Pemodelan Curah Hujan untuk Analisis Risiko Usaha Tani Padi di Tabanan

Abstract

Indonesia merupakan negara agraris yang bergantung pada kondisi iklim, khususnya curah hujan, dalam mendukung keberhasilan sektor pertanian. Risiko terjadinya banjir dan kekeringan akibat curah hujan yang ekstrem dan sulit diprediksi dapat mengancam keberlangsungan usaha pertanian. Penelitian ini bertujuan untuk menerapkan metode Seasonal Autoregressive Integrated Moving Average (SARIMA) dalam meramalkan curah hujan serta menganalisis risiko usaha tani padi di Kabupaten Tabanan. Hasil penelitian menunjukkan bahwa model terbaik untuk memodelkan data curah hujan adalah SARIMA(3,0,0)(0,1,1)24, dengan nilai mean absolute percentage error sebesar 9.0497% untuk data training dan 9.6964% untuk data testing. Risiko kegagalan panen akibat kekurangan air pada masa tanam pertama, kedua, dan ketiga masing-masing sebesar 1.43%, 37.5%, dan 0.84%. Sementara itu, risiko kegagalan panen akibat kelebihan air pada masa tanam pertama, kedua, dan ketiga masing-masing sebesar 0.56%, 0.2%, dan 1.28%.

Indonesia is an agricultural country that heavily relies on climatic conditions, particularly rainfall, to support the success of its agricultural sector. The risks of floods and droughts caused by extreme and unpredictable rainfall can threaten the sustainability of farming activities. This study aims to apply the Seasonal Autoregressive Integrated Moving Average (SARIMA) method to forecast rainfall and analyze the risks associated with rice farming in Tabanan Regency. The results indicate that the best model for forecasting rainfall data is SARIMA(3,0,0)(0,1,1)24, with a mean absolute percentage error of 9.0497% for the training data and 9.6964% for the testing data. The risk of crop failure due to drought in the first, second, and third planting seasons is 1.43%, 37.5%, and 0.84%, respectively. Meanwhile, the risk of crop failure due to flooding in the first, second, and third planting seasons is 0.56%, 0.2%, and 1.28%, respectively.