Disentangling the relative impacts of precipitation reduction and vapour pressure deficit (VPD) on plant water dynamics and determining whether acclimation may influence these patterns in the future is an important challenge. Here, we report sap flux density (F ), stomatal conductance (G ), hydraulic conductivity (K ) and xylem anatomy in piñon pine (Pinus edulis) and juniper (Juniperus monosperma) trees subjected to five years of precipitation reduction, atmospheric warming (elevated VPD) and their combined effects. No acclimation occurred under precipitation reduction: lower G and F were found for both species compared to ambient conditions. Warming reduced the sensibility of stomata to VPD for both species but resulted in the maintenance of G and F to ambient levels only for piñon. For juniper, reduced soil moisture under warming negated benefits of stomatal adjustments and resulted in reduced F , G and K . Although reduced stomatal sensitivity to VPD also occurred under combined stresses, reductions in G , F and K took place to similar levels as under single stresses for both species. Our results show that stomatal conductance adjustments to high VPD could minimize but not entirely prevent additive effects of warming and drying on water use and carbon acquisition of trees in semi-arid regions.

Productivity of conifers in seasonally snow-covered forests is high before and during snowmelt when environmental conditions are optimal for photosynthesis. Climate change is altering the timing of spring in many locations, and changes in the date of transition from winter dormancy can have large impacts on annual productivity. Sap flow methods provide a promising approach to monitor tree activity during the cold season and the winter-spring and fall-winter transitions. Although sap flow techniques have been widely used, cold season results are generally not reported. Here we examine the feasibility of using the Granier thermal dissipation (TD) sap flux density method to monitor transpiration and dormancy of evergreen conifers during the cold season. We conducted a laboratory experiment which demonstrated that the TD method reliably detects xylem water transport (when it occurs) both at near freezing temperature and at low flow rate, and that the sensors can withstand repeated freeze-thaw events. However, the dependence between sensor output and water transport rate in these experiments differed from the established TD relation. In field experiments, sensors installed in two Abies forests lasted through two winters and a summer with low failure. The baseline (no-flow) sensor output varied considerably with temperature during the cold season, and a new baseline algorithm was developed to accommodate this variation. The Abies forests differed in elevation (2070 and 2620 m), and there was a clear difference in timing of initiation and cessation of transpiration between them. We conclude that the TD method can be reliably used to examine water transport during cold periods with associated low flow conditions.

Bark beetle outbreaks are widespread in western North American forests, reducing primary productivity and transpiration, leading to forest mortality across large areas and altering ecosystem carbon cycling. Here the carbon isotope composition (δ 13 C) of soil respiration (δ J ) was monitored in the decade after disturbance for forests affected naturally by mountain pine beetle infestation and artificially by stem girdling. The seasonal mean δ J changed along both chronosequences. We found (a) enrichment of δ J relative to controls (<1 ‰) in near-surface soils in the first 2 years after disturbance; (b) depletion (1‰ or no change) during years 3-7; and (c) a second period of enrichment (1-2‰) in years 8-10. Results were consistent with isotopic patterns associated with the gradual death and decomposition of rhizosphere organisms, fine roots, conifer needles and woody roots and debris over the course of a decade after mortality. Finally, δ J was progressively more 13 C-depleted deeper in the soil than near the surface, while the bulk soil followed the well-established pattern of 13 C-enrichment at depth. Overall, differences in δ J between mortality classes (<1‰) and soil depths (<3‰) were smaller than variability within a class or depth over a season (up to 6‰).

No abstract

No abstract