Introduction
Plastic mulch is widely used in the cultivation of agricultural crops, especially summer crops, due to its potential to increase crop production and optimize water consumption. Plastic mulch protects agricultural products from adverse conditions such as cold, heat, heavy rain, wind, harmful insects, and diseases. Plastic mulch is mainly made of polyethylene, and due to their polymer-hydrocarbon nature, they are very difficult to decompose under natural conditions completely. However, plastic waste can cause soil pollution, reduced quality of agricultural fields, and decreased yield. Therefore, accurate and timely monitoring of the temporal and spatial distribution of plastic mulch in large areas is very important to evaluate its effects and prevent environmental problems. Conventional techniques for mapping plastic mulch, such as topographic measurements or photogrammetric techniques, are time consuming, costly, and require experienced personnel. Remote sensing is considered the only practical method for collecting this information on a large scale and an essential tool for managing the area under plastic mulch. High resolution satellite images are valuable new resources for accurate object detection and land data collection. Determine the valid threshold for detecting plastic mulches in fields using remote sensing technologies to manage the area under plastic cultivation and the possibility of estimating the amount of plastic mulch residues in the fields is the main goal of this research.
Method
In this study, 30 cucumber and watermelon fields with plastic mulch in Hamadan province were investigated. To identify the plastic mulch used for cucumber fields from mid-April to mid-July, and for watermelon fields from mid-June to mid-October Sentinel-2 satellite data from the Google Earth Engine (GEE) system were used. For this purpose, six Sentinel-2 satellite bands and 30-meter Scale mosaic images with cloud-free MOD09A1.006 Terra Surface Reflectance data were used. The bands used were Blue, Green, Red, NIR, SWIR1, SWIR2. Based on the difference in spectral characteristics between plastic mulch and other land objects (vegetation, water, bare soil, etc.), PMFI1, PMFI2, NDVI, NDWI, and PMI indices were extracted from Sentinel-2 satellite data to identify plastic mulch in the fields. After extracting PMFI1, PMFI2, NDVI, NDWI and PMI indices from Sentinel-2 satellite, the output data of GEE system was transferred to ArcGIS software. Since different indices have different reflectances of ground objects (soil, plastic mulch and vegetation cover). Therefore, according to the desired fields and crop planting date, spectral reflectances for ground objects in different months were examined and thresholds for plastic mulch range in different indices were determined. Then, after drawing boxplot of data, the interquartile range (range of first to third quartiles) of data was considered as plastic and data less and more than that were considered as areas without plastic mulch.
Results
The results of this evaluation showed that the evaluation indices had a significant impact on the accuracy of diagnosis, and the evaluation accuracy ranged from 11% to 97.5% depending on the indices examined. The highest accuracy of plastic mulch detection for watermelon fields was 97.5% and for cucumber fields was 91.8%, which was achieved using the NDWI indice. The NDWI index was more accurate than other plastic mulch detection indices in the study area due to its higher spectral reflectance in the NIR and the Green bands. According to the spectral bands of the Sentinel 2 satellite, the NDWI index allows some of the spectral characteristics of the underlying soil and the moisture in it to be transmitted more transparently, thus it was more accurate than other plastic cover detection indices. Among the other indices examined, the NDVI generally performed better than the spectral indices PMFI1, PMFI2, and PMI. The lowest accuracy of plastic mulch detection was for the PMI index, which was 11% for watermelon fields and 24.7% for cucumber fields.
Conclusions
The purpose of this research was to identify plastic mulch in cucumber and watermelon fields in Hamadan province. For this purpose, Sentinel-2 satellite images in the GEE cloud system were used. The results of this evaluation showed that the accuracy of plastic mulch detection varied from 11% to 97.5%, with the highest accuracy related to the NDWI indice for detecting plastic mulch in watermelon fields in June. NDWI index demonstrated the highest performance at the beginning of the growing season for both crops. However, in the mid-season, the spectral indices exhibited varying performance in detecting plastic depending on crop type: PMFI1 achieved higher accuracy for watermelon fields, while PMI performed better in cucumber fields. PMI index also had higher sensitivity to the type of crop cultivated than PMFI1 index. These findings underscore the high potential of remote sensing indices for accurately identifying plastic-mulched fields, which can significantly support decision-making in addressing environmental challenges and managing agricultural plastic waste.
Acknowledgements
The authors would like to acknowledge the Bu-Ali Sina University for finantcial support
Author Contributions
Majid Yousefian Hashemabadi: Data curation, Software, Writing
Hossein Bagherpour: Conceptualization, Methodology, Reviewing, Supervision
Mehraneh Khodamoradpour: Advisor, Conceptualization, Reviewing, Software

Data Availability Statement
Not applicable"
Ethical Considerations
This section states ethical approval details (e.g., Ethics Committee, ethical code) and confirms adherence to ethical standards, including avoidance of data fabrication, falsification, plagiarism, and misconduct.
Conflict of Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper
Funding Statement
The author(s) received no specific funding