Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 215
Review of NASA'S Distributed Active Archive Centers APPENDIX D User Survey The National Research Council's Committee on Geophysical and Environmental Data (CGED) is conducting a review of NASA's Distributed Active Archive Centers (DAACs), each of which manages a different kind of scientific data and serves a unique blend of user communities. The criteria for review focus on how well the DAACs are serving their current scientific user communities and how well positioned they are to serve the much larger community that will use Earth Observing System data and information. Such reviews are beneficial to data centers because they provide critical user feedback to the center and they educate the broader user community about the center's activities. In this sense, the CGED is serving a purpose similar to that of a visiting committee for an academic department. The review seeks to look at all aspects of the DAACs. In addition to site visits at all of the DAACs, the committee is meeting with representatives from ESDIS, NASA Headquarters, and the ECS contractor to learn about their interactions with the DAACs. Feedback from EOS investigators (both users and producers of data and information) is a critical part of the review, and we ask that you complete and return the following brief questionnaire: Questionnaire: 1a. Are you a user of the DAACs? ( ) yes ( ) no
OCR for page 216
Review of NASA'S Distributed Active Archive Centers 1b. If you answered yes to Question 1a, how do you interface with the DAAC? ( ) Web ( ) phone ( ) fax ( ) hard copy ( ) other: Please explain 1c. If you answered yes to Question 1a, are you: ( ) a data provider? ( ) exclusively a data user? ( ) both 2. Which DAAC(s) do you use? ( ) GSFC ( ) LaRC ( ) EDC ( ) ASF ( ) JPL (PO.DAAC) ( ) NSIDC ( ) ONRL 3. How often do you get data? ( ) daily ( ) weekly ( ) monthly ( ) yearly 4. Typical size of data set? ( ) <10 MB ( ) 10–100 MB ( ) 100 MB-1 GB ( ) 1–10 GB ( ) 10 GB 5. How do you get data? ( ) electronic transfer ( ) media 6. Ease in obtaining data in the form you want: ( ) very easy ( ) somewhat easy ( )average
OCR for page 217
Review of NASA'S Distributed Active Archive Centers ( ) somewhat difficult ( )very difficult 7. When problems arise, what do you do? ( ) contact friend ( ) contact help desk ( ) contact DAAC management ( ) other. Explain: 8. Overall satisfaction with the DAAC: ( ) excellent ( ) good ( ) average ( ) below average ( ) poor 9. Major kudos: Use the space below to list major kudos you would give the DAACs 10. Major criticisms: Use the space below to list major criticisms you would level against the DAACs 11. Have you authored any reviewed publications that use DAAC data? ( ) yes ( ) no Provide references (optional) We appreciate your help. Anne Linn National Research Council The CGED has been reviewing the operations of World and National Data Centers for more than 35 years. Among its most recent reports are 1992 Review of the World Data Center-A for Rockets and Satellites and the National Space Science Data Center (NRC, 1993) and 1993 Review of the World Data Center-A for Meteorology and the National Climatic Data Center (NRC, 1994.)
OCR for page 218
Review of NASA'S Distributed Active Archive Centers TABLE D.1. Summary of Survey Results User Category Number of Respondents Number of GSFC Answers Number of LARC Answers Number of EDC Answers Number of ASF Answers Number of JPL Answers Number of NSIDC Answers Number of ORNL Answers Sophisticated 163 103 47 32 27 87 39 21 Casual 46 10 0 2 1 39 4 2 Foreign 184 54 11 9 4 157 22 9 Total 393 167 58 43 32 283 65 32
OCR for page 219
Review of NASA'S Distributed Active Archive Centers FIGURE D.1. Method for accessing DAACs (see survey, Question 1b). Most respondents chose several access methods, so the total responses for each DAAC is greater than 100%.
OCR for page 220
Review of NASA'S Distributed Active Archive Centers FIGURE D.2. Number of DAAC's used (see survey, Question 3).
OCR for page 221
Review of NASA'S Distributed Active Archive Centers FIGURE D.3. Frequency of DAAC access (see survey, Question 3). Several respondents chose several time intervals, so the total responses for each DAAC may exceed 100%. Sophisticated users who selected ''never'' are data providers.
OCR for page 222
Review of NASA'S Distributed Active Archive Centers FIGURE D.4. Size of data sets retrieved (see survey, Question 4). Several respondents chose multiple answers, so the total responses for each DAAC may exceed 100%.
OCR for page 223
Review of NASA'S Distributed Active Archive Centers FIGURE D.5. Form of data retrieval (see survey, Question 5). Most respondents chose multiple answers, so the total responses for each DAAC generally exceed 100%.
OCR for page 224
Review of NASA'S Distributed Active Archive Centers FIGURE D.6. Source of help (see survey, Question 7). Several survey respondents chose multiple means of getting problems addressed, so the total responses for each DAAC commonly exceed 100%.
OCR for page 225
Review of NASA'S Distributed Active Archive Centers FIGURE D.7. Reviewed publications with DAAC data (see survey, Question 11).
OCR for page 226
Review of NASA'S Distributed Active Archive Centers REPRESENTATIVE COMMENTS GSFC DAAC DAAC management is more receptive to the scientists. There is less focus on glitzy stuff, like visualization, and more focus on providing access to data. Goddard DAAC certainly listens to users and wants to provide services that are user responsive. We use their interdisciplinary data sets not just for research but also for education (at the doctoral level). Whenever I have problems or questions the people at the help desk are very helpful and quick with their responses. When a major data revision occurred with the Seawifs data, no attempt was made to say "X file replaces the Y file you already have." We have to run a list of each tape that we have and compare with the listing of the new tape. This process is very time consuming. Very helpful in preparing for TRMM data, helping TRMM get visibility. Some of their products are hard to get into a format I can use. LaRC DAAC The LaRC DAAC Web site is difficult to navigate and find the necessary information and/or data. When I've had difficulty obtaining the data I ordered, some of the user services staff have been extremely helpful. There is no way to create a standing order of a data set. If I want all the data from a project, I must make individual orders. For example, ISCCP D1 has monthly data for at least 12 years; that's more than 144 orders. They are striving to provide close attention to the requirements for our flight project, and make their most senior technical and managerial staff directly accessible to us. EDC DAAC EDC has been very helpful in providing AVHRR and GLCC data sets to us, and has provided great customer service help to us. The speed of delivery of Landsat raw imagery is slow, and it is hard to obtain Level 0 data. There is evidence that the ordering systems are highly nonautomated. In fact I strongly suspect that at times they keep track of orders using lists created by pen and paper. When one is dealing with a major order, this means that at times they have no idea of even where they are in terms of fulfilling the orders. Sometimes we know better than they do, because their order-tracking system is so disaggregated and incompetent.
OCR for page 227
Review of NASA'S Distributed Active Archive Centers Sometimes overloaded and the response is slow in producing the routine products in large volume. Example, the NALC data set for the United States has taken nerarly a year to copy to CDs. When one DAAC manager was asked about why plans for processing data from one of the EOS instruments were apparently so badly behind, he responded that the ECS software was not likely to be ready on time so there was no need for them to be getting ready. This was about a year ahead of the scheduled launch. ASF DAAC The user interface (V0 IMS) is poorly suited to selecting data for repeat-pass SAR interferometry; these data must be selected according to viewed location, date of data collection, spatial baseline, and temporal baseline. The Web interface I have attempted to use catalogs data by location and date only, making cross-references to a database of baselines very difficult. Much of the data present in the archive is not visible on the Web interface. Perhaps other interfaces exist but we can't find them. ASF is doing a good job with radar interferometry software. The Alaska SAR Facility is an excellent resource for valuable, multitemporal SAR data. These data are making it possible to do significant new science. The data request system is working very well, and we get large amounts of data in a timely matter. ASF personnel also deal with us on a personal level and inform us when they have completed projects that may be of interest to us now or in the future. ASF just doesn't seem to be able to deal with users who want data outside their station mask. When we order data, every single time the order bounces the first few times. Sometimes we can tell why the order isn't accepted; usually we simply receive cryptic messages stating that the order isn't acceptable. Finally, after phone calls to the DAAC staff, which are returned when the people are in town, we can usually straighten out the order. Even if the order is OK, we often get strange disclaimers such as "the order may not have been placed even if it appears to have been placed correctly." On several occasions, we got multiple copies of data, most likely because we didn't know when the order is placed correctly. Getting data processed is excruciatingly slow at ASF. This is due to the processor software, not unwillingness. More thought should have been given earlier to a software fix. The ASF data consistently have errors in the formatting, and there are a new source of errors with each data set.
OCR for page 228
Review of NASA'S Distributed Active Archive Centers PO.DAAC Access via the Web site is well-organized and easy to navigate. Fast file transfer times. They have been helpful and flexible when I have dealt with them from my position as a data provider. Their performance on distributing and supporting the ERS-1/2 and NSCAT scatterometer data sets has been outstanding. Their flexibility and knowledge about the details of the data sets and processing (resulting from collocation and tight collaboration with JPL Project Offices and working scientists at JPL) is a hallmark of the PO.DAAC operation. Incredible array of products turned over to the real world in record time with good documentation and all by folks with nice manners. I remember the time I griped to Chris about the end-of-line terminators in the draft final TOGA CD, and he actually changed the whole thing. That's service. The DAAC is very responsive to e-mail questions, data arrive promptly, and when a revision was produced, all the information required was sent before I was fully aware I needed to know about it. Supply of data in CD-ROM is currently, for me at least, a very useful and time saving approach. There was a long startup time before systematic data distribution occurred, but I suppose that can be attributed to the new nature of the TOPEX data and the problems associated with turning production over to a contractor. The JPL PO.DAAC has done a superb job in keeping up with the Geophysical Data Records (GDRs) and revised GDRs for the TOPEX altimeter ever since the launch in 1992. The software provided together with the data for reading and manipulating purposes could improve, certain standards of portability and software quality control should be demanded. The databases provide students in my classes an opportunity to see and use data sets in problem assignments. It's a valuable resource in that it is readily available and in a form that easily integrates into programs they can use for analysis (spreadsheet, graphics, statistics, etc.). HDF is awkward. Some data sets have much more data than I need. Media formats are not always easy to transport from system to system. I would like to see a policy that every data set comes with a sample program to read, subset, and write an ASCII file of the subset. Subsetting should be by variable, time, space, every nth, first mth, last qth. Such a program should be easily modified by the user. I suggest two versions: FORTRAN and C. At a minimum these should be supported for a current PC operating system and a current Sun operating system.
OCR for page 229
Review of NASA'S Distributed Active Archive Centers NSIDC DAAC Good homepage, good interface for ordering data. NSIDC in particular is very much in touch with user needs. It understands the user's issues. NSIDC's and ASF's (for example) science background allow them to interact intelligently about the data being provided. They are always looking at how to best provide the information. User services at NSIDC is excellent! Very helpful, promptly address any questions or concerns. They have a lot of "in-house" expertise which I think is essential for a data center—that is, a DAAC must be more than just a "data warehouse," it has to be a resource center for the user community and a focus for acquiring data sets, possibly extracting higher-order data products. I think NSIDC excels at this. Communication with the user community is excellent through publications, the Web site, and the active involvement of NSIDC scientists and staff in research activities (the latter is very important). For example, NSIDC staff have attended all of our recent CRYSYS IDS annual meetings and made important contributions to the project. Need to have good visualization software for the data. They don't have a lot of imagination when it comes to developing methods to work with the data. ORNL DAAC Only one respondent provided written comments: DAACs have made strides providing value-added products such as ORNL's NPP products. These products show that good science and good data management are not mutually exclusive.
OCR for page 230
Review of NASA'S Distributed Active Archive Centers This page in the original is blank.
Representative terms from entire chapter: