Population Variability and Extinction Risk

Abstract: Population models generally predict increased extinction risk (ER) with increased population variability (  PV  ), yet some empirical tests have provided contradictory findings. We resolve this conflict by attributing negative measured relationships to a statistical artifact that arises because PV tends to be underestimated for populations with short persistence. Such populations do not go extinct quickly as a consequence of low intrinsic variability; instead, the measured variability is low because they go extinct so quickly. Consequently, any underlying positive relationship between PV and ER tends to be obscured. We conducted a series of analyses to evaluate this claim. Simulations showed that negative measured relationships are to be expected, despite an underlying positive relationship. Simulations also identified properties of data, minimizing this bias and thereby permitting meaningful analysis. Experimental data on laboratory populations of a bruchid beetle (Callosobruchus maculatus) supported the simulation results. Likewise, with an appropriate statistical approach (Cox regression on untransformed data), reanalysis of a controversial data set on British island bird populations revealed a significant positive association between PV and ER (p = 0.03). Finally, a similar analysis of time series for naturally regulated animal populations revealed a positive association between PV and quasiextinction risk (p < 0.01). Without exception, our simulation results, experimental findings, reanalysis of published data, and analysis of quasiextinction risk all contradict previous reports of negative or equivocal relationships. Valid analysis of meaningful data provides strong evidence that increased population variability leads to increased extinction risk.     

Air concentrations of currently used herbicides and legacy compounds in the Canadian prairies,subarctic, and arctic

Passive air samplers were installed in the summers of 2005 and 2007 for 90 days at four locations in the agricultural region of the Canadian Prairies and at five locations in the Canadian Subarctic and Arctic. The presence and masses of ten currently used herbicides and three legacy compounds in the polyurethane foam disks were quantified. Herbicides 2,4-D, bromoxynil and MCPA were detected at all locations in the Canadian Prairies and in both years because these herbicides are widely applied to control broadleaf weeds in cereal crops that are an integral part of Prairie agricultural production systems. MCPA was also detected at one location in the Arctic in 2007. The detection of the other seven herbicides in the 2 years combined ranged from no detections (atrazine only) to five detections for the relatively volatile herbicides trifluralin and triallate. Triallate was the only other herbicide detected in the Arctic (2005). Legacy compounds were either not detected (alachlor) or at levels near their detection level (γ-HCH and α-HCH). γ-HCH and α-HCH were more frequently detected in 2005 than in 2007 indicating that their concentrations in Canadian air have decreased over time. γ-HCH, widely used as an insecticide in Prairie oilseed production until 2002, was detected at larger concentrations in the Canadian Prairies than in the Subarctic and Arctic. α-HCH, a manufacturing by-product in technical HCH prior to 1971 in Canada, was not detected in the Canadian Prairies but was at detectable levels in the Subarctic and Arctic as the Arctic Ocean is reported to be a major source of α-HCH to the atmosphere. We conclude that some of the most widely used herbicides in Canadian agriculture today are commonly present in the air in regions where they are applied and that a portion of these herbicides may be traveling as parent molecules to the Canadian Arctic. To the authors’ knowledge, this is the first evidence of the presence of MCPA and triallate in Arctic air samples, perhaps because previous research has seldomly monitored for currently used herbicides in this region.     

Toxicity of a complex mixture of atmospherically transported pesticides to Ceriodaphnia dubia

The presence of several anthropogenic chemicals has been documented in the atmosphere of the Canadian prairies. The deposition of these chemicals as a mixture is of importancesince little is known of the combined effects of these chemicalson aquatic organisms. This study was designed to evaluate theacute and chronic toxicity of a complex mixture of nineatmospherically transported pesticides to Ceriodaphniadubia. The nine selected pesticides (bromoxynil, dicamba, 2,4-D,MCPA, triallate, trifluralin, pentachlorophenol, lindane, and4,4-DDT) were detected in appreciable quantities in dryatmospheric deposits. The concentration of each pesticide in themixture was based on maximum measured daily dry deposition ratesfor central Canada, except for pentachlorophenol, which wasestimated based on atmospheric concentrations. The 48-h LC50estimate for C. dubia exposed to the pesticide mixture was174.60 g L^-1 (340 times the measured total dry deposition concentration). The estimated NOEC and LOEC for bothsurvival and reproduction, as determined in the 7-d chronic toxicity test, were 51.3 (100 times) and 154 g L^-1 (300 times), respectively. A basic risk assessment, using the toxic unit approach, suggested that the toxicity of the pesticide mixture was mainly due to 4,4-DDT. Overall, this atmospherically transported complex mixture of pesticides appearsto pose a negligible toxicological risk to non-target aquatic invertebrates such as zooplankton.     

A comparison of flux chambers and ambient air sampling to measure gamma-hexachlorocyclohexane volatilisation from canola (Brassica napus) fields

The insecticide gamma-hexachlorocyclohexane (gamma-HCH) is primarily used in Canada in treatments of canola (Brassica napus) seed. It has been shown that gamma-HCH so applied will volatilise with 12-30% entering the atmosphere within 6 wk after the seed is planted. Both flux chambers and high-volume air samplers were used to measure gamma-HCH volatilisation from a canola field and the results from each method compared. Daily samples were collected from three flux chambers located on the field. gamma-HCH was found in the air of the chambers on the first day after planting. Volatilisation rates were low for the first 7d (40.0 mg ha(-1) wk(-1)) but increased during the second week (143.8 mg ha(-1) wk(-1)). This was consistent with previous studies. Weekly composite air samples, from three heights above the canola field, were used to calculate volatilisation rates from the field. These were 190 mg ha(-1) wk(-1) (week 1) and 420 mg ha(-1) wk(-1) (week 2). Soil temperatures in the open field were warmer than those under the flux chambers and this may have contributed to the higher ambient air measurements.     

What currency do scatter-hoarding gray jays maximize?

Gray jays (Perisoreus canadensis) cache thousands of food items during each summer for use during the subsequent winter. Previous work on the economics of gray jay scatter-hoarding behavior was based on the assumption that the jays maximize the rate at which they store food energy; alternative currencies were not considered. Here we develop and test models based on two currencies, net rate (net recoverable energy stored per unit time) and efficiency (recoverable energy stored per unit energy expended). Our experiment involved providing gray jays with two options. After collecting a single food item upon arrival at a feeding apparatus, a jay could wait for two additional food items to become (simultaneously) available and then transport all three items for storage in scattered arboreal sites. Alternatively, the jay could immediately transport the single item to a storage site and return to the source repeatedly for additional single-item loads. By incrementally increasing the amount of time jays were required to wait for multipleitem loads, we were able to determine how long jays would wait before switching from multiple- to single-item caching trips. In contrast with the finding in a variety of species that efficiency-maximization models provide a better account of foraging behavior, the net rate-maximization model was a better predictor of the jays' switching point than was the efficiency-maximization model (Figs. 2 and 3). We discuss these conflicting results in the context of recent theory that describes the conditions favoring rate- versus efficiency-maximizing behavior (Ydenberg et al. 1994). Offprint requests to: T.A. Waite     

Spring bloom sedimentation in a subarctic ecosystem

Results from a 5-yr study (1985 to 1989) in Auke Bay, Alaska show that termination of the spring bloom consistently occurred at limiting nitrate concentrations. Following nutrient exhaustion, phytoplankton sinking rates increased and displayed greater temporal variability. Threshold nitrate concentrations, approximating K_s values of the species present, were found to signal initiation of increased sedimentation. For Thalassiosira aestivalis, the threshold was ～2 μmol l^-1, while for Skeletonema costatum the threshold was ～1 μmol l^-1, suggesting genus-specific differences in sinking-rate sensitivity to nitrate exhaustion. Overall, sinking rates of the three principal genera ranked (high to low) Thalassiosira spp.> S. costatum>Chaetoceros spp., while the nitrate sensitivities of the sinking rates of the genera ranked (high to low) Thalassiosira spp.> Chaetoceros spp.> S. costatum. Thalassiosira spp. showed the most consistent sinking rate increases following nutrient impoverishment. During a bloom dominated by T. aestivalis, a decrease of cell sinking rate with depth coincided with a decrease in short-term nutrient stress as measured by intracellular nitrate pools. In addition, no correlation was found between chain length or aggregate formation and sinking rate for this species. Though we measured only small-scale cell-cell adhesion, not larger-scale marine snow formation, this supports the notion that the sinking rates of Thalassiosira spp. were controlled primarily by cell physiology. For S. costatum, however, shorter chains sank faster. The sinking behavior of the species studied here figures prominently in their pelagic ecology and in the carbon flux of coastal ecosystems, both of which are driven by short-term variability.     

Comparison of Streambed Sediment and Aquatic Biota as Media for Characterizing Trace Elements and Organochlorine Compounds in the Willamette Basin, Oregon

During 1992–93, 27 organochlorine compounds (pesticides plus total PCB) and 17 trace elements were analyzed in bed sediment and aquatic biota from 20 stream sites in the Willamette Basin as part of the U.S. Geological Survey's National Water-Quality Assessment Program. Data from each medium were compared to evaluate their relative effectiveness for assessing occurrence (broadly defined as documentation of important concentrations) of these constituents. Except for Cd, Hg, Se, and Ag, trace element concentrations generally were higher in bed sediment than in biota. Conversely, although frequencies of detection for organochlorine compounds in biota were only slightly greater than in bed sediment, actual concentrations in biota (normalized to lipid) were as much as 19 times those in sediment (normalized to organic carbon). Sculpin (Cottus spp.) and Asiatic clams (Corbicula fluminea), found at 14 and 7 sites, respectively, were the most widespread taxa collected during the study. Concentrations of trace elements, particularly As and Cu, were typically greater in Asiatic clams than in sculpin. In contrast, almost half of the organochlorine compounds analyzed were found in sculpin, but only DDT and its degradation products were detected in Asiatic clams; this may be related to the lipid content of sculpin, which was about three times higher than for clams. Thus, the medium of choice for assessing occurrence depends largely on the constituent(s) of interest.     

Uptake of atmospheric mercury by deionized water and aqueous solutions of inorganic salts at acidic,neutral and alkaline pH

Mercury (Hg) is well known as a toxic environmental pollutant that is among the most highly bioconcentrated trace metals in the human food chain. The atmosphere is one of the most important media for the environmental cycling of mercury, since it not only receives mercury emitted from natural sources such as volcanoes and soil and water surfaces but also from anthropogenic sources such as fossil fuel combustion, mining and metal smelting. Although atmospheric mercury exists in different physical and chemical forms, as much as 90% can occur as elemental vapour Hg0, depending on the geographic location and time of year. Atmospheric mercury can be deposited to aquatic ecosystems through both wet (rain or snow) and dry (vapour adsorption and particulate deposition) processes. The purpose of the present study was to measure, under laboratory conditions, the rate of deposition of gaseous, elemental mercury (Hg0) to deionized water and to solutions of inorganic salt species of varying ionic strengths with a pH range of 2-12. In deionized water the highest deposition rates occurred at both low (pH 2) and high (pH 12). The addition of different species of salt of various concentrations for the most part had only slight effects on the absorption and retention of atmospheric Hg0. The low pH solutions of various salt concentrations and the high pH solutions of high salt concentrations tested in this study generally showed a greater tendency to absorb and retain atmospheric Hg0 than those at a pH closer to neutral.     

An evaluation of the use of individual grass species in retaining polluted soil and dust particulates in vegetated sustainable drainage devices

A sustainable means of preventing polluted particulates carried in urban storm water entering rivers, groundwater and lakes is by employing vegetated sustainable drainage system (SUDS) devices, or best management practices to trap or biodegrade them. In the UK, a mixture of grass species is recommended for use in devices such as swales or filter strips. However, there is little evidence in support of the efficiency of the individual grasses or mixtures to deal with such contaminated material. A pot-based pollutant retention study was conducted using processed street dust from central Coventry, UK, as a simulated pollutant to be applied in different quantities to a variety of recommended grasses for vegetated SUDS devices. Analysis was conducted on compost cores, roots and shoots for heavy metals (Cd, Cu, Ni, Pb and Zn). Street dust mainly concentrated in the top compost layer for all grasses with only the finer material migrating down the profile. Analysis of roots indicated little accumulation, with ANOVA statistical tests indicating significant differences in heavy metal concentrations, with less in the compost and more in the shoots. Development of root systems on or near the surface possibly explains increased uptake of heavy metals by some species. Overall Agrostis canina and Poa pratensis showed the greatest accumulations compared to their controls although Agrostis capillaris syn.tenuis and Agrostis stolonifera also demonstrated accumulation potential. On ranking, Agrostis canina and Poa pratensis were highest overall. These rankings will assist in selecting the best grasses to address pollution of the urban environment by contaminated particulates.     

Environmental concentrations of agricultural herbicides: 2,4-d and triallate

The herbicides 2,4-D (2,4-dichlorophenoxyacetic acid) and triallate [S-2,3,3-trichloroallyl di-isopropyl(thiocarbamate)] are extensively used to control broadleaf and wild oat (respectively) weed infestations in Canadian cereal crops. In 1990, for example, more than 3.8 million kg of 2,4-D and 2.7 million kg of triallate were applied in the three prairie provinces (Alberta, Saskatchewan, and Manitoba). Maximum air concentrations of these two herbicides during the summers of 1989 and 1990 near Regina, Saskatchewan, were 3.90 ng m(-3) (2,4-D) and 60.04 ng m(-3) (triallate). Concentrations of these two herbicides were also measured in bulk atmospheric deposition (wet plus dry) and in farm pond water and associated surface film. Maximum measured levels of 2,4-D were 3550 ng m(-2) d(-1) (bulk deposition), 332 ng m(-2) (surface film), and 290 ng L(-1) (pond water). Maximum levels of triallate were 2300 ng m(-2) d(-1) (bulk deposition), 212 ng m(-2) (surface film), and 500 ng L(-1) (pond water). The highest quantities of the herbicides tended to be found during or immediately after the time of regional application. The movement of the herbicides in the environment will be discussed in relation to the four matrices studied.