Except for Sphagnum mosses of peatland habitats, reliable methods to assess moss productivity in arctic or boreal biomes give usually highly variable results. Therefore, ecosystem processes are poorly understood in these biomes where mosses are an important component of the system. The aim of this study was to compare three methods to estimate moss growth in polygon patterned fens: cranked wires, natural markers and artificial white marks (an alternative to the spray method). Precision of estimates was significantly higher when natural markers were used (coefficients of variation, CV, between 17 and 27%), compared to cranked wires (CV = 37%) or white marks (CV = 56%). Natural markers also provided estimates for growth of moss stems 32 to 113% higher than the other methods. Although cranked wires were calibrated shortly after snowmelt, some moss growth is still missed and consequently moss growth is underestimated. Accuracy of cranked wires was poor, mainly caused by frost heaving or permafrost activities that can affect wire position. Thus, this method should be avoided in arctic ecosystems. Even if white marks were painted on moss stems at the end of the growing season prior to the sampling year, lower estimates of moss growth were still found. We suspect some interference with moss growth processes during the marking process, at least when used with brown mosses. The natural marker method, which provides increment for an entire growing season, appears to be the most accurate method of the three. Additionally, it is also the easiest and the least time consuming method to use. Its main drawback is that relatively few species have natural growth marks and these species may not always be present among the targeted species under study. Also, measurements of stem growth on the same sample did not differ between observers, even if the second measurement was done 12 years later. In conclusion, when species with natural markers are present, this method should be used to assess moss growth. For arctic/sub-arctic studies where such species are lacking, the artificial white marks method should be refined further.