/84, and 8/85. Site 2. Site 2 was established on November 26, 1983. Soil samples were taken in accordance with the method mentioned for the previous site. The lime and fertilizer were hand applied and then disked in by the landowner. Wheat was then broadcast by a hand-held spreader and subsequently disked in. The wheat was harvested in early June. At that time the landowner expressed his desire to discontinue with the double-cropping trials. He wished to utilize the land for grazing his livestock on rye. He did allow continued sampling and monitoring of the pH in the field. Samples dates were 11/83, 1/84, 3/84, 5/84, 2/85, and 8/85. Site 3. The plots on this site were established on December 12, 1983. Soil samples were taken according to previously mentioned procedures. The limestone and fertilizer were hand-applied and immediately disked in.



PAGE 1

98 the total acidity measured by the AE and BaC12-TEA methods in the incubation study (Fig. 2-14). The similarity of results from both studies, especially in respect to total acidity measured, gives an indication that the AE method performs equally well on a wider range of Florida soils than were used in the incubation study. In summary, AE LR to pH 5.5, 6.0, and 6.5, was poorly correlated with the BaCl2-TEA reference method. This may be more related to the inapplicability of using the reference method for low target pH values than because of the AE method. When the AE total acidity was regressed with the reference method a very good correlation was found. The relationship was very similar to the one found in the incubation study, and indicates that the AE method provides an accurate index of the lime requirement of the 98 soils studied.



PAGE 1

82 The wheat was harvested in early June and on July 1, 1984 soybeans were planted. The soybeans were harvested on October 30. Wheat was once again planted in December with the same fertilizer additions as the previous season. After the wheat was harvested in May 1985, soybeans were planted in June 1985. As of the last sampling date the soybeans were still in the field. The field was sampled on the following dates; 11/83, 3/84, 5/84, 7/84, 11/84, and 8/85. The soils from sites 1 and 3 were also included in the incubation study reported on in Chapter III. Soil numbers 16 and 18 in the incubation study were taken from sites 1 and 3, respectively. Results and Discussion Observations of Field LR Over the Length of the Study Commonly used target pHw levels in Florida are 5.5, 6.0, and 6.5. For purposes of discussing the field results these target pHw levels will be used as reference points. Evaluation of the field data was complicated by the difficulty in determining actual field LR. The problems associated with determining the actual field LR and the decisions made to deal with these problems are discussed next. First, a decision had to be reached on when the field pHw had reacted sufficiently with the lime so that it could be considered to be at equilibrium. Observations indicated that at all of the sites pHw increased linearly up to six months after liming (Fig. 4-2, 4-3, 4-4). The initial decrease in



PAGE 1

65 SITE 2 4.48 Mg haI 601 C / 2.24 Mg ha-1 55 1.12 Mg ha-1 5|0_ L t0 Mg ha-1 11/83 1/84 3/84 5/84 2/84 8/85 SAMPLING DATE Figure 4-3. The effect of lime treatments on soil pHw, over the time period of the study, at site 2. Each point represents the average of four replications.



PAGE 1

CHAPTER IV FIELD STUDY Introduction Studies have shown that lime-soil reactions behave differently under field conditions than in laboratory incubation studies. In the field, liming materials are normally applied as agricultural limestone and are partially mixed with the soil by disking or plowing. In the laboratory, it is customary to use 320 mesh reagent grade CaCO3 and thoroughly mix it with the soil. The agricultural limestone (ag lime) is usually less effective and slower in neutralizing the soil acidity due both to its larger particle size and to nonuniformity of mixing with the field soil. Thus reaction under field conditions usually takes longer and is less complete than in laboratory incubation studies. So the amount of ag lime needed to accomplish a desired pH change is greater than the laboratory determined LR. The term 'liming factor' refers to the multiplication factor used to account for the difference in neutralization efficiency in the laboratory and in the field. This concept can be expressed by the following simplified equation: LR X LIMING FACTOR = FIELD LR 74



PAGE 1

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a thesis for the degree of Master of Science. Gerald Kider, Chairman Associate Professor of Soil Science I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a thesis for the degree of Master of Science. R. Dean Rhue Associate Professor of Soil Science I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a thesis for the degree of Master of Science. Peter E. Hildebrand Professor of Food and Resource Economics This thesis was submitted to the Graduate Faculty of the College of Agriculture and to the Graduate School and was accepted as partial fulfillment of the requirements for the degree of Master of Science. May 1986 Dean, College of Agriculture Dean for Graduate Studies and Research



PAGE 1

34 precision of the method as it resulted in a lowering of the correlation coefficient. SMP Single Buffer Method The SMP SB LR method has been evaluated by numerous researchers. Most of the researchers report that the SMP SB LR method underestimates the LR for low lime requiring soils (Yuan, 1975; Fox, 1980; Brown and Cisco, 1984). Most of the workers found the SMP SB LR method to be significantly correlated (r > 0.90) with their respective reference methods (Webber et al., 1977; Loynachan, 1981). The buffer exhibits relatively little buffering capacity from an initial buffer pH of 7.5 down to a pH of 6.9 (McLean, 1978). The vertical change in buffer pH in the pH range of'7.5 to 6.9 would be too great to indicate adequate lime for acid soils very low in cation exchange capacity (McLean et al., 1966). McLean et al. (1966) reported that the SMP SB was reasonably accurate for soils requiring >4000 lb of lime per acre. SMP Double Buffer Method McLean et al. (1978) found that through the use of the SMP DB method and regression equations the error of estimate for low LR soils could be significantly improved over the SMP SB method. However, they still recommended that the SMP SB was the most satisfactory compromise between simplicity of determination and reasonable accuracy for soils of a wide range of lime requirement.



PAGE 1

13 calculation of field lime requirement. The titration curve of the buffer was linear from pH 8.0 to 7.0. Mehlich (1976) developed a buffer for the rapid estimation of unbuffered salt-exchangeable acidity and lime requirement. This method was primarily calibrated against exchangeable acidity. The lime recommendations made are based on the neutralization of a portion of the exchangeable acidity rather than the amount needed to attain a target pH. This buffer exhibited linearity from pH 6.6 to 4.0. Nommik (1983) developed a buffer to measure titratable acidity and LR on Swedish soils. The buffer had a linear pH range from pH 7.00 to 5.20. He found the new buffer method was satisfactorily correlated with both the Yuan DB and the SMP methods. About twenty years after Woodruff modified his original buffer method, work supporting the change was finally published (Brown and Cisco, 1984). The new buffer was intended to more accurately reflect the contribution of aluminum. They cautioned that the method should not be adopted on Ultisols and Oxisols until it had been properly calibrated for those soils. The buffer method was designed for acid soils with a lime requirement <10 cmol (+) kg1. Double buffer Methods Yuan (1974) first proposed the double buffer concept. The method was based on two separate measurements in buffer solutions of the same composition but initially adjusted to



PAGE 1

45 Table 3-4. Individual lime rates applied to the 34 incubated soils. ESTL measured pHw and AE buffer pH. LR (lbs acre-1) was determined from published tables and converted to g CaCO3 100 g-1 soil to attain pHw6.5. pHw AE Buffer Soil (1:2) pH LR 1/3LR 2/3LR LR 4/3LR 5/3LR 100 lbs A-1 ----mg CaCO3 100 g-1 soil---1 5.3 7.50 3.3 37 73 110 147 183 2 5.0 7.65 2.6 29 58 87 116 145 3 5.5 7.55 2.7 30 60 90 120 150 4 5.7 7.70 1.5 17 33 50 67 83 5 5.0 7.65 2.6 29 58 87 116 145 6 4.3 7.35 6.0 67 133 200 267 333 7 5.2 7.40 4.2 47 93 140 187 233 8 5.8 7.70 1.4 16 31 47 63 78 9 6.1 7.70 0.9 10 20 30 40 50 10 6.2 7.50 1.5 17 33 50 67 83 11 4.6 7.65 3.0 33 67 100 133 167 12 4.4 7.45 5.1 57 13 170 227 283 13 5.9 7.55 1.9 21 42 63 84 105 14 5.8 7.65 1.6 18 35 53 71 88 15 5.7 7.55 2.3 26 51 77 103 128 16 5.9 7.65 1.5 17 33 50 67 83 17 5.3 7.65 2.3 26 51 77 103 128 18' 5.0 7.50 3.7 41 82 123 164 205 19 5.1 7.65 2.5 28 55 83 111 138 20 5.2 7.80 1.4 16 31 47 63 78 21 4.9 7.85 1.1 12 25 37 49 62 22 5.3 7.40 4.0 44 89 133 177 222 23 5.7 7.75 1.3 14 29 43 57 72 24 4.9 7.70 2.3 26 51 77 103 128 25 4.7 7.65 2.9 32 65 97 129 162 26 4.5 7.80 1.7 19 38 57 76 95 27 5.0 7.50 3.7 41 82 123 164 205 28 4.5 7.15 7.5 83 167 250 333 417 29 4.6 7.70 2.5 28 55 83 111 138 30 5.9 7.80 0.8 9 18 27 36 45 31 4.5 7.35 5.7 63 127 190 253 317 32 5.7 7.60 2.1 23 47 70 93 117 33 6.0 7.45 2.1 23 47 70 93 117 34 5.9 7.45 2.4 27 53 80 107 133



PAGE 1

Table 4-3. Comparison of various methods of predicting soil LR to pHw 6.0 and 6.5. to pHw 6.0 to pHw 6.5 Site IncuAE-F Yuan Current Fielda IncuAE-F Yuan Current Fielda bation DB AE LR Trial bation DB AE LR Trial ---------------------------------g CaCO3 kg-1 soil-----------------------------1 0.0 0.0 0.2 0.0 0.0 0.5 0.3 0.7 0.5 1.6 2 # 0.4 0.7 0.6 1.3 b 0.7 0.9 1.0 2.2c 3 0.0 0.0 0.3 0.0 0.0 0.4 0.4 0.8 0.6 2.0 a Lime requirement determined from regression equations b Soil was not used in the incubation study c Extrapolated beyond range of data



PAGE 1

49 The following relationship was calculated for data reported from 567 Florida soils: pHs = -0.488 + 0.954(pH,) (r = 0.90, n = 567). A very similar relationship was found between pHs and pHw for the 34 soils used in the study: pH = -0.676 + 0.974(pH,) (Fig. 3-3). These two equations predict an average difference of 0.8 pH between pHs and pHw, in the pH range of 4.0 to 6.5. An assumption is then made that most of these sandy soils under normal field conditions will have a pHs that is about 0.8 pH less than pHw.By subtracting 0.8 pH from target pHw values of 5.5, 6.0, and 6.5, the corresponding target pHs values of 4.7, 5.2, and 5.7 were obtained. Titration Curves Twenty-eight of the soils experienced a drop in pHw of the blank samples (no lime added) during incubation. Fifteen soils showed a decrease of at least 0.40 pH. Only one soil actually attained pHw 7.0. Evaluation of pHs values for the blank samples showed much less deviation from the original pHs.The mean pHw for blank samples for the 34 soils was 5.29 before incubation and 5.03 after incubation. The mean pHs for the blank samples was 4.47 prior to incubation and 4.49 after incubation. Dumford (1965) found that the average pH of 38 acid soils rose from 5.84 to 6.33 when they were leached with pure water, apparently due to the leaching out of salts, and then fell to 5.76 with the additions of 0.01M CaCI2.The lower



PAGE 1

30 of the various methods. The Adams-Evans method and the SMP SB method are included because they are currently being used by certain organizations in Florida. The Yuan DB is included because it was developed specifically for Florida soils. Since the SMP DB is reportedly an improvement of the SMP SB method it was also included in the review. The reference methods, target pH levels, and the soils used by the workers differed from experiment to experiment, making it difficult to compare results between experiments. However, general observations can be made and these are presented next. Adams-Evans Buffer Method Adams and Evans (1962) data showed that the AE method tended to underestimate the LR of three soils when compared to CaCO3-soil incubation. The difference between AE estimated and incubation predicted LR decreased as the LR decreased. Although McLean et al. (1966) did not publish data for the AE method, they did evaluate it. They mentioned that the AE method predicted less lime than the SMP SB method when CaCO3-soil incubation was used as the reference method. It can then be inferred from the published data that the AE LR underestimated the LR as predicted by the CaCO3-soil incubation. Fox (1980) reported a high correlation (r = 0.919) for the AE versus CaCO3-soil incubation. The AE overestimated the soil LR for soils requiring <4.87 meq CaCO3100 g-1soil and



PAGE 1

39 FLORIDA Figure 3-1. Location of Alachua (A) and Suwannee (S) Counties, Florida.



PAGE 1

CHAPTER III INCUBATION STUDY Introduction Rapid soil test methods for the determination of lime requirement provide indexes of the actual soil lime requirements that are found from field studies. Several factors influence the incomplete measurement of soil lime requirement not the least of which is the short time of soil-buffer contact. The first step in evaluating the suitability of a rapid soil test method to a particular group of soils is through lab correlation studies. In lab correlation studies the various rapid soil-test methods being evaluated are compared to one or more reference methods. The reference methods, which usually take a relatively long time to complete, are assumed to give a more accurate indication of the total amount of lime needed to attain a particular pH level. The ability of a soil-test method to be an accurate index of the reference method is then determined by regressing the results of the two methods. In this study, a commonly used reference method, soil-CaCO3 incubation, was used to evaluate both the AE and Yuan DB LR methods. The following objectives were sought to be determined from this study: 36



PAGE 1

106 Loynachan, T. E. 1981. Lime requirement methods for cold-region soils. Soil Sci. Soc. Am. J. 45:75-80. MacIntire, W. H. 1920. Report on the lime requirement of soils. J. Assoc. Off. Agric. Chem. 4:108-123. .MacIntire, W. H. 1915. A method for the determination of the immediate lime requirements of soils. J. Ind. Eng. Chem. 7:864-867. MacIntire, W. H., L. G. Willis, and J. I. Hardy. 1917. Lime requirement methods of soils. J. Assoc. Off. Agric. Chem. 3:144. Magdoff, F. R., and R. J. Bartlett. 1985. Soil pH buffering revisited. Soil Sci. Soc. Am. Proc. 49:145-148. McLean, E. O. 1976. Chemistry of soil aluminum. Commun. in Soil Sci. and Plant Anal. 7:619-636. McLean, E. 0. 1978. Principles underlying the practice of determining lime requirements of acid soils by use of buffer methods. Commun. in Soil Science and Plant Anal. 9:699-715. McLean, E. O. 1980. Testing soils for pH and lime requirement, p.77-95. In L. M. Walsh and J. D. Beaton (ed.). Soil Testing and Plant Analysis. Soil Sci Soc. of Am,. Madison, Wis. McLean, E. O. 1982a. Soil pH and lime requirement. In A. L. Page, R. H. Miller, and D. R. Keeney (ed.). Methods of Soil Analysis. Part 2. Agronomy 9:199-224. Am. Soc. of Agron., Madison, Wis. McLean, E. O. 1982b. Chemical equilibrations with soil buffer systems as bases for future soil testing programs. Commun. Soil Sci. Plant Anal. 13:411-433. McLean, E. 0., and J. R. Brown. 1984. Crop response to lime in the midwestern United States. In F. Adams (ed.). Soil Acidity and Liming. Agronomy 12:267-303. Am. Soc. of Agron., Madison, Wis. McLean, E. 0., S. W. Dumford, and F. Coronel. 1966. A comparison of several methods of determining lime requirement of soils. Soil Sci. Soc. Am. Proc. 30:26-30. McLean, E. 0., D. J. Eckert, G. Y. Reddy, and J. F. Trierweiler. 1978. An improved SMP lime requirement



PAGE 1

9 10 Figure 5-1. The location of the 98 soils used in the correlation study. Each number represents the amount of samples originating from the respective county.



PAGE 1

10 (Thomas-and Hargrove, 1984). They defined a 100% base saturated soil as one which had reached equilibrium with a surplus of CaCO3 at the partial pressure of CO2 existing in the. atmosphere and at a temperature of 25 C. They determined that .the pH of a base saturated soil would be approximately 8.2. Buffer Methods Schofield (1933) proposed a soil-buffer method which required two titrations. A lime and para-nitrophenol solution was added to a soil sample and allowed to equilibrate for at least 16 hours. Both the original solution and soil-solution were titrated. The difference in cubic centimeters between the two titrations was equal to the milligram equivalents of lime taken up per 100 g of soil. Mehlich (1938) reported that rapid and accurate analyses were needed in order to measure exchangeable H+, baseexchange reactions, and lime requirement on large numbers of soil samples. He proposed the triethanolamine acetate-barium hydroxide buffer at pH 8.15 for this. The Ba2+ would exchange the H+ on the soil colloids and the hydroxide would neutralize the resultant acidity. The base exchange could be measured by titration of the original buffer and soil-buffer solution and subtracting the two. Titratable acidity was determined electrometrically by titration with 0.2M HC1 to approximately pH 6.0. The difference between the titration



PAGE 1

17 1/ Al3+ + H20 <--> A(OH)2+ + H 2/ Al(OH)2+ + H20 <--> Al(OH)2+ + H 3/ Al(OH)2+ H20 <--> A1(OH)3 + H The Al3+ ion is predominant below pH 4.7, Al(OH)2+ between pH 4.7 and pH 6.5, and Al(OH)3 between pH 6.5 and pH 8.0 (Bohn et al., 1979). The reaction products of Al hydrolysis may remain in soil solution, be adsorbed as monomers to CEC sites of the soil, be adsorbed and then polymerized on surfaces of clay minerals, or be adsorbed and then complexed by organic matter (McLean, 1976). The H+ ions resulting from hydrolysis lower the pH of the soil solution and react with soil minerals to further break them down. The presence of Al and Fe on organic matter exchange sites causes the organic matter to exhibit a greater weakness as an acid. This results in less contribution to the CEC determined by extraction with a neutral salt solution, or effective cation exchange capacity (ECEC), of the soil, especially at low pH values (Hargrove and Thomas, 1984). Al forms rather stable complexes with the soil organic matter by reaction primarily with carboxyl groups and to a lesser extent with phenolic hydroxyl groups (Hargrove and Thomas, 1984). The amount of complex formed is dependent on the pH and the A13+ concentration in the soil solution (Hargrove and Thomas, 1984). Schnitzer and Skinner (1963) reported that Al was predominantly hydroxylated in organic matter as Al(OH)2+



PAGE 1

27 been due to the ability of the stronger BaCl2-TEA buffer's ability to react with more soil acidity than the AE buffer. Shoemaker et al. (1961) reasoned that buffers weaker than BaCl2-TEA would be expected to extract less soil acidity. Field Calibration Recommendations based on laboratory analyses are usually multiplied by a 'liming factor' when making field liming recommendations. The liming factor accounts for the decreased effectiveness of agricultural limestone applied in the field versus the finely ground CaCO3 used in laboratory and pot studies. The reduced effectiveness is due to the larger particle size of agricultural limestone and incomplete-mixing of lime in the field. The Adams and Evans method incorporates a liming factor of 1.5 based on previous data of Pierre and Worley (1928) and Schollenberger and Salter (1943). Thomas and Hargrove (1984) stated that since in practice, lime applications cannot be made that precisely, inaccuracies in the method are not likely to cause major problems. Adams (1984) supported the previous point of view, feeling that a high degree of precision in methods is usually wasted because of problems encountered in the field application of lime. These problems include soil variation, lime spreading irregularities, lime quality, and incomplete mixing of the lime with soil. McLean (1982b) countered with his argument that the optionof taking some additional simple steps with the double buffer method after taking the reading for the



PAGE 1

68 The BaCl2-TEA acidity overestimated the incubation LR at all three target pH levels. This difference was most pronounced at the lower target pH level but even for LR6.5 the BaCl2-TEA method estimated about twice as much acidity as the incubation method. The overestimation of CaCO3 incubation LR to pH 6.5 by the BaCl2-TEA method was expected. Similar results have been reported by several researchers comparing lime-incubation methods to BaCl2-TEA extractable acidity. Savant and Kibe (1971) in the study mentioned earlier, found that Ca(OH)2 titration to pH 6.5 to 7.0 (Puri, 1963), estimated about 57% of the total acidity determined mostly from BaCl2-TEA extractable acidity. Loynachan (1981) used BaCl2-TEA adjusted to 80% of the theoretical LR as proposed by Peech (1965). Loynachan found that the BaCl2-TEA method method overestimated the CaCO3 incubation LR to pH 6.5 by over 4 times for soils with high amounts of organic matter. The slope of 0.68 determined for this relationship was similar to the slope found in this study. Suggested Modifications The underestimation by the AE LR method can be corrected for in one of three ways. The first one would be to calculate the AE LR the same as is currently done and then use the regression equations (Fig. 3-11) to adjust for the underestimation of the AE LR method. The use of regression



PAGE 1

26 Brown and Cisco (1984) compared the Ca(OH)2-CaC12 titration of Benham (1970) to a CaCO3 incubation-cropping greenhouse method. The LR by the CaCO3 incubation-cropping method overestimated Ca(OH)2-CaC12 LR values of 2, 4, and 8 cmol (+) kg-1, by 194%, 136%, and 119%, respectively. Effectiveness of Buffers in Measuring Total Soil Acidity Total soil acidity is usually defined as the amount of acidity that must be neutralized to attain a pH at or near 8.2. This is the maximum pH attainable with CaCO3 in the presence of air having a CO2 content of 0.03% (Bradfield and Allison, 1933). The BaCl2-TEA method has been regarded as the rapid method that most closely approximates the titration method of Bradfield and Allison (Thomas, 1982). However, Shoemaker et al. (1961) found that the BaCl2-TEA buffer did not react with all of the extractable Al. Since Al is an important component of soil buffering, through Al hydrolysis, this means that the BaC12-TEA buffer does not react with all of the acidity present in the soil. Adams and Evans (1962) reported that their buffer measured slightly more acidity than was measured by 1M NH40Ac (pH 7.0) extraction. Acidity, by the latter method, was calculated by subtracting total bases from the NH40Ac measured CEC. The AE buffer measured less acidity than the BaCl2-TEA (pH 8.2) method for Florida soils (Yuan, 1974). This may have



PAGE 1

7 Soil-lime Titrations The years 1897 to 1920 were the pioneering era of the quantitative determination of lime requirement. The earliest methods of determining soil acidity were based upon the reaction of soil with excess CaCO3.The oldest of these methods is that of Tacke (1897) which consisted of suspending an excess of CaCO3 with the soil and aspirating the evolved CO2 three hours at room temperature. Shaw (1953) reported that, in 1900, Wheeler and his co-workers investigated the possibility of utilizing the evolved CO2, from the reaction of soil with a suspension of CaCO3 at boiling temperature, as a measure of lime requirement. They were unable, though, to determine a reasonable time limit within which the elimination of CO2 would be ended and observed. Veitch (1902) developed a lime-water method based on the Tacke procedure. The Veitch method consisted of a series of CaO equilibrations with the soil followed by boiling. The smallest amount of lime-water that gave the characteristic pink color, in the presence of phenolphthalein, was taken as the acidity equivalent of the soil. This method was designed to bring the soil pH up to neutrality because Veitch felt that an alkaline soil pH was necessary for optimum plant performance. The Veitch method was considered the most reliable for the first two decades of the 20th century, but at the same time was plagued by poor reproducibility and considered as too laborious (Shaw, 1953).



PAGE 1

3.0 a) 30r b) Sy = -0.18 + 1.33x ya = -0.03 + 1.22x .M (r=0.91, n=15, CV.=46%) a (r=0.93, n=31, C.V.=35%) so on O U l / u 1.0 -., 1.010o c 0 s % 1.0 1.0 2.0 3.0 INCUBATION LR 5.5 (g CaCO3 kg soil) INCUBATION LR 6.0 (g CaCO3 kg-1 soil) 3.00 c) OOLO 10 2.0 -3.0 INCUBATION LR 6.5 (g CaCO3 kg soil) Figure 3-13. Regression curves and statistics of Yuan DB versus incubation LR. a) to pH 5.5; b) to pHl 6.0; c) to pH 6.5.



PAGE 1

S3D a) 7 + 0 x b) y = 0.01 + 0.58x (A -(r=0.97, n=29, C.V.=19%) (r=0.94, n=19, CJ.=25%)r 0 20 20 I 00 S1.0 1D 0 a a o t 10 OD 1.0 20 3.0 0.0 10 3.0 INCUBATION LR 5.5 (g CaCO kg soil) INCUBATION LR 6.0 (g CaCO3 kg1 soil) -,'3Dc) S y = 0.09 + 0.54x (r=0.98, n=34, CV.=12%) S1.0 b) to pH 6.0; c) to pH 6.5.



PAGE 1

104 Carlisle, V. W., R. E. Caldwell, F. Sodek, III, L. C. Hammond, F. G. Calhoun, M. A. Granger, and H. L. Breland. 1978. Characterization data for selected Florida soils. Soil Science Research Report No. 78-1. University of Florida, Gainesville. Carlisle, V. W., M. E. Collins, F. Sodek, III, and L. C. Hammond. 1985. Characterization data for selected Florida soils. Soil Science Research Report No. 85-1. University of Florida, Gainesville. Carlisle, V. W., C. T. Hallmark, F. Sodek, III, R. E. Caldwell, L. C. Hammond, and V. E. Berkheiser. 1981. Characterization data for selected Florida soils. Soil Science Research Report No. 81-1. University of Florida, Gainesville. Christensen, H. R., and S. T. Jensen. 1926. On the quantitative determination of the lime requirement of the soil. Trans. Comm. Int. Soc. Soil Sci. A:94-115. Coleman, N T., S. B. Weed, and R. J. McCracken. 1959. Cationexchange capacity and exchangeable cations in Piedmont soils of North Carolina. Soil Sci. Soc. Proc. 23:146-149. Corey, R. B. 1973. Factors affecting the availability of nutrients to plants, p. 23-33. In L. M. Walsh and J. D. Beaton (ed.) Soil Testing and Plant Analysis. Soil Sci. Soc. Am., Madison, Wis. Dewan, H. C., and C. I. Rich. 1970. Titration of acid soils. Soil Sci. Soc. Am. Proc. 34:38-44. Dumford, S. W. 1965. An evaluation of several buffer techniques for determining the lime requirements of representative acid soils of the U.S.A. Unpublished M.S. thesis. Ohio State Univ. Dunn, L.E. 1943. Lime-requirement determination of soils by means of titration curves. Soil Sci. 56:341-351. Fiskell, J. G. A., and V. W. Carlisle. 1963. Weathering of some Florida soils. Soil and Crop Sci. Soc. Fla. Proc. 23:32-44. Fox, R. H. 1980. Comparison of several lime requirement methods for agricultural soils in Pennsylvania. Commun. in Soil Sci. and Plant Anal. 11(1):57-69. Gillman, G. P., R. C. Bruce, B. G. Davey, J. M. Kimble, P. L. Searle, and J. O. Sklemstad. 1983. A comparison of



PAGE 1

61 the average about 0.57 g CaCO3 kg-1 soil for every 1.0 g CaCO3 kg-1 soil predicted by incubation LR5.5. The AEmod method was also highly correlated with incubation LR6.0 and LR6.5 (r = 0.97, r = 0.98). However, the AEmod method also underestimated the incubation LR. For soils that were to be limed to pH 6.0 the AEmod method estimated an average of 0.59 g CaCO3 kg-1 soil for every 1.0 g CaCO3 kg-1 soil predicted by incubation. For soils that were to be limed to pH 6.5 the AEmod method estimated an average of 0.60 g CaCO3 kg-1 soil for every 1.0 g CaCO3 kg-1 soil predicted by incubation. Thus there appears to be no practical advantage in using the AEmod method versus the AE method. Yuan DB Method and CaCO3 Incubation The Yuan DB method did not perform as well as had been expected. Yuan (1974, 1975, 1976) showed that the method performed well when compared to BaCl2-TEA (pH = 8.2) acidity, Ca(OH)2 incubation to pH 7.0, and CaC03 incubation to various pH levels, respectively as reference methods. The Yuan DB overestimated incubation LR at all three target pH levels (Fig. 3-13). Although the Yuan DB method had a high r value (r = 0.91) for LR5.5, the method also had a relatively high coefficient of variation (C.V. = 46%). Comparing Fig. 3-13 and 3-11 it is obvious that the spread of points about the regression line for the Yuan DB is much greater than for the AE method. The Yuan DB method estimated



PAGE 1

CHAPTER V CORRELATION STUDY OF 98 SOILS Introduction The incubation study presented in Chapter III used 34 soils of relatively low organic matter and clay contents. These soils were accepted as representative of many agricultural soils of Florida. However the range of soil properties of Florida's soils extends beyond low organic matter and clay contents. For example, Ultisols of the panhandle contain fairly high clay contents. Histosols, found mostly in south Florida, have very high organic matter contents. The objective of this study was to evaluate the AE LR method on a wider range of Florida soils than was used in the incubation study. A reference method that is less timeconsuming than soil-CaCO3 incubation is BaCl2-TEA extractable acidity. Several researchers have used this method to evaluate more rapid lime requirement buffer methods (McLean et al., 1966; Yuan, 1974; Webber et al. 1977, Fox, 1980). Thus the BaC12-TEA method was used as the reference method to evaluate the AE buffer method on these soils. 92



PAGE 1

38 would allow the expression of desired pH levels in terms of pHs values that correspond to commonly used target pH, levels of 5.5, 6.0, and 6.5. This calculation was done to facilitate interpretation of the incubation data. For the incubation study, thirty-four surface horizon (0 to 15 cm) soils (13 Entisols, 6 Spodosols, and 15 Ultisols) were sampled from Alachua and Suwannee Counties (Fig. 3-1). The soils were characterized by the use of the Alachua and Suwannee County soil surveys (Soil Survey Staff, 1985, and 1965) (Table 3-1). Sampling areas were chosen to represent cultivated, previously cultivated, and native vegetation sites. All soils were relatively low in organic matter and clay contents and thus were representative of many agricultural soils of Florida. The soils were air dried and passed through a 2 mm sieve. Soil pHw was determined in triplicate using both 10 ml soil to 10 ml water (1:1 ratio) and 10 ml soil to 20 ml water (1:2 ratio). Soil pHs was also determined on the same sample in 0.01M CaC12 using 10 ml soil to 20 ml 0.01M CaC12 (1:2 ratio) (McLean, 1982a). Particle size analysis by the pipette method, organic carbon by acid dichromate digestion, 1M KC1 exchangeable Al, and BaCl2-TEA (pH 8.2) extractable acidity were also determined (Soil Survey Staff, 1972). Percent organic matter was calculated by multiplying percent organic carbon by 1.724. Mehlich I extractable P, K, Ca, Mg, and Al



PAGE 1

28 single buffer method should be weighed against the increased accuracy of measurement, especially for soils of low lime requirement. Calculation of Adams-Evans Lime Reauirement A brief explanation of the calculation of the AE LR will be presented. A more detailed example of the calculation is given by Rhue and Kidder (1984). Determination of the AE LR employs two major steps, one a laboratory measurement and the other a set of calculations. In the laboratory step a known amount of buffer is added to a known quantity of soil. The buffer is formulated with a beginning pH of 8.00 and has the property of a linear decrease in pH between 8.00 and 7.00. Thus the amount ofacidity neutralized by the buffer is determined by measuring the soil-buffer equilibrium pH. The amount of acidity neutralized by the buffer will be herein referred to as the AE total acidity (TA). The calculation step of AE LR involves determining the percent of AE TA that must be neutralized to attain a desired soil-water pH (pH,). This step is necessary because the soil buffer equilibrium pH will be higher than the pHw normally desired for crops. The buffer theoretically neutralizes more pH-dependent soil acidity at the higher soil-buffer pH (between 8.0 and 7.0) than would need to be neutralized at the lower pHwvalue desired for plants. This is especially



PAGE 1

107 method incorporating double-buffer and quick-test features. Soil Sci. Soc. Am. J. 42:311-316. McLean, E. O., M. R. Heddleson, R. J. Bartlett, and N. Holowaychuk. 1958. Aluminum in soils:I. Extraction methods and magnitudes in clays and Ohio soils. Soil Sci. Soc. Am. Proc. 22:382-387. McLean, E. O., H. E. Shoemaker, and W. R. Hourigan. 1960. Some effects of aluminum on lime requirement tests of soils. Int. Congr. Soil Sci. Trans. 7th (Madison, Wis.). p.142-151. Mehlich, A. 1938. Use of triethanolamine acetate-barium hydroxide buffer for the determination of some base exchange properties and lime requirement of soil. Soil Sci. Soc. Am. Proc. 3:162-166. Mehlich, A. 1942a. Rapid estimation of base-exchange properties of soil. Soil Sci. 53:1-14. Mehlich, A. 1942b. Base unsaturation and pH in relation to soil type. Soil Sci. Soc. Am. Proc. (1941). 6:150-156. Mehlich, A. 1976. New buffer pH method for rapid estimation of exchangeable acidity and lime requirement of soils. Commun. Soil Sci. Plant Anal. 7:637-652. Nommik, H. 1983. A modified procedure for rapid determination of titratable acidity and lime requirement in soils. Acta Agriculturae Scandinavica. 33:337-348. Peech, M. 1939. Chemical studies on soils from Florida citrus groves. Fla. Agric. Exp. Stn. Bull. 340. Peech, M. 1965. Lime requirement. In C. A. Black (ed.). Methods of Soil Analysis. Part 2. Agronomy 9:927-932. Am. Soc. Agron., Madison, Wis. Pierre, W. H., and S. L. Worley. 1928. The buffer method and the determination of exchangeable hydrogen for estimating the amounts of lime required to bring soils to definite pH values. Soil Sci. 26:363-375. Puri, A. N. 1963. Standardization of methods of soil analysis. Part I. Introductory and physico-chemical constants of Indian soils. Field Res. Stn. Bull. Series A. No. 2. Rhue, R. D., and G. Kidder. 1983. Procedures used by the IFAS extension soil testing laboratory and interpretation of



PAGE 1

BIOGRAPHICAL SKETCH Thomas S. Dierolf was born in Frankfurt, West Germany, on March 16, 1957. One of the first years of Tom's life was spent with relatives on his mother's side in East Germany. In December 1960, Tom immigrated to the U.S. with his parents. Tom spent his youth growing up in Montgomery, Illinois, a far western suburb of Chicago. Tom enrolled into Southern Illinois University in August 1975. Upon receiving his B.S. degree in plant and soil science in May 1980, Tom decided to join the job market prior to continuing his studies. Although he never expected to end up so far away from home, Tom went to Malaysia as a Peace Corps Volunteer in July 1980. The 27 months spent in Malaysia broadened Tom's outlook on life, and it was there that he decided to pursue a career in international agricultural development. Tom took the long way home from Malaysia and spent six months on the road travelling in Southeast Asia and the Pacific. Running low on funds and realizing that working in a grape vineyard was no way to see Australia, Tom accepted an offer of an assistantship at the University of Florida and began classes in August 1983. 112



PAGE 1

19 The major reason that soils exhibit different buffering capacities is that the type of base exchange material influences the pH-BS relationship. Mehlich (1942b) using BaCl2-TEA as the extractant, found kaolinite, which acts as a weak acid, to be only 65% BS at pH 7.0, whereas montmorillonite was 95% BS at pH 7.0. The BS values for organic matter were between the values for kaolinite and montmorillonite for almost the entire pH range. Base saturation generally increases with an increase in soil pH for a group of soils. Peech (1939) reported on the pH-BS relationship for Florida citrus soils. He determined CEC by pH 7.0 1M NH40Ac. Base saturation was determined by subtracting exchangeable bases from the CEC and dividing that by the CEC. He used a nonlinear relationship to show that soils were well buffered between pH 5.0 to 6.5. Base saturation was 25% at pH 5.0 and 90% at pH 6.5. Recent literature has shown that the pH-BS relationship is a linear one (Loynachan, 1981; Magdoff and Bartlett, 1985). In all of the pH-BS relationships shown, there is a wide scatter of points about the regression line, regardless of the shape of the regression line. Difficulties have been encountered when applying this relationship to very poorly buffered soils such as those commonly found in Florida (Yuan, 1974). Use of the pH-BS curve implies that all of the soils used have a titration curve similar to each other as well as to



PAGE 1

TABLE OF CONTENTS Page ACKNOWLEDGMENTS........................................... ii LIST OF TABLES ........................................... vii LIST OF FIGURES.......................................... ix ABSTRACT ................................................. xi CHAPTERS I INTRODUCTION........................................ 1 II REVIEW OF THE LITERATURE ............................ 4 Introduction........................................ 4 History of Lime Requirement Methods................. 5 Liming Practices................................ 5 Litmus Test ..................................... 6 Soil-lime Titrations............................ 7 Soil-lime Potentiometric Titrations............ 8 Buffer Methods ................................. 10 Double-buffer Methods........................... 13 Soil Acidity........................................ 14 Forms of Soil Acidity........................... 15 Buffering Capacity.............................. 16 Role of Al in Soil pH Buffering................ 16 Base Saturation................................ 18 pH-BU Relationship............................. 18 Measurement of Base Saturation................. 20 Characteristics of Florida Soils .............. 21 Methods of Testing Lime Requirement Methods.......... 22 Reference Methods............................... 22 Salt Accumulation. ............................ .23 Comparison of Reference Methods................ 25 Effectiveness of Buffers in Measuring Total Soil Acidity.................................. 26 Field Calibration.............................. 27 Calculation of Adams-Evans Lime Requirement.... 28 Previous Evaluations of Lime Requirement Methods.... 29 Adams-Evans Buffer Method....................... 30 Yuan Double Buffer Method....................... 31 v



PAGE 1

Table 3-3. Mean weights and standard deviations of soil contained in an 11 ml scoop. An excess amount of soil was scooped into the container, the side of the scoop was gently tapped three times, the soil was leveled off, and weighed. Mean of standard Mean of standard Soil 4 weighings deviation Soil 4 weighings deviation -----g 11 cm-3 ---------g 11 cm-3---1 14.77 0.43 18 12.88 0.16 2 15.14 0.17 19 13.94 0.13 3 14.96 0.26 20 14.67 0.12 4 15.97 0.14 21 15.23 0.04 5 14.83 0.12 22 13.20 0.13 6 14.24 0.19 23 15.80 0.09 7 14.43 0.17 24 14.60 0.27 8 13.97 0.15 25 14.65 0.17 9 14.55 0.16 26 14.28 0.23 10 15.38 0.14 27 13.96 0.20 11 13.23 0.63 28 12.70 0.19 12 13.32 0.10 29 15.33 0.12 13 14.86 0.23 30 15.68 0.17 14 15.36 0.13 31 14.23 0.24 15 14.89 0.09 32 13.64 0.12 16 15.54 0.12 33 13.81 0.09 17 13.92 0.22 34 14.16 0.12 Overall mean of 34 soils = 14.47 g standard deviation = 0.83 g range of means = 12.70 -15.97 g



PAGE 1

50 6.0 y = -0.676 + 0.974x (r = 0.99, n = 34) 5.0S0 4.0* * 3.0 4.0 -* * 4.0 5.0 6.0 PH, Figure 3-3. Plot of pHs versus pHw for the 34 soils used in the incubation study.



PAGE 1

3.0 -l o AE TA = 0.25 + 0.34(BaCl -TEA) 2 S2(r = 0.97, n = 34, C. =10%) --a O I I I W 0 1.0 20 30 4D 5.0 BaC12-TEA EXTRACTABLE ACIDITY (g CaCO3 kg-1 soil) Figure 3-14. The regression equation and statistics between the total amount of acidity measured by the AE buffer (TA) and BaCl -TEA extractable acidity.



PAGE 1

ACKNOWLEDGMENTS As with most accomplishments during one's lifetime, a work, such as this one, is possible only through the cooperation of many others besides the one whose sole name appears on the cover. It would not be feasible to thank everyone individually as the length of this thesis may then be doubled. Many thanks are extended to my major advisor Dr. G. Kidder, for always finding the time to work out any problems I encountered while I was at the University of Florida. He was especially helpful during the dreaded thesis writing stage. I also wish to thank the rest of the members of my supervisory committee, Dr. Dean Rhue and Dr. Peter Hildebrand, for reviewing the thesis and offering their helpful suggestions on improving the manuscript. I am grateful to Dr. Dean Rhue, Dr. Luther Hammond, the Soil Characterization Lab, and the Extension Soil Testing Lab for allowing me to make use of their already limited laboratory space and facilities. Thanks are also extended to all of the lab technicians who lent me a hand and put up with my usurpation of portions of the labs. Special thanks are offered to Bill Reve and Ed Hopwood for the extra consideration they gave me. iii



PAGE 1

18 Base Saturation The base saturation (BS) of a soil can be expressed in several ways. There are several factors which can influence the reported BS of a soil. Base saturation and its relationship to soil pH plays an integral part in the determination of lime requirement by the Adams-Evans lime requirement method. There are several factors that influence the measure of base saturation itself as well as the use of the soil pH-base saturation relationship that detract from its usefulness in lime requirement methods. Percent base saturation is a measure of the amount of exchangeable bases, mostly Ca, Mg, K, and Na, that occupy the cation exchange sites of a soil. The other sites are occupied by acidic cations, mostly Al and H. Base unsaturation (BU) is simply the complement of BS. pH-BU Relationship Adams and Evans (1962) used a BU (exchangeable acidity divided by exchangeable bases plus exchangeable acidity) versus pH in water relationship from 348 red-yellow Podzolic soils as a basis of their method. The relationship was used to describe the general buffering capacities of a group of soils. Even though they employed this generalized relationship, they also stated that since no constant relationship existed between soil pH and BU for all soils, pH was considered to be a measure of BS only for a particular soil.



PAGE 1

7.y = 7.71 -0.19x (r = 0.72, n = 31, CV. = 27%) 0C o o * ,o 8.0 I 1.0 20 3D -i INCUBATION LR (g CaCO3 'k, soil) Figure 3-15. The regression equation and statistics between the AE soil-buffer equilibrium pH and the incubation LR.



PAGE 1

2 this work with further field studies (Yuan et al., 1977; Yuan et al., 1978). When the decision to adopt a new LR method for routine use by the ESTL was made, the AE method was chosen over the Yuan DB method. Reasons included were that the AE method had been developed for low cation exchange capacity (CEC) soils such as those.which are commonly found in Florida. The neighboring states of Alabama and Georgia had also adopted the method thus providing for some interstate collaboration. Finally the AE method used less buffer and required less determinations than the Yuan DB method, making the AE method less expensive and more rapid then the Yuan DB method. Although Yuan (1975) examined the AE method he did not give the method the full attention needed if a method is to be adopted. Since the AE method was adopted by the Extension State Soil Testing Laboratory a critical examination of the buffer has not been performed. Prior to the adoption of a particular soil test method, in this case the LR method, several methods are usually evaluated in laboratory and/or greenhouse studies. These are also known as correlation studies. A representative and large number of soils for a region can be evaluated. This process allows for the selection of the method that provides the best index of the soil LR. The next step is calibration. Here field trials are conducted to relate the laboratory values to actual field



PAGE 1

86 soil pHw at site 1 (Fig. 4-2) may have been a result of the application of ammonium fertilizer. Although Yuan et al. (1978) observed continued pH, increases 2 years after liming, it was decided to include all pHw measurements taken after the six month equilibration period in the determination of a representative field pHw.The major reason that this decision was made is that a farmer is more concerned with the pHw values attained one or two years after liming and not the pHw that is theoretically obtainable. Another reason is that the further increases in pHw after the initial six month period are probably relatively minor compared to the increases observed during the initial period. Once a decision was reached on what the actual time period and respective pHw measurements were to be used, another problem had to be dealt with. More specifically, the noticeable fluctuation in pHw values after the initial six month period made determining actual pH, levels difficult. For example, Fig. 4-3 shows a variation of pHw for the no lime added treatment of about 0.5 pH. The wide pHw fluctuations observed for the above three trials is a well documented occurrence. Shuman et al. (1983) cite several references that have reported these observations. Hester and Shelton (1933) concluded that these seasonal fluctuations are attributed to a combination of rainfall, fertilization, and nitrification.



PAGE 1

88 Table 4-2.Average pHw by treatment for the three field sites. Site Treatment Averagea PH, Mg ag lime ha-1 1 0 6.18 1.12 6.31 2.24 6.34 4.48 6.60 2 0 5.27 1.12 5.71 2.24 5.88 4.48 6.35 3 0 6.08 1.12 6.34 2.24 6.33 4.48 6.49 a Average of three sampling dates for sites 1 and 2. Average of four sampling dates for site 3.



PAGE 1

110 Walker, T. W. 1952. The estimation of the lime requirements of soils. J. Soil Sci. 3:261-276. Webber, .M. D., P. B. Hoyt, M. Nyborg, and Diane Corneau. 1977. A comparison of lime requirement methods for acid Canadian soils. Canadian J. of Soil Sci. 57:361-370. Wheeler, H. J., and G. M. Tucker. 1896. Further observations upon the growth of barley upon an acid upland soil. Limed and unlimed. Rhode Island Agric. Exp. Stn. Annual Rep. 9:279-293. Woodruff, C. M. 1948. Testing soils for lime requirement by means of a buffered solution and the glass electrode. Soil Sci. 66:53-63. Yuan, T. L. 1959. Determination of exchangeable hydrogen in soils by a titration method. Soil Sci. 88:164-167. Yuan, T. L. 1970. Interpretation of soil pH in liming practice. Soil and Crop Sci. Soc. Fla. Proc. 30:200-210. Yuan, T. L. 1974. A double buffer method for the determination of lime requirement of acid soils. Soil Sci. Soc. Am. Proc. 38:437-440. Yuan, T. L. 1975. Lime requirement determination of sandy soils by different rapid methods. Soil and Crop Sci. Soc. Fla. Proc. 34:48-52. Yuan, T. L. 1976. Anomaly and modification of pH-acidity relationship in the double buffer method for lime requirement determination. Soil Sci. Soc. Am. J. 40:800-802. Yuan, T. L., N. Gammon Jr., and R. G. Leighty. 1967. Relative contribution of organic and clay fractions to cation-exchange capacity of sandy soils from several soil groups. Soil Sci. 104:123-128. Yuan, T. L., M. C. Lutrick, and W. K. Robertson. 1978. Response of soybeans and oats to lime, phosphorus, and potassium on a Paleudult. Soil Crop Sci. Soc. Fla. Proc. 38:116-121. Yuan, T. L., W. K. Robertson, and R. C. Littell. 1977. Soil pH adjustment by the double-buffer method for pearl millet (Pennisetum typhoides (Berm.) Staph and Hubbard) on an Ultic Haplaquod. Soil and Crop Sci. Soc. Fla. Proc. 37:208-212.



PAGE 1

Table 3-2. Chemical and physical characteristics of the 34 soils used in the incubation study. 1M KC1 Mehlich I Extractable Soil pHw pHw pHs exch. No. (1:1) (1:2) (1:2) O.C.O.M. Clay Silt Sand Al P K Ca Mg Al --------g 100 g-1 soil-----------------mg kg-l----------------1 5.66 5.61 4.88 1.1 1.9 3 3 94 0.11 120 16 397 197 393 2 4.82 4.86 4.04 0.8 1.4 5 2 93 0.51 7 6 36 15 128 3 5.56 5.61 4.78 0.8 1.7 5 4 92 0.28 22 13 125 72 389 4 5.89 5.89 5.06 1.0 1.8 6 1 93 0.02 90 20 440 235 267 5 4.88 4.82 4.05 1.2 2.0 3 1 96 0.54 13 10 71 35 162 6 4.01 4.17 3.15 2.2 3.8 1 2 97 0.61 3 12 92 61 84 7 5.24 5.27 4.70 1.9 3.2 5 3 92 0.24 55 46 396 236 710 8 6.00 5.92 5.11 1.2 2.1 3 4 92 0.00 27 12 570 331 176 9 6.32 6.32 5.69 1.3 2.2 5 4 91 0.00 97 48 1000 555 246 10 6.06 6.05 5.12 1.5 2.6 4 2 94 0.08 49 42 265 156 398 11 4.40 4.46 3.58 1.5 2.6 2 2 96 0.27 7 12 114 66 46 12 4.32 4.35 3.51 1.8 3.1 5 2 93 0.26 6 14 201 116 63 13 5.99 6.06 5.09 0.9 1.5 4 3 93 0.06 125 145 312 182 328 14 5.96 5.89 5.19 1.2 2.0 2 3 95 0.00 126 44 580 308 226 15 5.69 5.70 4.96 1.2 2.1 2 4 94 0.06 56 40 292 174 234 16 5.94 5.85 5.11 1.1 1.9 3 3 94 0.02 56 36 293 171 238 17 5.23 5.19 4.35 0.5 0.9 4 6 90 0.58 3 12 106 74 126 18 4.98 4.99 4.22 0.7 1.2 4 10 86 1.62 3 21 190 119 158 19 5.01 4.96 4.12 1.1 2.0 4 2 93 0.26 8 12 178 100 87 20 5.25 5.26 4.40 1.0 1.7 4 1 95 0.00 5 15 226 125 19 21 4.95 4.97 4.12 0.4 0.8 3 1 96 0.06 2 7 69 37 19 22 5.32 5.23 4.51 1.3 2.2 4 9 86 0.32 81 36 400 237 209 23 5.99 5.90 5.04 0.5 1.0 4 2 95 0.07 18 6 142 63 116 24 4.77 4.74 3.95 0.7 1.2 6 1 93 0.14 8 8 115 65 41 25 4.57 4.57 3.67 1.1 1.9 3 1 96 0.09 3 16 178 106 26 26 4.51 4.52 3.82 0.7 1.2 3 1 96 0.25 3 8 75 41 45 27 5.04 4.98 4.23 1.3 2.2 5 3 92 0.57 92 11 258 136 253 28 4.33 4.31 3.45 3.7 6.3 8 2 91 1.14 4 30 295 165 160 29 4.39 4.50 3.46 1.0 1.7 5 1 94 0.12 2 9 78 43 24 30 6.21 6.14 5.34 0.6 1.0 4 3 93 0.01 52 25 323 169 178 31 4.44 4.41 3.98 1.2 2.1 2 4 94 1.14 54 5 35 18 436 32 6.01 5.98 5.26 1.2 2.0 6 4 90 0.00 201 151 820 458 330 33 6.12 6.05 5.14 1.1 1.9 4 6 90 0.00 209 122 570 309 558 34 5.87 5.80 4.90 1.1 2.0 2 6 91 0.06 162 91 397 232 482



PAGE 1

111 Zelazry, L. W., F. G. Calhoun, and R. E. Caldwell., 1974. Soil Components contributing to acidity measured in selected Florida Paleudults. Soil and Crop Sci. Soc. Fla. Proc. 33:228-232.



PAGE 1

25 that less lime was required when soils and lime were not well mixed due to the presence of significant amounts of unreacted CaCO3 at all but the lowest lime levels. Comparison of Reference Methods As. researchers have moved to using relatively short term analyses as a measure of a soils lime requirement, the question remains as to what the true lime requirement of a soil is. Some workers have compared some of the reference methods or modifications of them. Yuan (1974) compared the BaC12-TEA method of Peech (1965) with a Ca(OH)2-CaCl2 one week incubation on twenty Florida soils. The mean IR determined by the BaCl2-TEA method was 4.54 T/A CaCO3 as compared to 4.47 T/A CaCO3 for Ca(OH)2-CaC12. McLean et al. (1978) reported that Ca(OH)2 titration to pH 7.2 with 72 hours of intermittent shaking gave values an average of 5% lower than those for CaCO3 incubation to pH 6.8 for 17 months (r = 0.99). Fox (1980) compared the BaC12-TEA method of Peech (1965), the Ca(OH)2 titration of McLean et al. (1978), and a six month CaCO3 incubation. The BaC12-TEA method overestimated the CaCO3 incubation LR below 9.28 meq CaCO3 100 g-1 soil and underestimated the LR above 9.28 meq CaCO3 100 g-1 soil. The Ca(OH)2 titration underestimated the CaCO3 incubation LR for all determinations. At 2 and 4 meq CaCO3 100 g-1 soil, the Ca(OH)2 method estimated only 29% and 39% of the CaCO3 LR, respectively.



PAGE 1

3 values. Field recommendations can then be made from these results. The overall objective of this thesis was to evaluate the AE and Yuan DB methods on Florida's sandy soils. These are the only two methods being evaluated because of the following reasons: 1/ There is a desire for collaboration between state soil testing laboratories, thus the desire to keep the tests as similar as possible, possibly sacrificing some accuracy. 2/ There may be a push in the future for DB methodology, so these methods should continue to be evaluated. The thesis is divided into five parts. Chapter II reviews the literature pertinent to the study. Chapter III presents the laboratory incubation study designed to evaluate the AE and Yuan DB methods. Chapter IV relates the laboratory data to actual field responses. Chapter V relates the results from the incubation study to a wider range of Florida soils to determine the wider applicability of the AE method. Finally Chapter VI summarizes the study and provides recommendations for further work.



PAGE 1

22 Zelazny et al. (1974) found clay content to give a higher correlation coefficient with total acidity of fifteen surface soils of Florida Paleudults than organic matter whereas organic matter gave a higher correlation coefficient with exchangeable acidity. Methods of Testina ,Lime Requirement Methods Due to the diversity of forms of soil acidity along with the influence of soil solution pH on their availability, changes in pH should be gradual so that all available acidity at a particular pH is neutralized (McLean, 1982a). This is an important consideration when attempting to use a quick-test buffer method, where the soil is subjected to sudden pH changes, as a measure of the effect of lime in the field where the changes are much more gradual. In correlation studies, the researcher is attempting to discern a lime requirement method that provides an accurate index of the lime needs of soil samples representative of the area of interest. Calibration studies provide field results that are used as the basis for recommendations (Hanway, 1973). Reference Methods McLean et al. (1966) listed the order.for lime requirement test methods,, with respect to increasing amount of time needed for completion as: pH measurement < titration < incubation < field studies. Early researchers relied primarily on field studies to base lime recommendations. Besides requiring



PAGE 1

4.48 Mg ha-1 SITE 3 ---. a 1.12 Mg ha-1 2.24 Mg ha-1 .a U1 60S0 Mg ha-1 5.51I 1-I I I 11/83 3/84 5/84 7/84 11/84 8/85 SAMPLING DATE Figure 4-4. The effect of lime treatments on soil pHw, over the time period of the study, at site 3. Each point represents the average of four replications.



PAGE 1

79 often has a clayey substrata that underlies a fine sand surface layer which was found at this site (Soil Survey Staff, 1965). This complex is also low in organic matter and in natural fertility, with a low available moisture capacity. Table 4-1 shows that even though this soil is under cultivation, soil pH and extractable base measures are relatively low. It was noted that the owner of the land where site 2 is located did not provide for a large amount of inputs. The particular field on which the experiment was located was used mostly for pasture and apparently few inputs were applied to the field. Site 3 was located on the Live Oak Agricultural Research Center (ARC) at Section 22 T.2S.-R.14E.. The soil was a Lakeland series described as a thermic, coated, Typic Quartzipsamment. Virgin sites of these soils are usually deep, well drained to excessively drained, strongly acid, coarse textured, and generally low in natural fertility and organic matter content (Soil Survey Staff, 1965,). Once again, soil pH and extractable bases were relatively high, in the trial area (Table 4-1), probably due to continued cultivation and the addition of proper soil amendments. During the process of locating suitable sites for the field experiments, great difficulty was encountered in obtaining fields with low pHw values. Most of the farmers in Suwannee County already had their soil pHwat or above 6.0.



PAGE 1

EVALUATION OF LIME REQUIREMENT METHODS FOR FLORIDA'S SANDY SOILS By THOMAS S. DIEROLF A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 1986



PAGE 1

101 A correlation study evaluated the AE method on a group of 98 soils with a wider range of physical and chemical properties than the 34 soils used in the incubation study. The BaC12-TEA method was used as a reference method for this group of soils. It was determined from this study that the BaC12-TEA method was not a reliable predictor of soil LR for target pH, values ranging from 5.5 to 6.5. However, the total acidity measured by the AE method was highly correlated with the reference method. Approximately the same relationship between AE total acidity and BaCl2-TEA acidity was found for both the incubation soils and for the group of 98 soils. This result suggested that the AE method provides an accurate index of lime requirement for a wide range of Florida soils. In summary, the laboratory study showed that the AE LR method provided a very reliable indication of the degree of soil acidity that must be neutralized to attain a particular pH. However, when applied under field conditions, the AE LR underestimated the field LR. The liming factor of 1.5, recommended by Adams and Evans and used by the ESTL, was shown to be too low. Values of 3 to 5 would be more appropriate. In Florida, there is not as much concern for liming soils with an initial pHw of 6.0 or above, thus liming factors of 5.0 are probably too high for most cases. The results of the study suggest that the following two changes would improve the accuracy of the LR recommendations made by the ESTL. First, the equation developed from the



PAGE 1

7.06. ...... 1".. .." 50 ... .... .... .. pHw = 7.34 -0.03(%BU). (r = -0.752, n = 567) I. .... . 20 40 60 80 100 BASE UNSATURATION (%) ----_ Q--.-----80 60 40 20 0 BASE SATURATION (%) Figure 3-2. Relationship between pHw and base unsaturation for 567 Florida topsoils (0 -15 cm).



PAGE 1

29 important for Florida's sandy soils where much of the buffering capacity is pH-dependent. The calculation employs the pH-BU relationship found for 348 Alabama Ultisols (Adams and Evans, 1962). The BU value for a particular pH can be determined from the curvilinear regression equation that was determined to give the best fit for the pH-BU relationship. Thus, knowing the initial pHw of the soil, the desired pH, of the soil, and the AE TA, the following equation of Adams and Evans (1962) is used to determine the AE LR: AE LR = AE TA x (Desired change in BU) Initial BU The AE TA is calculated from the soil-buffer equilibrium pH. The initial BU is calculated from the initial pHw of the soil and the regression equation for the pH-BU relationship. The desired change in BU is calculated by subtracting the BU value corresponding to the desired pH, from the BU value corresponding to the initial pHw of the soil. The resulting AE LR is the amount of CaCO3 that will neutralize the portion of AE TA necessary to attain the desired pHw.The LR may be presented in terms of agricultural limestone if a 'liming factor' is applied to the CaCO3 LR. Previous Evaluations of Lime Requirement Methods The various rapid soil test lime requirement methods have been evaluated by many workers and on a wide assortment of soils. This section will detail the strengths and weaknesses



PAGE 1

95 Table 5-1. Selected characteristics of 98 soils used in the correlation study. Soil Ordera Clay OM pH, Soil Ordera Clay OM pHw -q 100 q-1--q 100 q-1 1 E 0.7 1.16 5.3 50 A 5.7 5.65 4.8 2 I 3.3 11.38 4.5 51 S 2.2 2.38 4.8 3 E 0.6 1.14 4.5 52 A 16.9 3.22 5.1 4 E 1.9 1.03 5.5 53 U 5.5 3.43 5.1 5 E 0.3 0.84 4.7 54 U 18.9 5.38 4.3 6 S 3.2 6.97 4.5 55 I 2.6 2.93 4.7 7 A 4.5 6.24 4.4 56 A 1.8 2.72 5.9 8 E 1.5 1.24 5.0 57 S 1.4 2.69 4.3 9 E 1.3 1.78 4.6 58 A 10.2 4.48 4.9 10 A 8.7 2.84 5.5 59 U 4.0 2.76 4.7 11 S 0.3 0.93 4.5 60 S 0.8 2.21 4.5 12 E 0.5 0.86 5.3 61 E 0.5 1.09 5.4 13 S 1.2 2.33 4.3 62 S 0.5 1.52 4.5 14 E 1.5 2.02 4.6 63 S 1.8 1.38 5.3 15 S 1.3 1.12 4.8 64 S 2.3 3.10 4.8 16 A 1.6 1.22 4.9 65 E 3.2 1.53 5.2 17 S 2.1 0.40 4.8 66 E 2.3 1.67 4.8 18 E 0.0 1.10 4.8 67 U 3.9 1.38 5.7 19 E 1.4 0.95 5.3 68 U 13.6 4.28 5.3 20 U 2.4 1.03 6.0 69 E 0.2 1.48 5.4 21 U 1.5 2.38 4.4 70 U 17.1 1.95 6.1 22 U 2.3 1.05 5.5 71 U 12.0 2.31 5.6 23 I 14.0 2.93 5.2 72 U 7.9 3.07 5.5 24 U 3.6 3.34 5.1 73 U 6.4 1.72 5.4 25 U 7.4 2.48 4.9 74 E 2.0 3.65 5.0 26 E 3.1 0.95 5.6 75 S 1.5 0.69 5.0 27 U 3.2 1.10 5.9 76 S 1.2 1.76 5.3 28 E 0.2 1.21 4.9 77 S 0.4 1.10 4.6 29 E 2.2 1.76 5.1 78 A 0.7 0.97 4.4 30 U 7.9 1.76 6.0 79 S 0.6 1.62 4.3 31 U 11.0 5.65 4.5 80 E 0.9 2.02 5.7 32 U 4.4 2.55 5.0 81 S 0.9 1.17 4.4 33 U 16.3 1.41 5.4 82 S 0.4 0.52 4.7 34 U 9.6 1.33 5.9 83 A 16.6 3.10 5.7 35 S 3.6 6.03 4.5 84 A 1.5 1.52 4.8 36 U 4.6 2.81 4.7 85 U 4.5 2.57 4.6 37 A 0.8 1.24 4.5 86 E 0.6 2.50 4.6 38 M 2.7 2.69 5.0 87 U 1.2 0.83 6.0 39 E 2.3 1.57 5.8 88 S 1.1 4.40 4.4 40 S 1.2 1.74 4.4 89 A 1.5 2.52 5.2 41 S 0.5 1.88 4.4 90 A 0.7 1.38 5.2 42 S 0.5 2.17 4.5 91 S 1.4 3.71 4.4 43 U 8.5 1.17 6.1 92 S 2.8 3.09 4.6 44 A 35.1 3.33 6.1 93 E 3.3 7.03 5.5 45 U 13.8 0.88 5.5 94 E 5.8 9.67 5.7 46 U 19.2 2.52 5.3 95 M 19.5 14.91 5.0 47 A 1.5 1.21 4.8 96 E 3.0 1.74 4.6 48 S 5.5 2.71 5.0 97 S 1.6 1.43 4.4 49 S 4.0 6.26 4.4 98 S 1.2 2.10 4.4 aA=Alfisol, E=Entisol, I=Inceptisol, M=Mollisol, S=Spodosol, U=Ultisol



PAGE 1

47 Table 3-5. Periodic pHw (1:2) measurements taken from 14 of the experimental units over the length of the incubation study to determine when equilibrium pHw was reached. pHw (1:2) Lab Sample After After After Numbera 6 weeks 10 weeks 12 weeks 35 6.80 6.58 6.57 51 5.32 5.27 5.21 81 5.85 5.75 5.73 96 6.72 6.42 6.38 104 6.17 6.10 6.12 113 6.47 6.30 6.25 226 5.82 5.75 5.74 256 6.30 6.12 6.10 327 6.02 5.70 5.64 360 7.10 6.83 6.77 365 7.02 6.72 6.74 380 5.905.80 5.86 384 6.97 6.83 6.81 396 6.80 6.60 6.57 Mean pHw 6.38 6.20 6.18 a There were a total of 408 experimental units. 34 soil x 6 treatments x 2 replications.



PAGE 1

31 underestimated the soil LR for soils requiring >4.87 meq 100 g-1 soil. Yuan (1975) correlated the AE to BaC12-TEA method at pH 8.1 (r = 0.78) for 31 Florida soils. The AE only predicted 1/3 of the LR of the reference method. Yuan felt that only part of this difference was because the AE LR gives the amount of lime required to raise the soil pH to 6.5, while the BaCl2-TEA method estimates the total acidity. Most of the difference was apparently due to discrepancies in the AE LR method itself. Contrary to the findings of the previous papers, Tran and van Lierop (1981) found the average AE LR was 145% of that determined by the CaCO3-soil incubation. Using a wider soil:buffer ratio (1:4) expanded the buffer range, improved the correlation, and was only 114% of the CaCO3-soil incubation determined LR. Yuan Double Buffer Method Yuan (1974) compared the Yuan DB method to both the BaCl2-TEA method and the Ca(OH)2 titration-incubation using 20 Florida soils. The Yuan DB was highly and equally correlated to both reference methods (r = 0.97). Yuan DB method estimates of 1.0, 2.0, and 4.0 T/A compared to 1.38, 2.36, and 4.32 T/A as determined by the BaCl2-TEA method. The Ca(OH)2 method gave results comparable to the BaCl2-TEA method.



PAGE 1

71 The most satisfactory method of prediction of the incubation LR found in this study was one which used the AE soil-buffer equilibrium pH value and the difference. between initial pH and target soil pH (AE-F). When incubation LR, to pH 5.5, 6.0, and 6.5 (n = 85), was regressed on AE buffer pH and the difference between target and initial pH, the following equation (AE-F) resulted: Incubation LR = 17.4 -2.27 (AE buffer pH) + 0.68 (target pH -initial pH) (r = 0.94, C.V. = 27%, n = 85) Employing the change in pH and AE buffer pH (AE-F) appeared to be as accurate as using the AE method (Fig 3-16). Use of the latter correction factor eliminated the use of the pH-BU relationship that has been shown to be a very generalized relationship for Florida soils. The ESTL assumes that 15 ml soil is equal to 15 g soil and the lime recommendations are based on this. However, 15 ml soil was found to be equal to 20 g soil in this study. Hence, the recommendations given by the ESTL are actually 33% greater than the AE test predicts. Thus current ESTL recommendations already compensate for some of the underestimation of the AE method. Summary and Conclusions In summary the following conclusions are reached from this study of a group of surface soils representative of sand-textured Florida agricultural soils:



PAGE 1

100 relationship and adjusted for the AE method's underestimation of the incubation LR. The AE-F method uses the AE buffer pH value and the following equation to predict the incubation LR: Incubation LR = 17.4 -2.27 (AE buffer pH) + 0.68 (target pH -initial pH) A field study designed to evaluate the accuracy of the current ESTL lime recommendations was reported in Chapter IV. This study revealed that the current lime recommendations underestimated the actual field lime requirement. The problem was shown to be related to the liming factor as well as the underestimation of AE-LR. Assuming that the AE-F method accurately predicted soil LR as was reported from the incubation study, it is then evident from the field study that the current 'liming factor' of 1.5 as used by the ESTL is too low. Liming factors ranging from 3.2 to 5.3 were found by the field study. The lower liming factor of about 3 was found for the field trial that had a relatively low initial soil pHw.The higher liming factor of about 5 was found for the field trials that had a high initial soil pHw.The increase in the liming factor .was related to the decrease in pHw response to liming at a higher initial pHw.These results indicated that not only is the current liming factor of 1.5 too low, but that a range of liming factors, partly dependent on initial soil pH, are more appropriate.



PAGE 1

1.48 Mg ha-I SITE 1 65 2.24 Mg ha-1 ll.12 tMg ha-1 6J -0 Mg ha-1 C0 5550 .__ I I I I 11/83 1/84 3/84 5/84 11/84 8/85 SAMPLING DATE Figure 4-2. The effect of lime treatments on soil pHw over the time period of the study, at site 1. Each point represents the average of four replications.



PAGE 1

y = 0.07 + 0.95x (r = 0.96, n = 30, C. = 24%) O 20o (0 0 1.0 u k* 0 o0 S1.0 2.0 3.0 INCUBATION LRI 6.0 (g CaCO3 kg-1 soil) Figure 3-16. The regression equation and statistics between AE-F and incubation LR to pH 6.0. *^ * / ^, p o M / < / ^ and incubation LR to pH 6.0.



PAGE 1

77 SITE 3 SITE 2 SITE 1 Location of Suwannee County, Florida Figure 4-1. The location of the three field trials in Suwannee County, Florida.



PAGE 1

87 The wide pHw fluctuations were dealt with by averaging and the application of regression equations. The actual calculations are discussed in more detail in the following section. Calculation of Field Lime Requirement To provide an indication of the field lime'requirement the following calculations were made: 1/ The average pH, of four replications for each treatment level were calculated for each of the sampling dates (Fig. 4-2, 4-3, 4-4). 2/ For each treatment level, the average pHw of all measurements taken at least six months after the lime was added to the soil was calculated. 3/ The treatment levels were regressed with the average pHw values shown in Table 4-2. The following regression equations were calculated, with (x) being limestone in Mg ha-1, and (y) being the predicted pHw of the field soil: Site 1 ( y = 6.18 + 0.09x ) [1] Site 2 ( y = 5.35 + 0.23x ) [2] Site 3 ( y = 6.15 + 0.08x ) [3] These regression equations were then used to calculate the field LR. Comparison of Field and Laboratory Data In Table 4-3 are compared the various lime requirement determinations for the three soils. All LRs were converted to g CaCO3kg-1 soil to allow for direct comparison.



PAGE 1

90 Looking at the site 1 and 3 soils, the AE LR and the field trials both -concurred that no lime was needed to attain pHw 6.0. When target pHw 6.5 was desired the AE LR underestimated the field lime requirement for both sites 1 and 3. The underestimation by the AE LR method is more evident when site 2 field data are examined. As discussed in Chapter III, the incubation study showed that as incubation LR increased, the AE method estimated less of the incubation LR. It was also mentioned in Chapter III that the ESTL compensated for some of the underestimation of the AE LR method because of the assumption that 1 ml of soil was equal to 1 g soil. From the field data it appears that even with this correction factor the LR is still underestimated. The reason for the underestimation may be an incorrect assumption of a 1.5 liming factor for sandy Florida soils. As the field LR increased the AE-F method underestimated the field LR by increasingly larger amounts. If liming factors were determined from the AE-F and field data, they would range from 3.2 to 5.3. For site 2, the liming factor to attain either pH, 6.0 or pHw 6.5 was about 3. For sites 1 and 3, the liming factors were about 5. The higher liming factors for the soils at sites 1 and 3 were related to the high initial pHw of the soils. The greater slope for equation [1] then for equations [2] and [3] indicate that there was a greater pH, response to liming as the difference between target pHwand initial pHwincreases. Thus



PAGE 1

65 method. For example, for LR6.5 the AE method estimated only about 0.2 g CaCO3 for every 1.0 g CaCO3 kg-1 soil estimated by the reference method. The AE LR to lower pH levels would be expected to underestimate BaCl2-TEA LR since the BaCl2-TEA extractant is buffered at pH 8.2. More acidity would theoretically be extracted at the higher pH than would be neutralized by the incubation. As the incubation target pH LR increases it would begin to measure the same amount of acidity as the BaCl2-TEA method. The AEmod method was much better correlated with the reference method than the AE method (Table 3-7). For LR6.0 the correlation coefficient was slightly lower and the C.V. value higher than for LR5.5 and LR6.5. The AEmod method also severely underestimated the reference method. For LR6.5 the AEmod method estimated on the average 0.24 g CaCO3 for every 1.0 g CaCO3 kg-1 soil estimated by the reference method. The Yuan DB method was correlated as well with the reference method as was the AEmod method (Table 3-7). Yet the Yuan DB method also severely underestimated the reference method LR. For LR6.5 the Yuan DB method measured 0.5 g CaCO3 for every 1.0 g CaCO3 kg-1 soil predicted by the reference method. When the total amount of acidity measured by the BaCl2-TEA (pH 8.2) method was used as a reference method, lower r values and higher C.V. values were found for each of the three soil-test methods than when CaCO3incubation to pH



PAGE 1

57 in organic soils. Brown and Cisco (1984) found that for 14 soils, all but one showed a linear response up to pHw 7.0 to CaCO3 additions in a greenhouse pot trial. The AEmod LR values, using the pHW-BU relationship for Florida soils, were calculated for all 34 soils, six of which had pHw values <4.5. The original AE method regression of BU and pH did not include soils with such low pH values and thus was only used for 28 of the soils studied. To facilitate discussion of the data, the AE, AEmod, and Yuan DB tests will each be covered separately, first using the incubation as the reference method. The second section will cover the soil buffer methods using BaC12-TEA as the reference method. Finally the last section will present the findings of the two reference methods and how they compare to each other. AE Method and CaCO3 Incubation The lowest correlation coefficient for any of the three methods versus the three target pH levels was for the AE method to LR5.5 (r = 0.68) (Fig.3-11). The AE method estimated an average of only about 0.56 g CaCO3 kg-1 soil for every 1.0 g CaCO3 kg-1 soil (equivalent to 2.24 Mg CaCO3 ha-1 or 1 ton CaCO3 acre-1; also equivalent to 3.36 Mg ag lime ha-1 or 1.5 tons ag lime acre-1) predicted by incubation LR5.5. Although the AE method was highly correlated with incubation LR6.0 and LR6.5 (r = 0.96, r = 0.96), the AE method underestimated incubation LR at all practical levels



PAGE 1

108 results. Florida Cooperative Extension Service. Circular 596. Rhue, R. D., and G. Kidder. 1984. The lime requirement test in Florida. Notes in Soil Sci. No. 13. Soil Sci. Dept. IFAS, University of Florida, Gainesville. Sanchez, P. A. 1976. Properties and Management of Soils in the Tropics. J. Wiley and Sons, New York. Savant, N. K., and M. M. Kibe. 1971. Acidity in lateritic rice soils as estimated by various methods. Soil Sci. Soc. Am. Proc. 35:845-847. Schnitzer, M., and S. I. M. Skinner. 1963. Organo-metallic interactions in soils: I. Reactions between a number of metal ions and the organic matter of a podzol Bh horizon. Soil Sci. 96:86-93. Schofield, R. K. 1933. Rapid methods of examining soils: II. The use of p-nitrophenol for assessing lime status. J. Agric. Sci. 23:252-254. Schollenberger, C. J., and R. M. Salter. 1943. A chart for evaluating agricultural limestone. J. Am. Soc. Agron. 35:955-966. Sharp, L. T., and D. R. Hoagland. 1915. Acidity and adsorption in soils as measured by the hydrogen electrode. J. Agric. Res. 7:123-145. Shaw, W. M. 1953. Reaction of calcium carbonate with soils and determination of their calcium sorption capacities. J. Assoc. Off. Agric. Chem. 36:421-444. Shoemaker, H. E., E. O. McLean, and P. F. Pratt. 1961. Buffer methods for determining lime requirement of soils with appreciable amounts of extractable aluminum. Soil Sci. Soc. Am. Proc. 25:274-277. Shuman, L. M., F. C. Boswell, K. Ohki, M. B. Parker, and D. O. Wilson. 1983. Effects of HC1 acid and lime amendments on soil pH and extractable Ca and Mg in a sandy soil. Commun. in Soil Sci. Plant.Anal. 14:481-495. Soil Survey Staff. 1972. Soil Survey laboratory methods'and procedures for collecting soil samples. Soil Survey Investigations Report No. 1 (Revised). U.S. Dept. Agric., Soil Conservation Service, U.S. Government Printing Office, Washington, D.C..



PAGE 1

33 was more accurate for low lime requiring soils than for high lime requiring soils. It predicted 89% and 73% of the LR, respectively for 54 soils from the U.S. Tran and van Lierop (1982) used 37 acid, coarse-textured soils. They evaluated the Yuan buffer both as a double buffer and as a single buffer (Yuan SB). Linearity of the Yuan SB method increased as its. initial pH approached the desired soil pH. The precision was lower when the initial pH was 6.5 versus 6.0 or 7.0. They felt this could be due to a break in the titration curve at pH 6.2 (Yuan, 1976). For a target pH of 5.5 (LR5.5) or a target pH of 6.0 (LR6.0) the precision of the relationship was good for either the initial buffer pH of 6.0 or 7.0. The DB for LR(6.0) had a high correlation coefficient (r = 0.97) versus incubation with CaCO3 but only measured an average of 67% of the incubation LR. They proposed using a regression equation to adjust for the difference. The DB was less precise for LR(5.5) (r = 0.76). Yet the Yuan SB was as precise for LR(6.0) and more precise for LR(5.5) versus the Yuan DB. Van Lierop (1983) found that the Yuan DB performed poorly on organic soils. He felt this might be because many of the soil-buffer pH values ranged from pH 5.0 to 6.0 which fell on the curvilinear portion of the buffer curve. Altering the soil-buffer ratio to 1:5 vol/vol to raise the soil-buffer pH to the linear portion of the curve did not the improve the



PAGE 1

93 Materials and Methods A total of 98 surface horizon (0 to 15 cm) soils, representing 15 counties of Florida, were chosen from stored Soil Characterization Lab samples (Fig 5-1). Six soil orders. were represented as follows: 28 Spodosols, 26 Ultisols, 24 Entisols, 15 Alfisols, 3 Inceptisols, and 2 Mollisols (Table 5-1). Published data, (Calhoun et al., 1974; Carlisle et al., 1978, 1981, 1985) were used to obtain the soil classification, location (county), and % clay and % organic matter contents. Soil pHw (1:2) and AE buffer pH were determined, in duplicate, on the stored samples by the Extension Soil Testing Lab (ESTL) (Rhue and Kidder, 1983). Extractable acidity by the BaCl2-TEA (pH 8.2) method was determined, in duplicate, on the stored samples (Soil Survey Staff, 1972). The AE LR was determined to target pH levels of 5.5, 6.0, and 6.5, in accordance with the original published procedures (Adams and Evans, 1962). One ml of soil was assumed to weigh 1.32 g (as previously reported in Chapter III). The AE LR was evaluated by regression on BaCl2TEA extractable acidity. Results and Discussion The data in Table 5-2 gives the appearance that the AE LR method is not an accurate or precise predictor of LR, especially to pH 5.5 and 6.0. The improvement in the relationship between AE LR and the reference method as the AE LR



PAGE 1

54 6.01 6 5.0 4.0 0.0 1.0 2.0 3.0 -1 g CaCO3 kg soil Figure 3-8. Titration curves of final pHs versus the amount of CaCO3 added for soils 1, 2, 4, 6, and 9. 331 g CaCO kg soil Figure 3-9. Titration curves of final pHs versus the amount of CaCO added for soils 22, 26, 29, and 31. 29, and 31.



PAGE 1

96 Table 5-2.Regression statistics between AE LR and the BaCl2TEA reference .method. (y) Intercept Slope (x) n r C.V.(%) AE LR 5.5 0.00 0.08 BaCl2-TEA 83 0.47 140 AE LR 6.0 0.10 0.13 BaC12-TEA 83 0.72 56 AE LR 6.5 0.18 0.17 BaC12-TEA 83 0.85 33 AE TA 0.39 0.25 BaC12-TEA 98 0.95 16



PAGE 1

73 1/ The AE LR method estimated 0.56, 0.67, and 0.59 of the incubation LR to pH 5.5, 6.0, and 6.5, respectively. 2/ Use of the AEmod method did not show a practical increase in accuracy of LR prediction. 3/ The BaC12-TEA method for determining extractable acidity was highly correlated with the total acidity measured by the AE buffer. 4/ The Yuan DB method overestimated the incubation LR and was not as precise as the AE method. 5/ A regression equation, shown below, determined from the data found in this study gave a very satisfactory estimate of LR by using both the AE soil-buffer equilibrium pH and the difference between initial soil pH and target pH. Incubation LR = 17.4 -2.27 (AE buffer pH) + 0.68 (target pH -initial pH).



PAGE 1

105 methods used for determination of cation exchange capacity. Commun. Soil Sci. Plant Anal. 14:1005-1014. Hanway, J. J. 1973. Experimental methods for correlating and calibrating soil tests, p. 55-66. In L. M. Walsh and J. D. Beaton (ed.) Soil Testing and Plant Analysis. Soil Sci. Soc. Am., Madison, Wis. Hargrove, W. L., and G. W. Thomas. 1984. Extraction of aluminum from aluminum-organic matter in relation to titratable acidity. Soil Sci. Soc Am. J. 48:1458-1460. Hesse, P. R. 1971. A Textbook of Soil Chemical Analysis. Chemical Pub. Co., New York. Hester, J. B., and F. A. Shelton. 1933. Seasonal variation of pH in field soils a factor in making lime recommendations. Journal Am. Soc. Agron. 25:299-350. Hopkins, C. G., W. H. Knox, and J. H. Pettit. 1903. A quantitative method for determining the acidity of soils. USDA Bur. of Chem. Bull. 73. Horn, D. P., M. M. Alley, and P. M. Bertsch. 1982. Cation exchange capacity measurements. Commun. Soil Sci. Plant Anal. 13:851-862. Jenny, H. 1961. Reflections on the soil acidity merry-goround. Soil Sci. Soc. Am. Proc. 25:428-432. Jensen, S. T. 1924. Ueber die bestimmung der pufferwirkung des bodens. Intern. Mitt. fur Bodenkunde. 14:112. Jones, C. H. 1913. The determination of lime requirement by the calcium acetate method. Am. Fert. 39:28-29. Kamprath, E. J. 1970. Exchangeable aluminum as a criterion for liming leached mineral soils. Soil Sci. Soc. Am. Proc. 34:252-254. Keng, J. C. W., and G. Uehara. 1974. Chemistry, mineralogy, and taxonomy of Oxisols and Ultisols. Soil and Crop Sci. Soc. Fla. Proc. 33:119-126. Lathwell, D. L., and W. S. Reid. 1984. Crop response to lime in the northeastern United States. In F. Adams (ed.). Soil Acidity and Liming. Agronomy 12:305-332. Am. Soc. of Agron., Madison, Wis. Lin, C., and N. T. Coleman. 1960. The measurement of exchangeable aluminum in soils and clays. Soil Sci. Soc. Am. Proc. 24:444-446.



PAGE 1

51 average 0.01M CaC12 pH than for the unleached soils in pHw seems to indicate that the amount of salts accumulated was less than the concentration of 0.01M CaCl2.Thus by using pHs the salt effect can be accounted for. For this reason only results for incubation LR determined from pHs versus CaCO3 added curves will be reported as incubation LR. To facilitate the practical application of the following results, target pH values will be reported in terms of pH,, but will have been calculated from pHs values. Figures 3-4 thru 3-10 show the titration curves for all 34 soils. The curves for pHw and pHs were very similar except that the pHs values were lower. For the reason explained below, only pHs titration curves are shown. The titration curves were analyzed for both linearity and curvilinearity. All of the relationships were best described as linear. Table 3-6 lists the equations that were used to determine the LR values to the respective target pH values. Also shown are the comparison of r2 for linear and curvilinear equations. Due to the small differences in r2 between the linear and curvilinear equations as well as the small number of coordinate pairs (6) the linear equations were used in all of the cases. Linear increases in pH to CaCO3 additions have been reported by several other researchers. McLean et al. (1960) observed linear increases in pHw to CaCO3 additions to incubated so'ils. Van Lierop (1983) reported a linear increase in pHwto CaCO3additions



PAGE 1

a) b) 0 0 -o 00 y = 0.10 + 0.49x y = -0.04 + 0.70x S (r=0.68, n=13, CY.=29%) (r=0.96, n=23, CV.=21%) M 0 1.0s o *** -^o * co S y = 0.02 + 0.57x n .0 1.0 2.0 20 INCUBATION LR 6.5 (g CaCO3 kgsoil) Figure 3-11. Regression curves and statistics of AE LR versus incubation LR. a) to pH 5.5; b) to pH 6.0; c) to pH 6.5. a) to pH 5.5; b) to pH 6.0; c) to pH 6.5.



PAGE 1

16 BaC12-TEA at pH 8.2 (Thomas, 1982). Total acidity gives an indication of the amount of acidity that must be neutralized to attain a pH somewhere between the original soil pH and pH 8.2. This value would give an indication of the lime .requirement of a soil. Buffering Capacity The buffering capacity of a soil is its ability to resist pH change. The greater the buffering capacity of a soil the greater its lime requirement. There are several ways a soil can exhibit buffering capabilities. Most important in highly weathered soils are crystalline and noncrystalline oxides and hydrous oxides of Al and Fe, kaolinite, and allophane. The minerals have surface hydroxyls that protonate or deprotonate in response to pH fluctuations (Keng and Uehara, 1974). Acidic groups in organic matter, mostly carboxyls, also ionize at higher pH levels releasing H+ ions (Stevenson, 1982). The H+ ions released from both the mineral and organic matter can neutralize any added bases. Role of Al in Soil DH Buffering Aluminum hydrolysis is associated with both the mineral and organic fractions of the soil and can be a significant contributor to the buffering capacity of a soil. Hydrogen ions are subsequently released according to the following sequence of reactions occurring in acid soils (Tisdale and Nelson, 1975):



PAGE 1

Pare Figure 3-11. Regression curves and statistics of AE LR versus incubation LR. a) to pH 5.5; b) to pH 6.0; c) to pH 6.5....................... 58 Figure 3-12. Regression curves and statistics of AEmod versus incubation LR. a) to pH 5.5; b) to pH 6.0; c) to pH 6.5....................... 60 Figure 3-13. Regression curves and statistics of Yuan DB versus incubation LR. a) to pH 5.5; b) to pH 6.0; c) to pH 6.5........................ 62 Figure 3-14. The regression equation and statistics between the total amount of acidity measured by the AE buffer (TA) and BaC12-TEA extractable acidity............... 67 Figure 3-15. The regression equation and statistics between the AE soil-buffer equilibrium pH and the incubation LR...................... 70 Figure 3-16. The regression equation and statistics between AE-F and incubation LR to pH 6.0... 72 Figure 4-1. The location of the three field trials in Suwannee County, Florida ................... 77 Figure 4-2. The effect of lime treatments on soil pH,, over the time period of the study, at site 1. Each point represents the average of four replications........................... 83 Figure 4-3. The effect of lime treatments on soil pH,, over the time period of the study, at site 2. Each point represents the average of four replications........................... 84 Figure 4-4. The effect of lime treatments on soil pH,, over the time period of the study, at site 3. Each point represents the average of four replications........................... 85 Figure 5-1. The location of the 98 soils used in the correlation study. Each number represents the amount of samples originating from the respective county........................... 94 x



PAGE 1

76 field data showing the approximate lime levels needed to adjust soil pH to predetermined target pHw values. These results could then be compared to what was found under laboratory conditions and thus give an indication of the liming factor. Materials and Methods Description of Sites and Soils Of the four experiments initially laid out in the fall of 1983, three were located in Suwannee County (Fig. 4-1). The fourth experiment will not be reported on since only six months of data were recorded prior to the loss of the site due to experimental difficulties. Site 1 was on a farm located in Section 21 T.3S.-R.15E.. The soil on which the trial was performed belongs to the Blanton series (loamy, siliceous, coated Typic Quartzipsamment). Blanton soils are described as deep, light-colored, sandy soils principally on gentle slopes, that developed from thick beds of marine sand very low in silt and clay (Soil Survey Staff, 1965). Virgin sites of these soils are usually strongly acid, low in natural fertility, and low in organic matter content. As shown by Table 4-1, under cultivation this soil can have relatively high pH values and extractable base contents. Site 2 was located on another farm in Section 35 T.3S.-R.12E.. The soil belongs to the Blanton-Susquehenna complex. The Susquehenna series is described as a fine, montmorillonitic, thermic, vertic Paleudalf. This complex



PAGE 1

CHAPTER II REVIEW OF THE LITERATURE Introduction The lime requirement of a soil can be defined as the amount of lime or other base required to neutralize the undissociated and dissociated acidity in the range from the initial acid condition to a selected neutral or less acid condition (McLean, 1980). The key word in this definition is 'selected' such that the reason for selecting the desired pH is unrelated to the definition. This distinction is necessary because some definitions may include the reason for attaining a certain pH within the definition. This is true of the one given by Hesse (1971), where he relates lime requirement to the amount of lime needed for maximum economic return from a particular soil. The term pH is also left out of the definition because other workers such as Kamprath (1970) have reported that on highly weathered leached soils exchangeable Al was a valid criterion on which to base lime rates. The remainder of the literature review is divided into four sections with the purpose of presenting a general understanding of soil lime requirement (LR) methods used to predict soil LR. The first section covers a general history 4



PAGE 1

Table 4-1.Selected chemical and physical characteristics of the top 15 cm of the three soils used in the field study. pHw pHw pHs 1M KC1. Site (1:1) (1:2) (1:2) OC OM Clay Silt Sand Mehlich I extractable exch. Al P K Ca Mg Al ------g 100 g-------------------mg kg-1-----------1 6.0 5.9 5.2 1.2 2.0 2 3 95 226 126 44 580 308 0.00 2 5.0 5.0 4.1 0.5 0.8 5 2 93 84 27 20 46 27 0.19 3 5.9 5.8 5.1 1.1 1.9 2 3 94 238 56 36 293 171 0.02 00



PAGE 1

53 14 3 1112 6.0o5.04.0 0.0 1.0 2.0 -1 g CaCO3 kg soil Figure 3-6. Titration curves of final pHs versus the amount of CaCO3 added for soils 3, 5, 11, 12, and 14. 21 60 -13 0 19 10 w 5.0 4.0,, 0.0 1.0 2.0 g CaCO3 kg soil Figure 3-7. Titration curves of final pHs versus the amount of CaCO3 added for soils 10, 13, 19, 20, and 21.



PAGE 1

Table 3-7. Regression statistics of various lime requirement determinations versus the BaCl2-TEA (pH 8.2) extractable acidity reference method (g CaCO3 kg-1 soil). Dependent Intercept Slope Independent Variable Variable n r CV --%-AE to pH 5.5 0.22 0.02 BaC12-TEA 13 0.00 41 AE to pH 6.0 0.14 0.07 BaC12-TEA 24 0.19 78 AE to pH 6.5 0.20 0.12 BaC12-TEA 28 0.57 41 AEmod to pH 5.5 -0.02 0.14 BaC12-TEA 19 0.78 46 AEmod to pH 6.0 -0.12 0.20 BaCI2-TEA 30 0.70 62 AEmod to pH 6.5 0.01 0.24 BaCl2-TEA 34 0.80 36 Yuan to pH 5.5 -0.61 0.38 BaCl2-TEA 16 0.77 78 Yuan to pH 6.0 -0.54 0.46 BaC12-TEA 34 0.77 67 Yuan to pH 6.5 -0.33 0.60 BaCl2-TEA 34 0.84 39 Incub to pH 5.5 -0.08 0.22 BaCl2-TEA 20 0.67 66 Incub to pH 6.0 -0.11 0.28 BaCl2-TEA 31 0.60 72 Incub to pH 6.5 -0,08 0.41 BaC1R-TEA 34 0.76 44



PAGE 1

15 Americans once again to accept the Al-clay theory first advanced by Veitch fifty years before. Forms of Soil Acidity Soil acidity can be divided into three forms, active acidity, exchangeable acidity, and total acidity. Active acidity is expressed as the pH of a soil. If the soil pH is <7.0 the soil is considered acidic, conversely if the soil pH is >7.0 the soil is considered alkaline. Soil pH governs, among others, nutrient availability to plants (Corey, 1973), as well as microbial activity in the soil (Stotzky, 1972). Exchangeable acidity is defined as that which is replaceable by the cation of a neutral unbuffered salt such as KC1, CaC12, or NaCl (Lin and Coleman, 1960). Theoretically this value measures the amount of acidity present in the soil at the pH of the soil or at least at the pH of the soil-salt solution. Exchangeable acidity is due almost entirely to monomeric A13+ (Coleman et al., 1959; Dewan and Rich, 1970) Yuan (1959) observed that H+ dominated over A13+ in the exchangeable acidity for some Florida soils, where organic matter was an important contributor to the cation exchange capacity (CEC). Thomas and Hargrove (1984) felt this might have been more apparent than real because some of the H Yuan observed was probably from the hydrolysis of Al3+ held in nonexchangeable form by the organic matter. Total acidity is that which is neutralized at a designated pH. Opinion has shifted toward the use of



PAGE 1

109 Soil Survey Staff. 1965. Soil Survey of Suwannee County, Florida. USDA, Soil Conservation Service, U.S. Government Printing Office, Washington, D.C. Soil Survey Staff. 1985. Soil Survey of Alachua County, Florida. USDA, Soil Conservation Service, U.S. Government Printing Office, Washington, D.C. Ssali, H., and J.K. Nuwamanya. 1981. Buffer pH methods for estimation of lime requirement of tropical acid soils. Commun. Soil Sci. Plant Anal. 12:643-659. Stevenson, F. J. 1982. Humus Chemistry. John Wiley and Sons, New York. Stotzky, G. 1972. Activity, ecology, and population dynamics of microorganisms in soil. CRC Critical Reviews in Microbiology. 2:59-137. CRC Press, Cleveland, Ohio. Tacke, B. 1897. Ueber die bestimmung der freien humussauren. Chemiker Zeitung. 21:174. Thomas, G. W. 1982. Exchangeable cations. In A. L. Page, R. H. Miller, and D. R. Keeney (ed.). Methods of Soil Analysis. Part II. Agronomy 9:159-165. Am. Soc. of Agron., Madison, Wis. Thomas, G. W., and W. L. Hargrove. 1984. The chemistry of soil acidity. In F. Adams (ed.) Soil Acidity and Liming. Agronomy 12:3-56. Am. Soc. of Agron., Madison, Wisc. Tisdale, S. L., and W. L. Nelson. 1975. Soil Fertility and Fertilizers. 3rd edition. Macmillian Publishing Co., Inc., New York. Tran, T. S., and W. van Lierop. 1981. Evaluation and improvement of buffer pH lime requirement methods. Soil Science. 131:178-188. Tran, T. S., and W. van Lierop. 1982. Lime requirement determination for attaining pH 5.5 and 6.0 of coarsetextured soils using buffer-pH methods. Soil Sci. Soc. Am. J. 46:1008-1014. van Lierop, W. 1983. Lime requirement determination of acid organic soils using buffer-pH methods. Can. J. Soil Sci. 63:411-423. Veitch, F. P. 1902. The estimation of soil acidity and the lime requirements of soils. J. Am. Chem. Soc. 24:1120-1128.



PAGE 1

69 equations for this type of adjustment has already been proposed (McLean, 1982). The drawback to this approach is that the pH-BU relationship is still employed in the calculation. As recently as 1985, Magdoff and Bartlett questioned the use of the pH-BU relationship for use in predicting LR. The wide spread of data about the regression line found for Florida soils (Fig. 3-2) confirms the notion that the pH-BU relationship is a very general one. The second approach would eliminate the use of the pH-BU relationship, using only the soil-buffer equilibrium pH. Figure 3-15 shows the soil-buffer equilibrium pH values regressed against incubation determined LR. This regression equation can then be used to obtain the amount of lime needed according to the final soil-buffer equilibrium pH. It is evident, however that the variability using this approach is greater than if the first one were used. Tran and van Lierop (1981) also tried using the soil-buffer equilibrium pH reading to predict incubation LR. They had expanded the soilbuffer ratio from 1:2 to 1:4 however. At the 1:4 soil-buffer ratio they found no significant difference between using buffer pH or converting to AE LR by means of the pH-BU relationship. At the 1:2 soil-buffer ratio they found that using the pH-BU relationship improved the r value (r = 0.78**) over using the soil-buffer pH (r = -0.63**).



PAGE 1

CHAPTER VI SUMMARY AND CONCLUSIONS Laboratory and field studies were conducted to evaluate two soil-test methods used to determine soil lime requirement (LR). One of the methods was the Adams-Evans (AE) LR method which is currently being used by the University of Florida Extension Soil Testing Laboratory (ESTL) to make liming recommendations for the state of Florida. The other method was the Yuan double buffer (DB) method which was specifically designed to measure the LR of Florida's sandy soils. Chapter III reported on a laboratory study where two widely accepted reference methods, CaCO3-soil incubation and BaCl2-TEA extractable acidity, were used to evaluate the two soil-test methods. Results from this study revealed that the Yuan DB method was a poor predictor of the incubation LR. The AE method was very highly correlated with but largely underestimated incubation LR. A modification of the AE method, utilizing a pH versus base unsaturation (BU) relationship derived from a group of 567 Florida soils, did not offer a practical advantage over the AE method. An alternative method (AE-F) of calculating the LR was developed which eliminated the need for using the pH-BU 99



PAGE 1

97 target pH level increased most likely related to the high pH of the reference method. The best relationship was found when the total acidity (TA) measured by the AE buffer (acidity determined from the soil-buffer equilibrium pH) was compared to the BaCl2-TEA method. There is no convenient way, using the available data, to use the BaCl2-TEA method as a reference method for low target pH values. Although there is a significant correlation, the r values are low and the C.V. values are very high. Some workers have used the suggestion by Peech (1965) that only 80% of the BaCl2-TEA measured acidity need be neutralized to attain a pH of 7.0. This generalization is based on the pH-BS relationship. It was felt that it would be inappropriate to employ this type of generalization to a reference method for use in this study. The intent of this study was to relate it back to the incubation study in Chapter III. It was hoped that by comparing the data, from both studies, to the various target pH levels, the AE LR method could be evaluated on this wider range of soils. The data do show that the AE method does provide a good index of acidity as determined by the reference as shown by the low C.V. and high r values for the regression of total acidity measured by both methods against each other. Results from Chapter III also showed a poor correlation between the BaCl2-TEA method and AE LR to the low target pH levels. However, an excellent relationship was found between



PAGE 1

For helping me initiate and conduct the field trials in Suwannee County I am thoroughly indebted to Dr. Mickey Swisher who really went out of her way to accommodate me. I am thankful to Dr. Tito French and the rest of the people involved with the North Florida FSR/E program for providing logistical support. I am also grateful to the Andrews, Chamberlain, and Barr families for allowing me to conduct the trials on their farms. Friends are essential in helping us make our everyday life bearable. The friends I've made here are no exception, and they all will be sorely missed. I hope that I will be able to keep in touch with all of them in the future. I owe Steve Grant a lot of thanks for allowing me to type this thesis on his PC, although at times I wondered if it might of have been easier to hire a typist. Finally, there is a very special friend, Kate Gieger, whose love and friendship have helped make it all worthwhile. iv



PAGE 1

80 Experimental Design All of the trials were set up as a randomized complete block design (RCBD). Individual plot size was 6 x 10 meters. There were five treatment levels consisting of no lime, no lime + 15 kg ha-1 Mg, 1.12 Mg ha-1 lime, 2.24 Mg ha-1 lime, and 4.48 Mg ha-1 lime. The limestone applied was a bagged dolomitic limestone with the following analyses: Effective neutralizing power, 89.6%; CaC03 equivalent, 104%; 100% passed through 10 mesh; 80% passed through 50 mesh; 75% passed through 100 mesh; 1730 lbs = 1 ton standard liming material. There were four replications resulting in a total of 20 plots per experiment. The no lime + 15 kg ha-1 Mg treatment was included for a yield study and will not be discussed in the observation of soil pH, over the duration of the trials. Land Preparation. Use, and Analyses Site 1. The site was established on November 16, 1983. Soil cores were taken by walking a zigzag pattern up one half of the experimental plot and down the other half. Twenty cores to a depth of 15 cm were taken in all. The soil was mixed and bagged for future analyses. This sample would be representative of the entire experimental field prior to the initiation of the experiment. The agricultural lime was hand applied to the respective plots. A blanket application of NH4NO3, (NH4)2SO4, and KCl fertilizer was also applied. The lime and fertilizer were



PAGE 1

non-specific pH-base unsaturation (BU) relationship and corrected for the underestimation of incubation LR by the AE method. The alternative method used the same laboratory determinations as the AE method but used different calculations. The AE method was also evaluated using a group of 98 soils sampled from various areas throughout Florida. The soils represented a wider range of physical and chemical characteristics than did the soils used in the incubation study. Results indicated that the AE method provided an accurate index of the soil LR on the 98 soils studied. Field studies revealed that current lime recommendations made by the ESTL underestimated actual field LR. The AE-F method also underestimated field LR which implied that the liming factor of 1.5 used by the ESTL was too low. The field study indicated liming factors of 3 and 5 for soils with relatively low and high initial pHw, respectively. More field studies are needed to obtain better liming factors for Florida's soils. Results indicated that although the AE LR method was appropriate for use on Florida's sandy soils, two adjustments in the calculation of AE LR would improve the method. First, the equation developed from the incubation study should be employed instead of the one from the original AE LR method. Second, the current liming factor used by the ESTL should be increased to at least 3. '4j Chairman xii



PAGE 1

46 To determine when the soils had reached equilibrium pHw measurements were taken on 14 samples after 6, 10, and 12 weeks of incubation (Table 3-5). At the end of 12 weeks, soils were transferred into plastic cups and allowed to air dry. Duplicate soil pH, and pHs measurements were then taken. Lime requirements for different target pH levels were determined using simple regression and correlation techniques. Final pH, or pHs were regressed versus additions of CaCO3 for each soil. The amount of CaCO3 needed to attain a desired pH for each soil was calculated from the appropriate regression equation. Results and Discussion PHF and Base Unsaturation The best relationship between pHw and base unsaturation (BU) for the 567 soils in the cited soil characterization reports was found to be: pHw = 7.34 -0.03 (%BU) (r = -0.752, n = 567) (Fig. 3-2). This regression equation will be referred to as AEmod in the following discussion. The equation of the original AE method for 347 Ultisols was: pH, = 7.79 -5.55 (BU) + 2.27 (BU2) (Adams and Evans ,1962). The authors did not include the r2 or C.V. values in their paper. pHand pHs Commonly used target pHw values are 5.5, 6.0, and 6.5. To allow for interpretation of the incubation pHs measurements, target pHs values were calculated that correspond to the aforementioned pHwvalues.



PAGE 1

37 1/ To determine the precision and accuracy of the AE and Yuan DB LR methods using a soil-CaCO3 incubation as a reference method. 2/ To determine if a modification of the AE method could improve, if needed, the AE method for Florida sandy soils. 3/ To determine if a relatively faster reference method, BaCl2-TEA extractable acidity, could serve as a reference method for evaluating the AE method on a larger amount and wider range of soils. Materials and Methods Published data (Calhoun et al., 1974; Carlisle et al., 1978, 1981, and 1985) were used to develop a relationship between pHw and BU for Florida soils. Base unsaturation was calculated by dividing the BaCl2-TEA (pH 8.2) extractable acidity by the sum of the extractable acidity and 1M NH40Ac exchangeable base values, and multiplying by 100. With the exception of Histosols, all soils described in the cited reports were included in the analysis, resulting in 567 pairs of data. The regression equation for this relationship was substituted for the curvilinear regression equation that Adams and Evans (1962) found for Alabama Ultisols. From the same source of data, a relationship between pH, and pHs for Florida soils was also developed. Target pH values for Florida soils are only given in terms of pHw.The regression equation determined from the pHw-pHsrelationship



PAGE 1

In dedication to Helga, Curt, Barb, and Sylvia Dierolf



PAGE 1

REFERENCES Adams, F. A. 1984. Crop response to lime in the southern United States. In F. Adams (ed.) Soil Acidity and Liming. Agronomy 12:211-265. Am. Soc. of Agron., Madison, Wis. Adams, F. A., and C. E. Evans. 1962. A rapid method for measuring lime requirement of Red-Yellow Podzolic soils. Soil Sci. Soc. Am. Proc. 26:355-357. Baker, A. S., and Yeh Moon Chae. 1977. A laboratory quick test for predicting the lime requirement of acid mineral soils. Tech. Bull. 88. Washington State University, Pullman, Wash. Barber, S..A. 1984. Liming materials and practices. In F. Adams (ed.) Soil Acidity and Liming .Agronomy 12:171-209. Am. Soc. of Agron., Madison, Wis. Benham, K. E. 1970. Exchangeable acidity, salt pH, and base saturation in some Missouri soils. Unpublished M.S. thesis. Univ. of Mo.-Columbia. Bohn, H. L., B. L. McNeal, and G. A. O'Connor. 1979. Soil Chemistry. John Wiley and Sons, New York. Bradfield, R., and W. B. Allison. 1933. Criteria of base saturation in soils. Int. Soc. Soil Sci., Trans. Comm. II (Copenhagen) A:63-79. Brown, I. C. 1943. A rapid method of determining exchangeable hydrogen and total exchangeable bases of soils. Soil Sci. 56:353-357. Brown, J. R., and J. R. Cisco. 1984. An improved Woodruff buffer for estimation of lime requirements. Soil Sci. Soc. Am. J. 48:587-592. Calhoun, F. G., V. W. Carlisle, R. E. Caldwell, L. W. Zelazny, L. C. Hammond, and H. L. Breland. 1974. Characterization data for selected Florida soils. Soil Science Department Research Report No. 74-1. University of Florida, Gainesville. 103



PAGE 1

14 pH 6.0 and 7.0. This allowed the individual buffering capacity of each soil to be determined. The soil acidity to be neutralized was determined by the buffering property of a soil which, in turn, was defined by dividing the difference in soil acidity neutralized in the two buffer systems by the difference of the two equilibrium pH values. McLean et al. (1978) borrowed Yuan's double buffer concept and applied it to the original SMP buffer method. They adjusted the two buffer solutions to initial pH values of 6.0 and 7.5. They found the double buffer procedure (SMP DB) to be more promisingas an improved method, especially on soils of low lime requirement where the SMP SB method was known to lack accuracy. Soil Acidity The forms and sources of soil acidity influence the buffering capacity of a soil. Through an increased understanding of soil acidity, scientists have been able to improve the rapid lime requirement methods. Soil scientists spent the first half of the Twentieth Century debating the nature of soil acidity. Jenny (1961) described this debate as a merry-go-round, that began with Al-clay theories initiated by Veitch in 1904. In 1922, Van der Spek promoted the concept of H-clays. This theory dominated for thirty years. In the early and mid fifties, N. T. Coleman and co-workers as well as P. L. Low, convinced



PAGE 1

21 consistent with field conditions should be used for agronomic evaluation (Gillman et al., 1983). Characteristics of Florida Soils A general knowledge of some of the properties of Florida's soils can give an indication of their buffering characteristics. Most of Florida's topsoils are sandy. The particle size fraction is often composed of greater than 95% sand. Clay contents of the topsoils are resultantly also low. Using published data accumulated by the Soil Characterization Lab at the University of Florida (Calhoun et al., 1974; Carlisle et al., 1978, 1981, and 1985) several generalizations of the properties of Florida's surface soils according to soil order can be made. Most of the Entisols and Spodosols have a clay content of less than 4% and a CEC (pH 7.0 NH40Ac) of less than 10 meq 100 g'1 soil. The Ultisols tend to have a slightly higher clay content but the CEC is still relatively low. Organic matter content for the Ultisols and Entisols tends to range from 1 to 2%. The Spodosols tend to have a higher organic matter with most soils ranging from 2 to 3% organic matter. The clay fraction is usually dominated by kaolinite, halloysite, gibbsite, quartz, and vermiculite (Fiskell and Carlisle, 1963). The organic matter is the most important contributor to the CEC of acid, sandy Florida virgin topsoils (Yuan et al., 1967). They found that organic matter contributed from 66 to 96% of the CEC depending on the soil order.



PAGE 1

41 were determined by the University of Florida Extension Soil Testing Lab (ESTL) (Rhue and Kidder, 1983) (Table 3-2). The Adams and Evans buffer pH was determined in duplicate by the ESTL (Rhue and Kidder, 1983). As mentioned previously, the AE method theoretically measures more of the soil acidity than needs to be neutralized to attain a pHw of 6.5. The total amount of acidity measured by the AE buffer solution will be termed AE total acidity. For comparison purposes this value can be expressed either as the pH of the buffer-soil solution or as g CaCO3 kg-1 soil needed to neutralize all of the measured acidity. Calculations of AE LR were made on a weight basis to allow direct comparison to incubation LR data, which were also computed on a weight basis. Each of the 34 soils were scooped with an 11 cc container, tapped 3 times, leveled with a straight edge, and weighed. This was repeated four times for each soil (Table 3-3). The mean value was found to be 1.32 g cm"3 and was used as an average for all conversions of volume to weight. The Yuan double buffer (Yuan DB) method was also employed to determine soil LR on all 34 soils (Yuan, 1974, 1976). Separate 50 ml portions, in duplicate, of the Yuan DB with pH values adjusted to 6.0 and 7.0 were added to separate 5 g soil samples. The equation given by Yuan allowed for calculation of the LR to various pH levels.



PAGE 1

113 Prior to returning overseas to continue working in the international development field, Tom wishes to return to East Germany in an attempt to visit with relatives and trace some of his genealogy.



PAGE 1

44 Extractable acidity by the BaCl2-TEA (pH 8.2) method (Soil Survey Staff, 1972) was determined in duplicate on all of the soils. The acidity measured by this method was expressed as the amount of base required to neutralize the acidity in g CaCO3 kg-1 soil. The BaCl2-TEA method was also used as a reference method. Each of the 34 soils was incubated in duplicate with six increments of carefully weighed 320 mesh reagent grade CaCO3. The CaCO3 additions were 0, 1/3, 2/3, 1, 4/3, and 5/3 times the AE LR to attain pHw 6.5 (Table 3-4). The AE LR was determined separately for each soil. One hundred grams of each soil were placed in an unsealed polyethylene bag. This bag was placed inside another polyethylene bag to protect against accidental soil loss. The CaCO3 was added to the dry soil and the bag was vigorously hand-mixed for three minutes. Eight ml of deionized water were added to each bag to approximate field capacity. The soil was again mixed until all of the soil appeared moistened. The bags were stored on a lab bench at room temperature for the duration of the incubation period. After six weeks the bags were vigorously shaken to maximize soil-CaCO3 contact. To prevent large moisture losses while also providing for some gas exchange, outer bags were sealed but had a 5 mm hole punched in the side. Some bags were periodically weighed over the 12-week incubation period and minimal moisture loss of about 1 ml per sample was confirmed.



PAGE 1

75 Thus by knowing the LR, which can be determined by a buffer method, and the actual field LR, the liming factor can be determined. The AE method uses a liming factor of 1.5, which was obtained from work done by Pierre and Worley (1928), Schollenberger and Salter (1943), and research station data. Yuan (1976) recommended that a liming factor of 1.2 to 1.8 be employed for his method, on Florida soils. The exact value would be dependent on the limestone specifications, desired soil pH, and the amount of lime to be applied. The range of liming factors found by researchers varies widely. Walker (1952) found that 2 to 3 times as much lime was needed in the field as was calculated from complete mixing in the laboratory. Incubation studies are valuable in that they allow for the correlation of a soil test method for a large number of soils with a broad range of different characteristics. Final adoption of any soil test method should, however, be preceded by field calibration studies to determine what the actual liming factor should be for Florida soils. The field trials discussed in this chapter will attempt to give some indication of the applicability of a liming factor of 1.5, as is assumed by the AE method. Field experiments were conducted on two farms and a research station to determine the long-term effects of increasing rates of lime on soil pH. The main objective of this study was to obtain