Gap-filling strategies for annual VOC flux data sets

Abstract

Up to now the limited knowledge about the ex-change of volatile organic compounds (VOCs) between thebiosphere and the atmosphere is one of the factors whichhinders more accurate climate predictions. Complete long-term flux data sets of several VOCs to quantify the annualexchange and validate recent VOC models are basically notavailable. In combination with long-term VOC flux mea-surements the application of gap-filling routines is inevitablein order to replace missing data and make an importantstep towards a better understanding of the VOC ecosystem–atmosphere exchange on longer timescales.We performed VOC flux measurements above a mountainmeadow in Austria during two complete growing seasons(from snowmelt in spring to snow reestablishment in late au-tumn) and used this data set to test the performance of fourdifferent gap-filling routines, mean diurnal variation (MDV),mean gliding window (MGW), look-up tables (LUT) and lin-ear interpolation (LIP), in terms of their ability to replacemissing flux data in order to obtain reliable VOC sums. Ac-cording to our findings the MDV routine was outstandingwith regard to the minimization of the gap-filling error forboth years and all quantified VOCs. The other gap-fillingroutines, which performed gap-filling on 24 h average val-ues, introduced considerably larger uncertainties. The errorwhich was introduced by the application of the different fill-ing routines increased linearly with the number of data gaps.Although average VOC fluxes measured during the winterperiod (complete snow coverage) were close to zero, thesewere highly variable and the filling of the winter period re-sulted in considerably higher uncertainties compared to theapplication of gap-filling during the measurement period.The annual patterns of the overall cumulative fluxes forthe quantified VOCs showed a completely different behaviour in 2009, which was an exceptional year due tothe occurrence of a severe hailstorm, compared to 2011.Methanol was the compound which, at 381.5 mg C m−2and449.9 mg C m−2, contributed most to the cumulative VOCcarbon emissions in 2009 and 2011, respectively. In contrastto methanol emissions, however, considerable amounts ofmonoterpenes (−327.3 mg C m−2)were deposited onto themountain meadow during 2009 caused by a hailstorm. Otherquantified VOCs had considerably lower influences on the annual patterns.