Thermal state of ice-rich soils on the Tommot-Yakutsk Railroad right-of-way

Pavel Skryabin

Melnikov Permafrost Institute,Siberian Branch of the Russian Academy of Sciences,Yakutsk 677010,Russia

1 Introduction

Assessing the evolution of upper permafrost in response to climate change and infrastructure development is a high-priority research issue.Environmental disturbance associated with various types and scales of human activity (forest clearing,fires,embankments,etc.) leads to changes in permafrost conditions and triggers adverse geocryological processes.Permafrost research in support of the construction and operation of the Tommot-Yakutsk Railroad involves assessment of the ground thermal state along the railroad corridor and in the adjacent areas.The town of Tommot (population 8,000) and the city of Yakutsk(population 270,000) are located in eastern Siberia where the average daily high temperature in winter(October to April) is below-25 °C and the coldest temperature in January averages-42 °C.Yakutsk is approximately 450 km south of the Arctic Circle,and the railroad under construction will connect with Tommot 390 km farther south.

In 2007,the general contractor Proekttranstroy Company decided that geocryological and geotechnical monitoring was necessary during the railroad construction and operation stages within the section traversing ice-rich permafrost terrain (Pozinet al.,2007).The Melnikov Permafrost Institute in Yakutsk has been conducting an integrated research program along the railroad route between the stations of Olen and Kirim since 2005 (Figure 1).The study area is located on the Abalakh terrace of the Lena River in inter-alas terrain underlain by ice-rich permafrost with massive ice bodies occurring close to the surface.Engineering-geocryological investigations showed that this route segment would pose the most difficult construction,operation,and maintenance problems (e.g.,Skryabinet al.,2005;Varlamov,2006;Varlamovet al.,2006).

Figure 1 Research area map.1:geocryological stations,2:monitoring sites,3:railway

2 Materials and methods

This paper discusses the results of investigations carried out from 2007 to 2013 under public contracts with the Republic of Sakha (Yakutia).The railroad right-of-way (ROW) was cleared of trees during the winters of 2006 to 2008,and the embankment was constructed from April 2009 through September 2010.The study reported here involves long-term field observations using the method of terrain thermal physics.Eight transects were established at kilometer posts (KPs) across the ROW and adjacent undisturbed terrain with a total of 32 boreholes.Ground temperature monitoring continues in different permafrost terrain units:inter-alas terrain (KP 708.8,708.9,717.5,and 717.9),drainage depressions (KP 692.6 and 708.7),and sloping terrain (KP 693.2 and 693.4).

Our studies focus on the upper 10–15 m of permafrost within the depth of annual temperature variation.Field observations include the factors controlling upper permafrost temperatures (snow depth and density,soil texture,structure,and moisture content),thaw depth,ground temperature,and permafrost-related processes.

The commonly accepted indicators of changes in the thermal state of near-surface permafrost are active layer thickness (ξ) and mean annual ground temperature at 10-m depth (t).Ground temperatures are measured in backfilled boreholes with MMT-4 semiconductor thermistors.Thermistor cables record temperatures at depths of 0.3,1.0,2.5,5.0,7.5,and 10.0 m.Active layer depths are determined at the end of the thaw season (mid-September) by probing with a steel rod and a hand drill.Observations are made four times annually.

3 Results and discussion

Our understanding of heat transfer in the ground-land cover-atmosphere system stems from previous long-term studies conducted at the Yakutsk,Syrdakh,and Chabyda monitoring stations (Pavlov,1979;Skryabinet al.,1998).The relationships obtained help quantify the effects of air temperature,vegetation,snow,organic mat,and soil moisture and composition on the thermal state of permafrost.

The recent climate warming,which began in Central Yakutia in the second half of the 1960s,has been the strongest over the last 30 years.Mean annual air temperatures (averaged over a hydrologic year from October through September) recorded at the Yakutsk weather station were lower than normal only for two years since 1981.The 2007–2013 observation period was warmer than normal air temperatures,one winter with anomalously low snowfall,and five winters with higher than normal snowfall.

Long-term data from the Brolog personal weather station (Varlamovet al.,2002) indicate that the maximum 10-day snow depth is on average 43 cm,which is 1.5 times greater than at the Yakutsk weather station.Within the railroad route segment underlain by ice-rich permafrost,snow density increases from 120 to 230 kg/m3over the winter and averages 160 kg/m3.This thick,loose snow cover exerts an insulating effect on permafrost during seven months a year.

Data from eight transects established in different permafrost settings have been analyzed to estimate changes in the ground thermal state on the railroad ROW and adjacent undisturbed terrain.For illustration,the ground thermal regime within the layer of annual temperature fluctuations and its controls are discussed for the KP 693.4 transect.The study site at KP 693.4 is located on a relatively steep,east-facing slope.In 2007 three boreholes (B-13,B-14,and B-16)were established across the embankment.In 2009 boreholes were drilled in a larch forest stand to the left of the service road (B-17) and on the ROW immediately to the left of the embankment (B-15).

The ground thermal regime in the forest stand is affected by tree shading and snow accumulation.The canopy intercepts a portion of solar insolation and reduces the soil surface temperature and thaw depth.In summer,the larch forest is estimated to depress the surface temperatures by 4.3 °C,on average,in the inter-alas terrain type (Varlamovet al.,2002).

Maximum snow depths under the forest canopy vary over a wide range,from 41 cm in a low-snow winter to 61 cm in a high-snow winter.Because of weak winds,snow cover in the forest stands is very loose and has a low thermal conductivity,thus protecting the ground from strong cooling.Our investigations conducted on the east side of the Lena River have shown that the insulating effect of snow cover on soil surface temperatures in the inter-alas terrain type ranges from 2.5 °C to 15.4 °C,or 1.7–9.4 °C for every 10 cm of snow depth (Varlamovet al.,2002).

The ground thermal regime is controlled to a large extent by the thermal effects of surface covers during the winter and summer seasons.A moss/mountaincranberry cover has an insulating effect of 1.2–2.9 °C in winter and a cooling effect of 4.1–6.6 °C in summer,depressing the ground temperatures by 0.3–1.0 °C over an annual cycle (Varlamovet al.,2002).Field studies have shown that the moss/mountain-cranberry cover,5 to 15 cm in thickness,in the larch stand has a high moisture content(42%–62%) and a large thermal resistance.

Freezing and thawing of the active layer are strongly dependent on the soil moisture regime.Soil moisture and soil texture determine changes in the thermal properties of seasonally thawing sediments.The mean moisture content of the active layer is 24%–29%,and that of the permafrost is 27%–132%.Core drilling showed that ice wedges,2.8 to 4.2 m in thickness,occur very close to the surface,with their tops lying at 1.2 to 2.1 m.

Ground temperatures at 10-m depth range from-2.3 °C to-2.6 °C in the end of the warm season (Figures 2 and 3).Interannual variation int0is 0.1–0.3 °C,suggesting that its response to current variations in meteorological conditions is insignificant.

The active layer thickness in larch stands on both sides off-ROW was within 0.72–1.00 m over seven years of observations.The difference between maximum and minimumξvalues on the left and right sides was 0.23 and 0.11 m,respectively (Figure 4).Thus,the main factors controllingξare the shading effect of the trees and the insulating effect of the moss/mountain-cranberry cover.

Figure 2 Ground temperature variation in larch stand (B-17)and on-ROW (B-15),KP 693.4

Figure 3 Interannual ground temperature variation in larch stand (B-17) and on-ROW (B-15),KP 693.4

Figure 4 Active layer thickness variation in undisturbed(forest) and disturbed (on-ROW along the embankment)setting,KP 693.4

Clearing of the ROW is done mainly during the winter,with the surface covers left nearly intact.The resulting increases in incoming solar radiation,surface radiation balance,and ground heat flux lead to warmer permafrost temperatures compared to the forest sites.Clearing of the larch forests in inter-alas terrain in the northern section of the Tommot-Yakutsk Railroad caused an increase int0by 0.3–0.8 °C and inξby 0.4–0.8 m (Varlamov and Skryabin,2013).

The maximum snow depth on the ROW varied inter-annually within 42 to 61 cm,which was very similar to the forest site.The mean moisture content of the surface cover varied from 23% to 197%,while that of the active layer was 27% to 68%,or significantly lower than in the undisturbed terrain.These data indicate increased evaporation from the ground surface cleared of tree vegetation.

Ground temperature at the 10-m depth beneath the ROW to the left of the embankment increased above that in the undisturbed terrain by 0.3–0.8 °C (see figures 2 and 3).On the right side of the ROW,tree clearing without surface cover disturbance resulted in an increase inξby 0.2–0.4 m.On the left side of the ROW,the surface cover was disturbed by skidding and destroyed by burning.This resulted in a 1.7-to 2.5-fold increase inξ.Active layer depths on the ROW steadily increased from year to year.In the fifth year,ξwas 1.9 m on the left side of the ROW,resulting in thawing of the ice wedges lying at 1.2–2.1 m from the surface (see figure 4).Thaw settlement is evident here,threatening the embankment stability.

Based on the analysis of permafrost monitoring data from other sites along the railroad,changes in the main thermal parameters on the ROW have been estimated (Table 1).The deepest snow cover on the ROW,64–65 cm,develops in the drainage depressions due to drifting (KP 692.6 and 708.7).Clearing of forests in inter-alas areas (KP 708.8 and 708.9) reduces the average moisture content of the active layer to 27%–42%.The largest increase int0,by 0.7–1.2 °C,is observed on the ROW along the slopes (KP 693.4) and on the water divides in inter-alas areas (KP 708.8 and 708.9).The active layer is the deepest,1.9–2.3 m,in the drainage depression (KP 708.7) and in the inter-alas site (KP 708.8).The increases int0andξare accompanied by thaw settlement in the ice-rich terrain,which cannot be tolerated in railroad operation.

Table 1 Basic parameters of the ground thermal state

4 Conclusions

The results of the investigations can be summarized as follows:

1) The period of observations was characterized by warmer than normal mean annual air temperatures and higher than normal snowfall.However,the ground thermal regime in undisturbed areas showed little response to these variations in meteorological conditions.

2) Forest clearing and surface disturbance by ROW operations cause an increase in permafrost temperature,deepening of the active layer,thaw settlement,and water accumulation along the embankment.The active layer is the thickest along the sun-exposed left berm and the thinnest along the more shaded right berm.

3) Soil thaw depths on the ROW show a distinct increase from year to year,in places reaching the top of ice wedges and causing permafrost degradation.

4) To maintain embankment stability,it is recommended to place additional passive cooling devices,to establish more temperature boreholes,and to provide continuous drainage of water from the base of the embankment (KP 693.4,708.7,708.8,708.9,and 717.9).

5) Biological reclamation is recommended to restore the disturbed surface covers on the ROW.

The author is grateful to the Republic of Sakha(Yakutia) Government for financial support of field work during the period from 2005 to 2013 under Public Contracts 251,588,and 1090.The author would also like to thank the editors of SCAR for their constructive comments.

Pavlov AV,1979.Thermal Physics of Landscapes.Nauka,Novosibirsk,Russia.

Pozin VA,Korolev AA,Naumov MS (eds.),2007.Proceedings of the Workshop on Structural Reliability of the Tommot-Kerdem Railway in the Ice-Complex Area,11–12 September 2007,Yakutsk,Russia.Center Transstroiizdat,Moscow.

Skryabin PN,Varlamov SP,Skachkov YB,1998.Interannual Variability of the Ground Thermal Regime in the Yakutsk Area.SB RAS Press,Novosibirsk.

Skryabin PN,Varlamov SP,Skachkov YB,2005.Geothermal monitoring network between Olen and Nizhniy Bestyakh stations.In:Pozin VA,Korolev AA,Naumov MS (eds.).Design and Construction of the Tommot-Kerdem Railway Roadbed in Complicated Permafrost Engineering Conditions.Results of Geotechnical Investigations in 2005,Proceedings of the Scientific and Technical Council Meeting,7–8 December 2005,Yakutsk,Russia.Proekttransstroy,Moscow,pp.30–33.

Varlamov SP,2006.Ground ice contents in the northern section of the Tommot-Kerdem railway project (Olen station to Kerdem station).In:Earth Cryosphere Assessment:Theory,Applications and Prognosis of Alterations.Proceedings of International Conference,29–31 May 2006,Tyumen,Russia.Tyumen Petroleum University Press,Tyumen,(2),pp.212–214.

Varlamov SP,Skachkov YB,Skryabin PN,2002.Ground Temperature Regime in Permafrost Landscapes of Central Yakutia.Permafrost Institute,SB RAS Press,Yakutsk.

Varlamov SP,Skryabin PN,2013.Anthropogenic Impacts on the Ground Thermal Regime in Central Yakutia.LAP Lambert Academic Publishing,Saarbrücken,Germany.

Varlamov SP,Skryabin PN,Skachkov YB,et al.,2006.Geocryological conditions in icy permafrost terrain along the proposed Tommot-Kerdem Railway.Nauka i Obrazovanie,4(44):78–83.