1 Ecological Monographs, 79(4), 2009, pp Ó 2009 by the Ecological Society of America Sensitivity of the carbon cycle in the Arctic to climate c...
1 Chapter 10 Marine Life in the Arctic Rolf Gradinger 1, Bodil A. Bluhm 1, Russell R. Hopcroft 1, Andrey V. Gebruk 2, Ksenia Kosobokova 2, Boris Siren...
1 Celebrating a century at the University of Alaska Fairbanks2 Old Main, the Alaska Agricultural College and School of Mines first building, sits on T...
1 THE SUNDAY OF THE PASSION: PALM SUNDAY March 29, 2014 Year B & Daily Office Year Two (ALL SERVICES BEGIN in the PARISH HALL) THE LITURGY of the ...
1 Residential Lease This agreement is between thereinafter referred to as the TENANT, and Nature Cabins LLC thereinafter referred to as the LANDLORD, ...
1 THEMED ISSUE: Origin and Evolution of the Sierra Nevada and Walker Lane GEOSPHERE GEOSPHERE; v. 12, no. 2 doi: /ges figures; 2 tables; 1 supplementa...
1 Fairbanks, Alaska Solar & Weather Information Factsheet EEM Narrative Climatological Summary Fairbanks is located in the Tanana Valley of Interi...
1 University of Alaska Fairbanks Fall 2018 Physics 175x Introduction to Astronomy 4 Credits Instructor Dr. Mark Conde Breaking news in Astronomy: One ...
1 FAIRBANKS CITY COUNCIL AGENDA NO REGULAR MEETING FEBRUARY 24, 2014 FAIRBANKS CITY COUNCIL CHAMBERS 800 CUSHMAN STREET, FAIRBANKS, ALASKA PRELIMINARY...
1 You are receiving this because you have expressed an interest in Osher Lifelong Learning Institute at University of Alaska Fairbanks, or someone wh...
Permafrost Dynamics at the Fairbanks Permafrost Experimental Station Near Fairbanks, Alaska T.A. Douglas Cold Regions Research and Engineering Laboratory Fairbanks, AK M. Torre Jorgenson Alaska Biological Resources, Fairbanks, AK M.Z. Kanevskiy, V.E. Romanovsky, Y. Shur, K. Yoshikawa University of Alaska Fairbanks, Fairbanks, AK
Abstract The Fairbanks Permafrost Experimental Station was established in 1945 near Fairbanks, Alaska. In 1946 vegetation was removed from two plots (the Linell plots) to investigate the impacts of vegetation disturbance on permafrost degradation. We revisited the sites in 2007 to evaluate the permafrost table using probes and direct current electrical resistivity. The permafrost table has expanded downward to 9.8 m at a site where all surface vegetation and organic material was removed. The permafrost surface has remained at 4.7 m depth since 1972 at a second site where vegetation was removed but organic material was left intact. In 2005 a Circumpolar Active Layer Monitoring Network (CALM) site was established at an undisturbed plot nearby to provide a baseline assessment of the permafrost. The response of permafrost at the site to the hypothesized future climatic warming of the Alaskan Interior can be assessed, once a long-term record is available. Keywords: boreal forest; electrical resistivity; monitoring; permafrost degradation; vegetation disturbance.
Introduction Permafrost in thermal equilibrium can be drastically affected by disturbances such as clearing vegetation (Linell 1973, Nicholas & Hinkel 1996), forest fires (Viereck 1982, Burn 1998, Hinzman et al. 2003), or climatic change (Shur & Jorgenson 2007). The Fairbanks Permafrost Experiment Station near Fairbanks, Alaska (64.877°N, 147.670°W) was established by the U.S. Army in 1945 as a location where geotechnical, geophysical, and engineering studies could be performed on permafrost. The site has a rich history and was designated as a National Geotechnical Experimentation Site in 2003. In 1946 vegetation was removed from two plots at the site (the Linell plots) to investigate the impacts of vegetation disturbance on permafrost degradation. This paper summarizes the results of recent investigations focused on measuring the current state of permafrost at the site. Our results provide information on permafrost degradation where vegetation disturbance has lead to a change in the thermal regime either through engineering activities, fire, or climatic change. Permafrost at the study site is considered “warm” (the mean annual air temperature is -3.3°C) and mean annual temperatures in the area are increasing. As a consequence, future degradation of permafrost during climate warming can be assessed once a long-term record is available from the site.
(Linell 1973). Herein these plots are referred to as the Linell plots. One of the Linell plots was left undisturbed to preserve the subarctic taiga forest with dense white and black spruce. Vegetation was removed from the other two plots. One plot was stripped of trees by hand but the roots and organic mat were left intact (Fig. 1) while at the other plot all of the vegetation and surface organic material were removed (Fig. 2). Linell (1973) described permafrost degradation at the two disturbed sites 26 years later. In the completely disturbed site the thaw depth reached 6.7 m after 26 years while at the partially disturbed site the melting expanded to a depth of
Results Linell plots and Circumpolar Active Layer Monitoring site In 1946 three square plots 61 m on a side (3721 m2) were identified at the Fairbanks Permafrost Experiment Station to investigate the influence of vegetation removal on permafrost
Figure 1. The Linell plot where trees and shrubs were removed, but the organic mat and roots were left intact. From the U.S. Army Corps of Engineers, Permafrost Division, 1950.
374 Ninth International Conference on Permafrost
Figure 2. The completely disturbed Linell plot where trees, shrubs, the organic mat and roots were removed. From the U.S. Army Corps of Engineers, Permafrost Division, 1950.
Figure 3. The CALM site at the undisturbed plot in August 2006.
4.7 m. The results showed that permafrost degradation is linked to surface vegetation disturbance and that in warm permafrost random, mixed, low vegetation will not provide a stable thermal regime for permafrost. In 2005 a Circumpolar Active Layer Monitoring Network (CALM) site was established in the undisturbed Linell plot. Thaw depths have been measured yearly in early October at 121 locations on an 11 by 11 cell grid with 3 m spacing. This provides thaw depth measurements over a 33 by 33 m (1089 m2 area). Moss thickness and total thaw depth were measured and recorded at each probe location. In the fall of 2007 direct current electrical resistivity measurements were made along an 85 m long line at both of the disturbed Linell plots to quantify the state of permafrost 61 years after modification. A borehole was installed at the completely disturbed plot in 2007 to measure the current thaw depth and to ground truth the geophysical measurements. Climate, soils, vegetation and permafrost in Interior Alaska Fairbanks, located in Interior Alaska, has a continental climate with a mean annual temperature of -3.3°C and
Figure 4. A photograph in August 2007 of the Linell plot where in 1946 the surface vegetation was removed but the organic material was left intact.
typical monthly average temperatures of 20.2°C in the summer (July) and -31.7°C in the winter (January). Absolute extremes range from -51°C to 38°C (Jorgenson et al. 2001). The average annual wet precipitation is 407 mm and the typical average annual snowfall is 1.7 m. The Fairbanks Permafrost Experimental Station site sits on a gently sloping southward facing hill 6 km north of Fairbanks, Alaska. Soils in the study area consist of tan silt and wind blown loess near the surface and grey silt at depths below 1.4 m. Permafrost gravimetric percent moisture contents range from 26% to 41% for the frozen silts which makes this relatively low moisture content permafrost (Linell 1973). The thin (2 to 32 cm thick) surface peaty organic mat has a gravimetric moisture content as high as 258% (Linell 1973). Organic rich silt and peat layers are common as are layers and inclusions of charcoal. Vegetation at the Fairbanks Permafrost Experimental Station (Fig. 3) is typical of the Alaskan Interior- subarctic taiga forest with white and black spruce towering above a thick moss layer interspersed with low-bush cranberry and Labrador tea. Feather and sphagnum moss and woody debris cover the terrain surface (Hamilton et al. 1983). Forest succession following disturbance is evident at both Linell plots. At the two disturbed plots dead or dying shrubs covered in moss are being replaced by white and black spruce and birch trees. The trees are taller at the less disturbed site where the surface vegetation was removed but the organic material was left intact (Fig. 4). Permafrost in Interior Alaska is discontinuous, generally underlying north facing slopes and valley bottoms (Jorgenson et al. 2001). A detailed description of the types of permafrost in the Alaskan Interior is provided in Osterkamp et al. 2000). The thickest permafrost at the Fairbanks Permafrost Experimental Station, 60 m, is near Farmers Loop Road and the thickness decreases with increasing elevation. The mean annual temperature at 10 m depth from 1946 to 1972 was -0.5°C (Linell 1983).
Douglas et al. 375 Circumpolar Active Layer Monitoring site measurements Thaw depths at the CALM site from 2005 to 2007 are generally between 20 and 80 cm (excluding moss) in early October (Table 1). Moss thickness ranges from 0 to 32 cm (Fig. 5). Linell (1973) presents consistent thaw depth readings of ~85 cm at the undisturbed plot from 1946 to 1972. This suggests recent climatic warming in the Alaskan Interior (1.5°C warmer at Fairbanks from the 1960s to the 1990s (Osterkamp & Romanovsky 1999)) may not be greatly affecting thaw depths at the Fairbanks Permafrost Experiment Station. In all three years of measurements the shallower moss cover is associated with deeper thaw depths. Geophysical measurements at the Linell plots Direct current (DC) electrical resistivity measurements have been used to quantify the presence of permafrost at many locations worldwide (e.g., Gilmore & Clayton 1995, Hauck et al. 2003, Hauck & Kneisel 2006). The electrical resistivity of a soil is controlled by its mineralogy, porosity, moisture content, cation/anion concentration of moisture, temperature, and whether the pore water is frozen or thawed. The resistivity (ρ) values of frozen soil are generally 10‑ to 1000 times greater than those of unfrozen or brine-rich soils (Harada & Yoshikawa 1996). Direct current electrical resistivity measurements were run across the two disturbed Linell plots in September 2007 to quantify current thaw depths. We used two-dimensional Table 1. A summary of the summer average temperatures, active layer depth, and moss thickness from 121 points at the undisturbed Linell plot (CALM site) for 2005–2007. Year 2005 2006 2007 1
May-October mean air temperature (˚C)1 13.7 12.6 12.9
Mean thaw depth without moss (cm) 44.5 40.1 43.7
Mean moss thickness (cm) 13.0 13.5 13.6
Alaska Climate Research Center.
Figure 5. The relationship between moss thickness and thaw depth at the pristine Linell plot (CALM site) for 2005 to 2007.
resistivity profiling (IRIS instruments; Syscal pro R1 48-72 channel) for this investigation using a Wenner electrode configuration. The DC resistivity sounding employed four electrodes for measurement whereby a current (I) was delivered and received between the outer two electrodes and the resulting potential difference (V) was measured between the inner two electrodes. For this array on the ground surface, an apparent resistivity (ρa) between electrodes separated by distance (a) is: ρa = 2πa(V/I)
The inversion analysis was performed with changing values of resistivity and layer thickness by using the linear filter method for a one-dimensional investigation (Das & Verma 1980). We do not have measurements of pore water salinity or moisture content across the resistivity lines and this limits the confidence of our interpretation of the resistivity measurements. However, based on the data presented in Linell (1973), the soil moisture contents, the soil temperature, and our probing and borehole measurements we are confident the resistivity markers we interpret as the upper boundary of the permafrost table are accurate. For the acquisition of the two-dimensional apparent resistivity data we used multi-channel, equally spaced electrodes at 1.5 m for the minimally disturbed Linell plot and 2 m for the completely disturbed Linell plot. Each measurement was repeated up to 16 times, depending on the variance of the results. Two-dimensional model interpretation was performed using RES2DINV (Geotomo software) which performs smoothing and constrained inversion using finite difference forward modeling and quasi-Newton techniques (Loke & Barker 1996). Resistivity profiles across the disturbed Linell plots show that in one case the permafrost continues to degrade while in another a small zone of potentially refrozen material has been recreated since 1973 (Linell 1973). Figure 6 includes a DC resistivity cross section of the plot where surface vegetation was cleared by hand but the surface organic material was left intact. Undisturbed permafrost is present at 0 m on the line (x-axis). The dark mass to the right center of the plot corresponds with a low lying grassy area along the DC resistivity line with standing water and minimal shrubs or trees. This wet soil affects DC resistivity measurements and yields an apparent highly resistive mass at a depth of 3 m that is not present based on frost probing. This is signified by the bulbous mass with resistivities >200 Ω-m from 34 to 47 m along the section. To the left (east) the resistivities that are >200 Ω-m likely signify permafrost at 9 m depth. This depth to permafrost was confirmed at two locations with an expandable frost probe. Figure 7 includes a DC resistivity cross section of the plot where surface vegetation and organic material were removed in 1946. Undisturbed permafrost is present from 0 to 25 m on the line (x-axis). A borehole encountered permafrost at 9.8 m (Fig. 8) which corresponds with the tabular region (black) consistently yielding a resistivity of >450 Ω-m.
376 Ninth International Conference on Permafrost
Figure 6. A direct current resistivity profile across the Linell plot where, in 1946, the surface vegetation was removed but roots and organic material were left intact. The unit electrode spacing is 1.5 m.
Figure 7. A direct current resistivity profile across the Linell plot where, in 1946, the surface vegetation, roots, and organic material were removed. The unit electrode spacing is 2 m. The arrow denotes where a borehole encounters permafrost at 9.8 m depth (Fig. 8).
The DC resistivity values here are ~450 to 800 Ω-m when permafrost is encountered. This is on the low end for studies reporting resistivity values for permafrost. Hoekstra and McNeill (1973) report values of 300 to 7000 Ω-m for “Fairbanks silt” with the values approaching 800 Ω-m for “Fairbanks silt” at -5°C. This is within the range of values we measured for soils that are similar to Fairbanks silt. Harada et al (2000) report values of 2100 and 12,000 Ω-m for permafrost at the Caribou-Poker Creek Research Watershed (CPCRW) located 20 km north of our study site. Yoshikawa et al. (2006) report values of 1000 to 14,000 Ω-m for permafrost at the top of a pingo at the CPCRW and 600 to 10,000 Ω-m for frozen ground at a pingo located 4 km west of the Fairbanks Permafrost Experimental Station. Permafrost in northern Quebec (Seguin and Frydecki 1994) yields values between 1000 and 5000 Ω-m. DC resistivity values are strongly controlled by the moisture content and the salinity (or conductivity) of the permafrost. In the Fairbanks area the mean annual temperature is -3.3°C. The permafrost is roughly 0.5°C to 1°C (Fig. 8) which is considered warm for permafrost. As a consequence, there is likely unfrozen water in the pore spaces of silt rich sediment at the top of the
permafrost table. This could result in lower resistivity values than are measured at locations where colder temperatures and less unfrozen water are present. A summary of our measurements combined with the data reported by Linell (1973) is provided as Figure 9. Permafrost destruction at the site where surface vegetation was removed but organic material was left intact appears to have ceased since 1973. While probing along the DC resistivity line at the site we identified a thin (50 cm) frozen layer between 1 and 1.5 m depth. This likely signifies a zone of aggrading permafrost. At the site where all vegetation and surface organic material were removed the depth to permafrost has increased to 9.8 m depth (based on frost probing and a bore hole installed at this location in the fall of 2007 that yielded temperatures below freezing at 9.8 m depth in October 2007 and January 2008 (Fig. 8). If one assumes a maximum volumetric ice content of ~30%, the loss of 9.8 m of ice at the disturbed site and 4.7 m of ice at the partially disturbed site would lead to subsidence of ~2.9 m and ~1.4 m, respectively. This assumes all the ice was lost (as melt water) through evapotranspiration or subsurface flow which is not likely. The ground surface appears visibly
Douglas et al. 377
Figure 8. Borehole temperatures on October 19, 2007 (open boxes) and January 20, 2008 (open circles) at the Linell plot where surface vegetation, roots, and organic material were removed.
Figure 9. A cross section based on Linell, 1973. Vegetation removal from the two disturbed sites occurred in 1946. The bold line denotes the measurements collected in the fall of 2007, representing 61 years since disturbance.
depressed at the two disturbed Linell plots compared to the edges of the plots and to the control plot. However, the degree of subsidence cannot be ascertained beyond a visual estimation due to a lack of high resolution elevation measurements at the site prior to disturbance in 1946.
Discussion Comparing the results from this investigation with that of Linell (1973) it is apparent that at both of the plots where vegetation was removed the permafrost expanded downward for the first 26 years while the permafrost table eventually stabilized at the partially disturbed site. This is likely due to the reestablishment of a boreal forest at the site within 25 years. This forest succession did not lead to the upward migration of the permafrost table, most likely due to the fact that ambient temperatures in the area are relatively warm. However, the small frozen layer at roughly 1.5 m depth at this site may signify that an epigenetic (downward freezing) regime is currently in place. At the site where all the surface vegetation and organic material were removed the permafrost surface has migrated downward for the past 35 years. Vegetation is continuing to evolve and is currently transitioning from a shrub, birch, and willow forest to one with a higher density of spruce trees and moss. The downward migration of the permafrost surface has not been linear in nature (Fig. 10). In fact, the rate of permafrost degradation at both of the disturbed sites is best represented by a second order polynomial. Since the climate of Interior Alaska has warmed slightly over the past three decades it is possible some of the
Figure 10. Permafrost degradation at the disturbed Linell plots. Boxes denote the site where surface vegetation and organic material were removed, and circles denote the site where surface vegetation was removed but organic material was left intact.
permafrost degradation at the Linell plots is attributable to climatic change. The downward migration of the permafrost surface may be augmented by warming temperatures in addition to the changing thermal regime caused by vegetation disturbance. Thaw depths at the pristine Linell plot (the CALM site) are ~80 cm and are within the range of values presented by Linell (1973) for the period from 1946 to 1972. The recent establishment of a Circumpolar Active Layer Monitoring Network (CALM) site at the undisturbed Linell plot will provide a baseline of the current and future state of permafrost which can be assessed once a long-term record (>10 years) has been measured.
378 Ninth International Conference on Permafrost
Conclusions Based on the results from this and previous studies, it is evident that the downward degradation of permafrost is highly dependent on the type of surface degradation and time. This has ramifications for forest succession following fire or human-caused disturbances to vegetation in permafrost terrains. This study supports previous work showing that vegetation disturbance greatly affects permafrost stability. The regrowth of boreal forest vegetation at the partially disturbed Linell plot has lead to the cessation of permafrost degradation. In fact, a thin layer of frozen ground slightly below the typical thaw depth may be evidence of the reestablishment of permafrost. The site with greater vegetation disturbance has exhibited permafrost degradation since 1946, but the rate of degradation is not linear and is decreasing as the re-establishment of boreal forest continues. Resistivity values we measured for permafrost (~450 to 800 Ω-m) are relatively low compared to measurements in ice-rich permafrost. However, our site is characterized by low-moisture-content silts. Further studies of the thermal state and moisture content of the frozen silt at the site are warranted. Finally, ambient air temperatures have increased over the past few decades in the study area, and they are expected to continue to increase in the future. Though thaw depths in the permafrost at the CALM site appear to be somewhat stable since 1946, we will only be able to determine how future warming will continue to affect permafrost at the site with continued monitoring.
Acknowledgments Many people donated time, equipment, and information to support this collaborative study. Site history was provided from Charles Collins and Karen Henry of CRREL. Field assistance from Stephanie Saari, Art Gelvin, Allan Delaney, Charles Collins, and Tohru Saito has been invaluable. Beth Astley shared her geophysical information for Birch Hill and Farmers Loop. Comments from two anonymous reviewers and Editor D. Kane strengthened the manuscript.
References Burn, C.R. 1998. The response (1958–1997) of permafrost and near-surface ground temperatures to forest fire, Takhini River valley, southern Yukon Territory. Canadian Journal of Earth Science 35: 184-199. Das, U.C. & Verma, S.K. 1980. Digital linear filter for computing type curves for the two-electrode system of resistivity sounding. Geophysical Prospecting 28: 610-619. Gilmore, T.J. & Clayton, E.A. 1995. Mapping the top of the permafrost using surface direct current resistivity survey. Environmental Geology 30(1/2): 29-33. Hamilton, T.D., Ager, T.A. & Robinson, S.W. 1983. Late Holocene ice wedges near Fairbanks, Alaska, U.S.A.: Environmental setting and history of growth. Arctic and Alpine Research 15(2): 157-168. Harada, K. & Yoshikawa, K. 1996. Permafrost age and thickness near Adventfjorden, Spitsbergen. Polar Geography 20: 4, 267-281.
Harada, K., Wada, K. & Fukuda, M. 2000. Permafrost mapping by transient electromagnetic method. Permafrost and Periglacial Processes 11: 71-84. Hauck, C., Mühll, D.V. & Maurer, H. 2003. Using DC resistivity tomography to detect and characterize mountain permafrost. Geophysical Prospecting 51(4): 273-284. Hauck, C. & Kneisel, C. 2006. Application of capacitively-coupled and DC electrical resistivity for mountain permafrost studies. Permafrost and Periglacial Processes 17: 169-177. Hinzman, L.D., Fukuda, M., Sandberg, D.V., Chapin III, F.S. & Dash, D. 2003. FROSTFIRE: An experimental approach to predicting the climate feedbacks from the changing boreal fire regime. Journal of Geophysical Research 108(D1): 8153. Hoekstra, P. & McNeill, D. 1973. Electromagnetic probing of permafrost. National Academy of Sciences, National Research Council. Proceedings of Permafrost: North American contribution to the Second International Conference: 517-526. Jorgenson, M.T., Racine, C.H., Walters, J.C. & Osterkamp, T.E. 2001. Permafrost degradation and ecological changes associated with a warming climate in central Alaska. Climatic Change 48: 551-579. Linell, K.A. 1973. Long-term effects of vegetative cover on permafrost stability in an area of discontinuous permafrost. National Academy of Sciences, National Research Council. Proceedings of Permafrost: North American contribution to the Second International Conference: 688-693. Loke, M.H. & Barker, R. D. 1996. Rapid least-squares inversion of apparent resistivity pseudosections by a quasi-Newton method. Geophysical Prospecting 44(1): 131-152. Nicholas, J.R.J. & Hinkel, K.M. 1996. Concurrent permafrost aggradation and degradation induced by forest clearing, central Alaska, U.S.A. Arctic and Alpine Research 28(3): 294-299. Osterkamp, T.E. & Romanovsky, V.E. 1999. Evidence for warming and thawing of discontinuous permafrost in Alaska. Permafrost and Periglacial Processes 10: 17-37. Osterkamp, T.E., Vierick, L., Shur, Y., Jorgenson, M.T., Racine, C., Doyle, A. & Boone, R.D. 2000. Observations of thermokarst and its impact on boreal forests in Alaska, U.S.A. Arctic, Antarctic and Alpine Research 32(3): 303-315. Seguin, M.-K. & Frydecki, J. 1994. Semi-quantitative geophysical investigations of permafrost in northern Quebec. Journal of Applied Geophysics 32: 73-84. Shur, Y.L. & Jorgenson, M.T. 2007. Patterns of permafrost formation and degradation in relation to climate and ecosystems. Permafrost and Periglacial Processes 18: 7-19. U.S. Army Corps of Engineers, Permafrost Division. 1950. Investigation of Military Construction in Arctic and Subarctic Regions, Comprehensive Report 1945‑48; Main Report and Appendix III - Design and Construction Studies at Fairbanks Research Area. ACFEL Technical Report 28. Viereck, L.A. 1982. Effects of fire and firelines on active layer thickness and soil temperatures in interior Alaska. Proceedings of the Fourth Canadian Permafrost Conference. National Research Council of Canada Calgary: Ottawa, Ontario: 123-135. Yoshikawa, K. & eight others. 2006. Comparison of geophysical investigations for detection of massive ground ice (pingo ice). Journal of Geophysical Research 111(E06S19): 10 pp.