Mapping of selected soil properties in cashew growing soil in Puttalam District Sri Lanka

1. Jayawardhana HMAS, 2. Karunarathne SB and 3. Abeyasinghe DC

Understanding and quantification of soil spatial variability is important to manage soil in a sustainable manner. Geostatistics enables to quantify the spatial variability and create digital soil maps across the landscape exhibit the variations. This study was carried out to characterise the spatial dependency of key soil properties namely soil pH and Soil Organic Carbon (SOC) content in Red- Yellow Latosols in Puttalam district of Sri Lanka where predominant land use type was a cashew plantation. Study area (369.98 ha) was clustered using k- mean clustering algorithm in to 20 clusters and within each cluster two samples were randomly selected to carry out the field survey. Results revealed that the optimum spatial model for the soil pH and SOC contents were an Exponential model, and represented a strongly structured spatial dependency.

It was also reported that the spatial correlation (range parameter) of soil pH and SOC contents were reported as 99.9 m and 421.1 m respectively. Mapped soil pH indicated that large part of the area is in acidic condition, range from 5.0-5.5. A fairly low SOC distribution was exhibit in the majority of the area ranging from 1.0-1.2% for 0-0.3m depth intervals. The study laid the foundation for site specific input management and to select appropriate crops for intercropping within the cashew plantation. In addition, range value parameters obtained from the geostatistical analysis can be used as a guide to determine optimum sampling distance in future soil sampling in the study region.

Key words: Digital soil map, Spatial variability

_{Figure 1: Predicted maps for (a) Soil Organic Carbon % and (b) Soil pH}