Agriculture, Vol. 28 No. 1, 2022.
Original scientific paper
https://doi.org/10.18047/poljo.28.1.8
The Relationship of Environmental Factors and the Cropland Suitability Levels for Soybean Cultivation Determined by Machine Learning
Dorijan Radočaj
orcid.org/0000-0002-7151-7862
; Josip Juraj Strossmayer University of Osijek, Faculty of Agrobiotechnical Sciences Osijek, Vladimira Preloga 1, 31000 Osijek, Croatia
Tomislav Vinković
; Josip Juraj Strossmayer University of Osijek, Faculty of Agrobiotechnical Sciences Osijek, Vladimira Preloga 1, 31000 Osijek, Croatia
Mladen Jurišić
; Josip Juraj Strossmayer University of Osijek, Faculty of Agrobiotechnical Sciences Osijek, Vladimira Preloga 1, 31000 Osijek, Croatia
Mateo Gašparović
; University of Zagreb, Faculty of Geodesy, Kačićeva 26, 10000 Zagreb, Croatia
Abstract
The relationship between cropland suitability and the surrounding environmental factors has an important role in understanding and adjusting agricultural land management systems to natural cropland suitability. In this study, the relationship between soybean cropland suitability, determined by a novel machine learningbased approach, and three major environmental factors in continental Croatia was evaluated. These constituted of two major land cover classes (forests and urban areas), utilized soybean growth seasons per agricultural parcels during a 2017–2020 study period and soil types. The sensitivity analysis in geographic information system (GIS) using a raster overlay method, along with auxiliary spatial processing, was performed. The proximity of soybean agricultural parcels to
forests showed a high correlation with suitability values, indicating a potential benefit of implementing agroforestry in land management plans. A notable amount of suitable agricultural parcels for soybean cultivation, which were previously not utilized for soybean cultivation was observed. A disregard of crop rotations was also noted, with the same soybean parcels within the study period in three and four years. This analysis showed considerable potential in understanding the effects of environmental factors on cropland suitability values, leading to more efficient land
management policies and future suitability studies.
Keywords
land cover; crop rotation; soil types; land management; machine learning
Hrčak ID:
280043
URI
Publication date:
30.6.2022.
Visits: 1.487 *