Estimation of soil mechanical resistance parameter by using particle swarm optimization, genetic algorithm and multiple regression methods

• Determination of mechanical resistance in soils is difficult using penetrometer in fine textured.
• We tried to explore PSO, GA and MR to estimate mechanical resistance.
• The results revealed that PSO and GA models are promising approach for the estimation than MR model.
• The results also show that PSO model can estimate soil mechanical resistance more accurate than GA and MR.

Soil mechanical resistance directly affect seedling emergence and root growth. Tillage and irrigation managements decrease soil mechanical resistance in soil depth and increase root density in rhizosphere, wheat nutrient uptake and the yield. Soil mechanical resistance is high in the site of experiment due to high soil specific surface and cohesion between primary particles. Bulk density, pore size distribution and volumetric soil water content are important soil physical properties in relation to crop production through their effect on mechanical resistance. Determination of mechanical resistance in soils is difficult using penetrometer in fine textured soils particularly in hot and dry season. This study explores potentials of particle swarm optimization (PSO), genetic algorithm (GA) and multiple regression (MR) in the estimation of mechanical resistance value of soils. This research was carried out in a 3000 m2 piece of land at Gorgan University of Agricultural Science and Natural Resource Research Farm, Grogan, Iran. The land was ploughed with five tillage methods, namely conventional tillage with moldboard ploughing (MT), rototiller (RT), double disc (DD), chisel plow (CP) and no-tillage (NT). Then bulk density, volumetric soil water content and soil mechanical resistance were measured at six stages during growing season of wheat. Two physical properties of soils that include the soil bulk density (BD) and volumetric soil water content (Өv) were presented to the models as independent parameters to estimate soil mechanical resistance (Es). The performance of models was comprehensively evaluated some statistical criteria. The results showed that among the various tillage methods, Moldboard tillage (MT) reduced soil mechanical resistance which increased plant's root growth, water and nutrient uptake, head number per square meter, and wheat yield. Also, the results revealed that PSO and GA models are promising approach for the estimation of soil mechanical resistance in compare with MR model. The results also show that PSO model can estimate soil mechanical resistance more accurate than GA and MR with R2 = 0.932 and RMSE = 0.301.