In 2006, California passed landmark legislation (AB-32) to reduce emissions of greenhouse gases (GHGs) responsible for global climate change. AB-32 commits California to reduce total GHG emissions to 1990 levels by 2020, a reduction of 25 percent from current levels and to 80 percent below 1990 levels by 2050. To verify that GHG emission reductions are taking place, it will be necessary to measure emissions regionally. In 2003, we began an exploratory project with the California Energy Commission's Public Interest Energy Research Program to estimate whether atmospheric methods could be used to quantify GHG emissions on air basin to regional scales. Based on that work, we began a pilot project to implement GHG meaasurements in 2006. The California Greenhouse gas measurement project (CALGEM) is a collaboration with the Global Monitoring Division of the National Oceanic and Atmospheric Administration (NOAA) to implement measurements of greenhouse gas emissions in the San Francisco to Sacramento areas.
CALGEM is developing measurement and analysis techniques to estimate local-regional land surface emissions of major greenhouse gases (CO2, CH4, N2O, SF6, and halo carbons) using concentration measurements in combination with atmospheric inverse models. Tower locations in San Fracisco Bay and Sacramento Delta areas were chosen to capture the influences of a mixture of urban and rural land uses. In collaboration with the NOAA Global Monitoring Division, twice daily flask samples are collected at both towers and subsequently analyzed for all important GHGs (CO2, CO, CH4, SF6, and halo carbons). At the Sacramento Delta site, continuous CH4 and 222Rn analyzers from LBNL and CO2, and CO from NOAA are now also operational. Episodic 14CO2 sampling is planned for both sites. Measurement techniques include cavity-ringdown diode-laser spectroscopy, non-dispersive IR absorption, and low-background counting of naturally occurring radionuclides (e.g., 222Radon). Spatially explicit “prior” models for GHG emissions from California will be refined in cooperation with the California Air Resources Board. Mesoscale modeling is performed using a specialized version of Weather Research Forecasting (WRF) model. Surface influence functions, which quantify how much a unit flux of GHG from the land surface changes atmospheric GHG concentrations, are calculated using the Stochastic Time-inverted Lagrangian Transport (STILT) model. Probabilistic “best-estimates” of GHG emissions and their uncertainties are computed by optimal adjustment of the prior emissions estimates to provide a best match between measured and predicted atmospheric GHG concentration using classic Bayesian estimation. Future work will bring other gas species, measurements from other locations, refined meteorological modeling, and more sophisticated statistical analysis to improve the emission estimates and thereby verify the success of AB-32.