| Summary: | A better understanding of the water cycle has become even more crucial under the present condition of climate change. The stable isotopes of hydrogen and oxygen have been used for decades as powerful tracers to provide insights into the water cycle. While substantial understanding has been achieved, disputes remain on what processes set the observed isotope signal. Besides, albeit its great usefulness, no systematic isotope observations have existed in western Norway, which is a midlatitude location influenced by distinct weather systems (e.g. North Atlantic cyclones, cold air outbreaks) and swift precipitation formation. In this thesis, I present a systematic isotope observation of surface vapour and precipitation at the southwest coast of Norway between December 2016 and November 2019. The observation consists of high-resolution samplings for targeted weather events and long term (quasi-daily) routine samplings. To facilitate these observations, a stable water isotope laboratory with 3 laser spectrometers has been established. To ensure high-quality data acquisition, we thoroughly assess the instrument performance in many aspects. One important aspect is the correction of the mixing ratio dependency. In Paper I, we systematically investigate the mixing ratio dependency in a range from 500 to 23 000 ppmv. We find that the mixing ratio dependency systematically varies with the isotope composition of measured vapour. We refer this as isotope composition-mixing ratio dependency and have developed a scheme to correct for this dependency-introduced bias. Using in situ measurements from an aircraft measurement, we demonstrate the importance of the correction at low mixing ratios. Stability tests over up to 2 years indicate that the first-order dependency is a constant instrument characteristic that may be primarily related to spectroscopy. In Paper II, we present a case study of a 24-h land-falling "atmospheric river" event on 07 December 2016, with high-resolution paired measurements of near-surface vapour and ...
|
|---|