Background and Objectives
The frontline of agriculture and the agricultural environment that surrounds it are changing dramatically both in terms of space and time. It has become increasingly important to capture such changes in agricultural environment accurately in order to achieve sound and efficient management of farming. In terms of spatial changes, slight changes in just one plot of a rice paddy or upland field may require attention, while for different purposes overall changes in one continent may have to be considered. In terms of time, likewise, analysis of daily changes may be relevant, while long-term changes over decades and centuries may have important significance. We are conducting a leading-edge research to capture and analyze these spatiotemporal changes in agricultural environment by the use of remote sensing and geographical information system.
Remote sensing refers to the technology of measuring objects without touching them by the use of sensors mounted onto a satellite or an aircraft or those installed near farmlands. Geographic Information System ( “GIS” ) is the technology used for handling various types of geographical information by means of a computer.
Expected Outcomes
Accurate understanding of rice paddy and upland field conditions by remote sensing will permit more precise management of agrochemicals and fertilizer application (precision farming). On another front, the preparation of wide-area land use maps and cultivation history maps will be useful for better understanding and estimation of global warming and other wide-area changes in the environment as well as the impact that such environmental changes will have on food production. Moreover, analysis of the complex relationship between wildlife habitat and human land use will contribute to conservation of biodiversity and prevention of damage by wildlife.
Relevant Outcomes to Date
Measurement of community characteristics by multifrequency microwave remote sensing
Microwaves are electromagnetic waves with a longer wavelength than that of visible light. Since microwaves pass through clouds, they are an effective tool for remote sensing, particularly in Japan and other Asian countries in the monsoon region that see heavy clouds in the summer. Microwaves of various frequencies are emitted onto a grown rice community from a measuring unit situated by the rice paddy. When the intensity of the electromagnetic waves that are returned after scattering by the plant bodies is measured, we find that frequency differences represent different scattering processes (Fig. 1a). Consequently, we succeeded in estimating the surface area of rice leaf, total weight, stems density, ear weight, etc. by using microwaves of differing frequencies. Fig. 1b shows the ear weight estimated using the Ku band (frequency 15.95GHz) against actual measurements. As can been seen from the graph, the Ku band sensing offered an accurate estimate of the rice ear weight without the need to cut out rice samples.
This finding can be utilized for precise management of rice paddies and upland fields as well for future development of satellite-based remote sensing.
Fig. 1a Various scattering processes by microwaves of different frequencies
Fig. 1b Agreement of ear weight estimation by Ku band
Cloud-free estimation of rice-planted area by the satellite-mounted radar “SAR”
For the Japanese and other Asian peoples that live on rice, the accurate estimation of rice-planted area is an important task. Currently in Japan, the area is estimated by the laborious method of sampling surveys whereby the survey staff actually visit the sample site and do the measurement. We have successfully developed a method to accurately and efficiently estimate the rice-planted area by remote sensing. A type of radar called SAR emits microwaves from a satellite and measures the intensity of the electromagnetic waves that are returned after scattering on the earth surface. The rice-planted area is detected by comparing images taken on two different dates. The so-called “set-aside paddy” which is not planted but filled with water for the mere purpose of preventing growth of weeds can be differentiated (Fig. 2a). We have also used a geographical information system to improve the accuracy further by reducing erroneous findings on inclined landscape (Fig. 2b).
Fig. 2a Rice-planted area is detected by the differences in microwave scattering between the rice planting period and the rice growth period. The spots still in black in the image on the right represent “set-aside paddies.”
Fig. 2b The rice-planted area can be estimated with an accuracy comparable to that of the current survey method. (Estimated rice-planted area in 26 municipalities in Saga Plain)
Understanding land use alteration with the help of century-old Ready Survey Map
The so-called “Ready Survey Map” which was prepared by the Imperial Army in the early days of the Meiji Era (1880s) is the first map of Japan based on modern land survey techniques. It serves as a very valuable reference to visually reconstruct the agricultural environment of 100 years ago. Since no modern-day map projection system was adopted in the Ready Survey Map, however, it was difficult to make a quantitative comparison with today's land use map. We have successfully developed a method to accurately overlay the Ready Survey Map and the present-day topography map. Fig. 3a shows the reconstructed land use in the Ushiku district in the 1880s (the present-day projection system was applied). Comparison of this corrected Ready Survey Map against present-day land use map has made it possible to understand closely the complex changes in land use in this district (Fig. 3b). We will work to apply this method to increase understanding of the changes in wildlife habitats and other purposes.
Fig. 3a Land use in Ushiku district in the 1880s as reconstructed from the Ready Survey Map
Fig. 3b Land use alteration after the lapse of 100 years from 1880