Publication: Prediction of Greater Bangkok expansion using nighttime light image and CA-Markov Model
0
0
Issued Date
2025-01-17
Resource Type
Scopus ID
2-s2.0-85218428505
Journal Title
ESET 2024 - Conference Proceedings of the 2024 8th International Conference on E-Society, E-Education and E-Technology
Start Page
79
End Page
85
Rights Holder(s)
SCOPUS
Bibliographic Citation
ESET 2024 - Conference Proceedings of the 2024 8th International Conference on E-Society, E-Education and E-Technology (2025) , 79-85
Suggested Citation
Limgomonvilas T. Prediction of Greater Bangkok expansion using nighttime light image and CA-Markov Model. ESET 2024 - Conference Proceedings of the 2024 8th International Conference on E-Society, E-Education and E-Technology (2025) , 79-85. 85. doi:10.1145/3704217.3704234 Retrieved from: https://hdl.handle.net/20.500.14740/20815
Author(s)
Author's Affiliation
Corresponding Author(s)
Other Contributor(s)
Abstract
This study explores land-use changes in Greater Bangkok between 2013 and 2027 using nighttime light (NTL) data and CA-Markov modeling. NTL imagery from Suomi NPP VIIRS facilitates analysis of urban expansion trends. Unsupervised K-means classification categorizes land use into high-density urban, urban, and non-urban areas. Key findings reveal a continuous increase in high-density urban areas (0.08% annually) and a slight expansion of the urban areas (0.01% annually). Conversely, non-urban areas are decreasing (-0.02% annually), suggesting potential conversion to urban use. The CA-Markov model, informed by NTL data, predicts land-use changes for 2027. Predictions indicate a continuation of observed trends, with high-density urban areas experiencing the most significant growth. This study highlights the potential of NTL data and CA-Markov modeling for urban sprawl analysis and future land-use prediction. However, limitations include the spatial resolution of NTL data and the Markovian assumption of the CA-Markov model. Future research should incorporate additional NTL data, explore integration with other relevant datasets, and investigate alternative modeling approaches.