To delineate the drainage basin for a lake, would the pour point be the inlet or the outlet?

I am interested in the area draining into and affecting a particular lake. Delineating this drainage basin, should I use the lake's inlet or outlet to best capture the relevant drainage basin? I think including the lake is valuable in this exercise, as events on or in the lake itself also affect the lake.

I'd think outlet, but want to double check with you all. To complicate things, most lakes have more than one inlet and outlet. I guess I'd go with the lowest outlet in order to avoid missing any land which may drain into the lake.

• Few lakes actually have more than one outlet. Otherwise, if you use an inlet, you'd just get the drainage basin for that inlet. Jul 29, 2017 at 20:07
• I was wondering about that, how many lakes actually have more than one outlet. If most have only one significant outlet, than I guess using that single significant outlet as the delineation point for the entire watershed affecting that lake would be most appropriate.
– cr0
Jul 30, 2017 at 18:07
• It would be rare for a lake to have more than one outlet, but I think I stumbled across one on Google Earth in Nunavut Canada. An additional complication would be when a new outlet forms when lake level rises. That would be hard to handle in a GIS analysis, but probably isn't necessary to worry about. Aug 3, 2017 at 0:26

I think there are a few things to consider. First, a drainage basin is defined as the area upstream of the point to which all precipitation converges. Flow does not converge at the outlet - flow converges at the lake. Outlets don't contain any additional information of the upstream area draining into the lake. This implies that to find the area draining into a lake, you would necessarily have to proceed beginning from each and every inlet.

Here's another way to see this. Approximately 20% of all land drains to lakes with no outlets. These are referred to as endorheic lakes or endorheic basins. Even though there is no outlet, there is absolutely still a definable drainage basin area.

• Good point about the endorheic basins, in that case they must be delineated based on a or every inlet.
– cr0
Jul 30, 2017 at 23:00
• "Outlets don't contain any additional information of the upstream area draining into the lake." But if all the inlets of the lake drain into the lake, wouldn't the outlet the lake drains into capture the drainage basins of all of those inlets?
– cr0
Jul 30, 2017 at 23:01
• Basically I was trying to say that lakes don't contain any "additional" information over the outlet. But I now I need to think about this for a bit, because it might be a little more nuanced than I made it out to be. For instance, it's not clear to me whether the lake area itself should be included in the calculation of drainage area.
– Z W
Aug 1, 2017 at 1:02
• Right. I'm thinking the lake should be included, as a way to 1) ensure all areas draining into the lake are included, not just its major inlet, and 2) to represent the entire "domain" of the lake. For example, let's say we were delineating political boundaries according to watersheds. For the watershed of a lake, how would you delineate that watershed? You'd want to include the lake itself as well, since anything happening in or on the lake affects the lake as a resource and as a contributor to all downstream of the lake.
– cr0
Aug 1, 2017 at 16:54
• Yup i think you're right. And i think what this implies is that there is no "right answer", or that it's dependent upon the question. In your example of delineating political boundaries, I agree %100 with you. And as an example of where the lake should not be included is if you're trying to determine the relationship between lake size and drainage area. You'd think this would be discussed in textbooks but... i have yet to find it mentioned!
– Z W
Aug 2, 2017 at 21:02