Modeling suitable habitats of Parrotia persica (DC.) C.A.Mey. in the Hyrcanian Forests using environmental factors | ||
| تحقیقات جنگل و صنوبر ایران | ||
| Article 5, Volume 33, Issue 1 - Serial Number 99, August 2025, Pages 50-68 PDF (2.15 M) | ||
| Document Type: Research article | ||
| DOI: 10.22092/ijfpr.2024.366708.2173 | ||
| Authors | ||
| Hamed Asadi* 1; Hamid Jalilvand2; Mahya Tafazoli3; Seyedeh Fatemeh Hosseini4 | ||
| 1Corresponding author, Assistant Prof., Department of Forest Science and Engineering, Faculty of Natural Resources, Sari Agricultural Sciences and Natural Resources University, Sari, Iran | ||
| 2Prof., Department of Forest Science and Engineering, Faculty of Natural Resources, Sari Agricultural Sciences and Natural Resources University, Sari, Iran | ||
| 3Ph.D. in Forest Soil Science, Department of Forest Science and Engineering, Faculty of Natural Resources, Sari Agricultural Sciences and Natural Resources University, Sari, Iran | ||
| 4Ph.D. Student of Silviculture and Forest Ecology, Faculty of Natural Resources, Sari Agricultural Sciences and Natural Resources University, Sari, Iran | ||
| Abstract | ||
| Background and objectives: Parrotia persica (DC.) C.A. Mey., endemic to the Hyrcanian forests of Iran, has been classified as Near Threatened in the latest conservation status assessment by the International Union for Conservation of Nature (IUCN). Given the importance of this species in preserving genetic diversity, identifying its suitable habitat is a critical first step in the implementation of conservation and management strategies. As species distribution models are widely used for predicting and identifying suitable planting habitats, the objective of this study was to apply such models, based on machine learning algorithms and environmental variables including topography, soil, and climate, to predict suitable habitats for P. persica. Methodology: The study was conducted across the entire Hyrcanian forest region in Iran. Forest inventory data from northern Iran were used, including coordinates, altitude, slope, aspect, diameter at breast height, species type, and tree height for each sample plot. Meteorological data (precipitation and temperature) were obtained from the NASA POWER project. Soil variables, including bulk density, sand, silt, clay percentages, pH, nitrogen content, and organic carbon, were extracted from SoilGrids 2.0. Topographic indices such as the Topographic Wetness Index (TWI), Topographic Position Index (TPI), and curvature were derived in a GIS environment. After compiling presence-absence data and environmental variables, five modeling algorithms were applied to predict the distribution of P. persica: Random Forest, Support Vector Machine, k-Nearest Neighbor, Generalized Linear Model, and Maximum Entropy. The Maximum Entropy model was implemented using the dismo package, and the other models were run using the caret package in R. Variable importance was assessed using dismotools for Maximum Entropy and caret for the remaining models. Model performance was evaluated using cross-validation, and the area under the curve (AUC) statistic was used to assess predictive accuracy. Results: Among all models, the Random Forest algorithm had the highest AUC value (0.87), indicating the best performance. Altitude emerged as the most influential factor in predicting P. persica distribution across all models. The Random Forest and Maximum Entropy models identified soil sand percentage and precipitation, respectively, as the second most important variables. In contrast, Support Vector Machine, k-Nearest Neighbor, and Generalized Linear Models identified average air temperature as the second most important factor. Interpolated probability maps indicated a high likelihood of P. persica presence (over 70%) in low-altitude regions of the eastern and central Hyrcanian forests, while its presence was predicted to be very low in the western regions. Conclusion: All models demonstrated acceptable accuracy in predicting P. persica distribution, with the Random Forest model outperforming the others. Altitude and temperature were the most critical variables influencing its distribution. The highest presence probability was predicted in low-altitude, central parts of the Hyrcanian forests. These findings suggest the feasibility of identifying suitable mother trees and enhancing reproduction success through targeted breeding and silvicultural practices. Additionally, relatively undisturbed areas within suitable habitats could be designated as conservation reserves for this species. | ||
| Keywords | ||
| Maximum entropy model; random forest model; species distribution model; topographic factors | ||
| References | ||
|
- Alavi, S.J., Ahmadi, K., Hosseini, S.M., Tabari, M. and Nouri, Z., 2020. The importance of climatic, topographic, and edaphic variables in the distribution of yew species (Taxus baccata L.) and prioritization of areas for conservation and restoration in the north of Iran. Iranian Journal of Forest, 11(4): 477-492 (In Persian with English summary). - Alegria, C., Almeida, A.M., Roque, N., Fernandez, P. and Ribeiro, M.M., 2023. Species distribution modelling under climate change scenarios for Maritime pine (Pinus pinaster Aiton) in Portugal. Forests, 14(3): 591. - Araújo, M.B., Anderson, R.P., Márcia Barbosa, A., Beale, C.M., Dormann, C.F., Early, R., ... and Rahbek, C., 2019. Standards for distribution models in biodiversity assessments. Science Advances, 5(1): eaat4858. - Asadi, H., Jalilvand, H. and Moslemi, S.M., 2021. Vegetation classification of Darabkola forest and their relation to physiographic factors. Iranian Journal of Applied Ecology, 10(3): 17-33 (In Persian with English summary). - Bakhshi Khaniki, G. and Mohammadi, B., 2012. Ecological study of some species of the genus Salsola (Chenopodiaceae) in Golestan Province. New Cellular and Molecular Biotechnology Journal, 2(6): 45-52 (In Persian with English summary). - Bińka, K., Nitychoruk, J. and Dzierżek, J., 2003. Parrotia persica C.A.M. (Persian witch hazel, Persian ironwood) in the Mazovian (Holsteinian) Interglacial of Poland. Grana, 42(4): 227-233. - Chauvier, Y., Thuiller, W., Brun, P., Lavergne, S., Descombes, P., Karger, D.N., ... and Zimmermann, N.E., 2021. Influence of climate, soil, and land cover on plant species distribution in the European Alps. Ecological Monographs, 91(2): e01433. - Cutler, D.R., Edwards Jr., T.C., Beard, K.H., Cutler, A., Hess, K.T., Gibson, J. and Lawler, J.J., 2007. Random forests for classification in ecology. Ecology, 88(11): 2783-2792. - Damaneh, J.M., Ahmadi, J., Rahmanian, S., Sadeghi, S.M.M., Nasiri, V. and Borz, S.A., 2022. Prediction of wild pistachio ecological niche using machine learning models. Ecological Informatics, 72: 101907. - Fathollahzadeh, A., 2018. Response curve and species distribution model of beech, basswood, ironwood, maple and alder in Educational and Research Forest of Tarbiat Modares University. M.Sc. Thesis, Faculty of Natural Resources, Tarbiat Modares University, Nour, Iran, 69p (In Persian with English summary). - Goodarzi, Gh.R., Ahmadloo, F. and Sagheb-Talebi, Kh., 2013. Effects of Physiographic factors and some physical and chemical soil properties on distribution Amygdalus scoparia Spach. in 4 areas of Markazi Province. Journal of Wood and Forest Science and Technology, 19(3): 59-76 (In Persian with English summary). - Greener, J.G., Kandathil, S.M., Moffat, L. and Jones, D.T., 2022. A guide to machine learning for biologists. Nature Reviews Molecular Cell Biology, 23(1): 40-55. - Guisan, A., Thuiller, W. and Zimmermann, N.E., 2017. Habitat Suitability and Distribution Models: With Applications in R. Cambridge University Press, Cambridge, England, 462p. - Hamrani, A., Akbarzadeh, A. and Madramootoo, C.A., 2020. Machine learning for predicting greenhouse gas emissions from agricultural soils. Science of The Total Environment, 741: 140338. - Hijmans, R.J., Phillips, S., Leathwick, J., Elith, J. and Hijmans, M.R.J., 2017. Package ‘dismo’. Circles, 9(1): 68p. - Imani Rastabi, M., Jalilvand, H., Fallah, A. and Shahin Kaleybar, B., 2023. Genetic diversity of Parrotia persica (DC) C.A. Meyer in Hyrcanian forests. Iranian Journal of Forest, 15(3): 313-327 (In Persian with English summary). - Iturrate‐Garcia, M., O'Brien, M.J., Khitun, O., Abiven, S., Niklaus, P.A. and Schaepman‐Strub, G., 2016. Interactive effects between plant functional types and soil factors on tundra species diversity and community composition. Ecology and Evolution, 6(22): 8126-8137. - Khosropour, E., Heydari, M., Etemad, V. and Marvi Mohajer, M.R., 2011. Effect of physiography on asexual reproduction of Iron wood (Case study: Kheiroud forest). Iranian Journal of Forest and Poplar Research, 19(1): 154-162 (In Persian with English summary). - Kuhn, M., Wing, J., Weston, S., Williams, A., Keefer, C., Engelhardt, A., ... and Team, R.C., 2020. Package ‘caret’. The R Journal, 223(7): 48. - Mahjoob, Ch., 2006. Investigation on the spatial distribution of Persian Parrotia (Parrotia persica) in relation to landform and edaphic factors in Patom District of Kheyroudkenar forest. Department of Forestry, University of Tehran, Karaj, Iran, 79p (In Persian with English summary). - Marvie Mohadjer, M.R., 2005. Silviculture. University of Tehran Press, Tehran, Iran, 387p (In Persian). - Mirdeylami, S.Z. and Heshmati, Gh.A., 2014. Study of the forest vegetation on the basis of elevation gradient in Touskestan-Charbagh habitat, Golestan province. Journal of Wood and Forest Science and Technology, 20(4): 41-60 (In Persian with English summary). - Moghbel Esfahani, F., Alavi, S.J., Hosseini, S.M. and Tabari Kochaksarai, M., 2023. Determining the habitat suitability of Quercus castaneifolia C. A. Mey In order to plan restoration using species distribution modeling. Journal of Forest Research and Development, 9(3), 419-436 (In Persian with English summary). - Mohammadi, A., Alavi, S.J. and Hosseini, S.M., 2017. Predicting the habitat suitability of Wych elm (Ulmus glabra Huds.) in Kheyroud Forest. Journal of Wood and Forest Science and Technology, 24(3): 67-80 (In Persian with English summary). - Moridpour, A., Namiranian, M., Alavi, S.J. and Etemad, V., 2023. Identifying the most important factors affecting the distribution of Ash (Fraxinus excelsior L.) and detect potential habitats areas in Kherudkanar Nowshahr forest. Iranian Journal of Forest, 15(1), 69-85 (In Persian with English summary). - Noble, W.S., 2006. What is a support vector machine?. Nature biotechnology, 24(12): 1565-1567. - Oldfield, S. and Rivers, M.C., 2024. Parrotia persica. The IUCN Red List of Threatened Species 2024: e.T199803456A199803606. https://dx.doi.org/10.2305/IUCN.UK.2024-1.RLTS.T199803456A199803606.en - Palimkar, P., Shaw, R.N. and Ghosh, A., 2022. Machine learning technique to prognosis diabetes disease: Random forest classifier approach: 219-244. In: Bianchini, M., Piuri, V., Das, S. and Shaw, R.N. (Eds.). Advanced Computing and Intelligent Technologies. Lecture Notes in Networks and Systems, vol 218, Springer, Singapore, 645p. - Pecchi, M., Marchi, M., Burton, V., Giannetti, F., Moriondo, M., Bernetti, I., ... and Chirici, G., 2019. Species distribution modelling to support forest management. A literature review. Ecological Modelling, 411: 108817. - Pourbabaei, H., Heidari, M., Naghilou M. and Begim Faghir, M., 2015. Relationship between vegetation and environmental factors in the Anatolian oak (Quercus petraea L. subsp. iberica (Stev.) Krassiln) habitat: a case study of Asalem forests, Guilan. Journal of Plant Research, 28(1): 53-62 (In Persian with English summary). - Safarov, I.S., 1972. Ironwood Parrotia persica C.A.M. History, geography, taxonomy and bioecological peculiarities. Botanicheskii Zhurnal, 57: 932-944. - Scherrer, D. and Guisan, A., 2019. Ecological indicator values reveal missing predictors of species distributions. Scientific Reports, 9(1): 3061. - Schonlau, M. and Zou, R.Y., 2020. The random forest algorithm for statistical learning. The Stata Journal, 20(1): 3-29. - Sefidi, K., 2017. Structural diversity as component of biodiversity in forest ecosystems, case study from population of Persian ironwood (Parrotia persica C.A. Meyer) in the north Iran. Journal of Plant Research, 29(4): 805-818 (In Persian with English summary). - Sharma, A., Sachdeva, S.N. and Aggarwal, P., 2023. Predicting IRI using machine learning techniques. International Journal of Pavement Research and Technology, 16(1): 128-137. - Sobhani, P. and Danehkar, A., 2024. Modeling the distribution of Avicennia marina (Forssk.) Vierh. in the Khamir and Qeshm mangrove forests, Iran, using maximum entropy (MaxEnt). Iranian Journal of Forest and Poplar Research, 32(2): 97-111 (In Persian with English summary). - Sparks, A.H., 2018. nasapower: a NASA POWER global meteorology, surface solar energy and climatology data client for R. Journal of Open Source Software, 3(30): 1035. - Srivastava, V., Lafond, V. and Griess, V.C., 2019. Species distribution models (SDM): applications, benefits and challenges in invasive species management. CABI Reviews, 14: 1-13. - Steen, B., Broennimann, O., Maiorano, L. and Guisan, A., 2024. How sensitive are species distribution models to different background point selection strategies? A test with species at various equilibrium levels. Ecological Modelling, 493: 110754. - ter Steege, H. and Zagt, R., 2002. Density and diversity. Nature, 417(6890): 698-699. - Valavi, R., Guillera‐Arroita, G., Lahoz‐Monfort, J.J. and Elith, J., 2022. Predictive performance of presence-only species distribution models: a benchmark study with reproducible code. Ecological Monographs, 92(1): 01486. - Valizadeh, E., Asadi, H., Jaafari, A. and Tafazoli, M., 2023. Machine learning prediction of tree species diversity using forest structure and environmental factors: a case study from the Hyrcanian forest, Iran. Environmental Monitoring and Assessment, 195(11): 1334. - Wang, H. and Hu, D., 2005. Comparison of SVM and LS-SVM for regression. Proceedings of 2005 International Conference on Neural Networks and Brain. Beijing, China, 13-15 Oct. 2005: 279-283. - Wang, R., Wan, S., Chen, W., Qin, X., Zhang, G. and Wang, L., 2024. A novel finer soil strength mapping framework based on machine learning and remote sensing images. Computers and Geosciences, 182: 105479. - Xu, W.B., Svenning, J.C., Chen, G.K., Zhang, M.G., Huang, J.H., Chen, B., ... and Ma, K.P., 2019. Human activities have opposing effects on distributions of narrow-ranged and widespread plant species in China. Proceedings of the National Academy of Sciences, 116(52): 26674-26681. - Yosefzadeh, H., Tabari, M., Akbarinia, M., Akbarian, M.R. and Bussotti, F., 2010. Morphological plasticity of Parrotia persica leaves in eastern Hyrcanian forests (Iran) is related to altitude. Nordic Journal of Botany, 28(3): 344-349. - Zakeri Pashakolaei M., Alvaninejad, S. and Esmailzade, O., 2014. Relationship between plant biodiversity and topographical factors in forests of West Mazandaran (Case study: Research Forest of Tarbiat Modares University). Iranian Journal of Applied Ecology, 3(8): 1-16 (In Persian with English summary). - Zhong, Y., Xue, Z., Jiang, M., Liu, B. and Wang, G., 2021. The application of species distribution modeling in wetland restoration: A case study in the Songnen Plain, Northeast China. Ecological Indicators, 121: 107137. | ||
|
Statistics Article View: 871 PDF Download: 401 |
||
Journal management system. designed by sinaweb