PedoTransfer Functions (PTFs) are widely used to predict soil functioning in agricultural and environmental systems. The term PTF is used to describe equations that express dependencies of soil properties on basic soil attributes available from soil surveys. A prerequisite for the development of PTFs is the availability of a source database that contains potential predictors. Most PTFs available in the literature use soil texture, bulk density, and organic matter contents as predictors. Although the soil Water Retention Curve (WRC) is of great importance in present-day agricultural and environmental soil research, it is not an easily available soil property. The same can be stated for the Soil Resistance Curve (SRC). The main reason is that their measurement are expensive, time consuming, and labor intensive. Therefore, models need to be developed to predict the WRC and SRC from more easily measurable and more readily available soil properties. This study tested the hypothesis that pedotransfer functions (PTFs) that describe the influence of soil properties and management on the WRC and the SRC could be determined from the information available at the soil survey of the State of Santa Fe, Argentina. The objectives were: i) to determine the water release curves and soil resistance curves for the main soil series of the Santa Fe State, ii) to develop PTFs for both soil properties. Soil Series (10) were selected from the Soil Map (1:50000) taking into account the surface that they cover in the county and their productivity. They belong to the following soil types: Typic Argiudol, Aquic Argiudol, Typic Hapludol, Entic Hapludol. Undisturbed and disturbed soil samples were taken from the A and B horizons. Undisturbed samples (cores of 5 cm height, 5 cm diameter) were used to determine soil bulk density, soil water retention curve and soil penetration resistance curve. Disturbed soil samples were used to measure soil particle distribution, soil particle density, and organic carbon content. Cores were saturated with water and split into 15 groups. The following potentials were applied using a tension table: -0.001, -0.002, -0.003, -0.004, 0.005, 0.006, -0.008 MPa and –0.01 MPa. Pressure plates were used to equilibrate samples at potentials: -0.02, -0.03, -0.06, -0.1, -0.3, -0.6 and -1.5 MPa. After equilibrium, the samples were used to determine the soil resistance curve (SRC), and then oven dried at 105-110˚C. Soil water content and bulk density (Db) were also determined. Gravimetric water content was converted to volumetric water contents using the measured Db of each core, and the soil WRC was determined. Penetration resistance was measured using an electronic penetrometer provided with an electronic load cell of 20 kg capacity, with a cone of 4 mm diameter and semi-angle of 30º. The rate of penetration was set up to 1.0 cm min^-1. The measurements obtained from 1 to 4 cm of depth were averaged for each core. The electrical output was recorded by a data acquisition system. The soil water release data were fitted using the following function:

Θ = Θr + (Θs - Θr) / (1 + (a |Ψ|ⁿ))^(1+1/n)

Where Θ is the volumetric water content (cm³ cm^-3), Θr is the residual water content (cm³ cm^-3), Θs is the saturation water content (cm³ cm^-3), Ψ is the matric pressure (MPa), and a and n are constants. The influence of soil bulk density (Db), soil texture, organic matter, and all possible two-way interactions on the soil water release curve was incorporated in Θs, Θr, a, and n, as shown for a:

a = (a₀ + a₁ x₁ + a₂ x₂ + ... + a_i x_i-1xi)

Where a₀ ... a_i are the regression's coefficients, and x_i are Db, soil texture, and organic matter. Stepwise multiple linear regression was used to select the significant terms (P >0.15).

Soil resistance data were regressed against Θ and Db using the following model:

ln SR= ln d + e lnΘ + f lnDb

Where d, e, and f are constants and SR is the soil resistance (MPa). The influence of soil texture and organic matter, and their two-way interaction on d, e, and f was assessed using the same approach as was used for the water release curve, i.e.:

In SR=(d₀ + d₁ x₁ + d₂ x₂ + d₃ x₁ x₂) + (e₀ + e₁ x₁ + e₂ x₂ + e₃ x₁ x₂) ln Θ + (f₀ + f₁ x₁ + f₂ x₂ + f₃ x₁ x₂) lnDb

Where d_i, e_i, and f_i are the regression's coefficients and x₁ and x₂ are soil texture and organic matter, respectively. Stepwise multiple linear regression was used to select the significant terms (P > 0.15).

Results indicate that clay (CL), silt (SI) and sand (SA) contents varied from 11 to 62%, from 15 to 69%, and from 2 to 73%, respectively. Organic carbon (OC) varied from 0.8 to 48 g kg^-1, and the soil bulk density (Db) varied from 1.10 to 1.68 g cm^-3. Multiple regression analyses showed that WRC was related (R² = 0.96; F=503.8; p<0.0001) with particle size distribution, and OC. The SA and OC content influenced a, CL content affected Θs, and SI had an effect on Θr. The parameter n was affected by the CL and OC contents. On the other hand, SR was related (p<0.0001) positively with Db, CL and OC, and negatively with Θ. In addition, there was an interaction between OC and Θ (R² = 0.85). In summary, our findings show that using a set of relevant data was possible to develop PTFs, which in turn fit well the WRC and SRC data. The next step will consist in assessing the PTFs reliability by examining the correspondence between measured and estimated for data set other than the one used to develop the PTFs.