Technical Articles, Vol 26,3 Climate Change and Critical Zone Geophysics

Permafrost-Through-Canopy Investigation of Thermal and Ecohydrological Processes in Arctic Systems

By Baptiste Dafflon*, Sebastian Uhlemann, Susan S. Hubbard

Lawrence Berkeley National Laboratory, Berkeley, CA 94720, USA

Email: *

Published 29/11/2021


Decreasing uncertainty in how the Arctic water and carbon cycles will be impacted by a warming climate requires advances in quantifying and understanding above- and belowground hydro-biogeochemical processes and their controlling factors.  In this study, we summarize the results of our recent geophysical studies aimed at quantifying heterogeneity in Arctic subsurface properties and tractably identifying, characterizing and monitoring interactions occurring between permafrost-to-canopy compartments in cold and warm permafrost environments. To this end, we developed novel approaches for improving the characterization and monitoring of subsurface properties and their co-variability with surface properties through integration of geophysical methods, point-scale and remote sensing data, as well as for monitoring soil temperature at an unprecedented number of locations, and using such observations to infer soil thermal regimes and parameters. Our bedrock-to-canopy investigations led to high resolution estimates of ice-wedge, ice content, porosity, and salinity distributions important for model parameterization. They also documented and led to new insights about the significant co-variability between (micro-) topography, vegetation, snow distribution and subsurface characteristics and their impact on permafrost dynamics and carbon fluxes. Capitalizing on the co-variability concept, we identified regions in the landscape that had distinct carbon fluxes relative to neighboring regions, thereby providing an approach for rapidly integrating multi-scale, multi-type data for characterization and model parameterization.


Quantifying Arctic ecosystem properties and processes is critical for improving the prediction of how climate change will influence Arctic carbon and nitrogen cycling (Petrone et al., 2006), water distribution and resources, and infrastructure vulnerability (Hjort et al., 2018). The Arctic is a climatically sensitive region that is rapidly changing. It is warming two to three times faster than the global mean, perennially frozen ground is gradually thawing, and precipitation and the ratio of rain to snow is increasing (Bintanja, 2018; Biskaborn et al., 2019). With warming, a vast reservoir of carbon that is currently locked up in frozen permafrost (Tarnocai et al., 2009) may be exposed to microbial decomposition, leading to significant fluxes of greenhouse gasses from the land to the atmosphere (e.g., Schaphoff et al., 2013). However, an overall increase in ecosystem productivity due to warming might concurrently enhance carbon uptake (McGuire et al., 2009). The uncertainty associated with whether the Arctic will continue to serve as a carbon sink through the coming century has driven significant scientific investments, including the Next-Generation Ecosystem Experiments (NGEE Arctic) project that is funded by the U.S. Department of Energy (DOE). The NGEE Arctic project aims to improve climate model predictions through advanced understanding of coupled processes in Arctic terrestrial ecosystems.  

Key to decreasing uncertainty in how climate change will impact carbon and water cycling in the Arctic is improving our understanding of complex hydro-biogeochemical processes and their controls. Their complexity is caused by numerous interactions and feedback between the atmosphere, land surface, vegetation, soil, permafrost and bedrock properties (Jorgenson et al., 2010). In this regard, the quantification of the subsurface hydro-biogeochemical properties and the understanding of how they interact with surface properties across the landscape is critical, yet challenging to achieve at multiple scales across atmospheric, geologic, geomorphological and ecological gradients. Such multi-scale understanding requires integration of point-scale data (such as for example core samples analysis) with ground-surface geophysical data and remote sensing data (Figure 1). While permafrost characterization and data fusion strategies are still at an incipient stage of development, they are critical for quantifying interactions across permafrost through canopy compartments and across landscape relevant scales.

Figure 1. Improving the predictive understanding of the impact of climate change on carbon and water cycle across the Arctic ecosystem requires integration of various observational data (e.g., point-scale, geophysical, remote sensing).

To address this challenge, we used geophysical methods coupled with point-scale and remote sensing data to quantify heterogeneity in Arctic subsurface properties, as well as their controls on above- and belowground hydro-biogeochemical processes.  We primarily relied on Electrical Resistivity Tomography (ERT), frequency-domain Electromagnetic Induction (EMI), Ground Penetrating Radar (GPR), and Unmanned Aerial System (UAS) approaches to estimate properties such as ground elevation, snow thickness and vegetation indices, and point-scale measurements to quantify physical, thermal, hydrological and biogeochemical processes.  We also developed new characterization strategies, such as Distributed Temperature Profiling (DTP). Our studies were conducted near Utkiagvik, Alaska and on the South Seward Peninsula, Alaska; two end-members with regard to permafrost temperature. The Utkiagvik site is an ice-wedge polygon-dominated Arctic tundra in a coastal environment, where the mean annual air temperature is -11.3 oC, the mean annual precipitation is 173 mm, and the mean annual permafrost temperature can be as low as -9 oC at 16 m depth (Hubbard et al., 2013; Yoshikawa et al., 2004). The site exhibits low topographic relief, varying between about 2 and 6 m elevation, and is dominated by different types of polygons, including low-centered polygon (LCP), flat-centered polygon (FCP), and high-centered polygon (HCP) (e.g., Hinkel et al., 2001). These different types of polygons (~5-20 m in size), which represent various stages of ice wedge development and degradation, govern the local (micro-) topography (Leffingwell, 1915; Mackay, 2000). On the Seward Peninsula, our work was primarily performed in a watershed located along Teller road (mile #27) near Nome, which is underlain by discontinuous permafrost. The mean annual air temperature is -1.02 oC, the annual precipitation as rain is 451 mm, and the mean annual temperature of the permafrost is warmer than -3 oC (Léger et al., 2019; Uhlemann et al., 2021).  Below, we briefly review findings from studies focused on estimating Arctic ecosystem properties and covariance across permafrost, active layer, land surface and vegetation compartments; monitoring permafrost hydrological and thermal processes; and exploring how such processes may accelerate permafrost thaw. 

How do Arctic Tundra properties covary across ecosystem compartments, and can new approaches be developed to tractably characterize controlling features?

The vulnerability of permafrost to thaw depends on many factors including atmospheric forcing, topography, snowpack properties, vegetation characteristics, disturbance, surface water and soil moisture, soil thermal properties, and heat transferred from bedrock and sub-permafrost groundwater. These factors that control the active layer thickness (the portion of the soil column that thaws each summer; Figure 1) and its deepening vary over space and time. For example, thawing of ice-rich permafrost can lead to changes in soil bulk density, thermal parameter and (micro-) topography, which in turn can change snow accumulation, surface hydrology, soil moisture, methane emission, soil respiration and plant productivity and succession (Figure 1). 

At the Utkiagvik site, we acquired and integrated geophysical methods, point-scale measurements, laboratory core analysis, and UAS data to evaluate the spatial distribution of permafrost through canopy properties that influence or reflect active layer thickness and permafrost thaw in a polygonal-shaped tundra environment.  An extremely important aspect of this effort was to quantify subsurface properties, which have been the focus of much less investigation in Arctic systems due to the challenges associated with their characterization. Yet, information about active layer and permafrost properties are critically needed for parameterizing models used to predict how Arctic permafrost systems will respond to warming. For example, the organic matter content has a strong role on available carbon for microbial decomposition, as well as on the thermal and hydrological properties that influence the vulnerability of permafrost to thaw (e.g., Hinzman et al., 1991). Similarly, permafrost ice-content (which can be as high as 80% in ice-rich permafrost) and the distribution of ice-wedges influence the rate and geomorphic expression of permafrost thaw. We acquired ERT along 450 m long transects with 0.5 m electrode spacing, which provided high-resolution imaging of the active layer and permafrost structure in the top 5 m bgl (Dafflon et al., 2016; Hubbard et al., 2013). We acquired Ground Penetrating Radar (GPR) data to estimate thaw layer thickness, ice-wedge distribution and snow thickness distribution, as well as frequency-domain EMI data to map trends in active layer wetness and permafrost salinity across the site (Hubbard et al., 2013; Leger et al., 2017; Wainwright et al., 2017). Further, we collected cores down to 4 m depth at multiple locations to constrain the delineation of major subsurface features and the estimation of ice-content, salinity and unfrozen water content from the geophysical data (Figure 2) (Dafflon et al., 2016; Wu et al., 2018). 

The geophysical data integrated with core analysis enabled a first high-resolution glimpse of the spatial variability in the coastal Arctic near-surface permafrost, providing estimates of ice content, porosity, and salinity. Core analysis showed that water/ice saturated porosity in the top 2 m varied between 85% and 40%, and was negatively correlated with fluid salinity.  The salinity directly influenced ice formation and unfrozen water content and indirectly influenced the soil organic matter content. The fluid conductivity of thawed soil samples showed values as high as 5.1 mS/cm (approximately 10% of seawater salinity) at 0.75 m depth, and 30 mS/cm at 3.55 m depth (Dafflon et al., 2016). Based on the salinity-ice content relationship, a modified Archie equation and a simple model representing salt exclusion during ice formation, we estimated ice-content and unfrozen water content along the ERT transects (Figure 2). Results indicated a relatively continuous but depth-variable increase in salinity that lead to a partially unfrozen saline layer (cryopeg) located below the top of the permafrost. The unfrozen water content was spatially variable with maximum values around 15% at -5 oC.  Overall, the high salinity supported thaw of permafrost at temperatures below 0 oC, which can accelerate erosion and deformation in this coastal region. It can also support microbial activity; other studies have documented the presence of microbial activity in saline permafrost regions under subzero temperature conditions (Gilichinsky et al., 2005; Shcherbakova et al., 2009). 

Figure 2. ERT and cores laboratory analysis enabled estimation of major subsurface features in a polygonal-shaped tundra near Utkiagvik, Alaska. (a) Example of core analysis including wet bulk density, ice content, organic matter concentration and density, fluid conductivity from thawed samples, and comparison to X-ray computed tomography of the core wet bulk density (black line) and electrical conductivity extracted from the ERT data (red line). (b) UAV-inferred topography with the black line showing the location of the (c) ERT transect. (d) Petrophysical-based delineation of subsurface features including thaw layer (white); ice-wedge (white); very ice-rich permafrost (orange); organic-rich permafrost (green); permafrost with increasing salinity (light blue); and saline layer with limited interstitially segregated ice (dark blue). (e) estimated porosity (ice and water content) (modified from Dafflon et al., 2016).

We used the interpreted  permafrost-through-canopy properties to explore how the properties covaried over space, and to explore if the Arctic landscape could be subdivided into zones, each having a unique distribution of permafrost-through-vegetation properties that influenced carbon cycling (Wainwright et al., 2015). Our studies highlighted the significant control of (micro-) topography on above and belowground properties that influence carbon fluxes in this polygon-dominated Arctic tundra. We found strong covariance between (micro-) topography and snow distribution and hydrology, and in turn soil moisture, redox, microbial community and vegetation (Bisht et al., 2018; Gangodagamage et al., 2014; Hubbard et al., 2013; Tran et al., 2018; Wainwright et al., 2015; Wainwright et al., 2021; Wu et al., 2018). The recognition of this control opens the door for using topography information to estimate other properties that are generally more difficult to characterize, and to extrapolate results from study sites to regional scales. To test this concept, an algorithm was developed to automatically delineate polygons in Arctic ice wedge polygon regions, and compute attributes of each polygon (Wainwright et al., 2015). In addition, an unsupervised clustering approach was used to identify zones that provided the most unique spatial grouping of the above- and belowground measurements acquired at the site. Through comparison with chamber- and eddy-based flux data, the results showed that the identified polygon-based zones had distinct carbon fluxes (Wainwright et al., 2015; Wainwright et al., 2021). This important finding suggests that geophysical data and machine learning approaches can be useful for estimating regions that have distinct carbon fluxes relative to neighboring regions, which can guide field sampling and model parameterization. 

How do Arctic Tundra seasonal dynamics in above- and belowground properties covary, and can these be used to improve the estimation of soil thermo-hydrological properties?

Continuous monitoring of above- and belowground seasonal and annual dynamics is needed to gain a predictive understanding of spatiotemporal soil thermal and hydrological behaviors, and their drivers. At the Utkiagvik site, we installed a 35 m long autonomous monitoring system along a transect across three different types of polygons (HCP, FCP and LCP). An ERT system acquired data every day with an electrode spacing of 0.5 m to sense changes in thaw layer thickness and water content, as well as the dynamics in unfrozen water content and salinity in the permafrost. ERT data were complemented with several point-scale soil moisture and temperature sensors, and sporadic measurements of thaw and snow layer thickness. A pole-mounted camera was installed to monitor daily the Green Chromatic Coordinate (GCC) ─ proxy for the vegetation vigor and density ─ along the ERT transect with about 0.5 m resolution. The various datasets were used to evaluate permafrost dynamics, spatiotemporally variable above- and belowground interactions that influence carbon and water fluxes (Dafflon et al., 2017), soil thermal properties and organic matter content (Jafarov et al., 2020; Tran et al., 2017), and controls on active layer thickness (Tran et al., 2018).

The ERT data also highlighted specific polygon-type spatiotemporal dynamics in thaw layer thickness (TLT) and the drivers. The electrical conductivity extracted from the top 20 cm of ERT time lapse images were strongly correlated with TLT measurements (r=0.81) and TDR dielectric permittivity measurements (r=0.83) at the peak of the growing season. Analyzing the above datasets at various times during the growing season showed that electrical conductivity can be used as a proxy for TLT and the depth integrated amount of water in the thaw layer, which at this site increased during the growing season primarily due to the thickening of the thaw layer.  Spatiotemporal analysis of the ERT and TLT measurements showed that the TLT at HCP was smaller than that at LCP, and that both thawing and freezing occurred earlier at the HCP compared to the LCP.  The ERT data were further used with the TLT sporadic measurements to estimate TLTs with high temporal resolution and to compare them with hydro-thermal numerical simulations to assess the controlling factors (Tran et al., 2018). The 1-D hydrothermal simulations using the community land model (CLM) showed that both air temperature and precipitation jointly governed the temporal variations of TLT, while the topsoil organic content and polygon morphology were responsible for its spatial variations.  When the topsoil organic content and thickness increased, the TLT decreased. Meanwhile, LCP showed a thicker snow layer and saturated soil, which contributed to a thicker TLT and extended the time needed for TLT to freeze and thaw.  Overall, the data and model confirmed the strong impact of topography, snow distribution, soil moisture and organic matter content on the TLT dynamics. 

The coincident monitoring of soil and vegetation properties facilitated the evaluation of the control of electrical conductivity on the vegetation seasonal dynamics (Dafflon et al., 2017). While at the beginning of the growing season the vegetation green chromatic coordinate did not correlate with electrical conductivity of the top 20 cm, the correlation coefficient rapidly increased until the peak of the growing season (r = 0.89) on 22 July (Figure 3a). In addition, the data analysis showed that the green chromatic coordinate at the peak of the growing season was highly correlated with electrical conductivity almost every day from the beginning to the end of the growing season. We evaluated further the robustness of the developed relationships along a 470 m long transect, where we compared the green chromatic coordinate around the peak of the growing season obtained from UAV-imagery and the collocated electrical conductivity of the top 20 cm measured at the end of the season. Though not collocated in time, the resulting correlation coefficient between the green chromatic coordinate and the bulk conductivity of the top 20 cm of the long ERT transect is 0.75 (or 0.79 if the areas where ponds of surface water were present are removed) (Figure 3b-c). If this relationship can be documented or developed at other sites, this would open the door for using easily acquired optical data (ideally with topography information) to estimate growing season soil properties.

Figure 3. Strong coupling of soil top 20 cm Electrical Conductivity (EC) and plant Green Chromatic Coordinate (GCC) at the peak growing season across polygonal-shaped tundra near Utkiagvik, Alaska. (a) Correlation between soil top 20 cm EC extracted from ERT time-lapse data and collocated measurements of GCC using a pole-mounted camera with the 30 m long ERT monitoring transect in the field of view, from 1 July to 11 August 2014. (b-c) assessment of the obtained relationship at larger scale. (b) UAS-inferred GCC. Black line indicates the location of the ERT transect (Figure 2c), along which the (c) GCC and soil top 20 cm EC were extracted. The black and grey intervals indicate location of surface water and troughs, respectively. The resulting correlation between GCC and EC is 0.75 (and 0.79 if locations with surface water are removed) (modified from Dafflon et al., 2017).

Finally, the time-series of electrical conductivity, in conjunction with soil moisture and temperature data, were used to develop numerical approaches for the indirect estimation of thermal properties and/or carbon content and porosity. Several studies reported that inclusion of vertical organic carbon content profiles into a land surface model can considerably improve predictions of heat, water and carbon fluxes (e.g.,Nicolsky et al., 2007), yet the estimation of carbon content profiles remains challenging. As part of our work in Utkiagvik, we developed hydrological–thermal–geophysical inversion schemes that couple ecosystem with geophysical models in order to estimate thermal properties and in some cases organic and mineral content at several depths (Jafarov et al., 2020; Tran et al., 2017).  Results indicate that the approach is able to estimate organic and mineral content within the shallow active layer (top 0.3m of soil) with high reliability, while the uncertainty increases at larger depth (Tran et al., 2017). The uncertainty is reduced in the presence of sufficient temporal variations in temperature and moisture, as well as in cases where the soil porosity is functionally related to the organic and mineral content, which is often observed in organic-rich Arctic soil. Our results showed the promise in indirectly estimating soil properties from time-series data, yet also the need for denser temperature measurements, improved petrophysical relationships and advanced representation of freeze-thaw processes in models. 

How are transitional permafrost systems reshaped by thermohydrological dynamics?

Permafrost temperatures rising toward and above the freezing point cause increase in hydraulic conductivities in the soil to bedrock column, enhanced surface water-groundwater interactions, changes in solute composition and geomorphology, and thus a shift in carbon and water cycling (Hinzman et al., 2005; Ireson et al., 2013). While it is recognized that in discontinuous permafrost environments the development of perennially unfrozen ground (called taliks) can lead to the progressive disappearance of permafrost, there is little understanding of how subsurface heterogeneity influences these processes. 

At the Teller road site near Nome, we used geophysical methods to characterize regions with near-surface permafrost, taliks and absence of permafrost and to evaluate the controls on their associated thermal regimes. We relied on ERT and seismic refraction for permafrost and bedrock characterization; soil sampling and analysis for petrophysical analysis; UAS-based imaging for mapping of vegetation indices, surface elevation, snow thickness and plant height; and an in-house developed distributed temperature profiling (DTP). The DTP system consists of vertically resolved temperature probes that independently and autonomously recorded soil temperature at multiple depths, with 5 to 10 cm vertical resolution, high accuracy, and low physical footprint and energy consumption (Dafflon et al., 2021). The DTP system was designed to monitor soil or snow temperature at an unprecedented number of locations, or to be moved sequentially across the landscape for mapping soil temperature down to depths not impacted by daily temperature variations (Léger et al., 2019). In addition, a permafrost monitoring transect was established at the site to improve the understanding of spatial and temporal variability of above- and belowground thermohydrological processes (Uhlemann et al., 2021). The 127 m transect set perpendicular to the main slope gradient at the Teller site crosses two different vegetation types; one covered with graminoid and the other primarily with tall willow shrubs. An ERT system composed of 128 electrodes spaced at 1 m, provide daily measurement from March to September since 2018. At five locations along this transect soil temperature and moisture were monitored at various depths. Snow thickness was measured along the transect in late March 2018 and 2019. 

The DTP and ERT surveys at the Teller site provided insights on the heterogeneity in subsurface thermal regimes and the drivers. The mapping of soil temperature down to 0.8 m depth at hundreds of locations across the Teller site using the DTP systems enabled high-resolution identification and lateral delineation of near-surface permafrost locations from surrounding zones with no permafrost or deep permafrost table locations overlain by a perennially thawed layer (Figure 4a). The collocated acquisition of ERT data confirmed the presence of near-surface permafrost and their extent down to 15 to 20 m depth, with taliks surrounding them. UAV-based photogrammetric data indicated that these near-surface permafrost areas were primarily collocated under topographic highs (tens of cm higher than the surroundings) and areas covered with graminoids, where snow thickness is generally thin compare to the surrounding areas. Overall, the observed co-variability between above- and belowground properties opened the door to infer spatially continuous estimates of permafrost table depth across the watershed.  

The time-lapse ERT data highlighted the impact of snow thickness distribution on subsurface thermohydrological properties and processes (Figure 4b) (Uhlemann et al., 2021). Topographic lows and presence of tall shrubs led to increased snow accumulation, which insulated the ground and maintained soil temperatures above 0°C throughout the winter. Hence, snowmelt could readily infiltrate below the shrubs, while the frozen conditions of the graminoid due to thinner snowpack prevented infiltration of liquid water. This was confirmed by the change in electrical resistivity (with respect to a measurement prior to the event) and the shallow soil moisture sensors. It can be noted that large rain events in the summer months showed similar pattern, in that the two shrub units showed much faster variation in their electrical conductivity than in the graminoid. Rain events also clearly showed the presence of lateral heat flow at depths larger than 5 m, where a decreasing resistivity trend advanced from the western shrubs toward the graminoid. Based on the various measurements at the site, we associated the rapid changes in electrical conductivity occurring during the snowmelt and the rain events to heat advection dominating the thermodynamics. The expected increase in snow accumulation and rainfall may further increase the infiltration into the subsurface and the heat transport into the permafrost bodies.

Figure 4. Snowmelt is causing rapid changes in electrical resistivity, particularly underneath tall shrub where taliks are present. (a) Temperatures at 0.8m depth extracted from a DTP dataset collected on 17 July 2017, and overlaid on UAV-inferred visible imagery and topographic isolines for every 5m. Coldest temperature indicates near-surface permafrost locations, primarily located in graminoid covered areas. (b) Electrical resistivity (September 2017) and distribution of changes in electrical resistivity at two dates past the snowmelt events, relative to measurements prior to the start of snowmelt (May 1, 2018). The monitoring transect is parallel and close to transect B. Indicated snow thickness corresponds to measurements in April 2018 (modified from Leger et al., 2019 and Uhlemann et al., 2021).


This study summarized the results of our recent studies that were intended to (1) provide unprecedented information about subsurface Arctic permafrost systems and (2) to develop and test new approaches to tractably identify, characterize and monitor interactions occurring between compartments of complex Arctic ecosystems. 

Associated with the first goal, our bedrock-to-canopy investigations provided new understanding and quantification of:

  • The spatial variability in coastal Arctic permafrost, including high resolution estimates of ice-wedge distribution, ice content, porosity, and salinity;
  • The significant control of (micro-) topography on above- and belowground properties that influence carbon fluxes in polygon-dominated Arctic tundra;
  • A relationship between vegetation greenness and surficial electrical conductivity, opening the door for using visible imagery to estimate growing season soil properties;
  • The impact of snow on talik development and the associated increase in heat transport into the subsurface. 

Associated with the second goal, our studies demonstrated the value of:

  • Geophysical data and machine learning approaches for estimating regions that have distinct carbon fluxes relative to neighboring regions, which can guide field sampling and model parameterization;
  • Numerical approaches to ingest time-series of electrical conductivity, soil moisture and/or temperature to estimate thermal parameters and/or organic matter content; 
  • A new DTP system for identifying and delineating various soil thermal regimes, as well as monitoring soil or snow temperature at an unprecedented number of locations.

Future development in geophysical methods and remote sensing techniques and fusion of obtained disparate datasets are expected to further improve an integrated and multi-scale understanding of hydro-biogeochemical process across the warming Arctic.  


This material is based upon work supported by the Next Generation Ecosystem Experiment (NGEE-Arctic), funded by the U.S. Department of Energy, Office of Science, Office of Biological and Environmental Research under Award Number DE-AC02-05CH11231. We thank Sitnasuak Native Corporation for their guidance and for allowing us to conduct this research on the traditional homelands of the Inupiat people. 

Author Bios:

Baptiste Dafflon
Lawrence Berkeley National Lab
Baptiste Dafflon is a Staff Scientist in the Earth & Environmental Sciences Area at Lawrence Berkeley National Laboratory (LBNL).  After an MSc degree in Geophysics at the ETH-Zurich, he completed a PhD at the University of Lausanne and a one-year postdoc at Boise State University, before to join LBNL in 2011. His current research focuses on improving the quantification and understanding of above- and below-ground processes important for managing water resources, carbon cycle, natural hazards and environmental pollutions in a variety of environments. His work includes the development of distributed sensor network, remote sensing and geophysical methods to improve the quantification of subsurface hydro-biogeochemical properties, and the multi-scale estimation of subsurface and surface properties/fluxes using statistical methods and process-based models.

Sebastian Uhlemann
Lawrence Berkeley National Lab
Sebastian Uhlemann is a Research Scientist in the Earth & Environmental Sciences Area at LBNL. He obtained a joint M.Sc. degree in Applied Geophysics from TU Delft, ETH Zurich, and RWTH Aachen, and a PhD from ETH Zurich. Before joining LBNL in 2018, Sebastian worked as a Research Geophysicist at the British Geological Survey. His current research focuses on the development and application of geophysical techniques (geoelectrical and seismic) to understand hydrological processes that influence groundwater dynamics, slope instabilities, and interactions with plants and the atmosphere at a range of scales. This includes optimization of survey designs, integrated analysis of geophysical and environmental/hydrological data, and development of novel geophysical monitoring approaches.

Susan Hubbard
Lawrence Berkeley National Lab
Susan Hubbard is an Associate Lab Director and Senior Scientist at Berkeley Laboratory. She leads the Earth and Environmental Sciences Area, a premier organization working on some of the most pressing issues of our time with expertise in climate, ecosystem, water, critical mineral, energy geoscience and urban system science. Susan is an environmental geophysicist, with expertise in advancing geophysical and data integration approaches to quantify how watersheds and ecosystems are responding to abrupt and gradual disturbances. She is an elected member of the National Academy of Engineering, and a Fellow of the American Geophysical Union, the American Academy of Arts & Sciences, and the Geological Society of America. Susan received her PhD from University of California at Berkeley, where she holds an Adjunct Professor position, and previously was a geologist with the US Geological survey and a geophysicist in industry. 
Adjunct Professor position, and previously was a geologist with the US Geological survey and a geophysicist in industry. 


Bintanja, R., 2018, The impact of Arctic warming on increased rainfall. Scientific Reports, 8(1), 16001.

Bisht, G., Riley, W.J., Wainwright, H.M., Dafflon, B., Yuan, F. and Romanovsky, V.E., 2018, Impacts of microtopographic snow redistribution and lateral subsurface processes on hydrologic and thermal states in an Arctic polygonal ground ecosystem: a case study using ELM-3D v1. 0. Geoscientific Model Development (Online), 11(1).

Biskaborn, B.K., Smith, S.L., Noetzli, J., Matthes, H., Vieira, G., Streletskiy, D.A., Schoeneich, P., Romanovsky, V.E., Lewkowicz, A.G., et al., 2019, Permafrost is warming at a global scale. Nature Communications, 10(1), 264.

Dafflon, B., Hubbard, S., Ulrich, C., Peterson, J., Wu, Y., Wainwright, H. and Kneafsey, T.J., 2016, Geophysical estimation of shallow permafrost distribution and properties in an ice-wedge polygon-dominated Arctic tundra region. Geophysics, 81(1), WA247-WA263.

Dafflon, B., Oktem, R., Peterson, J., Ulrich, C., Tran, A.P., Romanovsky, V. and Hubbard, S.S., 2017, Coincident aboveground and belowground autonomous monitoring to quantify covariability in permafrost, soil, and vegetation properties in Arctic tundra. Journal of Geophysical Research: Biogeosciences, 122(6), 1321-1342.

Dafflon, B., Wielandt, S., Lamb, J., McClure, P., Shirley, I., Uhlemann, S., Wang, C., Fiolleau, S., Brunetti, C., et al.: A Distributed Temperature Profiling System for Vertically and Laterally Dense Acquisition of Soil and Snow Temperature. The Cryosphere Discuss., 1-29, 2021.

Gangodagamage, C., Rowland, J.C., Hubbard, S.S., Brumby, S.P., Liljedahl, A.K., Wainwright, H., Wilson, C.J., Altmann, G.L., Dafflon, B., et al., 2014, Extrapolating active layer thickness measurements across Arctic polygonal terrain using LiDAR and NDVI data sets. Water Resources Research, 50(8), 6339-6357.

Gilichinsky, D., Rivkina, E., Bakermans, C., Shcherbakova, V., Petrovskaya, L., Ozerskaya, S., Ivanushkina, N., Kochkina, G., Laurinavichuis, K., et al., 2005, Biodiversity of cryopegs in permafrost. Fems Microbiology Ecology, 53(1), 117-128.

Hinkel, K.M., Doolittle, J.A., Bockheim, J.G., Nelson, F.E., Paetzold, R., Kimble, J.M. and Travis, R., 2001, Detection of subsurface permafrost features with ground-penetrating radar, Barrow, Alaska. Permafrost and Periglacial Processes, 12(2), 179-190.

Hinzman, L.D., Bettez, N.D., Bolton, W.R., Chapin, F.S., Dyurgerov, M.B., Fastie, C.L., Griffith, B., Hollister, R.D., Hope, A., et al., 2005, Evidence and Implications of Recent Climate Change in Northern Alaska and Other Arctic Regions. Climatic Change, 72(3), 251-298.

Hinzman, L.D., Kane, D.L., Gieck, R.E. and Everett, K.R., 1991, Hydrologic and thermal properties of the active layer in the Alaskan Arctic. Cold Regions Science and Technology, 19(2), 95-110.

Hjort, J., Karjalainen, O., Aalto, J., Westermann, S., Romanovsky, V.E., Nelson, F.E., Etzelmüller, B. and Luoto, M., 2018, Degrading permafrost puts Arctic infrastructure at risk by mid-century. Nature Communications, 9(1), 5147.

Hubbard, S.S., Gangodagamage, C., Dafflon, B., Wainwright, H., Peterson, J., Gusmeroli, A., Ulrich, C., Wu, Y., Wilson, C., et al., 2013, Quantifying and relating land-surface and subsurface variability in permafrost environments using LiDAR and surface geophysical datasets. Hydrogeology Journal, 21(1), 149-169.

Ireson, A., van der Kamp, G., Ferguson, G., Nachshon, U. and Wheater, H., 2013, Hydrogeological processes in seasonally frozen northern latitudes: Understanding, gaps and challenges. Hydrogeology Journal, 21.

Jafarov, E.E., Harp, D.R., Coon, E.T., Dafflon, B., Tran, A.P., Atchley, A.L., Lin, Y. and Wilson, C.J., 2020, Estimation of subsurface porosities and thermal conductivities of polygonal tundra by coupled inversion of electrical resistivity, temperature, and moisture content data. The Cryosphere, 14(1), 77-91.

Jorgenson, M.T., Romanovsky, V., Harden, J., Shur, Y., O’Donnell, J., Schuur, E.A.G., Kanevskiy, M. and Marchenko, S., 2010, Resilience and vulnerability of permafrost to climate change. Canadian Journal of Forest Research, 40(7), 1219-1236.

Leffingwell, E., 1915, Ground-Ice Wedges, the dominant form of ground-ice on the north coast of Alaska. Journal of Geology, (23), 635-654.

Léger, E., Dafflon, B., Robert, Y., Ulrich, C., Peterson, J.E., Biraud, S.C., Romanovsky, V.E. and Hubbard, S.S., 2019, A distributed temperature profiling method for assessing spatial variability in ground temperatures in a discontinuous permafrost region of Alaska. The Cryosphere, 13(11), 2853-2867.

Leger, E., Dafflon, B., Soom, F., Peterson, J., Ulrich, C. and Hubbard, S., 2017, Quantification of Arctic Soil and Permafrost Properties Using Ground-Penetrating Radar and Electrical Resistivity Tomography Datasets. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 10(10), 4348-4359.

Mackay, J.R., 2000, Thermally induced movements in ice-wedge polygons, western Arctic coast: A long-term study. Geographie Physique Et Quaternaire, 54(1), 41-68.

McGuire, A.D., Anderson, L.G., Christensen, T.R., Dallimore, S., Guo, L., Hayes, D.J., Heimann, M., Lorenson, T.D., Macdonald, R.W., et al., 2009, Sensitivity of the carbon cycle in the Arctic to climate change. Ecological Monographs, 79(4), 523-555.

Nicolsky, D.J., Romanovsky, V.E. and Tipenko, G.S., 2007, Using in-situ temperature measurements to estimate saturated soil thermal properties by solving a sequence of optimization problems. The Cryosphere, 1(1), 41-58.

Petrone, K.C., Jones, J.B., Hinzman, L.D. and Boone, R.D., 2006, Seasonal export of carbon, nitrogen, and major solutes from Alaskan catchments with discontinuous permafrost. Journal of Geophysical Research: Biogeosciences, 111(G2).

Schaphoff, S., Heyder, U., Ostberg, S., Gerten, D., Heinke, J. and Lucht, W., 2013, Contribution of permafrost soils to the global carbon budget. Environmental Research Letters, 8(1).

Shcherbakova, V.A., Chuvil’skaya, N.A., Rivkina, E.M., Pecheritsyna, S.A., Suetin, S.V., Laurinavichius, K.S., Lysenko, A.M. and Gilichinsky, D.A., 2009, Novel halotolerant bacterium from cryopeg in permafrost: Description of Psychrobacter muriicola sp nov. Microbiology, 78(1), 84-91.

Tarnocai, C., Canadell, J.G., Schuur, E.A.G., Kuhry, P., Mazhitova, G. and Zimov, S., 2009, Soil organic carbon pools in the northern circumpolar permafrost region. Global Biogeochemical Cycles, 23.

Tran, A.P., Dafflon, B., Bisht, G. and Hubbard, S.S., 2018, Spatial and temporal variations of thaw layer thickness and its controlling factors identified using time-lapse electrical resistivity tomography and hydro-thermal modeling. Journal of Hydrology, 561, 751-763.

Tran, A.P., Dafflon, B. and Hubbard, S.S., 2017, Coupled land surface-subsurface hydrogeophysical inverse modeling to estimate soil organic carbon content and explore associated hydrological and thermal dynamics in the Arctic tundra. Cryosphere, 11(5), 2089-2109.

Uhlemann, S., Dafflon, B., Peterson, J., Ulrich, C., Shirley, I., Michail, S. and Hubbard, S.S., 2021, Geophysical Monitoring Shows that Spatial Heterogeneity in Thermohydrological Dynamics Reshapes a Transitional Permafrost System. Geophysical Research Letters, 48(6), e2020GL091149.

Wainwright, H.M., Dafflon, B., Smith, L.J., Hahn, M.S., Curtis, J.B., Wu, Y., Ulrich, C., Peterson, J.E., Torn, M.S., et al., 2015, Identifying multiscale zonation and assessing the relative importance of polygon geomorphology on carbon fluxes in an Arctic tundra ecosystem. Journal of Geophysical Research: Biogeosciences, 120(4), 788-808.

Wainwright, H.M., Liljedahl, A.K., Dafflon, B., Ulrich, C., Peterson, J.E., Gusmeroli, A. and Hubbard, S.S., 2017, Mapping snow depth within a tundra ecosystem using multiscale observations and Bayesian methods. The Cryosphere, 11(2), 857-875.

Wainwright, H.M., Oktem, R., Dafflon, B., Dengel, S., Curtis, J.B., Torn, M.S., Cherry, J. and Hubbard, S.S., 2021, High-Resolution Spatio-Temporal Estimation of Net Ecosystem Exchange in Ice-Wedge Polygon Tundra Using In Situ Sensors and Remote Sensing Data. Land, 10(7), 722.

Wu, Y., Ulrich, C., Kneafsey, T., Lopez, R., Chou, C., Geller, J., McKnight, K., Dafflon, B., Soom, F., et al., 2018, Depth-Resolved Physicochemical Characteristics of Active Layer and Permafrost Soils in an Arctic Polygonal Tundra Region. Journal of Geophysical Research: Biogeosciences, 123(4), 1366-1386.

Yoshikawa, K., Romanovsky, V., Duxbury, N., Brown, J. and Tsapin, A., 2004, The Use of Geophysical Methods to Discriminate between Brine Layers and Freshwater Taliks in Permafrost Regions. Journal of Glaciology and Geocryology, 26, 301-309.