نشریه مهندسی زراعی
سال چهل و ششم شماره 1 (بهار 1402)

Plants are usually exposed to a wide variety of abiotic stresses which can seriously inhibit plant growth and development. To address this, numerous strategies have been proposed and used by researchers, including increased irrigation rounds, cultivation of salinity- and drought-resistant genetically modified crops (GMO crops), and application of plant-growth-promoting rhizobacteria (PGPRs). PGPRs can act as an efficacious, long-lasting, and crucial option to ameliorate the negative impacts of abiotic stresses in crops. Plant growth promoting rhizobacteria can serve as key in sustainable agriculture by improving soil fertility, plant tolerance, crop productivity, and maintaining a balanced nutrient cycling. Consequently, search for new strains of PGPR for biofertilizer and development of microbial diversity map for any region is helpful. Hence, the present study investigates the effect of native salinity-resistant PGPRs on the physiological and biochemical characteristics and productivity of alfalfa plants in soil under salinity stress.

The present study was based on a completely randomized factorial design. The experiments were conducted on alfalfa plants (Bami variety) in four repetitions and under two levels of salinity (control and 200 mM sodium chloride and calcium chloride) with five strains of PGPRs from Sinorhizobium meliloti sp., two positive controls (i.e., 20 mg/kg of phosphorus and 70 mg/kg of nitrogen fertilizers), two negative controls (i.e., no fertilizer and no bacteria), and two treatments (including positive control and negative control). Growth parameters (dry weight of aerial parts, roots, and nodules), osmolytes (reducing sugars, soluble proteins, and proline), uptake of K+ and K+/Na+ ratio, and concentration of malondialdehyde (MDA) in alfalfa plants in non-saline and saline soils were measured at the end of 60-day experiments.

The analysis of variance (ANOVA) results revealed a significant effect of salinity on the dry weight of aerial parts and roots, the weight and number of nodes in each pot, the K+/Na+ ratio in roots and aerial parts, and the concentration of reducing sugars, proline, MDA, and soluble proteins in the aerial parts. The effect of PGPRs was also found to be significant on all the above traits. Under no salinity stress and compared to negative control plants, the dry weight of aerial parts in plants inoculated with superior PGPRs (SM89, SM16, and SM65) was raised by 2.3, 1.9, and 1.8 folds, while this increase in plants inoculated with mild (SM73) and weak (SM21) PGPRs was 1.4 and 1.2 folds, respectively. Under salinity stress (200 mM NaCl and CaCl2) and compared to negative control plants, the increase in dry weight of aerial parts of plants inoculated with superior PGPRs (SM89, SM16, and SM65) was increased by 4.2, 4, and 2.1 folds, while this increase in plants inoculated with mild (SM73) and weak (SM21) PGPRs was raised by 1.7 and 1.2 folds, respectively. Despite a drop in the growth of aerial parts in plants under salinity, salinity-resistant PGPRs were able to significantly enhance the growth of aerial parts of plants in saline conditions compared to the controls (receiving no fertilizer and PGPRs). Salinity stress reduced other growth parameters, the rate of K+ uptake, and the K+/Na+ ratio, while it contrarily increased the concentration of reducing sugars, soluble proteins, proline, Na+, and MDA in plants. Inoculation of alfalfa plants with two superior PGPRs (SM89 and SM16) was found to significantly improve growth parameters, uptake of K+, osmolytes, and K+/Na+ ratio in alfalfa plants under non-saline conditions and salt stress, compared to control plants (not inoculated with PGPRs and receiving no fertilizer). Ultimately, each inoculation of plants with all three superior PGPRs reduced the concentration of MDA and Na+ in alfalfa plants.

Experiments on the biochemical and physiological plant–PGPR interactions revealed that plant responses to stresses are largely controlled by microbial communication. PGPRs can trigger systemic resistance in plants through their metabolites, which function as extracellular signals, thereby enabling plants to survive under abiotic stresses. In the present study, microbial inoculation was found to significantly improve the physiological functioning of the plants. The results revealed that adding native salinity-resistant PGPRs to the soil can diminish the negative effects of salinity stress on alfalfa plants. Likewise, inoculation and enrichment of the plant's rhizosphere with beneficial and resistant microbiomes were efficient for sustaining the growth of plants under abiotic stresses such as salinity.

Salinity stress reduces the yield of agricultural products. Water or soil salinity is caused by the increase in the concentration of soluble salts and minerals in water and soil, which leads to the accumulation of salt in the root area, to the extent that it prevents water absorption and optimal plant growth. In general, tolerance to salinity is important during all stages of plant growth. Seed germination is the first stage of plant growth. Salinity stress reduces the percentage and rate of seed germination and also seedling growth. Crop yield is quantitatively and qualitatively dependent on the percentage, rate and uniformity of seed germination and also seedling growth. In recent years, a lot of attention has been paid to silicon due to its effect in reducing plant damage against some environmental stresses (such as drought, heat, heavy metals, salinity etc.). The studies show that silicon protects the plant against environmental stresses by stimulating growth and increasing the antioxidant enzymes activity. It has also been reported that the silicon is effective in increasing the chlorophyll content, stomatal conductance, photosynthesis rate and the resistance of plants under stressful conditions. Silicon increases the plant tolerance to the salinity by improving photosynthetic activity, increasing the relative selection of K+/Na+, increasing the soluble substances in the xylem, reducing sodium absorption, and mechanical protection against the toxicity of elements. Therefore, a research was carried out with the aim of investigating the effect of the silicon in increasing tolerance to salinity stress in camellia seedlings.

A laboratory experiment was carried out in 2021 at Campus of Agriculture and Natural Resources, Razi University, Kermanshah, Iran. The experiment laid out as a factorial based on a completely randomized design with three replications. The factors were camelina genotypes (Sohail cultivar and Line-84), salinity (four levels 0, -3, -6 and -9 bar) and silicon (five levels of 0, 2, 4, 6 and 8 mM). Salinity stress levels were prepared by different amounts of sodium chloride (NaCl) salt. Silicon factor levels were also prepared by different concentrations of sodium silicate (Na2SiO3). The experiment consisted of 120 petri dishes. Data analysis was done with MSTATC and SAS statistical softwares. Means were compared using Duncan's multiple range test (P≤0.05). Excel software was used to draw figures.

The results showed that with the increase in salinity intensity, the growth characteristics and the amount of soluble proteins of camellia seedlings decreased, but the activity of catalase, peroxidase and superoxide dismutase enzymes and the amount of malondialdehyde increased. The lowest activity of catalase was observed under non-salinity conditions (control). But the highest activity of catalase enzyme (104.4 µM/min. mg protein) belonged to the treatment of -9 bar salinity. The use of silicon increased the seedling growth, the amount of soluble proteins, the activity of antioxidant enzymes, and the amount of malondialdehyde in camelina seedlings. The highest germination rate (23.97 seeds/day) was obtained in the treatment of 8 mM silicon. With the increase in silicon concentration, the amount of soluble proteins increased, so that in the 2, 4, 6 and 8 mM treatments, compared to the control treatment, the amount of soluble protein increased by 4, 8, 10.75 and 10.9%, respectively. By increasing the concentration of silicon, the activity rate of catalase enzyme increased. The highest activity rate of peroxidase enzyme (35.38 µM/min. mg protein) was observed in 8 mM silicon, which was significantly different from other treatments. The lowest activity of peroxidase was related to the control treatment. Line-84 had 8.65% higher activity rate of superoxide dismutase than the Sohail cultivar. With increasing salinity stress and silicon concentration, the activity rate of superoxide dismutase increased. On average, in the treatments of 2, 4, 6 and 8 mM silicon, the activity rate of superoxide dismutase was increased 11, 27, 44 and 57%, respectively, compared to the control (without silicon). The highest amount of malondialdehyde (44.42 µM/g fresh weight) was observed in the treatment of 8 mM silicon.

The results of this experiment showed that the application of silicon, by increasing the activity of antioxidant enzymes, reduced the oxidative damage caused by reactive oxygen species and thus protected camellia seedlings against salt stress. In general, it seems that the use of silicon has been effective in reducing the adverse effects of salinity stress on growth and biochemical characteristics of camelina seedlings.

Various biomass sources such as crop residues have been proposed as feedstock for biochar production . Meanwhile, a large quantity of crop residues (rice) is produced as waste and they are either burnt or piled and abandoned at some locations in the fields. Burning of crop residues is resulting in substantial loss of nutrients, and may lead to air pollution and human health problems . An alternative approach is to apply crop residues to soil in the form of biochar. Bioavailability of nutrients exclusively micronutrients (Fe, Zn) is a serious problem in soils having high pH which ends in crops yield to decline and ultimately can lead to malnutrition in humans. The biochar modification with acid may increase the solubility of nutrients (P, , Fe, Zn, Cu,,Mn) present in biochar, thereby significant improvement in mineral nutrition of plants grown in calcareous soils. In the other hand, One of the ways to use and exploit saline lands is to use salinity-tolerant cultivars, such as the Quinoa (Chenopodium quinoa) plant. It is known that biochar increases soil pH, which may result in less availability of phosphorus and other micronutrients, such as Fe, Zn, and Mn, in alkaline and calcareous soils. Therefore, modifying biochar with acids can increase the availability of nutrients in biochar for different plants grown in calcareous soils. The objection of this study is to investigate the effect of normal biochar and acid-modified biochar from rice residues on the yield and yield components of quinoa plants (Gizavan number) in a calcareous soil affected by salt.

The soil used in the study was collected from 0-30 cm depth which passed through via 2-mm sieve after air-drying and its chemical and physical properties were determined. To achieve the aim of this study, the factorial experiment was carried out in a completely randomized design in 4 replications. Factors include 3 types of rice biochar (unmodified, modified by pre-acidic method and modified by post-acidic method) and different levels of biochar (0, 2, and 5% by weight). Then 10 quinoa seeds were planted in each pot at 2 cm depth which after the plant emerging and greening declined to 3 plants in each pot. The pots were randomly moved twice a week during growth to eliminate environmental effects. Irrigation and weeding operations were done by hand. After the end of the growth period (187 days), the plants were harvested. So vegetative growth parameters and yield components including shoots fresh and dry weight, plant height, stem diameter, panicle length, number of leaves, number of lateral branches, and 1000 grain weight were measured and then biological yield and harvest index were determined. The statistical results of the data were analyzed using SAS software (9.4) and the LSD test (at 5% level) was used for comparing the mean values.

As a result of adding biochar to soil, it becomes alkaline. Chemical modification of biochar using strong acids can reduce soil pH and improve the fertility of calcareous soils and increase vegetative parameters and yield components of quinoa. Based on the obtained results, the interaction effect of different types and levels of biochar on all investigated traits was significant at the level of 1%. The results showed that the highest height, fresh and dry weight, panicle length, number of lateral branches, and stem diameter were related to the 5% post-acidic rice biochar treatment and the lowest value was related to the control treatment. furthermore, the results showed that the highest amount of plant dry weight of 8.82 gr/pot, the height of 77.50 cm, and 1000 seed weight of 17.3 gr/pot was obtained from the treatment of 5% post-acidic rice biochar, compared to the treatment of 5% Unacidified rice biochar had an increase of (81.97), (56.77), (32.17) and (7.06) percent respectively. As a result of the high dry weight of shoots and the 1000 seed weight, the 5% post-acidic rice biochar treatment provided the highest biological yield at 16.05 and harvest index at 45.03.

Under the conditions of this study, acid-modified biochars (post-acidic and pre-acidic) enhanced vegetative growth characteristics and yield components of quinoa plants in calcareous soils affected by salt. Therefore, it is recommended to prepare biochar from acidic sources or to modify it with post-acidic and pre- acidic methods.

Despite the great use of digital soil maps, the problems of imbalance in classification disrupt the classification performance of many machine learning algorithms, and for this reason, it has attracted the attention of many researchers. Therefore, the aim of this research is to improve the classification of unbalanced soil data using resampling pretreatment technique in three forecasting models including Random forest (RF), Boosted regression trees (BRT) and Multinomial logistic regression (MNLR) in a part of the lands of Zanjan province in Iran.Sampling was done based on a regular grid pattern with 500 meters intervals, and 148 soil surfaces were randomly studied and classified. The region's soils at the subgroup level were in five classes with imbalanced distribution, including Typic Calcixerepts, Typic Haploxerepts, Gypsic Haploxerepts, Typic Xerorthents, and Lithic Xerorthents. Environmental covariates included geomorphological and geological maps, digital elevation model (DEM), and remote sensing (RS), selected by principal component analysis (PCA) and expert knowledge methods AND a number of environmental variables including geomorphological map information, Geological information and features extracted from the DEM were selected as the most effective environmental variables for predicting soil classes and as input to the model. Extraction of environmental covariates was done in ENVI and SAGA_GIS software and modeling of soil-landscape relationship was done using the aforementioned algorithms in Rstudio software. The resampling technique was applied to the minority and majority soil classes prior to modeling.The results showed that using original data that have imbalanced classes for mapping resulted in loss of the minority classes and relatively low Kappa agreement values and overall accuracy for RF (ovrall=65%, k=0.32) and BRT models (ovrall=60%, k=0.35). However, after resampling the data, two overall accuracy and Kappa coefficient statistics increased in all models. In addition, the BRT model provided an acceptable estimate by maintaining the minority classes and the Kappa coefficient of 0.64 and the overall accuracy of 75% in the spatial prediction of soil subgroups. The producer accuracy (PA) and user accuracy (UA) results showed that the two classes of Gypsic Haploxerepts and Lithic Xerorthents, which were excluded when training using imbalanced datasets in RF and BRT algorithms, showed significant improvement after balancing the data. Results show that they were well predicted in RF algorithm (UA =100%, 78%) and BRT algorithm (UA= 60% and 70%) using treated data. Also, these minority classes showed Producer accuracy in RF algorithm (PA= 75%, 88%) and BRT algorithm (PA=100%, 78%) in compared to zero accuracy when training using imbalanced data. On the other hand, the validation results of the MNLR algorithm showed that despite maintaining the minority classes after balancing the data, the minority classes were predicted with less accuracy. Results showed that modeling using imbalanced distribution of class observation caused uncertain maps with minority classes being lost and relatively poor accuracies. After data treatment, with over- and under-sampling, all models showed significant improvement in maintaining the minority classes, in evaluations. Data resampling technique can be a useful solution for dealing with imbalanced class observations to produce more certain digital soil maps.

Despite the great use of digital soil maps, the problems of imbalance in classification disrupt the classification performance of many machine learning algorithms, and for this reason, it has attracted the attention of many researchers. Therefore, the aim of this research is to improve the classification of unbalanced soil data using resampling pretreatment technique in three forecasting models including Random forest (RF), Boosted regression trees (BRT) and Multinomial logistic regression (MNLR) in a part of the lands of Zanjan province in Iran.Sampling was done based on a regular grid pattern with 500 meters intervals, and 148 soil surfaces were randomly studied and classified. The region's soils at the subgroup level were in five classes with imbalanced distribution, including Typic Calcixerepts, Typic Haploxerepts, Gypsic Haploxerepts, Typic Xerorthents, and Lithic Xerorthents. Environmental covariates included geomorphological and geological maps, digital elevation model (DEM), and remote sensing (RS), selected by principal component analysis (PCA) and expert knowledge methods AND a number of environmental variables including geomorphological map information, Geological information and features extracted from the DEM were selected as the most effective environmental variables for predicting soil classes and as input to the model. Extraction of environmental covariates was done in ENVI and SAGA_GIS software and modeling of soil-landscape relationship was done using the aforementioned algorithms in Rstudio software. The resampling technique was applied to the minority and majority soil classes prior to modeling.The results showed that using original data that have imbalanced classes for mapping resulted in loss of the minority classes and relatively low Kappa agreement values and overall accuracy for RF (ovrall=65%, k=0.32) and BRT models (ovrall=60%, k=0.35). However, after resampling the data, two overall accuracy and Kappa coefficient statistics increased in all models. In addition, the BRT model provided an acceptable estimate by maintaining the minority classes and the Kappa coefficient of 0.64 and the overall accuracy of 75% in the spatial prediction of soil subgroups. The producer accuracy (PA) and user accuracy (UA) results showed that the two classes of Gypsic Haploxerepts and Lithic Xerorthents, which were excluded when training using imbalanced datasets in RF and BRT algorithms, showed significant improvement after balancing the data. Results show that they were well predicted in RF algorithm (UA =100%, 78%) and BRT algorithm (UA= 60% and 70%) using treated data. Also, these minority classes showed Producer accuracy in RF algorithm (PA= 75%, 88%) and BRT algorithm (PA=100%, 78%) in compared to zero accuracy when training using imbalanced data. On the other hand, the validation results of the MNLR algorithm showed that despite maintaining the minority classes after balancing the data, the minority classes were predicted with less accuracy. Results showed that modeling using imbalanced distribution of class observation caused uncertain maps with minority classes being lost and relatively poor accuracies. After data treatment, with over- and under-sampling, all models showed significant improvement in maintaining the minority classes, in evaluations. Data resampling technique can be a useful solution for dealing with imbalanced class observations to produce more certain digital soil maps.

Phosphorus deficiency is one of the major problems of calcareous soils and a limiting factor for crop production in these soils and excessive use of phosphate fertilizers can cause pollution in soil and water. The use of organic amendments such as compost, biochar or a combination of them can be effective in improving the amount of available phosphorus of the soil. For this purpose, the effect of combined application of compost and biochar of sugarcane bagasse on phosphorus sorption and desorption were investigated.

Sugarcane bagasse and compost were provided from Debal Khozaei agro-industry Company. The oven-dried sugarcane bagasse 105 ºC were pass through 2 mm sieve and slow pyrolysed at 500 ºC using a laboratory muffle furnace with a heating rate of 5 ºC min-1 and presence of N2 flow to provide an anoxic condition. Some physico-chemical properties of samples were determined. The soil sample was collected from 0-30 cm of the campus of Shahid Chamran University of Ahvaz, Ahvaz, Khuzestan province, SW Iran (48° 65′91.12′′E31°30′53.82′′N). The studied soil classified as a Typic Haplocalcids. The air-dried soil samples were sieved (˂ 2 mm) and used for physico-chemical analysis. The incubation experiment was conducted with 7 treatments. The treatments including (1) control (without any amendments), (2) 100% sugarcane bagasse, (3) 100% sugarcane bagasse compost, (4) 100% sugarcane bagasse biochar, (5) 50% compost +50% biochar, (6) 75% compost + 25% biochar, (7) 25% compost + 75% biochar were added as 2% w/w to soil. The 200 g air-dried soil and treatments were mixed and kept in poly-ethylene containers for 120 days. Samples were incubated in 25  2 ºC and soil moisture was adjusted to soil field capacity using distilled water during the incubation time. At the end of incubation time, soil phosphorus sorption isotherm was measured. A 2.5 g of each treatment was transferred to a 50 ml centrifuge tube. Then, a 25 mL of CaCl2 0.01 M containing 0, 10, 20, 40, 60, 80, and 100 mg P L-1 (prepared from KH2PO4) was added to each centrifuge tube. Two drops of chloroform were added to each centrifuge tube to inhibit microbial growth. Samples were equilibrated at 25  1 ºC for 24 h on shaker at 150 rpm, and then centrifuged for 5 min at 3000 rpm and pass through 0.45-μm filter paper. The phosphate desorption was conducted on soil remaining in the filter immediately after sorption experiment. For this purpose, each treatment was resuspended with 25 ml of CaCl2 0.01 M solution without phosphate and shaken for 24 h. after collecting the supernatant, desorbed phosphate was measured. For assessing the adsorbed phosphate, the difference between the initial phosphate concentration and the phosphate concentration at equilibrium was calculated. The Langmuir, Freundlich and temkin isotherm models were used to describe the sorption of phosphate. In addition, some phosphorus buffering indices including maximum buffering capacity (MBC), standard buffering capacity (SBC), equilibrium buffering capacity (EBC) and standard phosphorus requirement (SPR) were obtained from P sorption equations at 0.2 mg P L-1 concentration in soil solution. The experimental data were fitted by Microsoft Excel-SOLVER and graphs were plotted by Microsoft Excel.

The soil was had loam texture, with low SOC content and high pH and calcium carbonate content, also the results of the characteristics of sugarcane bagasse compost and biochar showed that the compost had high salinity and the biochar had high pH and C/N ratio. The amount of phosphorus absorption increased with increasing the initial concentration of phosphorus in the treated soils. The highest and lowest amount of phosphorus absorption were in the control in compost treatments, respectively. In general, different levels of biochar and compost treatments caused a decrease in phosphorus absorption compared to the control treatment. The results showed that P sorption and desorption are described well by the Freundlich and Langmuir equations with a high correlation coefficient; however, the Temkin equation described the P sorption and desorption in the soils poorly. Biochar and compost treatments significantly decreased the Freundlich n parameter. Results showed that the effects of compost and 75% compost + 25% biochar were significantly greater than the effects of other treatments on the n parameter exponential adsorption equation. Application of different treatments of sugarcane bagasse compost and biochar application caused a significant increase in MBC (25.4-70.7%) and EBC (33.1-69.4%). The standard P requirements (SPR) were lower in soils treated than in control soil.

The results showed that the combined application of biochar-compost of sugarcane bagasse reduced the sorption and increased desorption of phosphorus. The maximum buffering capacity (MBC) and equilibrium buffering capacity (EBC), standard buffering capacity (SBC) and standard phosphorus requirement (SPR) in compost and compost 75% + biochar 25% showed more decrease than the control. In general, the results of this study indicate that the combined application of biochar-compost of sugarcane bagasse reduces phosphorus sorption in soil in calcareous soils, which can increase the availability phosphorus for plants.

The processes of soil erosion and sediment transport along rivers are the main causes of some socio-economic and environmental problems, such as a reduction in water quality, storage capacity of dams, destruction of aquatic habitats, failure of hydroelectric power plants, and soil degradation. Therefore, understanding the sedimentation status of watersheds is crucial for the effective management of soil and water resources. However, due to the lack of technical and human resources, continuous recording of sediment data is not possible in most sediment measuring stations, and sediment data are recorded only for a few days. In such a situation, a model that can estimate the amount of sediment load using auxiliary variables such as stream discharge and rainfall becomes crucial. Today, it is believed that techniques based on artificial intelligence have a much greater ability to uncover hidden relationships between variables than classical methods and are thus very useful and effective in modeling natural processes.

In this study, various machine learning techniques, including Artificial Neural Network (ANN), Adaptive Fuzzy-Neural Inference System (ANFIS), and Random Forest (RF), were used for sediment load modeling and sediment forecast for days without measurements. To achieve the research objectives, long-term meteorological and hydrometric data ranging from 2000 to 2020 were collected from related organizations and pre-processed before entering the model. The input variables for the models included 24-hour rainfall, flow rate, normalized difference vegetation index, maximum and minimum temperature, and daily suspended sediment as the dependent variable. Prior to modeling, the entire dataset was divided into two parts, training and testing, in a 70:30 ratio. Relationship modeling was performed using the training data, and model validation was conducted using the test dataset. The efficiency of the models was evaluated using two indicators, the coefficient of explanation (R2) and the root mean square error (RMSE). Additionally, morphometric parameters such as form factor (FF), drainage density (DF) coefficient, and relief ratio (RR) were utilized in modeling.

The hydrological analysis of the basin revealed that the highest annual amount of rainfall and erosivity index were recorded at the Sheyvand station in the east of the basin, while the lowest values were observed at the Ramhormoz station. The highest average monthly flow rate of 5.8 cubic meters per second was obtained at the Manjeniq station in April, and at the Mashin station, the highest average monthly flow rate of 8.8 cubic meters per second was recorded in December and January. Morphometrically, the studied basin belonged to the class of elongated basins, sloping basins in terms of relief, and the medium class in terms of drainage density. Analysis of the time series of NDVI index showed that the highest vegetation cover occurred in March, while the lowest values were recorded in September and October. The annual trend of the vegetation index indicated an overall improvement in vegetation cover in the region from 2000 to 2020, with the NDVI value increasing from 0.15 to 0.22. Among the different machine learning techniques studied, the Artificial Neural Network (ANN) model had the highest coefficient of explanation (R2=0.87) and the lowest RMSE for both sediment measuring stations in the region, making it the best model. The optimal inputs for the neural network model at Mashin station were daily average flow adjusted by the basin shape factor, daily rainfall, last day's rainfall, daily minimum temperature and daily maximum temperature. For the Manjeniq station, the optimal inputs were daily average flow, daily rainfall, last day's rainfall, cumulative rainfall for the past two days, and cumulative rainfall for the past three days. The NDVI index was removed from the model due to its low significance. The Random Forest (RF) model ranked second, and the Adaptive Fuzzy-Neural Inference System (ANFIS) model ranked third, with weak performance, especially for the Mashin station, where out-of-range errors occurred. Temporal analysis of sediment values showed that the highest sediment production occurred in December and January for Mashin station and in April for Manjeniq station. The highest production of sediment occurred in 2006 and 2002, and the trend of changes from 2011 to 2018 showed a decline, attributed to consecutive droughts and lack of rainfall. The annual average sediment production calculated using the values estimated with the neural network model was 88017 tons, equivalent to 1 ton per hectare per year.

Overall, this research demonstrated that machine learning methods, especially the neural network model, are highly effective for modeling and predicting sediment on a daily scale. These methods can compensate for the lack of sediment measuring facilities and equipment in most existing hydrometric stations in the country and eliminate the need for continuous sediment data and other water quality parameters.