"I suspect that we want to generate BIOCLIM variables for several different
time-steps. From what I've just read, it looks like most recent papers use
different periods based on the Worldclim (Bioclim download site), rather than a
specified year into the future. So, for example, species distributions are often
modeled for the reference period (1961-1990), and then future periods (2020s,
2050s, and 2080s). So, 2050s is given by the mean of 2040–2069, etc"
We'll use 1961-2000 as historical and then separate grids for 2020s, 2050s, 2080s.
Re @hpoulos 's email 6/6/2017
"I suspect that we want to generate BIOCLIM variables for several different time-steps. From what I've just read, it looks like most recent papers use different periods based on the Worldclim (Bioclim download site), rather than a specified year into the future. So, for example, species distributions are often modeled for the reference period (1961-1990), and then future periods (2020s, 2050s, and 2080s). So, 2050s is given by the mean of 2040–2069, etc"
We'll use 1961-2000 as historical and then separate grids for 2020s, 2050s, 2080s.