Assessing soil fertility is essential to help identify strategies with less environmental impact in order to achieve more sustainable agricultural systems. The main objective of this research was to assess two soil fertility evaluation approaches in paddy fields for rice cultivation, in order to develop a user-friendly and credible soil fertility index (SFI). The Square-Root method was used as a parametric approach, while the Joint Fuzzy Membership functions as a fuzzy method with adapted criteria definition tables, were used to compute SFI. Results indicated that both of the methods determined the major soil limiting factors for rice cultivation clearly, and soil fertility maps established using GIS (Geographic Information System) could be helpful for decision makers. The coefficients of determination (R2) for the linear regression between the two SFI values and rice yields were relatively high (0.63 and 0.61, respectively). Additionally, the two SFI were significantly correlated to each other (r = 0.68, p < 0.05). The study results demonstrated that both of the methods provide reliable and valuable information. Compared to the fuzzy method, the procedure of the parametric method is easier but may be expensive and time-consuming. However, the fuzzy method, with carefully chosen indicators, can adequately evaluate soil fertility and provide useful information for decision making.

The direct measurement of the soil water retention curve (SWRC) in a laboratory is time-consuming, difficult, and costly. Thus, many attempts have been made to predict the water retention curve indirectly from the physical and chemical properties of soil. The particle size distribution curve is one of the indirect methods used to predict the water retention curve. The Arya and Paris (AP) model predicts the soil water retention curve from soil particle size distribution (PSD) data. The AP model estimates pore radius from the radius of spherical particles by using a scaling parameter (\alpha). The objective of this study was to investigate the effect of predicting the scaling parameter with different methods to improve estimation of the SWRC. The evaluation of methods was done on 35 soil samples with different textures from the eastern region of Guilan Province in Iran. The results showed that estimated curves with different \alpha values gave different results that depended highly on the scaling parameter. Therefore, \alpha determination has a key role in estimating the soil water retention curve. The results also showed that a linear \alpha and a constant \alpha with a maximum coefficient of determination and minimum error are the best scaling parameters to estimate the SWRC.

Soil quality assessment, which considers numerous physical, chemical, and biological indicators, has long been a challenge for monitoring soil functions and ensuring sustainable resource use in agriculture. In this study, different indicator selection and weighting methods were compared to derive a reliable Soil Quality Index (SQI) in semi-arid agroecosystems. A total of 117 topsoil samples were taken from the Ap horizon within a 14,200 ha area of the Honam sub-catchment, southwestern Iran. Twenty-one soil indicators were measured and analyzed to assess the overall SQI. Soil indicator selection was performed using Principal Component Analysis (PCA), considering standard and norm value strategies, as well as component rotation. Four weighting approaches, including PCA, Coefficient of Variation (CV), correlation score (r), and Expert Opinion (EO), were applied to the Minimum Dataset (MDS) and Total Dataset (TDS) to compute the Integrated Quality Index (IQI), Nemoro (NQI), simple additive (IQIa), and Fuzzy Fertility Index (FFI). The performance of the SQI models was evaluated using the Sensitivity Index (SI) and their relationships with crop yield. The results showed that the combination of the norm value approach without component rotation was more effective in selecting the influential indicators for SQI determination. The Structural Stability Index (SSI), which integrates soil organic carbon and textural soil properties, was the key indicator with the highest contribution, ranging between 6.3% and 37.5% in most of the models. Among the evaluated approaches, the IQI-CV-MDS showed the highest sensitivity (SI = 6.8) and the strongest correlation (r = 0.53) with rainfed barley yield. The majority of the samples exhibited moderate SQI values, indicating a general risk of soil quality decline in the study area. The findings of this study highlight that appropriate indicator selection and weighting strategies are essential for improving the reliability of SQI assessments in semi-arid environments with diverse mountainous topography.