Dewfall during the dry season in an old growth conifer canopy: characterization of a potentially important water subsidy to trees and canopy biota

Abstract:

Recent research has highlighted increasing trends in atmospheric vapor pressure deficit (VPD) over much of the continental United States in recent decades, and future projections predict further increases, particularly in the Western U.S. This change is likely to modify important feedbacks between forest canopies and the microclimates that they create. One such feedback involves the deposition of dew on emergent portions of tall tree canopies. Emergent canopy layers can be simultaneously hotter, drier, brighter and windier than the layers below them during the day, but have a more negative net radiation balance on cloud-free nights and thus cool to lower temperatures. These nighttime conditions increase the likelihood of dewfall on the canopy layers that are the most moisture limited, which can be of particular importance for tree foliage, epiphytic bryophytes and lichens during dry seasons. While dry season dew deposition is likely an important component of vegetation health and is likely to be impacted by rising atmospheric VPD, the frequency and timing of these events are poorly understood.

In this study, we instrumented a 65-meter-tall old growth Douglas-fir tree (Pseudotsuga menziesii) with leaf wetness and meteorological sensors at intervals of 1.5, 10, 20, 30, 40 and 56 meters above ground. The measurement tree stands in a grove of old growth conifer trees at the H.J. Andrews Experimental Forest in Blue River, OR, where a Mediterranean climate regime brings approximately three month long dry seasons with periods of intense moisture deficit. Using a simple set of rules, we classified the leaf wetness sensor data at every height into classes of dry, wet by rain, wet by dew, frosted or wet by ambiguous source. By analyzing our three year data record, we found that time spent wet by dew is up to ten times longer at the top of the canopy than at the forest floor, and that multiple dew wetting events occur every late July and early August, when soil moisture deficits are at their peak and rain wetting is nearly non-existent. We paired these findings with field observations of shoot water potential to demonstrate the extent to which this dewfall is taken up by Douglas-fir foliage. Our findings are important for understanding how future increases in atmospheric dryness will change the frequency of dry season dew subsidies to vegetation.