Skip booklet index and go to page content

Taiga Plains Ecozone+ Evidence for Key Findings Summary

Theme: Biomes

Key finding 1

Theme Biomes

National key finding
At a national level, the extent of forests has changed little since 1990; at a regional level, loss of forest extent is significant in some places. The structure of some Canadian forests, including species composition, age classes, and size of intact patches of forest, has changed over longer time frames.

Ecozone+ key finding: Boreal forest is the dominant ecosystem type in the Taiga Plains. Fragmentation from roads and other linear development, resulting in loss of large intact blocks of forest, is most evident in northeastern BC. Climate-related changes in the treeline zone at the north of the ecozone+ include increased shrub growth, a small net increase in tree cover resulting from increased conifer cover at the northern part of the treeline zone balanced with reduction in coniferous forest in the south of the zone (1985-2006), and reduced growth rates, likely due to drought stress, of the majority of white spruce trees since the 1930s.

Spatial characteristics

The Taiga Plains ecozone+, with its large variations in latitude, elevation, and climate, varies in density of forest cover and degree of forest fragmentation. It is an area of frequent large wildfires and thus the vegetation is often a mosaic of uneven-aged forest at different stages of regeneration.Reference 13 Predominantly coniferous forest covers the valleys of the Mackenzie River and its tributaries, all the way to the Mackenzie Delta, although the lowlands of the Liard Valley tend toward mixed woods. Slightly higher elevations, such as the Cameron Hills, and regenerating burns are shrub-covered, while the highest elevations, primarily the eastern slopes of the Mackenzie Mountains, are characterized by tundra vegetation.

Much of this ecozone+ exhibits a proportion of forest greater than 50%. Lower forest densities are found immediately south of Great Slave Lake in the northern portion of Wood Buffalo National Park, the uplands near Norman Wells, in a large area west of Lac la Martre that burned in the mid-1990s, and in portions of the lower reaches of the Mackenzie Valley.Reference 13

Characteristics of Canadian forested regions were examined using remote sensing data.Reference 13 The habitat requirements for many species are strongly influenced by the spatial characteristics of land cover types. These spatial characteristics can include the proportion of particular land cover types in an area and the amount of fragmentation and connectivity of particular land cover types. The presence of edges, implying a certain degree of fragmentation, is important for many species, while others, notably woodland caribou, are adversely affected by fragmentation.

Two methods were used to examine forest spatial characteristics and provide a baseline for future trend analyses, both calculated from the Earth Observation for Sustainable Development dataset of pixels at 30 m spacing within 1 km2 cells (from the year 2000):

  1. forest density (proportion of land area that is forested): results are shown in Figure 5; and,
  2. edge density (the length of all edges between forested and non-forest pixels in each 1 km2 cell): The forest edge density in the Taiga Plains is higher than in many more southerly forests, with a typical value of 250 m/km2, increasing to 500-600 m/km2 in tundra areas flanking the eastern foothills of the Mackenzie Mountains.
Figure 5. Map showing forest density, Taiga Plains, 2000

Proportion of 30 m2 pixels that are forested in each 1 km2 unit is shown. The northern white section is an artefact of the methodology – it corresponds with the boundary of the Taiga Plains ecozone under the 1995 classification. Density is not the number of trees, but the appearance of land cover from space. Lower proportion of forest may represent stands of black spruce or jackpine or regenerating forest with sparse canopy cover, while the higher forest proportion areas may represent mature white spruce stands with a lower density of trees but a dense canopy cover.

Source: Ahern et al., 2011Reference 13
Long description for Figure 5

This map shows forest density in the Taiga Plains ecozone+. The map is derived from remote sensing imagery; each 1 km2 unit on the map is coded to reflect the proportion of pixels in the image that are forested, and the proportion of forest is derived from canopy cover. Most of the ecozone+ is a complex mosaic of forest density, except the southwest section of the ecozone+, which generally has high forest density.  The Mackenzie Delta region, areas to the west of Great Bear Lake and South of Great Slave Lake, and a large area in the centre of the ecozone+ have low (<30%) forest density.  Areas with lower proportion of forest may represent stands dominated by black spruce or jackpine, while mature white spruce with fewer trees will have dense canopy cover and will be depicted as high density forests.  A white space at the north edge of the ecozone+ is an artefact of different ecozone boundaries when the map was produced under the 1995 ecozone classification.

Top of Page

Intact forest blocks

Figure 6 shows areas of intact forest blocks in the Taiga Plains ecozone+. Note that northeastern BC, which has the lowest density of intact forest blocks, is a region with high forest density (Figure 5). This indicates that the lack of extensive areas of intact forest is due to fragmentation of the forest, rather than to large-scale land conversion or the presence of other natural land cover types.

Figure 6. Intact forest blocks, Taiga Plains ecozone+

Green shows contiguous blocks of forest undisturbed by anthropogenic features. Minimum block size is 10,000 ha. The northern white section is an artefact of the methodology – it corresponds with the boundary of the Taiga Plains Ecozone under the 1995 classification.

Source: based on data compiled by Lee et al., 2006Reference 25
Long description for Figure 6

This map depicts the distribution of intact forest blocks in the Taiga Plains ecozone+. To be considered intact, forest blocks must be a minimum of 10,000 ha in area and be undisturbed by anthropogenic features. The majority of the ecozone+ is covered by intact forest blocks, with the exception of the Peace Region in the northeast corner of BC. This region has a high density of forest, as indicated in Figure 4, but the low density of intact blocks indicates that forest fragmentation is prevalent, as land conversion is greatest in this area of the ecozone+.

The treeline zone

The following text box, which provides a broader perspective on treeline change, is excerpted from Canadian biodiversity: ecosystem status and trends 2010.Reference 3

Changes in the treeline zone

The term “treeline” is deceptive – there is not a sharp line where trees end, but rather a zone of transition from increasingly sparse trees to tundra. Treeline zones in Canada are both latitudinal, across the north of the country, and altitudinal, on the slopes of hills and mountains. The emerging picture is one of change, but not a uniform expansion of the treeline. In northern Quebec, trees in the forest-tundra zone have grown faster and taller since the 1970sReference 26 but distribution of trees has not changed greatly.Reference 27 In Labrador, treelines have expanded northward and up slopes over the past 50 years along the coast, but not inland.Reference 28 In the mountains of northwestern Canada, tree growth and density have changed more than the position of alpine treelines.Reference 29

A study on the treeline in western Canada found only a small net increase in tree cover, but major changes in vegetation within the treeline zone. Tree cover increased in the northern half of the zone, but this was mainly offset by decreases in the southern half, especially west of the Mackenzie Delta – likely related to drier conditions due to higher temperatures.Reference 30 The biggest changes were an increase in shrubs and, in the northwest of the treeline zone, a replacement of lichen cover and bare land with small, non-woody plants (herbs).

Figure 7. Vegetation changes in the treeline zone of Western Canada, between 1985 and 2006

Mean change across the zone over 22 years based on analysis of early spring and summer satellite images.

Source: data from Olthof and Pouliot, 2010Reference 31
Long description for Figure 7

This graph shows the average change in cover of bare ground, lichen, conifer, herb and shrub vegetation categories at the treeline zone over a period of 22 years. The inset map depicts the area analysed in the study, which extends along the treeline zone from the northern part of the Yukon across the Northwest Territories and into the southeastern Nunavut/northern Manitoba. Over the 22 year period, tree cover changed very little (<1%), but herb and shrub cover increased by 12.5% and 15%, respectively. Cover of bare ground decreased by about 9% and lichen cover decreased by approximately 3.5%. The data is based on analysis of early spring and summer satellite images.

Top of Page

Global perspective

Since 1900, treeline has advanced at 52% of the 166 sites examined around the world and has receded at only 1% of the sites.Reference 32

Several studies have shown strong correlations between summer temperatures and variation in vegetation patterns at the northern edge of the treeline zone in the Taiga Plains, indicating that increasing temperatures are likely to alter shrub abundance, vegetation structure, and species composition.Reference 33, Reference 34 There are indications that this change is underway. Lantz et al., 2010Reference 35 in a study along a transect from the Beaufort coast to the region of Fort McPherson in the south, found that green alder (a tall shrub) at sites at the northern edge of the forest-tundra transition zone showed patterns of recruitment markedly different from green alder at sites to the south. There was a higher proportion of younger shrubs in the northern transition zone, indicating recent colonization of the sites by green alder. The study also showed that green alder growth and reproduction were significantly greater on burned sites, with tall shrubs dominating burned sites in all vegetation zones. The combined effects of increased fire and warmer growing seasons are likely to result in continued northern movement of tall shrubs.

As in other regions of northwestern North America, white spruce stands in or near the treeline zone of the Taiga Plains show signs of decreased growth as the climate warms – perhaps due to the crossing of a physiological threshold for summer temperature and/or drought stress in the warmer summers.Reference 36-Reference 38 An analysis of annual growth, based on tree ring width, of 654 white spruce trees from 9 sites in the Mackenzie DeltaReference 30 showed that there was a high degree of similarity in growth rates among all trees up from 1600 to about 1930, when growth rates diverged. From about 1930 to the end of the study in 2003, growth rates for about 25% of the trees increased (this group is called positive responders), while growth rates for the remaining 75% of the trees (negative responders) declined (Figure 8). White spruce growth rates were compared with climate records from Inuvik (starting in 1927) and with Northern Hemisphere growing season temperatures (records starting in 1856). Annual growth of positive responders was strongly correlated with June temperatures in Inuvik and with North American growing season temperatures, while the low annual growth rates of negative responders were inversely related to temperatures from the current and, especially, the previous summer. This apparent slowing of growth in warmer summers was somewhat mitigated in years with higher April precipitation – an indication that both temperature stress and drought stress may be affecting white spruce in the region.

Figure 8. Growth of white spruce in the Mackenzie Delta, reconstructed from tree rings, 1600-2003, plotted with Northern Hemisphere growing season temperature anomalies, 1856-2003

The black line that starts in 1856 is the trend in temperature anomalies. The two groups of white spruce are the lines starting in 1600: the red line represents the average annual growth “negative responders” (trees showing an inverse relation between growth and summer temperatures after 1930) and the blue line represents the annual growth of “positive responders” (trees showing a positive correlation between growth and summer temperatures). The correlation coefficients show the strength of the relationships between the growth rates of the two groups of trees from 1600 to 1899 (r=0.91) and from 1900 to 2003 (r=0.11).

Source: adapted from Pisaric et al., 2007Reference 30. Reproduced with permission of John Wiley & Sons, Inc.
Long description for Figure 8

This line graph tracks the growth of white spruce in the Mackenzie Delta, reconstructed by tree rings spanning 1600-2003. Starting in 1856, the graph also plots the Northern Hemisphere growing season temperature anomalies. From 1600 to the early 20th century, white spruce is relatively stable with respect to annual tree ring growth. In the early 20th century, tree ring growth starts to show greater variation: while some trees (25%) showed increased growth in response to a significant increase in summer temperatures in the second half of the 20th century, the majority of trees (75%) had an inverse relationship to increased summer temperatures.

Top of Page

Key finding 3

Theme Biomes

National key finding
High loss of wetlands has occurred in southern Canada; loss and degradation continue due to a wide range of stressors. Some wetlands have been or are being restored.

Ecozone+ key finding: Wetlands are diverse and widespread in the ecozone+ and are vulnerable to anthropogenic threats including climate change. Periodic spring flooding along the Mackenzie River Basin, which maintains the diversity of delta lakes, has been shown to be more related to climate variables than to the influence of the upstream W.A.C. Bennett dam. There are, however, indications that spring flooding may be less frequent. Delta lakes are affected by the longer ice-free season but also by increased erosion from permafrost slumping, which causes abrupt changes in water quality.

Wetland types include extensive river deltas, floodplain lakes and ponds, meandering river channels bordered by wetlands, thermokarst lakes, peatlands, and marshes. The wetlands of the Taiga Plains provide habitat for hundreds of thousands of migrating and nesting water birds, as well as supporting a diversity of fish and providing habitat for mammals including moose, caribou, muskrats, and beavers. Wetlands are traditional hunting and fishing locations and important culturally to the residents of the Taiga Plains. This section focuses on aspects of wetlands undergoing change or vulnerable to change from anthropogenic threats.

As with the boreal forest in general, large blocks of undisturbed wetlands are important in maintaining wetlands biodiversity. Some species are not tolerant of disturbance and fragmented habitat – the extreme example for the ecozone+ being the whooping crane, a water bird that was driven almost to extinction from habitat loss. Disturbance and habitat alteration in nesting grounds may also be linked to reductions in species of waterfowl breeding in the Taiga Plains (see key finding on Species of special interest on page 54).

Ramsar sites

There are two Ramsar sites (wetlands designated as being of international significance) in the Taiga Plains ecozone+:

  1. Hay-Zama Lakes, in Alberta. This 486 km2 complex of lakes and wetlands, a staging area for migrating waterfowl, is protected as an Alberta park. Oil and gas exploitation, however, predated the creation of the park and is permitted to continue until reserves are depleted.Reference 39 It is a traditional harvesting area for the Dene Tha’.
  2. Whooping Crane Summer Range, in NWT and Alberta, within Wood Buffalo National Park. This 16,895 km2 area is the only remaining natural nesting area of the whooping crane; it contains thousands of water bodies including lakes, bogs, marshes, shallow ponds and streams.Reference 20

Ecozone+ wetlands and ponds are formed and maintained by low evapo-tranpiration rates and permafrost conditions as well as by the physical depressions left by glaciations, and thus are vulnerable to environmental shifts. While there are indications of changes in ponds in the Taiga Plains due to warming and thawing permafrost (discussed below), no widespread reduction in pond area has been observed in the ecozone+. Rising temperatures causing increased evapo-transpiration and changing permafrost have led to a regional trend of shrinking ponds in Alaskan boreal forestsReference 40 and there has been some reduction of total pond area in the Old Crow Flats in northern Yukon (Taiga Cordillera ecozone+).Reference 41 As wetlands dry out and fire frequency increases due to global warming, wetland plant species composition could change dramatically because of deep burning into the organic layers.Reference 42

Top of Page

Deltas and river-associated lakes and wetlands

The extensive Mackenzie River Delta (13,000 km2),Reference 43 with its 45,000 lakes, is partly in the forest and shrub zones of the Taiga Plains ecozone+ and partly across the tundra of the Arctic ecozone+. It is a productive aquatic ecosystem, unusually so for its latitude.Reference 44 Its high productivity is considered to be related to both the longer ice-free season of floodplain lakes (compared with nearby lakes) and their replenishment from the nutrient-rich river sediments.Reference 44 Many of the lakes of the Mackenzie Delta and throughout the continuous permafrost zone are thermokarst wetlands, formed in depressions on top of permafrost.

There are also two major freshwater deltas in the ecozone+: the Slave (the mouth of the Slave River, flowing into Great Slave Lake), and Mills Lake (a widening of the Mackenzie River at the mouth of the Horn River, near Fort Providence). The Slave River Delta, covering an area of 554 km2,Reference 45 is a stopover in spring and fall for birds on all four major continental flyways.Reference 45, Reference 46 Mills Lake (381 km2), is a major staging area for waterfowl during spring and fall migrations, a refuge for moulting diving ducks in summerReference 45 and a grazing area for wood bison.Reference 47

Flood regime

Periodic flooding associated with spring discharge and ice events creates and maintains the diversity of habitat provided by the lakes and wetlands of the Slave and Mackenzie river deltas.Reference 48Reference 49Reference 50Reference 51 A study of lakes and ponds in the Slave River Delta (2003 to 2005) indicated that the degree and frequency of river flooding is the dominant factor controlling water chemistry and plant and diatom plankton communities and biomass in each body of water.Reference 52

The Slave River is influenced by flow regulation from the W.A.C. Bennett dam and associated reservoir (see Lakes and rivers key finding). A study that reconstructed the frequency of spring break-up flood periods over 80 yearsReference 48 showed that floods have tended to be cyclical, alternating through periods of about a decade each of high and low flooding (Figure 9). Spring floods appear to be more related to climate-driven conditions in the upper reaches of the Mackenzie River Basin than to water regulation by the W.A.C. Bennett dam (which started in 1968). Periods of low flooding preceded flow regulation and periods of high flooding followed the onset of flow regulation. The authors predicted that floods will become less frequent with lowered snow pack and thus less headwater runoff resulting from climate change. There are indications that this trend to reduced spring flooding may have begun in this and other parts of the watershed.Reference 48

Figure 9. Flood events in the Slave River Delta, 1925-2005

Flood events are reconstructed from proxy data, as indicated (columns a through c) and from measured discharge (column d). Shaded boxes show periods of high flood frequency inferred from each proxy and the horizontal boxes outlined in dashed lines indicate the main periods of high flooding.

Source: Brock et al., 2010Reference 48. Used with permission of Canadian Water Resources Journal
Long description for Figure 9

This panelled graph presents three key geochemical and biological parameters that are related to flood frequency, measured in a sediment core in the Slave River Delta, 1925-2005. Discharge data on the Slave River is shown from 1960 to 2005, and was influenced by the completion of the W.A.C. Bennett dam in BC in 1968. Panel a) shows the carbon-to-nitrogen (C/N) ratios measured in the sediment core from 1925-2005; floodwater is associated with high C/N ratios and the high levels measured in the early and late 1970s and around 1995 correspond with high levels of discharge measured in the Slave River. Similarly, the diatom-based flood index, which assesses the relative abundance of diatoms know to be flood indicator species, corresponds roughly to high discharge periods; lower discharge periods have a substantially different index value. Panel c) shows data on the number of seeds of an aquatic plant (Sagittaria cuneata) that corresponds to a period of very low discharges in the 1950s. The sediment core data (C/N ratio and diatom index) suggests that a flooding event occurred in the early 1940s.

Top of Page

Permafrost degradation

Changes in spring melt and break up are also leading to changes in water quality due to increased erosion from permafrost slumping.Reference 46, Reference 53

Melting and slumping permafrost erodes and alters the physical configuration of wetlands as well as the water quality and shoreline and lake bottom characteristics.Reference 19 An analysis of aerial photographs at 23 study sites in the Mackenzie DeltaReference 54 showed an increase in thaw-slump activity from 1950 through 2004 (Figure 10). Retrogressive thaw slumping is a slope failure caused by thawing of ground ice and slumping of thawed soil, forming headwalls.Reference 55

Figure 10. Increase in retrogressive thaw slumps, Mackenzie Delta, 1973-2004 compared with 1950-1973
  • Average annual rates of slump growth (based on changes in area of disturbance from thaw slumping)
  • Average annual rates of headwall retreat

Error bars are +/- SE.

Source: Lantz and Kokelj, 2008Reference 54. Reproduced with permission of John Wiley & Sons, Inc.
Long description for Figure 10

These two bar graphs present data documenting an increase in retrogressive thaw slumps in the Mackenzie Delta during two time periods: 1950-1973 and 1973-2004. Increased retrogressive thaw slumping, measured by increases in slump area and slump headwall retreat, indicates increasing thawing of permafrost. Graph A shows a significant increase in slump growth (m2/year) in the 1973-2004 time period compared to 1950-1973. Slump headwall retreat (m/year) in graph B is on average twice that of the preceding time period, at just over 1 m/year between 1973 and 2004.

Top of Page

Impacts of changes in the deltas on biodiversity

Permafrost degradation was the main environmental factor explaining differences in water quality in a study of 73 lakes, about half of which were affected by retrogressive thaw slumping, in the Mackenzie Delta (both in the tundra region, Arctic Ecozone+, and in the region along the boundary with the Taiga Plains Ecozone+).Reference 53 The effect was mainly on water clarity and the concentration of ions, rather than on total organic carbon. The affected aspects of water quality are strong determinants of lake biotic communities. The authors concluded that the abrupt changes in lake chemistry brought about by thaw slumping can be expected to lead to abrupt shifts in aquatic food webs.

The wide ranges in extent of and in periods of connectivity between Slave and Mackenzie delta lakes and rivers are important in creating a diversity of habitats, which are then able to support many different communities of species of invertebrates, water birds, fish, and mammals.Reference 46, Reference 50 Changes in river flow (reduced flooding) combined with the rise in sea level associated with climate change may result in fewer types of wetlands, with a lowering of habitat diversity in the Mackenzie Delta.Reference 50

Top of Page

Key finding 4
Lakes and rivers

Theme Biomes

National key finding
Trends over the past 40 years influencing biodiversity in lakes and rivers include seasonal changes in magnitude of stream flows, increases in river and lake temperatures, decreases in lake levels, and habitat loss and fragmentation.

Ecozone+ key finding: The most widespread hydrological change is a trend to increased minimum and winter flows, both in the Mackenzie River as a whole (including tributaries upstream of the Taiga Plains) and in several smaller rivers monitored within the ecozone+. While upstream tributaries to the Mackenzie River are generally exhibiting trends to earlier peak flows, there is no clear trend in timing at most sites on smaller watercourses within the ecozone+. There are indications of a trend to increased streamflow variability within the ecozone+, with implications for riparian habitat.

The Taiga Plains Ecozone+ drains to the Arctic Ocean through the Mackenzie River catchment area. As Canada’s largest river basin (draining 20% of the nation’s area),Reference 56 the Mackenzie River drains a total area of 1,787,000 km2. The Mackenzie River Basin (Figure 11 ) includes a number of other important river systems, including the Athabasca, Peace, Liard, Slave, Arctic Red, and Peel rivers.Reference 57, Reference 58 The extensive Mackenzie River Delta is partly within the Taiga Plains Ecozone+ and partly within the Arctic Ecozone+. The Mackenzie basin has three major lakes: Lake Athabasca (along the Taiga Shield/Boreal Shield ecozones+ boundary), Great Slave Lake (partially within the Taiga Plains Ecozone+ and partially within the Taiga Shield Ecozone+), and Great Bear Lake, which is fully within the Taiga Plains Ecozone+. There are two major freshwater deltas: the Peace-Athabasca (flowing into Lake Athabasca, in the Boreal Shield Ecozone+) and the Slave (flowing into Great Slave Lake, within this ecozone+). Within the Taiga Plains Ecozone+ the Horn River also forms a delta at its confluence with the Mackenzie River. Lakes and wetlands of the Mackenzie and Slave deltas are discussed in the Wetlands key finding, above. Changes related to lake and river ice are discussed in the Ice across biomes key finding, below.

Figure 11. Sub-basins of the Mackenzie River Basin

Sub-basins: 1. Athabasca; 2. Peace; 3. Liard; 4. Peel; 5. Great Slave; 6. Mackenzie-Great Bear

Source: Mackenzie River Basin Board, 2004Reference 59. Reproduced with the permission of the Mackenzie River Basin Board.
Long description for Figure 11

This map shows the six sub-basins of the Mackenzie River basin: the majority of the Taiga Plains ecozone+ falls within the four more northern sub-basins: the Liard, Peel, Great Slave, and Mackenzie-Great Bear. The remainder falls within the two southern sub-basins: the Athabasca and Peace.

Top of Page

Trends in hydrology for the Mackenzie River

The total annual discharge from the Mackenzie River did not change from 1968 to 1999, despite significantly increased air temperatures in the river basin over this period.Reference 60 However, changes in timing and seasonal distribution of river discharge have occurred, with the strongest single effect of climate change being the trend to earlier peak flows at different locations throughout the basin, correlated with increasing spring temperatures.Reference 19 Trends towards an earlier onset of spring freshet were detected in analyses of the Mackenzie River and its main tributariesReference 61-Reference 63 and were shown to be correlated with increases in air temperatureReference 62 and with climate oscillations.Reference 61 Figure 12 shows this trend for the Liard River, measured at Upper Crossing, Yukon, upstream of the Taiga Plains Ecozone+.

Figure 12. Trend to earlier spring freshet date in the Liard River at Upper Crossing, upstream of the Taiga Plains Ecozone+, 1961-2005
Source: Burn, 2008.Reference 61 Reprinted with permission from Elsevier
Long description for Figure 12

This line graph presents the spring freshet date from 1961-2005 in the Liard River at Upper crossing, a water quality monitoring station upstream of the Taiga Plains Ecozone+. Trend lines from 35, 40 and 45 years of data are similar and indicate a consistent trend towards  an earlier spring freshet date, though the observed values for freshet data are variable from year to year.

Top of Page

Analyses of hydrometric data at the river basin scale demonstrate a widespread trend to increased minimum and winter flows: in the Liard, 1960-1999,Reference 62 and in the Mackenzie, Liard, Athabasca, Peace, Slave, and Peel rivers, 1960-2000.Reference 63 The increased winter flows in the Liard River (1960-1999) have been attributed in part to the Pacific Decadal Oscillation.Reference 62 Increased winter flows have also been linked to thawing of permafrost from climate change.Reference 19

Top of Page

Trends in hydrology within the ecozone+

Indicators of hydrological alteration

This section is based on Canada-wide analyses performed by Monk and Baird, 2011Reference 15 for the 2010 Ecosystem Status and Trends Report. Gauging sites that form the Reference Hydrological Basin Network (RHBN) that provides Canada’s contribution to the World Meteorological Organization monitoring program for climate changeReference 64 were examined for trends across a range of hydrological characteristics. Variables were calculated for 172 hydrometric sites across Canada for each hydrological year (1970 to 2005). The variables calculated are “indicators of hydrological alteration” which quantify aspects of streamflow of ecological significance.Reference 65 There was high variability in the results across the country. A main conclusion drawn by the authors was that the shortage of long-term continuous hydrometric records in Canada (particularly in the northern ecozones+) severely limits our ability to monitor current trends and project future trends in hydrological regimes. Many sites have been discontinued and the majority of sites in the national database have fewer than 18 years of data.

Eleven gauging sites with suitable data for this analysis were available for the Taiga Plains Ecozone+. Statistically significant trends in indicators of hydrological alteration at these sites are summarized in Figure 13 . Trends in monthly flows varied, with significant increasing trends during winter months at several sites. Six sites showed statistically significant increases in various measures of minimum runoff (which occurs during winter); five sites had significant increases in baseflow (the seven-day minimum flow divided by the mean annual flow). There were few significant changes in peak flows or in timing of maximum and minimum flows.

The increases in baseflow indicate that a greater component of streamflow is being supplied through groundwater at several of the sites; this is likely to be related to the increased degradation of permafrost throughout the region. One of the most profound projected consequences of permafrost thaw is a transition from surface-water-dominated rivers to groundwater-dominated rivers.Reference 66 Changes in water quality accompany an increase in baseflow, including increases in major ions from the more mineral-rich groundwater – though the overall impacts on nutrient levels and many other water characteristics remain uncertain.Reference 66

Nine sites had significant increasing trends in the number of hydrologic reversals (changes in the direction of the trend in streamflow), suggesting increased variability in runoff (Figure 13). This trend was also seen in many of the sites in neighbouring ecozones+. Ecological impacts of increased variability include: stranding of species in isolated habitat patches (falling levels); entrapment on islands and floodplains (rising levels); and drought and desiccation stress to stream-edge organisms,Reference 65, Reference 67 with the net effect of changes in riparian communities.

Figure 13. Number of sites displaying significant increasing and decreasing trends in indicators of hydrological alteration for the Taiga Plains Ecozone+

Trends shown are significant at the p<0.1 level. Parameters listed on the vertical axis refer to river discharge except: date min/max (decreasing date means earlier and increasing date means later for minimum and maximum annual discharge), low/high pulse # (number of discharge pulses), fall/rise rate (rate of change in discharge), and reversals (number of reversals)

Source: Monk and Baird, 2011Reference 15
Long description for Figure 13

This bar graph shows the number of sites displaying significant (p<0.1) increasing or decreasing trends in indicators of hydrological change in the Taiga Plains Ecozone+. Bars to the left of the centre line indicate sites with a decrease in that indicator and bars to the right indicate the number of sites with a significant increase. Most indicators show a trend of increase at more sites, including higher river discharge from November to April, and increased river discharge minimums, later maximum discharge and a higher number of reversals. Most parameters showed significant decrease in two sites or less, though low pulse duration and the rise rate of river discharge decreased in multiple sites.

Top of Page

Trends through the seasons at two hydrometric sites

This section is based on Canada-wide analyses performed by Cannon et al., 2011Reference 14 for the 2010 Ecosystem Status and Trends Report. Trends in hydrology over the course of the seasons (at five-day intervals) and the relationship to trends in climate were examined for Canada’s ecozones+. Analyses were based on climate and hydrology data from Environment Canada’s monitoring networks from 1961-2003 (thus using different subsets of the same data as the above analysis of indicators of hydrological alteration).

Only two sites in the Taiga Plains Ecozone+ had data records that met the study requirements for the period 1961-2003: 1) Hay River, where it drains northeast into Great Slave Lake in the NWT; and, 2) Muskwa River near Fort Nelson in BC, a main tributary of the Fort Nelson River, which in turn drains to the Liard River. With only two points for analysis, the results cannot be interpreted to apply to the entire ecozone+, but instead highlight the hydrologic and climatic changes at these two locations.

Both Hay River and Muskwa River are streams driven by snowmelt processes, but they differ in their annual discharge patterns. Hay River has a distinctively sharp streamflow peak during freshet, whereas the Muskwa River exhibits a broad summer freshet with higher flows lasting for several months. The shifts in climate variables for these rivers were similar, but the hydrologic responses differed due differences in snowmelt processes between the two streams.

Hay River

At the Hay River site, temperature increases were seen over the entire year except during the fall. Summer temperatures remained relatively unchanged, whereas winter and spring temperatures showed increases of up to 3°C, comparing the periods 1961-1982 and 1983-2003. The highest monthly precipitation occurred in July, when a 30% increase occurred between the two periods. Precipitation decreased before and increased marginally after the summer peak, except during December. The hydrograph (Figure 14) shows that the spring freshet occurs between April and May. The magnitude of this peak flow did not change much (comparing averages over 1961-1982 and 1983-2003). The timing of the peak, however, shifted to slightly earlier. The higher summer and fall flows were likely the result of greater precipitation. Winter flows increased between the two time periods, attributed to warmer winters enhancing snowmelt. Monk and Baird, 2011Reference 15 considered permafrost degradation to be a probable cause of the increase in winter discharge observed at the majority of Taiga Plain sites.

Figure 14. Changes in streamflow over the seasons, Hay River, comparing 1961-1982 and 1983-2003.

Measurements are from hydrometric station 07OB001.

Insert text here
Source: Cannon et al., 2011Reference 14
Long description for Figure 14

A line graph comparing streamflow in the Hay River shows that streamflow in the fall and winter was higher in 1983-2003 than 1961-1982. The spring freshet occuring in April shows the same discharge rate in both time periods; increase discharge in the later period starts in mid-summer.

Top of Page

Muskwa River

Between the periods 1961-1982 and 1983-2003, the only streamflow shift at Muskwa River was an additional 20-25 m3/s of discharge occurring from November to the end of March (Figure 15). This was likely related to the much warmer average winter air temperature in the second time period. Precipitation decreased at this site in winter and increased in the remaining seasons, especially during the July streamflow peak. There was, however, no clear relationship between these observed precipitation changes and the changes in streamflow.

Figure 15. Changes in streamflow over the seasons, Muskwa River, comparing 1961-1982 and 1983-2003

Measurements are from hydrometric station 10CD001.

Source: Cannon et al., 2011Reference 14
Long description for Figure 15

A line graph comparing streamflow in the Muskwa River shows little difference in streamflow between 1983-2003 and 1961-1982. Winter streamflow was higher in the more recent time period (1983-2003), from December to March. Streamflow in the Muskwa River peaks in June and July, but is high from April through September.

Top of Page

Key finding 7
Ice across biomes

Theme Biomes

National key finding
Declining extent and thickness of sea ice, warming and thawing of permafrost, accelerating loss of glacier mass, and shortening of lake-ice seasons are detected across Canada’s biomes. Impacts, apparent now in some areas and likely to spread, include effects on species and food webs.

Ecozone+ key finding: Changes in permafrost, well documented for this ecozone+, include: increased temperatures of permafrost, changes in active layer depth, reduction of the continuous permafrost zone, and thawing of discontinuous permafrost in some areas. This has resulted in landscape changes, including loss of frozen peat plateaus. River ice within the Mackenzie Basin shows trends to earlier break-up; datasets are poor for both river and lake ice within the ecozone+.

Top of Page


The forested peatlands that characterize much of the Taiga Plains are underlain by varying degrees of permafrost (Figure 16) and are vulnerable to climate warming.Reference 68 Both loss and warming of permafrost are occurring. One result has been a shrinking of the zone of continuous permafrost. A study that repeated, in 1988 to 1989, a 1964 permafrost survey along the Mackenzie Highway from Hay River south to Alberta showed that the northern fringe of discontinuous permafrost had migrated northward by about 120 km in 26 years.Reference 69

Figure 16. Permafrost zones of the Taiga Plains.
Source: Smith, 201112 based on Heginbottom et al., 1995Reference 70
Long description for Figure 16

This map shows the distribution of permafrost zones in the Taiga Plains Ecozone+. The southern part of the ecozone+ in BC, Alberta and southern NWT is classified as sporadic permafrost. The middle of the ecozone, from north of Great Slave late to the top of Great Bear Lake is covered by extensive discontinuous permafrost. The northernmost portion is largely continuous permafrost, with the exception of the Mackenzie River Delta, which has extensive discontinuous permafrost.

Top of Page

A permafrost monitoring network in the Mackenzie Valley provides records of permafrost temperature in the upper 20 to 30 m. This monitoring network is supplemented with results from research on permafrost and on changes in frozen peatlands. Trends, from south to north are:

  • Frozen peatlands are decreasing in parts of the southern Mackenzie Valley, with an average loss of frozen peatland area of 22% at four study sites over the latter half of the 20th century.Reference 71
  • Permafrost temperatures monitored at other sites in the south-central part of the valley, however, show little change (Fort Simpson and Northern Alberta in Figure 17). At these sites, permafrost is likely being preserved by an insulating layer of peat.Reference 72, Reference 73
  • Permafrost temperatures are increasing in the central Mackenzie Valley (shown for Norman Wells and Wrigley in Figure 17), where permafrost is both thicker (up to 50 m deep) and colder.Reference 74, Reference 75
  • Similar rates of permafrost temperature increase, 0.1 to 0.2°C per decade at a depth of 15 m, have occurred since the 1960s in the colder permafrost (-2 to -3°C) at spruce-forested sites in the Mackenzie Delta in the north of the ecozone+.Reference 76, Reference 77
Figure 17. Ground temperatures in the central Mackenzie Valley, 1984-2007

Measurements are at depths near 10 m. Note that the frequency of measurements was reduced in the mid-1990s at the two most southern sites.

Source: adapted from Smith et al., 2010Reference 78
Long description for Figure 17

This line graph depicts the following information:

Ground temperatures in the central Mackenzie Valley, 1984-2007.
Fort Simpson (10m)

Fort Simpson (10m)

Temperature °C
Northern Alberta (10m)

Northern Alberta (10m)

Temperature °C
Wrigley (12m)

Wrigley (12m)

Temperature °C
Norman Wells (10m)

Norman Wells (10m)

Temperature °C

At sites further north (Wrigley and Norman Wells), ground temperatures have gradually increased over recent decades: an increase of 0.1°C per decade at Wrigley  and a 0.3°C increase per decade at the Norman Wells site.

Top of Page

While these changes are consistent with trends in air temperature over the past few decades, changes in snow coverReference 79, Reference 80 and in wildfiresReference 81 are also affecting rates and locations of warming and thawing of permafrost. Fires may trigger a long-term process of permafrost decay,Reference 82 though in sphagnum-moss-dominated peatlands fires are infrequentReference 73 and in other peatlands fires rarely burn deeply enough to have lasting effects on permafrost.83 The distribution of peatlands, fire, and extent of snow cover are interrelated, leading to locally diverse permafrost dynamics.Reference 84

Degradation of permafrost leads to major impacts on landscapes with, for example, loss of frozen peat plateaus. A study of peatlands in the southern Taiga Shield concluded that peat plateaus in what is currently the discontinuous permafrost zone are unlikely to be sustained under a warming climate and that the landscape-level effects of this loss need further investigation.Reference 85

Top of Page

River and lake ice

Across Canada, lake ice break-up times show consistent trends to earlier dates, while freeze-up dates show fewer trends.Reference 86 Little data, however are available for lake sites within the Taiga Plains Ecozone+. In an analysis that used records from both in situ measurements and remote sensing observations, large lakes in Canada showed an overall trend toward earlier break-up of ice and later freeze-up over the period 1970 to 2004.Reference 87 Three sites in the Taiga Plains Ecozone+ were included in this analysis. Freeze-up was significantly later at two of the sites. There were no significant trends in ice break-up.

Seventeen river sites within the Mackenzie River Basin showed evidence of ice break-up occurring earlier over the period from 1970 to 2002, at a rate of approximately one day earlier per decade in the upstream basin.Reference 88 This analysis was based on hydrological events that correspond with ice break-up.

Spring thaw was measured in situ at sites in the Mackenzie River at Fort Simpson until the late 1980s when the monitoring program was discontinued (Figure 18 A). The mean thaw date moved forward by about a month from 1932 to 1988, with the date of thaw accounting for about 60% of the variability. As the spring thaw date is strongly related to April temperatures, and as April temperatures have continued to increase (Figure 18 B), it is probable that this trend has continued since 1988.Reference 19 The same degree of change has not occurred in the fall, with fall temperatures at Fort Simpson showing no significant trends.Reference 19

Figure 18. (A) Estimated Mackenzie River melt date, 1932 to 1988; and (B) five-year running average of April air temperatures, 1900 to 2005, Fort Simpson
Source: Mackenzie River Basin Board, 2010,Reference 19 based on Environment Canada data
Long description for Figure 18

These scatter plots show the estimated Mackenzie River melt date between 1932 and 1988 and five- year running average of April temperatures in Fort Simpson between 1900 and 2005. The Mackenzie River melt date (graph A) shows a significant trend towards earlier melt dates by almost a month. The five-year running average in April temperatures shows a significant increase in the last century, increasing from -4°C to approximately -1°C (graph B).

Top of Page