Calcareous soils have often more than 15% CaCO3. The best pedotransform functions may be soil water retention capacity and sodium adsorption ratio.

Here goes some good work..

Selection of a suitable model for the prediction of soil water content in north of Iran

Abstract :Multiple Linear Regression (MLR), Artificial Neural Network (ANN) and Rosetta model were employed to develop pedotransfers functions (PTFs) for soil moisture prediction using available soil properties for northern soils of Iran. The Rosetta model is based on ANN works in a hierarchical approach to predict water retention curves. For this purpose, 240 soil samples were selected from the south of Guilan province, Gilevan region, northern Iran. The data set was divided into two subsets for calibration and testing of the models. The general performance of PTFs was evaluated using coefficient of determination (R2), root mean square error (RMSE) and mean biased error between the observed and predicted values. Results showed that ANN with two hidden layers, Tan-sigmoid and linear functions for hidden and output layers respectively, performed better than the others in predicting soil moisture. In the other hand, ANN can model non-linear functions and showed to perform better than MLR. After ANN, MLR had better accuracy than Rosetta. The developed PTFs resulted in more accurate estimation at matric potentials of 100, 300, 500, 1000, 1500 kPa. Whereas, Rosetta model resulted in slightly better estimation than derived PTFs at matric potentials of 33 kPa. This research can provide the scientific basis for the study of soil hydraulic properties and be helpful for the estimation of soil water retention in other places with similar conditions, too. 

Citation: Esmaeelnejad, L.; Ramezanpour, H.; Seyedmohammadi, J.; Shabanpour, M. (2015). Selection of a suitable model for the prediction of soil water content in north of Iran. Spanish Journal of Agricultural Research, Volume 13, Issue 1, e12-002, 11 pages. http://dx.doi.org/10.5424/sjar/2015131-6111.

Evaluation of Pedotransfer Functions for Predicting the Soil Moisture Retention Curve

Abstract: The soil moisture retention curve (MRC) is time consuming and expensive to measure directly. Several attempts have been made to establish a relation between readily available soil properties, like particle-size distribution, organic matter content, and bulk density, and the soil moisture retention curve. Those relationships are referred to as pedotransfer functions (PTFs). The objective of this study was to evaluate some PTFs with respect to their accuracy in predicting the soil moisture retention curve. Five widely used and four more recently developed PTFs were selected for evaluation. Seven of the selected PTFs predict moisture retention function parameters, whereas the other two predict the moisture content at certain matric potentials. In order to quantify the prediction accuracy, the mean of the absolute value of mean differences (MAMD), the mean and the standard deviation of the root of mean squared differences (MRMSD and SDRMSD, respectively), and the mean of the Pearson correlation coefficient (Mr) were used. The evaluated PTFs were finally ranked based on these validation indices. The PTFs showed good to poor prediction accuracy with MAMD values ranging from 0.0312 to 0.0603 m3 m−3 and with MRMSDs between 0.0412 and 0.0774 m3 m−3 The SDRMSDs and Mrs ranged from 0.0212 to 0.0349 m3 m−3, and from 0.9468 to 0.9980, respectively. The validation indices computed by the PTF of Vereecken and coworkers gave the best results. Moreover, it predicts moisture retention function parameters, and therefore, this PTF is recommended most to predict the moisture retention curve from readily available soil pproperties.doi:10.2136/sssaj2001.653638x

Dear Alireza Amirian,

usually in pedotransfer functions the only variables used to predict soil water content at different suctions are: textural classification, organic carbon content, bulk density, particle density and depth of sampling. 

I think, beside these functions,information on energy status of water is also essential to quantitatively describe the net force of retention and flow of water in soil. For example, the water content at which water is 100% available in plants in sandy loam soil, may impose permanent wilting to plants in clay soil.This is due to higher energy with which water is retained in clay than in sandy loam soil.

Multiple Linear Regression (MLR), Artificial Neural Network (ANN) are the best models for predicting available water in calcareous soils?

This question is based on mis-information.   Compared to the vast amount of work in collecting soil density, texture, chemical composition - collecting available water data is easy and cheap.   It just requires two field trips at critical times.  If you know soil density then all you need is one auger sample after rain and another after harvest.  Simple quick and accurate !