Pemodelan Kebakaran Hutan di Provinsi Riau Dengan Metode Geographically-Temporally Weighted Regression

Laksana, Syahrul Eka Adi (2018) Pemodelan Kebakaran Hutan di Provinsi Riau Dengan Metode Geographically-Temporally Weighted Regression. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

[img]
Preview
Text
06211440000034-Undergraduate_Theses.pdf - Accepted Version

Download (2MB) | Preview

Abstract

Kebakaran hutan merupakan masalah serius yang dihadapi pada permasalahan pencemaran udara masa kini. Intensitas kebakaran hutan di Indonesia kini semakin meningkat. Dampak yang ditimbulkan bukan hanya pencemaran udara, namun juga pada perekonomian masyarakat Indonesia. Dampak ekonomi kebakaran hutan tahun 2015 lebih besar jika dibanding dengan kebakaran hutan tahun 1997, padahal luas hutan yang terbakar lebih kecil. Pada penelitian ini dilakukan untuk mengetahui pengaruh faktor-faktor alam yang diduga mempengaruhi kebakaran hutan di Provinsi Riau. Metode yang digunakan adalah Geographically-Temporally Weighted Regression (GTWR). Metode ini digunakan karena mempertimbangkan heterogenitas spasial dan temporal. Hasil penelitian menyimpulkan bahwa kebakaran hutan Provinsi Riau kebanyakan terjadi di daerah Riau bagian utara yaitu Kabupaten Rokan Hilir, Bengkalis, dan Kepulauan Meranti dengan luas kebakaran rata-rata 160.000 hingga 300.000 Ha. Hasil prediksi kebakaran hutan dengan metode GTWR lebih akurat daripada metode regresi linier berganda karena nilai R2-nya lebih tinggi dan MSE yang lebih kecil. ================================================================================================================== Forest fires are a serious problem in contemporary area pollution. Now, the intensity of forest fires in Indonesia is increasing. The impact is not only air pollution, but also on the economy of Indonesian society. The impact of the forest fires disaster in 2015 is greater than 1997 forest fires, whereas the forest area is smaller. The aim of this research is determining the effect of natural factors that alleged the affect of forest fires in Riau Province. One method used is Geographically-Temporally Weighted Regression (GTWR) because this method can consider spatial and temporal heterogenity. The results of this research concluded that Riau forest fires mostly occur in the northern part of Riau Province, namely Rokan Hilir, Bengkalis, and Kepulauan Meranti Regency with an average fire area is 160.000 until 300.000 Hectare. The prediction result of forest fires by GTWR method is more accurate than multiple linear regression because its R2 value is higher and MSE value is smaller.

Item Type: Thesis (Undergraduate)
Uncontrolled Keywords: El-Nino, GTWR, Forest Fire, Riau
Subjects: G Geography. Anthropology. Recreation > G Geography (General) > G70.212 ArcGIS. Geographic information systems.
G Geography. Anthropology. Recreation > G Geography (General) > G70.217 Geospatial data
G Geography. Anthropology. Recreation > GE Environmental Sciences
Q Science > QA Mathematics > QA278.2 Regression Analysis. Logistic regression
Q Science > QA Mathematics > QA76.9.D343 Data mining
S Agriculture > SD Forestry
S Agriculture > SD Forestry > SD387.F52 Fire management
Divisions: Faculty of Mathematics, Computation, and Data Science > Statistics > 49201-(S1) Undergraduate Thesis
Depositing User: Syahrul Eka Adi Laksana
Date Deposited: 30 Jun 2021 06:27
Last Modified: 30 Jun 2021 06:27
URI: https://repository.its.ac.id/id/eprint/56605

Actions (login required)

View Item View Item