Andrew Russell's CSIP Research

Andrew Russell

Summary of Russell et al. (2008)

Convective inhibition beneath an upper-level feature

This page represents a simple summary of some of the work that I have done looking at the upper-level influences on convective initiation in the UK. I have tried to use as little jargon as possible but, where there is a technical term, I have provided a link to the relevant page from Wikipedia. Please feel free to contact me with any questions or requests for further information.

This study considers the events that occurred on the 15th June 2005. The day was characterised by the passing of a cold front in the morning and the development of an isolated cluster of thunderstorms near to Oxford at around 1200hrs. The radar image to the right, showing rainfall intensity at 1200hrs, indicates the rainfall from the front to the east of the UK and, more importantly, the small, intense rainfall from the thunderstorms near Oxford (click on the image to enlarge it). The questions we ask here are: what role did the upper-level features play in iniating and/or inhibiting the development of this storm and; why was there only an isolated cluster of storms? A second paper has been written on this event (Morcrette et al. “Combination of mechanisms for triggering of an isolated thunderstorm: a case study of CSIP IOP 1”) which considers the role of the surface features in initiating the convection but here I am focusing, in particular, on the upper-level influences.

When meteorologists talk about "upper-level" features, they are generally referring to things that happen in the tropopause region - this is where the troposphere (the lowest part of the atmosphere where vertical motion is common and subsequently most "weather" occurs) meets the stratosphere (the dry part of the atmosphere above the tropopause, which is vertically stable). The tropopause is usually found at around 10km in the mid-latitudes and is, usually, quite level. However, as the plot to the left shows, the tropopause on the 15th June 2005 was not level - there was a "tropopause depression" - and there was a "tropopause fold" i.e. a region where the tropopause folds back on itself. As we shall see below, such stratospheric features can have significant impacts on our weather. This image was taken by an instrument called the Mesosphere-Stratosphere-Troposphere (or MST for short) radar located at Aberystwyth. The MST radar can identify areas where humidity changes in the atmosphere and, thus, it can be used to identify the tropopause and where stratospheric air penetrates into the troposphere (e.g. the fold).

The figure to the right shows a plot of potential temperature as measured from a series of weather balloons launched from Larkhill in the UK on the 15th June 2005. Put simply, potential temperature is a quantity derived from temperature and pressure that allows us to more easily compare air parcels from different heights in the atmosphere. What is most important in our profile of potential temperature is looking at the effect of the lowered tropopause, which can be seen at around 1000UTC and 300hPa. We can see that beneath the tropopause depression the lines potential temperature curve upwards - this is interpreted as a decrease in the stability of the troposphere and makes convection more likely, as also shown by the shaded regions where it has been calculated that there is energy available for this convection. There is also evidence here of the fold: the dry air behind the tropopause depression as recorded by the weather balloons and GPS equipment.

Given that this energy was available and that the tropopause depression was a relatively large feature, why was convection not more widespread? Looking in more detail at the the previous figure, we can also see that there was a dry layer (shown by the dash-dot contours) inhibiting the vertical motion of the air. Indeed, the plot to the left, which was taken from a small "wind-profiling" radar that also works by showing regions where moisture and potential temperature change, nicely shows this layer of dry air halting the convection. Such layers of dry air are sometimes called a "lid" as they act like the lid of, say, a pressure cooker and stop the vertical motion.

The extent of this dry lid can be seen in the image to the right, which shows a satellite image and moisture data from the Met Office's weather model. The dry air swept northward over the southern coast of England, eventually covering a large area. As this feature had such an impact on the weather this day, we next consider where the lid came from.

In order to do this, we use a tool called a back trajectory. Using wind data from atmospheric models, the back trajectory calculates where air parcels have come from and are quite accurate for periods up to about a week. The figure to the left shows back trajectories for both the dry lid and the tropopause depression with its associated fold. It can quite clearly be seen that both features travelled over the Labrador Sea, albeit at different times. Was this merely a coincidence?

We think that this was not just a coincidence but, in fact, is an interesting example of how the dyanmics and origin of the tropopause depression that helped to reduce tropospheric stabilty and encourage convection was intwinned with the lid that was largely responsible for inhibiting the convection. To back this claim up, the plot to the right (which essentially shows changes in the tropopause height over the Atlantic region) indicates that the two features we have discussed here both came from the same southward flow of stratospheric air into the troposphere over the Atlantic. In the five days before the storm near Oxford, we can see that depression in the tropopause that eventually travelled over the UK came from a larger, and unusually far south, depression in the tropopause. Such features are relatively common in the atmosphere and, with a little imagination, resemble the shape of a breaking wave in the sea. By comparing the previous figure to this one, we can see that the dry "lid" that moved over the UK on the 15th June 2005 came from the Labrador Sea region near the breaking wave: it descended from the height of the stratospheric feature at that point. We believe that the air that formed this lid flowed down the breaking wave feature and then relatively slowly travelled over the Atlantic to arrive at the same time over the UK as the tropopause depression. This is an interesting detail of atmospheric structure that has not previously been identified.

This work has been submitted with the title "Convective inhibition beneath an upper-level potential vorticity anomaly" to the Quarterly Journal of the Royal Meteorological Society for publication with Russell, A., Vaughan, G., Norton, E. G., Morcrette, C. J., Browning, K. A. and Blyth, A. M. listed as the authors. All of the pictures shown above (apart from the rainfall radar image) were taken from this paper and the abstract appears below:


Upper-level potential vorticity (PV) anomalies have the effect of reducing the convective stability of the troposphere through their impact on the vertical potential temperature profile, thus reducing convective inhibition (CIN) and increasing convective available potential energy (CAPE). Here, by contrast, we show the impact of a layer of stable air, which was intrinsically linked with a PV anomaly, that increased CIN; this layer descended and tracked beneath the small upper-level PV anomaly, which in this case was a shallow upper-level trough. Furthermore, it was found that this low-humidity, high-PV layer originated in the tropopause fold generated by the breaking Rossby wave that also produced the upper-level PV anomaly two days later. The CIN produced by this dry layer/lid was responsible for halting the development of widespread convection during the case presented here from the Convective Storm Initiation Project (CSIP).

Email: andrew.russell at