Mostafa et al. (2026) Coupling SVM classification and CA-Markov simulation: four decades of land cover evolution and 2056 projection in Egypt’s northern Nile Delta — dual-dimensional analysis of aquaculture expansion and agricultural land loss

Identification

• Journal: Frontiers in Environmental Science
• Year: 2026
• Date: 2026-09-21
• Authors: Wael Mostafa, Ali Abdullah Aldosari, Ahmed M. S. Kheir, Mohamed R. Abdelzaher, Shuisen Chen, Mohamed M. El-Khalafy
• DOI: 10.3389/fenvs.2026.1900117

Research Groups

• Geography Department, Faculty of Arts, Kafrelsheikh University, Kafrelsheikh, Egypt
• Geography Department, College of Humanities and Social Sciences, King Saud University, Riyadh, Saudi Arabia
• Agricultural Research Center, Soils, Water and Environment Research Institute, Giza, Egypt
• International Center for Agricultural Research in the Dry Areas (ICARDA), Maadi, Egypt
• Joint Laboratory on Low-Carbon Digital Monitoring, Guangdong Institute of Carbon Neutrality (Shaoguan), Shaoguan, China
• Botany and Microbiology Department, Faculty of Science, Kafrelsheikh University, Kafr El-Sheikh, Egypt

Short Summary

This study uses a combination of Support Vector Machine (SVM) classification and Cellular Automata-Markov (CA-Markov) simulation to model land use and land cover change in El-Hamoul District, Egypt's northern Nile Delta region. The results show that between 1986 and 2025, 192.5 km^2 (29.97% of the district) underwent net transition, with fish farms expanding by 359%, agricultural lands declining by 34.1 km^2, and urban areas growing by 189%.

Objective

• To map LULC accurately for five time intervals (1986, 1996, 2006, 2016, and 2025) through SVM classification using Landsat imagery via Google Earth Engine.
• To investigate the spatio-temporal trajectory of aquaculture-based land system transformation along with expansion to Lake Burullus area over six different classes of land cover.
• To estimate the point of exhaustion of bare land class and calibrate CA-Markov model through transition probability matrices of 1986–2025.
• To simulate future LULC scenarios up to 2056.

Study Configuration

• Spatial Scale: El-Hamoul District, Egypt's northern Nile Delta region (14% of Kafr El-Sheikh Governorate)
• Temporal Scale: 1986-2056

Methodology and Data

• Models used: Support Vector Machine (SVM) classification and Cellular Automata-Markov (CA-Markov) simulation
• Data sources: Landsat 5, 7, 8, and 9 imagery from five benchmark years (1986–2025)

Main Results

• Between 1986 and 2025, 192.5 km^2 (29.97% of the district) underwent net transition.
• Fish farms expanded by 359%, converting agricultural lands and bare lands.
• Agricultural lands declined by 34.1 km^2 (-7.0%), while urban areas grew by 189%.
• Bare lands were nearly eliminated, removing spatial buffer that once absorbed conversion pressures.

Contributions

• This study provides the first district-scale investigation on LULC dynamics and future simulation of El-Hamoul using SVM classification and CA-Markov spatial modeling.
• The results highlight the urgent need for spatially explicit land-use planning to balance aquaculture economics with wetland conservation and agricultural sustainability in one of Egypt's most ecologically sensitive deltas.

Funding

• This research was funded by [insert funding agency/program/project codes].

Citation

@article{Mostafa2026Coupling,
  author = {Mostafa, Wael and Aldosari, Ali Abdullah and Kheir, Ahmed M. S. and Abdelzaher, Mohamed R. and Chen, Shuisen and El-Khalafy, Mohamed M.},
  title = {Coupling SVM classification and CA-Markov simulation: four decades of land cover evolution and 2056 projection in Egypt’s northern Nile Delta — dual-dimensional analysis of aquaculture expansion and agricultural land loss},
  journal = {Frontiers in Environmental Science},
  year = {2026},
  doi = {10.3389/fenvs.2026.1900117},
  url = {https://doi.org/10.3389/fenvs.2026.1900117}
}