• Introduction
• Method
• Discussion
• Conclusion
• References

FIGURES

• Figures 1a, 1b, 1c
• Figures 2a, 2b, 2c

Introduction

Gravity waves in the atmosphere can be generated by deep convection as well as by orography. High resolution radiosonde data form the basis of a study to find whether convectively generated gravity waves are energetically important in the midlatitudes.

Several years of data from stations in the UK are used. Convective available potential energy (CAPE) is used as an indicator of the likely occurrence of deep convection. Gravity waves are detected through their effect on the ascent rate of radiosondes (typical vertical velocity perturbations are 1 ms^-1) and by examining the hodograph (u-v diagram) using the Stokes parameter technique for finding the polarization properties. Cases of correspondence between high values of CAPE and the existence of waves in the stratosphere are described. Shear is determined for the ascents. In the resulting climatology it is found that high values of CAPE occur in about 10% of all the gravity wave cases in the radiosonde ascents.

The energy content of gravity waves is determined using a spectral analysis. Using the total data set it is found that the energy content of gravity waves on average is NOT as important during convection as during strong shear events. This can also be seen in the associated calculations of the momentum transfer.

These two results suggest that convectively generated gravity waves represent a physical process in the midlatitudes that should not be neglected.

For the spectral analysis the high resolution radiosonde data was analysed following the work of Allen and Vincent (1995) and Vincent et. al. (1997). Climatologies of energy and of direction of propagation in both the horizontal and vertical directions which can be determined from high resolution radiosonde data are a valuable tool to do this.

Figure 1 : Energy density for high CAPE cases (solid) and low CAPE cases (dashed), for the total ascents for Camborne from 1990-1996 and for high (solid) and low (dashed) shear cases.

Figure 2 : Energy density for high CAPE cases (solid) and low CAPE cases (dashed), for the total ascents for Camborne from 1990-1996 and for high (solid) and low (dashed) shear cases.

Method

Criteria for inertial gravity waves, gravity waves, shear and convection were developed to automate the analysis of radiosonde ascent data. Relationships of the different criteria to each othere were considered. Positive identification of shear as an inertial gravity wave generation mechanism is possible if the origin of the waves is indicated by the Stokes parameter method and when Stokes parameters indicate a rotation direction of the wind velocity vector opposite for the troposphere and the stratosphere. If there are waves travelling away from a jet then shear can be related to inertial gravity waves in the climatology. Positive identification of shear as a generator of gravity waves is more difficult. Many of the case studies indicate the presence of shear in conjunction with CAPE. In this case, the relative importance of the two effects is difficult to measure. For this reason, the climatological study of the following section includes the analysis of the subsets of data where CAPE and shear are mutually exclusive.

Necessary conditions for the identification of convection as dominant wave generation mechanism are an unstable troposphere where there is a potential for convection to happen. A method to do this is using the CAPE. Also there must be stable regions present in the atmosphere where gravity waves can develop. Since moderate convection exist in a great number of ascents there will be cases where this condition is fulfilled. If cases of gravity waves are detected and a high or moderate CAPE and the other wave generating mechanisms can be excluded then there is a high probability that these waves are caused by convection. However, other mechanisms often cannot be excluded, as has been found in case studies (not shown here).

A thunderstorm case study has shown that there is a link between CAPE and convection. Here the thunderstorm case was identified independently using sferics and a high level of CAPE detected simultaneously. Much more work has to be done to quantify this link. However the case studies have indicated an intrinsic problem with an automated climatology based on radiosonde ascents: the lack of an indicator to exclude orography or synoptic scale wave generation.

TG shows when the atmosphere can support waves but does not explain the cause. With the TG analysis, stationary waves are indicated if solutions with the observed wavelength exist. Stationary waves must be orographic as long as any convection or other generation mechanism did not produce stationary waves. Convective sources of stationary waves are possible as is suggested by the thunderstorm case. It is however not possible to apply the TG analysis to over 30000 profiles since it requires manual interpretation and can therefore only be used for individual cases.

An important wave generation mechanism, orography, cannot be identified automatically for the climatology. Therefore it has to be remembered that the wave cases may include orographically generated waves.

For the spectral analysis of the radiosonde data the results for the determination of the different criteria have been used to produce the energy density plots.

Discussion

Atmospheric waves have many generation mechanisms. Shear and convection are amongst many others like mountains, turbulence fronts or geostrophic adjustment. In this chapter techniques have been described to indicate the existence of both gravity and inertial gravity waves and different wave generating mechanisms in radiosonde ascents. The association of wave generating mechanisms with gravity waves from radiosonde ascents is a very difficult task due to noise in the data and the uncertainty in the mechanisms.

These techniques have been combined in a new algorithm which allows an automated analysis of wave events. A significant innovation is the use of the Stokes parameter method. The Stokes parameter spectrum has been developed for the use in single ascents and was used here to routinely find inertial gravity waves in radiosonde ascents.

35261 profiles were used for the analysis. Chi-square analysis introduced significance to the indicators and a method has used to describe associations between those indicators.

The results of the climatological analysis are illuminating. Most important of these results are

1. Convection (as identified using the CAPE criterion) is present
   • with tropospheric gravity waves (as identified using the ascent rate criterion) in 4.20% of 8898 summer profiles and in 5.88% of 9530 winter profiles.
   • with stratospheric gravity waves in 1.56% (summer) and in 9.08% (winter).
   • with tropospheric inertial gravity waves (as identified using the Stokes parameter criterion) in 6.05% (summer) and in 6.77% (winter).
   • with stratospheric inertial gravity waves in 6.19% (summer) and in 8.58% (winter).
2. The chi-square analysis shows that convection is significant only for tropospheric gravity waves and for stratospheric inertial waves in summer and gravity waves in both the troposphere and the stratosphere but not for inertial gravity waves in winter.
3. An upper limit to the % of gravity waves produced by convection can be estimated from the % of cases where significance indicators occur together to be 4.20% in summer and 9.08% in winter. These percentages should be compared to the total gravity wave cases: 16.14% in summer and 57.49% in winter. The proportion of gravity waves for which convection could be an important generating mechanism is therefore: 26.0% in summer and 15.8% in winter.
4. Convection and gravity waves appear together more frequently in winter than in summer, but convection and inertial gravity waves have no seasonal trend.
5. There is no latitudinal trend for gravity waves or inertial gravity waves together with convection, even though there is for convection alone.
6. Chi-square shows that shear is significant for all types of waves apart from inertial gravity waves in the stratosphere due to stratospheric shear. In summer it is true apart from stratospheric inertial gravity waves due to tropospheric shear and in winter it is true for tropospheric inertial gravity waves due to tropospheric shear and stratospheric inertial gravity waves due to stratospheric shear.
7. Maximum numbers of shear generated inertial gravity waves are 20% in the stratosphere and 18% in the troposphere. For short period gravity waves it is 16% in the stratosphere and 15% in the troposphere.
8. Both gravity waves and inertial gravity waves appear much more frequent in winter together with shear than in summer.

For the interpretation of the above results the following uncertainties must be borne in mind:

1. CAPE as an indicator for a convective wave-generating mechanism:
   • Relationship between CAPE and convection needs to be quantified through further work
   • Spatial distribution of CAPE: Strong convection is localised and radiosonde ascents may not be indicative of important regional convective activity
   • Temporal relationship with waves: Uncertainty in the wave generating mechanism means that a temporal delay between the occurrence of convection and that of waves may be important.

   These points may lead to an underestimate of convection.

2. Lack of indicator for orographic wave generation: It is possible that where convection is present, and shear is present, waves detected can nevertheless be of orographic origin. Orographic origin, however, can only be excluded by individual case studies and could therefore not be included in this study.
3. Choice of thresholds: The threshold values were chosen to be high enough that there would exist cases for each of the categories. It turns out that seasonal changes are well illustrated by this choice of threshold values.
4. The chi-square significance measure allows only a quantitative interpretation.

Despite these uncertainties, the study gives important new information about the incidence of gravity waves in the UK, and their relationship with local environmental factors. The indication from this study is that convection may play an important role in gravity wave production even in the mid-latitude winter.

Conclusion

• Gravity waves are much more common in winter
• CAPE is not the dominant mechanism for gravity wave generation
• Convection does not generate inertial waves.
• Shear has been shown to be more important than convection in wave generation.

References

S. J. Allen and R. A. Vincent (1995) Gravity wave activity in the lower atmosphere: seasonal and latitudinal variations, Journal of Geophysical Research 100(D1):1327-1350

R. A. Vincent, S. J. Allen and S. D. Eckermann (1997) Gravity wave parameters in the lower stratosphere in Gravity wave processes, Springer-Verlag, ed. K. Hamilton, pp 7-26