1. Introduction
Climate is predicted to warm most dramatically at high latitudes in response to global atmospheric change (Houghton et al. 1996), a prediction that appears to be consistent with observed changes over the past three decades (Serreze et al. 2001, Chapman and Walsh 1993). In order to assess this consistency, and ultimately to make credible predictions of future change in northern latitudes it is vital to realistically simulate the governing processes and controls on the exchange of moisture and energy with the surface. Processes in the surface layer, which form the interface between the ground and the atmosphere, play a key role in controlling surface exchanges. In northern latitudes, mosses and lichens often dominate this surface layer where, we suggest, they are a significant factor in modulating the surface energy exchanges and the thermal, hydrologic, nutrient, and carbon regimes of soils (van Cleve et al. 1986).
Mosses (Bryophytes) are plants that consist of small, slender stalks and leaves with no vascular tissue or true roots. Mosses all reproduce by spores and can also form new plants from small fragments of stems and leaves that are broken off. Mosses have structures that resemble roots, stems, and leaves, but they lack true water- and food-conducting tissues. Lichens consist of fungal threads and microscopic green alga living together and functioning as a single organism. Lichens do not have roots, stems, and leaves, so they must receive their nutrients from rainfall. They are considered pioneer species in some habitats. Lichens are a group of composite organisms made up of a fungus and an alga living in symbiotic association. The fungus provides a structure that may protect the alga from drying under harsh conditions; the alga synthesizes and excretes a specific carbohydrate that is taken up and utilized as food by the fungus.
Beneath the moss–lichen layer is typically a layer of peat, whose properties also differ from the underlying mineral soil. Currently there is no explicit incorporation of moss, lichen, or peat layers in the National Center for Atmospheric Research Land Surface Model (NCAR LSM; Bonan 1996) that has been used to investigate land–atmosphere interactions in the Arctic (Lynch et al. 1999a,b; Eugster et al. 1997).
Mosses are ubiquitous in boreal forest and tundra ecosystems, which occupy 14% of the total global land area. In boreal forests, feather mosses (Hylocomium and Pleurozium spp.) dominate the groundcover and can occupy 50%–100% of forest floor area (Fig. 1a). In these stands mosses compose only a small fraction of the total ecosystem biomass but can contribute up to 50% of total annual net primary production (Viereck et al. 1986; Oechel and van Cleve 1986). In contrast, mosses in tundra ecosystems, especially peat mosses (Sphagnum spp.), may become the dominant vegetation in terms of total biomass and form a continuous cover over large areas of the landscape (Fig. 1b).
Mosses form a thick, insulating layer that alters the partitioning of incoming radiation between turbulent fluxes (sensible and latent heat), and ground heat flux, a determinant of soil and permafrost temperature regimes (Bonan et al. 1990; Dyrness 1982). Mosses are particularly important in the discontinuous permafrost zone, where the mean annual temperature is near 0°C (Nicholas and Hinkel 1996). If mosses and the underlying peat layer are removed by fire or mechanical disturbance, the active layer depth increases because of the increased heat conducted to the permafrost (Nicholas and Hinkel 1996; Mackay 1995; Dyrness et al. 1986). Ultimately, thawing of permafrost may lead to thermokarst (ground surface collapse) and inundation of lower-lying areas with water, with potentially large ecological and economic consequences (Nisbit 1989).
It has been estimated that these moss-dominated ecosystems (boreal forest and tundra) account for approximately 35% of the world's reactive soil carbon pool (McGuire et al. 1995). This soil carbon has accumulated because of low soil temperatures and/or poor drainage or is locked up in permafrost (Hobbie et al. 2000). Higher soil temperatures resulting from the disturbance of the moss layer or high-latitude warming could increase decomposition rates, potentially changing these systems from being a sink to a source of CO2 to the atmosphere (Ciais et al. 1995; Grulke et al. 1990) and creating a positive feedback to warming (Oechel et al. 1993; Tans et al. 1990; Billings et al. 1982; Post et al. 1982). Furthermore, an increased active layer depth may lower the water table, causing greater aeration and even warmer temperatures, in turn increasing decomposition of soil carbon. This could represent a significant additional source of CO2 to the atmosphere (Oechel and Vourlitis 1994). We do not address the direct effect of mosses and lichens on nutrient and carbon cycling; however, changes in the thermal and hydrological regime will have a major influence on these cycles.
Despite the potentially strong modulating effects of moss and lichen, they are not well represented in models of soil thermal dynamics or in models of land surface processes dealing with the regional effects of climate and vegetation change. At present the physical properties are generally poorly represented in land surface models utilized in climate models, with the exception of the Canadian Land Surface Scheme (Tilley and Lynch 1998; Verseghy 1991). This is surprising given that moss and lichen compose such a large proportion of the surface area in tundra and boreal systems. Current regional climate simulations for Alaska using the Arctic Regional Climate System Model (ARCSyM; Lynch et al. 1995) incorporate the NCAR LSM but the soil scheme is parameterized for bare mineral soil only. The surface properties of moss, lichen, and peat are sufficiently different from a mineral soil surface to warrant their inclusion in land surface models, particularly for northern ecosystems. In addition, the NCAR LSM does not currently allow for soil layers of differing texture types but specifies instead only one texture type for the whole column. In this study we investigate the effect of incorporating moss, lichen, and peat into the NCAR LSM and investigate the consequences for land–atmosphere exchanges. Both thermal and hydraulic properties are examined as well as the effect of mosses on the hydrology and exchanges of heat and moisture from the surface. Our research will provide information on how the thermal and hydrological regime may be altered through disturbance of the moss layer and the subsequent effect on the thermal characteristics of the surface layer.
a. Model description
The NCAR LSM is a one-dimensional model of energy, momentum, water, and CO2 exchange between the land and atmosphere. The model accounts for ecological differences among surface types within a grid cell (including tundra, lakes, and wetlands) and thermal and hydrological differences among soil types. We modified the soil thermal and hydrological modules of LSM in this study, as described below. A complete detailed technical description of the model is given by Bonan (1996).
1) Soil thermal and hydrology processes
The default soil column in LSM contains six layers with thicknesses of 0.10, 0.20, 0.40, 0.80, 1.60, and 3.20 m. In the standard configuration all six soil layers in the original model have identical properties that are specified with one set of parameters indicating the percentage of sand, silt, and clay for the entire soil column. The soil thermal and hydraulic properties are derived based on these percentages of sand, silt, and clay. Hence, soil textures do not vary with depth.
In the present study, the paradigm of specifying percent sand, silt, and clay values has been replaced by the specification of a single soil texture type for each one of the six layers in the soil column. Hence a more plausible soil profile can be defined. We added moss, lichen, and organic peat to the 12 U.S. Department of Agriculture soil texture types that were originally specified in the NCAR LSM (Wilson and Henderson-Sellers 1985). We specify the physical properties of each of these different texture types through a lookup table that provides all the thermal and hydraulic parameters for each texture type (Table 1). This is essential for the moss, lichen, and peat types, whose properties cannot be expressed as a function of the fraction of sand, silt, and clay.
2) Soil thermal regime
3) Hydrologic regime
b. Hydraulic characteristics
The hydraulic characteristics of the additional moss, lichen, and peat texture types is examined using the algorithms of (Clapp and Hornberger 1978) [Eqs. (3) and (4)] that are used in the NCAR LSM (Figs. 2a and 2b). These algorithms simulate the relationships between water content, soil water suction, and hydraulic conductivity and are subsequently used to model the movement and distribution of soil water through each layer of the soil column.
The relationship between soil water content and soil water suction, known as the water characteristic curve, determines how much moisture a soil will hold (θ) against a force such as gravity or extraction by plants (ψ) (Fig. 2a; Hinzman et al. 1991). This curve, as derived from Eq. (4) shows that, as soil water content decreases the matric potential decreases because the small amounts of water remain in the soil become more difficult to extract. In organic and other porous soil layers, water within the large soil pores drains easily, giving mosses and lichens a flat water characteristic curve with a high matric potential (less negative) at a wide range of water contents (Hinzman et al. 1991). In this study, for a given water content, the matric potential was smallest in loam followed by peat, sand, moss, and then lichen, suggesting that for a given soil water content that water is less tightly held in the moss and lichen (Fig. 2a).
For any given water content, lichen is parameterized with the highest hydraulic conductivity followed by moss, sand, peat, and loam. Consequently, the rate of moisture transfer in coarse texture types, for example mosses, for a given water content will always be greater than the loam column (Fig. 2b). This occurs because the soil has a larger saturated hydraulic conductivity and the hydraulic conductivity decreases more slowly with decreasing saturation (Fig. 2b), as also found by Wilson et al. (1987). Although these hydraulic relationships follow expected trends, actual field or laboratory measurements of the water characteristic curves for mosses are still needed for a more accurate parameterization of the model.
2. Model experiments
Simulations using our modified version of the NCAR LSM were performed for an arctic tundra site at the Imnavait Creek watershed of northern Alaska (69°25′N, 148°45′W). The watershed is underlain by continuous permafrost with a summer active layer depth of 50 cm (Hinzman et al. 1991). This site has been the focus of numerous research studies examining the role of permafrost-dominated tundra in the regional climate system (Lynch et al. 1999b; Eugster et al. 1997; Lynch et al. 1995; Hinzman and Kane 1992).
The land surface model was configured for a tundra ecosystem consisting of 70% “arctic grass,” 25% “deciduous shrub,” and 5% exposed “bare ground.” The ground albedo was held constant across the experiments as previous land surface studies have shown little sensitivity to soil albedo (Wilson et al. 1987). A color class of 3 was used following Dickinson et al. (1993), which is equivalent to a visible albedo of 0.2–0.10 for dry and saturated conditions, respectively, and a near-infrared albedo of 0.40–0.20 for dry and saturated conditions, respectively. This albedo range is similar to albedos measured over a moss-and lichen-dominated tundra ecosystem (Fig. 1b) of 0.195 (unpublished data). Spectral reflectances of moss and lichens have not been well documented, however, in general shortwave and near-infrared reflectances in mosses are less than other vascular plants (Bubier et al. 1997). Lichens have a greater shortwave reflectance than mosses and other plants and a near-infrared reflectance less than other vascular plants (Bubier et al. 1997). As a result, albedos of mosses and lichens are likely to be different from current parameterizations in tundra and boreal forest environments and the impact of these differing albedos on the surface energy balance needs to be addressed further. It should be noted that albedo changes will result in changes in the absolute magnitudes of fluxes but not the partitioning of energy.
A single annual time series of atmospheric forcing data for 1995 was constructed using a composite of observed data, proxy data from nearby stations, and European Centre for Medium-Range Weather Forecasts operational analysis. Simulations used the 1995 time series forcing data repeated over three consecutive annual cycles with the first two years allowed for spinup and the third year for analysis. Each simulation used identical forcing data, and hence any errors inherent to the construction of the composite forcing data are not of concern here.
Five experiments were performed (Fig. 3) using these forcing data. A “loam” experiment was performed using a homogenous column of loam soil, which is the standard current implementation of the NCAR LSM as coupled to the NCAR Climate System Model (Boville and Gent 1998) and the ARCSyM (Lynch et al. 1995). A second experiment comprising a homogeneous sand column (designated “sand”) was performed to compare the effect of varying the soil texture of a whole column (Fig. 3). A third experiment utilized a homogeneous peat column (designated “peat”). Two further experiments were performed in order to investigate the sensitivity of the simulation of Arctic soil profiles to different surface layer soil types using the same underlying soil profile. In each case, the profile consisted of two layers of peat underlying the surface followed by loam soil in the lowest three layers. The surface layer type was either moss or lichen, and they were designated as the “moss” or “lichen” experiments, respectively (Fig. 3). These experiments were analogous to those suggested by Morrill et al. (1999) in preliminary experiments with the Biosphere–Atmosphere Transfer Scheme model (Dickinson et al. 1993). All experiments were performed “offline” wherein the NCAR LSM responds to atmospheric forcing but does not feed back to influence the forcing. It should be noted that while the one-dimensional model has shown a strong impact on surface fluxes, it does not account for dynamical feedbacks with the atmosphere, which may have a modulating effect. However, these effects are unlikely to be eliminated entirely (Lynch et al. 1999a). The experiments in this study are a necessary first step in the process of analyzing the sensitivity of a climate model to a parameterization change, and it would be useful to assess the impact of mosses using regional climate simulations in the Arctic.
3. Results and discussion
Model simulations for all experiments showed all layers being frozen in winter with snow persisting until spring, which realistically follows actual conditions (Hinzman et al. 1991). As a result the wintertime thermal and hydraulic regimes were dominated by the properties of frozen water and snow and are not presented here. Simulations for all experiments showed that the surface layer was unfrozen only for the months of June, July, and August (data not shown). We present the July means for the third year of the simulations to illustrate how thermal and hydraulic properties vary with depth during midsummer.
a. Thermal regime
In LSM, the thermal regime of the model is determined by the specified values for thermal conductivity (Ksolids) and heat capacity (Csolids) of the substrate solids, the volumetric fraction of solids in the substrate (derived from the saturated water content, also called porosity), and the volumetric fraction of water in the substrate (Table 1). In July, the top two layers of the moss, peat, and lichen experiments were unfrozen, but the top three layers were unfrozen in the sand and loam experiments (Fig. 4). The depth of the active, or unfrozen, layer in the moss simulations was around 50 cm and is consistent with field observations (Hinzman et al. 1991) but simulations in the bare loam case were too deep (∼1 m).
The thermal conductivity of the top layer in the moss, lichen, and peat experiments was more than 4 times lower than for sand and loam experiments (Fig. 5a) because of the high air volume and lower water content in the surface layers of moss, lichen, and peat. The simulated thermal conductivity of the moss surface layer in July (0.37 W m−1 K−1) agrees well with field observations of mosses in arctic Alaska (0.1 W m−1 K−1 when dry to 0.6 W m−1 K−1 at saturation; Hinzman et al. 1991) and in Sphagnum or peat varied (0.04 to 1.1 W m−1 K−1 from dry to wet respectively; Brown and Williams 1972) but are higher than values reported by Sharrat (1997) for feathermoss on the floor of a black spruce forest near Fairbanks, Alaska (0.030 to 0.088 W m−1 K−1). Below the surface in the lower three layers of each experiment the thermal conductivity remained fairly constant because of the homogeneous loam texture in these lower layers. Differences in thermal conductivity between experiments were due to the variation in water content in the lower layers. The peat experiment had the highest water content in the lower layers and as a result had the highest thermal conductivity of all experiments, followed by moss, loam, lichen, and sand experiments (Fig. 5a).
The thermal conductivity of moss, lichen, and peat in the surface layer was low, and they effectively insulated the underlying soil. Along with differences in apparent heat capacity, they contributed to cooler simulated summer (July) soil temperatures in the lower layers than for a mineral soil surface (i.e., loam and sand experiments). For example, at 0.5-m depth the temperature was 6.9°C lower in the moss experiment than the loam experiment (Fig. 4). Conversely, this insulative effect produced warmer wintertime (January) soil temperatures for the moss, peat, and lichen experiments at depth with the temperature being 2.3°C higher in the moss experiment than the loam experiment at 0.5-m depth (Fig. 4). Land surface simulations appear to be sensitive to changes in thermal conductivity as observed by Bonan (1991) who found that doubling or halving thermal conductivity of the near-surface layer resulted in a 0.8°C increase or 1.1°C decrease in soil temperature, respectively, during the growing season in Alaska. Our study showed that the difference in thermal conductivity between moss (0.37) and loam (1.7) experiments contributed to simulated July surface soil temperatures of 11.85° and 12.25°C, respectively, a difference of 0.4°C.
The other determinant of the thermal regime is the apparent heat capacity (Cυ). Simulations for July show the surface layer was relatively dry and the apparent heat capacity was lowest for the lichen experiment, followed by moss, sand, peat, and loam experiments (Fig. 5b). The lower heat capacity for the surface moss layer as compared with the loam layer produced a 0.4°C-higher surface temperature for an equivalent energy input for the moss experiment in summer as compared with the loam experiment (Fig. 4). Moss surface temperatures were 0.8°C cooler than loam in winter (January), when the net heat flux was from the soil to the atmosphere. Simulated differences in the apparent surface layer heat capacity were driven by differences in porosity that were greatest for the lichen surface layer. In addition, July simulations showed that the third layer was thawing for the moss, peat, and lichen experiments and a large amount of energy was dissipated in the phase change from ice to water in that layer (Fig. 5b). Hence the apparent heat capacity for the third layer was very high. Similarly, for loam and sand the fourth layer was thawing and resulted in that layer having a very high apparent heat capacity (Fig. 5b).
The thermal and hydrological regimes are intricately related to each other and to the character and structure of the active layer and hence influence permafrost dynamics (Hinzman et al. 1991). For example, observations show that wet or frozen peat is highly conductive and promotes heat loss to the atmosphere when air temperatures are cold (Nicholas and Hinkel 1996; Stoutjesdijk and Barkman 1991). During dry conditions, the thermal effect reverses: dry peat insulates soil and permafrost from surface heat fluctuations and soils remain cool (Oechel and van Cleve 1986; van Cleve et al. 1983). This switching of thermal effect results in an overall cooling of soil and permafrost under moss cover. Although this pattern has been observed in moss-dominated ecosystems, there are few data on the thermal and hydraulic properties of mosses needed to accurately parameterize LSM for application in the Arctic.
b. Hydraulic characteristics
The simulated water content in each layer of the soil in LSM is a function of the hydrological balance among precipitation, evaporation, and runoff. The distribution of water among the layers depends on the hydraulic conductivity and porosity of the soil. Because of the high hydraulic conductivity specified for mosses and lichens, water infiltrated much more quickly into the lower layers, resulting in higher water contents in these lower layers, during the summer (July) simulations of the moss and lichen experiments (Fig. 5c). Greater surface evaporation (due to warmer surface temperatures) and high infiltration rates in the surface layer in the lichen and moss experiments resulted in lower water contents in the surface layer in comparison with the other experiments (Fig. 5c). Water contents tended to be evenly distributed among all soil layers for the sand and loam experiments because of the homogeneous nature of the texture type in these soil profiles. In summary, the new modifications to specify different properties for each soil layer alters the hydrologic regime of the entire soil column and provides a more plausible representation of the hydrology of the column. This can be used to improve simulations using ARCSyM as discussed later.
The low water content and high porosity of the surface layer in the moss and lichen experiments resulted in low matric potentials that would offset the tendency of these layers to drain quickly (Fig. 5d). The peat layers beneath the surface layer of the moss and lichen experiments were also drier than in other experiments, resulting in a lower matric potential for those layers (Fig. 5d). Below the third layer, matric potentials were similar among all experiments except sand because these lower layers were loam in all other experiments (Fig. 5d). Variations in matric potential in these lower layers arose solely from differences in water content. The sand experiment had a much higher matric potential throughout the profile, indicating that water was not held as tightly.
Hydraulic conductivity of the surface layer varied greatly among experiments, with the highest conductivities simulated for moss, followed by sand, lichen, peat, and then loam (Fig. 5e). This generally reflected the order of saturated hydraulic conductivities specified for specific texture types (Table 1) but was also controlled by the water content of the respective surface layers. The major exceptions were the lichen experiment that had a moderate simulated conductivity in July (Fig. 5e) due to its low surface water content and despite its high saturated conductivity (Table 1). Hydraulic conductivities in the second and sixth layers were lower than the surface layer because of the lower saturated hydraulic conductivity of the soil texture types that were specified for those layers (Fig. 5e). There was a slight increase in hydraulic conductivity for moss and lichen experiments in the first loam layer (fourth layer) due to an increased water content in that layer.
c. Surface energy exchanges and hydrology
The addition of moss and lichen not only affects the thermal and hydrologic regime in the simulations but also modulated the exchanges of energy to and from the surface. During winter (January), simulated mean monthly energy fluxes in all experiments were dominated by a loss of sensible and ground heat from the surface as it cooled. Mean monthly fluxes reached a summer maximum during June and July, with latent heat fluxes being dominant, followed by ground and then sensible heat fluxes (Fig. 6).
Surface energy exchanges were similar between the loam, sand, and peat experiments and then also between moss and lichen experiments. Here we present only the homogeneous loam column that is currently used in the ARCSyM model (Lynch et al. 1995) and the moss underlain by peat column, which best approximates the true soil profile in most tundra and boreal forest sites. The moss experiment showed a 57% reduction in ground heat flux in July in comparison with the loam experiment (Fig. 6), due to the lower thermal conductivity of the surface layer in the moss experiment. As a consequence of the reduced ground heat flux a larger proportion of incoming solar energy was directed into the turbulent fluxes of heat and water in the moss in comparison with the loam profile. For example, the sensible heat flux during July in the moss experiment was 67% greater than that in the loam experiment, and the latent heat flux, which was the largest of the surface fluxes, was 15% greater in the moss than the loam profile. The ground evaporation made up 44% of the total evapotranspiration in the moss experiment but only 37% in the loam experiment, suggesting that mosses are more effective contributors to total latent heat flux than a bare loam soil surface in these simulations.
Recent simulations using ARCSyM coupled with the existing NCAR LSM at Sagwon, Alaska, have shown that the modeled August latent heat fluxes were around 30% lower than observed values and that ground heat fluxes were around 3 times higher than observed values (Lynch et al. 1999b). Our results suggest that moss, lichen, and peat are climatically important and that simulated differences in ARCSyM fluxes could be improved by adding them to the NCAR LSM, which is coupled to ARCSyM.
Although the spatial extent of tundra and boreal forest ecosystems affected by moss and lichen is limited in a global sense and is unlikely to be an area of strong focus in a global climate model, a regional climate model such as ACRSyM may have a substantial area of tundra or boreal forest in the domain. In addition, the grid size of the regional model is smaller and hence any attenuation of these effects due to the mosaicking of vegetation types is less than in the global climate model. The spatial extent of the column in these experiments is the same as that in ARCSyM and hence is directly applicable. Although the experiments presented here are for a tundra vegetation, it is recognized that the absolute magnitude of the effect of mosses in a boreal forest system may be attenuated through shading by an overhead canopy. However, the relative sensitivities are the same as in the tundra experiments presented here. In addition, many boreal forest ecosystems have a sparse canopy, and hence the effects of mosses may still be quite important. It is also surface cover representation that is of increasing importance to realistic simulations of high-latitude climate.
The addition of mosses also affected the simulated hydrological balance, with the moss column having a lower July surface runoff (0.702 mm day−1) when compared with the homogeneous loam case (1.193 mm day−1). Wilson et al. (1987) also showed that modeled runoff decreased with increasing coarseness of the soil texture. Field measurements suggest that downslope movement of water during significant precipitation events occurs primarily in the surface organic layer (Hinzman et al. 1991). The decreased runoff for the moss column was due to the higher hydraulic conductivity of moss and the effective infiltration of precipitation into the soil profile. Simulated July infiltration rates were 3.696 (mm day−1) for moss and 3.231 (mm day−1) for loam. As a consequence, the simulated drainage through the profile in July was slightly higher in the moss (0.270 mm day−1) than the loam experiment (0.267 mm day−1).
4. Conclusions
The physical properties of mosses and lichens are sufficiently different from those of a bare loam soil that they warrant explicit parameterization in land surface models used in northern latitudes. The addition of variable soil texture types with depth in conjunction with the addition of moss, lichen, and peat layers resulted in a more plausible representation of northern soils in the NCAR LSM. The addition of a surface moss layer resulted in higher simulated winter soil temperatures and cooler summer temperatures. In addition, the distinct water characteristics of the moss surface resulted in simulated greater infiltration, lower surface runoff, and lower surface moisture contents. The insulating properties of mosses reduced simulated soil heat fluxes, resulting in greater energy available to be partitioned into latent and sensible heat fluxes. Therefore sensible and latent heat fluxes were greater in simulations with a surface moss layer. Last, further empirically based field and laboratory studies of the thermal and in particular hydraulic properties are still needed, and it will be important to examine the sensitivity of these parameters in simulated land–atmosphere processes.
Acknowledgments
This research is supported through the Arctic System Science (ARCSS) program of the National Science Foundation (OPP-9732126 and OPP-9732461). We thank Dr. Larry Hinzman for providing meteorological observations from Imnavait Creek. The helpful comments of Dr. Michaela Dommisse, Cath Copass, and Dr. Keith Olsson are appreciated.
REFERENCES
Billings, W. D., J. O. Luken, D. A. Mortensen, and K. M. Peterson, 1982: Arctic tundra: A source or sink for carbon dioxide in a changing environment? Oecologia, 53 , 7–11.
Bonan, G. B., 1991: A biophysical surface energy budget analysis of soil temperature in the boreal forests of interior Alaska. Water Resour. Res, 27 , 767–781.
Bonan, G. B., 1996: A land surface model (LSM version 1.0) for ecological, hydrological, and atmospheric studies: Technical description and user's guide. Tech. Note NCAR/TN-417+STR, National Center for Atmospheric Research, Boulder, CO, 150 pp.
Bonan, G. B., H. H. Shugart, and D. L. Urban, 1990: Sensitivity of some high-latitude boreal forests to climatic parameters. Climatic Change, 16 , 9–29.
Boville, B. A., and P. R. Gent, 1998: The NCAR Climate System Model, version one. J. Climate, 11 , 1115–1130.
Brown, R. J. E., and G. P. Williams, 1972: The freezing of peatland. National Research Council of Canada, Technical Paper of the Division of Building Research, Vol. 381, 24 pp. [Available from National Research Council of Canada, Division of Building Research, Ottawa, ON, Canada.].
Bubier, J. L., B. N. Rock, and P. M. Crill, 1997: Spectral reflectance of boreal wetland and forest mosses. J. Geophys. Res, 102 , 29 483–29 494.
Chapman, W. L., and J. E. Walsh, 1993: Recent variations of sea ice and air temperature in high latitudes. Bull. Amer. Meteor. Soc, 74 , 33–47.
Ciais, P., P. P. Tans, W. Trolier, J. W. C. White, and R. J. Francey, 1995: A large northern hemisphere terrestrial CO2 sink indicated by the 13C/12C ratio of atmospheric CO2. Nature, 269 , 1098–1102.
Clapp, R. B., and G. M. Hornberger, 1978: Emperical equations for some soil hydraulic properties. Water Resour. Res, 14 , 601–604.
Cosby, B. J., G. M. Hornberger, R. B. Clapp, and T. R Ginn, 1984: A statistical exploration of the relationships of soil moisture characteristics to the physical properties of soils. Water Resour. Res, 20 , 682–690.
Dickinson, R. E., A. Henderson-Sellers, and P. J. Kennedy, 1993: Biosphere–Atmosphere Transfer Scheme (BATS) version 1e as coupled to the NCAR Community Climate Model. Tech. Note NCAR/TN-387+STR, National Center for Atmospheric Research, Boulder, CO, 80 pp.
Dyrness, C. T., 1982: Control of depth of permafrost and soil temperature by the forest floor in black spruce/feathermoss communities. USDA Forest Service Research Note PNW-396.
Dyrness, C. T., L. A. Viereck, and K. Van Cleve, 1986: Fire in taiga communities of interior Alaska. Forest Ecosystems in the Alaskan Taiga, K. Van Cleve, et al., Eds., Springer-Verlag, 74–86.
Eugster, W., J. P. McFadden, and F. S. Chapin III, 1997: A comparative approach to regional variation in surface fluxes using mobile eddy correlation towers. Bound.-Layer Meteor, 85 , 293–307.
Grulke, N. E., G. H. Riechers, W. C. Oechel, U. Hjelm, and C. Jaeger, 1990: Carbon balance in tussock tundra under ambient and elevated atmospheric CO2. Oecologia, 83 , 485–494.
Hinzman, L., and D. L. Kane, 1992: Potential response of an arctic watershed during a period of global warming. J. Geophys. Res, 97 , 2811–2820.
Hinzman, L., D. L. Kane, R. E. Gieck, and K. R. Everett, 1991: Hydrologic and thermal properties in the active layer in the Alaskan Arctic. Cold Reg. Sci. Technol, 19 , 95–110.
Hobbie, S. E., J. P. Schimel, S. E. Trumbore, and J. R. Randerson, 2000: Controls over carbon storage and turnover in high-latitude soils. Global Change Biol, 6 , (Suppl. 1). 196–210.
Houghton, J. T., L. G. Meira Filho, B. A. Callandar, N. Harris, A. Kattenberg, and K. Maskell, 1996: Climate Change 1995: The Science of Climate Change,. Cambridge University Press, 584 pp.
Lunardini, V. J., 1981: Heat Transfer in Cold Climates. Van Nostrand Reinhold, 731 pp.
Lynch, A. H., W. L. Chapman, J. E. Walsh, and G. Weller, 1995: Development of a regional climate model of the western Arctic. J. Climate, 8 , 1555–1570.
Lynch, A. H., G. B. Bonan, F. S. Chapin III, and W. Wu, 1999a: The impact of tundra ecosystems on the surface energy budget and climate of Alaska. J. Geophys. Res, 104 , 6647–6660.
Lynch, A. H., F. S. Chapin III, L. D. Hinzman, W. Wu, E. Lilly, G. Vourlitis, and E. Kim, 1999b: Surface energy balance on the arctic tundra: Measurements and models. J. Climate, 12 , 2585–2606.
Mackay, J. R., 1995: Active layer changes (1968–1993) following the forest-tundra fire near Inuvik, N.W.T., Canada. Arct. Alp. Res, 27 , 323–336.
McGuire, A. D., J. M. Melillo, D. W. Kicklighter, and L. A. Joyce, 1995: Equilibrium responses of soil carbon to climate change: Empirical and process based estimates. J. Biogeogr, 22 , 785–796.
Morrill, J. C., K. J. Schaudt, and R. E. Dickinson, 1999: The lichen and moss algorithm: Improving the representation of the boreal forest in a land surface model. Preprints, 14th Conf. on Hydrology, Dallas, TX, Amer. Meteor. Soc., 400–401.
Nicholas, J. R. J., and K. M. Hinkel, 1996: Concurrent permafrost aggregation and degradation induced by forest clearing, Central Alaska, U.S.A. Arct. Alp. Res, 28 , 294–299.
Nisbit, E. G., 1989: Some northern sources of atmospheric methane: Production, history, and future implications. Can. J. Earth Sci, 123 , 2039–2046.
Oechel, W. C., and K. van Cleve, 1986: The role of bryophytes in nutrient cycling in the taiga. Forest Ecosystems in the Alaskan Taiga, K. Van Cleve et al., Eds., Springer-Verlag, 121–137.
Oechel, W. C., and G. L. Vourlitis, 1994: The effects of climate change on land–atmosphere feedbacks in arctic tundra regions. TREE, 9 , 324–329.
Oechel, W. C., S. J. Hastings, G. Vourlitis, M. Jenkins, G. Riechers, and N. Grulke, 1993: Recent change of Artic tundra ecosystems from a net carbon dioxide sink to a source. Nature, 361 , 520–523.
Post, W. M., W. R. Emanuel, P. J. Zinke, and A. G. Stangenberger, 1982: Soil carbon pools and world life zones. Nature, 298 , 156–159.
Serreze, M. C., A. H. Lynch, and M. P. Clark, 2001: The arctic frontal zone as seen in the NCEP–NCAR reanalysis. J. Climate, 14 , 1550–1567.
Sharrat, B. S., 1997: Thermal conductivity and water retention of a black spruce forest floor. Soil Sci, 162 , 576–582.
Stoutjesdijk, P., and J. J. Barkman, 1991: Microclimate, Vegetation and Fauna. Opulus Press, 216 pp.
Tans, P. P., I. Y. Fung, and T. Takahashi, 1990: Observational constraints on the global atmospheric CO2 budget. Science, 247 , 1431–1438.
Tilley, J. S., and A. H. Lynch, 1998: On the applicability of current land surface schemes for arctic tundra: An intercomparison study. J. Geophys. Res, 103 , 29 051–29 063.
van Cleve, K., C. T. Dyrness, L. Oliver, R. Schlentner, and L. A. Viereck, 1983: Productivity and nutrient cycling in taiga forest ecosystems. Can. J. For. Res, 13 , 747–766.
van Cleve, K., F. S. Chapin III, P. W. Flannagan, L. A. Viereck, and C. T. Dyrness, 1986: Forest Ecosystems in the Alaskan Taiga. Springer-Verlag, 230 pp.
Verseghy, D. L., 1991: CLASS—A Canadian land surface scheme for GCMS. I: Soil model. Int. J. Climatol, 11 , 111–113.
Viereck, L. A., K. Van Cleve, and C. T. Dyrness, 1986: Forest ecosystem distribution in the taiga environment. Forest Ecosystems in the Alaskan Taiga, K. Van Cleve et al., Eds., Springer-Verlag, 22–43.
Wilson, M. F., and A. Henderson-Sellers, 1985: A global archive of land cover and soils data for use in general circulation climate models. J. Climatol, 15 , 119–143.
Wilson, M. F., A. Henderson-Sellers, R. E. Dickinson, and P. J. Kennedy, 1987: Sensitivity of the biosphere–atmosphere transfer scheme (BATS) to the inclusion of variable soil characteristics. J. Climate Appl. Meteor, 26 , 341–362.
Moss and lichen growing in (a) a boreal white spruce forest and (b) a tussock tundra environment at Council, Alaska, (64°54.5′N, 163°40.5′W)
Citation: Journal of Climate 14, 15; 10.1175/1520-0442(2001)014<3324:TROASI>2.0.CO;2
Moss and lichen growing in (a) a boreal white spruce forest and (b) a tussock tundra environment at Council, Alaska, (64°54.5′N, 163°40.5′W)
Citation: Journal of Climate 14, 15; 10.1175/1520-0442(2001)014<3324:TROASI>2.0.CO;2
Moss and lichen growing in (a) a boreal white spruce forest and (b) a tussock tundra environment at Council, Alaska, (64°54.5′N, 163°40.5′W)
Citation: Journal of Climate 14, 15; 10.1175/1520-0442(2001)014<3324:TROASI>2.0.CO;2
The relationships between (a) matric potential (MPa) and the fraction saturated water content and (b) hydraulic conductivity (mm s−1) and fraction saturated water content, for each of the soil texture types used in the experiments
Citation: Journal of Climate 14, 15; 10.1175/1520-0442(2001)014<3324:TROASI>2.0.CO;2
The relationships between (a) matric potential (MPa) and the fraction saturated water content and (b) hydraulic conductivity (mm s−1) and fraction saturated water content, for each of the soil texture types used in the experiments
Citation: Journal of Climate 14, 15; 10.1175/1520-0442(2001)014<3324:TROASI>2.0.CO;2
The relationships between (a) matric potential (MPa) and the fraction saturated water content and (b) hydraulic conductivity (mm s−1) and fraction saturated water content, for each of the soil texture types used in the experiments
Citation: Journal of Climate 14, 15; 10.1175/1520-0442(2001)014<3324:TROASI>2.0.CO;2
The vertical profiles of the soil columns and constituent layers used for each of the five experiments
Citation: Journal of Climate 14, 15; 10.1175/1520-0442(2001)014<3324:TROASI>2.0.CO;2
The vertical profiles of the soil columns and constituent layers used for each of the five experiments
Citation: Journal of Climate 14, 15; 10.1175/1520-0442(2001)014<3324:TROASI>2.0.CO;2
The vertical profiles of the soil columns and constituent layers used for each of the five experiments
Citation: Journal of Climate 14, 15; 10.1175/1520-0442(2001)014<3324:TROASI>2.0.CO;2
The simulated vertical profiles of temperature through the six soil layers during Jan and Jul for the moss and loam experiments. The dotted vertical line represents the freezing point of water (273.15 K). Error bars illustrate ±1 std dev
Citation: Journal of Climate 14, 15; 10.1175/1520-0442(2001)014<3324:TROASI>2.0.CO;2
The simulated vertical profiles of temperature through the six soil layers during Jan and Jul for the moss and loam experiments. The dotted vertical line represents the freezing point of water (273.15 K). Error bars illustrate ±1 std dev
Citation: Journal of Climate 14, 15; 10.1175/1520-0442(2001)014<3324:TROASI>2.0.CO;2
The simulated vertical profiles of temperature through the six soil layers during Jan and Jul for the moss and loam experiments. The dotted vertical line represents the freezing point of water (273.15 K). Error bars illustrate ±1 std dev
Citation: Journal of Climate 14, 15; 10.1175/1520-0442(2001)014<3324:TROASI>2.0.CO;2
The simulated Jul vertical profiles of (a) thermal conductivity, (b) apparent heat capacity, (c) fraction saturated water content, (d) matric potential, (e) hydraulic conductivity, and (f) temperature, through the soil columns for each of the five experiments. The vertical depth axis is illustrated using a log10 scale
Citation: Journal of Climate 14, 15; 10.1175/1520-0442(2001)014<3324:TROASI>2.0.CO;2
The simulated Jul vertical profiles of (a) thermal conductivity, (b) apparent heat capacity, (c) fraction saturated water content, (d) matric potential, (e) hydraulic conductivity, and (f) temperature, through the soil columns for each of the five experiments. The vertical depth axis is illustrated using a log10 scale
Citation: Journal of Climate 14, 15; 10.1175/1520-0442(2001)014<3324:TROASI>2.0.CO;2
The simulated Jul vertical profiles of (a) thermal conductivity, (b) apparent heat capacity, (c) fraction saturated water content, (d) matric potential, (e) hydraulic conductivity, and (f) temperature, through the soil columns for each of the five experiments. The vertical depth axis is illustrated using a log10 scale
Citation: Journal of Climate 14, 15; 10.1175/1520-0442(2001)014<3324:TROASI>2.0.CO;2
The simulated mean daily summer (Jul) sensible, latent, and ground heat fluxes for all experiments. The ground evaporation for each experiment is also shown and is a subset of the total latent heat flux
Citation: Journal of Climate 14, 15; 10.1175/1520-0442(2001)014<3324:TROASI>2.0.CO;2
The simulated mean daily summer (Jul) sensible, latent, and ground heat fluxes for all experiments. The ground evaporation for each experiment is also shown and is a subset of the total latent heat flux
Citation: Journal of Climate 14, 15; 10.1175/1520-0442(2001)014<3324:TROASI>2.0.CO;2
The simulated mean daily summer (Jul) sensible, latent, and ground heat fluxes for all experiments. The ground evaporation for each experiment is also shown and is a subset of the total latent heat flux
Citation: Journal of Climate 14, 15; 10.1175/1520-0442(2001)014<3324:TROASI>2.0.CO;2
The different soil texture classes and the specified parameters for each texture type including the 11 original soil texture types in the NCAR LSM model along with their original percentage of sand, silt, and clay. Moss, lichen, and peat are additional soil texture types used in this study. Here Ksolids is the thermal conductivity of the soil solids, Csolids is the heat capacity of the soil solids, Ksat is the saturated hydraulic conductivity, ψsat is the saturated matric potential, θsat is the saturated volumetric water content, and b is the Clap and Hornberger constant
