Forest mortality is accelerating due to climate change and the largest trees may be at the greatest risk, threatening critical ecological, economic, and social benefits. Here, we combine high-resolution airborne LiDAR and optical data to track tree-level mortality rates for ~2 million trees in California over 8 years, showing that tree height is the strongest predictor of mortality during extreme drought. Large trees die at twice the rate of small trees and environmental gradients of temperature, water, and competition control the intensity of the height-mortality relationship. These findings suggest that future persistent drought may cause widespread mortality of the largest trees on Earth.

Traditional methods of carbon monitoring in mountainous regions are challenged by complex terrain. Recently, solar‐induced fluorescence (SIF) has been found to be an indicator of gross primary production (GPP), and the increased availability of remotely sensed SIF provides an opportunity to estimate GPP across the Western United States. Although the empirical linkage between SIF and GPP is strong, the current mechanistic understanding of this linkage is incomplete and depends upon changes in leaf biochemical processes in which absorbed sunlight leads to photochemistry, heat (via nonphotochemical quenching [NPQ]), fluorescence, or tissue damage. An improved mechanistic understanding is necessary to leverage SIF observations to improve representation of ecosystem processes within land surface models. Here we included an improved fluorescence model within the Community Land Model, Version 4.5 (CLM 4.5), to simulate seasonal changes in SIF at a subalpine forest in Colorado. We found that when the model accounted for sustained NPQ, this provided a larger seasonal change in fluorescence yield leading to simulated SIF that more closely resembled the observed seasonal pattern (Global Ozone Monitoring Experiment‐2 [GOME‐2] satellite platform and a tower‐mounted spectrometer system). We found that an acclimation model based on mean air temperature was a useful predictor for sustained NPQ. Although light intensity was not an important factor for this analysis, it should be considered before applying the sustained NPQ and SIF to other cold climate evergreen biomes. More leaf‐level fluorescence measurements are necessary to better understand the seasonal relationship between sustained and reversible components of NPQ and to what extent that influences SIF.

Drought conditions affect ozone air quality, potentially altering multiple terms in the O 3 mass balance equation. Here, we present a multiyear observational analysis using data collected before, during, and after the record-breaking California drought (2011−2015) at the O 3-polluted locations of Fresno and Bakersfield near the Sierra Nevada foothills. We separately assess drought influences on O 3 chemical production (PO 3) from O 3 concentration. We show that isoprene concentrations, which are a source of O 3-forming organic reactivity, were relatively insensitive to early drought conditions but decreased by more than 50% during the most severe drought years (2014−2015), with recovery a function of location. We find drought− isoprene effects are temperature-dependent, even after accounting for changes in leaf area, consistent with laboratory studies but not previously observed at landscape scales with atmospheric observations. Drought-driven decreases in organic reactivity are contemporaneous with a change in dominant oxidation mechanism, with PO 3 becoming more NO x-suppressed, leading to a decrease in PO 3 of ∼20%. We infer reductions in atmospheric O 3 loss of ∼15% during the most severe drought period, consistent with past observations of decreases in O 3 uptake by plants. We consider drought-related trends in O 3 variability on synoptic time scales by analyzing statistics of multiday high-O 3 events. We discuss implications for regulating O 3 air pollution in California and other locations under more prevalent drought conditions.