Priming is frequently overestimated in studies using 14C-labelled substrates due to underestimation of 14CO2 activity

Abstract

The priming effect defined as an increased turnover of soil organic matter (SOM) upon organic substrate addition to soils has been investigated for decades, but its underlying mechanisms remain elusive. It has been shown that 14CO2 trapped in NaOH can be rapidly released from the solution when mixing small volume, low molar or highly CO2-saturated NaOH samples with scintillation cocktails. Therefore, liquid scintillation counting can substantially underestimate the 14C-activity of such samples. In this study, we explore the implications of this methodological shortcoming for priming studies.In a literature review, we look at priming studies that use 14C-labelling to differentiate CO2 evolved from substrates and SOM published in 1993–2021. Out of 65 studies in total, 34 use scintillation cocktails that we had previously tested for outgassing of 14CO2. Out of the 21 studies that provide information about the relevant scintillation counting parameters that determine stability of the measurement, 33–43% use scintillation counting parameters that can be expected to underestimate 14C-activity. In an exemplary priming study, we examine a potential magnitude of error resulting from such incorrect measurements for priming determination and a method to correct the results even when the samples are no longer available to be re-measured. A sandy topsoil and subsoil were incubated with and without added nutrients and 500 (μg C) g−114C-labelled sucrose for 33 days, evolving CO2 was trapped in 0.35 M NaOH and scintillation counted with Ultima GoldXR. By repeatedly re-measuring similarly prepared calibration vials of known 14C-activity, we correct the measurements based on the time passed between preparation and measurement of the vial. We find that with these experimental parameters, 14C-activity is underestimated by 10–30% and consequently, priming is overestimated by 53–66 (μg C) g−1 soil. After correction of the data, we find 16–40 (μg C) g−1 primed carbon. Mass balancing reveals that 10.5–12.8% of 14C-activity of the substrate were recovered by the scintillation correction while 7.3–10.2% remain unaccounted for.These findings suggest that if the 14CO2 activity is underestimated during scintillation counting, then priming can be overestimated by a multiple of the true value and priming can even be found where there isn't any. This issue should not be ignored when discussing priming, considering the prevalence of this method in priming studies, the lack of reporting of relevant experimental parameters and the number of reported flawed experimental parameters found in the literature.