m. baghernejad

Understanding the particle size distribution (PSD) is of great importance for plant growth and soil management. In recent years, the science of soil has witnessed a significant increase in digital soil mapping (DSM) activities. In this regard, machine learning models (ML) have emerged as an alternative and tool for DSM, which are mainly used for data mining and pattern recognition purposes, and are now widely used for regression and classification tasks in all fields of science. Hence, this study was undertaken to spatially model sand, silt, and clay particles utilizing machine learning models such as Random Forest (RF), Support Vector Regression (SVR), and the Co-Kriging geostatistical model. Additionally, auxiliary variables with high spatial resolution were incorporated into the analysis. This investigation was conducted in a section of the Marvdasht plain, located in Fars province.

The present study was conducted in a part of Marvdasht plain located between 35.82´41°52' to 1.07´57°52' east longitude and 35.02´48°29' to 14.72´2°30' north latitude, and 40 km north of Shiraz with an area of about 50,000 hectares. After determining the study area boundaries, the positions of 200 sampling points were determined using the R software and the conditioned Latin hypercube sampling method. In other words, for soil feature modeling, 200 samples were taken from two depths of zero to 30 and 30 to 60 centimeters in the study area. Then, the samples were transferred to the laboratory, dried, and passed through a 2 mm sieve. Finally, the soil texture components were measured by the hydrometer method. The environmental variables used in this study are a wide range of representatives of soil-forming factors that were prepared as much as possible from sources with minimum cost and high accessibility. In total, 75 environmental variables were prepared, and the raster format related to all environmental variables, including 39 elevation and altitude variables and 36 remote sensing measurement variables, was extracted. Finally, the factor-tuning inflation variance and Boruta algorithm were used to select the optimal variables.

The minimum amount of clay was measured at 10.21% and 10.45%, respectively, and the maximum amount was 32.65% and 36.35% at the surface and subsurface depths. The average amount of clay in all samples was 37.91% and 35.61%. The average amount of sand was measured at 25.65% and 26.02% at the surface and subsurface depths, respectively. The maximum amount of sand was observed in the northern and higher parts of the study area and was equal to 54.68% and the minimum amount was predicted in the low-lying areas of the study area. Low-lying areas and sedimentary plains in the central part of the study area contained high amounts of silt. Four depth variables valley depths (VD), texture (TE), topographic wetness index (TWI), and clay index (CI) related to geomorphometric parameters and the normalized difference vegetation index (NDVI) variable related to remote sensing indices were selected as optimal variables. The RF model with R2 of 54.0% and 36.0% for predicting sand, 48.0% and 64.0% for predicting silt, and 52.0% and 49.0% for predicting clay at both surface and subsurface depths performed better than the SVR and Co-Kriging models. The most effective variable in predicting the spatial distribution of soil particles was VD with relative importance of 60% and 65% for predicting sand at the surface and subsurface depths, 70% for predicting silt at the surface depth, and 70% and 65% for predicting clay at both surface and subsurface depths, respectively. Only TE and TWI variables were more important than VD for predicting silt at subsurface depth. These results show that topographic variables are effective in the spatial variation of soil particles. Unlike clay, the highest amount of sand in both depths was observed in the northern part and the highest part of the study area, and the lowest amount was predicted in the low-lying areas of the study area.

In general, with the aim of this research, maps of the spatial distribution of soil texture components were prepared at both surface and subsurface depths using machine learning and geostatistical approaches along with environmental covariates in a part of Marvdasht plain. Among the selected environmental covariates, topographic attributes, especially the valley depth (VD), had the highest effect in justifying the spatial prediction of soil texture components. Also, the results of comparing the performance of machine learning models supported the higher efficiency of the RF model than other models. Therefore, the approach used in this study to prepare a map of soil texture components can be useful as a guide for mapping useful soil features in areas with similar climatic and topographic conditions.

Over the last three decades, there has been a general tendency for changing research methods on soil resource management from conventional and mainly qualitative methods to quantitative ones based on spatial correlation models, which are called Digital Soil Mapping (DSM). The present study was carried out in Shabankareh Plain (15,000 ha) with different physiographic units, in Bushehr Province, southern Iran. The target sites (172 points) were selected for soil sampling at depths of 0-30 cm. Soil texture classes DSM was produced by two methods. The first method was Conventional DSM, in which data of soil particles was obtained from laboratory analysis for each sampling point along with their geographical location. Also, the study area boundary was added to ArcGIS software in UTM format and was analyzed by operating Kriging or IDW estimators. The map produced by this method was a low quality digital map containing extra and scattered texture classes with unrealistically sharp boundaries. The second method used CoKriging of L8 multispectral imagery data (OLI bands) and soil samples analysis was operated. Results showed that using B1 band (0.433-0.453 µm) of Landsat 8 satellite imageries of the study area in April 2020 produced high quality digital maps. In this method, soil textures were the same as the ones in the study area. Salt accumulation and water content of surface soil were possible reasons indicating why satellite imageries in the other periods of year were not suitable for DSM. The highest and the lowest ranges of influence among soil texture parameters were 684 meters and 388 meter for clay and sand particles, respectively.

Soil classification in a standard system is usually defined based on information obtained from properties and their variations in different map units. The aim of this study was to compare soil genesis and morphological characteristics in different landforms with WRB and Soil Taxonomy (ST) Systems. From nine studied profiles, six profiles were selected as representative profiles and dug in Colluvial fans, Piedmont plain, and Alluvial plain physiographic units, respectively. Then, the soils were classified according to the pattern of the two systems. Also, variation analysis of variance (ANOVA) and comparing means were used to quantify interested soil properties. The results of soil physio-chemical properties at different landform positions were significant based on analysis of variance of the effect of physiographic units and soil depth at the level of 1 %. Soil classification results based on WRB indicated that WRB were recognized four reference soil groups (RSG) included Regosols, Cambisols, Calcisols, and Gleysols at the first level of WRB classification in comparison of ST with recognizing two order Entisols and Inceptisols could separate more soils. The soils were located on the alluvial plain with a high groundwater level in the WRB due to the creation of restrictive conditions for root development in contrast to the ST called “Aquepts” in the suborder level but in a WRB is classified as the “Gleysols” RSG. On the other hand, ST, unlike WRB, used the Shallow criteria at the family level to describe the shallowness of soils and the limitations of root development. Generally, the efficiency of each system varies despite the differences in their structure and depending on the purpose of using them.

The aim of this study was to assess the kinetics of nonexcheangable potassium release in surface and subsurface soil horizons, using organic and inorganic extractions, in Kohgilouye-va-Boyerahmad Province. Kinetics of K+ release was studied by successive extractions of K from 64 selective surface and subsurface soil samples, using 0.01 M CaCl2 and 0.01 M oxalic acid, for 1948 h, with two replicates. Nonexchangeable K+ release was fitted by Elovich, Pseudo-first order, Power function and Parabolic equations. Result showed that the average nonexchangeable K+ released (extracted by 1M HNO3) was 356 mg/kg, while those extracted by CaCl2 and oxalic acid after 1948 h were only 58% and 52% of the total amount of nonexchangeable K+ of the soils, respectively. In all soil samples, nonexchangeable K+ released by oxalic acid was less than that released by CaCl2, due to the high buffering capacity resulting from high carbonates in the soils. Potassium release rate in Elovich and Parabolic equations were significantly correlated with non-exchangeable potassium and some physical and chemical characteristics. Based on high Coefficients of determination (r2) and low Standard errors (SE), Elovich, Power function, First order and Parabolic equations were selected as the best equations for prediction of K+ release from the soils.

Sugarcane (Saccharum officinarum L.)‌ and rotational crops have been cultivated without potassium fertilizers for many years in southwestern Iran. Although potassium was removed from this soil, no response has been reported to K fertilizers by crops. This study was conducted to evaluate the effects of wetting and drying, and cultivation systems on potassium fixation in some Khouzestan soils. The results showed that expandable clay mineral was observed in cultivated soils but not in the non-cultivated soils. This may be related to irrigation, and cultivation practices in this soils. By adding K and sequential wetting and drying the amount of K-available increased in sugarcane, rotational cropping and uncultivated soils from 132.6, 226.2 and, 171.6 mgkg-1 to 266, 447 and 628, respectively. These results showed that more K can be fixed after cultivating soils, especially by sugarcane, but available K increased by adding K. All surface soils had higher K fixation capacity than subsurface ones. Also, by application K to these soils, the amount of K fixation decreased with increasing wetting and drying times. The K fixation increased by the increase of cation exchangeable capacity. Significant difference was observed between cultivation system and depth of sampling in 1% levels. High K fixation can be attributed to illite minerals depleted from K. Mica-like minerals formed after adding K, and wetting and drying cycles. Amount of K fixation by clay particle samples was more than the same soil samples but in both the same trend was observed for K fixation. Also, drying and wetting decreased K fixation in the last periods. It may be due to trapping K in the interlayer positions.

Chenges in the soil characteristics is rather continuously. A method that takes this continuity into account would present a realistic pattern of soil distribution either in taxonomic or geographical space. The fuzzy set theory provides such an approach. In this study, the robustness of fuzzy clustering in soil pattern recognition was evaluated in a subcatchment of western Iran. The clustering carried out on the basis of minimization of an objective function in assigning membership values to each pedon in each fuzzy class. Fuzziness exponent values from 1.15 to 1.5 were used. The following validation of the resulted clusters (classes), optimal number of classes in whole, morphological and particle-size subsets were determined 8, 4, and 5 respectively. Plots of membership values across the landscape indicated class overlap and considerable contiguity. Considering low differentiation of these young soils and the high similarity among their properties, the method indicated a high capacity in recognizing different soil types over the study area. Furthermore, there was relationships between the soil fuzzy classes and landform. Thus, the method is capable in continuous classification, which could be so important in construction of continuous soil maps at low aggregation levels, e. g., pedon.