بررسی کارایی مدل‌های آریما، جنگل تصادفی و اسپلاین مکعبی در پیش‌بینی شاخص (VCI) استان بصره عراق
محورهای موضوعی : کاربرد GIS&RS در برنامه ریزی
1 - دانشگاه تبریز
کلید واژه: شرایط پوشش گیاهی (VCI), الگوریتم جنگل تصادفی, اسپلاین مکعبی , آریما , تحلیل سری زمانی , استان بصره, عراق,
چکیده مقاله :
یکی از مهمترین چالشهای اساسی که بشر با آن روبرو است تغییر اقلیم و عواقب حاصل از آن مانند خشکسالی می باشد. شاخص های متعددی را برای پایش خشکسالی بکار گرفته اند که از آن جمله می توان به شاخص شرایط پوشش گیاهی (VCI)اشاره کرد. مساله اساسی در این راستا عدم توجه به دقت روش های مورد استفاده در پیش بینی تغییرات آتی شاخص در زمان می باشد. از اینرو هدف پژوهش حاضر ضمن بررسی روند تغییرات شاخص شرایط پوشش گیاهی (VCI) در استان بصره عراق برای بازه زمانی 2005 تا 2022 ، مقایسه سه روش اساسی در پیش بینی شامل تحلیل سری زمانی مبتنی بر آریما، اسپلاین مکعبی و الگوریم جنگل تصادفی در یادگیری ماشین بود. داده های مربوط به شاخص مورد پژوهش از سامانه گوگل ارث انجین استخراج شد و برای تجزیه و تحلیل داده ها از محیط نرم افزاری R و Spss و ArcGis استفاده شد. یافته های پژوهش نشان دادند که روند عمومی شاخص شرایط پوشش گیاهی (VCI)در استان بصره روندی کاهشی بوده و این شاخص کاهش چشمگیری را در سطح استان بصره علی الخصوص در مناطق جنوب شرقی این استان داشته است. همچنین نتایج این پژوهش نشان داد که داده های پیش بینی شده با داده های اندازه گیری شده از سنجده مادیس برای سال 2023 تفاوت معنا داری را دارد. نتیجه کلی پژوهش کارائی نسبی روشهای پیش بینی در شاخص های نظیر شرایط پوشش گیاهی (VCI)است.
One of the most significant challenges humanity faces is climate change and its resulting consequences, such as drought. Various indices are used to monitor drought, among which the Vegetation Condition Index (VCI) stands out. A crucial issue in this regard is the lack of attention to the accuracy of methods used for predicting future changes in the index over time. Therefore, the aim of this study is to analyze the trend of VCI changes in Basra province, Iraq, for the period from 2005 to 2022, and to compare three fundamental methods for prediction, including ARIMA-based time series analysis, cubic splines, and the Random Forest algorithm in machine learning. The data related to the studied index were extracted from the Google Earth Engine, and data analysis was conducted using R, SPSS, and ArcGIS software environments. The findings of the study indicated that the general trend of the VCI in Basra province is decreasing, and this index has significantly declined across the province, especially in the southeastern regions. Additionally, the results showed that the predicted data significantly differ from the measured data from the MODIS sensor for the year 2023. The overall conclusion of the study is the relative efficiency of prediction methods for indices such as VCI.
1. Abbasi, A., & Etemadi, H. (2023). Monitoring Drought in Bushehr Province Based on SPI and VCI Indices Using MODIS Sensor Images. Geographical
Space Quarterly, 82(Summer), pp. 179-198. DOI:10.52547/GeoSpa.23.2.179 (In Persian) 2. Alito, K. T., & Kerebih, M. S. (2024). Spatio-temporal assessment of agricultural drought using remote sensing and ground-based data indices in
the Northern Ethiopian Highland. Journal of Hydrology: Regional Studies, 52, 101700. https://doi.org/10.1016/j.ejrh.2024.101700 3. Alkaraki, K. F., & Hazaymeh, K. (2023). A comprehensive remote sensing-based Agriculture Drought Condition Indicator (CADCI) using machine
learning. Environmental Challenges, 11, 100699. https://doi.org/10.1016/j.envc.2023.100699 4. Arkhi, S., Barzegar Savasari, M., & Emadeddin, S. (2022). Evaluation of Remote Sensing-Derived Indices for Drought Assessment Using MODIS Images (Case Study: Qom, Isfahan, Chaharmahal and Bakhtiari, and Markazi Provinces). Journal of Geography and Environmental Hazards, No. 43, pp.
189-224. (DOI): 10.22067/geoeh.2021.72253.1102 (In Persian) 5. Cai, S., Zuo, D., Wang, H., Xu, Z., Wang, G., & Yang, H. (2023). Assessment of agricultural drought based on multi-source remote sensing data in a
major grain producing area of Northwest China. Agricultural Water Management, 278, 108142. https://doi.org/10.1016/j.agwat.2023.108142 6. Dutta, D., Kundu, A., Patel, N. R., Saha, S. K., & Siddiqui, A. R. (2015). Assessment of agricultural drought in Rajasthan (India) using remote sensing derived Vegetation Condition Index (VCI) and Standardized Precipitation Index (SPI). The Egyptian Journal of Remote Sensing and Space Science, 18(1),
53-63. https://doi.org/10.1016/j.ejrs.2015.03.006 7. Esmaeili, H., Mirmosavi, S. H., & Soheili, E. (2021). Time Series Analysis of Agricultural Drought in Darab County Using Remote Sensing and Google Earth Engine. Journal of Geography and Environmental Hazards, No. 40, pp. 175-192. 10.22067/geoeh.2021.69186.1029 (In Persian) 8. Fakhar, M. S., & Kaviani, A. (2023). Drought Assessment and Monitoring in Qazvin Plain Using MODIS-Based Indices in Google Earth Engine. Journal
of Irrigation and Drainage of Iran, 17(6), pp. 1089-1103. (In Persian) 9. Fakhar, M. S., & Nazari, B. (2024). Drought Analysis in Iran: Monitoring and Evaluating Spatial and Temporal Characteristics Using MODIS Indices.
Journal of Drought and Climate Change Research, 2(1), pp. 39-57. https://doi.org/10.22077/jdcr.2024.7011.1050 (In Persian) 10. Heydari Alamdarloo, E., Dehghan Rahimabadi, P., Khosravi, H., Rafiei, J., & Baraabadi, H. (2024). Probability Vulnerability Index (PVVI): A Method for Determining Desertification Risk. Journal of Remote Sensing and GIS in Natural Resources, 15(1), 1-19. https://sanad.iau.ir/fa/Article/902947
doi:10.30495/girs.2022.692984 (In Persian) 11. Luong, V. V., & Bui, D. H. (2023). Determination of the most suitable indicator area and remote-sensing-based indices for early yield warning for
winter–spring rice in the Central Highlands, Vietnam. Journal of Applied Remote Sensing, 17(1), 014504. https://doi.org/10.1117/1.JRS.17.014504 12. Malmir, M., Shayesteh, K., & Pazhoohan, I. (2024). Evaluation of Agricultural Drought Using Remote Sensing Data (Case Study: Tuyserkan County).
Journal of Agricultural Ecology, 16(3), pp. 513-531. (DOI): 10.22067/agry.2024.85155.1173 (In Persian) 13. Martín-Sotoca, J. J., Sanz, E., ..., & Tarquis, A. (2024). Relationship between vegetation and soil moisture anomalies based on remote sensing data: A
semiarid rangeland case. Remote Sensing, Published 11 September 2024. https://doi.org/10.3390/rs16183369 14. Meng, D., Bao, N., Tayier, K., Li, Q., & Yang, T. (2024). A remote sensing based index for assessing long-term ecological impact in arid mined land.
Environmental and Sustainability Indicators, 22, 100364. https://doi.org/10.1016/j.indic.2024.100364 15. Mirahasani, M. S., Mahini, A. S., Sefiyanian, A., Modarres, R., Jafari, R., & Mohammadi, J. (2017). Monitoring Regional Drought in the Zayandeh-Rud Watershed Based on Time Series Changes of MODIS VCI and SPI Indices. Journal of Geography and Environmental Hazards, No. 24, pp. 1-22.
10.22067/geo.v6i4.62601 (In Persian) 16. Mojarradi, B., Mirmiri, J., & Alizadeh, H. (2020). Evaluation of Vegetation Condition Index (VCI) Using the Modified Standardized Precipitation Index (MSPI) for Drought Monitoring and Zoning. Journal of Watershed Engineering and Management, 12(3), pp. 725-736.
https://doi.org/10.22092/ijwmse.2019.116643.1402 (In Persian) 17. Mokarram, M., & Pham, T. M. (2022). CA-Markov model application to predict crop yield using remote sensing indices. Ecological Indicators, 139,
108952. https://doi.org/10.1016/j.ecolind.2022.108952 18. Nazari-Pour, H., Karimi, Z., & Sedaghat, M. (2016). Evaluation of Hydro-Meteorological Drought Based on the Composite Drought Index and Its Prediction Using Markov Chain in the Sarbaz River Basin (Southeast Iran). Water and Soil Science (Isfahan University of Technology), 20(75), 151-169.
https://doi.org/10.18869/acadpub.jstnar.20.75.151 (In Persian) 19. Neghaban, S., & Makram, M. (2022). Investigation and Prediction of Drought Impacts on Maharloo Lake and Its Surrounding Land Uses Using
Remote Sensing. Journal of Climate Change Research, No. 10, pp. 71-81. 10.30488/ccr.2022.354398.1084 (In Persian) 20. Rezaei Moghaddam, M. H., Valizadeh Kamran, K., Rostamzadeh, H., & Rezaei, A. (2012). Evaluation of the Efficiency of MODIS Sensor Data in Estimating Drought (Case Study: Urmia Lake Watershed). Geography and Environmental Sustainability, 2(4), 37-52.
https://sanad.iau.ir/fa/Article/902833 doi:10.30495/girs.2021.680597 (In Persian) 21. Saati Zarei, S., & Ataieyan, B. (2021). The Impact of Rangeland Fires on Soil Organic Carbon Changes Using Remote Sensing-Based Indices. Journal
of Remote Sensing and GIS in Natural Resources, 12(3), 82-100. https://sanad.iau.ir/fa/Article/902833 doi:10.30495/girs.2021.680597 (In Persian) 22. Shabani, M. (2023). Evaluation of Remote Sensing-Based Indices in Monitoring Drought in Neyriz County. Journal of Remote Sensing and GIS in
Natural Resources, 13(4), 131-147. https://sanad.iau.ir/fa/Article/902907 doi:10.30495/girs.2022.690925 (In Persian) 23. Shamsipour, A. A., Alavipanah, S. K., & Mohammadi, H. (2010). Evaluation of the Efficiency of Vegetation and Thermal Indices from NOAA-AVHRR
Satellite in Drought Analysis of Kashan Region. Iranian Journal of Range and Desert Research, 17(3), [Issue 40, Fall 2010]. (In Persian) 24. Soleimani, K., Pourghasemi, H. R., & Alidadgan, F. (2019). Comparison of Shannon Entropy and Random Forest Data Mining Techniques in
Groundwater Potential Mapping of Jahrom. Journal of Desert Ecosystem Engineering, (24), 37-48. https://doi.org/10.22052/deej.2018.7.24.25 (In Persian) 25. Talebi, A., Goudarzi, S., & Pourghasemi, H. R. (2018). Assessing the Possibility of Landslide Hazard Mapping Using Random Forest Algorithm (Case
Study: Sardarabad Watershed, Lorestan Province). Journal of Environmental Hazards, 7(16), 45-64. https://doi.org/10.22111/jneh.2017.3213 (In Persian) 26. Wijesinghe, D. C., Withanage, N. C., Mishra, P. K., Ranagalage, M., Abdelrahman, K., & Fnais, M. S. (2024). An application of the remote sensing derived indices for drought monitoring in a dry zone district, in tropical island. Ecological Indicators, 167, 112681.
https://doi.org/10.1016/j.ecolind.2024.112681 27. Wongsai, N., Wongsai, S., & Huete, A. R. (2017). Annual seasonality extraction using the cubic spline function and decadal trend in temporal
daytime MODIS LST data. Remote Sensing, 9(1254). https://doi.org/10.3390/rs9121254 28. Xu, Z., Sun, H., Zhang, T., Xu, H., Wu, D., & Gao, J. (2024). The high spatial resolution Drought Response Index (HiDRI): An integrated framework for monitoring vegetation drought with remote sensing, deep learning, and spatiotemporal fusion. Remote Sensing of Environment, 312, 114324.
https://doi.org/10.1016/j.rse.2024.114324 29. Yin, G., He, W., Liu, X., Xia, Y., & Zhang, H. (2024). Wetting or greening? Probing the global trends in Vegetation Condition Index (VCI). International
Journal of Applied Earth Observation and Geoinformation, 129, 103822. https://doi.org/10.1016/j.jag.2024.103822 30. Luong, V. V., Bui, D. H., et al. (2023). Determination of the most suitable indicator area and remote-sensing-based indices for early yield warning
for winter–spring rice in the Central Highlands, Vietnam. Journal of Applied Remote Sensing, 17(1), 014504. https://doi.org/10.1117/1.JRS.17.014504