As others have pointed out, there's no need to store heat for six months. But you bring up a common enough viewpoint, one that I think bears re-examining: >I'd

As others have pointed out, there's no need to store heat for six months. But you bring up a common enough viewpoint, one that I think bears re-examining: >I'd imagine there's not enough water to do this inland.. We quite often forget that there's no physical limitation on the amount of water available inland. Water is constantly flowing from the oceans onto land, being recycled (mainly by trees pushing it back up into the airstream), and flowing back out and mixing with the ocean. So the amount of water per square meter of land is limited by only two variables: * The rate at which fresh water falls on land (liters/hectare/year) * The time it takes that water to flow to the ocean (years). The amount of water available on land (liters/hectare) is simply these two quantities divided by each-other. The flow is fundamentally limited by the energy balance of the sun, so there's not much we can do there. But the time is entirely determined by our ability to design infrastructure and land use patterns that store water on the land in useful catchments instead of draining it off rapidly. Farm-scale dams are one visible way of doing that, but it's much cheaper to store water in the ground via infiltration systems. The amount of water is inversely proportional to the amount of time it takes water to flow back to the ocean after it falls. So if we want to "wet up" the landscape, we should make it follow a loooong path with a lot of passive friction and storage buffers. Just to show my math here, in the USA on average there's 715 mm of rainfall per year, which works out to 19,000 gallons for the average 100 m^2 house footprint. And next year you get that much water again. In extremely arid areas like California it would even be possible to use salt water for thermal storage.