Dear Colleagues, you receive this eMail because you are either a user of the radiative transfer package libRadtran or because we think that you might be interested in this information. Should you not be interested in receiving further information, please let us know! Attached to this eMail is the 13th libRadtran Newsletter. The main issue of this Newsletter is to announce the new release libRadtran-1.6-beta The beta indicates that things are continuously changing and we hope to reach a stable attractor in the near future. Best wishes, Bernhard Mayer. /*---------------------------------------------------------------- * libRadtran Newsletter No. 13 * * March 4, 2011 * * Bernhard Mayer (bernhard.mayer_at_lmu.de) * Claudia Emde (claudia.emde_at_lmu.de) * Robert Buras (robert.buras_at_lmu.de) * Arve Kylling (arve.kylling_at_gmail.com) * Tim Dowling (dowling_at_louisville.edu) * * ### More info: http://www.libradtran.org ### *----------------------------------------------------------------*/ The main issue of this Newsletter is to announce the new release libRadtran-1.6-beta which will be remembered as the "Christmas? No. Groundhog? Hmm, probably meteorological spring? Whatever!"-release. The beta indicates that things are continuously changing and we hope to reach a stable attractor in the near future. But there have been that many news and improvements that we decided to release rather than waiting futher two years until stabilization is possibly reached. *** Major changes: * You might notice the beta in the version number. This has to do with the huge number of changes and new features introduced which will certainly need more testing. We count on your help! * A lot has been done "below the surface", as with previous releases. If you watch carefully, however, you will notice considerable changes with your favourite rte_solver disort: The disort solver has been converted from Fortran to C, manually and very carefully by Tim Dowling. C is on position 2 of the TIOBE Programming Community index http://www.tiobe.com/index.php/content/paperinfo/tpci/index.html which is an indicator of the popularity of programming languages. Fortran ranks at position 23, closely followed by "Scratch", "Alice", and "Groovy". In addition to popularity this implies the following important points: - memory allocation is reduced considerably using dynamic memory allocation; - for the same reason, computational speed is increased because considerable time was spent in the model to set variables to zero which aren't even required; in former times this could be prevented by manually adjusting libsrc_f/DISORT.MXD; with the new version this is achieved automatically; - finally, for reasons not yet fully understood, the number of numerical instabilities seems to be reduced considerably; actually, we haven't seen any numerical instabilities since the C version was introduced about three months ago The new C version is automatically invoked with "rte_solver disort" or "rte_solver disort2"; if you want the old Fortran version for compatibility reasons, please use "rte_solver fdisort1" or "rte_solver fdisort2". For details see the manual. * Similarely, a C version of the twostr solver has been added, "rte_solver ctwostr". * Also "below the surface" is a further improvement of the treatment of phase functions. The Legendre moments have been widely replaced by a direct specification of the scattering phase function where considerably less grid points are needed than in the Legendre representation. This process hasn't been completed yet. As a consequence, some combinations of input options (e.g. "wc_properties mie" together with "aerosol_files") might not work. Please refer to the manual or contact the developers if you encounter problems. * On popular demand we implemented a lidar solver, sslidar. As the name suggests, it simulates a lidar observation taking into account single scattering only. This is what essentially what all lidar simulators do and is is sufficiently precise to simulate both air-borne and ground-based lidars, as well as space-bourne lidars in the absence of clouds. For space lidars like CALIPSO and the coming ATLID on EarthCARE multiple scattering is certainly an issue. * New cloud optical properties are available for download in separate packages. The sampling of the droplet size distribution and the angular grid were optimized in order to avoid artifacts. These data are sufficiently accurate for almost any application. The resolution is fine enough to e.g. accurately simulate a rainbow which requires an accurate representation of the phase function at high wavelength resolution. * The mie tool has been restructured completely and memory requirements have been optimized. * A bug has been fixed in the HEY (Hong, Emde, Yang) parameterization of ice particles, in particular of aggregates and droxtals. While the scattering phase function was correct, the sign of the off-diagonal elements of the Mueller matrix was wrong, causing erroneous polarization signatures. * Clearly "above the surface" is the completely redesigned user manual. It should be more easy to navigate and provides extra sections describing details about the radiative transfer solvers etc. Please read it! * A new two-stream solver "rodents" has been added. It was developed as part of a simple climate model by the lectures (Bernhard Mayer and Robert Buras) in the first master course "meteorology" at LMU Munich. rodents is a delta-Eddington twostream code optimized to determine accurate upward and downward irradiances at top-of-atmosphere and surface; this is in contrast to the already available "twostr" which has been optimized for actinic fluxes and heating rates; please check which one is more appropriate and accurate for your application. * The graphical user interface has been improved considerably. It is still not complete, but a number of gimmicks has been added along with a set of examples to play with. Smaller things: * "output rgb" now also works in the thermal spectral range which allows you to simulate e.g. to visualize a glowing stovetop. You probably didn't even know about the exisitence of "output rgb". For an example, have a look at the rainbow application in the user area of http://www.libradtran.org * Added the extraterrestrial spectrum by Thuillier (2003) * Various minor bug fixes; probably removed segmentation faults at 1000 places, and introduced new ones at 3000 different places ...