Saskatchewan is one of the Prairie Provinces of western Canada (Fig. 1) with a total area of 651,036 km² of which 591,670 km² is land, while 59,366 km² is water (Anonymous 2008). GPS coordinates of the Province are as follows: 52.9399° N, 106.4509° W. From north to south, there are four ecological zones or biomes (Taiga Shield, Boreal Shield, Boreal Plain and Prairie) which are further subdivided into 11 ecoregions (Acton et al. 1998). The Taiga Shield ecozone stretches across part of Canada’s subarctic. This ecozone is dominated by the boreal coniferous forest. The Boreal Shield ecozone is widely forested with black spruce forming mixed stands with jack pine and tamarack. This ecozone separates the warmer Boreal Plain to the south from the colder Taiga Shield to the north. The Boreal Plain ecozone is characterised by mixwood and coniferous forest. The deciduous component is represented along the southern boundary adjoining the prairie grasslands. The Prairie ecozone is part of the Interior Plains of Canada, which are a northern extension of the Great Plains of North America.

Figure 1.  

Location of Saskatchewan depicting four ecozones or biomes in the Province with reference to Canada (Source: https://www.friendlyforest.ca/Page_files/Sask_Eco_Regions_Map.htm).

Saskatchewan has a continental climate, experiencing extremes in temperature and weather events, relatively low precipitation which falls mostly during summer months and considerable sunshine. Temperatures can range from -40°C in the winter to +35°C during summer. There is a significant difference in mean temperature and precipitation between the southwest and northeast parts of the Province. Generally, mean annual temperatures drop from south to north and from west to east within the Province. Conversely, precipitation levels ordinarily increase from south to north. Overall, there are warmer, drier conditions in the southwest and cooler moister conditions in the northeast of Saskatchewan.

The process of validation entailed taxonomic verification of Carex species against the Plants of the World Online (P.O.W.O. 2024) and the Database of Vascular Plants of Canada or VASCAN (Brouillet et al. 2010). The data filtering of the Carex specimen-based occurrences was conducted to eliminate multiple entries by cross-checking the Flora of Saskatchewan Association (FOSA 2022) dataset that contains 2207 records with the Global Biodiversity Information Facility (GBIF 2020) datasets that includes 940 records. To deal with the bias and mistakes in the occurrence data, several processes to manipulate the raw data were conducted. First, records that had either no geographical coordinates or unclear taxonomic information were excluded. Second, records that have duplicated coordinates were removed. Each validated species includes georeferenced points with an attribute table consisting of the sampling locations linked to the FOSA and GBIF datasets. In total, 2655 records of 105 species were validated for this study, 2084 records from FOSA and 571 records from GBIF (Fig. 2). It should be noted that all produced maps represent the record count (herbarium vouchers) and, as such, are a function of collecting bias. These maps may not be necessarily representative of the true Carex species distribution in the Province.

Figure 2.  

Occurrences of Carex species in Saskatchewan representing 2084 specimens in the Flora of Saskatchewan Association (FOSA) dataset and 571 specimens in the Global Biodiversity Information Facility (GBIF) datasets.

Moreover, the study integrated expert ecological knowledge into the validation process. Patterns observed in the interpolated maps were cross-referenced with published literature on Carex species in Saskatchewan. This approach allowed for qualitative validation of the spatial patterns, ensuring that the interpolated richness hotspots aligned with ecologically plausible areas given the environmental gradients across the Province. The integration of expert-driven validation adds a unique layer of confidence to the results, as it combines quantitative spatial analysis with ecological expertise to minimise misinterpretation of the interpolated data.

Environmental datasets are the independent variables of GIS layers developed for Saskatchewan (Kricsfalusy et al. 2015). These include raster and feature layers of topography (elevation), climate (precipitation, temperature and climate moisture index), soil type and ecological systems (ecozone and ecoregion). These variables (Table 1) were converted into point features, classified into ranges and spatially joined with the Carex occurrence point features such that each of the species’ occurrences carries the generated values for each of the environmental parameters. Thus, seven derivatives of validated species’ layers with environmental parameter features’ columns form the parameter for analysis. Each parameter was broken down into specific categories which were analysed using box plot of species occurrences, bar chart of total species count, bar chart of grids, species count map and hotspots map.

Applying GIS spatial and statistical tools, the patterns of the Carex specimen-based occurrences were assessed. A hotspot analysis that emphasises areas of species richness was completed, based on unique species counts per each grid and the number of grid cells occupied by each species. Saskatchewan was divided into 325 UTM grid cells of 50 km x 50 km (Kricsfalusy et al. 2015, Kricsfalusy 2021), over which the studied species were rendered for analysis. For the hotspot rendition, species richness (the unique species counts) of a grid cell provides information for the diversity indices of the cell. Species richness (r), frequency (Fr) and the Shannon-Weiner diversity index (H) were calculated (Begon et al. 1986) for each ecozone and ecoregion.

This study applied the ESRI ArcGIS platform to engage the environmental parameters’ raster and the Carex species points features in spatial analysis. In this study, the join attribute features refer to the environmental variable values associated with Carex occurrence points, while the target features represent the georeferenced Carex species occurrences with unique species counts and diversity metrics. The completed GIS analysis procedures began with Spatial Join exercises that connect analysis feature layers, based on their spatial relationship. The Join Attribute table includes the Join_ID column that shows the value of joined features attributes and the Target_ID that constitute the primary values for analysis. Two Spatial Join processes were conducted and during each of the sessions, the Join-One-to-One Spatial Join option and the Closest Match option were selected to ensure that each of the species feature points of occurrences assume the various values of the environmental parameters. Statistical analysis engaged the frequency column to produce the frequency for each unique combination of the specified attribute field before the Inverse Distance Weighting (IDW) of the geostatistical tool was used to produce interpolated surfaces of the unique species distribution pattern as in the case of diversity hotspot analysis.

Boxplots were used to compare the Carex specimen-based occurrences across categories of environmental variables, while bar graphs illustrated species richness and the number of grids within each variable category. Data visualisation for hotspots analysis is enhanced by the Spatial Join of validated species points derivatives with grid cells and are rendered using a spectrum of colour to indicate the pattern of the species spatial distribution. A more dense colour indicates a higher species count in grid cells which are linked to favourable conditions for Carex species. The interpolated surfaces used the colour gradient ranges from blue (low species richness) to red (high species richness), delineating areas with varying levels of diversity.

The obtained statistics data are presented in Suppl. material 1. A series of charts was developed to illustrate the distribution of the Carex specimen-based occurrences within each of the grid cells (Fig. 3a) with five ranges of occurrences and the unique species counts within the grid cells (Fig. 3b). The study has revealed that there are cells with a species record as high as 82 and species richness (the total number of species) as high as 44. The study has shown that Carex species occur in 253 grid cells (77%) of the total 325 grid cells of the Province.

Figure 3.  

Spatial distribution of the Carex specimen-based occurrences in Saskatchewan: a species occurrences with maximum of 82 collections within a grid cell; b species richness with maximum of 44 species within a grid cell; c interpolated species hotspots.

The Carex hotspots that indicate the areas of Saskatchewan with the highest species count are delineated not by the frequency of occurrences, but by the species richness within a given grid cell. The analysis of Carex species counts yielded five classes which were obtained using Jenks Natural Breaks classification method (Dagher-Kharrat et al. 2018). These classes represent the different levels of species cluster rate in each grid cell (Fig. 3b). The classes range from 1-6 counts (the lowest range that renders species sparseness) to 33-44 counts (the highest range of species hotspots). The blank grid cells indicate where there are no species occurrence records. The hotspot analysis considered the species ranges of 22-32 and 33-44 to provide information for areas with hotspots of Carex species that have been classified into three ranges. The interpolated surface (Fig. 3c) highlights areas with varying levels of diversity.

Elevation (ELEV). The box plot (Fig. 4a) highlights the variation in the Carex specimen-based occurrences across different elevations, with the middle elevation range (456-640 m) showing the highest median occurrence and the widest spread, suggesting it supports a diverse group of species. The low (206-455 m) and high (641-1386 m) elevation ranges exhibit more variability in the species occurrences, with some of them being highly prevalent at lower elevations. In the case of species count (Fig. 4b), the highest number of species (93) is observed in the 456-640 m range, followed by 206-455 m (87 species) and the lowest number in the 641-1386 m range (68 species). The middle elevation range (456-640 m) has five species in hotspots, while the low elevation range (206-455 m) has three species in hotspots and the high elevation range (641-1386 m) has no species in hotspots. Grids evaluation shows (Fig. 4c) their highest number (91 grids) in the middle elevation range (456-640 m) where eight grids are in hotspots. The low elevation range (206-455 m) has 84 grids with three grids in hotspots and none in the high elevation range (641-1386 m). The grid map (Fig. 4d) shows a species count in each grid cell which are linked to favourable elevation for Carex species. The interpolated diversity hotspot map (Fig. 4e) shows that the central part of Saskatchewan possesses the highest species number, which likely corresponds to the middle elevation range (456-640 m). Thus, all evaluations collectively suggest that Carex species are mostly distributed in the range of 456-640 m.

Figure 4.  

Relationship between Carex species distribution and elevation: a box plot of occurrences; b bar chart of total species count including hotspots and without hotspots; c bar chart of total number of grids including hotspots and without hotspots; d unique species count per grid; e interpolated species hotspots.

Temperature (MAT). The box plot (Fig. 5a) shows the variability in Carex species occurrences across different temperature ranges, with high temperatures (11-46°C) showing the widest species spread (1 to 60), suggesting this range supports a diverse group of sedges. For the medium temperature range (2-10°C), species occurrences vary from 1 to 72. For some species being highly prevalent under colder conditions as for low temperatures (<=1°C), the minimum occurrence is one, the maximum is 48. In the case of species count (Fig. 5b), the highest species number is observed in the medium temperature range (2-10°C) with 88 species, followed by the high temperature range (11-46°C) with 71 species and the lowest in the low temperature range (<=1°C) with 54 species. The medium temperature range (2-10°C) has three species in hotspots, while the high temperature range (11-46°C) has one species in hotspots and the low temperature range (<=1°C) has no species in hotspots. Grids evaluation (Fig. 5c) shows the highest number of grids present in the medium temperature range (2-10°C) with 269 grids, where six grids are in hotspots. The high temperature range (11-46°C) has 88 grids with four grids in hotspots and the low temperature range (<=1°C) has 40 grids with none in hotspots. The species count map (Fig. 5d) shows the number of species in each grid cell, indicating the highest species number linked to favourable temperatures for Carex species. The interpolated diversity hotspot map (Fig. 5e) reveals that the central part of the study area has the highest species number, likely corresponding to the medium temperature range (2-10°C). All evaluations collectively suggest that Carex species are mostly distributed in the medium temperature range (2-10°C).

Figure 5.  

Relationship between Carex species distribution and temperature: a box plot of occurrences; b bar chart of total species count including hotspots and without hotspots; c bar chart of total number of grids including hotspots and without hotspots; d) unique species count per grid; e interpolated species hotspots.

Precipitation (MAP). The box plot (Fig. 6a) presents the variability in Carex species occurrences across different precipitation ranges, with medium precipitation (397-461 mm) showing the highest median occurrence and the broadest spread, suggesting this range supports a diverse group of species. Low (282-396 mm) and high (463-547 mm) precipitation ranges exhibit more variability in species occurrences, with some species being highly prevalent under these conditions. In the case of species count (Fig. 6b), the highest species number is observed in the medium precipitation range (397-461 mm) with 93 species, followed by the low precipitation range (282-396 mm) with 89 and the high precipitation range (463-547 mm) with 71 species. The medium precipitation range (397-461 mm) has three species in hotspots, followed by the high precipitation range (462-547 mm) with two species and the low precipitation range (282-396 mm) with one species in hotspots. Grid evaluation (Fig. 6c) shows the highest number of grids present in the high precipitation range (462-547 mm) with 166 grids, where three grids are in hotspots. Both the low (282-396 mm) and medium (397-461 mm) precipitation ranges have 81 grids, with three grids in hotspots for the low range and four grids in hotspots for the medium range. The species count map (Fig. 6d) illustrates their number in each grid cell, indicating the highest species counts linked to favourable precipitation conditions for Carex species. The interpolated diversity hotspot map (Fig. 6e) demonstrates that the central part of Saskatchewan, likely corresponding to the high precipitation range (462-547 mm), possesses the highest species number. Therefore, all evaluations collectively suggest that Carex species are uniformly distributed across all precipitation ranges, with a slight preference for the high precipitation range in terms of grid density and species number in hotspots.

Figure 6.  

Relationship between Carex species distribution and precipitation: a box plot of occurrences; b bar chart of total species count including hotspots and without hotspots; c bar chart of total number of grids including hotspots and without hotspots; d unique species count per grid; e interpolated species hotspots.

Climate Moisture Index (CMI). The box plot (Fig. 7a) underscores the variability in Carex species occurrences across different climate moisture index ranges, with the 6-7 category showing the highest median occurrence and the broadest spread, suggesting this range supports a diverse array of species. Lower and higher climate moisture index ranges exhibit more variability in species occurrences, with some species being highly prevalent under these conditions. In the case of species richness (Fig. 7b), the highest species count is observed in the range of 6-7 with 81 species, followed by the range of 4-5 with 80 species, the range of 2-3 with 65 species and the lowest in the range of 8-9 with 63 species. The range of 4-5 has two species in hotspots, followed by the ranges of 2-3, 6-7 and 8-9, each with one species in hotspots. Grid evaluation (Fig. 7c) shows the highest number of grids are present in the range of 6-7 with 46 grids, where two grids are in hotspots. The range of 4-5 has 40 grids with two grids in hotspots, the range of 2-3 has 37 grids with three grids in hotspots and the range of 8-9 has the lowest number of grids, 34, with two grids in hotspots. The species count map (Fig. 7d) illustrates the species number in specific grid cells, indicating higher species counts linked to favourable climatic conditions for the study sedges. The interpolated diversity hotspot map (Fig. 7e) highlights the central part of the Province with the highest species number, likely corresponding to the range of 6-7. Thus, all evaluations collectively suggest that Carex species are mostly distributed in the areas with the climate moisture index range of 6-7 and significantly represented in the range of 4-5 as well.

Figure 7.  

Relationship between Carex species distribution and Climate Moisture Index: a box plot of occurrences; b bar chart of total species count including hotspots and without hotspots; c bar chart of total number of grids including hotspots and without hotspots; d) unique species count per grid; e interpolated species hotspots.

Soils (SOIL). The box plot (Fig. 8a) shows the variability in Carex species occurrences across different soil orders, with the Chernozemic soils showing the highest median occurrence and the broadest spread, suggesting this soil type supports a diverse range of species. Other soil orders exhibit varying degrees of species occurrences, with some showing significant variability and others showing more consistent, lower occurrences. Analysis of species distribution (Fig. 8b) shows that their highest count is observed in the Chernozemic order with 78 species, followed by the Luvisolic order with 75 species. The Brunisolic and Regosolic orders also show a significant species number with 68 and 69 species, respectively. The lowest species counts are observed in the Gleysolic (7 species) and Solonetzic (14 species) orders. The Chernozemic and Luvisolic orders each have two species in hotspots, while the Brunisolic and Organic orders each have one species in hotspots. Grid evaluation (Fig. 8c) reveals the highest number of grids present in the Brunisolic order with 77 grids, where one grid is in a hotspot. The Regosolic order follows with 54 grids, including one grid in a hotspot. The Luvisolic order has 48 grids, with four grids in hotspots and the Organic order has 43 grids, with one grid in a hotspot. The Chernozemic order has fewer grids (13), but a higher species count per grid, with three grids in hotspots. The Gleysolic, Solonetzic and Vertisolic orders have the lowest number of grids, with no hotspots. The species count map (Fig. 8d) illustrates their number in each grid cell, indicating the highest species counts linked to favourable soil conditions for the study sedges. The interpolated diversity hotspot map (Fig. 8e) highlights areas of the highest species number in the central part of Saskatchewan, likely corresponding to Chernozemic and Luvisolic soils. All evaluations collectively suggest that Carex species are mostly distributed in the Chernozemic and Luvisolic orders, with significant representation in the Brunisolic and Regosolic orders as well.

Figure 8.  

Relationship between Carex species distribution and soils: a box plot of occurrences; b bar chart of total species count including hotspots and without hotspots; c bar chart of total number of grids including hotspots and without hotspots; d unique species count per grid; e interpolated species hotspots.

Ecozone and Ecoregion (ECOZ/ECOR). The box plot (Fig. 9a) underscores the variability in Carex species occurrences across different ecoregions, with the Mid-Boreal Upland showing the highest median occurrence and the broadest spread, suggesting this ecoregion supports a diverse range of species. The Churchill River Upland and Aspen Parkland also show significant species occurrences with some variability. Ecoregions, such as the Cypress Upland, Mid-Boreal Lowland, Selwyn Lake Upland and Tazin Lake Upland, exhibit lower species occurrences with less variability. The Boreal Transition (71 species) and Mid-Boreal Upland (70 species) show the highest diversity, indicating these regions offer diverse habitats and favourable ecological conditions. Moderate species counts are observed in the Aspen Parkland (55 species), Moist Mixed Grassland (57 species) and Cypress Upland (42 species), while the lowest counts are in the Athabasca Plain (36 species), Churchill River Upland (28 species), Selwyn Lake Upland (31 species) and Tazin Lake Upland (30 species), likely due to harsher climates or less diverse habitats. The Boreal Plain ecozone has three species in hotspots, the Prairie two, the Boreal Shield one and the Taiga Shield zero (Fig. 9b). Grid evaluation (Fig. 9c) shows the highest number of grids present in the Prairie ecozone with 88 grids, where two grids are in hotspots. The Boreal Plain ecozone follows with 61 grids, including four grids in hotspots. The Boreal Shield ecozone has 47 grids, with one grid in a hotspot and the Taiga Shield ecozone has the lowest number of grids 29, with no grids in hotspots. The species count map (Fig. 9d) indicates the highest species number linked to favourable ecozone conditions for Carex species. The interpolated diversity hotspot map (Fig. 9e) identifies the central part of the Province with the higher species count, likely corresponding to the ecoregions within the Boreal Plain ecozone. This map visualises the trend observed on the other charts and maps, indicating that species distribution is more concentrated in certain ecozones. All evaluations collectively suggest that Carex species are mostly distributed in the Boreal Plain and Prairie ecozones, with significant representation in the Boreal Shield ecozone as well.

Figure 9.  

Relationship between Carex species distribution and major ecosystems: a box plot of occurrences; b bar chart of total species count including hotspots and without hotspots; c bar chart of total number of grids including hotspots and without hotspots; d unique species count per grid; e interpolated species hotspots.

The data on the distribution of the Carex specimen-based occurrences across multiple ecoregions within four distinct ecozones or biomes (Prairie, Boreal Plain, Boreal Shield and Taiga Shield) are summarised in Suppl. material 2. Each ecozone is further divided into specific ecoregions (totalling to 11), providing a detailed picture of Carex species distribution in Saskatchewan.

In the Prairie ecozone, a high number of Carex species is notable, with the Aspen Parkland having 55 species, Moist Mixed Grassland 57, Mixed Grassland 39 and Cypress Upland 42 species. Twenty-five species were noted as present within all ecoregions of this ecozone. These species show a broad ecological tolerance and adaptability to the various habitats. Several species, such as C. brevior, C. douglasii, C. duriuscula, C. gravida, C. meadii, C. petasata, C. raynoldsii, C. saximontana, C. simulata and C. tetanica are present only in the Prairie ecozone, which demonstrates their narrow ecological tolerance. The Moist Mixed Grassland has the highest biodiversity index (4.04), suggesting a diversified and balanced environment with a wide range of species. The Mixed Grassland has the lowest diversity score (3.66), indicating a larger dominance of certain species and fewer species overall. The Aspen Parkland and Cypress Upland have diversity levels of 4.01 and 3.74, respectively. Overall, the Prairie ecozone has high diversity indices, indicating a well-balanced diversity amongst its ecoregions with minor variance.

The Boreal Plain ecozone displays the highest Carex species count, particularly in the Mid-Boreal Upland, which hosts 71 species. The Mid-Boreal Lowland has 40 species and the Boreal Transition has 70 species. A large number of species (33) are present in all three ecoregions, highlighting their adaptability to the diverse conditions. On the other hand, the group of sedges with a limited distribution within this ecozone includes 13 species. Some other species, like C. cristatella, C. leptonervia, C. mackenziei, C. pedunculata, C. projecta and C. sterilis, are unique to the Boreal Plain ecozone. The Boreal Transition and Mid-Boreal Upland have the highest diversity indices of 4.26 and 4.25, respectively. These values suggest that both ecoregions hold a rich and balanced species composition, contributing to a high ecosystem stability and resilience. The Mid-Boreal Lowland, with a diversity index of 3.69, shows lower diversity compared to the other two ecoregions.

The Boreal Shield ecozone shows a moderate Carex species richness, with the Athabasca Plain possessing 36 species and Churchill River Upland having 28 species. Multiple species (14) are commonly found in both the Athabasca Plain and Churchill River Upland. At the same time, C. heleonastes and C. maritima are present only in this ecozone. The Athabasca Plain ecoregion has a slightly higher diversity index (3.58) compared to the Churchill River Upland, which has a diversity index of 3.45. This indicates that the Athabasca Plain has a more balanced species composition.

The Taiga Shield ecozone generally exhibits a lower Carex species count compared to others, with both the Selwyn Lake Upland and Tazin Lake Upland hosting 30 species. Several species (21) are prevalent in both ecoregions, demonstrating their ability to thrive under the harsher conditions of this ecozone. Such species as C. arctogena, C. bicolor, C. bigelowii, C. glacialis and C. supina subsp. spaniocarpa are found only in the Taiga Shield ecozone, known from one or both ecoregions.

There are 11 Carex species with a wide distribution in all four ecozones and at least in nine out of eleven ecoregions in the Province (Suppl. material 2). These are C. aquatilis, C. aurea, C. capillaris, C. diandra, C. disperma, C. foenea, C. inops subsp. heliophila, C. leptalea, C. rostrata, C. siccata and C. utriculata, indicating their high ecological flexibility. Two of these species (C. aquatilis and C. siccata) are widespread across all 11 ecoregions.

There are also 12 Carex species that have a restricted distribution, known only in one ecozone (Suppl. material 2). This group includes C. arcta, C. arctogena, C. bicolor, C. bigelowii, C. cristatella, C. heleonastes, C. leptonervia, C. mackenziei, C. maritima, C. petasata, C. raynoldsii and C. tetanica. These Carex species should be considered for conservation along with other sedges with low occurrence records in the Province.