Bulk Density Determination by Automated Three ... - NextEngine

Bulk Density Determination by Automated Three-Dimensional Laser Scanning Ann M. Rossi* Soil and Water Sciences Program Dep. of Environmental Sciences Univ. of California Riverside, CA 92521-0424

Daniel R. Hirmas Dep. of Geography Univ. of Kansas Lawrence, KS 66045-7613

Robert C. Graham Paul D. Sternberg Soil and Water Sciences Program Dep. of Environmental Sciences Univ. of California Riverside, CA 92521-0424

Soil bulk density is a ratio of the mass of solids to bulk soil volume. Values typically range from 1.2 to 1.6 g cm−3 depending on the texture, structure, degree of compaction, and shrink–swell characteristics (Hillel, 1998; Soil Survey Division Staff, 1993; Mason et al., 1957). Bulk density measurements are widely used in mass–volume conversions and calculating porosity and void ratios (Blake and Hartge, 1986). Bulk density values are also used as a measure of soil quality, indicating the ease of root penetration, water movement, and soil strength (Grossman and Reinsch, 2002). One commonly used procedure for determining bulk density is the clod method. This method is especially useful for soils containSoil Sci. Soc. Am. J. 72:1591-1593 doi:10.2136/sssaj2008.0072N Received 29 Feb. 2008. *Corresponding author ([email protected]). © Soil Science Society of America 677 S. Segoe Rd. Madison WI 53711 USA All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Permission for printing and for reprinting the material contained herein has been obtained by the publisher. SSSAJ: Volume 72: Number 6 • November–December 2008

MATERIALS AND METHODS Intact soil clods were collected from four sites representing a range of soil textures and structures. Sample locations and descriptions are given in Table 1. Hard, nonporous, granitic rock fragments from the Bishop Creek moraines in California were used as controls. Triplicate samples from each site were analyzed with both the three-dimensional scanning method (nondestructive) and the paraffin-coated clod method (destructive). Volumes and bulk density measurements from these two methods were analyzed in an analysis of variance using R 2.6.0 (R Development Core Team, 2007).

PEDOLOGY NOTE

This study examined the application of a relatively new automated three-dimensional scanning technology to bulk density determination of intact soil clods and rock fragments. The method uses an inexpensive commercially available three-dimensional desktop scanner. Measurements obtained by the scanning method were compared with those of the paraffin-coated clod method determined on the same clods. Results showed excellent agreement between volume and bulk density measurements obtained by the two methods across a wide range of textural classes. Use of the technology has several important advantages over the traditional clod method. The nondestructive nature of the scanning method makes it possible to use the same intact ped or clod for other purposes, such as thin sections for micromorphological analysis. In addition, high-resolution digital imaging opens up possibilities for new physical measurements of soil morphological features.

ing rock fragments (Cunningham and Matelski, 1968; Hirmas and Furquim, 2006) because extracting cores from these soils is problematic. Intact soil clods are coated with an impervious or semipermeable substance, such as liquid paraffin, saran, rubber, wax, or oil, and the clod volume is measured by water displacement (Blake and Hartge, 1986; Grossman and Reinsch, 2002; Soil Survey Staff, 1996). Preparing paraffin-coated clods can be difficult and labor intensive (Van Remortel and Shields, 1993). The process of removing the coating is difficult and destroys the clast, making any further analyses of the clod impossible. Sander and Gerke (2007) used a three-dimensional optical scanner to determine volume in measuring soil shrinkage characteristics. This technology can also be used to measure the bulk density of soils and geologic materials. The objective of this study was to evaluate the application of automated three-dimensional laser scanning technology to the determination of the bulk density of soil clods and rock fragments. A nondestructive method would have an advantage over the traditional clod method by allowing various further analyses on the same soil clod.

Three-Dimensional Laser Scanning Method Soil clods were oven dried at 105°C overnight and scanned using a commercially available desktop three-dimensional scanner (Fig. 1) (NextEngine Desktop 3D Scanner Model 2020i, NextEngine, Inc., Santa Monica, CA). This instrument is relatively inexpensive, with a cost roughly equivalent to a desktop computer and printer setup. A computer connected to the scanner is used to operate the scanner and store collected data. A total of 10 scans at 0.13-mm resolution was taken of each clod as follows. Clods were placed on the sample holder and scanned five times. Each scan was taken at 72° intervals so that a complete 360° scan of the sample was obtained. The sample holder is linked to the scanner and rotates automatically (Fig. 1), making the instrument simpler to operate than the one used by Sander and Gerke (2007). The five scans obtained by rotation of the sample, known as a scan family, were assembled to create a partial three-dimensional image of the sample using ScanStudio PRO software (NextEngine, Inc., Santa Monica, CA). Upon completion of the first scan family, the sample was repositioned on the sample holder to include the remaining unscanned portions of the clod. The sample was rescanned five times with a 72° rotation between each scan and the scans were assembled into a second scan family. Identical points on the surfaces of the two scan families were matched and used to align the images, creating a three-dimensional virtual model of the clod. In the process, digital color images were taken and automatically overlaid on the scanned image to create a three-dimensional color image. Clod volumes were calculated using 1591

Table 1. Sample locations and characteristics. Sample

Sampling location

Horizon Soil depth cm

Textural class

Gravel content %

Alfisol

North Mountain Experimental Area, Bt San Jacinto Mountains, CA

20–60

gravelly sandy clay loam

26

Vertisol Entisol Aridisol Rock

Carlsbad, CA Forest Falls, CA Southern Fry Mountains, CA Bishop Creek moraines, Bishop, CA

38–69 0–20 36–82 NA

sandy clay gravelly loamy coarse sand silt loam NA

0 22 0 NA

A3 A 2Btk2 NA†

† NA, not applicable.

the ScanStudio Pro software. Each clod was placed on a balance and the mass was recorded. Gravel was removed and weighed after volumes were determined using the paraffin-coated clod method (see below). Bulk densities (ρb) were calculated as

ρb =

W c −W g

(

V c − W g 2.65

)

[1]

where Wc is the mass of the clod, Wg is the weight of the gravel, Vc is the volume, and 2.65 g cm−3 is the assumed average particle density of the gravel. Volumes and bulk densities of the rock samples were determined using the same method.

W c −W r −W g

(

V c − W g 2.65

)

where Wr is the mass of soil lost from the clod during handling. Rock fragment volumes were obtained by displacement of water without paraffin coats because the samples were nonporous.

RESULTS AND DISCUSSION

After soil clods were scanned, volumes were reanalyzed using the paraffin-coated clod method (Grossman and Reinsch, 2002; Soil Survey Staff, 1996; Blake and Hartge, 1986). This method has been found to yield bulk density values indistinguishable from the similar saran-coated clod method (Shipp and Matelski, 1965). Oven-dried clods were secured in a hairnet and weighed. The mass of the hairnet was obtained and recorded separately. Soil material that separated from the clod during handling was collected in a tared dish and weighed. Each sample was repeatedly dipped in liquid paraffin until the entire clod was sealed. Clods were suspended until the paraffin solidified and masses were recorded. A 1000-mL beaker filled with approximately 600 mL of water was weighed on a top-loading balance and the mass was recorded. The clod was suspended and completely submerged in the beaker, keeping

Volume measurements by the three-dimensional scanner and the clod method in Fig. 2 showed excellent agreement (r2 = 0.999, P < 0.001). Bulk densities calculated from the volume measurements showed close agreement across the range of textures used in this study (Fig. 3). Both methods accurately measured the densities of the control (rock) samples at 2.60 g cm−3 (Fig. 3). Scanning time varies depending on the number of scans needed and the resolution settings used. With the resolution settings used in this study (0.13 mm), the entire scanning and image processing procedure took approximately 40 min per sample. Since sample holder rotation is automatic, the user only needs to position the sample once for each scan family. Under the resolution settings used in this study, each scan family took 15 min to scan. The paraffin-coated clod method is more labor intensive than the scanning method, although multiple samples can be processed simultaneously. The clod method becomes more complicated in coarse-

Fig. 1. The three-dimensional scanner and sample holder.

Fig. 2. Three-dimensional (3D) scanned volumes as a function of paraffincoated volumes; 1:1 line shown for reference.

Paraffin-Coated Clod Method

1592

ρb =

the clod away from the sides and bottom. The beaker was reweighed and the mass recorded. The change in mass is equivalent to the volume of water displaced by the clod. Gravel was separated from the paraffin-coated clod, weighed, and the mass recorded following Hirmas and Furquim (2006). Bulk density was calculated as

SSSAJ: Volume 72: Number 6 • November–December 2008

Fig. 3. Bulk density measurement comparisons using volume determinations from the three-dimensional (3D) scanning method and the paraffin-coated clod method; ALF = Alfisol, VERT = Vertisol, ENT = Entisol, ARID = Aridisol. Error bars show one standard deviation. Note: Rock samples were nonporous and did not require paraffin coating to determine volume by displacement.

textured samples because the paraffin coat must be removed from the clod to sieve the sample. Gravel and soil particles are cemented by the paraffin and must be separated by hand or by melting the paraffin (Hirmas and Furquim, 2006). Paraffin remaining on the gravel or gravel-sized cemented soil aggregates will contribute to the mass of gravel, resulting in underestimates of bulk density (Hirmas and Furquim, 2006). The three-dimensional scanning method eliminates the use of paraffin or other coatings, so gravel may be easily removed from the clod by sieving. In addition, inconsistencies in the paraffin coat can introduce errors into bulk density calculations. Incomplete coats allow water to enter soil clods, resulting in inaccurate volume measurements. Variations in paraffin temperature and soil texture can influence the depth of paraffin penetration into soil pores, changing the coating thicknesses (Soil Survey Staff, 1996) and affecting volume measurements. Since the scanning method is nondestructive, clods may be used for other analyses, such as for preparation of thin sections for micromorphology studies or porosity studies using computed tomography scanning. The destructive nature of the traditional clod method makes further analyses on the same clod impossible. Detailed three-dimensional images of peds can also be used to display soil structural units as in Fig. 4. This opens up the possibility for quantitative determination of ped properties that relate to structure type, size, and grade. Ped surface roughness can be assessed with the detailed images produced by this technology, allowing correlation with other physical properties such as linings and slickensides. In summary, the automated three-dimensional scanner provides a relatively inexpensive solution to the limitations of the traditional clod method. While the time required to analyze each clod is typically greater than with the paraffin-coated clod method, the three-dimensional scanner’s real advantage is that it preserves the clod for continued analyses. High-resolution three-dimensional imaging opens up new possibilities for quantitative soil morphological analyses.

SSSAJ: Volume 72: Number 6 • November–December 2008

Fig. 4. Three-dimensional images of scanned clods: (A) angular blocky ped, (B) subangular blocky ped, (C) columnar ped, (D) prismatic ped. Scale bars equal 1 cm.

ACKNOWLEDGMENTS This research was funded in part by the University of California Kearney Foundation of Soil Science. We thank NextEngine of Santa Monica, CA, for ScanStudio Pro software and for providing technical support. REFERENCES Blake, G.R., and K.H. Hartge. 1986. Bulk density. p. 363–375. In A. Klute (ed.) Methods of soil analysis. Part 1. 2nd ed. Agron. Monogr. 9. ASA and SSSA, Madison, WI. Cunningham, R.L., and R.P. Matelski. 1968. Bulk density measurements on certain soils high in coarse fragments. Soil Sci. Soc. Am. Proc. 32:109–111. Grossman, R.B., and T.G. Reinsch. 2002. Bulk density and linear extensibility. p. 201–228. In J.H. Dane and G.C. Topp (ed.) Methods of soil analysis. Part 4. SSSA Book Ser. 5. SSSA, Madison, WI. Hillel, D. 1998. Environmental soil physics. Academic Press, San Diego. Hirmas, D.R., and S.A.C. Furquim. 2006. Simple modification of the clod method for determining bulk density of very gravelly soils. Commun. Soil Sci. Plant Anal. 37:899–906. Mason, D.D., J.F. Lutz, and R.G. Petersen. 1957. Hydraulic conductivity as related to certain soil properties in a number of great soil groups: Sampling errors involved. Soil Sci. Soc. Am. Proc. 21:554–560. R Development Core Team. 2007. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. Sander, T., and H.H. Gerke. 2007. Noncontact shrinkage curve determination for soil clods and aggregates by three-dimensional optical scanning. Soil Sci. Soc. Am. J. 71:1448–1454. Shipp, R.F., and R.P. Matelski. 1965. Bulk-density and coarse-fragment determinations on some Pennsylvania soils. Soil Sci. 99:392–397. Soil Survey Division Staff. 1993. Soil survey manual. Agric. Handbk. 18. U.S. Gov. Print. Office, Washington, DC. Soil Survey Staff. 1996. Soil survey laboratory methods manual. Soil Surv. Invest. Rep. 42. Version 3.0. U.S. Gov. Print. Office, Washington, DC. Van Remortel, R.D., and D.A. Shields. 1993. Comparison of clod and core methods for determination of soil bulk density. Commun. Soil Sci. Plant Anal. 24:2517–2528.

1593