Magnetometer Investigation: The Lunar Wake

ISTP NEWSLETTER Vol 5, No. 1. Aug, 1995


Title Author

Welcome Message - Susan Sekira

The Challenge Ahead - Jim Willett

A Banner Year for ISTP - George Withbroe

ISTP Catalog of Preliminary Solar Wind Events - M.Peredo

ISTP Spacecraft Trajectory Plots Available via World Wide Web - M.Peredo

WIND/GEOTAIL Workshop - K W Ogilvie

The Goddard Modeling Group - David P. Stern

WIND/SWE Observations of The Lunar Wake - R. J. Fitzenreiter

An Early Finding from the WIND Magnetometer Investigation: The Lunar Wake - C. J. Owen

Upstream ULF waves associated with the lunar wake: Identification of a forewake region - W. M. Farrell

Solar Wind Protons and Alphas Measured by the SWE Experiment on Wind - Alan J. Lazarus, John T. Steinberg

Detection and Identification of a Single Commercial Earth-based Transmitter from Space - M. D. Desch, M. L. Kaiser

High School Students Use WIND SWE Data to Investigate Space Weather - T. G. Onsager

Automatic CDF Validation Software (ACVS) - G. Blackwell

Polar Thermal Vacuum Availability - Ed Nace

The Program Assistance Center (PAC) - Ann Strickling

Data Distribution Facility (DDF) Support for the ISTP Program - Ken Lehtonen

The DDF Architecture - Ken Lehtonen

IMP-8 Electronic Distribution Progress Report - Michael Anderson


Michael Cassidy

Contributing Editors:

Steven Curtis - Science Editor

Ken Lehtonen - Data Distribution Facility

Kevin Mangum - Central Data Handling Facility

Dr. Mauricio Peredo - Science Planning and Operations Facility

Dick Schneider - ISTP Project Office

Jim Willett - NASA Headquarters


Welcome to the first online edition of the ISTP Newsletter! This newsletter was established to highlight ISTP science and to provide key information to the user community. After its debut on the World Wide Web, the staff of the ISTP Newsletter will target subsequent issues on a bi-monthly basis and strive to improve upon each edition with suggestions from our readers.

The ISTP Newsletters are in Adobe Acrobat Portable Document Format (PDF). PDF readers for DOS, Windows, Mac, and UNIX are available at no charge from Adobe's WWW page or via anonymous ftp Each on-line edition of the ISTP Newsletter will be maintained on the ISTP Home Page.

If you have any comments or suggestions for the next newsletter, please send them to the Editor, Michael Cassidy. If you have comments or questions regarding the contents of any article, please address them to the individual authors.

Susan Sekira
ISTP CDHF Project Manager
Goddard Space Flight Center
Greenbelt, Md. 20771

The ISTP Newsletter is published bi-monthly by the staff of the ISTP CDHF (Central Data Handling Facility).

Editor Michael Cassidy


ISTP is the flagship mission for the space physics scientific community, our "great observatory" of the 1990s. In less than a year the culmination of more than a decade of work will be realized as all the elements of ISTP come together to pursue the prime objectives of this powerful set of missions. The prime phase for these missions ends in early FY98. What about extended missions?

Funding for extending ISTP mission operations and data analysis (MO&DA) beyond the prime phase requires new funds in the same way that development of new flight missions requires new funds. In earlier days mission extensions were virtually automatic. This is no longer so. Obtaining the necessary funds for an ISTP extension will be challenging in a highly constrained budgetary climate where federal deficit reduction is a major goal, and where both Congress and the White House have projected decreases in the funding levels for NASA over the next five years.

To meet this challenge we need to demonstrate the high value of ISTP with an outpouring of excellent results in the coming year. The Office of Space Science (OSS) budget for FY98 will be assembled in late spring/early summer of 1996 less than one year from now. A steady stream of high quality results will provide a strong justification of an extended mission in the FY98 budget. A second element in the development of a compelling case for the extension of ISTP is the approach of the next solar maximum when we can study the Sun-Earth connection under different conditions than at solar minimum when ISTP first comes into full operation. What better time to study the Sun-Earth connection than when the Sun moves from minimum to maximum?

Significant and important science results are fundamental in making the case for an extended mission. The FY98 NASA budget is assembled by early summer in 1996. We have less than one year to demonstrate the value of ISTP to the scientific community and public. Your help is critically needed.

Jim Willett
Operations & Data Analysis,
Code SS, Space Physics Division
NASA Headquarters,300 E Street SW
Washington DC 20456

phone: (202) 358-0888; fax: (202) 358-3987


In less than a year we plan to have all of the major elements of ISTP in place when Geotail, Wind, Polar, SOHO, and Cluster are operational, Geotail and Wind are already in orbit, with SOHO and Polar planned to be launched, respectively, in November and December,1995 and Cluster in January 1996. We also hope to have the Russian Interball spacecraft in orbit this fall and the Equator-S spacecraft in orbit by early 1997. This fleet of spacecraft will provide the scientific community with the most powerful tool yet assembled to study the physics of the Sun-Earth connection.

Getting to this point has required the dedicated efforts of a international partnership of scientific investigators, industrial teams, and governmental agencies. Many challenges, some of them severe, have been successfully overcome thanks to the hard work and teamwork by, and between, the various partners. Although we still face the final hurdle of launching many of the ISTP elements into orbit and bringing the spacecraft and instruments on line, we can give ourselves a pat on the back for a job well done thus far.

George Withbroe
Director, Space Physics Division
NASA Office of Space Science

ISTP Catalog of Preliminary Solar Wind Events

M. Peredo, D. Berdichevsky and S. Boardsen (Raytheon STX Corporation, ISTP/SPOF) R. Lepping, K. Ogilvie, J. Byrnes, and L. Burlaga (NASA GSFC/LEP) A. Lazarus, K. Paularena, and J. Steinberg (MIT, Center for Space Research)

The ISTP Science Planning and Operations Facility, in collaboration with ISTP investigators, is developing a catalog of preliminary solar wind events and features. The catalog also contains listings of times when WIND and IMP-8 were in the solar wind. Coverage begins on September 8, 1992, the start of ISTP science data collection.

Information is included on selected features of the solar wind at 1 AU identified from WIND and IMP-8 plasma and magnetic field measurements in the form of, Key Parameters (preliminary summary data at ~1 min time resolution produced quickly for survey purposes); as such, the catalog should not be used as a definitive source in formal scientific work. Researchers using the catalog should reference its contents with statements such as: The event from "Feb 26, 1995" has been identified as a candidate "Sector Boundary Crossing" worthy of further study . The primary intent for the catalog is to serve as an aid in identifying candidate periods for further study, such as may be the focus of coordinated data analysis efforts during ISTP or IACG Science Campaigns. A major goal is to help facilitate research that uses data sets from ISTP and related missions in three respects: 1) the study of the solar wind as a plasma laboratory of interest in itself, 2) studies of solar wind-magnetosphere coupling, and 3) studies of solar activity-solar wind linkage.

Clearly, the listing is not comprehensive in any sense. For example, coronal mass ejections (CME's) are known to affect magnetospheric activity. The suite of measurements available to us does not, however, include those usually required to identify CME's (e.g., bi-directional streaming of energetic electrons); thus, such events are not listed by that terminology, although some of the selected events may be associated with CME's.

The solar wind features are classified into several categories (e.g., interplanetary shock wave, extended period of strong negative Bz, etc.), which are believed to support the three types of studies listed above, without indicating which one is relevant. An extended event has separate start and end times listed for it, whereas a sharp discontinuous event (such as a shock ramp) has the same time listed for both start and end times (clearly, for follow-up studies researchers will need to examine data before and after the listed time). In any case, all times chosen are preliminary, serving as a guide for further examination. It must be emphasized that the employed terminology means only that the signature appears to be like that certain kind of event from a cursory examination and may be a candidate for further study to affirm or deny the preliminary estimation. For example, a shock-like profile will simply be put into the "shock" category, with no imprimatur implied. The categories used in the catalog are [editor's note: where x, y, and z are subscripts]:

· BzN = Northward Bz for extended period

· BzS = Southward Bz for extended period

· EyC = Change in Ey=VxBz

· HSS = High speed stream

· IMC = Interplanetary magnetic cloud

· IS = Interplanetary shock

· LSS = Low speed stream

· MISC = Miscellaneous

· PC = Pressure change

· SBC = Sector boundary crossing

In summary, the catalog is meant to be an aid in highlighting some aspects of the WIND and IMP8 solar wind data hopefully leading to new scientific endeavors, and we hope it inspires many.

The catalog is accessible via the World Wide Web at the URL listed at the beginning of the article. The tables below illustrate catalog entries for candidate events, and solar wind occupancy times.

             WIND                               IMP8
       Solar wind times                   Solar wind times
  Start time       End time          Start time       End time

                 1994  319 1900    1993  030 2200   1993  030 2400
1994  320 2200   1994  334 1800    1993  034 1400   1993  043 1300
1994  335 1300   1994  346 1000    1993  047 0800   1993  047 0900
1994  347 0800   1994  357 2200    1993  048 1400   1993  056 0300
1994  359 0700                     1993  059 0600   1993  068 0400
                                   1993  068 0500   1993  068 2000

Sample of Candidate Event Periods

Start Time     End Time  Catagory  S/C  Comments


93067 2130   93068 0600  IS/BzS     8   Magnetopause seen at geosync, ~5 hrs
                                        of Bz ~-15nT on Day 68 
94148 1400   94148 1400  IS         8   Np up to ~100/cc
94339 2100   94339 2100  IS         W   Followed by high speed stream

95085 0500   95085 1200  SBC        W   |B| ~ 12-18nT, N> 40/cc, strong 
                                        magnetospheric activity extends
                                        until day 88

Questions and comments on the catalog are welcome and may be sent via E-mail to M.Peredo (, or D.Berdichevsky (

ISTP Spacecraft Trajectory Plots Available via World Wide Web

M. Peredo and S. Boardsen (Raytheon STX Corporation, ISTP/SPOF)

The ISTP Science Planning and Operations Facility (SPOF) has added a Spacecraft Trajectory Plots option to its world wide web page (the SPOF home page is accessible at URL Presently, plots containing trajectories for WIND, GEOTAIL and IMP-8 are available; the effort will be extended to include plots for additional ISTP spacecraft.

The existing plots span 10-days each, and are organized as follows: all file names are of the form "orbit_yyddd00-yyddd00_??_t.gif, where the yyddd designations identify the starting and ending times for the plot, ?? is either "xy" or "xz" depending on which projection (X-Y)GSE or (X-Z)GSE [Editor's note: where GSE is a subscript] is displayed, and t indicates the type of the plot; t=0 indicates a close up view while t=1 is a distant view that allows display of the WIND orbit when it is near the L1 point. Average model bow shock and magnetopause boundaries are included for reference. Tick marks appear every day and tick labels every two days. The linestyle changes depending on whether the spacecraft are predicted to be in the solar wind (solid lines), the magnetosheath (dashed lines), or the magnetosphere (dotted lines). A sample plot, covering days 210 (July 29) through 220 (Aug 8) of 1995 is shown below.

Users without web access can obtain copies of the files via anonymous ftp from the CDHF. The files reside in the directory: SYS$PUBLIC:[SPOF.ORBIT_PLOTS].


Questions and comments on the orbit plots are welcome and may be sent via E-mail to M.Peredo (, or S.Boardsen (


K. W. Ogilvie

An informal WIND/GEOTAIL Workshop was held in Honolulu from the 16th to the 18th of May, 1995. Its purpose was to acquaint workers on each spacecraft data sets with those on the other, to select promising periods for joint data analysis and interperation, and to pick experimenter teams for this process.

After a day of presentations on recent unpublished work, the meeting divided into three subgroups, each with a Japanese and an American leader; at the end of each period, these leaders reported back to the entire meeting.

Although the meeting relied upon data plots, because of the location (chosen , incidentally, to even out the travel burden), progress was made on all subjects and sixteen topics in all were selected to be studied in the near future.

Informal minutes have been distributed to the attendees. Interested people can obtain a copy of this material by requesting it from K. W. Ogilvie, using email:

FAX: 301 - 286 - 1683
VOICE: 301 - 286 - 5904

WIND key parameters can also be requested at the same time. A second workshop is under consideration for December. This time both IMP-8 and theoretical simulations will be included, but attendance may have to be restricted by reason of facilities.

Announcements will be made in the next month.

K W Ogilvie
Goddard Space Flight Center
Code 692
Greenbelt, Md. 20771


The Goddard Modeling Group

by David P. Stern

Almost any interpretation of ISTP data needs information about the magnetic f ield. Where does the field line from an observation point reach the ionosphere, or the plasma sheet? Is the observed cusp in an expected position, or an unusual one? What pressure rise in the solar wind is needed to produce an observed increase in the dayside field?

Such questions are the concern of the Goddard modeling group, whose task is devising field models, improving them and advising their users. The group currently has four members. Nikolai ("Kolya") Tsyganenko handles data, fits models to them and devises appropriate representations for the various fields involved. Mauricio Peredo, in addition to guiding SPOF and duties at the National Space Science Data Center (NSSDC), also fits and disseminates models and adds new sets to the data base. Tom Sotirelis, finishing his post-doctoral stint, has been deriving models which satisfy pressure balance with the solar wind, including the tail and other current systems. And I have just developed a fairly realistic analytical model of the global field of Region 1 Birkeland currents. All four of us have described our recent work at the IAGA assembly in Boulder.

There exists a widespread view that the main purpose of models is to provide experimenters with neat packages to derive the magnetic field and answer with reasonable accuracy questions like the ones listed above. Actually, the primary purpose is quite differentto extract from the data the greatest amount of magnetic field information, and to learn about the current systems which produce it and their variation with tilt, solar wind pressure, etc. Only when the modeling package fits existing data can it be used operationally with any degree of confidence.

Modeling is not a simple task, and each of the 5 main field sourcesmain field, magnetopause, ring current, tail and Birkelandhas its own custom-fitted methods. To go beyond the simple use of computerized "black boxes" and find out what actually goes into them, you might consult the review "The Art of Mapping the Magnetosphere", JGR 99, 17,169, '94, which will further direct you to original papers. The field however is still developing and improved methods are constantly added, e.g. a boundary combining a half-ellipsoid and a cylinder (pictured below), with the magnetopause field described by a single scalar potential, fitted by least squares (JGR 100, 5599, '95). This is to be one component of a planned "Tsyganenko 1995" model replacing "all-purpose" expansions of older models.

If you use our models, please keep in mind that at all times we stand ready to advise and help you. You will learn more about our group and its work by logging onto it's web page:

After that, if any questions or requests remain, send them by E-mail, preferably to


David P. Stern
Goddard Space Flight Center
Code 695.0
Greenbelt, Md. 20771


R. J. Fitzenreiter and K. W. Ogilvie
NASA/Goddard Space Flight Center


Figure 1 (not to scale)

The positions of WIND and the moon through the lunar wake during a lunar swingby on 01 December, 1994, and during a more distant encounter on 27 December, 1994. The moon and lunar shadow are sketched schematically.

Two lunar eclipse events observed when WIND passed through the lunar wake are currently being studied. The WIND spacecraft passed through the wake at very different distances from the moon on 01 December, 1994 and 27 December, 1994, and make possible a study of the development of the lunar wake. These two events are being analyzed using plasma electron data from the Solar Wind Experiment (SWE). Plasma electrons have previously not been studied in the lunar wake. There will be several additional lunar wake encounters by WIND which should provide a good picture of the wake development.

Figure 1 is a sketch showing the orbits of WIND and the moon during periods including the two eclipses. On 01 December, WIND was far downstream from the moon, 77 Rm, whereas for the 27 December event, WIND was close to the moon, 4.5 Rm, the eclipse occuring during a lunar swingby [Editor's note: m is a subscript]. The lunar shadow formed by absorption of the solar wind plasma by the moon is shown schematically (not to scale).

The SWE electron data are shown for the 27 December eclipse in Figure 2. The energy spectrum in the top panel shows that the low energy electron flux (energy < 45 eV) drops off by a factor ~100 as the spacecraft moves to the center of the umbral region at 15:06 UT.

The drop off in higher energy flux is much smaller, less than a factor 10 at energies >45 eV. This change in plasma electron flux corresponds to the decrease in electron density by a factor 20 (second panel) and an increase in electron plasma temperature by a factor 4 (second panel). The temperature rise is due to the depletion of low energy electrons and the preferential penetration of higher energy electrons toward the center of the wake region. The bottom panel shows that the the magnitude of the electron plasma flow velocity increases slightly upon entering and leaving the wake region, and decreases to a minimum in the umbral region. The flow direction is within 10° of the anti-solar direction at the edge of the wake.

The electron plasma density during the 01 December lunar eclipse passage, 77 Rm behind the moon, is shown in Figure 3. Although density fluctuations are present, there is an overall gradual decrease and recovery in density which coincides with the time of the decrease and recovery of the WIND spacecraft solar array current through the eclipse. In addition, the sharp drop to the minimum electron density of 2 el/cm-³ at 15:19 UT occurs at the same time as the minimum in the solar array current, indicating the center of the wake. The nature of the electron distribution functions at the lowest energies suggests that the sharp decrease is not due to a change in the spacecraft electric potential. Comparing the minimum electron density for the two events shows that at this much greater distance, the wake is almost filled with electron plasma.


Figure 2

Electron plasma measurements through the lunar wake region during a close WIND encounter (4 Rm) with the moon. The energy spectrum of electron flux, and electron plasma density, temperature, and flow velocity are shown (top to bottom).

Figure 3

Electron density in the lunar wake during a distant WIND encounter (77 Rm) with the moon. The time of the minimum spacecraft solar array current is indicated by the dashed line.

Members of the SWE research team are K. W. Ogilvie, D. J. Chornay, R. J.Fitzenreiter, F. Hunsaker, J. Keller, J. Lobell, E. C. Sitter and Adolfo Vinas (Goddard Space Flight Center); R. B. Torbert, G. Needell, and T. G. Onsager (University of New Hampshire); A. J. Lazarus and J. T. Steinberg (Massachusetts Institute of Technology); J. D. Scudder (University of Iowa).

R. J. Fitzenreiter
Goddard Space Flight Center
Code 692
Greenbelt, Md. 20771

An Early Finding from the WIND Magnetometer Investigation: The Lunar Wake

C. J. Owen, R. P. Lepping, J. A. Slavin, W. M. Farrell, and J. B. Byrnes
NASA/Goddard Space Flight Center

The WIND spacecraft was launched on 1 November 1994 with a full compliment of advanced instrumentation to analyze the particles and fields of the solar wind magnetoplasma. Included was a magnetometer (MFI) to investigate the orientation and magnitude of the interplanetary magnetic field at a high sample rate. Following a series of four and 1/2 phasing orbits each lasting 12 days, the spacecraft made a close encounter past the moon on 27 December. This fortuitous event gave experimenters an opportunity to study the lunar wake.

Figure 1 shows the WIND spacecraft trajectory past the moon. Note that the closest approach occurred within 5 Rl. Also, the closest approach point occurred when the spacecraft was on the antisunward side of the body, which is the favorable geometry for the study of the wake.

Figure 2 shows the 3-second average values of the magnitude of the magnetic field from the WIND magnetometer. The transit through the umbra region is indicated by the vertical dashed lines. For comparison, the magnitude of the interplanetary magnetic field between 1400 UT and 1600 UT at the IMP-8 spacecraft is superimposed. The solar wind delay between the two spacecraft was typically less than 3 minutes, and is relatively small compared to the time scale of the event.


Figure 1 The WIND trajectory during the lunar swing by of 27 December 1994

Note that during the period of interest the magnetic field as measured by IMP-8 remains at a relatively constant value near 7 nT. With no perturbation, the WIND spacecraft should measure a comparable result. However, in the presence of the moon, the magnetic field magnitude shows some distinct alterations compared to the nominal situation. Specifically, between about 1415-1442 UT, the field magnitude was depressed by about 1 nT compared to the IMP-8 values, this in association with the spacecraft passage through the penumbra. Within the umbra (i.e., between 1442-1523 UT), the field magnitude rose above the nominal levels, but was very slightly depressed again as the spacecraft exited the umbra region and entered the penumbra.

The lunar wake was observed previously in the late 1960's by Explorer 35, as it orbited the moon within 3 Rl. Based upon numerous crossings, the typical wake magnetic signature in the penumbra and umbra regions became well defined, and is illustrated in Figure 3 (adapted from Whang and Ness, 1970). Specifically, there will be a significant decrease in the magnetic field in the penumbra regions. This signature is similar to that observed by the WIND spacecraft. The primary difference between WIND and Explorer 35 wake detections is that the features as observed by the latter are sharper, apparently due to the spacecraft's closer approach to the body.

The finding of the lunar wake by WIND is a serendipitous event. To our knowledge no one had predicted that wake effects would be observed during this lunar passage of the moon by WIND, apparently because of its relatively larger distance at closest approach.


Figure 2

The WIND magnetometer (MFI) measurements during the lunar swingby of 27 December 1994. Note that there is a significant perturbation relative to the IMP-8 measurements (superimposed), indicating the presence of the lunar wake.

Figure 3

An illustration of the expected magnetic signature asociated with a transit of the lunar wake. The expected B-field perturbations is based on measurements obtained during the numerous Explorer 35 Orbits. (Whang and Ness, 1970)


Whang Y.-C., and N. F. Ness, Observations and interpretation of the lunar mach cone, J. Geophys. Res., 75, 6002, 1970

C. J. Owen Goddard Space Flight Center
Code 696
Greenbelt, Md. 20771

Upstream ULF waves associated with the lunar wake: Identification of a forewake region

W. M. Farrell, R. P. Lepping, C. J. OWEN, and J. B. Byrnes
NASA/Goddard Space Flight Center

In order to achieve a favorable orbital trajectory, the WIND spacecraft made a close swingby of the moon of 27 December 1994. Although not a primary mission objective, this close lunar encounter allowed an opportunity to study the interaction of the large body with the solar wind, now with unprecedented temporal, spatial, and spectral resolution. Owen et al. [1995] recently obtained a distinct observation of the penumbra and umbra regions of the lunar wake based on the WIND magnetometer (MFI). Fitzenreiter et al. [1985] observed a distinct electron density dropout in the lunar shadow (i.e., umbra region) using the Solar Wind (SWE) experiment electron spectrometer. We report here on unusual ULF wave activity detected by the WIND magnetometer in a region just upstream from the wake. There was only one previous report of upstream ULF wave activity associated with the lunar wake, this made in the late 1960's by Ness and Schatten [1969] using the magnetometer onboard Explorer 35. The ULF activity detected by WIND and Explorer 35 is significant, given that it bares a striking resemblance to activity commonly found in foreshock regions adjacent to planetary and interplanetary shocks.

The WIND spacecraft made its closest approach to the moon on 27 December 1994 near 1500 UT at a distance of less than 5 lunar radii (Rl), when the spacecraft was antisunward of the moon. The entry into the lunar wake region, as defined by the magnetometer measurements, occurred near 1415 UT, and the spacecraft remained in the region for the following 85 minutes.


Figure 1

Figure 1 is a spectrogram displaying the ULF wave activity detected between 1 mHz and 5.43 Hz by the WIND magnetometer z-sensor data, the sensor oriented along the spacecraft spin axis. In the figure, the vertical axis represents increasing frequency, the horizontal axis represents increasing time, and the wave intensity is denoted by the color (blue, least intense; red, most intense). The entry and exit into the lunar wake is indicated in the diagram. The z -sensor is used for "quick-look" spectral analysis since the spin modulation of the magnetic field near 0.3 Hz is negligible.

As indicated in the spectrogram, prior to 1317 UT, much of the wave activity was that more typically observed in the solar wind, this being broadband Alfvenic turbulence that varies in frequency as 1/f 1.5 . However, at 1317 UT, a very unusual narrowband tone was detected between 1-2 Hz that persisted for nearly 60 minutes prior to the wake entry. The tone started initially near 1.2 Hz at 1317 UT, steadily rose in frequency to above 2 Hz at 1330 UT, and then steadily decreased in frequency to near 1.2 Hz for the last 45 minutes. During the period of time of wave observations, WIND was magnetically-connected to the lunar wake. The emissions ceased almost exactly as the spacecraft entered the wake region. The observed emission frequency lies well above the proton cyclotron frequency, but below both the electron cyclotron and electron plasma frequencies and thus is tentatively identified as the whistler mode. Confirmation of this mode has been made by subsequent determination of the emission k vector and theoretical analysis. Emissions at ULF frequencies in the whistler mode possess group velocities that are greater than the bulk flow speed of the solar wind. Thus, an emission generated at or near the wake edge is capable of propagating upstream of the wake.

Compared to the ULF emissions detected by the Explorer 35 magnetometer [Ness and Schatten, 1969], the WIND emissions are significantly weaker by a factor of 100, more narrow -banded, and were detected further from the moon. In fact, the Explorer 35 magnetomer with a detection threshold of +/- 0.1 nT would not be sensitive enough to detect the ULF emissions observed by the WIND magnetometer, these having integrated field strengths typically of 2.5 x 10-³ nT. It is only with the improved sensitivity of the WIND magnetometer with a detection threshold of +/- x 10-³ nT that an observation of these weak but important signals distant from the moon, was possible.

The importance of these upstream ULF waves is that they are conveying information: They are signaling the incoming solar wind about an upcoming magnetic structure, the lunar wake. This precursor phenomena commonly occurs ahead of other magnetic structures, such as terrestrial and interplanetary bow shocks. In these "foreshock" regions, information about the shock is carried upstream into the solar wind via interactions with outward propagating waves and particles. We thus denote this region upstream of the lunar wake as a "forewake" region, implying a philosophical similarity to the active foreshock regions; information about the shock is carried upstream into the solar wind via interactions with outward propagating waves and particles found upstream of collisionless shocks.


Owen et al,, ISTP Newsletter, this issue, 1995

Fitzenreiter and Ogilvie, ISTP Newsletter, this issue, 1995

Ness, N. F., and K. H. Schatten, Detection of interplanetary magnetic filed fluctuations stimulated by the lunar wake, J. Geophys. Res., 74, 6425, 1969.

W. M. Farrell
Goddard Space Flight Center
Code 695
Greenbelt, Md. 20771

Solar Wind Protons and Alphas Measured by the SWE Experiment on Wind

by John T. Steinberg and Alan J. Lazarus, MIT

The Faraday Cup detectors of the Solar Wind Experiment [1] provide measurements of the flow velocities, densities, and thermal speeds of positive ions of the solar wind. Preliminary data illustrate the experiment's capability for determining the vector flow velocity difference between protons and alphas.

Figure 1

Figure 1 shows results obtained on April 11, 1995 during a period of moderately fast solar wind speed. The top panel shows the speeds of the two ion components, note that the speeds of the protons are usually less that those of the alphas.

The next panel shows magnitude of the vectors representing the alpha velocity minus the proton velocity. Here it is clear that even though the two ions may have the same speed (e.g. near 1130) the vectors representing their velocities point in different directions. Also plotted in this panel is the Alven speed determined from the ion number density and the magnetic field magnitude [2]. The Alfven speed is a measure of the maximum value that the magnitude of the alpha -proton velocity differences should have.

The velocity difference vector should point along or counter to the magnetic field in the plasma because bulk motion perpendicular to the local field is not allowed for plasmas of the good conductivity of the typical solar wind.

The third and fourth panels from the top show the longitude and latitude of the flow compared to the angles characterizing the observed magnetic field direction. For this portion of data, the field lines are inward and velocity difference vector points counter to the field direction. (Hence, the field direction is plotted so that the velocity difference and field directions should track each other.)

These observations give us the opportunity of studying several interesting aspects of the velocity difference between alphas and protons. For example, why do the alphas generally have higher velocities? What are the conditions which lead to the less common occurrences of protons having the higher velocities? Is the velocity difference driven by Alfven waves propagating along the field and if so are the velocity differences limited by the Alfven speed as some earlier works have suggested? We look forward to studying these and other issues.

[1] The solar Wind Experiment is a joint effort of the Goddard Space Flight Center (K. W. Ogilvie P.I.), MIT, and University of New Hampshire.

[2] The magnetic field data were provided by the members of the Wind Magnetic Field Investigation Team, R. Lepping, P.I..

Detection and Identification of a Single Commercial Earth-based Transmitter from Space

M. D. Desch and M. L. Kaiser
NASA/Goddard Space Flight Center
Greenbelt, Md. 20771

The WAVES instrument onboard the WIND spacecraft is designed to observe naturally -occurring planetary and solar radio phenomena at frequencies from a fraction of a Hz to 14 MHz. WAVES also detects man-made signals from Earth-based transmitters around the globe. Over the dayside of the Earth, where WIND spends most of its time, only transmissions that are above the ionospheric cutoff frequency, usually about 6 to 7 MHz, are detected. Examples of these transmissions may be seen in the figure, where they appear as narrowband tones, almost exclusively at frequencies corresponding to the allocated shortwave bands (see panel to right of dynamic spectrum).

Usually, what is observed by the WAVES receiver at a given frequency is the combined effect of the transmissions from the commercial shortwave and amateur broadcast stations that are 'visible' to the spacecraft at a given time. Individual commercial broadcasts have a bandwidth of 5 kHz. Since almost the entire hemisphere of the Earth is visible, and because individual receiver passbands are 20 kHz, this can amount to a lot of stations.

However, on November 17, 1994, with WIND more than 200,000 km from earth, WAVES detected a highly unusual radio signal. Around midday, on 6.15 MHz, observing conditions were initially ideal. No solar storms were in progress, the planet Jupiter, normally an active radio source, was quiet, and almost all of the man-made interference was at higher frequencies. The only signal being recorded near 6 MHz was the ever-present galactic background noise. Abruptly at 1300 GMT, however, the situation changed. The receiver signal jumped by 30-40 dB, and remained high for exactly 30 minutes. At 1330, the output fell back to galactic background as sharply as it had risen. No similar signal enhancements were recorded on the receiver's adjacent channels, separated by 50 kHz. The transmission is visible in the figure as a fine horizontal red line centered at 6.15 MHz and 1315 hr and labeled 'BBC'.

There were several things about this signal that suggested it was not of natural origin. Most naturally-occurring radio emissions, whether they are from planetary or solar sources, turn on and off gradually, and when they are 'on' the signal is quite erratic. Our 6.15 MHz signal tended to remain well above the cosmic background for extended intervals. In addition, the signal was extremely narrowband, whereas most naturally-occurring emissions are relatively broadbanded. So we were fairly sure the signal was man-made, but we also thought that perhaps the transmission was from a single broadcast station on Earth but which one? Our instrument on WAVES is capable of performing direction finding to within about +/- 1° however the Earth at this distance subtended an angle of only 3° . So we could not pinpoint the source of the radio signals using direction finding. And further, half the globe was 'visible' to us at the time of the transmission.

Reference to one of the shortwave guides was very helpful at this point. We were able to find a station on 6.15 MHz that broadcast on a regular schedule each day from 13001330 GMT. It matched the time and frequency of our reception exactly. According to the guide, it was a BBC broadcast to Central America from the Voice of America's relay station in Delano, California. But was the transmission in the right direction to radiate the spacecraft?

At the time of the transmission, California was just coming up over the horizon as viewed from the spacecraft. Calculation showed that, as viewed from the transmitter in Delano, WIND was 18° above the horizon and about 64° East of due South. As it turned out, in order to transmit to Central America from Delano, the broadcast engineers had raised the main beam above the horizon by 18° and slewed the beam east of South by about 60°! The geometry was such that WIND was almost exactly in the main beam of the transmitter and that we were looking through the most tenuous portion of the normally shielding blanket of the Earth's ionosphere. These were ideal conditions for detecting such a signal.

Even so, only a percent or so of the total signal from the BBC got through to space: approximately 4000 Watts out of a total transmitter signal strength of about 250 kW. At the spacecraft, over 200,000 km away, the signal picked up by WIND's antennas was less than one-thousandth of a microwatt, a small amount of power but large enough to be easily detected. An amateur with a shortwave receiver on the spacecraft could have easily interpreted the signal as coming from the BBC in Spanish!

If our analysis was correct we should have detected this broadcast again the next day. But we didn't. Examination of the data has shown that we do not detect this signal every day despite the fact that it is a regularly scheduled broadcast. The reason is that subsequent to 17 November the orbit took the spacecraft more toward local noon (on the17th, WIND was near 9am local time, halfway between noon and the morning terminator). Therefore, when the station powered up on the 18th, WIND was situated such that the signal would have had to propagate through a thicker, denser ionosphere to reach the spacecraft. It was not until WIND had completely orbited the Earth and returned to the same 9am local time, on December 2, that we recorded the broadcast again. Clearly the conditions had to be perfect to make this reception possible.

Two papers, one submitted to Nature and one being prepared for submission to Radio Science, provide more detailed analysis of the WAVES observations of man-made signals above 1 MHz.


Figure 1

Mike Desch is a Co-investigator on the WAVES instrument and Mike Kaiser is the Principal Investigator. Both are radio astronomers in the Planetary Magnetospheres Branch of the Laboratory for Extraterrestrial Physics.

M. D. Desch and M. L. Kaiser
Goddard Space Flight Center
Code 695
Greenbelt, Md. 20771

High School Students Use WIND SWE Data to Investigate Space Weather

T. G. Onsager and the WIND/SWE Team (K. W. Ogilvie, R. J. Fitzenreiter, A. J. Lazarus, J. T. Steinberg, R. B. Torbert, G. Needell and J. D. Scudder)

Solar wind velocity measurements obtained by the SWE Faraday Cup are providing an exciting way for students to learn about the connection between the sun and our near-Earth space environment. High school students are participating in a one-month summer institute named Project SMART (Science and Mathematics Achievement through Research Training) at the University of New Hampshire. One of the research topics involves predicting space weather, and the students are finding the WIND data to be a vital part of their project.

The students involved in the space weather project use x-ray images from the Yohkoh satellite, WIND Faraday Cup data, the Kp index, and simple magnetometers that they build themselves. They use the data from these different sources to achieve a number of goals. First, they are excited by the beautiful images and the comprehensive data available to them, and they discover that the solar wind has measurable and important effects on our environment. Second, they are motivated to make many basic calculations based on the data that reinforce their math and science skills. And third, they use the data and their own calculations to predict space weather and to make decisions that could have important applications in today's society.

An example of the data the students use is shown in Figure 1. This figure contains three months of solar wind velocity measurements and the Kp index from earlier this year. These data illustrate some important properties of solar-terrestrial connections. The students see that the solar wind velocity is well correlated with the Kp index, and the activity level often repeats in 27-day intervals. From their investigation of Yohkoh images, they measure the rotation period of the sun (about 27 days) and notice distinct features such as coronal holes and active regions. By discovering the 27-day periodicity in the solar wind velocity and Kp, they learn that the sun is directly influencing geomagnetic activity. The students use the measured velocity of the solar wind to calculate how long it takes the solar wind to reach the Earth from the sun (about 3 to 6 days). They can then associate specific regions on the sun, such as coronal holes, with the properties of the solar wind that are measured a few days later. In addition, the students use their own simple magnetometers to discover that our magnetic environment is variable and directly influenced by the solar wind.

From the data and the results of their calculations, the students are then able to make simple space weather predictions. Their first prediction only assumes that tomorrow's geomagnetic ions will be the same as 27 days prior. They then study WIND data and Yohkoh images from the past few days and compare these with the the data from the previous solar rotation. By seeing if any new features have appeared in the Yohkoh images and how the solar wind properties have changed, they can modify their initial prediction. Finally, the students are occasionally able to view real-time WIND data, to assist in making "late-breaking" predictions that could warn satellite operators or power companies of CME's or other extreme events that would strike the Earth within about one hour.


Figure 1

(Top) solar wind speed measured by the WIND/SWE Faraday Cup and (bottom) the Kp index from February 1 through May 1, 1995. The Correlation between solar wind speed and geomagnetic activity is evident, as is the 27-day periodicity in the solar wind speed.

K. W. Ogilvie
Goddard Space Flight Center
Code 692
Greenbelt, Md. 20771

Automatic CDF Validation Software (ACVS)

Skeleton tables and CDF files are validated by the KPGS Integration Test Team using software that was developed by Jack Tsou. The ACVS will read a CDF created from a skeleton table by the CDFskeleton tool; validate entries in the header, the global attribute (GA) section, and the variable attribute (VA) section; and generate a report file containing the validation information. This report is sent to the user to use in making corrections to the file.

In the header section, if DATA_ENCODING and/or FORMAT are incorrectly defined, the ACVS reports it as an error in the skeleton table. If any of the 11 required global attributes in the global attributes section is missing, undefined, or incorrectly defined, it is reported as shown below:

    GA          missing                 <attribute name>

    GA          undefined               <attribute name>

    GA          defined incorrectly     <attribute name>

All remaining GAs included in the file are flagged as optional or other by the ACVS.

If any of the required variable attributes in the variable attributes section is missing, it is reported as shown below:

    VA          missing                 <attribute name>

All remaining VAs included in the file are flagged as optional or other by the ACVS.

The ACVS also checks the following information for all variables in the variable attribute section:

For integer or real variables:

  1. Checks that the values of FIELDNAM, LABLAXIS or LABL_PTR_1, and UNITS or UNIT_PTR do not exceed the maximum number of characters.
  2. Checks that the values of LABL_PTR_1, FORM_PTR and UNIT_PTR are included in the skeleton table as non-record variance (NRV) variables and have the same dimension variance (DV) and the same number of elements as their parent variables.
  3. Checks that the values of SCALEMIN and SCALEMAX are within (+/-) 10% of the range of VALIDMIN and VALIDMAX. For Epoch and Time_PB5, the software verifies the proper case and that the year field for VALIDMAX is 2020 and for SCALEMAX not greater than 2020.
  4. Checks that the number of values associated with any one of SCALEMIN, SCALEMAX, VALIDMIN, and VALIDMAX matches the size declared in the header when the values are unique. The software will allow a single value if the size in the header is greater than one and the values are the same.
  5. Checks that the value of VAR_TYPE is data.
  6. Checks that the value of FILLVAL follows the ISTP standard as defined in the ISTP standards and conventions document (see reference 1).
  7. Checks that the value of DICT_KEY follows the ISTP standard.
  8. For all record variance (RV) variables except Epoch, the ACVS checks that DEPEND_0 is included and that its value is the required Epoch variable.
  9. For DV variables, the ACVS checks that the variable attribute DEPEND_1 is present and that its value exists as an NRV variable in the skeleton table, unless the variable itself is NRV.
  10. Checks that the values of LABL_PTR_1, DEPEND_1, DELTA_PLUS_VAR, and DELTA_MINUS_VAR are included in the skeleton table and have the same DV and the same number of elements as their parent variable.
  11. If DELTA_PLUS_VAR and DELTA_MINUS_VAR are present, the ACVS checks that their values exist as variables in the skeleton table and have the same data types as their parent variables.

For character variables:

  1. Checks that the value of FIELDNAM does not exceed the maximum number of characters as defined in the ISTP standards and conventions document.
  2. Checks that the value of VAR_TYPE is metadata for NRV.
  3. Checks that the value of DICT_KEY follows the ISTP standards as defined in the ISTP standards and conventions document.

If any of these variables is missing , undefined, or defined incorrectly, it is reported as shown below:

<variable name>      missing               <attribute name>

<variable name>      undefined             <attribute name>

<variable name>      defined incorrectly   <attribute name>

In addition, the ACVS generates a summary at the end of the report that lists the following information:

          Required GA Present nn Out of 11
          Required GA Defined nn
          Required VA Present nn Out of 16
          Req. Variable Present nn Out of 2
          % complete variables = nn

where nn = actual number

(complete = all VAs are present and defined, % complete = number of complete variables / total variables)


  1. National Aeronautics and Space Administration (NASA), International Solar-Terrestrial Physics (ISTP) Key Parameter Generation Software (KPGS Standards and Conventions, Version 1.3, March 1994.

  2. NASA, NSSDC/WDC-A-R&S 9131, NSSDC CDF User's Guide, Version 2.4, January,1994

G. Blackwell
Goddard Space Flight Center
Code 560.7
Greenbelt, Md. 20771


Ed Nace

The Mission Operations and Systems Development Division of the ISTP ground segment supported the POLAR Thermal Vacuum Test from March 6, 1995 to April 20, 1995. Subsets of the TV test were identified, processed and made available on-line, and later distributed on physical media. Due to the large volume of data requested, and the manner in which the clocks were updated, it was too costly to level-0 process all the TV data in a timely manner. As a result, the ISTP Project Office solicited key processing periods from each of the PI teams. The PAC developed a "Data-map" which was placed in the CDHF News Bulletin Board (PAC News Group) to reflect data on-line status and to correlate filenames to specific time periods during the thermal vacuum test period. Data remained on-line at the CDHF for seven days (normal level-0 retention period), and then archived to the DDF (where it will remain until 30 days prior to the POLAR launch). Archived data is available on request (within 24 hours) via the CDHF User Data Interface.

Once all twenty data days were processed, a level-0 CD was generated according to the ISTP Project's authorization list for distribution to the PI-Teams. The CD-ROMS only contain copies of the specified PI-team's data request and do not represent what will be received during nominal operations. Data from other instruments must be retrieved electronically via the CDHF UDI software. If you are unfamiliar with the UDI process, please feel free to contact Ann Strickling at ISTP::PAC or at 301-286-9453.

Ed Nace
Greenbelt, Md 20771

The Program Assistance Center (PAC)

Ann Strickling

The Program Assistance Center (PAC) has recently expanded its hours, which is good news for the users on the west coast. The new support hours are Monday through Friday 7:30 AM to 9:00 PM DST. Joining the PAC staff is Montgomery (Monte) Raimond, who has over ten years operations experience. The phone mail on (301) 286-9453 continues to be monitored during the weekend day shifts to respond to critical events.

One of the best aspects of PAC is that when the user doesn't know exactly whom to address about a difficulty they have encountered, the PAC should be the first point of contact. If the question is outside the PAC expertise, the PAC staff will coordinate getting the answer from the individual who knows.

The PAC staff can be contacted at:

Ann Strickling: (301) 286-9453 with phonemail
Internet:, or

Monte Raimond (301) 286-2487
Internet:, or

Included in this and future newsletters will be answers to frequently asked questions about the ISTP User Interface (UI). If you have a topic that would be of interest to the user community, please let the PAC know. Also, if you see ways that would improve the UI, please notify the PAC because it's probably desirable to others as well. One of the changes made to the UI this year was creating standing requests which would execute multiple times in a day. This assisted the GEOTAIL users who were requesting the GEOTAIL quicklook data files, which are created about three times per day.

Frequently Asked Questions about the ISTP CDHF

Q. Why were no files transferred for my standing request?

A. The standing request is created for a particular mission and data type. When a new file is received on the CDHF and it matches the specified type, the file is flagged to be transferred. If no files have been sent, two things may have occurred. 1. No new file has been received on the CDHF. Sometimes the level-zero day group is delayed awaiting the retransmission of data. Check the PAC Newsgroup in the ISTP NEWS bulletin board to see if a notice on the delay has been posted. 2. Sometimes if the previous transfer request fails during the transfer process the previous job does not properly clear the system. The data base administrator will have to manually clear the system for the standing request to execute again.

Q. Why do I keep getting a warning message that I have exceeded the transfer quota?

A. The transfer quotas were calculated for approximately 10 % of the total data generated. The idea was that the user could see a portion of the data files soon after receipt with the bulk of the data delivered on the CD-ROM. Because the communication lines are not being impacted, the transfer quota is not currently being enforced, so only a warning message is being generated and transfers are not being affected. If at a future time the communication lines are impacted, the transfer quotas will have to be reviewed and re-evaluated.

Q. How long will it take to get a file that is off-line?

A. Data files are archived in another facility. How much time it takes to restore the data file on the CDHF depends on how busy the communication line is between the two facilities and the number of files that need to be restored. It should not take longer than 24 hours to restore the file on the CDHF. In the event of a critical support, please contact the PAC.

Data Distribution Facility (DDF) Support for the ISTP Program

Ken Lehtonen

The DDF has played a major role in the distribution of ISTP scientific data products to a worldwide community. From its humble beginnings as primarily a distributor of nine-track 6250 bpi magnetic tapes, the DDF now distributes the majority of its data products on Compact Disk-Read Only Memory (CD-ROM) media. To date, the DDF has distributed over 4000 CD -ROMs in support of the ISTP Program.

During the initial planning stages, the ISTP Project played a key role in helping to define the overall directory structure and scientific contents of each CD-ROM distributed to the user community. In addition, by adopting the ISO 9660 standard, a typical CD-ROM produced by the DDF can be read on multiple hardware/operating system platforms (e.g., Unix, MS-DOS, VMS, Macintosh, etc.). As the CD-ROM technology has evolved and matured, so has the DDF in adopting and applying the latest technology. For example, the cost to the ISTP Program for each Level-zero CD-ROM produced (minus shipping costs) last year was about $12 per unit; recently,the DDF negotiated a contract with a vendor reducing the cost per unit to $6 for a very high-grade CD-ROM specimen. The DDF is also in the process of significantly upgrading its hardware infrastructure to enable support for the upcoming Solar and Heliospheric Observatory (SOHO) and Polar Plasma Laboratory (POLAR) missions. (A CD-ROM duplicating system was procured that enables the reproduction of a full CD-ROM in 18 minutes compared to the normal 72 minutes.)

In the near future, the DDF will be expanding its capabilities for electronic distribution of data products (see accompanying articles below) using the TCP/IP File Transfer Protocol (FTP). To support the SOHO mission, for example, the DDF will send Quicklook data electronically to the SOHO Experiment Operations Facility located at Goddard.

The DDF Architecture

Ken Lehtonen

The Data Distribution Facility provides the data distribution and delivery functions for the GSFC's newly formed Mission Operations and Systems Development Division (comprised of the former Information Processing Division). By implementing generic distribution requirements, the DDF has evolved into a multi-mission system saving the added expense of building, operating, and maintaining unique facilities and capabilities for each mission. In its current configuration, the DDF is comprised of two halves: The Physical Distribution System (PDS) and the Electronic Distribution System (EDS).

The PDS portion of DDF distributes data on physical media such as CD-ROM and 8mm tape. The PDS consists of two DEC/VAX mainframes, the Data Recording Subsystem (DRS), and a Sun workstation with an 8mm tape drive. The DDF PDS is operational and currently supports ISTP's Geomagnetic Tail (GEOTAIL) and Interplanetary Physics Laboratory (WIND) missions and will soon support the SOHO and POLAR missions upon their launches. All CD -ROMs are produced by the DRS portion of PDS.

The current DRS consists of PC workstations with Philips 2X CD-write drives that interface with the VAX through a DEC Pathworks interface. The soon-to-be operational DRS+ consists of a Novell network of PC workstations with Philips CD-write drives, a PC file server, and a Control-PC. The DRS+ file server communicates with the PDS via the standard Network File System (NFS) protocol.

The DRS+ greatly improves upon the current DRS design in that it centralizes operations (Control PC), automates CD-ROM production, increases performance, and enhances verification of the data. These improvements are necessary to support the estimated 350 ISTP CD-ROMs that will be generated each week during the 1996 timeframe. Data products distributed on 8 -mm tapes (e.g., SOHO MDI Wideband) are created by the Sun workstation. The 8mm-tapes created on PDS follow the same directory structure and filenaming conventions used on the CD-ROM.

The EDS portion of DDF distributes data electronically. The EDS is a client/server type of architecture which consists of a network of Sun workstations. The DDF EDS will become operational this August upon the launch of its first supported mission, the X-Ray Timing Experiment (XTE). Some of the other missions EDS will support include SOHO, the Submillimeter Wave Astronomy Satellite (SWAS), and the Hubble Space Telescope (HST). The EDS is comprised of several key elements: the Mission Server Element (MSE) which performs all mission-specific processing such as receipt and transmission of messages, managing data ingest, and interfacing to a mass store system; the File Server Element (FSE) which provides short-term storage of data products ready for distribution to customer systems; the Mass Storage Element (MASE) which provides long-term storage of data products; the Media Generation Element (MGE) which will provide data products on physical media; the System Monitoring Element (SME) which provides the operational interface for all of EDS; and, the Firewall Element (FWE) which provides security between the operational network and public networks (e.g., the Internet).

Eventually, all of DDF will transition to the client/server type of architecture as implemented by the EDS. Portions of the PDS will become the MGE in the EDS architecture. By transitioning away from a mainframe type of architecture, the DDF can incorporate the latest technologies and standards to provide an integrated and flexible system for supporting multiple missions with an added range of physical and electronic distribution capabilities.

Ken Lehtonen
Goddard Space Flight Center
Code 514.1
Greenbelt, Md. 20771

IMP-8 Electronic Distribution - A Progress Report

Michael Anderson

The Data Distribution Facility is currently in the process of converting Interplanetary Monitoring Platform-8 (IMP-8) satellite data currently distributed on nine-track, 6250 bpi tapes to electronic transfer, replacing twenty plus years of magnetic tape dissemination. Eleven IMP-8 user products, all with varying file structures, are being tested to ensure a smooth transition.

The new method of data distribution involves reading the standard IMP-8 shipping letter and defining an expected file base. Files are uniquely renamed reflecting their product group, start time, and instrument descriptor. Data are then transferred electronically via the Network Systems Corporation NETEX Bulk File Transfer protocol (BFX) from a UNISYS 2200/424 to a common directory located on the DDF's VAX cluster. Upon receipt of the IMP-8 data, a special program is executed to assure that all required files have been received on disk; if so, it then transfers a copy of the data to long-term storage and issues electronic mail messages to the respective investigators informing them that their data are available for transfer from online disk. After this time period expires, the files are deleted from disk to conserve resources. Users requiring expired products may request that specific products be reloaded from long-term storage. These older products will also remain disk resident for 14 days.

To date, a majority of the IMP-8 experimenters have successfully tested this procedure. Using unique accounts on the DDF, users have successfully FTPed complete data sets and shipping letters to pre -determined test sites, using a variety of platforms. Electronic data sets were then compared to magnetic tapes containing the same identical data sets. Of sites completing comparison tests, all results have been very positive. Following confirmation from the remaining test sites, a period of parallel data distribution will be in effect, to be followed by an all-electronic transfer operation.

Michael Anderson
Goddard Space Flight Center
Code 517.6
Greenbelt, Md. 20771