Comparison of static and dynamic closed chambers for measurement of soil respirhtion under field conditions E. G. and Desjardins, R. L. 1992. Comparison of static and AVll*,!-"tg:d of soil respiration under field conditions. Can. I' Soil Sci. 72: ffi5-609' measurement chambers for The objective of this study *u, to .o*pure the dynamic closed and static chamber techniques for the prodlced measurlment of soil respiration under fiild conditions. The static chamber method consistently lower soil respiration vilues than did the dynamic closed system and the difference was larger at higher by both CO2 fluxes. A negative exponential model describes the relation between COz {uxes measured tecliniques. A eood fit was obtained for measurements on a sandy loam soil (Rr:0.61) and an organic soil (R):0.741 but parameter estimates were different for each soil'

Can. J. Soil. Sci. Downloaded from www.nrcresearchpress.com by MICHIGAN STATE UNIV on 01/14/17 For personal use only.

Rochette, P., Gregorich,

Key words: Carbon dioxide' enclosure, gas flux measurement et Rochette, P., Gregorich, E. G. et Desjardins, R. L. 1992. Comparaison des.chambres statique Le but 72: 605-609. SciJ. Soil Can. situ. dynamique pou.i" mesure de la respiration du sol in di cette etuOe etait de comparer les chahbres ferm6es statique et dynamique pour la mesure au champ que celles.de de la respiration du sol. Lis mesures prises par la chambre statique ont 6t6 plus basses la chambre dynamique et les diff6reni"s oni 6td plus grandes aux plus fortes valeurs de respiration du sol. Une iourbeixponentielle n6gative a 6te ihoisii pour d6crire la relation entre les valeurs de flux de CO2 donn6es par les deux syJtdmes. Des coefficients de d6termination 6lev6s ont 6t6 obtenus sableux (R'z =0.61) et un sol organique (R2:0.74). Les valeurs pour des *irur", prir", ,oi lstim6es des parambtres de l'6quation ont toutefois 6t6 diffdrentes pour chaque sol'

;; ig;

Mots cl6s: Dioxide de carbone, dchange gazeux

Soil respiration is an important process of carbon cycling in terrestrial ecosystems and

a useful index for heterotrophic metabolic activity in soils. Accumulation of CO2 in chambers is widely used to measure soil respiration under field conditions because it is simple and the presence of vegetation interferes with more direct micrometeorological techniques such as eddy correlation (Chahuneau et al. 1989). provides

Static chambers have been used to measure

soil respiration for more than 60 yr (Lundegardh 1927). These consist of inverted enclosures inserted into the soil in which the CO2 emitted at the soil surface is absorbed by an alkali solution. This technique is inexpensive, simple to use and has the advantage of integrating the flux over time (hours to days). The placement of the chamber on the soil surface for extended periods, however, disturbs the soil microclimate. Many factors

causing under- or over-estimations of CO2 fluxes have been identified (Anderson 1973; Gupta and Singh 1977; Sharkov 1984). More recently, dynamic chambers through

which air circulates have been used. In dynamic open systems, fresh air of a known

co2

concentration

is

admitted into the

chamber while an equal volume of air is withdrawn and analyze'dby infrared gas analyzers (Kanemasu et al. 1974; Ewel et al. 1987) or

chemically trapped (Witkamp 1969). Soil respiration is calculated using the flow rate and the difference in CO2 concentration between the air entering and leaving the

chamber. In dynamic closed systerns, air is circulated from the chamber to a gas analyzer and returned to the chamber. Soil respiration is calculated using the rate of increase of CO2 concentration. Reliable portable CO2 analyzers have recently been developed and successfully used to measure soil respiration in dynamic closed chambers (Norman et al. 1990: Hall et al. 1990; Rochette et

Can. J. Soil Sci. 72: 605-609 (Nov. 1992) 605

al' 1991).

606

CANADIAN JOURNAL OF SOIL SCIENCE

In comparison studies, dynamic open chambers yielded larger soil respiratlon estimates than static chambers 6Wittamp

1969; Cropper et al. 1985). The relationship

Can. J. Soil. Sci. Downloaded from www.nrcresearchpress.com by MICHIGAN STATE UNIV on 01/14/17 For personal use only.

between f,reld measurements using both techniques, however, could not be eaiilv defined

due to the high spatial variability of soil respiration. Ewel et al. (1987) overcame this

problem by taking six replicate measurements with each chamber. Their results showed a

logarithmic relationship:

Pt : (ln

Rd,o

- 3.98)/0.00653,

where R4,o and R, (mg CO2

soil

r-t n-t; are

respiration values measured using

d.ynamlc open and static chamber, respectively. No direct comparison of the statieand

closed dynamic chambers have yet been

reported. The objective of this study was to compare the dynamic closed and static chamber fechniques for the measurement of soil respiration under freld conditions. An approach for eliminating the problems related to spatial variability in such studies is also presented. Both dynamic and static chamberi consisted of a 2.9-L plexiglass cylinder with a cross

sectional area of 0.019 m2. The outside of the cylinder was coated with reflective paint. The dynamic system was connected to a

Ottawa. The sky was clear for the June sampling and hazy for the later sampling. Thirty pairs of sampling points were marked in a 4 x 10-m plot (Fig. 1). On rhe first day, a static chamber was placed on one point of each pair, for l0 h; the other point was monitored for 1 min every hour during l0 h using the dynamic system. On the second day, the chambers were switched and the measurements repeated. Soil respiration values obtained by each technique were averaged over the 2 sampling days for paired points and were then compared. Parameters of the nonlinear models relating respiration values obtained using both techniques were calculated using the Marquardt iterative method

(Statistical Analysis System Insritute, Inc. r98s). The static chamber method consistently produced lower soil respiration values than did the dynamic closed system and the difference was larger at higher fluxes

(Fig. 2). Ewel

et al. (1987) obtained similar results usins dynamic open chambers. Several factors ma! account for the difference between the two techniques. CO2 absorption by the NaOH solution may decrease over time, due to slow

diffusion (Freijer and Bouten l99l). The resulting increase in chamber air CO2 concentration lowers the CO2 gradient between the soil and the chamber, thereby reducing the

LI-6250 portable CO2 analyzer (LI-COR Inc., Lincoln, NE); the flow rate throush the chamber was approximately 1.3 L ;in-|. Details of the design and operation of the

measured fluxes. Cooler soil temperature inside the static chamber may also contribute to lower fluxes. In this study, soil tempera-

dynamic closed chamber are given elsewhere

chambers during the experiment, but not inside the static chambers. Measurements taken later in the year (12 September) under

(Rochette et al. 1991). Contaiiers with 25 mL

IM

NaOH were placed inside the static At the end of the measurement

chamber.

period, the containers were removed and the NaOH titrated wirh standard I M HCl. The surface area of NaOH solution was equal to 16% of the soil area covered by the chimber to ensure adequate absorption of CO2 by the alkali solution (Gupta and Singh 1977). Bothtechniques were used on a sandy loam Podzol (18 and 19 June) and an organic soil, classihed as a Terric Humisol (31 Julv and I August) at the Central Experimental Farm,

soil, classified as a Humo-Ferric

ture was measured hourly outside

the

clear conditions showed that the soil tempera-

ture at 20 mm under the static chamber was up to 4'C less than that measured at the same depth but outside the chamber. This temperature difference was the same for both studied soils. A negative exponential model was chosen to fit the respiration values measured by both techniques. It was assumed that neither system had a measurement bias in the absence of respiration. Therefore, the curves were forced through the origin. A good fir was obtained

ROCHETTE ET AL.

a)

o

\, a) a) a)

n

\, a) a) a) a)

a)

rf

0m

a) I

2so I

r^)

t

.)

r)

I

Fig. 1. Location of the 30 pairs of sampling points in the study plots. between the flux values measured by the two systems (Fig. 2). The variation observed is similar to that reported by Ewel et al. (1987) but our method needed a smaller number of measurements. A negative exponential model

was also

fit to the 60 individual pairs of

measurements taken on the sandy loam soil (18 and 19 June). The,coefficient of deter-

mination obtained (R2:0.21) was three times smaller and the standard error of estimate (SEE: 0.01) was 2.5 times larger than those obtained when values were averaged over the two sampling days (Fig. 2). These results indicated that the sampling design used

in this study reduced the effects of spatial variability of soil respiration observed in previous method comparisons (Witkamp 1969; Cropper et al. 1985). Chamber comparison showed different relationships on sandy loam and organic soils. If CO2 evolution in the static chamber was

limited only by the slow diffusion of CO2 into the alkali solution, we would expect the underestimation to be dependent on the magnitude of the flux alone. The different response curves on sandy loam and organic soiis therefore suggest that an interaction between microclimatic variables ([CO2] and temperature) and soil properties (biological and physical) may exist.

In conclusion, the comparison between

static and dynamic closed chambers for measuring soil respiration produced new

observations: (1) Soil respiration rate measurements were lower using the static chamber than the dynamic closed chamber method; (2) Paired sampling design separated differences in respiration between the chambers that

were otherwise masked by spatial variability;

(3) The same equation, but with different parameter estimates, described the relation between CO2 fluxes measured bY both

608

CANADIAN JOURNAL OF SOIL SCIENCE

-v, 0.20 o|

E

cf o ot

R3=

0.15

€

z Can. J. Soil. Sci. Downloaded from www.nrcresearchpress.com by MICHIGAN STATE UNIV on 01/14/17 For personal use only.

sand (o): Y = 0.083

SEE

l/t

tine

= 0.fi)40

muck (r): Y = 0.13 (1-EXP(€.22x)

*=

o J = o

(1-EXP(-12.5sx))

0.61

O.74

SEE = O.flXi8

0.10

tr,l

ct" fJ 0.05

o F

(t,

0.05 0.10 0.15 0.20 DYNAMIC CO2 EVOLUTTON (m9 CO, m'2s{) Fig. 2. Comparison of CO, evolution measured with alkali traps (static) and with a dynamic closed clamber (dynamic) on a sandy loam (sand) and an organic (muck) soil (n :30 for both soils; the dotted line is the best fit obtained by Ewel et al. (1987) for the comparison of alkali traps and a dynamic

open system).

techniques in each of the two soils studied. These results indicate that caution must be used when comparing soil respiration values

obtained

in

studies

in

which static

and

dynamic closed chambers are used. Further tests are required to evaluate the impact of [COz] build-up and temperature ihange inside the static chamber on the COr flux determination; and to explain the diflerent parameter estimates for the equation relating CO2 fluxes measured by the two techniquei in the sandy loam and organic soils. Anderson, J. M. 193. Carbon dioxide evolution from two temperate, deciduous woodland soils. J.

Appl. Ecol. l0: 361-378.

Chahuneau, F., Desjardins, R. L., Brach. E. and Verdon, R. 1989. A micrometeorolosical facility for eddy flux measurements of Co,-and H2O. J. Armos. Ocean. Technol. 6: I93-2b0. Cropper, W. P. Jr., Ewel, K. C and Raich, J.

W.

1985. The measurement of soil CO, evolution

in situ. Pedobiologia 28: 35-40. Ewel, K. C., Cropper, W. p. Jr. and Gholz, H.

L.. 1987. Soil CO, evolurion in Florida slash pine plantations. I. Changes through time. Can. J. For. Res. 17: 325-329.

Freijer, J. I. and Bouten, W. 1991. A comparison of field methods for measuring carbon dioxide evolution: Experiments and simulation. Plant and

Soil 135: 133-142. Gupta, S. R. and Singh, J. S. 1977. Effect of alkali, volume and absorption area on the measurement of soil respiration in a tropical sward. Pedobiologia l7 : 223-239.

Hall, A. J., Connor, D. J. and Whitfreld, D. M. 1990. Root respiration during grain fitling in sunflower: The effects of water stress. Plant and l2l: 57-66. Kanemasu, E. T., Powers, W. L. and Sij, J. W. 1974. Field chamber measurements of CO, flux from soil surface. Soil Sci. ll8: 233-237. Lundegardh, H. 1927. Carbon dioxide evolution of soil and crop growth. Soil Sci. 23: 417-453. Norman, J. N., Garcia, R. and Verma, S. B. l!)90. Soil surface CO2 fluxes on the Konza prairie. Proc. Symposium on FIFE, Feb.7-9, 1990. American Meteorological Society, pp. 29-31. Rochette, P., Desjardins, R. L. and Pattey, E. 1991. Spatial and temporal variability of soil respiration in agricultural fields. Can. J. Soil Sci.

Soil

10: 189-196. Sharkov, I. N. 1984. Determination of the rate of CO, production by the absorption method. Sov. Soil Sci. 16: 102-lll.

ROCHETTE ET AL.

-

MEASUREMENT OF SOIL RESPIRATION

Statistical Analysis System Institute, Inc. 19E5. SAS user's guide: Statistics, Version 5. SAS Institute, Inc., Cary, NC. 763 pp.

Can. J. Soil. Sci. Downloaded from www.nrcresearchpress.com by MICHIGAN STATE UNIV on 01/14/17 For personal use only.

Witkamp, M. 1969. Cycles of temperature and carbon dioxide evolution from litter and soil. Ecology 50:922-924.

P. Rochette, E. G. Gregorich' and R. L. Desjardins Centre for Land and Biological Resources Research, Research Branch, Agricultwre Canada, Ottawa, Ontario, Canadn KlA0C6. CLBRR Contribution No. 9- 12. Received I l February 1992, accepted 6 MaY 1992.