| Penetration resistance is an important soil physical property.Too high will cause co MPaction stress,which is not conducive to crop development,and too low will cause soil erosion.Current domestic and foreign research focuses on reclaiming soil or experimenting with fields,and does not reflect the differences between regions and soil types.Instead,it studies the spatial distribution of penetration resistance in large-scale grain-producing areas,quantitatively explores the driving forces of influencing factors,and reveals soil characteristics and penetration resistance.Mutual relations are of great significance for guiding the construction of fertile farming layers.In view of the unclear spatial characteristics of typical soil penetration resistance in dry farming areas in my country,the unclear role of influencing factors,and the unclear characteristics of structural fertility and other scientific and practical issues.This study takes the typical soil types in Northeast my country and the Huanghuaihai dry farming area as the research object.It uses classical statistics and geostatistics to explore the spatial distribution characteristics of soil penetration resistance.The prediction model of permeability resistance,quantitatively estimates the thickness of the cultivated layer through the detection of mutation points,systematically analyzes the characteristics of soil structure in dry farming areas,and clarifies the structural fertility index threshold.The main research conclusions obtained are as follows:(1)From the vertical section,all soil types follow the law of increasing first and then stably with the increase of soil depth in 0—40 cm.The overall average value is 1.402 MPa,and the average value of all soil types of chernozem is the smallest.It is 1.188 MPa,and the maximum is 1.706 MPa for cinnamon.From the horizontal section,for the overall soil,all soil layers are high-value areas in the southwest and low-value areas in the east.Typical soil types are distributed differently in each soil layer.The random variation of fluvo-aquic soil,cinnamon soil,and chernozem decreases with the depth of the soil layer,and the spatial variation of the overall soil and other soil types in each soil layer is not obvious.(2)Soil quality Water content,bulk density,silt content and clay content increase with soil depth;organic matter content and sand content decrease with depth.Under the same water content,black soil has the strongest water suction,and cinnamon soil is the weakest.The influence factors of penetration resistance are complex and diverse.Among them,water content and bulk density have the most significant i MPact on it.Penetration resistance decreases with water content and increases with bulk density.Among the prediction models,the extreme learning machine model has the highest applicability to the overall soil;for the fluvo-aquic soil and chernozem,the artificial neural network model has the best modeling effect;the transfer function(SPR=a*θgb*BDc,SPR soil penetration resistance,θg mass water content,BD bulk density)has the highest applicability for black soil and Shajiang black soil;it has the best modeling effect for cinnamon soil random forest model.(3)The Mann-Kendall trend analysis method can quickly,accurately and objectively estimate the thickness of the cultivated layer.The average thickness of the overall cultivated layer in the dry farming area is 17.9 cm,the highest is 18.8 cm in typical soil type,and the lowest is 15.5 cm in chernozem soil.The black soil has the best solid-liquid-gas three-phase structure and the best air permeability.The structural index thresholds of the fertile plow layer of each soil type in the study area are different.The overall soil fertile plow layer structural index thresholds are:water content of 14—18%,bulk density of 1.38—1.47 g/cm3,and clay-sand ratio of 6—8%,the thickness of the cultivated layer>20 cm,the GSSI is>98.05,the STPSD is<4.94,the matrix potential is 76.54—138.90 cm,and Ks is 51.37—80.08 cm/day.Figure[29]table[16]reference[106]... |