Forest Fish Project

The Mykiss Lake catchment in Algonquin Park was used to evaluate the impact of different flow routing algorithms on the calculation of the Topographic Index and its subsequent ability to identify shoreline habitat used by YOY brook trout.

The image above provides an overview of the Mykiss Lake catchment area with DEM, lakes, streams, and contours in the background.  The total catchment area is 322 hectares and is characterized by an elevation range of 115m from the headwater at 540 masl to the lake outlet at 425 masl.  In total 11 brook trout habitats were observed in the field.  The yellow circles on the image represent the spatial distribution of the habitats.

An analysis of 85 combinations of slope and flow routing algorithms indicated the Multiple Flow Direction (MFD) algorithm best represents the landscape patterns of topography and likely stream and seep locations around lakeshores.  The Deterministic 8-Direction (D8) algorithm used in earlier analysis (Borwick 2002) performed reasonably well and has been included in the following images for comparison with the MFD.  The other flow routing algorithms examined as part of this study included the Randomized 8-Direction (Rho8), the Kinematic Routing Algorithm (KRA), the Braunschweiger Relief Model, the Digital Elevation Model Network (DEMON), and the Deterministic Infinity (DINF).

Topographic Index Maps of Mykiss Lake Catchment

D8	MFD

The images above show spatially distributed topographic index values using the Deterministic Eight-Direction (D8) flow routing algorithm and the Multiple Flow Direction (MFD) flow routing algorithm.  Darker colours of blue represent those areas with higher likelihood of surface and near-surface saturation.  The areas with darker colours of blue generally have lower localized slopes and larger upslope contributing areas.

Spatial Distributions of “Upper 10%” of TI Values Around the Shoreline of Mykiss Lake

D8	MFD

The top 10% of the topographic index shoreline cells around Mykiss Lake typically contained most or all of the stream or seep habitats used by young-of-year brook trout.  The spatial location of the top 10% is different between the two flow routing algorithms.  The D8 approach produces a more dispersed, single-cell pattern while the MFD approach tends to aggregate and concentrate the top 10% of shoreline cells.  The main advantage of the MFD aggregation and concentration pattern is the minimization of falsely predicted brook trout habitats.