esc

Type at least two characters to start searching.

↑↓ select↵ openesc close

Share

Export Citation

APA
MLA
Chicago
Harvard
Vancouver
BIBTEX
RIS
Universitas Hasanuddin
Research output:Contribution to journal›Article›peer-review

Paddy field conversion prediction using CA-ANN method based on sentinel-2a imagery: A case study in Bajeng, Pallangga, and Somba Opu Districts, Gowa Regency

Saleh R.

Iop Conference Series Earth and Environmental Science

Published: 2026

Abstract

Gowa Regency is one of the areas that plays an important role in food security, especially rice production. However, rice production values have declined owing to several factors, such as urbanization, population growth, land use changes, and land conversion. This study aims to analyze the pattern of changes in paddy fields in Gowa Regency, specifically in Bajeng, Pallangga, and Somba Opu Districts, using Sentinel-2A imagery and evaluate the accuracy of predicting changes in paddy fields using the CA-ANN method. In this study, the Cellular Automata-Artificial Neural Network (CA-ANN) method was used, utilizing Sentinel-2A imagery as the basic data. Prediction maps of changes and areas of paddy and non-paddy fields for the year 2029 were obtained using a combination of true color, MSAVI, and NDVI images to conduct a comprehensive analysis. The Random Forest method revealed significant land use changes between 2019 and 2024, with a decrease in the paddy field area of 13.41%. The CA-ANN model was trained on 80% of land-change transition pixels, with the remaining 20% used as a held-out validation set. Model accuracy was verified through back-casting: the 2024 land use map was simulated from 2019 data and compared against the actual 2024 Random Forest classification, yielding an overall accuracy of 94.94% and a Kappa coefficient of 0.89. Predictions for 2029 show a continued decline, with a projected paddy field area of 5,355.85 ha. The CA-ANN method demonstrated good accuracy in predicting changes in paddy fields in the study area.

Other files and links

Fingerprint

Paddy fieldSciences
Normalized Difference Vegetation IndexSciences
Environmental scienceSciences
Agricultural engineeringSciences
PopulationSciences
Artificial neural networkSciences
Remote sensingSciences
MathematicsSciences
Field (mathematics)Sciences
StatisticsSciences
Random forestSciences
Land useSciences
Evaluation methodsSciences
Agricultural landSciences
Soil scienceSciences
Cohen's kappaSciences
Coefficient of determinationSciences
Correlation coefficientSciences
Field experimentSciences