A new method to obtain minute interval mass fluxes of H2O and CO2 using scintillometry and scalar turbulence measurements

The goal of this study is to test an alternative method for determining turbulent H2O and CO2 fluxes, which has a faster statistical convergence than the classical eddy-covariance method. This enables determining turbulent fluxes during strongly non-stationary conditions, e.g. in the intermittent stable boundary layer or rapidly changing cloud-cover. In our new method, we suggest a hybrid set-up that combines a point-sensor for scalar H2O and CO2 with a dual-beam laser scintillometer (DBLS). We used a LiCor7500 open-path fast-response H2O/CO2 sensor. The H2O/CO2 sensor forms the basis for estimating the turbulent exchange scale for H2O and CO2. The DBLS yields the friction velocity and stability. The data presented were gathered in May-June 2009 over a wheat field near Merken, Germany in the framework of the TransRegio32 program. With the DBLS installed in homogeneous areas, turbulence is averaged both in time and space, allowing short averaging flux intervals down to a couple of seconds. When determining the averaging-time-dependent systematic and random errors in the H2O and CO2 fluxes, we found that eddy-covariance is not apt to measure 1-minute averaged fluxes, whereas our new method is. Even for 1-minute averaging intervals, our new method does not have an averaging-time-dependent systematic error and the corresponding random error is only halve that of the eddy-covariance method. Thus, we demonstrate that our new method reproduces reliable estimates of H2O and CO2 fluxes for 1-minute averaging intervals and show that not only individual plants, but also at field scales wheat responds rapidly to changes in solar radiation.