ارزیابی تأثیر کشاورزی شهری در افزایش تاب‌آوری مناطق پرجمعیت شهری: رویکردی جامع مبتنی بر ذخیره و ترسیب کربن

ABSTRACT
The change in land use and land cover affected by the rapid expansion of urbanization and population growth in recent decades has created challenges such as carbon emissions in densely populated urban areas. Green infrastructures, such as urban agriculture, are significanting with the challenges above and promoting urban ecological resilience. In this research, land use and land cover maps of Tabriz metropolis with an area of 660.66 square kilometres for 2016 and 2023 were prepared using Sentinel-2 satellite images in Google Earth Engine. Using the InVEST model, evaluation and modelling of carbon storage in the studied area for the mentioned years was done. The simultaneous disclosure of the role of different uses with an emphasis on urban agriculture using remote sensing data and new technologies is one of the innovations of the mentioned research. The results show that the developed area of ​​Tabriz is facing an increase in the area of ​​human constructions, barren lands and green spaces and a decrease in urban agriculture, pastures and water areas in the period of 2016-2023. The InVEST model predicts the amount of total carbon content stored throughout the studied area in 2015 and 2012 as 2.43 and 2.27 million tons, respectively, which indicates a decrease in carbon storage due to land use change. Urban agricultural use accounts for the largest amount of carbon storage every two years and the reduction of the area of ​​this use is identified as the main cause of carbon emission. This research reveals the importance of urban agriculture in increasing ecological resilience and achieving the goals of sustainable urban development using remote sensing data and new technologies.
Extended Abstract
Introduction
Human activities exert significant anthropogenic environmental influences, significantly impacting ecosystem services (ESs) and capturing the interest of various scientific disciplines. Understanding the impact of land use/land cover (LULC) on the ecological environment is critical for the 21^st century. Enhancing green infrastructure is one approach, as photosynthesis helps absorb CO2 from the atmosphere. Consequently, diverse land-uses are crucial for carbon storage and reducing CO2 emissions. The rapid expansion of urbanization and population growth in recent decades has led to challenges such as increased carbon emissions in densely populated urban areas. ESs provided by ecological spaces and green infrastructures, such as urban agriculture (UA), play a crucial role in addressing these challenges and enhancing the ecological resilience of urban areas. Tabriz City, where UA activities are integrated with urbanization, was selected as a case study to model and map carbon storage using remote sensing (RS) data. Firstly, we need to generate LULC maps for various periods. This study aims to indicate how much carbon is stored on different types of land using Sentinel-2 MSI data from Google Earth Engine (GEE) and the Integrated Valuation of Ecosystem Services and Tradeoffs (InVEST) model. The significance of this study lies in its potential to contribute to urban planning strategies that prioritize carbon sequestration, aiding global efforts to mitigate the impact of human activities on the environment. 
 
Methodology
This research was conducted based on a multidisciplinary approach. In the first step, LULC maps for the city of Tabriz for 2016 and 2023 were generated using Sentinel-2 satellite images on the GEE web platform. The InVEST model was employed in the second step to create a comprehensive carbon storage map using reservoir data from similar studies and the generated LULC. In the final step, the carbon storage and sequestration for different land-uses, with an emphasis on UA, were analyzed and evaluated for the period 2016-2023. This research innovates by simultaneously revealing the role of different land-uses, particularly UA, through RS data (Sentinel) and new technologies (InVEST). Therefore, Sentinel-2 MSI, as a high-resolution satellite image, was initially used for LULC classification. A decisive reason for selecting Tabriz City in this study is to observe various land-uses and urbanization trends over the last decade across the study area. Regarding the GEE platform, six different land-uses were identified: built-up area, urban green space (UGS), UA, pasture, barren land, and water bodies. A spatiotemporal analysis was conducted using actual Google Earth images to validate the categorization's accuracy.
 
Results and discussion
The maps generated on the GEE platform were validated using the Kappa coefficient for 2016 and 2023. The validation process included analyzing 439 and 437 samples, respectively. According to the findings, the Kappa coefficient in 2016 was recorded at 0.87, with a total accuracy of 92.2%. In 2023, these values were 0.86 and 88.7%, respectively. According to LULC statistics, agricultural lands occupied the largest area in 2016, accounting for 34.9%. Built-up areas ranked next with 27.3%, while UA accounted for 17.6% of the total area. Pasture accounted for 12.9%, UGS for 5.9%, and water bodies for 1.4%. The highest variation in 2023 was observed in UA (-3.7%), followed by barren lands (+2.6%) and built-up areas (+1.5%). The InVEST model results indicated that the total carbon storage for the study area in 2016 and 2023 will be 2.43 million tons and 2.27 million tons, respectively. In 2016, the largest amount of carbon absorption was attributed to UA (1,071,198 tons), followed by pasture (585,477 tons), which accounted for 44% and 24% of the total carbon storage in the region, respectively. Due to their reduced size, these two land-uses will store less carbon in 2023 compared to 2016, making them significant factors influencing carbon emissions in the study area between 2016 and 2023. Despite its small area (5.9%), UGS stored about 532,258 tons of carbon in 2016, representing 22% of the total carbon storage. In 2023, this land-use will save about 92,210 tons of carbon compared to 2016 due to its increased size. Barren lands and human constructions have the lowest carbon absorption values in the studied area. While UA and UGS account for the highest carbon storage compared to their small size, UA is identified as the main cause of carbon emissions in the study area due to its area reduction.
 
Conclusion
The current research aimed to determine the carbon storage content in Tabriz City (East Azerbaijan Province) across six land-uses; it also assessed the carbon storage content under various land-uses based on results from the InVEST model and data derived from Sentinel-2 MSI. This study highlights the importance of RS in monitoring carbon storage across different land-uses. Integrating RS data with the InVEST model facilitates the efficient and cost-effective prediction of carbon storage across the study area. Our findings, supported by the Kappa index, indicate accurate LULC classification for the entire study area in 2016 and 2023. This research identified the spread of destructive human activities in densely populated urban areas as a significant factor affecting carbon storage and emissions throughout the study area. Future research should examine the impact of ecological and human LULC changes on the provision of ESs. This would promote effective land-use planning and ecosystem management for social well-being at the local level and enable the evaluation of proposed scenarios.
 
Funding
There is no funding support.
 
Authors’ Contribution
Conceptualization: B. Veisi Nabikandi, A. Hami; Data curation: B. Veisi Nabikandi, A. Hami; Formal analysis: B. Veisi Nabikandi; Investigation: B. Veisi Nabikandi; Methodology: B. Veisi Nabikandi; Project administration: B. Veisi Nabikandi, A. Hami; Software: B. Veisi Nabikandi; Supervision: A. Hami, K. Valizadeh Kamran; F. Emami Namin; Roles/Writing - original draft: B. Veisi Nabikandi; Writing - review & editing: B. Veisi Nabikandi.
 
Conflict of Interest
Authors declared no conflict of interest.
 
Acknowledgments
We are grateful to all the scientific consultants of this paper.