STUDI POTENSI VARIABEL ANGIN ZONAL SEBAGAI PREDIKTOR UNTUK WILAYAH BENUA MARITIM (STUDI KASUS SUMATERA)

Authors

  • Agus Safril Sekolah Tinggi Meteorologi Klimatologi dan Geofisika
  • Rista Hernandi Virgianto Sekolah Tinggi Meteorologi Klimatologi dan Geofisika

Keywords:

zonal wind, Singular Value Decomposition (SVD), Statistical Downscaling

Abstract

Rainfall prediction using global model (Global Circulation Model) has a low resolution that can not provide local information. To obtain local-scale information, we need to apply a statistical downscaling method. Reliability of predictions is obtained by selecting predictor variables that are closely related to rainfall. The selection of predictor variables can be done based on spatial and temporal considerations that affecting predicted rainfall for annual and semi-annual cycle.  The correlation analysis between predictor and rainfall were done by SVD (Singular Value Decomposition) method. The dominant mode of the expansion coefficient of the time series obtained from SVD is analyzed using the wavelet transform. The correlation results show that U850 zonal wind variable is closely related to rainfall in Sumatra with correlation  of 0.77 and 92% of covariance. The U850 millibar wind variable can capture monsoon patterns (1 and 0.5 year) and the inter-annual pattern is ENSO (El Nino Southern Ocillation) with periods of 2, 4, and 12 years. Correlation between zonal wind (observation) around Sumatra with Nino index. 3.4 (>0.6). The U850 mb predictor capability is more reliable in the southern part of North Sumatra compared to southern Sumatra. In general, predictors can be used for the annual and semi-annual of rainfall  in  maritime region.

Published

2019-04-29

How to Cite

Safril, A., & Virgianto, R. H. (2019). STUDI POTENSI VARIABEL ANGIN ZONAL SEBAGAI PREDIKTOR UNTUK WILAYAH BENUA MARITIM (STUDI KASUS SUMATERA). Jurnal Meteorologi Klimatologi Dan Geofisika, 5(1), 26–39. Retrieved from https://jurnal.stmkg.ac.id/index.php/jmkg/article/view/65