Where does the water in a lake come from, and how does water leave it? Water enters a lake from inflowing rivers, from underwater seeps and springs, from overland flow off the surrounding land, and from rain falling directly on the lake surface. Water leaves a lake via outflowing rivers, by soaking into the bed of the lake, and by evaporation. So much is obvious.

The questions become more complicated when actual volumes of water are considered: how much water enters and leaves by each route? Discovering the inputs and outputs of rivers is a matter of measuring the discharges of every inflowing and outflowing stream and river. Then exchanges with the atmosphere are calculated by finding the difference between the gains from rain, as measured (rather roughly) by rain gauges, and the losses by evaporation, measured with models that correct for the other sources of water loss. For the majority of lakes, certainly those surrounded by forests, input from overland flow is too small to have a noticeable effect. Changes in lake level not explained by river flows plus exchanges with the atmosphere must be due to the net difference between what seeps into the lake from the groundwater and what leaks into the groundwater. Note the word "net": measuring the actual amounts of groundwater seepage into the lake and out of the lake is a much more complicated matter than merely inferring their difference.

Once all this information has been gathered, it becomes possible to judge whether a lake’s flow is mainly due to its surface inputs and outputs or to its underground inputs and outputs. If the former are greater, the lake is a surface-water-dominated lake; if the latter, it is a seepage-dominated lake. Occasionally, common sense tells you which of these two possibilities applies. For example, a pond in hilly country that maintains a steady water level all through a dry summer in spite of having no streams flowing into it must obviously be seepage dominated. Conversely, a pond with a stream flowing in one end and out the other, which dries up when the stream dries up, is clearly surface water dominated.

By whatever means, a lake is constantly gaining water and losing water: its water does not just sit there, or, anyway, not for long. This raises the matter of a lake’s residence time. The residence time is the average length of time that any particular molecule of water remains in the lake, and it is calculated by dividing the volume of water in the lake by the rate at which water leaves the lake. The residence time is an average; the time spent in the lake by a given molecule (if we could follow its fate) would depend on the route it took: it might flow through as part of the fastest, most direct current, or it might circle in a backwater for an indefinitely long time.

Residence times vary enormously. They range from a few days for small lakes up to several hundred years for large ones; Lake Tahoe, in California, has a residence time of 700 years. The residence times for the Great Lakes of North America, namely, Lakes Superior, Michigan, Huron, Erie, and Ontario, are, respectively, 190,100,22,2.5, and 6 years. Lake Erie’s is the lowest: although its area is larger than Lake Ontario’ s, its volume is less than one-third as great because it is so shallow-less than 20 meters on average.

A given lake’s residence time is by no means a fixed quantity. It depends on the rate at which water enters the lake, and that depends on the rainfall and the evaporation rate. Climatic change (the result of global warming?) is dramatically affecting the residence times of some lakes in northwestern Ontario, Canada. In the period 1970 to 1986, rainfall in the area decreased from 1,000 millimeters to 650 millimeters per annum, while above-average temperatures speeded up the evapotranspiration rate (the rate at which water is lost to the atmosphere through evaporation and the processes of plant life).
The result has been that the residence time of one of the lakes increased from 5 to 18 years during the study period. The slowing down of water renewal leads to a chain of further consequences; it causes dissolved chemicals to become increasingly concentrated, and this, in turn, has a marked effect on all living things in the lake.

湖里的水从哪里来,又怎么流出的呢？湖中的水来自于河流的水,地下渗入的水和泉水，从四周地面流进来的水,还有直接降到湖面的雨水。湖中的水通过向外流的河流,渗透进河床以及蒸发离开湖泊。这些都是显而易见的。 当考虑到实际的水流量时就会变得更加复杂:水通过上述方式流进和流出的量是多少？了解河流的流进量和流出量是一件测量每一条流入和流出的河流其容量的事情。 和大气的交换是通过发现雨水中得到的水(通过雨量器大致测得)和蒸发损失的水（通过准确测量其他的水损失来源的模式测得）的差别来计算的。对于大多数的湖来说,特别是那些被森林环绕的湖,地面流入的水太少了以至于几乎没有能够感觉到的影响。河水和大气水量变化不能解释湖中水平面的变化， 是因为渗入湖中的地下水和渗出的地下水的净值不同。注意一下“净值”这个词:测量真正渗入和渗出湖中的地下水量比仅仅推断它们的不同要复杂的多。 一旦所有的这些信息都收集到了,那么判断一个湖的流量是由表面输入或输出决定的还是由地下水进出量决定的就变得可能了。如果主要是前者决定,那么湖泊就是一个表面水主导的湖,如果是后者,那么它就是一个渗透水主导的湖。有时候,常识会告诉你这两种可能性哪一种在起作用。 比如说一个丘陵地区的池塘在整个干燥的夏天尽管没有溪水注 入仍能保持稳定的水位，那么显然它是一个渗透水主导的池塘。相反,一个池塘有河流进和流出,随河水的干枯而干枯,那么这就是一个表面水主导的池塘。 不管怎么说，湖泊是在不停地流进和流出水；它的水不会停留在湖里，或者说不会长久的停留。 这个会增加湖泊的停留时间。 停留时间指的是特定水分子在湖中停留的平均时间长度，是通过计算湖水流量流出湖泊的速度计算出来的。停留时间是一个平均数；湖中特定分子（如果我们可以追踪它的路线的话）花费的时间取决于它的路线：它可能是最快最直接的水流的那一部分流过，或者它可能在在逆流中无限长的时间里打圈。 停留时间变化非常的大， 从小型湖的几天到大型湖泊的几百年。 加利福尼亚州的塔霍湖的停留时间就长达 700 年。北美五大湖也就是苏必利尔湖、密歇根湖、休伦湖、伊利湖和安大略湖的停留时间分别是 190 年、100 年、22 年、2.5 年和 6 年。伊利湖是最短的：尽管它的面积比安大略湖要大， 它的容量不及后者的三分之一， 因为它的平均深度还不到 20 米。 给定的湖泊的停留时间绝不是一个确定的值。 它取决于水流进湖的速度， 而这个速度取决于降雨量和蒸发速度。气候变化(全球变暖的结果？)严重影响着加拿大安大略湖西北部一些湖泊的停留时间。在 1970 年到 1986 年间，这个地区的降雨量由每年 1 000 毫米降到了 650 毫米， 而同时平均温度的上升加快了蒸散率 （这个速率指的是水蒸发到大气的速率以及植物生命的过程） 。结果是，在研究期间其中一个湖的停留时间从 5 年增加到 18 年。湖水的更新变慢导致了一系列后果；它使得溶解的化学物质不断变浓，这样反过来会对湖中的生物造成显著的影响。