Paddy Field Zone Delineation Using Apparent Electrical Conductivity and its Relationship to the Chemical and Physical Properties of Soil

Spatial variability and temporal variability of soil chemical and physical properties within a field is unavoidable. Meanwhile, laboratory soil test is usually time consuming and laborious. To satisfy the concept of precision farming, rapid and intensive soil sampling is necessary for describing the uncertainty within a field. Apparent or bulk soil electrical conductivity (EC,) technique for describing soil spatial variability is widely used. A sensor known as VerisEC can measure the average EC, of 0-30 cm (shallow EC,) and 0-90 cm (deep EC,) depths and locate its position by Differential Global Positioning System (DGPS) at every second. EC, includes soil salinity and soil texture. Soil texture has high correlation with soil cation exchange capacity (CEC) hence, soil nutrient contents. The main purpose of this study was to generate variability map of soil EC, within rice cultivation areas using VerisEC sensor for three seasons. The EC, values were then compared to some soil chemical and physical properties namely pH, EC, OM, OC, total S, total N, available P, CEC, Ca, Mg, K, Na, Al, Fe, total cation, BS, ESP, dry bulk density, moisture content, clay, silt, fine sand, coarse sand and sand, within classes after delineation. The study site was 145 ha paddy fields at Block C, Sawah Sempadan in the Tanjong Karang Rice Irrigation Scheme, Malaysia. The sensor was pulled by a tractor in a U-shape pattern between the field drains (speed = 15 km h-'). Disturbed and undisturbed of 236 sampling points were collected and recorded their positions by GPS (Trimble GeoExplorer3). Soil properties and EC, data were mapped using kriging technique on GS+ and ArcGIS. SPSS and SAS were used for their statistical analysis. The study showed that the EC sensor can determine soil spatial and temporal variability, where it can acquire the soil information quickly with less labour. Most of the soil properties and EC, changed from one season to other seasons, except total N. Much of the soil properties had the same mean values for seasons 1 and 3 such as K, moisture content, silt and coarse sand. Spatial variability of shallow and deep EC, had the same pattern for different seasons even though the mean values were different. Deep EC, showed the pattern of former canal routes clearly as continuous lines about 45 m wide at the northern and central parts of the study area. Low Na in zone 1 delineated by deep EC, may be due to deep soil profile to reach the parent material of marine alluvial, where it was a former water route. High fine sand and sand in zone 1 were found for all the seasons. The models of soil properties estimation based on EC, varied spatially and temporally from season to season and even from zone to zone. Most of them can be estimated better by deep EC, except, soil K and ESP. The selected models showed that the highest significant R* was found in fine sand and sand models with the consistency of the model throughout the study seasons. The relationship between yield and deep EC, was non significant for all the seasons

Text ITMA_2006_6.pdf Download (1MB)

Actions (login required)

View Item