Search

Understanding the transport of dissolved organic carbon (DOC) and nitrogen (N) as water flows through headwater basins is important for predicting downstream water quality. With increased recognition of climatic impact on nutrient transport, more studies are needed in headwater basins experiencing a Mediterranean-type climate, such as those of the Sierra Nevada, California. We analyzed water samples collected over 5 years from eight low-order and mixed-conifer watersheds to elucidate the temporal variation of water chemistry and evaluate their responses to prolonged drought and low-intensity forest thinning. We observed higher stream DOC concentrations in October compared to other months within water years prior to drought and thinning, suggesting the importance of antecedent moisture conditions on seasonal C export. In unthinned watersheds, stream DOC concentrations were lower (62%) and DOC aromaticity was higher (68 and 92%, depending on the index used) during drought compared to non-drought years. In thinned watersheds during drought years, stream water had higher DOC concentrations (66–94% in three consecutive years following thinning) and dissolved inorganic N (24%, in the third year following thinning) compared to unthinned watersheds during drought. Additionally, lower stream DOC concentrations were found in watersheds with higher elevations and lower drainage densities in the year with near-average precipitation; however, these correlations were not significant in years with greater or extremely low precipitation. Taken together, our results suggest that stream concentrations of DOC and dissolved N in Mediterranean headwater basins are extremely variable over time due to the high temporal climatic variabilities and periodic management practices.

As the largest biogeochemically active terrestrial reserve of carbon (C), soils have the potential to either mitigate or amplify rates of climate change. Ecosystems with large C stocks and high rates of soil C sequestration, in particular, may have outsized impacts on regional and global C cycles. Montane meadows have large soil C stocks relative to surrounding ecosystems. However, anthropogenic disturbances in many meadows may have altered the balance of C inputs and outputs, potentially converting these soils from net C sinks to net sources of C to the atmosphere. Here, we quantified ecosystem-level C inputs and outputs to estimate the annual net soil C flux from 13 montane meadows spanning a range of conditions throughout the California Sierra Nevada. Our results suggest that meadow soils can be either large net C sinks (577.6 ± 250.5 g C m−2 y−1) or sources of C to the atmosphere (− 391.6 ± 154.2 g C m−2 y−1). Variation in the direction and magnitude of net soil C flux appears to be driven by belowground C inputs. Vegetation species and functional group composition were not associated with the direction of net C flux, but climate and watershed characteristics were. Our results demonstrate that, per unit area, montane meadows hold a greater potential for C sequestration than the surrounding forest. However, legacies of disturbance have converted some meadows to strong net C sources. Accurate quantification of ecosystem-level C fluxes is critical for the development of regional C budgets and achieving global emissions goals.

Subsoil microbiomes play important roles in soil carbon and nutrient cycling, yet our understanding of the controls on subsoil microbial communities is limited. Here, we investigated the direct (mean annual temperature and precipitation) and indirect (soil chemistry) effects of climate on microbiome composition and extracellular enzyme activity throughout the soil profile across two elevation-bioclimatic gradients in central California, USA. We found that microbiome composition changes and activity decreases with depth. Across these sites, the direct influence of climate on microbiome composition and activity was relatively lower at depth. Furthermore, we found that certain microbial taxa change in relative abundance over large temperature and precipitation gradients only in specific soil horizons, highlighting the depth dependence of the climatic controls on microbiome composition. Our finding that the direct impacts of climate are muted at depth suggests that deep soil microbiomes may lag in their acclimation to new temperatures with a changing climate.

Biogeochemical theory and a substantial body of empirical data show that nitrogen (N), an atmospherically derived nutrient, limits plant growth on young substrates, while phosphorus (P), a rock-derived nutrient, limits plant growth on old substrates. In arid regions, water is also often a limiting resource to plant growth. We applied resource amendments of N, P, N + P, and water to blue grama (Bouteloua gracilis) growing on a 1 ky-old basaltic cinder substrate to test the hypothesis that N and water limit aboveground net primary production (ANPP) in a semi-arid climate, early in soil development. Contrary to our hypothesis, ANPP did not differ among treatments, suggesting that none of the resource amendments were limiting to blue grama growth. Unamended aboveground tissue N and P concentrations were three to five times lower at the 1 ky-old site than on older (55–3000 ky-old) substrates, suggesting differences in nutrient use efficiency across the substrate age gradient.

Soils are important sources and sinks of three greenhouse gases (GHGs): carbon dioxide (CO2), methane (CH4), and nitrous oxide (N2O). However, it is unknown whether semiarid landscapes are important contributors to global fluxes of these gases, partly because our mechanistic understanding of soil GHG fluxes is largely derived from more humid ecosystems. We designed this study with the objective of identifying the important soil physical and biogeochemical controls on soil GHG fluxes in semiarid soils by observing seasonal changes in soil GHG fluxes across a three million year substrate age gradient in northern Arizona. We also manipulated soil nitrogen (N) and phosphorus availability with 7 years of fertilization and used regression tree analysis to identify drivers of unfertilized and fertilized soil GHG fluxes. Similar to humid ecosystems, soil N2O flux was correlated with changes in N and water availability and soil CO2 efflux was correlated with changes in water availability and temperature. Soil CH4 uptake was greatest in relatively colder and wetter soils. While fertilization had few direct effects on soil CH4 flux, soil nitrate was an important predictor of soil CH4 uptake in unfertilized soils and soil ammonium was an important predictor of soil CH4 uptake in fertilized soil. Like in humid ecosystems, N gas loss via nitrification or denitrification appears to increase with increases in N and water availability during ecosystem development. Our results suggest that, with some exceptions, the drivers of soil GHG fluxes in semiarid ecosystems are often similar to those observed in more humid ecosystems.

The current paradigm of plant nutrient limitation during ecosystem development predicts a change from nitrogen (N) limitation when substrates are young to phosphorus (P) limitation when substrates are old. However, there are surprisingly few direct tests of this model. We evaluated this theory experimentally along a three million year semi-arid substrate age gradient using resource additions to intercanopy spaces dominated by the C4 bunchgrass Bouteloua gracilis. Unlike other gradients in subtropical and temperate ecosystems, soil water availability also increases strongly across this semi-arid system due to finer texture with substrate age. We found that aboveground net primary production (ANPP) of B. gracilis was limited by both water and N on the 55 ky substrate; not limited by N, P, or water on the 750 ky substrate; and limited by P alone on the 3000 ky substrate. Notably, measures of foliar nutrient concentration and N:P mass ratios were unable to predict nutrient limitations in these semi-arid systems. In unamended plots, mean ANPP declined dramatically at 3000 ky compared to the younger substrate age sites, presumably due to progressive limitation by P. This decline in ANPP late in ecosystem development is consistent with a reduction in soil total carbon and N storage at this site and provides a mechanism for successional retrogression in ecosystem structure and function. Our results unify biogeochemical theory across disparate ecosystems while illustrating the important water-nutrient interactions in these semi-arid ecosystems to further define the nature of nutrient limitations in terrestrial ecosystems.

The fire regime of ponderosa pine forests in the southwestern United States has shifted over the past century from historically frequent, low-intensity surface fires to infrequent, stand-replacing crown fires. We quantified plant and soil carbon (C) responses to this new fire regime and assessed interactions between changes in fire regime and changes in precipitation regime predicted by some climate models (specifically, an earlier monsoon rain season). We hypothesized that soil C pools and carbon dioxide (CO2) efflux rates would decrease initially following stand-replacing fires (due to low plant C inputs and the loss of the soil surficial organic (O) horizon), but then increase with time-after-fire (as plant C inputs increase). Water availability often limits soil biological activity in these forests, but we predicted that low soil C availability following fire would constrain soil CO2 efflux responses to precipitation. In a series of sites with histories of stand-replacing fires that burned between 2 and 34 years prior to sampling, burned patches had lower soil C pools and fluxes than adjacent unburned patches, but there was no evidence of a trend with time-after-fire. Burned forests had 7,500 g C m−2 less live plant biomass C (P < 0.001), 1,600 g C m−2 less soil total C (P < 0.001) and 90 g C m−2 less soil labile C (P < 0.001) than unburned forests. Lower soil labile C in burned patches was due to both a loss of O horizon mass with fire and lower labile C concentrations (g labile C kg−1 soil total C) in the mineral soil. During the annual drought that precedes summer monsoon rains, both burned and unburned patches had soil CO2 efflux rates ranging from 0.9 to 1.1 g CO2-C m−2 day−1. During the monsoon season, soil CO2 efflux in unburned patches increased to approximately 4.8 g CO2-C m−2 day−1 and rates in paired burned patches (3.4 g CO2-C m−2 day−1) were lower (P < 0.001). We also used field irrigation to experimentally create an earlier and longer monsoon season, and soil CO2 efflux rates at both burned and unburned plots increased initially in response to watering, but decreased to below control (plots without irrigation) rates within weeks. Watering did not significantly change cumulative growing season soil CO2 efflux, supporting our prediction that C availability constrains soil CO2 efflux responses to precipitation. This research advances our understanding of interactions among climate, fire, and C in southwestern forests, suggesting that climate-induced shifts toward more stand-replacing fires will decrease soil C for decades, such that a single fire can constrain future soil biological responses to precipitation regime changes.

Pinyon pine (Pinus edulis Engelm.) is an important tree species in the western United States that has experienced large-scale mortality during recent severe drought. The influence of soil conditions on pinyon pine response to water availability is poorly understood. We investigated patterns of tree mortality and response of tree water relations and growth to experimental water addition at four sites across a three million year soil-substrate age gradient. We measured recent pinyon mortality at four sites, and tree predawn water potential, leaf carbon isotope signature, and branch, leaf, and stem radial growth on 12 watered and unwatered trees at each site. Watered trees recieved fifty percent more than growing season precipitation for 6 years. Substrate age generally had a greater effect on tree water stress and growth than water additions. Pinyon mortality was higher on intermediate-aged substrates (50–55%) than on young (15%) and old (17%) substrates, and mortality was positively correlated with pinyon abundance prior to drought. These results suggest high soil resource availability and consequent high stand densities at intermediate-age substrates predisposes trees to drought-induced mortality in semi-arid regions. The response of tree water relations to water addition was consistent with the inverse texture hypothesis; watering reduced tree water stress most in young, coarsely textured soil, likely because water rapidly penetrated deep in the soil profile where it was protected from evapotranspiration.

Background and aims Soils can act as agents of natural selection, causing differential fitness among genotypes and/or families of the same plant species, especially when soils have extreme physical or chemical properties. More subtle changes in soils, such as variation in microbial communities, may also act as agents of selection. We hypothesized that variation in soil properties within a single river drainage can be a selective gradient, driving local adaptation in plants. Methods Using seeds collected from individual genotypes of Populus angustifolia James and soils collected from underneath the same trees, we use a reciprocal transplant design to test whether seedlings would be locally adapted to their parental soil type. Results We found three patterns: 1. Soils from beneath individual genotypes varied in pH, soil texture, nutrient content, microbial biomass and the physiological status of microorganisms. 2. Seedlings grown in local soils experienced 2.5-fold greater survival than seedlings planted in non-local soils. 3. Using a composite of height, number of leaves and leaf area to measure plant growth, seedlings grew ∼17.5% larger in their local soil than in non-local soil. Conclusions These data support the hypothesis that variation in soils across subtle gradients can act as an important selective agent, causing differential fitness and local adaptation in plants.

Severe wildfire may cause long-term changes in the soil-atmosphere exchange of carbon dioxide and methane, two gases known to force atmospheric warming. We examined the effect of a severe wildfire 10 years after burning to determine decadal-scale changes in soil gas fluxes following fire, and explored mechanisms responsible for these dynamics. We compared soil carbon dioxide efflux, methane uptake, soil temperature, soil water content, soil O horizon mass, fine root mass, and microbial biomass between a burned site and an unburned site that had similar stand conditions to the burned site before the fire. Compared to the unburned site, soil carbon dioxide efflux was 40% lower and methane uptake was 49% higher at the burned site over the 427-day measurement period. Soil O horizon mass, microbial biomass, fine root mass, and surface soil water content were lower at the burned site than the unburned site, but soil temperature was higher. A regression model showed soil carbon dioxide efflux was more sensitive to changes in soil temperature at the burned site than the unburned site. The relative importance of methane uptake to carbon dioxide efflux was higher at the burned site than the unburned site, but methane uptake compensated for only 1.5% of the warming potential of soil carbon dioxide efflux at the burned site. Our results suggest there was less carbon available at the burned site for respiration by plants and microbes, and the loss of the soil O horizon increased methane uptake in soil at the burned site.

Soils are one of the first selective environments a seed experiences and yet little is known about the evolutionary consequences of plant-soil feedbacks. We have previously found that plant phytochemical traits in a model system, Populus spp., influence rates of leaf litter decay, soil microbial communities and rates of soil net nitrogen mineralization. Utilizing this natural variation in plant-soil linkages we examined two related hypotheses: (1) Populus angustifolia seedlings are locally adapted to their native soils; and (2) Soils act as agents of selection, differentially affecting seedling survival and the heritability of plant traits. We conducted a greenhouse experiment by planting seedlings from 20 randomly collected P. angustifolia genetic families in soils conditioned by various Populus species and measured subsequent survival and performance. Even though P. angustifolia soils are less fertile overall, P. angustifolia seedlings grown in these soils were twice as likely to survive, grew 24% taller, had 27% more leaves, and 29% greater above-ground biomass than P. angustifolia seedlings grown in non-native P. fremontii or hybrid soils. Increased survival resulted in higher trait variation among seedlings in native soils compared to seedlings grown in non-native soils. Soil microbial biomass varied significantly across soil environments which could explain more of the variation in seedling performance than soil texture, pH, or nutrient availability, suggesting strong microbial interactions and feedbacks between plants, soils, and associated microorganisms. Overall, these data suggest that a “home-field advantage” or a positive plant soil feedback helps maintain genetic variance in P. angustifolia seedlings.

Understanding covariance of plant genetics and soil processes may improve our understanding the role of plant genetics in structuring soils and ecosystem function across landscapes. We measured soil nitrogen (N) and phosphorus (P) availability using ion exchange resin bags within three river drainages across Utah and Arizona, USA. The three drainages spanned more than 1,000 km in distance, 8° of latitude, and varying climatic regimes, but were similarly dominated by stands of Populus fremontii (S. Watts), P. angustifolia (James), or natural hybrids between the two species. Soil N availability was consistently greater in P. fremontii stands compared to P. angustifolia stands, and hybrid stands were intermediate. However, we found that the influence of overstory type on soil P availability depended on the river drainage. Our study suggests that, even with a near doubling of mean soil N availability across these drainages, the relative genetic-based effects of the dominant plant on N availability remained consistent. These results expand upon previous work by: 1) providing evidence for linkages between plant genetic factors and ecosystem function across geographic scales; and 2) indicating that plant genetic-based effects on nutrient dynamics in a given ecosystem may differ among nutrients (e.g., N vs. P).

Abiotic controls on net nitrification rates are well documented, but the potential effects of plants on this important ecosystem process are poorly understood. We evaluated four structural equation models to determine the relative importance of plant community composition, aboveground herbaceous production, and plant species richness on nitrifier abundance and net nitrification following restoration treatments in a ponderosa pine forest. Model selection criteria indicated that species richness was the best predictor of nitrifier abundance, but a model that included community composition effects also had some support in the data. Model results suggest that net nitrification was indirectly related to plant species richness via a positive relationship between species richness and nitrifier abundance. Community composition was indirectly related to nitrifier abundance through its relationship with species richness. Our model indicates that species-rich plant communities dominated by C3 graminoids and legumes are associated with soils that have high abundances of nitrifiers. This study highlights the complexity of deciphering effects of ecological treatments on a system response when multiple interacting factors are simultaneously affected. Our results suggest that plant diversity and composition can both respond to forest thinning, prescribed fire and fuel manipulations, and can be factors that might indirectly influence an ecosystem process such as nitrification. Ecological restoration treatments designed to increase plant diversity and alter community composition may have cascading effects on below-ground processes.

Research that connects ecosystem processes to genetic mechanisms has recently gained significant ground, yet actual studies that span the levels of organization from genes to ecosystems are extraordinarily rare. Utilizing foundation species from the genus Populus, in which the role of condensed tannins (CT) has been investigated aboveground, belowground, and in adjacent streams, we examine the diverse mechanisms for the expression of CT and the ecological consequences of CT for forests and streams. The wealth of data from this genus highlights the importance of form and function of CT in large-scale and long-term ecosystem processes and demonstrates the following four patterns: (1) plantspecific concentration of CT varies as much as fourfold among species and individual genotypes; (2) large within-plant variation in CT occurs due to ontogenetic stages (that is, juvenile and mature), tissue types (that is, leaves versus twigs) and phenotypic plasticity in response to the environment; (3) CT have little consistent effect on plant‐herbivore interactions, excepting organisms utilizing woody tissues (that is, fungal endophytes and beaver), however; (4) CT in plants consistently slow rates of leaf litter decomposition (aquatic and terrestrial), alter the composition of heterotrophic soil communities (and some aquatic communities) and reduce nutrient availability in terrestrial ecosystems. Taken together, these data suggest that CT may play an underappreciated adaptive role in regulating nutrient dynamics in ecosystems. These results also demonstrate that a holistic perspective from genesto-ecosystems is a powerful approach for elucidating

Herbivores can directly impact ecosystem function by altering litter quality of an ecosystem or indirectly by shifting the composition of microbial communities that mediate nutrient processes. We examined the effects of tree susceptibility and resis- tance to herbivory on litter microarthropod and soil microbial communities to test the general hypothesis that herbivore driven changes in litter inputs and soil microclimate will feedback to the microbial commu- nity. Our study population consisted of individual pinon pine trees that were either susceptible or resistant to the stem-boring moth (Dioryctria albovit- tella) and susceptible pinon pine trees from which the moth herbivores have been manually removed since 1982. Moth herbivory increased pinon litter nitrogen concentrations (16%) and decreased canopy precipi- tation interception (28%), both potentially significant factors influencing litter and soil microbial commu- nities. Our research resulted in three major findings: (1) In spite of an apparent increase in litter quality, herbivory did not change litter microarthropod abun- dance or species richness. (2) However, susceptibility to herbivores strongly influenced bulk soil microbial communities (i.e., 52% greater abundance beneath herbivore-resistant and herbivore-removal trees than susceptible trees) and alkaline phosphatase activity (i.e., 412% increase beneath susceptible trees relative to other groups). (3) Season had a strong influence on microbial communities (i.e., microbial biomass and alkaline phosphatase activity increased after the summer rains), and their response to herbivore inputs, in this semi-arid ecosystem. Thus, during the dry season plant resistance and susceptibility to a common insect herbivore had little or no observable effects on the belowground organisms and processes we studied, but after the rains, some pronounced effects emerged.

Soils that are physically disturbed are often reported to show net nitrification and NO3 loss. To investigate the response of soil N cycling rates to soil mixing, we assayed gross rates of mineralization, nitrification, NH4 consumption, and NO3 consumption in a suite of soils from eleven woody plant communities in Oregon, New Mexico, and Utah. Results suggest that the com- mon response of net NO3 flux from disturbed soils is not a straightforward response of increased gross nitrification, but instead may be due to the balance of several factors. While mineralization and NH4 assimilation were higher in mixed than intact cores, NO3 consumption declined. Mean net nitrification was 0.12 mg N kg -1 d -1 in dis- turbed cores, which was significantly higher than in intact cores (-0.19 mg N kg -1 d -1 ). However, higher net nitrification rates in disturbed soils were due to the suppression of NO3 consumption, rather than an increase in nitrification. Our results suggest that at least in the short term, disturbance may significantly increase NO3 flux at the eco

Can heritable traits in a single species affect an entire ecosystem? Recent studies show that such traits in a common tree have predictable effects on community structure and ecosystem processes. Because these 'community and ecosystem phenotypes' have a genetic basis and are heritable, we can begin to apply the principles of population and quantitative genetics to place the study of complex communities and ecosystems within an evolutionary framework. This framework could allow us to understand, for the first time, the genetic basis of ecosystem processes, and the effect of such phenomena as climate change and introduced transgenic organisms on entire communities.

We examined the impact of soil stress (low water and nutrient availabilities) and two keystone insect herbivores on pinyon pine (Pinus edulis) needle litterfall. We compared trees growing on two distinct soil types: volcanic cinders, which exhibit pronounced water and nutrient limitation, and sandy-loam soils, which have higher water-storage capacity and nutrient availability. Using two longterm herbivore removal experiments (15 and 18 years, respectively), we also examined the effects of the pinyon needle scale (Matsucoccus acalyptus, which attacks juvenile trees) and the stem-boring moth (Dioryctria albovittella, which attacks mature trees) on pinyon litterfall. These herbivores reach high densities on cinder soils but are absent or occur at much lower levels on sandy-loam soils. Four years of litterfall measurements showed four major patterns. First, independent of herbivory, needle litterfall was 20% lower under trees on high-stress cinder soils than on sandy-loam soils. Second, in agreement with the negative impact of scales on tree growth (that is, a 30% decline in stem growth), trees with scale infestations had 25% lower litterfall rates than trees resistant to scale; however, 15 years of scale-insect removal did not significantly increase needle litterfall. This implies possible intrinsic differences in litter production between scale-resistant and scale-susceptible trees. Third, in contrast with significant negative effects of moth herbivory on tree growth (that is, a 27% decline in stem growth), moth herbivory had no effect on needle litterfall. This, along with increased stem density in moth-susceptible trees, may be evidence of compensatory production. Fourth, there were strong year by soil type and year by scale herbivory interactions, such that in some years the effect on litterfall can be obscured or reversed by some other factor. In summary, soil stress has a strong and predictable effect on needle litterfall, whereas the relationship between insect herbivory and needle litterfall is weaker and depends on the individual herbivore. These effects, however, are mediated by other environmental factors that have considerable annual variation.

Although biogeochemistry is an integrative discipline, terrestrial and aquatic subdisciplines have developed somewhat independently of each other. Physical and biological differences between aquatic and terrestrial ecosystems explain this history. In both aquatic and terrestrial biogeochemistry, key questions and concepts arise from a focus on nutrient limitation, ecosystem nutrient retention, and controls of nutrient transformations. Current understanding is captured in conceptual models for different ecosystem types, which share some features and diverge in other ways. Distinctiveness of subdisciplines has been appropriate in some respects and has fostered important advances in theory. On the other hand, lack of integration between aquatic and terrestrial biogeochemistry limits our ability to deal with biogeochemical phenomena across large landscapes in which connections between terrestrial and aquatic elements are important. Separation of the two approaches also has not served attempts to scale up or to estimate fluxes from large areas based on plot measurements. Understanding connectivity between the two system types and scaling up biogeochemical information will rely on coupled hydrologic and ecological models, and may be critical for addressing environmental problems associated with locally, regionally, and globally altered biogeochemical cycles.

Rates and reactions of biogeochemical processes vary in space and time to produce both hot spots and hot moments of elemental cycling. We define biogeochemical hot spots as patches that show disproportionately high reaction rates relative to the surrounding matrix, whereas hot moments are defined as short periods of time that exhibit disproportionately high reaction rates relative to longer intervening time periods. As has been appreciated by ecologists for decades, hot spot and hot moment activity is often enhanced at terrestrial-aquatic interfaces. Using examples from the carbon (C) and nitrogen (N) cycles, we show that hot spots occur where hydrological flowpaths converge with substrates or other flowpaths containing complementary or missing reactants. Hot moments occur when episodic hydrological flowpaths reactivate and/or mobilize accumulated reactants. By focusing on the delivery of specific missing reactants via hydrologic flowpaths, we can forge a better mechanistic understanding of the factors that create hot spots and hot moments. Such a mechanistic understanding is necessary so that biogeochemical hot spots can be identified at broader spatiotemporal scales and factored into quantitative models. We specifically recommend that resource managers incorporate both natural and artificially created biogeochemical hot spots into their plans for water quality management. Finally, we emphasize the needs for further research to assess the potential importance of hot spot and hot moment phenomena in the cycling of different bioactive elements, improve our ability to predict their occurrence, assess their importance in landscape biogeochemistry, and evaluate their utility as tools for resource management.

THE importance of nitrate (NO–3) in the internal nitrogen cycle of undisturbed coniferous ecosystems has not been widely recognized1,2. Nitrate concentrations in soils from these forests tend to be low, and assays measuring net nitrification usually show exceedingly slow rates3,4. It may be, however, that microbial assimilation of NO–3 is substantial in these soils, and that net nitrification rates greatly underestimate gross rates5. Here we use a 15N isotope-dilution technique in intact soil cores to measure gross rates of nitrification and microbial assimilation of NO–3 in eleven undisturbed forest ecosystems of New Mexico and Oregon. We found that gross nitrification rates were surprisingly high in all of the forests examined. Net nitrification rates poorly predicted gross rates because the soil microbial communities had the capacity to assimilate almost all of the NO–3 produced. To our knowledge, this is the first report of gross nitrification and NO–3 assimilation rates in intact soil samples from a large number of contrasting forest ecosystems. Our results contradict previous assumptions that nitrification rates are low in mature coniferous forests and suggest that current models greatly underestimate the role of the microbial community in preventing NO–3 loss.

Soil microorganisms regulate the supply of nitrogen to plants and so are important controllers of plant productivity and ecosystem carbon sequestration. Johnson et al.1 report that exposure of a subarctic heath ecosystem to increased ultraviolet-B (UV-B) irradiation causes a drastic decline in the mass ratio of C:N in soil microorganisms, which would increase the amount of nitrogen stored in the microbial biomass and possibly alter the availability of nitrogen to plants. However, we argue that some of the authors' microbial C:N data are unrealistic, possibly because of an artefact of the technique used to measure microbial carbon and nitrogen concentrations. As a result, there is little reason to suppose that increased exposure of ecosystems to UV-B radiation will influence microbial nitrogen storage, plant nitrogen availability or rates of carbon sequestration.

Seasonal patterns and annual rates of N inputs, outputs, and internal cycling were determined for an old-growth mixed-conifer forest floor in the Sierra Nevada Mountains of California. Rates of net N mineralization within the forest floor, and plant N-uptake and leaching of inorganic N from the forest floor were 13, 10, and 9 kg-N ha-1 yr-1, respectively. The Mediterranean-type climate appeared to have a significant effect on N cycling within this forest, such that all N-process and flow rates showed distrinct seasonal patterns. We estimated the forest floor supplies less than one-third of the total aboveground plant N-uptake in this forest. The rate of net nitrification within the forest floor was always low (1 kg-NO3 --N ha-1 30d-1). Mean residence times for organic matter and N in the forest floor were 13 and 34 years, respectively, suggesting that this forest floor layer is a site of net N immobilization within this ecosystem. We examined the influence of the forest floor on mineral soil N dynamics by injecting small amounts of15N-enriched (NH4)2SO4 solutions into the surface mineral soil with the forest floor present (+FF) or removed (-FF). K2SO4-extractable NO3 --N, total inorganic-N, and total-N pool sizes in the mineral soil were initially increased after forest floor removal (after 4 months), but NO3 --N and total inorganic-N were not significantly different thereafter. Microbial biomass-N and K2SO4-extractable total-N pool sizes were also found to be larger in mineral soils without a forest floor after 1 and 1.3 years, respectively. Total15N-recovery was greater in the +FF treatment compared to the -FF treatment after 1-year (about 50% and 35%, respectively) but did not differ after 1.3 years (both about 35%), suggesting that the forest floor delays but does not prevent the N-loss from the surface mineral soil of this forest. We estimated using our15N data that fungal translocation from the mineral soil to the forest floor may be as large as 9 kg-N ha-1 yr-1 (similar in magnitude to other N flows in this forest), and may account for all of the observed absolute increase of N in litter during the early stages of decomposition at this site. Our results suggest that the forest floor acts both as a source and sink for N in the mineral soil.