search

christophe-rd / fuelinex

Fuelling Next Year's Tree Growth with Carbon and Nitrogen
0 stars 0 forks source link

Warm Fall Treatment -- Temperature decision #17

Open christophe-rd opened 3 months ago

christophe-rd commented 3 months ago

@lizzieinvancouver, @FrederikBaumgarten @wangxm-forest

I would like your opinion on deciding what temperature I should set the biochambers at for the Warm Fall Treatment.

I thought we could set them at the 30-year mean maximum September temperature, which is 24 C. The night temperature could be the mean minimum temperature, which is 8 C.

We didn't end up choosing the method of "choosing a temperature guideline according to past climate data", and I think 24 C might be a little high. Any thoughts?

Thanks!

FrederikBaumgarten commented 3 months ago

I think thats a reasonable amplitude. You can go also straight for 20deg and 10deg

lizzieinvancouver commented 3 months ago

I thought we could set them at the 30-year mean maximum September temperature, which is 24 C. The night temperature could be the mean minimum temperature, which is 8 C.

@christophe-rd Is that the daily maximum across the month? I think I might take the mean(max daily temp) and/or mean(max daily temp for first two weeks) and see what you get as 24 C does sound high.

I recall you set spring temperatures based on historical data (but tell me if I am wrong) so I do not recommend now switching to 20/10 C as it would seem a weird switch in design to me.

christophe-rd commented 3 months ago

Thanks @lizzieinvancouver @FrederikBaumgarten! 4 @lizzieinvancouver that is the daily maximum accross the month, for the last 30 years at Totem Field. To clarify, I took the maximum temperature values for each day and calculated the mean.

I intended to set the spring temperatures based on historical data, but after noticing that the chambers were actually too warm and caused the trees to flush, we stopped. E.g. the first week of CoolSpring was at 4.2C, the second at 5C, and the third at 6C. We stopped increasing the temperature weekly at 8.2C. So I don't think we will be able to argue that the design was based on historical data since more than half of the treatment length was not based on that.

I had trouble making this clear, so please let me know if I can clarify!

lizzieinvancouver commented 2 months ago

I intended to set the spring temperatures based on historical data, but after noticing that the chambers were actually too warm and caused the trees to flush, we stopped. E.g. the first week of CoolSpring was at 4.2C, the second at 5C, and the third at 6C. We stopped increasing the temperature weekly at 8.2C. So I don't think we will be able to argue that the design was based on historical data since more than half of the treatment length was not based on that.

@christophe-rd So you started with historically-based values, then stopped at 8.2C. Right? This makes me think we want to ground whatever you do this fall in historical values, but 24 still seems high. Can you confirm which math you did:

  1. for each year calculate: max(string of daily max temperatures for the month)
  2. for each year calculate: mean(string of daily max temperatures for the month)

? Thanks.

christophe-rd commented 2 months ago

@lizzieinvancouver, sorry for the confusion! The math I did is 1. for each year calculate: max(string of daily max temperatures for the month)

If we go for treatment temperature based on historical values, then I should select weekly values. What I did for spring, is I chose weekly temperatures from the previous/following month (this choice because it was first supposed to be a Warm Spring treatment which ended up with a Cool Spring treatment). E.g. The start of the treatment was on March 1st, then the temperature for that week was the temperature of the first week of April.

That being said, from the discussion we had yesterday @lizzieinvancouver, I just realized that you were right. What I did previously is I extracted the maximum value of a period instead of averaging the daily maximum values of that period. So for the first week of September, the value would be 20C instead of 25C! It makes more sense to me even though it might be slightly inconsistent with the Warm Spring treatments...

lizzieinvancouver commented 2 months ago

So for the first week of September, the value would be 20C instead of 25C! It makes more sense to me even though it might be slightly inconsistent with the Warm Spring treatments...

@christophe-rd So, we could start with 20 C based on this run of historical data? That sounds good to me; it does not have to be identical to what we did in spring.

christophe-rd commented 2 months ago

@lizzieinvancouver thanks! I forgot to mention, this week would be 22C since we want to fit historic values from the previous month (22C is for the first week of August).

I discussed this with @FrederikBaumgarten and as we discussed, here's a figure of what the biochamber schedule will look like.

lizzieinvancouver commented 2 months ago

@christophe-rd 22 C sounds good. And nice ramping!