Gathering predictors of biodiversity change and reconstructing land cover history in Central Apennines using machine learning and remote sensing data

Abstract

The selection of methods for image processing and software functionality is crucial for monitoring Earth's landscapes. This work presents the use of Machine Learning (ML) methods for remote sensing (RS) data processing. The aim is to perform cartographic analysis of land cover changes with a case of central Apennines, Italy. Technically, we present a ML-based classification method using GRASS GIS software integrated with Python library Scikit-Learn. Image processing using ML methods was investigated by employing the algorithms of GRASS GIS. The data are obtained from the United States Geological Survey (USGS) and include a time series of Landsat 8-9 OLI/TIRS satellite images. The operational workflow of image processing includes RS data processing. The images were classified into raster maps with automatically detected categories of land cover types. The approach was implemented by using a set of modules in scripting language of GRASS GIS, including for non-supervised classification used as training dataset of random pixel seeds. The ML classifiers were used to detect changes in land cover types derived from images. The results show different vegetation conditions in spring and autumn periods. Unlike the existing methods of image classification, ML considers the differences among the spectral reflectance of pixels when modelling topology of patches. Other advantages are that ML uses data on texture and spectral features to measure the similarity of neighbouring landscape patches during the process of generating random decision trees. This study demonstrated the benefits of ML for cartography, RS data processing and geoinformatics.

Keywords

• Artificial Neural Network
• Decision Tree
• Geospatial Data
• Ecology
• Random Forest
• GIS
• Italy

Makine öğrenimi ve uzaktan algılama verilerini kullanarak Orta Apeninler'deki biyolojik çeşitlilik değişiminin tahmin edicilerini toplamak ve manzara geçmişini yeniden oluşturmak

Öz

Görüntü işleme ve yazılım işlevselliği için yöntemlerin seçimi, Dünya manzaralarını izlemek için çok önemlidir. Bu çalışma, uzaktan algılama (UA) veri işleme için Makine Öğrenmesi (MÖ) yöntemlerinin kullanımını sunmaktadır. Amaç, İtalya'nın merkezi Apeninler örneği ile arazi örtüsü değişikliklerinin kartografik analizini gerçekleştirmektir. Teknik olarak, Python kütüphanesi Scikit-Learn ile entegre GRASS CBS yazılımını kullanan bir MÖ tabanlı sınıflandırma yöntemi sunuyoruz. MÖ yöntemlerini kullanarak görüntü işleme, GRASS CBS algoritmaları işletim araştırılmıştır. Veriler Amerika Birleşik Devletleri Jeolojik Araştırması (ABDJA)'den elde edilmiştir ve Landsat 8-9 OLI/TIRS uydu görüntülerinin bir zaman serisini içerir. Görüntü işlemenin operasyonel iş akışı, UA veri işlemeyi içerir. Görüntüler, otomatik olarak algılanan arazi örtüsü türü kategorileriyle raster haritalara sınıflandırılmıştır. Yaklaşım, rastgele piksel tohumlarının eğitim veri kümesi olarak kullanılan gözetimsiz sınıflandırma dahil olmak üzere GRASS CBS'in betik dilinde bir dizi modül kullanılarak uygulanmıştır. MÖ sınıflandırıcıları, görüntülerden türetilen arazi örtüsü türlerindeki değişiklikleri tespit etmek için kullanılmıştır. Sonuçlar, ilkbahar ve sonbahar dönemlerinde farklı bitki örtüsü koşullarını göstermektedir. Mevcut görüntü sınıflandırma yöntemlerinden farklı olarak, MÖ, yamaların topolojisini modellerken piksellerin spektral yansıması arasındaki farkları dikkate alır. Diğer avantajlar, ML'nin rastgele karar ağaçları oluşturma süreci sırasında komşu manzara yamalarının benzerliğini ölçmek için doku ve spektral özelliklerle ilgili verileri kullanmasıdır. Bu çalışma, MÖ'nin kartografi, UA veri işleme ve jeoenformatik için faydalarını göstermiştir.

Anahtar Kelimeler

• İtalya
• Yapay Sinir Ağı
• Karar Ağacı
• Coğrafi Veri
• Ekoloji
• Rastgele Orman
• CBS
• İtalya

Kaynakça

1. Agnoletti, M. (2007). The degradation of traditional landscape in a mountain area of Tuscany during the 19th and 20th centuries: Implications for biodiversity and sustainable management. Forest Ecology and Management, 249(1-2), 5–17. doi:10.1016/j.foreco.2007.05.032
2. Agnoletti, M., Emanueli, F., Corrieri, F., Venturi, M. & Santoro, A. (2019). Monitoring traditional rural landscapes. The case of italy. Sustainability, 11(21), 6107. doi:10.3390/su11216107
3. Antrop, M. (2005). Why landscapes of the past are important for the future. Landscape and Urban Planning, 70(1-2), 21–34. doi:10.1016/j.landurbplan.2003.10.002
4. Baldini, E. (2003). Notizie sull'olivicoltura Bolognese. Accademia Nazionale di agricoltura.
5. Ball, B.C. & Douglas, J.T. (2003). A simple procedure for assessing soil structural, rooting and surface conditions. Soil Use and Management, 19(1), 50–56. https://doi.org/10.1111/j.1475-2743.2003.tb00279.x
6. Barbati, A., R. Salvati, B. Ferrari, D. Di santo, A. Quatrini, L. Portoghesi, D. Travaglini, F. Iovino, & S. Nocentini. (2012). Assessing and promoting old-growthness of forest stands: Lessons from research in Italy. Plant Biosystems, 146, 167–174. https://doi.org/10.1080/11263504.2011.650730
7. Batey, T. (2009), Soil compaction and soil management – a review. Soil Use and Management, 25(4), 335-345. https://doi.org/10.1111/j.1475-2743.2009.00236.x
8. Boisvenue, C. & S. W. Running. 2006. Impacts of climate change on natural forest productivity—evidence since the middle of the 20th century. Global Change Biology, 12, 862–882. https://doi.org/10.1111/j.1365-2486.2006.01134.x

9. Brady, N. & R. Weil. 1999. The nature and properties of soils. Prentice-Hall.
10. Calfapietra, C., Barbati, A., Perugini, L., Ferrari, B., Guidolotti, G., Quatrini, A. & Corona, P. (2015). Carbon mitigation potential of different forest ecosystems under climate change and various managements in Italy. Ecosystem Health and Sustainability, 1(8), 1–9. https://doi.org/10.1890/EHS15-0023
11. Caparrini, F., Castelli, F. & Entekhabi, D. (2004). Estimation of surface turbulent fluxes through assimilation of radiometric surface temperature sequences. Journal of Hydrometeorology, 5(1), 145-159. https://doi.org/10.1175/1525-7541(2004)005<0145:EOSTFT>2.0.CO;2
12. Cavalli, M., Trevisani, S., Comiti, F. & Marchi, L. (2013). Geomorphometric assessment of spatial sediment connectivity in small Alpine catchments. Geomorphology, 188, 31–41 https://doi.org/10.1016/j.geomorph.2012.05.007
13. Ceotto, E. (1999). Exploring cropping systems of the low Po Valley using the systems approach [Unpublished master of science thesis]. Wageningen Agricultural University.
14. Corbari, C. & Mancini, M. (2014). Intercomparison across scales between remotely-sensed land surface temperature and representative equilibrium temperature from a distributed energy water balance model. Hydrological Sciences Journal, 59(10), 1830–1843. https://doi.org/10.1080/02626667.2014.946418
15. De Luca, A.I., Stillitano, T., Gulisano, G. & Toschi, T.G. (2024). A historical and social insight of olive groves and landscapes in Italy. In Muñoz-Rojas, J. & García-Ruiz, R. (Eds.), The olive landscapes of the Mediterranean: Vol. 36. Landscape Series (pp.133-141). Springer. https://doi.org/10.1007/978-3-031-57956-1_11
16. Diehr, J., Ogunyiola, A. & Dada, O. (2025). Artificial intelligence and machine learning-powered GIS for proactive disaster resilience in a changing climate. Annals of GIS, 31(2), 287-300. https://doi.org/10.1080/19475683.2025.2473596
17. Elshewy, M.A., Mohamed, M.H.A. & Refaat, M. (2024). Developing a soil salinity model from landsat 8 satellite bands based on advanced machine learning algorithms. J Indian Soc Remote Sens, 52, 617–632. https://doi.org/10.1007/s12524-024-01841-1
18. Isaia, M., Siniscalco, C. & Badino, G. (2014). From rural to urban: Landscape changes in north-west Italy over two centuries. Landscape History, 35(1), 73–76. https://doi.org/10.1080/01433768.2014.916914
19. Khan, Q., Liaqat, M. U. & Mohamed, M. M. (2021). A comparative assessment of modeling groundwater vulnerability using DRASTIC method from GIS and a novel classification method using machine learning classifiers. Geocarto International, 37(20), 5832–5850. https://doi.org/10.1080/10106049.2021.1923833
20. Klaučo, M., Gregorová, B., Koleda, P., Stankov, U., Marković, V. & Lemenkova, P. (2017). Land planning as a support for sustainable development based on tourism: A case study of Slovak rural region. Environmental Engineering and Management Journal, 16(2), 449-458. https://doi.org/10.30638/eemj.2017.045
21. Klaučo, M., Gregorová, B., Stankov, U., Marković, V. & Lemenkova, P. (2013). Determination of ecological significance based on geostatistical assessment: A case study from the Slovak natura 2000 protected area. Open Geosciences, 5(1), 28-42. https://doi.org/10.2478/s13533-012-0120-0
22. Lemenkova, P. (2019). Generic mapping tools and matplotlib package of python for geospatial data analysis in marine geology. International Journal of Environment and Geoinformatics (IJEGEO), 6(3). 225-237. https://doi.org/110.30897/ijegeo.567343
23. Lemenkova, P. (2020). Using R packages 'tmap', 'raster' and 'ggmap' for cartographic visualization: An example of dem-based terrain modelling of Italy, Apennine Peninsula. Zbornik radova - Geografski fakultet Univerziteta u Beogradu, 68, 99-116. https://doi.org/10.5937/zrgfub2068099L
24. Lemenkova, P. (2021). Distance-based vegetation indices computed by SAGA GIS: A comparison of the perpendicular and transformed soil adjusted approaches for the LANDSAT TM image. Poljoprivredna tehnika, 46(3), 49-60. https://doi.org/10.5937/PoljTeh2103049L
25. Lemenkova, P. (2023). Using open-source software GRASS GIS for analysis of the environmental patterns in Lake Chad, Central Africa. Die Bodenkultur: Journal of Land Management, Food and Environment, 74(1), 49-64. https://doi.org/10.2478/boku-2023-0005
26. Lemenkova, P. (2024a). Random forest classifier algorithm of geographic resources analysis support system geographic information system for satellite image processing: Case study of Bight of Sofala, Mozambique. Coasts, 4(1), 127-149. https://doi.org/10.3390/coasts4010008
27. Lemenkova, P. (2024b). Support vector machine algorithm for mapping land cover dynamics in Senegal, West Africa, using earth observation data. Earth, 5(3), 420-462. https://doi.org/10.3390/earth5030024 Lemenkova, P. (2024c). Approche cartographique par le SIG GRASS pour l'analyse de la structure du paysage au Libéria, Afrique de l'Ouest. Dynamiques environnementales, 53, 1-36. https://doi.org/10.4000/12n0l
28. Lemenkova, P. (2025a). Improving bimonthly landscape monitoring in Morocco, North Africa, by integrating machine learning with GRASS GIS. Geomatics, 5(1), 5. https://doi.org/10.3390/geomatics5010005 Lemenkova, P. (2025b). Automation of image processing through ML algorithms of GRASS GIS using embedded scikit-learn library of python. Examples and Counterexamples, 7, 100180. https://doi.org/10.1016/j.exco.2025.100180
29. Lemenkova, P. (2025c). Land cover analysis in the Yangtze River Basin for detection of wetland agriculture and urban dynamics in Wuhan area (China). Transylvanian Review of Systematical and Ecological Research, 27(1), 1-16. https://doi.org/10.2478/trser-2025-0001
30. Lemenkova, P. & Debeir, O. (2023). Time series analysis of landsat images for monitoring flooded areas in the Inner Niger Delta, Mali. Artificial Satellites, 58(4), 278-313. https://doi.org/10.2478/arsa-2023-0011
31. Lindh, P. & Lemenkova, P. (2022). Permeability, compressive strength and proctor parameters of silts stabilised by Portland cement and ground granulated blast furnace slag (GGBFS). Archive of Mechanical Engineering, 69(4), 667-692. https://doi.org/10.24425/ame.2022.141522
32. Lindh, P. & Lemenkova, P. (2023). Effects of water—binder ratio on strength and seismic behavior of stabilized soil from Kongshavn, Port of Oslo. Sustainability, 15(15), 12016. https://doi.org/10.3390/su151512016 Mačić, V. (2014). Anatomical features of cymodocea nodosa growing in Montenegro (Adriatic Sea). Journal of Black Sea / Mediterranean Environment, 20(3), 253-263.
33. Manzo, G., Tofani, V., Segoni, S., Battistini, A. & Catani, F. (2012). GIS techniques for regional-scale landslide susceptibility assessment: the Sicily (Italy) case study. International Journal of Geographical Information Science, 27(7), 1433–1452. https://doi.org/10.1080/13658816.2012.693614
34. Maselli, F., Chiesi, M. & Bindi, M. (2004). Multi-year simulation of Mediterranean forest transpiration by the integration of NOAA-AVHRR and ancillary data. International Journal of Remote Sensing, 25(19), 3929–3941. https://doi.org/10.1080/01431160310001653546
35. Onofri, S., Anastasi, A., Del Frate, G., Di Piazza, S., Garnero, N., Guglielminetti, M., …& Zucconi, L. (2011). Biodiversity of rock, beach and water fungi in Italy. Plant Biosystems - An International Journal Dealing with All Aspects of Plant Biology, 145(4), 978–987. https://doi.org/10.1080/11263504.2011.633117
36. Peres, D. J., Bonaccorso, B., Palazzolo, N., Cancelliere, A., Mendicino, G. & Senatore, A. (2023). A dynamic approach for assessing climate change impacts on drought: An analysis in Southern Italy. Hydrological Sciences Journal, 68(9), 1213–1228. https://doi.org/10.1080/02626667.2023.2217332
37. Renda, W. & Gıacobbe, S. (2021). First report of Alvania scuderii Villari, 2017 (Gastropoda: Mollusca) from Tyrrhenian Sea: Some biogeographic implications. Aquatic Research, 4(2), 208-213. https://doi.org/10.3153/AR21016
38. Salvati, L. & Colantoni, A. (2013). Land use dynamics and soil quality in agro-forest systems: A country-scale assessment in Italy. Journal of Environmental Planning and Management, 58(1), 175–188. https://doi.org/10.1080/09640568.2013.849235
39. Salvati, L., De Zuliani, E., Sabbi, A., Cancellieri, L., Tufano, M., Caneva, G. & Savo, V. (2016). Land-cover changes and sustainable development in a rural cultural landscape of central Italy: Classical trends and counter-intuitive results. International Journal of Sustainable Development & World Ecology, 24(1), 27–36. https://doi.org/10.1080/13504509.2016.1193778
40. Tarolli, P. & Straffelini, E. (2020). Agriculture in hilly and mountainous landscapes: Threats, monitoring and sustainable management. Geography and Sustainability, 1(1), 70-76. https://doi.org/10.1016/j.geosus.2020.03.003
41. Tecilla, G. & Cosner, A. (2024). The atlas of terraced landscapes of Trentino, Italy. In Agnoletti, M., Dobričič, S., Matteini, T. & Palerm, J.M. (Eds.), Cultivating continuity of the European landscape: Vol. 15. Environmental History (pp. 465-475). Springer. https://doi.org/10.1007/978-3-031-25713-1_51
42. Vessella, F. (2024). The role of historical cartography to evaluate the landscape evolution in the last two centuries: A case study in central Italy. Landscape Research, 50(1), 129–145. https://doi.org/10.1080/01426397.2024.2377169