Thinking about finding appropriate areas for power/tropical cyclones intersection and disruption analysis with @thomas-fred
Power network can be split on national boundaries to assume no interconnectors for flow modelling purposes
Then network components form independent functional units in system modelling
Can find the hull or bounding box of each component to give the area of interest
Storm tracks coild be buffered by some 100km (tbd) or after the windfield is generated, clip to a 20m/s (tbd) threshold and find the envelope, to give the area of interest for that storm
Then we can fairly quickly filter storms/network components to consider only relevant events
network component should be a simpler unit than piecing together unkown-up-front number of arbitrary grid boxes/slices and handling network linking properly
we should come back to the question of international interconnectors with a bit of search/research for where they are used and important, to improve on the gridfinder dataset
Thinking about finding appropriate areas for power/tropical cyclones intersection and disruption analysis with @thomas-fred