RECENT developments in water and carbon trading and biofuel production highlight the need to document the water and carbon balances of Australia’s cropping systems including sugarcane. This paper presents the results of studies of evaporation and CO2 exchange throughout the growing seasons of two sugarcane crops, a 1st ratoon crop at Murwillumbah where burnt-cane was practised and a 5th ratoon crop at Mackay where trash blanketing was employed. At both locations, a micrometeorological eddy covariance technique was employed to measure water vapour and CO2 exchange between crop and atmosphere and manual and automatic chambers to measure CO2 emission from the canopy floor. The measurement period extended from the time of fertilising to harvest and was 342 days long at Murwillumbah and 292 at Mackay. Evaporation from the Murwillumbah crop was 1281 mm and the net assimilation of CO2 was 132 t CO2/ha, with 38 t/ha coming from the canopy floor and 94 t/ha from the atmosphere. At Mackay, evaporation was 970 mm and net assimilation only 60 t CO2/ha, with the canopy floor contributing 10 t/ha and the atmosphere 50 t/ha. It is suggested that apart from the shorter season at Mackay, the differences in evaporation and CO2 exchange between the two crops was probably due to the presence of a near-surface water table and higher available soil water contents at Murwillumbah, and the age of the plants (1st ratoon versus 5th ratoon). Despite differences between crops in average daily evaporation rate, reference crop evapotranspiration was found to be a reasonably good estimator of crop evaporation, overestimating it by 10% at Mackay and underestimating by 10% at Murwillumbah. The very large difference in net assimilation between the crops was responsible for a drop in water use efficiency, from 103 kg CO2/ha assimilated per mm of water evaporated at Murwillumbah to 62 at Mackay.