Flux measurements and sampling strategies: Applications to methane emissions from rice fields

摘要

The emissions of methane from rice fields and other sources are often measured by placing chambers on the surface and taking sequential samples. Although static chambers pose several problems that affect the accuracy of the data, there are a few parameters that, if carefully chosen, can improve the reliability of the data and reduce the uncertainties. These parameters are the length of time the chamber is kept on the rice plants, the number of samples that are drawn to estimate the flux, the basal area and height of the chamber, the frequency of measurements during the growing season, and the number of plots sampled. In this paper we analyze a large data set to determine how these parameters can be chosen to improve data quality. The results show that, for individual flux measurements, extending the time the chambers are left on the plots improves precision more effectively than taking more sequential samples for each flux measurement. The exposure time cannot be extended too far, however, as it leads to a saturation effect so that the rate of accumulation in the chamber slows down, This can compromise the accuracy of the measurement. There is an optimum exposure time that balances these two effects. Many individual measurements are needed to characterize the emissions from the larger area of the fields and the seasonal patterns. For methane emissions from rice fields, the amplitude of the systematic seasonal cycle is usually large compared to the variability on shorter timescales. Consequently, reducing the sampling frequency increases the uncertainty of the seasonal flux very slowly. The spatial variability is large and random on the small scales of the basal area of the chambers. Reducing the number of plots sampled, therefore, has a major effect on the uncertainty of the seasonal average flux. [References: 3]