Me

EcoInformatics International

EcoInformatics International

Home Services Contact

 

MONITORING ECOSYSTEMS WITH SATELLITES

THE CHURCHILL-NELSON RIVERS DIVERSION IN MANITOBA

Contents

In September of 1969 we were doing fieldwork in Northern Manitoba as part of the Canada Land Inventory Project and the Manitoba Remote Sensing Program. Our fieldwork was concentrated in the Wekusko (63J), Grand Rapids(63G), Norway House (63H) and Cross Lake (63I) topographic map sheet areas.


Dave Klippenstein, standing, and Jean Thie, with a large White Spruce growing on a glacio-fluvial delta in the study area

During the fieldwork using helicopters, fixed wing aircraft, and boats we collected a wide range of vegetation, soil, and photographic data which I am reviewing now, almost 50 years later to study environmental change especially the melting of permafrost in palsas and peat plateaus so common in these area. Much of this data and imagery allows us to visualize not only the impacts of climate change on permafrost but also to visualize the impact of the major Churchill- Nelson Rivers Diversion on the environment like the impact of flooding.

An good example of this source of aerial photos is the following image taken from a helicopter in the fall of 1969: The amount of tree kill is very high because wetlands with permafrost were flooded. The original floating peat wetlands adjusted to the new water level (green arrow right)l, but the mainly Black Spruce forest on peat plateaus were flooded and killed (red arrow). It is interesting to see that the former collapse scars of the peat plateaus also float and adjusted to the new water level (green arrow left).

Unfortunately, the Google Earth below does not provide high resolution images for this area. The two slide sites are marked on the GE image below, taken in 1982 and showing a messy situation after the part of tree kill was removed.

The Apple Maps program provides very high resolution images for part of this area. In spite of the clouds the image shows that the floating wetlands are still floating. Dead trees have been removed along the shoreline. The logging road is visible on both the Google Earth and Apple Maps image. It did not exist at the time of the helicopter photo. Both satellite images show the very murky colour of the lake water. Much of the original shoreline was composed of lacustrine clay and silt deposits from Glacial Lake Agassiz. Since the flooding the erosion of this silt and clay have changed the lake.


On this Apple maps image the clouded area has been replaced by a different satellite image which also shows the high waterlevel situation in the reservoir. The red arrow shows the line where the two images are joined

The slide below provides a good visualization of the extent of the flooding damage. The the tree kill along the shorelines is quite obvious. Also the tree kill in surrounding wetlands is extensive (light blue arrow)

The earlier slides clearly show the extent of damage. It is less clear in the slide below. The water table here is visibly higher than the natural fluctuation, resulting in more tree kill than would be normal under the natural degradation cycle of permafrost in the peat plateaus.The red arrow points to the tree kill.

The two red arrows below show the location where the G3 and G4 photos were taken in 1969


An example of Significant shoreline erosion: Playgreen Lake

It has actually been more than a few decades that I have looked at this area again, that is looking at more than the melting of permafrost. I am systematically going through my slide collection taken during my field work in 1969 and 1970. It is an interesting puzzle to locate the images on Google Earth. Sometimes it proves too much of a challenge. The helicopter photo below shows the shoreline of Playgreen Lake, on a flight line bringing us back to Norway House. We were flying a Bell 47 helicopter which has such a marvellous view to take pictures.

Strangely enough I could not find the accurate location where the photo was taken, not on Google Earth, Apple maps or Bing. Fortunately, I still had a 1954 1:70'000 scale aerial phots for this area. The red dotted lines on the 1953 aerial photo below show the area covered by the colour photo taken from helicopter. (National Air Photo Library photo: A13933-20)

The Google Earth image below is dated 1984!!!. The aerial photo date is 1953. The amount of shoreline erosion is quite dramatic.

The longest red line represents about 450 meters in erosion; the shorter line on the right is about 175 meters in shoreline erosion. This means-for this stretch of shoreline an annual erosion rate of 5-15 meters. How much of this can be contributed to the Churchill- Nelson Diversion, in particular the Eight Mile Channel which was constructed in the early to mid-seventies.

The Google Earth overview and 'transparent overlay of the 1953 aerial photo gives a good overview of the situation. The red arrow shows the start of the Eight-Mile Channel constructed by Manitoba Hydro in the mid-seventies. The water flow in Playgreen lake can easily be determined by the surface water tones reflecting suspended sediments: the erosion from this area appears to flow directly in the Eight-Mile Channel.

How much of the erosion did take place between the 1953 and the date of the helicopter colour photo- 1969? This could be seen as an approximation of natural erosion processes. The yellow dotted line on the image below shows the shoreline location based on an interpretation and comparison of the 1969 and 1953 photos. Using the Google Earth measuring tool, the amount of erosion varies from about 66-84 meters, for the 16 year period. Or an average annual basis this would be about 4-5 meters per year.

The figure below shows a combination of two satellite images and the 1953 aerial photo. 1984 is the Google Earth base and the Apple Maps image is superimposed and made semi-transparent. The shoreline erosion in the later time period has slowed down significantly. The maximum measured was 50-60 meters (around the red bar) and 20-30 meters on the right side of the wetland. This would amount to between 1 and 2+ meters per year.

The 1974 (July 4) satellite image (below) processed by the Canada Centre for Remote Sensing gives a good perspective of the Playgreen Lake situation during the construction of the 8-mile channel -top yellow arrow- and the 2-mile Channel -bottom yellow arrow. The white arrow point to the eroding area described above. The flow of suspended sediments is clearly visible in this colour combination of Landsat bands 5,6 and 7. Natural erosion processes are certainly significant in this area of predominantly lacustrine clays and silts

In one of my early papers (Thie et al., 1974) I analyzed a 1973 satellite image and interpretation with provide an interesting background for this study. This image below was taken on 20 September 1973 -(CCRS number 1424-17082). The delineation on this image is based on shoreline configuration, water distribution and water reflectance. The continuous lines identify major physiographic regions. Broken line delineate different water spectral reflectance signatures that appear to have relations with physiographic sub-units. While most of the reddish water colours are related to suspended sediment levels, few of these near 'e' are shallow waters.


REFERENCES:

  1. Thie, J., Tarnocai, C., Mills, G.E., Kristof,S.J., 1974: A Rapid Resource Inventory for Canada's North by means of Satellite and Airborne Remote Sensing, in proceedings of the Second Canadian Symposium on Remote Sensing, University of Guelph, Guelph, Ontario, April 29 - May 1, 1974

The Manitoba Wildlands organization provided the map below showing, among others, the Churchill-NeLson Diversion area.

Jean Thie EcoInformatics International Inc. - Ottawa, Ontario and Wakefield, Quebec, Canada