H<sub>2</sub>O and δD profiles remotely-sensed from ground in different spectral infrared regions
We present ground-based FTIR (Fourier Transform Infrared) water vapour analyses performed in four different spectral regions: 790–880, 1090–1330, 2650–3180, and 4560–4710 cm<sup>−1</sup>. All four regions allow the retrieval of lower, middle, and upper tropospheric water vapour amounts w...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Copernicus Publications
2010-11-01
|
Series: | Atmospheric Measurement Techniques |
Online Access: | http://www.atmos-meas-tech.net/3/1599/2010/amt-3-1599-2010.pdf |
Summary: | We present ground-based FTIR (Fourier Transform Infrared) water vapour analyses performed in four different spectral regions: 790–880, 1090–1330, 2650–3180, and 4560–4710 cm<sup>−1</sup>. All four regions allow the retrieval of lower, middle, and upper tropospheric water vapour amounts with a vertical resolution of about 3, 6, and 10 km, respectively. In addition the analyses at 1090–1330 and 2650–3180 cm<sup>−1</sup> allow the retrieval of lower and middle/upper tropospheric δD values with vertical resolutions of 3 and 10 km, respectively. A theoretical and empirical error assessment – taking coincident Vaisala RS92 radiosonde measurements as a reference – suggests that the H<sub>2</sub>O data retrieved at high wavenumbers are slightly more precise than those retrieved at low wavenumbers. We deduce an H<sub>2</sub>O profile precision and accuracy of generally better than 20% except for the low wavenumber retrieval at 790–880 cm<sup>−1</sup>, where the assessed upper precision limit of middle/upper tropospheric H<sub>2</sub>O is 35%. The scatter between the H<sub>2</sub>O profiles produced by the four different retrievals is generally below 20% and the bias below 10%, except for the boundary layer, where it can reach 24%. These values well confirm the theoretical and empirical error assessment and are rather small compared to the huge tropospheric H<sub>2</sub>O variability of about one order of magnitude thereby demonstrating the large consistency between the different H<sub>2</sub>O profile retrievals. By comparing the two δD profile versions we deduce a precision of about 8 and 17‰ for the lower and middle/upper troposphere, respectively. However, at the same time we observe a systematic difference between the two retrievals of up to 40‰ in the middle/upper troposphere which is a large value compared to the typical tropospheric δD variability of only 80‰. |
---|---|
ISSN: | 1867-1381 1867-8548 |