Water gaps are openings or cuts at right angles to ridges and mountain ranges. They are usually defined in terms that presume they were made by a river or stream that was running before the mountain ridge was uplifted. The stream in these cases must erode down at the same speed in which the land is uplifted. There are at least 1,000 water gaps found around the world, most of them difficult to explain. Creationists have questioned the reasonableness of the standard geological explanations for water gaps, and note that flood runoff explains them better.
Wind gaps are like water gaps since they are notches caused by erosion rather than faulting or other causes. They do not have water flowing in them at present and are considered to be either an ancient water gap or a beginning one. Geologists generally think that wind gaps were originally water gaps but whose streams have been diverted somehow, so that only wind passes through the gap at present.
Oard cites three hypotheses for water gaps generally noted in the literature: 1) the antecedent stream model, 2) the superimposed stream model, and 3) stream piracy. However, generally, none have more than a slight amount of evidence. Water gaps are not always difficult to explain if they follow the most obvious route for water to take. The difficult gaps are those in which it seems that the water could have more easily gone a different direction than cutting through the barrier.
This idea presumes that a river was flowing across a region before the landscape was uplifted slowly enough to allow the river to cut its course faster than the land rises. For some time this was thought to be the way the Green River and Grand Canyon water gaps were cut. It generally takes a large river to keep up with the erosion necessary, and some geologists argue that any river erosion would be too slow to match uplift.
This theory generally cannot apply unless you can show that the river came before the uplift, which is usually difficult to do, and then show the speed of uplift was correct. A lake should form because of the barrier, but deposits that show a lake's presence usually have not been found. The chances are against one gap, and cannot hold up to several aligned gaps. In many cases, geologists have reinterpreted water gaps as being not from an existing stream but from superimposition. Generally antecedence is now considered a minor contributor to water gaps.
If a landscape is covered by sediments, creating a relatively flat surface, a stream that starts on the top of the surface can cut down along its course, eventually cutting through hard rock hidden below the sediments. Eventually all of the covering sediments are eroded leaving a stream that seemed to cut through barriers when it was really directed by covering sediments that are now missing.
Difficulties with the idea are that it has been hard to find evidence for the sediments that make the idea work. In the Rocky Mountains it is hard to say where the fill came from, and how it was moved away from several ranges, leaving streams cut in a straight line. The superimposed river must follow enough of a straight line to cut its course through rock, but at the same time meander enough to erode away the covering sediments from the resistant rocks. It is hard to imagine how both can happen at once, and there is usually a lack of evidence for the covering sediments.
In some cases it is thought that one faster eroding stream "steals" the discharge of another by crossing its channel. This creates a sharp change in channel direction, and can explain why a channel is left high and dry.
This has been difficult to examine, since the sharp changes in channel direction can have many different reasons besides capture. There have been cases where presumed stream capture is now proposed to be a breached dam from a lake.
Late Flood Origin
Catastrophic erosion, such as bursting ice dams, has been postulated by several creationists. Michael Oard believes it is better to look to the conditions just after the Flood as the most likely origin for water gaps. He says that "water and wind gaps can rapidly be cut during the Channelized Flow Phase of the Flood by currents ﬂowing perpendicular to mountains and ridges. An example of the cutting of a water and wind gap in a few days is provided by the Lake Missoula ﬂood. Worldwide water and wind gaps, like other global geomorphological mysteries, point to a global Flood."