ارتقاء دقت نقشه‌ها با گذار از مدل لجستیک به جمعی تعمیم‌یافته (پهنه‌بندی احتمال تخریب پوشش جنگلی در حوزه جنگلداری ناو اسالم در محیط GIS و R)

محورهای موضوعی : زیرساخت اطلاعات مکانی و طبقه بندی

Saeid Shabani ^{1 *} , مسعود علی دوست ² , Shahriar Sobhzahedi ³ , بهروز محسنی ⁴

1 - AREEO
2 - کارشناس ارشد پژوهشی مرکز تحقیقات کشاورزی و منابع طبیعی گیلان
3 - Rasht University
4 - بخش تحقیقات منابع طبیعی، مرکز تحقیقات و آموزش کشاورزی و منابع طبیعی استان گلستان، سازمان تحقیقات، آموزش و ترویج کشاورزی

تاریخ دریافت : 1404/02/06 تاریخ پذیرش : 1404/05/04 تاریخ انتشار : 1404/10/06

کلید واژه: تخریب جنگل, جنگل‌های هیرکانی, مدل جمعی تعمیم‌یافته, مدل‌سازی فضایي,

چکیده مقاله :

با افزایش روند تخریب پوشش جنگلی در جنگل‌های هیرکانی، نیاز به مدل‌سازی دقیق و پیش‌بینانه برای شناسایی مناطق بحرانی بیش از پیش احساس می‌شود. این پژوهش با هدف بهبود دقت نقشه‌های پهنه‌بندی احتمال تخریب پوشش جنگلی، عملکرد مدل لجستیک و مدل جمعی تعمیم‌یافته را در محیط GIS و R مقایسه کرده است. منطقه مورد مطالعه، حوزه جنگلداری ناو اسالم در غرب جنگل‌های گیلان است که با چالش‌های متعددی نظیر گسترش اراضی کشاورزی، جاده‌سازی و افزایش سکونت‌گاه‌های انسانی مواجه است. در این مطالعه، ۱۴ متغیر محیطی و انسانی مؤثر استخراج، پردازش و به دو مدل رگرسیونی وارد شدند. مدل جمعی تعمیم‌یافته توانست با برازش روابط غیرخطی، ساختار پیچیده و آستانه‌دار تأثیر متغیرهایی همچون فاصله از مناطق مسکونی، فاصله از جاده، نزدیکی به اراضی کشاورزی، اثر باد و ارتفاع از سطح دریای آزاد را آشکار سازد؛ در حالی که مدل لجستیک این روابط را صرفاً به‌صورت خطی تحلیل می‌کرد. نتایج نشان داد که مدل جمعی تعمیم‌یافته با سطح زیر منحنی عملکرد برابر 978/0 و شاخص کاپای 87/0 نسبت به مدل لجستیک (با سطح زیر منحنی عملکرد 895/0 و شاخص کاپای برابر 73/0) دقت بالاتری در پیش‌بینی مناطق تخریب‌شده دارد. همچنین، نقشه‌های تولیدشده توسط مدل جمعی تعمیم‌یافته از تفکیک فضایی بالاتری برخوردار بودند. یافته‌های پژوهش حاکی از آن است که استفاده از مدل‌های انعطاف‌پذیر مانند مدل جمعی تعمیم‌یافته می‌تواند ابزار مؤثری در پهنه‌بندی خطر، ارزیابی محیط‌زیستی و طراحی سیاست‌های حفاظتی باشد؛ به‌ویژه در اکوسیستم‌های ناهمگن و پیچیده مانند جنگل‌های هیرکانی، که تخریب آن‌ها تحت تأثیر تعامل عوامل طبیعی و انسانی است.

چکیده انگلیسی:

As the degradation of forest cover in the Hyrcanian forests intensifies, the need for accurate and predictive modeling to identify critical areas has become increasingly urgent. This study aims to improve the accuracy of deforestation risk mapping by comparing the performance of the Logistic Regression (LR) model and the Generalized Additive Model (GAM) in a GIS and R environment. The study area is located in the Nav forests, Rezvanshahr county in the western part of the Gilan province, which faces multiple challenges such as agricultural expansion, road construction, and increasing human settlements. In this study, 14 environmental and anthropogenic variables were extracted, processed, and entered into both regression models. The GAM model was able to reveal the non-linear relationships and complex, threshold-driven interactions between variables such as distance to residential area, distance to road, distance to agricultural land, wind effect, elevation lands, whereas the LR model analyzed these relationships linearly. The results showed that the GAM model achieved higher predictive accuracy for identifying deforested areas with an AUC of 0.978 and kappa index of 0.87, compared to the LR model's AUC of 0.895 and kappa index of 0.73. Additionally, the maps produced by the GAM model exhibited higher spatial resolution. The findings indicate that the use of flexible models like GAM can be an effective tool for risk zoning, environmental assessment, and designing conservation policies, especially in heterogeneous and complex ecosystems like the Hyrcanian forests, where degradation is influenced by the interaction of both natural and human factors.

منابع و مأخذ:

1. Bera D, Das Chatterjee N, Bera S. 2021. Comparative performance of linear regression, polynomial regression and generalized additive model for canopy cover estimation in the dry deciduous forest of West Bengal. Remote Sensing Applications: Society and Environment, 22: 100502.
2. Boonyanuphap J. 2005. Spatial model for determining risk area of deforestation. Suranaree Journal of Science and Technology, 12 (2): 145–159.
3. Brando PM, Balch JK, Nepstad DC, Morton DC, Putz FE, Coe MT, Silvério D, Macedo MN, Davidson EA, Nóbrega CC, Alencar A, Soares-Filho BS. 2014. Abrupt increases in Amazonian tree mortality due to drought–fire interactions. Proceedings of the National Academy of Sciences of the United States of America, 111: 6347–6352.
4. Cochrane MA. 2003. Fire science for rainforests. Nature, 421 (6926): 913-919.
5. Davison CW, Rahbek C, Morueta-Holme N. 2021. Land-use change and biodiversity: Challenges for assembling evidence on the greatest threat to nature. Global Change Biology, 27: 5414–5429.
6. de Oliveira BF, Bottino MJ, Nobre P, Nobre CA. 2021. Deforestation and climate change are projected to increase heat stress risk in the Brazilian Amazon. communications earth & environment, 2: 207.
7. Foley JA, DeFries R, Asner GP, Barford CC, Bonan G, Carpenter SR, Chapin FS, Coe MT, Daily GC, Gibbs H, Helkowski JH, Holloway T, Howard E, Kucharik CJ, Patz J, Prentice IC, Ramankutty N, Snyder PK. 2005. Global consequences of land use. Science, 309 (5734): 570-574.
8. Halperin J, LeMay V, Coops N, Verchot L, Marshall P, Lochhead K. 2016. Canopy cover estimation in miombo woodlands of Zambia: Comparison of Landsat 8 OLI versus RapidEye imagery using parametric, nonparametric, and semiparametric methods. Remote Sensing of Environment, 179: 170-182.
9. Hastie TJ, Tibshirani RJ 1990. Generalized Additive Models, 2nd ed. Chapman and Hall, London. 16 pp.
10. Hosonuma N, Herold M, Sy VD, Fries RSD, Brockhaus M, Verchot L, Romijn E. 2012. An assessment of deforestation and forest degradation drivers in developing countries. Environmental Research Letters, 7 (4): 044009.
11. Jellouli O, Bernoussi AS. 2022. The impact of dynamic wind flow behavior on forest fire spread using cellular automata: Application to the watershed BOUKHALEF (Morocco). Ecological Modelling, 468: 109938.
12. Kanade R, John R. 2018. Topographical influence on recent deforestation and degradation in the Sikkim Himalaya in India; Implications for conservation of East Himalayan broadleaf forest. Applied Geography, 92: 85–93.
13. Karami M, Sarvestan R. 2022. Locate solar panels using climate indices and GIS (Case study: Khuzestan). Journal of Environmental Science and Technology, 24 (2, 117): 169–180. (in Persian)
14. Knapp M, Strobl M, Venturo A, Seidl M, Jakubíkova L, Tajovský K, Kadlec T, Gonzalez E. 2022. Importance of grassy and forest non-crop habitat islands for overwintering of ground-dwelling arthropods in agricultural landscapes: A multi-taxa approach. Biological Conservation, 275: 109757.
15. Korhonen L, Hadi A, Packalen P, Rautiainen M. 2017. Comparison of Sentinel-2 and Landsat 8 in the estimation of boreal forest canopy cover and leaf area index. Remote Sensing of Environment, 195: 259-274.
16. Laurance WF, Goosem M, Laurance SGW. 2009. Impacts of roads and linear clearings on tropical forests. Trends in Ecology & Evolution, 24 (12): 659–669.
17. López S. 2022. Deforestation, forest degradation, and land use dynamics in the Northeastern Ecuadorian Amazon. Applied Geography, 145: 102749.
18. López-Bedoya PA, Bohada-Murillo M, Ángel-Vallejo MC, Audino LD, Davis ALV, Gurr G, Noriega JA, 2022. Primary forest loss and degradation reduces biodiversity and ecosystem functioning: A global meta-analysis using dung beetles as an indicator taxon. Journal of Applied Ecology, 59: 1572–1585.
19. Melin M, Korhonen L, Kukkonen M, Packalen P. 2017. Assessing the performance of aerial image point cloud and spectral metrics in predicting boreal forest canopy cover. ISPRS Journal of Photogrammetry and Remote Sensing, 129: 77-85.
20. Morreale LL, Thompson JR, Tang X, Reinmann AB, Hutyra LR. 2021. Elevated growth and biomass along temperate forest edges. Nature Communications, 12: 7181.
21. Netzel P, Tyminska L, Feleha DD, Socha J. 2024. New approach to assess forest fragmentation based on multiscale similarity index. Ecological Indiators, 158: 111530.
22. Ngo KM, Davies S, Hassan NFN, Lum S. 2016. Resilience of a forest fragment exposed to long-term isolation in Singapore. Plant Ecological Diversity, 9: 397–407.
23. Polansky L, Newman K, Nobriga M, Mitchell L. 2018. Spatiotemporal Models of an Estuarine Fish Species to Identify Patterns and Factors Impacting Their Distribution and Abundance. Estuaries and Coasts, 41 (2): 572-581.
24. Potts SE, Rose KA. 2018. Evaluation of GLM and GAM for estimating population indices from fishery independent surveys. Fisheries Research, 208: 167-178.
25. Sahana M, Hong H, Sajjad H, Liu J, Zhu AX. 2018. Assessing deforestation susceptibility to forest ecosystem in Rudraprayag district, India using fragmentation approach and frequency ratio model. Science of the Total Environment, 627: 1264–1275.
26. Sasikumar R, Indira S. 2019. Impact of climate factors through glm and gam methods. International Journal of Scientific & Technology Research, 8 (9): 3 pp.
27. Schwartz NB, Uriarte M, DeFries R, Bedka KM, Fernandes K, Gutiérrez-Vélez V, Pinedo-Vasquez MA. 2017. Fragmentation increases wind disturbance impacts on forest structure and carbon stocks in a western Amazonian landscape. Ecological Appications., 27: 1901–1915.
28. Sguotti C, Lynam CP, García-Carreras B, Ellis JR, Engelhard GH. 2016. Distribution of skates and sharks in the North Sea: 112 years of change. Global Change Biology, 22 (8): 2729-2743.
29. Shirk AJ., Cushman, SA, Waring KM, Wehenkel CA, Leal-Sáenz A, Toney C, Lopez-Sanchez CA. 2018. Southwestern white pine (Pinus strobiformis) species distribution models project a large range shift and contraction due to regional climatic changes. Forest Ecology and Management, 411: 176–186.
30. Silvério DV, Brando PM, Bustamante MMC, Putz FE, Marra DM, Levick SR, Trumbore SE. 2019. Fire, fragmentation, and windstorms: A recipe for tropical forest degradation. Journal of Ecology, 107 (2): 656–667.
31. Talebi KS, Sajedi T, Pourhashemi M. 2014. Forests of Iran, A Treasure from the Past, a Hope for the Future, Springer, Dordrecht, Germany, 149 pp.
32. Vásquez-Grandón A, Donoso PJ, Gerding V. 2018. Forest Degradation: When Is a Forest Degraded?. Forests, 9 (11): 726.
33. Vorpahl P, Elsenbeer H, Mrker M, Schrder S. 2012. How can statistical models help to determine driving factors of landslides? Ecological Modeling, 239: 27–39.
34. Worku A. 2023. Review on drivers of deforestation and associated socio-economic and ecological impacts. Advances in Agriculture. Food Science and Forestry, 11 (1): 1–12.
35. Yamamoto Y, Shigetomi Y, Ishimura Y, Hattori M. 2019. Forest change and agricultural productivity: Evidence from Indonesia. World Development 114: 196–207.