I'm a postdoc in the Department of Physical Geography and Ecosystem Science at Lund University. My research interests include climate variability and change, atmospheric dynamics, data assimilation, and inverse modeling of CO2 and related gases. At the moment my research focuses on developing methods to quantify fossil fuel emissions using mainly satellite observations.
|2013–2018||Ph.D. in Meteorology and Atmospheric Science, Department of Meteorology and Atmospheric Science, The Pennsylvania State University, University Park, Pennsylvania, United States.|
|2010–2012||M.S. in Atmospheric Sciences, Oceanography and Climate, Department of Meteorology, Stockholm University, Stockholm, Sweden.|
|2007–2010||B.S. in Meteorology, Department of Meteorology, Stockholm University, Stockholm, Sweden.|
More information can be found in my CV (PDF).
I'm interested in better our understanding of the climate system using a combination of observations, numerical models, and data assimilation methods. In particular, I want to understand what processes drive climate variability on intraannual to multidecadal time scales, and how climate change affects these processes. The research topics I am interested in include:
Our knowledge of carbon dioxide (CO2) emissions and removal by sinks is limited, largely due to a lack of observations of CO2 fluxes between the Earth's surface and the atmosphere. This information is crucial to better understand how the climate will change in the future. Furthermore, as the world's nations are moving towards international agreements to limit greenhouse gas emissions, it has become evident that we need to be able to verify the self-reported emission estimates by countries using independent data and methods.
In this project I will develop a data assimilation system based on the Ensemble Kalman Filter to constrain the CO2 fluxes in the contiguous United States by assimilating observed atmospheric CO2 concentration. Using this system, we will be able to assess the impact of different assumptions and sources of errors on the estimated CO2 fluxes, and answer questions such as: How much do uncertainties in the atmospheric transport contribute to the total uncertainties? Are we able to constrain CO2 fluxes on a regional scale using our current observations of CO2 concentration? How much do we benefit from satellite observations in terms of constraining the fluxes?
This work is part of the NASA-funded ACT-America project.
The aim of this project was to objectively classify different regions in the Arctic based on the interannual variability of sea ice concentration in each region, and to quantify the regional sea-ice extent variability and trends in each Arctic sub-region. We used Self-Organizing Map clustering to classify the Arctic sea-ice cover into nine regions in autumn and six regions in early winter. The regional sea-ice extents were detrended using an adaptive nonlinear method based on the Ensemble Empirical Mode Decomposition technique, which resulted in a better fit than a linear trend. Finally, we investigated the statistical relationship between the regional sea-ice variations and weather patterns in winter and found that sea-ice reductions in the Barents-Kara Seas and the Beaufort Sea are both related to unusually cold winters over the Eurasian continent, but are associated with distinctly different wintertime circulation patterns.
In this work we asked the question, can we find a robust atmospheric response when we gradually reduce the Arctic sea-ice cover? To find an answer, we ran a large number of experiments in a general circulation model with systematically perturbed Arctic sea-ice cover, and diagnosed the changes in the jet stream, planetary wave amplitude, teleconnection patterns, and extreme cold events over the continents. The results show that reductions in the Arctic sea-ice cover did not produce detectable changes in our metrics of the jet stream position and planetary wave amplitude, even in the simulations with the most severe sea-ice reduction that resembles the record-low Arctic sea-ice cover in September 2012. Some experiments show a statistically significant response in the frequency of large-scale circulation patterns, with a shift towards a circulation pattern that resembles the negative phase of the Arctic Oscillation, and more frequent extreme cold winter days in eastern Asia. However, the changes are not found in all experiments with reduced Arctic sea-ice cover and do not scale linearly with the sea-ice anomalies. This non-robust behavior of the atmospheric response to Arctic sea-ice loss suggests that the link between Arctic sea-ice anomalies and mid-latitude weather patterns is complex and depends on other factors, such as the state of the climate.
See Chen et al. (2016, Journal of Climate) for more information.
This project is based on my Master's thesis and examines the robustness of an atmospheric mode known as the Barents Oscillation (BO). The BO has a meridional structure and could be important for the heat transport into the Arctic, but has also been speculated to appear purely as an artifact from the Empirical Orthogonal Function (EOF) analysis due to a spatial shift of the leading mode, the Arctic Oscillation. In this work we investigated if the BO can be considered a real physical mode of variability, or if it is an artifact of the EOF analysis. We found that the BO is a robust mode that can be found even during time periods when the Arctic Oscillation was relatively stable, which suggests that the BO is not an artifact. On the project page you can find more information about the BO and obtain the BO index.
When trying to quantify the impact of climate variation and climate change, it is common to use a single scalar variable such as surface temperature. However, it is not given that a change in e.g. surface temperature will have the same impact everywhere; some ecosystems are relatively stable, while others are more sensitive to changes in the regional climate. This project explores a different approach using the Köppen climate classification to quantify the impact of climate variations and change on ecosystems in different regions and over various time scales. The results show that the Köppen classification can be used as a diagnostic tool to monitor changes in climatic conditions that are relevant for ecosystems.
Here I have listed various free software I have found useful in my research.