UBC Faculty Research and Publications

Quantifying summertime greenhouse gases fluxes from soils in various stages of restoration in the Burns… Christen, Andreas; Jassal, Rachhpal; Black, Andrew T.; Grant, Nick; Hawthorne, Iain; Johnson, Mark; Ketler, Rick; Lee, Sung-Ching; Merkens, Markus; Moore, R. Daniel; Nesic, Zoran; Schmid, Katrin; Stevenson, Alice; Wang, Yuexian Dec 2, 2014

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
52383-UBC_Burns_Bog_GHG_Report_141201.pdf [ 20.85MB ]
Metadata
JSON: 52383-1.0103598.json
JSON-LD: 52383-1.0103598-ld.json
RDF/XML (Pretty): 52383-1.0103598-rdf.xml
RDF/JSON: 52383-1.0103598-rdf.json
Turtle: 52383-1.0103598-turtle.txt
N-Triples: 52383-1.0103598-rdf-ntriples.txt
Original Record: 52383-1.0103598-source.json
Full Text
52383-1.0103598-fulltext.txt
Citation
52383-1.0103598.ris

Full Text

Quantifying summertime greenhouse gases fluxes from soils in various stages of restoration in the Burns Bog Ecological Conservancy Area Andreas Christen(1), Rachhpal Jassal(2), Andrew T. Black(2), Nick Grant(2), Iain Hawthorne(3), Mark Johnson(3), Rick Ketler(1), Sung-Ching Lee(1), Markus Merkens(4), R. Daniel Moore(1), Zoran Nesic(1,2), Katrin Schmid(1), Alice Stevenson(1), Yuexian Wang(1) (1) Department of Geography, University of British Columbia, Vancouver (2) Biometeorology Group, Faculty of Land and Food Systems, University of British Columbia, Vancouver (3) Institute of Resources, Environment and Sustainability, University of British Columbia, Vancouver (4) Metro Vancouver, Planning, Policy and Environment       Scientific report prepared for the  Planning, Policy and Environment Department of Metro Vancouver  December 2, 2014  Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 2 Scientific report by the University of British Columbia, Vancouver  Table of contents Table of contents ............................................................................................................. 2 Summary .......................................................................................................................... 3 1. Introduction .............................................................................................................. 5 1.1. Context of Burns Bog Ecological Conservancy Area ...................................................... 5 1.2. Why are fluxes of greenhouse gases from bogs relevant? ............................................. 6 1.3. Objectives and approach of this study ........................................................................... 7 2. Methods .................................................................................................................... 9 2.1. Study area characteristics ............................................................................................... 9 2.2. Measurement sites .......................................................................................................... 9 2.3. Instrumentation ............................................................................................................. 17 2.4. Flux calculations ............................................................................................................ 21 2.5. Modelling and up-scaling fluxes ................................................................................... 24 3. Results ..................................................................................................................... 29 3.1. Climate conditions ........................................................................................................ 29 3.2. Carbon dioxide ............................................................................................................. 31 3.3. Methane ........................................................................................................................ 38 3.4. Nitrous oxide ................................................................................................................ 41 3.5. Greenhouse warming potential of the different gases ................................................. 44 4. Discussion and Conclusions ..................................................................................... 45 4.1. Assessment of GHG fluxes ............................................................................................ 45 4.2. Limitations and future research ..................................................................................... 50 4.3. Management implications ............................................................................................ 51 Acknowledgements ....................................................................................................... 52 References ..................................................................................................................... 52 Appendix - Documentation of collars ............................................................................ 56    Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 3 Scientific report by the University of British Columbia, Vancouver  Summary Burns Bog Ecological Conservancy Area (BBECA) in Metro Vancouver is a part of a large disturbed raised bog ecosystem that is currently being managed to promote ecological recovery through rewetting. Restoration efforts aimed at raising the water table through damming of drainage structures have been in place since 2005. This study quantified summertime emissions of the three major long-lived greenhouse gases (GHGs) – carbon dioxide (CO2), methane (CH4) and nitrous oxide (N2O) from soils at sites representing different stages of recovery based on time since last disturbance within the BBECA. A sequence of four plots was chosen to represent an ecological recovery time series from a disturbed site (DS) cleared in 1998 with little vegetation cover, to a wet beakrush / three-way sedge ecosystem (BTS) disturbed by peat mining in the 1960s, a beakrush / sphagnum ecosystem (BS) disturbed by peat mining between 1930 and 1948, and a relatively undisturbed reference site consisting of a pine / sphagnum / low shrub ecosystem (PSLS). Between June 26 and August 11 2014, GHG fluxes between the soil surface and the atmosphere were measured using a portable chamber system, and syringe sampling and laboratory analysis resulting in ~500 flux measurements for CO2 and 60 flux measurements for each of CH4 and N2O. Overall, all four plots examined exhibited strong GHG emissions, dominated by CH4. The highest overall carbon dioxide equivalent GHG (CO2e) emissions were found in the water-saturated beakrush / threeway sedge (BTS) ecosystem (64.1 g CO2e m-2 day-1). This ecosystem experienced continuous anaerobic conditions with actively growing vegetation.   Figure 1 - Greenhouse warming potential for the three major GHGs emitted in Burns Bog from soils during summer in the four plots (DS, BTS, BS and PSLS). The area of the circles is proportional to the contribution of the selected species to total g CO2e in a given plot per day and square meter. The conversions are based on 100-year Global Warming Potentials (following the latest B.C. / Environment Canada practices based on IPCC AR4 values). Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 4 Scientific report by the University of British Columbia, Vancouver Table 1 - Average summertime emissions of the three major GHGs, and combined carbon-dioxide equivalent (Total CO2e emissions) measured from soils in the four plots in Burns Bog (DS, BTS, BS and PSLS). The conversion to CO2e is similar to Figure 1. Plot ID  DS  BTS  BS  PSLS  Ecosystem Type Disturbed, cleared bog  White beakrush / three-way sedge ecosystem disturbed by peat harvest in the 1960s  White beakrush / sphagnum ecosystem disturbed by peat harvest from ~1930 to 1950  Relatively undisturbed pine / sphagnum / low shrub ecosystem  CO2 net soil fluxes  (g CO2 m-2 day-1) 3.12 2.05 2.96 3.87 CH4 soil fluxes  (g CH4 m-2 day-1) 0.30 2.48 0.88 0.58 N2O soil fluxes  (g N2O m-2 day-1) 0.00004 0.00001 -0.00014 0.00008 Total CO2e emissions (g CO2e m-2 day-1) 10.68 64.14 24.88 18.48 All other plots in the sequence also exhibited significant CH4 emissions, between 1/8 to 1/3 of the emissions at BTS (Table 1). Overall, CH4 was responsible between 71% and 97% of net CO2e emissions in the plots (Figure 1). Measured CH4 fluxes were highly variable in space and time.  The GHG exchange of CO2 due to photosynthesis and respiration was of secondary importance compared to CH4. Overall, the soils in the plots were weak sources of CO2, however supplementary measurements using the eddy-covariance approach on a flux tower showed that at this time of the year the entire ecosystem (including tall grasses and trees) is a weak carbon sink, i.e. sequestering carbon – however, at a lower rate than most pristine wetlands (-3.3 g CO2 m-2 day-1). The measured ecosystem is not a system of high productivity, but with considerably limited respiration due to oxygen limitation. No significant emissions or uptake of N2O were found (Figure 1, Table 1). Although summertime measurements can be expected to be highest, full-year measurements of CH4 and CO2 are needed to more accurately quantify and account for the effect of rewetting and ecological recovery on GHG emissions on an annual basis, and to identify effective GHG emission mitigation strategies that can be considered in future restoration management of the BBECA.  Keywords: Burns Bog, carbon dioxide, greenhouse-gases, methane, nitrous oxide, photosynthesis, respiration, soils. Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 5 Scientific report by the University of British Columbia, Vancouver 1. Introduction 1.1. Context of Burns Bog Ecological Conservancy Area The Burns Bog Ecological Conservancy Area (BBECA) is located between the south arm of the Fraser River and Boundary Bay in the Municipality of Delta, British Columbia, Canada (Figure 2). The BBECA is part of a remnant ombrotrophic raised bog ecosystem recognized as one of Canada’s largest undeveloped natural areas retained within an urban area. The entire remaining bog is a unique ecosystem complex of global significance based on its chemistry (influenced by the nearby marine environment), form, location on a large estuarine delta, flora and large size (Hebda et al,. 2000; McDade 2000). It supports distinctive bog vegetation communities and recognized rare and endangered plant and wildlife species. While not pristine, the BBECA has retained enough of its ecological integrity to allow its restoration over time (Hebda et al., 2000; McDade, 2000). Historically, Burns Bog covered approximately 48 km2. Starting in the early 1900s, the encroachment of agriculture, industrial uses, numerous landfills and the development of transportation and utility corridors has altered and alienated close to 40% of the Bog. The hydrology and ecology of the remaining bog has been further disrupted by marginal and internal ditching, peat extraction and related activities and neighboring land use continues to be of concern.   Figure 2 - Geographic Location of the Burns Bog Ecological Conservancy Area (BBECA) in Delta, British Columbia. Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 6 Scientific report by the University of British Columbia, Vancouver In 2004, the government of Canada, the province of British Columbia, the Greater Vancouver Regional District (GVRD) and the Corporation of Delta purchased a large portion of the remaining undeveloped area (roughly 20 km2) to establish the Burns Bog Ecological Conservancy Area (BBECA). The BBECA contains about 14 km2 of disturbed wetland ecosystems with previous use of peat mining, agriculture or recreation, and about 6 km2 of relatively undisturbed raised peat bog (Metro Vancouver, 2007). The primary land management objective for the BBECA is the support of ecological recovery through restoration, conservation and protection as a raised bog ecosystem. However, this is challenging because the BBECA is located at the climatic limits for raised bogs. The restoration is primarily achieved by controlling the water table using a ditch-blocking program. A pilot study started in 2005 to control water outflow using weirs and peat dams in the southwestern sector of the BBECA. In this pilot study area, water levels and vegetation response are thoroughly monitored (Metro Vancouver, 2007). This project should inform and guide future restoration plans for the entire BBECA.  1.2. Why are fluxes of greenhouse gases from bogs relevant?  Generally, wetlands, including raised bogs, in their pristine state, are ecosystems that accumulate substantial amounts of atmospheric carbon dioxide (CO2) and sequester it over time in the form of peat in the catotelm, the permanently wet layers below the surface containing only dead organic matter. In water-saturated peat, dead plant material decomposes very slowly because of oxygen limitation (Bleuten et al., 2006), associated low water temperatures, and small microbial populations. Peatlands are the most space-efficient carbon (C) stock of all terrestrial ecosystems, and contain globally about 500 Gt of C. This is comparable to twice the carbon contained in all global forests, or 75% of the carbon currently in the atmosphere (Joosten, 2013). When draining wetlands, this organic carbon is exposed to atmospheric oxygen and consequently becomes available for oxidation by microbes, vulnerable to fires and removal by erosion. The decomposed organic matter of a bog under degradation will eventually enter the atmosphere in form of CO2. Drained peatlands are emission hot spots and it is estimated that globally anthropogenic peatland degradation and fires account for about 6% of all global anthropogenic emissions of CO2 (Joosten, 2013). Therefore, protecting and maintaining wetlands including bogs will ensure that the carbon stored in the peat remains sequestered.  However, under anaerobic conditions in bogs, microbial respiration can also result in the formation of methane (CH4) via a process called methanogenesis. Metanogenesis occurs in water-saturated soils with high organic material under warm temperatures. The amount and timing of CH4 emissions from wetland soils depend on the microbial communities that are responsible for methanogenesis and possibly subsequent CH4 oxidation in the soil. CH4 emissions also depend on the mechanisms that transport the CH4 formed in the lower layers of the peat to the surface and the atmosphere (Hendricks et al., 2010).  Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 7 Scientific report by the University of British Columbia, Vancouver Restoration efforts may change emissions and uptake of long-lived atmospheric greenhouse gases (GHGs) in rewetted areas. It has been reported that rewetting of previously disturbed wetlands can initially cause high to extreme emissions of CH4, and also – if the previous land cover was agricultural – emissions of nitrous oxide (N2O). At the same time carbon sequestration will be limited. However, it has been found that any negative effects are transient and pay off in the long-term when bogs return to their pristine state and re-enter a mode of sequestering atmospheric carbon. Work in the 1970s estimated the remaining peat volume in Burns Bog to be roughly 1.09 x 108 m3 and the dry mass of the organic (peat) material to be 3,949,440 metric tons (Biggs, 1976). This corresponds to roughly 2 Tg (2 Mt) of carbon (C) (Turunen, 2008). The source also estimated net primary productivity (NEP) based on weight of biomass of Sphagnum in revegetated mined areas as 129 g C m–2 year-1. Hummock sites were attributed an NEP of 168 g C m-2 year-1 and the NEP of depressional sites was estimated on average 70 g C m-2 year-1 (Biggs, 1976). To date, no direct emission or exchange measurements are available for any of the major long-lived GHGs from soils in the BBECA. Previous assessments and reports on Burns Bog (Catherine Berris Assoc., 1993; McDade, 2000) and the current management plan (Metro Vancouver, 2007) investigated and summarized the impact of the restoration on vegetation, wildlife, hydrology, water chemistry, and discuss the socio-cultural relevance of the bog. No data are available that would help to quantify the GHG emissions and carbon sequestration rates associated with pathways leading to either degradation, fires, or restoration. 1.3. Objectives and approach of this study The overall objective of this study was to quantify the fluxes of all three major long-lived GHGs - carbon dioxide (CO2), methane (CH4) and nitrous oxide (N2O) - from the soil surface of selected ecosystems in the pilot study area of the BBECA in summer of 2014. Four sites at different stages of recovery that are under the influence of increased water table elevations were chosen to quantify fluxes across a sequence from a recently disturbed site to a relatively intact pine / sphagnum / low shrub reference ecosystem. Summertime conditions were chosen because previous work established that warm water, high soil and air temperatures, and high solar irradiance generally favor emissions of CH4 and N2O and maximize the exchange of CO2 by photosynthesis and respiration (assuming no water limitation). These data are complemented by continuous CO2 flux measurements from an eddy-covariance tower that enables up-scaling of emissions. This report will make recommendations as to whether longer-term monitoring of GHG emissions is required, to quantify more precisely the land-cover change effects of the rewetting on emissions to be accounted for and managed in accordance with Metro Vancouver’s Integrated Air Quality and Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 8 Scientific report by the University of British Columbia, Vancouver Greenhouse Gas Management Plan (Metro Vancouver, 2011). Such future work could also help to identify GHG emission mitigation strategies in the management of the BBECA, and inform effective carbon sequestration strategies in combination with efforts to accelerate ecological recovery.   Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 9 Scientific report by the University of British Columbia, Vancouver 2. Methods 2.1. Study area characteristics Burns Bog exhibits the typical characteristics of a raised bog ecosystem: • a peat mound above the regional water table, • a persistent near-surface internal water mound, • acidic nutrient-poor water derived directly from rainfall, • a two-layered peat deposit consisting of an acrotelm and catotelm, and • widespread peatland vegetation communities dominated by peat moss (Sphagnum) and plants belonging to the Heather family (Ericaceae). Twenty-four ecosystem types were identified, mapped and described for Burns Bog (Madrone 1999) using terrestrial ecosystem mapping methodologies (Resources Inventory Committee 1998). Extensive white beakrush (Rhynchospora alba) - Sphagnum meadows covered 56% of the bog area classified by Madrone (1999) at the time of classification. Dominated by near complete cover of Sphagnum carpets, these sites have a noticeable yet minor component of tawny cotton-grass (Eriophorum virginicum), sundew (Drosera rotundifolia), and false asphodel (Tolfieldia glutinosa). Scattered scrub pine (Pinus contorta) and Labrador tea (Ledum groenlandicum) dot the landscape and bog cranberry (Oxycoccus palustris) trails over the Sphagnum carpets. Bog blueberry (Vaccinium uliginosum) is also common across these sites.   2.2. Measurement sites 2.2.1. Soil chamber plots Soil GHG fluxes were quantified by means of chamber measurements in a sequence of four plots, representing a progression of expected growth resulting from ecological recovery and adaptation to rewetting. The sequence included a recently disturbed surface (DS), a white beakrush/three-way sedge ecosystem (BTS) disturbed in the 1960s with a high water table, a moderately wet beakrush / sphagnum ecosystem (BS) disturbed in the 1930s and 1940s and a not directly disturbed (yet affected by drainage) pine / sphagnum / low shrub ecosystem (PSLS) at the end of the sequence. All plots were selected in the southwestern sector of the BBECA (Figure 2, Table 2).  Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 10 Scientific report by the University of British Columbia, Vancouver  Figure 3 - Ecological classification of Burns Bog and location of greenhouse gas flux monitoring sites.  Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 11 Scientific report by the University of British Columbia, Vancouver  Figure 3 - Disturbed site (DS), photographed on June 13, 2014 from the chamber measurement plot in Eastern direction. The disturbed site (DS) was prepared for commercial cranberry cultivation in 1998. At that time the field was cleared, existing vegetation was removed from the surface and a uniform, level field was established consisting of exposed catotelmic peat. Commercial cranberry was never planted on site and the field was left to undergo natural colonization by the surrounding plant communities after abandonment. Until 2010 the fields remained largely bare of vegetation, however, since then, the site has developed relatively sparse vegetation cover consisting primarily of low growing Labrador tea patches, widely spaced birch (Betula pendula) trees and commercial blueberry (Vaccinium corymbosum) bushes, as well as scattered herb and other shrub species characteristic of bog ecosystems (Figure 3). Sphagnum cover is extremely low and occurs as small hummocks of Sphagnum capillifolium often associated with some Labrador tea patches. The field is one of six that were last to be cleared before the Burns Bog Ecological Conservancy Area was established in 2004. The site has a high water table for most of the year, but, at the height of summer, surface cracking of the peat can be seen. As indicated by extremely low Sphagnum cover, this site cannot be considered a peat producer. Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 12 Scientific report by the University of British Columbia, Vancouver  Figure 4 - Beakrush / three-way sedge (BTS) site plot photographed on June 13, 2014. Vegetative cover of the white beakrush / three-way sedge (Dulichium arundinaceum) (BTS) site is noticeably incomplete with large patches of algae developing in some shallow wet pools. Scattered bog shrubs including Labrador tea, bog blueberry and sweet gale (Myrica gale) exist, primarily on drier hummocks (Figure 4). Sphagnum mounds are sometimes found within the shrub hummocks. This site is noticeably wetter than the other three sites of the sequence with the water table remaining very near the surface for much of the year. A blocked ditch adjacent to the sampling plot contributes to wetting the area. The Atkins-Durbrow Hydropeat method of peat removal was used to harvest peat from this sampling site between 1963 and 1966. This method involved cutting down trees, then blasting the peat surface with pressurized water to dislodge peat from exposed tree roots. The resulting slurry was then pumped in pipes to the processing plant. Remnant strips or baulks usually separated the harvested areas. These elevated baulks between the rectangular fields created by harvesting were covered by pine Sphagnum heath which in many cases is on an undisturbed surface. Fewer drainage ditches are associated with areas harvested by this method relative to areas requiring some level of drying/drainage prior to harvest using hand cutting or vacuum methods. Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 13 Scientific report by the University of British Columbia, Vancouver   Figure 5 - Beakrush / sphagnum (BS) site photographed on June 13, 2014 from the measurement plot in Eastern direction The white beakrush / sphagnum (BS) site (Figure 5) was harvested using the Atkins-Durbrow Hydropeat harvesting method sometime between the 1930s and 1948. The site has a high water table, often near (within 25 cm) surface for most of the year.  Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 14 Scientific report by the University of British Columbia, Vancouver   Figure 6 - Pine/sphagnum/low shrub (PSLS) site in June 2014. In the foreground visible are a turquoise large (L-) collar and a white small (S-) collar. The pine / sphagnum / low shrub (PSLS) site is relatively undisturbed, having never been cleared of vegetation for peat harvest or agriculture. This site is characterized by scattered stunted scrub pine    over a carpet of Sphagnum (Figure 6). Hollows and hummocks of Sphagnum tenellum and Sphagnum capillifolium mixed with ericaceous shrub species and patches of maritime reindeer lichen (Cladina portentosa) provide nearly complete cover. The site has a relatively high water table for most of the year but likely was compromised by a lowering of the water table for some period of time prior to the BBECA being established.     The BS and wetter sites of the PSLS are considered to be peat producers because they provide conditions suitable for peat accumulation and decomposition.   Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 15 Scientific report by the University of British Columbia, Vancouver Table 2 – Summary of ecosystem characteristics, historical disturbances, location of measurement plots, and numbers and size of collars for soil flux measurements in the BBECA. Site ID  DS  BTS  BS  PSLS  Flux Tower  Location of measurement plots (WGS-84) 123°00'02.97"W 49°06'37.20"N 123°00'01.42"W 49°07'08.80"N 122°59'47.76"W 49°07'09.37"N 123°01'02.96"W 49°06'34.01"N  122°59’05.87"W 49°07’47.20”N Ecosystem Type Disturbed  White beakrush / three-way sedge  White beakrush / sphagnum  Pine / sphagnum / low shrub  White beakrush / sphagnum  Disturbance Type Cleared bog Peat harvest Peat harvest Lowering of water table Peat harvest Peat harvest Method N/A Atkins-Durbrow Hydropeat Atkins-Durbrow Hydropeat N/A Atkins-Durbrow Hydropeat Period of Disturbance 1998 1963-1966 1948 or before Started 1930s 1957-1963 Measurements Chamber measurements (CO2, CH4, N2O) Eddy covariance (CO2) No of collars for chamber measurements 8 (S-), 6 (L-) 8 (S-), 6 (L-) 8 (S-), 6 (L-) 8 (S-), 6 (L-) N/A (a) S- indicates small (10 cm diameter) and L- indicates large (21cm diameter) collars. Only five of the six CH4 / N2O (L-) collars were used for sampling in this study. In the four plots (DS, BTS, BS and PSLS), a total of 56 collars were installed (of which 52 were used in this study) to allow the repetitive sampling of soil fluxes of CO2, CH4 and N2O at the same locations. In each of the four plots, a total of six large collars (21 cm diameter, 0.0692 m2, prefix “L-”) and eight small collars (10 cm diameter, 0.0157 m2, prefix “S-”) were installed in June 2014. Collar locations were chosen to represent different micro-environments with respect to water content, vegetation cover, and proximity to trees. The strategy when placing collars was to best capture the potential variability of the soil fluxes of CO2, CH4 and N2O at a given ecosystem (‘screening’) rather than obtain a representative spatial sample (‘area average’). Collars included small ground vegetation or grasses but no shrubs, taller grasses or trees. The small collars were used to measure CO2 fluxes and the large collars were used to measure CH4 and N2O fluxes. The measurement arrangement to measure the fluxes of the latter two gases with reasonable accuracy required a larger collar area to height ratio.  Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 16 Scientific report by the University of British Columbia, Vancouver The collars were installed in a way to minimally disturb the soil and the vegetation, although installation required cutting roots or Sphagnum down to the 5-cm depth with a sharp knife. The small (CO2) collars all protruded 4 cm above the local surface (see Appendix Table 17). The height of the large collars was adjusted to account for the varying height of vegetation and at the same time to maximize the area to height ratio. The heights ranged between 8 and 17 cm (see Appendix Table 16). All collars were installed in early to mid June 2014, about three to four weeks before the first flux measurements took place to allow for soil settling and vegetation regrowth.  2.2.2. Eddy-covariance tower As part of a related research project funded by Metro Vancouver and carried out by the University of British Columbia, a micrometeorological flux tower on a floating platform was installed over a white beakrush / sphagnum ecosystem in the central part of the BBECA (122°59’05.87"W, 49°07’47.20”N, WGS-84) in Summer 2014. The site was harvested between 1957 and 1963 using the Atkins-Durbrow Hydropeat harvest method. The site has a higher water table than comparable site BS and contains small open water ponds at times. Sphagnum carpets are discontinuous and pine and birch trees are dispersed and appear to be growing on the remnants of baulks.    Figure 7 – Photo of the flux tower seen from east on 25 July 2014. See also title page for a view to the south. Although the tower’s main research objective is the determination of evapotranspiration from this ecosystem, the tower’s instrumentation allowed the continuous measurement of fluxes of CO2 using the eddy-covariance (EC) method from 9 July to 11 August 2014. The EC CO2 fluxes were used to Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 17 Scientific report by the University of British Columbia, Vancouver complement and extrapolate the chamber measurements of CO2 fluxes in the four plots. The EC measurements were representative of the CO2 fluxes from an entire BS ecosystem including tall grasses and selected shrubs. The typical daytime 90% turbulent flux footprint of the tower extended up to 200 m from the tower and was estimated to be representative of ~20,000 m2 of the BS ecosystem compared to an area of 0.0692 m2 and 0.0157 m2 for the large and small chambers, respectively. No EC measurements of CH4 or N2O fluxes were made. 2.3. Instrumentation 2.3.1. Chamber measurements of soil CO2 fluxes Soil CO2 fluxes were measured with a non-steady state portable chamber system, developed and tested in previous research projects at the University of British Columbia (Jassal et al. 2005, 2007). The measurement head was a chamber with a volume of 1.4 x 10-3 m3 (height: 15.6 cm, diameter: 10.7 cm), measuring flux from a surface area of 79 cm2 (the inner diameter of the collar). Both opaque (PVC) and transparent (acrylic) chambers were used, depending whether soil respiration (opaque), or net ecosystem exchange (NEE, i.e. respiration minus photosynthesis) (transparent) was of interest. The chamber was placed on each small collar for 2 minutes. A foam gasket provided a seal between the collar and the chamber. A pump (flow rate of 600 cm3 min-1) circulated air from the chamber into a portable, battery operated infrared gas analyzer (IRGA) (LI-800, LI-COR Inc., Lincoln, NE, USA) and back into the chamber through a closed circuit. The IRGA measured CO2 and water vapour concentrations at 1-second intervals during the run. The data from the IRGA were digitized using a 21X data logger (Campbell Scientific Inc. (CSI), Logan, UT, USA) and stored on a memory module to later calculate the rate of change in the CO2 mixing ratios (d[CO2]/dt) in the chamber over the period of the measurement (see Section 2.3). Simultaneously, a thermocouple was used to measure soil temperature at the ~5-cm depth just outside the collar. For measurements with the transparent chamber, a quantum sensor (LI-190, LI-COR Inc.) measured photosynthetic photon flux density (PPFD in µmol m-2 s-1) during the period of the measurement. Start and end time of each chamber measurement were recorded. The IRGA was powered for at least 15 minutes to warm it up before the first measurement.  Prior to this study, the IRGA was calibrated using a two-point calibration in the laboratory while the system operated in standard measurement mode. Two calibration gases (N2 for 0 ppm and a 414.07 ppm CO2 in dry air for the span gas) were subsequently injected into a tee junction at the end of the sample tube. The calibration gas flow rate was maintained slightly over the system’s sample flow rate so that the excess flow from the calibration tank could escape through the open end of the tee. The span gas cylinder was calibrated against standards from the Greenhouse Gases Measurement Laboratory (GGML), Meteorological Service of Canada, and is traceable back to NOAA/CMDL Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 18 Scientific report by the University of British Columbia, Vancouver calibration facilities in Boulder, USA. The typical standard deviation (noise) of the IRGA was < 0.7 ppm. With the opaque chamber, a total of 314 measurements of soil respiration were made on 16 days during the study period (see also Figure 12). Using the transparent chamber, a total of 64 light response curves were measured on two days during the study period using the shade cloth methodology (Marini and Sowers, 1990). For the light response curves, chamber measurements were repeated four times in a sequence at the same collar while progressively covering the chamber (and the quantum sensor) with cloth of decreasing transparency following same methodology as in Hum (2006). Starting with an un-shaded run where the sun entered the vegetation in the collar unhindered, a light (39.4% transparency), medium (10.8% transparency) and completely dark cloths (0% transparency) were mounted successively over the chamber to simulate different light levels. Those runs were completed within 20 minutes to avoid substantial changes in solar altitude, incoming PPFD and other environmental controls. 2.3.2. Chamber measurements of CH4 and N2O fluxes CH4 and N2O fluxes were measured using syringe sampling in the 2-cm-diameter (large, L-) collars over a period of 30 minutes following a procedure similar to that in Jassal et al. (2011). During the 30 minutes, collars were covered with a transparent acrylic lid equipped with a foam gasket. A battery-operated fan ensured mixing within the chamber headspace, while a septum in the lid allowed samples to be withdrawn from the headspace. The first syringe sample (A) was taken with a 30-ml syringe (21G) at the beginning of a 30 min period but without the lid. Immediately after the first sample was taken, the lid was put on the collar. Subsequent samples (B, C, D, and E) were taken from the headspace, with the needle inserted through the septum (again with the same 30 ml syringe / needle, when possible using the same pierce hole). These were taken at roughly 5 (B), 10 (C), 20 (D), and 30 (E) minutes after sample A. For each lid, the septum was replaced at the end of a sampling day. Syringe samples were immediately transferred into evacuated 12-ml exetainers (Labco Ltd., Buckinghamshire, UK). All exetainers were stored and transported in a cooled box and, on the same day, were transferred to a temperature-controlled, cooled storage room at the University of British Columbia. Exetainers were stored upside down in water to prevent any leaking before analysis in the laboratory. Over the study period, 3 measurements were made per collar, resulting in a total of 300 samples (5 samples per run, 5 collars per plot, 4 plots). The three dates for measurements (17/18 July, 23/24 July and 11 August, 2014) were chosen to represent different soil wetness conditions. Soil volumetric water Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 19 Scientific report by the University of British Columbia, Vancouver content was the highest on 17/18 July and the lowest on 11 August. The samples on 24/25 July were taken immediately after a rain event of 11 mm (see Figure 12).  Figure 8 – Photo of a closed large collar (L-) while performing a run to measure CH4 and N2O effluxes. The coiled black tube allows for pressure equalization while minimizing diffusion out of the chamber. The analysis of CH4 and N2O in the exetainers was performed using an Agilent 7890A (G3440A, Agilent Technologies, Santa Clara, California, USA) gas chromatogram system (GC) with a flame ionization detector (FID) and an electron capture detector (ECD).  Table 3 – Specifications and system accuracy for the GC system used to analyze samples of CH4 and N2O.  Methane (CH4) Nitrous oxide (N2O) Inlet SS purged-packed SS purged-packed Backflush Column SS 12 ft HaySep Q 80/100 (Col 1) SS 12 ft HaySep Q 80/100 (Col 1) Analyte Separation Column SS 6 ft HaySep Q 80/100 (Col 2) SS 6 ft HaySep Q 80/100 (Col 2) Analyte Delivery Column SS 6 ft HaySep Q 80/100 (Col 2) Fused Silica (Col 3) Detector/ Temp. (oC) FID/ 250 ECD / 300 Oven Temp. (ºC) 200 200 Air / H2 flow rate (ml min-1) Air: 350  H2: 60 Not applicable Retention time (min) 3.06 6.5 Calibration Range a 1.11-5 ppmv 0-1 ppmv System Accuracy b 16 ppbv 2 ppbv Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 20 Scientific report by the University of British Columbia, Vancouver a Range of calibration standards injected into vials using 20-ml medical syringe with a 23-gauge needle. b Determined from middle of range calibration standard value sampled using Combi-Pal autosampler. Samples were injected by a Combi-Pal autosampler (CTC Analytics, Zwingen, Switzerland) capable of sampling 100 exetainers. Samples were drawn from the exetainers using a 2.5 ml N2 purged glass syringe, with HD-Type PTFE tipped syringe plunger and 23-gauge needle (CTC Analytics AG, Zwingen, Switzerland) and injected into the GC stainless steel, heated (110ºC) purged-packed inlet using N2 (99.999%) as a carrier gas with a flow rate of 21 ml min-1 at constant flow (column 1) and 22.3 ml min-1 (column 2) at constant pressure. 2.3.3. Eddy-covariance measurements of CO2 fluxes Ecosystem fluxes of CO2 were measured using an EC system on the micrometeorological flux tower (see Section 2.2.2). The EC system consisted of an ultrasonic anemometer-thermometer (CSAT-3, CSI), and a H2O/CO2 open-path infrared gas analyzer (Li-7500, Licor Inc.), which was installed at a height of 1.8 m on a boom extending out from the flux tower’s scaffold (Section 2.2.2). The system measured vertical wind speed w (in m s-1) and CO2 molar density ρc (in mmol m-3) at 10 Hz. Data were sampled on a data logger (CR1000, CSI) and stored on a CompactFlash Module (NL115, CSI). Prior to deployment, the IRGA was calibrated in the laboratory using a two-stage calibration process with N2 for 0 ppm and span gas (similar to the portable chamber system IRGA, Section 2.3.1) 2.3.4. Climate measurements Measurements of environmental variables for extrapolating and modelling fluxes over the season were made. In each of the four plots, an automatic soil climate logger (CR10X, CSI) was equipped with continuously recording soil temperature sensors to measure soil temperature at 5 cm below the surface (Type T-thermocouples, custom made at the University of British Columbia) and two TDR sensors (CS616, Campbell Scientific Inc.) vertically installed to integrate soil volumetric water content, θw from the surface to a depth of 30 cm. Those systems were continuously operated for the entire study period. At the flux tower, a four-component net radiometer (model CNR-1, Kipp & Zonen, Delft, Netherlands) provided continuous, unobstructed short-wave irradiance !↓ (in W m-2), representative for the study period. !↓ was used to model photosynthesis (see Section 2.5.2).    Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 21 Scientific report by the University of British Columbia, Vancouver 2.4. Flux calculations 2.4.1. Calculating fluxes from CO2 chamber measurements Digital data from the portable chamber systems contained CO2 mixing ratio, water vapour concentrations, temperature and photosynthetically active radiation at 1 Hz. The data were imported into a custom-developed software/user interface developed at the University of British Columbia1, based on IDL Version 8.1 (Excelis Inc., McLean, VA, USA). The start and end-time of each run were entered based on field notes and the data were tagged for specific plot and collar IDs. Then the software was used to convert CO2 concentration into mixing ratios and plot the mixing ratio vs. time relationships for each chamber run and fit a linear regression over the 2-minute period following the chamber placement on the collar (Figure 9). Data from the first 20 seconds of the run were discarded to avoid inadequate mixing and the effects of pressure fluctuations during and shortly after chamber placement. The slope s of the linear regression (in ppm s-1 or µmol mol-1 s-1) was then used to calculate the molar flux of CO2, !!"! (in µmol m-2 s-1) as follows:  !!"! = !!!!!  (Eq. 1) where A is the surface area of the collar (0.0079 m2), V is the chamber volume (0.0014 m3), and m is the molar density of air (in mol m-3) at the time of measurement. The molar density was calculated considering the measured water vapour concentration in the collar. The software calculated the root mean square error (RMSE) between the linear regression and the measurements (in ppm) of the 1Hz data during the selected period as follows:  RMSE =1! ([CO!]!,!−[CO!]!,!)!!!!! (Eq. 2) Where [CO!]!,!!is the measured CO2 mixing ratio over the time of the chamber measurement (discarding 20 seconds at the start), and [CO!]!,! is the CO2 mixing ratio calculated based on the linear regression. The median RMSE of all chamber runs was 0.9 ppm, and values as low as 0.7 ppm were                                                 1 The source code and a complied version of the analysis software is available for download through  https://github.com/achristen/UBC-Portable-CO2-Chamber. Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 22 Scientific report by the University of British Columbia, Vancouver obtained. A threshold RMSE of 2 ppm was chosen as acceptable. Runs with a RMSE > 2 ppm were discarded (7 out of 522 measurements). None of the measurements were affected by water vapour condensation. Measurements with the opaque chamber that resulted in negative flux values > -0.05 µmol m-2 s-1 were also discarded because they are physically implausible.   Figure 9 - Example of data from the portable CO2 chamber system over a period of 2 minutes. Data points with round markers are times when the chamber was closed. The green points were used for calculating a linear regression (in blue), while data points in red are discarded (first 20 seconds) to avoid influences from pressure effects when closing the chamber. The black line shows the measurements before and after the chamber head was placed on the collar and were not used. 2.4.2. Calculating fluxes from CH4 and N2O chamber measurements A similar procedure was used to calculate CH4 and N2O fluxes from the measured mixing ratios in the laboratory (Section 2.3.2). However, no continuous trace of increasing mixing ratios were available, instead only five measurements over the period of 30 minutes in each run were used for regression (Figure 10). For CH4, it was found that concentration increased substantially in some cases. Maximum mixing ratios for CH4 in the chambers climbed to an unexpected value of 1473 ppm, which corresponds to an increase by a factor of 800 from the start of the measurement. Compared to CO2, for which the increase in mixing ratio was always less than a factor of 2 between start and end times of a chamber run, the substantial increases observed in CH4 in some cases caused non-linear increases with time (see example in Figure 11). A logarithmic function was fitted through the five mixing ratios:! CH! (!) = !!!ln!(! + !)+ !!! (Eq. 3) Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 23 Scientific report by the University of British Columbia, Vancouver where t is the time (s) since the start of the chamber run, and a1 and a0 are empirical fitting parameters. b is a time offset (s) that satisfied b > 0. The rate of change r was then calculated for 10-s time increments and the average of all increments from 1 to 500 s were used to obtain an average !!→!""! (in ppm s-1 or µmol mol-1) to calculate the fluxes   !!"!,!!! = !!!!!→!""!!  (Eq. 4) Here, the area A is 0.0692 m2, but V is specific to each collar as V = Ah, where h is the height of the collar (in m, see Table 16). The RMSE between measured and fitted mixing ratios was calculated – as stated above for CO2 (Eq. 2) but for the logarithmic function.  Figure 10 - Example of time course of CH4 mixing ratios measured in a chamber volume over a period of 30 min. The blue dots are the measured mixing ratios, the dotted line is the linear regression and the full line is the logarithmic fit to the five points. In this case there is no substantial difference between the log and linear fits. Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 24 Scientific report by the University of British Columbia, Vancouver  Figure 11 – Same as Figure 10 but for a case with substantial initial rise in CH4 mixing ratios where the response is clearly not linear. Note that in this case the logarithmic curve is a better fit. 2.4.3. Eddy covariance flux calculations for CO2  Fluxes of CO2 were calculated based on 10-Hz data of stored vertical wind and carbon dioxide molar densities over blocks of 30 minutes for the period between 9 July and 11 August 2014. Before flux calculation, the coordinate system of wind components was corrected for tilt (double rotation). Fluxes are corrected for air volume and density fluctuations (Webb et al. 1980) and for high frequency flux losses based on path averaging of the sensors (Moore, 1986). Data was quality checked following procedures outlined in Crawford et al. (2009).  2.5. Modelling and up-scaling fluxes  For the soil fluxes of CH4 and N2O, averaged statistics were calculated per collar and per plot without considering potential temporal variability. For soil fluxes of CO2, the strong diurnal variability due to environmental controls (i.e. photosynthetically active radiation and soil temperature) inhibited a straight averaging of the daytime measurements. The CO2 release due to soil respiration (Rs in µmol m-2 s-1) and the CO2 uptake due to photosynthesis (P in µmol m-2 s-1) were separately modelled, and then aggregated into a long-term net CO2 flux (!!"!). Section 2.5.1 discusses how Re is modelled and extrapolated to the study period, and 2.5.2 describes the analogous procedure for P.  2.5.1. Modelling and extrapolating soil respiration During the study period, a total of 309 successful measurements of Rs were made with an opaque chamber (where P is zero). All Rs values from a particular collar were plotted against simultaneously Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 25 Scientific report by the University of British Columbia, Vancouver measured soil temperature, T. A total of 51 additional measurements from September and October 2014 were added to include measured Rs at lower T. Although θw can be another important control on Rs in wetlands (Luo and Zhou, 2006), it changed little in the layer from the surface to the 30-cm depth over the study period except in the PSLS plot. For simplicity and due to the limited number of measurement over a larger range of θw, Rs was modelled only as a function of T. All Rs measurements at a collar were used to fit an empirical temperature-dependency curve based on the work of Lloyd and Taylor (1994):  !! = !!"#!exp! !1!!"# − !!− !1! − !! (Eq. 5) where Rref is the reference respiration (in µmol m-2 s-1) at T = Tref. T is the soil temperature (in ºC), Tref is the reference temperature set to 20ºC, and T0 is the temperature at which all respiration ceases (set to -46.02ºC). Using a gradient-expansion algorithm to compute a non-linear least squares fit, Rref and the empirical parameter, E, were determined for each collar (8 collars at 4 plots = 32 sets of parameters). While Rref was not constrained, E was only allowed to vary within the range of 150 to 600 K. Individual collar values for Rref and E were then used to extrapolate Rs (t,c) based on (Eq. 5) to each time step t and collar c using the continuously measured soil temperatures T in each plot (see Section 2.3.4). The two soil temperatures measured continuously by thermocouples in each plot were averaged for each time step. This modelling was done at the 5-minute scale over the entire study period (26 June to 11 August, 2014). The study period average !! was then calculated per plot as the averaged Rs over all time steps (t) and all eight collars (c) at each plot: !! !=1!!8 ! !!(!, !)!!!!!!!! (Eq. 6)  where N is the number of time steps in the study period. 2.5.2. Modelling and extrapolating photosynthesis For each collar, two light response curves were measured at four different photosynthetic photon flux densities (PPFD in µmol m-2 s-1) each using shading cloth of different transparency over a transparent chamber (see section 2.3.1 for details). This resulted in a total of 64 light response curves based on 256 values (4 values per curve, 2 curves per collar, 8 collars, 4 plots). Out of the 256 measurements, 230 were successful and 26 were discarded due to measurement problems / poor quality. For each light response curve, the rate of photosynthesis P (in µmol m-2 s-1) at a given PPFD was calculated as Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 26 Scientific report by the University of British Columbia, Vancouver !(!!"#) = − !!"#!(!!"#) − !!"#!(!!"#!!)  (Eq. 7)  where !!"#!(!!"#) is the measured flux of CO2 at the given light level (PPFD). !!"#!(!!"#!!) is the measured flux of CO2 with a completely dark cover (PPFD = 0 µmol m-2 s-1) in the same response curve. The minus sign is introduced to retrieve positive !(!!"#) for uptake of CO2. If (Eq. 7) resulted in a negative !(!!"#) then P was set to 0 µmol m-2 s-1. Slightly negative P values occurred for 17 out of the 230 valid measurements due to measurement noise for collars with essentially no photosynthesis (bare soil or standing water). For each of the 64 light response curves, a non-rectangular hyperbolic function (Ögren and Evans, 1993) was fitted through measured PPFD and corresponding !(!!"#) following (Eq. 7) as follows: !(!!"#) !=!!PPFD+ !! − !!PPFD+ !! ! − 4!! !PPFD!!!2!  (Eq. 8)  where Pm is the maximum P at light saturation (in µmol m-2 s-1), C is the curvature and ! is the maximum quantum yield (in µmol mol-1). Using a non-linear least squares fit, Pm was determined for each curve. The curvature was fixed to C = 0.7 and the maximum quantum yield was determined to be ! = 0.01 based on measurements on the flux tower (see discussion of Figure 16). For each collar the two retrieved Pm values from the two different dates were averaged into a single Pm per collar. To model P for each collar and time step over the entire study period (June 26 to August 11, 2014), first a continuous time trace of PPFD was constructed. Continuous PPFD was determined from !↓ measured at the flux tower (see Section 2.3.4) converted to PPFD using !!"# = !! ↓, where j = 2.01 mol J-1. The conversion factor j was determined on long-term simultaneous measurements of PPFD and !↓ at the University of British Columbia’s Climate Station (Crawford and Christen, 2014). For time steps when no measurement of !↓!was available at the flux tower (this is before the tower was established on July 9, 2014) !↓ for a station 12 km north of the BBECA on a tower in the City of Vancouver (‘Vancouver-Sunset’, 49°13' 34.0" N 123°04'42.2"W) was used instead. The study period average ! for a plot was then calculated as the averaged P over time (t) and all eight collars (c) at that plot using and equation similar to (Eq. 6), but replacing Rs by P. Study-period averaged !!"! (in µmol m-2 s-1) was calculated as for each plot as:  !!"! = !! -! (Eq. 9) Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 27 Scientific report by the University of British Columbia, Vancouver 2.5.3. Gross ecosystem photosynthesis and ecosystem respiration Net fluxes !!"!!measured in the collars at the four soil plots represent the effects of soil respiration Rs and photosynthesis P of ground level plants. In contrast, fluxes measured by EC at the flux tower quantify net ecosystem exchange (NEE) or net ecosystem productivity (NEP = - NEE) integrated over the turbulent source area including tall vegetation. From measurements of NEE by EC, similarly to (Eq. 8), an ecosystem-scale light response curve was constructed. Gross ecosystem photosynthesis (GEP) was determined by replacing P by GEP in (Eq. 8) using irradiance data continuously measured at the flux tower. For each time-step, GEP was modeled and together with measured NEE, ecosystem respiration Re was solved as the residual:  !! = NEE+ GEP (Eq. 10) 2.5.4. Calculating GWP and CO2e The effect of the different long-lived greenhouse gases was compared considering the GHG warming potential (GWP) for each gas by converting the molar fluxes of all long-lived GHGs measured to CO2 equivalent mass fluxes as follows:  CO!e!(s) = GWP!! !!!! (Eq. 11) where CO2e (s) is the equivalent mass flux of the species s (CO2, CH4 and N2O) expressed in g CO2e m-2 s-1. GWPs is the mass-based GWP (g g-1). In this report, we use 100-yr GWPs following Environment Canada's inventory practices2, which are based on IPCC’s fourth assessment report and are 1, 25, and 298 for CO2, CH4 and N2O, respectively. !! is the study-period averaged molar flux for a plot converted to mol m-2 day-1 for the species, and ms is its molar mass (in g mol-1). The total CO2e emissions from soils in a particular plot were then calculated as the sum of the three CO!e!(s) values for CO2, CH4 and N2O:                                                 2 https://www.ec.gc.ca/ges-ghg/default.asp?lang=En&n=CAD07259-1 (accessed Nov 29, 2014) Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 28 Scientific report by the University of British Columbia, Vancouver  CO!e = !!!"!!!!"! + 25! !"!!!"! + 298! !!!!!!! (Eq. 12) !!"! and !!!"were calculated as the average of all individual measurements at all collars, and then averaged over all collars in a plot. !!"! was calculated using (Eq. 9) based on the procedure described in Sections 2.4.1 and 2.5.2.  Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 29 Scientific report by the University of British Columbia, Vancouver 3. Results 3.1.  Climate conditions The study period covered 47 days from 26 June to 11 August 2014. This period was characterized by generally dry weather with high !↓ which was on average 23.24 MJ m-2 day-1 with 27 days experiencing clear sky conditions with none or very few high clouds. Total precipitation during this period was 22.2 mm and precipitation occurred on only 8 days (Figure 12, top). Values of θw were high (water table at the surface) in all the plots at the beginning of the period. Over the entire period, the highest values of θw were measured in BTS, followed by DS, PSLS and lowest at BS. θw decreased towards the end of the study period (Figure 12, top). Plot BS experienced the highest differentiation between drier microenvironments (hummocks) and wet microenvironments in local depressions. Most collars in the BTS plot remained water saturated over the entire study period, whereas many collars in DS and PSLS developed a shallow dry surface layer. Averaged soil temperatures, T at the 5 cm depth (average of two thermocouple measurements, see Section 2.3.4) ranged between 17.9 and 21.5 ºC depending on the plot (Table 4). Highest T was measured in the disturbed plot (DS), which was likely due to the low albedo of the dark exposed bare soil and the lack of shading vegetation (see Figure 3) Lowest average T was found in the wet and highly vegetated BS ecosystem with only 17.9ºC.   Table 4 - Average soil climate conditions from 26June, 2014 to 11 August , 2014. In each plot, soil temperature and soil volumetric water content was measured twice and values represent the average of the two measurements. Soil temperature was measured at the 5-cm depth, and soil volumetric water content for the 0 - 30 cm depth. Plot Soil temperature (ºC)  Soil volumetric water context (%) Mean Mean Daily Min Mean Daily Max Mean Absolute Min Absolute Max DS 21.5 17.0 26.1  75 62 87 BTS 20.2 16.2 23.4  78 76 85 BS 17.9 14.3 20.6  69 49 87 PSLS 19.0 16.0 22.2  72 64 82  Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 30 Scientific report by the University of British Columbia, Vancouver   Figure 12 - Precipitation and soil volumetric water content (top panel) and soil temperature (bottom panel) over the period of GHG flux measurements in summer 2014. The green and yellow dots indicate dates when manual soil GHG flux measurements were made to determine CO2, CH4 and/or N2O fluxes. Soil temperature and soil volumetric water content are ensemble averages of all four soil flux plots (two replications each at DS, BTS, BS and PSLS). For soil temperature, the shaded region shows the daily minimum and maximum of the ensemble soil temperature. The dotted line shows the maximum and minimum out of all eight measurements on each day. Precipitation is a composite series between the Environment Canada Climate Station “Delta Burns Bog”and the measurements from the flux tower after 9 July 2014, both located within the BBECA.  101520253035Soil Temperature (ºC)Precipitation (mm day-1)Soil Volumetric Water Content (%)0246810 100908070605040CO2 Sampling CH4 / N2O Samplingmaxmaxminmin10 1131302928272625243029282726252423 23222120191817161514131211 98765432110987654321August 2014July 2014June 2014Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 31 Scientific report by the University of British Columbia, Vancouver 3.2.  Carbon dioxide 3.2.1. Soil respiration The total of 309 soil respiration (Rs) measurements at all plots range between close to zero and 4.57 µmol m-2 s-1. Figure 13 shows the cumulative frequency distributions of all measured Rs values sorted by plot. The highest individual Rs measurements with an average of 1.46 (max: 4.57) µmol m-2 s-1 were found in the most recently disturbed ecosystem DS, which was also the plot with warmest soil temperatures over the study period (21.5 ºC, see Section 3.1). This was followed by PSLS with an average measured Rs of 1.33 (max: 4.125) µmol m-2 s-1 and BS with an average measured Rs of 1.06 (2.01) µmol m-2 s-1. The lowest average Rs of 0.68 (max: 2.10) µmol m-2 s-1 was found in BTS, which was the plot that experienced lowest soil temperatures (17.9 ºC) and highest water saturation (standing water). In terms of frequency distributions, it is evident from Figure 13 that BTS experienced consistently lower Rs compared to the three other plots in all classes, while the other three plots mainly differed in the higher value classes.  Figure 13 - Cumulative frequency distributions of all individual soil respiration (µmol m-2 s-1) measurements (all plots and all collars) between 23 July and 7 August 2014. Only data from opaque chambers are included. The number of efflux measurements considered in each plot is given in brackets.   −1 0 1 2 3 4 50.00.20.40.60.81. 0Normalized cumulative frequency 75%50%25% DS (n=67)BTS (n=72)BS (n=60)PSLS (n=88)Soil respiration (CO2 flux density in    mol m-2 s-1)Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 32 Scientific report by the University of British Columbia, Vancouver Table 5 - Summary statistics of measured soil respiration at the four plots. Plot Measured soil respiration (µmol m-2 s-1)  No of samples Mean 25th percentile 50th percentile 75th percentile DS 1.46 0.68 1.39 1.98  67 BTS 0.68 0.25 0.63 1.03  72 BS 1.06 0.80 1.15 1.34  60 PSLS 1.33 0.74 1.13 1.61  88    Figure 14 - Distribution of measured soil respiration. Shown are statistics of all soil carbon dioxide (CO2) flux measurements with an opaque chamber (all plots and all collars) between23 July and 7 August 7, 2014. Table 6 lists all Rs measurements separated by collar. Generally the highest Rs values were measured in collars with active grass or herb vegetation (e.g. DS: S-1, S-4, S-5, PSLS: S-1, S-7, S-8), and lowest values were measured in collars that are located in standing water (e.g. BTS: S-1, S-2) or bare soil (e.g. BTS: S-7, DS: S-2, S-3). Collars with Sphagnum generally exhibited lower Rs (see photos of all collars in Table 17).  67 72 60 88DS BTS BS PSLS0.51. 01. 52.02.5 1590%50%75%meanno of samples25%10%Soil respiration (CO 2 flux density in    mol m-2 s-1 )Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 33 Scientific report by the University of British Columbia, Vancouver Table 6 - Individual collar soil respiration (µmol m-2 s-1) measurements from all plots.  Plot Collar Measured soil respiration No of samples Mean  (µmol m-2 s-1) Std. dev.  (µmol m-2 s-1) Min  (µmol m-2 s-1) Max  (µmol m-2 s-1) DS S-1 2.52 0.63 1.61 3.53 10  S-2 0.78 0.31 0.37 1.49 12  S-3 0.85 0.36 0.45 1.68 10  S-4 2.08 0.50 1.54 2.98 8  S-5 1.65 0.45 1.03 2.31 10  S-6 3.20 1.15 1.46 4.57 9  S-7 0.65 0.25 0.38 1.23 10  S-8 1.48 0.63 0.20 2.39 10  All 1.65  0.20 4.57 79 BTS S-1 0.43 0.35 0.05 1.08 13  S-2 0.45 0.44 -0.06 1.39 13  S-3 1.75 0.61 0.98 2.80 10  S-4 0.94 0.24 0.67 1.28 9  S-5 1.23 0.32 0.70 1.71 9  S-6 0.55 0.48 -0.03 1.44 8  S-7 0.66 0.47 0.07 1.29 9  S-8 0.93 1.01 -0.06 2.62 8  All 0.87  -0.06 2.80 79 BS S-1 0.81 0.66 0.17 2.25 8  S-2 1.25 0.27 0.90 1.71 8  S-3 1.13 0.35 0.68 1.71 6  S-4 1.42 0.63 0.19 2.03 7  S-5 1.39 0.46 0.84 2.00 7  S-6 1.02 0.32 0.72 1.59 9  S-7 1.47 0.43 0.67 2.02 9 Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 34 Scientific report by the University of British Columbia, Vancouver Plot Collar Measured soil respiration No of samples Mean  (µmol m-2 s-1) Std. dev.  (µmol m-2 s-1) Min  (µmol m-2 s-1) Max  (µmol m-2 s-1)  S-8 1.31 0.52 0.75 2.36 9  All 1.22  0.17 2.36 63 PSLS S-1 1.60 0.92 0.75 3.70 10  S-2 1.17 0.38 0.66 2.02 10  S-3 0.91 0.46 0.40 1.78 11  S-4 0.68 0.75 0.06 2.61 11  S-5 1.24 0.59 0.26 2.32 11  S-6 1.44 0.59 0.76 2.60 11  S-7 3.39 1.34 1.69 6.51 12  S-8 2.18 0.91 0.96 3.80 12  All 1.58  0.06 6.51 88 All plots 1.33    309  Modelled !! values (extrapolated and averaged over the entire study period based on the procedure outlined in Section 2.5.1) for each plot are shown in Table 7. The highest extrapolated !! was found in the recently disturbed DS ecosystem, and the lowest !! was modelled for the most water saturated BTS ecosystem. Variability between collars in each plot was large. Table 7 - Summary statistics of modelled soil respiration in the four plots over the period 26 June to 11 August 2014 in µmol m-2 s-1 and in g CO2 m-2 day-1. Plot Modelled soil respiration  Mean of all collars (µmol m-2 s-1) Mean of all collars (g CO2 m-2 day-1) Mean of collar with lowest Rs in plot (g CO2 m-2 day-1) Mean of collar with highest Rs in plot (g CO2 m-2 day-1) DS 1.57 5.97 2.24 12.48 BTS 0.90 3.42 1.99 6.59 BS 1.03 3.90 2.93 4.68 PSLS 1.46 5.57 2.58 11.59 Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 35 Scientific report by the University of British Columbia, Vancouver 3.2.2. Photosynthesis Photosynthesis, P was modelled based on information from 64 individual light response curves (Figure 15) for each of which Pm was determined (see section 2.5.2). The highest P values were observed in the DS and PSLS plots, in particular in collars with grassy and herb vegetation (e.g. DS; S-1, S-4, S-5; BTS: S-3, S-4, PSLS: S-7, S-8). Collars with Sphagnum had generally low rates of photosynthesis (e.g. BS: S-3, S-7, PSLS: S-2, S-3) as did collars with bare soil (e.g. BTS S-8 with P=0).  Figure 15 - Light response curves measured in all four plots and all eight collars for two replications each. Table 8 - Summary statistics of modelled photosynthesis in the collars of the four plots over the period 26 June to 11 August, 2014 (24 hours) in µmol m-2 s-1 and in g CO2 m-2 day-1. Plot Modelled photosynthesis Mean of all collars (µmol m-2 s-1) Mean of all collars (g CO2 m-2 day-1) Mean of collar with lowest P in plot (g CO2 m-2 day-1) Mean of collar with highest P in plot  (g CO2 m-2 day-1) DS 0.75 2.85 0.57 5.48 BTS 0.36 1.37 0.00 3.16 BS 0.25 0.94 0.21 2.32 PSLS 0.45 1.70 0.70 4.11 DS−1DS−2DS−3DS−4DS−5DS−6DS−7DS−807/29 08/07BTS−1BTS−2BTS−3BTS−4BTS−5BTS−6BTS−7BTS−807/30 08/05PSLS−1PSLS−2PSLS−3PSLS−4PSLS−5PSLS−6PSLS−7PSLS−807/29 08/07BS−1BS−2BS−3BS−4BS−5BS−6BS−7BS−807/30 08/05Photosynthesis P (  mol m-2 s-1 )Photosynthesis P (  mol m-2 s-1 )PPFD (  mol m-2 s-1)0 500 1000 1500 2000−2−10123450 500 1000 1500 2000−2−10123450 500 1000 1500 2000−2−10123450 500 1000 1500 2000−2−1012345Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 36 Scientific report by the University of British Columbia, Vancouver P determined from light response curves in the collars represents only the low ground vegetation (low grasses, herbs, Sphagnum) that grows within the collars. However, all of the ecosystems probed contained tall vegetation (tall grasses, shrubs, trees) capable of photosynthesizing more effectively. Figure 16 shows EC measured ecosystem CO2 fluxes in the source area of the flux tower. For PPFD > 169 µmol m-2 s-1, the ecosystem becomes a sink for CO2 where P > Re. Based on a best fit of (Eq. 8), P at the flux tower was determined to be 2.27 µmol m-2 s-1 or 8.62 g CO2 m-2 day-1 (for the period 9 July 9 to 11 August 2014).  Figure 16 - Light response curve determined from all individual 30 minute eddy covariance measurements of NEE (=FCO2 measured by EC) at the flux tower for the period 9 July 2014 to 11 August 2014. The lowest class is at 0 µmol m-2 s-1 (night). The dotted curve is a best fit for (Eq. 8) with parameters Pm = 5.78 µmol m-2 s-1, φ = 0.0098 and C = 0.7 and offset by 1.50 µmol m-2 s-1 (respiration at PPFD = 0 µmol m-2 s-1). The u* threshold was 0.08 m s-1. Number of valid half-hour measurements averaged is indicated below each box. 3.2.3. Modelled and measured net ecosystem exchange Figure 17 shows net fluxes, !!"!!averaged over the study period in relation to plot averaged !! and ! following (Eq. 9). A positive !!"! indicates that soils in the ecosystem were a source (emission) of CO2, negative would mean there is CO2 uptake. Soils in all four plots were sources of CO2 during the study period. The highest !!"!!values (emissions) were found in PSLS (+3.88 g CO2 m-2 day-1), followed by the recently disturbed DS ecosystem (+3.12 g CO2 m-2 day-1) and then BS (+2.97 g CO2 m-2 day-1). The lowest !!"!!values!were found in the BTS ecosystem (+2.05 g CO2 m-2 day-1).  113 44 42 40 43 47 45 30  40  35 36 38 48  39 54  57 71  82 15 10 200 400 600 800 1000 1200 1400 1600 1800 2000PPFD (μmol m − 2 s − 1 )−6−4−202CO2 molar flux density (μmol m−2  s−1 )Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 37 Scientific report by the University of British Columbia, Vancouver  Figure 17 – Modelled net flux (!!"#!)!separated into Rs and P for the four plots. These values of !!"!!at the soil plots are in contrast to the EC measured net ecosystem exchange (NEE) of the flux tower source area, where average NEE was -0.87 µmol m-2 s-1 or -3.30 g CO2 m-2 day-1 over the study period, indicating that this system was a net carbon sink when considering tall vegetation in addition to soils (Figure 18). Ecosystem respiration, Re, for the flux tower was estimated, as Re = NEE + GEP, to be 1.40 µmol m-2 s-1 or 5.32 g CO2 m-2 day-1 (where GEP was modelled based on Section 2.5.3).  Figure 18 – Ensemble-averaged diurnal course of EC measured net ecosystem exchange (NEE) for the flux tower source area between 9 July 2014 and 9 August 2014. Number of valid half-hour measurements averaged is indicated below each box.+0.82 +0.54 +0.78 +1.02Rs± Standard deviationP (ground only)Net flux (ground only)DS BTS BS PSLS-10123CO2 Flux (μmol m−2  s−1 ) between collarsChristen et al. – Greenhouse gas fluxes from soils in Burns Bog 38 Scientific report by the University of British Columbia, Vancouver 3.3.  Methane A total of 60 soil CH4 efflux measurements were made on 17-18 July, 24-28 July and 11 August 2014 in all four plots with five replications each (five different collars) following the methodology outlined in Sections 2.3.2 and 2.4.2. Out of the 60 samples, 13 samples were rejected due to step increases in CH4 over time and poor fits. Those step changes could have been caused by disturbances while taking samples and/or by ebullitive emissions. The measured soil CH4 fluxes of the remaining samples ranged over five orders of magnitude with a highly skewed distribution. This justifies the use of cumulative frequency distributions (Figure 19) and median values (Figure 20) rather than the use of mean values to assess ecosystem differences. Highest median (50th percentile) effluxes were measured in BTS with 77 (Range: 18 - 15,000) nmol m-2 s-1, followed by BS with 43 (2 - 3000) nmol m-2 s-1, and PSLS with 41 (4 - 3500) nmol m-2 s-1. The lowest median soil CH4 fluxes were found in the most recently disturbed ecosystem (DS) with 19 (7 - 3000) µmol m-2 s-1.    Figure 19 - Cumulative frequency distributions of individual soil methane (CH4) flux measurements (all plots and all collars). Data from 17-18 and 24-28 July and 11 August 2014 are included. In order to visualize the large range of measured efflux values, the x-axis is shown on a logarithmic scale. Two points were negative and not shown.  100 101 102 103 104 1050.00.20.40.60.81.0DS (n=12)BTS (n=11)BS (n=13)PSLS (n=11)75%50%25%Normalized cumulative frequencySoil CH4 flux density (nmol m-2 s-1)Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 39 Scientific report by the University of British Columbia, Vancouver Table 9 - Summary statistics of soil CH4 flux measurements at the four plots. Plot Soil CH4 flux density (nmol m-2 s-1)  No of valid samples Mean 25th percentile 50th percentile 75th percentile DS 456 10 19 312  12 BTS 2564 56 77 2398  11 BS 270 21 43 70  13 PSLS 357 16 41 204  11 The highest value of ~15000 nmol m-2 s-1 was recorded in BTS in collar L-3 on 11 August 2014 (grass in standing water), followed by collar L-2 at BTS (standing water without above water vegetation). In all plots, spatial and temporal variability was extremely high (Table 12). The lowest fluxes were measured on 24-25 July 2014, immediately after a rain event, and the highest fluxes at the end of a 17-day dry period on 11 August 2014.   Figure 20 - Distribution of soil methane (CH4) flux measurements (all valid runs from all collars, n=47). Data from 17-18, 24-28 July and 11 August, 2014 are included. In order to visualize the large range of measured efflux values, the y-axis is shown on a logarithmic scale.    1590%50%75%meanno of samples25%10%DS BTS BS PSLS110100100010000CH4 Flux Density (nmol m-2 s-1 ) 12 11 13 11Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 40 Scientific report by the University of British Columbia, Vancouver Table 10 - Individual collar soil methane (CH4) flux measurements from all plots. Plot Collar 17/18 July 2014  24/25 July 2014  11 August, 2014  Average CH4 flux (nmol m-2 s-1) RMSE  (ppm) CH4 flux (nmol m-2 s-1) RMSE  (ppm) CH4 flux (nmol m-2 s-1) RMSE  (ppm)  CH4 flux (nmol m-2 s-1) DS L-1 11.0 0.01  12.1 (a) 0.24  [12.0] (b) [1.07]  12  L-2 3.1 (a) 0.18  [0.9] (b) [0.22]  3139.6 (a) 80.43  1571  L-3 25.9 0.08  27.2 0.13  254.2 1.51  102  L-4 9.8 0.01  9.0 (a) 0.22  [-3.6] (b) [0.20]  9  L-5 1497.1 17.16  2.3 0.03  486.6 (a) 9.57  662  All 309     13     1293     471 BTS L-1 7813.8 (a) 145.60  [440.1] (b) [48.30]  3185.1 20.81  5499  L-3 1610.7 4.31  76.5 0.17  15119.1 112.49  5602  L-4 [144.6] (b) [7.58]  24.3 0.12  [116.1] (b) [16.67]  24  L-5 70.4 (a) 0.95  42.0 0.11  71.7 0.53  61  L-6 168.7 1.04  [56.0] (b) [0.31]  17.8 0.05  93  All 2416     48     4598     2256 BS L-1 42.8 0.37  7.3 (a) 0.22  22.7 0.12  24  L-2 [484.3] (b) [58.93]  2965.8 64.45  [65.1] (b) [2.47]  2966  L-3 89.7 0.12  68.1 0.81  20.7 0.33  60  L-5 36.9 0.71  9.7 (a) 0.20  68.9 0.62  39  L-6 113.4 1.38  9.0 0.13  49.3 0.47  57  All 71     612     40     629 PSLS L-1 41.4 0.30  25.0 0.06  49.6 (a) 3.20  39  L-2 [447.7](b) [13.54]  8.0 0.08  4.3 (a) 0.36  6  L-3 115.0 5.91  11.2 0.19  [7.2] (b) [0.46]  63  L-5 [1.2] (b) [0.37]  [-2.7] (b) [0.38]  20.7 0.15  21  L-6 3670.9 47.40  660.7 (a) 18.05  293.7 0.68  1542  All 1276     176     92     334 All plots 1018     212     1506   923 Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 41 Scientific report by the University of British Columbia, Vancouver (a) Values that have a likelihood of minor disturbances during sampling. Four out of five samples are increasing over time, but one of the five samples might show slightly unexpected values (e.g. decrease). However the overall goodness of fit of all five samples is acceptable. Those values are included in calculations of averages. (b) Values in square brackets are from runs with unsuccessful fits, or cases with a high likelihood of disturbance by the sampling and not used in calculations of averages. 3.4. Nitrous oxide A total of 60 soil N2O flux measurements were made on 17-18 July, 24-28 July and 11 August, 2014 in all four plots with five replications each (five different collars) following the methodology outlined in Sections 2.3.2 and 2.4.2. N2O was measured simultaneously with CH4, so all samples match those reported in Section 3.3. The N2O fluxes in all plots were generally small and ranged between -0.178 (weak uptake) and +0.185 nmol m-2 s-1 (weak emission). Figure 21 shows the cumulative frequency distributions of all N2O flux measurements sorted by plot. The highest median (50th percentile) effluxes were measured in PSLS with +0.025 nmol m-2 s-1. BS showed a median uptake of -0.018 nmol m-2 s-1 with some significantly negative values. The other two plots (BTS, DS) showed median values close to zero (Table 11). Out of the 60 measurements, 27 measurements were negative (uptake).  Figure 21 - Cumulative frequency distributions of all individual nitrous oxide (N2O) fluxes (all plots and all collars). Data from 17-18, 24-28 July and 11, August 2014 are included. −0.2 −0.1 0 .0 0.1 0 .20.00.20.40.60.81. 0Normalized cumulative frequencyDS (n=15)BTS (n=15)BS (n=15)PSLS (n=15)75%50%25%Soil N2O flux density (nmol m-2 s-1)Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 42 Scientific report by the University of British Columbia, Vancouver Table 11 - Summary statistics of soil N2O flux measurements in the four plots. Plot  Soil N2O flux density (nmol m-2 s-1)  No of samples Mean 25th percentile 50th percentile 75th percentile DS 0.011 -0.019 -0.004 0.022  15 BTS 0.003 -0.049 0.000 0.071  15 BS -0.033 -0.118 -0.018 0.034  15 PSLS 0.022 -0.004 0.025 0.053  15    Figure 22 - Distribution of soil nitrous oxide (N2O) flux measurements (all plots and all collars). Data from 17-18, 24-28 July and 11 August 2014 are included.   15 15 15 15DS BTS BS PSLS−0.15−0.10−0.050.000.05N 2O Flux Density (nmol m-2 s-1 ) 1590%50%75%meanno of samples25%10%Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 43 Scientific report by the University of British Columbia, Vancouver Table 12 - Individual collar nitrous oxide (N2O) flux measurements from all plots.  Plot  Collar 17/18 July 2014  24/25 July 2014  11 August 2014  Average N2O flux (nmol m-2 s-1) RMSE  (ppb) N2O flux (nmol m-2 s-1) RMSE  (ppb)  N2O flux (nmol m-2 s-1) RMSE  (ppb)  N2O flux (nmol m-2 s-1) DS L-1 -0.019 1.940  0.057 2.020  -0.004 1.410  0.01   L-2 0.031 0.340  -0.040 5.920  -0.005 1.200  -0.005  L-3 0.185 4.310  0.015 2.520  -0.012 5.240  0.063  L-4 -0.011 5.480  0.022 2.580  -0.023 0.880  -0.004  L-5 -0.040 6.580  0.002 4.410  0.012 3.460  -0.009  All 0.029   0.011   -0.006   0.011 BTS L-1 -0.038 7.900  0.077 4.800  -0.049 2.450  -0.003  L-3 0.071 5.310  -0.082 6.090  -0.037 3.090  -0.016  L-4 0.154 6.340  -0.072 9.460  -0.034 1.330  0.016  L-5 0.023 3.870  -0.089 8.000  0.007 1.400  -0.020  L-6 0.040 2.670  0.078 4.480  0.000 3.400  0.039  All 0.050   -0.018   -0.023   0.003 BS L-1 [0.026] [13.27]  -0.178 7.620  0.025 2.410  -0.042  L-2 0.124 2.500  0.055 3.320  0.034 1.870  0.071  L-3 -0.038 6.040  -0.078 1.260  -0.150 3.270  -0.089  L-5 -0.108 2.010  -0.118 5.430  0.009 2.430  -0.072  L-6 -0.155 4.540  0.068 3.670  -0.018 0.770  -0.035  All -0.030   -0.050   -0.020   -0.033 PSLS L-1 0.024 3.560  0.089 6.150  -0.090 3.390  0.008  L-2 0.025 3.490  0.087 3.110  -0.009 2.710  0.034  L-3 -0.027 4.180  0.008 6.830  -0.004 0.410  -0.008  L-5 0.041 1.920  0.062 5.240  0.023 1.520  0.042  L-6 0.028 3.900  0.053 4.720  0.025 1.630  0.035  All 0.018   0.060   -0.011   0.022 All plots 0.017   0.001   -0.015   0.0007  Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 44 Scientific report by the University of British Columbia, Vancouver 3.5. Greenhouse warming potential of the different gases The combined greenhouse gas emissions from all three long-lived greenhouse gases (GHG) are shown as carbon dioxide equivalents (CO2e) following the calculations in Section 2.5.3 in Figure 23. Soils in all four plots exhibited strong GHG emissions dominated by CH4. The highest overall CO2e emissions originated from the wet BTS ecosystem (64.1 g CO2e m-2 day-1). All other plots exhibited between 1/8 and one third of the BTS emissions with 10.7 (DS), 18.5 (PSLS) and 24.9 (BS) g CO2e m-2 day-1. Overall, CH4 was responsible for 97% of all emissions in BTS, 71% in DS, 79% in PSLS and 88% in BS. The rest was attributable to CO2 emissions. The effect of N2O was negligible and ranged between –0.17% (BS) and +0.16% (PSLS) of the total CO2e emissions.  Figure 23 - Calculated total GWP in g CO2e m-2 day-1 by ecosystem plot.   CO2CH4N2O+64.14+10.68+24.88+18.48DS BTS BS PSLSCO2e Flux (g CO 2e m−2  day−1 )020406080Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 45 Scientific report by the University of British Columbia, Vancouver 4. Discussion and Conclusions 4.1. Assessment of GHG fluxes Methane (CH4) is the most relevant long-lived GHG emitted by soils in the rewetted sector of the BBECA. Emissions of CH4 are substantial in all plots. The presented soil flux measurements of CH4 from the four plots provide clear evidence that the rewetted sector of the BBECA - in its current phase of restoration - is a source of CH4. Overall, the 50th percentile flux of all measurements and plots was ~40 mg CH4 m-2 day-1 (~15 – 210 mg CH4 m-2 day-1, 25 – 75th percentiles). Fluxes in the BTS ecosystem were consistently higher than other three plots with extreme values up to 20 g CH4 m-2 day-1 (average ~5 g CH4 m-2 day-1). The BTS ecosystem experienced a high water table (at the surface) all summer and visible ebullition (formation of bubbles) of gases (Figure 24). It is known that CH4 fluxes from wetlands are highest in microenvironments where the water table is at the surface and a ready supply of fresh biomass is available. The average fluxes at the BTS ecosystem, where both conditions were present, experienced averaged fluxed by a factor 5 higher than the averages of the other three ecosystems studied.    Figure 24 – Close-up photograph of possible ebullition, bubbles of gases forming at the water surface at the BTS plot (Photographed on June 13, 2014). Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 46 Scientific report by the University of British Columbia, Vancouver The large variability in CH4 fluxes over several orders of magnitude between the five collars at each plot and between measurement dates is a challenge for proper quantification of fluxes from only 47 runs that were considered in this study. The averages presented in this work are associated with large errors, and cumulative frequency distributions are better suited to study differences between plots. The large spatial variability in CH4 emissions could be due to high spatial variability in microbial communities as observed in earlier studies (e.g. Hendriks et al., 2010; Schrier-Uijl et al., 2010; Teh et al., 2011; Baldocchi et al., 2012). In cases where anaerobic conditions are present, it is known that vegetation type also controls the rate of CH4 emissions. Firstly, the availability of new living plant material in the micro-environment is important (Couwenberg, 2009). Although DS also experienced a high water table like BTS, the lack of vegetation at DS failed to provide fresh biomass thereby possibly limiting CH4 production. BS and PSLS showed drying over the summer and had a more variable microtopography where aerobic (hummocks) and anaerobic conditions existed (depressional zones). Secondly, CH4 emissions vary greatly due to differences in the effectiveness of plant species in transporting methane from the water-saturated zone through their aerenchyma (gas conductive plant tissues) into the atmosphere (Couwenberg, 2009). This could be a plausible explanation why most extreme CH4 fluxes are measured in collars that contained both anaerobic conditions and vegetation (beak rush), e.g., collars BTS L-3 and L-5 as opposed to collars with anaerobic conditions but no vegetation (DS: L-5, BTS: L-4 and L-6). However, substantial CH4 emissions were also found from bare, water saturated collars (DS L-2 and BTS L1). Several collars exhibited step changes in the chamber CH4. Those samples have been removed, due to possible disturbance by the experimenters but the same pattern could occur due to ebullitive delivery when intermittent bubbles reach the water surface. Table 13 - Comparison of average CH4 fluxes (and median values in square brackets) measured in this study to averages reported in the literature on fluxes from wetlands in temperate / boreal North America. Column “period” refers to the period of the measurements, where “YR” means year-round, “SA” are seasonal (summer season) averages, and “SP” are sporadic measurements limited in time.    Period Average [and median] of CH4 fluxes Study Location Ecosystem (nmol m-2 s-1) (mg CH4 m-2 day-1) This study (DS) BC, Canada Ombrotr. bog SP 471 [43]  652 [59] This study (BTS) BC, Canada Ombrotr. bog SP 2256 [77] 3126 [106] This study (BS) BC, Canada Ombrotr. bog SP 629 [41] 872 [57] This study (PSLS) BC, Canada Ombrotr. bog SP 334 [19] 463 [26] Nadeau et al. (2013) QC, Canada Ombrotr. bog SN 41.6 57.6 Moore and Knowles (1990) QC, Canada Ombrotr. bog YR 0.2 0.3 Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 47 Scientific report by the University of British Columbia, Vancouver    Period Average [and median] of CH4 fluxes Study Location Ecosystem (nmol m-2 s-1) (mg CH4 m-2 day-1) Roulet et al. (1992) QC, Canada Ombrotr. bog YR 1.9 - 4.2 2.6- 5.8 Shannon and White (1994) MI, USA Ombrotr. bog YR 3.1 - 49.4 4.3 - 68.4 Moore et al. (1994) ON, Canada Ombrotr. bog SP 5.8 8.1 Chasar et al. (2000) MN, USA Ombrotr. bog SP 20.2 - 173.9 28 - 241 Dise (1993) MN, USA Ombrotr. bog YR 85.1 118.0 Moore and Knowles (1990) QC, Canada Poor fen YR 19.4 26.9 Roulet et al. (1992) ON, Canada Poor fen YR 2.2 3.0 Moore et al. (1994) ON, Canada Poor fen SP 8.9 12.3 Dise (1993) MN, USA Poor fen YR 130 180 Treat et al. (2007) NH, USA Poor fen YR 147 - 305 204 - 423 Moore and Knowles (1990) ON, Canada Rich fen YR 5.9 8.2 Chasar et al. (2000) MN, USA Rich fen SP 120 – 206 166 – 286 Knox et al. (2014) CA, USA Restored wetland YR 926 1283 Knox et al. (2014) CA, USA Restored wetland wetland YR 1270 1760 Table 13 compares CH4 fluxes recorded in various wetlands in temperate/boreal North America to the four plots in the current study. The average CH4 fluxes reported in this study are substantially higher than to-date reported CH4 fluxes in pristine Northern peatlands but comparable to the fluxes reported from two restored coastal wetlands in the Sacramento-San Joaquin Delta (Knox et al., 2014). There are several reasons for the high fluxes in the BCECA. Firstly, the pilot study area in BBECA has been recently rewetted similar to the study sites in Knox et al. (2014). In the first phase after rewetting (5 – 20 years), CH4 emissions are known to be high with rewetted bogs becoming a major GHG source - a ‘hot spot’ (Joosten, 2013). Secondly, measurements were made during the warmest time of the year (and day), when microbial activity is under most favorable conditions, while other studies in Table 13 represent either seasonal or year-round averaged fluxes. Currently there is no information available on CH4 fluxes in the cold season in the BBECA.  The GHG exchange of carbon dioxide (CO2) due to photosynthesis and respiration was of second order importance compared to CH4. Overall, soil and ground vegetation were a net source of CO2 in all plots where respiration dominates over photosynthesis. However, the ecosystem-wide fluxes determined at the flux tower showed that the ecosystem overall is a moderate carbon sink at this Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 48 Scientific report by the University of British Columbia, Vancouver time of the year. The EC flux measurements include the effect of tall grasses and shrubs that are not included in the collar measurements and more reliably to determine the net fluxes. BTS with its water saturated soil had consistently lower respiration than the other three plots due to limited oxygen availability. Table 14 compares measured CO2 fluxes at the flux tower in the BBECA to CO2 fluxes measured over other land cover in the region at the same time of the year. An unmanaged grassland site 15 km to the West of Burns Bog in the Fraser River Delta (Westham Island, Delta, BC) showed about twice the net uptake (NEE) of carbon at this time of the year. Ecosystem respiration (Re) and photosynthesis (P) were higher by a factor of 3 and 2.6, respectively, at the grassland site compared to Burns Bog. A mature 55 year-old forest on Vancouver Island (200 km NW) showed a typical NEE in July (2002 - 2006 average) that is close to the one in Burns Bog, but actual rates of Re and P are higher by a factor of 7 and 5, respectively, than Burns Bog. And in comparison to a young forest plantation (Buckley Bay, 150 km W), Burns Bog sequesters almost twice the carbon, even though Re is 4.5 and P is 3 times larger in the young and growing forest. This comparison highlights that the studied site in the BBECA is not an ecosystem of high productivity (P is smallest at all sites listed in Table 14), but one with considerably limited respiration that makes the sequestration of assimilated carbon more efficient. Table 14 – Comparison of summertime eddy covariance-measured ecosystem CO2 net fluxes (NEE) , and ecosystem respiration (Re) and ecosystem photosynthesis (P), over different vegetated land cover in the Vancouver region. Site Land use / cover Period Daily CO2 fluxes (g CO2 m-2 day-1) FNEE Re P Burns Bog,  Delta, BC (1) Raised bog July 9 – Aug 11, 2014 -3.30 5.32 8.62 Westham Island,  Delta, BC (2) Unmanaged grassland July 1 – July 31, 2009 -6.46 16.64 23.10 Campbell River (DF49), Vancouver Island (3) Douglas-fir forest (~55 yrs) July average 2002-2006 -3.54 38.94 42.48 Buckley Bay (HDF88), Vancouver Island (3) Douglas-fir forest (~15 yrs) July average 2002-2006 -1.77 24.19 25.96 (1) Measured at the Burns Bog EC flux tower (this study, see Section 2.3.3). (2) Measured during the CFCAS EPiCC project, unpublished data (http://ibis.geog.ubc.ca/~achristn/infrastructure/westhamisland.html) (3) EC Flux measurements reported in Krishnan et al (2009) Fig. 7 (before fertilization in 2007). Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 49 Scientific report by the University of British Columbia, Vancouver The comparison of strictly mid-summer data in Table 14 is limited and an extrapolation to annual dynamics and annual total values is not possible at this time. Also forest ecosystems on Vancouver Island are severely limited in water at this time of the year while being productive on a year-round basis. Full-year measurements in the BBECA are needed to assess and compare land cover effects in the region on an annual basis (for the other sites in Table 14 year-round EC data are available).   Table 15 – Comparison of summertime ecosystem CO2 net fluxes (NEE = - NEP) measured in various wetlands in temperate climates of North America, Europe and Asia to values determined in this study. Entries are sorted by NEE. Study Location Ecosystem Study period NEE  (g CO2 m-2 day-1) Ta (ºC) Coursolle et al. (2006) AB, Canada Boreal bog Aug -33.3 23.6 Zhou et al. (2009) Panjin, China Tidal wetland Jul 18 -26.7 24.5 Aurela et al. (2001) Finland Mesotrophic flark fen Jul -21.6 13.8 Shurpali et al. (1995) MN, USA Peatland May – Oct -19.9 14.9 Coursolle et al. (2006) ON, Canada Peatland Aug -14.3 26.3 Lafleur et al. (2003) ON, Canada Ombrotrophic bog Sep 15 -11.0 22.3 Knox et al. (2014) CA, USA Restored wetland Jul -10 18 Knox et al. (2014) CA, USA Restored wetland Aug -8 21 Neumann et al. (1994) ON, Canada Water-stressed bog Jun 25 –Jul 28 -6.7 - Coursolle et al. (2006) SK, Canada Boreal fen Aug -5.5 26.5 This study (2014) BC, Canada Ombrotrophic bog Jul 9 – Aug 11 -3.3 19.1 Sottocornola et al. (2005) Ireland Atlantic bog Jul -3.3 13.6 Coyne and Kelle (1975) AK, USA Wet meadow Jul -2.0 - Table 15 compares the NEE measured at the flux tower between July 9 and August 11 to other NEE measurements over various wetlands (most of them undisturbed) reported in the literature. All sites in Table 15 are carbon sinks (negative NEE). Burns Bog is amongst the weakest sinks compared to many similar ecosystems under comparable climatic conditions. This reflects the effects of the disturbance in the BBECA, and the location of the flux tower in an open area without significant shrubs and trees. Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 50 Scientific report by the University of British Columbia, Vancouver No significant emissions or uptake of N2O was found. N2O is almost certainly not relevant year round because the nutrient poor bog has no significant sources of nitrogen. If there are no significant fluxes recorded in summer then it is very likely that there will be no emissions in the cold season. 4.2. Limitations and future research The following recommendations can guide further work to (i) quantify more precisely the current GHG fluxes in the rewetted sector of the BBECA, (ii) identify GHG emission mitigation strategies in the management of the bog, and (iii) inform effective carbon sequestration strategies for the bog throughout the long-term restoration process:  • To capture the spatially highly variable fluxes of CH4 calls for spatially integrated, ecosystem-level flux measurements. The eddy-covariance (EC) approach is a reliable and appropriate method to retrieve spatially integrated annual total CH4 emissions over a larger area. Hence EC measurements of CH4 would be the first priority to better capture fluxes for quantifying and reporting emissions (Objective I). Together with continuous soil climate and water table monitoring, ideally year-round EC measurements would allow the identification of dominant climatic and hydrologic controls on CH4 fluxes in the specific context of the rewetted areas of BBECA and aid to make more detailed recommendations for mitigating emissions (Objective II). The existing flux tower (Section 2.2.2) could be an excellent platform to install additional EC instrumentation to measure CH4 fluxes if technical limitations (power) can be solved and its operation can be extended into 2015. Currently the planned operation period of the flux tower ends December 31, 2014.  • The summertime chamber measurements likely provide an upper limit of the flux magnitudes for both CO2 and CH4, and it is not possible to derive sound annual averages. To serve the need for annual totals in emissions reporting (Objective I) a year-round monitoring system (such as flux tower) is required. In addition to EC measurements, it is proposed to repeat CH4 chamber measurements at the same collars in the cold season (including cases with snow). CO2 fluxes have been already monitored into fall 2014 as part of a student project. It is expected that including these lower fluxes will lead to more realistic long-term estimates. Ideally there will be three more sampling dates: in January, February, and March 2015. Because the magnitude of the CH4 fluxes exceeded initial expectations, there is no need to make measurements over 30 minutes in future syringe sampling measurements (both summer and winter). Sampling over a period of 3 to 10 minutes is sufficient. This would ensure greater linearity and more samples would fall within the calibrated range of the GC.  Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 51 Scientific report by the University of British Columbia, Vancouver • Eddy covariance measurements of CO2 proved highly useful in this study and are in many regards superior over the sporadic and highly localized chamber measurements due to their spatially integrating signal that includes the response from low and tall vegetation. To assess the carbon sequestration potential of the BBECA, year-round measurements of CO2 fluxes by EC are needed. It is proposed to run the flux tower with the current EC system until at least July 8, 2015 to have a dataset that covers a full year and consequently carbon sequestration rates in g C m-2 year-1 can be compared between the BBECA, other ecosystems in Metro Vancouver and elsewhere.  • To more precisely quantify the role of photosynthesis across the restoration sequence of the near-surface plants, larger transparent chambers could be used. Measurements of photosynthesis by light-response curves in the small 9 cm collars was challenging and could have been impacted by shading from the collar rims. Also the small chamber head did not allowed the inclusion of larger grasses, sedges and bushes in the measurement volume. A design of larger (maybe automated) chambers could ensure more consistent measurements and partitioning of R and P.  • There is no immediate need to further monitor N2O fluxes. • It is proposed to archive all data measured at the flux tower in publicly accessible form from measured at the flux tower in standard format (e.g. through the global Fluxnet platform, and with Metro Vancouver), so potential future studies in the long-term can fully compare and exploit those first data as part of a long-term restoration sequence. 4.3. Management implications These first measurements of GHG emissions in the rewetted sector of the BBECA, Metro Vancouver, confirm high CH4 emissions, while only a weak carbon uptake is observed. However, it is expected, based on other studies, that after a period of 10-20 years, CH4 emissions will stabilize at a lower level and C uptake will increase. The exact duration of this first period is uncertain. Seasonally managing the water table could be an effective strategy for mitigating CH4 emissions in this initial 5-20 year rewetting period, but such a strategy would require assessment on vegetation and wildlife. Previous work showed that CH4 emissions clearly correlate with water table in the top 20 cm. With a water table below 20 cm of the surface, CH4 emissions are decreasing to negligible small values (Moore and Roulet, 1993) With ample oxygen in a narrow zone just above the water table, methanotrophic bacteria can oxidize CH4 produced in the saturated zone. If no oxygen is available above the saturated zone, Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 52 Scientific report by the University of British Columbia, Vancouver CH4 emissions increase rapidly when the water level is less than 20 cm below the surface (Couwenberg, 2009).  Acknowledgements This study was funded by Metro Vancouver through a research agreement with the University of British Columbia (PI: A. Christen / A. Black, 15 June – 31 Aug, 2014). We thank Dr. Conor Reynolds and all the staff involved at Metro Vancouver and the BBECA for providing the opportunity to pursue this research and his enthusiastic support. We appreciate the support of laboratory and administrative staff at UBC. References Aurela M, Laurila T, Tuovinen JP (2001) ‘Seasonal CO2 balances of a subarctic mire’. Journal Geophysical Research, 106:1623–1637. Baldocchi DD, Detto M, Sonnentag O, Verfaillie J, The JA, Silver WL, Kelly NM (2012) ‘The challenges of measuring methane fluxes and concentrations over a peatland pasture’, Agriculture Forest Meteoroogy., 153, 177–187, doi:10.1016/j.agrformet.2011.04.013. BC Ministry of Environment (2013) ‘2013 B.C. Best Practices Methodology for Quantifying Greenhouse Gas Emissions - Including guidance for public sector organizations, local governments and community emissions’ 44pp. http://www.env.gov.bc.ca/cas/mitigation/pdfs/ BC-Best-Practices-Methodology-for-Quantifying-Greenhouse-Gas-Emissions.pdf (Accessed Nov 25, 2014) Bleuten, W, Borren W, Glaser PH, Tsuchihara T, Lapshina ED, Mäkilä M, Siegel D, Joosten H, Wassen MJ (2006) ‘Hydrological Processes, Nutrient Flows and Patterns of Fens and Bogs’, in Caldwell MM, Heldmaier G, Jackson RB [Eds.] (2006) ‘Wetlands and Natural Resource Management’, Springer ISBN 9783642069727 Biggs W (1976). ‘An Ecological And Land Use Study Of Burns Bog, Delta, B.C.’ MSc Thesis in Plant Science, University of British Columbia, Vancouver. Catherine Berris Associates Inc. (1993) 'Burns Bog Analysis'. British Columbia, Ministry of Environment, Lands, and Parks. Chasar LS, Chanton JP, Glaser PH and Siegel DI (2000) ‘Methane concentration and stable isotope distribution as evidence of rhizospheric processes: comparison of a fen and bog in the glacial lake Agassiz peatland complex’. Annals of Botany, 86(3): 655–663. Coursolle C, Margolis HA, Barr AG, Black TA, Amiro BD, McCaughey JH, Flanagan LB, Lafleur PM, Roulet NT , Bourque CPA, Arain MA, Wofsy SC, Dunn A, Morgenstern K, Orchansky AL, Bernier PY, Chen JM, Kidston J, Saigusa N, Hedstrom N (2006) ‘Late-summer carbon fluxes from Canadian forests and peatlands along an east–west continental transect’. Canadian Journal of Forest Research, 36: 783–800. Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 53 Scientific report by the University of British Columbia, Vancouver Couwenberg J (2009) ‘Methane emissions from peat soils (organic soils, histosols) - Facts, MRV-ability, emission factors’, Technical report produced for UN-FCCC meetings in Bonn, August 2009. Greifswald University / Wetlands International. 16p. Coyne PI, Kelley JJ (1975) ‘CO2 exchange in the Alaskan tundra: meteorological assessment by the aerodynamic method’. Journal of Applied Ecology, 12: 587-611. Crawford B, Christen A (2014) ‘Spatial source attribution of measured urban eddy covariance carbon dioxide fluxes’. Theoretical and Applied Climatology, DOI 10.1007/s00704-014-1124-0. Crawford B, Christen A, Ketler R (2009) ‘Processing and quality control procedures’. EPiCC Technical Report No. 1, 11pp. http://hdl.handle.net/2429/45079 Dise NB (1993) ‘Methane emission from Minnesota peatlands: Spatial and seasonal variability’. Global Biogeochemical Cycles, 7: 123–142. Hebda RJ, Gustavson K, Golinski K, Calder AM (2000) ‘Burns Bog Ecosystem Review Synthesis for Burns Bog, Fraser River Delta, South-western British Columbia, Canada’. Environmental Assessment Office, Victoria, B.C. Hendriks DMD, Van Huissteden J, Dolman AJ (2010) 'Multi- technique assessment of spatial and temporal variability of methane fluxes in a peat meadow'. Agr Forest Meteor, 150:757–774. Hum, A (2006). ‘Photosynthetic response of Pseudotsuga menziesii seedlings to variations in light conditions in a variable retention stand’. Unpublished BScF thesis, Faculty of Forestry, UBC. Marini, RP, Sowers, DL (1990). ‘Net photosynthesis, specific leaf weight, and flowering of peach as influenced by shade’. Hortscience, 25(3): 331-334. Jassal RS, Black TA, Cai T, Morgenstern K, Li Z, Gaumont-Guay D, Nesic Z (2007) ‘Components of ecosystem respiration and estimates of net primary productivity of an intermediate-aged Douglas-fir stand’. Agricultural and Forest Meteorology, 144: 44-57. Jassal RS, Black TA, Novak MD, Morgenstern K, Nesic Z, Gaumont-Guay D (2005) ‘Relationship between soil CO2 concentrations and forest-floor CO2 effluxes’. Agricultural and Forest Meteorology, 130: 176-192. Jassal RS, Black TA, Roy R, Ethier G (2011) ‘Effect of nitrogen fertilization on soil CH4 and N2O fluxes, and soil and bole respiration’. Geoderma 162: 182-186. Joosten H (2013): ‘For peat’s sake – bogs and climate change’. Recorded presentation at the Environmental Protection Agency of Ireland, accessed Nov-18-2014 at http://www.epa-pictaural.com/s/climateChange/Joosten1.php?key=v110101 Knox SH, Sturtevant C, Matthes JH, Koteen L, Verfaillie J, Baldocchi D (2014). ‘Agricultural peatland restoration: effects of land-use change on greenhouse gas (CO2 and CH4) fluxes in the Sacramento-San Joaquin Delta’. Global Change Biology, doi: 10.1111/gcb.12745. Krishnan P, Black TA, Jassal RS, Chen B, Nesic Z (2009). Interannual variability of the carbon balance of three different-aged Douglas-fir stands in the Pacific Northwest. J. Geophys. Res. 114, G04011, doi:10.1029/2008JG000912. Lafleur PM, Roulet NT, Bubier JL, Frolking S, Moore TR (2003) ‘Interannual variability in the peatland-atmosphere carbon dioxide exchange at an ombrotrophic bog’. Global Biogeochemical Cycles, 17:1036. Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 54 Scientific report by the University of British Columbia, Vancouver Lloyd J, Taylor JA (1994) ‘On the temperature dependence of soil respiration’. Functional Ecology, 8: 315-323.  Luo Y, Zhou X (2006) ‘Soil respiration and the environment’. Elsevier, Inc., San Diego. Madrone Consultants Ltd. (1999). ‘Burns Bog Ecosystem Review. Plants and Plant Communities’. Report prepared for Delta Fraser Properties Partnership and the Environmental Assessment Office in support of the Burns Bog Ecosystem Review, with additional work on publicly owned lands conducted for the Environmental Assessment Office in association with the Corporation of Delta. Marini RP, Sowers DL (1990). ‘Net photosynthesis, specific leaf weight, and flowering of peach as influenced by shade’. Hortscience, 25(3): 331-334. Matthes JH, Sturtevant C, Verfaillie J, Knox S, Baldocchi D (2014). ‘Parsing the variability in CH4 flux at a spatially heterogeneous wetland: Integrating multiple eddy covariance towers with high-resolution flux footprint analysis’. Journal of Geophysical Research, 119: 1322-1339. McDade, GJ (2000) ‘Burns Bog Ecosystem Review’. Report of the special advisor to the Minister of the Environment, British Columbia’ Metro Vancouver (2007) ‘Burns Bog Ecological Conservancy Area Management Plan July 2007’, 46 p. Metro Vancouver (2011) ‘Metro Vancouver Integrated Air Quality and Greenhouse Gas Management Plan’, 41 pp. http://public.metrovancouver.org/about/publications/Publications/ IntegratedAirQualityGreenhouseGasManagementPlan-October2011.pdf Moore TR and R Knowles (1990) ‘Methane emissions from fen, bog and swamp peatlands in Quebec’. Biogeochemistry, 11(1): 45–61. Moore TR and Roulet NT (1993) ‘Methane flux: Water table relations in northern wetlands’. Geophysical Research Letters, 20, 587–590. Moore TR, Heyes A, Roulet NT (1994) ‘Methane emissions from wetlands, southern Hudson Bay lowland’. Journal Geophysical Research, 99: 455–467. Nadeau DF, Rousseau AN, Coursolle C, Margolis HA, Parlange MB (2013) ‘Summer methane fluxes from a boreal bog in northern Quebec, Canada, using eddy covariance measurements’. Atmospheric Environment, 81: 464-474. Neumann, HH, Hartog GD, King KM, Chipanshi AC (1994) ‘Carbon dioxide flux over a raised open bog at the Kinosheo Lake tower site during the Northern Wetlands Study (NOWES)’. Journal Geophysical Research, 99(D1): 1529–1538. Ögren E, Evans JR (1993) ‘Photosynthetic light-response curves, 1. The influence of CO2 partial pressure and leaf inversion’. Planta 189, 182e190. Resources Inventory Committee (1998). ‘Standard for Terrestrial Ecosystem mapping in British Columbia’. Ecosystems Working Group, Terrestrial Ecosystems Task Force, Resources Inventory Committee, Victoria, BC. Roulet NT, Ash R, Moore TR (1992) ‘Low boreal wetlands as a source of atmospheric methane’. Journal Geophysical Research, 97: 739–749. Schrier-Uijl, AP, Kroon PS, Hensen A, Leffelaar PA, Berendse F, Veenendaal EM (2010), ‘Comparison of chamber and eddy covariance-based CO2 and CH4 emission estimates in a Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 55 Scientific report by the University of British Columbia, Vancouver heterogeneous grass ecosystem on peat’, Agriculture Forest Meteorology, 150(6), 825–831, doi:10.1016/j.agrformet.2009.11.007. Shannon RD and White JR (1994) ‘A three-year study of controls on methane emissions from two Michigan peatlands’. Biogeochemistry, 27(1): 35–60. Shurpali, NJ, Verma SB, Kim J (1995) ‘Carbon dioxide exchange in a peatland ecosystem’. Journal of Geophysical Research, 100(D7): 14319-14326. Sottocornola M, Kiely G (2005) ‘An Atlantic blanket bog is a modest CO2 sink’. Geophysical Research Letters, 32:L23804. Teh, YA, Silver WL, Sonnentag O, Detto M, Kelly M, Baldocchi DD (2011) ‘Large Greenhouse Gas Emissions from a Temperate Peatland Pasture’, Ecosystems, 14(2), 311–325, doi:10.1007/s10021-011-9411-4. Treat CC, Bubier JL, Varner RK, Cril PM (2007) ‘Timescale dependence of environmental and plant mediated controls on CH4 flux in a temperate fen’. Journal of Geophysical Research, 112: G01014. Turunen J (2008) ‘Development of Finish peatland area and carbon storage’ Boreal Environmental Research, 13, 310-334. Zhou L, Zhou G Jia Q (2009) ‘Annual cycle of CO2 exchange over a reed (Phragmites australis) wetland in Northeast China’. Aquatic Botany, 91: 91-98.   Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 56 Scientific report by the University of British Columbia, Vancouver Appendix - Documentation of collars Large collars to measure methane and nitrous oxide fluxes A total of 24 PVC collars with a diameter of 21 cm were installed in the four plots, and 20 collars were used to measure CH4 and N2O exchange. The inset graphs in the column “Rank of fluxes”show the relative rank of the averaged soil fluxes of CH4 and N2O for this collar. The ranking is shown relative to all 20 collars measured, and values to the right denote average CH4 and N2O soil fluxes in nmol m-2 s-1. Table 16 - Documentation of all large (L-) PVC collars used for CH4 and N2O monitoring Plot Collar  Height (cm) Rank of fluxes  (nmol m-2 s-1) Photo June 20, 2014 Photo August 11, 2014 DS L-1 12.4    DS L-2 9.4    DS L-3 12.6    DS L-4 12.1    Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 57 Scientific report by the University of British Columbia, Vancouver Plot Collar  Height (cm) Rank of fluxes  (nmol m-2 s-1) Photo June 20, 2014 Photo August 11, 2014 DS L-5 8.2    BTS L-1 9.1    BTS L-3 17.1    BTS L-4 11.7    BTS L-5 14.6    BTS L-6 13.7    Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 58 Scientific report by the University of British Columbia, Vancouver Plot Collar  Height (cm) Rank of fluxes  (nmol m-2 s-1) Photo June 20, 2014 Photo August 11, 2014 BS L-1 16.9    BS L-2 13.8    BS L-3 13.3    BS L-5 13.9    BS L-6 14.7    PSLS L-1 9.9    Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 59 Scientific report by the University of British Columbia, Vancouver Plot Collar  Height (cm) Rank of fluxes  (nmol m-2 s-1) Photo June 20, 2014 Photo August 11, 2014 PSLS L-2 12.3    PSLS L-3 12.6    PSLS L-5 9.3    PSLS L-6 12.9       Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 60 Scientific report by the University of British Columbia, Vancouver Small collars to measure carbon dioxide fluxes A total of 32 PVC collars with a diameter of 10 cm were installed in the four plots to measure CO2 exchange. The inset graphs in the column “Rank of fluxes”show the relative rank of the averaged soil respiration Re and gross primary productivity GPP for situations with a photosynthetically active irradiance of > 1000 µmol m-2 s-1. The ranking is shown relative to all 32 collars measured, and values to the right denote average Re and GPP in µmol m-2 s-1 respectively. Data for this ranking is taken from the two sets of light response curves.  Table 17 - Documentation of all small (S-) PVC collars used for CO2 monitoring Plot Collar Height (cm) Rank of fluxes  (µmol m-2 s-1) Photo June 20, 2014 Photo August 11, 2014 DS S-1 4.0    DS S-2 4.0    DS S-3 4.0    DS S-4 4.0    Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 61 Scientific report by the University of British Columbia, Vancouver Plot Collar Height (cm) Rank of fluxes  (µmol m-2 s-1) Photo June 20, 2014 Photo August 11, 2014 DS S-5 4.0    DS S-6 4.0    DS S-7 4.0    DS S-8 4.0    BTS S-1 4.0    BTS S-2 4.0    BTS S-3 5.0    Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 62 Scientific report by the University of British Columbia, Vancouver Plot Collar Height (cm) Rank of fluxes  (µmol m-2 s-1) Photo June 20, 2014 Photo August 11, 2014 BTS S-4 4.0    BTS S-5 4.0    BTS S-6 4.0    BTS S-7 4.0    BTS S-8 4.0    BS S-1 4.0    BS S-2 4.0    Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 63 Scientific report by the University of British Columbia, Vancouver Plot Collar Height (cm) Rank of fluxes  (µmol m-2 s-1) Photo June 20, 2014 Photo August 11, 2014 BS S-3 4.0    BS S-4 4.0    BS S-5 4.0    BS S-6 4.0    BS S-7 4.0    BS S-8 4.0    PSLS S-1 4.0    Christen et al. – Greenhouse gas fluxes from soils in Burns Bog 64 Scientific report by the University of British Columbia, Vancouver Plot Collar Height (cm) Rank of fluxes  (µmol m-2 s-1) Photo June 20, 2014 Photo August 11, 2014 PSLS S-2 4.0    PSLS S-3 4.0    PSLS S-4 4.0    PSLS S-5 4.0    PSLS S-6 4.0    PSLS S-7 4.0    PSLS S-8 4.0    

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.52383.1-0103598/manifest

Comment

Related Items