Hungarian Geographical Bulletin

Evaluation of the applicability of potential evapotranspiration models in Hungary

2026-04-01

Marcell Imre, Noémi Sarkadi, Ervin Pirkhoffer, Szabolcs Czigány

One of the most challenging problems in hydrometeorology is the quantification of potential evapotranspiration (PET) rates. The aim of this study was to identify PET models that can reliably approximate the FAO Penman– Monteith reference evapotranspiration or available PET data provided by the Hungarian Meteorological Service (HungaroMet) for Hungary while requiring fewer meteorological input variables. Nevertheless, an understanding of PET values and trends can offer invaluable insights into the drought sensitivity of an area. We analysed the performance of 18 PET models for Hungary based on meteorological data from 2010 to 2022 and identified and ranked the most relevant ones. The PET values were calculated at 16 meteorological stations using different models and subsequently ranked according to six distinct statistical indicators. As a basis for comparison, data from the nearest pan-evaporation measuring station and FAO Penman-Monteith (FAO-PM) values were calculated. PET provided by HungaroMet was used as the reference potential evaporation value. Model performances were ranked on a 1–120 scale. Our results showed that the temperature-based Oudin model had the most accurate performance, but in general, the radiation-based models were the most reliable. The spatial distribution of the data indicates that the performance of the PET models is somewhat inferior in the eastern and western regions of the country in comparison to that observed in the central areas. Our results are likely applicable to the temperate zone of similar subhumid climates.

Slope-driven edge analysis of high-resolution LiDAR data for automated detection of cultural terraces in Slovenia

2026-04-01

Lenart Štaut, Rok Ciglič, Blaž Repe

Cultural terraces were often constructed to improve agriculture. Some terraces are still in use, while others have been abandoned. Knowledge of their locations is important for their preservation or potential reuse. There have been several attempts worldwide to create a register of terraces. In Slovenia, a suitable register has not yet been created due to heavy overgrowth and significant differences in cultural terrace types across different regions of the country. This research proposes detecting terraces using a LiDAR digital elevation model, geoinformation tools, and additional spatial data. The method detects sharp changes in slope data and creates polygons where such changes are detected in close proximity. The main advantage of the method is that it does not require any training samples yet still provides accurate results despite the diversity of terraced areas. We applied the method in Slovenia and achieved an accuracy of 91 percent, a precision of 76 percent, and a recognition value of 66 percent in one test area, and 92, 47, and 65 percent in another designated test area. To achieve higher accuracy, the input settings can be adapted to regional characteristics, which confirms earlier findings that terraces in Slovenia exhibit high diversity.

Manuscript reviewers 2023–2025

2026-03-30

Ferenc Gyuris