Prairie Skink (Plestiodon septentrionalis): COSEWIC assessment and status report 2017

Official title: COSEWIC Assessment and Status Report on the Prairie Skink (Plestiodon septentrionalis) in Canada 2017

Committee on the status of Endangered Wildlife in Canada (COSEWIC)
Special concern 2017

COSEWIC assessment summary

Assessment summary : November 2017

Common name: Prairie Skink

Scientific name: Plestiodon septentrionalis

Status: Special Concern

Reason for designation: The Canadian distribution of this species is restricted to a small area of mixed-grass prairie on sandy soils in Manitoba and is isolated from the rest of the species’ range in the USA by over 100 km. Its prairie habitat has been historically lost and fragmented mainly due to agricultural activities. Aspen succession and invasion by exotic plants continue to degrade remaining habitats. Several new localities have been discovered within the known range since the last assessment as a result of increased survey efforts, and habitat management is ongoing within portions of the skink’s range on federal and provincial lands. Change in status from the previous assessment results from a different interpretation of status assessment criteria by COSEWIC. While the species is deemed to no longer be at risk of imminent extinction, it could become Threatened if factors affecting it are unmitigated.

Occurrence: Manitoba

Status history: Designated Special Concern in April 1989. Status re-examined and designated Endangered in May 2004. Status re-examined and designated Special Concern in November 2017.

COSEWIC executive summary

Prairie Skink
Plestiodon septentrionalis

Wildlife species description and significance

Prairie Skink is a small, slender lizard with body (snout-vent) length up to 85 mm; the tail can be approximately as long as the body. Prairie Skink is brown with four light stripes along the length of the body and extending onto the tail. Males have reddish orange colouration on the head and throat during the breeding season, and juveniles have bright blue tails. There are three subspecies, but only Northern Prairie Skink occurs in Canada. Northern Prairie Skink’s complete association with the mixed-grass prairie sandhill ecosystems of southwestern Manitoba make it an indicator of this rare landform.

Distribution

Globally, Prairie Skink occurs in a narrow band from southern Manitoba south to coastal Texas with the distribution of the Northern Prairie Skink subspecies extending south to Kansas. In Canada, its distribution is limited to a small area of southwestern Manitoba in the Brandon and Lauder sandhills. It is disjunct from populations in the United States, separated by approximately 125 km.

Habitat

Prairie Skink is associated with mixed-grass prairies and savannas and limited to areas with sandy soils. The skinks require heterogeneous habitat structure for foraging and nesting. Such habitats provide a range of temperatures for thermoregulation, gestation and incubation, and for shelter from predators. The skinks preferentially select warm micro-sites with cover objects and leaf litter. In Manitoba, the skinks occur in mixed-grass prairie with patches of grasses/shrubs or open areas near forest edges. They also use edges of deciduous and mixed-wood forest and have been documented to move through these forest types.

Biology

Prairie Skinks spend over seven months of the year (September – April) in hibernation. They burrow beneath the frost line in sandy soils for overwintering. In Manitoba, they become active in mid- to late April and breed in spring. Home ranges are fairly small (typically <100 m in greatest dimension). Females lay a clutch of 1 – 18 eggs (~5 eggs on average) from late June to early July and remain with the eggs until hatching in late July or early August. The young reach sexual maturity at 1 – 3 years and can live up to approximately seven years of age. The generation time is thought to be 3 – 5 years.

Prairie Skinks are prey for a variety of birds, mammals, and snakes. If captured by a predator, they can detach their tail as a defence mechanism. The severed tail can twitch and move for over 15 minutes, giving the skink time to escape. Prairie Skinks feed on a variety of invertebrates, with crickets and spiders being common food items.

Population size and trends

Estimating the size of the Canadian population is challenging because the full extent of the species’ distribution is incompletely known, the distribution is patchy, and densities vary greatly among sites and years. Extrapolation from density estimates from mark-recapture studies suggests that the population probably consists of at least 10,000 adults with ~98% of the population in the Brandon Sandhills and the remainder in the Lauder Sandhills. Limited information suggests that declines might have occurred at monitored sites in the Brandon Sandhills between 1989 and 2007, but broad scale population trends within the past three generations are unknown. Approximately 28% of the species’ Canadian distribution and 20% of the proposed Critical Habitat under the Species at Risk Act are located within CFB Shilo, where the skinks are currently afforded protection from habitat modification and military training activities. As the species is conservation dependent and requires active habitat management over a significant proportion of its small Canadian range and the main threats to mixed-grass prairie habitat persist, the population could decline if the current level of protection is not maintained and threats across its distribution are not adequately mitigated.

Threats and limiting factors

Prairie Skinks are at the northern limits of their distribution in southern Manitoba and are habitat specialists, increasing their inherent vulnerability to perturbations. The major threats to Prairie Skinks are habitat degradation from invasive plants and vegetation succession of prairie to forest due to fire suppression, and residential development. Roads, railroads, utility and service lines, all-terrain-vehicle trails, grazing, military exercises, and predation by domestic cats may also be threats at some sites. The impact of chemicals applied to crops adjacent to skink habitat and the effect of climate change on skink populations are potential threats but with unknown impacts. Cultivation has destroyed significant amounts of skink habitat in the past, but this threat to new areas has mostly ceased.

Protection, status, and ranks

In Canada, Prairie Skink is listed federally as Endangered and is in Schedule 1 under the Species at Risk Act. In Manitoba, the species is also listed as Endangered under the provincial Endangered Species and Ecosystems Act. A recovery strategy has been proposed as part of national-level planning in Canada. Provincially, a draft Manitoba provincial action plan and recovery strategy have been prepared. Globally, the Prairie Skink’s IUCN status is Least Concern.

A significant portion of suitable Prairie Skink habitat occurs on protected lands in Canadian Forces Base Shilo (28%) and Spruce Woods Provincial Park (13%). However, while the land in these areas may be protected from development, threats from invasive plants, vegetation succession, and soil compaction still exist.

Technical summary

Scientific name: Plestiodon septentrionalis

English name: Prairie Skink

French name: Scinque des prairies

Range of occurrence in Canada: Manitoba

Demographic information

Generation time (usually average age of parents in the population; indicate if another method of estimating generation time indicated in the IUCN guidelines (2011) is being used):

3 – 5 years

Is there an [observed, inferred, or projected] continuing decline in number of mature individuals?

Yes, observed and projected decline. Eight element occurrences are now probably extirpated, which suggests past declines, but the timing of the declines is uncertain. Habitat succession and invasive plant species are ongoing threats and could result in continued decline of suitable habitat and consequently, the number of mature individuals.

Estimated percent of continuing decline in total number of mature individuals within [5 years or 2 generations]:

Unknown

[Observed, estimated, inferred, or suspected] percent [reduction or increase] in total number of mature individuals over the last [10 years, or 3 generations]:

Unknown for the past 10 – 15 year period; the known IAO has increased due to increased search effort and discovery of new localities within the known range.
Since 1965, eight element occurrences have most likely been lost; however, the time frame of the disappearances is uncertain and may precede the past 10 – 15 year period.

[Projected or suspected] percent [reduction or increase] in total number of mature individuals over the next [10 years, or 3 generations]:

Unknown

[Observed, estimated, inferred, or suspected] percent [reduction or increase] in total number of mature individuals over any [10 years, or 3 generations] period, over a time period including both the past and the future:

Unknown

Are the causes of the decline a) clearly reversible and b) understood and c) ceased?

Causes of decline are partly reversible and reasonably understood, but they have not ceased.

Are there extreme fluctuations in number of mature individuals?

No. Annual and multi-year fluctuations occur, but they are unlikely to be extreme (i.e., an order of magnitude).

Extent and occupancy information

Estimated extent of occurrence (EOO):

3785 km²

Index of area of occupancy (IAO) (Always report 2x2 grid value):

316 km²

Is the population “severely fragmented” i.e., is >50% of its total area of occupancy in habitat patches that are (a) smaller than would be required to support a viable population, and (b) separated from other habitat patches by a distance larger than the species can be expected to disperse?:

a) Unknown but possible due to natural and anthropogenic habitat fragmentation, specific habitat requirements of the skinks, and shrinking open habitat on the sandhills due to aspen encroachment. Fragmentation is difficult to assess quantitatively given the paucity of information on subpopulation sizes and viability within habitat fragments; ~70% of the element occurrences (as defined by Manitoba Conservation Data Centre) appear to support viable subpopulations, which does not support severe fragmentation, but many uncertainties exist (see Population Fragmentation for analyses and assumptions).
b) Yes; Prairie Skinks have limited dispersal abilities (movements are typically <100 m). Individuals are capable of movements of approximately 500 m, but these are not common. Habitat patches (as defined by element occurrences polygons) are separated by >1 km.

Number of “locations”^* (use plausible range to reflect uncertainty if appropriate):

Unknown but probably > 10. The greatest plausible threats are habitat loss, deterioration, and fragmentation from invasive plants and aspen encroachment into grassland habitat. While these threats are range-wide, exposure rates and impacts on skinks are likely to vary depending on habitat features and land management, making grouping of occurrences into threat-based locations difficult. It is also uncertain whether these processes occur sufficiently rapidly to qualify as contributing to threat-based locations. The species occurs within approximately 26 land parcels that are under different ownership.

Is there an [observed, inferred, or projected] decline in extent of occurrence?

Not within the past 3-generation period, but a 37% historical decline in EOO has been observed since 1965, resulting from the probable loss of the three most northern element occurrences. Skinks have not been observed at the site near Gladstone since 1965 and have not been observed at the other two sites since 1988.

Is there an [observed, inferred, or projected] decline in index of area of occupancy?

Not documented within the past 3-generation period, but loss of element occurrences has been observed since 1965. Observed increase in known IAO over the past 3 generations reflects increased search effort and not range expansion.

Is there an [observed, inferred, or projected] decline in number of subpopulations?

Yes, observed decline, but the timing is uncertain and could have occurred before the past 15 year period. Ten element occurrences (assumed to represent subpopulations) have not been confirmed since 2000, and eight of these appear to have disappeared and may now be extirpated, with last observations in 1965 (n=1) and in 1988–1999 (n=7).

Is there an [observed, inferred, or projected] decline in number of “locations”*?

Yes, observed decline, but the timing is uncertain and could have been before the past 15 year period. Eight element occurrences are now probably extirpated, and might correspond to locations, each with a site-specific set of threats.

Is there an [observed, inferred, or projected] decline in [area, extent and/or quality] of habitat?

Yes, observed decline in area, extent and quality of habitat due to threats.

Are there extreme fluctuations in number of subpopulations?

No

Are there extreme fluctuations in number of “locations”?

No

Are there extreme fluctuations in extent of occurrence?

No

Are there extreme fluctuations in index of area of occupancy?

No

^* See Definitions and Abbreviations on COSEWIC website and International Union for Conservation of Nature (IUCN) (Feb 2014) for more information on this term.

Quantitative analysis

Probability of extinction in the wild is at least [20% within 20 years or 5 generations, or 10% within 100 years]: Not done due to insufficient data

Threats (actual or imminent, to populations or habitats, from highest impact to least)

Was a threats calculator completed for this species? Yes, on September 1, 2016.

1. Other ecosystem modifications (7.3) – Invasive plants degrading habitat. Predicted impact: Medium-Low
2. Fire suppression (7.1) - allowing aspen encroachment into mixed grass prairie habitat. Predicted impact: Predicted impact: Low
3. Housing and urban areas (1.1) – Loss of mixed grass prairie habitat for houses, lawns, and gardens. Predicted impact: Low

The threats calculator assessment presumed that threats from military activities are mitigated, as currently is the case within the proposed Critical Habitat for the species. Potential reduction in the species’ protection on federal lands where these activities occur would elevate the projected threat from this source.

What additional limiting factors are relevant?

Specific habitat requirements: Prairie Skinks are limited to mixed-grass prairie and forest edge habitat on sandy soils. The species is conservation-dependent within CFB Shilo, which encompasses ~20% of the proposed Critical Habitat for the species under the Species at Risk Act.

Rescue effect (immigration from outside Canada)

Status of outside population(s) most likely to provide immigrants to Canada.

In the US states bordering Manitoba, Prairie Skink is listed as imperilled/vulnerable (S2S3) in North Dakota and secure (S5) in Minnesota.

Is immigration known or possible?

No. Nearest US populations are separated from Canadian populations by a large distance (~125 km) and large stretches of unsuitable habitat.

Would immigrants be adapted to survive in Canada?

Probably

Is there sufficient habitat for immigrants in Canada?

Unknown

Are conditions deteriorating in Canada?⁺

Probably

Are conditions for the source population deteriorating?⁺

Unknown

Is the Canadian population considered to be a sink?⁺

No

Is rescue from outside populations likely?

No

⁺ See Table 3 (Guidelines for modifying status assessment based on rescue effect).

Data-sensitive species

Is this a data sensitive species? No

Current status

COSEWIC: Endangered

Year Assessed: 2004

COSEWIC Status History: Designated Special Concern in April 1989. Status re-examined and designated Endangered in May 2004. Status re-examined and designated Special Concern in November 2017.

Additional Sources of Information: Draft documents for the Prairie Skink recovery strategy in Canada and Critical Habitat identification methods in Manitoba

Status and reasons for designation

Status: Special Concern

Alpha-numeric codes: Not applicable

Reasons for designation: The Canadian distribution of this species is restricted to a small area of mixed-grass prairie on sandy soils in Manitoba and is isolated from the rest of the species’ range in the USA by over 100 km. Its prairie habitat has been historically lost and fragmented mainly due to agricultural activities. Aspen succession and invasion by exotic plants continue to degrade remaining habitats. Several new localities have been discovered within the known range since the last assessment as a result of increased survey efforts, and habitat management is ongoing within portions of the skink’s range on federal and provincial lands. Change in status from the previous assessment results from a different interpretation of status assessment criteria by COSEWIC. While the species is deemed to no longer be at risk of imminent extinction, it could become Threatened if factors affecting it are unmitigated.

Applicability of criteria

Criterion A (Decline in Total Number of Mature Individuals): Does not meet criteria. Although declines are projected based on continuing threats, the magnitude of population decline is unknown.

Criterion B (Small Distribution Range and Decline or Fluctuation): Does not meet criteria. While the EOO and IAO are below threshold values for Endangered, only 1 sub-criterion (b) is met as there are probably greater than 10 locations. Severe fragmentation cannot be demonstrated (sub-criterion “a”) and population fluctuations are not extreme (sub-criterion “c”).

Criterion C (Small and Declining Number of Mature Individuals): Does not meet criteria. Population size is unknown but is probably over 10,000 adults.

Criterion D (Very Small or Restricted Population): Does not meet criteria. Population is neither very small nor very restricted.

Criterion E (Quantitative Analysis): Not done due to lack of data.

Preface

This document is an updated status report on Prairie Skink, Plestiodon (Eumeces) septentrionalis (COSEWIC 2004). New evidence suggests that the species formerly included within the genus Eumeces is not a monophyletic group. As a result, Plestiodon is now accepted as the genus for all North American species (north of Mexico), formerly referred to as Eumeces (Crother 2012).

Since the previous status assessment, increased survey efforts have improved our knowledge of the distribution of Prairie Skink. From 2006 – 2015, 21 new sites were discovered within the known range of the species. Other sites appear to have been lost, including eight sites with last observations in 1965 (n=1) and in 1988 – 1999 (n=7). These possibly extirpated sites include three northernmost localities, resulting in a 37% reduction of the range. Ongoing research by Rutherford and colleagues (Rutherford pers. comm. 2016, 2017) and by the Canadian Wildlife Service (Didiuk pers. comm. 2016, 2017), including mark-recapture studies, continue to expand our knowledge of the skinks’ population biology and habitat use.

A proposed Recovery Strategy for Prairie Skink in Canada has been posted, including description of Critical Habitat (Environment and Climate Change Canada 2016). A Draft Provincial Action Plan and Draft Provincial Recovery Strategy have also been prepared (Prairie Skink Recovery Team 2007; Manitoba Prairie Skink Recovery Team 2014).

COSEWIC history

The Committee on the Status of Endangered Wildlife in Canada (COSEWIC) was created in 1977 as a result of a recommendation at the Federal-Provincial Wildlife Conference held in 1976. It arose from the need for a single, official, scientifically sound, national listing of wildlife species at risk. In 1978, COSEWIC designated its first species and produced its first list of Canadian species at risk. Species designated at meetings of the full committee are added to the list. On June 5, 2003, the Species at Risk Act (SARA) was proclaimed. SARA establishes COSEWIC as an advisory body ensuring that species will continue to be assessed under a rigorous and independent scientific process.

COSEWIC mandate

The Committee on the Status of Endangered Wildlife in Canada (COSEWIC) assesses the national status of wild species, subspecies, varieties, or other designatable units that are considered to be at risk in Canada. Designations are made on native species for the following taxonomic groups: mammals, birds, reptiles, amphibians, fishes, arthropods, molluscs, vascular plants, mosses, and lichens.

COSEWIC membership

COSEWIC comprises members from each provincial and territorial government wildlife agency, four federal entities (Canadian Wildlife Service, Parks Canada Agency, Department of Fisheries and Oceans, and the Federal Biodiversity Information Partnership, chaired by the Canadian Museum of Nature), three non-government science members and the co-chairs of the species specialist subcommittees and the Aboriginal Traditional Knowledge subcommittee. The Committee meets to consider status reports on candidate species.

Definitions (2017)

Wildlife species

A species, subspecies, variety, or geographically or genetically distinct population of animal, plant or other organism, other than a bacterium or virus, that is wild by nature and is either native to Canada or has extended its range into Canada without human intervention and has been present in Canada for at least 50 years.

Extinct (X)

A wildlife species that no longer exists.

Extirpated (XT)

A wildlife species no longer existing in the wild in Canada, but occurring elsewhere.

Endangered (E)

A wildlife species facing imminent extirpation or extinction.

Threatened (T)

A wildlife species likely to become endangered if limiting factors are not reversed.

Special concern (SC)
(Note: Formerly described as “Vulnerable” from 1990 to 1999, or “Rare” prior to 1990.)

A wildlife species that may become a threatened or an endangered species because of a combination of biological characteristics and identified threats.

Not at risk (NAR)
(Note: Formerly described as “Not In Any Category”, or “No Designation Required.”)

A wildlife species that has been evaluated and found to be not at risk of extinction given the current circumstances.

Data deficient (DD)
(Note: Formerly described as “Indeterminate” from 1994 to 1999 or “ISIBD” [insufficient scientific information on which to base a designation] prior to 1994. Definition of the [DD] category revised in 2006.)

A category that applies when the available information is insufficient (a) to resolve a species’ eligibility for assessment or (b) to permit an assessment of the species’ risk of extinction.

The Canadian Wildlife Service, Environment and Climate Change Canada, provides full administrative and financial support to the COSEWIC Secretariat.

Wildlife species description and significance

Name and classification

Prairie Skink (Plestiodon septentrionalis, Baird 1859) was formerly known as Eumeces septentrionalis. However, genetic and other evidence indicate that Eumeces as formerly known is not a monophyletic group (Griffith et al. 2000; Schmitz et al. 2004; Brandley et al. 2005; Smith 2005), and Plestiodon is accepted as the genus for all North American species formerly known as Eumeces (Crother 2012).

The classification of Prairie Skink is as follows:

Class: Reptilia

Order: Squamata

Family: Scincidae

Genus: Plestiodon, Duméril and Bibron 1839

Species: P. septentrionalis, Baird 1859

The species includes three subspecies (Crother 2012), one of which, Northern Prairie Skink (P. s. septentrionalis, Baird 1859), occurs in Canada. The other two subspecies are the Southern Prairie Skink (P. s. obtusirostris, Bocourt 1879) and Pallid Skink (P. s. pallidus, Smith and Slater 1849). Some authors consider the northern and southern subspecies to be separate species (Collins and Taggart 2009). Fuerst and Austin (2004) sequenced two regions of the mitochondrial genome and found sequence divergence ranging from 6.7 – 7.0%, suggesting that the northern and southern forms exist in allopatry with little to no gene flow; however, the sample sizes were too small to adequately address relationships between these taxa. Thus, Crother (2012) recognizes a single species with three subspecies until thorough geographical sampling has been completed.

Morphological description

Prairie Skinks are small, slender lizards, reaching approximately 85 mm in snout-vent length (SVL); most adults are less than 80 mm SVL. The tail can be approximately the same length as the body. Prairie Skinks are brown with four light stripes along the length of the body and extending onto the tail (see cover photo). Males have a reddish orange patch on the head and throat during the breeding season. Some authors have suggested that female Prairie Skinks grow larger than males, but the difference appears to be slight in Canadian populations (Scott 2005; Rutherford 2015). Juveniles have bright blue tails, which in most individuals fades by the time they reach approximately 45 mm SVL, after their first winter (Nelson 1963).

Population spatial structure and variability

In Canada, Prairie Skinks are found in the Brandon Sandhills (formerly Carberry Sandhills) and Lauder Sandhills in Manitoba. Prairie Skinks from the Brandon and Lauder sandhills are separated by ~80 km of unsuitable habitat, and genetic exchange among skinks from the two areas is unlikely. The habitat separating the two sandhills has heavier soil types that are not suitable for overwintering by skinks (Cook 1964). Skinks tend to be aggregated within both areas, forming local subpopulations (Bredin 1989; Manitoba Prairie Skink Recovery Team 2014). The majority of the Canadian population is found in the Brandon Sandhills area. Skinks from the Brandon Sandhills are further separated into 39 extant element occurrences^{Footnote 1}  (Manitoba Conservation Data Centre unpubl. data 2015). These clusters of occurrences may correspond to subpopulations, i.e., isolated units with minimal interchange of individuals. However, genetic analyses using microsatellites among element occurrences did not detect differences that could be linked to anthropogenic habitat fragmentation but did detect two genetic clusters, north and south of the Assiniboine River (McFadden 2010; Siu 2011). Finer scale genetic differentiation may not have been detectable due to the small number of loci examined, small sample sizes, and length of time since separation of subpopulations. One small subpopulation exists in the Lauder Sandhills area (MB CDC unpubl. data 2015). Genetic analyses have not been conducted for this subpopulation.

The Prairie Skink population in Manitoba probably became isolated from the more continuous range of the species to the south of the international border some time after the end of the Hypsithermal period, approximately 4000 years ago. Fuerst and Austin (2004) examined the population genetic structure of Prairie Skinks and postulated that northern populations originated from a single colonizing source that subsequently experienced modest gene flow. Canadian populations were not genetically distinct from populations in the US, but some haplotypes identified from Manitoba samples were not found elsewhere.

Designatable units

Prairie Skink occurs in one COSEWIC Terrestrial Amphibian and Reptile Faunal Province, the Prairie/Western Boreal faunal province. While it occurs in two distinct and separate areas, Lauder and Brandon sandhills, there is no information on genetic, morphological, or behavioural differences between skinks from these two areas that would suggest local adaptations and satisfy the criterion of significance required for designatable units. The post-glacial expansion of skinks into Canada was most likely from one general area south of the border. Thus, there is insufficient evidence for more than a single designatable unit.

Special significance

Prairie Skink is the only lizard that occurs in Manitoba and a unique member of the prairie fauna. It is also the only Canadian reptile unique to Manitoba. The skink’s complete association with the mixed-grass prairie sandhill ecosystems makes it an indicator of this rare landform. The species reaches its northern limits of distribution in southern Manitoba and is disjunct from populations in the US; also, it contains unique haplotypes. Therefore, Canadian populations may have valuable genetic diversity, which may become increasingly important as climate change proceeds

Distribution

Global range

Prairie Skink occurs in a narrow band from southern Manitoba south to coastal Texas (Figure 1). The Northern Prairie Skink subspecies is found from southern Manitoba south to Kansas.

Canadian range

In Canada, the distribution of Prairie Skinks is limited to a small area of southwestern Manitoba disjunct from the US range (Cook 1964). The nearest US population is in northeast North Dakota, approximately 125 km from the nearest Canadian locality. Within Manitoba, Prairie Skinks occur in two discrete areas: Brandon Sandhills and Lauder Sandhills (Figure 2).

The Brandon Sandhills occur on the post-glacial Upper Assiniboine Delta deposits. Within this area, Prairie Skinks are limited to areas of Stockton Loamy Sands (1400 km² ) and Miniota Soils (370 km²) (Bredin 1989). Historically, the Brandon Sandhills subpopulations spanned approximately 69 km north-south from Neepawa to Glenboro and 74 km east-west from 17 km east of Brandon to Treherne. However, there are no recent observations (since 2000) for eight element occurrences. If the records pre-dating the year 2000 are excluded, then the current north-south span is reduced to ~51 km from Edrans to Glenboro. However, sufficient data are unavailable to conclusively determine whether skinks still occur in the Neepawa area, which contains fragments of suitable habitat. Many occurrences are within the Canadian Forces Base Shilo (CFB Shilo) and Spruce Woods Provincial Park (SWPP). New sites continue to be found with increased survey effort, including 21 new sites discovered within the past decade (2006 – 2015; Table 1). However, these are thought to reflect increased survey effort rather an expansion of the area of occupancy.

Note. First observed and last observed dates were tallied by decade to examine occurrence persistence.

Within the Brandon Sandhills, Prairie Skink occurrences are naturally fragmented, reflecting the distribution of appropriate soil types (Figure 3). There are three main areas with appropriate soil types that align with skink occurrences:

1. area south of the Assiniboine River
2. east-west area extending south of Carberry and north of the Assiniboine River. The Douglas Marsh extends more than 15 km into the east-west area, separating northern and southern occurrences of skinks in this area
3. north-south section north of Carberry, consisting of habitat patches of various sizes. A number of these patches are separated by a 20 km long band of clay loam (Wellwood Association) north of Carberry

There are also a number of small patches of habitat with suitable soil, particularly east of Carberry, which are isolated or nearly isolated from other patches of Stockton Loamy Sand and Miniota Sand.

In the Lauder Sandhills, Prairie Skinks have been found only in one small area located in the Lauder Sandhills Wildlife Management Area. Surveys in 2015 – 2016 have increased the area of recorded occurrences to an area ~1.6 km in diameter (<1 ha was estimated in COSEWIC 2004). Suitable unsurveyed habitat was also observed in the vicinity (Didiuk and Murray unpubl. data 2016). This subpopulation, occurring on sandy soils of the Souris Till Plain, is at the western and southern limit of the species’ range in Canada and ~80 km from the Brandon Sandhills subpopulations.

Extent of occurrence and area of occupancy

The Prairie Skink Recovery Team has archived 858 records from 1919 to 2014, representing 604 separate sites with Prairie Skink records. Twenty-eight records were excluded from the EOO and IAO calculations because their accuracy could not be verified (e.g., coordinates did not match site description) or they were from historical sites where the species is considered probably extirpated. An additional 175 records from 2008 to 2015 from 131 unique sites (78 sites not included in the previous dataset) were provided by the Manitoba Conservation Data Centre (MB CDC).

The above records  represent a total of 48 element occurrences or clusters of records. Eight element occurrences identified prior to 2005 are now most likely extirpated (referred to as “historical” by MB CDC), with skinks last observed from 1965 to 1999 (Figure 1; MB CDC unpubl. data 2015). Sites are considered to be historical/probably extirpated if (i) presence has not been reconfirmed in 40 or more years, (ii) sufficient surveys have been conducted and no individuals were observed, or (iii) extirpation is suspected due to a known major disturbance or general habitat loss/degradation (Tomaino et al. 2008). The exact time of the extirpations is unknown, but there have been no new records from any of these sites since the previous status assessment in 2004.

The current EOO was calculated at 3785 km², based on records from 2001 to 2015 (Appendix 1). This value is 37.3% smaller than the historical EOO, including all records. If the localities deemed extirpated are excluded, the comparison of the EOO for two periods, pre-2001 and 2001 – 2015, reveals no differences, based on calculations from the currently available dataset (Appendix 1). The EOO was given as considerably smaller (1770 km²) in COSEWIC (2004), but it was not calculated from the actual records and excluded the Lauder Sandhills records.

The current IAO was calculated at 316 km², based on records from 2001 to 2015 (Appendix 1). Recalculation of the IAO for the pre-2001 period, based on the current data set, results in values of 96 km² and 60 km² with and without the probably extirpated localities, respectively. This represents an increase of 3.3 or 5.3 times in the known IAO since the previous assessment and reflects increased knowledge of the species’ distribution within both the Brandon and Lauder sandhill areas due to survey efforts.

Search effort

Since the previous COSEWIC (2004) status assessment, there has been a significant increase in survey efforts, which have resulted in the documentation of new sites within both the Brandon and Lauder sandhills. The majority of surveys were conducted by Dr. Pamela Rutherford (Brandon University) and her students, with additional surveys by Manitoba Conservation, Manitoba Conservation District (MBCD) Association, and Nature Conservancy Canada. Rutherford and co-workers conducted 1325 surveys at 50 sub-sites on ten large properties from 2006 to 2015. Each sub-site was surveyed 1 – 4 times/active season of the skinks. Most surveys were in the Brandon Sandhills, focusing on CFB Shilo and SWPP.

Colin Murray and co-workers (MBCD) surveyed 115 sites from 2011 to 2015 at nine localities in the following areas: Oak Lake, Routledge, St. Lazare, Stockton, Bede to Lauder West, Lauder Wildlife Management Area and surroundings, and SWPP and surroundings. Skinks were located using plywood cover boards, which were checked annually 1 – 5 times/site each year during the field season. These surveys included collaboration with Canadian Wildlife Service as part of the Lauder Skink Project. Skinks were detected at several sites within the SWPP and its vicinity (Glenboro-Spruce Woods South) and within the Lauder Wildlife Management Area and its vicinity (Lauder East), but not at the other localities.

Devon Baete (Manitoba Habitat Heritage Corporation) conducted cover board monitoring at 28 sites from 2010 to 2011. Nine sites were within a 7 km² area south of SWPP within 2 km of the park boundary. The other 19 sites were in the Stockton area along a 25 km stretch approximately 5 km southwest of CFB Shilo. Sites were checked 1 – 8 times from August 2010 to August 2011. Skink presence was detected at six sites in SWPP. Baete has continued annual monitoring of cover boards from 2012 to 2017 and continues to find skinks in the SWPP area but not at the Stockton sites.

There are also observations from the general public in the Manitoba Herps Atlas (Nature North 2016a). However, Prairie Skinks are cryptic and as a result the full extent of their distribution is still incompletely known. Even areas that are known to support the species are not thoroughly surveyed due to logistic concerns. For example, areas within the Live Fire Range of CFB Shilo contain potentially high-quality habitat but could not be accessed due to safety reasons (Rutherford pers. comm. 2017a).

Habitat

Habitat requirements

Prairie Skinks are associated with mixed-grass prairie and savannah habitats (Breckenridge 1943; Somma and Cochran 1989). In Canada, their distribution is limited to the Prairie Ecozone (Ecological Stratification Working Group 1995). All extant verified sites occur in the Aspen Parkland Ecoregion in the Parkland Prairies Ecoprovince. One site that was farther northeast occurred in the Lake Manitoba Plain in the Eastern Prairies Ecoprovince, but this site is now considered to be unoccupied.

In Canada, Prairie Skinks are found only in areas with sandy soils and appear to prefer loamy sands with 75 – 85% sand content (Cairns 2007). Sandy soils are thought to allow skinks to burrow beneath the frost line and may also facilitate nest construction by females. Skinks appear to be more numerous on south- and west-facing slopes (Bredin 1989; Scott 2005). These slopes tend to be dominated by grass cover, whereas north- and east-facing slopes are dominated by trees and have lower ground temperatures (Scott 2005).

Foraging and nesting require heterogeneous habitat structure to provide a range of temperatures for thermoregulation, gestation and incubation, and shelter from predators. In Manitoba, Prairie Skinks use mixed-grass prairie habitat with patches of grasses/shrubs or open areas near forest edges (Manitoba Prairie Skink Recovery Team 2014). They may also use edges of deciduous and mixed-wood forest and have been documented to move through these forest types . Prairie Skinks appear to use open grassland habitat more in spring and move to areas with denser vegetation, such as juniper (Juniperus sp.) shrub land or aspen (Populus sp.) woodland edges, during warmer conditions in summer (Rutherford unpubl. data 2014 cited in Manitoba Prairie Skink Recovery Team 2014).

Scott (2005) compared Prairie Skink abundance among three land-cover types (forest, shrub, and grass) and in relation to aspect and found skinks to be associated with vegetation characteristic of native mixed-grass prairies, such as bunch and clonal grasses, pteridophytes (ferns and fern allies), prostrate shrubs, spreading grasses, sedges, and moss/lichen, and areas with high solar absorption and leaf litter cover. Scott (2005) suggested that areas dominated by grass and shrubs provide higher quality habitat for skinks than do treed areas because they allow skinks to maintain higher body temperatures. Krause Danielsen et al. (2014) examined habitat selection in exurban private land (low-density rural residential developments, 2 – 8 ha) and found that Prairie Skinks were more often found in native prairie habitat than on mowed lawn or in garden-flower beds. Again, temperature was suggested as the factor driving this pattern; mean daily maximum temperatures were lower in both lawns and gardens than in prairie habitat. Vegetation type (grasses, forbs, or small shrubs) was not a significant variable in site selection. Larkin (2011) found no difference in Prairie Skink densities or ground temperature (2 cm below the surface) between plots that were or were not invaded by non-native Leafy Spurge (Euphorbia esula), but the percentage coverage or stem density of the plants were not incorporated into the study design. It appears that Prairie Skinks may be able to adapt to vegetation composed of native and non-native species as long as vegetation structure provides appropriate microhabitat conditions (Krause Danielsen et al. 2014). However, whether Leafy Spurge patches can provide suitable habitat for skinks is likely to depend on patch density and environmental context (see Threats and limiting factors).

Prairie Skinks use natural and artificial cover objects on the ground surface for shelter from predators and temperature extremes. Females nest under cover objects or in subterranean burrows (Nelson 1963). A variety of cover types are used for refuges, such as flat rocks, fallen trees, clumps of vegetation, sheets of metal, old boards, plywood, shingles, discarded carpets, and other objects (Bredin 1989; Somma 1990). Nelson (1963) found that despite an abundance of boards and logs, Prairie Skinks in Minnesota were most commonly found under leaves, grass or buried in the sand. However, Larkin (2011) and Krause Danielsen et al. (2014) both found that in Manitoba, Prairie Skinks selected cover objects more strongly than other microhabitat components (e.g., forbs, grasses, low shrubs, leaf litter, bare ground). The importance of cover objects may depend on the habitat context. For example, the habitat at SWPP was unburned and had a thicker layer of grass litter compared to habitat at CFB Shilo, which is burned periodically (Scott 2005). At CFB Shilo, skinks appear to move very little and individuals have been found under the same piece of cover where they were originally captured up to three years later (Bredin 1989). At SWPP, skinks moved farther than observed elsewhere (e.g., 16 m in a 4 h period), and movements were not centred around retreat sites (Scott 2005). With the abundance of ground cover at SWPP, skinks may be able to move without being constrained by the thermal environment (Scott 2005). Leaf litter may also provide refuges; Krause Danielsen et al. (2014) found that Prairie Skinks selected sites with a high percentage of leaf litter.

Prairie Skinks appear to tolerate and possibly even benefit from some human modifications to the landscape. Skinks are often numerous in areas with lots of debris, such as in old dumps (Bredin 1988). They can also use woodpiles, debris, porches, and other such structures as cover and nest sites. Prairie Skinks also appear to be common in some areas that are grazed by cattle, which help maintain openness of the habitat (COSEWIC 2004).

Habitat trends

North American mixed-grass prairie habitat has been declining throughout the 20th century (Samson and Knopf 1994). Numerous factors are responsible for such decline in the Brandon Sandhills, including cultivation, urbanization, road construction, grazing, succession to Aspen Parkland (linked to broad climatic fluctuations and likely exacerbated by fire suppression), and the invasion of exotic plants (e.g., Leafy Spurge).

Potato farming has been responsible for much habitat loss in the Brandon Sandhills. From 1961 – 2000 in the Carberry area alone, over 7000 ha of land was converted to potato farming (Town of Carberry 2016). Potato fields border CFB Shilo and may have resulted in the loss of skink habitat. Conversion of land to potato fields appears to be primarily a past threat, because the area seeded for potato farming in Manitoba has stabilized in recent years. During 2014, 25,495 ha of land in Manitoba was seeded for potato production; this area was <20% than land seeded for potatoes in 2009 (Statistics Canada 2012; Agriculture and Agri-Food Canada 2015).

In the past, tree planting programs have reduced skink habitat by increasing forest cover and destroying prairie soils. For example, in 1994 the Manitoba Agro Woodlot Program ploughed almost 3 ha of mixed grass prairie, which previously had supported a population of Prairie Skinks in Spruce Woods Provincial Forest. From the mid-1980s to mid-1990s, scouts and guides planted approximately 80,000 Scots Pine (Pinus sylvestris) trees, non-native to this area, thereby eliminating considerable area of mixed-grass prairie, which is required by Prairie Skinks. The importance of mixed-grass prairie habitat is now recognized; thus, tree planting programs in the region are unlikely to resume. In SWPP, efforts are made to prevent the spread of pine seedlings from plantations (Manitoba Conservation and Water Stewardship 2012).

Mixed-grass prairie also is being lost to succession to Aspen Parkland. Tree-ring studies in SWPP have shown that spruce regeneration and tree growth have fluctuated over the period of 1850 – 2000 (Chhin and Wang 2002; Chhin et al. 2004). Regeneration was found to correlate negatively with high daily July temperatures; the latter is associated with lack of moisture retention in the soil and heightened risk of fire. Periods of decreased recruitment were seen from the early 1920s to the late 1930s. While there have been brief periods of decreased recruitment in the early 1960s, and the late 1990s, the general trend since the 1940s and the late 1990s has been increased recruitment, increasing the rate of vegetation succession. Fire suppression has also played a role in continued forest succession. During 1946 – 1994, over 22% of 1000 ha of crown land was lost to Trembling Aspen (Populus tremuloides) encroachment (Mansell and Moore 1999). Succession to Aspen Parkland has primarily occurred on unused portions of CFB Shilo and throughout SWPP (Bredin 1993). SWPP adopted a prairie management plan in 1996, and 16 prairie sites within the park are maintained through regular burning (Manitoba Conservation and Water Stewardship 2012). Rotational grazing is used to reduce encroachment in other areas of SWPP (Manitoba Conservation and Water Stewardship 2012). Succession is not occurring on active portions of the range at CFB Shilo, largely due to grass fires set accidentally as a result of military exercises. Controlled burns are also used in some areas by Range Control at CFB Shilo (Bredin 1999). Pasture lands grazed by cattle still experience aspen succession in the absence of fire; the amount of land covered by Aspen increases on the average 5.3% per year in grazed pastures in Manitoba (Oliver 2008). The rate of succession might possibly slow down in the future because higher temperatures and more frequent and intense droughts are predicted for Manitoba (Lemmen and Warren 2004), which could reduce survival and establishment of White Spruce (Picea glauca) and Aspen in this region (Chhin and Wang 2002; Hogg et al. 2002; Chhin et al. 2004).

Agriculture is permitted in some areas within SWPP: cattle grazing (two areas), haying (31 fields), and cropping (one permit) occur on former agricultural lands in the park (Manitoba Conservation and Water Stewardship 2012). However, agricultural land within SWPP is expected to decrease over time because no new areas are allocated for haying or cultivation, and fields where hay production is no longer feasible are being returned to natural conditions (Manitoba Conservation and Water Stewardship 2012).

Rutherford et al. (2012 and in prep.) attempted to calculate habitat loss for Prairie Skinks for lands outside of CFB Shilo and SWPP. They estimated that 62% of the land within the extent of occurrence (as known at that time) of Prairie Skinks has suitable sandy soils that allow burrowing (125,000 ha of 203,000 ha). Vegetation data were limited, but land-use data from 1966 and 2000 were available. Using these data, Rutherford et al. (2012) estimated that during 1966 – 2000, agricultural activities caused a loss of 19% of habitat with suitable substrates for Prairie Skinks. During this same period, the Brandon Sandhills area experienced a 26% change in habitat from woodland to grassland (Rutherford et al. in prep.). This increase in grassland may provide new habitat for Prairie Skinks; however, this change is primarily from landowners clearing woodland for grazing, and the suitability of grazed habitat for Prairie Skinks is unknown (Rutherford et al. in prep.). Accordingly, the area and/or quality of habitat available for Prairie Skinks will probably continue to decline, given that the main threats to mixed-grass prairie persist.

Biology

Seasonal activity and overwintering

Prairie Skinks are terrestrial, but they can dive into water to avoid capture. They have not been observed climbing trees or shrubs (Nelson 1963), unlike the congeneric Five-lined Skink (Plestiodon fasciatus). Skinks often thermoregulate while under cover and hence are rarely seen. They require loose substrates to allow them to bury sufficiently deep to avoid freezing during winter.

In Manitoba, Prairie Skinks enter hibernation by mid-September and hibernate for over 7 months of the year. They emerge from hibernation in mid- to late April (Bredin 1988). Males generally emerge first, and juveniles tend to emerge 3-4 weeks after adults (Nelson 1963). Prairie Skinks are diurnal and most active from late morning to mid-afternoon during spring and summer months. They tend to intersperse short periods of rest with periods of activity (Nelson 1963). Prairie Skinks spend the night under cover objects or buried 5 – 8 cm within the substrate (Nelson 1963).

In the Brandon Sandhills, all capture sites in spring and fall were approximately in the same area as in summer (Rutherford 2015), but there are no data on actual hibernation sites in Canada. In the US, Prairie Skinks were documented to hibernate communally or individually at depths of up to 74 cm in Minnesota and 137 cm in Iowa (Scott and Sheldahl 1937; Nelson 1963). Under experimental conditions in enclosures in Minnesota, skinks successfully overwintered when they buried themselves 30 cm below the ground surface (Breckenridge 1943). In Minnesota and Iowa, Prairie Skinks hibernated in gravel deposits (Scott and Sheldahl 1937; Breckenridge 1943). Fidelity to specific overwintering sites has been observed (Nelson 1963).

Life cycle and reproduction

Mating occurs in spring. Males develop orange colouration on the chin at this time and are aggressive toward other males. Females lay a clutch of 1 – 18 eggs, with clutch size increasing with the body size of females (Somma 1987a; Bredin 1988); however, the number of eggs laid by a particular female can vary markedly from one year to the next (Nelson 1963). During drought years, females laid an average of only 1.8 eggs (n = 5) compared with an average of 8.4 eggs (n = 8) in wetter years (Bredin 1988). Mean clutch size was 4.4 eggs (range: 1 – 7, n = 9) from 2007 – 2014 in Manitoba (Rutherford 2015). Eggs are laid from late June to early July and hatch from late July to early August (Bredin 1988). Incubation period ranged from 28 – 52 days and averaged 34 days for ten nests in Manitoba (Breckenridge 1943; Nelson 1963; Bredin 1988). Hatching success under laboratory conditions was less than 40% (Somma and Fawcett 1989).

Females may nest communally, and up to three nests have been found under a single cover object in Manitoba (Bredin pers. obs. in COSEWIC 2004). Females remain with their eggs until hatching. Eggs are often, but not always, laid under cover objects. In Minnesota, females tracked with radioactive tags constructed subterranean nests 3 – 9 cm below the surface in open sandy areas (Nelson 1963).

Hatchlings average approximately 25 mm SVL (Breckenridge 1943; Nelson 1963; Bredin 1989). Sexual maturity is reached at 55 – 65 mm SVL (Breckenridge 1943; Bredin 1989; Rutherford 2010). Some individuals reach breeding size at one year of age and would breed during the following spring when they are close to two years old (Rutherford 2015). Most individuals reached breeding size by the age of two years, and by age three all animals were capable of breeding (Rutherford 2015). Therefore, females would breed for the first time after their second to fourth winter. Females bred annually and frequently returned to the same general area to nest in Minnesota (Nelson 1963). However, in a given year, only 19% of marked adult females reproduced at study sites in the Brandon Sandhills, which suggests that individual females may not reproduce annually in Manitoba (Rutherford 2010).

Survival

In Manitoba, few marked skinks have been captured across more than two years, although one skink first caught as an adult was re-captured five years later, suggesting a potential lifespan of at least seven years (Bredin 1999). Rutherford (2010) examined the body size distribution of 569 Prairie Skinks from four study sites in the Brandon Sandhills, which also suggested longevity of about seven years. During 2007 to 2011, mark-recapture methods were used to calculate annual survivorship of Prairie Skinks at seven study areas within CFB Shilo, Manitoba; estimates were 0.65 for juveniles and 0.78 for adults (Rutherford 2015). The majority of reproducing adults are likely 3 – 6 years old, and generation time is probably somewhere between 3 – 5 years.

Like most reptiles, Prairie Skinks are apt to be most vulnerable at the egg and hatchling/juvenile stages. However, brooding of eggs by the female may increase nesting success compared with species with no parental care. Egg mortality is significantly reduced when the female skink remains with her eggs (Somma and Fawcett 1989). Egg mortality increases with very low (5 – 10%) or very high (25 – 30%) soil moisture contents (Somma and Fawcett 1989). Hatchling size is positively correlated with soil moisture levels (Somma 1989).

Physiology and adaptability

Through basking, Prairie Skinks can maintain a body temperature higher than the ambient air or substrate temperature (Nelson 1963). Body temperatures of wild skinks in Minnesota averaged 29.7°C (range 22 – 35°C) during the active season, and the lethal thermal limit was 41 – 44°C (Nelson 1963). Prairie Skinks from Manitoba selected temperatures averaging 33.5 ± 0.8°C in the laboratory, but their body temperatures were considerably lower (25.3 ± 0.2°C) in the field during May – October (Rutherford unpubl. data in Manitoba Prairie Skink Recovery Team 2014). This discrepancy may reflect the use of microsites that are not thermally optimal to avoid predation (Scott 2005) and/or shortage of warmer microsites in the landscape.

Prairie Skinks readily use anthropogenic cover-objects for refuges, which have been deployed with success as a survey method. They may be able to tolerate some human activities, provided that essential habitat features, such as refuges and suitable substrates, are retained.

Dispersal and migration

Home ranges are assumed to be small based on displacement distances of marked individuals. Bredin (1989) used mark-recapture data from 1981 to 1983 to examine movement distances in Manitoba; the longest movement observed was ~18 m (straight-line distance). Bredin (1989) found that Prairie Skinks made limited movements over the course of the active season and were frequently found under the same cover within a year, and sometimes even from one year to the next. Nelson (1963) found that the behaviour of Prairie Skinks in Minnesota varied greatly among individuals; some individuals remained within areas of ~30 m in diameter for up to four years, while others used areas >100 m in diameter within ca. two weeks. Radio-telemetry has been attempted, but Prairie Skinks lose their transmitters easily. In the Brandon Sandhills, Rutherford (2013, 2015) tracked seven adult Prairie Skinks for 3 – 8 days from July 5 to August 9, 2012 and 11 adults for 7 – 23 days from May 27 to June 29, 2013; minimum convex polygons of use areas for these skinks ranged from 15 – 7016 m² in 2012 and from 0 (all observations in the same place) – 6914 m² in 2013.

Prairie Skinks may occasionally make longer movements. Movements up to 500 m have been observed, but these are uncommon (Rutherford pers. comm. 2017b). A single skink was observed along a railway line in a bog over 1 km from the edge of known habitat (Bredin 1989), which could represent a long-distance dispersal movement. Alternatively, the open railway corridor could have provided suitable habitat for skinks to occupy this linear habitat (Didiuk pers. comm. 2016). There are no data on dispersal movements of hatchling Prairie Skinks. Five-lined Skink hatchlings have been observed to move over 100 m (Seburn 1993).

Interspecific interactions

Prairie Skinks feed on a variety of small invertebrates, and it is unlikely that food is a limiting resource. In Minnesota, adults fed mainly on crickets, grasshoppers and spiders (Breckenridge 1943; Nelson 1963). Breckenridge (1943) examined stomach contents of 37 adult Prairie Skinks, and by frequency of occurrence the diet consisted of 29.5% Araneae, 27.3% Orthoptera, and 28% of a combination of Homoptera (mainly leaf hoppers), Coleoptera, and Lepidoptera (mainly larvae). Nelson (1963) examined stomach contents for 28 hatchling Prairie Skinks from Minnesota; 142 food items were found and the identified invertebrates consisted of Araneae (46%), Orthoptera (15%), Diptera (15%), Homoptera (13%), Lepidoptera (2%), Acari (1%), Coleoptera (1%), and Hemiptera (1%). The smallest food item was a mite ~1 mm in diameter and the largest a 18 mm long caterpillar, but most food items were 4 – 9 mm in length (Nelson 1963). Ants, a common soil invertebrate, were found only in one sample from an adult skink by Breckenridge (1943); none of the hatchlings examined by Nelson had consumed ants (1963). Cannibalism can also occur, and Breckenridge (1943) found that one adult female had consumed a yearling skink.

Captive feeding experiments suggest that Manitoba skinks prefer spiders and crickets (Bredin 1989). Grasshoppers and mealworms were consumed only after several days of fasting. Adult beetles, caterpillars, and a chopped meat mixture were all refused.

Skinks, particularly juveniles, are preyed upon by a variety of birds, mammals, and snakes, and occasionally even by other Prairie Skinks (Breckenridge 1943). American Kestrels (Falco sparverius) may be a significant predator in some areas; a pair of kestrels was observed to feed numerous skinks to nestlings (in COSEWIC 2004). Domestic cats have also been reported to kill skinks (Bredin 1989; Krause Danielsen et al. 2014). Prairie Skinks can detach their tail as a defence mechanism. The severed tail can twitch and move for over 15 minutes (Bredin 1981), potentially giving the skink time to escape. The proportion of individuals with missing or re-grown tails could be an indicator of local predation pressure. At Cedar Creek Natural History Area in Minnesota, Pitt (2001) observed tail loss in 7.6% of the individuals examined (n = 108). Nelson (1963) reported tail loss rates of 3% in hatchlings, 35% in juveniles, and 77% for adults at his study site in Minnesota. At SWPP in Manitoba, 31% of females (n = 13), 11% of males (n = 9), and no juveniles (n = 8) were missing tails (Scott 2005).

Population sizes and trends

Sampling effort and methods

Three mark-recapture studies have been conducted in Manitoba. Bredin (1989) calculated the density of skinks at a study area at CFB Shilo (study area A) based on mark-recapture of 108 skinks in 1981 –1983. Rutherford (2015) estimated population size for five sites (including Bredin’s study area A) at CFB Shilo in 2007 – 2011, using the Closed Captures model in MARK v6.1 (Table 2); sample sizes at two additional mark-recapture sites were insufficient for analysis. Skinks were caught by hand using cover boards or walking surveys, measured, marked, and released at their original capture sites. Each of five sites had 5 – 7 clusters of cover-boards (¾” thick plywood, 1’ x 2’). A cluster of cover-boards contained 5 – 10 boards (most had 10 cover-boards). The boards within a cluster were placed ~30 cm (1’) from each other in a circular pattern, resulting in ~60 cover-boards/site. The sites were surveyed four times each field season (May – September). All sites were in grassland habitat. Skinks were grouped into young-of-the-year, juvenile, or adult (SVL > 55 mm), but the data were pooled for population size estimation to increase the sample size. Didiuk and Murray (unpubl. data 2016) conducted a mark-recapture study of Prairie Skinks in two 36-hectare study areas, one each in SWPP and CFB Shilo, in 2016. Each study area had 2500 cover boards arranged within a grid, where the grid points were in the centre of a 25 m by 25 m area and sampled approximately 0.0625 ha of habitat. The boards were visited 15 times during the field season from early May to late September 2016. Skink densities were calculated using only Grass, Grass/Open Shrub, Edge of Woods, Open Sapling Patches, and Open Tee Patches. Tree/shrub, Conifer, and Opening in Woods habitat types were excluded. Therefore there were 354 grid cells (22.1 ha, 1408 boards) at the CFB Shilo grid, and 453 grid cells (28.3 ha, 1806 boards) at the Spruce Woods Provincial Park grid.

Note. Insufficient sample sizes at Areas D and G prohibited population size estimation.

Abundance

Bredin (1989) calculated a density of 48.8 skinks/ha at study area A in Manitoba for 1981 – 1983 but did not give an estimate of the local population size. Rutherford (2015) calculated a population size of 150 skinks (95% confidence limits: 135 – 178) and density of 26.8 skinks/ha (Rutherford unpubl. data 2014) in the same study area for 2007 – 2011. Rutherford’s (2015) population estimates for her other four study areas ranged from 33 – 99 skinks/study area (95% confidence limits: 28 – 116) (Table 2). Population densities from these four study areas were estimated to be 1.3 – 2.8 (mean ± SE = 1.8 ± 0.3) skinks/ha (Rutherford unpubl. data 2014). The average density from these study areas may not be representative for the majority of areas occupied by skinks, because Rutherford conducted her research in areas where skinks could be found in relatively high densities (Rutherford pers. comm. 2017b).

Estimating the size of the Canadian population is challenging because the full extent of the species’ distribution is incompletely known, the distribution is patchy, and densities may vary greatly among sites and years. However, a crude estimate can be derived through extrapolation from densities obtained from mark-recapture studies. An average population density of 1.8 skinks/ha was calculated using Rutherford’s (unpubl. data 2014) density estimates from study areas B, C, E, and F; study area A (density = 26.8 skinks/ha) was not included in the average because it had unusually high density of skinks, higher than observed anywhere else in Manitoba (Rutherford pers. comm. 2017b). Preliminary analysis of Didiuk and Murray’s (2016) study resulted in an apparent density of 7.7 adults/ha at SWPP and 3.34 adults/ha at CFB Shilo for a mean of 5.37 adults/ha. These values are based on observations of 218 and 62 adults, excluding recaptures but including escapees (with unknown recapture status), in grassland habitats of the two study areas, respectively. The number of potential recaptures of skinks seen but that escaped capture was estimated.

The area occupied by the Canadian population of Prairie Skink was calculated by creating a minimum convex polygon for each element occurrence deemed extant. This polygon layer was merged with 100 m buffer polygon layer to calculate area of occupied habitat; this width was based on the maximum documented dimension of individual home ranges (Nelson 1963). The resulting polygon layer covered a total of ~9000 ha with 8849 ha in the Brandon Sandhills and 129 ha in the Lauder Sandhills. Using density estimates from Rutherford’s study, the adult population size was estimated by multiplying the average density for the four study areas (1.8 skinks/ha) by the proportion of captured individuals classified as mature (0.62) by the area occupied (ha) estimates for all element occurrences, except for study area A, where the observed density (26.8 skinks/ha) was used. Adult population size estimates using this method are 10,660 adults for the Brandon Sandhills and 144 for the Lauder Sandhills, for a total of 1151 adult skinks. Using densities from Didiuk’s study, the estimates are considerably higher at 47,519 and 693 adults/ha for the two areas, respectively.

The above values may be overestimates because unsuitable habitat and habitat of lower quality were included in the area occupied estimate, and skink densities may be relatively high in the prime habitat in the protected areas where the mark-recapture studies were conducted. Furthermore, population fluctuations, as inferred from other species of skinks (see Fluctuations and Trends), could result in periodically lower or higher population sizes on multi-year basis. However, while there is considerable uncertainty, given that there is additional unsurveyed habitat, it is likely that the Canadian population exceeds 10,000 adults, with approximately 98% of the population in the Brandon Sandhills and only 2% in the Lauder Sandhills.

Fluctuations and trends

The population density calculated for study area A was lower in 2011 than in 1989 (Bredin 1989; Rutherford unpubl. data 2014), but density estimates were calculated using different techniques for the two periods (Rutherford unpubl. data 2014). Also, population densities for this species can have high annual variability. Population densities calculated for a Five-lined Skink population in Kansas were found to vary among years from 81.5 – 227.2 skinks/ha (Fitch 1954 in Bredin 1989). Fluctuations could be due to differences in yearly reproductive success, which is influenced by climatic factors (Bredin 1989).

The persistence of known element occurrences can be used to infer population trends. Many new element occurrences (n = 21 of a total of 48, as determined by MB CDC) have been documented since 2005 as a result of increased survey effort (see Table 1 for number of element occurrences first observed and last observed for each decade). Of the 27 pre-2005 element occurrences, 19 are considered to be extant (70%) and 17 of these have had at least one observation from 2005 – 2015 (63%). Eight occurrences identified prior to 2005 (30%) are now probably extirpated, with skinks last being observed between 1965 and 1999.

Results of the threats calculator assessment suggest a continuing decline (see threats and limiting factors). An overall threat impact of medium to low, as projected for this species, indicates a decline of 8 – 30% for medium and 0 – 10% for low impact threats over the next three generations from threats acting over the next ten years. However, the projected impacts could be higher than these results indicate because threats were evaluated assuming that the current level of protection will continue. The species is conservation-dependent within CFB Shilo, which comprises ~28% of the species’ Canadian range and 20% of the proposed Critical Habitat, where mitigation measures are currently implemented in light of the species’ endangered status. Were the mitigation measures from military activities withdrawn or reduced in the future, declines in habitat quantity and quality would occur.

Population fragmentation

The Prairie Skink’s distribution is patchy, partially because the species requires a distinct combination of sandy soils and mixed-grass prairie habitat, but also because of loss of suitable habitat due to human activities and forest encroachment. Despite living in naturally patchy habitats, skinks have limited dispersal abilities and are unlikely to move >1 km across unsuitable habitat. Aspen encroachment and human activities are increasing isolation of suitable habitat patches. COSEWIC (2004, p.18) considered the population severely fragmented because “most individuals (are) found within small and relatively isolated (geographically or otherwise) populations between which there is little exchange, i.e., < 1 successful migrant/year”.

On a broad scale, the two sandhills occupied by the skinks could be considered habitat patches. At this scale, the skink population is not severely fragmented because the Brandon Sandhills contain most of the population, and there are many occurrences with apparently good viability. However, the habitat is also patchy within the Brandon Sandhills, including sand dunes and patches of grassland interspersed with unfavourable woodland areas and anthropogenic barriers to movement. The element occurrences layer created by MB CDC can provide some insight into the degree of population fragmentation at the landscape scale as they represent relatively isolated clusters of observations defined by their distance to other such clusters and habitat suitability in the intervening areas. Therefore, it may be assumed that they correspond to subpopulations inhabiting patches of suitable habitat. Of 40 extant element occurrences identified by MB CDC (unpubl. data 2017), viability was estimated to be excellent for nine, good for eight, and poor for one occurrence; viability of the remaining occurrences either could not be assessed (18 occurrences) or was not assessed (4 occurrences) (Table 3; Appendix 2; see Tomaino et al. 2008 for viability classification criteria). Seventeen of the 40 extant occurrences were deemed to have good to excellent viability, covering 70% of the total area of occupancy, including the area of polygons that were not assessed for viability (Table 3). However, this analysis does not provide accurate area estimates for subpopulation sizes because the area mapped for each element occurrence reflects the precision of the data points (precise data points have small polygons surrounding them, whereas imprecise data points have large polygons to cover the uncertainty for the point), not area occupied in each patch. However, a second calculation using the area occupied estimates from the Abundance section does provide similar values (83% using the minimum convex polygon + 100 m buffer area estimates).

Note. See Appendix 2 for details by EO.

The above analyses imply that the population is not severely fragmented at the landscape level in the strictest sense of the definition of this term, which requires that most of the population is in habitat patches that are smaller than expected to support a viable population over the long term. However, the viability of 22 of the 40 (55%) element occurrences could not be assessed; also, the extent of the area occupied by these clusters may be underestimated because most had only a few skink observations. It could be argued that “verified extant” status clusters might not represent viable populations because of presumed low densities and small (but unknown) population sizes. Furthermore, the viability assessments for element occurrences should be interpreted with caution and, in some cases, may be based only on habitat quality or threats; for example, one site assessed to have good viability does not have a single valid record. There is also uncertainty about how well the element occurrences match with COSEWIC’s definition of subpopulations. The element occurrences that are separated by barriers or >1 km of unsuitable habitat can be considered to be separate subpopulations, but the element occurrences that are separated by >5 km of suitable habitat might not qualify because survey effort has not been great enough to confirm absence of skinks in the intervening habitat. Furthermore, data were not available for detailed habitat mapping to estimate patch size independent of the above clusters.

In conclusion, the available information does not allow to determine with confidence whether the population is severely fragmented. However, the data suggest that there are several subpopulations that continue to persist and are presumed to have good viability.

Rescue effect

The nearest US population of Prairie Skinks is located in northeast North Dakota, approximately 125 km from the nearest Canadian site. The Canadian population is genetically similar to the US populations (Fuerst and Austin 2004), and immigrants can be expected to survive here. Nonetheless, rescue from the US populations is deemed non-existent because the sites are separated by long distances and by large stretches of unsuitable habitat. In addition, nearby North Dakota populations are classified as vulnerable (S2S3), although Minnesota populations are considered secure (S5).

Threats and limiting factors

Limiting factors

Prairie Skinks are at the northern limit of their distribution in southern Manitoba. Within this area, their distribution is restricted to areas of mixed-grass prairie on sandy soils. In winters with poor snow cover, skinks may be killed in their hibernacula when sub-zero temperatures penetrate more deeply. Overwinter mortality could be a significant source of mortality, especially for hatchling Prairie Skinks (Bredin 1989). Weather during the active season may also be a limiting factor. Prolonged cold and wet weather could reduce reproductive activity, hatching success of eggs, or the ability of adults and young to forage and store fat reserves, which could in turn reduce overwinter survival or reproduction the following year.

Historical loss of mixed-grass prairie habitat though cultivation may be the most important limiting factor for Prairie Skinks in southwest Manitoba. Significant amounts of habitat were converted to cultivation historically and to potato fields in the more recent past, but the area seeded for potato farming in Manitoba has stabilized in recent years (see Habitat Trends). Tree planting programs destroyed mixed-grass prairie habitat in the past, but the importance of mixed-grass prairie habitat is now recognized, and efforts are underway to inventory and protect the remaining parcels (Province of Manitoba 2015c).

Threats

The IUCN Threats Calculator was applied to Prairie Skinks by a panel of experts (Appendix 3). The overall threat impact was deemed to be “medium – low”, based on one medium – low impact and one low impact threat. In addition, impacts of three threat categories were scored as unknown. The threats determined to apply are discussed below in their perceived order of importance.

Natural ecosystem modifications (threat impact = medium - low)

Invasive plants (other ecosystem modifications)

Exotic, invasive plants are a threat to the native mixed-grass prairie habitat, and several species can be found in Manitoba. Leafy Spurge, Smooth Brome (Bromus inermis), and Purple Loosestrife (Lythrum salicaria) are problems in SWPP (Manitoba Conservation and Water Stewardship 2012), of which the first two occur in Prairie Skink habitats.

Leafy Spurge has been identified as a potential threat to Prairie Skinks (Bredin 1989) and has now invaded every Prairie Skink site examined (Rutherford pers. comm. 2017a; Figure 5). This Eurasian invader is a herbaceous, deep-rooted perennial (family Euphorbiaceae) that can grow up to one metre in height. It can reproduce vegetatively or by seed and is associated with anthropogenic disturbances, such as vehicle tracks (Belcher and Wilson 1989). Leafy Spurge was first observed in North America in 1827 in Massachusetts (Britton 1921). In Manitoba, it was first noted in 1900, and Bird (1961) wrote that Leafy Spurge “is now widely established”. This is particularly well demonstrated in the Spruce Woods Forest Reserve, where many hundreds of patches now occur. Significant efforts to eradicate this species have been undertaken in Canada and the US, but the species continues to be a problem (Team Leafy Spurge 2005; Province of Manitoba 2015a). Leafy Spurge currently infests approximately 52,600 ha in Manitoba (Province of Manitoba 2015a).

Leafy Spurge is considered to be a threat to Prairie Skinks because dense stands reduce structural heterogeneity of the habitat and could limit thermoregulatory options. Dense patches of Leafy Spurge have an average of 96 stems/m² (Bredin 1988). Three cases have been documented where Prairie Skinks abandoned sites following colonization by this plant (Bredin 1988). However, the actual effect on Prairie Skinks is unclear. Larkin (2011) found that thermal conditions under Leafy Spurge were not significantly different from adjacent prairie habitat. Rutherford (2010) found comparable abundance of Prairie Skinks between spurge and control patches. The density of Leafy Spurge patches and the thermal environments of adjacent habitat patches may influence whether Leafy Spurge patches can provide suitable habitat for Prairie Skinks. Additional research investigating the suitability of Leafy Spurge habitats for Prairie Skinks in Manitoba is currently underway (Didiuk pers. comm. 2016). Preliminary results from that study suggest that skinks occur in areas where Leafy Spurge is present at various densities but avoid habitat patches with dense concentrations (>70% coverage) of Leafy Spurge (Didiuk pers. comm. 2017). At present, Leafy Spurge appears to be expanding its distribution within CFB Shilo and SWPP (Rutherford pers. comm. 2017c), but the occurrence of high density patches appears to be constrained by moisture conditions and terrain features at present (Didiuk pers. comm. 2017).

Leafy Spurge is managed using chemical control, cultivation, planting competitive crop species, mowing, burning, tilling, grazing using sheep (Ovis aries) and goats (Capra aegagrushircus), and biological control. The use of flea beetles (Aphthona sp.) has been most effective (Province of Manitoba 2015a). Use of flea beetles began in 1983, and two species (A. nigriscutis and A. cyparissiae) have now been released at 900 and 250 sites in Manitoba, respectively (Province of Manitoba 2015a). Flea beetles have been released at over 300 sites in SWPP (Province of Manitoba 2015b). Although it takes time for populations to become established, flea beetles have removed 95% of the spurge at some of the earliest release sites (Province of Manitoba 2015a). Goat grazing was used for three years to help control the spread of Leafy Spurge at one site in SWPP (Manitoba Conservation and Water Stewardship 2012). The goats have been effective at reducing Leafy Spurge, but many years of grazing would be needed to effectively control it (Province of Manitoba 2015b). Chemicals were used to control Leafy Spurge in SWPP from 1983 – 2009 (Manitoba Conservation and Water Stewardship 2012).

Relatively little is known of the impacts of the introduced Smooth Brome on Prairie Skink, but high abundance could have negative effects similar to those postulated for Leafy Spurge. Brome currently occurs in relatively isolated patches, but it continues to expand its occurrence in skink habitats, and the patches have a relatively large basal area when compared to Leafy Spurge. Ongoing work suggests that skinks continue to use habitats with Brome patches (Rutherford pers. comm. 2016). However, invasive plant patches may need to be interspersed with native prairie to provide adequate habitat, and significant knowledge gaps remain on the effects of this and other invasive plants on skinks.

This threat was assessed as pervasive in scope (71 – 100% of the population affected) because Leafy Spurge was found at all sites examined, and moderate-slight in severity (1 – 30% decline expected at affected sites over the next three generations). A range was used for severity because there are significant knowledge gaps on whether, or under what conditions, skinks can co-occur with Leafy Spurge and other invasive plants.

Fire suppression

Prairie Skinks require heterogeneous habitat conditions and are associated with vegetation characteristic of native mixed-grass prairie (Scott 2005). Fire suppression is a threat to mixed-grass prairie habitat because it results in the succession towards Aspen Parkland. Succession is believed to have caused the disappearance of Prairie Skinks from at least four areas (Bredin unpubl. data 2003 in COSEWIC 2004). Fire suppression may also reduce habitat quality by increasing the layer of thatch (dead plant material), which insulates the ground. Repeated burning causes a reduction of litter with each additional burn and results in higher soil surface temperatures during the growing season (Shay et al. 2001). However, the effect of decreased thatch on Prairie Skinks is not clear. Increased insulation could reduce the active season for Prairie Skinks. Pitt (2001) found that Prairie Skinks were more abundant in old fields that were burned regularly in Minnesota than in habitats that were not burned. However, Krause Danielsen et al. (2014) found that Prairie Skinks selected habitat with a higher percentage of leaf litter in Manitoba, so the environmental context may dictate whether the effects of litter are positive or negative.

Grazing by cattle may slow down Aspen succession but is not sufficient to prevent it (see Habitat Trends). Climate change could possibly slow down the rate of succession in the future, because higher temperatures and increase in drought conditions are predicted for Manitoba (Lemmen and Warren 2004), which could reduce the survival and establishment for trees in this region (Chhin and Wang 2002; Hogg et al. 2002; Chhin et al. 2004).

There are fire restrictions on the land surrounding SWPP. Within the park, controlled burns are conducted to maintain major prairie sites (Manitoba Conservation and Water Stewardship 2012), but the park is not attempting to reclaim prairie lost to forest encroachment in the past (Oliver pers. comm. 2003 cited in COSEWIC 2004). Disturbance is essential to prevent succession of prairie habitat to Aspen Parkland; however, the benefits of fire may be influenced by its intensity, frequency, and timing.

This threat was assessed as large in scope (31 – 70% of population affected) because forest encroachment is a problem throughout much of the skinks’ range, and slight in severity (1 – 10% decline expected at affected sites).

Residential and commercial development (impact = low)

Residential development affects Prairie Skinks though vegetation and soil disturbance, resulting in loss and fragmentation of habitat. Residential development often results in conversion of native prairie into gardens and Kentucky Bluegrass (Poa pratensis) lawns. These habitats are cooler on average and have less variable substrate temperatures than native prairie (Krause Danielsen et al. 2014). While Prairie Skinks can survive and reproduce in residential land, they are more abundant in prairie habitats (Krause Danielsen et al. 2014). Multiple threats exist in residential landscapes (such as mowed lawns that are unsuitable, cat predation, and road networks), thus cumulative impacts are of a concern. Mitigation through stewardship may be possible. Prairie Skinks can survive in habitats with some exotic species and artificial cover as long as the vegetation structure and microhabitat conditions are suitable and provide a heterogeneous landscape with opportunities for thermoregulation and shelter (Krause Danielsen et al. 2014). Residential development is occurring adjacent to SWPP and CFB Shilo. This threat was assessed as small in scope (1 – 10% of the population exposed) because ongoing new developments might affect only approximately 1% of skink sites. The severity was assessed as slight (1 – 10% decline predicted over the next three generations) because skinks can use these environments, but cumulative impacts could cause local extirpations.

Other threats (negligible impacts)

Threats assessed as negligible at present could become more important in the future and be locally important. Paved roads are considered to be a barrier to movements of Prairie Skinks (Didiuk pers. comm. 2016). Prairie Skinks may cross gravel roads, wide sandy trails, and railway tracks but probably with increased risk of mortality compared to natural habitat (Didiuk pers. comm. 2016). Human modified landscapes, such as pasture land, roadsides, utility and service lines, all-terrain-vehicle (ATV) trails, and military activities may have both positive and negative effects on skink habitat, depending on the extent and intensity of the activities. ATV activity, on and off trails, is intensive and ongoing in Lauder Sandhills and can result in soil compaction, collapse of burrows, and mortality of skinks. Cattle, vehicles, and heavy foot traffic can also cause soil compaction (McKernan 1984). Military training exercises may also scour and excavate the soil. Soil compaction and excavation could reduce suitability for burrowing and potentially destroy overwintering sites. However, activities in landscapes disturbed by humans often create open habitat which can provide increased thermoregulatory opportunities for skinks. The positive or negative effect of these activities depends on habitat context and the intensity of impacts, and the overall threat impact for all the above was scored as “negligible”.

The threat from military training activities within CFB Shilo, which encompasses approximately 28% of the species’ distribution and 20% of the proposed Critical Habitat, was assessed as negligible assuming that current level of protection afforded to the skink is maintained. While species listed as Special Concern receive some level of protection and are subject to monitoring activities, it is likely that the current level would be reduced if legal protection were not required over the next ten years. Presently, localized habitat destruction has been noted recently from certain military training activities (tank exercises) in skink habitat, although such activities appear to be rare. Negative effects also result from bomb craters, vehicles on roadsides, and hole which is frequently associated with military training activities on the base. There are also positive effects from opening of habitat and maintaining native grassland on the base.

Collection is not perceived to be a significant threat to this species in Manitoba but does occasionally occur; campers in SWPP have been observed collecting Prairie Skinks (Scott and Rutherford pers. obs. in Manitoba Prairie Skink Recovery Team 2014). Domestic cats can add mortality pressure to skink populations. Krause Danielsen et al. (2014) interviewed rural landowners who had skink populations on their property and found that 50% of landowners who owned cats had seen evidence of predation on Prairie Skinks.

Potential threats (unknown impacts)

The effects of pesticides and fungicides applied to crops (e.g., potatoes) on Prairie Skinks are unknown. Prairie Skinks cannot survive in potato fields, but skinks living in adjacent areas could be exposed to these applications.

Climate change is predicted to affect the Prairie region of Manitoba with increased temperatures and drought, more extreme weather events (heavy rains and longer and more frequent droughts), milder and shorter winters, and significantly less snowfall (Lemmen and Warren 2004). Winter mortality has been suggested as a limiting factor for Prairie Skink populations (Bredin 1989). Shorter winters may be beneficial for Prairie Skinks, but below-surface temperatures might fluctuate more dramatically with less snow cover. Winter temperature fluctuations could be hazardous for Prairie Skinks because warmer conditions would increase metabolic rates and could deplete energy reserves more rapidly; exposure to sub-zero temperatures can kill individual skinks. Threats posed by climate change were assessed as unknown.

Number of locations

Ecosystem modification from invasive plants, Leafy Spurge and Smooth Brome in particular, is considered to be the major plausible threat to Prairie Skinks in Canada. Leafy Spurge is pervasive throughout Prairie Skink’s range in Canada but is still invading new sites and spreading at sites where it is already present. However, the impact of Leafy Spurge on Prairie Skinks is still unclear. The density of Leafy Spurge patches and habitat context (e.g., the thermal environments of adjacent habitat patches) may influence whether Prairie Skinks and Leafy Spurge can co-exist. Another important threat affecting the species’ entire range over the long-term is succession of mixed-grass prairie habitats to Aspen-dominated ecosystems in the absence of fire or other equivalent disturbances. While both threats (invasive plants and succession) are range-wide, exposure rates and impacts are likely to be variable due to site-specific conditions and management practices. Prairie Skinks are also exposed to other, site-specific threats from various sources throughout their range.

Prairie Skink occurs on approximately 26 land parcels, including CFB Shilo, SWPP, three wildlife management areas, one nature preserve, and approximately 20 other land parcels (crown land or privately owned). Threats from invasive plants and succession do not follow landownership but operate at a broader scale, which would result in less than 26 threat-based locations. The exact number cannot be calculated with any certainty because of uncertainties in spread of invasive plants and succession and management practices on the privately owned land parcels in particular. Furthermore, It is uncertain whether these processes occur sufficiently rapidly to qualify contributing to threat-based locations.

Protection, status and ranks

Legal protection and status

Prairie Skink was listed as Endangered in Schedule 1 of the Species at Risk Act (SARA) in 2005. Under this Act individuals and their residence are protected from harm. Additionally, the Minister of the Environment is responsible to have a recovery strategy prepared within one year for species listed as Endangered.

A proposed Recovery Strategy (Environment and Climate Change Canada 2016) has been posted for national-level planning. Provincially, a draft Manitoba Provincial Action Plan and Recovery Strategy have been prepared (Manitoba Prairie Skink Recovery Team 2014).

Efforts to protect Prairie Skink habitat are underway. The proposed Recovery Strategy for the species includes Critical Habitat description, defined as the suitable habitat within a 100 m radius of occupied sites, for 569 sites (Environment and Climate Change Canada 2016). This includes 116 polygons in the Brandon Sandhills and one polygon in the Lauder Sandhills, which cover areas of 5.13 km² and 0.08 km², respectively (Environment and Climate Change Canada 2016).

Several actions to meet conservation objectives have been completed or are underway (Environment and Climate Change Canada 2016).

Monitoring and assessment

Surveys have identified several new sites, and documented sites have been mapped by MB CDC. Surveys at CFB Shilo and Nature Conservancy of Canada are designed to monitor the persistence of skinks at selected study areas.

Habitat assessment, management, conservation and protection

Prescribed burns are conducted to maintain prairie habitat at designated sites in SWPP, CFB Shilo, and areas managed by Nature Conservancy of Canada. Sandhill habitat has been secured in the Lauder Sandhills area. Nature Conservancy of Canada lands are actively managed for the protection of Prairie Skinks.

Research

Microhabitat selection has been examined within native mixed-grass prairie and rural residential developments. Examination of impact of invasive plants on skink populations is in progress.

Communication, collaboration and engagement

Community outreach has been conducted by SWPP, researchers, and through a website (Nature North 2016b).

Prairie Skink is listed as Endangered in Manitoba under the Endangered Species and Ecosystems Act (Manitoba Wildlife Branch 2016). Under this act it is unlawful to kill, injure, possess, disturb or interfere with the species; destroy, disturb or interfere with the habitat of the species; damage, destroy, obstruct or remove a natural resource on which the species depends for its life and propagation.

Non-legal status and ranks

Prairie Skink’s IUCN Status is Least Concern (IUCN 2015), and its Global Heritage Status Rank is G5 (secure; NatureServe 2017). In the United States, the species is also considered to be secure (National Status of N5); in Canada Prairie Skink is critically imperilled (N1; NatureServe 2017). NatureServe (2017) ranks the Northern Prairie Skink subspecies as G5 and N5 in United States (status last reviewed in 1996) and N1 in Manitoba (status last reviewed in 2016). COSEWIC (2004) assessed Prairie Skink as Endangered. Sub-nationally, Prairie Skink (entire species) has the following ranks (NatureServe 2017): Manitoba (S1), Arkansas (S2), Iowa (S3-vulnerable), Kansas (S4-apparently secure), Louisiana (S1), Minnesota (S5), Missouri (SNR-unranked), Nebraska (S5), North Dakota (S2S3), Oklahoma (S4), South Dakota (S5), Texas (S5), and Wisconsin (S3).

Habitat protection and ownership

A significant portion of suitable Prairie Skink habitat in the Brandon Sandhills occurs on protected lands (CFB Shilo: 34,650 ha, 28%, and SWPP: 16,800 ha, 13%), while the remaining 59% of suitable habitat is primarily on private lands (73,500 ha) (Rutherford unpubl. data 2014). SWPP covers 268 km² and is protected from development, but threats such as aspen encroachment, invasive plant species, and collection occur in this area. CFB Shilo has restricted access, which protects the skinks from many threats, but vehicles and military exercises could impact populations. The Assiniboine Corridor Wildlife Management Area and Whitemud Watershed Wildlife Management Area consist of fragmented parcels (Figure 6), but some Prairie Skink occurrences are present in these areas. A large block of mixed-grass prairie habitat is protected by Nature Conservancy Canada. Prairie Skinks are not present in any national parks.

The Lauder Sandhills Prairie Skink population is small (~1.6 km in diameter) and isolated (~80 km from the Brandon Sandhills population) and occurs entirely within the Lauder Sandhills Wildlife Management Area. The Lauder Sandhills Wildlife Management Area covers an area of ~31 km² and was originally established to protect White-tailed Deer (Odocoileus virginianus) winter habitat (Province of Manitoba 2015d). Designated vehicle routes have been established to minimize the impact of vehicles (Province of Manitoba 2015d), but recreational use by ATVs is common.

It is important to note that although Prairie Skinks occur in protected areas, this does not mean that the habitat is maintained for skinks in all areas. As noted under Habitat Trends, forest encroachment is a serious threat at SWPP and the surrounding land. However, prescribed burns are conducted at SWPP at sites designated by the Prairie Management Plan, at CFB Shilo, and at sites managed by the Nature Conservancy of Canada. The Nature Conservancy of Canada applies a multi-species at risk management and recovery planning framework, and Prairie Skink is included as a focal species (Environment and Climate Change Canada 2016).

Community outreach work to engage the public in Prairie Skink’s conservation occurs through a “Save our Skinks” website and Skinkfest celebrations at SWPP (Nature North 2016b). Researchers have also worked directly with landowners to protect habitat (Krause Danielsen 2012).

Acknowledgements and authorities contacted

Errol Bredin wrote the original version of the COSEWIC report on Prairie Skink in 1989. David Seburn, Carolyn Seburn, and Errol Bredin updated the report in 2004. A draft Status Appraisal Summary was prepared for Prairie Skink by Pamela Rutherford and Drew Hoysak in 2014, providing information for this report. Jenny Wu of the COSEWIC Secretariat provided assistance with the species’ extent of occurrence (EOO) and index of area of occupancy (IAO) calculations and mapping. Members of the Amphibians & Reptiles Specialist Subcommittee provided helpful revisions and review comments. Andrew Didiuk and Pamela Rutherford generously shared their unpublished data and results of work in progress to facilitate this assessment.

Authorities contacted

Doug Collicutt, Biologist, Manitoba Herp Atlas, Winnipeg, MB.

Andrew Didiuk, Wildlife Biologist, Canadian Wildlife Service – Prairie and Northern Region, Saskatoon, SK.

Chris Friesen, Biodiversity Information Manager, Manitoba Conservation Data Centre, Winnipeg, MB.

Neil Jones, Scientific Project Officer and ATK Coordinator, COSEWIC Secretariat, Canadian Wildlife Service, Environment Canada, Gatineau, QC.

Allison Krause Danielsen, Regional Wildlife Biologist, Manitoba Conservation, Brandon, MB.

Colin Murray, Project Biologist and Geomatics, Manitoba Conservation Data Centre, Winnipeg, MB.

Rebekah Neufeld, Assistant Conservation Biologist, Nature Conservancy Canada, Brandon, MB.

Sherry Punak-Murphy, Base Biologist, Canadian Forces Base Shilo, Shilo, MB.

Pamela Rutherford, Associate Professor, Brandon University, Brandon, MB.

Jenny Wu, Scientific Project Officer, COSEWIC Secretariat, Canadian Wildlife Service, Environment Canada, Gatineau, QC.

Information sources

Agriculture and Agri-Food Canada. 2015. Potato Market Information Review 2013 to 2014 (PDF). Web site [accessed September 2017].

Belcher, J.W., and S.D. Wilson. 1989. Leafy Spurge and the species composition of a mixed-grass prairie. Journal of Range Management 42:172-175.

Bird, R.D. 1961. Ecology of the Aspen Parkland of western Canada in relation to land use. Canadian Department of Agriculture, Research Branch, Publication 1066:1-155.

Brandley, M.C., A. Schmitz, and T.W. Reeder. 2005. Partitioned Bayesian analyses, partition choice, and the phylogenetic relationships of Scincid lizards. Systematic Biology 54:373-390.

Breckenridge, W.J. 1943. The life history of the black-banded skink Eumeces septentrionalis sepentrionalis (Baird). American Midland Naturalist 29:591-606.

Bredin, E.J. 1981. Distribution of the northern prairie skink in Manitoba. Manitoba Department of Natural Resources, Biological Services MS Report No. 81-17.iv + 38 pp.

Bredin, E.J. 1988. Additions to the life history of the northern prairie skink. Unpubl. report to World Wildlife Fund, Wild West Program, Project #W74.

Bredin, E.J. 1989. COSEWIC status report on the northern prairie skink, Eumeces septentrionalis, in Canada. Committee on the Status of Endangered Wildlife in Canada. Ottawa 1-48 pp.

Bredin, E.J. 1993. The Carberry Sandhills of southwestern Manitoba: the need for active management. pp. 300-306, in G.L. Holroyd, H.L. Dickson, M. Regnier, and H.C. Smith (eds.). Proceedings of the Third Prairie Conservation and Endangered Species Workshop. Natural History Occasional Paper Number 19. Provincial Museum of Alberta, Edmonton, Alberta.

Bredin, E.J. 1999. Field monitoring results: reptiles and amphibians CFB Shilo training ranges, 1998 to 1999. Unpubl. report to CFB Shilo, Shilo, Manitoba.

Britton, N.L. 1921. The Leafy Spurge becoming a pest. Journal of the New York Botanical Gardens 22:73-75.

Cairns, N.A. 2007. Habitat suitability model for the Northern Prairie Skink (Eumeces s. septentrionalis) in Manitoba. Unpubl. report to Manitoba Conservation. 19 pp.

Chhin, S., and G.G. Wang. 2002. Spatial and temporal pattern of white spruce regeneration within mixed-grass prairie in the Spruce Woods Provincial Park of Manitoba. Journal of Biogeography 29:903-912.

Chhin, S., G.G. Wang, and J. Tardif. 2004. Dendroclimatic analysis of white spruce at its southern limit of distribution in the Spruce Woods Provincial Park, Manitoba, Canada. Tree-Ring Research 60:31-43.

Collins, J.T., and T.W. Taggart. 2009. Standard common and current scientific names for North American amphibians, turtles, reptiles and crocodilians. Sixth Edition. Center for North American Herpetology. iv + 44 pp.

Cook, F.R. 1964. The northern Prairie Skink in Manitoba: a possible relict population. National Museum of Canada Natural History Papers 24:1-11.

COSEWIC. 2004. COSEWIC assessment and update status report on the Prairie Skink Eumeces septentrionalis in Canada. Committee on the Status of Endangered Wildlife in Canada. Ottawa. vi + 22 pp.

Crother, B.I. (ed.). 2012. Scientific and standard English names of amphibians and reptiles of North America North of Mexico, with comments regarding confidence in our understanding. 7th Edition. Herpetological Circular No. 39. Society for the Study of Amphibians and Reptiles, Salt Lake City, Utah. iv + 92 pp.

Didiuk, A.B., pers. comm. 2016. Email correspondence to C. Browne. February, March, and September 2016. Wildlife Biologist, Canadian Wildlife Service – Prairie and Northern Region, Saskatoon, Saskatchewan.

Didiuk, A.B., and C. Murray, unpubl. data. 2016. Unpubl. data provided by A. Didiuk to Ruben Boles and Kristiina Ovaska. April 2017. Wildlife Biologist, Canadian Wildlife Service – Prairie and Northern Region, Saskatoon, Saskatchewan.

Didiuk, A.B., pers. comm. 2017. Unpubl. data provided during Amphibians & Reptiles Specialist Subcommittee meeting in Regina, Saskatchewan. September 2017. Wildlife Biologist, Canadian Wildlife Service – Prairie and Northern Region, Saskatoon, Saskatchewan.

Ecological Stratification Working Group. 1995. A national ecological framework for Canada. Agriculture and Agri-Food Canada, Research Branch, Centre for Land and Biological Resources Research and Environment Canada, State of the Environment Directorate, Ecozone Analysis Branch, Ottawa/Hull. Report and national map at 1:7,500,000 scale.

Environment and Climate Change Canada. 2016. Recovery strategy for the Prairie Skink (Plestiodon septentrionalis) in Canada [Proposed]. Species at Risk Act Recovery Strategy Series. Environment and Climate Change Canada, Ottawa. v + 38 pp.

Fitch, H.S. 1954. Life history and ecology of the five-lined skink, Eumeces fasciatus. University of Kansas Publications, Museum of Natural History 8(1):1-156.

Fuerst, G.S., and C. Austin. 2004. Population genetic structure of the Prairie Skink (Eumeces septentrionalis): Nested clade analysis of post Pleistocene populations. Journal of Herpetology 38:257-268.

Griffith, H., A. Ngo, and R.W. Murphy. 2000. A cladistic evaluation of the cosmopolitan

Genus Eumeces Wiegmann (Reptilia, Squamata, Scinidae). Russian Journal of Herpetology 7:1-16.

Hogg, E.H., J.P. Brandt, and B. Kochtubajda. 2002. Growth and dieback of aspen forests in northwestern Alberta, Canada, in relation to climate and insects. Canadian Journal of Forest Research 32:823-832.

IUCN. 2015. The IUCN Red List of Threatened Species. Version 2015.2. Web site:[accessed August 2015].

Krause Danielsen, A. 2012. Using landowner knowledge and field captures to determine habitat use by the northern Prairie Skink (Plestiodon septentrionalis) on exurban land in southwestern Manitoba. M.N.R.M. thesis, University of Manitoba, Winnipeg, Manitoba. 137 pp.

Krause Danielsen, A., P. Rutherford, and N. Koper. 2014. The importance of vegetation structure and artificial cover for Prairie Skinks (Plestiodon septentrionalis) on exurban land. Journal of Herpetology 48:67-73.

Larkin, J. 2011. Microhabitat preferences of the Prairie Skink (Plestiodon septentrionalis) in southwestern Manitoba. Master of Natural Resources Management thesis, University of Manitoba, Winnipeg, Manitoba. 77 pp.

Lemmen, D.S., and F.J. Warren. 2004. Climate Change Impacts and Adaptation: A Canadian Perspective. Natural Resources Canada report, Ottawa, Ontario. 201 pp.

Manitoba Conservation and Water Stewardship. 2012. Spruce Woods Provincial Park Management Plan. Report prepared for Spruce Woods Provincial Park by Manitoba Conservation and Water Stewardship, Parks and Natural Areas Branch, MB. 28 pp.

Manitoba Prairie Skink Recovery Team. 2014. Manitoba’s provincial conservation and recovery strategy for the Northern Prairie Skink (Plestiodon septentrionalis) (draft). 19 pp.

MB CDC (Manitoba Conservation Data Centre). 2017. data. 2015. Data provided to C. Browne.

Manitoba Wildlife Branch. 2016. Species At Risk. Web site: [accessed December 2016].

Mansell, T., and J. Moore. 1999. Mixed-grass prairie inventory of Manitoba – Interim status report. Manitoba Natural Resources, Winnipeg, Manitoba.

McFadden, W.C.J. 2010. The Assiniboine River separates two genetically distinct clusters of the Northern Prairie Skink (Plestiodon septentrionalis). B.Sc. Honours thesis, Department of Biology, Brandon University, Brandon, Manitoba. 28 pp.

McKernan, J.M. 1984. Effects of military training on mixed-grass prairie at Shilo, Manitoba, Canada, and utility of remedial seeding measures. M.Sc. thesis, University of Manitoba, Winnipeg, Manitoba. 123 pp.

Nature North. 2016a. The Manitoba Herps Atlas. Web site: [accessed March 2016].

Nature North. 2016b. Save Our Skinks! Web site: [accessed February 2016].

NatureServe. 2017. NatureServe Explorer: An online encyclopedia of life [web application]. Version 7.1. NatureServe, Arlington, Virginia, USA: NatureServe. Web site: [accessed July 2017].

Nelson, W.F. 1963. Natural history of the northern prairie skink, Eumeces septentrionalis septentrionalis (Baird). Ph.D. dissertation, University of Minnesota, Minneapolis, Minnesota, USA. 134 pp.

Oliver, G., pers. comm. 2003. Correspondence to D. Seburn, C. Seburn, E. Bredin cited in COSEWIC (2004). Co-ordinator, Mixed-Grass Prairie Stewardship Program, Box 508, Carberry, Manitoba.

Oliver, G. 2008. Introduction. pp. 9-13. in R. Moss, B. Gardiner, A. Bailey and G. Oliver (eds.). A Guide to Integrated Brush Management on the Western Canadian Plains, Manitoba Forage Council, Brandon, Manitoba.

Pitt, W.C. 2001. Density of prairie skinks (Eumeces septentrionalis) in old-field habitats. American Midland Naturalist 146:86-93.

Prairie Skink Recovery Team. 2007. Draft national action plan for the Prairie Skink (Plestiodon septentrionalis). Winnipeg, Manitoba. 12 pp.

Province of Manitoba. 2015a. Integrated management of Leafy Spurge. Web site: [accessed January 2016].

Province of Manitoba. 2015b. Prairie management at Spruce Woods Provincial Park. Web site: [accessed January 2016].

Province of Manitoba. 2015c. Critical Wildlife Habitat Program. Web site: [accessed January 2016].

Province of Manitoba. 2015d. Wildlife Management Areas Western Region. Web site: [accessed January 2016].

Rutherford, P. 2010. Impact of habitat modification on Prairie Skinks. 2009 Sustainable Development Innovation Fund Final Report to Manitoba Sustainable Development. 30 pp.

Rutherford, P. 2013. Species at Risk Surveys at CFB Shilo. Unpubl. report prepared for CFB Shilo, Shilo, Manitoba. 10 pp.

Rutherford, P. 2015. Identifying Critical Habitat of Northern Prairie Skink in Manitoba. Permit Number: SARA-PNR-2012-0187. Report Addendum May 13, 2015. 17pp.

Rutherford, P., pers. comm. 2016. Phone communications during threats assessment conference call. December 2016. Associate Professor, Brandon University, Brandon, Manitoba.

Rutherford, P., pers. comm. 2017a. Email correspondence to K. Ovaska. April 2017. Associate Professor, Brandon University, Brandon, Manitoba.

Rutherford, P., pers. comm. 2017b. Email correspondence to C. Browne. January and February 2017. Associate Professor, Brandon University, Brandon, Manitoba.

Rutherford, P., pers. comm. 2017c. Information provided during Amphibians & Reptiles Specialist Subcommittee meeting in Regina, Saskatchewan. September 2017. Associate Professor, Brandon University, Brandon, Manitoba.

Rutherford, P., unpubl. data. 2014. Data provided to C. Browne. January and February 2017. Associate Professor, Brandon University, Brandon, Manitoba.

Rutherford, P.L., D.J. Wiseman, and N.A. Cairns. 2012. Shifts in land use in mixed-grass prairie in Southwestern Manitoba: Impacts on the Endangered Northern Prairie Skink. North American Prairie Conference, Winnipeg, Manitoba.

Rutherford, P.L., D.J. Hoysak, and S. Pratt. Microhabitat selection of the Northern Prairie Skink (Plestiodon septentrionalis). Manuscript in preparation (as of February 2017).

Samson, F., and F. Knopf. 1994. Prairie conservation in North America. BioScience 44:418-421.

Schmitz, A., P. Mausfeld, and D. Embert. 2004. Molecular studies on the genus Eumeces Wiegmann, 1834: phylogenetic relationships and taxonomic implications. Hamadryad-Madras 28:73-89.

Scott, J.L. 2005. Activity and habitat use of prairie skinks, Eumeces septentrionalis, in Manitoba. Master of Environment thesis. University of Manitoba, Winnipeg, Manitoba. 155 pp.

Scott, T.G., and R.B. Sheldahl. 1937. Black-banded skink in Iowa. Copeia 1937:192.

Seburn, C.N.L. 1993. Spatial distribution and microhabitat use in the five-lined skink (Eumeces fasciatus). Canadian Journal of Zoology 71:445-450.

Shay, J., D. Kunec, and B. Dyck. 2001. Short-term effects of fire frequency on vegetation composition and biomass in mixed prairie in south-western Manitoba. Plant Ecology 155:157-167.

Siu, J. 2011. Evaluating the population structure and conservation in the northern range of the Prairie Skink, Plestiodon septentrionalis. B.Sc. Honours thesis, Department of Biology, Queen’s University, Kingston, Ontario. 46pp.

Smith, H.M. 2005. Plestiodon: a replacement name for most members of the genus Eumeces in North America. Journal of Kansas Herpetology 14:15-16.

Somma, L.A. 1987a. Reproduction of the prairie skink, Eumeces septentrionalis, in Nebraska. Great Basin Naturalist 47:373-374.

Somma, L.A. 1989. Influence of substrate water content on neonate size in the prairie skink, Eumeces septentrionalis. Great Basin Naturalist 49:198-200.

Somma, L.A. 1990. Observations on the nesting ecology of the prairie skink (Eumeces septentrionalis) in Nebraska. Bulletin of the Chicago Herpetological Society 25:77-80.

Somma, L.A., and P.A. Cochran. 1989. Bibliographic and subject index of the prairie skink, Eumeces septentrionalis (Baird)(Sauria: Scincidae). Great Basin Naturalist 49:525-534.

Somma, L.A., and J.D. Fawcett. 1989. Brooding behaviour of the prairie skink, Eumeces septentrionalis, and its relationship to the hydric environment of the nest. Zoological Journal of the Linnean Society 95:245-256.

Statistics Canada. 2012. Canadian Potato Production. Service bulletin, Catalogue no. 22-008-X, Vol. 10, no. 2, 14 pp.

Team Leafy Spurge. 2005. Team Leafy Spurge, developing IPM solutions for land managers. Web site: [accessed February 2016].

Tomaino, A., J. Cordeiro, L. Oliver, and J. Nichols. 2008. Key for ranking species element occurrences using the generic approach. Web site: [accessed December 2015]

Town of Carberry. 2016. The Potato Era. Web site: [accessed February 2016; archived].

Biographical summary of report writers

Constance Browne is a freelance biologist and a Research Associate at the New Brunswick Museum. She specializes in conservation biology, population biology, landscape ecology, habitat/resource selection, spatial ecology, and herpetology. She has an M.Sc. from Lakehead University where she investigated the status of turtle populations in Point Pelee National Park, and a Ph.D. from the University of Alberta where she studied habitat use of the Western Toad in north-central Alberta and the influence of scale. She has published scientific journal articles, reports, and other work on turtles, snakes, anurans, and other species.

Collections examined

No specimens were examined, but the following databases were queried for distributional records:

Manitoba Conservation Data Centre (MB CDC). 2017. Prairie Skink (Plestiodon septentrionalis), distribution records up to the year 2015 for Manitoba. Received on 19 January 2017.

Manitoba Herp Atlas. Prairie Skink (Plestiodon septentrionalis), distribution records up to the year 2014 for Manitoba. Available at: MHA Interactive Database (Public Version). Accessed on 22 October 2015.

Prairie Skink Recovery Team. Prairie Skink (Plestiodon septentrionalis), distribution records up to the year 2014 for Canada. Received on 22 October 2015.

Appendix 1

The extent of occurrence (EOO; top panel) and index of area of occupancy (IAO; 2x2 km grid cells; bottom panel) for Prairie Skink from 2001 - 2015 (current) and pre-2001 (historical). Sites presumed extirpated are indicated. Maps and calculations by Jenny Wu (COSEWIC Secretariat)

Appendix 2

Summary of element occurrences (EO) and their estimated viability according to Manitoba Conservation Data Centre files. Tomaino et al. (2008) ranking criteria were used to estimate viability

Note. Tomaino et al. (2008) ranking criteria were used to estimate viability.

Appendix 3. IUCN threats calculator results for Prairie Skink completed by a panel of experts on 1 September 2016

Species or ecosystem scientific name

Plestiodon septentrionalis

Element ID

N/A

Elcode

N/A

Date

9/1/2016

Assessor(s)

Connie Browne (status report writer), Briar Howes (Amphibians and Reptiles SSC), Kristiina Ovaska (Facilitator and status report Co-chair), Pamela Rutherford (recovery team), Andrew Didiuk (Canadian Wildlife Service), Cynthia Pazskowski (Amphibians and Reptiles SSC)

References

COSEWIC status report, draft August 2016; Recovery Strategy 2016

Assigned overall threat impact

CD = Medium - Low

Impact adjustment reasons

N/A

Overall threat comments

Generation time 3-5 years (3 generations 9 - 15 years)