Abstract. Land surface modelers need measurable proxies to constrain the quantity of carbon dioxide (CO₂) assimilated by continental plants through photosynthesis, known as Gross Primary Production (GPP). Carbonyl sulfide (COS), which is taken up by leaves through their stomates and then hydrolysed by photosynthetic enzymes, is a candidate GPP proxy. A former study with the ORCHIDEE land surface model used a fixed ratio of COS uptake to CO₂ uptake normalized to respective ambient concentrations for each vegetation type (Leaf Relative Uptake, LRU). COS leaf fluxes were then computed from GPP, and the resulting concentrations were transported with an atmospheric model which included all other known COS fluxes as inputs. Modelled COS concentrations could then be compared to COS measurements from the NOAA air sampling tower network. The LRU approach is known to have limited accuracy since the LRU ratio changes with variables such as Photosynthetically Active Radiation (PAR): while CO₂ uptake slows under low light, COS uptake is not light limited. However, the LRU approach has been popular for COS-GPP proxy studies because of its ease of application and apparent low contribution to uncertainty for regional scale applications. In this study we refined the COS-GPP relationship and implemented in ORCHIDEE a mechanistic model that describes COS uptake by continental vegetation. We compared the simulated COS fluxes against measured hourly COS fluxes at two sites, and studied the model behaviour and links with environmental drivers. We performed simulations at global scale, and estimated the global COS uptake by vegetation to be −756 Gg S yr⁻¹, in the middle range of former studies (−490 to −1335 Gg S yr⁻¹). Based on the mechanistic approach in ORCHIDEE, we derived new LRU values for the different vegetation types, ranging between 0.92 and 1.72, close to recently published averages for observed values of 1.21 for C4 and 1.68 for C3 plants. We transported the COS using the monthly vegetation COS fluxes derived from both the mechanistic and the LRU approaches, and evaluated the simulated COS concentrations at NOAA sites. Although the mechanistic approach was more appropriate when comparing to high-temporal-resolution COS flux measurements, both approaches gave similar results when transporting with monthly COS fluxes and evaluating COS concentrations at stations. In our study, uncertainties between these two approaches are of second importance as compared to the uncertainties in the COS global budget, which are currently a limiting factor to the potential of COS concentrations to constrain GPP simulated by land surface models on the global scale.

中文翻译：

---

羰基硫：在土地表面模型和叶片相对吸收方法中比较植被吸收的机械表示

摘要。陆地表面建模者需要可测量的代理来限制大陆植物通过光合作用吸收的二氧化碳（CO ₂）的量，称为总初级生产（GPP）。羰基硫（COS）是候选GPP代理，其通过叶片的气孔被叶片吸收，然后被光合酶水解。以前使用ORCHIDEE地表模型进行的研究使用了固定的COS吸收量与CO ₂比率每种植物类型的摄入量均归一化为各自的环境浓度（叶片相对摄入量，LRU）。然后从GPP中计算出COS叶通量，并用大气模型运输所得的浓度，其中包括所有其他已知COS通量作为输入。然后可以将建模的COS浓度与NOAA空气采样塔网络的COS测量值进行比较。已知LRU方法的精度有限，因为LRU比率随光合有效辐射（PAR）等变量而变化，而CO ₂在弱光条件下吸收会减慢，COS的吸收不受光的限制。但是，由于LRU方法易于应用且对区域规模应用的不确定性影响很小，因此在COS-GPP代理研究中很受欢迎。在这项研究中，我们完善了COS-GPP关系，并在ORCHIDEE中实现了一种描述大陆植被吸收COS的机制模型。我们将模拟的COS通量与两个地点的每小时小时COS通量进行了比较，并研究了模型行为以及与环境驱动因素的联系。我们在全球范围内植被进行模拟，并估计全球COS摄取是-756千兆克小号年^-1，在前人研究的中间范围（-490至-1335千兆克小号年^-1）。基于ORCHIDEE中的机械方法，我们得出了不同植被类型的新LRU值，范围在0.92至1.72之间，接近了最近发布的C4植物观察值1.21和C3植物观察值1.68的平均值。我们使用来自机械方法和LRU方法的每月植被COS通量来运输COS，并评估了NOAA站点模拟的COS浓度。尽管与高温分辨率的COS通量测量相比，机械方法更合适，但当使用每月COS通量运输并评估站点的COS浓度时，两种方法都得出了相似的结果。在我们的研究中，与COS全球预算中的不确定性相比，这两种方法之间的不确定性具有第二重要性，