Patterns and correlates of giant sequoia foliage dieback during California’s 2012–2016 hotter drought

Hotter droughts – droughts in which unusually high temperatures exacerbate the effects of low precipitation – are expected to increase in frequency and severity in coming decades, challenging scientists and managers to identify which parts of forested landscapes may be most vulnerable. In 2014, in the middle of California’s historically unprecedented 2012–2016 hotter drought, we noticed apparently drought-induced foliage dieback in giant sequoias (Sequoiadendron giganteum Lindl. [Buchholz]) in Sequoia and Kings Canyon national parks, California. Characteristics of the dieback were consistent with a controlled process of drought-induced senescence: younger (distal) shoots remained green while older (proximal) shoots were preferentially shed. As part of an ongoing interdisciplinary effort to understand and map sequoia vulnerability to hotter droughts, we reviewed historical records for evidence of previous foliage dieback events, surveyed dieback along trail corridors in eight sequoia groves, and analyzed tree-ring data from a high- and a low-foliage-dieback area. In sharp contrast to the greatly elevated mortality of other coniferous species found at low and middle elevations, we estimate that <1% of sequoias died during the drought. Foliage dieback was notably elevated in 2014 – the most severe single drought year in our 122-year record – but much lower in subsequent years. We found no historical records of similar foliage dieback during previous droughts. Dieback in 2014 was highly variable both within and among groves, ranging from virtually no dieback in some areas to nearly 50% in others. Dieback was highest (1) at low elevations, probably due to higher temperatures, reduced snowpack, and earlier snowmelt; (2) in areas of low adult sequoia densities, which likely reflect intrinsically more stressful sites; and (3) on steep slopes, probably reflecting reduced water availability. Average sequoia ring widths were narrower at the high-dieback than the low-dieback tree-ring site, but for reasons that remain unclear the sites did not differ in their proportional ring-width responses to past droughts. Collectively, our results suggest that giant sequoia vulnerability to hotter droughts may be spatially quite variable, and that at least some of that variability can be explained by metrics related to site water balance. Future research will focus on integrating our results with physiological and remote-sensing data, including tracking sequoias as they recover from the drought.