Testing of the consistency of the sieving (wash-over) process of waterlogged sediments by multiple operators

Abstract

The sieving process has a considerable influence on the subsequently retrieved archaeobotanical data. As known from earlier work, the wash-over method is the most suitable method to extract plant macroremains from waterlogged sediments. This paper presents an experiment in which it was tested if different sievers using this method produced comparable results.Some systematic differences between sievers were found in the larger fractions (≥2mm), namely the varying presence of small remains. This problem can be avoided if detailed instructions are given to the sievers and guidelines for counting remains are used during analysis. In the small fraction (>0.35mm), differences between sievers were not substantial anymore. In addition to differences caused by the sieving technique we could also show that the patchy pattern of clumpy waterlogged sediments complicates a statistically relevant subsampling. We can state that only large differences between samples should be interpreted in palaeoeconomic terms, but that it is no disadvantage if several sievers work on the same project.It is our purpose to raise awareness of the fact that the methodology has a strong impact on the results obtained and should therefore always be revealed on a detailed level, especially if data from one site will later be used for comparisons with other sites.