search

sdtaylor / phenology_dataset_study

1 stars 1 forks source link

check on Crimmins et al. values #18

Closed sdtaylor closed 6 years ago

sdtaylor commented 6 years ago

At first glance, the AGDD values (which should match my gdd_fixed model values) are pretty off (much higher in their paper).
Their sample size from NPN dataset seems to be a lot lower. for example 461 observations for acer rubrum when I have > 1000

sdtaylor commented 6 years ago

I implemented that papers stricter NPN subsetting but was still getting numbers way off. Then I figured out they match pretty closely if I convert my numbers from celsius to fahrenheit. So their reportedd AGDD values from that model are likely in fahrenheit, but this isn't mentioned anywhere.