This publication shows the results of the analysis of agricultural abandonment and recultivation within the western provinces of Ukraine, detected by classification of satellite images. For this research we defined agricultural abandonment as absence of any kind of agricultural utilization of land parcel for at least 4 years in 2007-2012 while it was cultivated for at least 5 years in previous 6-year period (2001-2006). To define agricultural recultivation, as a process of bringing back of fallow land, a reverse rule was applied than to abandonment. Polarization of agricultural land use in our study displays the hot-spot of land use change disparities when both agricultural abandonment and recultivation manifest within the same administrative district. Our findings suggest that the volume of abandoned agricultural lands within the study area in the period from 2007 to 2012 reached 200 thousand hectares, while the process of recultivation of former abandoned land in Western Ukraine was recorded only on about 70 thousand hectares. This means prevailing of agricultural abandonment due to less profitability of agriculture here in comparison to other parts of Ukraine. Underlying drivers of this general trend in land use might be the fall of cattle breeding in the region, which resulted in substantial decreasing of utilization of pastures and hay fields for as a sources of feed. Northern districts of Rivne and Volyn provinces experienced severe land abandonment whilst agricultural recultivation occurred the most on the south of Khmelnytskyi and east of Chernivtsi provinces. To analyze the proximate drivers of land use change we employed machine learning method, namely booster regression trees, which are a powerful non-parametric regression approach and can capture complex, non-linear relationships between response and predictor. The statistical models that were built using R software allowed us for the first time to investigate at the drivers of two opposite land use trends and showed a significant difference in the predictors that shape agricultural abandonment and recultivation. Accessibility variables played more important role in explaining agricultural recultivation. However, probability of land abandonment as was shown of our modeling substantially increase in the districts with high share of pasture and hay field.

Key words: farmland recultivation, farmland abandonment, boosted regression trees algorithm, land use polarization, spatial drivers.

Література (References):

  1. Земельний кодекс України : Кодекс України, Закон від 25.10.2001 № 2768-III [Електронний ресурс]. – Режим доступу:
  2. Картограми якісного стану ґрунтів України / Державна установа «Інститут охорони ґрунтів України» [Електронний ресурс]. – Режим доступу:
  3. Мапа зерносховищ України / Аграрна біржа України [Електронний ресурс]. – Режим доступу:
  4. Штовба С. Д. Інтелектуальні технології ідентифікації залежностей. Лабораторний практикум : електронний навчальний посібник / Штовба С.Д., Мазуренко В.В. – Вінниця : ВНТУ, 2014. – 113 с.
  5. Alcantara C. Mapping the extent of abandoned farmland in Central and Eastern Europe using MODIS time series satellite data / C. Alcantara, T. Kuemmerle, M. Baumann et al. // Environ. Res. Lett. – 2013. – Vol. 8. – No. 3. –
  6. Cramer V. A. What’s new about old fields? Land abandonment and ecosystem assembly / V.A. Cramer, R. J. Hobbs, R. J Standish // Trends Ecol. Evol. – 2008. – Vol. 23. – P. 104–112.
  7. Elith J. A working guide to boosted regression trees / J. Elith, J.R. Leathwick, T. Hastie // J. Anim. Ecol. – 2008. – Vol. 77. – P. 802–813.
  8. Estel S. Mapping farmland abandonment and recultivation across Europe using MODIS NDVI time series / S. Estel, T. Kuemmerle, C. Alcántara et al. // Remote Sens. Environ. – 2015. – Vol. 163. – P. 312–325.
  9. Friedman J.H. Greedy function approximation: a gradient boosting machine/ J.H. Friedman // Ann. Statist. – 2001. – Vol. 29. – P 1189–1232.
  10. Friedman J.H. Multiple additive regression trees with application in epidemiology / J.H. Friedman, J.J. Meulman // Stat. Med. – 2003. – Vol. 22. – P. 1365–1381.
  11. Griffiths P. Agricultural land change in the Carpathian ecoregion after the breakdown of socialism and expansion of the European Union / P. Griffiths, D. Müller, T. Kuemmerle, et al. // Environ. Res. Lett. – 2013. – Vol. 8. – No. 4. –
  12. Hartvigsen M. Land reform and land fragmentation in Central and Eastern Europe / M. Hartvigsen // Land Use Policy. – 2014. – Vol. 36. – P. 330–341.
  13. Hastie T. The Elements of Statistical Learning: Data Mining, Inference, and Prediction / Hastie T., Tibshirani R., Friedman J.; 2nd ed. – New York: Springer-Verlag, 2009 – pp. 745.
  14. Hengl T. SoilGrids1 km—global soil information based on automated mapping / T. Hengl, J.M. de Jesus, R.A. MacMillan et al. // PLoS One – 2014. – 9 (12): e105992.
  15. Hijmans R.J. Dismo: species distribution modeling / R.J. Hijmans., S. Phillips, J.R Leathwick et al. – 2013 [Electronic source]. – Access mode:
  16. Hijmans R.J. Very high resolution interpolated climate surfaces for global land areas / R.J. Hijmans, S.E. Cameron, J.L. Parra, P.G. Jones, A. Jarvis. // Int. J. Climatol. – 2005. – Vol. 25. – P. 1965–1978.
  17. Jarvis A. Hole-filled SRTM for the globe Version 4 [Електронний ресурс] / A. Jarvis, H.I. Reuter, A. Nelson, E. Guevara // CGIAR-CSI SRTM 90m Database, 2008. – Режим доступу:
  18. Kraemer R. Long-term agricultural land-cover change and potential for cropland expansion in the former Virgin Lands area of Kazakhstan / R. Kraemer, A.V. Prishchepov, D. Müller et al. // Environmental Research Letters. – – Vol. 10. – 054012.
  19. Kuemmerle T. Cross-border comparison of land cover and landscape pattern in Eastern Europe using a hybrid classification technique / T. Kuemmerle, P. Hostert, K. Perzanowski et al. // Remote Sens. Environ. – 2006. – Vol. 103. – P. 449-464.
  20. Landscapes in transition: An account of 25 years of land cover change in Europe / European Environment Agency. – 2017 – 84 pp.
  21. Laurance W. F. Agricultural expansion and its impacts on tropical nature / W.F. Laurance, J. Sayer, K.G. Cassman // Trends Ecol. Evol. – 2014. – Vol. 29. – P. 107–116.
  22. Müller D. Changing rural landscapes in Albania: cropland abandonment and forest clearing in the postsocialist transition/ D. Müller, D.K. Munroe. // Ann. Assoc. Am. Geogr. – 2008. – Vol. 98. – P. 855–876.
  23. Müller D. Comparing the determinants of cropland abandonment in Albania and Romania using boosted regression trees / D. Müller, P.J. Leitão, T. Sikor // Agric. Syst. – 2013. – Vol. 117. – P. 66–77.
  24. Müller D. Lost in transition: determinants of post-socialist cropland abandonment in Romania / D. Müller, T. Kuemmerle, M. Rusu et al. // J. Land Use Sci. – 2009. – Vol. 4. – P. 109–129.
  25. Prishchepov A.V. Determinants of agricultural land abandonment in post-Soviet European Russia / A.V. Prishchepov, D. Müller, M. Dubinin et al. // Land Use Policy. – 2013. – Vol. 30. – P. 873–884.
  26. Smaliychuk A. Recultivation of abandoned agricultural lands in Ukraine: patterns and drivers // A. Smaliychuk, D. Müller, A. V. Prishchepov et al. // Global Environmental Change. – 2016. – Vol. 38. – P. 70-81.
  27. Stefanski J. Mapping and monitoring of land use changes in post-Soviet western Ukraine using remote sensing data / J. Stefanski, O. Chaskovskyy, B. Waske // Applied Geogaphy. – 2014. –Vol. 55. – P. 155–164.
  28. Zomer R.J. Climate change mitigation: a spatial analysis of global land suitability for clean development mechanism afforestation and reforestation / R.J. Zomer, A. Trabucco, D.A Bossio et al. // Agric. Ecosyst. Environ. – 2008. –Vol. 126. – P. 67–80.