Evaluation of Parametric and Non-Parametric Models in Estimating Canopy Cover Density of Zagros Forests Using Remote Sensing and Machine Learning

Alimahmoodi Sarab, Sajad; Moayery, MohamadHadi; Mirzavand, MOhammad; Shataee Jouibary, Shaban; Rashki, Alireza

doi:10.22059/ije.2025.397227.1873

Evaluation of Parametric and Non-Parametric Models in Estimating Canopy Cover Density of Zagros Forests Using Remote Sensing and Machine Learning

Document Type : Research Article

Authors

¹ Assistant Prof., Research Institute of Forests and Rangelands, Agricultural Research Education and Extension Organization (AREEO), Ahvaz, Iran.

² Associate Prof., Faculty of Forest Sciences, Gorgan University of Agricultural Sciences and Natural Resources, Gorgan, Iran

³ Assistant Prof., School of Energy Engineering and Sustainable Resources, College of Interdisciplinary Science and Technology, University of Tehran, Tehran, Iran

⁴ Professor., Dept. of Forestry, Gorgan University of Agricultural Sciences and Natural Resources, Gorgan, Iran.

⁵ Department of Faculty of Natural Resources and Environment, Ferdowsi University of Mashhad, Mashhad, Iran

10.22059/ije.2025.397227.1873

Abstract

Research Topic: Evaluation of Parametric and Non-Parametric Models in Estimating Canopy Cover Density of Zagros Forests Using Remote Sensing and Machine Learning
Objective: This study aims to compare parametric and non-parametric methods for estimating the percentage of forest canopy cover in a section of the Zagros ecosystem.
Method: In order to achieve the research objective, field sampling was conducted to determine the percentage of canopy cover, and high-resolution satellite imagery was utilized. The vegetation indices TSAVI, NDVI, and WDVI were calculated. Subsequently, the values derived from the vegetation cover indices at the sample plots were extracted using the Zonal Statistics function in ArcGIS. Multiple linear regression and artificial neural networks were employed to estimate vegetation density. To compare the performance of these two models, the metrics RMSE, RMSE%, and R² were utilized.
Results: The results indicated that the MLR model achieved an R² value of 0.54 and an RMSE% of 10.4 at a 0.05 confidence level, while the MLP model yielded an R² of 0.82 and an RMSE% of 4.5.
Conclusions: The comparative analysis demonstrated that the artificial neural network (MLP) provided more accurate estimates with lower error rates than the multiple linear regression (MLR) method in predicting vegetation density.

Keywords

Main Subjects

Biodiversity

References

Amiryousefi, M.R., Zarei, M., Azizi, M., & Mohebbi, M. (2012). Modelling some physical characteristics of pomegranate (Punica granatum L.) Fruit during ripening using artificial neural network. Journal of Agricultural Science and Technology, 14, 857-867.

Arekhi, S., & Adibnejad, M. (2011). Efficiency assessment of the of Support Vector Machines for land use classification using Landsat ETM+ data (Case study: Ilam Dam Catchment). Iranian Journal of Range and Desert Research (IJRDR), 18 (3), 420-440. (In Persian)

Baret, F., Guyot, G., & Mojor, D.J. (1989). TSAVI: A vegetation index which minimizes soil brightness effectson LAI and APAR estimation PP. 1355-1358. In Proc. IGARRS 89 Can. Symposium of Remote Sensing12th, Vancouver, BC, Canada.

Cain, S.A. (1938). The species-area curve. American Midland Naturalist, 19, 573-581.

Clevers, J.G.P.W. (1978). The deviation of simplified reflectance model for estimation of leaf area index.

Fathipour Jalilian, A., & Najba, M. (2009). Neural network in SPSS. Kian publication. 151 p. (In Persian)

Feiznia, S., Asgari, M.H., & Moazzami, M. (2008). Investigating the applicability of Neural Network method for estimating daily suspended sediment yield (Case study: Zard Drainage Basin, Khozestan Province). Journal of the Iranian Natural Res, 60(4), 1199-1210. (In Persian)

Foody, G.M., Boyd, D.S., & Cutler, M.E.J. (2003). Predictive relations of tropical forest biomass from Landsat TM data and their transferability between regions, Remote Sensing of Environment, 85, 463– 474.

Goodarzi Mehr, S., Abbaspur, R.A., Ahadnejad, V., & Khakbaz, B. (2012). Comparison of Support Vector Machine, Neural Network, and Maximum Likelihood Methods for the Separation of Lithological Units. Iranian Journal of Geology, 6(21). (In Persian)

Guo, Y., De Jong, K., Liu, F., Wang, X., & Li, C. (2012). A Comparison of Artificial Neural Networks and Support Vector Machines on Land Cover Classification, Springer-verlag Berlin Heidelberg, Communications in Computer and Information Science ISICA, CCIS, 316, 531-539.

Huang, J.L., & Klemas, V. (2012). Using remote sensing of land cover change in coastal watersheds to predict downstream water quality. J. Coast. Res. 28, 930–944.

Karami, O., Fallah, A., Shataei, Sh., & Latifi, H. (2017). Investigation on the feasibility of mapping of oak forest dieback severity using Worldview-2 satellite data (Case study: Ilam forests). Iranian Journal of Forest and Poplar Research, 25(3), 452-462. DOI: 9.2017.ijfpr/22092.1. (In Persian)

Kia, M. (2010). Neural network in Matlab. Kian publication. 226 p. . (In Persian)

Kouchpideh, N., & Keykhosravi, D. (2008). History of Natural Resource Evolution in Iran - Forests, Rangelands, and Watershed Management Organization Pooneh Publications. 96 p. (In Persian)

Leboeuf, A., Fournier, R.A., Luther, J.E., Beaudoin, A., & Guindon, L. (2012). Forest attribute estimation of northeastern Canadian forests using QuickBird imagery and a shadow fraction method. Forest Ecology and Management, 266, 66-74.

Lu, D., Mausel, P., Brondizio, E., & Moran, E. (2004). Relationships between forest stand parameters and Landsat TM spectral response in the Brazilian Amazon Basin. Forest Ecology and Management, 198, 149-167.

Miller, S.D. (2003). A consolidated technique for enhancing desert dust storm with MODIS, Geographical research letters NO, 20pp. 12-1-12-4.

Mirzavand, M., Ghasemieh, H., Akbari, M., & Sadatinejad, J. (2015). Artificial Neural Network (ANN) Model for Ground Water Quality Simulation (Case study: Kashan aquifer). Journal of Range and Watershed Management. 68(1), 159-171. (In Persian)

Mokhtari, M., & Najafi, A. (2015). Comparison of Support Vector Machine and Neural Network Classification Methods in Land Use Information Extraction through Landsat TM Data. Jwss, 19 (72), 35-45. (In Persian)

Noshadi, H., Namiranian, M., Attarod, P., & Hoseinzadeh, J. (2014). Effect of Physiographic Factors on Mortality of Persian Oak in the Middle of Forests (Case study: Ilam). Journal of Forest and Wood Products, 67(1), 73-84. (In Persian)

Nourian, N., & Shataee Jooybari, Sh. (2014). Capability investigation the different spatial resolution satellites with emphasis on spectral indices for separation of planted pure stands. J. of Wood & Forest Science and Technology, 12 (3), 149-166. (In Persian)

Otukei, J. R., & Blaschke, T. (2010). Land cover change assessment using decision trees, support vector machines and maximum likelihood classification algorithms. International Journal of Applied Earth Observation and Geoinformation, 12, S27-S31.

Parma, R., Shataee, Sh., Khodakarami, Y., & habashi, H.. (2011). Capability Investigation on ETM+ Data for Forest Type Mapping in the Zagros Forests (Case Study Ghalajeh Forests, Kermanshah Province. Journal of Wood and Forest Science and Technology, 18(1), 1-76. (In Persian)

Pir Bavaghar, M. (2011). Evaluation of capability of IRS-P6 satellite data for predicting quantitative attributes of forests (case study: Northern Zagros forests). Iranian Journal of Forest, 3(4), 277-289. (In Persian)

Rahdari,V., Maleki, S., Soffianian, A.R., Khajeddin, S.J., & Pahlevanravi, A. (2013a). Change detection of canopy cover percentage using satellite data during 1972 to 2008 (case study: Mouteh Wild Life Refuge). Iranian Journal of Range and Desert Research (IJRDR). 20(3), 508- 521. (In Persian)

Rahdari,V., Soffianian, A.R., Khajeddin, S.J., & maleki, S. (2013b). Identification of Satellite Image Ability for Vegetation Cover Crown Percentage Mapping in Arid and Semi Arid Region (Case study: Mouteh wild life sanctuary). Journal of Environmental Science and Technology, 15(4), 43-54. (In Persian)

Rezae Arshad, R., Sayyad, GH., Mazloom, M., Shorafa, M., & Jafarnejady, A. (2012). Comparison of Artificial Neural Networks and Regression Pedotransfer Functions for Predicting Saturated Hydraulic Conductivity in Soils of Khuzestan Province. Jwss, 16 (60), 107-118. URL: http://jstnar.iut.ac.ir/article-1-2310-fa.html (In Persian)

Richardson, A. J., & Wieg, C. L. (1977). Distinguishing vegetation from soil background information. Photogrammetric Engineering & Remote Sensing, 43, 1541– 1552.

Rouse, J. W. (1974). Monitoring the vernal advancement of retrogration of natural vegetation. NASA/GSFS Type III, Final Report, Greenbelt, MD. 371 pp-NDVI.

Shataee, Sh., Kalbi, S., & Fallah. A. (2012). Forest attributes imputation using machine-learning methods and ASTER data: comparison of k-NN, SVR and random forest regression algorithms. International Journal of Remote Sensing, 33(19), 6254-6280 (In Persian).

Shojaeeian, A., Mokhtari Chelche, S., Keshtkar, l., & Soleymani rad, E. (2015). Comparing the Performance of Parametric and NonparametricMethods in Land Cover Classification using Landsat-8 Satellite Images (Case study: A part of Dezful city). Quarterly of Geographical Data, 24(93), 53-64. (In Persian)

Vafaei,S., Pourhashemi, M., Pirbavaghar, M., & Jafari, E. (2016). Applying artificial neural network and multiple linear regression models for estimation of forest density in Marivan forests. Iranian Journal of Forest, 7(4), 539-555. (In Persian)

Vali, A.A., Moayeri, M., Ramesht, M.H., & Movahedinia, N. (2010). A Comparative Performance Analysis of Artificial Neural Networks and Regression Models for Suspended Sediment Prediction. Case Study Alexander Cachment in Zayand Roud Basin. Physical Geography Research (JPHGR), 71(3), 21-30. (In Persian)

Zobeiri, M. (2005). Forest inventory. University of Tehran Press. (In Persian)

Zobeiri, M., & Majd, A. (2022). Introduction to Remote Sensing and Its Applications in Natural Resources. University of Tehran Press. (In Persian)