Implications of dioecy for sustainable forest management Stan Van Uden*, Glenn H. Stewart*" and Richard P. Duncan*** Abstract The dense podocarp forests of New Zealand, and in South West- land in particular, are unusual in that they are dominated by dioecious trees. Females and males of dioecious species perform different reproductive functions and consequently may use resources in different ways. This differential resource use can result infemales and males differing in growth rate, age at matu- rity, and life span, which in turn can result in difSerences between the sexes in size distribution, age distribution and number within a population. DifSerences in growth rate and size in particular could result in one sex being harvested at a greater rate than the other, altering the natural sex ratio - an outcome that may have important implications for the long-term sustainability of managed dioecious forests. We review the dlfferences between the sexes for the two species studied in New Zealand (kahikatea and rimu) and pre- sent additional rimu sex data from dcfferently-aged landforms in Saltwater Forest, South Westland. Female dominated, male dom- inated, and 1:l ratios of the sexes have all been found, and in some instances d~fferences in growth rate reported. We discuss the implications of dioecy and dlfferences in growth and mortal- ity between the sexes, and suggest areas of research required, for sustainable forest management. Introduction The Forests Amendment Act (1993) requires management of the remaining indigenous forests of New Zealand for timber pro- duction to be based on ecologically sustainable silviculture. Sus- tainable harvest from indigenous forest is aimed at maintaining near-natural forest structure and composition, and emphasis is placed on forests retaining both their natural values and their regenerative capabilities in perpetuity (Ministry of Forestry 1997). In most forests there is no need to take account of gender when selecting trees for harvesting since most forests are domi- nated by monoecious species, where both female and male repro- ductive structures occur together on each individual. In New Zealand, however, extensive areas of forest are dominated by dioecious conifers, where the female and male reproductive struc- tures occur on separate individuals. Of the native conifer tree species harvested in New Zealand for timber production, eight are strictly dioecious, two are sometimes dioecious, and only one is strictly monoecious (Table I). We searched the local and international literature and could find little published research on the implications of dioecy for production forest management, or on how harvesting may affect the population structure of dioecious species. This suggests that there is a significant gap in our understanding of dioecious for- est management. In this paper we examine the possible implica- tions of dioecy for sustainable forest management in New Zealand. * Muster of Applied Science student, Depurtment of Plunt Science, Lincoln University, PO Box 84, Lincoln, Cunterbury, New Zeulund. ** Reader in Ecology, Department cfPlunt Science, Lincoln University, PO Box 84, Lincoln, Cunterbury, New Zeulund. *** Lecturer in Ecok~gy, Department of Plunt Science, Lincoln University, PO Box 84, Lincoln, Cunterbury, New Zeulund. Table 1 New Zealand conlfei ipecm ihowmg w h~h species are dmmous. monoecxous or both and whxh species are eQ hariested . - - Famdy, genus, species and common name D~wemus/monoec~ous Harvested - - - - - - - Araueariaeeae Agathrj australis (kaun)" monoec~ous small amounts Cupressaceae Llbocedrui bidciilii (kalkawaka, cedar) dmecmui or monoecmus small amounts Libocedrusplcirnosa (kawaka) dmcms or monoecms no Podocarpaeeae Dacr3carpur dncgdioides (kahiatca) dioec~ous Dac~diurn cupressinum (nmu) dmcms Hoiocorpus brdwiiirr (bog pme) d~oec~ous Hniociirpus brforrnis (pink pme) dmc~ous Hniocarpus kirkri (monoao) dioecmus Leprdotharnnus inrermsdrus (yellow-sliver pine) dmecmus, rarely monoecmus Leptdothomnui laxifolius (pygmy pine) dmcmus or monoecmus Monoao colensoi (sliver pme) dmecmus Ph)llociudur aspienrfiliui rar nlpinw (mountnn taataa) monoeclous Ph>iiociadus giaucus (toatoa) dmecmus 01 monoecmus Phyilocladur iricl~omanoides (lanekaha. celery p m ) dmeclous or monoeclous Padocarpur acutfolius dmecmus Podocarpus ctinningharni~ (Hall's totara) dmc~aus Podocnrpus nrualrs (mow totara) dmecmus Podocarpui ioiara (totara) dmecms P,umnopiiys ferruginea (miro) dmcms Prurnnopirys iaxifolia (matal) dmecmus . - --- - -. - - -- -- - - small amounts no no no no no small amounts no no small amounts small amounts small amounts "0 small amounts small amounts small amounts Dioecy and population structure In dioecious species, reproductive costs can be greater for females than males because females are required to put considerable effort into producing and maturing seeds and associated dispersal stmc- tures, a process that can span several growing seasons. If more resources are put into reproduction, fewer are available for other functions such as growth and defence, resulting in the females of many dioecious species (1) growing slower, (2) maturing later, and (3) dying earlier than males (Lloyd & Webb 1977). These differences can manifest themselves as differences in the popu- lation structure of the sexes in terms of: (1) Size and age distributions. Because they grow slower and die earlier females may be on average smaller and younger than males (Lloyd & Webb 1977). (2) Spatial distribution. Males may have higher survival rates and so be more common in harsher environments, such as dry or low nutrient sites, whereas females may occur more fre- quently in fertile or wetter sites (Freeman, Klikoff & Harper 1976). The scale at which females and males show spatial segregation will depend on the environmental conditions that limit growth and survival and may be influenced by soil type, elevation, topography, drainage patterns, and fertility (Free- man, Klikoff & Harper 1976, Lloyd & Webb 1977, Grant & Mitton 1979). (3) Sex ratios. Differential survival will result in the longer-lived sex outnumbering the other. In particular, if females die ear- lier then the male population will be on average older and the overall population will be male biased (Shea, Dixon & Sharitz 1993). Males have been found to exceed females in both growth rate and longevity, and consequently in both size I. Nomencluture,follows Allun (1 96I),fi)r Aruucuriuceue cmd Cupressaceue and Edgar and Connor ( I 983)j'i)r Podocurpuceue except,for Monouo colen- soi which,follows Molloy (199%). N.Z. FORESTRY FEBRUARY1998 39