Parameterise free-living N fixation

[hrlai]

The equation for free-living nitrogen fixation rate is
$$F_r Q_{10}^{(T_s - T_r)/10}$$
where $F_r$ is the fixation rate at reference temperature [kg N m^-2 day^-1], $Q_{10}$ the Q10 temperature quotation [unitless], $T_s$ the soil temperature [K], and $T_r$ the reference temperature [K]. This is Equation set 10 from Lin et al. (2000).

Lin et al. took values from Granhall et al. (1981), which are assumed and not empirically parameterised. I will try to do the latter.

In VE we convert fixation rate to per soil volume but this is not needed here.

[hrlai]

Hi @jacobcook1995 I'm trying to revisit these constants, but stuck at deciding whether to stick with $Q_10$, which is a power function of temperature, or the unimodal function that takes into account of nitrogenase denaturation. The latter could entail lots of recoding in VE?

[jacobcook1995]

So, whether this would entail a lot of recoding in the model is depends what the change would be. If it's just replacing one thermal response function with another it would be less than a day of work for me. What would be tricky is if the change needed additional intermediate variables introduced, i.e. if the change meant that nitrogenase abundance had to be (explicitly) tracked. That would be a lot more work. (Broadly this is true across the model, changes to specific functions used are easy to do, increasing the number of variables tracked is hard).

Guess my general inclination is that it's a better use of your time to track down data for the better model now, rather than a model that I think we are both fairly unhappy with and then the data for the better model at a later date. Though if I would have to introduce several new variables to the model that would give me pause

[hrlai]

So we will try:

1. Find out which "trait group" nitrogenaese / N fixation belong to in Fig. 5 of Kontopoulos et al. (2024)
2. Fit the top models to datasets, e.g., https://doi.org/10.1038/nature07028 or https://doi.org/10.1101/2025.06.17.660177 or https://doi.org/10.1016/j.tim.2023.12.007