1 Introduction

Soil has various roles and duties in agroecosystem e.g., the ability of its production, carbon and water storage, cycling of nutrients, and filtering of water (Brady and Weil 2016). Therefore, depending on the aims when using the soil, its quality can be great importance. During the last few decades, several definitions have been provided for soil quality, which most of them have focused on the soil’s ability to perform the activities (Doran and Parkin 1996). For example, Karlen et al. (1998) and de Paul Obade and Lal (2016) defined soil quality as the soil’s permanent ability for performing its role as a living biological system in the ecosystem under different applications, so that it can maintain its biological fertility, while having the ability to improve water and air quality, and ensure the health of humans, animals, and plants. From the agricultural view, soil quality is its ability to produce or soil fertility. Due to the vulnerability of the agricultural soils to environmental changes and human activities, attention to soil quality has a high strategic and economic importance in many countries (Wander et al. 2002; Sun et al. 2022).

Due to the vulnerability of agricultural soils from biological, environmental, and human health perspectives, providing a suitable method for measuring soil quality of cropland is essential to achieve sustainable management and predicting the risks of damage to the soil (Armenise et al. 2013; Qi et al. 2009). However, due to heterogeneity and the variety of soil changes and their consequences, selecting a suitable method for measuring soil quality still provides an immense challenge. Among various methods currently used to measure soil quality, SQI has the most significant number of applications due to its flexibility, quantitative nature, and suitability for various soils (Biswas et al. 2017), as well as its ability to combine different physical, chemical and biological characteristics of the soil. Further examples of successful use of this approach for measuring soil quality are available worldwide. Some of the most important examples included investigating the effects of deforestation (Shao et al. 2020), irrigation with wastewater (Jahnay and Rezapour 2020), soil erosion (Santos-Francés et al. 2019), land use (Choudhury and Mandal 2021), soil’s nutrients (Zhang et al. 2021), morphological properties of soil (Vasu et al. 2021), and soil desertification and degradation (Kaya et al. 2022) on SQI. However, the impact of long-term and continuous cultivation of cereals on SQI have not been investigated.

Cereals (e.g., wheat) are a group of plants that produce the primary food source for the majority of the people of the world as well as are being used for feeding animals, birds, and industrial applications (Grote et al. 2021). Nodaway, more than 70% of the total 1 billion Hectares of the soil under cultivation around the world are used for the cultivation of cereals and approximately half of the needed food of humans, especially in Asia, are directly satisfied using cereals (Grote et al. 2021). This indicates the critical role of cereals, mainly wheat, in food security around the world. Therefore, protecting and maintaining the quality of soils under cereal cultivation around the world is undoubtedly essential for protecting and preserving the ecosystem and achieving sustainable management of these soils.

Previous studies have investigated the relationship between each soil characteristic (physical, chemical and biological characteristics), and the cultivation of cereals. However, there is very little information regarding the soil quality index, which provides an integrated view of different soil qualities with yield components of wheat, especially for soils in the Mediterranean climate, which covers a vast area of Asia and Europe.

The main aims of the current study were 1. Evaluating SQI for soils under constant and long-term wheat cultivation using credible models; 2. Comparing the changes in SQI for different soils under continuous and long-term wheat cultivation, and adjacent uncultivated soils, and 3. Evaluating the relationship between SQI and yield components of wheat.

2 Materials and methods

2.1 Field description of the study area

This study was conducted in the Piranshahr region (36° 30′–36° 50′ North, 45° 05′–45° 25′ East), located at Western Azerbaijan Province of Iran (Fig. 1). The total area of this region is 37524 ha and includes Piranshahr, Jaldiyan, and Pasveh areas. The altitude of the area ranged from 1300 to 1650 m above sea level. The examined soils are developed mainly on alluvial deposits with the plateau and alluvial fan landforms (a slope 3–12%) along with slight to medium erosion. The average annual temperature and rainfall of the region was 10.5 °C and 500–600 mm, respectively, resulted in a sub-humid Mediterranean type climate (Rezapour 2014). Its minimum and maximum temperature is 8.5 °C and 31.8 °C recorded in January and August, respectively. The soils of the region have high cultivation potential and have been under comprehensive cultivation operation (especially wheat) for more than six decades. Annually, a significant amount of chemical and organic fertilizers (an average of 90 kg of urea fertilizer and 50 kg of triple phosphate fertilizers per hectare) are used to maintain the agricultural yield.

Fig. 1
figure 1

Location of soil profile sites in the study region

Full size image

2.2 Soil sampling

During field operations, based on a semi-detailed study, a total of 40 profiles were dug, evaluated, sampled, and analyzed. The soils were classified (Keys to Soil Taxonomy 2014) in three orders of Mollisols, Inceptisols, and Vertisol with a total area of 21,184, 8877.1, and 7462.9 ha, respectively. For each soil order, one profile was selected and analyzed as the control soil from adjacent rangelands (uncultivated soil). The features of the control soil include the type of materials, physiographic characteristics, slope, slope direction, and drainage were similar to the profile of the cultivated soils. Therefore, any difference between the cultivated soils and the adjacent uncultivated soils can only be due to cultivation activities and human management.

2.3 Laboratory analyses

Soil samples were air-dried and prepared to analyze the various physicochemical properties including particle size distribution, pH, electrical conductivity EC, organic carbon (OC), cation exchange capacity (CEC), calcium carbonate equivalent (CCE) and active carbonate calcium (ACC), total nitrogen, available P and K, soluble and exchangeable cations, available water (AW), porosity, and bulk density (BD) using standard methods (Sparks et al. 1996). Sodium absorption ratio (SAR) and exchangeable sodium percentage (ESP) were calculated using Eqs. 1 and 2, respectively (Sparks et al. 1996).

$${\text{SAR}} = \frac{{{\text{Na}}^{ + } }}{{\sqrt {\frac{1}{2}({\text{Ca}}^{2 + } + {\text{Mg}}^{2 + } { })} }}$$
(1)
$${\text{ESP}} = \frac{{{\text{Na}}^{ } }}{{\text{CEC }}} \times 100$$
(2)

2.4 Soil degradation indices

Soil Degradation Index (SDI) was calculated using the following equation:

$${\text{SDI}} = \frac{{P_{{\text{c}}} - P_{{\text{f}}} }}{{P_{{\text{f}}} }} \times 100$$

where, Pc is the value of each characteristic for cultivation soils and Pf is the same property for adjacent uncultivated soils. This equation is used for soil characteristics that are negatively affected by cultivation activities. A negative SDI value (< 0%) indicates soil degradation, with larger negative values indicating higher levels of degradation (Zhao et al. 2014).

2.5 Selected indicators and soil quality indices

A total of 18 soil characteristics which were sensitive to human management and cultivation activities (Biswas et al. 2017; Mamehpour et al. 2021) were selected as total data set (TDS). These characteristics included clay, silt, sand, pH, EC, OC, CEC, CCE, ACC, total N, available P and K, SAR, ESP, porosity, BD, and soil depth. To select the minimum data set (MDS) of the characteristics affecting soil quality, all soil parameters were investigated using the principal component analysis (PCA) method, and principal components (PC) were selected. By selecting the MDS, the number of soil characteristics decreases, and only effective and key variables remain. During PCA calculations, PCs with factor loadings of less than 1 are PCs that are not statistically significant and are therefore eliminated from the analysis. In PCs with eigenvalues higher than 1, the PC with the highest eigenvalue is selected and eigenvalues with 10% difference from the first principal component (PC1) are selected as the minimum number of components affecting soil quality. To prevent repetition, if two PCs have a correlation coefficient of higher than 0.6 with each other, the PC with the higher eigenvalues is selected. In contrast, for coefficient correlation of less than 0.6, both PCs are selected (Andrews and Carroll 2001). Since the measured characteristics have different units and scales, making their addition and multiplication impossible, all data are converted into dimensionless criteria using the standard scoring functions. Three criteria are specified in soil quality according to their functions, which include “the lower, the better", "the upper, the better", and the optimum range (Karlen et al. 1998; Andrews et al 2003). The more-is-better function is used for soil characteristics where higher values indicate higher soil quality (such as OC). In comparison, the less-is-better function is used for variables where higher values decrease the soil quality (such as BD), and optimum level function is used for soil characteristics where values higher or lower than the optimum value result in the depredation of soil quality (such as soil pH). The equations for these functions are as follows:

Function 1: standard scoring function for upper limit (more-is-better):

$$f\left( x \right) = \left\{ {0.9*\frac{x - L}{{U - L}} + 0.1} \right\},$$
$$L \le x \le U$$
$$x \le L$$
$$x \ge U$$

Function 2: standard scoring function for lower limit (less-is-better):

$$f(x) = \left\{ {1 - 0.9*\frac{x - L}{{u - L}}} \right\},$$
$$x \ge L$$
$$L \le x \le U$$
$$x \ge U$$

Function 1: standard scoring function for optimum limit (optimum level):

$$f \left( x \right) = \left\{ {\begin{array}{*{20}c} {0.9*\frac{x - L1}{{L2 - L1}} + 0.1} \\ {0.9*\frac{x - U1}{{U2 - U1}} - 1} \\ \end{array} } \right.$$
$$L1 \le x \le L2$$
$$L2 \le x \le U1$$
$$L1 \le x \le U2$$

In these equations, x is the measured value for each characteristic, f(x) is the scoring of the factors between 0.1 and 1, L is the minimum threshold, and U is the maximum threshold.

Further, the Integrated Quality Index (IQI) and Nemoro Quality Index (NQI) were calculated.Using the following equations (Doran and Parkin 1996; Qi et al. 2009):

$${\text{IQI}}w = \mathop \sum \limits_{i = 1}^{n} w_{i} N_{i}$$

in which Wi is the weight value, Ni is the score of each criterion, and n is the total number of criteria. The ratio of the communality value of each criteria to the sum of the communality values was considered as a weight of each soil criteria.

$${\text{NQI}} = \sqrt {\frac{{p_{{{\text{ave}}}}^{2} + p_{{{\text{min}}}}^{2} }}{2}} \times \frac{n - 1 }{n}$$

In which Pave and Pmin are the average and minimum scores of the selected criteria.

2.6 Validation of soil quality indices

To validate the SQI, sensitivity index (SI) and correlation analysis between SQI models as well as between SQI models with wheat yield components were determined, and the best possible SQI model for the studied area was introduced. The sensitivity index was calculated as the ratio of the maximum SQI of each model to its minimum (Masto et al., 2008). Due to being more sensitive to environmental conditions and anthropogenic impacts, higher ratios of SI are preferable.

Using the SPSS 19 software package (SPSS INC., Chicago, USA), all statistical analyses were carried out. Mean comparison was performed with the independent-test (in case of two means), i.e. each soil order under cultivation was compared with control soil order.

3 Results and discussions

3.1 General qualitative characteristics of soils

Table 1 shows some of the statistical data related to the main characteristics of the studied soils. The range of particle size distribution was from 26.8 to 52.78% for clay, 10–63.2% for silt, and 13.4–50.75% for sand. These data indicated a significant change in the soil texture ranging from sandy loam to clay. Changes in the soil BD was in the range of 1.33–1.57 with the mean value of 1.42 gcm−3 indicated a medium class (1.3 < BD < 1.6 gcm−3) for bulk density in the majority of the soil samples, which is a typical class for agricultural soils with different texture classes (Hazelton and Marphy 2016; Brady and Weil 2016). The averages of pH and CCE values of the soils were 7.32 (a range of 6.56–8.04) and 11.66% (a range of 0.9–26.88%), respectively, which demonstrates that the investigated soils are mostly alkaline and calcareous. The content of soil EC with the CV of 37.5% was in the range of 0.24–1.4 dS m−1, which indicates all the studied soils have low salinity. Furthermore, the mean values for SAR and ESP were 2.63 and 1.77%, respectively, which shows a lack of sodium-related problems in the soils. The mean value of 0.86, 0.145%, and 24.6 cmol kg−1 was observed for OC, total nitrogen and CEC of the soils, respectively, indicating a week class for OC (0.6 < OC < 1%) and total N (0.05 < N < 0.15%), and a medium- class for CEC (12 < CEC < 25 cmol kg−1) (Hazelton and Marphy 2016; Wang et al. 2022). The available K of the soil had the mean value of 165.31 mg kg−1, which was very high category (> 160 mg kg−1; Havlin et al. 2005). In comparison, available P of the soil with the mean value of 6.96 mg/kg was weak to medium (a range of 4–7 mg kg−1) (Havlin et al. 2005). Furthermore, the average values of 14.91 and 43.3% for AW and porosity were found. The majority of soil characteristics had a coefficient of variations higher than 35%, which indicates a wide range of changes in soil characteristics in the studied area (Wilding and Dress 1983) due to human management and cultivation activities (Table 1).

Table 1 Some statistical characteristics of the cultivated soils
Full size table

Soil degradation index (SDI) was calculated based on the characteristics that were sensitive to agricultural operations (e.g., clay, pH, EC, OC, CEC, total N, available P, available K, SAR, ESP, AW, BD, and porosity). The mean value of SDI the Vertisols (− 26.35) was lower (more negative) than Inceptisols (− 20.72) and Mollisols (− 17.96), indicating that long-term cultivation operations had more destructive effects on Vertisols than the other soil orders (Fig. 2). The SDI values of OC, EC, total, available P and K, SAR, ESP were lower than their mean value in each three soil orders, implying those soil indicators were exposed to the most degradation by agricultural operations. Previous studies also reported the adverse effects of natural resource changes (such as forest and pasture) to cropland on soil quality degradation using SDI, e.g., Zhao et al. (2014) and Raeisi and Beheshti (2022).

Fig. 2
figure 2

Soil degradation of soil properties for different soil types

Full size image

3.2 Soil Quality Index

3.2.1 Total data set

Considering the 18 soil characteristics, SQI was calculated using the IQI-TDS (IQI-T) and NQI-TDS (NQI-T) models. Among 18 soil characteristics, CCE and clay had the highest weights, and soil porosity had the lowest weight, with the weight of other features being placed in the middle (Table 2).

Table 2 The communality and weight values of each soil attribute for TDS and MDS
Full size table

For the cultivated soils, IQI-T and NQI-T were in the range of 0.41–0.58 and 0.28–0.39 with a mean value of 0.5 and 0.33, respectively. These data for the uncultivated soils were from 0.56 to 0.71 with a mean of 0.66 for the IQI-T model and 0.42–0.51 with mean value of 0.4 for the NQI-T model. Comparing these results indicates that cultivation operations have resulted in a significant decrease in soil quality indices in the ranges of 17–21.3% for IQI-T and 20–27% for NQI-T (Fig. 3). Similar findings have been reported by Ji et al. (2020), Mamehpour et al. (2021), Raiesi and Beheshti (2022), and Li et al. (2022), which reported that human activities such as cultivation operations and floor irrigation, using the combination of chemical and organic compounds, and land use change could negatively affect soil quality indices. In the considerable area of the cultivated soils (Fig. 4), the SQI for both models showed a high grade II (0.49 < IQI-T < 0.53 and 0.32 < NQI-T < 0.34) (Fig. 4) (Table 3). However, grade I or very high grade (IQI-T > 0.53 and NQI-T > 0.34) was observed in most adjacent uncultivated soils using both models. This indicates that cultivation operations have resulted in a drop in the SQI grade of significant area of soils by one grade compared to their adjacent uncultivated soils. The soils of grade I have very suitable and unlimited conditions for the growth of most agricultural products. In contrast, the soils of grade II offer suitable conditions with some limitations for plant growth (Qi et al. 2009).

Fig. 3
figure 3

The comparison of the mean values of IQI-TDS (a), NQI-TDS (b), IQI-MDS (c), and NQI-MDS (d) between cultivated and uncultivated soils in different soil types. Different letters indicate significant differences in SQI model for each soil type at P < 0.05 confidence interval

Full size image
Fig. 4
figure 4

Soil quality grades distribution for different SQI models in the study region

Full size image
Table 3 Categorization of soil quality grades using different methods
Full size table

3.2.2 Minimum data set (MDS)

To determine the MDS and key factors affecting soil quality index, PCA was carried out on the TDS. The PCA method is a common and accepted approach for determining MDS and widely used in previous studies (e.g., Andrews et al. 2004; Govaerts et al. 2006; Karlen et al. 2008; Mamehpour et al. 2021). The MDS data set has been introduced to reduce the number of indicators used in assessing SQI, resulted in reduce the process of laboratory analysis and cost (Karlen et al. 1998; Andrews et al. 2004). Table 4 shows the PCA data for all of the characteristics investigated in the current study. As can be seen, a total of 5 PCs with eigenvalue > 1, ranging from 1.33 to 5.73, explain 74.59 of total data variances. In PC1, which explains 31.74 of total data variance, the most significant weights belonged to OC, TN, AK, AP, and BD. In PC2, which described 16.56% of the total variances, the main factors included clay and sand fractions. In PC3, PC4, and PC5, which explained 10.55%, 8.11%, and 7.51% of the variances, respectively, SAR and ESP (for PC3), CCE and ACC (for PC4), and silt fraction (for PC5) were significant. Considering the correlation coefficients (Fig. 5) between soil factors presented in Table 5 and their weights, five factors, including OC (PC1), clay (PC2), SAR (PC3), CCE (PC4), and silt content (PC5) were selected as MDS. These results indicate a drop from 18 soil variables associated with the TDS to five in the MDS which can result in a decrease of more than 70% in the cost and time of soil analysis using MDS.

Table 4 Results of principal component analysis (PCA) of soil quality properties
Full size table
Fig. 5
figure 5

Correlation matrix between the different soil variables

Full size image
Table 5 Correlation coefficients between four SQI models
Full size table

Among the variables included in the MDS, OC has the most significant weight and contribution (43%) in the IQI determined by MDS (IQI-M), followed by clay 22.1%), SAR (14%), CCE (10.96%), and silt (10.1%). OC, clay, and SAR have a combined contribution of 80% in IQI- M, which have been widely reported as effective and sensitive factors for the development of SQI (Tian et al. 2020; Karaca et al. 2021; Zhang et al. 2021; Roy et al. 2022). These three factors have a crucial impact on the combination of soil physicochemical and biological properties, soil fertility- productivity, and yield components of crop (Brady and Weil 2016). For example, OC plays an essential role in the cycling and storage of soil nutritional elements and the development of soil structure as well as acts as the main source of food for soil heterotrophic microorganisms (Brady and Weil 2016). Likewise, SAR is subject to the degradation of soil physical properties (e.g., degradation of soil structure and restricted soil water–air circulation and permeability) and has an adverse effect on soil microbial population as well as root growth and, therefore, reduces the growth and performance of crop yield (Yeilagi et al. 2021; Mamehpour et al. 2021). The IQI is calculated using the following equation based on the MDS and weights for each factor:

$$\begin{aligned} {\text{IQI-M}} & = \left( {0.{43}} \right){\text{ OC}} + \left( {0.{222}} \right){\text{ clay}} + \left( {0.{141}} \right){\text{ SAR}} \\ & \quad + \left( {0.{11}} \right){\text{ CCE}} + \left( {0.{1}0{1}} \right){\text{ silt}} \\ \end{aligned}$$

Given the cultivated soils, the IQI-M values were in the range of 0.39–0.66 with a mean value of 0.55, while NQI-M values were in the range of 0.25–0.41 with the mean value of 0.33. These values for the adjacent uncultivated soils were 0.6–0.75 with a mean of 0.68 for IQI-M and 0.38–0.49 with a mean of 0.42 for NQI-M. For both cultivated and uncultivated soils, the most significant IQI-M and NQI-M values were observed in the Inceptisols, followed by Mollisols and Vertisol. This is similar to the behavior observed for IQI-T and NQI-T. These results indicate the importance of the effect of soil type on the SQI. Furthermore, Inceptisols and Mollisols often had the quality grade of II to I, while the majority of Vertisols showed grade III to II, which is similar to the data found by Stevenson et al. (2015) and Mamehpour et al. (2021). Such data are almost comparable to those recorded for IQI-T and NQI-T. As a result, most the cultivated soils of the region had a quality grade of II–I regarding all SQI models and provides suitable conditions for the growth and development of agricultural crops mainly cereals.

Compared to the adjacent uncultivated soils, a 17–22% decrease in IQI-M and a 21–26% decrease in NQI-M was observed in the agricultural soils, resulting in a decrease in the SQI grade from grade I to grade II in most of the study soils. The most significant decrease in SQI as a result of cultivation operations (24% decrease) was observed in Vertisols, followed by Inceptisols (21% decrease) and Mollisols (19.5% decrease). Vertisols are the soils with unique characteristics (e.g., high expansion and shrinkage potential), make up 2.4% of all the soils worldwide, and found in more than 80 countries (Boul et al. 2011). These soils, when wet, sticky and plastic and, in contrast, are very hard when dried. Such a trend makes it extremely problematic to perform any tillage, cultivation, or engineering operations. (Brady and Weil 2016). Therefore, the significant decrease in the quality of these soils as a result of long-term cultivation operations is not unexpected because Vertisols are susceptible to cultivation operations, and performing advanced and scientific soil management operations are vital to prevent their degradation. Kraemar et al. (2021) from Argentina, Wang et al. (2022) from China, and Garg et al. (2022) reported that Vertisols have a high sensitivity and problematic issues to cultivation operations for soybeans, wheat, corn, and rice.

3.3 The relationship between SQI and wheat performance indicators

Linear regression analysis (Fig. 6) showed that all SQI models have a positive and significant correlation with the yield components of wheat. Therefore, any of the models can monitor the wheat performance components despite having different coefficients of variations (R2). Several studies (e.g., Tian et al. 2020; Mendes et al. 2021; Roy et al. 2022) have also found a significant correlation between the yield components of various agricultural products and SQI with different coefficients of variations. Regression equations showed that IQI-T, NQI-T, IQI-M, and NQI-M models could explain 59%, 39%, 53%, and 35% of variations in wheat biological yield and 57%, 37%, 51%, and 33% of variations in wheat grain yield, respectively (Fig. 6), showing that the IQI scenario present a more accurate evaluation of the yield components of wheat than the NQI scenario. The difference between IQI and NQI models can be because of a combination of scoring and weights are used for soil properties in calculating the IQI model, while the NQI model is calculated based on the mean and values of minimum scores for soil properties (Vasu et al. 2016; Rezapour et al. 2021). Although the most significant correlation coefficient between SQI and the yield components of wheat was observed in the IQI-T model, we suggest the IQI-M model has better performance for assessing SQI compared to the other modes due to: (1) there was a positive and significant correlation between IQI-T and IQI-M models (Table 5), while the correlations between the other models were less significant and (2) the sensitivity index (SI) for the IQI-M model was the highest (1.73), followed by IQI-T (1.64), NQI-M (1.42), and NQI-T (1.39), indicating the higher sensitivity of the IQI model to soil management activities when compared to other models. The sensitivity index is a valuable and helpful tool that can be used to distinguish the differences between different models of soil quality assessment, as shown in previous studies (Yeilagi et al. 2021; Mamehpour et al. 2021). The studies conducted by Santos-Francés et al (2019) and Zhou et al (2020) on cropland in Spain and China, respectively, highlighted the performance of the IQI-MDS method for predicting SQI using the SI analysis. Furthermore, evaluating soil quality using the IQI-M model can result in a significant decrease in the number of soil factors measured in the lab for SQI determination, resulting in a considerable reduction in analysis time and cost. Therefore, the correlation coefficients between different SQI, SI, time and money savings, and regression coefficient between SQI and the yield components of wheat confirmed that IQI-M model was an effective tool and good criterion for evaluating changes in land use as well as the effect of different soil management systems on soil quality.

Fig. 6
figure 6

The biological yield (BY) and grain yield (GY) of wheat versus different SQI models

Full size image

4 Conclusion

In the current study, the SQI of cropland with a cultivation history of more than six decades was investigated using four SQI models (IQI-T, IQI-M, NQI-T, and NQI-M). Among 18 soil variables, five variables including OC, clay, SAR, CCE, and silt were selected as MDS based on the ANOVA and PCA results. Compared to the uncultivated soils, a 17% to 27% decrease in soil quality was observed in the cultivated soils. The most significant reduction in SQI was observed in Vertisols, indicating its high sensitivity to agricultural and management operations compared to other soil types. All SQI models used in the current study showed a significant correlation (with a correlation coefficient in the range of 0.33–0.59) with the yield components of wheat. Although the IQI-TDS was established as the most accurate model for soil quality assessment in the study region, IQI-M could be a good alternative as it adequately performed the TDS method; in addition to its reliability and economic feasibility. This soil quality model can help local smallholder farmers to be aware of the potential for low soil quality in future following current conventional farming systems and to consider suitable agricultural management operations for maintaining soil quality in highly intensive agroecosystems. Nevertheless, a potential limitation of this soil quality model is that it might be specific to the soil and location. Another potential limitation of this study is that the soil biological indicators were not considered and can be an interesting subject of study in the future. Hence, its application with additional studies in the other ecosystems and soil orders can be further recommended to provide a more accurate test. The SQI assessment would have important implications for resource management and can be of particular importance for soil owners and policy makers to assess the sustainability of management practices after long-term continuous cultivation.