Page 104 - Contributed Paper Session (CPS) - Volume 7

CPS2042 Anna Christine D. et al.

Improving paddy rice statistics using area sample
frame technique
Anna Christine Durante ,Pamela Lapitan, David Megill, Lakshman Nagraj Rao
Asian Development Bank

Abstract
Agricultural surveys typically use a list frame as the main sampling frame to
identify and access the elements of the target population. List frames are built
on the basis of information from administrative records and data from recently
conducted censuses. While the efficiency of survey implementation may be
affected by the overall survey process, frame quality has a major effect on the
quality of statistics produced. In most developing countries, a complete frame
with updated and comprehensive holding information may not be available,
leading to under-coverage and biased estimates. To circumvent these issues,
the Asian Development Bank evaluated the usefulness of an area sampling
frame developed using technological advances in satellite imagery and
geographic information system techniques for estimating paddy rice area and
production in a study that was piloted in three major rice producing provinces
in Southeast Asia—Savannakhet, Lao PDR; Ang Thong, Thailand; and Thai Binh,
Viet Nam. The study employed a three-stage stratified sampling technique for
which the sampling frame used was developed using freely available MODIS
and Landsat data on land cover and land use. Direct estimates of total paddy
rice area and production are calculated from area frame using two methods–
–one involving measurement of plot size using a global positioning system
instrument and the other utilizing a digitized map of farmer-identified plot
boundaries on a high- resolution Google Earth image. A third method
involving the calculation of ratio estimates using independent mesh-level
measures is compared with the first two methods involving direct estimates,
and with the estimates generated from administrative data from the countries.
Our study finds that ratio estimation significantly improves the level of
precision of paddy rice statistics. Significant deviations for paddy rice area,
production, and yield are also observed between official statistics and the
statistics generated through direct estimation. Nonetheless, the estimates are
likely more reliable when compared to official estimates since satellite-based
estimation methodology is transparent, reduces under-coverage and allowed
for calculation of confidence interval. It is also necessary to improve the land-
use stratification of the frame by using higher-resolution satellite images and
a greater power of discrimination in the models used for defining the strata.
Sentinel 1 and Sentinel 2, which have better resolution than existing freely

91 | I S I W S C 2 0 1 9