A new algorithm for cloud identification over the Arctic using AATSR/SLSTR and its application for ACLOUD/PASCAL campaigns
S. Jafariserajehlou, L. Mei, M. Vountas, V. V. Rozanov, & J. P. Burrows
Institute of Environmental Physics, University of Bremen, Otto-Hahn-Allee 1, 28359 Bremen, Germany
The remoteness of the Arctic region has led to lim-ited number and marginal coverage of ground-based measurements of geophysical parameters in this region and highlighted the role of space-borne observations to investigate parameters and feedback mechanisms which contribute to Arctic research. However, clouds represent one of the major sources of error in satellite-based retrievals of snow properties, aerosol, trace gases as well as cloud properties and etc. In fact, a precise cloud detection method as a prerequisite in satellite-based retrievals plays an important role in reliability of final results and could hamper the usage of them for further analyses.
In this study, a new cloud detection algorithm based on time-series measurements is developed and ap-plied to Advanced Along-Track Scanning Radiometer (AATSR), one payload on the European Environ-mental Satellite (Envisat). Furthermore, the de-veloped algorithm is successfully applied to the Sea and Land Surface Temperature Radiometer (SLSTR) onboard the Sentinel-3 platforms to estimate cloud properties during ACLOUD/PASCAL campaigns [Wendisch et al.,2018].
The main idea behind this method is that clouds have larger spatial variability and less stability com-pared to cloud free conditions. Therefore, the stability of ground scenarios in cloud-free conditions is char-acterized by Pearson Correlation Coefficient (PCC) values, calculated between the last measurement and time-series data.
One central aim of this work is cloud masking for aerosol retrieval over the Arctic. To avoid misclassi-fication of heavy aerosol loadings with cloud, PCC analysis has been designed for a wavelength which is affected little by aerosol particles, whereas the Top Of Atmosphere (TOA) reflectance is affected by clouds.
Furthermore, additional information from thermal in-frared channels of the above mentioned instruments
has been utilized to separate cloudy and cloud-free pixels to produce a cloud mask with 1×1 km2spatial resolution. Moreover, a simple land classification step is added to derive five surface types: snow, land, ice, cloud and ocean.
The results of applying this algorithm to case stud-ies over the Arctic region and the validation against
1. European Space Agency (ESA) standard cloud product from AATSR L2 nadir cloud flag, 2. One of existing methods based on clear-snow
spectral shape,
3. Surface synoptic observations (SYNOP), 4. Moderate Resolution Imaging
Spectrora-diometer (MODIS) are presented.
Acknowledgements
This work was supported by the SFB/TR 172 “Arc-tiC Amplification: Climate Relevant Atmospheric and SurfaCe Processes, and Feedback Mechanism (AC)3”
funded by the German Research Foundation (DFG, Deutsche Forschungsgemeinschaft: 40101200).
References
M. Wendisch, et al. [2018]: The Arctic cloud Puzzle:
Using ACLOUD/PASCAL Multi-Platform Obser-vations to Unravel the Role of clouds and aerosol particles in Arctic Amplification. to be submitted to Bull. Am. Meteorol. Soc.
S. Jafariserajehlou, L. Mei, M. Vountas, V. V. Roz-anov, J. P. Burrows [2018]: Cloud identification algorithm for aerosol retrieval over the Arctic using AATSR/SLSTR time series measurements. to be submitted to Atmos. Meas. Tech.
Understanding of polar atmospheric dynamics by measurements of surface air pressure using O2-band differential absorption radar
Bing Lin1, Qilong Min2, Steve Harrah1, Yongxiang Hu1, & Roland (Wes) Lawrence3
1NASA Langley Research Center, Hampton, VA 23681, USA;
2State University of New York at Albany, Albany, NY 12222, USA;
3National Institute of Aerospace, Hampton, VA 23666, USA
Polar regions have very high sensitivities on global environmental variations and could enhance extreme weather conditions in middle latitudes, which would significantly affect people’s daily life and public safety.
Continuous monitoring polar weather conditions is a key in understanding polar dynamics, improving weather predictions, and minimazing polar impacts.
Although many major meteorological variables have being measured over the regions from space, the key atmospheric dynamic variable, air pressure, can only be observed in very limited surface stations. There is a significant gap in the measurement of air pressure in various spatiotemporal scales: from small, local to regional and large and from hourly, daily to weekly and even longer ones. There is no operational space capability available for direct air pressure remote sens-ing over polar regions. This effort tries to develop a feasible active microwave approach that measures surface air pressure from space using a Differential-absorption BArometric Radar (DiBAR) operating at 50-55 GHz O2 absorption band for weather forecasts.
The measured data will enable numerical weather pre-diction models constraining their assimilated dynamic fields close to actual meteorological conditions and improving the weather forecasts of not only polar
re-gions but also the globe. For example, this approach will increase our knowledge on polar vortex dynamics, monitor their changes and variations in real-time, and predict their impacts accurately. Analyses show that with the proposed space DiBAR the errors in instant-aneous (averaged) pressure estimates can be as low as≈4mb (≈1mb) under all weather conditions.
NASA Langley research team has made substantial progresses in advancing the DiBAR concept and tech-nology since it developed a decade ago. The feasibility assessment clearly shows the potential of surface ba-rometry using existing radar technologies. The team has also developed a DiBAR system design, fabricated a Prototype-DiBAR (P-DiBAR) for proof-of-concept, conducted laboratory, ground and airborne P-DiBAR tests. The flight test results are consistent with the in-strumentation goals. Observational system simulation experiments (OSSEs) for space DiBAR performance based on the existing DiBAR technology and capab-ility show substantial improvements in weather pre-dictions. Satellite DiBAR measurements will provide an unprecedented level of the prediction and know-ledge on polar and global weather conditions. The development of the DiBAR concept will be presented.
Characteristics and genesis conditions of January polar lows: Microwave satellites, radiative transfer simulations and arctic system reanalysis
A. Radovan1, S. Crewell1, M. Mech1, & A. Rinke2
1University of Cologne, Institute of Geophysics and Meteorology, Pohligstr. 3, 50969 Cologne, Germany;
2Alfred Wegener Institute, Helmholtz Centre for Polar and Marine Research, Telegrafenberg A43, 14473 Potsdam, Germany
Polar lows (PLs), often called “hurricanes of the Arctic” are intense, high-latitude maritime cyclones that bring heavy precipitation, (mostly in the form of snow), and whose winds are above gale force. Their intense winds combined with large amounts of snow, can cause significant infrastructural damage to coastal communities and disruption of shipping routes. How-ever, their small horizontal scale (less than 1000 km) and short life time (sometimes only 3 h) makes them hard to predict. Therefore, improved understanding and prediction is of high importance. Satellite ob-servations in the microwave range that have a good coverage of the Arctic region offer high potential due to theirs sensitivity to snow. In this study, two such satellite instruments, namely Advanced Microwave Sounding Unit –B (AMSU-B) and Microwave Humid-ity Sounder (MHS) have been used. The investigation of PLs is done for the period of 12 years (January, 2000 – December, 2011) over which 33 January cases were reported. Arctic System Reanalysis version 1 (ASRv1) is used for the analysis of atmospheric
gen-esis conditions of PLs and compared with AMSU-B and MHS observations. For the latter, radiative trans-fer simulator called PAMTRA (Passive and Active Microwave Radiative Transfer Model) that is able to simulate microwave brightness temperatures (TB) in the 1 – 800 GHz range has been employed. We found that AMSU-B and MHS are performing well in representing the PLs, where channels around strong water vapour line, namely 183.31, ±1, ±3, ±7 and 190.31 GHz, are showing strong depression in PL con-vective cores. The depression at times can be more than 40 K for the 183.31 ±7 and 190 GHz channels.
Generally, simulations show good agreement with the AMSU-B and MHS observations, though not all cores of multi-low PL are resolved. Possible explanation for that could be coarser resolution of the ASR as well as the parameterization of the precipitation processes.
Furthermore, we investigate PL cases originating in different geographical area and the amount of snowfall they bring.