Feuchte Schichten in der freien Troposphäre - Untersuchung eines blinden Flecks im globalen Beobachtungssystem mit HALO während EUREC4A
Zusammenfassung der Projektergebnisse
The question of whether passive satellite observations can detect and vertically resolve Elevated Moist Layers (EMLs) was investigated. EMLs are significant mid-tropospheric humidity anomalies that emerge from detrainment of moist air from thunderstorm clouds in the tropics in heights of about 5 km, where the atmosphere is typically dry. Increased moisture in these altitudes has strong implications on the local radiation budget due to the greenhouse effect of water vapor. This can affect the formation of thunderstorm clouds, how intense they are and how long they persist. Previous work suggested that EMLs are not uncommon in the tropical atmosphere, in particular in the inner tropics where airmasses are on average rising. Stevens et al. (2017) suggested that EMLs may be inherently difficult to detect using passive satellite observations, denoting a possible reason for why they have up to now not received more attention. The first scientific goal addressed in the project was to establish what makes EMLs particularly challenging to resolve in satellite based retrievals and assess whether the cause for this difficulty can be resolved. For this purpose, the retrieval setup of Stevens et al. (2017) for the IASI satellite instrument (Infrared Atmospheric Sounding Interferometer) was reproduced and their results regarding the EML observed during the NARVAL-2 measurement campaign were reproduced, with the result that the EML is not captured by the satellite retrieval. In our first publication in Atmospheric Measurement Techniques we show that by introducing more independent temperature information to the retrieval that is available from the spectral band of CO2 covered by IASI, the EML can in fact be resolved well. The key here is that for the water vapor retrieval an assumption about the temperature structure is required. EMLs go along with temperature inversions that, if not accounted for, can mask the spectral signal of the moist layer. Having shown that EMLs can in principle be detected by passive satellite observations, we introduced a new method for identifying and characterizing EMLs in our first study, where we test the method on model-based retrievals where the reference atmospheric state is perfectly known. When applying the method to a test dataset of tropical ocean atmospheres, we find that the retrieval detects about 80 % as many EMLs as present in the reference dataset and that EML strength and thickness are underestimated by about 17 % and 15 %, respectively. This analysis sets an upper limit to what we would expect operational satellite retrievals to be able to resolve in terms of EMLs. In summary, we find that there is no inherent limitation in passive satellite based observations to resolve EMLs. In a second study, published in Atmospheric Chemistry and Physics we evaluate how operational satellite retrievals and reanalysis data resolve EMLs. We do so by collocating these datasets with 2146 radiosondes in space and time. The radiosondes were launched from Manus Island over the course of 4 years. By applying the method for identifying and characterizing EMLs to the different collocated datasets, we find that reanalysis data shows the best agreement with the reference regarding EMLs and highlight different issues of the satellite retrieval products that motivate algorithm refinements and revisits of post-processing steps for the products in the future.
Projektbezogene Publikationen (Auswahl)
-
Are elevated moist layers a blind spot for hyperspectral infrared sounders? A model study. Atmospheric Measurement Techniques, 14(11), 7025-7044.
Prange, Marc; Brath, Manfred & Buehler, Stefan A.
-
Elevated Moist Layers: Exploring a Blindspot in the Global Satellite Observing System, PhD thesis, Universität Hamburg, Hamburg, 2022
Prange, M.
-
How adequately are elevated moist layers represented in reanalysis and satellite observations?. Atmospheric Chemistry and Physics, 23(1), 725-741.
Prange, Marc; Buehler, Stefan A. & Brath, Manfred
