Model SPASIAL INDEKS LUAS DAUN (ILD) PADI MENGGUNAKAN DATA TM-LANDSAT UNTUK PREDIKSI PRODUK PADI

Abstract

The spatial model for irrigated paddy yield acreage and yield prediction use the Landsat-TM of remote sensing data which has been producted by LAPAN using the Vegetation Index (VI) as a single parameter. Verification of the model mentioned above has also been done for Java Island showing that the accuracy result is acceptable for the operational although there are some limitation of the model. The objective of this research is to develop a spatial model for the paddy yield acreage and the yield prediction using Landsat-TM data, based on another parameter i.e the single parameter of Leaf Area Index (LAI), or using both parameter of LAI and VI to improve the accuracy prediction, compared to the accuracy using the single parameter of VI. The spatial model based on the Leaf Area Index (LAI) reduces dynamic factor of the parameter which control the growth stage of the paddy in the field such as the soil moisture (level of water) and the weather condition such as the temperature and the moisture radition, pests and diseases. In this research phase, the profile of LAI against the paddy age based on the field measurement shows that the LAI value increases a long with the vegetative growth and reaches the peak value of 4,567 at the maximum vagatative index (8-9weeks after the planting time). Furthermore, the LAI value decreases a long with the generative growth. The LAI value at the maximum vegetative phase can be used to predict the paddy production. The relation between the LAI and the spectral bands combination of Landsat-TM can be obtained by using the Power Regression Model as follows: LAI=0,2219*(TM4/TM3)2,1005(R2=0,95) where LAI means the value Leaf of Area Index on the paddy object at the paddy field area, which represents the pixel in the image spatial distribution. While TM3 means the digital number (gray level value) of the pixel in the spectral band 3 of Landsat-TM image data which represent the paddy object at the paddy field area, and TM4 means the digital number of the pixel in the spectral band 4 of Landsat-TM image data, which represent the paddy object at the paddy field area. The research also shows the application example or the model or the algorithm whivh is obtained in this research by using Landsat-TM. The LAI spatial of the paddy field area in Kabupaten Subang/Sukamandi West Java can be produced.