Dioecious woody plants in riparian zones often exhibit highly biased sex ratios, which may be explained by sex differences in water use efficiency and resource requirements. The sex ratio in willows (Salix sitchensis) colonizing primary successional sites at Mount St. Helens is slightly biased (54% female), but highly biased in nearby secondary successional sites (69%). These willows are highly attacked by stemboring weevils in upland sites, but less so in riparian sites. We investigated drought stress and nutrient dynamics in these sites to better understand differences in weevil attack. We hypothesized that upland plants and especially females would exhibit signs of drought stress, while males would exhibit greater water use efficiency (WUE).

We measured pre-dawn and mid-day leaf water potential (LWP), using a pressure chamber, on un-attacked stems of male and female willows in upland and riparian habitats on four warm dates in summer 2008 (n=88 plants).  We also measured long-term (growing season) WUE (based on δ¹³C),  δ¹⁵N, leaf %C and %N in leaves of these same plants, and in N-fixing Lupinus lepidus from the same sites.

Results/Conclusions

There were no consistent differences between sexes in LWP or transpiration, although at some upland sites on some dates males had a more negative pre-dawn LWP or mid-day LWP, indicating greater drought stress for those individuals. However, there were no sex differences or interaction effects in any of the leaf characteristics (%C, %N, δ ¹³C,  and δ¹⁵N). Either males and females do not differ physiologically, or the roots of these plants may be sufficiently deep that neither males or females suffer significant drought stress. Studies of experimental plantings may reveal greater sex differences and are currently underway.

LWP was consistently more negative in upland habitats (indicating greater drought stress), but long-term WUE (δ¹³C) did not differ between habitats. Leaves from upland plants were significantly richer in N, and exhibited a highly significant enrichment in ¹⁵N  (means: d¹⁵N Riparian =-4.18, Upland=-1.66; p<0.0001). Leaves of L. lepidus growing in the upland area had mean δ¹⁵N = 0.42. The greater leaf %N and δ¹⁵N in upland willows may reflect a high availability of recently-fixed N from insect-killed Lupinus lepidus (mean δ¹⁵N = 0.42), relative to riparian plants just tens of meters away. The increased N may be related to high levels of herbivory in upland sites.