search

forc-db / Global_Productivity

Creative Commons Attribution 4.0 International
2 stars 0 forks source link

Hypothesis table #34

Closed beckybanbury closed 4 years ago

beckybanbury commented 5 years ago

@teixeirak a rough outline of a potential hypothesis table is saved here: is this the kind of thing you had in mind? Some of the hypotheses I took from our original word doc plans for the paper; these might have changed and it would be good to discuss if so!

teixeirak commented 5 years ago

I just pushed a new version of this table. @beckybanbury, please see what you think.

beckybanbury commented 5 years ago

@teixeirak I like this! I think it makes sense to put seasonality all together, as that brings in growing season but also allows us to include the fact that temperature seasonality is such a good predictor. And I like the way it is structured to include the analysis of the effect of climate variables within the growing season.

It'll be good to talk about this in person whenever you're in the office next!

teixeirak commented 5 years ago

Good! Why don't you work on filling out more of this table (after finalizing the growing season analysis)? Also feel free to adjust as you see needed.

The convention I was using was that "yes" or "no" indicated an unequivocal result (significant in all relevant analyses), whereas parentheses indicate a trend that is not (always) significant.

beckybanbury commented 5 years ago

Do you mean hypothesis 2.1. to be C flux increases exponentially with MAT across the latitudinal gradient? I would have thought it would make more sense to either predict a linear relationship between productivity and temperature, or a saturating relationship?

beckybanbury commented 5 years ago

For hypothesis 2.3. "There is a postive interaction between temperature and precipitation" - have you phrased it this way to distinguish between the directions of the interaction between ANPP and BNPP? Both BNPP fluxes showed a negative interaction - do we want to include this as an alternative hypothesis. Alternatively we could state this hypothesis as "there is a interaction between temperature and precipitation".

teixeirak commented 5 years ago

Do you mean hypothesis 2.1. to be C flux increases exponentially with MAT across the latitudinal gradient? I would have thought it would make more sense to either predict a linear relationship between productivity and temperature, or a saturating relationship?

Let's change that to linear. (There could be some reasons to test exponential, but easier not to get into that.)

beckybanbury commented 5 years ago

Do you think it's worth including a hypothesis related to vapour pressure deficit under hypothesis 3: how do carbon fluxes relate to other climate variables? It isn't a particularly significant variable in terms of proportion of variability it explains, but it shows a clear trend, and could be interesting to mention.

teixeirak commented 5 years ago

Yes, sounds good (at least if you keep it in the figure, which I think we should).