Soil organic carbon mapping using remote sensing technique and multivariate regression model / Muhammad Radhi Abdul Rahman

Organic were the terms use to represent the materials that combined with or derived
from living organisms. The quantity of organic matter in soil is frequently used as an
indicator of the possible sustainability in a soil system. Soil organic matter was
significant part in nutrient cycle and fixing soil structure. Organic carbon in soil was
important to build up good health in soil environment and vital in supplying the needs
of the ecosystem. This project aims to identify the Soil Organic Carbon distribution
based on multivariate regression model. This project was used satellite imagery, SPOT
5 to estimate SOC distribution using remote sensing technique and soil sampling in the
Ladang Harumanis, UiTM Aran, Perlis. There were nine soil samplings were picked
randomly collected using a handheld Global Positioning System (GPS) unit to location
the position of the sampling points. The satellite data derived spectral indices, NDVI
and BSl were used to assess spatial distribution of SOC in the study area by testing in
the multivariate regression model. The result of regression analysis between the
observed and predicted SOC using = 0.10 value was showed only 10% accurate
because of the lack of number of soil samples and same land use type which no really
soil variations that reflected this result. This information of this study can gave
advanced understanding by using the remote sensing approach which had many
advantages regarding conventional approach before would be important technique thus
increase the effectivity of the soil management method.