Water vapor, CO2, and temperature profiles in and above a forest-accuracy assessment of an unattended measurement system

Abstract

The possibility of a global climate change has increased research interest in the least understood parts of the climate system. One of those parts is the boundary between the land surface of the earth and the lowest part of the planetary boundary layer. The structure of this layer and the exchange processes in it are still incompletely understood for a variety of situations and surfaces, especially in the boreal zone and during the dark parts of the day and the year. Progress in this area requires new data measured continuously and unattended with high accuracy and long-term reliability. A measurement system for profiles of temperature, humidity, and carbon dioxide was designed to meet the above goals. The system used thermocouples and a Li-Cor gas analyzer combined with an array of tubing to suck air from different heights. Turbulent fluctuations of water vapor and carbon dioxide concentrations were smoothed by continuous-flow mixing chambers without moving parts. Half-hourly mean differences in temperature, humidity, and CO2 were measured to better than 0.03 K, 0.015 g kg-1, and 0.5 μmol mol-1, respectively. These accuracies were confirmed by comparisons with a thermometer-interchange (reversing) system and CO2 profiles theoretically deduced from eddy-correlation fluxes. Daytime temperature and humidity differences over the full height interval (24.5-87.5 m), as well as over the roughness sublayer part (24.5-58.5 m), commonly exceeded the estimated errors by five times. The CO2 differences could only be measured reasonably accurately over the entire height interval (24.5-87.5 m) and then only exceeded the error by a factor of 2-3. Temperature and humidity measurements were sufficiently accurate for studies of flux-profile relationships over a forest. The CO2 profiles were accurate only for rough flux estimates and may be especially useful for nighttime studies.