My last post used the idea that an air parcel moving at 5 mph over a 5 mile long solar heater would be heated for an hour. This might be enough to allow the moist San Francisco Bay air to rise up through the temperature inversion and be transported to where it might produce some useful rain in the summer months and early fall. The 5 mile length would be aligned with the frequently-observed west-to-east wind direction. A 1 mile solar absorber width would allow an air parcel that deviates from the west-to-east direction to remain over the absorber and be heated for an hour. The 5 mile length would be broken into 3 segments of 1.67 miles each, separated by two navigation channels 0.33 miles wide. The total length is 5.67 miles. A 5 mph air parcel would still be heated for an hour, but there would a slightly reduced tolerance to angular deviation from an exact west-to-east direction.
The three rafts might be moored in San Francisco Bay near the airport at San Bruno, but away from the coast so that near-shore activities are not impacted. When drawn to scale on a map, the solar absorber looks pretty small and might be tolerated. Raft segments about 9.5 ft wide by 150 ft long would be deployed from reels on workboats with their length oriented west-to-east. (Maybe north-south might have advantages, but it is too early to know.) We don't want any gaps between the raft segments, so we need a means of stitching them together side-to-side. This might be done by an operator on a cart that uses four cylinders as wheels that ride on top of the raft. The cylinders might be 3 ft diameter by 4 ft length to provide enough buoyancy to keep the rafts from being pushed too deep. The cylinders would slide on their respective shafts so that the raft segments could be pulled toward or pushed away from each other. The operator would then insert clips (which have yet to be invented) into receptacles along the edges of the raft segments. (More invention needed.)
With this tentative layout and deployment description, my next step may be to make better performance predictions. The rafts will work all day long, day after day, so the performance should be better than what I calculated for a few hours on a summer afternoon. Here's hoping.