Rice monitoring and production estimation using multitemporal RADARSAT

Rice monitoring and production estimation has special significance to China , as rice is the staple grain and accounts for 42% of the crop production i n this country. Radar remote sensing is appropriate for monitoring rice bec ause the areas where this crop is cultivated are often cloudy and rainy. Sy nthetic Aperture Radar (SAR) is thus anticipated to be the dominant high-re solution remote sensing data source for agricultural applications in tropic al and subtropical regions. IP also provides revisit schedules suitable for agricultural monitoring. This paper presents the results of a study examin ing the backscatter behavior of rice as a function of time using multitempo ral RADARSAT data acquired in 1996 and 1997. A rice-type distribution map w as produced, showing four types of rice with different life spans ranging f rom 80 days to 120 - 125 days. The Life span of a rice crop has significant impact on the yield, as well as on the taste and quality of the rice, with the longer growing varieties having the best taste and the highest product ivity. The rice production of three counties and two administrative regions , totaling 5000 km(2), was estimated in this study. The accuracy of the ric e classification was found to be 91% (97% after postclassification filterin g) providing confidence that multitemporal RADARSAT data is capable of rice mapping. An empirical growth model was then applied to the results of the rice classification, which related radar backscatter values to rice life sp ans. These life spans could then be used to sum up the production estimates , which were obtained from agronomic models already in use for rice by loca l agronomists. These models related the yield of rice to their life span ba sed on empirical observations for each type of rice. The resulting producti vity estimate could not be compared to any other existing data on yield pro duction for the study-area, but was well received by the local authorities. Based on the studies carried out in the Zhaoqing test site since 1993, it is suggested that rice production estimates require three radar data acquis itions taken at three different stages of crop growth and development. Thes e three growth stages an: at the end of the transplanting and seedling deve lopment period, during the ear differentiation period, and at the beginning of the harvest period. Alternatively, if multiparameter radar data is avai lable, only two data acquisitions may be needed. These would be at the end of the transplanting and seedling development period, and at the beginning of the harvest period. This paper also proposes an operational scenario for rice monitoring and production estimation. (C) 2001 Elsevier Science Inc. All rights reserved.