On the performances of empirical regressions for the estimation of bulk snow density

Abstract

Grain size variation across surfaces of three gravel bars was studied at the River Opava. Sediment sorting at the scale of individual gravel bars is still an imperfectly explained phenomenon. The aim of the study was to examine the relation of grain-size pattern to within bar position (lateral and longitudinal distance to thalweg and elevation above channel bottom) and vegetation cover. Grain size analysis was performed by combination of grid count and dry sieving of fine fractions at sample plots aligned to transects crossing the bar surfaces. Descriptive statistics have been computed (median, sorting, skewness, kurtosis) and data were analysed by means of correlation (Spearman rank correlation), principal component analysis (PCA), and redundancy analysis (RDA) to elucidate the effect of controlling variables upon the grain size spatial pattern. Median grain size was found to be weakly correlated to lateral and longitudinal distance to thalweg. Sediment coarsening with increasing elevation above the channel bottom, probably mediated by vegetation, was detected. Vegetation coverage proved to be a factor explaining much of the variability in data. Grain size median and sorting were both affected by vegetation coverage. RDA analysis revealed that vegetation coverage, elevation above the channel bottom, and lateral distance to thalweg were the variables most affecting the grain size pattern (altogether explaining 33.1% of data variability). However, the role of particular variables differed between gravel bars. Field evidence from the studied river reach suggests that variables controlling within bar grain size variability are strongly site specific.