1
SLEMA Comments on the
2006 Wildlife Effects Monitoring Program
Annual Report
Anne Gunn
February 2008
2
Overall comments: The 2006 monitoring report is essential a cut and paste of the 2005
annual report with brief additions of 2006 numbers. While reviewing the 2005 report for
SLEMA in 2007, various suggestions to improve clarity of data presentation and
identifying omissions were offered to correct errors and improve reporting clarity. The
suggestions were neither refuted nor accepted thus the same problems remain – see
below.
De Beer’s intention is every 3 years to undertake detailed analyses and discussion of the
wildlife monitoring data. The annual report format is to include cumulative data from
previous years but this is inconsistent. For example there is no comparison of Wildlife
Incidences with previous years.
In 2007, I commented that in the 2005 annual monitoring report was that there is no
specific explanation of how the project-specific data will contribute to describing
cumulative effects. The 2006 report is similar in that there is no even summary of how the
annual monitoring relates to cumulative effects assessment. Another comment on the
2005 annual monitoring report was that in the 2006 report, there is no mention of how
the Snap Lake monitoring contributes to regional monitoring.
Another omission from both the 2005 and 2006 annual reports is that there is no mention
of environmental variability. Some indexes to the extent of environmental variation are
an essential component of cumulative effects (to assist discriminating between project
related and environmental effects). Conspicuous features such as timing snow melt,
freeze-up and an indication of exceptionally hot or wet weather need to be reported
as they can influence the wildlife behaviour and abundance. They are also essential to
measure Cumulative Effects. I wonder why even such aspects as satellite measures
such as NDVII are not under consideration.
Specific comments:
1. The same comment made for the 2005 report can be repeated for the 2006 report
with reference to the use of endpoints for detecting mine-related effects. De Beers
refers to the use of measurable endpoints or indicator variables (such as abundance,
distribution, probability of occurrence). There is no explanation of how the endpoints will
be used to separate project-related effects from natural environmentally caused ones.
The endpoint for abundance is the range of values measured 1999-2004
(preconstruction) and the data annually vary which raises the question of whether it is
reasonable to expect to separate project-related effects from natural environmentally
caused ones using this design. The number of sample units is partly the design problem –
7 transects is too few.
Additionally, this comment on the objectives was made in 2005 and is still applicable for
2006 as the three listed objectives for caribou (page 16) identical to 2005 and are
vague. The objectives need to be broken down into measurable components with
testable research hypotheses.
3
The 2006 survey design is unchanged despite a comment in 2007 on the 2005 annual
monitoring report “In particular the effects at the other two mines appear to include a
distributional change at 20-25 km – the specific hypothesis for Snap Lake needs to
include testing for this effect (which may require modification of the current design for
aerial surveys)”
For points 2-9, the same comment was also made for the 2005 report and it applies to
the 2006 report – so I have repeated the comments
2. Table 3.1
The dates of migration are dependent on the dates of the first and last aerial surveys –
the table should include the dates for the ‘triggers’ for when the surveys were
undertaken (satellite-collared caribou, camp sightings).
3. The report is incorrect in stating that precision is increased by reducing strip width.
Reducing transect width will reduce bias (visibility bias). Reducing strip width also lowers
the % coverage which is then a reduction in precision.
4. The survey altitude varies 60 m which will have a strong effect on bias – some effort is
needed to try and standardise the aircraft altitude.
5.Page 13 – “snow tracks “ were obtained during post-calving migration which seems
unusual – but is not mentioned in the results. Some idea of the time period over which
the tracks will have accumulated (time since last snowfall or strong wind) is needed to
weight the ranking of the tracks as low, medium etc. Does the low number of tracks in
2005 reflect snow conditions as there were fewer tracks but more caribou compared to
2004 (also late May).
6. Page 13
The emphasis on reporting the aerial surveys is on ‘groups’ whereas mean numbers of
individuals would be more useful
7. Page 14 – for the sightings of the larger groups, an idea of weather conditions would
be useful – whether the conditions were conducive to insect harassment.
8. Table 3-2. The number of surveys each year is variable enough that the caribou
numbers in the table are not comparable between years. The data needs to be
presented and then graphed as a mean +/- SE.
9. Page 19
The sample unit for the behavioural data is unclear – in the Methods N is number of
scans whereas the Results suggests N = number of caribou groups. The time of day for
the scans should be included and some idea of weather.
Additional comments for 2006
4
In the 2005 report, the Ahiak herd of caribou was mentioned as potentially being in the
regional study area. Why was this omitted from the 2006 report?
Section 2.8
There is no comment or explanation why the number of fox incidents was so high. In
2005, a single fox incident was reported compared to 41 in 2005. Either there was a
breakdown of food disposal or an ecological suggestion for why there were so many
fox-related incidents (more foxes?). There are no details such as were the incidences
throughout the year or concentrated at a specific time.
Page 20
The snow track survey results do not follow the trend on the aerial sightings – some
comment or suggested explanations would be useful.
Page 32
Were there any reasons – weather or a dry year as a comment on the reduction in bear
sightings in riparian habitats especially given the high number of plots sampled?
Page 42
The tabular comparison of the 2006 monitoring results with the Environmental
Assessment Impact predictions do not take into account whether the monitoring design
especially for caribou abundance is adequate to detect the predicted impacts.