Identification of Agricultural Land Types in Abbas Plain Using Time Series Analysis of Sentinel 2 Satellite Imagery
Subject Areas : Environmental managment
Mahdi
Rezaei
1
(PhD Student, Remote Sensing and GIS, Faculty of Natural Resources and Environment, Science and Research Branch, Islamic Azad University, Tehran, Iran.)
Hosein
Agha mhammadi zanjir abad
2
(Assistant Professor, Remote Sensing and GIS, Faculty of Natural Resources and Environment, Science and Research Branch, Islamic Azad University, Tehran, Iran. *(Corresponding author))
Zahra
Azizi
3
(Assistant Professor, Remote Sensing and GIS, Faculty of Natural Resources and Environment, Science and Research Branch, Islamic Azad University, Tehran, Iran.)
Alireza
Vafaee nejhad
4
(Associate Professor, Faculty of Civil, Water and Environmental Engineering, Shahid Beheshti University of Tehran, Tehran, Iran.)
Saeid
Behzadi
5
(Assistant Professor in Surveying Engineering, Faculty of Civil Engineering, Shahid Rajaee Teacher Training University, Tehran, Iran.)
Keywords: Time series, NDVI, Irrigated Lands, Image, Sentinel,
Abstract :
Background and Objective: Agricultural land is a crucial source of fresh water, with semi-deep and deep wells serving as the primary water supply for irrigated lands. Effective management and monitoring of these lands is essential for sustainable water consumption. However, identifying unlicensed agricultural wells among licensed ones through land visits can be challenging due to various factors such as the large number of lands, weather conditions, and difficult access to remote areas. Material and Methodology: Remote sensing and geographic information systems can provide a quick and highly accurate solution for discovering and monitoring these lands. In this study, we utilized the NDVI index and time series data set of Sentinel 2 images (bands 4 and 8) for March, April, and May to classify a portion of Abbas Dehlran plain's lands in the water year 2022 using the SVM method. We validated our results using cadastre data. Findings: Our classification accuracy based on cadastre maps was 98.4% for irrigated lands and 86.7% for rainfed lands, indicating high accuracy in identifying different types of agricultural land. Discussion and Conclusion: Our study demonstrates that time series algorithms applied to satellite images can effectively identify illegal irrigated lands along with their corresponding water supply wells by evaluating results based on existing land cadastres in the region. This approach can help improve management practices for sustainable water consumption in agricultural areas.
_||_