Surveying Dog and Wildlife Fecal Contamination

In June and August of 2011 feces of dogs, coyotes and other animals (mostly Canada goose Branta canadensis and white-tailed deer Odocoileus virginianus) were collected from 16 parks and one bird sanctuary in the city (Fig. 1). The 17 areas were selected to include the four sectors of the city (NW, NE, SW, SE), the main habitat types (grasslands, forests, and shrublands), and the different dog management bylaws. The sampled parks also included two of the largest parks in the city, namely Nosehill Park and Fishcreek Provincial Park, and most of the large-sized parks (> 100 ha) of the city, where a lot of dog-related activities took place. The selected parks ranged in sizes from 1 to 1348 ha. Despite the overall very good socio-economic conditions in Calgary, the four sectors of the city are somehow descriptive of different socio-economic conditions, with the NE sector being the one at lower income. The dog bylaws consist of the following: dogs not permitted (“no dog”); only dogs on leash allowed (“leash on”); dogs allowed without leash (“leash off”); and parks with both “leash on” and “leash off” areas (“mixed”).Figure 1

Map showing the City of Calgary, and the locations of sampling points within the dog fecal contamination surveys carried out in 2011 in the city of Calgary (AB, Canada). The map was created with ArcGIS10 (ESRI, CA, USA), using geographic data created by the City of Calgary. The blank map showing territory governed by provinces and territories of Canada was obtained from Wikipedia under CC BY-SA 3.0.

Within the 17 areas, we used a random sampling design to identify sites for monitoring fecal contamination: we assigned random points in the sampled parks roughly proportional to their size (15.94 points/km², range 13.22–18.67). Exceptions were made for small parks (Taradale, Martindale, Meadowlark, and West Hillhurst), where two sampling points were assigned regardless of its size. Exceptions were also made for very large parks (Fishcreek Provincial Park and Nosehill Park) due to logistics, where two sub-areas of the parks were designated wherein the sampling points were assigned. Random points were generated using a random point algorithm implemented in ArcGIS10 (ESRI, CA, USA).
Each random point was marked with a 25 cm metal bar in the ground and a sampling plot was defined as the circular area within a 10 m radius surrounding the point. Plots were then visited twice in June and twice in August 2011; the first visit to clear the areas of feces, and the second visit one week later to collect the feces. This was done to assess the weekly rate of fecal deposition. The feces were identified to species by size, shape, content, and texture³³ . In Calgary there were only two canid species (dog and coyote), and in previous studies our group developed significant experience in identifying coyote feces and tested our efficiency with molecular tools^{34 (link)–36 (link)} and because coyote feces were normally visually distinct from those of dogs due to their content of animal hair, bones, and plant matters, dog feces identification was relatively simple. Each fecal sample was bagged, tagged with a unique identifier, weight, and the species. The feces were then brought to the lab and placed at − 80 °C for 48 h to inactivate any Echinococcus multilocularis eggs^{37 (link)}. The feces were then weighed and kept at − 20 °C.