Carbon dioxide emissions from active subaerial volcanoes represent 20–50% of the annual global volcanic CO₂ flux (Barry et al., 2014). Passive degassing of carbon from the flanks of volcanoes, and the associated accumulation of dissolved inorganic carbon (DIC) within nearby groundwater, also represents a potentially important, yet poorly constrained flux of carbon to the surface (Werner et al., 2019). Here we investigate sources and sinks of DIC in groundwaters in the Lassen Peak region of California. Specifically, we report and interpret the relative abundance and isotopic composition of helium (³He, ⁴He) and carbon (¹²C, ¹³C, ¹⁴C) in 37 groundwater samples, from 24 distinct wells, collected between 20 and 60 km from Lassen Peak. Measured groundwater samples have air-corrected ³He/⁴He values between 0.19 and 7.44 R_A (where R_A = air ³He/⁴He = 1.39 × 10⁻⁶), all in excess of the radiogenic production value (~0.05 R_A), indicating pervasive mantle-derived helium additions to the groundwater system in the Lassen Peak region. Stable carbon isotope ratios of DIC (δ¹³C) vary between −12.6 and − 27.7‰ (vs. VPDB). Measured groundwater DIC/³He values fall in the range of 2.2 × 10¹⁰ to 1.1 × 10¹². Using helium and carbon isotope data, we explore several conceptual models to estimate surface carbon contributions and to differentiate between DIC derived from soil CO₂ versus DIC derived from external (slab and mantle) carbon sources. Specifically, if we use ¹⁴C to identify soil-derived DIC (assuming decadal-to-centennial groundwater ages and a soil CO₂ ¹⁴C activity equal to that of the atmosphere), we calculate that a hypothetical external carbon source would have an apparent δ¹³C signature between −10.3 and − 59.3‰ (vs. VPDB) and an apparent C/³He between 7.0 × 10⁹ and 1.0 × 10¹². These apparent δ¹³C and C/³He values are substantially lighter than and greater than canonical MORB values, respectively. We suggest that >95% of any external (non-soil-derived) DIC in groundwater must thus be non-mantle in origin (i.e., slab derived or assimilated organic carbon). We further investigate possible sources of external DIC to groundwater using two idealized conceptual approaches: a pure (unfractionated) source mixing model (after Sano and Marty, 1995) and a scenario that invokes fractionation due to calcite precipitation. Because the former model requires carbon contributions from an organic source component with unrealistically low δ¹³C (~ − 60‰), we suggest that the second scenario is more plausible. Importantly, however, we caution that all conceptual models are dependent on assumptions about initial ¹⁴C activity. Thus, we cannot rule out the possibility that the true fraction of non-surface-derived DIC in these samples is lower or negligible, despite the pervasive mantle-derived He isotope signatures throughout the region. Following the ¹⁴C approach to deconvolving sources of DIC, we determine that the maximum passive carbon flux could be up to ~2.2 × 10⁶ kg/yr, which is lower than previous magmatic carbon flux estimates from the Lassen region (Rose and Davisson, 1996). We find that the passive dissolved carbon flux could represent a maximum of ~4–18% of the total Lassen geothermal CO₂ degassing flux (estimated to be ~3.5 × 10⁷ kg/yr Rose and Davisson, 1996; Gerlach et al., 2008), which is still more than an order of magnitude smaller than soil gas CO₂ flux estimates (7.3–11 × 10⁷ kg/yr) for nearby volcanoes (Sorey et al., 1998; Gerlach et al., 1999; Evans et al., 2002; Werner et al., 2014). We conclude that passive dissolved carbon fluxes should be combined with geothermal fluxes and soil gas fluxes to obtain a complete picture of volcanic carbon emissions globally. Our approach highlights the utility of measuring helium isotopes in concert with the full suite of noble gas abundances, tritium, δ¹³C and ¹⁴C, which when interpreted together can be used to better elucidate the various sources of DIC in groundwater.

活跃的地下火山的二氧化碳排放量占全球每年火山 CO ₂通量的20-50% （Barry 等，2014）。来自火山侧翼的碳的被动脱气，以及附近地下水中溶解无机碳 (DIC) 的相关积累，也代表了一种潜在的重要但对地表的碳通量限制很差（Werner 等人，2019 年）。在这里，我们调查了加利福尼亚拉森峰地区地下水中 DIC 的来源和汇。具体而言，我们报告并解释了氦 ( ³ He, ⁴ He) 和碳 ( ¹² C, ¹³ C, ¹⁴C) 来自 24 口不同井的 37 个地下水样品，收集于距离拉森峰 20 到 60 公里的地方。测量的地下水样品的空气校正³ He/ ⁴ He 值介于 0.19 和 7.44 R _{A 之间}（其中 R _A  = 空气³ He/ ⁴ He = 1.39 × 10 ^-6），所有这些都超过了放射生成值（~0.05 R _A )，表明拉森峰地区地下水系统中普遍存在地幔衍生的氦。DIC (δ ¹³ C) 的稳定碳同位素比率在 -12.6 和 - 27.7‰（相对于 VPDB）之间变化。实测地下水 DIC/ ³ He 值在 2.2 × 10 ¹⁰范围内到 1.1 × 10 ¹²。使用氦和碳同位素数据，我们探索了几种概念模型来估计表面碳贡献并区分源自土壤 CO _{2 的}DIC 与源自外部（板坯和地幔）碳源的 DIC。具体来说，如果我们使用¹⁴ C 来识别土壤衍生的 DIC（假设十年到百年的地下水年龄和土壤 CO ₂ ¹⁴ C 活动等于大气的活动），我们计算出假设的外部碳源将具有明显的δ ¹³ C 特征介于 -10.3 和 - 59.3‰（vs. VPDB）之间，表观 C/ ³ He 介于 7.0 × 10 ⁹和 1.0 × 10 ^{12 之间}。这些表观δ¹³ C 和 C/ ³ He 值分别显着小于和大于规范 MORB 值。我们建议，地下水中>95% 的任何外部（非土壤衍生的）DIC 必须是非地幔来源的（即，板坯衍生或同化的有机碳）。我们使用两种理想化的概念方法进一步研究了地下水的外部 DIC 的可能来源：纯（未分馏）源混合模型（在 Sano 和 Marty 之后，1995 年）和由于方解石沉淀而调用分馏的情景。因为前一个模型需要来自具有不切实际的低 δ ¹³的有机源成分的碳贡献C (~ − 60‰)，我们认为第二种情况更合理。然而，重要的是，我们警告所有概念模型都依赖于关于初始¹⁴ C 活动的假设。因此，我们不能排除这些样品中非表面衍生的 DIC 的真实分数较低或可以忽略不计的可能性，尽管该地区普遍存在地幔衍生的 He 同位素特征。按照¹⁴ C 方法对 DIC 源进行解卷积，我们确定最大被动碳通量可能高达 ~2.2 × 10 ⁶  kg/yr，低于先前来自拉森地区的岩浆碳通量估计值（Rose 和 Davisson， 1996）。我们发现被动溶解碳通量最多可占拉森地热 CO2 总量的约 4-18%₂脱气通量（估计为 ~3.5 × 10 ⁷  kg/yr Rose 和 Davisson，1996 年；Gerlach 等人，2008 年），仍然比土壤气体 CO ₂通量估计值（7.3-11 × 10 ⁷  kg/yr)（Sorey 等人，1998 年；Gerlach 等人，1999 年；Evans 等人，2002 年；Werner 等人，2014 年）。我们得出结论，被动溶解碳通量应与地热通量和土壤气体通量相结合，以获得全球火山碳排放的完整图景。我们的方法突出了测量氦同位素与全套惰性气体丰度、氚、δ ¹³ C 和¹⁴的效用C，当它们一起解释时，可以用来更好地阐明地下水中 DIC 的各种来源。