Uncertainties and limitations of IBC module

Here are notes on some key uncertainties and limitations when using IBC within LEAP.

GEOS-Chem Adjoint coefficients for PM2.5: The GEOS-Chem Adjoint coefficients quantify the sensitivity of PM2.5 concentrations in the target country to NO_x, SO₂, NH₃, BC and OC emissions in grid squares across the world. These sensitivities are calculated for a base set of emissions, for the year 2010. The coefficients are applied in IBC to look at changes in PM2.5 concentrations in the target country that result from changes in emissions in the target country, and across the world. They are linear coefficients, which means that a change in emissions results in a linear increase/decrease in PM2.5 concentrations in the target country. The methodology therefore does not account for non-linear changes in target PM2.5 concentrations resulting from non-linear chemical reactions in the atmosphere, e.g. combination between NO_x, SO₂ and NH₃ to form secondary inorganic aerosol.
GEOS-Chem Adjoint coefficients for Ozone: The GEOS-Chem Adjoint coefficients quantify the sensitivity of ozone concentrations in the target country to NO_x, VOC, CO and CH₄ emissions in grid squares across the world. These sensitivities are calculated for a base set of emissions, for the year 2010. The coefficients are applied in IBC to look at changes in ozone concentrations in the target country that result from changes in emissions in the target country, and across the world. They are linear coefficients, which means that a change in emissions results in a linear increase/decrease in ozone concentrations in the target country. Ozone formation depends on the relative emissions of VOC and NO_x, with ozone increasing due with increasing NO_x emissions in VOC-limited regimes, and with increasing VOC emissions in NO_x-limited regimes. Hence changes in the relative emissions of NO_x and VOCs will result in non-linear changes in ozone concentrations. This interaction is not taken into account in the application of the ozone coefficients in IBC.
Health impact assessment methodology: The health impact assessment estimate premature mortality associated with PM_2.5 and ozone exposures, using concentration-response functions that have been used by the Global Burden of Disease project. These concentration-response functions are based on health effects research that has been carried out in North America and Europe. It is assumed that the same relationships apply in other regions of the world, including those where PM_2.5 and ozone concentrations are much higher, and where the composition of PM_2.5 may differ.
Crop impact assessment methodology: Agricultural crop yield loss is estimated using concentration-response functions that quantify the relationship between wheat, maize, rice or soy yield and ozone exposure. These functions are based on experiments carried out in North America, that assessed this relationship for ozone exposure between 20 and 90 ppb. The relationship between yield loss and ozone exposures above 90 ppb is uncertain. To avoid unrealistic estimates of ozone-induced yield loss, for ozone concentrations above 90 ppb, there is assumed to be no increase in yield loss. This means that for mitigation scenarios to show a benefit for ozone crop yield loss, ozone concentrations must be reduced to below 90 ppb, e.g. a reduction from 110 ppb to 100 ppb in ozone exposure does not result in an estimated reduction in yield loss.
Other: Significant but hard to quantity uncertainties exist in the activity levels, energy intensities and emission factors used in any LEAP area. Currently LEAP's calculations are deterministic and do not reflect any uncertainty in these values. However, you can use LEAP to perform sensitivity analyses or even connect it with tools such as Oracle Crystal Ball that allow uncertainties to be explored using techniques such as Monte Carlo analysis. Bear in mind also that all future values are inherently uncertain since they depend on policy choices that may or may not be made.