Моделовање Запремине Стабала Различитих Клонова У Засадима Топола На Подручју Војводине

The aim of this study was to develop mathematical models (tables) for calculating the volume of clones I-214, M1 and Deltoid in plantations on different soil types and at different planting distances in the alluvial plains of the Sava, Danube, Tisa and Tamis rivers in the Vojvodina region. Based on preliminary research, the samples were structured in such a way that their implementation would provide data for the development of models for calculating volume (and indirectly volume increment) within acceptable margins of error. However, as the data collection took place exclusively after the regular felling in the plantations of the mentioned clones without the possibility of a targeted selection of missing trees in certain treatment segments and due to time and spatial limitations in data collection after the storms of 2023, which caused significant damage to the forests in the region especially poplar plantations, as well as the need for their urgent rehabilitation, the planned sample was realized with 102.6% for I-214, 51.5% for M1 and 46.1% for Deltoid clones, with relatively poor distribution across river alluviums, soil types and diameter and height classes. Due to this sampling structure, it was not possible to build models (tables) at the level of clone, alluvium, soil type and plant spacing, but they were built based on the combined data for the whole area (Vojvodina), with certain exceptions - for I-214 (clone spacing and plant form) by applying the Schumacher-Hall linear mixed hierarchical (cluster) model; for M1 (clone) by applying classical regression analysis and for Deltoid clones (clona-alluvium of the Sava, alluvium of the Danube) by ap?plying the mixed linear hierarchical model, as in the case of I-214. Theoretically, the reliability of the developed volume tables strongly depends on the size and structure of the sample. However, their verification on real (not on empirical) trees and stands in practice will indicate the extent of the error and, in this context, on the degree of their usability.