Introduction
treamer is a radiative transfer model that can be used for computing either
radiances (intensities) or irradiances (fluxes) for a wide variety of
atmospheric and surface conditions. Its interface is extremely flexible and
easy to use. Streamer's major features are:
- Fluxes (irradiances) may be computed using two or more streams, either broadband or narrow band. For more than two streams, a discrete ordinate solver (DISORT) is used.
- Radiances (intensities) may be computed for any polar and azimuthal angles, using 4 or more streams. TOA (top-of-atmosphere) albedos or brightness temperatures are output along with the radiances.
- Upwelling and downwelling, shortwave, longwave, and net fluxes, cloud radiative effect ("cloud forcing"), and heating rates can be computed.
- There are 24 shortwave and 105 longwave bands.
- Gas absorption is parameterized with overlapping gases. Gaseous absorption may be turned on or off.
- Liquid and ice cloud optical properties, five aerosol optical models, and four aerosol vertical profiles are part of the model database. Clouds may be specified as some combination of particle size, water content, optical and geometrical thicknesses, or the optical properties and phase function can be input. A variety of different ice particle shapes (column, aggregate, etc.) are available. Mixed-phase clouds (liquid and ice or ice and ice) can also be specified.
- There are seven standard atmospheric profiles. Either standard or user-defined profiles can be used, or total column amounts of water vapor, ozone, and/or aerosols can be specified. Standard profiles include tropical, mid-latitude, subarctic, and arctic.
- Each computation is done for a "scene", where the scene can be a mixture of up to 10 individual cloud types occurring individually, up to 10 overlapping cloud sets of up to 10 clouds each, and clear sky.
- Various built-in surface types may occur within the scene: open ocean (sea water), meltponds, bare ice, snow, vegetation (four types), and dry sand. Spectral albedo and bidirectional reflectance models (BRDF) are included, or BRDF data can be input.
- A simple user command language provides looping structures for up to ten variables at a time, and output can be customized. There is an interactive (interpreted) mode and a Web interface.
Input and Output Overview
Processing is controlled by a set of options and by input data. The options define the characteristics common to all the data, while the data provides information for each scene or case (e.g., a pixel or observation in time). Output includes surface and top of the atmosphere radiative fluxes, surface albedo, and cloud radiative effect for flux calculations, or radiances and TOA albedo or brightness temperature. Either of two files may be written: one with labeled results, and/or one that is user-customizable.
Limitations, Problems, Peculiarities
- Unless the cloud phase function is input by the user, it is approximated using the asymmetry parameter and the Henyey-Greenstein function. This is fine for fluxes but can result in significant errors when radiances are being computed.
- No atmospheric refraction or spherical geometry is included so that results for solar zenith angles greater than about 70 degrees are subject to error.
- The only gases considered are water vapor, oxygen, carbon dioxide, and ozone.
- The flux/radiance that Streamer computes for the scene is the sum of the fluxes/radiances computed for each cloud part (individual clouds and overlap) weighted by their area fraction.
- Shortwave calculations are not done when the solar zenith angle is 90 degrees or more. No shortwave calculations are done beyond a wavelength of 4 um; no longwave calculations are done at wavelengths less than about 3.4 um.
- While different surface types can occur within a scene, they do not occupy different portions of the scene; i.e., their reflectance signatures are used to create a weighted average albedo for the entire scene.
Questions?
Streamer may be obtained via anonymous ftp as described in the next section. If you have questions or bug reports, contact:
Jeff Key
NOAA/NESDIS/ASPT
1225 W. Dayton St.
Madison, WI 53706
Phone: (608) 263-2605
Fax: (608) 262-5974
email: jkey@ssec.wisc.edu
Are you on the mailing list? If you have requested information via e-mail then you are. If not, and you plan to use Streamer, please register at http://stratus.bu.edu.
Continued work on this program is largely unfunded. I'll be happy to answer questions about things that are not in the User's Guide, and will fix bugs in a reasonably short period of time. Keep in mind, however, that I may not be able to provide an immediate response.
What's New
Version 3 includes DISORT version 2, the capability of specifying a surface bidirectional reflectance function or a built-in BRDF model, a facility to input cloud optical properties including a phase function, new cloud optical property models including eight ice particle shapes, the option to compute cloud water content from optical and geometrical thicknesses, an interactive mode of execution with new commands, a smoke aerosol model, and much more! See the revision history in the Reference Guide for detailed information.
Disclaimer
This program is distributed as "freeware". Except when otherwise stated in writing the program is provided "as is" without warranty of any kind, either expressed or implied, including, but not limited to, the implied warranties of merchantability and fitness for a particular purpose. The entire risk as to the quality and performance of the program is with you. Should the program prove defective, you assume the cost of all necessary servicing, repair or correction. In no event unless agreed to in writing will the author or any other party who may modify and/or redistribute the program be liable to you for damages, including any general, special, incidental or consequential damages arising out of the use or inability to use the program. This includes, but is not limited to, the loss of data or data being rendered inaccurate or losses sustained by you or third parties or a failure of the program to operate with any other programs.
A Note on "Cautions"
There are a number of cautions given throughout this manual. These are not meant to discourage you from using Streamer, but rather to inform those users with little experience in radiative transfer of things that this model does not do well. Ultimately you have to decide if what you are trying to do is reasonable, but these cautions should help steer you away from those aspects of the model that have the greatest uncertainty.
Referencing Streamer in Publications
Streamer is comprised of code and data from a variety of sources. Some of it is original, some of it is not. While Streamer is a unique compilation, it does not constitute a new approach to radiative transfer and has therefore not been published in any atmospheric science journal. So, if you are presenting results obtained using Streamer, how do you reference it? The most complete way would be to reference the individual pieces: the discrete ordinate solver described in Stamnes et al. (1988), the two-stream method following Toon et al. (1989), the water cloud optical properties from Hu and Stamnes (1993), the ice cloud optical properties from Ebert and Curry (1992), the gas absorption data described in Tsay et al. (1989), etc. See the Component Summary section of the User's Guide: Reference for complete details. If you want a single reference, use
Key, J. and A.J. Schweiger, 1998, Tools for atmospheric radiative transfer: Streamer and FluxNet, Computers & Geosciences, 24(5), 443-451.
or simply reference this User's Guide:
Key, J., 2001, Streamer User's Guide, Cooperative Institute for Meterological Satellite Studies, University of Wisconsin, 96 pp.
Comparison with Other Models
A variety of radiative transfer models exist, both for the calculation of radiative fluxes and the simulation of radiances measured by satellite sensors. However, those used for the calculation of radiative fluxes are generally components of climate models (cf., Ellingson et al., 1991) and are not well documented or easy to use. For this reason they will not be discussed further in this paper. Radiative transfer models that are well documented, reliable, and available to the scientific community include LOWTRAN (Kneizys et al., 1988), MODTRAN (Snell et al., 1995), and 6S (Vermote et al., 1994). While other general-purpose models exist, these three have been widely used in remote sensing applications. They are all medium or high spectral resolution band models and incorporate thorough treatments of gas absorption. However, the cloud models in LOWTRAN/MODTRAN are not very easy to modify, and the two-stream approximation for multiple scattering can result in significant errors under certain conditions. The 6S model does not include clouds but is very flexible for clear sky satellite simulations. None of these models computes fluxes directly, and the user interfaces are somewhat crude. Some of the major similarities and differences between these models are listed in Table 1.