2.31. Carbon Isotopes

CLM includes a fully prognostic representation of the fluxes, storage, and isotopic discrimination of the carbon isotopes ¹³C and ¹⁴C. The implementation of the C isotopes capability takes advantage of the CLM hierarchical data structures, replicating the carbon state and flux variable structures at the column and PFT level to track total carbon and both C isotopes separately (see description of data structure hierarchy in Chapter 2). For the most part, fluxes and associated updates to carbon state variables for ¹³C are calculated directly from the corresponding total C fluxes. Separate calculations are required in a few special cases, such as where isotopic discrimination occurs, or where the necessary isotopic ratios are undefined. The general approach for ¹³C flux and state variable calculation is described here, followed by a description of all the places where special calculations are required.

2.31.1. General Form for Calculating ¹³C and ¹⁴C Flux

In general, the flux of ¹³C and corresponding to a given flux of total C (\({CF}_{13C}\) and \({CF}_{totC}\), respectively) is determined by \({CF}_{totC}\), the masses of ¹³C and total C in the upstream pools (\({CS}_{13C\_up}\) and \({CS}_{totC\_up}\), respectively, i.e. the pools from which the fluxes of ¹³C and total C originate), and a fractionation factor, \({f}_{frac}\):

(2.31.1)\[\begin{split}CF_{13C} =\left\{\begin{array}{l} {CF_{totC} \frac{CS_{13C\_ up} }{CS_{totC\_ up} } f_{frac} \qquad {\rm for\; }CS_{totC} \ne 0} \\ {0\qquad {\rm for\; }CS_{totC} =0} \end{array}\right\}\end{split}\]

If the \({f}_{frac}\) = 1.0 (no fractionation), then the fluxes \({CF}_{13C}\) and \({CF}_{totC}\) will be in simple proportion to the masses \({CS}_{13C\_up}\) and \({CS}_{totC\_up}\). Values of \({f}_{frac} < 1.0\) indicate a discrimination against the heavier isotope (¹³C) in the flux-generating process, while \({f}_{frac}\) \(>\) 1.0 would indicate a preference for the heavier isotope. Currently, in all cases where Eq. is used to calculate a ¹³C flux, \({f}_{frac}\) is set to 1.0.

For ¹⁴C, no fractionation is used in either the initial photosynthetic step, nor in subsequent fluxes from upstream to downstream pools; as discussed below, this is because observations of ¹⁴ C are typically described in units that implicitly correct out the fractionation of ¹⁴C by referencing them to ¹³C ratios.

2.31.2. Isotope Symbols, Units, and Reference Standards

Carbon has two primary stable isotopes, ¹²C and ¹³C. ¹²C is the most abundant, comprising about 99% of all carbon. The isotope ratio of a compound, \({R}_{A}\), is the mass ratio of the rare isotope to the abundant isotope

(2.31.2)\[R_{A} =\frac{{}^{13} C_{A} }{{}^{12} C_{A} } .\]

Carbon isotope ratios are often expressed using delta notation, \(\delta\). The \(\delta^{13}\)C value of a compound A, \(\delta^{13}\)C_A, is the difference between the isotope ratio of the compound, \({R}_{A}\), and that of the Pee Dee Belemnite standard, \({R}_{PDB}\), in parts per thousand

(2.31.3)\[\delta ^{13} C_{A} =\left(\frac{R_{A} }{R_{PDB} } -1\right)\times 1000\]

where \({R}_{PDB}\) = 0.0112372, and units of \(\delta\) are per mil (‰).

Isotopic fractionation can be expressed in several ways. One expression of the fractionation factor is with alpha (\(\alpha\)) notation. For example, the equilibrium fractionation between two reservoirs A and B can be written as:

(2.31.4)\[\alpha _{A-B} =\frac{R_{A} }{R_{B} } =\frac{\delta _{A} +1000}{\delta _{B} +1000} .\]

This can also be expressed using epsilon notation (\(\epsilon\)), where

(2.31.5)\[\alpha _{A-B} =\frac{\varepsilon _{A-B} }{1000} +1\]

In other words, if \({\epsilon }_{A-B} = 4.4\) ‰ , then \({\alpha}_{A-B} =1.0044\).

In addition to the stable isotopes ¹²C and ¹³C, the unstable isotope ¹⁴C is included in CLM. ¹⁴C can also be described using the delta notation:

(2.31.6)\[\delta ^{14} C=\left(\frac{A_{s} }{A_{abs} } -1\right)\times 1000\]

However, observations of ¹⁴C are typically fractionation-corrected using the following notation:

(2.31.7)\[\Delta {}^{14} C=1000\times \left(\left(1+\frac{\delta {}^{14} C}{1000} \right)\frac{0.975^{2} }{\left(1+\frac{\delta {}^{13} C}{1000} \right)^{2} } -1\right)\]

where \(\delta^{14}\)C is the measured isotopic fraction and \(\mathrm{\Delta}^{14}\)C corrects for mass-dependent isotopic fractionation processes (assumed to be 0.975 for fractionation of ¹³C by photosynthesis). CLM assumes a background preindustrial atmospheric ¹⁴C /C ratio of 10^-12, which is used for A:sub::abs. For the reference standard A\({}_{abs}\), which is a plant tissue and has a \(\delta^{13}\)C value is \(\mathrm{-}\)25 ‰ due to photosynthetic discrimination, \(\delta\)¹⁴C = \(\mathrm{\Delta}\)¹⁴C. For CLM, in order to use the ¹⁴C model independently of the ¹³C model, for the ¹⁴C calculations, this fractionation is set to zero, such that the 0.975 term becomes 1, the \(\delta^{13}\)C term (for the calculation of \(\delta^{14}\)C only) becomes 0, and thus \(\delta^{14}\)C = \(\mathrm{\Delta}\)¹⁴C.

2.31.3. Carbon Isotope Discrimination During Photosynthesis

Photosynthesis is modeled in CLM as a two-step process: diffusion of CO₂ into the stomatal cavity, followed by enzymatic fixation (Chapter 2.9). Each step is associated with a kinetic isotope effect. The kinetic isotope effect during diffusion of CO₂ through the stomatal opening is 4.4‰. The kinetic isotope effect during fixation of CO₂ with Rubisco is \(\sim\)30‰; however, since about 5-10% of carbon in C3 plants reacts with phosphoenolpyruvate carboxylase (PEPC) (Melzer and O’Leary, 1987), the net kinetic isotope effect during fixation is \(\sim\)27‰ for C3 plants. In C4 plant photosynthesis, only the diffusion effect is important. The fractionation factor equations for C3 and C4 plants are given below:

For C4 plants,

(2.31.8)\[\alpha _{psn} =1+\frac{4.4}{1000}\]

For C3 plants,

(2.31.9)\[\alpha _{psn} =1+\frac{4.4+22.6\frac{c_{i}^{*} }{pCO_{2} } }{1000}\]

where \({\alpha }_{psn}\) is the fractionation factor, and \(c^*_i\) and pCO₂ are the revised intracellular and atmospheric CO₂ partial pressure, respectively.

As can be seen from the above equation, kinetic isotope effect during fixation of CO₂ is dependent on the intracellular CO₂ concentration, which in turn depends on the net carbon assimilation. That is calculated during the photosynthesis calculation as follows:

(2.31.10)\[c_{i} =pCO_{2} -a_{n} p\frac{\left(1.4g_{s} \right)+\left(1.6g_{b} \right)}{g_{b} g_{s} }\]

where \(a_n\) is net carbon assimilation during photosynthesis, \(p\) is atmospheric pressure, \(g_b\) is leaf boundary layer conductance, and \(g_s\) is leaf stomatal conductance.

Isotopic fractionation code is compatible with multi-layered canopy parameterization; i.e., it is possible to calculate varying discrimination rates for each layer of a multi-layered canopy. However, as with the rest of the photosynthesis model, the number of canopy layers is currently set to one by default.

2.31.4. ¹⁴C radioactive decay and historical atmospheric ¹⁴C and ¹³C concentrations

In the preindustrial biosphere, radioactive decay of ¹⁴C in carbon pools allows dating of long-term age since photosynthetic uptake; while over the 20\({}^{th}\) century, radiocarbon in the atmosphere was first diluted by radiocarbon-free fossil fuels and then enriched by aboveground thermonuclear testing to approximately double its long-term mean concentration. CLM includes both of these processes to allow comparison of carbon that may vary on multiple timescales with observed values.

For radioactive decay, at each timestep all ¹⁴C pools are reduced at a rate of –log/\(\tau\), where \(\tau\) is the half-life (Libby half-life value of 5568 years). In order to rapidly equilibrate the long-lived pools during accelerated decomposition spinup, the radioactive decay of the accelerated pools is also accelerated by the same degree as the decomposition, such that the ¹⁴C value of these pools is in equilibrium when taken out of the spinup mode.

For variation of atmospheric ¹⁴C and ¹³C over the historical period, \(\mathrm{\Delta}\)¹⁴C and \(\mathrm{\Delta}\)¹³C values can be set to either fixed concentrations or time-varying concentrations read in from a file. A default file is provided that spans the historical period (Graven et al., 2017). For \(\mathrm{\Delta}\)¹⁴C, values are provided and read in for three latitude bands (30°N–90°N, 30°S–30°N, and 30°S–90°S).