The delta is dominated by approximately 25,000 lakes. These lakes are not static features, but are constantly changing. Changes are caused by the deposition of sediments within the lakes and in the areas surrounding the lakes, the draining of lakes as main river channels cut into the banks separating the channel from the lakes, the division of large lakes into a number of smaller lakes as deltas are formed in these large lakes, the enlarging of lakes due to melting of permafrost in the lake shoreline, and the creation of new lakes as channels are abandoned. It is, in fact, the large number of lakes and the great variety of lake size, shape and hydrology that make the Mackenzie Delta so dynamic and productive. Only northern deltas have such a large number of lakes, as most temperate and tropical deltas have large areas of marshes and swamps, and few lakes.

Given the great diversity and importance of lakes in the Mackenzie Delta there are a number of important questions, including:

1. Why are there so many lakes in the delta?
2. Why don't the delta lakes dry up? Why do the lake levels change over time?
3. How often are lakes flooded by the Mackenzie River?
4. What would happen if lakes were not flooded by the Mackenzie River?
5. What causes changes in the water level in the channels of the Mackenzie River?
6. Do storm surges affect channel or lake levels all through the delta?
7. What controls the thickness of lake ice?
   

In the following sections we will try to briefly answer these questions. Other questions we could ask include:

1. Are there different types of lakes?
2. How often does the water level fall low enough so that the channels connecting the lakes to the Mackenzie River dry up?
3. How much sediment accumulates in a lake each year?
4. How long would it take for a lake to fill up with sediment?
   

Why are there so many lakes in the delta?

Unfortunately, we really cannot answer this question to any great degree of certainty. However, since most temperate and tropical deltas do not have large numbers of lakes we expect that their occurrence in the Mackenzie Delta is somehow related to the cold conditions of this area. One hypothesis is that the small number of lakes in temperate areas is the result of the rapid infilling of lake basins with organic material (primarily plant remains) due to the high productivity of these areas. For example, in the Mississippi and Fraser Deltas, there are over 5m of peat accumulation in many areas. Such peat accumulations do not occur in the Mackenzie Delta. It seems likely then that the continued existence of lakes in the Mackenzie Delta is due to the slow accumulation of organic material within the lakes.

Why don't the delta lakes dry up?

Figure 5. Diagram of water cycle for a delta lake showing all of the components of the water balance and the location of permafrost in the area around the lake.

A delta lake receives water from a number of sources (Figure 5), including rainfall and snowfall directly onto the lake surface, floodwater from the Mackenzie River, and runoff from the land surrounding the lake. A delta lake may lose water by evaporation from the lake surface, groundwater flow out of the bottom of the lake, or as flow out of the lake through a channel. Although groundwater may either enter or leave a delta lake through the unfrozen area beneath the lake (called a talik), it is usually a very small amount since the lakes are underlain by silts and clays that have a low permeability to water. The addition of all of these sources and losses of water is called the water balance. If the water balance is positive, the lake is gaining more water than it is losing and the lake level will rise. If the balance is negative, the level will fall. Lakes typically undergo periods when the balance is positive and periods when it is negative. If the balance is negative for a long period of time, the lake will eventually disappear. Measurements at the study lakes (Figure 6) have allowed us to determine the magnitude of all of the water balance components mentioned above. Briefly, these measurements were conducted as follows: snow added directly to the lake over the winter is estimated by measuring the depth and density of snow on the lake ice, while rainfall onto the lake during the summer is measured by a tipping bucket recorder. Runoff into the lake is determined by measuring water flow in small rills draining the land area around the lake, and by measuring groundwater flow through the thin unfrozen layer overlying the permafrost. Evaporation from the lake is estimated by standard equations that use air temperature, net radiation, and lake water and bed temperature. Groundwater flow beneath the lake is measured using narrow pipes installed down to various depths in the lake bed. Water flowing into or out of the lake through the lake channel is determined using a device that measures the water depth, velocity and flow direction (either into or out of the lake) within the lake channel.

Figure 6. Photo of field measurements at NRC Lake, showing a tipping bucket rain gauge, an anemometer and instruments for measuring air temperature, relative humidity solar and net radiation. Also visible is a solar panel for recharging the batteries required to run the scientific equipment.

This research has shown that for Skidoo and South Lakes the channel flow in and out of the lake is the largest part of the water balance. For NRC Lake, however, the channel flow is only important during the spring since it is dry during the summer. Since there is usually little runoff from the lake basin to the lake and evaporation is usually greater than summer rainfall, the lake level falls over the summer period. If flooding by the Mackenzie River does not occur, lake level will continue to decline over a number of years. This suggests that without flooding by the Mackenzie River, lakes would disappear within about 10 years.

How often are lakes flooded by the Mackenzie River?

How often a lake is flooded by the Mackenzie River depends on where the lake is located in the delta, and how high the lake is perched above the channel. NHRI has measured how high lakes are perched at three locations down the east side of the delta, and at locations between Inuvik and Aklavik (see Figure 2 for locations). In all, we have measured approximately 3,500 lakes. In the upper delta towards Point Separation, the lakes are perched up to 7 m above the normal late winter main channel levels, while in the lower delta near Reindeer Station, the highest lakes are only perched up to 3 m above the normal late winter levels. Near Horseshoe Bend, lakes are perched higher than on either side of the delta (Figure 7). For one of the study areas near Inuvik, the elevation of the perched lakes was compared to the records of water level in East Channel. Through this comparison, one is able to see how often during the last twenty years various lakes have been flooded. In the Inuvik area the largest flood since 1964 occurred in 1972, and during that year all lakes in the study area were flooded. The lowest flooding occurred in 1984 when only the lowest 2/3 of lakes were flooded in the area near Inuvik. This work has shown that 2/3 of lakes are flooded every year in the spring, while the remaining 1/3 of lakes are flooded between every two and five years. In the upper portion of the delta, many lakes are perched higher, and they therefore flood less frequently. In the highest lakes in this area, flooding may only occur every 10 years on average. Some of these lakes undergo large declines in water level and in fact may nearly dry up before being flooded again.

Figure 7. Changes in sill elevation for lakes in the East, Middle and Peel Channel areas along the cross delta transect shown in Figure 2. This diagram shows that, although there are low and high elevation lakes in all areas, there are more high elevation lakes in the area of Middle Channel that in the areas of the east and west side of the delta. A similar diagram for areas in the upper and lower delta, would show that the number of high elevation lakes decreases as you go from near Separation Point to the Beaufort Sea.

What would happen if lakes were not flooded by the Mackenzie River?

As discussed previously, one third of lakes in the Inuvik area may not be flooded during the spring period, and they are not flooded during the summer period. What would happen to these lakes if spring flooding no longer happened? Although it may seem unlikely that this could occur, it must be remembered that after many extensive floods over the period prior to 1972, the Peace-Athabasca Delta underwent a period of over 20 years (1972 to 1996) when flooding did not occur and many lakes dried up. This had a major impact on the wildlife of the delta and on the people who use the delta. Recent research has shown that the reasons for this decline in flooding are probably related to both the construction of a dam on the Peace River, and decreased snowfall and snowmelt runoff to the Peace River.

Mackenzie Delta studies have clearly shown that there is very little snowmelt runoff from the land area surrounding the lakes. This is because most of the water is absorbed by the frozen soil and subsequently evaporates or transpires from these soils. Moreover, evaporation from the lake surface is usually larger than the total of summer rainfall and winter snowfall. The result is that in most years, the lake water balance is negative and therefore the lake level declines. This was observed for a lake in the upper delta that was flooded in 1982, but not flooded in the following years, and as a result the lake level gradually declined.

These results seem to suggest that many of the perched lakes would disappear without frequent flooding from the Mackenzie River. In addition, since lake ice is usually 0.6 to 1.5 m in thickness, and the high perched lakes are often only 1 to 2 m in depth, even a small change in lake depth would result in the lakes freezing to the bed over their entire area, making it unsuitable for muskrats over the winter period.

What causes changes in the water level in the channels of the Mackenzie River?

As those who boat in the Mackenzie Delta know, the water levels in these channels are constantly changing (Figure 8). In autumn, water levels gradually decline due to reduced runoff from the upstream areas of the Mackenzie River basin. However, when the channels begin to freeze in October, water moves more slowly through the channel because of the formation of ice. As a result, the water level in the channel rises. This causes an increase in the amount of water stored in the channel and the channel discharge actually decreases for a brief period of time, after which it increases and then stabilizes for much of the winter. This is often the period with the lowest discharge of the year. For the remainder of the winter, the channel water levels continue at a low level. In the small channels of the delta, discharge may cease as the channels freeze to the bed and as a result, the discharge of the large channels may increase as flow is diverted from these smaller channels. The rising water levels in these larger channels, may overflow into low lying lakes during this period.

Figure 8. Example of water level for East Channel at Inuvik for a typical year (in this case 1974). The highest water level for the year occurred during the spring breakup in late May and early June, and lower water levels occurred during the summer period. Note, however, that there was a period of high water levels during August due to a rain storm upstream in the Mackenzie River basin. Also shown for comparison are four important water levels: the highest water level on record, the mean spring break, the mean summer peak and the mean summer low.

In late April and early May, the water level in the channels begins to rise as the first snowmelt runoff from the southern portions of the Mackenzie Basin reaches the delta. The first sign of the arrival of this water is the occurrence of overflow water on the ice of the channels. Water levels continue to rise as the crest of the flood-wave approaches the delta. As discharge and water levels increase, ice jams may form in the larger channels, especially in Middle Channel, causing the water level to rise dramatically. Two locations that are very prone to the formation of major ice jams are Middle Channel just downstream of Point Separation and the Horseshoe Bend area. Other channels are prone to the development of smaller ice jams (see this page).

In the summer, water levels may rise in response to rainstorms upstream of the Mackenzie Delta. The largest summer floods occur due to rain in the Mackenzie Mountains and in the headwaters of the Liard River. During the highest summer floods, only two thirds of lakes in the Inuvik area of the Mackenzie Delta are flooded.

In the summer, water levels in the delta are also affected by changes in sea level in the Beaufort Sea. Since the maximum tidal range of the Beaufort Sea is only about 0.37 m, channel water levels are only affected by tidal activity in the area seaward from a line approximately from Tununuk to 10 km south of Shallow Bay and then west to the western edge of the delta (Figure 1). This area affected by tidal activity covers about 30% of the entire delta. In addition to this tidal activity, when there are periods with high winds from the north and north-west, sea level of the Beaufort Sea may rise by up to 2.4 m. These events are called storm surges (see this page).

Even with a storm surge of about 1.5 m, water level in the Delta channels rises in response as far upstream as Tsiigehtchic.

What controls the thickness of lake Ice?

The thickness of ice on the lakes of the Mackenzie Delta is very important in determining the suitability of lakes as muskrat habitat and in deciding if the ice is safe to travel on. Ice thickness is controlled by both the air temperature and the snow depth, with the ice being thickest with colder temperatures and thinner snow depths. In fact, snow depth is usually as important as air temperature in controlling ice thickness. In years where there is little snow, or on lakes which are windswept, the ice will be thicker. Observations suggest that if water depth below the ice cover is less than approximately 0.5 m, the lakes are not used by muskrats. Such conditions may occur in very shallow lakes, during winters when the lake level is low, or during years when the ice is thicker than normal. Therefore to determine the impact of climate change on the delta lakes, measurements of both lake level and ice thickness are required.