Clustering BMKG Stations in Central Java Based on Meteorological Characteristics Using K-Means Clustering

Tri Ayu Mulyani; Agung Prabowo; Niken Ayu Larasati; Setyo Luthfi Okta Yohandoko

Tri Ayu Mulyani Mathematics Study Program, Mathematics Department, Faculty of Mathematics and Natural Sciences, Universitas Jenderal Soedirman, Purwokerto, Indonesia.
Agung Prabowo Statistics Study Program, Mathematics Department, Faculty of Mathematics and Natural Sciences, Universitas Jenderal Soedirman, Purwokerto, Indonesia.
Niken Ayu Larasati Mathematics Study Program, Mathematics Department, Faculty of Mathematics and Natural Sciences, Universitas Jenderal Soedirman, Purwokerto, Indonesia.
Setyo Luthfi Okta Yohandoko Mathematics Study Program, Mathematics Department, Faculty of Mathematics and Natural Sciences, Universitas Padjdajaran, Sumedang, Indonesia.

Abstract

The Meteorology, Climatology, and Geophysics Agency (BMKG) is a trusted provider of weather data. The meteorological characteristics assessed at BMKG stations include elevation, air temperature, humidity, rainfall, and the number of rainy days. This research aims to categorize BMKG stations according to these meteorological characteristics by employing the K-Means Clustering method. An analysis of data from 34 BMKG stations in Central Java, collected between 2019 and 2023, shows an average elevation of 101.6536 meters above sea level, an average air temperature of 22.5965 degrees Celsius, humidity levels at 66.9224 percent, total rainfall measuring 1,848.3924 mm, and an average of 119.5294 rainy days. The clustering process led to the formation of three distinct clusters: Cluster 1, which includes 22 BMKG stations; Cluster 2, containing 6 BMKG stations; and Cluster 3, which also comprises 6 BMKG stations. The assessment of clustering effectiveness using the Davies-Bouldin Index resulted in a DBI value of 1.02575, suggesting that while the clustering outcomes are satisfactory, they are not fully optimized.