search

EcoClimLab / ForestGEO-tree-rings

Repository for analysis of tree-ring data from 10 globally distributed forests (Anderson-Teixeira et al., in press, Global Change Biology)
2 stars 2 forks source link

confirm that missing DBH trends are non-sig relationships #57

Closed teixeirak closed 4 years ago

teixeirak commented 4 years ago

@ValentineHerr , could you please confirm that when species are missing from the DBH plots (e.g., LITU at SCBI and ABAL at Zofin for ∆r) this means the relationship is not significant?

image

ValentineHerr commented 4 years ago

yes. Technically it means that AICc weights of models in which the variable was included did not add up enough to pass the threshold. (Note that when I have more time to think of this, I may decide to change this rule). You can see the model weights for each species-variable combinations in these types of figures: https://github.com/EcoClimLab/ForestGEO-climate-sensitivity/blob/master/results/log_core_measurement_dbh/SCBI/GLS_Sum_of_AICweights_SCBI.png (basically if it is not a strong cyan color, the variable doesn't have strong evidence that it is important for the species).

teixeirak commented 4 years ago

Okay, thanks.