Lab: Interpreting SAR radar Imagery
"Assignment and Information"
1. Objectives:
2. Historical Background.
Prior to Magellan, the highest resolution radar images of Venus came from Goldstone and Arecibo ground-based radar installations. Interpretation of these images led various investigators to identify circular radar dark features as giant impact craters or else circular geological structures that might be either large impact craters or volcanoes. Some planetologists concluded on this basis that Venus was as heavily cratered as the Moon or Mars and thus had a very old surface.
The higher-resolution Magellan data showed there to be only about 900 impact craters on Venus, and none of them of the postulated size. Even the first definite impact crater to be identified proved to be nonexistent.
The question arises what they were seeing in these images? In one case (e.g.,Lakshmi Planitae) they were misidentifying a 2 km high plateau as a giant sunken impact crater. But in other cases they do not seem to be misinterpreting extant geological structures or features. No satisfactory explanation for most of these erroneous interpretations has been given in published reports.
In this lab you will use Magellan radar images to figure out what, if anything, they were seeing in the earlier radar images which were misinterpreted as likely giant impact craters or volcanoes.
Ishtar Terra with a portion of the 2 km. High Lakshmi Planitae plateau in the foreground that initially was mistaken for a giant impact crater.
3. Resources.
The following primary resources are available to you to aid you in your investigation:
4. Accessing Resources.
Most of the resources just listed can be accessed by links in thisdocument, through Netscape, are assigned course materials, or will be supplied in hard copy. Your instructor will inform you how to access various Hypercard stacks, the special NASA version of NIH Image, and the Magellan radar image files.
5. Assigned Task:
6. Practical Problems to Surmount.
The Magellan radar images are archived on 126 CD-ROMS in an order that reflects the mapping strategy and history of the Magellan mission. Thus the images are not archived in a manner that groups images of the same region together. The only practical way to find the relevant images is to use the Magellan HyperMap which catalogs known features and the radar images using an efficient search image. Basically it allows you to:
To find a specific latitude and longitude location in the images, you will haveto identify catalogued features and use their locations (given by Magellan HyperMap) as latitude and longitude markers in the images, then extrapolate, interpolate, or otherwise estimate the relative location of the latitude and longitude corresponding to the region you are interested in examining. A gridded map showing main features on Venus is available which will help in estimating locations on MIDRs.
By the time you have figured out how to find specified latitudes and longitudes in the Magellan images, you should be pretty good at interpreting them.
7. Strategies for Proceeding.
Part of doing real science is figuring out how to work with the data bases and the software resources. Doing so is a central part of this lab assignment. Some suggestions for proceeding will be given, but you will have to invent your own procedures. If you get really stuck, there is a sample tutorial you can access, but it should be used as a last resort and accessing it does carry a grade penalty. (However, that grade penalty can be made up by doing the extra-credit task.)
Before you begin serious work on the assignment, you might familiarize yourself with radar images and their interpretation. The Magellan at Venus Hypercard Tutorial can help you do that. It shows what the main geological structures/features on Venus look like when radar imaged and gives you some background on the features. To open it just click on the icon for its Hypercard stack. If you want more practice recognizing geological features in radar images, you can look at the following images from Welcome to the Planets archives.
Look at maps of Venus identifying main features. Two are given: A gridded pre-Magellan map and an ungridded labeled composite Magellan radar image.
You should learn how to use NIH Image. Once you have opened the software, try bringing up a Magellan image (your instructor will give you information about their file locations and how to access those files over the network) by using the Import and by using the Open commands in the File Pull-Down menu. Use it on files marked "BROWSE.LBL" Find the "Tools" palette and experiment with the magnifying glass tool by clicking on the tool and using it as a modified mouse cursor. If you place it on a specific part of a displayed image, it will give you a close-up view centered on that part by clicking on the mouse. If you hold down on the "option" key as you click the image will be a far-away wide-angle view. Experiment using the "select" tool (the dotted square box) which picks out a rectangular portion of the image if you drag it with the mouse down. Figure out how to remove an unwanted selection vs. deleting or copying a selected portion. Learn how to change the brightness and contrast of the displayed image by using the "B" and "C" sliderbars in the "Map" box beneath the tools palette. Play with "Scale to fit" under the options menu which displays the entire image on your screen but distorts it differently in vertical and horizontal directions. Experiment with some of its other features if you want.
You also should open the Magellan HyperMap and become familiar with it. Next to the Magellan images, this is your most important resource.
There are other resources listed above that might be helpful in your investigation. However, I would suggest you do as much as you can with the Magellan images and Magellan HyperMap before you turn to these resources. They include www access to the full Magellan radar data set (as opposed to the selected files your instructor has made available locally on carnap.umd.edu), several archived www data bases of catalogued features, and the Brown University Data Base which is the most comprehensive of these resources.
8. Magellan MIDR Images.
The Magellan images you will be using are known as MIDRs (Mosaiked Radar Images). They exist in different resolutions which are a function of extent of compression (by averaging adjacent pixels). F-MIDRS are full resolution, and C-1, C-2, and C-3 MIDRS respectively are once, twice, and thrice compressed. The greater the compression, the larger extent of surface used. For the purposes of this lab you will find the F and C-1 MIDRS most useful.
Each Magellan MIDR CD-ROM contains multiple image files with labels like C1-MIDR.45N329;2 or F-MIDR.30S332;301.
The labels respectively tell us that the file is a MIDR image at specified resolution, that the image centers on a location whose longitude is so many degrees N or S of the equator, and at a specified longitude (e.g., `45N329' centers on latitude 45 degrees N, longitude 329 degrees). Thus the label tells you coordinates for the centerpoint of the image. The number after the semicolon tells you which mapping cycle (1, 2, 3) the image is from, and when more than one image exists for a cycle additional digits differentiate them. Actually the label is for a file containing a mosaiked browse image together with a number of constituent detailed images. For the purpose of this lab, we will use the browse images.
The combination of the label indicating centerpoint and the Magellan HyperMap indication of the coordinates of the four corners of the browse image provide you with the basis for working out a coordinate system for the image and estimating the locations of specific features. From knowledge of the planetary radius (6052 km) you can figure out how many degrees correspond to how many km at F or C-1 resolution, and thus approximately determine the size of features. (If you plan also to use the "Scale to Fit" command in Image you will need to do these calculations for both the normal and the "Scale to Fit" displays since the latter has different vertical and horizontal distance scales from normal images. Also, the Magellan HyperMap indicates the diameters of cataloged impact craters and the like, thus giving you other bases for estimating diameters.
After you have familiarized yourself with the available resources-especially Image and Magellan HyperMap you should try to develop a scheme for determining locations and distances on F and C-1 images. (You may need to adapt the scheme for different images depending on what sorts of features they include, the nature of gaps in the image coverage, etc.)
Many of the images have blank spots due to some failure of the imaging on that cycle. In some case you may have to use multiple images from different cycles in order to see enough of the region you are interested in or to identify features. Dealing with data gaps and incompleteness is a normal part of doing real science using real data.
8. Solved Example is Available (for a price).
As was mentioned, figuring out how to work with the data given the available scientific instruments is a big part of doing real science. However, even real scientists sometimes need assistance in figuring out how to proceed. Such assistance usually carries a price: Sharing authorship of a paper with those who assist you, or perhaps even doing less significant work.
If you get really stuck and can't figure out how to proceed, you can get help by going through a sample solution for one of the other potential craters in the table. However, if you do, you will have to pay a full grade penalty. You will have to balance the relative merits/rewards of doing it on your own vs. getting expert guidance. If you do access the sample, that fact automatically will be registered in the course grade records. The grade penalty can be erased by doing the Extra Credit third feature.
When you are ready to start working on the lab, click here.
