The abalist program: computing American Birding Association list totals |  |
The abalist program: computing American Birding Association list totals | |
The ABA has a large set of rules for which bird observations are countable. The observer is responsible for interpreting and applying these rules. The way records are coded in birdnotes files tells us these things about observations that are important to the abalist program:
Their taxonomic identity, that is, what species, genus, or other taxon was observed. The xnomo system allows observations to be pegged to all taxonomic ranks, including a rank deeper than species that we call form rank. Forms include taxonomic subspecies, color morphs, and other fine distinctions within species.
Whether the observer was certain of the identification.
If the identification is in question, the observer
codes the record with the q='?' attribute.
Whether the observed individual is part of a countable
population. For example, Whooping Cranes were regular
in New Mexico for a time in the 1970s-1990s, but they
were not an established breeding population, and hence
not countable under ABA rules. The observer flags such
records with the q='-'
attribute in the birdnotes system.
Although the ABA totals are generally referred to as “species totals,” they may include observations pegged to other taxonomic ranks:
Observations at the form level are counted as their containing species. For example, if either of both of Myrtle and Audubon's Warblers were observed, those count as their containing species, Yellow-rumped Warbler.
Observations at higher ranks than species can be counted only if there are no observations pegged to lower ranks inside them. For example, if a list includes both “goose sp.” and Snow Goose, that counts as only one species. But if the observer reported “goose sp.” and saw no other kind of goose, that counts as one species.
It is also of interest to the author which observation was the first for a given species or other taxon. So, after assembling a list of observed taxa, here are the rules for counting the “species” in that list. We step through the list in phylogenetic order and apply these transformations to the tree of observed taxa:
If a form was observed but its parent species was not, promote the observation to the species level.
If a form was observed but its parent species was also observed, retain whichever observation was earlier, and peg it to the species level.
For any taxon at species level and higher, remove all ancestor taxa. For example, if there is an observation pegged to a given species, remove any observation from its containing genus, subfamily, family, etc.
After the tree is transformed by these rules, the number of nodes remanining is the species count for ABA purposes.
Here are some examples. In this table, the left-hand column shows the raw observations, and the right-hand column shows the transformed observation set.
| Before | After |
|---|---|
|
|
|
|
|
|
|
|
|
|
|
|
In the first example, American Green-winged Teal is a form of the species Green-winged Teal. Since its parent species is missing, it is promoted to species.
The second example has Snow Goose and two of its forms. Because the species-level observation is first, it takes precedence over its forms.
In the third example, the white Snow Goose record is promoted to species level because it predates the other form.
The fourth example shows genus Aechmophorus and two of its species, Western and Clark's grebes. The genus is deleted because there are records below it.