نوع مقاله : مقاله پژوهشی
نویسندگان
1 گروه مهندسی بیوسیستم، دانشکده کشاورزی، دانشگاه شهید چمران اهواز، ایران
2 گروه مهندسی علوم خاک، دانشکده کشاورزی، دانشگاه شهید چمران اهواز، ایران
چکیده
کلیدواژهها
موضوعات
عنوان مقاله [English]
نویسندگان [English]
Land use changes, as a major challenge in the field of environment and sustainable development, have significant effects on the carbon cycle and greenhouse gas emissions, especially carbon dioxide. Monitoring and analyzing the trend of land use and vegetation changes, as a key indicator for assessing the environmental status, is of particular importance. This study aimed to investigate the impact of land use changes on CO₂ emissions in Ahvaz city during the years 2014 to 2020. In this research, images extracted from Google Earth and deep learning methods were used for land use classification. Data related to CO₂ emissions were extracted and analyzed from the Giovanni database. The overall classification accuracy for the years 2014 and 2020 was 97.55 and 98.86 percent, respectively, and the kappa coefficient for the years 2014 and 2020 was 94.36 and 96.06 percent, respectively. The results showed that agricultural land (77.47%) and man-made land (55.89%) increased significantly during the study period, while barren and rocky land decreased. Analysis of CO₂ concentration also indicated a 28% increase in 2020 compared to 2014. A moderate negative correlation was observed between barren land and CO₂ concentration. On the other hand, agricultural land (0.75) and man-made areas (0.43) showed a positive correlation with CO₂ concentration, which is likely due to human activities and energy consumption.
کلیدواژهها [English]
Land use changes are considered one of the serious environmental and sustainable development challenges in the 21st century. These changes result from population growth, urban development, and increasing demand for natural resources, negatively impacting climate and human health. Identifying trends in land use and vegetation changes as indicators of environmental status is one of the key approaches in environmental studies [9]. Undoubtedly, one of the most important consequences of land use changes is its impact on the carbon cycle and greenhouse gas emissions, especially CO₂. This research examines the impact of land use changes on CO2 emissions in Ahvaz County using Google Earth imagery and deep learning algorithms. The main objective of this study is to qualitatively and quantitatively analyze land use changes and their relationship with CO2 emissions. The results of this study can assist in sustainable land management and reducing greenhouse gas emissions, providing tools for policymakers.
This study investigates the impact of land use changes on CO2 emissions in Ahvaz County, the capital of Khuzestan Province, from 2014 to 2020. The general research stages include extracting and preprocessing images from Google Earth, classifying land use using deep learning algorithms, and assessing the accuracy of the algorithms. For analysis, three images from February 2014 and three images from March 2020 were selected. The deep learning method, particularly convolutional neural networks (CNN), was utilized for land use classification. To evaluate the classification accuracy, statistical indices and methods, such as Pearson correlation coefficient, were employed. CO2 data were extracted from NASA’s Giovanni portal, and changes in CO2 concentrations were analyzed in relation to land uses. Pearson correlation coefficient was used to examine the relationship between land use changes and carbon dioxide concentration.
The results indicate that the overall classification accuracy for 2014 was 97.55%, and for 2020, it reached 98.86%. In 2014, barren lands and pebble lands comprised 46.13% and 36.39% of the total area, respectively. These two land types were predominant due to prevailing climatic and geological conditions. Specifically, pebble formations appear in the southwestern and eastern regions, limiting vegetation growth conditions. Additionally, in 2020, barren lands constituted 40% and grassland lands 30.35% of the total area, revealing a significant decrease compared to 2014. In contrast, agricultural lands increased to 21.74%, and human-made lands grew to 6.75%. Agricultural and man-made lands accounted for 12.25 and 4.33 percent of the area of the Ahvaz county in 2014, respectively. Changes in land use particularly affect environmental dynamics and resource utilization. The application of deep learning techniques in this research is introduced as an effective tool for analyzing and planning resource management, sustainable development, and environmental protection. These results can serve as a foundation for future policymaking in land use development and natural resource conservation. In 2014, the average concentration of CO2 was 4.47e-08 ppm, with a standard deviation of 1.41e-09 ppm, while in 2020, these values increased to 6.34e-06 ppm and 1.74e-09 ppm, respectively. The increase in the maximum CO2 concentration was more pronounced than the average, reaching 80.53%, particularly around industrial centers and the metropolitan area of Ahvaz. The correlation between land use and CO₂ concentration indicates differing impacts of land use types; barren lands exhibit a moderate negative correlation, suggesting that CO₂ increases with the reduction of these areas. Agricultural lands demonstrated a strong positive correlation with a coefficient of 0.75 regarding the concentration of carbon dioxide.
The results of this study demonstrate that the deep learning method is effective in classifying agricultural lands based on Google Earth images, with classification accuracy around 98 percent. From 2014 to 2020, barren and pebble lands decreased by 14.75%, while agricultural and human-made lands increased by 77.47% and 55.89%, respectively. Pearson correlation analysis indicates a moderate negative correlation for barren lands and a strong positive correlation for agricultural lands concerning CO2 emissions. This study also notes that from 2014 to 2020, the concentration of CO2 increased by 28.34%, reflecting heightened industrial and agricultural activities and changes in emissions sources. This trend may be due to population growth, economic development, and changes in environmental policies. Overall, the conversion of barren and pebble lands into agricultural and human-made areas has led to an increase in CO₂ concentrations.
Conceptualization, K. Andekaeizadeh., A. Asakereh and S. Hohati.; methodology, K. Andekaeizadeh. and A. Asakereh; software, K.A. and S.H; validation, K. Andekaeizadeh., A. Asakereh and S. Hojati.; formal analysis, K.Andekaeizadeh and A. Asakereh.; investigation, K. Andekaeizadeh.; resources, K. Andekaeizadeh. and A. Asakereh; data curation, A. Asakereh.; writing—original draft preparation, K. Andekaeizadeh; writing—review and editing, A. Asakereh and S.Hojati; visualization, K. Andekaeizadeh.; supervision, A. Asakereh.; project administration, A. Asakereh.; funding acquisition, A. Asakereh. All authors have read and agreed to the published version of the manuscript. All authors contributed equally to the conceptualization of the article and writing of the original and subsequent drafts.
All data generated or analyzed during this study are available from the corresponding author on request
The authors express their gratitude to the Vice Chancellor for Research and Technology of Shahid Chamran University of Ahvaz, Iran, for providing financial support through the research grant (No. SCU.AA98.29747).
The authors avoided data fabrication, falsification, plagiarism, and misconduct.
The author declares no conflict of interest.
)