Embedded Files
Figure 8: Welcome sign at a Ducks Unlimited Canada conservation site.
Study design
Study design
I conducted an observational study using natural variation to model the effects of pond consolidation on water boatman communities in Alberta's Prairie Pothole Region. Observational units were quarter sections (160-acre square of land) containing pond groups under varying levels of consolidation. I chose the quarter section as the spatial bounds for pond groups because it is a common land management unit in Alberta and is large enough to contain high pond habitat and size diversity while small enough to feasibly sample across. To measure pond consolidation, we created an area-based consolidation index (CI) that indicates how pond area is distributed among shapes:
CI = ∑Ai2/(∑Ai)2
, where Ai is the area of ith pond. Bound between zero and one, CI increases as pond count decreases and as pond area becomes unevenly distributed among ponds (Fig. 10). Using ArcGIS Online, I measured CI for 24 candidate Ducks Unlimited Canada (DUC) wetland conservation properties (Fig. 8). We visited these quarter sections in May to verify pond sizes and conditions. We discovered that all ponds we visited that were approximately south of the Central Parkland natural subregion were dry in May, restricting the spatial extent of the study to the northwest portion of the Prairie Pothole Region.
I selected 14 quarter sections (Fig. 9) such that pond consolidation varied while land use and total pond area (habitat amount) remained as equivalent as possible among quarter sections. All of the selected DUC properties were used for grazing cattle and growing hay. The riparian areas of the ponds were not mowed. I randomized the geographic locations of the quarter sections along the gradient of pond consolidation to minimize spatial correlation (Fig. 9). To control for water boatman dispersal between ponds during the summer (to achieve independence of observational units), the quarter sections were at least 15 km apart from each other. This distance was based on previous water boatman flight distance estimates (Popham, 1964).
Figure 9: Locations of the 14 study quarter sections (represented by triangles and circles and numbered from lowest to highest consolidation index score) in the context of the Prairie Pothole Region and Alberta.
Figure 10: Six study quarter sections arranged on a pond consolidation gradient according to consolidation index score. Summer pond area is coloured in blue.
Data collection
Data collection
We visited each quarter section in the spring (May) of 2024 to verify pond presence and sizes, then we returned to them in the summer (late July and early August) of 2024 to sample water boatmen before their fall wetland-river migration. During the summer visits, we took six water boatman samples from each quarter section for a total of 84 samples. We sampled as many ponds as were available in each quarter section, to a maximum of six. For each sample, we used a D-frame aquatic dip net and performed a two-minute zig-zagging travelling net sweep (Fig. 11), paralleling the shoreline and targeting representative habitat for each pond, similar to the Canadian Aquatic Biomonitoring Network protocol (Government of Canada, 2002). We aimed to maximize the habitat diversity captured with the six samples for each quarter section. For each pond we sampled, we also measured water salinity in three locations that were at least 10 metres apart and averaged these for each pond, then averaged the pond salinity averages for each quarter section. We repeated sampling at nine of the 14 quarters sections in the fall (late October and early November), after fall migration, to determine the migratory status of some water boatman species.
After each net sweep, we carefully washed extraneous plant material of all aquatic insects and removed it, then placed the remaining net contents into a 1-litre plastic bottle and preserved it propylene glycol. In the laboratory, we washed aquatic invertebrate material from the samples using a quality-controlled elutriation process (Alberta Biodiversity Monitoring Institute, 2015). We extracted and counted water boatmen from the washed samples by visual examination under dissecting microscopes and manual picking of individual specimens. Due to extremely high water boatman abundances, we subsampled some samples (at the species-level, when possible) and extrapolated the counts. We placed extracted water boatmen into 95% ethanol and subsequently identified the adults to species using a new taxonomic key I am developing (in progress). We measured the salinity of each pond we sampled using an EcoSense pH/EC1030A pen, then assigned each pond salinity value to its corresponding sample(s), then calculated a mean sample salinity for each quarter section.
Figure 11: The author performing an aquatic net sweep to collect water boatmen.
Statistical analysis
Statistical analysis
I used boxplots to explore the distribution of the consolidation index scores and mean sample salinites among quarter sections. I used a Principal Coordinates Analysis using water boatman species-counts converted to Bray-Curtis distances to examine similarites in species composition among quarter sections. I used a direct unconstrained gradient analysis to examine associations between water boatman species composition, consolidation index, and salinity by scaling the consolidation and salinity data (as they have different units of measurement) and converting these to Euclidean distances, then ordinating the quarter sections by these distances using non-metric multidimensional scaling and overlaying vectors for the species-counts. Finally, I used permutational Analysis of Variance to test the effects of consoldation index and salinity on the abundance of four key migratory water boatman species.
Page updated
Report abuse