Nature, Science, and Macro Imagery

Lightroom keywords for taxa

I hate keywording.  I really do.  I like post-processing in general but sitting down and coming up with all the possible ways someone might be searching for an image is tedious.  Especially as a nature photographer, since there are so many different things about an animal that might be potentially interesting to someone.

Take the name of the animal, like this Andrena bee. Someone may need a photo to actually represent the genus Andrena. Or they might just be looking for any mining bee — in that case, are they looking for the term “mining bee”, or for the family Andrenidae? Or even the sub-family Andreninae? What if they’re looking for any bee? Any hymenopteran? Any insect?

I hate having to put all those in, and honestly I often half-ass it and settle for “bee, andrena, insect”. That’s probably fine, especially if it’s a photo I’m just putting up on Flickr, but I don’t like leaving those other possibilities off the table just because I’m bored of typing in family and order names, and want to get to the next photo.

Lightroom has a way to automate this, through the creative use of nested keywords and synonyms.  A keyword can contain other keywords: Let’s say you have a tree that looks like this:

food
        meat
        plant
                fruit
                vegetable
                grain
        dairy

The benefit of building this structure is that when you export an image with a keyword nested inside another, Lightroom walks up the keyword tree and adds every root keyword it finds.  So tagging a photo of an apple as “fruit” would also automatically tag it as “plant” and “food”.  The tree is one-way: tagging it as “plant” does not add “fruit”, “vegetable”, or “grain”.

Synonyms are related to trees.  Instead of an up-and-down structure, it’s side-by-side.  So you could add the synonyms “veggie” and “greens” to “vegetable”, and they would all be added to any image tagged with “vegetable”.

 Essentially what I’ve started doing is to build a taxonomic tree of nested keywords, all starting under “animal”.  Inside “animal” is the keyword “arthropod”, which contains “hexapod”, and then “insect”, and “Hymenoptera”.  I’ve started building a structure below this of all the kinds of bees I might conceivably find — I’ve only got it flesh out to about the tribe level so far, and in some places not much below the family, but here’s what the hymenopteran part of the tree looks like when I export it as a text file:

({brackets} show synonyms of the keyword they’re inside, so “hymentopteran” is a synonym of “Hymenoptera”.)

Hymenoptera      
    hymenopteran}                      
    Aculeata                              
        bee                  
            {anthophila}
            {apidae}
            {apiformes}
            {apoidea}
            Andrenidae
                {andrenid bee}
                {solitary bee}
                {burrowing bee}
                {ground-nesting bee}
                {miner bee}
                {mining bee}
                Andreninae
                    Andrena
            Apidae
                Apinae
                    Anthrophorini
                        {anthrophorine bee}
                    Apini
                        {honey bee}
                        Apis
                            Apis mellifera
                    Bombini
                                       {bumble bee}
                    Centridini
                    Emphorini
                    Eucerini
                        {long-horned bee}
                    Euglossini
                        {orchid bee}
                    Exomalopsini
                    Melectini
                Nomadinae
                    {cuckoo bee}
                Xylocopinae
                    {carpenter bee}
                    Ceratina
                        {small carpenter bee}
                        Calloceratina
                            Calloceratina cobaltina
                        Ceratinula
                            Ceratinula arizonensis
                            Ceratinula cockerelli
                        Zadontomerus
                            Zadontomerus calcarata
                            Zadontomerus dupla
                            Zadontomerus strenua
                    Xylocopa
                      
&nbs
p;Xylocopa virginica
                            {Eastern carpenter bee}
            Halictidae
                {halictid}
                {halictid bee}
                {sweat bee}
                Halictinae
                    Augochlorini
                        Augochlora
                        Augochlorella
                        Augochloropsis
                    Halictini
                        Agapostemon
                            {green metallic bee}
                            {green sweat bee}
                            {green}
                            {metallic}
                            Agapostemon virescens
                        Halictus
                        Lasioglossum
                        Mexalictus
                        Sphecodes

For myself, I’ve chosen to use Latin names as the base keywords, with common names and outdated scientific names as synonyms.  Where a common name can apply to more than one species, I put it as high up on the tree as I can.  In some cases, like for Agapostemon (a bright green bee with a reflective thorax), I added certain adjectives that will be relevant to any photo I ever take of one as synonyms.

So, I tag the image above as “Andrena”.  When I export it, Lightroom automatically adds all the root keywords and synonyms it finds.  It ends up tagged as “Aculeata; Andrena; Andrenidae; Andreninae; Hexapoda; Hymenoptera; andrenid bee; animal; anthophila; apidae; apiformes; apoidea; arthropod; bee; burrowing bee; groundnesting bee; hexapod; hymenopteran; insect; miner bee; solitary bee;” — all from one single tag.

I’m also fleshing out structures for other groups of animals that I frequently shoot, and will do the same for behaviors later.  All my spiders are auto-tagged “predator” now, for example.  It’s an up-front investment in time, but great for days like today, where it’s raining and I’m waiting on FedEx to bring me my replacement flash.  And I’ll be perfectly honest: I love automating repetitive tasks, I love taxonomy and phylogeny, and I love directional trees (graph theory is my favorite kind of math).  It’s a total confluence of geekery, that, for once, is actually somewhat productive.