Komparasi Model Cellular Automata Dalam Memprediksi Perubahan Lahan Sawah Di Kabupaten Purworejo

Abstract

Penelitian ini bertujuan untuk mengkomparasikan akurasi metode ANN dan LR dalam memprediksi perubahan lahan sawah di Kabupaten Purworejo. Adapun data masukan yang dibutuhkan adalah peta lahan sawah tahun 2008, 2015 dan 2019 hasil interpretasi visual citra satelit resolusi tinggi dan faktor pendorong perubahan lahan sawah. Hasil penelitian menunjukkan bahwa model prediksi lahan sawah yang dibangun dari ANN dan LR secara umum memiliki akurasi yang sama-sama baik. Tetapi jika dilihat dari total nilai false alarm dan misses, model CA yang dibangun dari ANN lebih baik dari LR. Hasil penelitian ini juga menunjukkan bahwa dalam rentang tahun 2008 sampai 2019, luasan lahan sawah di Kabupaten Purworejo berkurang sekitar 194.01 Ha. Kata kunci: artificial neural network; cellular automata; logistic regression; perubahan penggunaan lahan; sawah. This research aims to compare the accuracy of the ANN and LR methods in predicting changes in paddy fields in Purworejo Regency. The input data required is a map of paddy fields in 2008, 2015 and 2019 as a result of visual interpretation of high-resolution satellite imagery and the driving factors for changes in paddy fields. The results showed that the paddy field prediction model built from ANN and LR generally has the same accuracy. But if it is seen from the total value of false alarms and misses, the CA model from ANN is better than LR. This study shows that from 2008 to 2019, the area of paddy fields in Purworejo Regency decreased around 194.01 Ha. Keywords: artificial neural network; cellular automata; land-use changes; logistic regression; paddy field.

Projected land cover changes in 2031 in Awangpone watershed, South Sulawesi, Indonesia

“…The driving factor is an independent variable in building a transition probability model, in previous studies the driving factors used are existing factors that will cause changes in land cover, in this case the distance factor of existing land to the surrounding land use [3]. Distance an influential factor in spatial dynamics, including changes in land cover.…”

“…Independent variables that have a high correlation value are assumed to have the same characteristics, so they can give rise to multicollinearity that can interfere with the regression process [7]. There is a factor that has a high correlation value or is above 0.7 then one of the factors can be eliminated to prevent multicollinearity [3]. The results of the calculation of Evaluating correlation can be seen in table 2.…”

Modeling Land Use/Land Cover Change in Berau Pantai Forests, Berau Regency, East Kalimantan Province

“…Visual classification has better accuracy for classifying land cover than digital classification (Kosasih et al 2019). This is because visual interpretation can distinguish between objects in an image based on human judgment, which can lead to better interpretation of complex objects, but visual interpretation is not efficient in terms of processing time (Fariz et al 2020).…”

Analysis of Land Cover Change and Projection of Settlement Land in Sepaku District, North Penajam Paser Regency

“…Land cover data for 2019 was obtained from the visual interpretation of Landsat-8 satellite imagery. The mapping process is carried out by updating the 2009 land cover map based on the appearance of satellite imagery in 2019 (Fariz et al, 2020).…”

“…Therefore, this research needs to be developed based on the limitations that we have conveyed. In addition, land change studies should also be analyzed projectionally using Cellular Automata analysis, such as research by Nurhidayati et al (2017), Fariz et al (2020) and Permatasari et al (2021), so that land cover changes can be seen spatially.…”