Maximum bulk density (BD_max) and critical water content (θ_c) (i.e., Proctor parameters) are valuable parameters to evaluate soil compactness and optimum moisture of workability for tillage. There are two novelties in the present study: First, no study has been conducted so far to estimate the Proctor parameters from CaCO₃, saturated and field water contents in agricultural lands using state-of-the-art methods. Second, no study has been done to compare the estimation accuracy of linear (LR) and nonlinear (NLR) regression, support vector machine (SVM), and artificial neural networks (ANNs) methods in estimating Proctor parameters in agricultural soils. In total, 105 soil samples were taken from agricultural lands of Ardabil plain, northwest of Iran. Pedotransfer functions (PTFs) were constructed using SVM, ANNs, LR and NLR methods to estimate BD_max and θ_c from readily available soil properties including organic carbon (OC), CaCO₃, particle size distribution (PSD), bulk (BD), and particle (D_p) density, total porosity (n), penetration resistance (PR), and saturated (θ_s) and field (θ_f) water contents. The results of the LR, NLR, ANNs, and SVM estimations showed that θ_s, θ_f, OC, and D_p were the most suitable estimators in estimating BD_max and θ_c. The values of root mean square error (RMSE) criterion in the best LR, NLR, ANNs, and SVM PTFs were obtained 2.3, 3.29, 2.19 and 3.09 g g⁻¹ for θ_c and 0.05, 0.07, 0.05 and 0.07 g cm⁻³ for BD_max in the testing data set, respectively. Overall, Proctor parameters of agricultural soils could be accurately estimated by the ANNs compared with the LR, NLR and SVM.

最大堆积密度（BD_最大值）和临界水含量（θ _Ç）（即，普罗克特参数）是有价值的参数来评估土壤紧凑性和可加工性为耕作最佳含水量。本研究有两个新颖之处：首先，到目前为止，尚未进行任何研究来从CaCO ₃估算Proctor参数。使用最先进的方法处理农田中的饱和，饱和和田间含水量。其次，尚未进行任何研究来比较线性（LR）和非线性（NLR）回归，支持向量机（SVM）和人工神经网络（ANN）方法在估计农业土壤中Proctor参数时的估计准确性。总共从伊朗西北部的Ardabil平原的农业土地上采集了105个土壤样品。使用SVM，人工神经网络，LR和NLR方法来估计BD构建Pedotransfer函数（的PTF）_最大和θ _Ç由容易获得的土壤的性质，包括有机碳（OC），碳酸钙₃，粒度分布（PSD），本体（BD），和粒子（D _p）密度，总孔隙度（N），贯入阻力（PR），和饱和的（θ_小号）和场（θ _˚F）的水含量。在LR的结果，NLR，人工神经网络，支持向量机和估计表明，θ_小号，θ _˚F，OC，以及d _p是在估计BD最合适的估计_最大值和θ _Ç。根均方误差（RMSE）准则在最佳LR值，NLR，人工神经网络，和SVM的PTF获得2.3，3.29，2.19和3.09克克^-1为θ _Ç和0.05，0.07，0.05和0.07克厘米^-3 BD_最大在测试数据集中。总体而言，与LR，NLR和SVM相比，ANN可以准确估算农业土壤的Proctor参数。