Water-efficient sensing method for soil profiling in the paddy field

The comprehensive and reliable perception of moisture in a soil profile is essential to irrigation. To establish an efficient method for sensing soil moisture, field trials were conducted to analyze the spatial and temporal variations of moisture in a paddy soil profile by using coefficients of variation. The results showed that soil layers at shallow depths undergo more extensive changes in the coefficients of variation. Moisture perception is most sensitive within a depth range of 0-60 cm in the vertical soil profile of a paddy field. By using the clustering algorithm of Euclidean distance, the paddy soil profile was divided into three categories based on soil depth. The first category includes depths ranging from 10-20 cm, the second is from 30-40 cm, and the third is from 50-100 cm. Path analysis indicated that the most sensitive depths for moisture sensing in a paddy soil profile were 20 cm, 30 cm, and 50 cm, whereas the most sensitive depths for moisture sensing by time stability analysis were 20 cm, 40 cm, and 60 cm. Based on the multiple regression of sensitive depths, the results of quantitative inversion indicated that the time stability analysis results were 0.962 when the Cor(y, (y) over bar) of path analysis was 0.980, and the time stability analysis was 0.61 when the root mean square error (RMSE) of path analysis was 0.40. The relative error range between the measured and predicted values of path analysis was less than that of time stability analysis. These findings suggest that it is feasible to effectively sense the moisture of the entire vertical soil profile based on the sensitive depth. The present study has also determined that path analysis is superior to time stability analysis.