Vertical distribution of shipping emissions
Currently in HERMESv3_BU we are assuming that all shipping emissions occurring in port areas are released into the lowest vertical layer of the model. This assumption is hardcoded in the code as follows (line 597 of the shipping_port_sector.py
script):
dataframe['layer'] = 0
Several studies suggest that this assumption can lead to an overestimation of concentrations in the vicinity of the port areas (e.g. Badeke et al. (2022).
FMI recommends to inject emissions from ships between 30 - 100 m. This would mean that in the CALIOPE system, emissions should be mostly injected between the 2nd and 3rd layer, considering the current definition of vertical levels in the system:
Ilayer,height_magl
1,39
2,78
3,119
This is not a priority, but considering the general overestimation we report in the Barcelona Port area, at some point we should perform some sensitivity simulations to see the impact of changing the vertical distribution of these emissions (e.g. we could start by simply assuming all emissions are occurring in the 2nd layer and not the first). Of course, a proper implementation would entail changing the code of HERMESv3_BU.
For HERMESv3_GR and the processing of shipping emissions reported in CAMS-REG, this would imply adding a new vertical profiles for the GNFR_G (shipping) sector, following FMI's recommendation. In this case, no changes in the code will be needed.
This discussion may be of interest for Plan O3, depending on the contribution we estimate from shipping emissions to O3 formation in the Iberian Peninsula.
FYI: @ojorba , @ctena , @senciso , @kdeolive, @flope3 , @rgaratac , @cperez