| dc.description.abstract | Recent forest and land wildfires in Indonesia not only result in ecosystem loss but also economic loss and health and pollution effect. The other hand, fire weather is believed to be an important aspects for fire occurrences. Fire weather directly affects fuel temperature, accelerating its easiness to burn. Thus, it is important to clarify the effects of air temperature to forest fire incidence. A statistical analysis and forecasting for air temperature is needed to predict the air temperature condition in the future. Thisstudyused AutoregressiveIntegratedMovingAverage(ARIMA) model for air temperature forecasting in Palangkaraya, Central Kalimantan. Palangkaraya has become the fire dense areas due to several high fire occurrences in recent years. The result showedthat theAR(5) model could be usedfor daily airtemperaturemodeling foreight forecasting days with aMAPEvalue of3.11%. ARIMA(0,1,1) could be usedfor weekly airtemperatureforecastingwith aMAPEvalue of2.1%. Therefore these models could be used for air temperatureforecasting, althoughthey produced relativelyconstantvalueof temperature forecastingon27oCand28oC. For the monthly temperature, model MA (1) could be used with a MAPE value of 2.08%. Forecasting for the next 6 months also tend to resulta constant temperature at 26.8 oC, but MA (1) has lowest MAPE value, indicating that MA(1) is the best model for predicting the air temperature of Palangkaraya in the future. | en |
