Mapping Climate Variables as an Effort to Mitigate Forest Fire Disaster in Central Kalimantan

Abstract

Climate change and global warming have had far-reaching impacts at the global, national and local levels. Impacts that have emerged in recent decades include tropical storms, changes in weather patterns, floods, landslides, melting ice caps in the north and south, sea level rise, drought and land and forest fires. Land and forest fires are getting worse not only in Central Kalimantan, at least recorded in October 2021 there were 2,375 fire hotspots with an area burned based on the Central Kalimantan Province Regional Disaster Management Agency (BPBD) report in the field, which reached around 642.8 ha. The first step in mitigating land and forest fires is to make predictions or preliminary calculations related to rainfall, temperature and humidity that occur in Central Kalimantan. One of the methods that can be used in predicting weather is by using machine learning. This research focuses on machine learning, where this method will help automatically recognise complex patterns and make intelligent decisions based on data. Machine learning can learn patterns from historical data contained in the database from previous years to predict rainfall, temperature and humidity that occur in Central Kalimantan. The prediction results of the Mechine Learning calculation can be made into a distribution map based on the predicted value of the intervals.