Multi-annual crop maps reveal cropping patterns in the Vojvodina region (Serbia) Article Swipe

<p>Cropping patterns give useful information in agricultural<br>production from the perspective of logistics and pricing.<br>In this study, a machine learning classification model based<br>on Random Forest algorithm and Sentinel-2 imagery was used<br>to provide spatial maps of the crops in the Vojvodina region<br>(Serbia), a predominantly agricultural area with 77% cropland<br>and intensive production of arable crops. Ground truth data<br>on geolocation of the fields was collected during the seven-year<br>period (2016–2022) to train the classification algorithm. The<br>most common crops identified are wheat (together with barley),<br>maize, soybean, sugar beet, sunflower and rapeseed. Crop maps<br>resulting from the classification were analyzed. By multiplying<br>the pixel count by the size of each pixel the percentage of area<br>covered by most common crops within the region was calculated<br>annually and averaged across all years. Multi-annual trends<br>were also examined. For the sake of clarity, the most frequent<br>rotations were analyzed at both the county and municipality level.<br>The spatial distribution analysis highlighted variability in total<br>area coverage by certain crops in the region. While some crops<br>maintained a significant area coverage across all counties, others<br>exhibited localized cultivation. Maize occupies the most areas<br>in all counties (29–39%), however there are cases where wheat<br>is equally present as maize (26–33%) and others where wheat<br>is less cultivated in favor of soybean and sunflower. Counties<br>with the most arable area covered by sunflower (18–21%) are<br>in the eastern part of the Vojvodina region, experiencing slightly<br>drier conditions than counties in the western part of the region<br>where soybean takes up 22–30% of arable land. Lastly, sugar beet<br>and rapeseed together cover 2–9% of arable area in all counties<br>and are consequently less present in rotation patterns. Notable<br>variations in crop transition frequencies and spatial distribution<br>patterns were revealed among all counties in the study region.<br>Certain crop transitions demonstrated consistency across all<br>counties, indicating widespread adoption of specific rotation<br>schemes, others showed variability, reflecting local agronomic<br>practices. The transition between maize and wheat emerged as<br>the most frequent across all counties throughout the observed<br>years, however at varying scales. It is observed that the counties<br>in northern and southern parts of the region practice maizewheat<br>transition most frequently (18–32% of all transitions).<br>Sunflower in transition with maize and wheat is more present<br>in the eastern part of the region (> 20%), while soybean and<br>maize transition is the most prevalent in counties in central<br>and western Vojvodina (> 15%). Another notable observation<br>is that some municipalities within one county practice different<br>crop rotation patterns. A typical example is Juˇznobaˇcki county,<br>where western municipalities show more arable areas covered by<br>soybean, hence the more frequent transitions with maize. On the<br>other side, it is observed that the eastern parts of the same county<br>incorporate more wheat in rotating crops. The findings of this<br>study contribute to our understanding of crop rotation practices<br>in Vojvodina region which is important in guiding agricultural<br>management decisions and promoting sustainable agriculture.</p>