Carbon Cycling Linkages of Permafrost Systems [CYCLOPS]
Principal Investigator: Williams, Mathew (3)
Licence Number: 15133
Organization: University of Edinburgh
Licenced Year(s): 2014 2013 2012
Issued: Jul 30, 2012
Project Team: Iain Hartley (Investigator, University of Exeter), Gareth Phoenix (Investigator, University of Sheffield), Julian Murton (Investigator, University of Sussex), Richard Essery (Investigator, University of Edinburgh)

Objective(s): To develop and evaluate a detailed process-based model of vegetation-soil-permafrost interactions using data collected through directed field campaigns in the discontinuous and sporadic permafrost zones of western Canada.

Project Description: The ultimate aim of the proposed project is to develop and evaluate a detailed process-based model of vegetation-soil-permafrost interactions using data collected through directed field campaigns in the discontinuous and sporadic permafrost zones of western Canada. The research team will (i) elucidate the role of plant biodiversity in controlling permafrost status in contrasting ecosystems with differing soil profiles; (ii) determine the extent to which the recovery of key plant communities determines the resilience of permafrost to fire disturbance; and (iii) determine the effects of permafrost thaw on carbon dioxide versus methane fluxes in free-draining ecosystems versus peatland systems

In NWT the research team will set up 7 field sites. Four sites will be in uplands; of these 2 will be in spruce forest, selected for contrasting burn history (undisturbed versus fire disturbed). The other 2 sites will be in birch forest, again selected for contrasting burn history. Three sites will be in poorly drained lowlands, sampling an extensive stable peat plateau, a collapse lawn, and a permafrost-free peatland. Initial deployment of equipment will be in August/September 2012, with main field campaigns in summer 2013 and 2014. Equipment will be removed at the end of summer 2014. Some targeted sampling of soils and vegetation is planned, with small harvests and some soil coring to collect material for chemical analysis. All data will be archived with Canadian partners and local agencies on request.

At each site, five 20 by 20 m plots, representative of the main vegetation communities, will be established. In the central plot, the research team will log hourly soil temperature (at 0.03, 0.15, 0.5, 1.0 and 2.0 m), in accordance with the protocol of the Circumpolar Active Layer Monitoring (CALM) program, soil moisture (0-0.3 m), soil heat flux (2 locations), air temperature and down-welling shortwave radiation. In all other 20 x 20 m plots, soil temperature and moisture (at multiple depths to 1 m, including the standard 0.15 m, CALM protocol), and thaw and water table depths will be measured manually each week, using mechanical probing, linked to thermistor and Delta-T PR2 probes. Within these plots, locations for monitoring net ecosystem (understorey) CO2 exchange, CH4 fluxes, below-ground respiration and 14CO2 release will be established, and tree growth will be measured in the central plot. These data will be linked with plant biomass measurements (understorey, moss and tree biomass) and soil sampling to 2 m. To further quantify links between vegetation community, organic horizon depth and permafrost extent, a further thirty “extensive sampling” plots will be established through a combination of cyclic sampling, random distribution and targeted sampling (to capture in-site variation). Here, soil temperature (multiple depth), surface soil moisture and thaw depth will be monitored bi-weekly, and linked to vegetation surveys and soil profile descriptions. Weather station data are available close to each site.

Plant biomass, C and N stocks. Tree biomass (present at most sites) will be determined from stem diameter maps within the five 20 x 20 m plots combined with standard allometric equations, with tree-ring cores (central plot only) to assess recent changes in woody-biomass production. Understorey (field layer) and bryophyte peak biomass and C stocks will be assessed mid-season by harvests of five 50 x 50 cm plots per site (25x25cm for bryophytes) plus a soil core for fine roots. These will be sorted into stem, leaf and fine root components and sub-samples returned to the UK for C and N analysis, lignin and cellulose content, and leaf mass per area for model calibration. In the 30 “extensive sampling” plots, leaf area index (LiCor LAI-2000) of the tree canopy and understorey (separately) will be quantified. Cover abundance of the dominant species (including moss species) will be assessed by eye in 1m2 quadrats (sub-divided into 4 sections). This cover survey approach is most appropriate since it is rapid, and estimates cover well when targeting the main species. Depth of the moss layer (along with species identity) will also be measured at the 30 extensive sampling points.

Soil and permafrost C stocks and decomposability
Sampling will take place adjacent to the five plant biomass harvest plots. To avoid compression, surface moss and organic soils above the frozen layer will be sampled from shallow pits (depths up to 40 cm) or using a large cross section box corer (depths > 40 cm). A large capacity Russian corer will be used for any deeper unfrozen peat. Frozen material will be sampled with a CRREL permafrost corer provided by project partner Wolfe. Cores will be divided by horizon, and C stocks determined down to a depth of 2 m for model initialization. Soil/peat core samples will be returned to the UK for detailed lab analyses.

CO2 and CH4 fluxes
Net CO2 exchange and photosynthesis curves:
For C flux budgeting, understorey net ecosystem exchange will be determined mid-season enclosing each biomass harvest/vegetation survey plot location (immediately prior to harvest) using 1m2 squared roving chambers attached to a PP-Systems EGM-4 IRGA. NEE will be quantified 5 times through the season on the central plot, to gain temporal data for model evaluation (and so will occur beside the harvested quadrat, rather than enclosing it). The same approach, though using a 25x25 cm chamber will be used to quantify moss photosynthesis for model data. These measurements will be supported by ACi curves taken from detached branches of dominant tree species assessed within the central intensive plot.

Partitioning below-ground respiration: On five dates during the growing season, rates of CO2 production will be measured, using an IRGA (EGM-4, PP Systems), at surface collars and collars inserted 40 cm into the soil profile severing all near surface roots (deep collars). Both collar types will be maintained free of above-ground plant biomass. Rates of respiration measured at the deep collars represent heterotrophic respiration, and subtracting rates measured at the deep collars from those measured at the shallow collars, allows rates of autotrophic soil respiration to be calculated. The partitioning will allow calibration of heterotrophic and autotrophic contributions to modeled soil effluxes over time.

CH4 fluxes: CH4 fluxes will be measured from vegetated collars, with floating collars being deployed in the wettest parts of the landscape. Chambers will be connected to the collars and methane concentrations measured soon after closure and then after a set period dependent on production rates. Headspace concentrations will be measured using a CH4 analyser (DP-IR, Heath Inc.). This analyser is extremely portable allowing a large number of locations to be measured simultaneously, and has been tested extensively in the field, and laboratory, by project partner Garnett. Fluxes will be measured on 5 dates, to provide evaluation data for seasonal model simulations.

14CO2 production and the release of 'old' C: The 14C content of the CO2 released through heterotrophic respiration will be measured using collars inserted 40 cm into the soil [whole soil (WS) collars]. The research team will also determine the contribution of CO2 released from deep in the profile. The research team will extract 40 cm-deep cores, place them into cylinders with sealed bottoms, and reinstall, thus excluding CO2 produced below 40 cm (near-surface collars; NS). To determine the isotopic signature of CO2 produced at depth, probes with hydrophobic filters that allow soil air to be sampled from below 40 cm will also be inserted into the soil profile. The contribution of deep C can then be calculated based on the difference in the 14C content of the CO2 respired from the two collar types (WS versus NS) and the 14C content of the CO2 collected by the probe. Samples for 14C analysis will be collected over an 8 week period using a passive sampling method. By permitting integration over an extended period of the growing season, together with the relative ease of setting up the collection, the passive sampling method extends the spatial extent of our 14CO2 survey. Isotopic fractionation caused by this method has been quantified, and 14C results are unaffected. These data provide model constraints on the impact of permafrost thaw for decomposition rates of previously frozen C.

The work will be of interest to communities concerned about sustainability of their way of life, water resources, grazing, and changes to permafrost and thermokarst. The research team will engage with local communities, to explain the research and to learn more about their perspectives on, and experiences with, climate change.
The research team will inform the public of our activities through information displays and talks at the Yellowknife Northern Heritage Centre.
On the visits to Yellowknife the research team will also make contact with local media and schools to offer interviews and classroom visits.
The project results will be on the website, allowing the broader community to view and access the research directly.

The fieldwork for this study will be conducted from August 19, 2012 to September 19, 2012.