|
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
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
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):
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.
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
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.
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:
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.
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.
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.
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
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
|
RSS
HIDE
