Sampling Design Using Remote Sensing in Korea
Pyong Namkung1, Jong-Seok Byun2, Joon-Suk Lee3, Chan-soo Lim4
1Department of Statistics, Sungkyunkwan University, Seoul, Korea; 2Department of Statistics and Information, Hanshin University, Osan, Gyeonggi-Do, Korea; 3Team of Statistics, PDS Korea, Seoul, Korea; 4Department of Statistics, Sungkyunkwan University, Seoul, Korea

US, EU, Japan, India and China estimate arable land and crop area using remote sensing and aerial photograph. Ministry of Environment, Korea Maritime Institute, Korea Forest Service and National Academy of Agriculture Science of Korea have used remote sensing and aerial photograph. Among them, LUCAS was estimating arable land and crop area for 2 stage sampling method in EU.

Recently, Korea use 1 meter resolution because we have KOMPSAT-2. So, we thought that a weak point of field survey could be enhanced by sampling design using high resolution remote sensing data to estimate arable land.

Remote sensing data is two dimension grid. So, we considered what proper grid type and sampling design was. And then, arable land data of Ministry of Enviroment, which has 2.5 meter resolution and purpose surveying distribution of the whole nation land, was used sampling frame. Remote sensing data was used for estimation.

Therefore square grid was selected, stratified two stage cluster sampling was designed. Remote sensing data of Kimpo was used for estimation.

Likewise, we suggest proper sampling design of arable land estimation of Kimpo by comparing coefficients of variation, differences and s of some sampling methods.

Keywords: Grid; Remote Sensing; Remote Sensing data; Stratified two stage cluster sampling design

Biography: I am living in Korea and is in the process of attending in Sungkyunkwan universities.

Currently is a doctor completion.