Performing ponding infiltration runs with non-circular sources could represent a good means to sample completely an area of interest. Regardless of the shape of the source, predicting the expected reliability of the collected data by infiltrometers should facilitate soil hydraulic characterisation and also allow a more conscious use of the field data. The influence of the shape of the infiltration source (i.e., circular or square) and the analysis procedure of the steady-state infiltration data on the saturated hydraulic conductivity, Ks, of a sandy-loam soil was tested in this investigation. Circular and square surfaces sampled with the pressure infiltrometer (PI) yielded similar estimates of Ks (i.e., differing by a factor of 1.05-1.16, depending on the calculation method) when an equivalent radius was considered to geometrically describe the square source. With the simplified falling head (SFH) technique, the shape of the source was irrelevant (i.e., circular and square sources yielding Ks values that differed by a factor of 1.19), as theoretically expected. For the steady-state PI experiment, the twoponding depth approach yielded two times smaller Ks values than the one-ponding depth (OPD) approach, probably due to lower steady-state flow rates than those expected for the second phase of the two-level run. The conclusions were that: i) simple infiltrometer experiments (PI, SFH) can be carried out with square sources; and ii) the simplest PI run (OPD approach) is expected to yield the most reliable predictions of Ks. Sampling other soils is advisable in an attempt to make these conclusions of general validity.
Hydraulic conductivity field characterization from the joint inversion of hydraulic heads and self-potential data
