Northern Prairie Wildlife Research Center
2. Have changes in western Canadian boreal forests resulted in reduced reproductive success of scaup?
Sources of Possible Landscape-Level Impacts
The boreal forest herein refers to Alaska, Yukon, Northwest Territories, northeastern British Columbia, northern Alberta, northern Saskatchewan, and northern Manitoba, including the Saskatchewan River Delta. We considered 9 possible landscape-level impacts to breeding scaup:
Fire.--The northern boreal forest has evolved as a fire-dependent ecosystem, not only in North America (Kronenberg et al. 1998, Larsen 1997, Lee et al., 1997, McCullough et al. 1998), but throughout the circumpolar boreal eocozone (e.g., Conrad and Ivanova 1997, Stocks et al. 1998). Fire is perceived to play a significant role in determining the structural complexity of boreal mixed-wood ecosystems (Kronenberg et al. 1998, Lee et al. 1997). Estimates of fire frequency in the ecosystem vary, but fires typically are viewed as having occurred on a relatively frequent (i.e., ≤100 yr) interval (Kronenberg et al. 1998, Larsen and McDonald 1998, Lee et al. 1997). Due to the extent of fires across this landscape and size of individual fires (often thousands of hectares), effects of fire on waterfowl habitat is potentially widespread.
Alterations in fire frequency have occurred in this century as fire suppression activities increased to protect merchantable timber resources (Conrad and Ivanova 1997, Thompson et al. 1998). Furthermore, enhanced fire suppression has resulted in a shift in fire extent and severity in recent decades (Thompson et al. 1998). Climatic changes may increase fire frequency, either by increased fire probabilities or changes in vegetation (Kronenberg et al. 1998). Changes in climate also have been projected to alter fire frequency and intensity (Stocks et al. 1998, Thompson et al. 1998). Whether fires affect reproductive success of scaup is presently unknown. A review of fire history would determine if fire frequency and extent have changed during periods and in areas of scaup population declines. Assessment of the interaction between fire and scaup reproductive success would be warranted if a correlation was detected between areas of declining scaup numbers and fire frequency or extent.
Logging.--In the Northwest Territories, logging activity is minimal and likely has little impact on scaup populations. However, logging recently may have had an effect in northeast British Columbia, northern Alberta, northern Saskatchewan, and northern Manitoba, especially in the past 10 years. It is during this period that technological development has allowed and promoted harvest of aspen (Populus tremuloides) and balsam poplar (P. balsamifera) in these areas. Additionally, government attitudes toward development of resources in the boreal mixed-wood region have resulted in the recent expansion of logging activities (Province of Alberta 1998). Exploratory analyses of remotely-sensed data in the boreal mixed-wood region of east-central Alberta (Lac LaBiche) confirm the recent expansion of logging activity (Dr. I. Creed, University of Western Ontario, pers. comm.). Broad-scale, long-term declines in scaup populations probably are not a result of logging activities in the boreal mixed-wood ecosystem given their recent (i.e., since 1990) development. Whether logging affects nesting habitat or wetland quality is unknown. Certainly, the agricultural transition zone has recently expanded northward in concert with logging activities along the southern edge of the boreal forest, particularly in central Alberta. This expansion may be contributing to declines in scaup populations there. The extent of conversion of boreal mixed forest to agriculture could be examined using LANDSAT-TM images, and assessed relative to breeding waterfowl survey data. However, this area is insufficiently large to have caused range-wide declines observed for scaup.
Climate.--The Boreal Forest ecosystem is predicted to experience the greatest warming due to global climate change (Environment Canada 1995, Rouse et al. 1997). Most anticipated effects on ecosystem processes are linked to changing biogeochemical processes of plant growth and soil, changes in vegetation distribution patterns, and changes in net primary productivity (Plöchl and Cramer 1995). However, few studies have examined potential response of non-peat wetlands to climate change in this ecosystem (e.g., Marsh and Lesack 1996). Given predictions of significant changes in hydrology, temperature, and precipitation patterns, potential effects on boreal wetland ecosystems likely will be broad-scale and significant. The magnitude and direction of changes and long-term effects on scaup populations are not known. We recommend retrospective analyses to assess potential effects of climate change on wetland extent and quality in the boreal ecosystem.
Oil and gas developments.--Exploration and development for oil and gas has been wide-spread across the region, but effects are mostly localized. Exploration effects are largely transitory, but oil and gas extraction facilities come with permanent infrastructure, including roads and well pads.
Acid rain.--Contrary to the situation documented for wetland-dependant wildlife in eastern boreal ecosystems (Hansen 1987, Longcore et al. 1993), acidification in western Canadian boreal forests is anticipated to be localized and minimal (ADPR 1989). Acid neutralizing capacity (ANC; the ability of soils and bedrock to ameliorate acidic deposition) in much of the western boreal ecosystem generally is moderate to high (Anonymous 1988), largely due to extensive glacial overburden found there. Furthermore, wind patterns resulting in the long-range transportation of SO2 and NOx emissions from high volume sources common to northeastern North America (Haines 1981) are not common to the western boreal ecosystem. Thus, point source deposition is of greater importance in this region. Wetlands located in areas immediately downwind of industrial sources (e.g., metal smelting facilities in Thompson and Flin Flon, MB) that lack sufficient ANC may be affected, but these effects are anticipated to be localized.
Hydro-electric development.--At present, we are unaware of large-scale projects planned for the western Canadian boreal forest region. Significant local effects on wetlands and wetland-dependant wildlife are possible, but past developments are unlikely to have caused range-wide declines in the scaup population.
Mining.--Mining occurs largely outside of traditional waterfowl survey areas, and effects usually are localized. Past activities are not presumed to be sufficient to have resulted in the significant declines observed in the scaup population. Several new mining interests have been developed in areas of the Northwest Territories, but minimal impacts to scaup habitat are anticipated.
Predators.--There is no evidence of changes in predator communities or predator efficiency that suggest scaup are suffering increased losses to predation. However, rigorous data to test this are not available.
Confounding aspects of prairie drought cycles.--We hypothesize that prairie drought may contribute to lower reproductive success of scaup (see discussion above under Breeding biology: climate change; Afton and Anderson in review). During prairie drought, wetland availability and quality are markedly reduced. This could cause shifts in scaup distribution during spring migration as they seek suitable wetland habitat, and reduced availability of food; these factors might affect acquisition of nutrient reserves. Reproductive success of boreal scaup could be reduced if they were not able to acquire sufficient nutrient reserves during spring migration. The significance of prairie-parkland food resources to scaup breeding farther north is unknown.
Reproductive effort in prairie-parkland and boreal forest habitats.--Comparison of data collected in the YKSA (Trauger 1971), Alaska (B. Grand, D. Esler, P. Flint, and T. Fondell, unpubl. data), and Erickson, Manitoba (Afton 1984) suggests basic reproductive parameters of lesser scaup were similar among boreal and parkland areas. Data are lacking for scaup breeding in tundra and for other boreal forest areas. Additional studies in a range of areas, conducted simultaneously to examine within-year differences, would enable biologists to determine whether reproductive effort and success differ between scaup breeding in prairie-parkland and boreal forest, especially in areas where the population is declining. Studies should compare several reproductive parameters including clutch size, nesting effort, nest and hen success, role of nutrient reserves, and duckling survival.
There are several constraints to addressing these issues. First, we need to clarify where, specifically, the largest declines have occurred. We recommend a continuation of the work initiated by MacCluskie et al. and D. Caswell. This information is particularly critical for development of intensive studies of breeding biology and to directly assess effects of habitat changes noted above. We lack comparative information on temporal and spatial patterns of spring and fall migration of Boreal Forest and Prairie-Parkland populations. Possibly, differences in these factors lead to differential hunting pressures, exposure to contaminants, and differential scaup reproductive success and recruitment. Banding, mark/resighting, and telemetry studies could address these issues and better describe spatial and temporal patterns of migration.
The following recommendations are not prioritized.