Air pollution is an environmental problem that often occurs and needs to be addressed to produce healthy air quality. Nitrogen Dioxide (NO2) as one of the main pollutant components that contributes to the quality of the air produced. NO2 is caused by exhaust fumes in motor vehicles. Kriging is part of the geostatistical method used to estimate a value at an unsampled location point based on the sampled location points in the vicinity. This study provides a way to determine the level of NO2 pollution concentration in an area by using the method of spatial data geostatistics, namely the ordinary kriging method. Ordinary kriging method can provide an estimate to estimate the value at a measurement sample point, using a semivariogram model of regionalized variables. The data used in this study is secondary data obtained from the Department of Environment and Forestry of Aceh Province. Determination of the best model is done by adjusting the experimental semivariogram value with several values from the theoretical semivariogram model, namely the Spherical model, the Exponential model and the Gaussian model which has the smallest Mean Square Error (MSE). Based on the analysis that has been done, the Exponential model is the best model for prediction using ordinary kriging. The area with the highest prediction of air pollution cases in the estimation of air content (NO2) in Aceh Province is in the Lhokseumawe region, while the area with the lowest prediction is in the Simeulue region.
Keywords: Motor Vehicles, NO2, Ordinary Kriging, MSE