Porównanie oceny zwarcia drzewostanu za pomocą metody wizualnej i zwarciomierza

摘要

W fitosocjologii ocena zwarcia drzewostanu należy do metod, za pomocą których badane są cechy analityczne fitocenoz leśnych. Pośrednio mówi ona o warunkach świetlnych panujących na dnie lasu. Tradycyjnie zwarcie warstwy drzew oceniane jest subiektywnie przez badacza, który szacuje, jaki procent nad powierzchnią zdjęcia fltosocjologicznego zajmują korony drzew [Pawłowski i in. 1977]. Metoda ta jest szybka i tania, lecz może być obarczona trudno mierzalnym błędem. Ocena wizualna zwarcia warstwy drzew, powszechnie stosowana w badaniach fitosocjologicz-nych, należy do często krytykowanych [Kercher i in. 2003; Klimes 2003; Vittoz, Guisan 2007], dlatego też podejmowane są próby stworzenia alternatywnej, bardziej obiektywnej i równie nieskomplikowanej metody do powszechnego zastosowania [Jennings i in. 1999].%The visual estimates of tree canopy cover belong to the fundamental measurements of analytical features of forest phytocoenoses. However, it is the subjective method, which results in error difficult to control. The objective of the research was to present a canopyscope and to compare results of canopy cover estimation using it with visual estimates. We conducted several experiments in oak-hornbeam, pine and spruce forest stands in southern Poland. We engaged in fieldwork well- and less-experienced students, i.e. biologists skilled in methods of phytosociology and students of nature engineering as well as three professional phytosociologists. They performed a comparative research in terms of correlation analyses, analysis of coefficients of variation and intra-class correlation (ICG) between-observers and between-methods using both ways of canopy cover estimates. The correlation between canopyscope score and results of visual estimates was highest in the group of phytosociologists (rs=0.82, p<0.001) followed by well-experienced students (rs=0.72, p<0.001) and less-experienced persons (rs=0.62, p<0.001). For purpose of comparison of both methods two variants of visual method were applied: optional visual cover estimate, and adjusted visual method where percentage cover of tree layer was noted using 4% intervals likewise in canopyscope method. The results of visual estimate method were characterized by more than threefold higher variation (CV%=9.72 and GV%=9.46 for optional variant and adjusted variant respectively) in comparison with canopy-scope method (CV%=2.26) in repeating measurements on the same plot. In terms of repeatability, analysis yielded 0.873 (p<0.0001) and 0.622 (p=0.00682) values of ICC for canopyscope and adjusted visual estimate respectively between two phytosociologists. The result of intra-class correlation for comparison between methods within each rater separately revealed low and non-significant value of ICC. We suggest using canopyscope because this method is more precise, reliable and repeatable than visual estimation. Moreover, canopyscope is easier and more convenient to use when compared to objective methods as hemispherical photography of tree canopy and image analysis.