ANALISIS GEOSTATISTIK UNTUK PEMETAAN PERUBAHAN KUALITAS AIR TANAH KAWASAN KARST KABUPATEN GUNUNGKIDUL

Abstract

Gunungkidul Regency has a karst area of approximately 807 km2 or 53% of the total area of its territory. There is a tendency for expansion in karst mining leading to a number of potentials, including damage to the water system which is a pollution of karst water sources. Temperature, turbidity, Total Dissolve Solid (TDS), PH, hardness, manganese, iron, and chloride are parameters affecting groundwater quality. Measurement of the concentration of each parameter is performed through a long process and expensive costs. Therefore, not all measurements are performed in the entire area of Gunungkidul. Hence, it is important to interpolate the eight parameters using the geostatistical method. Geostatistical kriging method is an estimation method that reduces the error of variance estimation by a cross-correlation between primary and secondary variables. The best semivariogram for a five-year period with the smallest RMSE value is the temperature in 2018 using an gaussian model, turbidity in 2018 using a IDW model, Total Dissolve Solid (TDS) in 2017 using a gaussian model, PH in 2016 using a linear exponential, hardness in 2019 using a exponential model, manganese in 2017 using a circular model, iron in 2017 using a exponential  model, and chlorides in 2015 using a RBF. Monitoring points of groundwater quality using these eight parameters have different variances so that five parameters are producing more than one RMSE value. To resolve this, besides comparing several interpolation methods, natural logarithmic transformations and the correlation of actual values with estimates were also performed. The correlation between the actual value and the estimation indicates that the estimation produced by the non-transformed data is more accurate than the transformed data. The estimated results of each parameter are visualized in the form of a map so that changes in groundwater quality every year can be seen. Besides the maps, the results of this study are shown in graphs of changes in the form of cross-sections of each parameter from 2015 to 2019. Visualization of changes in the quality level groundwater is expected to give input for relevant agencies in the conservation of water resources. Keywords:  Karst Mining, Mapping, Geostatistics, Groundwater Quality.