Michael Tjernström


Contact information

Phone: +46-(0)8-16 3110
Mobile phone: +46-(0)70-2056631
Email: michaelt@misu.su.se
Room: C652
Website: http://people.su.se/~tjern/index.html

Research interests

  • Boundary Layer and Mesoscale Dynamics
  • Arctic Studies
  • Climate Studies
  • Global Atmospheric Dynamics

Research projects

Boundary-Layer Meteorology
The atmospheric boundary layer, the layer of the atmosphere closest to the surface, is categorized by three-dimensional small-scale chaotic and highly non-linear motions (winds) – we call this turbulence. Although seemingly random, correlations between the vertical wind-speed fluctuations and fluctuations of horizontal winds, temperature and moisture fluctuations, averaged over time, constitute net vertical fluxes of momentum and of sensible and latent heat; this flux couples the surface to the atmosphere. In my research I am in particular concerned with two types of turbulence: turbulence under stably stratified conditions (increasing temperature with height; inversion) and turbulence in clouds, in particular in the Arctic (see Arctic Meteorology), generated by cloud-top radiative cooling. In stable conditions, turbulence is generated by wind shear (vertical gradients of wind-speed) and dissipated by both the stability and molecular dissipation (viscosity). This results in a negative feedback between turbulent mixing and the vertical gradients: Stability → reduced mixing → increased stability → even less mixing, etc.; this kills the turbulence very fast. Fortunately, the vertical wind-speed gradient also increases, which generates more turbulence. Complicating this is that as stability increases, part of the turbulent kinetic energy is transferred into wavy motions (buoyancy or gravity waves). The main problem is how to describe this in numerical models that cannot resolve these fine-scale motions directly. Here my research follow two paths: Resolving the interactions between turbulence and waves, and “Probabilistic Parameterization”, where the relationship between turbulent fluxes and vertical gradients is not deterministic, but instead determined by probability distributions.

Mesoscale Dynamics
The mesoscale is a spatial scale sufficiently large scale that the rotation of the Earth (the Coriolis force) cannot be neglected, but the flow is strongly ageostrophic (no balance between the wind and the horizontal pressure gradient). There are many types of mesoscale motions and some of these have very little in common, for example: Gravity waves (see Boundary-Layer Meteorology), sea and land breezes, ana- and katabatic flows, convection and low-level jets (sharp maximums in wind speed at a very low altitude). My research is particularly directed on low-level jets of two different types: nocturnal low-level jets associated with nocturnally increased static stability, and thus partial frictional decoupling, and coastal low-level jets. In the first case, the increased wind-speed gradient gives rise to additional turbulence which can be an important source of mixing in the stable boundary layer (see Boundary-Layer Meteorology). In the second case, the low-level wind speed maximum is driven by the land/sea temperature contrast, where land is warmer than the marine layer. The geostrophically adjusted response to such a contrast is an along-coast flow, with the warm land to the left on the northern hemisphere. If there is substantial coastal terrain as high or higher than the depth of the marine layer this flow is enhanced; a semi-geostrophic flow develops where the alongshore wind speed can become very strong. As this strong wind band passes a coast with capes and bays, hydraulic phenomena occurs that can lead to winds well above 30 ms-1 only 50-100 m above the surface.

Arctic meteorology
The Arctic is experiencing the fastest climate change on Earth; the annually averaged Arctic near-surface temperature is increasing 2-4 times faster than the global average temperature and the summer sea ice is disappearing fast. In a global sense, the Arctic climate us determined by the balance of heat transported into the Arctic from farther south by atmosphere and ocean and heat lost to space by longwave radiation inside the Arctic. Locally, however, the melting of the sea ice is determined by the surface energy balance and the leading terms here are the radiation fluxes, and those are very strongly affected by clouds. Low-level clouds dominate the Arctic; in summer it is cloudy ~90% of the time and in winter 50-70%, but models have a very hard time describing this. The clouds strongly affect the vertical structure of the boundary layer. Cloud free winter conditions leads to strong surface inversions, with large static stability (see Boundary-Lauer Meteorology), while cloudy conditions give rise to a shallow well mixed structure which is very similar to summer conditions. Contrary to other climate regimes, low clouds warm the surface. This is because longwave radiation dominates. In winter the sun is of course below the horizon, but even during much of the summer the longwave radiation rules since the surface reflectivity is about as high as the cloud reflectivity. In my research I mostly use field experiment data from summertime icebreaker based expeditions to the central Arctic Ocean but also some modeling.


I have been teaching most of the core courses in dynamic meteorology over the years, including Dynamic Meteorology, Mesoscale Meteorology and Boundary-Layer Meteorology; the two latter I have also taught at the graduate level. I have also recently been teaching a number of overview courses, on Meteorology and on Climate. I also frequently give public lectures on climate and climate change, and especially on the Arctic.

Publications and CV

 CV: Michael Tjernström (438 Kb)