You must have a javascript-enabled browser and javacript and stylesheets must be enabled to use some of the functions on this site.
 
ESA
     
Introduction
 
envmail@esa.int






 

Odin-SMR measurements of tropical upper tropospheric water

Bengt Rydberg(1), Patrick Eriksson(1) and Donal Murtagh(1)

(1) Chalmers University of Technology, Chalmers University of Technology, SE-412 96 Gothenburg, Sweden

Abstract

Although upper tropospheric water vapour and cloud ice mass, and additionally the upper tropospheric water diurnal cycle, are keys in the prediction of future climate, the knowledge of these parameters in the present atmosphere is limited. This mainly as a result of that traditional sensors are not very adequate to measure these parameters. However, in the last years a series of microwave instruments (Odin-SMR, AURA-MLS, and CloudSat) capable of measuring upper tropospheric humidity and cloud ice have been launched. Odin-SMR is a passive limb-sounding radiometer operating at around 500 GHz and can measure both quantities. A new well qualified retrieval algorithm has been developed, which handle the formerly not very well controlled beam filling problem in a general way. The retrieval algorithm uses CloudSat data as a priori information on cloud structure variability on scales and dimensions not resolved by Odin-SMR measurements. Thus, the retrieval algorithm is a fruitful example on the indirect usefulness of CloudSat and upcoming EarthCARE measurements. A sophisticated retrieval characterisation showed that Odin-SMR can measure profiles of upper tropospheric humidity and ice water content with a vertical resolution of ~5km with a precision of ~20% and ~65% respectively, and with a conservative estimated accuracy of ~30% and ~50% respectively. The maximum measurement response, for both humidity (~0.7) and ice water content (~1), is found between 10.5-15km in altitude. Results of retrieved humidity and ice water content at altitude layers of high response have been successfully compared to AURA-MLS and CloudSAT respectively. Yearly mean humidity fields were shown to differ with less than 10%. Mean values of tropical ice water content agree within 50% and probability density functions of retrieved ice water content values compare well over the entire range of values. As Odin-SMR measurements of upper tropospheric water now are well characterised and have been successfully compared to AURA-MLS and CloudSAT it can be considered for further studies. The diurnal cycle of upper tropospheric water is examined by combining results from Odin-SMR, AURA-MLS and CloudSat. These instruments have fixed tropical observation times at 6:00 am/pm (Odin-SMR) and 13:40 am/pm (AURA-MLS and CloudSat) and each region in the tropics is observed with approximately 6 hour intervals. Strongest diurnal variations of cloud ice amount are found over land regions associated with strong convection, with a pronounced maximum in the afternoon. Over oceans the variations have a lower amplitude and the maximum is generally find during early morning. Obtained results can potentially be used to evaluate the diurnal cycle of convection and associated clouds in climate models, which tend to have phase errors of several hours over tropical land. These phase errors leads to an associated error in cloud amounts and accordingly downward solar radiation, both of which can have detrimental effects on the simulated mean climate of temperature, soil moisture and precipitation.