Haze trajectory clusters in South Sumatra in 2015 A simulation using Opentraj

Haze is a negative consequence of forest and land fires that caused many health problems and contributed on global warming. This study aims to generate haze trajectory from peatland fires in South Sumatra in October 2015. We used the HYPLIT model to generate haze trajectories with hotspot sequences as the initial points of trajectory. The simulation was performed using the package Opentraj which is available in R. The experimental results show that haze from fires in South Sumatra affected not only people in this province but also in neighboring provinces namely Jambi and Riau. Haze also reached West Sumatra, North Sumatra and moved to Indian Ocean. K-Means clustering on the dataset with 119,808 haze positions results two dense clusters with the average of height 65.06 m AGL and 22.21 m AGL. Dense clusters cover about 27% to 29% members of all haze trajectories. Haze in two dense clusters are located at low height near initial points of the simulation which is dangerous for human health.