oEPA
United States
Environmental Protection
Agency
Office of Acid Deposition,
Environmental Monitoring and
Quality Assurance
Washington DC 20460
EPA/600/4-87/041 b
December 1987
Research and Development
Direct/Delayed Response
Project: Field
Operations and
Quality Assurance
Report for Soil
Sampling and
Preparation in the
Southern Blue Ridge
Province of the United
States
Volume II. Preparation
-------
EPA/600/4-87/041b
December 1987
Direct/Delayed Response Project:
Field Operations and Quality Assurance Report
for Soil Sampling and Preparation in the
Southern Blue Ridge Province of the United States
Volume II Preparation
M.F. Haren and R,D, Van Re mo rt el
A Contribution to the
National Acid Precipitation Assessment Program
U.S. Environmental Protection Agency
Begion 5, Library (5PL-16)
230 S. Dearborn Street,. Baom
if* £0604
U.S. Environmental Protection Agency
Office of Modeling, Monitoring Systems, and Quality Assurance
Office of Ecological Processes and Effects Research
Office of Research and Development
Washington, D.C. 20460
Environmental Monitoring Systems Laboratory, Las Vegas, Nevada 89193
Environmental Research Laboratory, Corvallis, Oregon 97333
-------
Notice
The information in this document has been funded wholly or in part by the United States
Environmental Protection Agency under Contract Number 68-03-3249 to Lockheed Engineering &
Sciences Company. It has been subject to the Agency's peer and administrative review, and it has
been approved for publication as an EPA document.
The mention of trade names or commercial products does not constitute endorsement or
recommendation for use.
This document is one volume of a set which fully describes the Direct/Delayed Response
Project, Southern Blue Ridge and Northeast soils surveys. The complete document set includes the
major data report, quality assurance plan, analytical methods manual, field operations reports, and
quality assurance reports. Similar sets are being produced for each Aquatic Effects Research
Program component project. Colored covers, artwork, and the use of the project name in the
document title serve to identify each companion document set.
The correct citation of this document is:
Haren, M. F. and R. D. Van Remortel. 1987. Direct/Delayed Response Project: Field Operations and
Quality Assurance Report for Soil Sampling and Preparation in the Southern Blue Ridge
Province of the United States EPA 600/4-87/041b. U.S. Environmental Protection Agency, Las
Vegas, Nevada. 24 pp.
-------
Abstract
The Direct/Delayed Response Project Soil Survey includes the mapping, characterization,
sampling, preparation, and analysis of soils in order to assess watershed response to acidic
deposition within various regions of the United States. Soil samples from the Southern Blue Ridge
Province were transported to preparation laboratories for processing before delivery to analytical
laboratories. This document summarizes procedural and operational compliance with the protocols
used at the preparation laboratories. Deviations from the protocols and difficulties encountered
are identified and discussed. Recommendations are made for program improvement.
A review of the soil data suggests that the integrity of the soil samples was maintained
during the preparation activities. In most cases, laboratory personnel adhered to protocols.
This report was submitted in partial fulfillment of contract number 68-03-3249 by Lockheed
Engineering and Sciences Company, Inc. under the sponsorship of the U.S. Environmental Protection
Agency. The report covers a period from March, 1986 to December, 1986, and work was completed
as of October, 1987.
-------
-------
CONTENTS
Page
Notice ii
Abstract iii
Figures vii
Tables viii
Acknowledgments ix
1 Introduction 1
Overview 1
Objectives 2
2 Sample Preparation Methods and Analysis 5
Sample drying 5
Moisture determination 5
Crushing and sieving 5
Rock fragment determination 5
Soil homogenization 5
Qualitative test for inorganic carbon 5
Bulk density determination 6
3 Preparation Laboratory Operations 7
Sample receipt and storage 7
Equipment inventory 7
Sample drying 8
Moisture determination 8
Crushing and sieving 8
Rock fragment determination 8
Soil homogenization 9
Qualitative test for inorganic carbon 9
Bulk density determination 9
Sample shipment 9
Record keeping 10
4 Quality Assurance/Quality Control 11
Design components 11
Training 11
Communications 11
Data quality objectives 11
On-site systems audits 12
-------
Data evaluation 12
Quality assurance samples 13
Method of estimating analytical precision 13
Precision results for rock fragment determination 14
Precision results for bulk density determination 14
Completeness results 15
5 Conclusions and Recommendations 16
General recommendations 16
Sample receipt 16
Equipment Inventory 16
Sample drying 17
Moisture determination 17
Soil homogenization 17
Rock fragment determination 17
Qualitative test for inorganic carbon 17
Bulk density determination 18
Sample shipment 18
Quality assurance and quality control samples 18
Record keeping 19
Design components 19
References 24
Appendix A Preparation Laboratory Manual for the
Direct/Delayed Response Project 25
-------
Figures
Number Page
1 Form 101 - Master data form 3
2 Form 102 - Shipping form 4
3 Sample receipt form 20
4 Bulk density raw data form 21
5 Bulk sample processing form 22
-------
Tables
Number Page
1 Precision estimates for rock fragments 14
2 Precision estimates for bulk density 14
-------
Acknowledgments
Critical peer reviews by the following individuals are gratefully acknowledged: W. Banwart,
University of Illinois, Urbana, Illinois; and J. S. Lohse, Illinois Department of Agriculture, Bureau
of Farmland Protection, Springfield, Illinois.
The following individuals provided guidance in the development of statistical analyses:
T. Starks, Environmental Research Center, University of Nevada, Las Vegas, Nevada; and
J. E. Teberg and M. J. Miah, Lockheed Engineering & Sciences Company, Las Vegas, Nevada.
Critical reviews by the following individuals were instrumental in the preparation of this
document and are gratefully acknowledged: C. J. Palmer, Environmental Research Center, University
of Nevada, Las Vegas, Nevada; J. C. Foss, University of Tennessee, Knoxville, Tennessee; B. R.
Smith, Clemson University, Clemson, South Carolina; J. J. Lee, U.S. Environmental Protection
Agency, Environmental Research Laboratory, Corvallis, Oregon; D. S. Coffey, Northrop Services, Inc.,
Corvallis, Oregon; and J. K. Bartz, M. D. Best, G. E. Byers, W. H. Cole, and M. L Papp, Lockheed
Engineering & Sciences Company, Las Vegas, Nevada.
-------
-------
Section 1
Introduction
Overview
The U.S. Environmental Protection Agency
(EPA), in conjunction with the National Acid
Precipitation Assessment Program (NAPAP),
has designed and implemented a research
program to predict the long-term response of
watersheds and surface waters in the United
States to acidic deposition. Based on this
research, each watershed system studied will
be classified according to the time scale in
which it will reach an acidic steady state,
assuming current levels of acidic deposition.
The Direct/Delayed Response Project (DORP)
was designed as the terrestrial complement to
the aquatic resources program.
As part of the DORP, the services of two
preparation laboratories were obtained through
interagency agreements to receive and process
soil samples collected from the Southern Blue
Ridge Province (SBRP) of the United States
and to perform preliminary analyses on these
samples. Laboratories located at the Univer-
sity of Tennessee in Knoxyille, Tennessee and
at Clemson University in Clemson, South
Carolina were selected for these tasks be-
cause of the proximity of each laboratory to
the sampling sites and analytical experience
with soils of the region.
Each laboratory was supervised by a
university faculty member and a laboratory
manager. The manager was responsible for
ensuring that the integrity of the soil samples
was maintained after the samples were deliv-
ered to the preparation laboratory. Both
laboratory managers had received university
degrees in soil science and the other labora-
tory personnel had received prior training in
soil science. All participants were required to
comply with specified protocols, as outlined in
Appendix A.
Soils processed at the preparation labo-
ratories were collected from sampling sites
located in Georgia, South Carolina, North
Carolina, and Tennessee and from special
interest watersheds in North Carolina and
Virginia. Upon receipt of bulk soil samples
and clod samples from the sampling crews,
laboratory personnel performed the following
analyses on the samples:
• Moisture determination (air-dry mois-
ture content)
• Rock fragment determination (2- to
4.75-millimeter and 4.75- to 20-milli-
meter fractions by
weight)
• Qualitative test for inorganic carbon
• Bulk density determination by clod
analysis
Laboratory personnel prepared analytical
samples derived from homogenized, air-dry
bulk samples. The analytical samples were
labeled and were organized according to their
parent pedons. Multiple batches were assem-
bled, each containing no more than 37 routine
and duplicate samples, two natural audit pairs,
and a preparation duplicate. The duplicates
and audit pairs were included in each batch
for quality assurance (QA) purposes. The
samples were randomized within each batch
by the laboratory manager. The assembled
batches were shipped to various analytical
laboratories contracted by EPA for additional
analyses.
-------
The preparation laboratories were re-
sponsible for completing Form 101 (see Figure
1) for each batch of samples. This data form
contained the results of all analyses performed
on the soil samples by preparation labora-
tories prepared a shipping form, Form 102 (see
Figure 2), that was included with each batch
of samples shipped to the analytical labora-
tories. The shipping form identified each
sample by a batch and sample number accom-
panied by a mineral or organic soil designa-
tion. This procedure disguised the originating
pedon and horizon of each field sample, as
well as the identity of the QA samples.
All information, e.g., sample labels,
received from the sampling crews and raw
data from the preparation laboratory analyses
were documented in log books. The protocols
specified that completed log books were to be
sent to EMSL-LV QA staff for use during the
data verification process.
QA and quality control (QC) measures
were applied to maintain consistency in soil
preparation protocols and to ensure that the
soil sample analyses would yield results of
known quality. Laboratory personnel received
training in the preparation procedures and
analytical methods. QA representatives from
EMSL-LV and ERL-C conducted on-site sys-
tems audits of the preparation laboratories.
Weekly communication between the QA staff
and laboratory personnel was established to
identify, discuss, and resolve issues.
Objectives
This document reports the results of the
preparation laboratory operations and QA
program for the SBRP Soil Survey. Information
concerning the specified protocols for the
preparation laboratories can be found in
Appendix A.
The following sections contain detailed
information concerning the preparation labora-
tory methods, operations, and data quality. A
series of recommendations for improving both
the quality and efficiency of preparation labo-
ratories in future DDRP soil surveys are
included.
-------
NATIONAL ACID DEPOSITION SOIL SURVEY (NADSS)
FORM 101
DATE RECEIVED
BY DATA MGT.
O O M M M V ¥
Set ID
Data Sa
Dale RE
Dale Pr
No. ol S
SAMPLE
NO.
01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
16
17
13
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
Balr.h ID
CrpwID
Prop 1 ah ID
1 an Sel Snnl To
Dale Shinned
Cnived
pp Cnmptplpd
amoles ._ .__
sire 10
Signature ol Preparation
Comments:
SAMPLE CODE
SET
ID
COARSE
FRAGMENTS
%
CF
AID DRIED
MOISTURE
S
W H50
INORO.
CARBON
(1C)
1* fES
N ' NO
BULK
DENSITY
VCC
ahnr.ilnry Manarjnr'
COLO - (»t. c
Figure 1. Form 101 - Master data form.
-------
NATIONAL ACID DEPOSITION SOIL SURVEY (NADSS)
SHIPPING FORM 102
DATE RECEIVED
BY DATA MGT.
~
U U M V V
O O M M U f V
Prop 1 .ih in Ham Rnrmvort
Batch ID , Data Shipped
Analylip.nl I ah in
SAMPLE NO.
01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
16
17
10
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
AIR-OHIEO MOISTURE
•*
W USD
Signature of Preparation Laboratory Manager:
Comments:
INORGANIC CARBON (1C)
T . ves
N » NO
COARSE FRAGMENTS SHIPPED?
(CHECK T IF YES)
Figure 2. Form 102 - Shipping form.
-------
Section 2
Sample Preparation Methods and Analysis
The detailed methods and analytical
procedures used in soil preparation activities
for the SBRP soil survey are given in the
protocols (Appendix A). Brief explanations of
the specified methods and procedures for the
preparation laboratory tasks are given below,
including one correction to the protocols. The
procedures are discussed in sequential order
of performance by the preparation laboratories.
Sample Drying
Bulk samples are spread out on large
sheets of paper to air dry. Where necessary,
several sheets of paper are placed underneath
the samples to absorb excess moisture and
the paper is replaced periodically for faster
drying. Laboratory personnel occasionally stir
the samples to encourage uniform drying.
Moisture Determination
A 15- to 20-gram air-dry subsample is
placed in a drying dish and is weighed. The
dish is placed in a drying oven for a minimum
of 16 hours, 105 *C for mineral samples and
60 "C for organic samples. The sample is
allowed to cool in a desiccator for 30 minutes
and is reweighed. This procedure is repeated
until the difference between successive daily
moisture contents is less than 2.5 percent.
The final air-dry moisture content is recorded.
(Note: Field moisture was not a measured
parameter for the SBRP survey.)
Crushing and Sieving
After recording the weight of the air-dry
bulk sample, the soil peds are crushed to
allow passage of the less than 2-mm soil
fraction through a No. 10 sieve. A wooden
rolling pin or a rubber stopper is used to crush
the peds, depending on the consistence of
rock fragments in each sample.
Rock Fragment Determination
Rock fragments remaining from the
sieving procedure are retained on either one of
two sieves. Fragments retained on the No. 4
sieve constitute the 4.75- to 20- millimeter
coarse pebble fraction (fragments larger than
20 millimeters in diameter were sieved by the
sampling crews). Fragments retained on the
No. 10 sieve constitute the 2.0- to 4.75-milli-
meter fine pebble fraction. Gravimetric analy-
sis is used to determine the percentage by
weight of each fraction in the soil samples.
Soil Homogenization
A Jones-type riffle splitter is used for
homogenizing the samples. The less than 2-
millimeter fraction is deposited evenly across
the baffles of the riffle splitter and is chan-
neled into two receiving pans. After seven
passes through the riffle splitter, the entire
sample is combined for a final pass through
the riffle splitter. The material in one pan is
placed into a plastic sample bag for archiving,
and the material in the other pan repeatedly
passes through the riffle splitter until a one-
kilogram subsample is obtained.
Qualitative Test for Inorganic
Carbon
One gram of air-dry soil is placed in the
well of a porcelain spot plate, is saturated
with deionized water, and is stirred to release
any entrapped air. The sample is observed
through a microscope or a stereoscope in
order to detect an effervescent reaction when
three drops of 4N HCI are added. Two types
-------
of QC samples are used to determine the
detection limit and to qualitatively calibrate the
test.
Bulk Density Determination
Replicate soil clods, usually three per
horizon, are collected by the sampling crews
(Blume et al., 1987). Initially, the clods are
weighed at the laboratory and are dipped in a
1:7 saran:acetone mixture. The clods are
suspended from a line, are allowed to dry
briefly, and are reweighed. The dipping proce-
dure is repeated until each clod is assumed to
be impervious to water.
Approximately 800 milliliters of deionized
water in a one-liter beaker is de-gassed by
boiling, is allowed to cool to room tempera-
ture, and is tared on a balance. Each clod is
submerged in the water and the increase in
weight is recorded. The clods are oven-dried
for 48 hours, cooled, and reweighed. A two-
hour heat treatment in a 400 *C muffle furnace
allows the saran to vaporize, and the clods are
cooled and again are reweighed. Each clod is
crushed and is passed through a 2- millimeter
sieve to determine percent by weight of rock
fragments. This figure is used to adjust the
bulk density calculation for rock fragment
content.
Bulk density is defined as the mass of
dry soil per unit volume, including pore space,
and expressed as grams per cubic centimeter
(g/cm3). Bulk density normally ranges from 1.0
to 1.8 g/cm3 in mineral soils (USDA-SCS, 1983).
Laboratory personnel at the Clemson
preparation laboratory discovered that one of
the bulk density algorithms was given incor-
rectly in the protocols. The algorithm used to
estimate the air-dry saran weight was origi-
nally written as:
MTS =
X (M. -
a - 1
where: MTS = air-dry saran weight
X = total number of coatings (field
and laboratory)
Ma = clod weight after final coating
M, = initial clod weight after
unpacking
a = number of laboratory coatings
The correct algorithm is as follows:
X (M. - MO
MTS =
a
All participants were informed of the
error as soon as it was discovered. The
laboratory data forms immediately were amen-
ded to show the corrected values.
6
-------
Section 3
Preparation Laboratory Operations
This section describes preparation labo-
ratory operations, including difficulties with the
procedures and deviations from the protocols.
The laboratories are characterized as Labora-
tory A and Laboratory B to prevent disclosure
of their identity, except where necessary to
clarify an activity.
Sample Receipt and Storage
Due to budgeting and logistical difficul-
ties, neither preparation laboratory was admin-
istratively operational when sampling began.
The laboratory director at Laboratory B retired
from the position soon after preparation activ-
ities were initiated, causing some initial confu-
sion with sample tracking and processing.
Each laboratory provided the sampling
crews with convenient access to cold storage.
A sample receipt log book was kept at each
facility to allow sampling crews to log in the
samples. Each laboratory was responsible for
checking that all samples delivered by the
sampling crews were recorded in the log book.
Entries were checked for legibility and
accuracy.
The temperature of the storage facilities
was maintained at the contract-specified 4 *C,
with one exception. Laboratory B had a tem-
porary refrigeration failure that was reported
to the QA staff immediately. The refrigeration
unit was repaired within approximately six
hours, and the highest temperature measured
during the failure was 14 °C. The lapse in
cooling may have had some effect on micro-
bial activity in field-moist samples stored at
the facility, although studies were not con-
ducted to identify this effect. Analytical data
for the affected samples processed at Labora-
tory B will be flagged in the verified data base,
denoting possible contamination. If time
permits, the data will be evaluated during
verification to determine whether or not there
are suspect values that indicate contamination
may have occurred. Although the effects are
expected to be negligible, organic transforma-
tions involving carbon, nitrogen, or sulfur may
have occurred. This issue will be discussed in
the QA report for the analytical data.
In order to provide an equal distribution
of incoming samples and to improve the
efficiency of available cold storage and labor,
34 bulk samples and 72 corresponding clods
were transferred from the University of Ten-
nessee to Clemson University. The samples
were packed in coolers and were transported
by truck. The samples were returned to cold
storage within eight hours. This transfer
provided each laboratory with approximately
400 bulk samples and their respective clods.
There was some confusion concerning
the proper dispensation of practice samples
that were delivered to Laboratory B at the
request of the laboratory director. These
practice samples had been collected by one of
the sampling crews in preparation for routine
sampling. The samples were delivered to
Laboratory B and were logged into the sample
receipt log book. The samples were not
identified as practice samples, therefore labor-
atory personnel treated the samples as if they
were routine samples and processed them
accordingly. The error was discovered after
the samples had undergone soil analysis at
the analytical laboratories. Data for the prac-
tice samples were later removed from the data
base.
Equipment Inventory
The laboratory managers were respon-
sible for tracking the distribution of equipment
-------
to the sampling crews. Each preparation
laboratory stored the equipment in a locked
room with limited access. The sampling crews
usually obtained supplies while samples were
being delivered to the cold storage facility.
Laboratory managers generally were notified in
advance of a sampling crew's arrival to ensure
that the crews had access to the supply room.
The sampling crews were asked to record in
the equipment log book all supplies taken.
A complete inventory was performed
monthly at each preparation laboratory. Equip-
ment shortages were reported to EMSL-LV
during the weekly conference call. Laboratory
personnel were tasked with providing the
saran and acetone used in mixing a solution
for coating clods in the field. They also were
responsible for providing gel packs used in the
refrigeration of bulk samples during transport.
It was decided at the training workshop
that certain sampling hardware would be
distributed to the sampling crews via the Soil
Conservation Service (SCS) state offices
instead of the preparation laboratories. After
sampling was completed, difficulties were
encountered in tracking and recovering this
equipment because the laboratory managers
were not provided with a master inventory
sheet listing all equipment provided to the
crews. Additional difficulties occurred when
sampling crews utilized both preparation
laboratories as supply points.
After the soil preparation activities were
completed, leftover supplies were inventoried
and were sent to EMSL-LV for storage. Cam-
eras used to photograph the sampling sites
were turned over to the SCS state offices by
some of the sampling crews. Although there
were instances of misplaced sampling hard-
ware, most of the equipment eventually was
located. A complete list of the types of equip-
ment supplied to the crews is provided in
Appendix A.
Sample Drying
Although there were no deviations from
the specified protocols for sample drying,
concerns were raised about air-drying the soils
during humid or rainy weather. Drying areas
for both laboratories were located in buildings
lacking humidity control: one used a green-
house and the other used a prefabricated
metal structure. Samples occasionally took as
long as three weeks to dry. Long-term expo-
sure of the samples during humid weather
provided a greater opportunity for contamina-
tion. Both laboratories loosely covered the
drying samples with kraft paper to reduce
airborne contamination.
Moisture Determination
One significant deviation from the speci-
fied protocols for the air-dry moisture deter-
mination occurred. Laboratory B obtained a
convection oven after the sample processing
had begun. Therefore, the initial air-dry mois-
ture determinations at this laboratory were
made by an alternate method approved in
advance by QA staff. The alternate method
was to weigh a known quantity of sample and
allow it to dry overnight under the same condi-
tions as the bulk sample. The sample then
was reweighed. If the change in sample
weight was less than 2.5 percent absolute
over a two-day period, it was presumed that
the soil was ready for processing. If not, the
procedure was repeated until the variation in
weight was less than 2.5 percent.
Crushing and Sieving
There were no deviations from the speci-
fied protocols for sample crushing and sieving.
Laboratory A generally used the rolling pin
method. Because of the soft, weathered state
of rock fragments in its samples, Laboratory B
generally used the specified alternate method,
which utilized a rubber stopper to crush indivi-
dual soil peds. This procedure allowed the
passage of soil through the sieves without
disturbing the fragments. The rolling pin
method occasionally was used by Laboratory
B for those samples containing hardened peds
or few fragments, and a notation of this was
made in the sample processing log book.
Rock Fragment Determination
There were no deviations from the speci-
fied protocols for rock fragment determination.
Laboratory B washed all rock fragments in
deionized water to remove adhered soil before
calculating the percentages. Both laboratories
bagged and labeled the fragments and
awaited further instructions for shipment.
8
-------
Soil Homogenization
There were no deviations from the speci-
fied protocols for soil homogenization. During
October, 1986 it was discovered that there was
insufficient audit material remaining to con-
tinue processing one-kilogram audit samples.
As a result, the protocols were modified and
the laboratory managers were advised to begin
preparing 500-gram analytical samples instead
of one-kilogram samples. All batches shipped
from Laboratory A contained 500-gram sam-
ples. Because many of the batches from
Laboratory B already had been shipped, only 5
of the 14 total batches at Laboratory B con-
tained 500-gram samples.
Both laboratory managers voiced concern
about rehomogenizing the bulk samples to
obtain subsamples for mineralogical study.
Repetitious soil homogenization could mis-
represent the particle-size distribution because
of the potential for loss of clay and fine silt
particles during riffle-splitting. The list of
samples selected for mineralogical analysis
was sent to the preparation laboratories after
soil processing was well underway. Bulk
samples had to be retrieved from storage,
rehomogenized, subsampled, relabeled, and
returned to storage. This exercise was very
time consuming and might have been avoided
through better planning and communication.
Qualitative Test for Inorganic
Carbon
There were no deviations from the speci-
fied protocols for the inorganic carbon test.
Inorganic carbon was not detected in any
samples collected.
Bulk Density Determination
There were no deviations from the speci-
fied protocols for the bulk density determina-
tion. Clod analysis was chosen as the meth-
od for determining bulk density, despite a few
disadvantages. Obtaining replicate clods from
dry, loose, or extremely wet soils and from
horizons containing many rock fragments often
can be difficult or impossible. Another concern
is that the bulk density values that are ob-
tained may be higher than the average bulk
density of the soil horizon they represent,
because sampling could be biased toward the
collection of firmer, more coherent clods cap-
able of withstanding disturbance during sam-
pling and transport.
Variations in bulk density measurements
could occur when clods are allowed to air dry
after multiple saran coatings. Because the
saran is permeable to water vapor, the clod
weight could be affected by moisture loss
from prolonged drying. This situation did
occur at Laboratory A during the early stages
of sample processing, although the actual
number of affected clods is unknown. Thirty-
nine clods at Laboratory B were discarded
because a weak coating solution used in the
field rendered the clods unsuitable for analysis
upon arrival at the laboratory.
According to the log books, three floating
clods were identified. Each clod was forcibly
submerged to allow a bulk density determina-
tion to be made.
Sample Shipment
The preparation laboratories were provi-
ded with shipping materials used for packag-
ing the samples and with a express mail
charge number for overnight shipment of the
samples to the analytical laboratories. Labor-
atory personnel were instructed to assemble
batches containing as many as 37 routine
samples and field duplicates. Each batch also
contained one preparation duplicate prepared
by the laboratory and two pairs of natural
audit samples provided by QA staff at EMSL-
LV. The laboratory managers were responsible
for randomizing the samples within each batch
and assigning sample numbers. The individual
samples were tagged with the appropriate
label B and were packed in cardboard boxes
for shipment to the designated analytical
laboratory.
QA representatives routinely called one or
two days in advance to notify the preparation
laboratory manager that a shipment was due
for delivery to an analytical laboratory. Labor-
atory B expressed some dissatisfaction with
the amount of advance notice received. In a
few cases, notification was given the day
before batches were to be shipped. Labora-
tory A was not directly affected because their
batches generally were sent in two mass
shipments.
-------
Because of insufficient detail in the
protocols, there was some confusion in the
handling of the audit samples. Later written
instructions stated that the audit samples
should be identical in appearance to the rou-
tine samples, although the audit samples had
arrived at the preparation laboratories with
numbers on both the outer canvas bags and
inner plastic bags. The integrity and ano-
nymity of the audit samples were jeopardized
because Laboratory A left the audit samples in
their original bags and Laboratory B re-bagged
the audit samples.
Record Keeping
Both preparation laboratories were
provided with log books to use for recording
data. Each laboratory manager was instructed
to organize log books containing the following
information:
• Label A - This log book contained
labels that originally were filled out by
the sampling crews and were affixed
to the inner bag of the bulk samples.
Laboratory B placed these labels on
looseleaf paper and organized them
into a binder. Laboratory A did not
submit their labels, and later efforts
to locate them were unsuccessful.
• Clod Label - This log book contained
labels that were filled out by the
sampling crews and affixed to the
prepared clods. Laboratory A did not
submit their labels, and later efforts
to locate them were unsuccessful.
• Sample Receipt - This log book was
filled out by the sampling crews upon
delivery of samples to each
laboratory.
• Equipment - The sampling crews
were asked to list the field supplies
they obtained at the laboratories. The
laboratory managers were responsible
for tracking and confirming the equip-
ment inventories.
• Percent Moisture - This log book
contained raw data from the air-dry
moisture analyses.
• Percent Rock Fragments - This log
book contained raw data from the
rock fragment analyses.
• Bulk Density - This log book con-
tained raw data from the bulk density
analyses.
• Inorganic Carbon - This log book
contained raw data from the test for
inorganic carbon.
• Sample Processing - This log book
tracked the progress of the soil sam-
ples through the various preparation
activities.
Because a standard format for each log
book was not specified, there was variation
between the laboratories. As a result, verifica-
tion of the data took more time than was
expected. Laboratory A assigned internal
laboratory numbers to their samples, making
it difficult for QA personnel to match these
numbers to the original sample codes.
10
-------
Section 4
Quality Assurance/Quality Control
A specified QA/QC program must be
followed during the course of survey activities
to ensure that the resulting data are of known
quality. The QA/QC program for the SBRP soil
survey consisted of design and evaluative
components that aided survey participants in
obtaining samples and producing data that
meet the needs of end users.
The QA/QC design for the preparation
laboratories included training personnel in the
protocols to be followed, establishing a com-
munications network, assessing data quality,
and performing on-site systems audits. The
data were evaluated systematically using
analytical data from the clod replicates and
from the QA duplicate samples that were
included in each batch of routine samples.
The following sections explain aspects of the
QA/QC program in relation to preparation
laboratory activities.
Design Components
Training
Preparation laboratory personnel at-
tended a regional workshop held in Knoxville,
Tennessee, from March 18 through 20, 1986.
The purpose of the workshop was to review
the laboratory protocols and discuss key
activities. Training of the participants estab-
lished a basis for consistency between the
laboratories, thereby increasing the likelihood
that the data would be of comparable quality.
Communications
Weekly conference calls assisted in
keeping the preparation laboratories operating
efficiently and consistently by providing a
forum that allowed potential difficulties to be
identified, discussed, and resolved. Prepara-
tion laboratory managers, QA personnel, and
scientists involved in the study of soil miner-
alogy participated in these calls. Issues
discussed during the conference calls included
supply shortages and clarification of proce-
dures, e.g., sample labeling, record keeping,
and drying of clods. The laboratory manager
at Laboratory A stated that the conference
calls were not as beneficial as routine calls for
dealing with specific issues.
Data Quality Objectives
Data quality objectives for the prepara-
tion laboratories were not established during
the SBRP soil survey. The preparation labora-
tories were assessed according to the follow-
ing data characteristics:
• Precision and accuracy - These are
quantitative measurements that esti-
mate the amount of variability and
bias inherent in a given data set.
Precision refers to the level of agree-
ment among repeated measurements
of the same parameter. Accuracy
refers to the difference between an
estimate based on the analytical
results and the true value of the
parameter being measured.
• Representativeness - This refers to
the degree to which the collected data
accurately reflect the population or
medium that is sampled.
• Completeness - This refers to the
amount of data that is successfully
collected with respect to the amount
intended in the design. A defined
percentage of the intended amount
must be successfully collected for
11
-------
conclusions based on the data to be
valid. Lack of data completeness
may reduce the precision of esti-
mates, may introduce bias, and may
reduce the level of confidence in the
conclusions.
• Comparability - This refers to the
similarity of data from different
sources included in a single data set.
If more than one laboratory is ana-
lyzing samples, uniform procedures
must be used to ensure that the data
from different sources are based on
measurements of the same
parameter.
These five data quality characteristics
were identified in the DDRP QA Plan, and their
application to preparation laboratory activities
were stated (Bartz et al., 1987). A brief des-
cription of the specific characteristics follows:
• Precision and Accuracy - The prepara-
tion laboratory combines sets of field
samples into one batch containing a
maximum of 39 routine and duplicate
samples. After processing, i. e., air
drying, crushing, sieving, and ho-
mogenization, one bulk sample is split
into two subsamples which are term-
ed preparation duplicates. Compari-
son of physical and chemical data for
these duplicates allows evaluation of
the subsampling procedure.
• Representativeness - Each bulk sam-
ple is processed by a preparation
laboratory to obtain a homogeneous
sample. Homogenization is accomp-
lished by passing the sample through
a Jones-type riffle splitter at least
seven times. The riffle splitter also is
used for subsampling. All samples
not being processed are stored at 4
°C by the preparation laboratory.
• Completeness - Each batch of sam-
ples sent to a contractor analytical
laboratory includes the preparation
duplicates.
• Comparability - All preparation labora-
tories process bulk samples accor-
ding to the protocols. Strict adher-
ence to protocols should result in
comparability among preparation
laboratories.
Precision is estimated in this QA report
by evaluating field duplicate data from the
rock fragment analyses and replicate data
from the bulk density analyses. Additional
precision estimates using analytical data from
the field duplicates and from the preparation
duplicates will be documented in the QA report
on the analytical laboratory data.
Accuracy cannot be assessed because:
(1) true values for the parameters in question
are not known, (2) quality control calibration
samples were not used in the determination of
bulk density, (3) it was not known how to
provide quality control calibration samples to
evaluate the rock fragment analyses, and (4)
it was not known how to provide audit sam-
ples for the rock fragment and bulk density
analyses. Relative accuracy based upon an
interlaboratory study will be addressed in the
QA report on the analytical laboratory data.
Representativeness will be addressed in
the QA report on the analytical data by using
data from the preparation duplicates to as-
sess the subsampling procedure performed by
each laboratory.
Completeness is measured by each
laboratory's performance of the analyses and
processing tasks for all samples assigned.
Comparability is evaluated using the
precision data from preparation duplicates.
This evaluation will be documented in the QA
report on the analytical laboratory data.
On-Site Systems Audits
Pre-sample audits of both preparation
laboratories were conducted during the Spring
of 1986. These visits were designed to as-
sess the laboratory facilities, equipment, and
staff before any samples were received or
processed. Laboratory A was audited by a
QA representative from EMSL-LV, and Labora-
tory B was audited by a QA representative
from ERL-C.
Laboratories A and B later were audited
by a QA representative from EMSL-LV in June,
1986. At Laboratory A, the QA auditor ob-
served sample drying and toured the cold
12
-------
storage and laboratory facilities. At Laborato-
ry B, the auditor observed the early stages of
sample preparation, including the sample
drying, sample crushing, and soil homogeniza-
tion procedures.
Laboratory A had not been covering their
drying samples with paper, thereby leaving the
samples vulnerable to airborne contamination.
The laboratory manager was instructed to
loosely cover all samples in the future. Labor-
atory B was not calibrating the thermometer
used for recording the temperature of the
water in bulk density calculations. The labora-
tory manager was instructed to check the
accuracy of the thermometer at least once a
month. It also was discovered that Labora-
tories A and B did not have the required Class
S weights for calibration of the balance.
These weights were obtained shortly
thereafter.
The log books were inspected by the QA
auditor, who observed pencil entries in the
sample receipt log book at Laboratory B. To
correct this deficiency, a black pen was at-
tached to the log book and was used there-
after by the sampling crews. Laboratory A
occasionally used blue ink instead of the
specified black ink. Although both laboratories
set up their log books with different formats
and headings, the books contained all required
data.
Data Evaluation
Quality Assurance Samples
Three types of QA samples were in-
cluded in each batch of samples submitted to
the analytical laboratory: (1) field duplicates,
(2) preparation duplicates, and (3) natural
audit samples. Data from the QA samples will
be evaluated in the QA report for the analytical
laboratory data. An explanation of each type
of QA sample follows.
One horizon per sampling crew per day
was sampled in duplicate as specified in the
field protocols (Blume et al., 1987). The first
sample of the pair is considered the routine
sample, and the second sample is referred to
as the field duplicate. The field duplicate
underwent the same preparation steps as its
associated routine sample. This allows an
estimate to be made of the combined effects
of sampling error, horizon variability, and
preparation laboratory error.
One sample per batch was chosen by
the laboratory manager to be processed and
then split into two subsamples. On the Form
101 sent to EMSL-LV, one of the pair retained
the routine sample code and the other was
assigned the preparation duplicate designa-
tion. Analytical data from the preparation
duplicates allow the range of physical and
chemical characteristics for splits of the sam-
ple material to be determined. Statistical
analyses of the data will allow an estimate to
be made of the within-laboratory variability due
to subsampling.
Two natural audit pairs supplied by QA
staff were included in each batch that was
sent from a preparation laboratory to an
analytical laboratory, but the samples did not
undergo any processing at the preparation
laboratory. These samples were used to
assess the performance of the analytical
laboratories. Specific information on audit
samples can be found in Bartz et al. (1987).
Method of Estimating Analytical
Precision
Data for the rock fragment and bulk
density determinations were grouped by labor-
atory into data sets. A scatter plot of horizon
standard deviation versus horizon mean was
generated to evaluate the relationship between
precision and concentration. The data for both
parameters displayed a random pattern of
standard deviation, indicating that precision
was independent of concentration. On this
basis, a completely randomized design model
was selected for the statistical estimation of
precision (Steel and Torrie, 1960). The model
that represents data collected at a specific
sampling site can be demonstrated, as
follows:
y,, = u + h, + ey
where: yN = the variable of interest for the jth
observation from the ith horizon
represented
u = the general mean of the
population
13
-------
h, = the effect of the ith horizon on
the variable of interest
By = the random error of analytical
measurement for the jth observa-
tion from the ith horizon
represented
The model was used to perform statis-
tical analyses of (1) rock fragment data from
the field duplicates and (2) bulk density data
from the replicate clods. A root mean square
error statistic was used to estimate the pooled
standard deviation (Sp) across all pedons and
horizons for each laboratory data set, and the
coefficient of variation (CV) was derived by
dividing Sp by the mean of the data set (x)
and multiplying by 100.
Precision Results for Rock Fragment
Determination
Field duplicate data for the rock fragment
determination were analyzed by the completely
randomized design model to provide overall Sp
and CV values for each preparation laboratory.
Summary statistics incorporating these values
are provided in Table 1.
Because of the simplicity of the method
used for determining the percentage of rock
fragments in the bulk soil samples, a greater
amount of the imprecision can be attributed to
spatial horizon variability or sampling bias and
a lesser amount to preparation laboratory
bias. The field sampling imprecision may be
an indication of within-horizon rock fragment
variability or improper field duplicate sampling
technique.
Precision Results for Bulk Density
Determination
Data from the bulk density determination
were analyzed by the completely randomized
design model to provide overall Sp and CV
values for the sets of replicate clods at each
preparation laboratory. In addition, CV values
were generated for two data groups repre-
senting the sets of clods which exhibited a
mean bulk density that was either greater than
or less than the mean bulk density of each
laboratory's clods. Summary statistics for the
Sp and CV values are given in Table 2.
The CV values less than and greater than
the mean were evaluated in order to determine
Table 1. Precision Estimate* for Rock Fragments
Laboratory
Field Duplicates
Number of Pairs
Rock Fragments
Mean (percent)
Sp
CV
A
B
43
61
7.1
13.6
1.371
1.673
19.33
12.32
Table 2. Precision Estimates for Bulk Density
Laboratory
Number of
Horizons
Bulk Density
Mean (g/cm1)
Sp
CV
CV (x)
A
B
262
267
1.28
1.27
0.159
0.118
12.47
9.30
21.09
12.82
7.77
7.18
14
-------
whether or not the statistical relationship of
higher CV values at lower concentrations, in
this case at lower bulk densities, would hold
true for bulk density data. The CV values
presented in Table 2 appear to confirm this
relationship.
The Sp values suggest a consistency of
bulk density values within a given horizon.
Audit reports indicated that the sampling
crews were able to choose representative
clods from each horizon and that the labora-
tories were consistent in their use of measure-
ment techniques. However, the exact percent-
age of error contributed to the Sp values by
horizon variability, sampling bias, or laboratory
imprecision cannot be determined because (1)
inherent spatial variability made it impossible
to sample identical field clods or to provide an
audit sample for measurement of potential
sampling bias, and (2) the preparation labora-
tories were not provided with audit samples
to allow estimation of laboratory bias.
The following types of sampling errors
could contribute to sampling bias for the bulk
density replicates:
• Collection of replicates from transi-
tional zones or adjacent horizons
• Mislabeling of clods
• Inconsistent saran coating procedure
• Variability relating to the coherency of
clods
Based on the field experience of the SCS
sampling crews and the fact that the various
field audit reports did not indicate major devia-
tions from the protocols, sampling bias is not
presumed to have been a significant factor
affecting the Sp values. Some variability in the
use of saran was mentioned in a few audit
reports, although the Sp values would not be
affected if the coating procedure was consis-
tent for all replicates within a horizon.
Bias may be introduced at the prepara-
tion laboratory because of measurement or
method errors, such as the following:
• Transcription errors, such as mis-
recorded weights or sample codes
• Inconsistent saran coating procedures
• Improper clod handling, e.g., compac-
tion
• Incomplete drying
• Loss of material during sieving
• Incorrect numeric calculations
• Faulty weights, e.g., clod tags and
hairnets not subtracted, or balance
not calibrated
Because audit samples were not provi-
ded, interlaboratory bias could not be esti-
mated. Therefore, it is difficult to quantify the
potential effect of preparation laboratory bias
on the Sp and CV values.
Completeness Results
The requested analyses and soil proces-
sing steps were performed on 100 percent of
the bulk samples and clods received by the
preparation laboratories. This satisfied the
maximum theoretical level of completeness.
Preparation duplicates were created for each
batch of samples sent to the analytical labora-
tories for a 100 percent level of completeness.
15
-------
Section 5
Conclusions and Recommendations
The conclusions and recommendations
discussed below have been summarized from
information supplied by preparation laboratory
personnel, QA staff, and other survey partici-
pants. The recommendations are presented
for consideration and possible implementation
in future surveys. Many of the recommenda-
tions are amendments to the existing proto-
cols and are based on the information pro-
vided in Section 3.
General Recommendations
In future surveys, the preparation labora-
tories should be on-line and operational before
field sampling begins. This would allow an
opportunity to assess the overall preparation
function of the laboratories and to clarify the
protocols in advance of soil processing
activities.
The protocols should be rewritten to
include an opening section that addresses
general laboratory procedures. The following
practices will help to improve laboratory safety
and ensure the integrity of the soil samples:
• It should be specified that eating,
drinking, and smoking is prohibited in
the soil drying and processing areas.
• Inexpensive rubber or plastic gloves
should be worn during all bulk sample
processing activities.
• An air-filtering mask or respirator
should be worn while homogenizing
soils in order to prevent excessive
inhalation of airborne soil particles.
• Samples should be crushed, sieved,
and subsampled under an exhaust
hood to limit airborne contamination.
• The saran and acetone should be
mixed only under an operable fume
hood. A respirator also should be
worn.
Sample Receipt
Preparation laboratory personnel should
be available to receive samples from the
sampling crews and to check the sample
labels against the log book entries. If this is
not feasible, the check should be performed
the following day. All discrepancies should be
communicated to the QA manager within three
days. A computer spreadsheet should be kept
current, showing the status of all samples
delivered to the laboratories.
The delivery of practice samples to a
preparation laboratory is not encouraged. If
such samples are delivered, the laboratory
manager should ensure that the practice
samples are separated from the routine sam-
ples. Under no circumstances should practice
samples be logged into the sample receipt log
book.
Equipment Inventory
The preparation laboratories should be
the sole distribution and recovery points for all
sampling equipment provided by EMSL-LV. A
master inventory of equipment should be kept
current and should be inspected by the QA
auditor during the systems audits. The where-
abouts of missing equipment should be deter-
mined as quickly as possible.
16
-------
Sample Drying
Implementation of the following modifica-
tions to the sample drying procedure could
expedite drying considerably:
• Stirring the samples every 24 hours
would encourage rapid drying.
• In addition to the paper placed under-
neath the samples, a single sheet
should be placed loosely over the
samples. The paper beneath the
samples should be replaced daily.
• Small, clip-on, multi-speed electric
fans can be used to facilitate air
movement in the drying area. The
fans should not be allowed to blow
directly on the samples.
Moisture Determination
Many participants stated that a field
moisture determination would be very helpful
for correlating the physical and chemical
status of the soil at the time of sampling.
Three different methods for making this deter-
mination have been identified, as follows:
(1) Field moisture of clods: This method
would utilize the same clods sampled for bulk
density determination. To prevent evaporation,
clods would require greater care in packaging
and refrigeration before being analyzed in the
laboratory. A weight measurement would be
made for each clod upon arrival at the labora-
tory, and would be used to generate percent
field moisture values upon completion of the
bulk density analyses. Sampling crews would
identify saturated soil horizons on the field
data forms.
(2) Field moisture of bulk samples:
Laboratory personnel would select a repre-
sentative 20-gram mixed subsample from each
bulk sample upon its arrival at the laboratory.
The subsample would be placed in a capped
drying container prior to weighing and oven-
drying. Sampling crews would identify satu-
rated soil horizons on the field data forms.
(3) On-site field moisture: Sampling
crews would select replicate moisture samples
from each soil horizon sampled and place each
sample in a capped drying container. The
samples would be weighed and oven-dried at
the preparation laboratory. Saturated horizons
would be identified on the field data forms.
Moisture data for saturated soils would
be tagged automatically in the data base,
irrespective of the method used to determine
field moisture. The gravitational water content
of a saturated sample is expected to vary
widely, depending on the soil textural class,
length of storage, and representativeness of
the bulk sample and subsample taken.
Review of the air-drying procedure has
generated the following recommendation:
• When the laboratory manager believes
that a soil sample is air dry, 10-gram
subsamples should be analyzed for
percent moisture on two successive
days.
Soil Homogenization
Both laboratory managers expressed
displeasure with the amount of time that
elapsed before a list of the soils chosen for
mineralogical study was provided. Every effort
should be made to ensure that this list is
provided to the preparation laboratories before
the analytical samples are split from the bulk
samples. This would eliminate the need for
laboratories to locate and homogenize the
samples twice, thereby saving time and main-
taining sample integrity.
Rock Fragment Determination
Dispensation of the rock fragments
following analysis should be decided before
soil processing begins. This action will avoid
lengthy or costly storage of the fragments at
the preparation laboratories.
Qualitative Test for Inorganic
Carbon
Many participants considered the inor-
ganic carbon test to be inadequate for identify-
ing the soils that should undergo further
analysis for carbonate. It has been suggested
that a pH determination in 0.1M CaCI2 on field-
moist samples would be a better way of
identifying those soils likely to contain inor-
ganic carbon in measurable concentrations.
17
-------
This analysis also would provide valuable data
that cannot be obtained after the samples
have been air-dried. A pH value of 6.0 or
higher is a possible pH at which to target
individual samples for further analysis. If
selected, this procedure should be performed
as soon as the samples arrive at the prepara-
tion laboratory. The procedure outlined in
Cappo et al., 1987 is recommended, except
that a field moist sample would be used.
Bulk Density Determination
To avoid some of the difficulties encoun-
tered with the bulk density determinations, the
following recommendations are made:
• Audit samples and quality control
calibration samples should be used
during clod analysis (see recommen-
dations under QA/QC samples).
• Clods should be air dry before being
analyzed for bulk density. The air-dry
state can be assumed when the clod
weight is constant, using criteria
similar to those applied to the bulk
samples.
• Clod weights should be recorded
immediately before coating with saran
and immediately after the saran
coating has dried in the laboratory.
• A method for measuring the bulk
density of floating clods should be
added to the protocols. Floating
clods should be identified on the bulk
density raw data form and in the bulk
density log book.
• The density of rock fragments within
each clod should not always be
assumed to be 2.65 g/cm3 as sug-
gested in the protocols. In areas with
a variety of lithologies or weathering
characteristics, an appropriate low or
high density value should be used
where necessary.
Sample Shipping
Both laboratory managers stated that a
shipping schedule would have helped to coor-
dinate labor and space requirements for sam-
ple processing. Laboratory personnel often
rushed to get batches shipped because of
insufficient notification, usually one-day notice.
A tentative schedule should be in place before
the initiation of field sampling, and the sched-
ule should be finalized immediately following
the award of analytical laboratory contracts.
All necessary shipping materials, e.g.,
forms and labels, should be delivered to each
preparation laboratory well in advance of the
first shipment of batches to an analytical
laboratory. This will allow an opportunity to
review the forms and to resolve any issues
relating to the shipping procedure or the
sample packaging. Information about the
audit samples should be transmitted by EMSL-
LV at least one week before batch shipment,
which will allow the laboratory manager time
to assemble the batches and prepare Forms
101 and 102. The Form 101 for each batch
should be sent to EMSL-LV on the same day
the batch is shipped.
Quality Assurance and Quality
Control Samples
Guidelines for the packaging of audit
samples should be better defined in the proto-
cols. EMSL- LV should demonstrate the sam-
ple packaging procedure during the pre-sample
audit visit so that laboratory personnel do not
have to repackage audit samples. To ensure
anonymity of the samples at the analytical
laboratories, preparation laboratory personnel
should prepare the routine and duplicate
samples to be indistinguishable from the audit
samples.
Calibration samples should be utilized for
QC checks during the bulk density determina-
tion. It has been suggested that minerals of
known density, e.g., quartz at 2.65 g/cm3,
would be appropriate. The analyses could be
monitored at regular intervals by using mineral
samples that compare favorably in size and
weight to the clods. While this particular
technique has not been tested at EMSL-LV,
calibrated mineral samples of the desired size
and density are readily available from labora-
tory supply outlets. The use of graded sand
samples also has been suggested as a QC
check.
A synthetic audit sample should be
utilized to estimate precision and interlabora-
tory bias during the bulk density determination.
18
-------
A set of plastic eggs that spans a wide range
of displacement capabilities has been sug-
gested as an audit medium, although proto-
types of synthetic eggs have not been devel-
oped or tested.
Record Keeping
Verification of data from the SBRP soil
survey was difficult because the formats of
the sample receipt and raw data log books
varied considerably between Laboratories A
and B. Formats for log books and for record
keeping should be specified in the protocols
and should be reiterated during the pre-sample
audit visit. The assignment of internal labora-
tory numbers to the samples should be
discouraged.
Standardized forms have been developed
to record raw data from the preparation labor-
atories. The sample receipt form (Figure 3)
would include column headings for field sam-
ple code, dates of sample collection and
sample receipt, sample condition, number of
replicate clods collected, and designation of
field duplicates and paired pedons (Coffey et
al., 1987). The bulk density raw data form
(Figure 4) would include computerized column
headings for field and laboratory weights,
water displacement, rock fragments, and
special features of the replicate clods. The
bulk sample raw data form (Figure 5) would
summarize data from the moisture, rock frag-
ment, and inorganic carbon (or pH) determina-
tions. Each bulk sample form would contain
data for one sample. The completed forms
could be sorted alphabetically by state, numer-
ically by pedon number, and could be used as
an index for the Form 101 data.
Laboratory personnel could enter the raw
data into a data base file via a compiled data
base software program, which could be acces-
sed through a personal computer without
actually using the data base software. (Note:
A program similar to this was generated by
ORNL and used by the University of Tennessee
laboratory.) Ideally, each data entry screen
would display the same format as the raw
data forms. The final data could be calculated
and printed on the Form 101 automatically.
Entry or procedural errors could be identified
before a batch is sent to an analytical labora-
tory. Data verification could be completed
shortly after the final batch of soil samples is
analyzed at the preparation laboratory. The
software should be delivered and demon-
strated during the pre-sample audit or the
training workshop.
Design Components
It was suggested that a more extensive
training workshop be conducted. The work-
shop should be held when all laboratory per-
sonnel are able to attend and before soil
sampling has begun. Laboratory personnel
could be instructed on the preparation proce-
dures, record keeping, and packaging samples
for shipment. Also, the QA auditor could
demonstrate the use of the data entry com-
puter program and could help in setting up the
sample drying area.
The documentation and dissemination of
the information discussed during conference
calls would be improved if the following sug-
gestions were implemented:
• The QA laboratory auditor should be
the moderator for all conference calls.
A staff assistant should act in the
place of the QA auditor only when
necessary.
• All participants in each conference call
should be identified.
• Each conference should be tape
recorded in order that the major
points of the discussion can be clar-
ified if necessary.
• All pertinent information should be
documented by the moderator in a log
book set up specifically for this
purpose.
• Log book notes should be compiled
monthly and should be typed and sent
to all participants.
After preparation laboratory operations
are underway, the conference calls may be
reduced to bi-weekly or monthly intervals. The
conference calls should not exclude calls of an
urgent nature, because open communication
and swift resolution of difficulties is in the
best interest of all soil survey participants.
19
-------
u
*"*•"
Crtw
10
Site
ID
D. It
Cellrclid
0|t<
KM
• /
•r
Cor^ltlon
VtifOry (k/D)
Sloid/Unilord IS/U)
l.g SflU (IS)
Undtr VoluM (UI)
Wdl
Si«pt
-------
IN)
All Heights In ±
SMI CODC
0.01 9
«tf
raws
ncio op
nciB wi
Ufl_OP
I>B_HT
UULK ULIIS
CLOO_HaIJ
in IIAU u
TCKf
AIA
CLOO OO
n^pRAa
-
PREP.
CIIRA
LAB.
riOAT
t'ATC
C*g« /
AMAiin
COMUTS IHIlUkS
Figure 4. Bulk density raw data form.
-------
SAMPLE ID:
SITE ID:
SET ID:
BATCH ID:
BULK SAMPLE RAW DATA
DATE SAMPLED:
DATE REC'D:
PROCESS START:
PROCESS COMPLETE:
SOIL TYPE: M / 0
Initials:
INORGANIC CARBON: YES / NO
Initials:
SAMPLE DRYING:
Date Weight Initials
/ . g
/ • g
/ . g
/ . g
/ . g
TOTAL AIR DRY WT:
Date: / /
g
Initials:
ROCK FRAGMENT WT:
Date: / /
2 to 4.75 mm: . g
4.75 to 20 mm: . g
Initials:
ENTERED IN COMPUTER: Date:
Initials:
COMMENTS:
Figure 5. Bulk cample raw data form.
22
-------
Each set of systems audits should be
performed by the same QA auditor, thereby
ensuring uniformity in the evaluation of prepa-
ration laboratory activities. The second round
of audits should be conducted after sample
processing is underway. Audit reports should
be submitted to the QA manager within two
weeks after returning from an audit, and
telephone contact should be made within two
days if serious discrepancies have been
identified.
23
-------
References
Bartz, J. K., S. K. Drous6, K. A. Cappo,
M. L Papp, G. A. Raab, L J. Blume,
M. A. Stapanian, F. C. Garner, and
D. S. Coffey. 1987. Direct/Delayed
Response Project: Quality Assurance Plan
for Soil Sampling, Preparation, and
Analysis. U. S. Environmental Protection
Agency, Las Vegas, Nevada.
Blume, L. J., M. L. Papp, K. A. Cappo,
J. K. Bartz, D. S. Coffey, and K. Thornton.
1987. Soil Sampling Manual for the
Direct/Delayed Response Project Soil
Survey. Appendix A. In Coffey, D.
S., J. J. Lee, J. K. Bartz, R. D. Van
Remortel, M. L Papp, and G. R. Holdren.
1987. Field Operations and Quality
Assurance Report for Soil Sampling and
Preparation in the Southern Blue Ridge
Province of the United States, Volume 1:
Sampling. U. S. Environmental Protection
Agency, Las Vegas, Nevada.
Cappo, K. A., L J. Blume, G. A. Raab,
J. K. Bartz, and J. L. Engels. 1987.
Analytical Methods Manual for the
Direct/Delayed Response Project Soil
Survey. EPA 600/8-87/020. U. S.
Environmental Protection Agency, Las
Vegas, Nevada.
Coffey, D. S., J. J. Lee, J. K. Bartz, R. D. Van
Remortel, M. L Papp, and G. R. Holdren.
1987. Field Operations and Quality
Assurance Report for Soil Sampling and
Preparation in the Southern Blue Ridge
Province of the United States, Volume I:
Sampling. U. S. Environmental Protection
Agency, Las Vegas, Nevada.
Steel, R. G. D., and J. H. Torrie. 1960.
Principles and Procedures of Statistics.
McGraw-Hill Book Company, New York.
481 pp.
USDA-SCS. 1983. National Soils Handbook,
Parts 600-606. U. S. Government Printing
Office, Washington, D.C.
24
-------
Appendix A
Preparation Laboratory Manual for the
Direct/Delayed Response Project Soil Survey
by
J. K. Bartz, D. S. Coffey, and L J. Blume
The following protocols were used by preparation laboratory personnel during the Southern
Blue Ridge Province Soil Survey. This appendix was Part III of the draft "Soil Sampling and
Preparation Laboratory Manual for the Direct/Delayed Response Project Soil Survey. The draft did
not undergo a full external review and was not formally released by EPA. It is presented here
without editorial correction.
The protocols are preceded by a table of contents from the draft manual. Parts I and II of
the manual are presented as Appendix A in Coffey et al. (1987), referenced as follows:
Coffey, D. S., J. J. Lee, J. K. Bartz, R. D. Van Remortel, M. L Papp, and G. R. Holdren. 1987. Field
Operations and Quality Assurance Report for Soil Sampling and Preparation in the
Southern Blue Ridge Province of the United States, Volume I: Sampling. U. S. Environmental
Protection Agency, Las Vegas, Nevada.
25
-------
-------
Section T of C
Revision 4
Date: 5/86
Page 1 of 4
Table of Contents
Section Page Revision
Part I. Overview
1.0 Introduction 1 of 2 4
Part II. Field Operations
2.0 Field Personnel and Equipment 1 of 5 4
2.1 Personnel 1 of 5 4
2.1.1 Field Crews 1 of 5 4
2.1.2 USD A Soil Conservation Service,
Soils Staff 1 of 5 4
2.1.3 Regional Coordinator/Correlator 2 of 5 4
2.1.4 Quality Assurance/Quality Control
Representative 2 of 5 4
2.2 Field Equipment 2 of 5 4
2.2.1 Site Selection Equipment 2 of 5 4
2.2.2 Excavation Equipment 3 of 5 4
2.2.3 Soil Description Equipment 3 of 5 4
2.2.4 Photographic Equipment 4 of 5 4
2.2.5 Clod Sampling Equipment 4 of 5 4
2.2.6 Sampling Equipment 4 of 5 4
2.2.7 Transportation Equipment 5 of 5 4
2.3 Use of Field Equipment 5 of 5 4
3.0 Selection of Pedon to be Sampled 1 of 4 4
3.1 Identifying a Suitable Pedon for Sampling 1 of 4 4
3.2 Procedure for Locating a Suitable Pedon 1 of 4 4
3.3 Locating a Suitable Pedon of a Map Unit
Inclusion 3 of 4 4
3.4 Paired Pedons 4 of 4 4
4.0 Pedon Excavation 1 of 3 4
4.1 Standard Excavation 1 of 3 4
4.1.1 Pit Size 1 of 3 4
4.1.2 Steps in the Pit 2 of 3 4
-------
Section T of C
Revision 4
Date: 5/86
Page 2 of 4
Table of Contents (Continued)
Section Page Revision
4.2 Excavation of Soils with Water Tables 2 of 3 4
4.3 Excavation of Organic Soils 3 of 3 4
4.4 Soils Difficult to Excavate 3 of 3 4
5.0 Site and Profile Description 1 of 3 4
5.1 Profile Preperation 1 of 3 4
5.2 Photographs of Profile and Site 1 of 3 4
5.3 Thick Horizons 2 of 3 4
5.4 Field Descriptions 2 of 3 4
5.5 Documents 3 of 3 4
6.0 Field Sampling Procedures 1 of 6 4
6.1 Sampling the Pedon 1 of 6 4
6.1.1 Field Sampling Protocol 1 of 6 4
6.1.2 Important Points Concerning Soil
Sampling 1 of 6 4
6.2 Sample Size 1 of 6 4
6.3 Sampling Procedure 2 of 6 4
6.3.1 Stratified Horizons 2 of 6 4
6.3.2 Field Duplicates 2 of 6 4
6.4 Sampling Clods for Bulk-Density Determination 2 of 6 4
6.4.1 Procedure 3 of 6 4
6.4.2 Transport of Clods 3 of 6 4
6.5 Filling Sample Bag 3 of 6 4
6.6 NADSS Label A 4 of 6 4
6.7 Delivery 5 of 6 4
Part III. Preparation Laboratory
7.0 Preparation Laboratory Personnel and Equipment 1 of 3 4
7.1 Personnel 1 of 3 4
7.2 Equipment 1 of 3 4
-------
Section T of C
Revision 4
Date: 5/86
Page 3 of 4
Table of Contents (Continued)
Section P^tge Revision
8.0 Receipt and Storage of Samples 1 of 1 4
8.1 Bulk Soil Samples 1 of 1 4
8.2 Clods for Bulk Density 1 of 1 4
9.0 Sample Processing 1 of 6 4
9.1 Air Drying 1 of 6 4
9.1.1 General Considerations 1 of 6 4
9.1.2 Procedure 1 of 6 4
9.2 Crushing and Sieving 2 of 6 4
9.2.1 General Considerations 2 of 6 4
9.2.2 Procedure 3 of 6 4
9.2.3 Calculation of Percent Rock Fragments 4 of 6 4
9.3 Homogenization and Subsampling 4 of 6 4
9.3.1 General Considerations 4 of 6 4
9.3.2 Procedure for Analytical Samples 5 of 6 4
9.3.3 Procedure for Mineralogical Samples 5 of 6 4
9.4 Documentation 6 of 6 4
10.0 Formation and Shipping of Batches 1 of 2 4
10.1 Analytical Samples 1 of 2 4
10.1.1 Procedure 1 of 2 4
10.2 Mineralogical Samples 2 of 2 4
11.0 Analytical Procedures 1 of 5 4
11.1 Rock Fragments 1 of 5 4
11.1.1 Procedure 1 of 5 4
11.1.2 Calculations 1 of 5 4
-------
Section T of C
Revision 4
Date: 5/86
Page 4 of 4
Table of Contents (Continued)
Section Pagg
11.2 Moisture 1 of 5 4
11.2.1 Procedure 1 of 5 4
11.2.2 Calculations 2 of 5 4
11.3 Inorganic Carbon 2 of 5 4
11.3.1 Procedure 2 of 5 4
11.3.2 Internal Quality Control 3 of 5 4
11.4 Bulk Density 3 of 5 4
11.4.1 Procedure 3 of 5 4
11.4.2 Assumptions 4 of 5 4
11.4.3 Calculations 5 of 5 4
12.0 References 1 of •) 4
Appendices
A Strategy of Site Selection and Sampling
Information for the Northeastern United States 1 of 10 4
B Strategy of Site Selection and Sampling
Information for the Southeastern United States 1 of 2 4
C Field Data Form and Legends 1 of 59 4
D Preparation Laboratory Forms 1 of 3 4
E List of Northeast Soils by Sampling Class 1 of 6 4
F List of Southern Blue Ridge Soils by
Sampling Class 1 of 12 4
-------
Acknowledgements
Revision 4
Date: 5/86
Page 1 of 1
Acknowledgments
Contributions provided by the following individuals were greatly appreciated: S. Bodine, D.
Lammers, M. Johnson, J. Lee, B. Jordan, M. Mausbach, R. Nettleton, W. Lynn, F. Kaisacki, B.
Waltman, W. Hanna, B. Rourke, G. Raab, and J. Warner.
The following people were instrumental in the timely completion of this manual: Computer
Sciences Corporation word processing staff at the Environmental Monitoring Systems Laboratory-
Las Vegas, C. Roberts at the Environmental Research Laboratory-Corvallis, J. Engels, M. Faber, and
G. Villa at Lockheed Engineering and Management Services Company, Inc., and Mary Lou Putnam
of Donald Clark Associates.
-------
-------
Section 7.0
Revision 4
Date: 5/86
Page 1 of 3
Part III. Preparation Laboratory
7.0 Preparation Laboratory Personnel and Equipment
7.1 Personnel
Personnel at the preparation laboratory will be responsible for distributing equipment and supplies
provided by EMSL-LV to the field crews for maintaining an adequate inventory of supplies for the
duration of the survey, and for returning all provided equipment and unused supplies to EMSL-LV
at the close of the survey.
The preparation laboratory is also responsible for obtaining the acetone necessary for mixing the
saran:acetone that is used by both the field crews and the preparation laboratory for the collection
of clods and for the analysis of samples for bulk density.
The preparation laboratory is responsible for keeping complete documentation for the tracking of
samples during storage and through the preparation procedures. The laboratory may be required
to provide this documentation to the EPA project officer or designee during the course of the
project. Required documentation must be submitted to the EPA upon conclusion of the survey.
The preparation laboratory is responsible for receipt of bulk soil samples and clods from the field
crews and is also responsible for following the specified protocol for all aspects of storage.
processing, subsampling, batching, and shipping of samples. The laboratory is responsible for
performing the specified analytical procedures, i.e., bulk density, percent air-dry moisture, percent
rock fragments, and the qualitative test for inorganic carbon and for keeping accurate and legible
data books. The data books will be bound, kept in black ink, dated and signed by the analysts,
with all erroneous entries initialed and crossed out so that they remain legible. The laboratory is
responsible for reassignment of field samples to analytical batches and for keeping supporting
documentation which includes the NADSS label A logbook and NADSS forms 101 and 102.
The manager of the preparation laboratory will be a person experienced in processing soil samples
and organized in the tracking of samples through the laboratory.
7.2 Equipment
The preparation laboratory will be equipped to perform the specified protocol. A partial listing of
equipment follows; items marked with an asterisk will be provided by EMSL-LV:
• Logbooks*
• Plastic and cloth sample bags*
• Twist ties for sample bags*
• Brown or white paper
-------
Section 7.0
Revision 4
Date: 5/86
Page 2 of 3
• Drying trays
• Dehumidifier
• Wooden rolling pin
• Sieves, 2-mm mesh and 4.75-mm mesh
• Jones-type, riffle splitter with 1.25-cm openings
• Scale or two-pan balance
• NADSS labels A and B*
• NADSS forms 101 and 102*
• Pens, permanent ink*
• Bottles, 500-ml plastic*
• Shipping boxes*
• Packing material
• Strapping tape*
• Air courier account number*
• Porcelain spot plate
• Squeeze bottle or eyedropper for deionized water
• Microscope or stereoscope (10x or higher)
• 4 N HCI
Quality control detection limit sample, i.e., soil spiked with 1 percent (wt/wt) CaCO, or
• Quality control calibration sample (QCCS) test soil spiked with5% (wt/wt) CaCO3
• QCCS test soil spiked with 5% (wt/wt) CaMgCO3
• Top-loading balance
• Analytical balance
• Drying oven
• Muffle furnace
-------
Section 7.0
Revision 4
Date: 5/86
Page 3 of 3
• Laboratory equipped with house air, fume hoods, safety equipment
-------
-------
Section 8.0
Revision 4
Date: 5/86
Page 1 of 1
8.0 Receipt and Storage of Samples
8.1 Bulk Soil Samples
Samples are received at the preparation laboratory as soon as possible after sampling. Keep a
logbook to record the following information: date received; time received; who delivered samples;
who received samples; condition of samples, noting only problems specifically by sample code, e.g.,
if the samples were not held at 4°C prior to delivery or if a sample container broke and
contamination is possible; set ID numbers; and total number of samples. The field crew is
responsible for registering samples upon delivery unless laboratory personnel decide to accept this
responsibility.
Unless the samples are taken immediately to the processing area and are spread out to air-dry,
place the samples in cold storage upon receipt. Return samples to cold storage after drying and
whenever they are not undergoing processing as described in Section 9.0. Maintain the cold
storage locker at 4*C and monitor the temperature on a daily basis. Record any deviations in the
temperature in a logbook and identify those samples affected by the change in temperature by set
ID and, if necessary for clarity, by sample code.
Keep samples organized so that a particular sample may be located easily. Shelving of some kind
is necessary to maximize the use of available space and to facilitate organization of the samples.
8.2 Clods for Bulk Density
Clods are received at the preparation laboratory along with the bulk samples. A separate logbook
is kept to record the receipt of clods.
Remove the label from each clod and staple it in the logbook. Immediately relabel the clod with
the sample code and assign a replicate number to provide a unique identity for each clod. Also,
record the date of receipt, the condition of each clod, the number of saran coatings, and the
assigned replicate number.
-------
-------
Section 9.0
Revision 4
Date: 5/86
Page 1 of 6
9.0 Sample Processing
Specific areas will be designated for sample processing. During all steps of processing, sample
integrity is of the greatest importance. This is protected by maintaining the unique identity of each
sample by labeling with the sample code, by documenting each sample through all processing, and
by avoiding physical and chemical contamination during each processing step.
9.1 Air Drying
9.1.1 General Considerations
The area designated for air-drying samples must have adequate bench space to allow several sets
of samples, i.e., approximately 60 samples, to be drying simultaneously. Also, it is important that
there be a method to reduce relative humidity in the room, e.g., by removing moist air via a fume
hood or by reducing humidity with a dehumidifier. If available, a temperature- and humidity-
controlled greenhouse environment is ideal for drying samples. The area or room must be secure
and have limited access.
In evaluating an area for air-drying samples, sources of potential contamination, including growing
plants, other sample types, dust, and chemicals such as fertilizers or pesticides, will be identified
and eliminated. Handling samples without wearing gloves is a source of contamination as is the
talc which is used on some plastic or rubber gloves.
Soils high in clay may harden irreversibly if allowed to dry completely prior to crushing; therefore,
crush such soils before they reach a constant moisture content. Then continue air-drying after
the crushing procedure (described in Section 9.2) and determine the moisture content after the
air-drying is completed. It may be desirable to hold partially dry soils that are high in clay in cold
storage until crushing may be scheduled.
9.1.2 Procedure
9.1.2.1 Spread the sample on a new sheet of white or brown paper large enough to allow paper
to be exposed on all sides of the sample. The paper may be placed on a tray or directly
upon the bench top.
9.1.2.2 Occasionally stir the soil with a clean stainless-steel spoon to facilitate drying. Water may
condense on the tray or bench top, creating an undesirable situation that would maximize
conditions for microbial growth. An arrangement allowing for air circulation below the
sample is preferred. For example, frames with screen bottoms may be used in place of
trays, or the bench top may be made of steel netting. As an alternative, place several
layers of paper below the paper containing the sample and replace those layers when they
become damp. Any observations of fungal or algal growth should be noted by sample code
in the sample tracking log book.
9.1.2.3 Allow the sample to air-dry until it achieves a constant moisture content, i.e., until it is in
equilibrium with the relative humidity. This step may require a period of two days or more
than three weeks, depending upon the character of sample.
-------
Section 9.0
Revision 4
Date: 5/86
Page 2 of 6
Test each sample for moisture content as described in Section 11.2. Constant air-dry
moisture content is achieved when the moisture content is not reduced by more than an
absolute 2.5 percent in two consecutive days. After the sample passes this test, it may
undergo further processing immediately or be placed in a new plastic bag, sealed, and
placed in cold storage until further processing is possible.
9.1.2.4 On rainy days or during periods of high relative humidity, a sample may adsorb moisture;
therefore, determinations of moisture may show an increase in moisture content. Do not
rebag a sample that has increased in moisture content because of such weather
conditions. Allow additional, less humid days for it to dry. If necessary, replace the damp
paper below the sample.
9.2 Crushing and Sieving
9.2.1 General Considerations
A fume hood with inspected and properly functioning fans, vents, and filters, is the preferred
location for crushing and sieving samples. It is a relatively small area that can be covered with
protective layers of paper and that can be cleaned easily. Also, the exhaust removes small soil
particles that would otherwise contaminate the work area and that would be inhaled by the
technician.
An area near a source of compressed air is desirable because the air can be used to clean the
surfaces of the equipment, i.e., wooden rolling pin, 4.75-mm mesh sieve, and 2-mm mesh sieve,
after the processing of each sample. However, water and oil tend to accumulate in the conduit and
to spray out occasionally from in-house, compressed-air lines. This would represent a source of
contamination. Therefore, the air must be passed through a trap to collect the offending water and
oil
Such a trap can be assembled by using an Erlenmeyer flask, a small piece of cotton cloth or
several Kimwipes, a one-holed rubber stopper sized to fit the flask, a piece of glass or plastic
tubing long enough to extend through the rubber stopper and well past the side arm of the flask,
and two lengths of vinyl tubing. Place the cloth or Kimwipes into the flask. Insert the glass or
plastic tubing through the rubber stopper and stopper the flask. Connect the side arm of the flask
to the air nozzle with a short length of vinyl tubing. The second length of vinyl tubing should be
long enough to reach from the position of the flask to the work area. Attach it to the exposed end
of the glass or plastic tubing which is in the rubber stopper. If water or oil is observed in this
length of vinyl tubing when using air to clean the equipment, the cloth or Kimwipes should be
replaced, and the tubing should be cleaned out before continuing. Occasionally, the air pressure
may be too high for this apparatus, and this will cause the rubber stopper to blow out. This
problem can be corrected by adjusting the delivery of air from the nozzle.
In addition to cleaning the equipment with compressed air, the surfaces may be wiped with
Kimwipes if soil is adhering to the wooden rolling pin or to the frame of the sieve. Used Kimwipes
should be discarded after cleanup for each sample processed to avoid contamination. Sieves which
have been cleaned with compressed air will be rinsed with deionized water and will be thoroughly
air-dried or oven-dried at the end of each day. Brass sieves should be removed from the drying
oven as soon as they are dry because brass will oxidize at temperatures as low as 60"C.
-------
Section 9.0
Revision 4
Date: 5/86
Page 3 of 6
Large pieces of stems and roots that are obviously not part of the soil should be removed when
encountered in the processing described in Section 9.2.2.
Caution: Rock fragments in soils of the Southeast are likely to be relatively soft either because
they are of sedimentary origin or because they have been subjected to intensive
weathering. Using a wooden rolling pin as specified in the procedure below will crush
these soft rock fragments. This will result in an underestimation of percentage rock
fragments in the laboratory as compared to the field estimates and also will result in
a sample of questionable integrity because it contains crushed rock material. When
rock fragements are soft and easily crushed, the procedure may be modified according
to the alternatives specified in the appropriate subsections below.
9.2.2 Procedure
9.2.2.1 Record the weight of the whole sample prior to crushing and sieving. Any sample lost
during the processing is most likely to be from the less than 2-mm fraction.
9.2.2.2 Spread a small portion of sample on an approximately 60-cm by 60-cm piece of brown or
white paper. Another layer of paper may be placed over the sample, if desired. Crush the
sample with the rolling pin; apply only enough force to disaggregate the clods or soil
structure, but not so much force that weathered rocks are crushed. Go to Section 9.2.2.3.
Alternative: Place a portion of sample in the 2-mm sieve and gently push the soil through
the sieve with a rubber stopper. Attempt to include soil adhering to the rock
fragments. Continue until the entire sample is processed. Go to Section
9.2.2.5.
9.2.2.3 After each portion is crushed, set the portion aside and repeat with another aliquot until
the entire sample is prepared. Go to Section 9.2.2.4. Alternatively, each portion may be
sieved as described in Section 9.2.2.4 before continuing with another portion of sample.
9.2.2.4 Either nest the two sieves with the 2-mm mesh sieve on the bottom or use only the 2-mm
mesh sieve for this step. Place the sieve(s) onto an approximately 45-cm by 45-cm piece
of brown or white paper. Pour aliquots of soil onto the sieve (s). Move the sieve (s) from
side to side, gently tapping the side of the sieve(s) to facilitate the passage of sample
through the mesh. Do not press sample through the sieve (s). If peds are retained by the
sieve(s), remove that material from the sieve(s) and crush as described in Section 9.2.2.3.
Continue sieving until all material has been processed. Save the less than 2-mm material
in plastic sample bags and return to cold storage or immediately continue with processing
as described in Section 9.3.
9.2.2.5 If both the 4.75-mm and 2-mm sieves were used in Section 9.2.2.4, continue at Section
9.2.2.6.
If only the 2-mm mesh sieve was used in Section 9.2.2.4, sieve all rock material which was
retained by the 2-mm mesh sieve through the 4.75-mm mesh sieve. Catch the material
which passes the sieve on a piece of paper.
-------
Section 9.0
Revision 4
Date: 5/86
Page 4 of 6
Alternative: A significant amount of soil may adhere to the soft rock fragments. Wash
the fragments with deionized water while they are on the sieve. Then
continue with Section 9.2.2.6.
9.2.2.6 Weigh the material retained by the 4.75-mm sieve. Record the weight as that of the 20-
to 4.75-mm rock fragments.
9.2.2.7 Weigh the material retained by the 2-mm sieve. Record the weight as that of the 4.75- to
2-mm rock fragments.
9.2.2.8 Combine the rock fragments in a plastic sample bag and label with the appropriate sample
code.
If the qualitative test for inorganic carbon as described in Section 11.3 is positive, crush the
rock fragments to pass a 2-mm mesh sieve. Use a Jones-type riffle splitter first to
homogenize the sample and then to obtain a 100-g subsample. Crush the entire subsample
to pass an 80-mesh sieve. Place it in a plastic sample bag labeled with the appropriate
NADSS label. If crushing of rock fragments is done prior to sample batching, the 100-g
subsample is labelled initially with NADSS Label A which is later replaced with NADSS Label
B. If sample batching is done prior to the determination of inorganic carbon, the 100-g
subsample is labeled with NADSS Label B. Place the plastic bag in a cloth sample bag.
Keep this subsample with the appropriate 1-kg analytical subsample for shipment to the
analytical laboratory as described in Section 10.0.
Save all rock fragment samples until instructed to discard them or to ship them to another
location. Cold storage is not required for rock fragments.
9.2.3 Calculation of Percent Rock Fragments
The percentage of rock fragments in each size faction, i.e., 20- to 4.75-mm and 4.75- to 2-mm, is
calculated as follows:
([weight of size fraction] \
100
[total weight] j
9.3 Homogenization and Subsampling
9.3.1 General Considerations
A fume hood is the preferred location for homogenization and subsampling the less than 2-mm
fraction. It should be prepared as described in Section 9.2.1.
A Jones-type riffle splitter with 1.25-cm openings is used for both homogenization and subsampling.
The riffle splitter will be cleaned between each sample with compressed air as described in Section
9.2.1. The surfaces may be cleaned with a brush, if necessary.
It will be cleaned with deionized water and allowed to thoroughly air-dry after each day of use.
Prior to use, the riffle spitter should be inspected to verify that it is dry.
-------
Section 9.0
Revision 4
Date: 5/86
Page 5 of 6
A scale or two-pan balance must be available to weigh the 1-kg analytical sample or the 500-g
mineralogical sample.
Because (1) only a few horizons will be chosen for mineralogical analyses from all routine samples
collected and (2) the identification of those horizons is made after all samples have been collected,
samples for mineralogical analyses will be split at some time after those for analytical analyses
are split.
9.3.2 Procedure for Analytical Samples
9.3.2.1 Position the receiving pans on each side of the riffle splitter. Pour the less than 2-mm
material evenly across the baffles of the riffle splitter. Repeat six times in succession for
the material in each receiving pan to ensure that the sample is thoroughly homogenized.
9.3.2.2 Transfer the entire sample to one receiving pan or pour it onto a new square of brown or
white paper,
Place the receiving pans on each side of the riffle splitter. Pour the sample evenly across
the baffles. Place the material from one receiving pan into a plastic sample bag. With the
material in the other receiving pan, repeat the procedure as required until a subsample of
approximately 1 kg is obtained.
It is important that the subsample be obtained entirely by splitting with the riffle splitter.
An alternate method of using the riffle splitter to obtain the 1-kg subsample follows: After
an easily handled amount of sample is split out, a subsample of 1 kg may be obtained by
splitting the material and allowing the material to remain in both pans. Then the material
from one pan is passed through the splitter again, so that one pan now contains 3/4 of
the material and the other pan contains 1/4 of the material. Place the smaller amount of
material in the sample bag and continue to split in this manner until the 1-kg sample Is
obtained. If only a small amount of material is needed to attain the 1-kg subsample,
successive splits of the material may be made so that the pan containing the lesser
amount of material contains 1/8, 1/16, 1/32, and so on, and the pan containing the greater
portion of material contains 7/8, 15/16, 31/32, and so on.
9.3.2.3 Place the 1-kg subsample in a plastic sample bag and secure with a twist tie. Label with
NADSS Label A or, if an analytical batch is being assembled as described in Section 10.0,
NADSS Label B may be placed on the 1-kg subsample. The sample may be double-bagged
in plastic for security. Place the plastic bag in a cloth bag and return the subsample to
cold storage as soon as possible.
9.3.2.4 Label remaining sample material with NADSS Label A Place the plastic bag in a cloth bag
and return the sample to cold storage for archiving. Sets will be packed according to
sample code for permanent archiving.
9.3.3 Procedure for Mineralogical Samples
9.3.3.1 Rehomogenize the less than 2-mm material as described in Section 9.3.2.1.
9.3.3.2 Split out a 500-g subsample as described in Section 9.3.2.2.
-------
Section 9.0
Revision 4
Date: 5/86
Page 6 of 6
9.3.3.3 Place the 500-g subsample in a 500-mL air-tight plastic bottle. Label the bottle with the
sample code and return the sample to cold storage until it is shipped.
9.4 Documentation
In addition to data books used for entry of raw data for determination of percent moisture and
percent rock fragments, records will be kept to document the date(s) of sample processing. Also
the weight of the less than 2-mm material obtained after sieving, the percent rock fragments in the
20- to 4.75-mm fraction, and the percent rock fragments in the 4.75- to 2-mm fraction will be
entered into the NADSS Label A logbook described in Section 10 11 4
-------
Section 10.0
Revision 4
Date: 5/86
Page 1 of 2
10.0 Formation and Shipping of Batches
10.1 Analytical Samples
A batch of samples for chemical and physical analysis by a contractor laboratory consists of a
maximum of 39 routine samples and field duplicates. Depending upon the number of samples in
each set, i.e., a group of samples taken by one field crew in 1 day, a batch may include up to six
sets In addition, one sample per batch is chosen as the preparation duplicate and is split into two
subsamples according to the procedure in Section 9.4. Two audit samples supplied by the quality
assurance staff will be specified for each batch. Therefore, one batch including routine samples
field duplicates, a preparation duplicate, and two audit samples will contain a maximum of 42
samples.
10.1.1 Procedure
10111 After deciding which sets will be included in the analytical batch, randomly choose one
sample to be the preparation duplicate. The horizon-type chosen as the preparation
duplicate should vary from batch to batch so that all horizon-types are chosen at least
once Rehomogenize the archived sample material for the chosen sample by seven
successive passes through a Jones-type riffle splitter. Then split out a subsample as
described in Section 9.4.
10 11 2 Randomize the samples in the batch. Assign sample numbers from 1 through 42, as
needed. Record the site ID, sample code, and set ID on NADSS Form 101.
The sample code for a preparation duplicate will begin with three alpha characters, PLD,
and end with the same last 10 digits as the routine sample from which it was split. For
example, the preparation duplicate split from routine sample R11NH01300405 would be
designated as PLDNH01300405.
The sample code for an audit sample would begin with the first three characters of the
horizon designation and would be followed by two zeros, the three-digit audit number
specified by the quality assurance staff, a hyphen, and four zeros. For example, the audit
sample prepared from soil sampled from an argillic, maximum B horizon would begin with
the three characters, B+2. If the horizon designation includes less than three characters,
the spaces are filled in with zeros.
Set NADSS Form 101 aside until the remaining data for weight percent rock fragments,
weight percent air-dry moisture, soil type, presence of inorganic carbon, and bulk density
may be entered from the appropriate data books. For sample tracking purposes, it is
important that this form be sent as soon as possible. After it is completed, Form 101 will
be submitted to the following address:
Lockheed Engineering & Sciences Company
1050 East Flamingo, Suite 200
Las Vegas, Nevada 89119
-------
Section 10.0
Revision 4
Date: 5/86
Page 2 of 2
10.1.1.3 Fill out NADSS Label B for each sample. This label includes only the batch number and
the sample number assigned above.
10.1.1.4 Remove NADSS Label A from each sample and replace it with NADSS Label B. Staple
NADSS Label A in a logbook and initial the label so that the writing overlaps both the label
and the page. Indicate in the logbook the newly assigned batch and sample numbers.
1°'1'1'5 fJI!,2tN^StF°T I?2' Remove the white (ori9inal> c°Py and send jt via overnight
courier to the Sample Management Office at the following address:
Sample Management Office (Viar)
300 North Lee Street
Alexandria, Virginia 22314
The remaining copies are left intact and are included inside the packed box(es).
10.1.1.6 Each sample should now be labeled with only the newly assigned batch ID and sample
number. Place each plastic bag of sample in a cloth bag. Secure the cloth bag so that
any sample that might leak from the plastic bag will be contained by the cloth baa Pack
the samples securely, by using additional packing materials as needed.
10.1.1.7 Store the packed boxes under refrigeration at 4 eC until they are to be shipped.
10.1.1.8 Send the box(es) containing the designated analytical batch to the contractor laboratory
specified by the quality assurance staff. An overnight air courier service will be specified
for all shipments, and an account number will be supplied for billing shipments to a third
party.
10.2 Mineralogical Samples
nfJJStr for .minerflo9'cal analysis will be combined into batches of 20 routine samples, 3
£252! • duphcate!' a"d ? audit samP|es at EMSL-LV. The subsamples will be shipped via
fo^SlT ST-10 tnef°lowin9 address where they will be combined into batches and
forwarded to the designated mineralogical laboratory:
Lockheed Engineering & Sciences Company
1050 East Flamingo Road, Suite 200
Las Vegas, Nevada 89119
-------
Section 11.0
Revision 4
Date: 5/86
Page 1 of 5
11.0 Analytical Procedures
11.1 Rock Fragments
The term "rock fragments" is defined in Soil Taxonomy as "particles 2 mm in diameter or larger
and includes all sizes that have horizontal dimensions less than the size of a pedon." Only the 20-
to 4 75-mm fraction and the 4.75- to 2-mm fraction are determined in the preparation laboratory.
The percentage of rock fragments is determined on the basis of the weight of sample brought into
the laboratory for processing.
//. /. / Procedure
11111 Record the weight of the whole sample prior to crushing and sieving. Any sample lost
during the processing described in Section 9.2 is most likely to be from the less than 2-
mm fraction.
11.1.1.2 Weigh the material retained by the 4.75-mm sieve. Record the weight as that of the 20-
to 4.75-mm rock fragments.
11.1.1.3 Weigh the material retained by the 2-mm sieve. Record the weight as that of the 4.75- to
2-mm rock fragments.
11.1.2 Calculations
The percentage of rock fragments in each size faction, i.e., 20- to 4.75-mm and 4.75- to 2-mm, is
calculated as follows:
([weight of size fraction] \
100
[total weight] J
11.2 Moisture
Moisture content is determined on a 15- to 20-g grab sample taken after the air-dry sample
described in Section 9.1.2.2 has been thoroughly mixed. Two convection-type drying ovens are
preferred for this procedure if both mineral and organic soils are tested because mineral soils are
dried at ± 5 *C and organic soils are dried at 60 *C ± 5 *C. Each oven must equilibrate at the
appropriate temperature for 24 hours prior to use. If only one drying oven is available, allow 24
hours for the temperature to equilibrate after an adjustment is made from one temperature to the
other. The range of the thermometers used to measure oven temperature will be 0 to 200 C.
11.2.1 Procedure
11.2.1.1 Thoroughly mix the air-dried soil. Transfer a 15- to 20-g sample to an aluminum weighing
dish which is not to exceed 2 g in weight. Handle the weighing dish with forceps or finger
cots.
-------
Section 11.0
Revision 4
Date: 5/86
Page 2 of 5
11.2.1.2 Record the initial weight of sample and weighing dish to the nearest 0.01 g. Because the
aluminum weighing dishes are manufactured to be a nearly-consistent weight, the average
weight of 10 aluminum weighing dishes may be used as the tare weight.
11.2.1.3 Dry the sample overnight or for 16 hours in a drying oven which is equilibrated at the
appropriate temperature.
11.2.1.4 Remove the sample from the oven and allow it to cool for at least 30 minutes in a
desiccator. Then record the oven-dried weight of the sample and weighing dish.
11.2.1.5 On the following day, transfer a second 15- to 20-g sample from the central portion of the
air-dried material to an aluminum weighing dish. Repeat steps 11.2.1.2 through 11.2.1.4.
11.2.1.6 If the calculated moisture contents differ by more than an absolute 2.5 percent, allow the
sample to continue air-drying and retest it at a later time. However, if the calculated
moisture contents do not differ by more than 2.5 percent in two consecutive days, the air-
dry sample may undergo processing as described in Section 9.0.
11.2.2 Calculations
I [air-dried weight - oven-dried weightn
Percent moisture = 100
\ [oven-dried weight] /
11.3 Inorganic Carbon
The less than 2-mm fraction is tested qualitatively for the presence of carbonate minerals, i.e.,
inorganic carbon. If carbonate minerals are present in the soil at levels of approximately 1 percent
or higher, effervescence is observed after the addition of 4 N HCI.
When a sample is found to contain inorganic carbon by using this test, both the less than 2-mm
material and the 20- to 2-mm rock fragments are submitted to the analytical laboratory for analysis
Refer to Section 9.2.2.8. '
11.3.1 Procedure
11.3.1.1 Place 1 g of less than 2-mm soil material in a well of a porcelain spot plate. Thoroughly
moisten the soil with a few drops of deionized water; stir with a clean glass rod to remove
entrapped air.
11.3.1.2 Add three drops of 4 N HCI and immediately observe the treated sample under a light
microscope or stereoscope. Be careful not to get acid on lens or lens sealant. If this
occurs, wash off acid and clean and dry lenses. Record presence or absence of
effervescence for each sample according to sample code.
11.3.1.3 Repeat the treatment and the observation with a second 1-g sample.
-------
Section 11.0
Revision 4
Date: 5/86
Page 3 of 5
11.3.2 Internal Quality Control
Both a quality control (QC) detection limit sample and a quality control calibration sample (QCCS)
are required for this procedure. The purpose of the QC samples is to ensure that the technician
is able to distinguish effervescence.
The QC detection limit sample is prepared by spiking an aliquot of non-carbonate soil with 1 percent
(wt/wt) CaCO3 or CaMgfCO^. The QC detection limit sample is analyzed prior to sample analysis
and after each 10 or fewer samples.
The QCCS is prepared by spiking an aliquot of non-carbonate soil with 5 percent (wt/wt) CaCO3
or CaMg(CC>3)2. The QCCS is analyzed prior to sample analysis and after each twenty or fewer
samples.
11.4 Bulk Density
Density is defined as mass per unit volume expressed in units of g/cm3. The bulk density of a soil
is defined as the mass of dry soil per unit volume including the pore space. The bulk volume is
determined prior to drying the soil to constant weight, at 105 *C for mineral soils and at 60 C for
organic soils.
Bulk density of mineral soils generally ranges between 1.0 and 2.0 g/cm3. Soils that are porous
will have low bulk density values, and soils that are compacted will have high bulk density values.
With increasing organic matter content, soils generally exhibit a decrease in bulk density because
(1) organic matter is less dense than mineral particles of the same size and (2) organic matter
promotes granular soil structure with a resulting increase in porosity.
11.4.1 Procedure
11.4.1.1 Weigh the clod as received from the field crew. Record the initial weight as m,. Clods
weighing more than 100 g are weighed to the nearest 0.1 g; clods weighing less than 100
g are weighed to the nearest 0.01 g.
11.4.1.2 Briefly dip the clod into a 1:7 (wt/wt) saran-acetone mixture. Then suspend the clod from
a line and allow the clod to dry.
11.4.1.3 Reweigh the clod and record this weight as m2.
11.4.1.4 Repeat steps 11.4.1.2 and 11.4.1.3 as necessary until a coating of saran that is impervious
to water is obtained. Record each additional weight as m3, m4 m,.
11.4.1.5 Degas approximately 800 mL of deionized water in a 1-L beaker by boiling until no rising
gas bubbles are observed. Cover the beaker with a watch glass and allow the water to
cool to room temperature.
11.4.1.6 Place the beaker of degassed water on a balance and record the tare weight. Also, record
the temperature of the water so that the density of the water may be obtained from Table
11.1.
-------
Section 11.0
Revision 4
Date: 5/86
Page 4 of 5
Table 11.1. Density* of Water
4C
0
10
20
30
40
50
60
70
80
90
0.9999
0.9997
0.9982
0.9957
0.9922
0.9881
0.9832
0.9778
0.9718
0.9653
0.9999
0.9996
0.9980
0.9954
0.9919
0.9876
0.9827
0.9772
0.9712
0.9647
1.0000
0.9995
0.9978
0.9951
0.9915
0.9872
0.9822
0.9767
0.9706
0.9640
1.0000
0.9994
0.9976
0.9947
0.9911
0.9867
0.9817
0.9761
0.9699
0.9633
1.0000
0.9993
0.9973
0.9944
0.9907
0.9862
0.9811
0.9755
0.9693
0.9626
1.0000
0.9991
0.9971
0.9941
0.9902
0.9857
0.9806
0.9749
0.9686
0.9619
1.0000
0.9990
0.9968
0.9937
0.9898
0.9852
0.9800
0.9743
0.9580
0.9612
0.9999
0.9988
0.9965
0.9934
0.9894
0.9848
0.9795
0.9737
0.9673
0.9605
0.9999
0.9986
0.9963
0.9930
0.9890
0.9842
0.9789
0.9731
0.9667
0.9598
0.9999
0.9984
0.9960
0.9926
0.9885
0.9838
0.9784
0.9724
0.9660
0.9591
Also the specific gravity or unit weight of water in grams per milliliter.
11.4.1.7 Suspend the clod over the beaker. Lower the clod gently into the water until it is totally
submerged. Record the weight displayed on the balance as weight of water supporting
clod.
11.4.1.8 Suspend the clod in a drying oven at the appropriate temperature for 48 hours: 105 *C for
mineral soils and 60 *C for organic soils. Moisture from within the clod will diffuse through
the saran coating.
11.4.1.9 Remove the clod from the oven and allow it to cool in a desiccator. Weigh the clod to
obtain the oven-dried weight.
11.4.1.10 For mineral soils, place the clod in a container that will withstand 400 *C. Place the clod
and container in a muffle furnace that is equilibrated at 400 *C and allow the saran to
vaporize from the clod over the next 2 hours. The muffle furnace must be placed in a
fume hood: saran is a carcinogen.
Remove the container and contents from the muffle furnace and allow to cool in a
desiccator. Weigh only the contents of the container. Record the weight of the soil
particles.
11.4.1.11 Pass the sample through a 2-mm mesh sieve. Weigh both the rock fragments and the
less than 2-mm fine earth fraction. Record the weights as mrt and m(f, respectively.
11.4.2 Assumptions
The following assumptions are made in this procedure:
• The weight of each field-applied saran coating is equivalent to each coating applied
in the laboratory.
• Saran has not infiltrated the clod.
• The specific gravity of air-dried saran is 1.30 g/cm3.
-------
Section 11.0
Revision 4
Date: 5/86
Page 5 of 5
• Saran loses 15 percent of its weight upon oven-drying for 48 hours.
• The density of the rock fragments is 2.65 g/cm3.
11.4.3 Calculations
11.4.3.1 Calculate the weight of the saran coatings before oven-drying;
/[number of coats (final clod wt - initial clod wt)]
Weight of saran - — . _ ~~
\ [(number of coats in lab) - 1]
11.4.3.2 Calculate the weight of the saran coating after oven-drying:
OD Weight of saran = (0.85) (Weight of saran)
11.4.3.3 Calculate the volume of the saran coatings:
\
Wt of saran
Saran vol
1.30
11.4.3.4 Calculate the volume of water displaced by the clod:
Wt of water
Vol of water =
(density of water)
11.4.3.5 Calculate the volume of rock fragments in the clod:
'wtof RF
Vol of RF =
2.65
11.4.3.6 Calculate the volume of the less than 2-mm fraction and pore space:
Vol of fines = [Vol of water - (Vol of RF + Vol of saran)] and pores
11.4.3.7 Finally, calculate the field moist bulk density:
[OD clod wt - (wt of RF + OD saran wt)]
Bulk Density =
Vol of fines and pores
-------
-------
Section 12.0
Revision 4
Date: 5/86
Page 1 of 1
12.0 References
1. USDA/SCS. 1983. National Soils Handbook. Part 600-606. U.S. Government Printing
Office, Washington D.C.
2. USDA/SCS. 1984. SCS National Soil Survey Manual. U.S. Government Printing Office,
Washington D.C.
3 Mausbach, M., R. Yeck, D. Nettleton, and W. Lynn. 1983. Principles and Procedures for
Using Soil Survey Laboratory Data. National Soil Survey Laboratory. Lincoln, Nebraska.
4. USDA/SCS. 1981. National Handbook of Plant Names. U.S. Government Printing Office,
Washington, D.C.
5 USDA/SCS 1984b. Soil Survey Laboratory Methods and Procedures for Collecting Soil
Samples. Soil Survey Investigations Report No. 1. U.S. Government Printing Office,
Washington D.C.
------- |