Differences in water use between mature and post-fire regrowth stands of subalpine Eucalyptus delegatensis R. Baker

Abstract

We estimated plot level water use from sap flux measurements over a 9-month period, in post-fire regrowth and mature plots of Eucalyptus delegatensis R. Baker (Alpine ash) in high elevation catchments near Falls Creek, Victoria, Australia, seven years after a major stand-replacing bushfire. Water use was more than double (460±100mmyear−1 more) in regrowth as compared to mature plots, whereas sap flux was similar between age classes. This difference in water use reflected 72% greater sapwood area index and 35% greater leaf area index in regrowth than in mature plots. A small part of the difference in water use can be attributed to nocturnal transpiration, which was greater in regrowth than in mature plots (10.3±0.8% vs 7.3±0.8% of diel totals). As evaporative demand was 41% greater in mature than in regrowth plots, these data suggest mean transpiration rate and stomatal conductance per unit leaf area were approximately 1.6 and 2.3 times greater, respectively, in the regrowth. However, mid-day leaf water potential and photosynthetic capacity were similar in both age classes. Evaporative demand was the primary environmental driver of water use in all cases, whereas soil moisture was not a strong driver of either water use or canopy conductance (estimated as sap flux/evaporative demand). Together, our results suggest (a) stand water use rapidly recovers after fire in these high elevation forests and quickly surpasses rates in mature stands, confirming projections by Kuczera (1987) for lower-elevation Mountain Ash (E. regnans) forests and highlighting the potential impact of tree water use on water yield in the first decade of forest regeneration after fires, and (b) stomatal conductance and sapwood area/leaf area ratio are both less in tall, older Alpine ash trees, whereas leaf water status and photosynthetic capacity appear to be sustained – consistent with predictions from optimisation theory but not Pipe-Model Theory.