Soil quality is fundamental in ensuring healthy forests. Mechanical operations such as logging may have a greater impact on forest floors if proper procedure is not followed. For example, minimizing soil compaction during harvesting and mechanical site preparation operations on forested lands is critical for maintaining the productive capacity of a site (Powers and others 2005). Compaction increases soil bulk density and soil strength, decreases water infiltration and aeration porosity, restricts root growth, increases surface runoff and erosion, and alters heat flux (Greacen and Sands 1980, Williamson and Nielsen 2000). Further, in forested ecosystems, the presence and availability of cations is governed through the interplay of numerous natural processes, including atmospheric additions, mineral weathering, soil formation, plant uptake and growth, forest stand dynamics, and leaching losses (Likens and others 1998). However, mounting evidence indicates that a variety of anthropogenic factors are altering biogeochemical cycles and depleting base cations such as calcium (Ca) and magnesium (Mg) from terrestrial ecosystems.

The sections that follow outline various indicators of soil quality and function. Impacts of wildfire and intentional management are reviewed as are potential approaches to monitoring the quality and functionality of forest soils. The effect calcium has on forest health and productivity is also assessed.

Government agencies, industrial landowners, and private landowners often strive to maintain soil quality after site management activities in order to maintain site productivity, hydrologic function, and ecosystem health. Soil disturbance resulting from timber harvesting, prescribed fire, or site preparation activities can cause declines, improvements, or have no effect on site productivity and hydrologic function. In many cases, detailed soil resource data can be used to determine the stress level and ecosystem health of stands and may be one method used to determine the risk of disease or insect outbreak. Currently, organic matter accumulations in many forests exceed historic levels. Fire suppression or fire exclusion has produced numerous overstocked stands. When this condition is combined with increased climatic variation, drought, and type conversion, these stands have a high risk for catastrophic wildfire. The resulting large, high-intensity, and high-severity fires could contribute to changes in soil quality and lead to outbreaks of insects and diseases in many ecosystems. Changes in ecosystem processes can also be associated with changes in overstory properties that alter forest stand resilience. For example, loss of western white pine to blister rust infection in the Northwestern United States has caused a type conversion to forest species that are not tolerant of root diseases, are not fire resistant, and sequester nutrients in the surface mineral soil and tree crown that can later be lost through logging or fire. These relationships, and others, can be used in conjunction with soil resource data bases to assess susceptibility to threats and to help develop management strategies to mitigate disturbances. Development of monitoring strategies that use common methods that can be utilized by a variety of land management agencies and specialists is a key component for relating forest health to soil changes after fire or other land management activities.

Soil quality and function are interrelated concepts that represent the range of soil properties and their associated ecological processes. The National Forest Management Act of 1976 and related legislation direct U.S. Department of Agriculture Forest Service managers to maintain the productivity potential of national forest land. The British Columbia Ministry of Forests uses professional assessment to evaluate the impacts of management practices on organic matter (OM) losses (British Columbia Ministry of Forests Forest Practices Code 1997). Even with these mandates and laws, the concept of soil productivity has not been well defined, and the impact of timber removal or fire on the productive potential of soils is not well understood or easily measured (Powers and others 2005). Soil quality has been defined as the capacity of a soil to function within an ecosystem to sustain biological productivity, maintain environmental quality, and promote plant and animal health (Doran and others 1996). In addition, soil health definitions include maintaining the integrity of nutrient cycling and resilience to disturbance or stress (O?Neill and others 2005). Tree or stand growth has often been used as an indicator of soil productivity changes, but growth reductions attributable to management practices may take >20 years to become manifest in many North American ecosystems (Morris and Miller 1994). The forest floor is likely a key element in maintaining healthy ecosystems, but it is also the one most impacted by fire and forest management (Tiedemann and others 2000). Maintaining site organic matter at or near the ecosystem baseline levels may help reduce nutrient losses (McNabb and Cromack 1990), insect (Fellin 1980b) and disease (McDonald and others 2000) outbreaks, and may ultimately reduce many forest health problems. For example, Page-Dumroese and Jurgensen (2006) describe baseline organic matter levels in 13 undisturbed forests around the Northwestern United States. The levels include measurements of downed wood, forest floor and mineral soil organic matter (OM), carbon (C) and nitrogen (N), and they can be used to determine when a site has excess or deficient organic matter stores. Carbon accumulation, as measured by forest floor depth or amounts of downed wood, can be a useful indicator of forest health because forests with OM levels above their historic baseline levels are at risk from increased insect and disease activities or high-intensity fires (Oliver and others 1994).

Active fire suppression during the 20^th century has increased OM volume on the soil surface in forest stands that historically had supported a regular fire-return interval (Oliver and others 1994). It has been suggested that active fire suppression, together with selective harvesting of seral species, has resulted in a shift in dominance to shade-tolerant Douglas-fir (Pseudotsuga menziesii) and grand fir (Abies grandis) in many western forests (Mutch and others 1993, Swetnam and others 1995), and a build-up of fuels in other forest types such as ponderosa pine (Covington and Sackett 1984, DeBano and others 1998). A consequence of the advance in succession in some Western forests and the suppression of fire in others is the increased accumulation of aboveground biomass and nutrients in standing live trees, standing dead trees, downed wood, and forest floor (Keane and others 2002, Major 1974). Particularly in the Western United States, this increase in OM biomass in fire-suppressed or fire-excluded forests has led to forest floor C accumulation far in excess of normal conditions. In the absence of fire, critical nutrients are tied up in this excess plant debris, possibly causing the site to become nutrient limited (Harvey 1994). Accumulations of woody residue and surface OM from fire suppression activities are also undesirable because of the increased risk from high-severity wildfires and slower OM decomposition rates (Covington and Sackett 1984).

Deep accumulations of organic material (those in excess of decomposition) are generally lost through fires (Oliver and others 1994). If the fires are frequent and of low severity, few organic matter (or nutrient) losses occur (Neary and others 1999, Page-Dumroese and Jurgensen 2006), but infrequent, high-severity fires can be catastrophic to soil productivity and forest health if significant amounts of biomass have accumulated (Habeck and Mutch 1973). Neary and others (1999) outline the threshold temperatures for biological disruptions in soils. The cumulative impact of a catastrophic fire may directly affect belowground processes because it can alter nutrient inputs (soil macro- and microfauna), increase soil temperatures, increase erosion, alter evapotranspiration rates, and decrease moisture availability (Neary and others 1999). These detrimental impacts may also exacerbate insect and disease outbreaks (Harvey and others 1989, Jurgensen and others 1990).

The Healthy Forest Restoration Act of 2003 was designed to help alleviate the accumulation of OM by using partial cuts and prescribed fires to remove small-diameter trees and surface OM from many forest stands. Prescribed fire and harvesting operations are important variables in determining soil OM losses because they both influence the removal of organic matter, C, and N on the soil surface and influence the amount of OM within the mineral soil profile. However, frequent repeated burns and multiple entries by mechanical equipment to reduce wildfire risk may impact ecosystem processes, soil quality and productivity, and site sustainability at a variety of scales, (e.g., a cutting unit or an entire watershed).

Prescribed fire, as a site preparation method or for underburning intact stands, is a major component of the restoration effort to reduce fuel levels in many forested ecosystems (McIver and Starr 2001). Prescribed fires produce a wide range of fire intensities, depending on fuel loads, fuel moisture content, slope position, and slope aspect (Brown and others 1991, Huffman and others 2001, Little and Ohmann 1988, Oswald and others 1999, Vose and others 1999). Fire severity is a term used to describe the impact of fire on both above- and below-ground stand components (DeBano and others 1998, Keane and others 2002). Various burn indices have been developed to evaluate fire effects on ecosystem processes and soil productivity (Neary and others 1999), but three classes are commonly used: low severity ?a nonlethal, low intensity surface fire in which patches of surface OM are lost, moderate severity?a patchy fire that creates a mixed mosaic of fire intensities and all small-diameter (<7.6 cm) OM is consumed, and high severity ?a stand replacement fire that kills more than 90 percent of the trees, consumes most surface OM, and some OM in the mineral soil has been lost (mineral soil changes color) (Keane and others 2002). Information is needed on the impact of prescribed fire on soil OM content and distribution to evaluate the effects of fire management practices on residual fuel loads, soil erosion potential, and long-term site productivity (Elliot 2003, Neary and others 2000).

Because of the range of fire conditions possible in any given stand, the range of soil conditions will also be variable (Landsberg 1994). Often, reduced productivity or health of the remaining stand is influenced by the amount of injury to remaining trees, crowns, and roots, reductions in microorganisms in the surface mineral soil, and changes in C and other nutrient pools (Klemmedson and Tiedemann 1995, Page-Dumroese and others 2003). In addition, the impacts of prescribed fire are dependent on forest type and past management (Schoennagel and others 2004). Historically, dry forests of the Western United States, (e.g., Pinus ponderosa, dry Pseudotsuga menziesii, etc.), which had a relatively short fire return interval and low fire intensity (Agee 1998), did not usually develop disease problems (like Armillaria spp) when the fire return interval remained short (McDonald and others 2000). However, because these dry stands are water limited and have relatively shallow forest floors (Page-Dumroese and Jurgensen 2006), as fire exclusion and suppression increase, so does the stress and competition between trees for limited water and nutrient resources (McDonald and others 2000).

The combined effects of harvest operations and prescribed burning on the remaining forest slash may severely impact mesofauna living in the forest floor by either directly killing them or removing their desired food source (Fellin 1980b). This includes both pests and beneficial insects. For instance, predators and parasites of spruce budworm that live in the forest floor may help to regulate budworm numbers at low levels so that existing populations do not reach epidemic proportions (Fellin 1980b). Stands affected by low-severity fires, which can leave many unburned areas of the forest floor, may provide a favorable location for maintaining important insects in the forest floor and ensuring that decomposition and nutrient cycling processes continue (Fellin 1980b).

Results of studies of repeated prescribed fires on soil quality and forest health are mixed. Annual or bi-annual prescribed fires have been shown to reduce pools of C, N, and sulfur (S) in the forest floor after 30 years (Binkley and others 1992). Burning on a 2-year interval for 20 years reduced N from both the forest floor and mineral soil (Wright and Hart 1997). However, a study of prescribed fire at intervals of 1, 2, 4, 6, 8, and 10 years in ponderosa pine resulted in an increase of soil C and N (Neary and others 2002). These long-term changes in nutrient status may or may not affect long-term site productivity (Jurgensen and others 1997), but on some sites they could affect insect and disease outbreaks (Harvey and others 1989). For example, the introduction of white pine blister rust (Cronartium ribicola) in the Western United States has reduced the number of 5-needled pines, (e.g., Pinus monticola or P. lambertiana) in many ecosystems (Monnig and Byler 1992). These pines, along with ponderosa pine (P. ponderosa) and western larch (Larix occidentalis) tend to be broadly adapted species (Rehfeldt 1990) and are relatively tolerant to many native pests. However, in the absence of fire, they are strongly reduced and are more susceptible to non-native pest outbreaks (Harvey 1994). This shift from more tolerant species has also reduced nutrient cycling within the surface organic matter and mineral soil (Harvey 1994). Typically, more carbon is held in aboveground biomass when there is a compositional shift toward less tolerant species, and this can result in more plant stress when available moisture is low (McDonald 1990).

Soil displacement (removal of surface organic and mineral soil) is most often measured by the amount of forest floor removed. Loss of surface OM either by equipment or through accelerated erosion may produce detrimental changes if it is moved off-site, is unavailable to tree roots, or if mineral soil removal results in exposing subsoil horizons. However, careful placement of harvesting and yarding layout in ground-based units could mitigate some detrimental displacement. For instance, McIver and others (2003) noted that when displacement along the edge of trails is included, (i.e., displacement caused by the harvester moving close to trails to cut logs) in the inventory, displacement can be extremely variable (5 percent-43 percent). However, if soil displacement along trail edges was excluded from the site inventory, displacement was no greater than USDA Forest Service guidelines of 15 percent. The report by McIver and others (2003) and the information in Curran and others (2005) both stress that it is imperative to have uniform terms for describing soil disturbance to improve our techniques for tracking the consequences of forest practices on soil productivity and forest health.

Often soil erosion is not a significant problem on slopes that have some soil cover. Using information about slope, amount of the hillside with some soil cover, and local precipitation values, Page-Dumroese and others (2000) noted that in many cases, soils with at least 50-percent soil cover did not produce more than 2-4 mg ha^-1 of soil erosion. In some cases, removal of soil from the upper slope to somewhere downslope may reduce upslope productivity, but increase downslope productivity. However, if soil is moved offsite, productivity is reduced permanently (Elliot and others 1998). Both onsite and offsite soil movement results in lower soil productivity for part of the slope because of loss of nutrients, water holding capacity, and rooting depth, and it may also impair forest health on that portion of the landscape. Combining soil-cover loss with compaction can accelerate erosion rates above the natural soil formation rates (Elliot and others 1998). Soil loss is only one problem associated with accelerated erosion. Often N, C, and cation exchange capacity are also moved offsite with the moving soil (Page-Dumroese and others 2000). Erosion rates are usually highest immediately after soil is disturbed mechanically or as a result of fire (Robichaud and Brown 1999).

Minimizing soil compaction during harvesting and mechanical site preparation operations on forested lands is critical for maintaining the productive capacity of a site (Powers and others 2005). Compaction increases soil bulk density and soil strength, decreases water infiltration and aeration porosity, restricts root growth, increases surface runoff and erosion, and alters heat flux (Greacen and Sands 1980, Williamson and Nielsen 2000). These changes can lead to substantial declines in tree growth and forest health (Froehlich and others 1986, Gomez and others 2002) or, conversely, have little impact (Powers and others 2005). Significant changes in soil physical properties occur more often on fine-textured soils than on coarse-textured soils (Page-Dumroese and others 2006), and knowing basic site conditions like texture and soil moisture content along with designating skid trails or providing operator training may help reduce undesirable soil conditions and maintain long-term productivity (Quesnel and Curran 2000).

Most forests are dependent on a variety of biological processes to regulate nutrients and cycle organic matter. For instance, forest diseases such as Armillaria and Annosus root diseases are a key ecosystem process to recycle carbon (Harvey 1994), but they can also expand to epidemic proportions if conditions are favorable. Trees not adapted to a site, wounded during thinning operations, growing on compacted soils, or in areas of disturbed hydraulic function, are at risk from both disease infection (Goheen and Otrosina 1998, Wiensczyk and others 1997) and insect outbreaks (Larsson and others 1983). In a study in loblolly pine plantations, Annosum was positively correlated with areas of higher bulk density that had stressed numerous trees (Alexander and others 1975). Low vigor, stressed forest stands are also susceptible to insect attacks (Larsson and others 1983). The recent outbreak of mountain pine beetle (Dendroctonus ponderosae) in British Columbia, Canada, and the Pacific Northwest of the United States have caused significant changes in water flow, soil moisture, and groundwater levels in many forest stands (Uunila and others 2006). These changes in water abundance and timing can affect the health of subsequent stands by changing annual water yields, times of low flow and peak flows (Uunila and others 2006).

The risks posed by atmospheric deposition to forest health are complex because of distinct regional patterns of deposition and temporal variations in air quality (McLaughlin and Percy 1999). However, air pollution stresses in North American forests and their impact on insect and disease incidence are a major consideration in both the United States and Canada (McLaughlin and Percy 1999), particularly in the eastern part of North America. One potential threat from changes in air quality on many forested lands is the altered nutrient balance and soil acidity (Adams and others 2000), which then affects insect and disease outbreaks. Soil impacts and threats to forest health can be caused by nitrogen (N) saturation (Aber and others 1989), depletion of basic cations due to increased leaching (Federer and others 1989), and altering nutrient availability (Robarge and Johnson 1992). Combined, all these nutrient stresses may cause long-term declines in site productivity (Likens and others 1996) and may increase the number and severity of disease or insect outbreaks (McLaughlin and Percy 1999). Altered soil processes by deposition of air pollutants also alters the way forests grow and respond to biotic and abiotic stresses within their regional environments. The potential for large-scale forest health changes becomes greater over time (McLaughlin and Percy 1999).

Current soil monitoring efforts primarily address changes in soil quality through measures of compaction, pH, water infiltration, hydrologic function, water availability, and plant-available nutrients, but do not address ecological function of a site (van Bruggen and Semenov 2000). Ecological function may be difficult to assess because of season, management, or climate variables. Potential approaches to assess ecological function can include techniques that extract DNA to determine microbial species composition (Zhou and others 1996), determine substrate utilization, (i.e., Biolog) to trace microbial communities (Bossio and Scow 1995), or decomposition of a wood substrate (Jurgensen and others 2006). In addition, levels of root pathogens or disease suppression organisms could also be considered as potential bioindicators of soil health (Hoeper and Alabouvette 1996, Visser and Parkinson 1992).

In North America, several forest monitoring programs, [i.e., Forest Health Monitoring (FHM, U.S.A.), Acid Rain National Warming System (ARNEWS, Canada), North American Maple Project (NAMP, joint United States and Canada)), Forest Inventory and Analysis (FIA, U.S.A.)] have been developed as a result of studies that indicated widespread changes in forest health (McLaughlin and Percy 1999). Many of these tree monitoring programs also include soil indicators in their protocols (O?Neill and others 2005). In addition, the Montréal Process (an international effort to develop criteria and indicators of forest sustainability) is working to provide a common framework for describing soil changes by using two important soil characteristics: significantly diminished organic matter and significant compaction (Montréal Process Working Group 1997). The USDA Forest Service is mandated to monitor management impacts on soil productivity, site resiliency, and long-term productivity (Johnson and Todd 1998, Jurgensen and others 1997, Page-Dumroese and others 2000) and is part of an international, multiagency effort to pursue correlation and common definitions of practical standards for soil disturbance and maintenance of forest productivity capacity (Curran and others 2005).

There are numerous tools available to monitor a forest site or soil for changes over time or after wildfire, prescribed fire, or mechanical disturbances. Together, forest and soil monitoring data can provide a backdrop for current conditions and trends in forest health. Successful forest and soil monitoring programs must:

? be simple to use

? have repeatable methods that are useable by nonspecialists

? provide meaningful results

? be comprised of collaborative efforts between both scientists and various government agencies

? provide a compelling link between soil properties and forest growth or health

Solutions to the current forest health problems are neither clear-cut nor easily managed (Tiedemann and others 2000). However, maintenance of the forest floor during prescribed fire or mechanical forest treatments will help to maintain soil quality, soil health, and ultimately forest health (Fellin 1980, Page-Dumroese and Jurgensen 2006). The indirect influence of changing microclimatic conditions of the forest floor will depend on the kinds of site treatment and the condition of the previous stand. Overcrowded and fire-suppressed stands will likely differ in their response to treatment than more open-grown stands. Similarly, stands with seral species will respond differently to nutrient, water, insect, and disease stressors than those stands occupied by less tolerant, narrowly adapted species (Harvey 1994).

Forest health can be successfully managed if we shift the focus from just ?tree health? to entire ecosystem health. Current levels of pest problems are not entirely due to a single organism, but the complex of environmental conditions and stand histories that determine stand resilience to pest outbreaks. Healthy and resilient forest soils can also help limit the extent of forest pest problems.

Growing evidence from around the globe indicates that anthropogenic factors including pollution-induced acidification, associated aluminum mobility, and nitrogen saturation are disrupting natural nutrient cycles and depleting base cations from forest ecosystems. Although cation depletion can have varied and interacting influences on ecosystem function, it is the loss of calcium (Ca) that may be particularly limiting to tree health and productivity. Ca plays unique roles in plant cell function, including environmental signal transduction processes that allow cells to sense and respond to stress. Considering this, Ca depletion could impair plant response systems and predispose trees to reduced growth and increased decline. Controlled experiments with red spruce and other tree species provide mechanistic support for the hypothesis that Ca deficiencies predispose trees to decline. Importantly, several examples of species declines in the field also suggest that injury is often greater when Ca depletion and stress exposure co-occur.

Connections between contemporary species declines and Ca depletion highlight the need for monitoring forests for indicators of change, including Ca loss. Direct measures of soil and plant Ca concentrations provide one traditional means of assessing that Ca status within forests. Although often valuable, substantial variation among soils and species and a lack of comparative historical data provide obstacles to the use of these measures for evaluating Ca depletion across the landscape. An alternative approach for assessing Ca depletion is to model critical loads and exceedances of pollutant additions that lead to net losses in Ca pools and likely disrupt ecosystem Ca cycles within forests. For example, spatial associations of Ca cycling and loss to broad-scale data on forest health and productivity were recently conducted for portions of the Northeastern United States. A steady-state ecosystem process model was coupled to extensive spatial databases and used to generate maps identifying forest areas likely to experience Ca depletion. Sustainable Ca supplies in forest ecosystems are functions of forest type, timber extraction intensity, prior land use, atmospheric deposition rates, and site factors including climate, hydrology, soil mineral type and weathering rates. Considering the unique vulnerability of Ca to leaching loss and its vital role in supporting tree stress response systems, the model focused on how changes in Ca pools may influence forest health conditions. Initial comparisons within New England indicated that the model-based Ca deficiency metric was a good predictor of field-based indicators of forest health and productivity. Models such as this show promise for evaluating the threat Ca depletion poses to forest health and productivity in an integrated and spatially explicit manner in North America. This approach has already proven valuable to policy makers and managers in Europe when evaluating alternative pollution reduction or mitigation options.

Cations are naturally occurring, positively charged elements that are important constituents of soils and surface waters and play unique and critical roles in biological systems. Among many functions, cations serve as important co-factors influencing the activity of biomolecules, act to modify charge balances within cells and organelles, and serve as signaling agents that help regulate cell physiology (Buchanan and others 2000, Marschner 2002). In forested ecosystems, the presence and availability of cations is governed through the interplay of numerous natural processes, including atmospheric additions, mineral weathering, soil formation, plant uptake and growth, forest stand dynamics, and leaching losses (Likens and others 1998). However, mounting evidence indicates that a variety of anthropogenic factors are altering biogeochemical cycles and depleting base cations such as calcium (Ca) and magnesium (Mg) from terrestrial ecosystems. Chief among these drivers of cation loss are processes directly or indirectly associated with atmospheric pollution.

Through industrial activity and the increased combustion of fossil fuels over the past century, humans have dramatically increased gaseous emissions of sulfur dioxide (SO₂), nitrogen oxides (NO_x), and ammonia (NH₃) and particulate emissions of acidifying compounds (Driscoll and others 2001). Recent pollution controls have reduced emissions of sulfur (S) -based compounds in Europe and North America, resulting in moderate reductions in S deposition, but there has been little change in nitrogen (N) deposition (Driscoll and others 2001, UNECE 2005). In contrast to North America and Europe, with rapid economic development and economic growth Asia? and most notably China?have significantly increased fossil fuel combustion in recent years (Liu and Diamond 2005). As a result, emissions of SO₂, NO_x,, NH₃, and associated compounds have increased greatly in the region (Carmichael and others 2002, Liu and Diamond 2005, Richter and others 2005). In fact, pollutant deposition of S and N compounds now affects a quarter of China?s land area, making China one of the countries most affected by these pollutants (Feng and others 2002, Jianguo and Diamond 2005).

Through the atmospheric conversions of SO₂ and NO_x to the acids H₂SO₄ and HNO₃, as well as the release of H⁺ during the oxidation of NH₄⁺ by soil microbes, S- and N-based pollutants act to acidify forest systems (Driscoll and others 2001). Among other impacts, this acidification increases the leaching of base cations from soils ( Kirchner and Lydersen 1995, Likens and others 1996, Likens and others 1998, Schulze 1989), and enhances the availability of aluminum (Al), which reduces base cation availability for plant uptake (Cronan and Scholield 1990, Lawrence and others 1995). In addition to the atmospheric production of acids from pollutant constituents, N inputs can lead to N saturation (the availability of N in excess of biological demand), which can deplete cations as excess N leaches from forest soils (Aber and others 1998). It has even been hypothesized that pollution-associated climatic warming could enhance rates of natural acidifying process, further exacerbating soil cation loss (Tomlinson 1993). In addition to pollution-associated cation loss, a side effect of existing pollution controls has been the reduced emission of particulates that contain base cations such as Ca (Hedin and others 1994). Reduced inputs and increased removals of cations from forests have resulted in net depletions that have been documented in a variety of ways, including long-term changes in stream chemistry, the analysis of archived soils, and the chemical analysis of tree xylem cores.

Long-term data of stream water chemistry at watersheds such as those at the Hubbard Brook Experimental Forest, New Hampshire, have documented changes consistent with the pollution-induced leaching of base cations from soils (Likens and others 1996, Likens and others 1998). Early stream monitoring revealed an increasing flush of base cations (Ca and Mg) that paralleled SO₄^2- and NO^3- concentrations?evidence that pollutant inputs were leaching stored soil cations into surface water soils (Likens and others 1996, Likens and others 1998). However, after 1970, mass balance calculations identified ever-reducing concentrations of cations, particularly Ca, coincident with decreases in SO₄^2- and NO^3-?a pattern suggesting the depletion of available cations following long-term leaching (Likens and others 1996, Likens and others 1998). The connection between cation losses and pollutant inputs was reinforced by data indicating that these same trends in stream chemistry occur in old-growth forests where the potentially confounding effects of land use disturbance (and associated acidification) were avoided (Martin and others 2000). Furthermore, European data indicates that the largest losses of Ca and Mg occur at sites with the most acid loading (Kirchner and Lydersen 1995).

Calculated reductions in soil cation storage inferred from the chemical analysis of stream water have recently been bolstered by studies from the United States and Europe that directly measured reductions in soil Ca storage following long-term exposures to acidic deposition. Bailey and others (2005) measured the cation concentration of soils at four forested sites in the Allegheny Mountains of Pennsylvania in 1997 and compared these to data from archived soil samples from these same sites collected in 1967. At all four sites there were significant reductions in Ca and Mg concentrations and pH over the two sample periods, and, at most sites, documented losses of Ca and Mg were much larger than could be accounted for by biomass accumulation?suggesting leaching losses as a more likely cause. In a separate analysis, Lawrence and others (2005) measured the cation contents of soil samples collected in 1926, 1964, and 2001 near St. Petersburg, Russia. They found that concentrations of exchangeable Ca in the upper 30 cm of soil decreased about tenfold from 1926 to 1964 but remained stable thereafter. In contrast, exchangeable Al showed a small decrease in the upper 10 cm of soil from 1926 to 1964, but a tenfold increase in the upper 30 cm from 1964 to 2001. They interpreted these results as reflecting a two-stage acidification process: (1) from 1926 to 1964 when inputs of acidity were neutralized by the replacement of exchangeable Ca by H, and (2) from 1964 to 2001 when the neutralizing of continued acidic inputs shifted from cation exchange to weathering of solid phase Al (Lawrence and others 2005). Here, too, changes in soil Ca concentrations were not attributable to biomass accumulation of Ca, but appeared better related to pollution-induced soil Ca depletion.

Consistent with measured reductions in soil Ca, several studies have noted reduced Ca concentrations in the stemwood of trees following the advent of elevated pollution loading (Bondietti and others 1990, Likens and others 1996, Shortle and others 1995). An initial increase in Ca concentration is often noted within wood for the decades with the greatest increases in acidic deposition that likely mobilized soil cations, increasing their availability for root uptake and leaching loss, (e.g., Shortle and others 1995). However, the reduction in stemwood Ca in recent decades may better reflect the long-term depletion of Ca from soils (Shortle and others 1995). Importantly, reductions in Ca concentrations within wood also suggest that pollution-induced changes in soil Ca levels are being transferred to living organisms.

In addition to pollution-associated depletion, tree harvests have the potential to exacerbate cation depletion within forests if they contribute to net cations losses that exceed long-term inputs (Adams 1999, Federer and others 1989, , Huntington 2000, Mann and others 1988, Nykvist 2000). Sequestration in above-ground woody biomass is an important cation sink within forest systems (Federer and others 1989, Mann and others 1988), and this is particularly true for Ca, which is highly concentrated in woody cell walls (Marschner 2002). Because of this, tree harvests can lead to the disproportionate removals of Ca relative to other cations (e.g., Adams 1999, Federer and others 1989). Harvests can also affect nutrient cycling through increased site acidification and leaching (Federer and others 1989), and reduced stocking following harvest may diminish stand-level transpiration and associated Ca uptake, further promoting Ca loss via leaching (Hornbeck and others 1993). In addition, varying methods of harvest can differentially alter Ca loss. For example, in one study, whole-tree (stems and branches) harvests removed up to 530 kg/ha, whereas sawtimber sales (bole wood only) removed about 442 kg/ha (Mann and others 1988). The frequency of tree harvest may also influence overall cation removal. Calculations from one study estimated a 15 percent loss of Ca due to leaching even with no harvest, a 28 percent loss of Ca with one harvest (at 80 years), and a 41 percent Ca loss for an equal intensity harvest performed in two stages: once at 40 years and once at 80 years (Adams 1999).

Although the depletion of any of the essential base cations can have varied and interacting influences on ecosystem function, the loss of Ca may be particularly limiting to tree health because the unique distribution and physiology of Ca suggests that the depletion of this key cation could specifically weaken plant stress response systems and predispose trees to decline.

In contrast to many cations, Ca is highly compartmentalized within plant cells and tissues, and this partitioning is a defining characteristic of its physiological function. Although Ca is an essential micronutrient, it is toxic in its free form within the cell cytoplasm because it precipitates with inorganic phosphate, (e.g., Bush 1995, Knight 2000). Thus, in order to assure phosphate availability for energy metabolism and other essential processes, Ca is actively pumped from the cytoplasm and is sequestered in inaccessible locations and chemical forms, including insoluble oxalate crystals outside the plasma membrane (Fink 1991). Because Ca can only exist in very low concentration in the cytoplasm, it is functionally immobile in the phloem, (which relies on cytoplasmic transport). Thus, unique to other cations, Ca cannot be redistributed within plants to overcome localized deficiencies.

Localized concentrations of Ca support at least two important functions: (1) they add to the structural stability of cell walls and membranes, and (2) labile Ca is a key constituent in the pathway that allows cells to sense and respond to environmental stimuli and change (Marschner 2002). This second function appears particularly relevant to tree health concerns relevant to Ca depletion. Ca serves as an important second messenger in the perception and transduction of environmental and stress signals (Bush 1995, Pandey and others 2000, Roos 2000, Sanders and others 1999). Because extremely little free Ca exists in the cytoplasm of cells, environmental stimuli that temporarily alter the permeability of the plasma membrane allow labile Ca to flow into cells along a steep concentration gradient (Sanders and others 1999). Once in the cytoplasm, Ca quickly binds to Ca-specific proteins such as calmodulin, which then initiate a chain of physiological modifications, (e.g., changes in enzyme activity, gene transcription, etc.) that help cells adjust to the environmental conditions that triggered the response cascade. This entry of Ca into the cytoplasm acts as a messenger of environmental information for cells and appears to be an essential first step in triggering a wide range of physiological responses needed by plants to successfully adjust to environmental change or defend against pests and pathogens. Numerous independently conducted studies have concluded that Ca plays a critical message perception and transduction role in response to an array of environmental stresses, including low temperature (DeHayes and others 1997, 1999; Monroy and others 1993), drought (Sheen 1996), fungal infections (Hebe and others 1999), and insect infestations (McLaughlin and Wimmer 1999).

Schematic representation

Given the fundamental role Ca plays in plant stress response systems, biological Ca depletion could create a scenario analogous to the suppression of animal immune systems (Schaberg and others 2001). For example, there are numerous circumstances, (e.g., HIV infection, chemotherapy treatment, etc.) that impair the normal function of human immune systems. An immuno-compromised person may appear, feel, and ostensibly function as if they were healthy. Nonetheless, when exposed to a disease agent, they can experience declines in health that are exaggeratedly large relative to a person with a fully functioning immune system. In this same way, it is possible that depletions of biologically available Ca could suppress the ability of plants to adequately sense and respond to changes in their surroundings and make them more vulnerable to decline. This suppression would predispose plants to disproportionate decline following exposure to perhaps even normal levels of stress, (e.g., pathogens or drought) that would otherwise pose no catastrophic threat if biological response systems were fully functional. Importantly, under this scenario plants might initially appear to be normal and healthy even though their biological response systems were compromised (Schaberg and others 2001).

Although based on basic understandings of the distribution and physiology of Ca in plants, experimental evidence that Ca deficiencies could reduce stress tolerance in trees has only recently surfaced. This evidence was first documented for the well-studied phenomenon of winter injury in red spruce (Picea rubens Sarg.), but was later shown to be relevant to other tree species and stresses other than freezing injury.

Red spruce winter injury is the reddening and mortality of the foliage in late winter followed by its abscission in late spring (DeHayes 1992). Injury is caused by freezing and is likely the result of various stresses, including low temperatures (DeHayes and others 1990), freeze-thaw cycles (Hadley and Amundson 1992, Lund and Livingston 1998), and rapid freezing (Perkins and Adams 1995). The current-year foliage of red spruce is more vulnerable to injury than older foliar age classes or foliage from sympatric species because it is less cold tolerant (DeHayes and others 1990). In addition, certain anthropogenic inputs such as acidic or prolonged N deposition can further reduce foliar cold tolerance and increase the risk of freezing injury (Schaberg and DeHayes 2000). Heavy foliar loss and potential bud mortality due to winter injury disrupts the carbon economies of trees, leading to growth declines and potential mortality (DeHayes 1992, Lazarus and others 2004). Winter injury was linked to the widespread decline of red spruce observed in the Northeastern United States from the 1960s through the 1980s (Friedland and others 1984, Johnson 1992), and severe winter injury events persist within the region (Lazarus and others 2004).

Beginning in the late 1980s, a series of studies were published showing that acid mist exposure significantly reduced the cold tolerance of red spruce current-year foliage, increasing the risk of foliar winter injury, (e.g., DeHayes and others 1991, Fowler and others 1989, Vann and others 1992). The physiological mechanism for this acid-induced reduction in cold tolerance remained unresolved, however, until a new method for measuring Ca specifically associated with cellular membranes was used in conjunction with controlled acid mist exposure experiments. Using these new methods for measuring membrane-associated Ca (mCa), a series of experiments documented that acid mists preferentially leached mCa from the outside of mesophyll cells, whereas other cations and forms of Ca were leached less, presumably because they were concentrated within the protective membrane barrier of cells (DeHayes and others 1999, Jagels and others 2002, Jiang and Jagels 1999, Schaberg and DeHayes 2000,Schaberg and others 2000). Furthermore, these studies showed that this loss of mCa destabilized membranes, depleted a source of Ca needed for stress signaling, reduced foliar cold tolerance, and predisposed trees to the secondary freezing injury responsible for decline (DeHayes and others 1999, Schaberg and DeHayes 2000, Schaberg and others 2000). Later work verified that soil-based Ca depletion initiated the same mechanistic sequence of physiological disruptions documented for foliar Ca leaching (Schaberg and others 2002).

The bulk of experimental evidence elucidating the influence of Ca depletion on tree nutrition and stress response has involved winter freezing injury of red spruce. However, recent evidence indicates that the same basic mechanism of physiological disruption documented for this species and syndrome are pertinent to other tree species and stressors. For example, Schaberg and others (2001) treated red spruce, eastern hemlock (Tsuga Canadensis (L.) Carr.), balsam fir (Abies balsamea (L.) Mill.), and eastern white pine (Pinus strobus L.) seedlings with acid mist and compared the nutritional and physiological responses of the newly evaluated species to those well-documented for red spruce. Although there was insufficient tissue to make all measurements on each species, results showed that acid mist reduced mCa levels (in eastern hemlock), decreased cell membrane stability (in balsam fir), and reduced foliar cold tolerance (in white pine) similar to red spruce (Schaberg and others 2001). In a separate experiment with red spruce, Borer and others (2005) examined the influence of acid mist exposure on stomatal closure following tissue desiccation? a stress response to drought that is also dependent on Ca signaling (Knight 2000). Red spruce seedlings were exposed to pH 3 or 5 mists and then measured for foliar Ca concentrations and rates of stomatal closure as foliage desiccated following shoot harvest. As with past experiments, acid mist exposure reduced the Ca available in foliage, but here the loss of Ca was also accompanied by a 15 percent slower rate of stomatal closure as tissues desiccated (Borer and others 2005). Results of experimental trials like this support the theory that anthropogenic Ca depletion could deplete biological Ca pools enough to suppress stress response systems and predispose trees to decline.

Controlled experiments like the ones outlined above have provided valuable insights into the biological mechanism through which Ca depletion may influence tree physiology and health. However, such studies by themselves do little to inform us of the threat Ca depletion may pose to native forests. Instead, evidence from numerous field studies has supported experimental findings and implicated Ca depletion as a contributing factor in the real-world decline of tree species in the United States and abroad.

The role that foliar winter injury has played in red spruce decline in Northeastern United States and adjacent Canada has been long understood (see DeHayes 1992). In addition, experimental evidence has provided a detailed understanding of the mechanism through which acid deposition can deplete biologically available Ca and predispose foliage to damage (see previous section). However, it was only following the severe winter injury event of 2003 that evidence indicated that acid deposition exposure in the field influences winter injury expression across the landscape. Lazarus and others (2006) measured the degree of foliar winter injury among dominant and codominant red spruce trees at multiple elevations (plots) at 23 sites in Vermont and adjacent States and used regression analyses to evaluate how injury varied with plot elevation, latitude, longitude, slope, and aspect. They found that injury was significantly greater in western portions of the study area, west-facing slope, and higher elevations? areas that have historically received higher levels of acidic and N deposition (Lazarus and others 2006). Although these findings support the hypothesis that acidic or N deposition or both act on a landscape scale to exacerbate winter injury, it was an ancillary evaluation that more specifically implicated Ca depletion as a modifier of injury expression in 2003. Hawley and others (2006) measured foliar nutrition and winter injury of red spruce on two watersheds at the Hubbard Brook Experimental Forest in New Hampshire: one a reference watershed that has undergone considerable Ca loss attributed to acid deposition-induced leaching (Likens and others 1996, Likens and others 1998), and another watershed that was fertilized with CaSiO3 in 1999 to replace lost Ca. Dominant and codominant red spruce on the Ca-addition watershed had significantly more Ca in their foliage and experienced about one-third the foliar injury of comparable trees on the reference watershed (Hawley and others 2006).

Sugar maple (Acer saccharum Marsh.) decline has been documented throughout parts of the Northeastern United States and Quebec over many recent decades (Allen and others 1992a, Kelley 1988, Mader and Thompson 1969, Wilmot and others 1995). These declines have been characterized using various measures, including crown deterioration, increased leaf chlorosis, and reduced growth. Stress factors such as drought (Payette and others 1996), freezing (Robitaille and others 1995), and insect defoliation (Allen and others 1992b) have been implicated with the decline and mortality of sugar maple. Regardless of stressor, decline has also been associated with deficiencies or imbalances of various elements including N, phosphorous (P), K, Mg, manganese (Mn), or Ca (Bernier and Brazeau 1988, Horsley and others 2000, Mader and Thompson 1969, Ouimet and Fortin 1992, Paré and Bernier 1989, Wilmot and others 1995). Although the specific elements associated with decline can vary among sites, deficiencies in Ca have been highlighted as a potential contributor to sugar maple decline in recent studies throughout the region, (e.g., Ellsworth and Liu 1994, Ouimet and Camiré 1995, Wilmot and others 1996), in part because experimental additions of Ca or lime or both have been shown to reduce decline symptoms (Long and others 1997, Moore and others 2000, Wilmot and others 1995).

Proposed mechanism

Consistent with these observations, Schaberg and others (2001) hypothesized that sugar maple decline may be another example of Ca depletion?s influence on tree stress response systems and health. Variations in maple decline symptoms (crown condition and basal area growth) coinciding with differences in soil Ca status across a range of sites are consistent with thi