Downscaling inventory

[gkuhl]

We currently do not redistribute the low-resolution inventory data within high-resolution model grid using additional information, such as land/sea masks, population density or country borders, which can be used for improving the spatial allocation of area sources along the coastline, at city scale, or between multiple countries.

This can be implemented as a pre-processing of the inventory before we compute the interpolation in the main function. In the pre-processing, we need to split the inventory cells (polygon with four corner points) into smaller grid cells (polygons with many points) using the intersection with the additional information. The interpolation is then computed with the arbitrary polygons and can be used for projecting the pre-processed inventory to the model grid.

Known issues:

• Since the splitting can be different for different sectors, we have to pre-process the inventory by sector and also compute a new interpolation for each pre-processed inventory.