Performance Comparison of Random Forest and Support Vector Machine Algorithms for Land Use Change Monitoring in the Samian Watershed (2015–2024) Using Remote Sensing Data in Google Earth Engine

Asghari Saraskanroud, Sayyad; Samadi Shalveh Alia, Fatemeh

Performance Comparison of Random Forest and Support Vector Machine Algorithms for Land Use Change Monitoring in the Samian Watershed (2015–2024) Using Remote Sensing Data in Google Earth Engine

Document Type : Origional Article

Authors

Sayyad Asghari Saraskanroud ¹

Fatemeh Samadi Shalveh Alia ²

¹ Professor , Department of Physical Geography, Faculty of Social Sciences, University of Mohaghegh Ardabili, Ardabil, Iran

² MSc Student Remote Sensing and Geographic Information Systems (GIS) Department of Physical Geography ، Faculty of Social Sciences، University of Mohaghegh Ardabili ، Ardabil ، Iran

Abstract

Background and Objective: Land use changes represent a critical environmental challenge, significantly impacting natural resources, ecosystems, and hydrological processes. This study aims to comparatively evaluate the performance of two machine learning algorithms—Random Forest (RF) and Support Vector Machine (SVM)—for land use mapping and analyzing temporal changes between 2015 and 2024 in the Samian Watershed, Ardabil Province, with an approximate area of 4236 km².
Methodology: Satellite imagery from Landsat 8 and 9, along with Sentinel-2, were utilized within the Google Earth Engine platform for land use classification. The RF and SVM classifiers were applied to produce land use maps consisting of eight classes: water, residential, irrigated agriculture, rainfed agriculture, snow, forest, dense rangeland, and sparse rangeland. Accuracy assessment was conducted using confusion matrices and related accuracy metrics. Global datasets (Dynamic World and GHSL) were employed for sampling and model training.
Results and Findings: Comparative analysis revealed that the RF algorithm outperformed SVM, achieving an overall accuracy and Kappa coefficient exceeding 99%. Significant land use changes were observed during the study period, including a notable increase in irrigated agriculture and residential areas, alongside a decrease in rainfed lands, snow cover, and surface water bodies. Overall, due to its high accuracy and stable performance, RF is recommended as the superior method for monitoring land use changes within big data environments such as Google Earth Engine.

Keywords

Land Use

Watershed

Samian

Landsat

Sentinel

Machine Learning

Subjects

geographical information system

Acharya, S., Hori, T., & Karki, S. (2023). Assessing the spatio-temporal impact of landuse landcover change on water yield dynamics of rapidly urbanizing Kathmandu valley watershed of Nepal. Journal of Hydrology: Regional Studies, 50, 101562.‏ https://doi.org/10.1016/j.ejrh.2023.101562

Aşci, E., & Genç, L. (2024). Determination of the Effects of Various Spectral Index Combinations on Seasonal Land Use and Land Cover (LULC) Changes Using Random Forest (RF) Classification Case Study: Southeast Marmara Region 2016-2020. Türkiye Uzaktan Algılama Dergisi, 6(1), 12-25.‏ https://doi.org/10.51489/tuzal.1395189

Asghari Saraskaneroud, S., & Sharifi Toularod. (2023). Detecting land use and forest area changes using remote sensing (Case study: Astara County). Land Management Journal, 11(2), 195–208. https://doi.org/10.22092/lmj.2023.362257.332 (in Persian)

Asghari Saraskaneroud, S., Aghayari, L., & Pirouzi, E. (2017). Land use changes and their effect on erosion using RS and GIS (Case study: Nir County). Remote Sensing and GIS in Natural Resources, 8(4), 49–62. (in Persian)

Asghari Saraskaneroud, S., Pasandeh, A. H., & Khavarian, [Initials]. (2024). Detecting land use changes during 2013–2024 using Landsat 8 image processing and impact analysis (Case study: Miandoab City). Journal of Urban and Regional Sustainable Development Studies, 5(4), 330–346. (in Persian) https://www.srds.ir/article_222189.html?lang=fa

Avcı, C., Budak, M., Yağmur, N., & Balçık, F. (2023). Comparison between random forest and support vector machine algorithms for LULC classification. International Journal of Engineering and Geosciences, 8(1), 1-10.‏ https://doi.org/10.26833/ijeg.987605

Brandolini, F., Domingo-Ribas, G., Zerboni, A., & Turner, S. (2021). A Google Earth Engine-enabled Python approach for the identification of anthropogenic palaeo-landscape features. Open Research Europe, 1, 22.‏ 10.12688/openreseurope.13135.2

Chandra, M. A., & Bedi, S. S. (2021). Survey on SVM and their application in image classification. International Journal of Information Technology, 13(5), 1-11.‏

Ebrahimi, B., Pasandi, M., & Neil Foroushan, H. (2023). Estimating land use changes under traditional irrigation in Khansar County. Water and Soil Science, 27(4), 77–91. https://doi.org/10.47176/jwss.27.4.27532 (in Persian)

Esfandiari Darabad, & Nezafat Takleh. (2025). Analysis of the impact of E and L indices on water sustainability and security in the Samian watershed. Journal of Urban and Regional Sustainable Development Studies, 6(4), 15–30. (in Persian) https://www.srds.ir/article_216659.html?lang=en

Esfandiari Darabadi, R., Rafiei, M., & Farzaneh, H. (2021). Land use/cover changes in Zarrineh River Basin and their impact on soil erosion using the WLC hydrogeomorphology model. Hydrogeomorphology, 8(29), 45–68. https://doi.org/10.22034/hyd.2021.47572.1600 (in Persian)

Farkhzadeh, B., Nour, G., & Aghabigi, A. (2023). Modeling land use change trends in Gahvareh region, Kermanshah Province. Journal of Rainwater Catchment Systems, 11(2), 48–62. https://doi.org/10.1001.1.24235970.1402.11.2.4.2 (in Persian)

Fazeli Farsani, A., Ghazavi, R., & Farzaneh, M. R. (2015). Performance assessment of land use classification algorithms using image fusion techniques (Case study: Beheshtabad sub-basin). Remote Sensing and GIS Applications in Natural Resources, 6(1), 91–105. (in Persian)

Ghodsi, K., Kheirkhah Zarkesh, M., Mir Masoud, Q., & Baghershameh, B. (2020). Comparing SVM and Random Forest methods for land use and crop mapping using multi-temporal Sentinel-2 images. Iranian Journal of Remote Sensing and GIS, 12(4), 73–92. https://doi.org/10.52547/gisj.12.4.73 (in Persian)

Golshan, M., Kavian Ataollah, A., Esmaeili Ouri, A., & Ziegler, A. D. (2018). Modeling runoff and sediment yield using hydrogeomorphic characteristics in Samian watershed, Ardabil Province. Watershed Science and Engineering, 12(43), 117–126. https://doi.org/10.1001.1.20089554.1397.12.43.13.6 (in Persian)

Gorelick, N., Hancher, M., Dixon, M., Ilyushchenko, S., Thau, D., & Moore, R. (2017). Google Earth Engine: Planetary-scale geospatial analysis for everyone. Remote sensing of Environment, 202, 18-27.‏ https://doi.org/10.1016/j.rse.2017.06.031

Guo, F., Fan, L., Zhang, C., & Xue, S. (2024). A novel space–spectrum array tile probability random-forest model enhances LULC mapping accuracy on Google Earth Engine: An experiment in Ordos, China. Ecological Informatics, 81, 102607.‏ https://doi.org/10.1016/j.ecoinf.2024.102607

Gupta, R. P. (2017). Remote sensing geology. Springer.‏

Haji Elyasi, N., Naseri, B., Badi’i, & Seyed Peyman. (2024). Evaluating land use changes based on integration of machine learning and spectral angle mapper using variable training samples: Case study Anzali wetland watershed. Journal of Natural Geography Research, 56(4), 39–56. https://doi.org/10.22059/jphgr.2025.384462.1007848 (in Persian)

Hasan, R., Kapoor, A., Singh, R., & Yadav, B. K. (2024). A state-of-the-art review on the quantitative and qualitative assessment of water resources using google earth engine. Environmental Monitoring and Assessment, 196(12), 1266. doi: 10.1007/s10661-024-13315-5

Hezbavi, A., & Nazila. (2024). Assessment and optimization of hydrological connectivity for water resources management in Samian watershed. Spatial Planning, 14(3), 115–138. https://doi.org/10.22108/sppl.2024.139234.1749 (in Persian)

Irani, A., Hiraad, & Rasouli, [Initials]. (2024). Analyzing the relationship between land use changes and landscape metrics: Case study of Godarchai watershed. Spatial Geography Journal, 14(1), 165–188. (in Persian)

Izadi Fard, H., Rathi Nezami, M. R., & Khavarian, [Initials]. (2024). Urban development and characteristics of short- and medium-term hydrological drought in Samian watershed, Ardabil Province. Geography and Environmental Studies, 51(46), 46. https://doi.org/10.71740/ges.2024.979123 (in Persian)

Izadi Fard, H., Rathi Nezami, M. R., & Khavarian, [Initials]. (2023). Relationship between surface runoff coefficient changes and land use/cover changes in Samian watershed, Ardabil Province. Journal of Dryland Geographical Studies, 14(51), 24–41. https://doi.org/10.22034/jargs.2023.373960.0 (in Persian)

Jahanbakhshi, F., & Ekhtesasi, M. R. (2018). Evaluating the performance of three image classification methods (Random Forest, Support Vector Machine, and Maximum Likelihood) in land use mapping. Journal of Remote Sensing and GIS in Iran, 22(4), 235–247. https://doi.org/10.1001.1.24763594.1397.22.4.29.4 (in Persian)

Jahdi, [Initials]. (2023). Modeling land use changes and future predictions using CA-ANN simulation in Shenarood watershed, Siahkal. Geography and Environmental Studies, 12(46), 164–179. (in Persian)

Jiang, M., Cheng, T., Dong, K., Xu, S., & Geng, Y. (2020). Fault diagnosis method of submersible screw pump based on random forest. Plos one, 15(11), e0242458.‏https://doi.org/10.1371/journal.pone.0242458

Karimzadeh Motlagh, L., Lotfi, M., Pour Manafi, & Ahmadi Zadeh, S. (2022). Assessment and prediction of land use changes using CA-Markov model. Geography and Environmental Planning, 33(2), 67–84. https://doi.org/10.22108/gep.2022.130601.1458 (in Persian)

Khalaji, & Saba. (2021). Effects of land use changes on land use planning. Journal of Space and Place Research, 19(1), 5–18. https://doi.org/10.1001.1.25386050.1400.1400.19.1.2 (in Persian)

Kuemmerle, T., Levers, C., Erb, K., Estel, S., Jepsen, M. R., Müller, D., ... & Reenberg, A. (2016). Hotspots of land use change in Europe. Environmental research letters, 11(6), 064020.‏ doi:10.1088/1748-9326/11/6/064020

Lin, C., & Doyog, N. D. (2023). Challenges of retrieving LULC information in rural-forest mosaic landscapes using random forest technique. Forests, 14(4), 816.‏ https://doi.org/10.3390/f14040816

Loukili, I., Laamrani, A., El Ghorfi, M., & El Moutak, S. (2025). Monitoring land changes at an open mine site using remote sensing and multi-spectral indices. Heliyon, 11(2).‏ doi:10.1016/j.heliyon.2025.e41845

Mazaheri, M. R., Esfandiari, M., Masih Abadi, M. H., & Kamali, A. (2013). Monitoring temporal land use changes using remote sensing and GIS techniques (Case study: Jiroft, Kerman Province). Remote Sensing and GIS Applications in Natural Resources, 4(2), 25–39. (in Persian)

Moradzadeh, H., Hezbavi, A., Esmaeili Ouri, A., Mostafazadeh, R., ... & Nazila. (2022). Assessing spatial heterogeneity of hydro-erosion disturbance index in Samian sub-watersheds. Hydrogeomorphology, 9(31), 117–136. https://doi.org/10.22034/hyd.2022.51186.1634 (in Persian)

Noszczyk, T. (2019). A review of approaches to land use changes modeling. Human and Ecological Risk Assessment: An International Journal, 25(6), 1377-1405.‏ doi:10.1080/10807039.2018.1468994

Noubaharan, K., Abtahi, A., & Mahmoudi, S. (2019). Trends of land use changes in Marvdasht plain, Fars Province. Water and Soil Conservation, 8(3), 68–82. (in Persian)

Rana, V. K., & Suryanarayana, T. M. V. (2020). Performance evaluation of MLE, RF and SVM classification algorithms for watershed scale land use/land cover mapping using sentinel 2 bands. Remote Sensing Applications: Society and Environment, 19, 100351.‏ https://doi.org/10.1016/j.rsase.2020.100351

Rathi Nezami, H., Izadi Fard, H., Khavarian, [Initials], & Mofateh, R. (2023). Land use/cover changes in Samian watershed and their relationship with surface water quality. Environmental and Water Engineering, 9(3), 413–426. https://doi.org/10.22034/jewe.2022.341548.1783 (in Persian)

Sellami, E. M., & Rhinane, H. A. S. S. A. N. (2023). A new approach for mapping land use/land cover using google earth engine: a comparison of composition images. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 48, 343-349. doi.org/10.5194/isprs-archives-XLVIII-4-W6-2022-343-2023

Shekrian, F., & Soleimani, K. (2024). Land use changes based on satellite imagery in Haraz plain. Iranian Journal of Remote Sensing and GIS, 16(1), 113–128. https://doi.org/10.48308/gisj.2023.232979.1176 (in Persian)

Taloor, A.K., Kumar, V., Kumar, S.G., Singh, A.K., Kale, A.V., Sharma, R., Khajuria, V., Raina, G., Kouser, B., & Chowdhary, N.H. (2020). Land use land cover dynamics using remote sensing and gis techniques in western doon valley uttarakhand India. Geoecology of Landscape Dynamics, Advances In Geographical And Environmental Sciences, 4(1), 37-51.

Tavakoli, M., & Naeem Abadi, N. (2019). Urban sprawl and land use changes in peri-urban areas of Nishapur. Peri-urban Space Development, 1(2), 151–165. https://doi.org/10.1001.1.26764164.1398.1.2.12.4 (in Persian)

Tesfaye, W., Elias, E., Warkineh, B., Tekalign, M., & Abebe, G. (2024). Modeling of land use and land cover changes using google earth engine and machine learning approach: implications for landscape management. Environmental Systems Research, 13(1), 31.‏

Van Asselen, S., & Verburg, P. H. (2013). Land cover change or land‐use intensification: simulating land system change with a global‐scale land change model. Global change biology, 19(12), 3648-3667.‏https://doi.org/10.1111/gcb.12331

Wang, J., Bretz, M., Dewan, M. A. A., & Delavar, M. A. (2022). Machine learning in modelling land-use and land cover-change (LULCC): Current status, challenges and prospects. Science of the Total Environment, 822, 153559.‏https://doi.org/10.1016/j.scitotenv.2022.153559

Wang, Y., & Ling, J. (2024). Analysis of spatio-temporal evolution and influencing factors of land dividends in China. PloS one, 19(11), e0309786.https://doi.org/10.1371/journal.pone.0309786

Wu, J., Qin, C. X., Yue, Y., Cheng, S. P., Zeng, H., & He, L. Y. (2024). Comprehensive effects of climate, land use/cover and management practices on runoff and nutrient variations in a rapidly urbanizing watershed. Chemosphere, 349, 140934.‏https://doi.org/10.1016/j.chemosphere.2023.140934

Youneszadeh, S., Salajeghe, S., Mesbahzadeh, T., Tajrishi, [Initials], Khalighi, S., & Taheri Shahraini. (2020). Evaluating the capability of Dyna-CLUE model in analyzing and predicting future land use patterns in Siminerud watershed. Ecohydrology, 8(1), 73–87. https://doi.org/10.22059/ije.2020.309703.1381 (in Persian)

Yousefi Taleghani, M., Molaei Heshjin, R., Nasrollah, & Rezaei, [Initials]. (2022). Explaining factors affecting land use changes in peri-urban villages of Rasht. Peri-urban Space Development, 7(4), 47–66. https://doi.org/10.1001.1.26764164.1401.4.1.3.4 (in Persian)

Zarei, H., Hezbavi, A., Mostafazadeh, R., Roouf, & Esmaeili Ouri, A., & Abazar. (2020). Comparing vulnerability of sub-watersheds of Samian based on changes in climatic components. Journal of Natural Geography Research, 52(2), 217–236. https://doi.org/10.22059/jphgr.2020.283909.1007406 (in Persian)

Zhu, L., Suomalainen, J., Liu, J., Hyyppä, J., Kaartinen, H., & Haggren, H. (2018). A review: Remote sensing sensors. Multi-purposeful application of geospatial data, 19, 19-42.‏