Research Article
Artificial Neural Networks for Estimating Soil Water Retention Curve Using Fitted and Measured Data
Table 1
Descriptive statistics of soil properties assumed as input and output parameters in the developed ANNs (all 228 soil samples).
| Soil property | Input parameters | Min | Max | Mean | SD |
| Bd (g cm−3) | 0.46 | 1.95 | 1.41 | 0.26 | Pd (g cm−3) | 1.98 | 2.93 | 2.63 | 0.09 | (cm3 cm−3) | 0.26 | 0.92 | 0.47 | 0.10 | Sand (%) | 0.00 | 98.90 | 26.30 | 25.58 | Silt (%) | 1.10 | 95.60 | 40.84 | 18.79 | Clay (%) | 0.00 | 87.60 | 32.82 | 21.49 | (cm) | 0.00 | 15.62 | 5.20 | 2.50 |
| Matric potentials (cm) | Output parameters | Min | Max | Mean | SD |
| Measured | 0.008 | 0.838 | 0.314 | 0.140 | 0 | 0.139 | 0.837 | 0.449 | 0.117 | −10 | 0.139 | 0.826 | 0.427 | 0.108 | −20 | 0.139 | 0.789 | 0.408 | 0.106 | −30 | 0.108 | 0.749 | 0.391 | 0.108 | −50 | 0.066 | 0.707 | 0.364 | 0.113 | −100 | 0.043 | 0.631 | 0.329 | 0.118 | −200 | 0.037 | 0.597 | 0.299 | 0.118 | −500 | 0.010 | 0.576 | 0.265 | 0.112 | −1000 | 0.003 | 0.568 | 0.241 | 0.108 |
|
|