Development of the Empirical Model for Rice Field Distribution Mapping Using Multi-temporal Landsat Etm+ Data: Case Study in Bali Indonesia

Information about an area and its distribution of planted and harvested paddy rice field is essential for food security management, water resource management, and for the estimation of gas emissions. Rice field has specific coverage characteristics throughout its life time. Its coverage is mixed proportionally in accordance with its age, between water, soil, and rice vegetation. Landsat ETM+ has a good spatial, spectral, and temporal resolution for rice growth monitoring and production estimation. The study was done in Tabanan Regency, Bali Province, Indonesia. The objectives of the study were: (1) to develop the rice growth vegetation index (RGVI), (2) to map the rice distribution and its age, and (3) to quantitative compare the rice field area’s analysis results with the reference data. The results of this study show that an exponential equation is the best relationship form between a rice field spectral and the rice age. The visible bands of Landsat ETM+ (Band 1, Band 2, and Band 3) show a weak exponential relationship with the rice age. However, the reflective infrared bands of Landsat ETM+ (Band 4 and B5) and the entire vegetation index show a strong exponential relationship with the rice age. Using multiple bands of Landsat ETM+ as a vegetation index gives a better result than using a single band. RGVI is the new vegetation index developed in this study with the equation RGVI = 1 – ((b1+b3)/(b4+b5+b7)) where b1, b3, b4, and b5 are the bands of Landsat ETM+. The RGVI that is evaluated in this study is the better vegetation index compared to the existing vegetation index. The relationship between rice age and RGVI shows the highest determination coefficient (R2) with the equation y = 14.265e, where y and x are the rice age and RGVI respectively. Quantitative comparison of the rice field analysis results and reference data at district level shows a linear relationship with the equation y = 0.920x 3.841 and R2 = 0.971, where y is rice field area of reference data and x is the rice field area of the analysis result of the Landsat ETM+. The standard of this estimation is 43.04 ha. * Corresponding author. Email address: [email protected]