C14 experiments (Post GMAC)

After my presentation at the GMAC and CT meetings, there are a few correlation ideas that I would like to try out. They are:

postGMAC-01

17 FF regions. Within the US, nine regions broken up according to EIA classification. 500 km hybrid correlation. Land disequilibrium over 23 regions; 10 TRANSCOM regions, and N Am Boreal broken up into CT2013 regions. Regionally coherent correlation, error is 50% of net flux.

postGMAC-02

11 FF regions. Within the US, three regions: Eastern US, Central US and Western US. 600 km hybrid correlation. Land disequilibrium over 23 regions; 10 TRANSCOM regions, and N Am Boreal broken up into CT2013 regions. Regionally coherent correlation, error is 50% of net flux.

postGMAC-03

11 FF regions. Within the US, three regions: Eastern US, Central US and Western US. 600 km hybrid correlation. Land disequilibrium over 12 regions; 11 TRANSCOM regions and ocean. Regionally coherent correlation, error is 50% of net flux.

postGMAC-04

17 FF regions. Within the US, nine regions broken up according to EIA classification. 500 km hybrid correlation. Land disequilibrium over 12 regions; 11 TRANSCOM regions and ocean. Regionally coherent correlation, error is 50% of net flux.

For all four experiments, I am also running a version without 14CO2 data. Also, there is one version where the fossil fuel flux is not being estimated, also without 14CO2.

[raglan_road]

For the second set of experiments, we will try two variations.

postGMAC-01-loosebio

Scott thinks that the biosphere is still too tight, because of which the 14C data cannot "speak". Therefore I will try a variation where the NEE uncertainty is 0.5 * respiration instead of 0.25. Needless to say, there will be two versions, one with 14CO2 data and one without.

postGMAC-01-flatprior

The fossil fuel flux has some seasonality in the prior, but perhaps not enough. We want to know, if there was no seasonality at all, would the 14C data recover it entirely? To check, we will run two inversions -- one with and one without 14C data -- where the fossil fuel flux has no seasonality.

[raglan_road]

The postGMAC-01-loosebio settings look the most promising, but are tweaked a bit.

postGMAC-05

1. Increase land disequilibrium uncertainty to 160% at the grid scale
2. Increase ocean disequilibrium uncertainty to 50% at the grid scale
3. Use the 'loose biosphere' setting for NEE uncertainty
4. Impose annual total FF emission constraint over the US, EU25 and Canada

postGMAC-06

postGMAC-05 + prior NEE and fires from SiBCASA GFED4.

postGMAC-05-unevenerr

With postGMAC-05, but the land disequilibrium uncertainty only increased to 160% over boreal N America, and ocean disequilibrium uncertainty only increased to 50% over the Southern ocean.

postGMAC-07

postGMAC-01 with the 'loose biosphere' NEE uncertainty setting, and annual total FF emission constraints on the US, Canada and EU25.

postGMAC-08

Same as postGMAC-07, except we assume a Δ¹⁴C measurement error of 0.1 per mil. The idea is to impose a stricter constraint on the inversion to fit the radiocarbon data, and see what fluxes result.

postGMAC-09

Same as postGMAC-07, except using John Miller's new production fields.

postGMAC-10

Same as postGMAC-09, with addition of UCI BRW data (via datathief).

postGMAC-10-tighterror

Same as postGMAC-10, with a Δ¹⁴C measurement error of 0.1 per mil.

postGMAC-10-brw

Same as postGMAC-10, but this time the observations have been forced to mid-afternoon.

postGMAC-11

Same as postGMAC-09, but with real Barrow data from Xiaomei.

[raglan_road]

Warning

There seems to have been a mistake in parsing the UCI data, so times have been inadvertantly duplicated. This mistake exists in postGMAC-11, which is now complete. postGMAC-12 is still running, so we are going to terminate that and start afresh, once we find the error.