Soil Science Society of America Journal

Soil Sci. Soc. Am. J. 83:327–331 doi:10.2136/sssaj2018.10.0378 Received 9 Oct. 2018. Accepted 30 Dec. 2018. *Corresponding author (viktor.polyakov@ars.usda.gov). © 2019 The Author(s). Re-use requires permission from the publisher.

A Simple Automated Laser Profile Meter

Soil Physics & Hydrology Note

An automated profile meter with vertical precision of ±0.2 mm was built to measure soil surface elevations. It consists of a laser distance meter mounted on an automated linear motion system. Operation and data logging is con-ducted wirelessly using a Bluetooth enabled mobile device. The profile meter performed well under various lighting conditions and on a variety of natural materials, with the exception of specular surfaces. It is inexpensive (<$1500), consists of readily available components, portable, and easy to setup in the field (<10 min). When measuring soil random roughness the profile meter was compared with a ground based terrestrial Light detection and ranging (lidar). The lidar systematically overestimated random roughness by approxi-mately 15% compared with the profile meter.

Abbreviations: lidar, light detection and ranging; SfM, structure from motion.

Surface roughness is an important soil characteristic that affects runoff, in-filtration, surface storage, and evaporation. It controls overland flow gen-eration (Darboux et al., 2002), hydraulics resistance and sediment transport

(Abrahams et al., 1998). Roughness is a dynamic property, which evolves in time and space (Cremers et al., 1996) under anthropogenic, climatic, and biological fac-tors (Eltz and Norton, 1997). Typically it increases under tillage and decreases under rainfall impact (Bertol et al. 2010; de Oro and Buschiazzo, 2011). Researchers char-acterize soil roughness using a variety of indexes (Vermang et al., 2013) ranging from simple standard deviation (Kuipers, 1957) to fractal dimension (Chi et al., 2012).

To calculate these indexes, soil micro-topography needs to be measured quickly and accurately. Available measurement methods can be classified as contact, such as pin meter (Kuipers, 1957) or chain (Saleh, 1993), and non-contact or remote. Among the latter are photogrammetry (Bretar et al., 2013; Taconet and Ciarletti, 2007), Structure from Motion (SfM; Snapir et al., 2014), laser scanning (Huang and Bradford, 1990), terrestrial and airborne lidar (Turner et al., 2014), ultrasonic (Robichaud and Molnau, 1990), and infrared (Mankoff and Russo, 2013) techniques.

Contact methods have limited resolution and precision ( Jester and Klik, 2005) and cannot be used on fragile surfaces. Non-contact methods can be tech-nically or operationally complex, time-consuming or costly. They often require calibration (Huang and Bradford, 1990; Mankoff and Russo, 2013) or substantial data post-processing (Bretar et al., 2013), both of which add to the uncertainty of measurements. In addition, data acquisition by non-contact methods in the field poses challenges for some instruments. For example, ground-based terrestrial li-dar has a low viewing angle resulting in increased data noise and data gaps (Eltner and Baumgart, 2015), three-dimensional laser scanners have been reported to per-form poorly on dark and wet surfaces (Rieke-Zapp et al., 2007) and in very bright sunlight ( Jester and Klik, 2005), infrared sensors suffer from soil radiant heat (Mankoff and Russo, 2013), and SfM techniques don’t work well in areas with low color contrast, high surface homogeneity, or poor lighting.

V. Polyakov* M. Nearing

Southwest Watershed Research Center USDA-ARS Tucson, AZ 85719

Core Ideas

•The proposed laser profile meter is simple yet offers good accuracy and precision.

•Data obtained by the profile meter compared well with those obtained by lidar.

•Profile meter provides a practical alternative to more expensive or operationally complex systems.

Published April 4, 2019

328 Soil Science Society of America Journal

Considering these challenges, there is a need for an accu-rate, affordable, simple to operate, and easily portable instrument for micro-topography measurements. While estimation of sur-face roughness requires accurate high resolution elevation data, three-dimensional coverage is not always necessary. Hence, the goal of this technical note is to describe the capability and use of a simple laser profile meter built using readily available off-the-shelf components and software.

MeThODsexperimental equipment and setup

The profile meter consists of a laser distance meter (DISTO E7500i)1 by Leica (2018) mounted on a linear motion system (Stage Zero) by Dynamic-Perception (2018). The entire appara-tus has adjustable support legs that allow it to be placed approxi-mately 0.5 m over the studied surface (Fig. 1). During operation the laser meter moves along a straight line and acquires profile elevation data in automatic mode.

The E7500i uses a Class 2 laser with a 635-nm wavelength and 0.03° beam divergence. The instrument has a range of 0.05 to 200 m and a manufacturer specified measuring tolerance of ±1 mm at a 10-m distance. Short-range accuracy and precision is discussed further below. In addition to single push of a but-ton measurements, the E7500i features a continuous laser mode where data is acquired automatically at a set time interval. Other critical features are Bluetooth connectivity (SIG Corp., Bellevue, WA) and real time data logging into a Microsoft Excel spread-sheet (Microsoft Corp. Redmond, WA). Data logging capabil-ity is enabled through DISTO Transfer BLE application avail-able for Android and Microsoft Windows operating systems (Microsoft Corp. Redmond, WA).

The Stage Zero linear motion system consists of a rail, belt driven carriage with various mounting options, NMX motion controller, stepper motor, and a battery pack. The NMX is a 3-axis time-lapse and real time motion controller with Bluetooth module. It is programmed and operated wirelessly via software available for Android and iOS operating systems. This particu-lar motion system was chosen because of superior rigidity and straightness of the linear rail (part # 25–2576) available in vari-ous lengths up to 6 m (80/20 Inc., 2018). The downward deflec-tion of the rail loaded with the carriage, battery, and laser meter does not exceed 0.1 mm over 1.5 m span.

Laser Meter PerformanceThe ability of the E7500i to acquire elevation data was test-

ed on a variety of surfaces over a range of reflectivity, transpar-ency, and textures at different angles of attack and with full or partial reflection. The frequency of measurements in automatic mode was calculated by counting the number of data points ob-tained over 60-s intervals.

The accuracy of the E7500i was estimated using a digital caliper (accurate to 0.01 mm). A target was attached to a mov-

able jaw of the caliper secured 500 mm away from the laser meter. The laser beam was pointed at the target while the target was in-crementally moved over the range of 20 mm. The measurements (n = 30) by laser meter and caliper were recorded and compared.

The precision of the laser meter was estimated under static and dynamic conditions. Under static conditions the distance to an object (flat matte surface) 0.5 m away was measured re-petitively while the instrument was stationary. Under dynamic conditions similar measurements were obtained while the instru-ment was traversing back and forth on the linear rail over a flat horizontal surface located 0.5 m away.

To further characterize the reproducibility of laser mea-surements, smooth (silty clay) and rough (gravelly) soils were scanned at 5-mm resolution over 1300-mm long profiles. The scans were repeated five times over the same transect on each of the soils. Random roughness of the surface was calculated as the standard deviation of the slope-adjusted elevation measurements. Roughness, and other statistics of the two surfaces were compared.

The profile meter was used in soil surface evolution experi-ments on artificial plots under simulated rainfall (Nearing et al., 2017). Eighty-four 2-m long elevation transects at 5 mm resolu-tion were obtained. These transects represented different stages of erosion at a variety of slopes and slope positions. On the same plots the soil surface was measured using a Riegl VZ-400 terres-trial lidar (Riegl Laser Measurement Systems, 2018a). The scans were conducted from six positions with the instrument located approximately 2 m above soil surface. Tie-points were used to align the scans done from each position into a single point cloud with average point density of 1 × 105 m-2 (LiLi, M.A. Nearing, M.H. Nichols, V. Polyakov and M.L. Cavanaugh, unpublished data, 2018). RiSCAN PRO software (Riegl Laser Measurement Systems, 2018b) was used to clip 2 m by 0.02 m bands of point data that spatially matched profile meter transects. These bands were used to calculate random roughness and compare it with the profile meter data.

setup and OperationBoth instruments (the laser meter and the linear rail) can be

operated simultaneously from a single mobile device with split screen capability (Android 6.0 and above) or from two separate mobile devices. The frequency of data acquisition by the laser in automatic mode is constant. Hence, the distance between data points in the profile is controlled through velocity of the carriage. The travel time (s) of the laser along transect is determined as:

t = L/l/n [1]

where L is transect length (mm), l is desired horizontal reso-lution (mm), and n is the frequency of data acquisition by the E7500i in automatic mode (Hz).

The experimental procedure was as follows. First, the pro-file meter was positioned over the surface and two elevated targets (square blocks) were placed at the opposite ends of the profile. For simplicity and repeatability, the targets can be per-1 Trade names and company names are included for the benefit of the reader and do not

imply endorsement by the USDA.

www.soils.org/publications/sssaj 329

manently attached to the rail support legs. The NMX software was launched and the rail carriage was moved to the beginning and to the end of the profile above the targets. These two posi-tions were set as start and stop and the travel time (Eq. [1]) was entered into the program. The DISTO Transfer application was initiated and the laser was started in automatic mode. Then the carriage was launched and on completion of the profile the data file was saved. The total setup time in the field was approximately 5 to 10 min.

ResuLTsProfile Meter Performance

The cross-section of the E7500i laser beam was approxi-mately 2 by 4 mm at the 0.5-m distance. When partial reflection occurred, such as part of the beam reflects from the edge of an object and part from the object behind it, the instrument record-ed the larger of the two reflections. The laser meter could reliably measure the height of an isolated particle wider than 3 mm.

Distance measurement could be obtained at angles between 90 and 8° to the surface. The laser meter was able to acquire data from a wide variety of artificial and natural objects, both organic (plant parts, surface litter) and mineral (rough, smooth, or opaque rock fragments, glazed ceramic, clay). It performed equally well in the dark and under bright sunlight. The laser me-ter failed on surfaces with specular reflection, such as glass mir-ror or polished metal, and on objects submerged under water >2 mm deep. The latter is an important consideration when deal-ing with water flow or ponding on the soil surface. In automatic

mode, the E7500i obtained measurements with a frequency of 1.872 ± 0.011 Hz (p < 0.05).

The accuracy of the E7500i at the 0.5-m distance was esti-mated at ±0.07 mm (p < 0.05, n = 31) on flat surface. Precision of the instrument is defined as reproducibility of the repeated measurements. In the static test, 95% of the repeat measurements were within ±0.2 mm from the mean value at approximately 0.5 m distance (Table 1). Note, that the tests were conducted on dif-ferent occasions and average elevations are not directly compa-rable. The precision was largely unaffected by the laser moving along the rail. However, the movement caused outliers that in-creased the total range of values from 0.5 mm under static to 0.9 mm under dynamic conditions, even though the standard devia-tions of the flat surface were similar (Table 1).

Good reproducibility of transect measurements was achieved on soil surfaces (Table 1). The average elevation of fine and gravelly soils was 497.23 ± 0.06 and 501.33 ± 0.05 mm respectively after five consecutive scans. Random roughness of these respective surfaces was estimated at 2.92 ± 0.06 and 8.66 ± 0.08 mm. The uncertainty on irregular surface results from the combination of instrument uncertainty, vibration, and horizon-tal misalignment of consecutive scans. A better horizontal align-ment of multiple scans over the same transect might be desired when monitoring surface elevation change over time. This can be achieved by placing permanent reference blocks at either end of the profile and use those to align the digital profiles.

Fig. 1. Profile meter consisting of e7500i laser meter (a), carriage (b), stepper motor (c), controller (d), rail (e), rail support legs (f), and battery (g).

330 Soil Science Society of America Journal

Field experiment

A comparison of random roughness values obtained by the laser profile meter and by Riegl VZ-400 terrestrial lidar scanner during a rainfall simulation experiment (Nearing et al., 2017) is presented on Fig. 2 for 84 scans. Random roughness ranged from 2 to 8 mm and increased nonlinearly with cumulative soil ero-sion. The lidar overestimated random roughness over the entire range by approximately 15% compared with the profile meter. However, the slope of the regression line between the two data-sets is not significantly different from 1 and the intercept is not significantly different from 0 (p < 0.05). The discrepancy be-tween the two methods might be attributed to the nature of the lidar point cloud, which was a composite of six scans. Multiple scans from different positions are needed to compensate for oblique field of view of the instrument and properly sample el-evations in narrow depressions, such as between soil clods. Any vertical misalignment of the individual scans as well as random

noise inherent in the lidar data (Eltner and Baumgart, 2015) increase variability. In contrast, minimal noise was observed in profile meter data. This could be seen on surfaces with regular geometry, where outliers are easily identifiable.

CONCLusIONA simple portable non-contact profile meter was built to

measure soil surface elevations. It consists of off-the shelf com-ponents (laser distance meter and automated linear rail) and is operated wirelessly from a mobile device, such as a mobile phone, using free software. The profile meter is relatively cheap (<$1500 not including a mobile device) and can be constructed quickly without the need for expert knowledge. Low cost of hardware makes the instrument accessible to a wider scientific and educa-tion community.

The profile meter has good vertical accuracy (±0.07 mm) and precision (±0.2 mm) and can be programed to sample at any horizontal resolution. It works reliably on a wide variety of natu-ral materials under any lighting conditions. It requires no eleva-tion calibration or post processing of the output data. An impor-tant advantage over similar systems is vertical orientation of the laser beam and the optical detector, which provides orthogonal view thus minimizing data gaps in narrow depressions or around protruding elements of the soil. Measurements of long transects (up to 5.5 m) are possible but may require additional support or stiffening of the linear rail. The profile meter can be transported and operated by one person. It takes <10 min to set up and <2 min to measure 200 elevation points.

While the proposed apparatus measures elevations only along a single profile, there are many applications where such datasets will satisfy research needs, for example, calculating ran-dom roughness of soil, measuring and monitoring channel and rill cross-sections, quantifying soil cover, etc. When performing repetitive scans of the same cross-section, such as monitoring temporal changes of the surface, extra steps are needed for hori-zontal alignment of the digital profiles.

Soil roughness obtained by the profile meter compared well with those obtained by a ground based terrestrial lidar. Our ap-paratus provides a simple and practical alternative to more ex-pensive or operationally complex systems.

Fig. 2. Comparison of random roughness obtained by the laser profile meter and terrestrial lidar scanner on a soil plot collected for a rainfall simulation experiment. each dot (n = 84) represents a 1.8-m transect consisting of approximately 360 elevation points.

Table 1. Profile meter performance. Laser precision under static and dynamic conditions, and repeatability of profile data acquired on smooth and gravelly soil surface from a distance of 0.5 m to the laser.

Flat surface soil profile

static Dynamic Fine Gravelly

No. of points 284 640

No. of repeat transects 5 5

Elevation, mm

Average 472.76 ± 0.01* 481.64 ± 0.01 497.23 ± 0.06 501.33 ± 0.05

Minimum 472.5 481.3 491.2 ± 0.13 479.1 ± 0.34

Maximum 473.0 482.2 503.6 ± 0.09 518.8 ± 0.36

Range 0.5 0.9 12.4 ± 0.14 39.7 ± 0.60

Standard deviation 0.10 0.09

Random roughness, mm 2.92 ± 0.02 8.66 ± 0.08* Mean standard error at p < 0.05.

www.soils.org/publications/sssaj 331

ReFeReNCes80/20 Inc. 2018. The industrial erector set, Catalog 22. Available at: http://

catalogs.8020.net/?bookcode=egt18flx. (Accessed 10 Aug. 2018; verified 4 Mar. 2019). 80/20 Inc., Columbia City, IN.

Abrahams, A.D., G. Li, C. Krishnan, and J.F. Atkinson. 1998. Predicting sediment transport by interrill overland flow on rough surfaces. Earth Surf. Process. Landf. 23:1087–1099. doi:10.1002/(SICI)1096-9837(199812)23:12<1087::AID-ESP934>3.0.CO;2-4

Bertol, I., W.A. Zoldan, A.P. Gonzalez, F.T. Barbosa, and R.D. Werner. 2010. Sediment transport in runoff on rugous soil surface submitted to simulated rainfall. Sci. Agric. 67:591–597. doi:10.1590/S0103-90162010000500013

Bretar, F., M. Arab-Sedze, J. Champion, M. Pierrot-Deseilligny, E. Heggy, and S. Jacquemoud. 2013. An advanced photogrammetric method to measure surface roughness: Application to volcanic terrains in the Piton de la Fournaise, Reunion Island. Remote Sens. Environ. 135:1–11. doi:10.1016/j.rse.2013.03.026

Chi, Y., J. Yang, D. Bogart, and X. Chu. 2012. Fractal analysis of surface microtopography and its application in understanding hydrologic processes. Trans. ASABE 55:1781–1792. doi:10.13031/2013.42370

Cremers, N., P.M. VanDijk, A.P.J. DeRoo, and M.A. Verzandvoort. 1996. Spatial and temporal variability of soil surface roughness and the application in hydrological and soil erosion modelling. Hydrol. Processes 10:1035–1047. doi:10.1002/(SICI)1099-1085(199608)10:8<1035::AID-HYP409>3.0.CO;2-#

Darboux, F., C. Gascuel-Odoux, and P. Davy. 2002. Effects of surface water storage by soil roughness on overland- flow generation. Earth Surf. Process. Landf. 27:223–233. doi:10.1002/esp.313

de Oro, L.A., and D.E. Buschiazzo. 2011. Degradation of the soil surface roughness by rainfall in two loess soils. Geoderma 164:46–53. doi:10.1016/j.geoderma.2011.05.009

Dynamic-Perception. 2018. Stage Zero digital system with NMX motion controller. Available at: http://www.dynamicperception.com/Stage-Zero-Digital-System-w-NMX-Motion-Controller-p/bdl-0-nmx1.htm. (Accessed 10 Aug. 2018, verified 4 Mar. 2019). Dynamic Perception, Ann Arbor, MI.

Eltner, A., and P. Baumgart. 2015. Accuracy constraints of terrestrial Lidar data for soil erosion measurement: Application to a Mediterranean field plot. Geomorphology 245:243–254. doi:10.1016/j.geomorph.2015.06.008

Eltz, F.L.F., and L.D. Norton. 1997. Surface roughness changes as affected by rainfall erosivity, tillage, and canopy cover. Soil Sci. Soc. Am. J. 61:1746–1755. doi:10.2136/sssaj1997.03615995006100060028x

Huang, C.H., and J.M. Bradford. 1990. Portable laser scanner for measuring soil surface roughness. Soil Sci. Soc. Am. J. 54:1402–1406. doi:10.2136/sssaj1990.03615995005400050032x

Jester, W., and A. Klik. 2005. Soil surface roughness measurement—Methods, applicability, and surface representation. Catena 64:174–192. doi:10.1016/j.catena.2005.08.005

Kuipers, H. 1957. A reliefmeter for soil cultivation studies. Neth. J. Agric. Sci. 5:255–262.

Leica. 2018. Leica DISTO E7500i, the original laser distance meter. Available at: https://lasers.leica-geosystems.com/sites/default/files/leica_media/product_documents/DISTO/prod_docs_e7500i_d510/leica-disto-e7500i-user-manual_792320_us.pdf. Leica Geosystems, Heerbrugg, Switzerland.

Mankoff, K.D., and T.A. Russo. 2013. The Kinect: A low-cost, high-resolution, short-range 3D camera. Earth Surf. Process. Landf. 38:926–936. doi:10.1002/esp.3332

Nearing, M.A., V.O. Polyakov, M.H. Nichols, M. Hernandez, L. Li, Y. Zhao, and G. Armendariz. 2017. Slope-velocity equilibrium and evolution of surface roughness on a stony hillslope. Hydrol. Earth Syst. Sci. 21:3221–3229. doi:10.5194/hess-21-3221-2017

Riegl Laser Measurement Systems. 2018a. Riegl Laser Masurement Systems. RIEGL VZ-400i. Available at: http://www.riegl.com/nc/products/terrestrial-scanning/produktdetail/product/scanner/48/. (Accessed 13 June 2018; verified 4 Mar. 2019). RIEGL Laser Measurement Systems, Horn, Austria.

Riegl Laser Measurement Systems. 2018b. RiSCAN PRO 2.0, Companion software to 3D terrestrial scanners. Available at: http://www.riegl.com/products/software-packages/riscan-pro/. (Accessed 18 Sept. 2018; verified 4 Mar. 2019). RIEGL Laser Measurement Systems, Horn, Austria.

Rieke-Zapp, D., J. Poesen, and M.A. Nearing. 2007. Effects of rock fragments incorporated in the soil matrix on concentrated flow hydraulics and erosion. Earth Surf. Process. Landf. 32:1063–1076. doi:10.1002/esp.1469

Robichaud, P.R., and M. Molnau. 1990. Measuring soil roughness changes with an ultrasonic profiler. Trans. ASAE 33:1851–1858. doi:10.13031/2013.31549

Saleh, A. 1993. Soil roughness measurement: Chain method. J. Soil Water Conserv. 48:527–529.

Snapir, B., S. Hobbs, and T.W. Waine. 2014. Roughness measurements over an agricultural soil surface with Structure from Motion. ISPRS J. Photogramm. Remote Sens. 96:210–223. doi:10.1016/j.isprsjprs.2014.07.010

Taconet, O., and V. Ciarletti. 2007. Estimating soil roughness indices on a ridge-and-furrow surface using stereo photogrammetry. Soil Tillage Res. 93:64–76. doi:10.1016/j.still.2006.03.018

Turner, R.M., R. Panciera, M.A. Tanase, K. Lowell, J.M. Hacker, and J.P. Walker. 2014. Estimation of soil surface roughness of agricultural soils using airborne LiDAR. Remote Sens. Environ. 140:107–117. doi:10.1016/j.rse.2013.08.030

Vermang, J., L.D. Norton, J.M. Baetens, C. Huang, W.M. Cornelis, and D. Gabriels. 2013. Quantification of soil surface roughness evolution under simulated rainfall. Trans. ASABE 56:505–514. doi:10.13031/2013.42670