There are two issues, harnessing and market. You need both to be viable.
A market, makes most of the 70% of the land mass covered by water unusable, because even if you could harness the power, it is impossible to get the power to consumers. There are no practical storage alternatives, either.
Ocean buoys do harness ocean currents to meet their individual power needs.
Wave power does exist, but this is on a small scale, mostly experimental, close to land.
Tidal power has great potential by either harnessing tides as they flow in and out or using dams to capture hide tide and harnessing outflow. But they have to overcome damage by mother-nature or deal with impact on marine animals / environment.
Many rivers are ruled out because there are used by ships/boats, which would require locks to allow passage or damming would create a flood plane, which would impact existing infrastructure (farm land, communities).
Other rivers are ruled out because of spawning fish. Fresh-water spawning fish, like salmon, can pass dams by fish ladders, but dam flood plane is impacted because fish need calm areas to spawn.
Countries with remote rivers (like Canada - 9% of land mass is fresh water - James Bay, Churchill Falls) have dammed them up and harnessed them. This creates large amounts of power, which is shipped via transmission lines to urban regions. They take up a lot of real estate, which impacts local communities, either having to relocate or impacting hunting / fishing / recreation territory.
Smaller rivers/lakes with a significant water head are harnessed as appropriate. Wikipedia list of generating facilities for Newfoundland and Labrador, Canada. Not all are on the island of Newfoundland (Churchill Falls), but the hydro-electric facilities (~38 facilities) are not enough to supply the power needs of ~400,000 people. This is supplemented by Diesel Turbine facility to meet peak power demand.
Micro-hydro exists. Any flow of water can potentially be harnessed. Water-wheels to grind wheat. There may be legislative hurdles to overcome. After all you are impacting a community resource.
Canada, despite it's large geographic area and considerable hydro-electric facilities, still supplements peak power requirements with atomic, wind, solar and diesel (for remote regions).
Pumped Storage is another alternative. Usually used to balance the grid load, pumped storage pumps water to a higher height during periods of low power usage to be delivered during peak power periods. A nuclear power plant is kept at optimum efficiency, so during non-peak times, extra power is available to pump water uphill. This makes sense because the extra Pumped Storage capacity is available in times when peak demand occurs (breakfast, lunch, supper).
To answer the question: Most significant hydro-electric resources are either harnessed or planned to be harnessed. There is a significant investment (money, resources) to harness these renewable resources, but the return on investment is worth the investment.
The Grand Canyon would make an ideal dam site (remote, formed by water, no population [people or herds]), except for two things: rainfall & desert. It does get some rain, but it is mainly dry because it is in a desert zone. If it had the water flow, it would dammed and harnessed.