A technique for measuring CO2 and water vapor profiles within and above plant canopies over short periods

A multiport air sampling apparatus was developed to measure vertical profil es of CO2 and water vapor concentrations within and above the plant canopy over time intervals of as short as 5 min. Air is drawn continuously from si x heights through tubings of equal length and diameter, each connected to a three-way solenoid valve. Flow through five of the six intake lines is dri ven by a large purge pump while the air from the selected sampling height i s drawn by a small sample pump and pushed through an infrared gas analyzer (IRGA) for the simultaneous measurements of CO2 and water vapor concentrati ons. A data logger controls the solenoid valves and switches the air from e ach height sequentially through the IRGA. The internal volume of the appara tus is minimized to ensure a fast response in TRGA output when the sample a ir switches from one height to another. The response was shown to be essent ially complete within 2 s or less. By delaying signal recording for 2 s aft er switching, one cycle of sequential measurements over the six heights is completed in 1 min, and 5 min mean profiles are obtained as the mean of fiv e 8 s measurement periods spaced 1 min apart for each height. The short int erval between each measurement period is essential when the scalar concentr ations fluctuate or change rapidly. We compared the water vapor pressure wi thin the upper part of a cotton canopy measured with our apparatus, with th at measured by a precision, shielded and ventilated psychrometer made of pl atinum resistance thermometers. The results showed excellent agreement over the daily cycle between the two instruments for the 30-min mean data, and very good agreement for the 5-min mean data, showing coincidence of nearly all the 5-min peaks and troughs between the two methods. Sample diurnal pat terns measured by the apparatus over a maize field spanning two consecutive days with contrasting wind speeds are given. The delay of nearly 2 s in re sponse is mostly due to the internal volume of the apparatus. Taking the ti me required to clear this volume upon switching to a new port into account, it is possible to make CO2 measurements of slightly reduced accuracy at a rate of 1 s per port. The potential use of the apparatus for measuring at s everal heights over very short time intervals (in terms of seconds) is disc ussed in the context of the requirements for studying coherent turbulence s tructures and calculating the storage term in CO2 flux determinations. (C) 1999 Published by Elsevier Science B.V. All rights reserved.