METODE ROBUST KRIGING UNTUK MENGESTIMASI DATA SPASIAL BERPENCILAN (Studi Kasus: Pencemaran Udara Gas NO2 di Kota Semarang)

Published: 30 Aug 2016.
Open Access

Citation Format:
Abstract

Kriging is a geostatistical analysis used to estimate the value of the function at an unsampled point by computing a spatial correlation in the neighbourhood of the sample point. Interpolation can produce less precise predictive value if there are outliers among the data. Outliers defined as extreme observation value of the other observation values. Robust kriging is development method of ordinary kriging which transform weight of classic semivariogram thus become semivariogram that robust to outlier of the data. This research aims to estimate the concentrate of Nitrogen Dioxide (NO2) in Semarang using robust kriging method. The spatial data used in this research is coordinates point and concentrate of Nitrogen Dioxide (NO2). This method compare between robust semivariogram and theoretical semivariogram (such as spherical, exponential, and gaussian models) to determine the best estimator of the theoretical semivariogram model. The analysis showed that the best theoretical semivariogram model is exponential model. The estimation of Nitrogen Dioxide concentration conducted at 177 urban communities in Semarang.

Keywords: kriging, outliers, robust kriging, robust semivariogram

Keywords: kriging; outliers; robust kriging; robust semivariogram

Article Metrics: