Next Article in Journal
Analysis of Ionospheric VTEC Retrieved from Multi-Instrument Observations
Previous Article in Journal
Harmful Risk of Bioaerosol Pollution at Major Indoor Sites of a Summer Campus in Guilin City
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Atmosphere
Volume 15
Issue 6
10.3390/atmos15060698
atmosphere-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Variation in Soil CO2 Fluxes across Land Cover Mosaic in Typical Tundra of the Taimyr Peninsula, Siberia

by
Alexey Panov
1,*,
Anatoly Prokushkin
1,
Mikhail Korets
1,
Ilya Putilin
1,
Galina Zrazhevskaya
1,
Roman Kolosov
1 and
Mikhail Bondar
2
1
V.N. Sukachev Institute of Forest of the Siberian Branch of Russian Academy of Sciences—Separated Department of the KSC SB RAS, 660036 Krasnoyarsk, Russia
2
Joint Directorate of Taimyr Nature Reserves, 663305 Norilsk, Russia
*
Author to whom correspondence should be addressed.
Atmosphere 2024, 15(6), 698; https://doi.org/10.3390/atmos15060698
Submission received: 17 April 2024 / Revised: 21 May 2024 / Accepted: 6 June 2024 / Published: 9 June 2024
(This article belongs to the Special Issue Carbon Fluxes in the Pan-Arctic Region)
Browse Figures
Versions Notes

Abstract

:
Increased warming in the Arctic is of great concern. This is particularly due to permafrost degradation, which is expected to accelerate microbial breakdown of soil organic carbon, with its further release into the atmosphere as carbon dioxide (CO2). The fine-scale variability of CO2 fluxes across highly mosaic Arctic tundra landscapes can provide us with insights into the diverse responses of individual plant communities to environmental change. In the paper, we contribute to filling existing gaps by investigating the variability of CO2 flux rates within different landscape units for dominant vegetation communities and plant species across typical tundra of the southern part of the Taimyr Peninsula, Siberia. In general, the variability of soil CO2 flux illustrates a four-fold increase from non-vascular vegetation, mainly lichens and mosses (1.05 ± 0.36 µmol m−2 s−1), towards vascular plants (3.59 ± 0.51 µmol m−2 s−1). Barren ground (“frost boils”) shows the lowest value of 0.79 ± 0.21 µmol m−2 s−1, while considering the Arctic “browning” phenomenon, a further substantial increase of CO2 flux can be expected with shrub expansion. Given the high correlation with top soil temperature, well-drained and relatively dry habitats such as barren ground and non-vascular vegetation are expected to be the most sensitive to the observed and projected temperature growth in the Arctic. For mixed vegetation and vascular species that favor wetter conditions, soil moisture appears to play a greater role. Based on the modeled seasonal pattern of soil CO2 flux and precipitation records, and applying the rainfall simulations in situ we outlined the role of precipitation across enhanced CO2 emissions (i.e., the “Birch” effect). We found that a pulse-like growth of soil CO2 fluxes, observed within the first few minutes after rainfall on vegetated plots, reaches 0.99 ± 0.48 µmol m−2 s−1 per each 1 mm of precipitation, while barren ground shows 55–70% inhibition of CO2 emission during the first several hours. An average additive effect of precipitation on soil CO2 flux may achieve 7–12% over the entire growing season, while the projected increased precipitation regime in the Arctic may strengthen the total CO2 release from the soil surface to the atmosphere during the growing season.

1. Introduction

The ongoing warming of the Arctic, with observed temperature increases at more than twice the rate of the Northern Hemisphere [1,2,3], is of great importance [4,5]. This is primarily due to the large amount of carbon stored in perennially frozen soil (permafrost) becoming accessible for mobilization in deglaciated soils [5,6,7]. Rising air temperatures and permafrost degradation are expected to accelerate the microbial breakdown of soil organic carbon, with its further release as carbon dioxide (CO2) into the atmosphere [8,9] and partial lateral export to the Arctic Ocean [10]. At the same time, shifts in the duration and extent of snow cover [11] and snow-free periods, followed by higher temperature records, are thought to alter hydrological cycles [12,13] as well as vegetation distribution and growth (i.e., tundra “greening”) [14], which in turn may increase carbon uptake [15]. The balance of these opposing effects is expected to modulate the shape of future carbon turnover in the Arctic. Observed and projected shifts in plant communities towards shrubs and late-successional species [16], disturbance and vegetation stress leading to biomass losses [17], and nutrient limitation [18] introduce further uncertainties in assessing the fate of the Arctic carbon cycle.
Recent efforts to calculate the terrestrial carbon balance in the Arctic have produced rather contradictory results, e.g., [5,19]. Measurements across the Arctic suggest that tundra ecosystems act as a weak net CO2 sink [20,21] or a source of CO2 to the atmosphere [22,23], that the balance is close to zero [24], or that annual variability may mask the true long-term carbon dynamics [25]. A meta-analysis based on 54 studies in the tundra [26,27] found an increase in CO2 sources, mainly due to higher ecosystem respiration rates, resulting in increased carbon release to the atmosphere under warmer conditions. Among other environmental and climatic factors, a significant part of the uncertainty and variability in the Arctic carbon cycle is related to the high heterogeneity of tundra landscapes, which is evident at multiple scales [28,29]. Different environmental conditions and vegetation types build up a mosaic of ecosystems at the landscape scale, while these ecosystems in turn form a patchy structure of diverse plant communities. At this fine (synusiae) scale, a diversity of environmental conditions is evident within a few meters [30]. Such environmental diversity is thought to partially offset climatological effects and mitigate climate-driven transformation of CO2 fluxes [31]. However, the fine-scale variability of CO2 fluxes remains a scientific challenge [27] whose study may provide insights into understanding the diverse responses of individual plant communities to ongoing processes.
Observations of CO2 fluxes at the level of individual plant communities and fine-scale spatial mosaics (100–10.000 square centimeters) [27] are mostly made using chamber measurements. Chambers are widely used instruments due to their autonomy of power supply, mobility, and relatively robust operation, although manual measurements in the field can be quite laborious [27]. However, they remain extremely scarce in the pan-Arctic [32], which is particularly true in the high Arctic latitudes of Siberia, with ~3 million square kilometers of tundra ecosystems underlain by permafrost, amounting to nearly 50% of the tundra biome globally, e.g., see [33]. Despite their global importance in the carbon cycle, there are only a few sites where chamber measurements of soil CO2 fluxes are systematically performed in the Siberian Arctic, notably the Cherskii Northeast Science Station at the mouth of the Kolyma River, Northeast Siberia [34], and the Tiksi Observatory in Sakha (Yakutia), Northeast Siberia [35].
In this paper, we contribute to filling the existing gaps by investigating (i) the variability of CO2 flux rates at the fine scale within fixed sampling points across the different landscape units, (ii) the spatio-temporal variations of CO2 fluxes within the dominant vegetation communities at the site scale, and (iii) the differences in CO2 release between non-vascular and vascular plant species. In addition, based on the calculated flux of top soil temperature dependencies, we simulated the seasonal variations of soil CO2 flux and attempted to outline the effect of precipitation on significant pulse-like CO2 emissions (i.e., the “Birch” effect—a rapid growth of CO2 to a peak value followed by an exponential rise during rainfall events), based on the seasonal flux pattern and precipitation simulations. To achieve our goals, we performed chamber measurements of CO2 fluxes in situ and under manipulations, coupled with records of meteoclimatic variables in diverse landscapes and vegetation across typical tundra in the southwestern coast of the Taimyr Peninsula (Siberia) during the growing seasons in 2021–2022. The reported study is part of the long-term research program linked to accurate continuous atmospheric observations of carbon dioxide and methane in the area, which have been operational since 2018 at the “DIAMIS” (Dikson Atmospheric Measurement Integration Station) station [36,37].

2. Materials and Methods

2.1. Study Area

The study area is located within the Meduza Bay of the Gulf of the Yenisei River, on the southwestern coast of the Taimyr Peninsula (Siberia), a northernmost extension of the Eurasian continent (Figure 1). The area is a part of the Great Arctic Reserve, and a scientific facility is located at the Willem Barentsz Biological Station (https://eu-interact.org/field-sites/willem-barents-biological-station/ (accessed on 18 March 2024)), ~30 km southwest of the Dikson settlement. The landscape of the area is mainly formed by low hills (up to 60 m a.s.l.), ridges, and exposures of bare rocks. Between the hills and along the streams, marshy depressions are typical. The soil is perennially frozen. The Meduza River is the main stream in the area and is the only permanent stream during the summer; it freezes again by late fall. Several other streams cross the area, starting to flow at the end of June and mostly running dry during the summer [38]. The Köppen climate classification [39] refers the study area as having a “polar” or “tundra” climate. Summer in the area is short, with an average air temperature of +3.8 °C, while winter is windy, with frequent intense blizzards, but relatively mild, with an average air temperature of −24.2 °C. The further meteoclimatic details of the area have been previously presented elsewhere [36,37]. According to the Circumpolar Arctic Vegetation Map [40], the study area can be classified as the southern part of the Arctic tundra, almost bordering the northern hypoarctic tundra.
Six vegetation types are reported across the area [38]; on most of the territory, moss tundra, frost-heaved tundra, and marsh tundra prevail. These vegetation types within the highly heterogeneous tundra landscapes are diverse in terms of soil thermal regime, hydrology, geomorphological processes, and biogeochemical functions [14]. In particular, moss tundra is found mostly on the lower parts of the slopes and is dominated by mosses and dwarf willows, while frost tundra is widespread on the higher slopes and hilltops. On the other hand, marsh tundra dominates in the flooded areas, with sedges, cotton grass, and other grasses growing in the depressions and along the rivers.

2.2. Research Plots

In order to capture the dominant land cover types across the study area, three representative sites were selected and processed for further observations. The multispectral (NIR-R-G-NDVI) unmanned aerial vehicle (UAV) imagery (resolution 0.16 m) was applied for definition of the dominant land cover classes: 4-band orthophoto images were segmented (Multiresolution approach—Trimble eCognition) and classified (IsoCluster method—ESRI ArcGIS). The base land cover classes were identified and aggregated according to in situ description of the vegetation. The final land cover maps were compiled as a union of the base class’s polygons to build the legend of the 6 dominant land cover types (Figure 2, Table 1).
Across the study sites, permanent research plots were established in moss (Figure 2a), frost-heaved (Figure 2b), and marsh tundra (Figure 2c), representing diverse topography, vegetation, and active layer depths. On the research plots, continuous meteoclimatic observations have been operational since August 2021, based on the widely used [41] Vantage Pro2 weather stations (Davis Instruments, Hayward, CA, USA) (Figure 1). Weather variables continuously recorded by meteorological sensors installed at 2 m a.g.l. and inside the soil surface include the following: wind parameters (speed and direction), atmospheric pressure, air temperature, relative humidity, solar radiation, precipitation, soil temperature, and moisture records at different depths. A Davis 6410 anemometer (Davis Instruments, Hayward, CA, USA) measures wind speed and direction. The instrument is resistant to hurricane-force winds, which is critical for the study area. An MPL3115A2 barometric pressure sensor (NXP B.V., Eindhoven, The Netherlands) is used for accurate observations of air pressure and altitude. Air temperature is measured by a DS18B20 digital temperature sensor module (HK Shanghai Group Ltd., Shenzhen, China), while a HIH-5031 low-voltage humidity sensor (Honeywell International Inc., Charlotte, NC, USA) continuously collects data on relative humidity. A Davis 6450 solar radiation sensor (Davis Instruments, Hayward, CA, USA) is used for solar radiation measurements. A stand-alone Davis 6466M rain gauge (Davis Instruments, Hayward, CA, USA) with a mounted wind-protection system measures precipitation amounts. Additionally, soil temperature (0, 2, 5, 10, 15, 20, 30, 40, 50, 60, 70, 80, 90, and 100 cm depths) and soil moisture records (10, 30, and 60 cm depths) are collected using DS18B20 digital temperature sensor probes (Maxim Integrated Products Inc., San Jose, CA, USA) and interface capacitive V1.2 soil moisture sensors (Zhiwei Robotics Corp., Shanghai, China), respectively. An internal data logger of the weather station is used for programmed logging of meteorological variables, with an optionally adjustable resolution, for routine measurements during the year (hourly) or field manipulations (minutely). The weather stations are continuously operational throughout the year (including low-sun and no-sun or “polar night” periods) by means of solar panels and are powered by low-temperature batteries with intelligent charge controllers.
Figure 3 illustrates the seasonal patterns of the following meteoclimatic variables for two vegetation types—moss tundra and frost-heaved tundra: air temperature (Figure 3a,b) as well as soil temperature (Figure 3c,d) and soil moisture (Figure 3e,f) at 10 cm depth. The plots in these two ecosystems are rather contrasting in terms of seasonal freeze–thaw depth. Unfortunately, due to a technical failure during data acquisition in 2021, continuous records for the marsh tundra are only available since October 2022 and are therefore not discussed here. The available patterns demonstrate the relatively comparable records of mean air temperature (Figure 3a,b) throughout the seasons for both vegetation types: −21.4 °C and −21.9 °C in winter (DJF), and +6.9 °C and +7.4 °C in summer (JJA), respectively, for moss and frost-heaved tundra.
On the other hand, top soil temperature (10 cm depth) (Figure 3c,d), which is very sensitive to seasonality and air temperature extremes, shows a high seasonal discrepancy between these vegetation types. In particular, we observed significantly lower soil temperatures in winter in the moss tundra site (−19.4 °C), located at the bottom of the slope, compared to the frost-heaved tundra site (−14.9 °C), located on the hill. This could be attributed to the persistent strong winds in the area [37], which greatly expose flat areas of moss tundra, which can be snow-free for most of the winter [38], while gravel ridges above the frost-heaved tundra retain snow that insulates soil from the atmosphere. Personal communication with the local state officer at the Willem Barentsz Biological Station confirms our suggestion. In summer (JJA), the mean values of soil temperature for both vegetation types are comparable: +2.1 °C and +2.3 °C for moss and frost-heaved tundra, respectively.
The extreme temperatures at top soil depths over the observed vegetation types determine the seasonal thickness of the active layer, which appears to be essentially shallower in moss tundra (Figure 4a) than in frost-heaved tundra (Figure 4b) with large gravel inclusions in the soil, thus modulating soil biodiversity, root growth, water cycles, and vegetation composition. As shown in Figure 4, a large portion of the active soil layer in both vegetation types remains unfrozen for an extended period of time, lasting 121 days (1 July–29 October) and 105 days (6 July–22 October), respectively, for moss and frost-heaved tundra, including an early colder period when the soil temperature fluctuates around 0 °C, the so-called “zero curtain”: 19 days (10–29 October) for moss and 14 days (8–22 October) for frost-heaved tundra. Microbial communities have been also found to be active during the “zero curtain” [42].
Seasonal patterns of the subsurface soil moisture regime also provide valuable information on shallow soil hydrology, which is mostly related to the multiple above-surface processes and vegetation dynamics, and in particular to the duration and timing of snow cover and spring melt. In general, the two plots demonstrate similar patterns (Figure 3e,f), which mostly differ during the spring thaw. In particular, the plot in moss tundra that was snow-free since the winter showed a faster thawing of the topsoil and its thermal transition towards 0 °C due to free convection with air, which forms a massive water input into the deeper soil and, hence, a sharp spike in soil moisture (Figure 3e). For the same duration, spring melt in the snowier frost-heaved tundra demonstrates a smoother pattern (Figure 3f).

2.3. Experimental Design and Methods

In general, we define the following hierarchy of sampling site categories: (i) tundra vegetation type (VT), i.e., moss, march, sedge, etc., which characterizes an entire plant community; (ii) vegetation life form (VF) (vascular vs. non-vascular), and (iii) land cover class (LCC), which deals with a fine-scale mosaic of specific vegetation types, including barren ground, lichen, moss, and dominant mixed units. We designed three different measurement algorithms aimed at observing (1) the temporal variation of CO2 flux rates, by studying the daily (24 h) patterns at fixed measurement points in the different landscape units (point scale); (2) the spatio-temporal variability of CO2 fluxes within the dominant vegetation communities over the study site (site scale), by repeatedly measuring fluxes across the ecosystem mosaic; (3) the differences in CO2 release for different vegetation types (plant-specific) by measuring at points covered by individual plant species. For plant-specific sampling, we grouped the vegetation into three classes: non-vascular (NV) vegetation, vascular (VS) plants, and an intermediate group where vascular plants are interspersed with non-vascular species—the mixed class (MIX). Assignment to a particular vegetation class was made on the basis of the dominant vegetation in a sampling point, but this does not completely exclude the partial presence of subdominant plants, for example, the occurrence of vascular species in points assigned exclusively as NV.
Chamber measurements were performed repetitively in 2021–2022 during the peak of the growing season in the Arctic: late July–mid-August. For each plot, we used 8–10 plastic collars (inner diameter 10 cm). Measurements of CO2 fluxes from the soil surface were made using the LI-6400 portable photosynthesis system (LI-COR Biosciences Inc., Lincoln, NE, USA) coupled with the LI-6400-09 soil CO2 flux chamber (LI-COR Biosciences Inc., Lincoln, NE, USA) and an auxiliary type T handheld thermocouple penetration probe HPS-NP-T-316G-12-SMP-M (Omega Engineering, Inc., CT, USA). The instrument has two absolute CO2 and two H2O non-dispersive infrared analyzers, located in the sensor head. The bandwidth of the analyzers is 10 Hz. Measurements were made using collars inserted ~3 cm into the soil and covered by the chamber. The analyzed soil area was 71.6 cm2. Before each measurement cycle, air in the chamber was scrubbed down to remove the CO2 in the closed system down. The scrubber was then switched off, and the CO2 concentration in the chamber headspace increased, resulting in CO2 flux from the soil surface. Data on the CO2 concentration achieved in the ambient state were logged by the instrument. The preliminary flux data, fitted with a regression throughout the measurement, were further utilized to calculate the soil CO2 flux under the target ambient CO2 values. The three automatic repetitions of the cycle for each of the collars were set in a measurement protocol, resulting in accurate values of soil CO2 flux. During statistical processing, the raw records on soil CO2 flux calculated by the instrument were further processed and averaged regarding assigned sampling categories, i.e., VT, VF, and LCC. The statistical analysis was performed with StatPlus package (AnalystSoft Inc., VA, USA) and R Statistical Software (R Core Team 2024, Vienna, Austria).
CO2 flux measurements were coupled with records of top soil temperature at 10 cm depth, and an exponential equation was further used to calculate the relationship between soil temperature and CO2 flux rate and estimate the temperature sensitivity of soil CO2 flux, as the Q10 value, according to the approach given in [43]. Based on the calculated dependencies and continuous meteoclimatic records, we simulated the seasonal patterns of soil CO2 flux for different vegetation types and land cover classes and tried to estimate the rainfall effect via enhanced pulse-like CO2 emissions (i.e., the “Birch” effect) based on both the modeled seasonal flux variations and the rainfall simulations in the field.

3. Results and Discussion

3.1. Soil CO2 Flux Rates across Typical Tundra

The mean soil CO2 flux (Fsoil) at the peak of the growing season (late July–mid-August) in the study area reaches 2.23 ± 1.26 µmol m−2 s−1 (Figure 5a) and varies between different vegetation types (F = 3.03; p = 0.000), ranging from low rates observed within the frost-heaved tundra (1.14 ± 0.56 µmol m−2 s−1), intermediate values—in moss tundra (1.81 ± 0.97 µmol m−2 s−1), and the highest records measured in marsh tundra (3.59 ± 0.51 µmol m−2 s−1).
In general, the observed range of CO2 fluxes is consistent with previously reported estimates, in particular for typical tundra on the Chukotskiy Peninsula, Russian Far East [44], middle-Arctic tundra in Yakutia, northeastern Russia [35], high-Arctic tundra in Svalbard, Norway [5,45,46], tundra heath in northwestern Finland [47], shrub tundra near Kangerlussuaq, West Greenland [48], high-Arctic environments on Melville Island, Nunavut, Canada [49], and permafrost-affected Alaskan tundra landscapes [50]. The variability of Fsoil is modulated by the soil microclimate, which varies spatially over short distances [27], and in this regard, soil temperature is considered to be an essential driver [51] that modulates soil respiration and decomposition processes.
An observed increasing trend in soil flux towards marsh tundra (Figure 5a) is not linearly correlated with corresponding changes in top soil temperature at 10 cm depth (ST10) over these VTs (Figure 5b,c), while the overall temperature dependence shows a bell-like shape with an optimum ST10 value of 5.6 ± 2.4 °C (Figure 5c). ST10 shows statistically significant differences (F = 3.03; p = 0.000) between the VTs. In particular, the minimal mean ST10 (4.1 ± 2.4 °C) was recorded in the moss tundra, with a lower active layer thickness (Figure 4a) compared to the frost-heaved tundra (Figure 4b) and an abundant moss layer that maintains low soil temperatures. Higher ST10 values were observed in the two contrasting VTs, specifically in the overwetted marsh tundra (6.0 ± 0.9 °C) and the essentially drier frost-heaved tundra (6.9 ± 2.7 °C) (Figure 5b). The surface water level in the marsh tundra can convectively transfer heat and spread it within the shallow layers [52], while the large seasonal active layer thickness (Figure 4b), the relatively sparse vegetation, and the multiple “frost boils”—small circular mounds of barren soil resulting from frost actions and cryoturbation processes—provide an easy and deeper thermal convection between the soil surface and atmospheric air in the frost-heaved tundra.
The variability of soil CO2 flux over the assigned land cover classes (LCCs) (F = 2.25; p = 0.000), grouped into vegetation life forms (VFs), illustrates an increasing trend from non-vascular (NV) vegetation, especially lichens and mosses, towards vascular (VS) plants (Figure 5d). In general, mean Fsoil values do not exceed 1.05 ± 0.36 and 2.06 ± 0.92 µmol m−2 s−1 for NV and mixed non-vascular and vascular vegetation (MIX), respectively, while within VS plants soil flux reaches up to 3.59 ± 0.51 µmol m−2 s−1. The observed values are consistent with recently reported estimates for tundra landscapes of Svalbard, Norway, where vascular species showed ~three-fold significantly higher CO2 emission rates than non-vascular vegetation [5]. In turn, Fsoil rates across the LCCs, including barren ground, measured in this study ranged from the lowest rates recorded on the frost boils (0.79 ± 0.21 µmol m−2 s−1) and lichen synusiae (0.98 ± 0. 32 µmol m−2 s−1)—the well-drained and relatively dry habitats—to the much higher values found on the moisture-favored sedge synusiae (3.59 ± 0.51 µmol m−2 s−1), where the wet conditions and larger vegetation may have led to increased productivity and production of plant root exudates in the vegetation [49]. The coupled importance of water levels and vascular plant abundance in relation to soil CO2 flux across the Arctic tundra has also been shown previously by other studies [53]. In particular, the recent analysis in [50] noted that higher summer flux rates were often observed in the Alaskan tundra where subsurface soils were relatively wet, but not oversaturated. A post hoc Fisher’s LSD test shows statistically significant differences in Fsoil (F = 2.25; p = 0.000) for LCCs, except for the groups lichen –moss, lichen–barren ground, and barren ground–moss. The mounds of barren ground and NV plots showed comparable mean soil flux rates, not exceeding 0.97 ± 0.17 µmol m−2 s−1. On the other hand, the absolute values and the variability of the soil CO2 flux increased significantly for the MIX vegetation forms with a mosaic of moss and sedge (1.81 ± 0.90 µmol m−2 s−1) and VS plants (2.39 ± 0.84 µmol m−2 s−1), finally showing a sharp increase towards the marsh tundra (3.59 ± 0.51 µmol m−2 s−1), where sedge prevails (Figure 5d).
In contrast to the observed growth of soil CO2 flux rates within the LCCs according to moisture status (Figure 5f), the mean ST10 (F = 2.25; p = 0.000) showed a rather decreasing trend (Figure 5e), starting from the relatively high values of 8. 2 ± 1.8 °C and 7.9 ± 2.8 °C, respectively recorded in the barren ground and lichen vegetation, and reached minimum values in the LCC, where moss groups dominate: moss (3.6 ± 1.8 °C) and moss/sedge (3.2 ± 1.4 °C). In general, the post hoc test shows statistically different ST10 between most of the LCCs, except lichen–barren ground and moss–moss/herbs.

3.2. Soil CO2 Fluxes and Soil Temperature Relationships: Simulating the Seasonal Patterns

Although the general scatterplot shows no linear correlation between Fsoil and ST10 (Figure 5c), Fsoil–ST10 relationships were found for the relatively dry habitats across the studied LCCs (Figure 5f). These dependencies are well illustrated by the temporal patterns of soil flux variations observed during daily (24 h) point sampling: soil CO2 flux was well correlated with near-surface ST10 for the barren ground (frost boils) (Figure 6a) and NV (moss and lichen) (Figure 6b,c). With a lower R2 value, a temperature dependence was also observed for some of the MIX plots (moss and herbs) (Figure 6d), while for the other part of the MIX (moss and sedges) and VS (sedge) groups we did not observe reliable correlations (Figure 6e,f). Based on the exponential regressions of the measured soil flux rates against ST10, Q10 equaled 3.77 (R2 0.7858) for barren ground (Figure 6a), 2.43 (R2 0.7166) for lichen vegetation (Figure 6b), 9.72 (R2 0.841) for moss vegetation (Figure 6c) and 3.55 (R2 0.562) for the MIX groups with moss vegetation and herbs (Figure 6d). Thus, except for the increased Q10 value found for moss vegetation, where ST10 fluctuated around and slightly below zero at night, the mean Q10 for the other LCCs (with ST10 above zero) does not exceed 2.99 (3.25—with barren ground included in the calculations), which is in the range of previously reported mean Q10 value (2.72) across the tundra biome [54].
The barren ground and NV plots (lichens and mosses) show the stronger Fsoil–ST10 relationships, assuming soil temperature as the main modulating factor for soil respiration under increased soil drainage coupled with high temperatures during the peak season [55]. Conversely, for MIX and VS vegetation growing under wetter conditions, soil moisture appears to play a greater role in controlling both plant and microbial activities. This is consistent with a recent analysis of heterogeneous landscapes in the Alaskan Arctic tundra [56], which also showed that the response of soil CO2 flux to soil temperature increased as site wetness decreased.
Based on the Fsoil–ST10 relationships (R2 0.5), and assuming them valid throughout the growing season (GS), we simulated the seasonal patterns of soil CO2 flux for barren ground (Figure 7a,b), non-vascular vegetation (NV): lichen (Figure 7c,d), moss (Figure 7e,f), and MIX (moss and herbs) (Figure 7g,h) across the study plots in moss and frost-heaved tundra.
As these are the simulated GS patterns, our aim here is not to discuss absolute values but rather to illustrate the shape, timing, and some peculiarities of soil respiration across different LCCs in the studied vegetation types of the typical tundra. In general, the seasonal pattern (JJASON) of soil CO2 flux was followed by the GS, which is rather short in the Arctic and triggered by stable values of positive soil temperature (Figure 3c,d). The pattern is similar for LCCs and VTs. Biogenic activity, which primarily consists of microbial respiration (the heterotrophic flux) and plant root respiration (the autotrophic flux), usually starts around mid-July (12–13), an early season, and an increase in soil flux can be observed from the low rates when ecosystems are still dormant (ST10 > 0 °C) to the maximum records observed in mid-August (14–15), a peak season. During the early GS, the positive soil temperatures (Figure 4) promote the onset of the active soil layer, starting from the uppermost horizons, making the soil organic matter immediately accessible to microbial communities and thus providing a startup of active soil CO2 release. Furthermore, previous studies [57] have found that spring respiration can contribute up to 40% of the Arctic GS net CO2 flux, depending on the year. In turn, higher Fsoil rates in warmer soils during the peak season may be driven not only by increased microbial decomposition of soil organic matter but also by belowground autotrophic respiration, which is considered to be a strong source of CO2 in thawing permafrost systems [50].
The variability of absolute rates of soil flux during GS is rather site-specific (Figure 5), as previously reported by [5,35,44]. Finally, the variations in soil CO2 flux drop to the essentially lower rates by the third decade of September (19–20), a late season, despite the highest depth of soil thaw reached by that time (Figure 4). In late GS, presumably due to the conductive processes in the active soil layer [58], the uppermost soil already starts to freeze (Figure 3c,d and Figure 4), while thawing in the deeper soil layers reaches its maximum (Figure 4). This reduces the availability of organic matter in the upper soil layers for microbial degradation, which explains the significantly lower soil flux rates (Figure 7). Therefore, a greater amount of unfrozen soil achieved by the late season and potentially accessible organic matter content for microbial respiration is expected to be insufficient to promote higher flux rates if not coupled with positive upper soil temperatures. The limiting role of near-surface soil temperatures is expected, especially in the early and late GS periods, as previously reported [46]. However, a prolonged period of microbial activity that may persist into late October–November was also noted, as some soil flux might still be expected (Figure 4) until complete soil freezing towards the end of the “zero curtain” period [41,42], when soil temperature drops from values fluctuating around 0 °C to stable negative records. For instance, in the study area, we repeatedly detected some presumably biogenic signals of atmospheric CO2 in air masses travelling upwind from the surrounding tundra ecosystems towards the “DIAMIS” station in late autumn [37]. Eventually, after complete soil freezing, microbial activity decreases to meagre rates compared to GS values, and soil flux intensity decreases to lower values (Figure 7), which are continuously observed throughout the non-growing season and, in total, as noted by [59], may represent a rather significant source of CO2. However, our chamber measurements of mid-winter soil CO2 fluxes in 2024 in the study area (unpublished data) have so far not shown any prominent values, probably due to the persistent strong winds and thin snow cover (6–8 cm depth), with patchy exposed and completely frozen soil areas, where ST10 records are comparable with negative air temperatures.

3.3. Soil CO2 Fluxes and Precipitation: Simulating Immediate and Seasonal Patterns

Since soil CO2 flux rates are very sensitive to alterations in precipitation amounts, rapid growth to a peak value followed by an exponential rise can be observed periodically during rainfall events [43]. Though such sporadic pulse-like cases of rapidly increasing CO2 emission rates are more likely to occur in soils under seasonally dry climates (i.e., the “Birch” effect) [43,60], the response of Arctic environments under the projected transition from a snow-dominated to a rain-dominated precipitation regime [61] may also be a crucial issue. For instance, recent studies reported the increased rainfall-driven soil CO2 emissions in Siberian boreal forests [62], and further investigations made in Alaska [63] noted that deep soil temperatures were warmer in the years with high rainfall events compared to the years with average precipitation.
To simulate the behavior of soil flux after rainfall, we modelled rainfall events of different intensities for the well-drained and relatively dry LCCs across the studied vegetation types: barren ground, lichen, and moss vegetation. Figure 8 illustrates the temporal patterns of soil CO2 flux after simulated wetting events of 5 mm (moderate to heavy rainfall) and 10 mm (extreme rainfall).
The observed patterns illustrate the pulse-like effect on soil CO2 flux and a contrasting behavior between barren ground and vegetated plots after rainfall: immediate suppression of flux intensity at frost boils (Figure 8a,b) and sharp peaks of CO2 emissions within vegetated plots (Figure 8c–f). In particular, for lichen synusia, we found spikes up to 6.38 and 6.26 µmol m−2 s−1, demonstrating a four-fold increase of Fsoil for 5 and 10 mm of added precipitation, respectively. A greater increase of soil flux was observed on moss vegetation: 9.71 (5 mm) and 11.7 µmol m−2 s−1 (10 mm), an increase of 6 to 8 times for the peaks, respectively. In general, the values are consistent with recently reported rates of 5–11-fold increases in soil CO2 flux after a single rainfall in Siberian boreal ecosystems [62]. For both plots, observed spikes occurred within the first 3 to 5 min and were followed by an exponential growth, reaching a steady state after 60 to 90 min. Thus, an average additive soil flux over vegetated plots triggered by moderate to heavy (extreme) rainfall in the peak phase can reach up to 0.99 ± 0.48 µmol m−2 s−1 per 1 mm of precipitation, but this lasts only a few minutes. In contrast to vegetated plots, barren ground showed the opposite pattern: a 55–70% inhibition of CO2 flux observed for first 6 h after moderate rainfall (5 mm), which is expected to last longer (>7 h) when extreme rainfall events (10 mm) occur. Hence, part of the additive soil CO2 flux (0.07 ± 0.04 µmol m−2 s−1 per 1 mm) can be compensated for by patches of barren ground interspersed with plant communities across heterogeneous tundra landscapes. Although the rate is comparatively small, this inhibition effect is rather long-lasting and remains for hours before returning to pre-rainfall values.
Based on previously simulated seasonal patterns of soil CO2 fluxes (Figure 7) and continuous precipitation records across the GS, we calculated cumulative CO2 fluxes for the studied LCCs, particularly for barren ground, lichen, and moss vegetation over moss and frost-heaved tundra (Figure 9).
Our patterns illustrate the GS soil CO2 flux without and with rainfall-driven pulses taken into account in our estimates. Depending on the ecosystem, the difference reaches an average of 7–12% throughout the GS and ranges from 0.57 ± 0.39 to 0.40 ± 0.28 mol m−2 day−1 for moss and frost-heaved tundra, respectively. This could be attributed to essentially lower (~20%) precipitation across the GS over moss tundra (117 mm) compared to frost-heaved tundra (90.2 mm). In turn, moss vegetation showed the most pronounced effect of precipitation on soil flux within these VTs: 1.03 mol m−2 day−1 or 12% (moss tundra) and 0.73 mol m−2 day−1 or 8% (frost-heaved tundra). In general, applying our estimates, we can assume an average additive soil flux across the GS of up to 0.08 ± 0.06 and 0.07 ± 0.05 µmol m−2 s−1 for each 1 mm of precipitation for moss and frost-heaved tundra, respectively.
However, the in situ measured behavior of Fsoil after the simulated precipitation for barren ground (Figure 9) partly contradicts the model CO2 flux pattern across the GS (Figure 8) showing essentially lower additive values than other LCCs though still slightly increased due to precipitation. The possible reasons for the observed discrepancy are twofold: the substantially rounded GS model values of Fsoil, which may hide both the growth and inhibition of soil CO2 flux, and the simulation of rather high and simultaneous precipitation inputs compared to the much smoother natural precipitation impact on soil CO2 flux. Nevertheless, estimates of different scenarios of the precipitation-induced effect on Fsoil are of a great importance given the warmer and wetter climate projected in the Arctic. For bare ground this effect is even more complicated compared to the vegetated LCCs. As previously reported, e.g., [64], the climate-driven responsiveness of Arctic landscapes is substantially determined by the extent and frequency of disturbances, which form mineral-rich seedbeds with comparatively warm soils (ST10 = 6.9 ± 2.7 °C; this study) and low competition; these are immediately recruited by shrubs—the major component of Arctic “browning” [64], the counterpart to Arctic “greening” [14]. The expansion of shrubs and other vegetation towards competitively favorable frost-heaved circles may greatly modify soil CO2 flux rates for these non-vegetated LCCs. Based on the revealed estimates in this study, we may expect a substantial growth of CO2 flux for barren ground, which constitutes an essential part of Arctic landscapes (~10% based on [51], and up to 20% observed in our study area (Table 1)), from the lowest observed values across the Arctic landscapes towards the much higher rates along with shrub recruitment. Moreover, the projected increased precipitation may strengthen the additive effect on total CO2 release from the soil surface to the atmosphere during the GS in the Arctic.

4. Conclusions

The mean soil CO2 flux in the typical tundra landscapes of the southern Taimyr Peninsula, Siberia, at the peak of the growing season reaches 2.23 ± 1.26 µmol m−2 s−1, demonstrating an almost four-fold increase from non-vascular vegetation (1.05 ± 0.36 µmol m−2 s−1) towards vascular plants (3.59 ± 0.51 µmol m−2 s−1). Barren ground (so-called “frost boils”), which covers ~20% of the study area, shows the lowest mean CO2 flux of 0.79 ± 0.21 µmol m−2 s−1. However, the expected expansion of shrubs towards competitively favorable frost-heaved circles may lead to a substantial growth of CO2 flux across the entire tundra landscape.
The well-drained and relatively dry habitats across the studied land cover classes (barren ground and non-vascular vegetation (mosses, lichens)) are found to be the most temperature-sensitive land cover classes across tundra landscapes. For mixed vegetation and vascular plants that favored wetter conditions, soil moisture plays a greater role in modulating soil CO2 flux rates.
On the other hand, based on the simulated precipitation of 5 and 10 mm, we found spikes up to 6.38 and 6.26 µmol m−2 s−1, respectively, for lichen synusia, demonstrating a four-fold increase compared to the pre-rainfall values. An even greater increase of soil flux was observed for the moss vegetation: 9.71 (5 mm) and 11.7 µmol m−2 s−1 (10 mm), a six- to eight-fold increase, correspondingly. Hence, for vegetated plots (lichens, mosses), an average additive soil CO2 flux induced by moderate to heavy rainfall in the peak phase can reach up to 0.99 ± 0.48 µmol m−2 s−1 per 1 mm of precipitation, but the effect lasts only a few minutes. Barren ground shows the opposite pattern, with 55–70% inhibition of CO2 flux observed within the first few hours after rainfall, which can reduce the total soil efflux in tundra landscapes.
Our simulated seasonal patterns of soil CO2 flux for well-drained and relatively dry habitats assume an additive precipitation effect of 7–12% over the entire growing season, depending on the ecosystem. This implies an average additive soil flux across the growing season of 0.08 ± 0.06 and 0.07 ± 0.05 µmol m−2 s−1 per 1 mm of precipitation for moss and frost-heaved tundra, respectively. The projected increased precipitation may strengthen the additive effect on total CO2 release from the soil surface to the atmosphere during the growing season in the Arctic.
The reported study is a part of a long-term research program in the Central Siberian domain of the Arctic, complementing accurate continuous atmospheric observations of carbon dioxide and methane in the area, which have been operational at the “DIAMIS” station since 2018. Besides the scientific insights presented here, ground-based measurements across typical ecosystems in the footprint area of the observation station are intended to determine the links between variations in atmospheric signatures and carbon dynamics over the local tundra landscapes.

Author Contributions

Conceptualization, A.P. (Alexey Panov) and A.P. (Anatoly Prokushkin); methodology, A.P. (Alexey Panov), G.Z., and M.K.; software, A.P. (Alexey Panov) and I.P.; validation, A.P. (Alexey Panov), A.P. (Anatoly Prokushkin), and G.Z.; formal analysis, A.P. (Alexey Panov), I.P., and R.K.; investigation, A.P. (Alexey Panov), A.P. (Anatoly Prokushkin), I.P., and G.Z.; resources, M.B.; data curation, A.P. (Alexey Panov) and A.P. (Anatoly Prokushkin); writing—original draft preparation, A.P. (Alexey Panov); writing—review and editing, A.P. (Alexey Panov), A.P. (Anatoly Prokushkin), and G.Z.; visualization, A.P. (Alexey Panov) and I.P.; supervision, A.P. (Alexey Panov) and A.P. (Anatoly Prokushkin); funding acquisition, A.P. (Alexey Panov) and A.P. (Anatoly Prokushkin). All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the Russian Science Foundation, Krasnoyarsk Territory, and the Krasnoyarsk Regional Fund of Science, project #24-27-20064, https://rscf.ru/en/project/24-27-20064/ (accessed on 20 February 2024).

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

The data presented in this study are available on request from the corresponding author. The data are not publicly available due to privacy.

Acknowledgments

The authors would like to thank Igor Kornienko (the Great Arctic Reserve, Dikson) for his technical support regarding the measurements at the Willem Barentsz Biological Station. We also acknowledge the work of Alexander Tsukanov for his help in the field, technical suggestions, and improvements.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. IPCC: The Ocean and Cryosphere in a Changing Climate. Available online: https://www.ipcc.ch/srocc/home (accessed on 20 February 2024).
  2. Arctic Report Card: Update for 2020. Available online: https://arctic.noaa.gov/Report-Card/Report-Card-2020 (accessed on 20 February 2024).
  3. Coates, K.S.; Holroyd, C. The Palgrave Handbook of Arctic Policy and Politics; Palgrave Macmillan: Cham, Switzerland, 2020; p. 555. [Google Scholar]
  4. Overland, J.E.; Hanna, E.; Hanssen-Bauer, I.; Kim, S.-J.; Walsh, J.E.; Wang, M.; Bhatt, U.S.; Thoman, R.L.; Ballinger, T.J. Surface air temperature. In Arctic Report Card 2019; Richter-Menge, J., Druckenmiller, M.L., Jeffries, M., Eds.; NOAA: Silver Spring, MD, USA, 2019. Available online: https://www.arctic.noaa.gov/Report-Card (accessed on 20 February 2024).
  5. Magnani, M.; Baneschi, I.; Giamberini, M.; Raco, B.; Provenzale, A. Microscale drivers of summer CO2 fluxes in the Svalbard High Arctic tundra. Sci. Rep. 2022, 12, 763. [Google Scholar] [CrossRef]
  6. Schuur, E.A.G.; McGuire, A.D.; Schadel, C.; Grosse, G.; Harden, J.W.; Hayes, D.J.; Hugelius, G.; Koven, C.D.; Kuhry, P.; Lawrence, D.M.; et al. Climate change and the permafrost carbon feedback. Nature 2015, 520, 171–179. [Google Scholar] [CrossRef] [PubMed]
  7. Strauss, J.; Schirrmeister, L.; Grosse, G.; Fortier, D.; Hugelius, G.; Knoblauch, C.; Romanovsky, V.E.; Schädel, C.; Schneider von Deimling, T.; Schuur, E.A.G.; et al. Deep Yedoma permafrost: A synthesis of depositional characteristics and carbon vulnerability. Earth-Sci. Rev. 2017, 172, 75–86. [Google Scholar] [CrossRef]
  8. Jeong, S.J.; Bloom, A.A.; Schimel, D.; Sweeney, C.; Parazoo, N.C.; Medvigy, D.; Schaepman-Strub, G.; Zheng, C.; Schwalm, C.R.; Huntzinger, D.N.; et al. Accelerating rates of Arctic carbon cycling revealed by long-term atmospheric CO2 measurements. Sci. Adv. 2018, 4, eaao1167. [Google Scholar] [CrossRef] [PubMed]
  9. Feng, J.; Wang, C.; Lei, J.; Yang, Y.; Yan, Q.; Zhou, X.; Tao, X.; Ning, D.; Yuan, M.M.; Qin, Y.; et al. Warming-induced permafrost thaw exacerbates tundra soil carbon decomposition mediated by microbial community. Microbiome 2020, 8, 3. [Google Scholar] [CrossRef]
  10. Kusch, S.; Rethemeyer, J.; Ransby, D.; Mollenhauer, G. Permafrost organic carbon turnover and export into a High-Arctic fjord: A case study from Svalbard using compound-specific 14C analysis. J. Geophys. Res. Biogeosci. 2021, 126, e2020JG006008. [Google Scholar] [CrossRef]
  11. Mudryk, L.; Brown, R.; Derksen, C.; Luojus, K.; Decharme, B.; Helfrich, S. Terrestrial snow cover. In Arctic Report Card 2019; Richter-Menge, J., Druckenmiller, M.L., Jeffries, M., Eds.; NOAA: Silver Spring, MD, USA, 2019. Available online: https://www.arctic.noaa.gov/Report-Card (accessed on 20 February 2024).
  12. Mernild, S.H.; Malmros, J.K.; Yde, J.C.; Knudsen, N.T. Multi-decadal marine- and land-terminating glacier recession in the Ammassalik region, southeast Greenland. Cryosphere 2012, 6, 625–639. [Google Scholar] [CrossRef]
  13. Overeem, I.; Syvitski, J.P.M. Shifting discharge peaks in arctic rivers, 1977–2007. Geogr. Ann. Ser. A Phys. Geogr. 2010, 92, 285–296. [Google Scholar] [CrossRef]
  14. Berner, L.T.; Massey, R.; Jantz, P.; Forbes, B.C.; Macias-Fauria, M.; Myers-Smith, I.; Kumpula, T.; Gauthier, G.; Andreu-Hayles, L.; Gaglioti, B.V.; et al. Summer warming explains widespread but not uniform greening in the Arctic tundra biome. Nat. Commun. 2020, 11, 4621. [Google Scholar] [CrossRef]
  15. Hudson, J.M.G.; Henry, G.H.R.; Cornwell, W.K. Taller and larger: Shifts in Arctic tundra leaf traits after 16 years of experimental warming. Glob. Change Biol. 2011, 17, 1013–1021. [Google Scholar] [CrossRef]
  16. Leffler, A.J.; Klein, E.S.; Oberbauer, S.F.; Welker, J.M. Coupled long-term summer warming and deeper snow alters species composition and stimulates gross primary productivity in tussock tundra. Oecologia 2016, 181, 287–297. [Google Scholar] [CrossRef] [PubMed]
  17. Abbott, B.W.; Jones, J.B.; Schuur, E.A.G.; Chapin, F.S., III; Bowden, W.B.; Bret-Harte, M.S.; Epstein, H.E.; Flannigan, M.D.; Harms, T.K.; Hollingsworth, T.N.; et al. Biomass offsets little or none of permafrost carbon release from soils, streams, and wildfire: An expert assessment. Environ. Res. Lett. 2016, 11, 034014. [Google Scholar] [CrossRef]
  18. Andresen, C.G.; Lougheed, V.L. Arctic aquatic graminoid tundra responses to nutrient availability. Biogeosciences 2021, 18, 2649–2662. [Google Scholar] [CrossRef]
  19. Schuur, T. Permafrost and the global carbon cycle. In Arctic Report Card 2019; Richter-Menge, J., Druckenmiller, M.L., Jeffries, M., Eds.; NOAA: Silver Spring, MD, USA, 2019. Available online: https://www.arctic.noaa.gov/Report-Card (accessed on 20 February 2024).
  20. Lund, M.; Falk, J.M.; Friborg, T.; Mbufong, H.N.; Sigsgaard, C.; Søgaard, H.; Tamstorf, M.P. Trends in CO2 exchange in a high Arctic tundra heath, 2000–2010. J. Geophys. Res. 2012, 117, G02001. [Google Scholar] [CrossRef]
  21. Hickman, J. Carbon sinks and sinking tundra. Nat. Geosci. 2014, 7, 784. [Google Scholar] [CrossRef]
  22. Euskirchen, E.S.; Bret-Harte, M.S.; Shaver, G.R.; Edgar, C.W.; Romanovsky, V.E. Long-term release of carbon dioxide from Arctic tundra ecosystems in Alaska. Ecosystems 2017, 20, 960–974. [Google Scholar] [CrossRef]
  23. Commane, R.; Lindaas, J.; Benmergui, J.; Luus, K.A.; Chang, R.Y.-W.; Daube, B.C.; Euskirchen, E.S.; Henderson, J.M.; Karion, A.; Miller, J.B.; et al. Carbon dioxide sources from Alaska driven by increasing early winter respiration from Arctic tundra. Proc. Natl. Acad. Sci. USA 2017, 114, 5361–5366. [Google Scholar] [CrossRef] [PubMed]
  24. Lüers, J.; Westermann, S.; Piel, K.; Boike, J. Annual CO2 budget and seasonal CO2 exchange signals at a high Arctic permafrost site on Spitsbergen, Svalbard archipelago. Biosciences 2014, 16, 6307–6322. [Google Scholar] [CrossRef]
  25. Euskirchen, E.S.; Bret-Harte, M.S.; Scott, G.J.; Edgar, C.; Shaver, G.R. Seasonal patterns of carbon dioxide and water fluxes in three representative tundra ecosystems in northern Alaska. Ecosphere 2012, 3, 1–19. [Google Scholar] [CrossRef]
  26. Belshe, E.F.; Schuur, E.A.; Bolker, B.M. Tundra ecosystems observed to be CO2 sources due to differential amplification of the carbon cycle. Ecol. Lett. 2013, 16, 1307–1315. [Google Scholar] [CrossRef]
  27. Virkkala, A.-M.; Virtanen, T.; Lehtonen, A.; Rinne, J.; Luoto, M. The current state of CO2 flux chamber studies in the Arctic tundra: A review. Prog. Phys. Geog. 2018, 42, 162–184. [Google Scholar] [CrossRef]
  28. Fletcher, B.J.; Gornall, J.L.; Poyatos, R.; Press, M.C.; Stoy, P.C.; Huntley, B.; Baxter, R.; Phoenix, G.K. Photosynthesis and productivity in heterogeneous arctic tundra: Consequences for ecosystem function of mixing vegetation types at stand edges. J. Ecol. 2012, 100, 441–451. [Google Scholar] [CrossRef]
  29. Virtanen, T.; Ek, M. The fragmented nature of tundra landscape. Int. J. Appl. Earth Obs. Geoinf. 2014, 27, 4–12. [Google Scholar] [CrossRef]
  30. Aalto, J.; le Roux, P.C.; Luoto, M. Vegetation mediates soil temperature and moisture in Arctic-Alpine environments. AAAR 2013, 45, 429–439. Available online: http://www.jstor.org/stable/24551600 (accessed on 4 March 2024). [CrossRef]
  31. Post, E.; Forchhammer, M.C.; Bret-Harte, M.S.; Callaghan, T.V.; Christensen, T.R.; Elberling, B.; Fox, A.D.; Gilg, O.; Hik, D.S.; Høye, T.T.; et al. Ecological dynamics across the Arctic associated with recent climate change. Science 2009, 325, 1355–1358. [Google Scholar] [CrossRef] [PubMed]
  32. Virkkala, A.-M.; Natali, S.M.; Rogers, B.M.; Watts, J.D.; Savage, K.; Connon, S.J.; Mauritz, M.; Schuur, E.A.G.; Peter, D.; Minions, C.; et al. The ABCflux database: Arctic–boreal CO2 flux observations and ancillary information aggregated to monthly time steps across terrestrial ecosystems. Earth Syst. Sci. Data 2022, 14, 179–208. [Google Scholar] [CrossRef]
  33. Sachs, T.; Giebels, M.; Boike, J.; Kutzbach, L. Environmental controls on CH4 emission from polygonal tundra on the microsite scale in the Lena River delta, Siberia. Glob. Change Biol. 2010, 16, 3096–3110. [Google Scholar] [CrossRef]
  34. Göckede, M.; Kwon, M.J.; Kittler, F.; Heimann, M.; Zimov, N.; Zimov, S. Negative feedback processes following drainage slow down permafrost degradation. Glob. Change Biol. 2019, 25, 3254–3266. [Google Scholar] [CrossRef] [PubMed]
  35. Juutinen, S.; Aurela, M.; Tuovinen, J.-P.; Ivakhov, V.; Linkosalmi, M.; Rasanen, A.; Virtanen, T.; Mikola, J.; Nyman, J.; Vaha, E.; et al. Variation in CO2 and CH4 fluxes among land cover types in heterogeneous Arctic tundra in northeastern Siberia. Biogeosciences 2022, 19, 3151–3167. [Google Scholar] [CrossRef]
  36. Panov, A.; Prokushkin, A.; Kübler, K.R.; Korets, M.; Urban, A.; Bondar, M.; Heimann, M. Continuous CO2 and CH4 observations in the coastal arctic atmosphere of the western Taimyr peninsula, Siberia: The first results from a new measurement station in Dikson. Atmosphere 2021, 12, 876. [Google Scholar] [CrossRef]
  37. Panov, A.; Prokushkin, A.; Semiletov, I.; Kübler, K.; Korets, M.; Putilin, I.; Urban, A.; Bondar, M.; Heimann, M. Atmospheric CO2 and CH4 fluctuations over the continent-sea interface in the Yenisei River sector of the Kara Sea. Atmosphere 2022, 13, 1402. [Google Scholar] [CrossRef]
  38. Tulp, I.; Bruinzeel, L.; Jukema, J.; Stepanova, O. Breeding waders at Medusa Bay, Western Taimyr, in 1996. In WIWO Report 57; WIWO: Zeist, The Netherlands, 1997; Volume 57, pp. 7–8. [Google Scholar]
  39. McKnight, T.L.; Hess, D. Climate Zones and Types: The Köppen System, Physical Geography: A Landscape Appreciation; Prentice Hall: Hoboken, NJ, USA, 2000; pp. 235–237. [Google Scholar]
  40. Walker, D.A.; Raynolds, M.K.; Daniëls, F.J.A.; Einarsson, E.; Elvebakk, A.; Gould, W.A.; Katenin, A.E.; Kholod, S.S.; Markon, C.J.; Melnikov, E.S.; et al. The Circumpolar Arctic vegetation map. J. Veg. Sci. 2005, 16, 267–282. [Google Scholar] [CrossRef]
  41. Jenkins, G. A comparison between two types of widely used weather stations. Weather 2014, 69, 105–110. [Google Scholar] [CrossRef]
  42. Schimel, J.; Mikan, C. Changing microbial substrate use in Arctic tundra soils through a freeze-thaw cycle. Soil Biol. Biochem. 2005, 37, 1411–1418. [Google Scholar] [CrossRef]
  43. Meyer, N.; Welp, G.; Amelung, W. The temperature sensitivity (Q10) of soil respiration: Controlling factors and spatial prediction at regional scale based on environmental soil classes. Glob. Biogeochem. Cycles 2018, 32, 306–323. [Google Scholar] [CrossRef]
  44. Zamolodchikov, D.G.; Karelin, D.V.; Ivaschenko, A.I.; Oechel, W.C.; Hastings, S.J. CO2 flux measurements in Russian Far East tundra using eddy covariance and closed chamber techniques. Tellus B Chem. Phys. Meteorol. 2003, 55, 879–892. [Google Scholar] [CrossRef]
  45. Elberling, B. Annual soil CO2 effluxes in the High Arctic: The role of snow thickness and vegetation type. Soil Biol. Biochem. 2007, 39, 646–654. [Google Scholar] [CrossRef]
  46. Cannone, N.; Ponti, S.; Christiansen, H.; Christensen, T.R.; Pirk, N.; Guglielmin, M. Effects of active layer seasonal dynamics and plant phenology on CO2 land-atmosphere fluxes at polygonal tundra in the High Arctic, Svalbard. Catena 2019, 174, 142–153. [Google Scholar] [CrossRef]
  47. Seppälä, O. Spatial and Temporal Drivers of Soil Respiration in a Tundra Environment. Master’s Thesis, University of Helsinki, Helsinki, Finland, 2020. Available online: http://hdl.handle.net/10138/313261 (accessed on 4 March 2024).
  48. Bradley-Cook, J.I.; Virginia, R.A. Landscape variation in soil carbon stocks and respiration in an Arctic tundra ecosystem, west Greenland. AAAR. 2018, 50, S100024. [Google Scholar] [CrossRef]
  49. Wagner, J.; Hung, J.; Neil, A.; Scott, N. Net greenhouse gas fluxes from three High Arctic plant communities along a moisture gradient. Arct. Sci. 2019, 5, 185–201. [Google Scholar] [CrossRef]
  50. Watts, J.; Natali, S.; Minions, C.; Risk, D.; Arndt, K.; Zona, D.; Euskirchen, E.; Rocha, A.; Sonnentag, O.; Helbig, M.; et al. Soil respiration strongly offsets carbon uptake in Alaska and Northwest Canada. Environ. Res. Lett. 2021, 16, 084051. [Google Scholar] [CrossRef]
  51. Virkkala, A.-M.; Niittynen, P.; Kemppinen, J.; Marushchak, M.; Voigt, C.; Hensgens, G.; Kerttula, J.; Happonen, K.; Tyystjärvi, V.; Biasi, C.; et al. High-resolution spatial patterns and drivers of terrestrial ecosystem carbon dioxide, methane, and nitrous oxide fluxes in the tundra. Biogeosciences 2024, 21, 335–355. [Google Scholar] [CrossRef]
  52. Heitman, J.L.; Horton, R. Coupled heat and water transfer in soil. In Encyclopedia of Agrophysics; Gliński, J., Horabik, J., Lipiec, J., Eds.; Encyclopedia of Earth Sciences Series; Springer: Dordrecht, The Netherlands, 2011. [Google Scholar] [CrossRef]
  53. Wickland, K.; Jorgenson, M.; Koch, J.; Kanevskiy, M.; Striegl, R. Carbon dioxide and methane flux in a dynamic Arctic tundra landscape: Decadal-scale impacts of ice wedge degradation and stabilization. Geophys. Res. Lett. 2020, 47, e2020GL089894. [Google Scholar] [CrossRef]
  54. Johnston, A.S.A.; Sibly, R.M. The influence of soil communities on the temperature sensitivity of soil respiration. Nat. Ecol. Evol. 2018, 2, 1597–1602. [Google Scholar] [CrossRef]
  55. Zona, D.; Lipson, D.; Paw, U.K.T.; Oberbauer, S.; Olivas, P.; Gioli, B.; Oechel, W. Increased CO2 loss from vegetated drained lake tundra ecosystems due to flooding. Glob. Biogeochem. Cycles 2012, 26, GB2004. [Google Scholar] [CrossRef]
  56. Hashemi, J.; Zona, D.; Arndt, K.; Kalhori, A.; Oechel, W. Seasonality buffers carbon budget variability across heterogeneous landscapes in Alaskan Arctic Tundra. Environ. Res. Lett. 2021, 16, 035008. [Google Scholar] [CrossRef]
  57. Arndt, K.; Lipson, D.; Hashemi, J.; Oechel, W.; Zona, D. Snow melt stimulates ecosystem respiration in Arctic ecosystems. Glob. Change Biol. 2020, 26, 5042–5051. [Google Scholar] [CrossRef] [PubMed]
  58. Watanabe, T.; Matsuoka, N.; Christiansen, H.H.; Cable, S. Soil physical and environmental conditions controlling patterned-ground variability in a continuous permafrost site, Svalbard. Permafr. Periglac. Process. 2017, 28, 433–445. [Google Scholar] [CrossRef]
  59. Natali, S.M.; Watts, J.D.; Rogers, B.M.; Potter, S.; Ludwig, S.M.; Selbmann, A.K.; Sullivan, P.F.; Abbott, B.W.; Arndt, K.A.; Birch, L.; et al. Large loss of CO2 in winter observed across the northern permafrost region. Nat. Clim. Change 2019, 9, 852–857. [Google Scholar] [CrossRef]
  60. Singh, S.; Mayes, M.A.; Kivlin, S.N.; Jagadamma, S. How the Birch effect differs in mechanisms and magnitudes due to soil texture. Soil Biol. Biochem. 2023, 179, 108973. [Google Scholar] [CrossRef]
  61. McCrystall, M.R.; Stroeve, J.; Serreze, M.; Forbes, B.C.; Screen, J.A. New climate models reveal faster and larger increases in Arctic precipitation than previously projected. Nat. Commun. 2021, 12, 6765. [Google Scholar] [CrossRef] [PubMed]
  62. Makhnykina, A.; Vaganov, E.A.; Panov, A.; Koshurnikova, N.; Prokushkin, A. The pulses of soil CO2 emission in response to rainfall events in Central Siberia: Revisiting the overall frost-free season CO2 flux. Forests 2024, 15, 355. [Google Scholar] [CrossRef]
  63. Neumann, R.; Moorberg, C.; Lundquist, J.; Turner, J.; Waldrop, M.; McFarland, J.; Euskirchen, E.; Edgar, C.; Turetsky, M. Warming efects of spring rainfall increase methane emissions from thawing permafrost. Geophys. Res. Lett. 2019, 46, 1393–1401. [Google Scholar] [CrossRef]
  64. Frost, G.; Epstein, H.; Walker, D.; Matyshak, G.; Ermokhina, K. Patterned-ground facilitates shrub expansion in Low Arctic tundra. Environ. Res. Lett. 2013, 8, 015035. [Google Scholar] [CrossRef]
Atmosphere 15 00698 g001
Figure 1. Study area on the southwestern coast of the Taimyr Peninsula, located within the Willem Barentsz Biological Station, with continuously running weather stations mounted on research plots in moss (1), frost-heaved (2), and marsh tundra (3).
Figure 1. Study area on the southwestern coast of the Taimyr Peninsula, located within the Willem Barentsz Biological Station, with continuously running weather stations mounted on research plots in moss (1), frost-heaved (2), and marsh tundra (3).
Atmosphere 15 00698 g001
Atmosphere 15 00698 g002aAtmosphere 15 00698 g002b
Figure 2. UAV-based images of the representative sites with research plots in moss (a), frost-heaved (b), and marsh (c) tundra, located within the Willem Barentsz Biological Station. Stars mark the location of weather stations for continuous meteorological observations.
Figure 2. UAV-based images of the representative sites with research plots in moss (a), frost-heaved (b), and marsh (c) tundra, located within the Willem Barentsz Biological Station. Stars mark the location of weather stations for continuous meteorological observations.
Atmosphere 15 00698 g002aAtmosphere 15 00698 g002b
Atmosphere 15 00698 g003
Figure 3. Seasonal patterns of air temperature at 200 cm height (a,b) and soil temperature (c,d) and moisture (e,f) at 10 cm depth for moss and frost-heaved tundra, correspondingly, for September 2021–September 2022. For air and soil temperatures (a–d) values above 0 °C—positive (red), below 0 °C—negative (blue).
Figure 3. Seasonal patterns of air temperature at 200 cm height (a,b) and soil temperature (c,d) and moisture (e,f) at 10 cm depth for moss and frost-heaved tundra, correspondingly, for September 2021–September 2022. For air and soil temperatures (a–d) values above 0 °C—positive (red), below 0 °C—negative (blue).
Atmosphere 15 00698 g003
Atmosphere 15 00698 g004
Figure 4. Seasonal gradients of soil temperature (ST) across top 100 cm soil layer for moss (a) and frost-heaved tundra (b) for September 2021–September 2022.
Figure 4. Seasonal gradients of soil temperature (ST) across top 100 cm soil layer for moss (a) and frost-heaved tundra (b) for September 2021–September 2022.
Atmosphere 15 00698 g004
Atmosphere 15 00698 g005
Figure 5. Patterns of soil CO2 flux rates (a,d) and soil temperature at 10 cm depth (b,e) for frost-heaved, moss, and marsh tundra and different land cover classes within these vegetation types, where 1—barren ground, 2—lichen, 3—moss, 4—moss and herbs, 5—moss and sedge, and 6—sedge. Scatter plots of soil CO2 efflux rates vs. soil temperature at 10 cm depth (c,f).
Figure 5. Patterns of soil CO2 flux rates (a,d) and soil temperature at 10 cm depth (b,e) for frost-heaved, moss, and marsh tundra and different land cover classes within these vegetation types, where 1—barren ground, 2—lichen, 3—moss, 4—moss and herbs, 5—moss and sedge, and 6—sedge. Scatter plots of soil CO2 efflux rates vs. soil temperature at 10 cm depth (c,f).
Atmosphere 15 00698 g005
Atmosphere 15 00698 g006
Figure 6. Dependency of soil CO2 flux rates and soil temperature at 10 cm depth for barren ground (a) and different vegetation groups (b–f) across the frost-heaved, moss, and marsh tundra. Corresponding exponential curves and Q10 coefficients are presented.
Figure 6. Dependency of soil CO2 flux rates and soil temperature at 10 cm depth for barren ground (a) and different vegetation groups (b–f) across the frost-heaved, moss, and marsh tundra. Corresponding exponential curves and Q10 coefficients are presented.
Atmosphere 15 00698 g006
Atmosphere 15 00698 g007
Figure 7. Temporal patterns of soil CO2 flux, modelled for growing season (JJASON) based on the Fsoil–ST10 relationships revealed for barren ground and different vegetation groups within the moss (a,c,e,g) and frost-heaved (b,d,f,h) tundra.
Figure 7. Temporal patterns of soil CO2 flux, modelled for growing season (JJASON) based on the Fsoil–ST10 relationships revealed for barren ground and different vegetation groups within the moss (a,c,e,g) and frost-heaved (b,d,f,h) tundra.
Atmosphere 15 00698 g007
Atmosphere 15 00698 g008
Figure 8. Temporal patterns of Fsoil behavior after simulated rainfall event of 5 mm (moderate to heavy) and 10 mm (extreme) for barren ground (a,b), lichen (c,d), and moss (e,f). Dashed red line indicates the pre-rainfall value of Fsoil.
Figure 8. Temporal patterns of Fsoil behavior after simulated rainfall event of 5 mm (moderate to heavy) and 10 mm (extreme) for barren ground (a,b), lichen (c,d), and moss (e,f). Dashed red line indicates the pre-rainfall value of Fsoil.
Atmosphere 15 00698 g008
Atmosphere 15 00698 g009
Figure 9. Simulated patterns of cumulative soil CO2 fluxes across the GS (2021–2022), with rainfalls included in estimations (blue curves) for moss (a) and frost-heaved tundra (b), where (1) moss, (2) lichen, (3) barren ground. Red arrows demonstrate an additive effect of precipitation to soil CO2 fluxes.
Figure 9. Simulated patterns of cumulative soil CO2 fluxes across the GS (2021–2022), with rainfalls included in estimations (blue curves) for moss (a) and frost-heaved tundra (b), where (1) moss, (2) lichen, (3) barren ground. Red arrows demonstrate an additive effect of precipitation to soil CO2 fluxes.
Atmosphere 15 00698 g009
Table 1. Dominant land cover classes of the selected sites across the study area.
Table 1. Dominant land cover classes of the selected sites across the study area.
Site 1Site 2Site 3
Land Cover ClassArea, m2%Area, m2%Area, m2%
1. Ground and rocks81659.112,33211.099909.0
2. Moss/Bare ground11,34212.615,70614.122,86020.6
3. Moss28,14731.237,05233.239,24135.3
4. Moss/Herbs28,61431.829,28426.211,93810.8
5. Sedge998611.111,86710.612,41911.2
6. Wetland38364.353724.814,60113.1
90,090100111,614100111,049100
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Panov, A.; Prokushkin, A.; Korets, M.; Putilin, I.; Zrazhevskaya, G.; Kolosov, R.; Bondar, M. Variation in Soil CO2 Fluxes across Land Cover Mosaic in Typical Tundra of the Taimyr Peninsula, Siberia. Atmosphere 2024, 15, 698. https://doi.org/10.3390/atmos15060698

AMA Style

Panov A, Prokushkin A, Korets M, Putilin I, Zrazhevskaya G, Kolosov R, Bondar M. Variation in Soil CO2 Fluxes across Land Cover Mosaic in Typical Tundra of the Taimyr Peninsula, Siberia. Atmosphere. 2024; 15(6):698. https://doi.org/10.3390/atmos15060698

Chicago/Turabian Style

Panov, Alexey, Anatoly Prokushkin, Mikhail Korets, Ilya Putilin, Galina Zrazhevskaya, Roman Kolosov, and Mikhail Bondar. 2024. "Variation in Soil CO2 Fluxes across Land Cover Mosaic in Typical Tundra of the Taimyr Peninsula, Siberia" Atmosphere 15, no. 6: 698. https://doi.org/10.3390/atmos15060698

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop