Using desktop hydrologic data to predict fish presence in streams in northern British Columbia

Byrd, B. (2014) Using desktop hydrologic data to predict fish presence in streams in northern British Columbia. [USQ Project]

[img]
Preview
Text
Byrd_2014.pdf

Download (2725Kb) | Preview

Abstract

Identification of fish-bearing streams is a key part of many environmental assessments in

Canada in general, and specifically in British Columbia (BC), where fish and fish habitat are
highly valued components of the natural environment. Pre-field identification of likely
fish-bearing and non-fish-bearing streams has the potential to reduce cost and effort related to
field inventories, and to expedite the project design process.

Previous research has considered desktop level hydrologic, geologic and land-use data from single
catchments with good results, but in some cases did not maintain simi- lar predictive success for
distant catchments. This research drew from three distinct catchments, with the aim of developing a
model that will be more generally applicable. Data on fish presence/absence, watershed area, and
mean and maximum monthly flows was collected from 2055 stream crossing points as part of the
environmental assessment for the Prince Rupert Gas Transmission (PRGT) project. Canadian Digital
Elevation Data was used to identify the elevation and derive the slope for each site. Parameters
derived from this data were assessed using logistic regression to develop a model for predicting
fish-bearing status.

The final model included the following parameters: watershed area, field gradient (as a proxy for
higher-quality desktop slope values), number of months per year with maxi- mum flow ≥ the 80th
percentile of maximum monthly flows, and latitude. The model achieved good predictive success for
non-fish-bearing streams (79% to 91% correctly identified) but performed less well for
fish-bearing streams (65% to 66% correctly iden- tified). The contrast between levels of
predictive success was thought to be strongly influenced by the quality of the underlying data,
where, for regulatory reasons, the actual status of streams classified as non-fish-bearing was
likely far more certain than
the status of streams classified as fish-bearing.


Statistics for USQ ePrint 27284
Statistics for this ePrint Item
Item Type: USQ Project
Item Status: Live Archive
Additional Information: Bachelor of Engineering - Honours (Environmental Engineering) project.
Faculty / Department / School: Current - Faculty of Health, Engineering and Sciences - School of Civil Engineering and Surveying
Supervisors: Brodie, Ian; Ottenbreit, Kirby
Date Deposited: 21 Aug 2015 04:52
Last Modified: 21 Aug 2015 04:55
Uncontrolled Keywords: fish, stream crossing, presence/absence, hydrology, terrain mapping, logistic regression.
Fields of Research : 09 Engineering > 0907 Environmental Engineering > 090799 Environmental Engineering not elsewhere classified
URI: http://eprints.usq.edu.au/id/eprint/27284

Actions (login required)

View Item Archive Repository Staff Only