Three 8.6 m3 volume flux chambers (1.9 m height and 2.1 m
square) were used for this study. Each chamber was equipped
with an internal fan (0.5m3 min1) tomix the inside air and an
external fan (2.0 m3 min1) to draw air through the chamber.
The flow rate was set to exchange the chamber air every
4.3 min. Chambers were placed on replicate piles for each
treatment for 20 min to allow the chambers to reach equilibrium
conditions prior to measurements. Flux measurements
were made immediately prior to mixing, one and three days
after mixing, and once or twice a week thereafter. Gas samples
were taken from the flux chamber outlet and analysed for CO2,
CH4, and N2O using a photoacoustic field gas monitor
(INNOVA 1412; Lumasense Technologies, Ballerup, Denmark).
In order to test the accuracy of the gas sampling system, recovery
tests were conducted. Pure CO2 gas was released into
each chamber at a known flow rate while monitoring gas
concentration at the outlet. Comparison of the measured gas
concentrations with calculated values revealed very high accuracy
(recovery rate above 95%; data not shown). Prior to flux
chamber measurements, temperature values were determined
in the centre of each pile (approximately 60 cm from
the pile surface) using long stem thermometers (Reotemp Instruments,
San Diego, USA). Cumulative gas emission values
were calculated by trapezoidal integration of the averaged
emission results and airflow rates.