I’m currently working on a similar circumstance and would like to advance my approach as well as understand more about applying memory palaces by exposing what I’m working on.
Plant families contain large characteristic descriptors of all the genera it contains. Rarely will it describe environmental factors (soil, light, etc), unless the family is a very restricted distribution.
Genus will further restrict physical descriptors, with some generalities of environmental factors.
This repeats through species, cultivars and varieties.
The memory palace I’m using is my childhood home. It has an exposed north-facing lawn, a south-facing lawn with a forest canopy, the west has a slope, east is flat. Truthfully, it is rotated a bit northeast, but I rotate it to make this easier.
The northernmost aspect of the lawn I associate with full sun, marginal, disturbed (tough) conditions, as it moves towards the house, the soil is a bit more moist, semi-shaded, and finally fully sheltered.
The interior of the house I retain for tropicals or funky plants with special needs.
The backyard for part shade, and filtered light (from the forest canopy), and as it approaches the forest properly it is qualified as woodland conditions.
East and West to location plants that have a preference for afternoon or morning shade.
There are certain features like raised beds, or areas I use to denote other special conditions (clay soil, loam, etc).
I have 356 genera, with probably 450 species/var/cultivar specifications to make.
For me, it seems the most important thing is to generalize first and specify later. If I can say that junipers predominately love full sun, dry conditions, I can locate them in the northernmost placement of the lawn. If one specific juniper species needs something different, it is located elsewhere.
I think, I’m not 100%, this will create the least friction to learning patterns first and individuation later.
For visual ID, I have an Anki card deck of images containing plant parts, habits, etc. where the family/genus/species/etc is on the back. I decided to add the botanical descriptions of plant parts on the front side to help coach a verbal way of referencing what I’m staring at. I’m not sure if this is information overload, but it seems to challenge me away from simply remembering the image without processing the content.
I’m thinking of having the loci # of each plant below the image to cue its cultural specifications in the memory palace.
As the Anki card is random, I would then need to reference my map to find that location and retrieve the cultural conditions.
I have not implemented a concrete approach to associating families > genera > species > etc. yet. I tend to try to find mnemonics where that support some aspect of the plant’s behaviour. Right now I’m just hoping that a lot of osmosis and overlap from putting all this together
I would appreciate any input on this, as through a few environmental and institutional circumstances I’ve lived a long time believing I simply have poor memory and intellectual abilities.