A new approach to evaluate regional methane emission from irrigated rice paddies: Combining process study, modeling and remote sensing into GIS
Sass, Ronald L.
Doctor of Philosophy
A large seasonal variation in methane emission from Texas rice fields was observed in most of the growing seasons from 1989 through 1997. In general, the pattern showed small fluxes in the early season of cultivation and reached maximum at post-heading time, then declined and stopped after fields were drained. The amount of methane emission positively relates to the aboveground biomass, the number of effective stems and tillers, and nitrogen addition. The day-to-day pattern of methane emissions was similar among all cultivars. The seasonal total methane emission shows a significant positive correlation with post-heading plant height. The total methane emission from Texas rice fields was estimated as 33.25 x 109 g in 1993, ranging from 25.85 x 109 g/yr to 40.65 x 109 g/yr. A mitigation technique was developed to obtain both high yield and less methane emission from Texas rice fields. A new approach was also developed to evaluate regional to large-scale methane emission from irrigated rice paddies. By combining modeling, ground truth information and remote sensing into a Geographic Information System (GIS)---a computer based system, the seasonal methane emission from a large area can be calculated efficiently and more accurately. The methodology was tested at the Richmond Irrigation District (RID) site in Texas. The average daily methane emission varied from field to field and even within a single field. The calculated seasonal total methane emission from RID rice fields was as low as 3.34 x 108 g CH4 in 1996 and as high as 7.80 x 108 g CH4 in 1998. To support the application of the estimation method in a worldwide study, an algorithm describing the mapping of irrigated rice paddies from Landsat TM data was demonstrated. The accuracy in 1998-supervised classification approached 95% when cloud cover was taken into account. Model uncertainty and data availability are the two major potential problems in worldwide application of the new approach. A potential alternative model is proposed which allows estimation of regional methane emission from rice plant height.
Ecology; Atmospheric sciences; Environmental science; Biology; Physics