Effects of loose soil layer thickness and compaction on soil subsidence after irrigation

Soil preparation is one of the most critical steps in rice production. The dry-land leveling operations have also been continuously expanding in paddy fields. However, the differential soil subsidence can often occur during irrigation after dry tillage. Field levelness can be significantly reduced on the efficiency of the rice growth and yield. In this study, a systematic investigation was implemented on the influencing factors on the differential soil subsidence. A series of the field tests were carried out to determine the effects of the loos soil layer thickness and compaction on the soil subsidence after irrigation. The relationship was then established between these factors and soil subsidence. The soil subsidence was controlled using moderate compaction (defined as compaction thickness). A nonlinear regression model was constructed between loose soil layer thickness, compaction thickness, and soil subsidence (or expansion). A container test was finally conducted to validate the feasibility. Field tests revealed that there was the significant differential soil subsidence that caused by dry tillage and subsequent irrigation. In terms of the terrain changes, the rotary tillage increased the average terrain height of the field by 47 mm, whereas, the post-irrigation decreased 14 mm. Additionally, the areas with the greater height after rotary tillage also exhibited the significantly higher subsidence after irrigation. In the soil physical properties, the Mann-Kendall test was used to locate the abrupt variations in the soil penetration resistance, indicating the thickness of the loose soil layer. The field tests showed that the loose soil layer was typically 0–50 mm thick before rotary tillage, thus increasing to 125–175 mm afterward. There was the negligible subsidence after irrigation before tillage. But there was the significant subsidence after tillage and irrigation. As such, the thickness of the loose soil layer was the primary factor on the differential soil subsidence. The thickness of the loose soil layer was taken as the key variable with the real-world needs from the perspective of the practical production. Furthermore, the agricultural machinery that equipped with sensors was utilized to real-time measure the thickness of the loose soil layer for the soil subsidence. The container tests show that there was the relationship between loose soil layer thickness and soil subsidence. There was the significant positive linear correlation between them, with a determination coefficient of 0.97. Once the loose soil layer thickness was 50 mm, the subsidence was nearly negligible; When the thickness increased to 300 mm, the soil subsidence reached 36.15 mm, thus exceeding the requirement of the high-standard farmland leveling (±30 mm). Additionally, the soil subsidence was controlled after moderate compaction, defined as the compaction thickness. Container test results showed that the post-irrigation subsidence of the loose soil layer decreased gradually, as the compaction thickness increased. Once the compaction thickness exceeded the critical value, the soil expansion began to occur. A nonlinear regression analysis was conducted for the relationship between loose soil layer thickness, compaction thickness, and soil subsidence (or expansion). A determination coefficient of 0.964 and a root mean square error of 2.4 mm were achieved for the excellent accuracy. According to the principle of subsidence uniformity, the model validation tests demonstrated that there were the absolute errors between calculated and measured subsidence of 0.64 and 0.25 mm, respectively, with the root mean square errors of 1.43 and 0.39 mm, respectively, indicating the high precision and reliability. The findings can provide a scientific basis to develop and apply the soil preparation using intelligent technologies. The precise control of the post-irrigation soil subsidence and field levelness can be expected to enhance the production efficiency and profitability. Moreover, the quantification of the soil subsidence can also offer the spatial and temporal guidance for the precision field of the ridge-planted crops.