Early detection of drought impact on rice paddies in Indonesia by means of Niño 3.4 index

Abstract

El Nino events have been frequently marked by drought occurrences with severe consequences for agricultural production in Indonesia. Paddy drought occurs almost every year and extends during El Nino phenomena. The Nino 3.4 index is commonly used as an important tool for managing a food security policy. However, there are no details regarding the impact of El Nino on drought-induced paddy damage. We developed the Paddy Drought Impact Index (PDII), which is the ratio of drought-induced paddy damaged area to the total paddy area planted in order to investigate the impact of drought on paddies among 335 districts in Indonesia. Unlike other agricultural drought indices, this index represents real-life percentage of drought-induced paddy damage to indicate each district’s relative severity to drought, which can be easily understood by practical users. The connection between the Nino 3.4 index and PDII was assessed using cross correlation analysis. Scatter plots of best lag time Nino 3.4 index against PDII were examined. The findings show that with 2 months lag of Nino 3.4 prior to PDII, March and June Nino 3.4 indices can be used to predict May–July and August–October PDII, respectively. Critical thresholds of the March Nino 3.4 index were found to range from 0.0 to 0.5 °C, which is associated with a 0.57 probability of weak El Nino occurrence during the subsequent 5 months. On the other hand, a higher probability of 0.67 for occurrences of moderate El Nino is associated with the critical thresholds of June Nino 3.4 index, which ranges from 0.5–1.0 °C. This study has found that the potential impact of drought due to the weak and moderate El Nino occurrences in Indonesia is such that yields are reduced by about 40 % in average. We also found that the most drought-prone areas are located in West Java for both May–July and August–October and in South Sulawesi for August–October.