The Analyse of Oil Palm Productivity Prediction Using Anomaly of Sea Surface Temperature at Nino-3,4

Abstract

Yield variability of oil palm (fresh fruit bunch, FFB) is determined by interaction between genetics, ages, environmental (soil and climate), and technical culture factors. Climate factor may explain most of the yield variability if other factors are not very variable. One of the main climate factors affecting the yield of palm oil in Indonesia is rainfall. This study was conducted to evaluate impact of rainfall variability on production of palm oil in 26 big plantations in Sumatra Island with assumption that conditions of non-climatic factors such as varieties (genetic), technical culture is relatively homogenous. As many studies suggest that rainfall variability in Indonesia is strongly influenced by ENSO phenomenon, we analyse the potential use of ENSO index (based on anomaly of sea surface temperature/ASST at Nino-3,4) to predict palm oil yield in the plantations. The prediction models was developed using a number of steps. First is to remove the effect of age and seasonality of palm oil yield from the data series using non linear equation and Fourier regression (called as detrended yield). Second is to develop relationship between the anomaly of the detrended yield and anomaly of ASST in region 3,4. Third is to construct model to predict palm oil production in the plantation using stochastic spreadsheet based on equations developed in the previous steps. The result of analysis showed that the non-linear function can explain well the relationship between yield and age. The model could explain between 60% and 90% of the variability of palm oil production across the 26 plantations. Furthermore, the study suggests that the ENSO impact on yield variability was only significant in seven plantations located in areas that have south moonsonal rainfall type (South part of Sumatra such as part of Jambi, Lampung and South Sumatra). It was found that the ASST in region 3,4 is potential to be used for predicting production of palm oil in the seven plantations. The result of validation showed that there is no significant different between observed and predicted data (tvalue < ttable; α=1%).

Produktivitas kelapa sawit (tandan buah segar, TBS) merupakan fungsi dari faktor genetik, umur, lingkungan (tanah dan iklim), dan kultur teknis (seperti pemupukan). Genetik bahan tanaman dapat diasumsikan homogen dan kultur teknis dilakukan optimal, sehingga keragaman produksi dapat dijelaskan oleh kondisi iklim dan umur. Kondisi iklim yang paling mempengaruhi produktivitas tanaman kelapa sawit di Indonesia adalah curah hujan. Penelitian ini dilakukan untuk mengevaluasi pengaruh dari keragaman curah hujan terhadap produktivitas tanaman kelapa sawit pada 26 perkebunan di Pulau Sumatera dengan asumsi bahwa kondisi dari kondisi non-iklim seperti genetik tanaman dan kultur teknis relatif seragam. Beberapa penelitian menunjukkan bahwa keragaman curah hujan di Indonesia diperngaruhi kuat oleh fenomena ENSO, sehingga akan dianalisis potensi untuk pemanfaatan indeks ENSO (anomali SST di Nino-3,4) untuk memprediksi produktivitas tanaman kelapa sawit di perkebunan. Model prediksi disusun dalam beberapa tahap.Yang pertama yaitu untuk menghilangkan pengaruh umur dan musiman produktivitas kelapa sawit dari data produktivitas dengan menggunakan persamaan non linier dan regresi Fourier.