资源环境科技发展态势分析平台

The first part presents a summary of existing methods for collecting land area, production, and yield data in the region. The second part discusses measurement errors associated with the existing data collection methods. The third part presents ways to address these measurement errors using remote sensing technology by showcasing results from three methodological research activities undertaken by the Asian Development Bank in three countries: the Lao People’s Democratic Republic, Thailand, and Viet Nam.

This report concludes with a summary of how other innovations, such as drones, computer-assisted personal interviewing, and artificial intelligence hold promise in transforming the field of agricultural statistics.

Highlights

• The importance of agricultural development in achieving poverty reduction is undisputed. Yet, limited efforts have been made to improve the accuracy and timeliness of agricultural statistics.
• Recognizing this challenge, the Asian Development Bank (ADB) joined the Global Strategy to Improve Agricultural and Rural Statistics (GSARS) as an implementing partner. Specifically, ADB piloted the use of remote sensing technology as an alternative to existing methods for generating key paddy rice statistics in three countries: the Lao People’s Democratic Republic, Thailand, and Viet Nam.
• As a first step, a systematic comparison of existing objective and subjective methods to estimate plot area, rice production, and yield was conducted. Significant differences were observed between the estimates derived from objective and subjective data collection methods.
• Given that the subjective methods are expensive and time consuming, three methodological studies were conducted to explore the viability of using satellite data as an alternative to traditional methods.
• In the first study, plot boundaries were traced on high-resolution Google Earth images to estimate area, which was compared with GPS derived plot area estimates. The second study employed a novel data fusion technique in Thai Binh province, Viet Nam, to generate a spatially disaggregated rice yield map. The utility of land-use maps developed from satellite data while constructing a sampling frame was explored in the third study.
• The results of the three methodological studies provide strong evidence in favor of satellite data as a viable alternative to existing methods to generate paddy rice statistics.
• Other technological innovations such as drones, computer-assisted personal interviewing (CAPI), and artificial intelligence hold much promise for the future of agricultural statistics.

This is a special supplement to the Key Indicators for Asia and the Pacific 2018.