Detection of soil compaction using soil electrical conductivity

Conventional methods for soil compaction mapping, such as penetrometers, although accurate, work as stop-and-go providing point measurements. This process is both time consuming and labour intensive. On-the-go electrical Conductivity (EC) measurements such as electromagnetic induction (e.g. EM38) are affected by key soil properties including texture, moisture content and compaction, so offer a possible rapid alternative for compaction detection. Therefore, the aim of this work is the detection of the within-field variability of soil compaction using soil electrical conductivity for improved soil management. A methodology for identification of within-field variability and for comparison of the data collected by contact and contact-less EC sensors, soil compaction sensor (which contains of eight instrumented wedge faces attached to the leading edge of a tine) and cone penetrometer was developed and a randomised block design experiment was performed. The data was evaluated statistically, the maps of spatial variability were created and the areas for targeted soil loosening were determined. A key finding was the development of application maps for targeted soil loosening, based on soil electrical conductivity measurements. The practical utilisation of this method assumes the presence of two maps that would be compared. The initial map has to be created for the soil in desirable loosened conditions and will be used as a standard for further comparisons. Further map will characterize the conditions of the compacted soil. It is recommended that both maps would be obtained in similar conditions to minimize the effect of soil moisture. The application map can be then created either manually using visual comparison of both initial (loosened soil condition) and further map (compacted areas), or using an appropriate geostatistical tool, in this case a ratio of the two maps. It was found out that the EC readings collected by the Conductometer at depth range of 0-0.3 m are able to distinguish the soil areas with no compaction above 0.3 m and the soil compacted within whole profile. The EC readings obtained at depth range of 0-0.9 m can distinguish the soil zones with no compaction above 0.6 m from the rest of the field. Using the EC data obtained at both depth ranges it is possible to determine three different environments within the field: one with no compaction above 0.3 m, one with no compaction above 0.6 m and one containing the soil compacted within whole profile. The best results for determining the areas with different depths of soil compaction were obtained by soil compaction sensor and cone penetrometer. However the absolute values of soil compaction sensor were affected by small changes in soil texture. The slow data collection speed of point penetration resistance measurement technique practically limits the spatial resolution of final data set. The soil compaction sensor has to be attached on the frame of the subsoiler and although it provides precise readings, it has high energy requirement. The high labour and time requirements are the main disadvantages of penetration resistance measurements. Second most precise determination of the areas with different soil compaction presence was provided by the EC data collected by the Conductometer which is able to collect data on-the-go with much lower power requirements. EM38 operated in horizontal mode distinguished the areas with no compaction above 0.3 m and areas with soil compacted at whole profile with less precision. The same instrument operated in vertical mode was not sensitive enough to measure any differences in soil bulk density.