Thomas Stevens

Associate Senior Lecturer

Uppsala University
Country flagSweden

Research Interests

ApplyKite TwinLearns as you apply

Make your next move.

Prepare your application

Review drafts before you send them.

Explore related searches

About

Thomas Stevens is an Associate Senior Lecturer at Uppsala Universitet in Sweden. His research focuses on paleoclimate reconstruction, dust aerosol deposition, and sediment dynamics, as evidenced by his recent articles on glaciation impacts and climate proxies. He has contributed significantly to the understanding of loess and subaerial sediments, exploring their implications in various geographical contexts, including Central Asia and Svalbard. Dr. Stevens' work integrates geological and environmental data to enhance our knowledge of historical climate variations.

Articles (27)

Year:

2026

Year:

2026

Year:

2026

Greenland ice cores demonstrate that transient last glacial cold climate events (stadials) were associated with greatly enhanced atmospheric dust loading. Detailed radiocarbon dating of loess in central Europe reveals concurrent increases in dust activity in dust‐emitting regions. However, the causes of these changes in dust emission and the role of dust in rapid climate change remain unclear. We address this uncertainty through multi‐proxy analysis of loess sources in Hungary. Our results demonstrate that loess particles were dominantly produced by subglacial grinding processes in the eastern Alps. These particles were released along with stadial Alpine Ice Sheet‐driven meltwater pulses, turning major river systems into efficient dust sources for loess deposition in Europe. Concurrent strengthened anticyclonic circulation over northern Europe would have caused wider dispersal of Alpine Ice Sheet‐derived dust as far as Greenland. Resultant continental‐hemispheric scale changes in atmospheric dust loading likely amplified concomitant North Atlantic climate cooling and ice sheet decay during Greenland stadials.

Year:

2025

Environmental magnetism, including the use of magnetic susceptibility (MS), has formed the backbone of analyzing past terrestrial climate dynamics recorded in loess deposits world‐wide. However, the nature of MS signal response and frequency dependence ( χ FD ) varies between loess sequences, which can limit the applicability of the approach. Here, we explore how measuring MS using multiple alternating‐current field frequencies can transform our understanding of the past climate record in loess. We compare loess MS data measured at 15 different frequencies from diverse environments across the Northern Hemisphere, and provide high‐resolution data for a late Quaternary loess‐paleosol section in Tajikistan. Additionally, we study the magnetic mineral composition in selected Tajik loess samples using rock magnetic methods and Raman spectroscopy and assess the usefulness of calculating superparamagnetic nanoparticle (SP) size distributions from multi‐frequency magnetic susceptibility. Our results demonstrate that using this approach to determine χ FD has several advantages over widely used dual‐frequency approaches: (a) Inclusion of a wider grain size range of magnetism‐bearing SP particles in measuring χ FD , allowing for a more complete analysis of the environmental drivers behind the MS signal; (b) Higher sensitivity of the χ FD palaeoclimate proxy to climatic changes; and (c) Increased statistical meaningfulness of the χ FD proxy by allowing quantification of uncertainty. Our approach is particularly beneficial for understanding sites characterized by low‐susceptibility samples and potentially diverse processes of magnetic enhancement. However, we advocate its routine use even in more typical loess sequences due to its greater sensitivity to climatic changes and better understanding of inherent proxy uncertainties.

Year:

2025

Central Asia represents a key region for our understanding of early human dispersal patterns, because it served as a migration corridor that linked the Levant and southern Caucasus with Northeast Asia. However, no Early Paleolithic sites in Central Asia are anchored with reliable age constraints, including the thick loess-paleosol sections in Tajikistan that recorded early human activities and environmental changes over multiple glacial-interglacial cycles. This lack of absolute age constraints is presently the key factor limiting our understanding of the early human dispersal history. Here, we provide a new comprehensive description and the first absolute age framework for the Early Paleolithic Karatau Culture; defined by the rich lithic assemblages found in paleosols in the Khovaling Loess Plateau (KLP), Tajikistan. Age constraints are provided through multi-method analysis of three loess-paleosol sections in the KLP, combining luminescence ages, cosmogenic 26 Al- 10 Be concentrations, and magnetostratigraphic boundaries into a probabilistic inverse age-depth model. This model shows that the Karatau Culture flourished with the onset of Marine Isotope Stage (MIS) 15, thrived during MIS 13 and 11, but disappeared around onset of MIS 10. We frame the archaeological occupations within local and regional ecological settings to better understand the drivers of Pleistocene human migrations.

Year:

2025

Reconstructing past wind directions is critical in constraining synoptic scale climate regimes in the geologic past. Such wider scale understanding of past climate helps frame the climatic and environmental conditions that past human populations were exposed to. This is particularly important in regions such as Сentral Asia, at the interface of a range of climate zones. Palaeolithic tools are extensively embedded within the loess-palaeosol sequences of the Khovaling Loess Plateau, Southern Tajikistan, yet little is known about the environmental conditions these hominin groups faced.Magnetic techniques applied to loess-palaeosol sequences are particularly insightful for investigating dust transport dynamics and variations in the intensity of hydroclimate. Anisotropy of magnetic susceptibility (AMS) in particular measures the magnetic fabric, i.e., the orientation of magnetic particles in an oriented sample, which can provide insights into palaeowind directions and post-depositional processes. Here we apply AMS to three different loess-palaeosol sections of the Khovaling Loess Plateau, approximately spanning the past 1 Ma, in order to assess if this technique allows us to reconstruct palaeowind directions for this region of Central Asia.Our results reveal the first palaeowind directions reconstructed for the Khovaling Loess Plateau. We present AMS data from three sites on the Khovaling Loess Plateau and demonstrate the potential and limitations of this data for reconstructing palaeowind directions. We investigate the question to which degree local winds, responsible for loess deposition on the Khovaling Loess Plateau, are related to broader atmospheric circulation patterns. We also explore how AMS data can be combined with other palaeoenvironmental proxies, such as frequency-dependent magnetic susceptibility as an indicator of palaeoprecipitation, and how this technique can be used to assess sediment reworking.

Year:

2025

Aeolian parabolic dunes in Arctic Fennoscandia are widely scattered across diverse biomes from tundra to mountain birch and pine forest, with many systems in more sparsely vegetated locations still showing signs of current degradation or partial activity. Dune stratigraphy reveals that these dunes have also undergone multiple phases of Holocene activity and reactivation, following initial dune formation immediately post deglaciation. Multiple buried soil, charcoal and homogenous dune sand layers demonstrate that dune stability is repeatedly interrupted by multiple fire events and dune reworking episodes during the Holocene. As such, these systems have great potential as archives of Arctic landscape, climate and fire history in a highly sensitive environment, and over a range of biomes.However, few of these dunes, especially in Sweden, have been independently dated using modern luminescence and radiocarbon techniques in order to constrain the timing of phases of dune stability, fire activity and reworking. As such, there is limited understanding of whether fire frequency or landscape degradation episodes have changed over the Holocene, whether there are wider patterns in these events over Arctic Fennoscandia, and if so what the climatic or anthropogenic drivers of such changes may be. Luminescence dating of aeolian sands constrains the timing of dune sand movement (or rather the cessation of movement), allowing direct dating of reworking events, while radiocarbon dating of charcoal constrains the timing of fires that may have marked the end of periods of stability, and the initiation of dune degradation and reworking.Here we address this by applying detailed quartz optical stimulated luminescence (OSL) and feldspar post infrared - infrared stimulated luminescence (pIR-IRSL) dating to cross bedded (initial dune form) and homogenous sand (reactivation layers) units in multiple dunes in Arctic Sweden and Finland over a diverse range of biomes. We also apply detailed AMS 14C dating to multiple charcoal fragments from charcoal bands exposed in the stratigraphy of these dunes, which often separate dune reactivation sand units. In this way, not only can we cross check multiple independent chronological techniques, but we can also construct detailed local dune development and reworking histories across a wide area of Arctic Europe. Combining these histories allows testing of potential wider climatic and human drivers for fire and landscape changes in Arctic Fennoscandia over the Holocene. Our results show that the choice of best luminescence dosimeter mineral varies substantially by location, and that scattered charcoal fragments exposed in dune sands are often reworked from previous fires. However, our detailed, multi-technique approach allows development of detailed fire and reactivation histories for these dunes, and reveals substantial differences in these histories across biome type and location. Despite these differences, some wider trends are also apparent, notably a substantial increase in fire and dune activity after the Holocene Optimum and in recent centuries. We discuss possible causes for these changes.

Year:

2025

The postglacial landscape of Northern Fennoscandia is characterised by abundant depositional landforms, including eskers and aeolian sand dunes. Stratigraphic markers in sand dunes, such as buried soils or charcoal bands, record past environmental and climatic changes, preserving a history of landscape instability and sand movement caused by climate, fire, and land use changes. Although the shape and orientation of the dunes reflect the dominant wind direction during their formation, the mapping of aeolian sand dunes in Fennoscandia remains limited to a few studies and is generally not conducted at the polygon level. To systematically analyse the location, shape, and orientation of dunes over large areas, scalable and reproducible semi-automated mapping approaches based on remote sensing data are needed, with results validated in the field.While analysing and interpreting digital elevation models, derived hillshades, satellite imagery, and aerial photographs, we observed features we referred to as ‘double-dune ridges’; dunes appearing in direct proximity to each other that mimic each other’s shape and orientation. Ground-truthing fieldwork showed that the ridges are unlikely to represent two separate, quasi-parallel dunes but rather the remnants of single dunes where the crests have been eroded and excavated. In their present form, these mostly parabolic dunes can be described by stable, vegetation-covered windward (stoss) and leeward (lee) sides, with open to low pioneer vegetation covering areas where the former crest once existed.We discuss this example of dune degradation in Northern Fennoscandia and illustrate how an interdisciplinary approach can provide valuable insights into the dynamics involved. We strongly believe that integrating detailed mapping with chronostratigraphic, tree ring, and remote sensing data is essential for understanding the processes leading to the excavation of the former crest of dunes and, more broadly, to landscape evolution and degradation.

Year:

2025

Aeolian dunes are formed by the accumulation of wind-blown sand and its movement over time. While their shape may reflect wind directions during the period of formation, stratigraphic features in aeolian dunes record past climate and environments. In northern Fennoscandia, abundant dune fields serve as detailed records of Holocene arctic climate change. Specifically, buried soil and charcoal layers observed in mostly parabolic dunes preserve a rich history of past changes in climate, environment, fire history and land use. Understanding the evolution of the dunes in the face of these changes is crucial to project how Arctic environments respond to future climate change. Such studies require knowledge on the location, size, and shape of the aeolian sand dunes, which involves mapping them as polygon features. However, manual mapping of these landforms is very time-consuming. Semi-automatization can be used to develop a transferable, reproducible and scalable mapping approach, overcoming this issue.Here we utilize a semi-automatic method involving machine learning models to map aeolian sand dunes in two study areas in northern Finland. While we rely on manual mapping based on a 2 m digital elevation model (DEM), its hillshade and Google satellite imagery for establishing a training dataset, segmentation and classification of sand dune objects is carried out using DEM derivatives such as slope, convergence index, general curvature, and topographic position index through support vector machines and random forest algorithms. We validate the training dataset and mapping results during field campaigns. Our semi-automatic object-based mapping method enables a mostly correct identification of dune objects, including attribute information on their size, shape, and morphological characteristics, compared to our field investigation. False positives occur at locations of great similarity between parabolic dunes and other topographic features. Future steps include reducing the false positives and transferring the approach to additional areas to ultimately develop a method for automated, regional-scale mapping of aeolian sand dunes in Arctic Fennoscandia.

Year:

2025

Year:

2025

Year:

2024

Year:

2024

Year:

2024

Year:

2024

Year:

2024

Year:

2024

Year:

2023

Year:

2023

Year:

2022

Year:

2022

Year:

2022

Year:

2022

Year:

2022

Year:

2022

Year:

2022

Year:

2021

Collaborators (33)

Torsten Vennemann

Professor

University of Lausanne

SWITZERLAND
View details

Susanne Gier

Universität Wien

AUSTRIA
View details

Jesper Olsen

VIP, Associate Professor

Aarhus University

DENMARK
View details

mark bateman

Professor

University of Sheffield

UNITED KINGDOM
View details

Aske Lohse Sørensen

Aarhus University

DENMARK
View details

Jan-Pieter Buylaert

Senior scientist

Technical University of Denmark

DENMARK
View details

Kaja Fenn

Lecturer

University of Liverpool

UNITED KINGDOM
View details

Zdzisław Jary

professor

University of Wrocław

POLAND
View details

Dávid Molnár

Senior lecturer

University of Szeged

HUNGARY
View details

Dominika Niezabitowska

Institute of Geophysics, Polish Academy of Sciences

POLAND
View details

Mads Faurschou Knudsen

-

DENMARK
View details

Daniel Hölbling

Senior Scientist & Research Group Leader

University of Salzburg

AUSTRIA
View details

Johan Arnqvist

Associate professor

Uppsala University

SWEDEN
View details

Zhongshi Zhang

Peking University

CHINA
View details

Eduardo Garzanti

University of Milano-Bicocca

ITALY
View details

SERGIO ANDO'

Associate Professor

University of Milano-Bicocca

ITALY
View details

Andrew Murray

Professor

Aarhus University

DENMARK
View details

Frank Lehmkuhl

Prof. Dr

RWTH Aachen University

GERMANY
View details

László Makó

Assistant professor

University of Szeged

HUNGARY
View details

Urs Kloetzli

Professor

Universität Wien

AUSTRIA
View details

Grzegorz Adamiec

Associate Professor

Silesian University of Technology

POLAND
View details

Johanna Salminen

University of Helsinki

FINLAND
View details

Katja Bohm

University of Helsinki

FINLAND
View details

Lorena Abad

Lecturer

University of Salzburg

AUSTRIA
View details

B. Almqvist

Uppsala University

SWEDEN
View details

Peter Svedlindh

Uppsala University

SWEDEN
View details

Anu Kaakinen

Professor

University of Helsinki

FINLAND
View details

Hui Tang

Senior Researcher

Finnish Meteorological Institute

FINLAND
View details

Kathryn Fitzsimmons

Associate Professor of Physical Geography

MONASH UNIVERSITY

AUSTRALIA
View details

Lele Shu

Associate Professor

Chinese Academy of Sciences

CHINA
View details

David K. Wright

Associate Professor / Professor

University of Oslo

NORWAY
View details

Anna Bird

Senior Lecturer

University of Hull

UNITED KINGDOM
View details

Jacek Skurzyński

University of Wrocław

POLAND
View details