Impact of Atmospheric Nitrogen Deposition on Carbon Storage Pathways Along a Nitrogen Deposition Gradient in Coastal Sage Scrub Soils

Transcription

1 Mission Kearney Foundation of Soil Science: Soil Carbon and California's Terrestrial Ecosystems Final Report: , 1/1/ /31/2004 Impact of Atmospheric Nitrogen Deposition on Carbon Storage Pathways Along a Nitrogen Deposition Gradient in Coastal Sage Scrub Soils *David E. Crowley 1, Kun Liu 1, and Edith B. Allen 2 Summary Carbon storage in soil is favored by degradation of plant litter through fungal degradation pathways as compared to those involving bacteria. Environmental factors affecting the fungal:bacterial ratios in soils thus can potentially influence the accumulation of soil carbon. The objectives of this research was to examine the carbon storage pathways involving the fungal and bacterial biomass components, and the extent to which these pathways shift in response to soil nitrogen in the semiarid coastal sage scrub plant communities of Southern California. Experiments were conducted on preexisting field plots in coastal sage vegetation along a previously characterized nitrogen deposition gradient. The activities of bacteria and fungi were measured using a combination of selective antibiotic inhibition techniques in conjunction with substrate-induced respiration (SIR) assays. Relative shifts in fungal:bacterial biomass were performed using phospholipid fatty acid analysis (PLFA). Results showed that the microbial communities from all soils sampled along the nitrogen deposition gradient were highly similar with respect to their relative composition of fungi, bacteria, and actinomycetes. Fungal-specific biomarkers comprised on average one third of the total PLFA, but likely underestimate the actual percent composition of the community. SIR assays indicated that fungi were responsible for the majority of the soil respiration and carbon turnover irrespective of differences in total or soluble soil nitrogen. Soils containing high nitrogen had higher respiration rates per unit biomass, which suggests that they will not accumulate carbon at the same rates as the soils with low nitrogen. As there was no apparent effect of relatively small differences in nitrogen between soils, an additional experimental treatment was established to examine the impact of supplemental nitrogen provided as 100 ppm ammonium nitrate on sites under low and high atmospheric nitrogen deposition. At this time, the overall impact of nitrogen deposition on altering fungal:bacterial ratios in arid grassland ecosystems is not significant. However, both seasonal and annual variation in soil moisture and temperature, as well as changes in vegetation from CSS to grassland may contribute to long-term changes that were not apparent in the short-term period over which this study was conducted. The present study provides baseline data for monitoring future changes in the relationship between soil carbon and microbial communities in Southern California. Objectives Objective 1: Examine the extent to which microbial activity, measured as soil respiration, is associated with the fungal and bacterial biomass components of the soil under CSS vegetation along an atmospheric nitrogen deposition gradient. *Principal Investigator 1 Department of Environmental Sciences, University of California, Riverside, CA 2 Dept. of Botany and Plant Sciences, University of California, Riverside, CA

2 Objective 2. Determine whether fungal-dominated decomposition results in higher carbon storage than in soils where decomposition occurs through bacterial decomposition pathways. Approach and Procedures Nitrogen has been shown to have significant effects on microbial activity and carbon storage in forest and agricultural ecosystems by causing shifts in fungal:bacterial activity that favor bacterial degradation pathways that become dominant under high-nitrogen conditions. This phenomenon has not been examined for the semiarid coastal sage scrub plant communities of Southern California. Much of the biomass in these systems is belowground where plant roots must scavenge for water and nutrients, and soil carbon levels are extremely low in the 0.1% range. Over the past few decades, invasions of European annual grasses have begun to occur in these areas and provide large inputs of carbon through fine root turnover. This represents another variable that may influence carbon storage as microbial communities may be influenced by differences in the substrate quality and relative recalcitrance of woody roots from the CSS vegetation and grasses. With the low organic matter content of these soils, storage of inorganic carbon in the soil is critical for maintaining soil structure which allows root growth and water infiltration, and for providing slowly mineralizable nitrogen for plant uptake. If atmospheric deposition of nitrogen does in fact cause a decrease in soil carbon storage, this may have a significant long-term impact on the ability of these soils to support coastal sage scrub plant communities which depend on deep water infiltration in the soil profile. To study the influence of atmospheric nitrogen deposition on CSS vegetation, this research examined soil respiration rates and microbial activity for soils collected along a nitrogen deposition gradient at five locations under coastal sage scrub in Southern California. The key parameters that were measured included the fungal and bacterial biomass, total microbial biomass, respiration per unit biomass (metabolic quotient), and substrate-induced respiration assays that allowed estimates of the relative amounts of soil microbial respiration for bacteria and fungi. The composition of the microbial community was determined using phospholipid fatty acid (PLFA) assays, which analyze for the presence of biomarkers that discriminate and quantify different groups of bacteria and fungi. Total phospholipids and fattys were extracted from soil and were purified using silicic acid column chromatography. They were methylated and separated by gas chromatography. Identification and quantification was achieved by comparison with a standard reference mixture. Among the fatty acids that are commonly found in soils, the quantity of palmitic acid is strongly correlated with a number of measures of microbial biomass (Zelles et al. 1992). Other fatty acids markers include i15:0, a15:0 and i16:0 signature fatty acids, which serve as markers for Gram-positive bacteria (Ratledge and Wilkinson 1988), and 16:1ω5c for methane-oxidizing bacteria (Tunlid and White 1992). Fungal biomass is calculated as the sum of all extracted lipids that are characteristic of eukaryotes (Zelles 1999; Bailey et al. 2002). These include the fatty acid 18:1w9 and the sum of all polyunsaturated fatty acids. Many bacteria and fungi in soils are adapted to extreme starvation and may exist in spore forms or are relatively nonresponsive to carbon inputs. There also are large differences in fungal and bacterial activity that reflect the relative drought tolerance of different groups of 2

3 microorganisms. Fungi and filamentous, high GC, actinomycetes are generally more tolerant of dry soil conditions and are metabolically active at a lower water potential as compared to gramnegative and low GC, gram-positive bacteria. To quantify the potential biodegradation activity of the fungal and bacterial biomass components, the response patterns for these two broad groups were quantified using substrate-induced-respiration (SIR) assays in which glucose or another carbon source is added to soil. The soil was incubated for up to 72 hours to observe the response pattern for release of CO 2. Typically the response pattern was biphasic, with the first phase consisting of a rapid increase followed by a plateau that indicates the carbon use and respiration that is derived from the preexisting biomass. The plateau period was then followed by another increase after 48 to 72 hours, which reflects respiration from new microbial growth. To measure the active biomass, only the CO 2 that is produced in a short period prior to active microbial growth was measured. Typically this is six hours in agricultural soils, but as observed in our results with low organic matter soils from semiarid grasslands, this period needed to be extended for much longer periods of time, up to 48 hours. Differentiation of bacterial and fungal respiration with the SIR assay was accomplished by use of selective inhibitors (Velvis 1997). To differentiate fungal and bacterial activity, each soil is subjected to four treatments: no addition of antibiotics; soil treated with a fungicide; soil treated with a bactericide, and soil treated with both fungicide and bactericide. As a further control treatment, all SIR data were normalized by subtraction of identically treated control samples without added glucose to determine the basal respiration. The response patterns following the addition of glucose and measurement of the total carbon respired as CO 2 were determined by means of a multichannel gas sampling unit that was connected to a respirometer and a data logger. To examine the microbial activity and community structures of soils along the nitrogen deposition gradient, replicate intact soil cores were taken from five sites previously characterized from an earlier study that was conducted between 1995 and 1998 (Padgett et al. 1999) (fig. 1). The five sites that were sampled, varying from high to low nitrogen, included Jurupa Hills (JH), Motte Reserve (MR), Lake Mathews (LM), Lake Skinner (LS) and Santa Margarita (SM). The soil cores were transported to the lab, where they were brought to field capacity and allowed to equilibrate for 10 days. One set of cores from each site was subjected to substrate-induced respiration assays with selective inhibitors as described by Bailey et al Each SIR assay included four treatments: no inhibitor, bacterial antibiotic (streptomycin), fungal antibiotic (cyclohexamide), and a combination of both inhibitors. Additional cores from each site were subjected to PLFA analysis, organic matter analysis, and nitrogen analysis. PLFA was run using standard procedures involving fatty acid extraction, separation of phospholipids fatty acids by column chromatography, and derivitization for GC analysis. 3

4 70 60 LSD NO3 LSD NH4 Extractable N (mg/g) NH4 NO JH MR WR BG PE MO LM HE LS SM Figure 1. Nitrate and ammonia nitrogen concentrations at 10 different locations along a nitrogen deposition gradient Coastal Sage Scrubland in Riverside County. Key: JH, Jurupa Hills; MR, Motte Reserve; WR, Waterman Rd; BG, UCR Botanic Garden; PH, Pedley Hills; MO, Mockingbird Reservoir; LM, Lake Mathews; HE, Hemet; LS, Lake Skinner; SM, Santa Margarita (Data from Padgett et al. 1999). Additional intact soil cores (2.5 x 10 cm) with five replicates from each site were adjusted to field capacity, allowed to equilibrate for 10 days, and were then maintained under steady state conditions in the laboratory to measure their basal respiration rates. After two weeks of measurement, the soil biomass was measured using the fumigation-incubation method. Respiration rates per unit biomass were calculated to obtain the metabolic quotient values for each soil (Anderson and Domsch 1975). The soil assays were conducted at three separate intervals, once in fall 2003, and in the spring and summer of Soil samples were also collected for analysis in November 2004, but are still being analyzed, and the data for these last samples are not included in this report. Results Results of the PLFA analyses showed that the microbial community composition was highly similar at all five locations that were sampled along the nitrogen deposition gradient (fig. 2). Among the PLFA that were detected, there were marker PLFA present for bacteria, actinomycetes, and fungi that were used to obtain an estimate of the relative biomass of each of these groups in the different soils. 4

5 PLFA abundance (% of total) :0 ISO 15:0 2OH 16:1 2OH 16:1 w9c ISO 17:1 AT 9 17:1 w8c 18:1 w9c 18:3 w6c 19:0 cyclo c :0 Jurupa Regional Park Lake Mathews Motte Reserve Rancho Santa Margarita Reserve Lake Skinner Figure 2. PLFA abundance (% of total) in soils across a nitrogen deposition gradient. Based on the biomarker signatures for the different microbial groups that were compared, the microbial communities were shown to be highly similar with respect to their relative proportions of fungi, bacteria, and actinomycetes (fig. 3). The overall quantity of biomarker for fungi was approximately 1/3 that of the total biomass. Similar data were obtained for all three sampling periods in fall 2003, and March and June Data for March 2004 are shown in figure 4. Caution should be used in interpreting differences in PLFA values as absolute values for biomass, since the fungal markers are thought to underestimate the actual fungal biomass. Bacteria to fungi biomarker ratios varied from 1.9 to 2.4 across the different locations, and showed no correlation with nitrogen content. There were considerable differences in the amount of ground cover by vegetation at the different locations, which may confound the comparisons of relatively small changes in the ratios of bacteria and fungi. For example, the Jurupa Hills area had extensive ground cover in grasses as compared to the Motte Reserve. To better address this question, we have since set up nitrogen fertilization experiments at each site to monitor shifts in the ratios of bacteria and fungi that are independent of vegetation effects. These studies will require several years of site monitoring to determine whether there are significant shifts and the rate at which such shifts occur. 5

6 Figure 3. Relative abundances of bacteria, actinomycetes, and fungi based on selected PLFA biomarkers for these microorganisms. Vertical bars indicate the total peak heights and provide an estimate of differences in biomass between the different soils that were collected along the nitrogen deposition gradient. Data are shown for September F:B ratio MR JRP LM LS SMR Figure 4. Fungal bacterial ratio of PLFA markers for microbial communities under CSS vegetation in March Vertical bars indicate one standard deviation. Similar data were obtained for other sampling periods (not shown). 6

7 Soil Respiration Soil respiration rates varied by approximately two fold depending on the time of year that the soils were sampled. All of the soils sampled at any particular time were taken to the laboratory as intact soil cores, and were hydrated to field capacity for 10 days prior to measuring respiration rates. The conditions for measurements of soil respiration rates thus were highly standardized and the observed differences almost certainly reflect differences in the amount of carbon provided to the soil microorganisms by death and turnover of fine roots. Soils sampled in the spring following the winter rains contained abundant fine roots, which underwent decomposition during incubation. During the summer months, soil respiration values declined. Substrate-induced respiration assays conducted at each sampling date indicated that the fungi were responsible for 40 to 50% of the carbon respiration from these soils. There were no differences between the soil locations in the SIR values for fungi and bacteria. Similar results were obtained at each sampling interval Fungi:Bacteria MR JRP LM LS SMR Figure 5. Soil respiration (%) associated with fungi in soils sampled in June Metabolic Quotients Soil respiration rate per unit biomass, or metabolic quotient, has been recognized for some time as one of the most important parameters for describing the potential for soils to store carbon (Anderson and Domsch 1979). This factor has been shown to be an important predictor of changes in carbon storage for soils that have been impaired by addition of heavy metals or acid rain, but has not been well examined with respect nitrogen. In our first year s data, we obtained the first indication that this may be the most important difference among soils that are exposed to different levels of nitrogen deposition, although there was some variability among the five sites that were tested. The two high-nitrogen soils both had high-respiration rates per unit biomass, whereas the Lake Mathews and Santa Margarita soils, which contain moderate- and low-levels of nitrogen, both had metabolic quotients that were approximately three-fold lower than in the high-nitrogen soils (fig. 6 top). However, subsequent analyses indicate that the metabolic quotient was variable at each site depending on the time of year that it sampled. This may reflect differences in 7

8 substrate quality at each site associated with different types of vegetation and seasonal differences in fine root turnover. Figure 6. Measurements of the respiration per unit biomass, also known as the metabolic quotient (Q). Top graph is for Oct Bottom graph is for June This is an indication that carbon may be lost from these soils at a higher rate. However, the metabolic quotients were highly variable between sampling intervals, which may reflect differences in substrate quality at each site associated with different types of vegetation and plant phenology. Discussion Carbon storage in soils is the result of decomposition of plant litter that accumulates as partially degraded plant materials and as microbial byproducts arising from microbial metabolism and biosynthesis. This process is carried out by a milieu of microorganisms. The rate at which decomposition occurs is dependent upon both the size and activity of the soil biomass, which is the combined weight of all of thousands of species of bacteria, Archaea, fungi, and soil microfauna that live in the soil. Among the most broad system parameters that can be used to predict carbon storage, the two most important from a microbiological perspective are the fungal: bacterial ratio and the respiration rate per unit biomass. Results of the research reported here demonstrated that in the soils we studied, the soil fungi are not only an important weight 8

9 component of the biomass, but that they are also responsible for approximately half of the substrate inducible respiration following a carbon pulse with glucose. The initial indication based on this two-year study indicate that there were no differences in the F: B ratios observed among the soils along the nitrogen deposition gradient. However, since differences in vegetation and soil moisture also may influence F:B ratios (Wardle and Parkinson 1990), an independent approach is needed in which supplemental fertilizer is used in plots at each site to determine the threshold ranges at which shifts in F:B ratio and carbon storage might occur. Lastly, further studies should focus on the differences in the metabolic quotients that occur with respect to nitrogen, as this proved to be the most variable parameter among the different sites that were examined. In terrestrial ecosystems, the availability of nitrogen often plays a critical role in determining the rate and extent of litter decomposition (Agren et al. 2001). Until recently, the linkage between carbon and nitrogen cycling has been of interest in agricultural ecosystems where nitrogen is applied as fertilizers or in organic amendments, and in managed forests where nitrogen cycling is critical to forest productivity. In the larger picture, nitrogen excesses are viewed as having a positive effect on carbon storage that may help alleviate the greenhouse effect caused by elevated carbon dioxide concentrations. However, over the past few decades there are increasing instances in which anthropogenic activities have resulted in nitrogen excesses in natural ecosystems, particularly in the vicinity of urban areas where nitrogenous pollutants have resulted in nitrogen pollution and nitrogen saturation, leading to denitrification and water pollution. One of the detrimental effects of a nitrogen surplus is that excess nitrogen may increase the ratio of bacteria to fungi in the soil, which has been hypothesized to decrease the total quantities of carbon that are retained in the soil. To date there are still very few studies in the literature that have examined the impact of excess nitrogen on carbon storage pathways, and the different sensitivities of fungal and bacterial-dominated pathways to different levels of nitrogen. Plant materials contain relatively large amounts of carbon as compared to nitrogen, and have much higher C:N ratios than fungi (15:1) and bacteria (5:1). Under low nitrogen conditions, fungi have several competitive advantages over bacteria. First, their lower C:N ratio results in a lower nitrogen demand, whereas bacterial activity is typically limited by nitrogen during the degradation of high C/N plant materials. Under high carbon conditions, this results in immobilization of nitrogen in the bacterial biomass that must undergo turnover in order to support successive generations of bacteria. Another advantage that fungi have under low nitrogen conditions is their ability to form extensive hyphal networks that can reallocate nutrients to exploit new carbon sources. The degree to which bacteria become more competitive and displace fungi as an increasingly predominant component of the biomass in response to nitrogen is just now beginning to be examined for different ecosystems. Among the factors that can potentially affect F:B ratios, disturbances caused by tillage have been particularly well investigated. Tillage results in a combination of effects that work together to lower the F:B ratio. This includes destruction of fungal hyphal network, exposure of protected organic matter, and release of mineral nitrogen that is caused by death and turnover of the microbial biomass. Soil aggregates, in particular, help to protect partially degraded, particulate organic matter that becomes sequestered in the soil aggregates. In general fungi are thought to play a more important role in soil aggregate formation than bacteria. It has frequently been observed that fungi tend to dominate the microbial biomass 9

10 in undisturbed systems, whereas tillage and other environmental disturbances favor bacterialdominated systems (Bardgett et al. 1996; Bardgett and McAlister 1999). In comparison to our knowledge on how tillage affects F:B ratios and carbon storage, much less is know about the effects of nitrogen alone, since studies on nitrogen in managed systems typically are confounded by interactions with grazing, ph, tillage, or other disturbance effects (Bardgett et al. 1996; Bayer et al. 2001; Bailey et al. 2002). Nonetheless, the literature is suggestive that nitrogen does in fact have significant effects on carbon storage that are mediated by changes in microbial community composition. In a 12-year study of the effects of nitrogen fertilization on Minnesota grasslands, increased nitrogen was associated with decreased biomass carbon and low carbon storage (Wedin and Tilman 1996). In a complementary study that provides further insight into reasons for such decreases in carbon storage, analysis of soils from meadow grasslands in England showed that F:B biomass ratios were consistently higher in unfertilized grasslands as compared to fertilized grasslands (Bardgett and McAlister 1999). The strongest support for the hypothetical link between nitrogen and carbon storage is the data provided by Bailey and coworkers (2002) in which they show that F:B ratios and carbon storage are directly linked in a variety of ecosystems including a desert mountain, tallgrass prairie, Douglas fir forest, pine plantation, and agricultural land under different tillage systems. By pairing disturbed or fertilized sites at each location, they further show that in all but one case, F:B activity ratios were significantly correlated with carbon storage (r 2 =0.85). Differences in the F:B activities were greatest in the prairie soils, where the fungal respiration component was 13.5 times greater than that of the bacteria, and stored carbon was measured at 50 g C kg soil. In comparison, adjacent farmed prairie soil had an F:B activity ratio of 0.8 and stored only 36 g C kg soil. In one treatment in that study, that directly examined the effects of nitrogen, fertilization of a pine plantation resulted in both a decrease in the F:B ratio and stored carbon. Based on this work, Bailey and coworkers propose that site characteristics that favor fungi also favor enhanced storage of C in soil. Future studies on the role of nitrogen in the CSS ecosystem should use paired plots at different locations to examine the sensitivity of these soils to a range of nitrogen concentrations. Our research indicates that site-specific factors may obscure the detection of changes in soil respiration and microbial activity that occur in response to nitrogen deposition. The carbon dynamics of these systems are highly dependent on seasonal and soil moisture patterns, and may further vary with the extent of grass or weed ground cover associated with the CSS vegetation. The use of SIR and PLFA techniques confirmed the importance of fungi in organic matter decomposition in this system, and provide detailed insight into the dynamics of carbon turnover in soils with respect to vegetation and environmental variables that control carbon accumulation. References Agren, G.I., E. Bosatta, A.H. Magill Combining theory and experiment to understand effects of inorganic nitrogen on litter decomposition. Oecologia 128: Anderson, J.P.E., K.H. Domsch Measurement of bacterial and fungal contributions to respiration of selected agricultural and forest soils. Canadian Journal of Microbiology 21:

11 Bailey, V.L., J.L. Smith, H. Bolton Fungal-to-bacterial ratios in soils investigated for enhanced C sequestration. Soil Biology & Biochemistry 34: Bardgett, R.D., P.J. Hobbs, A. Frostegard Changes in soil fungal-bacterial biomass ratios following reductions in the intensity of management of an upland grassland. Biology and Fertility of Soils 22: Bardgett, R.D., E. McAlister The measurement of soil fungal:bacterial biomass ratios as an indicator of ecosystem self-regulation in temperate meadow grasslands. Biology and Fertility of Soils 29: Bayer, C., L. Martin-Neto, J. Mielniczuk, C.N. Pillon, L. Sangoi Changes in soil organic matter fractions under subtropical no-till cropping systems. Soil Science Society of America Journal 65: Padgett, P.E., E.B. Allen, A. Bytnerowicz, R.A. Minich Changes in soil inorganic nitrogen as related to atmospheric nitrogenous pollutants in southern California. Atmospheric Environment 33: Ratledge, C., S.G. Wilkinson Microbial lipids, vol. 1. New York: Academic Press Inc. Tunlid, A., D.C. White Biochemical analysis of biomass, community structure, nutritional status, and metabolic activity of microbial communities in soils. In Soil Biochemistry, ed. Stotzky, G., Bollag, J.M., New York: Marcel Dekker. Velvis, H Evaluation of the selective respiratory inhibition method for measuring the ratio of fungal: Bacterial activity in acid agricultural soils. Biology and Fertility of Soils 25: Wardle, D.A., D. Parkinson Response of the soil microbial biomass to glucose, and selective inhibitors, across a soil moisture gradient. Soil Biology & Biochemistry 22: Wedin, D.A., D. Tilman Influence of nitrogen loading and species composition on the carbon balance of grasslands. Science 274: Zelles, L Fatty acid patterns of phospholipids and lipopolysaccharides in the characterisation of microbial communities in soil: A review. Biology and Fertility of Soils 29: Zelles, L Fatty acid patterns of phospholipids and lipopolysaccharides in the characterisation of microbial communities in soil: A review. Biology and Fertility of Soils 29: This research was funded by the Kearney Foundation of Soil Science: Soil Carbon and California's Terrestrial Ecosystems, Mission ( The Kearney Foundation is an endowed research program created to encourage and support research in the fields of soil, plant nutrition, and water science within the Division of Agriculture and Natural Resources of the University of California. 11