The main purpose of this analysis is to develop the methodology for grasping agricultural production comprehensively, timely and exactly, applying satellite data overlaid geographic information. Classification of agro-environmental conditions was attempted using Landsat TM data in North-East Thailand where the production is low and the problem of soil salinization is most serious. Landsat TM data obtained in early dry season (Dec. 18, 1987) and in early wet season (June 11, 1988) were analysed in this experiment. A land use map and a biomass distribution map were created from Landsat image analysis for each season. Then average and difference of biomass between the two seasons were calculated. In addition, soil organic matter distribution map was produced from TM image acquired in early and mid dry season (Feb. 9, 1990). On the other hand, elevation and land inclination angle in topograph map were classified into some levels and evaluated from view point of suitability of crop production. Also soil fertility map which was published in 1973 from Land Development Department of Thai was used as an existing information. Thus, the following thematic maps were prepared through Landsat data analysis and existing data transformation. Landsat origin: Land use map, Average biomass distribution map, Difference biomass distribution map, Soil organic matter distribution map. Existing data origin: Elevation class map, Land inclination class map, Soil fertility map. In the next step, we examined ground truth data and agricultural census of this area to give reasonable weights on the produced maps from the view point of land suitablity for rice and upland crops cultivation, respectively. Overlaying these weighted thematic maps, and finally produced agricultural productivity maps for each paddy and upland field. The properiety and accuracy of the evaluation maps was examined by the agricultural exparts working in this area and proved that the results were reasonable.

View full abstract

Producer price of sugar beet is determined on quality and quantity basis, while it was determined by quantity only until 1985. This change has drew farmer's attention to the strategy to estimate and enhance sugar rate as well as root weight. We estimated the sugar rate and the root weight at Tokachi Area, Hokkaido, Japan, using Landsat TM data obtained around harvesting time in 1986. The following equation were derived to estimate the sugar rate and root weight successfully : Sugar rate=0.172TM1o+0.342TM3o-3.103 (n=22, r=0.826**) Root weight=2.664TM1s+0.849TM4s-1.395TM5s-198.839 (n=22, r=0.909**) where TM1o and TM3o are Band1 and Band3 derived Landsat TM data on October 18 respectively. TM1s, TM4s and TM5s are Band1, Band4 and Band5 derived Landsat TM data on September 16 respectively. The sales per unit area was also estimated; that in brown andsol is high and that in alluvial soil or black andsol is low. These results indicate that estimation of sugar rate, root weight, and sales by Landsat TM data are useful to improve productivity of sugar beet by appropriate fertilization, for example.

View full abstract