You are likely to get a sufficient buildup of charge for lightning to occur when you have strong turbulence and ice crystals present, which requires strong convection currents at high altitude.
Sufficiently strong convection currents are much more common over land, where the sun can heat ground rapidly.
Another factor is global circulation patterns. Convection causes air to rise at the equator, and to sink at the poles. (actually we get three convection cells, so we also get descending air at 30 degrees latitude, and rising air at 60 degrees). This means that we tend to get high rainfall, storms, and lightning at the equator and at 60 degrees, low rainfall and few storms at 30 degrees and at the poles.
Instead of one big cable, how about using a large number of thin wires? Each elevator is attached to more wires than the minimum needed to lift it. This gives you a number of advantages:
Avoids Red Mars-type scenarios. Each wire could be made thin enough that in the event of failure it would either burn up in the atmosphere or lose most of its kinetic energy to air resistance before hitting the ground.
Redundancy. If one wire is severed by meteorites, space debris, etc. nothing goes flying off, nothing gets dropped except for the one damaged wire.
Stability. Not all the wires have to be attached to the same point on the ground. Have bundles of wires attached over a wide ground area, tensioned by a counterweight (space station) just beyond geosynchronous orbit, and you have a fair degree of protection against orbital instabilities, tidal forces (yes they might matter on something this size!) and coriolis effects from accelerating payloads to/from orbital velocity.
Scalablity/Maintainance. You could start small, and add new bundles of wires later, perhaps to new ground stations. You could also replace individual wires without affecting the structure as a whole. It might be necessary to have a rolling replacement program for wires to correct for degradation caused by micrometeorites, radiation, etc.
Slashdot Mirror
Sufficiently strong convection currents are much more common over land, where the sun can heat ground rapidly.
Another factor is global circulation patterns. Convection causes air to rise at the equator, and to sink at the poles. (actually we get three convection cells, so we also get descending air at 30 degrees latitude, and rising air at 60 degrees). This means that we tend to get high rainfall, storms, and lightning at the equator and at 60 degrees, low rainfall and few storms at 30 degrees and at the poles.
http://shrimp.tripod.com