Authors
1
PhD student, Soil Science, Agricultural Faculty, Shiraz University
2
Associate Professor., Soil Science Department, Agricultural Faculty, Shiraz University
3
Associate Professor., Soil and Water Research Institute of Iran
4
Associate Professor., Soil Science Department, Agricultural Faculty, Guilan University
Abstract
Soil quality is not directly measurable, and it is determined from soil quality indicators. The slope of the water retention curve at its inflection point is one of these indices. The S associated with different factors, for examples, organic carbon, bulk density, texture and soil structure. The purpose of this study was to compare three models for determining soil physical quality index and assessing the physical quality of paddy soils using this index. So, 35 distributed soil samples were collected from paddy soils in eastern and western parts of Guilan province, and texture, bulk density, organic carbon, saturation water content, water content at field capacity and wilting point were measured. The Brooks and Corey, van Genuchten, and Groenevelt and Grant models were used to determine the S index. The RETC program was used to estimate the parameters of the van Genuchten model, which were later used to estimate the slope of the water retention curve at its inflection point. By applying the MATLAB software, the parameters of Brooks and Corey (1964) and Groenevelt and Grant (2004) models were determined and then S index was obtained. The indicators of geometric mean error ratio (GMER), geometric standard deviation of error ratio (GSDER), mean error (ME) and normalized root mean square error (NRMSE) were used to evaluate the goodness of fits for the models. The Groenevelt and Grant model could estimate the SMRC with the minimum values of ME and NRMSE at all tensions. Furthermore, the GMER and GSDER values of this model at all tensions were closer to 1 as compared to the other models. The Groenevelt and Grant model with ME=-0.015 and NRMSE=0.4 was the best model compared with the other models. Also, the geometric mean error ratio and geometric standard deviation of error ratio (GMER=0.85, GSDER=1.33) were lower in Groenevelt and Grant model. Besides, results showed that physical quality of the studied soils was good. The significant correlations between S and clay, water content at wilting point, and bulk density were negative (P<0.01); whereas its correlation with available water and the other soil attributes were positive (P<0.05). Although some investigators believe that S cannot be a good index of soil physical quality, the determined S index of the studied paddy soils showed significant correlation with attributes affecting the physical quality of soils. Therefore, by applying the S quality index, it could be possible to compare the physical conditions of different soils and the effect of various management practices on them instead of measurements of several attributes affecting the soil physical quality. The physical quality index of the studied paddy soils was greater than 0.035, which indicated that the quality of the soils was good for paddy production.
Keywords