High resolution vegetation assessment with beta-diversity – a moving window approach

Abstract

Monitoring designs are often suffering from the inherent non-stationarity of the monitored systems. To overcome this limition, we propose a sampling designs based on high resolution mapping and spatial analyses with double spatial scaling process. Applying for vegetation, we record the presence of plant species along 26 m or 52 m long belt transects of 520 (or 1040) units of 0.5 cm x 0.5 cm contiguous microquadrats. Beta diversity (represented as the diversity of species combinations) is estimated by subsequent computerised samplings from the baseline transect data sets. Beta diversity is scaled with changing resolutions across a range of scales from 5 cm x 5 cm to 5 cm x 500 cm. Second, it is also scaled using moving window. Local maximum of beta diversity is repeatedly calculated in 5 m extent observational windows shifted along the transect with 1 m lag, and the spatial variability of vegetation is visualized by the related beta-diversity profile. Using field example, we demonstrate that beta diversity, when applied with our methodology, is a sensitive indicator, and it can reveal more information than alpha or gamma diversity.