Time variation of the estimated CH4 flux. Prior flux estimates are shown in red, and post-assimilation flux estimates are shown in blue. Note that only the biomass burning prior flux estimates vary from year to year; other prior flux estimates are constant with time. The post-assimilation estimates, on the other hand, may show considerable variability due to sensitivity of microbial methane production and uptake to year-to-year changes in temperature and precipitation. 1 σ uncertainty bounds are shown as light red (prior) and light blue (post-assimilation) shaded areas. The post-assimilation error bounds decrease from the prior error bounds because of the existence of observational constraints. In regions where there are few observations, the error bounds do not narrow significantly. Note that microbial sources of methane, such as wetlands and agriculture, are temperature-sensitive and therefore tend to be largest during summer. Units are TgCH4/yr.
Estimated Prior and Post-assimilation CH4 fluxes in units of TgCH4/yr. Red numbers in each cell are the prior fluxes and uncertainties, blue numbers are post-assimilation fluxes and uncertainties.
Time variation of estimated CH4 flux anomalies. Anomalies are calculated by subtraction of mean flux values for each month over the entire period of the assimilation. The heavy line shows smoothed flux anomalies computed by applying a running 6-month average. 1 σ uncertainty bounds are shown as light blue shaded areas. Units are TgCH4/yr.
Posterior CH4 flux anomalies and uncertainties in units of TgCH4/yr. Anomalies are calculated by subtraction of mean monthly values