Abstract

There are several resulting arguments from the research done on climate variation in Indonesia stating that the observed affects are through various phenomena such as ENSO, monsoon, dipole mode event, and MJO. However, the magnitude of the effect varies for each region in Indonesia. This research aims to identify the relationship among the global climate features (GCFs) in the Nino3.4 (5°S–5°N, 120–170°W) with the local climate features (LCFs) in the Aceh regions which represented by: I(2–3°N, 95–98°E), II(3–4°N, 95–98°E), III(4–5°N, 95–98°E), and IV(5–6°N, 95–98°E) using canonical correlation analysis (CCA) in the ENSO phenomena. The analysis shows that global GCFs variations have strong correlation with LCFs variations with the correlation values, 0.893, 0.899, 0.900, and 0.901, respectively. The result show that when there is a global change in any feature of GCFs, the same change also appears in each feature of LCFs. The canonical loading shows that there are original variables which have strong correlation with the first canonical global variable (X1) with correlations 0.987, 0.969, 0.987, and 0.865,respectively, and the local wind (Y1) with correlations 0.974, 0.952, 0.979, and 0.845, respectively. All the other climate features have weak correlations with the first canonical variables. From the MANOVA, we can conclude that the climate features (wind, SST, SSTA, and SLP) affect climate changes in both study regions. Our results also reveal that LCFs are significantly affected in the Nino3.4 99.5% and in I, II, III, and IV for given correlations 99.8, 99.7, 99.6, and 99.5%, respectively.

Highlights

  • There are several resulting arguments from the research done on climate variation in Indonesia stating that the observed affects are through various phenomena such as ENSO, monsoon, dipole mode event, and MJO

  • From the MANOVA, we can conclude that the climate features affect climate changes in both study regions

  • Our results reveal that local climate features (LCFs) are significantly affected in the Nino3.4 99.5% and in I, II, III, and IV for given correlations 99.8, 99.7, 99.6, and 99.5%, respectively

Read more

Summary

Dua matriks hasil dalam analisis adalah

Di mana, Σxx dan Σ yy adalah dalam matriks varians kovarians, Σ xy ( Σ yx ) adalah antar matriks kovarians. Untuk menemukan kombinasi linier dari m variabel independen dan sebagai dasar penduga terhadap Σ xx , Σ yy , Σ xy dan Σyx (Dillon et al 1984). Dalam beberapa kasus variabel X dan Y dalam susunan matriks data dibakukan sehingga matriks varians-kovarians adalah matriks korelasi. Sampel sebagai dasar penduga terhadap bobot kanonik dinotasikan adan b (Dillon et al.1984). Gambar 1 Nilai korelasi spasial antara kejadian El-Nino dengan curah bobot kanonik dan matriks korelasi hujan wilayah Indonesia (sumber: Harijono 2008). S S 1 S S 1 yy yx xx xy di mana, Rxx dan R yy matriks korelasi dalam kumpulan, a j dan b j adalah vektor bobot kanonik dari kumpulan variabel X dan Y pada variabel kanonik ke-j. Untuk hipotesis nol bahwa variabel dependen p tidak berkorelasi dengan variabel independen m, H0 : yx 0 H1 : yx 0

Bartlett mendefinisikan
BAHAN DAN METODE
HASIL DAN PEMBAHASAN
Nilai korelasi antara fitur perubahan iklim di wilayah
Selanjutnya nilai korelasi antara fitur iklim di wilayah
Perubahan Iklim
Fitur iklim
UCAPAN TERIMA KASIH
DAFTAR PUSTAKA
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call