Long-term hydrometeorological time-series analysis over the central highlands of West Papua
In one sentence. A NARX neural network predicts the 12-month drought index SPI-12 in West Papua from the MEI ENSO index, capturing ENSO-linked rainfall patterns.
Abstract
This article introduces an innovative data-driven approach to examining the long-term temporal rainfall patterns in the central highlands of West Papua, Indonesia. Through the utilization of wavelet transforms, we identified signs of a negative temporal correlation between the El Niño-Southern Oscillation (ENSO) and the 12-month Standardized Precipitation Index (SPI-12).Building upon this cause-and-effect relationship, we employed dynamic causality modeling, utilizing the Nonlinear Autoregressive with Exogenous input (NARX) model, to predict SPI-12. In this predictive framework, the Multivariate ENSO Index (MEI) was employed as an attribute variable. Consequently, this dynamic neural network model effectively captured common patterns within the SPI-12 time series.The implications of this study extend significantly to the advancement of data-driven precipitation models for regions characterized by intricate topography within the Indonesian Maritime Continent (IMC).
Labels
- Topic
- Sub-seasonal to seasonal, Climate analysis & attribution
- Type
- New model or method
- Method
- RNN / LSTM / ConvLSTM
- Time scale
- Seasonal to annual
- Domain
- Regional
- Data
- Station / in-situ
- Status
- Research
- Identifiers
- DOI:10.31223/x50k7x OpenAlex:W3150692956
- Added
- 2026-10-22
Labels and summary are generated by Claude and reviewed weekly. Wrong? Edit this entry.