Please note that all abstracts and articles in the ADS are copyrighted
by the publisher, and their use is free for personal use only.
What is the ADS Abstract Service?We are an abstracting service offering a search interface into the scientific and technical literature covering astronomy, planetary science, physics, and the arXiv e-prints. You can use our service to search the contents of hundreds of publications in these fields and then optionally access the fulltext of articles of interest to you if and when it is available online. Note, however, that in many cases access to this content is restricted by the publisher, and there is nothing we can do to help you gain access to it.
Searching the Databases:
Additional information about using the ADS abstract services is available from the abstract service help page. If you have other questions, please feel free to send email to us at firstname.lastname@example.org
How complete is the Abstract Service?The abstracts database contains data from several sources, including NASA's Scientific and Technical Information group (STI), journal publishers SIMBAD, NED, and typed from table of contents. The original primary source of abstracts has been NASA's STI, which provided abstracts from 1975 through about the middle of 1995. Since then, we have been receiving the majority of our abstracts directly from the journal editors. Starting in 2007 we have been able to further complete our coverage of historical records thanks to our access to CrossRef, which collects and makes available metadata records for most of the content produced by Scientific journals.
How up-to-date is the Abstract Service?There should not be any significant delay in getting a paper from one of the major journals into our system, which is updated weekly. Records which we receive directly from journal publishers typically go into the system on or before the publication date.
How far back do the abstracts go?For any journal we have scanned and placed online, we are complete back to Volume 1. A complete listing of these is available from our Article Service. For those journals which we have not yet scanned, we typically go back to about 1975 when the STI coverage begins.
How complete are the citations/references in the Abstract Service?In early 1997 we started compiling for a subset of the bibliographic entries in the astronomy database their list of references and citations (the references of a paper are the papers cited by it; the citations of a paper are the papers which cite it). The original dataset used to create these cross-references was purchased from the Institute for Scientific Information (now Thomson Reuters), producer of the Science Citation Index, and consists of references from articles published by the major astronomical journals between 1980 and 1998 to articles in the same date range.
In 1999 we started extracting reference lists from the full-text of papers available in the ADS article service or provided to us by the journal publishers. In October 2007 we gained access to CrossRef's metadata, which includes references from many articles in the physical sciences. As of June 2009 we have parsed and identified over 36 million references from all the sources of bibliographic metadata available to us. While the addition of these references has greatly improved our coverage, users should realize that we only have been able to generate these lists for a well-defined set of publications and in well-defined time spans. In addition, references may be incomplete due to our inability to match them with 100% accuracy (e.g. in press, private communications, author errors, some conference series, etc.). Anyone using the citations for analysis of publishing records should keep this in mind.
In our effort to expand the coverage of references in our databases, we strongly encourage the submission of lists of references appearing in a given publication from any of the following sources:
How can I get a list of my citations?Perform an author search using the ADS Abstract Service Search form. Make sure to increase the default number of returned items so that all your papers are listed in the resulting list of papers. Now go to the bottom of the page and click on ``Get citation lists for all articles in above list." The resulting screen will contain all papers citing one or more of your papers. The score displayed next to each citing paper is the number of times one of the original papers has been cited by the paper in question. The number of citing papers as well as the total number of citations are displayed at the very top of the list.
You can also choose to view the list of papers citing just one or more papers by checking the buttons next to the bibcodes of each paper and then clicking on ``Get citation lists for selected articles." When the input query was an author-based one, you will also have the opportunity to remove self-citations by checking the corresponding checkbox at the bottom of the results list.
Why is a citation from an e-print missing?Starting in 2005, we have begun integrating citations obtained from parsing the reference sections of e-prints from the ArXiv. As with references from published papers, there is always the possibility that our procedures may be unable to successfully extract and identify the references. If you think we have missed a citation to your paper from an e-print, you should first check that e-print's list of references as parsed by ADS to see if your paper is amongst them. If it isn't, then you can help us by sending us the relevant information as outlined below.
For more information, please see our help page on citations.
Can you add these citations to your database?The answer to this question is (a qualified) yes. We will accept user-submitted lists of citations currently missing from our databases. Please use the Reference submission form to provide us the proper information. We also welcome feedback on the accuracy of our existing reference lists. If you come across cases where we have missed or mis-identified a reference, please let us know and we will fix the relevant records. However, please note that we will not match a record in our database with a reference which does not have sufficient information to allow its identification with high confidence. In particular, please note the following rules concerning references to papers which are in the process of being published:
What journals are abstracted?The NASA database contains abstracts from hundreds of journals, publications, colloquia, symposia, proceedings, and internal NASA reports. Complete coverage can not be guaranteed for any of the sources. For a listing of journals included in the abstract service, see the journal list.
Why isn't this paper in there?Often when a paper is missing from one of the databases, it is often because it was published in a journal which does not provide us with article metadata.
We welcome the submission of missing papers, but request that they be submitted to us through the Abstract Submission Form. If you have many abstracts to submit, you may also email them to us, provided they are formatted in the required format.
How do you evaluate journals for inclusion?The ADS evaluates all journals for coverage using the criteria listed below. We seek comprehensive, but not all-inclusive, coverage of the field of astronomy, and we take into consideration relevancy of content, standards of presentation, and regularity of publication. We evaluate journals on an ongoing basis and add or delete journals regularly. ADS reserves the right to exclude content deemed inappropriate for its collection or inconsistent with its editorial policies.
To be accepted for coverage, a journal must:
In addition, e-journals must:
Features that support acceptance:
Relevance: An article that includes any of the following may be relevant:
How do you determine if a publication is refereed or not?ADS considers articles refereed when they appear in journals that participate in peer review. Peer review is the process of having articles reviewed by experts in the field before publication. ADS staff check journal websites and author instructions to verify the peer review status. Sometimes periodical directories are consulted, such as Ulrich's International Periodicals Directory.
Non-refereed materials, such as conference proceedings, circulars and bulletin entries, are screened only by an editor and not sent out for peer review. The ADS does not consider materials that are verified only by an editor as refereed.
Occasionally there may be refereed articles in a non-refereed journal or non-refereed articles in a refereed journal (e.g. announcements or conference abstracts). ADS staff will mark the individual articles as refereed or non-refereed once this difference is known.
How can I get the full text of the paper I've found?The ADS is primarily an abstract service, however when we know that a certain article has been published online, we do provide a link to the fulltext, both in the available items column (as an "E" or "F") as well as at the top of the page displaying the abstract. If we do not provide that link, we do not have any further information. For journal articles, you might want look at the publisher's website directly. Back to top
Why is this reference incomplete/wrong?Since we do not write the abstracts, we can not assume responsibility for incorrect references. However, we would be pleased to correct any entry in the database which is incomplete or incorrect. Please use the Feedback Form to provide this information to us.
How are readshelp for a paper calculated?For each paper, a "read" is counted if an ADS user runs a search in our system and then requests to either view the paper's full bibliographic record or download the fulltext. To see the readership history for a paper, click on the "Readshelp History" link at the top of its full bibliographic record display page. Please note that in computing readership numbers we attempt to remove log entries generated by robots, users coming to an ADS record from an external search engine, and multiple clicks from the same user. Given the uncertainties, biases, and potential for abuse involved in collecting usage data, we discourage using our readership counts as a measure of a paper's scientific value.
One useful ADS feature that readership data makes possible is the creation, for each article, of its list of "Also Read" papers. This is the list of papers that have been read, in the last 90 days, by those users who read that particular paper, ranked by readshelp. The list therefore offers an additional way for a user to find papers that are likely to be relevant to the original record, based on usage patterns.
Please note that the list of "Also Read" papers includes, for e-print records, readership statistics from both the ADS and the arXiv itself, which in some cases may increase the total number of readshelp by a factor of 10 or so. The reason we add records from arXiv's anonymized logs is because this gives much better results when computing article co-readership lists. arXiv usage is not otherwise included in the article readership data displayed by ADS.
Why are some abstracts in the ADS database different from the published ones?Some of the abstracts in the ADS database (most abstracts in the time span 1975-1995) were given to us by the STI program at NASA, which in most cases rewrote the abstracts to avoid infringing the publisher's copyrights, often shortening the author's original abstract. In general, there is no way to be certain whether an abstract which originated from STI has been rewritten or not.
Abstracts that we receive directly from the publishers and the original authors (called "author" abstracts) are instead a faithful and accurate reproduction of text that was published in the paper's abstract (except for math and non-ASCII symbols, of course). Individuals who wish to submit original abstracts to be included into our system can do so through the Abstract Submission Form (see also previous question).
Why are proposals indexed in the ADS?ADS is now indexing proposals from several major institutions, including Chandra, HST, and IUE. Although we recognize that these are not published literature, they were already publicly available through the websites at those institutitions, and by including them in the ADS we are allowing users quick access to the proposal data, which was otherwise difficult to get to. Successful observing proposals are thoroughly scrutinized by the telescope committees, and therefore represent significant scientific content.
Anyone wishing to exclude these from their searches can enter the string "-?????prop" in the box next to "Select/deselect publications" in the filter section of the main query form.
Why is there an "E" in the bibcode?ADS currently includes abstracts from some conference proceedings prior to their publication. When conference editors send us abstracts which are submitted to them to be included in meeting abstracts books, we include these with an "E" in the 14th column of the bibliographic code (e.g. 1998bllp.confE...1U). Since these typically do not have page numbers, we use a "counter" in place of the page number and sequentially number the abstracts as received from the conference proceeding editors.
When these are later published in proceedings with page numbers, the E abstracts are replaced with real printed proceedings abstracts containing page numbers. Any E abstracts which are not matched with the printed versions are deleted at this point (presumably because the author did not attend the conference or withdrew the paper). It is therefore not adviseable to create hyperlinks to one of such entries, since they are to be considered "temporary" bibliographic references.
We invite all conference proceeding editors to submit their abstracts to ADS by sending email to email@example.com either before the conference, after they are published, or (preferably) both.
Can I archive my proceedings articles through the ADS?As more conferences are published electronically, editors are turning to the ADS for long-term archiving of their full text proceedings. We can agree to this under the following conditions:
1) The conference is originally published elswhere (e.g. through a university or observatory publication, or through another publisher).
2) We receive the pdf or postscript files as from the conference proceedings editors as described in our conference help pages.
3) The data have been published at least three years ago.
Why are the BAAS entries in the system in two different formats?Starting with the 180th AAS meeting (June, 1992), the AAS began collecting abstracts for the BAAS in electronic format. We have put the abstracts for this meeting and all subsequent meetings into our database. However, the abstracts which are submitted electronically do not contain page numbers. It was therefore necessary for us to find another way to differentiate papers, since it was impossible to create unique entries on volume and author initial.
We decided to use a different journal abbreviation (AAS), together with the information contained in these versions -- the meeting number and the session number to create the bibliographic code. Therefore, most recent BAAS abstracts are in the following format:
where 187 is the meeting number and 9004 is session #90.04.
However, we also receive these abstracts from Simbad, who enter them into the system from the printed BAAS booklet, so those references include the volume and page number (1995BAAS...27.1417J). We make every effort to match these two references so that each abstract is only in the system once. Note that a "bibcode query" for either of these formats will find the appropriate paper.
This is also true for the DPS and SPD meeting abstracts, though their meeting number is the same as the volume number. Therefore, recent DPS and SPD meeting abstracts (i.e. the 28th meeting, session #12.16), have bibliographic codes such as:
We apologize for the confusion caused by these two different naming systems, but we must work with the data we have, which is sometimes incomplete.
How do I link together articles such as an errata with their original article?If you know of articles which are associated with each other (e.g. an article and its erratum or a series of articles), we can include this information in the ADS. Please use this form to provide us with this information. Back to top
How can I tell which journals are in which database?The split between categories is not made by journal, it is made by the subject of the article. For example, there are ApJ articles in all three databases. It is necessary to keep the databases separate in order to keep the searches working as well as possible. If we were to combine the databases, we would lose the importance of the frequency of the word in a given dataset. A word which is very common in one database (and therefore given less weight in a search) may be uncommon in another database, in which case the word would be given a lot of weight. The searching algorithm is one of the real attributes of the service which would be severely degraded if we merged all the categories.
Who owns the copyright to these abstracts??Most of the abstracts are copyrighted by the journals which have kindly provided them to us. It is illegal to download abstracts for any reason other than personal use without permission of the copyright holder. For more information, please read our page detailing the Terms and Conditions regulating the use of our resources.
Why am I getting a page without results after a search?Some Windows users have reported problems in seeing the list of results from an abstract query. The page that is displayed in their browser consists only of a header and a footer showing the query parameters. This problem is typically caused by one of Norton's security programs that act as firewalls between the browser and the content that is returned by our server. Some of these programs filter out part of our result pages in an attempt to weed out advertisements (this is what we get for having "adshelp" as part of our hostnames).
The solution to the problem depends on the product that you are using, but typically involves configuring it so that it "trusts" content coming from our site. Sometimes clicking on options such as "enable pop-ups from this site" (when available) is enough, other times the solution is more involved. We recommend you consult your product's documentation.
Additional help in the form of step-by step instructions contributed by our users is available
Why does my search generate so many results?By default, our search engine returns any reference containing one or more of the words (and their synonyms) that have been specified in any of the search fields. If you want to narrow your search you may do so by modifying the settings of the abstract query form as appropriate. Here are some of the available options that can be used for this purpose:
How can I perform a boolean search?By default, the abstract service "OR's" words together. Therefore, a search on two authors returns papers written by either author1 OR author 2. The list is ranked by score so that papers written by both authors are listed at the top since they have a score of 1.
If you would like to see only papers written by both authors (or which contain all words which you have entered in any other field), you need to change the combination method to "AND" above the text entry field.
In addition, there are now two additional options for search logic:
SIMPLE LOGIC QUERIES -- The simple logic recognizes '+' and '-' before the words. To require a word to be found in a search, it needs a '+' in front of it. A '-' before a word indicates that only references that do NOT contain that word should be returned. AND , OR, NOT are stopwords and will be ignored in the simple logic.
Note that if you want to exclude a word, it is best if it is not the first or only word listed in the search. This requires every abstract in the system to be returned, before the excluded word can be found and omitted from the returned abstracts. These searches can therefore take a long time.
FULL BOOLEAN QUERIES -- This allows more complex queries than just combining all words with "OR" or "AND". The allowed boolean operators are: "and", "or", "not", "(", and ")". They can be used in any combination (as long as "(" and ")" match). For example the query "(redshift or survey) and not galaxy" finds all references that contain either "redshift" or "survey", but not "galaxy". The order of precedence of the operators is "(...)", "not", "and", "or". For larger numbers of search terms, this type of search will be slower than regular searches, especially if the "not" operator is used. Regular scoring is done on any terms that are combined with "or". "and" and "not" combinations are scored as 1.
How can I search for a phrase like "black holes"?The abstract service supports searches for "phrases" by entering them between double quotes.
How can I find papers in which XXX is first author?To force an abstract query to return only papers which have a particular author as first author, add a carat ("^") to the author's name in the search field. For example, searching for ^JONES, C will return papers in which C. Jones is the first author.
Why do my search parameters disappear from the query form?Under certain circumstances, your browser may display a blank abstract query form when you use your back button after searching. This is usually a symptom of the fact that your browser is not caching the search form properly, presumably because the cache directory is full. We have had several users report this problem when using Netscape 2.x and 3.x under UNIX platforms. A simple solution to clear this problem under Netscape Navigator 3.x is the following:
Why am I not able to store a default search form?ADS allows you to modify your default search parameters and store them in a form that can be used on a regular basis. If you modify the search parameters at a later date, the page dislpaying this form needs to be refreshed so that the new settings can take place. Usually clicking on the ``Reload'' or ``Refresh'' button is enough to force your browser to load the new parameters. However, sometimes this does not seem to be sufficient. Here are a few things you can do to correct that:
Why can't I search on keywords?The keywords formerly used in the abstract service were those assigned by NASA's STI group when they rewrote the abstracts. However, as more and more of our data came from other sources, it became apparent that keyword searches were giving both incomplete and outdated results.
We have now indexed keywords from all sources into the abstract text words. For best results on searching, enter words ("keywords" and others) in the abstract text field.
If for some reason, you want to do a NASA/STI keyword search, we still provide that capability through our Abstract Legacy Form, but understand that results will be outdated (since we no longer receive abstracts from NASA/STI) and incomplete (since keywords from other sources are not included).
How can I link to my papers and articles?ADS allows you to compose URLs corresponding to queries that can be used to generate the list of papers published by a particular author. All you have to do is compose the appropriate URL as detailed below. (Please note that the URLs shown here have been broken up in several lines for legibility; you should always use a single line when creating a URL).
If you want to specify the author middle initial in addition to the first
initial, use exact author matching (&aut_xct=YES):
If you know the
of the paper you want to link to:
What are cookies and why do you use them?HTTP cookies allow us to customize many of our site's functionalities to the user's preferences. If you are unable to set up a user account in ADS or use our login system it is probably because you have configured your browser not to accept any cookies. Please see your browser's settings to correct this.
If you choose not to let our server set a cookie with you, you will not be able to take advantage of all of the features on our system. You will still be able to use the server for basic searching, provided you select the search page which disables cookies:
Can I search by affiliation?We have not integrated an affiliation field in the main abstract service search form because affiliations found in the ADS databases are inconsistently formatted, contain a lot of noise, and most importantly only exist for about half of the entries in the database, which means that a search by affiliation would generate very biased results.
However, given the amount of requests we have had on this subject, in 2005 we introduced a separate query form allowing a user to search for different affiliation spellings in the database and subsequently retrieve any records containing them. The search form is available at:
For researchers wishing to analyze affiliations on a large-scale basis, we can provide this information on a collaborative basis. Note however that because of the limitations of this type of search we recommend that people use author searches instead when compiling bibliometric studies for particular institutions.
How do I filter my results list?For example: you have a list with records (as the result of query or the contents of a private library) and you want to see just the refereed publications in this list. To filter out your refereed papers, select all your publications and click on the button "Get form to query selected articles only". In the resulting form, put a year that is before your first paper in the start year box for "Publication Date" and select "All refereed articles" in the "FILTERS" section. Pressing "Send Query" will result in a list of all your refereed papers. If you want to filter by additional or other criteria, you need to specify these in the query form. Back to top
I need to generate the number of citations from a certain period of time to my papers. How do I do that?Let's assume that you have created a list which contains all your papers (and only your papers). This first step can be as easy as running an author query or creating a private library which contains only your papers (make sure you get all of them!) To find out the number of citations published in a particular time range, first you should select all the records in your original list. Go to the bottom of the page and in the section named "Find papers related to the above articles" specify the period for which you need the citations in the month and year boxes following the text "Publication Date between". For example, if you need citations for the period 09/2015 through 10/2016, you specify these dates. Press the button "Get citation lists for selected articles". The resulting page will display the records for papers citing your articles in the requested time range. The heading above this list will read "Retrieved 200 abstracts, starting with number 1. Total number selected: XXX. Total citations: YYY". This means that your papers have accumulated YYY citations from a total of XXX papers. Additional hints: if you need to perform this computation on just a subset of your original list (e.g. just the refereed papers) you can first filter out the unwanted records as described in the FAQ How do I filter my results list? above.
Why won't the articles print for me?We recommend downloading the PDF version of a paper whenever possible, and then configuring your system to display and print PDF files locally. When downloading PDF files, occasional printing problems may occur when the articles are long or contain graphics which take a lot of space. Often this is due to a lack of disk space on your system or printer spool area. Please clean up these areas in case you are having trouble.
Additional issues related to printing the fulltext documents served by ADS are described in our article retrieval help page. Back to top
Why are the article files that I print so big?First of all, you should realize that some of the full-text files that you can download from ADS are made available by the journal publishers rather than by the ADS article service itself, so when you run into problems please make sure you tell us what article (bibcode) you are trying to print and what the symptoms are. If you are downloading PDF files either from the ADS article service or the publishers' sites, it is important that you configure your PDF viewer to take advantage of your local printing capabilities to avoid creating huge printing jobs.
If you use Adobe Acrobat Reader on a UNIX machine to view your PDF files and your printer supports PostScript level 2 (most modern printers do), then you should make sure that you select the following settings in acrobat's printing menu (which pops up when you click on the "Print" button):
Why do I get prompted for username/password when following a link from the ADS?All "E" and "F" hyperlinks available in a list of results generated by an ADS query point to documents which are available on web sites maintained by other institutions, which may enforce some kind of access restriction to the documents. The typical case is following links pointing to an electronic version of a paper made available by the publisher. If the publisher is restricting access based on IP access (for institutional subscribers) or username/password (for individual subscribers), then you may be denied access or prompted to enter a username and password if you're not a subscriber. Often, your institution's library may be able to provide you with a username and password. For detailed information about these issues, please see our access help page.
I notice that the record ADS displays for the paper I published contains my email address and I am concerned about its potential abuse.We take several steps in order to protect the privacy of the people whose emails appear in our records. Among them:
How can I search for a particular journal by volume?There are two ways to do a search for articles appearing in a particular volume of a journal. A "Table of Contents" search is available for the most popular journals in each of the three databases (Astronomy, Instrumentation, and Physics and Geophysics).
For all other journals, you can do a "BIBCODE_QUERY", linked off the top level search form, provided you know the journal abbreviation (listed in the journal list).
Is it possible to retrieve the data in BibTex form?At the bottom of the list of results from a query are options for returning the data in a variety of formats, including BibTex. You must click on the checkbox next to each abstract you want to save, then choose where you would like the results (screen, printer, or file), then choose to return BibTex format.
Please note that our BibTeX entries make use of the AASTeX macro definitions for the most popular astronomical journals. If you're not using the AASTeX package then you will need to include these macros in your LaTeX source.
Note also, that it is possible that the listing of authors may be out of order due to errors in the data which were supplied to us. If you find an error, we would appreciate hearing about it so that we can correct it.
We have also written a short document describing how to easily incorporate the BibTeX entries generated by ADS in your papers, and how to customize the available style files in order to create hyperlinks from each reference back to ADS.
Why are there errors in the bibtex returned by the ADS?The bibtex generated by ADS works best for the journal data which is represented in a fairly standard way in the ADS. For conference data, however, the bibtex which we can generate automatically usually contains errors because the data contained in the ADS are not formatted uniformly. This makes it difficult for us to easily extract the editor and publisher information to format the entries correctly in Bibtex. Users should be aware that using the bibtex entries from the ADS most often requires some editing.
What other ways are available for accessing the ADS databases?We support a number of different interfaces to those individuals and institutions who have a need to access the ADS databases in an automated way or without using a web browser. Please read the document discussing all the available options for querying and linking to the ADS.
Can ADS do everything Simbad does?ADS only has available a small subset of the information that you can gain by using Simbad when searching information about an astronomical object. In particular, the only use of the Simbad database in the Abstract Service is in the object name resolution, and it is done in a transparent way as far as the user is concerned. No other object measurements from Simbad are available.
Are there any mirror sites of the ADS?The ADS databases are mirrored and available from an increasingly large number of sites. Please see our mirror page for current information.
How should I acknowledge ADS?If you wish to acknowledge us in a publication, kindly use a phrase such as the following: ``This research has made use of NASA's Astrophysics Data System Bibliographic Services'' Thanks!
If you are using the ADS as a tool for bibliometric studies, please make sure you have an in-depth understanding of the system, its features and limitations, by reading and citing as appropriate the relevant published literature.
adshelp at cfa.harvard.edu