Plant transpiration is a major conduit for the transfer of water from the land to the atmosphere (Jasechko et al., 2013). Recent studies have advanced our understanding of how complex and sensitive the leaf-to-atmosphere link is to localized feedbacks, such as slope exposure and associated conditions (e.g., Amitrano et al., 2019;Harrison et al., 2020;Link et al., 2014). Microclimates have been highlighted in particular as a distinct control on long-term ecosystem vulnerability, separate from evolving macroclimate (De Frenne et al., 2013. The urgency of understanding the vulnerability of forest ecosystems to changing climate conditions in arid or Mediterranean climates has been underscored by mass tree die-offs in western North America (Asner et al., 2016;Fettig et al., 2019;D. J. N. Young et al., 2017), and worsening wildfire regimes around the world. Yet, the influence of natural microclimatic variations, such as those associated with topographic position, on plant-water relations and physiology is often omitted in models used to forecast ecosystem vulnerability, due to a lack of data (Mencuccini et al., 2019;Pappas et al., 2016).Our aim is to understand how microclimates caused by topographic complexity may feed back into spatial variations in tree physiology and ultimately variations in tree vulnerability to fire or mortality under stressful conditions. To do this, we embarked on a study at the University of California's Angelo Coast Range Reserve in Northern California, a site which is now NSF's Eel River Critical Zone Observatory. The focus is