Statistical upscaling of ecosystem CO2 fluxes across the terrestrial tundra and boreal domain: Regional patterns and uncertainties

The regional variability in tundra and boreal carbon dioxide (CO₂ ) fluxes can be high, complicating efforts to quantify sink-source patterns across the entire region. Statistical models are increasingly used to predict (i.e., upscale) CO₂ fluxes across large spatial domains, but the reliability of different modeling techniques, each with different specifications and assumptions, has not been assessed in detail. Here, we compile eddy covariance and chamber measurements of annual and growing season CO₂ fluxes of gross primary productivity (GPP), ecosystem respiration (ER), and net ecosystem exchange (NEE) during 1990-2015 from 148 terrestrial high-latitude (i.e., tundra and boreal) sites to analyze the spatial patterns and drivers of CO₂ fluxes and test the accuracy and uncertainty of different statistical models. CO₂ fluxes were upscaled at relatively high spatial resolution (1 km² ) across the high-latitude region using five commonly used statistical models and their ensemble, that is, the median of all five models, using climatic, vegetation, and soil predictors. We found the performance of machine learning and ensemble predictions to outperform traditional regression methods. We also found the predictive performance of NEE-focused models to be low, relative to models predicting GPP and ER. Our data compilation and ensemble predictions showed that CO₂ sink strength was larger in the boreal biome (observed and predicted average annual NEE -46 and -29 g C m^-2  yr^-1 , respectively) compared to tundra (average annual NEE +10 and -2 g C m^-2  yr^-1 ). This pattern was associated with large spatial variability, reflecting local heterogeneity in soil organic carbon stocks, climate, and vegetation productivity. The terrestrial ecosystem CO₂ budget, estimated using the annual NEE ensemble prediction, suggests the high-latitude region was on average an annual CO₂ sink during 1990-2015, although uncertainty remains high.

Spatial heterogeneity and environmental predictors of permafrost region soil organic carbon stocks

Large stocks of soil organic carbon (SOC) have accumulated in the Northern Hemisphere permafrost region, but their current amounts and future fate remain uncertain. By analyzing dataset combining >2700 soil profiles with environmental variables in a geospatial framework, we generated spatially explicit estimates of permafrost-region SOC stocks, quantified spatial heterogeneity, and identified key environmental predictors. We estimated that Pg C are stored in the top 3 m of permafrost region soils. The greatest uncertainties occurred in circumpolar toe-slope positions and in flat areas of the Tibetan region. We found that soil wetness index and elevation are the dominant topographic controllers and surface air temperature (circumpolar region) and precipitation (Tibetan region) are significant climatic controllers of SOC stocks. Our results provide first high-resolution geospatial assessment of permafrost region SOC stocks and their relationships with environmental factors, which are crucial for modeling the response of permafrost affected soils to changing climate.

Large loss of CO2 in winter observed across the northern permafrost region

Recent warming in the Arctic, which has been amplified during the winter^1-3, greatly enhances microbial decomposition of soil organic matter and subsequent release of carbon dioxide (CO₂)⁴. However, the amount of CO₂ released in winter is highly uncertain and has not been well represented by ecosystem models or by empirically-based estimates^5,6. Here we synthesize regional in situ observations of CO₂ flux from arctic and boreal soils to assess current and future winter carbon losses from the northern permafrost domain. We estimate a contemporary loss of 1662 Tg C yr^-1 from the permafrost region during the winter season (October through April). This loss is greater than the average growing season carbon uptake for this region estimated from process models (-1032 Tg C yr^-1). Extending model predictions to warmer conditions in 2100 indicates that winter CO₂ emissions will increase 17% under a moderate mitigation scenario-Representative Concentration Pathway (RCP) 4.5-and 41% under business-as-usual emissions scenario-RCP 8.5. Our results provide a new baseline for winter CO₂ emissions from northern terrestrial regions and indicate that enhanced soil CO₂ loss due to winter warming may offset growing season carbon uptake under future climatic conditions.

Ecosystem carbon response of an Arctic peatland to simulated permafrost thaw

Permafrost peatlands are biogeochemical hot spots in the Arctic as they store vast amounts of carbon. Permafrost thaw could release part of these long-term immobile carbon stocks as the greenhouse gases (GHGs) carbon dioxide (CO₂ ) and methane (CH₄ ) to the atmosphere, but how much, at which time-span and as which gaseous carbon species is still highly uncertain. Here we assess the effect of permafrost thaw on GHG dynamics under different moisture and vegetation scenarios in a permafrost peatland. A novel experimental approach using intact plant-soil systems (mesocosms) allowed us to simulate permafrost thaw under near-natural conditions. We monitored GHG flux dynamics via high-resolution flow-through gas measurements, combined with detailed monitoring of soil GHG concentration dynamics, yielding insights into GHG production and consumption potential of individual soil layers. Thawing the upper 10-15 cm of permafrost under dry conditions increased CO₂ emissions to the atmosphere (without vegetation: 0.74 ± 0.49 vs. 0.84 ± 0.60 g CO₂ -C m^-2  day^-1 ; with vegetation: 1.20 ± 0.50 vs. 1.32 ± 0.60 g CO₂ -C m^-2  day^-1 , mean ± SD, pre- and post-thaw, respectively). Radiocarbon dating (¹⁴ C) of respired CO₂ , supported by an independent curve-fitting approach, showed a clear contribution (9%-27%) of old carbon to this enhanced post-thaw CO₂ flux. Elevated concentrations of CO₂ , CH₄ , and dissolved organic carbon at depth indicated not just pulse emissions during the thawing process, but sustained decomposition and GHG production from thawed permafrost. Oxidation of CH₄ in the peat column, however, prevented CH₄ release to the atmosphere. Importantly, we show here that, under dry conditions, peatlands strengthen the permafrost-carbon feedback by adding to the atmospheric CO₂ burden post-thaw. However, as long as the water table remains low, our results reveal a strong CH₄ sink capacity in these types of Arctic ecosystems pre- and post-thaw, with the potential to compensate part of the permafrost CO₂ losses over longer timescales.

Tundra landscape heterogeneity, not interannual variability, controls the decadal regional carbon balance in the Western Russian Arctic

Across the Arctic, the net ecosystem carbon (C) balance of tundra ecosystems is highly uncertain due to substantial temporal variability of C fluxes and to landscape heterogeneity. We modeled both carbon dioxide (CO₂ ) and methane (CH₄ ) fluxes for the dominant land cover types in a ~100-km² sub-Arctic tundra region in northeast European Russia for the period of 2006-2015 using process-based biogeochemical models. Modeled net annual CO₂ fluxes ranged from -300 g C m^-2  year^-1 [net uptake] in a willow fen to 3 g C m^-2  year^-1 [net source] in dry lichen tundra. Modeled annual CH₄ emissions ranged from -0.2 to 22.3 g C m^-2  year^-1 at a peat plateau site and a willow fen site, respectively. Interannual variability over the decade was relatively small (20%-25%) in comparison with variability among the land cover types (150%). Using high-resolution land cover classification, the region was a net sink of atmospheric CO₂ across most land cover types but a net source of CH₄ to the atmosphere due to high emissions from permafrost-free fens. Using a lower resolution for land cover classification resulted in a 20%-65% underestimation of regional CH₄ flux relative to high-resolution classification and smaller (10%) overestimation of regional CO₂ uptake due to the underestimation of wetland area by 60%. The relative fraction of uplands versus wetlands was key to determining the net regional C balance at this and other Arctic tundra sites because wetlands were hot spots for C cycling in Arctic tundra ecosystems.

Nongrowing season methane emissions-a significant component of annual emissions across northern ecosystems

Wetlands are the single largest natural source of atmospheric methane (CH₄ ), a greenhouse gas, and occur extensively in the northern hemisphere. Large discrepancies remain between "bottom-up" and "top-down" estimates of northern CH₄ emissions. To explore whether these discrepancies are due to poor representation of nongrowing season CH₄ emissions, we synthesized nongrowing season and annual CH₄ flux measurements from temperate, boreal, and tundra wetlands and uplands. Median nongrowing season wetland emissions ranged from 0.9 g/m² in bogs to 5.2 g/m² in marshes and were dependent on moisture, vegetation, and permafrost. Annual wetland emissions ranged from 0.9 g m^-2  year^-1 in tundra bogs to 78 g m^-2  year^-1 in temperate marshes. Uplands varied from CH₄ sinks to CH₄ sources with a median annual flux of 0.0 ± 0.2 g m^-2  year^-1 . The measured fraction of annual CH₄ emissions during the nongrowing season (observed: 13% to 47%) was significantly larger than that was predicted by two process-based model ensembles, especially between 40° and 60°N (modeled: 4% to 17%). Constraining the model ensembles with the measured nongrowing fraction increased total nongrowing season and annual CH₄ emissions. Using this constraint, the modeled nongrowing season wetland CH₄ flux from >40° north was 6.1 ± 1.5 Tg/year, three times greater than the nongrowing season emissions of the unconstrained model ensemble. The annual wetland CH₄ flux was 37 ± 7 Tg/year from the data-constrained model ensemble, 25% larger than the unconstrained ensemble. Considering nongrowing season processes is critical for accurately estimating CH₄ emissions from high-latitude ecosystems, and necessary for constraining the role of wetland emissions in a warming climate.

The positive net radiative greenhouse gas forcing of increasing methane emissions from a thawing boreal forest-wetland landscape

At the southern margin of permafrost in North America, climate change causes widespread permafrost thaw. In boreal lowlands, thawing forested permafrost peat plateaus ('forest') lead to expansion of permafrost-free wetlands ('wetland'). Expanding wetland area with saturated and warmer organic soils is expected to increase landscape methane (CH₄ ) emissions. Here, we quantify the thaw-induced increase in CH₄ emissions for a boreal forest-wetland landscape in the southern Taiga Plains, Canada, and evaluate its impact on net radiative forcing relative to potential long-term net carbon dioxide (CO₂ ) exchange. Using nested wetland and landscape eddy covariance net CH₄ flux measurements in combination with flux footprint modeling, we find that landscape CH₄ emissions increase with increasing wetland-to-forest ratio. Landscape CH₄ emissions are most sensitive to this ratio during peak emission periods, when wetland soils are up to 10 °C warmer than forest soils. The cumulative growing season (May-October) wetland CH₄ emission of ~13 g CH₄  m^-2 is the dominating contribution to the landscape CH₄ emission of ~7 g CH₄  m^-2 . In contrast, forest contributions to landscape CH₄ emissions appear to be negligible. The rapid wetland expansion of 0.26 ± 0.05% yr^-1 in this region causes an estimated growing season increase of 0.034 ± 0.007 g CH₄  m^-2  yr^-1 in landscape CH₄ emissions. A long-term net CO₂ uptake of >200 g CO₂  m^-2  yr^-1 is required to offset the positive radiative forcing of increasing CH₄ emissions until the end of the 21st century as indicated by an atmospheric CH₄ and CO₂ concentration model. However, long-term apparent carbon accumulation rates in similar boreal forest-wetland landscapes and eddy covariance landscape net CO₂ flux measurements suggest a long-term net CO₂ uptake between 49 and 157 g CO₂  m^-2  yr^-1 . Thus, thaw-induced CH₄ emission increases likely exert a positive net radiative greenhouse gas forcing through the 21st century.

A simplified, data-constrained approach to estimate the permafrost carbon-climate feedback

We present an approach to estimate the feedback from large-scale thawing of permafrost soils using a simplified, data-constrained model that combines three elements: soil carbon (C) maps and profiles to identify the distribution and type of C in permafrost soils; incubation experiments to quantify the rates of C lost after thaw; and models of soil thermal dynamics in response to climate warming. We call the approach the Permafrost Carbon Network Incubation-Panarctic Thermal scaling approach (PInc-PanTher). The approach assumes that C stocks do not decompose at all when frozen, but once thawed follow set decomposition trajectories as a function of soil temperature. The trajectories are determined according to a three-pool decomposition model fitted to incubation data using parameters specific to soil horizon types. We calculate litterfall C inputs required to maintain steady-state C balance for the current climate, and hold those inputs constant. Soil temperatures are taken from the soil thermal modules of ecosystem model simulations forced by a common set of future climate change anomalies under two warming scenarios over the period 2010 to 2100. Under a medium warming scenario (RCP4.5), the approach projects permafrost soil C losses of 12.2-33.4 Pg C; under a high warming scenario (RCP8.5), the approach projects C losses of 27.9-112.6 Pg C. Projected C losses are roughly linearly proportional to global temperature changes across the two scenarios. These results indicate a global sensitivity of frozen soil C to climate change (γ sensitivity) of -14 to -19 Pg C °C(-1) on a 100 year time scale. For CH4 emissions, our approach assumes a fixed saturated area and that increases in CH4 emissions are related to increased heterotrophic respiration in anoxic soil, yielding CH4 emission increases of 7% and 35% for the RCP4.5 and RCP8.5 scenarios, respectively, which add an additional greenhouse gas forcing of approximately 10-18%. The simplified approach presented here neglects many important processes that may amplify or mitigate C release from permafrost soils, but serves as a data-constrained estimate on the forced, large-scale permafrost C response to warming.

A pan-Arctic synthesis of CH4 and CO2 production from anoxic soil incubations

Permafrost thaw can alter the soil environment through changes in soil moisture, frequently resulting in soil saturation, a shift to anaerobic decomposition, and changes in the plant community. These changes, along with thawing of previously frozen organic material, can alter the form and magnitude of greenhouse gas production from permafrost ecosystems. We synthesized existing methane (CH₄ ) and carbon dioxide (CO₂ ) production measurements from anaerobic incubations of boreal and tundra soils from the geographic permafrost region to evaluate large-scale controls of anaerobic CO₂ and CH₄ production and compare the relative importance of landscape-level factors (e.g., vegetation type and landscape position), soil properties (e.g., pH, depth, and soil type), and soil environmental conditions (e.g., temperature and relative water table position). We found fivefold higher maximum CH₄ production per gram soil carbon from organic soils than mineral soils. Maximum CH₄ production from soils in the active layer (ground that thaws and refreezes annually) was nearly four times that of permafrost per gram soil carbon, and CH₄ production per gram soil carbon was two times greater from sites without permafrost than sites with permafrost. Maximum CH₄ and median anaerobic CO₂ production decreased with depth, while CO₂ :CH₄ production increased with depth. Maximum CH₄ production was highest in soils with herbaceous vegetation and soils that were either consistently or periodically inundated. This synthesis identifies the need to consider biome, landscape position, and vascular/moss vegetation types when modeling CH₄ production in permafrost ecosystems and suggests the need for longer-term anaerobic incubations to fully capture CH₄ dynamics. Our results demonstrate that as climate warms in arctic and boreal regions, rates of anaerobic CO₂ and CH₄ production will increase, not only as a result of increased temperature, but also from shifts in vegetation and increased ground saturation that will accompany permafrost thaw.

Temperature and peat type control CO2 and CH4 production in Alaskan permafrost peats

Controls on the fate of ~277 Pg of soil organic carbon (C) stored in permafrost peatland soils remain poorly understood despite the potential for a significant positive feedback to climate change. Our objective was to quantify the temperature, moisture, organic matter, and microbial controls on soil organic carbon (SOC) losses following permafrost thaw in peat soils across Alaska. We compared the carbon dioxide (CO2 ) and methane (CH4 ) emissions from peat samples collected at active layer and permafrost depths when incubated aerobically and anaerobically at -5, -0.5, +4, and +20 °C. Temperature had a strong, positive effect on C emissions; global warming potential (GWP) was >3× larger at 20 °C than at 4 °C. Anaerobic conditions significantly reduced CO2 emissions and GWP by 47% at 20 °C but did not have a significant effect at -0.5 °C. Net anaerobic CH4 production over 30 days was 7.1 ± 2.8 μg CH4 -C gC(-1) at 20 °C. Cumulative CO2 emissions were related to organic matter chemistry and best predicted by the relative abundance of polysaccharides and proteins (R(2) = 0.81) in SOC. Carbon emissions (CO2 -C + CH4 -C) from the active layer depth peat ranged from 77% larger to not significantly different than permafrost depths and varied depending on the peat type and peat decomposition stage rather than thermal state. Potential SOC losses with warming depend not only on the magnitude of temperature increase and hydrology but also organic matter quality, permafrost history, and vegetation dynamics, which will ultimately determine net radiative forcing due to permafrost thaw.