PREMISE OF THE STUDY: The Pacific Northwest of North America experiences relatively mild winters and dry summers. For the tall coniferous trees that grow in this region, we predicted that loss in the hydraulic conductivity of uppermost branches would be avoided because of difficulty reversing accumulated emboli in xylem that is always under negative pressure. METHODS: To test this hypothesis, we measured native percent loss in hydraulic conductivity (PLC; the decrease of in situ hydraulic conductivity relative to the maximum) monthly throughout 2009 in branches at the tops (∼50 m) of four species in an old growth forest in southern Washington. KEY RESULTS: Contrary to our prediction, freeze-thaw cycles resulted in considerable native PLC. Branches showed hydraulic recovery in the spring and after a moderate increase in native embolism that was observed after an unusually hot period in August. The September recovery occurred despite decreases in the leaf and stem water potentials compared to August values. CONCLUSIONS: Recoveries in branches of these trees could not have occurred by raising the water potential enough to dissolve bubbles simply by transporting water from roots and must have occurred either through water absorption through needles and/or refilling under negative pressure. Excluding the August value, native embolism values correlated strongly with air temperature of the preceding 10 d. For three species, we found that branches with lower wood density had higher specific conductivity, but not greater native PLC than branches with higher wood density, which calls into question whether there is any hydraulic benefit to higher wood density in small branches in those species.
PREMISE OF THE STUDY: The Pacific Northwest of North America experiences relatively mild winters and dry summers. For the tall coniferous trees that grow in this region, we predicted that loss in the hydraulic conductivity of uppermost branches would be avoided because of difficulty reversing accumulated emboli in xylem that is always under negative pressure. METHODS: To test this hypothesis, we measured native percent loss in hydraulic conductivity (PLC; the decrease of in situ hydraulic conductivity relative to the maximum) monthly throughout 2009 in branches at the tops (∼50 m) of four species in an old growth forest in southern Washington. KEY RESULTS: Contrary to our prediction, freeze-thaw cycles resulted in considerable native PLC. Branches showed hydraulic recovery in the spring and after a moderate increase in native embolism that was observed after an unusually hot period in August. The September recovery occurred despite decreases in the leaf and stem water potentials compared to August values. CONCLUSIONS: Recoveries in branches of these trees could not have occurred by raising the water potential enough to dissolve bubbles simply by transporting water from roots and must have occurred either through water absorption through needles and/or refilling under negative pressure. Excluding the August value, native embolism values correlated strongly with air temperature of the preceding 10 d. For three species, we found that branches with lower wood density had higher specific conductivity, but not greater native PLC than branches with higher wood density, which calls into question whether there is any hydraulic benefit to higher wood density in small branches in those species.