Intermittent streams comprise much of the global river network, and are expected to become more prevalent with climate change. Characterizing the expansion and contraction of intermittency in stream networks, and understanding how sensitive these dynamics are to climatic variability, is critical for predicting the trajectory of hydrologic regimes in a changing climate. Here, we consider the spatial patterns of stream intermittency, focusing on wetted channel conditions at the end of the dry season, and identify land cover, physiographic, and climate variables that influence surface water presence and variability across years. We trained statistical models with wetted channel mapping data from 25 streams over 7 years to predict both the spatial and interannual variability of the wetted channel network. We then used the models to assess intermittent stream dynamics across the Russian River watershed in northern California, USA. We found that an average of 3.7% of the stream network was reliably dry, while 16.1% was reliably wet at the end of the dry season, with the remainder of the network exhibiting variability in wetted conditions in response to antecedent precipitation. Both climatic and landscape characteristics controlled the extent of the wetted network, particularly antecedent precipitation at seasonal and annual time scales, highlighting the role of hydrologic memory in this system. Given predictions of increased climate volatility, an improved understanding of the spatial patterns and stability of dry season conditions in intermittent streams can inform climate risk assessments and strategies for protecting biodiversity and the ecosystem services that intermittent streams support.

中文翻译：

---

间歇性河流干燥对气候变异的空间模式和敏感性

断断续续的溪流构成了全球河流网络的大部分，预计随着气候变化将变得更加普遍。表征河流网络中间歇性的扩张和收缩，并了解这些动态对气候变化的敏感程度，对于预测气候变化中水文状况的轨迹至关重要。在这里，我们考虑河流间歇性的空间模式，重点关注旱季结束时的湿润河道条件，并确定影响地表水存在和多年变化的土地覆盖、地貌和气候变量。我们使用来自 7 年多的 25 条河流的湿通道映射数据训练统计模型，以预测湿通道网络的空间和年际变化。然后，我们使用这些模型来评估美国加利福尼亚州北部俄罗斯河流域的间歇性河流动态。我们发现平均 3.7% 的河流网络是可靠干燥的，而在旱季结束时 16.1% 是可靠湿润的，网络的其余部分表现出湿润条件的可变性，以响应前期降水。气候和景观特征控制了湿润网络的范围，特别是季节和年度时间尺度的前期降水，突出了水文记忆在该系统中的作用。鉴于气候波动加剧的预测，