Elsevier

Geoderma

Volume 262, 15 January 2016, Pages 20-34
Geoderma

Testing a new automated single ring infiltrometer for Beerkan infiltration experiments

https://doi.org/10.1016/j.geoderma.2015.08.006Get rights and content

Highlights

  • Testing a cheap and easy to use infiltrometer

  • Automating data collection for Beerkan infiltration experiments

  • Assessing the accuracy of BEST methods

  • Characterizing water infiltration at the surface of contrasting soils

Abstract

The Beerkan method along with BEST algorithms is an alternative technique to conventional laboratory or field measurements for rapid and low-cost estimation of soil hydraulic properties. The Beerkan method is simple to conduct but requires an operator to repeatedly pour known volumes of water through a ring positioned at the soil surface. A cheap infiltrometer equipped with a data acquisition system was recently designed to automate Beerkan infiltration experiments. In this paper, the current prototype of the automated infiltrometer was tested to validate its applicability to the Beerkan infiltration experiment under several experimental circumstances. In addition, the accuracy of the estimated saturated soil hydraulic conductivity, Ks, and sorptivity, S, was assessed by applying different BEST algorithms to the data obtained with the infiltrometer. At this purpose, both analytically generated and real experimental data were used. The analytical assessment showed that the use of the infiltrometer along with BEST methods could lead to accurate estimates of the considered soil properties in most cases, which validated the design of the infiltrometer and its combination with BEST algorithms. Loamy soils and high initial water contents led to misestimating Ks and S or to failure of BEST algorithms, but advices about the infiltrometer design were developed to alleviate such problems. A comparison between the automated procedure and the original BEST procedure was made at three field sites in Sicily (Italy). Other experiments were carried out in an infiltration basin located in the pumping well field of Crépieux-Charmy (Lyon, France), in order to assess the ability of the automated infiltrometer to check clogging effects on Ks. The experiments showed that the automatic data collection increased measurement speed, allowed a more efficient data handling and analysis, and reduced sensitivity of the calculated hydraulic parameters on the applied BEST algorithm.

Introduction

The saturated soil hydraulic conductivity, Ks, and the soil sorptivity, S, are important soil properties controlling water infiltration and movement into the unsaturated soil profile. Saturated hydraulic conductivity depends strongly on soil texture and structure whereas sorptivity also depends on the initial and final soil water contents and, when present, the water depth at soil surface (Touma et al., 2007). Both soil hydraulic properties thus exhibit strong spatial and temporal variations and a large number of determinations are required to assess the magnitude of the variation within the selected area (Logsdon and Jaynes, 1996). Assessment of simple and rapid field techniques is therefore important to obtain reliable data with a sustainable effort.

The Beerkan Estimation of Soil Transfer (BEST) parameters procedure by Lassabatere et al. (2006) is very attractive for practical use since it allows an estimation of both the soil water retention and the hydraulic conductivity functions from cumulative infiltration collected during a ponded field experiment and a few routinely laboratory determinations. Lassabatere et al. (2006) suggested to measure the infiltration time of small volumes of water repeatedly poured on the soil surface confined by a ring inserted to a depth of about 1 cm into the soil. BEST considers a zero ponded infiltration model which was assumed to be appropriate for an infiltration run performed with small, but positive, pressure heads. This assumption was supported by numerical tests carried out by Touma et al. (2007). Yet, several problems arise with this method, including (i) the need for an operator over the whole duration of the experiment; (ii) the need to reach steady state infiltration, which can be extremely long in certain cases; and (iii) the experimental error in the measured infiltration times and the variable skillness among operators. Moreover several algorithms were developed to analyze the infiltration data, i.e., BEST-slope (Lassabatere et al., 2006), BEST-intercept (Yilmaz et al., 2010) and BEST-steady (Bagarello et al., 2014b), but the relative performance of the alternative algorithms has not yet been tested.

Automatic data collection increases measurement speed, permits measurement at short time intervals, improves measurement precision, allows for more efficient data handling and analysis, and reduces the amount of effort involved and the potential for errors that may occur when manual procedures are applied (Madsen and Chandler, 2007, Dohnal et al., 2010). Nevertheless, the advantages of simplified methodologies, such as BEST, are their simplicity and cheapness. The use of expensive devices or time consuming procedures contradicts the original purpose of these simplified methodologies and monitoring equipment often contains proprietary technology with prohibitive cost. Yet, rapid advances in electronic technologies have allowed researchers and practitioners access to low-cost, solid-state sensors and programmable microcontroller-based circuits (Fisher and Gould, 2012).

Recently, Di Prima (submitted for publication) developed a method to automate data collection with a compact infiltrometer under constant head conditions. The device, maintaining a small quasi-constant head of water (i.e., 2–3 mm) on the infiltration surface, is equipped with a differential pressure transducer to measure the stepwise drop of water level in the reservoir, and, in turn, to quantify cumulative infiltration into the soil. The data acquisition system has been designed with low cost components and it is based on the open source microcontroller platform, Arduino. Total measurable cumulative infiltration and the increment between two successive experimental points are fixed, since they depend on the capacity of the Mariotte reservoir and the radius of air entry tube, respectively. The very limited cost of the system could represent a step towards a cheaper and more widespread application of accurate and automated infiltration rate measurement. However, the current version of the infiltrometer has not been tested yet against a wide range of experimental conditions in terms of soils and initial water contents.

The main objective of this paper was to check the usability of the device to automatize the Beerkan infiltration experiment and to analyze the infiltration data to characterize soil hydraulic properties. The focus is put on the derivation of saturated soil hydraulic conductivity and soil sorptivity by using the combination of the automated infiltrometer and the three BEST algorithms. The proposed combination is assessed by using both analytically generated and field data and with regard to reliable predictions of the saturated soil hydraulic conductivity and soil sorptivity.

Section snippets

Automated infiltrometer

The automatic infiltrometer by Di Prima (submitted for publication) (Fig. 1) allows to maintain a small constant water head on a soil surface confined by a 150 mm inner diameter ring using a Mariotte bottle for water supply. Depending on the surface roughness, the Mariotte bottle can be regulated in height so that the surface confined by the ring is entirely submerged under a practically negligible water depth, i.e., 2–3 mm. The bottle has an inner diameter of 94 mm and a height of 520 mm, allowing

Testing the infiltrometer with analytically generated data

As a first step, analytically generated data were used to assess the accuracy of the BEST predictions of Ks and S obtained by the infiltration of 130 mm of water, sampled every 5 mm, through a 150 mm diameter soil surface. Then, a sensitivity analysis was performed to investigate the influence of total cumulative infiltration and infiltration increments. The Beerkan infiltration experiments were modeled for five soils (sand, sandy loam, sandy clay loam, loam and clay), using the parameters listed

Comparison between BEST-deduced parameters and targeted values

As a first step, a total infiltration of 130 mm of water with 5-mm increments and through a 150 mm diameter source was considered (Fig. 3). Cumulative infiltration could be calculated for all cases and exhibited usual shapes, with a concave part corresponding to the transient state and a linear part at the end of the curves related to the steady state. Note that the time for infiltration of a given water volume increased with the initial effective saturation degree. A lower initial Se value

Conclusions

The BEST method is an alternative technique to conventional laboratory or field measurements for rapid and low-cost estimation of soil hydraulic properties that is based on trustworthy and robust analytical solution. In this paper, the potential of a new automated infiltrometer, allowing infiltration under a practically null constant head of water at the soil surface, was tested. The relative performance of the three existing BEST algorithms, i.e., BEST-slope, BEST-intercept and BEST-steady, to

Acknowledgments

This study was supported by grants from the “Ambassade de France en Italie” and the Università degli Studi di Palermo (Dottorato di Ricerca in Sistemi Agro-Ambientali, indirizzo Idronomia Ambientale).

References (47)

  • J.R. Philip

    Theory of infiltration

    Adv. Hydrosci.

    (1969)
  • X. Xu et al.

    Analysis of single-ring infiltrometer data for soil hydraulic properties estimation: Comparison of BEST and Wu methods

    Agric. Water Manag.

    (2012)
  • M.D. Ankeny et al.

    Design for an automated tension infiltrometer

    Soil Sci. Soc. Am. J.

    (1988)
  • V. Bagarello et al.

    Estimating field-saturated soil hydraulic conductivity by a simplified Beerkan infiltration experiment

    Hydrol. Process.

    (2014)
  • V. Bagarello et al.

    Comparing alternative algorithms to analyze the Beerkan infiltration experiment

    Soil Sci. Soc. Am. J.

    (2014)
  • U. Buczko et al.

    Water repellency in sandy luvisols under different transformation stage in northeast Germany

    Geoderma

    (2002)
  • R.F. Carsel et al.

    Developing joint probability distributions of soil water retention characteristics

    Water Resour. Res.

    (1988)
  • F.X.M. Casey et al.

    Improved design for an automated tension infiltrometer

    Soil Sci. Soc. Am. J.

    (2002)
  • J. Constantz et al.

    An automated technique for flow measurement from Mariotte reservoirs

    Soil Sci. Soc. Am. J.

    (1987)
  • S. Di Prima

    Automatic analysis of multiple Beerkan infiltration experiments for soil hydraulic characterization

  • S. Di Prima

    Automated single ring infiltrometer with a low-cost microcontroller circuit

    Comput. Electron. Agric.

    (2015)
  • M. Dohnal et al.

    Improving hydraulic conductivity estimates from minidisk infiltrometer measurements for soils with wide pore-size distributions

    Soil Sci. Soc. Am. J.

    (2010)
  • D.K. Fisher et al.

    Open-source hardware is a low-cost alternative for scientific instrumentation and research

    Mod. Instrum.

    (2012)
  • Cited by (62)

    • A novel frequency-domain integrated sensor for in-situ estimating unsaturated soil hydraulic conductivity

      2022, Journal of Hydrology
      Citation Excerpt :

      Therefore, accurate, low-cost and in-situ measurement of Ku remains challenged. To date, in-situ Ku can be measured by tension disc infiltrometer and pressure ring infiltrometer (Angulo-Jaramillo et al., 2000) or indirectly determined with fitted parameters of soil water characteristic curves (SWCC) based on hydraulic models (e.g. Campbell, 1974; van Genuchten, 1980) and the measured saturated soil hydraulic conductivity (Ks) (Burdine, 1953; Mualem, 1976; van Genuchten, 1980; Ankeny et al., 1991; Shani and Or, 1995; Kosugi and Nakayama, 1997; Šimůnek et al., 1998a; Angulo-Jaramillo et al., 2000; Assouline, 2001; Di Prima et al., 2016; Tian et al., 2019) in the field. The accuracy of parameters of the SWCC significantly affected the determination of Ku.

    View all citing articles on Scopus
    View full text