Landsat 8 satellite imagery analysis for rice production estimates (Case study: Bojonegoro regencys)

Bojonegoro as the mainstay of rice producers in the province of East Java, have a mission to realize the dream become a national food basket. In 2012, Bulog's Bojonegoro Bulog to be the highest regional subdivisions throughout Indonesia. Seeing this potential, it is necessary to attempt to monitor the stability of agricultural production on a regular basis. By integrating the technology of remote sensing using Landsat satellite imagery 8 to identify a growth phase and forecasting models Autoregressive Integrated Moving Average (ARIMA) to predict the productivity of rice, are expected to provide a solution and ease of repeated and continuous monitoring with wide area coverage. Identify the growth phase carried out in 9 phases. Of the linear regression between growth stage rice plants with vegetation index values are used, the value of the coefficient of determination (R2) of 0.7229 for NDVI algorithms and algorithms MSAVI amounted to 0,879. Used reflectance values of wave band SWIR2 (1.57μm-1.65μm) to help distinguish each growth phase of the identification algorithm MSAVI where to phase 3, 4, 5 has a reflectance SWIR2 above 0.15, while the phases 7, 8, 9 has reflectance SWIR2 under 0.15. Forecasting process rice productivity obtained seasonal ARIMA (1,0,0) 3. So that it can be seen Forecast Figures (ARAM) rice productivity for subround III in 2013 amounted to 66.21 quintal per hectare. Results highest estimate of 169,595.385 tons for tillering phase (15 weeks ahead of harvest) and amounted 72246.878 tons for seedling phase (13-14 harvest next week). So it can be seen that when the study was conducted, Bojonegoro located in the growing season.