Introduction

The inclusion of organic soils in Canadian soil taxonomy was first proposed in 1955 by the Subcommittee on Soil Classification of the National Soil Survey Committee (NSSC 1955). A preliminary schema was adopted for field testing in 1960, although it was deemed insufficient and in need of further refinement (NSSC 1960). By 1965, three of the four great groups currently recognized in the Canadian System of Soil Classification (CSSC) (SCWG 1998) were included in the first published approximation, the System of Soil Classification for Canada (CDA 1970). These represented wetland organic soils. The second edition, 4 years later, saw the addition of the Folisol great group, or the upland organic soils (CDA 1974). A detailed description of the evolution of the Organic order of great groups and subgroups is outlined in Kroetsch et al. (2011).

For classification purposes, Organic soils of the Fibrisol, Mesisol, and Humisol great groups have a control section that extends from the surface to a depth of 1.6 m or to a lithic contact, which is divided into three tiers: surface (0–40 cm), middle (40–120 cm), and bottom (120–160 cm; SCWG 1998). The great group classification is based on the dominant material in the middle tier, except when a terric, lithic, or hydric substratum is present within the middle tier, in which case the dominant material in the middle and surface tiers are both considered (SCWG 1998). Organic horizons in wetland organic deposits are designated with the letter O and are further refined with four lowercase suffixes: f (fibric), m (mesic), h (humic), or co (coprogenous). Whereas the first three suffixes refer to the degree of decomposition of the organic material, the fourth suffix, “co”, refers to a specific type of limnic deposit, coprogenous earth, or sedimentary peat, which is formed in an aquatic environment from aquatic organisms and fecal material derived from aquatic animals (SCWG 1998; USDA 2015). Gyttja, the Swedish word for slime, is also used to describe coprogenous earth and was first used as a scientific term by von Post (1862). Other types of limnic deposits are also recognized within the Organic order, including diatomaceous earth and marl. These layers are designated as C and Ck horizons, respectively, and are the only two named layers of Organic soils that are not designated with O, L, F, or H. Coprogenous earth can be either organic (>30% organic matter) or inorganic (≤30% organic matter), while diatomaceous earth and marl are inorganic materials primarily composed of siliceous shells of diatoms and shells of aquatic animals and calcium carbonate precipitated in water, respectively (SCWG 1998), making limnic layers an intriguing exception in the Organic order. It should also be noted that for classification purposes, soils with marl or diatomaceous earth >40 cm thick occurring at the surface, or soils with >40 cm of these materials within the upper 80 cm of the control section, are excluded from the Organic order.

Assessment of the mineralogy of limnic materials, and specifically microscopic observations, are limited (Ismail-Meyer et al. 2018). Nonetheless, micromorphological analyses of gyttja have been conducted in Canada. Aquatic plant or animal residues and lacustrine sponge spicules were observed in eutrophic gyttja by Parent et al. (1980) on thin sections of limnic materials in Quebec. The same authors described calcareous gyttja, which contained, amongst other components, identifiable shells, aragonite, and pyrite grains. Calcite is also commonly found in calcareous gyttja in Poland (Jarnuszewski and Meller 2018). In British Columbia, the micromorphological study of various types of gyttja revealed the presence of plant debris, diatoms, and black pyrite grains (Lévesque et al. 1987; Fox and Tarnocai 1990).

In the CSSC, limnic materials in Organic soils are captured at the subgroup level within the Fibrisol, Mesisol, and Humisol great groups. As an example, the Limnic Fibrisol subgroup is defined as:

In addition, limnic material may be present in other subgroups, namely the Terric and Hydric subgroups. Limnic materials are not found in the Folisol great group of the Organic order, since these are upland organic soils. At the soil family level, criteria for Organic soils include characteristics of the surface tier, reaction, soil temperature, soil moisture regime, particle-size class of terric layer, limnic material, and depth (SCWG 1998). As such, these soils can be further differentiated to describe the type of limnic material at the soil family level using the terms coprogenous, diatomaceous, or marl. Currently, the recognition of limnic materials is not possible when a soil profile meets the requirements for the Terric subgroup. As an extreme example, Table 1 provides a profile description for a soil that would be classified as a Terric Humisol, with no recognition of the important limnic layer, even though the middle tier consists only of coprogenous earth.

Table 1.

Profile description of site PTBO18_1246, a Terric Humisol dominated by coprogenous earth based on SCWG (1998).

Fundamentally, the recognition of limnic layers in Organic soils in the CSSC is insufficient, especially when dealing with coprogenous materials. This was recognized early in the development of the Organic order by R.E. Smith (Research Branch, Canada Department of Agriculture, Winnipeg, Manitoba) at the Second Meeting of the Eastern Section of the National Soil Survey Committee, but it was acknowledged that it was too early to investigate at that time (CSSC 1971). Two key issues are apparent in the current version of the CSSC (SCWG 1998). First, fibric, mesic, and humic organic materials are recognized as organic horizons (i.e., Of, Om, and Oh) and form the basis for the soil great groups of the Organic order. Despite coprogenous materials being recognized as an organic horizon (i.e., Oco), they are not considered at the great group level. Secondly, although limnic materials are recognized at the subgroup level, they are only captured in deep organic soils where the organic material extends beyond the control section (160 cm). Therefore, in soils with significant limnic deposits and a terric layer within the control section, effectively the limnic materials are ignored. For these reasons, we provide evidence of soils from three study areas in Ontario and Quebec where limnic deposits, primarily coprogenous earth, are not adequately captured by the current CSSC (SCWG 1998), estimate the possible extent of these soils, support these taxonomic considerations with soil management implications, and propose revisions to the classification of Organic soils in the CSSC.

Study areas and approaches

The study includes a field investigation component and a mapping component. For the field investigations, soil profiles from three different regions were collected from existing projects: Middlesex County, Ontario; Peterborough County, Ontario; and the Plain of Montreal, Quebec, which includes Châteauguay, Laprairie, Napierville, and St-Jean counties (Fig. 1). Middlesex County is located in southwestern Ontario and is roughly centered around the city of London, Ontario, located at 42.98°N latitude and 81.25°W longitude (Fig. 1). The organic deposits in this area are sparse and located mainly southeast of London and are described in the Soil Survey of Middlesex County (Hagerty and Kingston 1992). The wetlands in Middlesex County are all classified as swamps dominated by deciduous species (Hagerty and Kingston 1992; National Wetlands Working Group 1997; Ontario Ministry of Natural Resources and Forestry 2019). Peterborough County is located in east central Ontario approximately 140 km northeast of Toronto at 44.46°N latitude, 78.17°W longitude (Fig. 1). Organic soils in the southern portion of the county are distributed mainly as linear depressional features within the drumlinized till plain or in level areas underlain by glaciolacustrine deposits and are described in the Soils of Peterborough County (Gillespie and Acton 1981). Three sites were investigated in Peterborough County, and all three were classified as swamps (National Wetlands Working Group 1997; Ontario Ministry of Natural Resources and Forestry 2019). The Plain of Montreal study area is located approximately 30 km south of the City of Montreal, centered approximately at 45.15°N latitude and 73.62°W longitude (Fig. 1). The organic soils in this area are distributed mainly in five large deposits, which occur across a broad region that transects numerous soil survey reports. These include the Soil Survey of Napierville (Lamontagne et al. 2014), the Soil Survey of Huntingdon and Beauharnois (Mailloux and Godbout 1954), the Soil Survey of St-Jean (Lamontagne et al. 2001), the Soil Survey of Châteauguay (Baril and Mailloux 1950), and the Soil Survey of Laprairie (Lamontagne et al. 2000). The wetlands in this region are classified as treed peatlands (Canards Illimités Canada and Ministère de l'environnement et de la lutte contre les changements climatiques 2019a, 2019b), which corresponds to swamps in the national system (National Wetlands Working Group 1997), but were also described in the soil surveys as basin bogs or shore swamps (Lamontagne et al. 2000, 2001, 2014). The development of these peatlands can be complex. Due to the presence of limnic materials at the base of the wetlands, and forest peat at the surface, these wetlands likely developed following the trajectory described in Anderson et al. (2003). After deglaciation, large depressions were filled with water and limnic deposits accumulated in the lake bottoms, and terrestrialization of the waterbody began with formation and encroachment of vegetative mats. Following mat development and thickening, larger vegetation, such as shrubs and trees, could be supported, and the process of paludification then became the primary mechanism of peat development.

Fig. 1.

Index map (a) showing the location of the Montreal Plain, Middlesex, and Peterborough study areas with respect to Ontario and Quebec, and maps showing organic deposits mapped within the Montreal Plain (b), Middlesex County (c), and Peterborough County (d) study areas that contain limnic materials. Topographic base maps courtesy of the Ontario GeoHub, Ontario Ministry of Natural Resources and Forestry. Map projection: GCS WGS 1984. [Colour online.]

At each soil inspection, soil profiles were described (Day 1983) and classified as per the CSSC (SCWG 1998). Given that soil profiles were collected under different projects, soil physical and chemical analysis varied considerably. Where available, organic matter content, bulk density, soil pH, calcium carbonate equivalent, cation exchange capacity, total organic carbon, total carbon, total nitrogen, carbon to nitrogen ratio, water content at saturation and field capacity, electrical conductivity, coefficient of linear extensibility (COLE), shrinkage, and hydraulic conductivity are presented. Analyses and method references are provided, by study area, in Table 2.

Table 2.

Summary of soil properties, units of measurement, and method reference for the Middlesex, Peterborough, and Montreal Plain study areas.

For the mapping component, we acquired the detailed soil survey (DSS) geodatabase for provinces made available through the Canadian Soil Information Service (CanSIS), and additionally, where available, acquired DSS mapping directly through the provincial authorities (Soil Data Distribution Package, British Columbia; Agricultural Region of Alberta Soil Inventory Database (AGRASID), Alberta; Saskatchewan Soil Information Database (SKSID), Saskatchewan; Soils Agricultural Interpretations Database (SoilAID), Manitoba; Ontario Soil Survey Complex (OSSC), Ontario; Études pédologiques, Quebec). We then searched through all records in the available data sets for descriptions of any soil series classified as a Limnic subgroup of either Humisol, Mesisol, or Fibrisol great groups, and any example soil profiles containing Oco or other limnic horizons (i.e., marl and diatomaceous earth). In addition, soil series known to the authors as containing limnic deposits in shallow organic soils within the study areas that would not be currently recognized as limnic materials were also identified. Polygons within the geodatabases that contained these soils were then extracted to create maps and estimate the potential geographic extent and distribution of these soils in Canada.

Field observations

Morphological descriptions of soil profiles (n = 29) are provided in Table 3. Only one profile is provided from the Middlesex project, three from Peterborough, and 25 from the Plain of Montreal. The profile from Middlesex is unique in that it is the only profile described with diatomaceous earth and marl whereas only two profiles from the Plain of Montreal contained a marl layer. All other profiles contain coprogenous earth. All profiles are classified as Terric subgroups, meaning they have a terric layer at least 30 cm thick below the surface tier. Due to the presence of a terric layer and the absence of a lithic contact in any of the profiles, the control section for all profiles extends from the surface to 160 cm. For classification purposes, when the terric layer is within the middle tier, the dominant material in both the middle and surface tier is given consideration to determine the great group; however, when the terric layer is below the middle tier, only the materials within the middle tier are used to determine the great group. There are 11 profiles classified as Terric Humisols, five as Terric Mesisols, one Terric Mesic Humisol, and 12 profiles that cannot be classified as soils of the Organic order (Table 3).

Table 3.

Soil profile descriptions for soils with limnic layers in the Middlesex, Peterborough, and Montreal Plain study areas.

The thickness of humic, mesic, and limnic materials to be considered for great group classification for each profile was quite variable (Table 3). The average thickness across all profiles was 50.8, 16.4, and 46.8 cm for the humic, mesic, and limnic materials, respectively. All 29 soil profiles had limnic materials as either dominant (11 profiles) or subdominant (18 profiles) within the material to be considered for great group classification (depends on the depth to terric layer), while humic materials were dominant in 13 profiles and subdominant in eight profiles. Mesic materials were less common and were dominant in only five profiles, and subdominant in two profiles.

In total, 28 limnic material samples from the soil profiles were analyzed for various chemical and physical properties (Table 4). Since there were so few marl and diatomaceous earth samples, they were excluded from the analysis, leaving only coprogenous earth materials. Organic matter content ranged from 4.5% to 81.8%, highlighting the fact that coprogenous materials can be either organic or inorganic. Bulk density was below 0.34 g cm⁻³ with the exception of one site; this low bulk density is similar to that of humic organic materials. The mean pH was neutral; however, two sites in the Napierville area were acidic (pH < 5.5). A mean calcium carbonate equivalent of 24.2% for sites in Peterborough suggests the materials there are strongly calcareous; however, it should be noted that two sites in Quebec had this analysis, and both had no calcium carbonate. This, coupled with the pH results, suggests the biogeochemistry of the environments at the time of deposition was quite different between the Ontario and Quebec locations. Despite differences in the other soil properties, the C:N ratio was within a narrow range (11.6–16.8) across all sites and horizons. Soil electrical conductivity was only measured at some of the sites in Quebec and seemed to indicate the materials were slightly saline; however, the nature of the salinity was not examined. It should be noted that the Quebec study area was once inundated by the Champlain Sea; therefore, saline materials are not surprising. In terms of water retention characteristics, the coprogenous earth had on average 86.4% and 72.8% water content at saturation and at field capacity, respectively, and hydraulic conductivity of 0.0 cm h⁻¹, demonstrating the imperviousness of the material. Two specialized physical analyses, COLE and shrinkage, were determined for only two of the Oco samples. Both parameters are indications of the shrinking potential of the material. COLE was 0.5 cm cm⁻¹, which indicates a 50% reduction in the material and shrinkage was 3.6 m³ Mg⁻¹, both confirming the potential for major subsidence if the material is allowed to dry.

Table 4.

Summary of physical and chemical analyses from coprogenous earth horizons from the Peterborough and Montreal Plain study areas.

In addition to the analytical data from this study, data are reproduced from Hamel, Malouin, Ruel et Associés (1972), which has the most detailed information about the coprogenous materials in the peatlands of the Plain of Montreal, with emphasis on their suitability for vegetable crop production (Table 5). These data show close alignment to those collected in this study and provide additional analyses not completed on the more recent sample collection efforts, including the pyrophosphate index, unrubbed fiber, and rubbed fiber content, which are analyses specifically used to determine the level of decomposition of materials in organic soils. Based on their results, the Oco horizons sampled in their study are typically at a medium level of decomposition. Photos of coprogenous materials and soil profiles containing these materials are provided in Fig. 2.

Fig. 2.

Photos of soil profiles containing limnic horizons and coprogenous materials. Soil profile from Peterborough County (a), mixing of coprogenous material into surface plow layer in Napierville County (b), layered nature of coprogenous material from Napierville County after air-drying (c), close-up of two coprogenous layers from a profile in Peterborough County with the lower layer containing shells (d), close-up of layering of dark and light bands in the coprogenous horizon (e), sloughing of humic material into the coprogenous layer above an installed tile drain (f), soil profile from Peterborough County (g) and burgundy colored core of coprogenous material from plain of Montreal area (h). Photo credits: D. Saurette (d, g); R. Deragon (e, f, h); L. Lamontagne (a, b, c). [Colour online.]

Table 5.

Summary of physical and chemical analyses from coprogenous earth horizons from the Montreal Plain study area reported in Hamel, Malouin, Ruel et Associés (1972).

Despite the fact that limnic materials were found to be dominant or subdominant in all of the profiles, based on the current rules for classifying soils of the Organic order in the CSSC, none of these profiles would include the term “limnic” in any of the five recognized taxonomic levels. Twelve of the 29 soil profiles cannot be classified into the Organic order at all, despite the fact they clearly belong there. The reason they cannot be classified as the Organic order is due to a lack of thickness of humic (Oh) or mesic (Om) materials. Based on the key to soil orders, soils of the Organic order must have organic horizons (>17% organic carbon) that extend from the surface to a depth ≥60 cm for fibric materials or a depth ≥40 cm for mesic and humic materials (SCWG 1998). The 12 profiles that cannot be classified do not meet these requirements (Om + Oh <40 cm). Furthermore, if these soils were keyed out as Organic order soils by modifying the key to the Organic order, 11 of the profiles could not be classified as a great group within the Organic order because the dominant material in the middle tier, or middle and surface tiers, is limnic material, whereas the key only allows for fibric, mesic, humic, or folic materials to determine the great group. Finally, if the rules at the great group level were ignored, and these soils were keyed into one of the subgroups within the three wetland Organic great groups, the classification would fail to recognize the limnic materials once again because the Terric subgroup takes precedence over the Limnic subgroup.

The extent of soils with limnic materials is not extensive. Based on available information from the provincial and federal soil survey data, we estimated 32057 ha of organic soils potentially with limnic deposits across Canada in five provinces (Fig. 3). Quebec holds the largest area with 18111 ha, followed by Ontario (6192 ha), British Columbia (3961ha), Manitoba (3791 ha), and Saskatchewan (2 ha). It is certainly worth noting that these estimates are based on known, mapped soil units, which contain limnic materials, and that additional areas are likely to exist. For example, DSS mapping efforts by the private sector to support project planning in Alberta have mapped limnic deposits near Edmonton, Alberta (Total E&P Canada 2007; Vujnovic et al. 2000), despite the fact the AGRASID database and the CanSIS DSS for Alberta do not contain soils with limnic materials. Agricultural suitability was certainly a key driver for soil survey, and as such organic soils were never surveyed and described as intensively as mineral soils. Proof of this is apparent in many soil survey reports where organic soils were either reported simply as “muck” or “peat” or treated as a “Miscellaneous Mapping Unit” (Gillespie and Wicklund 1971; Hoffman 1974; Hoffman et al. 1963). In such areas, a more detailed survey of organic deposits would likely yield the discovery of additional acreages of organic soils with limnic materials, especially given the fact that these materials have been described in adjacent surveys/projects.

Fig. 3.

Map of Canada showing soil polygons where limnic materials have been mapped as per provincial soil survey data, detailed soil surveys from the Canadian Soil Information Service (CanSIS), and the information contained in the Soil Name File of CanSIS. Topographic base map courtesy of the Ontario Ministry of Agriculture, Food and Rural Affairs. Map projection: Canada Lambert Conformal Conic, GCS North American 1983. [Colour online.]

Proposed classification revisions

To address these shortcomings, we propose a series of modifications to the Organic order of the CSSC to be considered for inclusion in the fourth edition (in development):

It should be noted that no changes are proposed for the Folisol great group because these are upland organic soils, and limnic deposits only occur in wetland organics since these materials are deposited in standing water. Revised keys to the Organic soil order, great groups, and subgroups are provided in Appendix A, where changes to the keys are denoted by underlined and italic text. In addition to the revisions to the keys, upon acceptance, content will be created to amend Chapter 9: Organic Order, in light of these suggested changes. Furthermore, we propose a slight clarification to the definition of the Middle Tier (SCWG 1998, p. 11):

If these proposed changes were adopted, the resulting classification for all the soil profiles described in this study would change, as outlined in Table 3. In summary, we would have 13 profiles classified as Terric Limnic Humisol, nine profiles as Terric Humic Limnisol, five as Terric Limnic Mesisol, and two as Terric Mesic Limnisol.

Discussion

The CSSC is not the only system that recognizes limnic deposits in organic soils for taxonomic purposes and, likewise, is not the only system that might lack clarity when classifying profiles composed dominantly of limnic materials. In Soil Taxonomy, for instance, the Limnic subgroup can be applied to each great group within the Histosol order, the equivalent of the Organic order in the CSSC (USDA-NRCS 1999). In Soil Taxonomy, however, the Limnic subgroup appears before the Terric subgroup (the opposite of the sequence in the CSSC), and therefore any Histosol with a limnic layer, or layers, with a total thickness of ≥5 cm, regardless of the presence of a terric layer, is assigned to the Limnic subgroup of the corresponding great group (USDA-NRCS 1999). This, of course, creates the opposite problem faced in the CSSC and does not resolve the issue of accommodating both Terric and Limnic in a combined subgroup. The World Reference Base (WRB) system of soil classification uses two levels to classify soils: the Reference Soil Group (RSG) and Qualifiers, of which the latter are separated into principal and supplementary qualifiers. The principal qualifiers are intended as integral to allow further characterization of a soil, while the supplementary qualifiers are intended to provide additional characterization (IUSS Working Group WRB 2015). Limnic materials are represented only as a supplementary qualifier across many of the RSGs, including the Histosols, whereas “fibric”, “hemic” (mesic), and “sapric” (humic) are designated as principal qualifiers (IUSS Working Group WRB 2015). One difference with the WRB is that “terric” is not used as a qualifier for the Histosol Reference Soil Group.

It should be noted that, in Canada, considerably less research has been conducted to quantify and characterize the properties and origins of coprogenous earth materials as has occurred in Northern European countries such as Finland (Berglund 1996), Sweden (Larsson 1990), Poland (Jarnuszewski and Meller 2018, 2019; Łachacz et al. 2009), and Germany (Schulz et al. 2019). Schulz et al. (2019) classify six types of gyttjas, or coprogenous earth, including detritus, algal, calcareous, sand, silt, and clay gyttjas, differentiated based on their organic matter content, lime (CaCO₃) content, and silicate fraction. In Poland, gyttjas, or lake bottom deposits, have been differentiated based on their ash content without CaCO₃, organic matter content by loss-on-ignition, and CaCO₃ content (Łachacz et al. 2009; Łachacz and Nitkiewicz 2021), which results in a classification triangle that includes 11 different classes, with the sum of the three components adding up to 100. It is evident that more research is needed to better classify the types of coprogenous earth that exist in Canada, and that advanced systems in Europe would provide much insight.

In addition to taxonomic considerations for recognizing limnic materials at a higher taxonomic level, many agricultural considerations would also justify the proposed modifications to the CSSC. Organic soils make up approximately 10% of the Canadian land mass and support important agricultural regions in Canada (Lévesque et al. 1981). A primary example is the Saint-Lawrence plain south of Montreal, which has 145 000 ha of deep organic soils and 24 000 ha of shallow organic soils (Grenon 1988). Specialty crop production in organic soils, focused on root and leafy vegetables, is critically important to the agricultural sector in this region since as early as 1936, with high-value crops destined for markets in Montreal and New York State (McKibbin and Stobbe 1936), generating 50% of agricultural revenues on only 3%–4% of the agricultural land base in the province (Groupe AGÉCO 2007; Parent and Gagné 2010). A significant portion of the organic deposits in the region is underlain by coprogenous earth, and they are at increased risk of being exposed at the surface as a result of subsidence of the peat due to drainage and continued losses of surface organic materials from microbial decomposition (1.25 cm year⁻¹) and wind and water erosion (1.25 cm year⁻¹). Parent et al. (1982) concluded that wind erosion rates could be as high as 4.53 cm year⁻¹ in these organic soils when beneficial management practices were not implemented to protect the soils.

Coprogenous materials are reported to have characteristics that could be deleterious to agricultural production. These include being plastic and gelatinous when wet, hard when (irreversibly) dry, saline, and containing potentially high levels of sulfur (Kroetsch et al. 2011). In a detailed assessment of the agricultural potential of organic soils in the Plain of Montreal, Hamel et al. (1972) noted that the underlying gelatinous material (sedimentary peat) made tile drainage very difficult, limited agricultural production when too close to the surface, had negative effects on soil fertility, compromised production once incorporated into the surface peat, and offered very little possibility for continued agricultural production once the overlying peat was gone. They also noted that once dry, the material cracks irreversibly, hardens, and comes apart like thin leaves. The installation of tile drainage is problematic when an irregular limnic layer is present in a field since tiles must sit above the impervious material. Common drainage systems are sometimes inefficient in the presence of such a layer and might require the use of surface drains such as trenches, or complex, multileveled drainage systems. In northeast Poland, many of the extensive gyttja lands have been excluded from agricultural production due to the difficulty in farming them (Łachacz and Nitkiewicz 2021). More recently in Canada, the presence of a coprogenous or mineral layer found close to the surface was found to affect soil's physical and chemical properties in the plain of Montreal (Deragon et al. 2022). For instance, the coprogenous material was found to have salinity levels high enough to reduce crop yields. Crops, particularly high-value vegetable crops, are negatively affected by different levels of salinity. For example, carrots (1.0 mS cm^-1), onions (1.2 mS cm⁻¹), celery (1.8 mS cm⁻¹), lettuce (2.0 mS cm⁻¹), and spinach (2.0 mS cm⁻¹) are sensitive and moderately sensitive to soil salinity (Machado and Serralheiro 2017). Possible yield loss can be expected when soil salinity exceeds these thresholds, which was frequent in the study area (Table 4). Water retention, and indirectly gas diffusion, was correlated to the presence of those relatively impervious coprogenous layers. Kroetsch et al. (2011) further support this claim by noting the plasticity and gelatinous appearance of limnic material when wet, and the shrinkage and hydrophobicity of the material when dry. Vegetable roots require sufficient aeration and a growing medium that can store and supply water throughout the growing season, which a limnic layer would fail to deliver. The bearing capacity of these soils can also be problematic for heavy agricultural machinery if a limnic layer is near the soil surface; the consistence is such that machinery can sink dangerously into the limnic material (Hamel et al. 1972).

A more accurate soil classification could provide better information for land transactions. Indeed, a Limnic great group, and better integration at the subgroup level, would add crucial information important to understanding the long-term agricultural potential of an organic soil containing limnic materials. With an average selling price of $63000 ha⁻¹ (2019–2021) in Napierville County, cultivated organic soils are a scarce resource (data extracted and filtered from the Registre financier du Quebec). Inevitably, soils classified as Limnic Terric Humisols or as Terric Limnisols would identify the presence of limnic materials potentially close to the surface and indicate reduced land value, because once exposed or incorporated into the plow layer, thick limnic layers would be an impediment to further agricultural land use (J. Caron, personal communication, 2022). Figure 2b shows that dried coprogenous material does not mix well in the surface plow layer. This behavior can affect the seeding operation and can lead to non-uniform germination. While a thin limnic layer could be mixed with the underlying mineral soil to allow cultivation after the depletion of the peat, a thick limnic layer could not be so easily incorporated into the underlying material. Moreover, the underlying mineral soil would likely have fertility issues and undesirable physical properties for specialty crop production, such as low air content. Yield losses could be expected, and therefore land value would decrease. A metric such as the residual maximum peat thickness (MPT)—the thickness of the arable peat layer—has been proposed to estimate the remaining years of production considering average soil loss rates (e.g., 2.5 cm year⁻¹) if no conservation practices are applied (Deragon et al. 2022). Deragon et al. (2022) observed statistically significant changes in chemical and physical properties of peaty layers as a function of the MPT, where soils with an MPT <60 cm showed important signs of soil degradation, while soils with an MPT >60 cm showed fewer signs of degradation. Not only would the MPT reflect the potential of a field, but it would also allow the assessment of economically justifiable soil conservation practices. Wojahn and Illner (1962) determined a critical peat thickness of 80 cm should be maintained over limnic materials due to the possibility of lateral flow of the limnic materials into surface drainage ditches. Therefore, organic soils with limnic layers should not be considered a priority management zone in a soil conservation context, especially if the MPT is less than 60 cm (Deragon et al. 2022) to 80 cm (Husemann 1947; Wojahn and Illner 1962) in thickness.

Recommendations and conclusion

Limnic materials are poorly captured in the CSSC and are even ignored at all five taxonomic levels in some instances, despite the fact these limnic materials can be dominant within a soil pedon. Furthermore, their composition is poorly documented in Canada, and further studies are required to better understand the nature and properties of these materials in the future. Numerous soil profile descriptions were provided from Ontario and Quebec to highlight the deficiencies in the CSSC, and a review of DSSs confirms the potential for over 32000 ha of organic soils with limnic deposits, which we conclude is likely an underestimate. In addition to taxonomic considerations, we also demonstrate the importance of modifying the CSSC for agronomic and financial reasons, including negative impacts of limnic materials on crop production. Based on our analysis, we recommend the addition of the Limnisol great group to the Organic order, and additional integration of limnic materials at the subgroup level, specifically as it relates to recognizing limnic and terric together. Furthermore, we provide revised keys to the Organic order, its great groups, and subgroups in light of the proposed changes.