Tests of Tropospheric Models
Date
2024
Authors
Curtis, Simon André
Editors
Advisors
Reid, Iain
MacKinnon, Andrew
Hamilton, Murray
MacKinnon, Andrew
Hamilton, Murray
Journal Title
Journal ISSN
Volume Title
Type:
Thesis
Citation
Statement of Responsibility
Conference Name
Abstract
The work in this thesis is the use of high temporal resolution ground based measurements to the applications of validating the cloud products from new reanalysis models, and the study of the northern Australian Low-level Nocturnal Jet. These two topics are linked by their connection to tropospheric turbulence and boundary layer interactions. Numerical weather prediction models may be refined through the use of the reanalysis technique, by comparing the model predictions against actual measurements. Two recently released reanalysis models (ERA5 and BARRA-AD) are among the first to provide output data with hourly temporal resolution, so similarly high temporal resolution measurements are needed to validate them. Data from a cloud lidar operated at Adelaide, Australia is used to provide hourly cloud base heights and cloud cover fractions to effect this validation. This validation is valuable because it involves measurements obtained in the southern hemisphere, which is underrepresented in such validations. It also includes mid-level cloud base heights which are rarely used in validation studies. The addition of a synoptic classification database allows the model biases to be split into different weather conditions, providing direction into what processes are introducing these biases. These data are compared with the model outputs and assessed for seasonal and synoptic related patterns. The ERA5 model is found to slightly underestimate low-level cloud base height by around 100 metres in spring and summer, although it overestimates mid-level cloud base heights in autumn and winter. However, in autumn and winter, there is also a frequent erroneous prediction of cloud between 100 and 500 metres above ground level. This is found to correlate with cyclonic weather conditions. In contrast, overestimated mid-level cloud base heights are found to correlate with non-cyclonic (high pressure or frontal) conditions. The ERA5 model is also found to strongly underpredict low and mid-level cloud cover fractions all year round, with the mid-level cloud cover fractions being more strongly underestimated during cyclonic conditions. The BARRA-AD model is found to predict cloud base heights that are on average 200 metres too low all year round, except in the case of mid-level clouds during a warm front where the prediction shows no significant offset from the observations. The BARRA-AD model is also found to underestimate low and mid-level cloud cover fractions, but to a smaller degree than the ERA5 model. Large underestimations of low-level cloud cover fractions are found to correlate with rainfall producing weather systems. Applying different temporal and spatial averaging to the results of both models indicates that cloud structures are not accurately represented on an hourly basis, but this does not appear to be due to a clear offset in the predicted location/timing of the cloud. These results indicate that issues seen in previous models still exist in the case of ERA5, while the BARRA-AD model has mostly resolved this issue by calculating non-parameterised convection, this has introduced new biases. Five years of wind measurements from a wind profiler at Tennant Creek, Australia are used to study the northern Australian low-level nocturnal jet. Jet-like behaviour of the wind at Tennant Creek is found to occur on 41% of nights, strongly associated with the dry season. Analysis of these events finds three distinct phenotypes of jet and an automated identification algorithm is developed. Jet events that clearly fit with the accepted theory and behaviour are found to occur 9% of the time, while a phenotype that does not exhibit an inertial oscillation is found to occur 15% of the time. An unexpected result is that of an additional phenotype that exhibits just a rotation of the wind vector in the lower boundary layer, but no net increased in wind speed. This is found to occur 7% of the time, is more common in the wet season and associated with low pressure systems. This provides a more comprehensive analysis of the northern Australian low-level nocturnal jet and paves the way for further studies to deepen our understanding of the extent and behaviour of this jet. The jet phenotypes identified here point towards new mechanisms that could increase our understanding of boundary layer dynamics. Through these two areas of work, high resolution measurements have allowed for a more detailed analysis of complicated meteorological phenomena. This helps improve our understanding of them and points towards areas where our models and theories need refinement, along with where further understanding should be sought.
School/Discipline
School of Physics, Chemistry and Earth Sciences
Dissertation Note
Thesis (Ph.D.) -- University of Adelaide, School of Physics, Chemistry and Earth Sciences, 2024
Provenance
This electronic version is made publicly available by the University of Adelaide in accordance with its open access policy for student theses. Copyright in this thesis remains with the author. This thesis may incorporate third party material which has been used by the author pursuant to Fair Dealing exceptions. If you are the owner of any included third party copyright material you wish to be removed from this electronic version, please complete the take down form located at: http://www.adelaide.edu.au/legals