Secondary ice production (SIP) is a crucial phenomenon for explaining the formation of ice crystal clouds, especially when addressing the discrepancies between observed ice crystal number concentrations and ice nucleating particles (INPs). In this study, we investigate parameterizations of three SIP processes (Hallett-Mossop, fragmentation of freezing drops, and fragmentation due to ice–ice collision) by simulating a deep convective cloud observed during the HAIC/HIWC campaign with the 3D bin microphysics scheme DESCAM (DEtailed SCAvening and Microphysics model). The simulated mean cloud properties, including particle size distributions and ice crystal number concentration are compared with in situ probe observations obtained during the campaign. Simulation excluding SIP shows a large underestimation of small ice crystals (< 1 mm diameter) for temperatures warmer than ‐30∘C. In our results, incorporating Hallett-Mossop and fragmentation due to ice–ice collision processes leads to ice crystal number concentrations close to observed values, thereby reducing discrepancies by two orders of magnitude. Our simulations also indicates that fragmentation of freezing drops affect minimally the properties of the cloud at its mature stage. Furthermore, we investigate the impact of fragments sizes resulting from SIP processes and show that the size of fragments generated from fragmentation due to ice–ice collision significantly influences the shape of ice particle size distribution. Employing various parameterizations of the ice crystal sticking efficiency reveals a notable impact on cloud properties. This study shows that SIP mechanisms are important and have to be considered for cold and mixed-phase clouds. However their parameterization lack reliability, highlighting the need for better quantifying these mechanisms. The companion paper, investigates the effects of SIP processes on the formation and the evolution of the deep convective system.

中文翻译：

---

使用 3D bin 微物理模型研究深对流云中的二次冰产生：第一部分 - 微物理过程表示的敏感性研究


二次冰产生 （SIP） 是解释冰晶云形成的关键现象，尤其是在解决观察到的冰晶数浓度和冰成核粒子 （INP） 之间的差异时。在这项研究中，我们通过使用 3D bin 微物理方案 DESCAM（去尾 SCAvening 和微物理模型）模拟在 HAIC/HIWC 活动期间观察到的深对流云，研究了三个 SIP 过程（Hallett-Mossop、冻结液滴的碎裂和冰-冰碰撞引起的碎裂）的参数化。将模拟的平均云特性（包括粒径分布和冰晶数浓度）与活动期间获得的原位探针观测结果进行了比较。不包括 SIP 的模拟显示，当温度高于 -30∘C 时，小冰晶（<，直径 1 毫米）被大大低估了。在我们的结果中，结合 Hallett-Mossop 和冰-冰碰撞过程引起的碎裂导致冰晶数浓度接近观测值，从而将差异减少两个数量级。我们的模拟还表明，冰滴的碎片化对云在成熟阶段的特性影响最小。此外，我们研究了 SIP 过程产生的碎片大小的影响，并表明冰-冰碰撞导致的碎片产生的碎片大小显着影响冰粒径分布的形状。采用冰晶粘附效率的各种参数化揭示了对云特性的显着影响。这项研究表明，SIP 机制很重要，必须考虑用于冷相云和混合相云。 然而，它们的参数化缺乏可靠性，突出了更好地量化这些机制的必要性。姊妹论文研究了 SIP 过程对深对流系统的形成和演变的影响。