search

APSIMInitiative / ApsimX

ApsimX is the next generation of APSIM
http://www.apsim.info
Other
136 stars 164 forks source link

Can CO2 defaults be brought up to date? #6617

Open zur003 opened 3 years ago

zur003 commented 3 years ago

The Weather model currently defaults to a value of 350 ppm if no value is explicitly provided. This is the level that was in the atmosphere in 1989, but it's now somewhere above 410, and continuing to rise rapidly. Would it be feasible to use fit a rough model of CO2 vs. date using historical data, and maybe project values out to 2050 or so to use for defaults? What implications would this have on existing simulations?

hut104 commented 3 years ago

I've been thinking about this recently as well.

It will change results, and will need a bit of work. Note that some of our datasets within the test set go back to the 1930's, so the impact may be even larger than expected. I don't think we need to extrapolate into the future, but a simple system to pull historical data into a met timeseries, which is perhaps automated (ie grabs CO2 data from a feed somewhere) might be what we need.

From: Eric Zurcher @.> Sent: Tuesday, 6 July 2021 4:47 PM To: APSIMInitiative/ApsimX @.> Cc: Subscribed @.***> Subject: [APSIMInitiative/ApsimX] Can CO2 defaults be brought up to date? (#6617)

The Weather model currently defaults to a value of 350 ppm if no value is explicitly provided. This is the level that was in the atmosphere in 1989, but it's now somewhere above 410, and continuing to rise rapidly. Would it be feasible to use fit a rough model of CO2 vs. date using historical data, and maybe project values out to 2050 or so to use for defaults? What implications would this have on existing simulations?

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://github.com/APSIMInitiative/ApsimX/issues/6617, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AC2UVWTUJUGZG474KMNFGQDTWKRG3ANCNFSM4735K3PQ.

BrianCollinss commented 3 years ago

I once complemented the CapeGrim monthly data (1976-) with that from Muana (1958-), fitted a polyline and used the polyline to adjust CO2 in APSIM simulations. One can assume these numbers belong to the 15th day of each month and use the fit to estimate daily values. I have a Manager script in Classic to do all this. Happy to share.

image image (X-Axis: month ID from January 1958).

AnnaBlanke commented 1 year ago

Hi! This question has been asked a while ago and I would like to know what´s the current status. If I do not provide CO2 in the met file, is it then still set to a default value of 350ppm? Or is it updated for recent past/ future periods? I am currently comparing simulation runs with and without CO2 values in the met files and would like to better understand what APSIMX does if CO2 is not explicitly set in the met files. Thanks a lot!

hol353 commented 1 year ago

Updating CO2 is on our (long) list of things to fix. If you don't specify CO2 then 350ppm is assumed, which is the value that is used when our models are calibrated and validated (except those models that have observed FACE (CO2) data in their validation dataset).

AnnaBlanke commented 1 year ago

OK, thanks a lot for the update!