Bernet, Leonie (2020). Detecting trends of stratospheric ozone and tropospheric water vapour at mid-latitudes using measurements from multiple techniques. (Thesis). Universität Bern, Bern
|
Text
20bernet_l.pdf - Thesis Available under License Creative Commons: Attribution (CC-BY 4.0). Download (34MB) | Preview |
Abstract
This thesis investigates and quantifies changes in stratospheric ozone and tropospheric water vapour at mid-latitudes since the mid-1990s. Recent studies have shown that estimates of such changes from various ground-based measurement techniques are not always consistent. A possible reason for these differences may be inhomogeneities in the data. Data inhomogeneities arise from modifications in the instrument setup, measurement failures, problems or adjustments in the calibration and retrieval procedures, or from temporal sampling biases. To explain differences in observed changes, data inhomogeneities have first to be identified by intercomparing various datasets. In a second step, the inhomogeneities can be considered in the trend estimation to obtain optimal estimates of the true changes. This thesis aims to obtain more consistent trend estimates of stratospheric ozone profiles and tropospheric water vapour at mid-latitudes. For this purpose, we compared ozone and integrated water vapour (IWV) time series from various measurement techniques. The observations were intercompared to identify anomalies, biases, and inhomogeneities in the data. Trends in recent decades were then estimated by considering these irregularities in the trend estimation. To this end, two advanced trend analysis methods were tested and applied on the data. The trend models use the full error covariance matrix of the observations, which can be adapted to account for data correlations and inhomogeneities. We used stratospheric ozone observations measured by ground-based microwave radiometers, lidars, and ozonesondes, as well as satellite and reanalysis model data.We found good agreement between various ozone datasets. However, we also identified some anomalies and inhomogeneities in the ozone data and showed that they affect the trend estimates. Stratospheric ozone trend profiles are presented for northern (central Europe) and southern mid-latitudes (New Zealand). In both hemispheres, we observe a recovery in ozone concentrations in the upper stratosphere after the turn-around of ozone-depleting substances (ODSs) in 1997. We found trends that generally lie between 1% and 3% per decade, providing a confirmation of ozone recovery in the upper stratosphere as expected from reduced ODS emissions. In the lower stratosphere, we found inconsistent trends, suggesting that further research on lower-stratospheric ozone changes is required. Observations of IWV were used from a microwave radiometer, from a Fourier-transform infrared spectrometer, from a network of ground stations of the Global Navigation Satellite System (GNSS), and from reanalysis model data. We present trends derived from these IWV measurements in Switzerland. They show that IWV increased by 2% to 5% per decade since 1995, which is generally consistent with rising temperature. Also, we show that the advanced trend model used is well suited to reduce trend biases caused by GNSS-antenna updates. In conclusion, this thesis presents optimized trends of stratospheric ozone and IWV for recent decades at mid-latitudes. Further, it helps to better understand inconsistencies between trend estimates from multiple techniques by investigating the effect of data irregularities on the trends. Consequently, the results of this thesis contribute to a better understanding of ozone and water vapour changes in a changing climate.
| Item Type: | Thesis |
|---|---|
| Dissertation Type: | Cumulative |
| Date of Defense: | 13 November 2020 |
| Subjects: | 600 Technology > 620 Engineering |
| Institute / Center: | 08 Faculty of Science > Institute of Applied Physics |
| Depositing User: | Hammer Igor |
| Date Deposited: | 12 Jan 2021 17:43 |
| Last Modified: | 13 Nov 2021 01:30 |
| URI: | https://boristheses.unibe.ch/id/eprint/2410 |
Actions (login required)
 |
View Item |