Penerapan Model Statistik Multivariat dalam Studi Kerawanan Longsorlahan di Daerah Aliran Sungai Kodil, Jawa Tengah

Elok Surya Pratiwi, Danang Sri Hadmoko

Abstract

Tingginya jumlah kerugian yang ditimbulkan dari bencana longsorlahan di DAS Kodil Jawa Tengah membuktikan bahwa upaya manajemen bencana di daerah tersebut masih lemah. Penelitian ini dilakukan sebagai salah satu upaya mengurangi resiko bencana yakni dengan cara merekonstruksi peta kerawanan longsorlahan. Peta kerawanan longsorlahan direkonstruksi menggunakan salah satu metode indirect mapping untuk mengurangi tingkat subjektivitas yang tinggi  yakni metode Logistic Regression Model. Hasil peta kerawanan longsorlahan DAS Kodil menunjukkan bahwa 11,9% luas area termasuk dalam kategori daerah kerawanan tinggi dan 5,8% luas area termasuk dalam kategori daerah kerawanan sangat tinggi. Tingkat akurasi peta kerawanan longsorlahan yang dihasilkan adalah sebesar 69,5%, dimana variabel yang paling berpengaruh terhadap kejadian longsorlahan di daerah penelitian adalah jarak buffer jalan.


High numbers of loss from the previous landslide hazard proved that hazard management in Kodil Watershed Central of Java was still low. One of the way to decrease the hazard risk occurrence are with analyzing dynamic of landslide distribution observed from building landslide susceptibility map. Landslide susceptibility map was built using one of indirect method that is Logistic Regression Model (LRM) to reduce the subjectivity result. The result of landslide susceptibility map reveals that 11,9% of area in Kodil Watershed have high susceptibility rate, even more 5,8% area have very high susceptibility rate. Landslide susceptibility map built by LRM reached an accuracy as high as 69,5%, where variable distance from road performs as the biggest influence for landslide occurrence in that area.

Keywords

kerawanan longsorlahan; logistic regression model; dinamika temporal longsorlahan

Full Text:

PDF

Refbacks

  • There are currently no refbacks.