During daylight hours, the isotope composition of leaf water generally approximates steady-state leaf water isotope enrichment model predictions. However, until very recently there was little direct confirmation that isotopic steady-state (ISS) transpiration in fact exists. Using isotope ratio infrared spectroscopy (IRIS) and leaf gas exchange systems we evaluated the isotope composition of transpiration and the rate of change in leaf water isotopologue storage (isostorage) when leaves were exposed to variable environments. In doing so, we developed a method for controlling the absolute humidity entering the gas exchange cuvette for a wide range of concentrations without changing the isotope composition of water vapour. The measurement system allowed estimation of 18O enrichment both at the evaporation site and for bulk leaf water, in the steady state and the non-steady state.We show that non-steady-state effects dominate the transpiration isoflux even when leaves are at physiological steady state. Our results suggest that a variable environment likely prevents ISS transpiration from being achieved and that this effect may be exacerbated by lengthy leaf water turnover times due to high leaf water contents.

In Mediterranean climates, the season of water availability (winter) is out of phase with the season of light availability and atmospheric moisture demand (summer). We investigate the seasonality of evergreen tree transpiration in a Mediterranean climate, using observations from a small (4000 m 2 ), forested, steep (32 ) hillslope, in the northern California Coast Range. We analyze 3 years of half-hourly measurements from 39 sap flow sensors in 26 trees, six depth profiles of soil moisture measured by TDR, and spatially distributed measurements of micrometeorology from five locations. The sap flow measurements show that two common evergreen tree species have different seasons of peak transpiration. Douglas-firs (Pseudotsuga menziesii) maintain significant transpiration through the winter rainy season and transpire maximally in the spring, followed by a sharp decline in transpiration in the summer dry season. Pacific madrones (Arbutus menziesii), and to a lesser extent other broadleaf evergreen species (Quercus wislizeni, Notholithocarpus densiflorus, Umbellularia californica), in contrast, transpire maximally in the summer dry season. The seasonal patterns are quantified using principal component analysis. Markov chain Monte Carlo estimation of response to environmental variables shows that the difference in transpiration seasonality arises from different sensitivities to atmospheric evaporative demand and root-zone moisture. The different sensitivities to atmospheric evaporative demand also create species differences in transpiration variability at synoptic time scales. Using the sap flow measurements and a regional forest inventory, a bottom-up regional transpiration estimate is constructed. The estimate suggests that sensitivity of Douglas-fir transpiration to water stress suppresses dry season evapotranspiration at the regional scale.

Obtaining the d‐excess parameter from oxygen and hydrogen stable isotope composition of meteoric waters has the potential power to reconstruct changes in atmospheric water pools (e,g. sources, origins and overall balance) and the climatic conditions that prevail during surface evaporation. Recently, plant and ecosystem scientists turned their attention using d‐excess information to inform questions at these scales. Here, we use the d‐excess parameter to evaluate the influence of forest canopies on atmospheric humidity within a mixed evergreen forest in coastal California. We found that during the day, when transpiration was at a maximum, the d‐excess of atmospheric water vapour exceeded model predictions for the background atmosphere into which the ecosystem evapotranspiration mixes. At night when transpiration was minor, the d‐excess of atmospheric water vapour was on average less than model predictions for an ocean derived water vapour source. The observed diurnal fluctuations around the d‐excess of the modelled background water vapour provided a strong evidence that during the day as the land surface warms and the boundary layer grows plants alter the isotope composition of atmospheric humidity via non‐steady state isotope effects. In contrast, at night equilibrium isotope effects dominate as the atmosphere stabilizes. These day and nighttime fluctuations around the d‐excess of ocean derived water vapour highlight the importance of plant transpiration for the isotope hydrology of near surface humidity and subsequently for the isotope composition of condensate like dew, an important water input to this ecosystem. Copyright © 2013 John Wiley & Sons, Ltd.