Intercomparison Experiment of Water-Insoluble Carbonaceous Particles in Snow in a High-Mountain Environment (1598 m a.s.l.)

Abstract

The harmonization of sampling, sample preparation and laboratory analysis methods to detect carbon compounds in snow requires detailed documentation of those methods and their uncertainties. Moreover, intercomparison experiments are needed to reveal differences and quantify the uncertainties further. Here, we document our sampling, filtering, and analysis protocols used in the intercomparison experiment from three laboratories to detect water-insoluble carbon in seasonal surface snow in the high-mountain environment at Kolm Saigurn (47.067842° N, 12.98394° E, alt 1598 m a.s.l.), Austria. The participating laboratories were TU Wien (Austria), the University of Florence (Italy), and the Finnish Meteorological Institute (Finland). For the carbon analysis, the NIOSH5040 and EUSAAR2 protocols of the OCEC thermal-optical method were used. The median of the measured concentrations of total carbon (TC) was 323 ppb, organic carbon (OC) 308 ppb, and elemental carbon (EC) 16 ppb. The methods and protocols used in this experiment did not reveal large differences between the laboratories, and the TC, OC, and EC values of four inter-comparison locations, five meters apart, did not show meter-scale horizontal variability in surface snow. The results suggest that the presented methods are applicable for future research and monitoring of carbonaceous particles in snow. Moreover, a recommendation on the key parameters that an intercomparison experiment participant should be asked for is presented to help future investigations on carbonaceous particles in snow. The work contributes to the harmonization of the methods for measuring the snow chemistry of seasonal snow deposited on the ground.

Accuracy of Manual Snow Sampling, Depending on the Sampler’s Cross-Section—A Comparative Study

“…The device is estimated to have an error rate of 5-11%, depending on the snow conditions [26]. Previous field comparisons showed that the Federal Sampler yielded values that were 57% lower than the site's mean SWE, as measured by other devices [14]. This difference in the results between the two Federal Samplers and the density profile cutters compares very well with the differences reported by Leppänen et al [27].…”

“…Proksch et al [13] concluded that snow densities measured by different methods, including the different density cutters and micro-CT, agreed to within 9%. A previous comparative study of SWE samplers showed that the devices provided slightly different uncertainties since they are designed for use in different snow conditions [14]. The study showed that the uncertainty induced by instrumental bias was generally less than 10%.…”