Add something about spatial error filtering
With version 2 of the website, we redesigned the checklists The checklists are created from the database algorithmically at the state, county, major basin, and HUC8 levels. Before this recent revision, checklists included all species, even when their occurrences in the area of interest were based on what we suspected might be erroneous records. That was done because we felt strongly about providing users all of the available data, even when we might be dubious about some of the records. Our new design continues to provide all of the data, so our checklists can still be considered , and more recently Native Conservation Areas (NFCA's). Checklists are built on all of our occurence records with the exception of those that are non-specimen, captive and with spatial error larger than xxx. We consider the checklists the most complete list based on total specimen evidence, but we now . We indicate (via rose-colored highlight) which species we believe to be possibly likely erroneous, thus allowing users to easily exclude them from their workflows as appropriate. The new checklists also include statistics (N occurrence records that we examined and that we deem to be likely erroneous) to better communicate communicating the full justification (or relative lack thereof) for inclusion of each species on the list (N occurrence records that we examined and that we deemed to be likely erroneous). Users can use these data to make their own decisions about what species should be included in the checklist. Checklists for hydrological units now include labels indicating which species are non-native. These determinations are for their application of the data.
Checklists also include native status for each species, which are algorithmically derived “on the fly” based on whether or not an occurrence falls within the nativity layers that we created for each species. Future updates to our nativity layers will thus automatically be accounted for in these checklists.
We will continue working on checklists and will soon add a new type of checklist based strictly on our nativity layers. Since our existing checklists are based strictly on occurrences, they may lack species for which we lack, or have only spotty data. For example, we are well aware of our relatively sparse data for American Eel (Anguilla rostrata), which we are confident is due to an artifact of sampling biases or rareness. But we also know, based on life history and other data, that this species occurs in many places from which we lack specimens. Producing checklists based on our nativity layers should skirt such problems and provide consistently more taxonomically complete checklists. Future versions will include checklists based strictly on our nativity layers.
Partial view of current FoTX website’s species checklist for the Guadalupe River basin. Rose-colored rows are species recorded but considered suspect (may be removed following specimen examination). Note “Non-native” labels for those species. To the right is a summary of the basis for inclusion of each species.