Temperature monitoring in Bakchar bog (West Siberia)

Russian Geology and Geophysics 50 (2009) 579–586 www.elsevier.com/locate/rgg

Temperature monitoring in Bakchar bog (West Siberia) E.A. Dyukarev a,*, E.A. Golovatskaya a, A.D. Duchkov b, S.A. Kazantsev b a b

Institute of Monitoring of Climatic and Ecological Systems, Siberian Branch of the RAS, 10/3 prosp. Akademicheskii, 634055, Tomsk, Russia Trofimuk Institute of Petroleum Geology and Geophysics, Siberian Branch of the RAS, 3 prosp. Akad. Koptyuga, Novosibirsk, 630090, Russia Received 10 June 2008; accepted 27 August 2008

Abstract We report the results of continuous temperature monitoring (812 days, from 28 June 2005 to 26 September 2007) in a 80 cm layer of peat soil in Bakchar bog (West Siberia), at sampling rates of 60 min in wintertime and 15 min in summertime. Both annual and daily temperature patterns are controlled by water table position and weather conditions. Wintertime soil temperature patterns are disturbed by the formation of a seasonal frozen layer with its thickness (freezing depth) depending on the time when steady snow cover sets up and on soil moisture. During the period of frozen layer thawing, the temperature of peat becomes sensitive to peat moisture and water table position as well as to the air and peat surface temperature. The warm-season soil temperature patterns bear effects of peat warming by rainwater percolation, both in nightand daytime. The patterns with soil warming during rainfall and phase change during seasonal freezing-thawing cycles record disturbances to conductive heat transfer. © 2009, IGM, Siberian Branch of the RAS. Published by Elsevier B.V. All rights reserved. Keywords: soil temperature patterns; autonomous digital temperature recorder; peatland soil; seasonal frozen layer; rainwater percolation; Bakchar bog

Introduction Soil temperature is critical for many biotic and abiotic soil processes, such as plant growth and productivity, soil organic matter decomposition and mineralization, emission of greenhouse gases (Golovatskaya et al., 2008; Moore and Dalva, 1993; Vomperskii, 1994), release of dissolved organic carbon (Prokushkin and Guggenberger, 2007), etc. Soil temperature patterns are controlled by current climate, physiography and Earth’s orbit, as well as by air-soil thermal interaction driven by geobotanic and geomorphologic factors (Pavlov, 1979). Below-zero air temperatures in the cold season produce a seasonal frozen layer in soils, which causes the respective changes to the thermal regime. The frozen layer increases the surface runoff during snow melting and rainfall, and prevents moisture from percolating down the soil profile. Peat (organic) and mineral soils have different temperature patterns. Peat soil is a complex organic-mineral system with high water content and porosity, and with large amounts of underdecomposed organic matter (Romanov, 1961). Peatlands are the main source of greenhouse gases (CO2, CH4, and * Corresponding author. E-mail address: [email protected] (E.A. Dyukarev)

N2O). Increase in these gases, which are produced by temperature-dependent microbial metabolism (Glagolev et al., 2008; Golovatskaya et al., 2008; Lafleur et al., 2005), has been a largely discussed but still poorly understood cause of the ongoing climate change (Kabanov, 2000; Semenov, 2004). On the other hand, warming of soil associated with climate warming increases emission of greenhouse gases. Methane and carbon dioxide production continues in wintertime as well, and their emission occurs through the seasonal frozen layer and snow (Pannikov and Dedysh, 2000). Soil temperature monitoring in peatlands is important because the current climate warming is especially rapid in northern areas (Houghton, 2001; Ippolitov et al., 2007), the main pool of peat (and carbon). Peatland ecosystems have been estimated to store 120 to 455 billion tons carbon (Gorham, 1991; Vomperskii, 1994). The resources of peat carbon in Russia amount to 215 billion tons (Botch et al., 1995), of which up to 70 billion tons are sequestered in peatlands of West Siberia (Sheng et al., 2004). Climate change or human activity may cause this great amount of carbon to partly release into the atmosphere as CO2 or CH4 and thus to change the atmospheric carbon budget. In addition to the climate change implications, high-resolution monitoring of soil temperatures can provide clues to the

1068-7971/$ - see front matter D 2009, IGM, Siberian Branch of the RAS. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.rgg.2008.08.010

580

E.A. Dyukarev et al. / Russian Geology and Geophysics 50 (2009) 579–586

complex patterns of heat transfer from near-surface air into rocks. Below we discuss the instruments, methods, and results of soil temperature monitoring at a site of the Vasyugan mire in southern West Siberia. Preliminary results of the study were reported in (Dyukarev et al., 2006).

Background and instruments Air and soil temperatures were measured in an oligotrophic pine–dwarf shrub–sphagnum bog (low ryam) at the Bakchar bog site (Bakchar district, Tomsk region). The low ryam microtopography is made up of 30 to 50 cm high flat moss hummocks occupying about 70% of the area and narrow (up to 2 m) hollows between them. The local vegetation belongs to a pine-shrub-sphagnum community with low Pinus silvestris f. litwinowii in the suppressed tree layer. The tree layer consists of 2–3 m high and ∼3 cm thick trees, with a projective cover of 30%. The shrub layer, abundant in hummocks, with a total projective cover of 60–70%, is composed of Chamaedaphne calyculata Linnaeus, Andromeda polifolia and Vaccinium uliginosum, with Ledum palustre L. and Oxycoccus microcarpus on hummocks. The grass layer has a projective cover less than 5 % and consists of Eriophorum vaginatum L., Rubus chamaemorus L. and Drosera rotundifolia clumps. The moss layer includes common Sphagnum fuscum Klinggr. on hummocks (95%) and Sph. angustifolium and Sph. magellanicum in hollows (Golovatskaya and Porokhina, 2005). The low-ryam peat deposit is 2 m thick and has a mixed structure with low peat below high peat (up to 1.5 m thick) of medium (magellanicum) and low (fuscum) degrees of decomposition, separated by a thin transitional layer of woody-moss peat (Kabanov, 2003). Air temperatures (Ta) were measured by an Onset corporation HOBO Water Level Logger (USA) at every 15 min. Soil temperatures (Ts) were sampled at eight depth levels below the surface by an autonomous digital temperature recorder designed at the Trofimuk Institute of Petroleum Geology and Geophysics, Novosibirsk (Duchkov et al., 2005; Kazantsev and Duchkov, 1992). See Fig. 1 for the instrument layout and Table 1 for technical specifications. The digital recorder (Fig. 1) consists of a 16-bit analog-digital converter (ADC) and a microcontroller (single-crystal micro-computer). The microcontroller controls the ADC operation as well as switch of sensors and data storage and exchange with the external PC. Incoming signals from temperature sensors (thermistors) enter ADC through a switch and become stored as a digital code in a permanent flash storage for up to 20,000 eight-channel sessions. The basic ATR modification has eight measurement and two calibration channels. The recorder is started by a special integral timer (DS 1305). The sampling interval is programmed and can range from tens of seconds to tens of hours. After reading, preprocessing, and saving all data, the microcontroller switches the station to the standby mode. Drain is about 20 mA in the operating mode and about

Fig. 1. Autonomous digital temperature recorder: Functional diagram.

50 µA in the standby mode. Power supply to the recorder is from a 3 V battery. Data are read on a PC connected to the recorder through a standard RS232 COM port. Memory fill is programmed; stored data are read and saved as table files. Before starting the recorder, the user specifies the sampling interval and cleans the memory if necessary. Once the PC is off, the recorder takes up the operating mode automatically. The MMT-6 10,000 Ohm thermistors were calibrated against 0.01 °C mercury thermometers on a special laboratory stand and, correspondingly, the absolute temperatures were measured to a precision of 0.01–0.02 °C. According to our Table 1 Specifications of autonomous digital temperature recorder Parameter

Specification

Number of thermistors ADC, bit Temperature precision Temperature sensitivity Operation temperatures Memory

1 to 16 (basic layout) 16 to 0.02 °C to 0.002 °C from –20 to +60 °C permanent flash, up to 20,000 eightchannel sessions 3–6 V Li-ion battery USB or RS232 port up to 12 months ∅ = 40 mm, L = 200 mm; 1 kg

Power Read-out Autonomous run period size; weight (in a water-proof container)


experience (Duchkov et al., 2005), the precision for relative temperature variations was 0.002–0.003 °C. The temperature recorder was set into soil using a special probe consisting of a small (400 mm in diameter and 200 mm long) tight container with a 1.0 m long steel supporting rod and a steel tube mounted on its bottom. The tube housed eight thermistors at different distances from the container. The probe was placed on a small hummock with its elevation corresponding to the average low-ryam surface, and the thermistors measured soil temperatures at 2, 5, 10, 15, 25, 40, 60, and 80 cm below the surface. The site was of hummock-hollow microtopography, with up to 20 cm high hummocks and 30 cm deep hollows (relative to the average surface elevation). The temperatures were monitored continuously for 812 days (from 28 June 2005 to 26 September 2007), at sampling rates of 60 min in wintertime and 15 min in summertime.

Results Daily means. See Fig. 2 for daily temperature patterns of air and soil at different depths below the peat surface. The two cold seasons of the experiment differed strongly in weather conditions. The winter of 2005–2006 was extremely cold, with the lowest daily mean air temperature of –44.1 °C

581

on 13 January 2006. The snow cover was the thickest (56 cm) on 4 April 2006 and had went off completely by 26 April 2006. The following winter (2006–2007) was, on the contrary, unusually warm, with air temperatures almost never below –25 °C, except a single event of –31.3 °C on 22 February 2007. The snow cover persisted from 21 November 2006 to 11 April 2007 and was the thickest (74 cm) on 19 March 2007. The warm seasons were about the climate norm in 2005 and 2007 and very hot in 2006, with the highest mean daily temperature of 27.5 °C (on 23 June 2006); the summer of 2007 was much wetter (416 mm of moisture) than the long-term average (285 mm), which caused abnormally high water table position during the whole warm season. The measured soil temperature patterns at different depths generally followed the annual trend of air temperatures (Fig. 2). The warm-season patterns recorded the transfer of heat from the sun-heated soil surface down to deeper layers, with high-frequency temperature oscillations decaying with depth. The peat temperature oscillations become damped with depth due to high heat insulation from the porous moss layer from above and to high heat capacity of bog water standing 10–20 cm below the surface. The diurnal temperature variations can propagate to depths of 15–25 cm in peat soils and

Fig. 2. Daily mean temperature patterns. Ta is air temperature, Ts is soil temperature at depths 2 to 80 cm (T2, T5, T10, T15, T25, T40, T60, T80), SDP is snow depth (cm), FD is soil freezing depth (cm).

582


much deeper (to 60–80 cm) in mineral soils (Pavlov, 1979). Although being quite thin, the layer of variable daily temperatures is important for plant and microbial activity. The yearly temperature wave commonly decreases in amplitude and shifts in phase with depth. For instance, the amplitude of the annual air temperature variations (estimated as the difference of daily means) at the monitoring site was 64.6 °C while those for soil were 26.4, 19.1, 8.6, and 5.9 °C at depths of 2, 10, 40, and 80 cm, respectively. The mean air temperature reached the maximum on 11 July on average while the soil peaks lagged to 14, 25 July, 12 August, and 6 September at 2, 10, 40, and 80 cm below the surface, respectively. The 10 °C isotherm penetrated the deepest in the first half of August (to 46, 16, and 66 cm in 2005, 2006, and 2007, respectively). In the dry and hot summer of 2006, peat remained cooler despite the hot air temperatures because the overlying moss layer was dry and thus less heat conductive. On the contrary, abundant moisture in the wet summer of 2007 maintained a high water table position and a high moisture in the upper peat layer, which provided a higher thermal conductivity and, hence, warmer soil temperatures. At a depth of 80 cm, soil warmed up gradually during the warm season to reach the maximum in the end of September (7.9, 4.8, and 9.2 °C in 2005, 2006, and 2007, respectively). In the warm seasons (see logs 1, 3 and 5 in Fig. 3), the temperature decreased with depth and the temperature gradient became negative, i.e., heat transport was down from the warm surface soil. The gradient in the upper 25 cm of peat was much higher in its absolute value (–0.17...–0.22 °C/cm) than in the soil below (–0.01...–0.06 °C/cm). In the summer of 2006, soil was 1–6 °C colder than in 2005 and 2007. The upper 25 cm of soil began cooling since the mid-summer, and the gradient changed to positive in latest September when the daily mean air temperature dropped to below 8–10 °C. The gradient became stably positive at all sampled depths in the beginning of October, and then the whole soil layer set into cooling. In the cold seasons (logs 2 and 4 in Fig. 3), soil grew warmer with depth, with the gradient decreasing downward from 0.16 °C/cm at 2–5 cm to 0.02 °C/cm at 60–80 cm below the surface. Note that the deep thermal gradient in this area is about 100 times lower than that (0.02−0.03)⋅10−2 °C/cm) (Fotiadi, 1987).

Fig. 3. Temperature logs, averaged over different time intervals: 1, 28.06.2005– 30.09.2005; 2, 01.10.2005–29.03.2006; 3, 01.04.2006–30.09.2006; 4, 01.10.2006–29.03.2007; 5, 01.04.2007–30.09.2007.

patterns (Pavlov, 1979). The reason is that the frozen layer, and the respective phase change, isolates the unfrozen ground below from any external forcing during the cold season, and the soil-air heat exchange resumes only after the frozen layer has thawn completely. The freezing front defined by the depth of the 0 °C isotherm moved, at a mean rate of 12.7 mm/day, till 12 December 2005, when soil had frozen to a depth of 8 cm, and then the freezing rate slowed down to 3 mm/day; the minimum temperature at 2 cm below the surface was –4.7 °C on 24 February 2006 Table 2 Freezing-thawing rate of seasonal frozen layer (SFL) Winter

Period

2005–2006

SFL base 06.12.2005–12.12.2005 13.12.2005–27.02.2006 28.02.2006–06.06.2006 07.06.2006–13.06.2006* SFL top 26.04.2006–05.05.2006 06.05.2006–13.06.2006 SFL base 14.11.2006–30.11.2006 01.12.2006–20.03.2007 21.03.2007–29.04.2007* SFL top 10.04.2007–14.04.2007 15.04.2007–29.04.2007

Freezing-thawing dynamics of peat soil In the cold season of 2005–2006, the temperature patterns became notably smoother (20 times) at all sampled depths after the stable snow cover had set in (on 29 October 2005), and the peat deposit began to cool down slowly. The highly insulating snow cover and the heat stored in wet deep peat layers during the warm season held the soil temperature positive as long as 6 December 2005. The warming effect of snow acted all the cold season long until the thawing period. Heat exchange between soil and air in areas of seasonally frozen ground is controlled primarily by summer weather

2006–2007

* SFL base moved upward (thawing from below).

Rate, mm/day

12.7 3 0.3 7.8 14.3 2.6 22.4 0.6 1.1 33.2 10.6


583

Fig. 4. Daily temperature patterns of soil (Ts) at depths 2, 5, 10, 15 cm and air (Ta) during thawing of seasonal frozen layer in spring 2007. Data points on curves are for 12:00 of each day. Soil temperatures above 1 °C are in log-log scale.

(Table 2). After the peat surface had freed from snow by 26 April 2006, the frozen layer began to degrade rapidly. The upper 15 cm of peat thawed by 5 May 2006, but the highly insulating dry moss layer above kept the ground frozen below that depth as long as mid-June. The frozen layer reached 32.6 cm deep on 6 June 2006 and set into rapid thawing (at 7.8 mm/day) only after 7 June. The cold season of 2006 began with repeated snowfall-melt events when thin snow stayed no more than two or three days while the daily mean air temperature dropped to –9 °C. The air temperature changes and the lack of snow caused rapid cooling of the peat deposit which froze to a depth of 15 cm, then thawed down and froze up again. By the time when the snow cover became stable (21 November 2006), peat froze as deep as 25 cm, and the freezing depth changed slowly during the winter season, with the maximum (39.4 cm) recorded on 20 March 2007. The top 2 cm of soil cooled to –10 °C on 30 November 2006 when the daily mean air temperature dropped to –23.6 °C and snow was thin. The soil thawing pattern traceable in the continuous Ta and Ts data was as follows. In the spring of 2007, the mean daily air temperatures became stably positive after 1 April 2007, but the snow cover still persisted and the soil temperatures remained below zero. From 8 to 10 April, the upper 15 cm of peat stayed at 0 °C while the air temperature in the daytime rose to 20 °C (Fig. 4.). It was the time of rapid snow melting on the surface accompanied by heat absorption (cooling), which provided the so-called zero curtain effect (Outcalt et al., 1990). After 11 April, the topmost 2 cm of peat warmed up to above zero in the daytime, the snow cover had degraded by that date, and soil began to thaw. It thawed during the day

and warmed up to 1–3 °C but cooled down to 0 °C in the nighttime from April 14 to 16. Between 19 and 26 April, the upper peat layer became as warm as 8–10 °C due to warm daytime air temperatures, and the frozen layer was degrading rapidly, though there were brief freezing events to a small depth in the nighttime. The frozen layer had thawed completely by 29 April 2007, and the soil temperature became positive from that time on. The zero curtain effect in the upper peat layer occurred also in the spring of 2006, when soil at depths of 2, 5, 10, and 15 cm remained at 0 °C from 19 to 27 April.

Warming of peat soil during percolation of rainwater Soil is a multi-phase capillary porous system with both conductive and convective (water transport) heat transfer (Chudnovskii, 1976). In thermophysics soil is commonly assumed to be a quasi-uniform body in which effective heat transfer parameters allow for all possible transport mechanisms. However, this simplification may be far from reality, for instance, in the case of rainwater percolation through soil. The soil temperature patterns of summer seasons we obtained through high-density (5–15 min) monitoring included a number of moisture percolation events. See Fig. 5, a for the example typical Ta and Ts daily patterns in the upper soil layer on 30 July 2007. The air temperature rose from 9.2 °C at 4:00 to 29.5 °C at 16:00, warming being especially rapid between 6:30 and 10:30. It remained more or less stable in the daytime and dropped to 17 °C from 18:00 to 22:00. The soil temperature at depths from 2 to 15 cm was the lowest at

584


Fig. 5. Temperature patterns of soil (Ts) at depths 2, 5, 10, 15, 25, 40 cm and air (Ta). Sampling at every 15 min, data points at 60 min. a, 30.07.2007: normal daily pattern; b, 02.07.2006: soil warming by rainwater percolation.

6–7 a.m., and the highest at 16:10, 17:00, 18:20, 19:30, and 21:30 for the depths 2, 5, 10, 15, and 25 cm, respectively. The soil temperature pattern is generally conformable with the air trend but looks smoother and shifted (the shift corresponds to the lag of the thermal wave front moving through soil). This is a typical warm-season temperature behavior in the absence of rainfall when peat warms up uniquely by conduction. Percolation of rainwater through peat commonly disturbed the air and soil temperature patterns in different ways: air cooled down abruptly while soil warmed up. One such event was recorded on 2 July 2006 (Fig. 5, b). The air temperature at the site was 29 °C at noon but dropped to 18–19 °C when rain began at 13:00. Some time later (at 14:00), the percolating rainwater made soil warmer. The warming was the most intense at depths of 10 and 15 cm (8.2 °C in 15 min at 10 cm), less intense at 2 and 5 cm, and the lowest at 25 cm (3.1 °C for several hours). The topmost soil (2 and 5 cm) did not warm up much because percolating rainwater normally carries both its own heat and the heat it takes off the topsoil on its

way down. This soil warming mechanism is common to peatland ecosystems with a highly porous moss surface and poorly decomposed peat (Romanov, 1961). The temperature of the upper 15 cm of moistened peat balanced out in an hour and became equal to the air temperature; then soil cooled down slowly to temperatures slightly higher than before the rainfall. The wet peat below the water table (at 40 cm) kept the temperature invariable due to its high thermal inertia. In the course of the experiment, from 2005 through 2007, we recorded 13 events of soil warming by rainwater (Table 3). The soil temperature rose for 3 °C on average at 10–15 cm below the surface and for 0.02 to 2.1 °C at 25 cm, in 15–30 min. The topmost peat cooled down (for 0.3–2.1 °C) or sometimes warmed up (for 5.6 °C), mostly in the nighttime when the soil surface was colder than air. The vertical temperature gradient corresponding to heat flux into soil (Lykov, 1978) became negative during rainfall as soil warmed up. See Fig. 6 for gradient patterns at different periods of the rainfall event on 30 June 2007 which illustrate (together with Fig. 5, b) the soil temperature dynamics. Before

Table 3 Soil temperature changes at different depths during rainwater percolation (dt is step size) Date

02.07.2005 20.07.2005 30.07.2005 02.08.2005 04.08.2005 13.06.2006 25.06.2006 02.07.2006 17.06.2007 23.06.2007 24.06.2007 30.06.2007 22.07.2007

Time

15:30 21:15 15:15 01:00 18:15 01:30 00:45 14:00 19:15 14:35 21:35 15:10 17:55

dt, min

30 30 45 15 30 45 15 30 15 10 15 15 10

Temperature change, °C 2 cm

5 cm

10 cm

15 cm

25 cm

–1.57 0.87 –1.11 1.92 –1.37 1.08 –2.1 2.45 –1.9 –0.33 –1.34 –1.03 –0.29

1.84 1.69 2.33 2.38 1.39 1.11 –0.23 5.61 –0.83 0.9 –0.55 0.78 0.06

5.53 2.87 6.1 3.33 4.62 1.07 2.28 9.47 1.58 2.87 1.13 1.84 0.29

4.59 2.88 4.88 2.74 3.5 0.62 1.68 8.64 3.63 4.04 2.78 0.78 0.14

0.25 0.26 0.21 0.13 0.09 0.02 0.13 0.53 2.09 2.12 0.97 0.02 0.02


Fig. 6. Soil temperature gradient in different periods of rainfall on 30.06.2007. Gradient axis has a break from –0.75 to 1. Time: 1, 14:00; 2, 15:15; 3, 15:20; 4, 15:25; 5, 15:35; 6, 15:40; 7, 16:00.

rain (curve 1 in Fig. 6), the gradient was the lowest at 2–5 cm below the surface (–1.1 °C/cm), grew to –0.1 °C/cm with depth, and remained almost invariable below the bog water table (between 40 and 80 cm). When it began raining, the gradient increased rapidly in the topmost soil (curves 2–4 in Fig. 6) but decreased to –0.7 °C/cm at 10 and 15 cm, where most of heat dissipated. In 20 min (curves 5 and 6 in Fig. 6), the warming front moved deeper to 15–25 cm; the gradient in the top layer increased slightly 45 min after the rain began (curve 7 in Fig. 6), and then started to return to its original state. The porous upper soil layer not necessarily warmed up in every rainfall event, but only when peat was already warm and rain was heavy enough to maintain fast gravity-driven flow of moisture and respective heat transport.

Conclusions The results of the reported continuous temperature monitoring in a 1 m thick layer of peat soil run for 812 days (from 28 June 2005 to 26 September 2007), at sampling rates of 60 min in wintertime and 15 min in summertime can be summarized as follows. The annual temperature amplitude changed depthward from 26.4 °C in the top layer to 5.9 °C at a depth of 80 cm. The daily temperature changes affected peat to a depth of 15–25 cm. Both annual and diurnal temperature patterns turned out to be controlled primarily by water table position and weather. The cold-season patterns were influenced by the presence of a seasonal frozen layer in the peat

585

soil. The freezing depth depended on the time when steady snow cover set in and on peat moisture. Soil kept positive temperatures as long as mid-December after the snow cover became stable but froze up rapidly when snow was thin or absent while the air was cold (winter of 2006–2007). The maximum freezing depths in 2006 and 2007 were 32.6 and 39.4 cm, respectively. In 2006 the frozen layer persisted to mid-June. When it began to thaw, the soil temperature pattern became largely controlled by the peat moisture and the water table position, besides the effects from air and peat surface. The warm-season patterns showed effects of soil warming by rainwater percolation, both in night- and daytime. We never read reports of such events from peatland ecosystems in the literature, possibly, because these events are relatively rare and the long-term continuous temperature monitoring in peatlands is not a common practice. The temperature patterns disturbed by soil warming during rainfall and by phase change during seasonal freezing-thawing cycles record disturbances to conductive heat transfer. The reported temperature monitoring results have general and specific implications. The former are, for instance, the matters of quantitative analysis of heat transport and interaction of solar and terrestrial heat fluxes, and the latter is peat soil thermophysics. In particular, we used them to estimate effective thermal diffusivity of peat in the summer season of 2005 which varied from 4.1 to 6.7 cm2/s (Dyukarev et al., 2006). The study was partly supported by grant 08-05-00426-a from the Russian Foundation for Basic Research and was carried out as part of Integration Project 27 of the Siberian Branch of the Russian Academy of Sciences.

References Botch, M.S., Kobak, K.I., Vinson, T.S., Kolchugina, T.P., 1995. Carbon pools and accumulation in peatlands of the former Soviet Union. Global Biogeochemical Cycles 9 (1), 37–46. Chudnovskii, A.F., 1976. Thermophysics of Soils [in Russian]. Nauka, Moscow. Duchkov, A.D., Kazantsev, S.A., Duchkov, A.A., 2005. Experimental study of bottom temperature variations in the deep-water part of the South Baikal basin. Dokl. Earth Sci. 403A (6), 943. Dyukarev, E.A., Golovatskaya, E.A., Duchkov, A.D., Kazantsev, S.A., 2006. The thermal regime of the active layer during the warm season: An experimental study, in: Environment Monitoring and Rehabilitation. Proc. V Intern. Symp. Tomsk, pp. 101–103. Fotiadi, E.E. (Ed.), 1987. The Lithospheric Thermal Field of Siberia [in Russian]. Nauka, Novosibirsk. Glagolev, M.V., Golovatskaya, E.A., Shnyrev, N.A., 2008. Greenhouse gas emission in West Siberia. Contemporary Problems of Ecology 1 (1), 136–146. Golovatskaya, E.A., Porokhina, E.V., 2005. Botany and fundamentals of plant ecology, in: Dyrin, V.A. (Ed.), Biological Productivity of Peatland Ecosystems. Learner’s Book [in Russian]. Tomsk University, Tomsk. Golovatskaya, E.A., Dyukarev, E.A., Ippolitov, I.I., Kabanov, M.V., 2008. Influence of landscape and hydrometeorological conditions on CO2 emission in peatland ecosystems. Dokl. Earth Sci. 418 (1), 187–190. Gorham, E., 1991. Northern peatlands: role in the carbon cycle and probable responses to climatic warming. Ecol. Applications, No. 1, 182–195. Houghton, J.T., Ding, Y., Griggs, D.J. (Eds.), 2001. Climate Change 2001: The Scientific Basis. Contribution of Working Group I to the Third

586


Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge. Ippolitov, I.I., Kabanov, M.V., Loginov, S.V., 2007. Spatiotemporal scales of warming observed in Siberia. Dokl. Earth Sci. 413 (2), 248–252. Kabanov, M.V. (Ed.), 2000. Great Vasyugan Mire. Current State and Evolution [in Russian]. IOA SO RAN, Tomsk. Kabanov, M.V. (Ed.), 2003. Vasyugan Mire: Environment, Structure, and Functioning [in Russian]. 2nd edition. TsNTI, Tomsk. Kazantsev, S.A., Duchkov, A.D., 1992. An autonomous temperature recorder, in: Kononov, V.I., Yudakhin, F.N., Svalova, V.B. (Eds.), Geothermics of Seismic and Aseismic Regions [in Russian]. Nauka, Moscow, pp. 365– 373. Lafleur, P.M., Moore, T.R., Roulet, N.T., Frolking, S., 2005. Ecosystem respiration in a cool temperate bog depends on peat temperature but not water table. Ecosystems 8, 619–629. Lykov, A.V., 1978. Heat and Mass Exchange. A Handbook [in Russian]. Energiya, Moscow. Moore, T.R., Dalva, M., 1993. The influence of temperature and water table position on carbon dioxide and methane emissions from laboratory columns of peatland soils. J. Soil Science 44, 651–664. Outcalt, S.I., Nelson, F.E., Hinkel, K.M., 1990. The zero-curtain effect: heat

and mass transfer across an isothermal region in freezing soil. Water Resources Research 26 (7), 1509–1516. Pannikov, N.S., Dedysh, S.N., 2000. Cold season CH4 and CO2 emission from boreal peat bogs (West Siberia): Winter fluxes and thaw activation dynamics. Global Biogeochemical Cycles 14, 1071–1080. Pavlov, A.V., 1979. Thermophysics of Landscapes [in Russian]. Nauka, Novosibirsk. Prokushkin, A.S., Guggenberger, G., 2007. Climate control of dissolved organic matter transport from permafrost drainage basins in Central Siberia. Meteorologiya i Gidrologiya 6, 93–105. Romanov, V.V., 1961. Hydrophysics of Peatlands [in Russian]. GIMIZ, Leningrad. Semenov, S.M., 2004. Greenhouse Gases and the Earth’s Modern Climate [in Russian]. Meteorologiya i Gidrologiya, Moscow. Sheng, Y., Smith, L.C., MacDonald, G.M., Kremenetski, K.V., Frey, K.E., Velichko, A.A., Lee, M., Beilman, D.W., Dubinin, P., 2004. A high-resolution GIS-based inventory of the west Siberian peat carbon pool. Global Biogeochemical Cycles 18, GB3004. Vomperskii, S.E., 1994. Role of peatlands in the carbon cycle, in: Shilov, I.A. (Ed.), Peatland Ecosystems and their Sustainable Use [in Russian]. Nauka, Moscow, pp. 5–37.

Editorial responsibility: M.I. Epov