This site requires Javascript. Please enable it.

Taxonomic Name Resolution Service v4.0

How to Optimize Your Search

While we have designed the TNRS to handle most botanical data formats found in a majority of datasets, there are a few issues that should be considered in order to increase the probability that the TNRS correctly matches your name.

TNRS Settings

The TNRS has seven settings that affect how your names are resolved. These settings reflect the fact that there is no one best way to resolve names. The best settings for you depend on the requirements of your study and the amount of error present in your names.

  • Processing mode

    Setting Perform name resolution Parse names only
    Consequence Parses names (breaking them down into their component parts). Also attempts to match the name to one or more taxonomic sources. Only breaks names into their component parts. For example, separating the authority from the taxon name does not attempt to correct or match names. None of the remaining settings apply if you choose this setting.
    Examples Any study where the user wants to standardize and correct taxonomic names. Most users will choose this setting.
    • User is migrating taxonomic data to a new database and does not want to change or correct names. Name and author are currently concatenated in the old database and need to be separated to migrate to the new database.
    • User has a list of species and subspecies and needs to separate the genus and specific epithets from the rank indicator and subspecific epithet.
  • Allow partial matching?

    Setting Yes No
    Consequence Allow matching to higher taxa than the one submitted. For example, if you submit a species, the TNRS will return the best-matching genus if no species is found. Will only match taxa at the rank submitted. For example, if you submit a species and no species can be found, a genus match will not be returned.
    Examples Ecological studies where 'morphospecies' are used (for example, Quercus sp.1). Validation of species lists, such as from a museum database. Comparison of different data sets where species must be matched.
  • Constrain match by higher taxa?

    Setting Yes No
    Consequence Find the best match only within the genus (or family) submitted, even if a better overall match is available in another genus (or family). Find the best overall match, even if it is in a different genus from the one submitted.
    Examples Lists where higher taxa are unlikely to be incorrect. Where it jumps to distant taxa are most likely errors and/or would have severe consequences. For example, resolving names of georeferenced specimens used to map species ranges. Lists where higher taxa may be incorrect. Allowing matches to species in other higher taxa may aid in the discovery of the correct name.
  • Which classification sources?

    Setting Single Many
    Consequence Reduces the number of matches found for each name submitted, as matches from different sources are often variants of the same name. For example, the name with or without the author. Also, it allows the user to only use a trusted taxonomic source. Using >1 source, you will be more likely to find matches for all of your names. However, results may be confusing due to multiple matches and variant spellings of the same name by different sources.
    Examples
    • Taxonomic studies including only species from a higher taxon covered adequately by a single source. A list of species in the family Asteraceae would best be resolved using the Global Compositae Checklist. An ecological study from a particular region covered adequately by one list.
    • A species list from an agency required to use a particular source. For example, a US government agency might be required to use USDA Plants to ensure that their taxonomy conforms to other government agency species lists.
    Ecological or phylogenetic studies from broad geographic regions where no one source is likely to have all names. For example, an ecological study including both Old and New World species should use multiple taxonomic sources.
  • Sort by source?

    This setting only applies if you are using >1 taxonomic source to resolve your names.

    Setting Yes No
    Consequence If name is found in the highest ranked source, it will be shown as the best match even if a better scoring match is found in another source. A match from a lower-ranked source will only be shown if no match is found in the higher-ranked source. The highest scoring match will always be shown as the best match, regardless of source.
    Examples User knows some sources to be generally preferable to other sources. For example, for an ecological study that contains species from many families, the Global Compositae Checklist (GCC) might be ranked first to ensure that any Asteraceae are always resolved first according to this source. Species in other families would not be found in GCC, but might be resolved by the remaining sources. User does not believe any one source to be better than the other and simply wants the best possible spelling match in all cases.
  • How sensitive a match accuracy should I use?

    Setting Low High
    Consequence Allows matching to name quite different from the one submitted. This ensures most names will be matched and also return more possible matches for each name. Will only match to names very similar or identical to the name submitted. Fewer names will be matched, but matches will be of higher certainty.
    Examples Very large or divers taxonomic lists or lists of low quality are likely to have many errors or old synonymous names. User should closely inspect results. For smaller or high quality lists, or studies where no match is preferable to an incorrect match. For example, validating large numbers of georeferenced specimen observations which will be used to model species ranges.
  • Higher classification

    This setting only affects the family classification of matched and accepted names.

    Setting APGIII NCBI
    Consequence Family classification conforms exactly to Chase, M. & J. Reveal. 2009. A phylogenetic classification of the land plants to accompany APGIII. Botanical Journal of the Linnean Society 161:122-127. This is also the family classification used by Tropicos. Family classification as used by NCBI Taxonomy. Similar to APGIII, but with updates based on more recent phylogenetic studies.
    Examples Some community phylogenetic programs, such as phylomatic require APGIII families. Using exactly APGIII families may also be desirable if comparing results with a previous study that followed the APGIII classification. A study requiring a more updated family classification would use this setting.


Additional Help

  1. Make sure that the first letter of each genus name is capitalized.

  2. Make sure that each specific epithet (and subspecific epithet, if applicable) is all lowercase.

    In the second example below, the incorrectly capitalized epithet "Rubra" prevents the name from matching to species.

    Optimize Your Search Example
  3. Try submitting names without author strings at the end of the name.

    In the example below, the difference in spelling between the author of the name submitted and the name matched lowers the overall match score, relative to submitting the name alone, without the author string.

    Author String Example

    Be careful, however. Including the author name can sometimes increase match accuracy, especially if homonyms exist (the same name used by different authors to refer to different species). In the example below, the name retured by default for the Campylopus, without the author, is the accepted genus Campylopus Brid., a moss.

    This is acceptable if the moss is the intended meaning. However, Campylopus was also later (and incorrectly) used by the author Spach to refer to a genus of trees in the family Clusiaceae. Even though the second name is an illegitamate posterior homonym, it clearly refers to the tree and not the moss; by including the author the intended meaning is preserved.

    Increase Match Accuracy

    Note: Including the family name with the name will also distinguish between homonyms in different families (see recommendation #8 below)

    You should consider this tradeoff when determining whether to omit or include the author string.

  4. The TNRS currently does not fuzzy match hybrid names. If a submitted name is mispelled, the TNRS will not find it.

    In the example below, the first name is spelled correctly; the second name is not.

    Misspelled

  5. Try to strip unwanted characters before submitting your names.

    While the TNRS automatically strips out odd characters that are not used in botanical nomenclature (such as punctuation, underscores, +, %, #, etc.), names that include many such characters produce in lower match scores, potentially preventing the TNRS from discovering the correct name.

    Note the image below showing the effect of submitting misspelled names with and without odd characters:

    Odd Characters

  6. Change the match score threshold.

    By default, names will be returned to you in a way that maximizes the number of returned matches (lowest setting). By increasing the match threshold, you can increase the exactness of the names being returned for your submitted name. However, this may lead to names being missed, especially in the case of badly misspelled names.

  7. Inspect your results!

    It is especially important you inspect results flagged with warnings or names with more than one possible match. Names with more than one match that have the hyperlinked words "(+x more)" after the name matched ("x" is the number of additional possible matches).

    Click on the hyperlink to see a list of all possible matches. Sometimes your intended name will be a lower-ranked match. This is especially important if you have set higher taxonomic constraint to ON.

Large jobs

Whereas the TNRS application can handle millions of names, this web site has not been optimized for such large jobs. If your input list is over 500,000 names long, please contact support for assistace.