| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Environmental and Resource Sciences, College of Agriculture, Univ. of Nevada, Reno, NV 89512
b Dep. of Biological Sciences, Northern Arizona Univ., Flagstaff, AZ 86011-5640
c Plant Protection Service, P.O. Box 9102, 6700 HC, Wageningen, the Netherlands
d Smithsonian Environmental Research Center, Edgewater, MD 21037
Corresponding author (dwj{at}unr.edu)
Received for publication May 17, 2000.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODS AND MATERIALSRESULTSDISCUSSIONSUMMARY AND CONCLUSIONSREFERENCES |
|---|
Abbreviations: OTC, open-top chamber • PRS, plant root simulator
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODS AND MATERIALSRESULTSDISCUSSIONSUMMARY AND CONCLUSIONSREFERENCES |
|---|
Studies on the effects of elevated CO2 on P also have produced conflicting results. Norby et al. (1986) found an increase in soil-extractable P with elevated CO2 in a greenhouse study with white oak (Quercus alba) L. and speculated that elevated CO2 increased phosphatase activity. On the other hand, Johnson et al. (1995) found reduced soil-extractable P levels under elevated CO2 in a greenhouse study of ponderosa pine growing in a poor soil, but no effects of elevated CO2 were found on either plant P uptake or soil-extractable P when the plants were grown on a richer soil. Johnson et al. (1995) concluded from these two studies that the effects of elevated CO2 on soil P "were inconsistent and no general conclusions can be drawn." In a field study of ponderosa pine, Johnson et al. (1997) found statistically significant effects of elevated CO2 on extractable P in various treatment combinations and at various times during the 6-yr experiment; however, these effects were inconsistent among treatments and years, and in part reflected pretreatment differences.
In this paper, we summarize the results of three years of investigation into the effects of CO2 on soils from an open-top chamber study in a Florida scrub oak ecosystem. Previous results from this site have shown that elevated CO2 caused increased fine root biomass and negative effects on soil C and N availability. In a pilot study that preceded the current study, Day et al. (1996) found that elevated CO2 caused greater root length densities in a minirhizotron study. Hungate et al. (1999) found that elevated CO2 had no effect on microbial biomass N, but caused decreased N mineralization, nitrate leaching, and increased specific NH+4 immobilization (NH+4 immobilized per unit microbial N) in soils during the first 14 mo. The increased specific NH+4 immobilization was explained by increased root growth combined with decreased quality of C input to the soil (Hungate et al., 1999). Similarly, Shortemeyer et al. (2000) found no effects of elevated CO2 on microbial biomass C and N, but lower C accumulation in buried, homogenized soil bags during a 1-yr period. Shortemeyer et al. (2000) found lower soluble C, ninhydrin-reactive N, and microbial activity in rhizosphere soil at one sampling date, however, and hypothesized that this was caused by N limitations, which were in turn caused by increased N uptake by plants.
Based on the literature cited above and the fact that the soils at this site were very poor in nutrients, we hypothesized that elevated CO2 would cause (i) increases in soil pCO2 and soil respiration and (ii) reduced levels of soil-available N and P.
|
METHODS AND MATERIALS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODS AND MATERIALSRESULTSDISCUSSIONSUMMARY AND CONCLUSIONSREFERENCES |
|---|
Experimental Design
The study site was burned in August 1995 with a few remaining areas burned in January 1996 prior to siting of 16 open-top chambers (OTCs). Of the 16 OTCs, 8 were maintained at current ambient CO2 and 8 at ambient plus 350 µL L-1 (elevated). The OTCs were octagonal in design with the largest diameter 3.66 m and the sides 1.4 m long. Each chamber was 3.3 m high with the frustum at the midpoint providing a chamber volume of 18.9 m3. Eight unchambered control plots of identical surface area were also established. The plots were blocked according to preburn aboveground biomass, species composition, and proximity. Each block consisted of one of each of the eight ambient OTCs, eight elevated OTCs, and eight unchambered control plots.
Soil Leaching Measurements
Two methods to measure soil nutrient leaching were employed during the experiment, but only one proved viable. Initially (January 1996), we installed ceramic cup lysimeters to collect soil solution. However, we found that damage due to intensive solar radiation and animals precluded the reliable operation of these lysimeters, and after approximatley 4 mo without complete collections, they were abandoned. Beginning in February 1997, resin lysimeters were used to collect cumulative soil N and P leaching. Resin lysimeters have decided advantages over traditional soil solution collection lysimeters in terms of cost and maintenance, especially under harsh field conditions such as those encountered in this study. Questions have been raised as to the effects of collection efficiency, effects on soil water flow, and microbial transformations on resin flux estimates (Kjønass, 1999; Schnabel, 1983; Schnabel et al., 1993; Susfalk, 2000; Torbert and Elkins, 1992). Nontheless, resin lysimeters have been used successfully to obtain indices of leaching in several previous studies, and microbial transformations have generally been found to be minimal (Kjønass, 1999; Schnabel et al., 1993; Susfalk, 2000). The resin lysimeters used in this study consisted of a 5.5-cm-long, 4-cm-i.d. PVC pipe within which a resin bag was sandwiched between layers of washed silica sand. Ten grams of oven-dried Rexyn I-300 (H-OH) resin (Fisher Scientific, Fair Lawn, NJ) were placed in a section of nylon pantyhose, using cable ties to secure each end. This resin bag was placed on a 20-g layer of moist, washed silica sand at the bottom of the tube and covered with another 20-g layer of silica sand. In order to keep the sand in the tube until installation, the bottom of each PVC tube was covered with cheesecloth held in place with a rubber band. The lysimeters were installed by excavating a small hole and tunneling beneath the A horizons at approximately 15 cm depth. The resin lysimeters were installed in February 1997, and removed and replaced in December 1997, December 1998, and December 1999. After collection, resins were removed from the lysimeters, placed into 250-mL Erlenmeyer flasks, and extracted with 100 mL of 1.0 M KCl with shaking for 1 h. The extract was filtered (Whatman No. 1) and stored at 4°C until analysis. The extracts were analyzed for NH+4, NO-3, and ortho-P by automated colorimetric analysis at the Desert Research Institute (Reno, NV). Three 10-g replicates of untreated resins were extracted in the same way and served as blanks. Fluxes were calculated from the amount of NH+4, NO-3, and ortho-P extracted from the resins (minus blanks) divided by the surface area of the lysimeters (12.6 cm2). The flux for Year 1 was annualized by assuming that fluxes for the month of January 1997 were equal to one-twelfth those for the entire year (i.e., measured flux was multiplied by 1.091). The data for 1997 were reported previously (Hungate et al., 1999); here we update the data set with data from 1998 and 1999.
Measurement of Soil Nutrient Availability
To avoid the unacceptable effects of destructive soil sampling, less intrusive methods were used. In 1997, soil-available P was measured using anion exchange membranes (Cooperband and Logan, 1994). A 39-cm2 square of anion exchange membrane (BDH [Darmstadt, Germany] Product no. 55164) was converted into the bicarbonate form, which is known to adsorb more orthophosphate than the chloride form (Sibbesen, 1978). This was accomplished by four 1-h sequential rinses of 200 mL 0.5 M NaHCO3 (pH adjusted to 8.5) with intermittent stirring in a 250-mL flask. The membrane was placed into a slit in the surface soils (two replicates per chamber) in February 1997. A length of fishing line was sewn through each membrane and tied to guy wires in the chambers to facilitate relocation. The membranes were retrieved in December 1997 and extracted with 30 mL of 0.5 M NaCl by shaking for 1 h. The NaCl solution was then decanted, and replaced with a fresh 30 mL of 0.5 M NaCl and shaken for another hour. After this extraction process, the NaCl solutions were combined and analyzed for orthophosphate on an Alpkem RFA 300 colorimeter (OI Corp., College Station, TX) using EPA600/479020, a molybdate and ascorbic acid method (Fishmann and Friedman, 1985). Because of the loss of some membrane material during removal from the soil, the extractable P values were expressed as milligrams P per gram of membrane material. Due to the considerable difficulties encountered in installing and relocating membranes as well as the loss of membrane material during retrieval, the use of membranes was discontinued after this collection.
In July 1999, soil N and P availability were measured using plant root simulator (PRS) probes (Western Ag Innovations, Saskatoon, SK, Canada), which consist of either anion- or cation-exchange membranes conveniently imbedded in plastic stakes for easy installation and recovery. In this case, the probes were installed for a period of 2 wk before recovery. Upon recovery, the probes were extracted with 40 mL of 1 M KCl by shaking for 1 h. Extractant from the cation exchange probe was analyzed for NH+4, and the extractant from the anion exchange probe was extracted for NO-3 and ortho-P using automated colorimetric analysis.
Homogenized soil bags (David et al., 1990; Johnson et al., 2000a) were buried in the chambers to assess changes in soil C, N, and P with greater precision and less disturbance than would be possible with conventional soil sampling. Homogenized samples from the C horizon were placed in 1-mm mesh bags, labeled, and inserted in the A horizons of each chamber (three per chamber) in February 1997. One bag from each chamber was retrieved in December 1997, and the rest were left for later recovery. Soils from the bags were analyzed for C and N on a Perkin–Elmer (Norwalk, CT) CHN Analyzer for C and N and for extractable P with 0.5 M HCl plus 1 M NH4F (Olsen and Sommers, 1982). The homogenized bags were not intended to provide estimates of actual rates of soil change (because the disturbance of homogenizing soils precludes this), but rather to provide a means of measuring relative treatment effects on soil chemical properties. David et al. (1990) used this technique to detect very small changes in soil chemical properties in response to acidification treatments to a Spodosol in Maine.
Measurement of Soil pCO2 and Calculation of Soil Carbon Dioxide Efflux
The measurement of soil pCO2 and calculation of soil CO2 efflux were performed according to the methods outlined by Johnson et al. (2000a). Between April 1997 and January 2000, soil pCO2 concentrations were monitored approximately monthly from gas wells established at 15 cm depth in each chamber (in triplicate). There was a sampling gap between November 1998 and June 1999 due to personnel changes and various logistical considerations. The gas wells consisted of 4-mm Tygon tubing inserted to the proper depth in the soil and fitted at the surface with a stoppered, female end of a plastic union. The three tubes all exited the OTC at 20 cm above the soil surface through an acrylic panel, removing any need to open or enter the OTCs during sampling. Samples for CO2 analyses were obtained with Hamilton gas syringes (50 mL; Wilmad Corp., Buena, NJ) from the section of tubing between the large syringe and the union. During gas collections, stoppers were removed and 15 mL of soil gas were withdrawn from each well, completely evacuating the tubing. A second 50-mL sample was then removed for analysis. Carbon dioxide concentrations were measured on a LiCOR (Lincoln, NE) 6262 infrared gas analyzer using peak heights compared with a CO2 standard gas (Boggs Gases, Titusville, FL). These measurements were made between three and five 5-mL volumes from each 50-mL sample. For each sampling period, soil temperature was recorded at 1-, 10-, and 50-cm depths in one OTC and one unchambered control plot. Soil moisture was recorded between 0 and 15 cm in each OTC and control plot.
Soil moisture was measured by time domain reflectometry in each plot using Campbell (Logan, UT) CS615 soil moisture reflectometers. Probes were calibrated individually by measuring probe output at four known soil moisture contents ranging from field capacity to residual water content. We placed A horizon soil in four 30-cm-diam. x 60-cm-deep PVC cylinders, adjusted water content in each (residual, 25%, 75%, and 100% of field capacity), and then measured the output of each probe when placed in each cylinder. We then developed a third-degree polynomial to describe the relationship between probe output and volumetric water content. Probes were connected to a multiplexer–datalogger array programmed to record volumetric water content every 11 min. For calculating soil CO2 flux, we used the average water content for the days in which soil pCO2 was measured, as described below. Cumulative soil respiration was estimated using the profile method (De Jong and Schappert, 1972; De Jong et al., 1974; Johnson et al., 1994, 2000a; Mattson, 1995). For the diffusion coefficient, we used the Moldrup et al. (1996) formulation, which depends upon commonly measured soil properties (total soil porosity, moisture content, percent clay, and percent fine sand). Cumulative CO2 flux was calculated by trapezoidal integration of respiration values (Cotrufo et al., 1994; Johnson et al., 2000a) for the periods April 1997 through November 1998 and June 1999 through January 2000.
Statistical Analyses
Statistical analyses were performed using Microsoft Excel (Microsoft, 1998) for Student's t-tests and DataDesk (DataDesk, 1997) software for repeated measures ANOVA. For treatment effects on soil bags, membranes, PRS probes, and pCO2, Student's t-tests were used. For the resin lysimeter leaching data, repeated measures analysis of variance (ANOVA) was used.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODS AND MATERIALSRESULTSDISCUSSIONSUMMARY AND CONCLUSIONSREFERENCES |
|---|
|
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODS AND MATERIALSRESULTSDISCUSSIONSUMMARY AND CONCLUSIONSREFERENCES |
|---|
Treatments could have caused reduced P availability in soils initially as a result of (i) increased P immobilization by microbes, (ii) increased P adsorption to soils, (iii) increased P uptake by vegetation, or (iv) any combination of the above. We do not believe that microbial immobilization was a factor, given the results of Hungate et al. (1999) and Shortemeyer et al. (2000). They collectively found that there is either no effect or a reduction in microbial activity and soluble organic C with elevated CO2. As suggested by Shortemeyer et al. (2000), increased plant uptake of N under elevated CO2 may have reduced the supply of N to microbes, thereby reducing microbial activity due to N limitation. Microbial P uptake would, therefore, be expected to be lower under elevated CO2 as well. Certainly, the pCO2 and soil respiration data gave no indication of higher microbial activity under elevated CO2. Adsorption of P to soils could have changed if pH decreased under elevated CO2; unfortunately, we have no data to test this hypothesis but consider it unlikely. We do not as yet have data on P uptake by the biomass, but have made preliminary calculations based on biomass estimates thus far and weighted-average P concentrations from Schmaltzer and Hinkle (1996) for the same vegetation type. These calculations suggest that P accumulation in aboveground vegetation at the end of the 1999 growing season was approximately 7 kg ha-1 in the elevated CO2 treatment and approximately 4 kg ha-1 in the ambient CO2 treatment. Initial values for soil-extractable P (to a depth of 60 cm) combined with bulk density values from Schmaltzer and Hinkle (1996) yield values of 7 to 8 kg ha-1 (B. Hungate, unpublished data, 2000). Some of the P accumulated in aboveground biomass could have been mobilized from surviving roots; on the other hand, roots themselves may have accumulated P from soil sources. Thus, these values suggest that plant uptake could have caused a substantial decline in soil-available P levels over this period. Belowground P uptake would add very substantially: Schmaltzer and Hinkle (1996) estimated that saw palmetto rhizomes contained approximately twice as much biomass, N, and P as aboveground biomass in 2- to 4-yr-old scrub oak stands similar to the one studied here. Although plant P uptake during the first year of growth would have been very low (on the order of 1 and 2 kg ha-1 in the ambient and elevated treatments, respectively), it still could have contributed to the observed treatment effects on soil-available P pools when belowground uptake is included. Thus, we hypothesize that plant uptake was the primary factor accounting for the initial treatment effects on soil-available P and for the observed overall declines in P leaching in all treatments.
The results reported in this and previous papers from this site suggest that soil-available N under elevated CO2 is either equal to or (more often) lower than under ambient CO2. Hungate et al. (1999) reported that soil exchangeable NH+4 was lower under elevated CO2 in March 1997. Shortemeyer et al. (2000) reported lower N availability under elevated CO2 in the summer of 1998, and the PRS probe data reported here clearly indicate lower N availability in 1999. As discussed earlier, N leaching is a function of both N availability and water flux, the latter of which may have been greater with elevated CO2 because of reduced evapotranspiration (B. Hungate, personal communication, 2000). Lower NO-3 leaching under the elevated CO2 treatment in 1997 was probably due to lower soil-available N, and the lack of differences in N leaching in other cases may have been due to the offsetting effects of N availability and soil water flux. The greatly reduced NH+4 leaching in 1998 (compared with 1997 and 1999) may have been due to the drought that year: total precipitation in 1998 was 839 mm compared with 1380 mm in 1997 and 1403 mm in 1999. No reductions in the leaching of NO-3 or ortho-P during 1998 were noted, however.
The results for N and P leaching in this study contrast to those obtained by Körner and Arnone (1992) for artificial tropical ecosystems. These authors found increased N and P leaching with elevated CO2 and attributed this to stimulated microbial activity, causing nutrient release in excess of plant demand. On the other hand, Torbert et al. (1996) found reduced NO-3 leaching with elevated CO2 in agroecosystems, and attributed it to reduced N release from crop residues and increased N retention in soil organic matter. In this study, we hypothesize that differences in N uptake (which would have amounted to approximately 10 kg ha-1 during the first season and accumulated to a difference of approximately 90 kg ha-1 by the end of the third growing season) were the major factor causing the reduction in N leaching (which amounted to less than 0.01 kg ha-1) and availability.
The lack of response in soil pCO2 or calculated respiration at any time during this study is unusual for elevated CO2 experiments (Allen et al., 2000; Canadell et al., 1996; Edwards and Norby, 1999; Hungate et al., 1999; Johnson et al., 1994; Körner and Arnone, 1992; Verburg et al., 1998; Vose et al., 1995), and difficult to explain in view of the probable increase in root biomass (Day et al., 1996). Part of the reason for the lack of response in soil pCO2 and respiration may have been the presence of a substantial amount of live root biomass that survived the fire and remained basically intact as the experiment began (especially saw palmetto rhizomes), masking any current treatment effects on fine root biomass or microbial activity. These surviving root systems may not yet have been affected by elevated CO2, and respiration from them could still reflect prefire biomoass. On the other hand, the results of Hungate et al. (1999) and Shortemeyer et al. (2000) for this site suggest that treatment effects on microbial activity were either negligible or negative, and thus an increase in soil respiration from microbial sources does not seem likely. Further investigation is needed to reconcile the lack of soil pCO2 and respiration response with other findings in this study.
|
SUMMARY AND CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODS AND MATERIALSRESULTSDISCUSSIONSUMMARY AND CONCLUSIONSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODS AND MATERIALSRESULTSDISCUSSIONSUMMARY AND CONCLUSIONSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  P. J. Drohan, D. J. Merkler, and B. J. Buck Suitability of the Plant Root Simulator Probe for Use in the Mojave Desert Soil Sci. Soc. Am. J., August 4, 2005; 69(5): 1482 - 1491. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. W. Johnson, P. S. J. Verburg, and J. A. Arnone SOIL EXTRACTION, ION EXCHANGE RESIN, AND ION EXCHANGE MEMBRANE MEASURES OF SOIL MINERAL NITROGEN DURING INCUBATION OF A TALLGRASS PRAIRIE SOIL Soil Sci. Soc. Am. J., January 1, 2005; 69(1): 260 - 265. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Klamer, M. S. Roberts, L. H. Levine, B. G. Drake, and J. L. Garland Influence of Elevated CO2 on the Fungal Community in a Coastal Scrub Oak Forest Soil Investigated with Terminal-Restriction Fragment Length Polymorphism Analysis Appl. Envir. Microbiol., September 1, 2002; 68(9): 4370 - 4376. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Crop Science | |||
| Vadose Zone Journal | Journal of Plant Registrations | ||||
| Journal of Natural Resources and Life Sciences Education |
Soil Science Society of America Journal |
