Members of our research team have been traveling to the Feshie since 2000 in order to complete annual topographic surveys of the Glen Feshie study site.
Including our most recent survey in the spring of 2013, we now have
over a decade of surveys, allowing us to analyze the evolution of the
Feshie and determine how the mechanisms by which the stream braids might
vary from year-to-year, depending on the magnitude and number of
floods. As an example, we'll present the results of a 2013 study that
examined the evolution of the Feshie using repeat surveys from
2003-2007. The hydrologic regime of the Feshie is one
marked by consistent snowmelt floods in the spring and less consistent
high flows in the warmer months due to summer storms. The level of the
river often rises and falls quickly, reflective of the steep topography
in the catchment which can deliver water - and move it through the study
reach - rapidly following snowmelt or a summer rainstorm. Below is the
hydrograph from the five-year period that we studied, from 2003-2007.
Click on any of the figures below for a larger view. Each of the figures
below is from: Wheaton
JM, Brasington J, Darby SE, Kasprak A, Sear DA, Vericat D. 2013.
Morphodynamic signatures of braiding mechansisms as expressed through
change in sediment storage in a gravel-bed river. Journal of Geophysical Research: Earth Surface 118, 759-779. DOI: 10.1002/jgrf.20060  Instantaneous discharge hydrograph for Feshie Bridge during 5 year study period. The dashed vertical lines represent the start and stop dates for the five individual topographic surveys. The shaded periods represent the relatively dry flow years. The numbers in circles represent the top four ranked flows for each epoch. Dashed line represents average peak instantaneous discharge (135.8 m3s−1) over 19 year flow record. Shaded interval shows estimated bankfull discharge (28–42 m3s−1) at Glen Feshie study site. Daily flow data are from Scottish Environmental Protection Agency (SEPA). Figure from Wheaton et al., 2013; copyright 2013, Wiley and Sons. Note
that we have divided the period of record into four periods, some of
which (e.g. 2007-2006 and 2005-2004) are distinctly 'wetter' than others
(e.g. 2006-2005 and 2004-2003), in that they contain a larger number of
high flows. We might hypothesize that these wetter periods may be
marked by a greater degree of geomorphic change on the Feshie, or at
least may be denoted in that they cause different types of geomorphic
change. To
evaluate the mechanisms of change on a yearly basis, we can use the
repeat topographic surveys that we collected in the field. Surveys have
typically been completed using a combination of terrestrial laser
scanning (TLS) and global positioning system (GPS), along with the use
of a control network surveyed using a total station. You can find a poster that specifically discusses the collection of TLS data here.  From
these topographic surveys, we can create digital representations of the
landscape, or digital elevation model (DEM). The high resolution of
these DEMs makes it possible to identify individual features of the
channel - in the case of the Feshie, we are able to delineate various
types of bars. Moreover, by comparing two successive years' DEMs, we can
difference the two, allowing us to locate areas which have changed
(either via erosion or deposition).  Illustration of primary geomorphic units and bars classified in the reach and how they can change from year to year in response to a combination of braiding mechanisms and other mechanism of change. The bar classifications were used to help infer mechanisms of change and categorize erosion and deposition patterns. Figure from Wheaton et al., 2013; copyright 2013, Wiley and Sons. The
"DEMs-of-Difference", or DoDs, (below) show the amount of elevation
change between two successive surveys on the Feshie. Areas in red have
undergone erosion, and areas in blue have seen deposition of sediment
since the previous survey. All the DoDs have been 'thresholded' at 95%
confidence; that is, the only colored areas are those where we are 95%
confident the changes in elevation are real, and not due to inherent
errors in our surveying equipment.  Thresholded DEM of Difference (DoD) maps (top) and their elevation change distributions (bottom) for each epoch. The DoDs shown reflect the changes after an uncertainty analysis and discarding changes with less than a 95% probability of being real (thresholded). Hillshades derived from the more recent DEM in each epoch are shown behind the thresholded DoDs for context (i.e., where hillshade is visible, the DoD changes have been determined not to be distinguishable from noise). The volumetric elevation change distributions show the unthresholded DoD in grey, and the thresholded distribution in red and blue for the erosional and depositional fractions respectively. Figure from Wheaton et al., 2013; copyright 2013, Wiley and Sons. While
understanding the magnitude and direction of the sediment imbalance
(that is, how much more sediment was gained than was lost, or vice
versa) for any given period is important, we are also able to exploit
the high resolution of the DoDs in order to infer the actual mechanism
of change. Four main 'braiding mechanisms' have been discussed in the
literature (by Ashmore, 1992), and we have expanded on these for a total
of eleven mechanisms.  We
then segregated the changes in each DoD period depending on the
mechanism causing them. Below are segregated DoDs using the 2007-2006
(relatively wet, more floods) and 2004-2003 (relatively dry, fewer
floods) periods. The
mechanisms of braiding have been color coded, and the distributions of
elevation change resulting from the four mechanisms noted by Ashmore
(1992) have been separated in the histogram plots. Note that in the
dryer 2004-2003 period, the overall area of the braidplain which has
undergone change is relatively small, and that a relatively large
proportion of the changes are due to chute cutoff, while the
remaining three mechanisms play a lesser role in forcing geomorphic
change. In contrast, the wetter 2007-2006 period saw a much greater
areal extent of the braidplan altered. In addition, chute cutoff played a reduced role in terms of forcing geomorphic change, while transverse bar conversion was responsible for a much greater amount of deposition than in the 2004-2003 epoch.  Segregated elevation change distributions for a relatively dry period (2004-2003 surveys; top) and a relatively dry period (2007-2006 period; bottom). Braiding mechansisms are inferred and denoted on hillshaded DEMs on right. Figure modified from Wheaton et al., 2013; copyright 2013, Wiley and Sons. These
results suggest that not only do the volumes of geomorphic change have
some dependency on the magnitude and frequency of floods in a given
year, but also that the individual mechanisms by which the Feshie
evolves depend on the flood regime. The entire study, along with a more
thorough discussion of the influence of hydrograph on braiding
mechansisms at annual timescales, is available at: Wheaton JM, Brasington J, Darby SE, Kasprak A, Sear DA, Vericat D. 2013. Morphodynamic signatures of braiding mechansisms as expressed through change in sediment storage in a gravel-bed river. Journal of Geophysical Research: Earth Surface 118, 759-779. DOI: 10.1002/jgrf.20060 |
Home > Our Research >
