Net Ecosystem Exchange of Carbon dioxide in a Temperate Poor Fen: a Comparison of Automated and Manual Chamber Techniques

Abstract

We used five analytical approaches to compare net ecosystem exchange (NEE) of carbon dioxide (CO 2 ) from automated and manual static chambers in a peatland, and found the methods comparable. Once per week we sampled manually from 10 collars with a closed chamber system using a LiCor 6200 portable photosynthesis system, and simulated four photosynthetically active radiation (PAR) levels using shrouds. Ten automated chambers sampled CO 2 flux every 3 h with a LiCor 6252 infrared gas analyzer. Results of the five comparisons showed (1) NEE measurements made from May to August, 2001 by the manual and automated chambers had similar ranges: À10.8 to 12.7 lmol CO 2 m À2 s À1 and À17.2 to 13.1 lmol CO 2 m À2 s À1 , respectively. (2) When sorted into four PAR regimes and adjusted for temperature (respiration was measured under different temperature regimes), mean NEE did not differ significantly between the chambers (p < 0.05). (3) Chambers were not significantly different in regression of ln( À respiration) on temperature. (4) But differences were found in the PAR vs. NEE relationship with manual chambers providing higher maximum gross photosynthesis estimates (GP max ), and slower uptake of CO 2 at low PAR (a) even after temperature adjustment. (5) Due to the high variability in chamber characteristics, we developed an equation that includes foliar biomass, water table, temperature, and PAR, to more directly compare automated and manual NEE. Comparing fitted parameters did not identify new differences between the chambers. These complementary chamber techniques offer a unique opportunity to assess the variability and uncertainty in CO 2 flux measurements.

Summertime CO2 fluxes and ecosystem respiration from marine animal colony tundra in maritime Antarctica

“…A 30 min enclosure is a long time for NEE measurements according to the studies in other ecosystems, such as Arctic tundra and temperate grasslands (Burrows et al, 2005;Corradi et al, 2005). For our measurements, a good concentration-time linear relation (R 2 > 0.97) was observed from NEE and ER determinations at all the regular sites (See Fig.…”

Spatial variation in CO2 exchange at a northern aapa mire

“…The peatlands in the studies by Bubier et al (2003), Tuittila et al (2004) and Riutta et al (2007a) are fens with a water table varying between -35 and ?5 cm (Burrows et al 2005 for the peatland in Bubier et al (2003), -48 and ?10 cm (Tuittila et al 2004) and -45 and 0 cm (Riutta et al 2007a) below moss surface. In the Mer Bleue bog, water table varied between 30 and 75 cm below the hummock tops (Lafleur et al 2005), a range similar to that of Kaamanen strings (Fig.…”

CO&lt;sub&gt;2&lt;/sub&gt; flux determination by closed-chamber methods can be seriously biased by inappropriate application of linear regression

“…Jensen et al, 1996;Xu and Qi, 2001;Pumpanen et al, 2003Pumpanen et al, , 2004Reth et al, 2005;Wang et al, 2006). Also, it is often applied to quantify the net CO 2 exchange between the atmosphere and low-stature canopies typical for tundra (Vourlites et al, 1993;Christensen et al, 1998;Oechel et al, , 1998Oechel et al, , 2000Zamolodchikov and Karelin, 2001), peatlands (Alm et al, , 2007Tuittila et al, 1999;Bubier et al, 2002;Nykänen et al, 2003;Burrows et al, 2004;Drösler, 2005;Laine et al, 2006), forest understorey vegetation (Goulden and Crill, 1997;Heijmans et al, 2004) and agricultural crop stands (Dugas et al, 1997;Wagner et al, 1997;Maljanen et al, 2001;Steduto et al, 2002). Advantageously, the closed-chamber method is relatively low in cost and power consumption, simple to operate and can therefore be used in remote, logistically difficult areas.…”

“…Vourlites et al, 1993;Oechel et al, 1993Oechel et al, , 1998Oechel et al, , 2000Jensen et al, 1996;Alm et al, 1997Alm et al, , 2007Goulden and Crill, 1997;Christensen et al, 1998;Tuittila et al, 1999;Maljanen et al, 2001;Xu and Qi, 2001;Bubier et al, 2002;Nykänen et al, 2003;Pumpanen et al, 2003;Burrows et al, 2004;Heijmans et al, 2004;Drösler, 2005;Reth et al, 2005;Laine et al, 2006;Wang et al, 2006). Usually, the authors justify the use of linear regression by keeping the closure time short and assuming the concentration change over time to be still in the linear range.…”