Sains Malaysiana 43(11)(2014): 1791–1800

Pengelompokan Kabur dalam Perantauan Kecenderungan Kemarau

di Semenanjung Malaysia

(Fuzzy Clustering for Regionalization of Drought Proneness in Peninsular Malaysia)

WAHIDAH SANUSI¹*, ABDUL AZIZ JEMAIN² & WAN ZAWIAH WAN ZIN²

¹Jurusan Matematika, Fakultas Matematika dan, Ilmu Pengetahuan Alam

Universitas Negeri Makassar, 90224, Parangtambung Makassar, Sulawesi Selatan

Indonesia

²Pusat Pengajian Sains Matematik, Fakulti Sains dan Teknologi, Universiti Kebangsaan Malaysia

43600 Bangi, Selangor, Malaysia

Received: 22 May 2013/Accepted: 2 April 2014

ABSTRAK

Dalam kajian ini, pendekatan pengelompokan kabur Gustafson-Kessel (GK) telah digunakan untuk mengelaskan 35 stesen hujan di Semenanjung Malaysia ke dalam rantau homogen. Pertama, algoritma pengkelasan kabur GK digunakan untuk mengenal pasti rantau awal. Kemudian, diuji keserasian dan kehomogenan rantau berkenaan. Akhir sekali, penyesuaian rantau dilakukan untuk mendapatkan rantau homogen. Hasil kajian mendapati 35 stesen hujan kajian boleh dibahagikan kepada enam rantau yang homogen. Rantau 1 meliputi bahagian barat laut dan utara Semenanjung Malaysia, rantau 2, 3 dan 4 meliputi bahagian barat, rantau 5 meliputi bahagian barat daya dan rantau 6 meliputi bahagian timur. Hasil kajian ini juga memperlihatkan bahawa berdasarkan nilai purata Indeks Kerpasan Piawai (SPI) skala masa satu bulan, rantau 2 lebih sering mengalami keadaan kemarau melampau. Walau bagaimanapun, berdasarkan SPI skala masa satu bulan, peristiwa kemarau terjadi secara rawak dalam semua rantau yang dianalisis, bahkan semua rantau tersebut pernah mengalami kejadian kemarau melampau dalam tempoh masa setahun. Hasil kajian ini turut menunjukkan bahawa pendekatan pengelompokan kabur Gustafson-Kessel boleh digunakan untuk membina rantau homogen.

Kata kunci: Indeks Kerpasan Piawai (SPI); pengelompokan kabur Gustafson-Kessel; perantauan; ujian kehomogenan; ujian keserasian

ABSTRACT

In this study, the Gustafson-Kessel (GK) fuzzy clustering method is used to classify the 35 rainfall stations in Peninsular Malaysia into homogeneous regions. First, the GK fuzzy clustering algorithm is applied to identify the initial region. The next step is to test the discordancy and homogeneity of corresponding region. Finally, adjustment of region is done to obtain the homogeneous region. The results showed that, for thirty five rainfall stations studied, these stations could be grouped into six homogeneous regions. The first region covers the northwestern and northern of Peninsular Malaysia, region 2, 3 and 4 cover the western, region 5 covers the southwestern and region 6 covers the eastern. The study also indicates that, based on the average Standardized Precipitation Index (SPI) value for one-month time scale, region 2 experiences more frequent extreme drought condition. However, based on the SPI, drought events randomly occurred in all regions, moreover these regions experience drought events within a year. The results also showed that GK fuzzy clustering method could be applied to construct a homogeneous region

.

Keywords: Discordancy test; Gustafson-Kessel fuzzy clustering; homogeneity test; regionalization; Standardized Precipitation Index (SPI)

REFERENCES

Babuska, R., van der Veen, P.J. & Kaymak, U. 2002. Improved covariance estimation for Gustafson-Kessel clustering. Fuzzy System 2: 1081-1085.

Burn, D.H. 1989. Cluster analysis as applied to regional flood frequency. Journal of Water Resources Planning and Management 115: 567-582.

Cancelliere, A., Mauro, G.D., Bonaccorso, B. & Rossi, G. 2007. Drought forecasting using the standardized precipitation index. Journal of Water Resources Management 21: 801-819.

Doring, C., Lesot, M.J. & Kruse, R. 2006. Data analysis with fuzzy clustering methods. Computational Statistics & Data Analysis 51: 192-214.

Hoel Le Capitaine & Carl Frelicot. 2009. A fuzzy modeling approach to cluster validity. FUZZ-IEEE. Korea. August 20-24. pp. 462-467.

Hosking, J.R.M. & Wallis, J.R. 1997. Regional Frequency Analysis. An Approach Based on L- Moment. UK: Cambridge University Press.

Kaymak, U. & Setnes, M. 2000. Extended Fuzzy Clustering Algorithms (No. ERS-2000-51-LIS). Erasmus Research Institute of Management (ERIM).

Lin, G.F. & Chen, L.H. 2006. Identification of homogeneous rantau for regional frequency analysis using the self-organizing map. Journal of Hydrology 324: 1-9.

Liu, H.C., Jeng, B.C., Yih, J.M. & Yu, Y.K. 2009. Fuzzy C-means algorithm based on standard mahalanobis distances. Proc. International Symposium on Information Processing. pp. 422-427.

Modarres, R. 2006. Regional precipitation climates of Iran. Journal of Hydrology (NZ) 45(1): 13-27.

Moreira, E.E., Paulo, A.A., Pereira, L.S. & Mexia, J.T. 2006. Analysis of SPI drought class transitions using loglinear models. Journal of Hydrology 331: 349-359.

Pal, N.R. & Bezdek, J.C. 1995. A cluster validity for the fuzzy-c-means model. IEEE Transactions on Fuzzy Systems 3(1): 370-378.

Paulhus, J.L.H. & Kohler, M.A. 1952. Interpolation of missing precipitation datas. Mon. Wea. Rev. 80: 129-133.

Sadri, S. & Burn, D.H. 2011. A fuzzy C-Means approach for regionalization using a bivariate homogeneity and discordancy approach. Journal of Hydrology 401: 231-239.

Sayang, M.D., Suhaila, J., Wan Zin, W.Z. & Jemain, A.A. 2010. Spatial trends of dry spells over Peninsular Malaysia during monsoon seasons. Theor. Appl. Climatol. 99: 357-371.

Soltani, S. & Modarres, R. 2006. Classification of spatio-temporal pattern of rainfall in Iran using a hierarchical and divisive cluster analysis. Journal of Spatial Hydrology 6(2)Fall: 1-12.

Suhaila, J., Sayang, M.D. & Jemain, A.A. 2008. Revised spatial weighting methods for estimation of missing rainfall data. Asia-Pacific Journal of Atmospheric Sciences 44(2): 93-104.

Xie, X.I. & Beni, G. 1991. A validity measure for fuzzy clustering. IEEE. Trans. on Pattern Analysis and Machine Intelligence 12: 841-847.

Zhang, Q., Xiao, M. & Chen, X. 2012. Regional evaluation of the meteorological drought characteristics across the Pearl River Basin, China. American Journal of Climate Change 1: 48-55.

*Corresponding author; email: w_sanusi@yahoo.com