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

Abstract

The effects of irregular population growth, migration mobility, and vegetation dynamics by humans can lead to changes in Land Use and Land Cover (LULC). Changes in LULC are particularly significant in coastal areas associated with industrial activities. The southeastern Marmara region, which is one of Turkey's industrial coastal areas, is also affected by the surrounding changes. The study area was selected to determine LULC change and classification accuracy using Sentinel-2 vegetation indices combinations. In the study area, the Gemlik-Bursa Northern Interchange Investments Area and TOGG (Turkey's Automobile Initiative Group) factory are located. The study area was determined by creating a 5-km buffer zone from the coast to the mainland covering Armutlu district of Yalova province and Osmangazi, Mudanya, and Gemlik districts of Bursa province. Random Forest (RF) classification technique was applied both to the original bands and to 21 new band combinations that are derived from Sentinel-2 multispectral satellite imagery for 3 seasons in 2016 and 2020. The new band combinations used for classification were created by adding the normalized vegetation indices, the original bands and the bands obtained from the simple ratio formula. In 2016, the highest accuracy results for the winter, spring, and summer seasons were observed for the OI12 (82.93%), ORF (84.44%), and ORF (84.67%) indices, while in 2020 were observed for the OI5 (85.89%), ORF (84.75%), and OI6 (84.63%) indices. In Southeast Marmara, investment decisions taken at national level have led to population growth in the region. Although it was observed that there was no significant change in classification accuracy with the addition of spectral features to the original bands such as NDVI and SR, we believe that future testing of the data with different statistical and machine learning methods provide higher accuracy.

Keywords

• Remote Sensing
• LULC
• Vegetation İndices
• South – Eastern Marmara Region

Spektral İndeks Kombinasyonlarının Rastgele Orman (RO) Sınıflandırması Kullanarak Mevsimsel Arazi Kullanımı ve Bitki Örtüsü (AKBÖ) Değişiklikleri Üzerindeki Etkilerinin Belirlenmesi: Güneydoğu Marmara Bölgesi Örneği 2016-2020

Abstract

Düzensiz nüfus artışı, göç hareketliliği ve insanların vejetasyon dinamiklerine etkileri Arazi Kullanım ve Bitki Örtüsü (AKBÖ) değişimlerine yol açabilmektedir. AKBÖ değişiklikleri sanayi ile ilişkili kıyı bölgelerinde oldukça önemlidir. Türkiye’nin önemli kıyı alanlarından olan Güneydoğu Marmara alanı da çevredeki değişimlerden etkilenmektedir. Çalışma alanı, Sentinel-2 tabanlı bitki örtüsü indeksleri kombinasyonlarını kullanarak gerek AKBÖ değişimini gerekse sınıflandırmanın doğruluğunu belirlemek amacıyla seçilmiştir. Çalışma alanında Gemlik- Bursa Kuzey Kavşağı yatırım alanı ve yeni inşa edilen TOGG (Türkiye'nin Otomobili Girişim Grubu) fabrikası yer almaktadır. Çalışma alanı, Yalova ili Armutlu ilçesi ve Bursa ili Osmangazi, Mudanya ve Gemlik ilçelerini kapsayan alanda kıyıdan anakaraya 5 km’lik tampon bölge oluşturularak belirlenmiştir. Rastgele Orman (RO) sınıflandırma tekniği, 2016 ve 2020 yıllarında 3 sezon boyunca Sentinel-2 multispektral uydu görüntülerinden elde edilen indeksler kullanılarak orijinal bantlara ve 21 yeni bant kombinasyonuna uygulanmıştır. Sınıflandırma için kullanılan yeni bant kombinasyonları, normalize edilmiş bitki örtüsü indeksleri (NDVI), orijinal bantlar ve basit oran (SR) formülünden elde edilen bantlar eklenerek oluşturulmuştur. En yüksek doğruluk sonuçları 2016 yılı kış, ilkbahar ve yaz mevsimleri için OI12 (%82,93), ORF (%84,44) ve yine ORF (%84,67) indekslerinde gözlemlenirken, 2020 yılında OI5 (%85,89), ORF (%84,75) ve OI6 (%84,63) indekslerinde gözlemlenmiştir. Güneydoğu Marmara'da ulusal düzeyde alınan yatırım kararları bölgede nüfus artışına yol açmıştır. NDVI ve SR gibi orijinal bantlara spektral özelliklerin eklenmesiyle sınıflandırma doğruluğunda önemli bir değişiklik olmadığı gözlemlenmiş olsa da verilerin gelecekte farklı istatistiksel ve makine öğrenimi yöntemleriyle test edilmesinin sınıflama doğruluğunu daha fazla artırabilir.

Keywords

• Uzaktan Algılama
• AKBÖ
• Vejetasyon İndeksler
• Güney-Doğu Marmara Bölgesi

References

1. Ahamed, T., Tian, L., Zhang, Y. & Ting, K. C. (2011). A Review of Remote Sensing Methods for Biomass Feedstock Production. Biomass and Bioenergy, 35. https://doi.org/10.1016/j.biombioe.2011.02.028
2. Akturk, E. & Altunel, A. O. (2019). Accuracy Assesment of a Low-Cost UAV Derived Digital Elevation Model (DEM) in a Highly Broken and Vegetated Terrain. Measurement, 136, 382-386. https://doi.org/10.1016/j.measurement.2018.12.101
3. Asci, E., Inalpulat, M. & Genc, L. (2021). Identification of Residential Development Impacts on Agrıcultural Lands Using Landsat Imageries: Case Study of Bursa, Nilufer (1990-2020). III. Balkan Agricultural Congress (AGRIBALKAN), Edirne, Turkey.
4. Baeza, S. & Paruelo, J. M. (2020). Land Use/Land Cover Change (2000-2014) in The Rio De La Plata Grasslands: An Analysis Based on MODIS NDVI Time Series. Remote Sensing, 12(3). https://doi.org/10.3390/rs12030381
5. Barnes, E. M., Clarke, T. R., Richards, S. E., Colaizzi, P. D., Haberland, J., Kostrzewski, M., Waller , P., Choi, C., Rilye, E., Thomson, T., Lascano, R. J., Li, H. & Moran, M. S. (2000). Coincident Detection of Crop Water Stress, Nitrogen Status and Canopy Density Using Ground Based Multispectral Data. Proc. 5th Int. Conf. Precis Agric, 1619(6).
6. Batunacun, Nendel, C., Hu, Y. & Lakes, T. (2018). Land-Use Changea and Land Degradation on The Mongolian Plateau from 1975 To 2015—A Case Study From Xilingol, China. Land Degradation and Development, 29(6), 1595–1606. https://doi.org/10.1002/LDR.2948
7. Breiman, L. (2001). Random Forests. Machine Learning, 45, 5-32. https://doi.org/10.1023/A:1010933404324
8. Cavur, M., Duzgun, H. S., Kemec, S. & Demirkan, D. C. (2019). Land Use and Land Cover Classification of Sentinel-2A: St.Petersburg Case Study. International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives, 42(1/W2). https://doi.org/10.5194/isprs-archives-XLII-1-W2-13-2019

9. Chan, J. C. W. & Paelinckx, D. (2008). Evaluation of Random Forest and Adaboost Tree-Based Ensemble Classification and Spectral Band Selection for Ecotope Mapping Using Airborne Hyperspectral Imagery. Remote Sensing of Environment, 112(6), 2999–3011. https://doi.org/10.1016/J.RSE.2008.02.011
10. Chandra Pandey, P., Koutsias, N., Petropoulos, G. P., Srivastava, P. K., & Ben Dor, E. (2019). Land Use/Land Cover in View of Earth Observation: Data Sources, Input Dimensions, and Classifiers-A Review of the State of The Art. Geocarto International, 36, 957–988. https://doi.org/10.1080/10106049.2019.1629647
11. Chehata, N., Guo, L. & Forests, R. (2009). Airborne Lidar Feature Selection for Urban Classification Using Random Forests. Laser Scanning, XXXVIII(3/W8), Paris, France.
12. Colkesen, I., Ozturk, M. Y., Kavzoglu, T. & Sefercik, U. G. (2021). Determination of sea surface mucilage formations using multitemporal Sentinel-2 imagery. In Proceedings of the the 42nd Asian Conference on Remote Sensing (ACRS2021), Can Tho City, Vietnam, 22-24.
13. Demarchi, L., Canters, F., Cariou, C., Licciardi, G. & Chan, J. C. W. (2014). Assessing The Performance of Two Unsupervised Dimensionality Reduction Techniques on Hyperspectral APEX Data for High Resolution Urban Land-Cover Mapping. ISPRS Journal of Photogrammetry and Remote Sensing, 87. https://doi.org/10.1016/j.isprsjprs.2013.10.012
14. Dewidar, K. M. (2010). Detection of Land Use/Land Cover Changes for the Northern Part of The Nile Delta (Burullus Region), Egypt. International Journal of Remote Sensing, 25(20), 4079–4089. https://doi.org/10.1080/01431160410001688312
15. El-naggar, A. M. (2018). Determination of optimum segmentation parameter values for extracting building from remote sensing images. Alexandria engineering journal, 57(4), 3089-3097. https://doi.org/10.1016/j.aej.2018.10.001
16. Genc, L. (2002). Comparison of Landsat MSS and TM imagery for long term forest land cover change assessment. Doctoral Thesis, University of Florida, USA, 177p (in English).
17. Ghimire, B., Rogan, J., & Miller, J. (2010). Contextual Land-Cover Classification: Incorporating Spatial Dependence in Land-Cover Classification Models Using Random Forests and The Getis Statistic. Remote Sensing Letters, 1(1), 45–54. https://doi.org/10.1080/01431160903252327
18. Gitelson, A. A., Kaufman, Y. J., Stark, R. & Rundquist, D. (2002). Novel Algorithms for Remote Estimation of Vegetation Fraction. Remote Sensing of Environment, 80(1). https://doi.org/10.1016/S0034-4257(01)00289-9
19. Goel, E. & Abhilasha, E. (2017). Random Forest: A Review. International Journal of Advanced Research in Computer Science and Software Engineering. https://doi.org/10.23956/ijarcsse/v7i1/01113
20. Guan, H., Yu, J., Li, J. & Luo, L. (2012). Random Forests-Based Feature Selection For Land-Use Classification Using Lidar Data and Orthoimagery. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, XXXIX-B7. https://doi.org/10.5194/isprsarchives-xxxix-b7-203-2012
21. Herrmann, I., Pimstein, A., Karnieli, A., Cohen, Y., Alchanatis, V. & Bonfil, D. J. (2011). LAI Assessment of Wheat and Potato Crops by Venμs and Sentinel-2 Bands. Remote Sensing of Environment, 115(8). https://doi.org/10.1016/j.rse.2011.04.018
22. Hütt, C., Koppe, W., Miao, Y. & Bareth, G. (2016). Best Accuracy Land Use/Land Cover (LULC) Classification To Derive Crop Types Using Multitemporal, Multisensor, and Multi-Polarization SAR Satellite Images. Remote Sensing. https://doi.org/10.3390/rs8080684
23. Jamali, A. & Abdul Rahman, A. (2019a). Evaluation of Advanced Data Mining Algorithms in Land Use/Land Cover Mapping. International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives, 42(4/W16). https://doi.org/10.5194/isprs-archives-XLII-4-W16-283-2019
24. Jamali, A. & Abdul Rahman, A. (2019b). Sentinel-1 Image Classification For City Extraction Based on The Support Vector Machine and Random Forest Algorithms. International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives, 42(4/W16). https://doi.org/10.5194/isprs-archives-XLII-4-W16-297-2019
25. Joshi, R. R., Warthe, M., Dwivedi, S., Vijay, R. & Chakrabarti, T. (2011). Monitoring Changes in Land Use Land Cover of Yamuna Riverbed in Delhi: A Multi-Temporal Analysis. International Journal of Remote Sensing, 32(24), 9547–9558. https://doi.org/10.1080/01431161.2011.565377
26. Radhika, K. & Varadar, S. (2016). A Tutorial on Classification of Remote Sensing Data. International Research Journal of Engineering and Technology (IRJET), 3(8), 881-885.
27. Kavzoglu, T., Colkesen, I. & Yomralioglu, T. (2015). Object-Based Classification with Rotation Forest Ensemble Learning Algorithm Using Very-High-Resolution Worldview-2 Image. Remote Sensing Letters, 6(11). https://doi.org/10.1080/2150704X.2015.1084550
28. Kumar, V. & Agrawal, S. (2019). Agricultural Land Use Change Analysis Using Remote Sensing And GIS: A Case Study of Allahabad, India. International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives, 42(3/W6), 397–402. https://doi.org/10.5194/ISPRS-ARCHIVES-XLII-3-W6-397-2019
29. Le Maire, G., François, C. & Dufrêne, E. (2004). Towards Universal Broad Leaf Chlorophyll Indices Using PROSPECT Simulated Database and Hyperspectral Reflectance Measurements. Remote Sensing of Environment, 89(1). https://doi.org/10.1016/j.rse.2003.09.004
30. Main, R., Cho, M. A., Mathieu, R., O’Kennedy, M. M., Ramoelo, A. & Koch, S. (2011). An Investigation Into Robust Spectral Indices for Leaf Chlorophyll Estimation. ISPRS Journal of Photogrammetry and Remote Sensing, 66(6). https://doi.org/10.1016/j.isprsjprs.2011.08.001
31. Meinel, G. & Neubert, M. (2004). A Comparison of Segmentation Programs for High Resolution Remote Sensing Data. International Archives of Photogrammetry and Remote Sensing, 35(Part B), 1097-1105.
32. Mendoza, G. A. & Martins, H. (2006). Multi-Criteria Decision Analysis in Natural Resource Management: A Critical Review of Methods and New Modelling Paradigms. Forest Ecology and Management, 230, 1–22. https://doi.org/10.1016/j.foreco.2006.03.023
33. Mukhawana, M. B., Kanyerere, T. & Kahler, D. (2023). Review of In-Situ and Remote Sensing-Based Indices and Their Applicability for Integrated Drought Monitoring in South Africa. Water, 15. https://doi.org/10.3390/w15020240
34. Myint Htun, A., Shamsuzzoha, M. & Ahamed, T. (2023). Rice Yield Prediction Model Using Normalized Vegetation and Water Indices from Sentinel-2A Satellite Imagery Datasets. Asia-Pacific Journal of Regional Science, 7, 491–519. https://doi.org/10.1007/s41685-023-00299-2
35. Pal, M. (2005). Random Forest Classifier for Remote Sensing Classification. International Journal of Remote Sensing, 26(1), 217–222. https://doi.org/10.1080/01431160412331269698
36. Panigrahy, R. K., Ray, S. S. & Panigrahy, S. (2009). Study on the utility of IRS-P6 AWIFS SWIR band for crop discrimination and classification. Journal of the Indian Society of Remote Sensing, 37, 325-333. https://doi.org/10.1007/s12524-009-0026-6
37. Penuelas, J., Pinol, J., Ogaya, R., Filella, I., Pen Ä Uelas, J., Pin, J. & Ol, Ä. (1997). Estimation of Plant Water Concentration By The Reflectance Water Index WI (R900/R970). International Journal of Remote Sensing, 18(13), 2869–2875. https://doi.org/10.1080/014311697217396
38. Perumal, K. & Bhaskaran, R. (2010). Supervised Classification Performance of Multispectral Images. Journal of Computing, 2(2), 124-129.
39. Phiri, D., Simwanda, M., Salekin, S., Nyirenda, V. R., Murayama, Y. & Ranagalage, M. (2020). Sentinel-2 Data for Land Cover/Use Mapping: A Review. Remote Sensing, 12(14). https://doi.org/10.3390/rs12142291
40. Rodriguez-Galiano, V. F., Ghimire, B., Rogan, J., Chica-Olmo, M. & Rigol-Sanchez, J. P. (2012). An Assessment Of The Effectiveness Of A Random Forest Classifier For Land-Cover Classification. ISPRS Journal of Photogrammetry and Remote Sensing, 67(1), 93–104. https://doi.org/10.1016/J.ISPRSJPRS.2011.11.002
41. Sathian, S. & Brema, J. (2023). Assessment of Vegetative Cover Dynamics During Pre and Post Covid-19 Period Using Sentinel-2A Imageries in the Western Ghats, South India. Journal of Metrology Society of India, 14. https://doi.org/10.1007/s12647-023-00683-5
42. Scornet, E. (2015). Random Forests and Kernel Methods. IEEE Transactions on Information Theory, 62(3), 1485-1500. https://doi.org/10.1109/TIT.2016.2514489
43. Sharma, R. & Joshi, P. K. (2016). Mapping the Environmental Impacts Of Rapid Urbanization in The National Capital Region of India Using Remote Sensing Inputs. Urban Climate, 15(2016), 70-82. https://doi.org/10.1016/j.uclim.2016.01.004
44. Tesfaye, A. A. & Gessesse Awoke, B. (2021). Evaluation of The Saturation Property of Vegetation Indices Derived from Sentinel-2 in Mixed Crop-Forest Ecosystem. Spatial Information Research, 29, 109-121. https://doi.org/10.1007/s41324-020-00339-5
45. Tucker, C. J. (1979). Red and Photographic Infrared Linear Combinations for Monitoring Vegetation. Remote Sensıng of Environment, 8, 127–150. https://doi.org/10.1016/0034-4257(79)90013-0
46. Tucker, C. J. (1980). Remote Sensing of Leaf Water Content in the Near İnfrared. Remote Sensing of Environment, 10(1), 23–32. https://doi.org/10.1016/0034-4257(80)90096-6
47. Whiteside, T. G., Boggs, G. S. & Maier, S. W. (2011). Comparing Object-Based and Pixel-Based Classifications for Mapping Savannas. International Journal of Applied Earth Observation and Geoinformation, 13(6), 884–893. https://doi.org/10.1016/j.jag.2011.06.008
48. Wu, C., Niu, Z., Tang, Q. & Huang, W. (2008). Estimating Chlorophyll Content from Hyperspectral Vegetation Indices: Modeling and Validation. Agricultural and Forest Meteorology, 148(8–9). https://doi.org/10.1016/j.agrformet.2008.03.005
49. Wu, T., Luo, J., Gao, L., Sun, Y., Dong, W., Zhou, N., Liu, W., Hu, X., Xi, J., Wang, C. & Yang, Y. (2021). Geo-Object-Based Vegetation Mapping via Machine Learning Methods with an Intelligent Sample Collection Scheme: A Case Study of Taibai Mountain, China. Remote Sensing. https://doi.org/10.3390/rs13020249
50. Xianju, L., Gang, C., Jingyi, L., Weitao, C., Xinwen, C. & Yiwei, L. (2017). Effects of RapidEye Imagery’s Red-edge Band and Vegetation Indices on Land Cover Classification in an Arid Region. Chinese Geographical Science, 27(5), 827–835. https://doi.org/10.1007/s11769-017-0894-6
51. Yulianti, E. (2019). Multi-Temporal Sentinel-2 Images for Classification Accuracy. Journal of Computer Science, 15, 258–268. https://doi.org/10.3844/jcssp.2019.258.268
52. Zaidi, S. M., Akbari, A., Abu Samah, A., Kong, N. S. Gisen, J. I. A. (2017). Landsat-5 Time Series Analysis for Land Use/Land Cover Change Detection Using NDVI and Semi-Supervised Classification Techniques. Polish Journal of Environmental Studies, 26(6), 2833-2840. https://doi.org/10.15244/pjoes/68878
53. Zarco-Tejada, P. J., Miller, J. R., Noland, T. L., Mohammed, G. H. & Sampson, P. H. (2001). Scaling-Up and Model Inversion Methods with Narrowband Optical Indices for Chlorophyll Content Estimation in Closed Forest Canopies with Hyperspectral Data. IEEE Transactions on Geoscience and Remote Sensing, 39(7). https://doi.org/10.1109/36.934080
54. Zhang, T., Su, J., Liu, C., Chen, W. H., Liu, H., & Liu, G. (2017). Band selection in sentinel-2 satellite for agriculture applications. In 2017 23rd international conference on automation and computing (ICAC), Huddersfield, UK, 1-6.