In investigating heavy metal contents at specific areas, the method of selecting sampling points is important. A general method is, according to the law , random sampling of zigzag-form in the selected field. In this work, we studied whether the measured values obtained from a certain method of selecting sampling points is a representative of heavy metal contents in the selected field or not. The selected field for this study is located on the lower Yangsan-river; Gasan-li, Mulgum-myon, Yangsan-gun, KyoungNam. 1 km away from the main stream of Nakdong river. The heavy metals investigated were Cd, Cr, Cu, Fe, Mn, Ni, Pb, and Zn. The inductively coupled plasma(ICP) atomic emission spectrometer was used to measure these metals quantitatively. The number of total sampling points were 24. We compared the total mean values with the mean obtained from various methods of selecting sampling points.

Observational date is integral in our understanding of present climate, its natural variability and any change due to anthropogenic effects. This study incorporates a brief overview of sampling requirements using data from the first ISLSCP Field Experiment (FIFE) in 1987 which was a multidisciplinary field experiment over a 15 km grid in Konza Prairie, USA. Sampling strategies were designed for precipitation and soil moisture measurements and also detecting land cover type. It was concludes that up to 8 raingages would be needed for valuable precipitation measurements covering the whole FIFE catchment, but only one soil moisture station. Results show that as new gages or station are added to the catchment then the sampling error is reduced, but the improvement in error performance is less as the number of gages or stations increases. Sampling from remotely sensed instruments shows different results. It can be seen that the sampling error at larger resolution sizes are small due to competing error contribution from both commission and omission error.