Influence of agroforestry practices on the structure and spatiality of shea trees (Vitellaria paradoxa C.F. Gaertn.) in central-west Burkina Faso

This article examines the role of humans in shaping a shea tree (Vitellaria paradoxa C.F. Gaertn.) population in Prata, central-west Burkina Faso. Four management regimes or land uses were considered: fields cultivated by indigenous Gurunsi farmers, fields cultivated by migrant Moose farmers, fallows, and bush lands. The structure of the shea population differed between Gurunsi and Moose fields and between these fields and fallow and bush lands. The size class distribution of V. paradoxa in fallows and bush lands was skewed towards the lower classes and the slope of the distributions was negative and significant, indicating the occurrence of recruitment. In comparison, Gurunsi fields carried young and mature trees with 95 % of individuals in the 15.5–90 cm range, whereas Moose fields carried no specimen with dbh <16.5 cm. The slope of the size class distribution was slightly negative for Gurunsi fields and slightly positive for Moose fields, but non significant in both cases. Vitellaria paradoxa densities did not significantly differ between Gurunsi (35 stems/ha) and Moose (26 stems/ha) fields, but were higher in fallows (172 stems/ha) and in the bush (161 stems/ha) than in cultivated fields. Nearest neighbour distances were progressively greater from uncultivated fields to Gurunsi and finally Moose fields, whereas shea trees were increasingly aggregated from Moose to Gurunsi fields to fallows and bush lands. In Prata, shea tree management is thus associated with ethnicity and/or with a host/migrant status that confers different farm sizes and levels of tenure security to farming households. Gurunsi and Moose farmers cited productivity, spacing and shading effects as the main factors influencing their decision to conserve specific shea trees in their fields. Results signal favorable prospects for V. paradoxa regeneration in Prata’s fallows and bush lands and to a lesser extent in Gurunsi fields.     

Patterns of tree growth in a single tree selection silver fir–European beech forest

The basal area and height growth of trees and saplings in silver fir–European beech single stem selection forest were studied with regard to their social status and crown parameters of size, coverage, shading and vitality. On 24 permanent research plots (20 m × 20 m each) all trees [diameter at breast height (dbh) ≥10 cm] and saplings (≥1.3 m tall and dbh <10 cm) were surveyed. Repeated measurements of dbh (N = 1,608) and height (N = 1,135) 10 years apart enabled the calculation of annual basal area increment (BAI) of trees and saplings, and annual height (HI) increment of saplings. To obtain the growth characteristics for individual trees and saplings, their social status and crown parameters were assessed by rank. In the multivariate general linear model for BAI, social status, crown size and crown coverage of individuals as the predictors, and dbh² as the covariate, explained 70% of total variability. Similarly, social status, crown size, crown coverage and crown shading had a significant impact on the HI of saplings, explaining 70% of total variability. Among the observed variables, social status determined according to the individual’s position in vertical stand structure was, in addition to dbh, the most important predictor for both BAI and HI. Significant differences were observed between the BAI and HI models for the main tree species (European beech and silver fir), indicating their different growth characteristics. The applied method could be used as a supplement to the more widely used approaches for studying basal area and height growth of individual trees in selection forest stands.     

Factors influencing stump sprouting by pondcypress (Taxodium distichum var. nutans (Ait.) Sweet)

Sprouting can be an important mechanism of regeneration in harvested cypress (Taxodium sp.) wetlands; however, limited information exists on factors influencing sprouting. We examined the influence of harvesting practices, tree size, and site conditions on stump sprouting in eight pondcypress (Taxodium distichum var. nutans (Ait.) Sweet) wetlands in Florida, USA. Stumps were cut at three heights: 20 cm below high water level, at high water, and 40 cm above high water. Half of each site was harvested using a sawhead feller and the other half using shearhead. We monitored sprouting for 2 years post-harvest. Stump height affected first year sprouting, but the effect disappeared during year 2. Felling equipment did not affect sprouting; however, the shearhead caused greater stump damage. Sprouting declined with increasing parent tree size; percentage of stumps with sprouts after 2 years was 68% for trees 10–14 cm dbh and 28% for trees 20–24 cm dbh. Stump sprouting ranged among sites from 23 to 54% after 2 years. Results indicate that sprouting may be an important, but inadequate, form of regeneration. At sites with short hydroperiods, dense understories, and/or predominance of trees greater than 20 cm dbh, reduced sprouting may occur, and planting seedlings may be beneficial.     

Long-term spatial dynamics in an old-growth deciduous forest

In 1926, 1976, and 1986 stems (≥ 10 cm diameter at 1.37 m (dbh)) were measured and mapped in a 4 ha interior study area on the Davis-Purdue Research Forest in east-central Indiana. Spatial pattern type was determined using the Clark and Evans's index, and Ripley's L function. The G function was used to examine spatial dispersion at intertree distances (< 20 m). Spatial pattern of combined size classes shifted from weakly non-random in 1926 to uniform in 1976 and 1986. Neighborhood density was depressed for 2 m in 1976 and 5 m in 1986. Peak in neighborhood density at 6 m suggested trees were spaced 6 m apart. Subcanopy (10–25 cm dbh) tree spatial pattern shifted from aggregated in 1926 to uniform in 1976. The spatial pattern of canopy (> 25 dbh) trees was uniform between 1926 and 1986. Density-dependent mortality and ingrowth processes are maintaining uniform spatial distributions. Initial neighborhood tree distribution was higher around trees which died in the periods 1926–1976 and 1976–1986 than for contemporary survivors, i.e. trees which survived this period had fewer neighboring trees within 6 m at the beginning of the period than did trees which died. The higher initial neighborhood densities around mortality trees than survivors supports density-dependent mortality. Ingrowth was inhibited in a 6 m zone proximite to established trees for both the 1926–1976 and 1976–1986 periods.     

Tree microhabitats at the stand scale in montane beech–fir forests: practical information for taxa conservation in forestry

Recent studies have highlighted the key role of tree microhabitats in forest habitat complexity and have suggested using them as surrogates for local taxonomic biodiversity. However, few practical guidelines have been published to help foresters in managing microhabitats at the stand scale. This paper provides scientific background information to help to develop such guidelines. We surveyed trees in nine long-unmanaged beech–fir forests to model tree microhabitat occurrence and diversity at the tree level. Data were upscaled to a size range of tree cluster, i.e., at the tree population scale, by aggregating observed values of microhabitat occurrence. Accumulation curves were used to estimate the minimum number of trees required to make all the microhabitat types available. Two managed forests were then studied to quantify management effects on microhabitats. Diameter at breast height (dbh) and tree species, respectively, explained 16 and 10 % of the variations in the number of microhabitat-bearing trees, and 21 and 10 % for the number of microhabitat types. Beech trees and firs with a dbh of less than dbh 50 and 65 cm, respectively, did not ensure the provision of all microhabitat types. At least 20 ha of unmanaged forest were necessary to conserve all the microhabitat types. Current management practices have reduced the number of microhabitat-bearing beeches both by reducing the number of very large trees (dbh > 67.5 cm) and by tree selection within mid-size diameters. For fir, only the logging of very large trees (dbh > 62.5 cm) negatively affected microhabitats. These figures may inspire guidelines for conservation-friendly forestry.     

Regeneration dynamics and sylvigenesis in the moist deciduous forests of southwest India

Sustained management of natural forest depends on their ability to regenerate. The pace at which older trees are replaced by younger ones is important in this respect. This study was conducted to assess the demographic details of natural regeneration dynamics of south Indian moist deciduous forests. The status of natural regeneration has been assessed in eight localities of varying levels of disturbance in Thrissur Forest Division of Kerala State of southwest India. Enumeration of trees and their regeneration were analyzed at three levels of organization; whole-stand, stratum, and species.The trees of the forest were grouped into upper, middle and lower vertical strata. Species richness was highest in the middle stratum. Upper stratum species were commercially important. The mean basal area for the Forest Division was 12.83 m²/ha. and average number of trees per hectare (>=20 cm dbh) was 150. Regeneration of the upper stratum species was higher at about 10% gap in the canopy.The growing stock of established seedlings and saplings (height > 100 cm and dbh <=10cm) was very low compared to unestablished seedlings (height <=100 cm). Owing to very low survival probability in the sapling stage (dbh >=1 cm to <=10 cm), a low numbers of poles (dbh >=10 cm to <=20 cm) in the upper stratum was observed. Presence of fire, grazing, browsing and over expoitation by humans are the major constraints of natural regeneration recognized in these forests.     

Oak/Douglas-fir/sheep: A three-crop silvopastoral system

A small scale agroforestry study which was begun in 1952 was revisited in 1985 to evaluate the long-term influence of site preparation and grazing on tree growth and survival in a Douglas-fir (Pseudotsuga menziesii)-white oak (Quercus garryana)-sheep silvopastoral system. In 1952–1953, two-year-old Douglas-fir seedlings were planted at the rate of 2500 trees/ha under three levels of site preparation: (1) no treatment, (2) oak thinned by 50%, and (3) oak clearcut. From 1954 to 1960, yearling ewes grazed one-half of each of the three thinning treatments for 3–4 weeks each spring. The conifers have been undisturbed since grazing was discontinued in 1960. Survival of planted conifers averaged 64% in 1985 and did not vary among either site preparation or grazing treatments. From 1964 to 1985, trees on the thinned and clearcut plantations grew an average of 1060 and 990 cm in height, respectively, compared to 900 cm on the unthinned plantation. Diameter at breast height (dbh) averaged 3.8 and 5.6 cm greater on thinned or clearcut plantations, respectively, than on the unthinned control by 1985 (p < 0.05). Conifers on grazed plantations had increased height and dbh growth during the first 12 years of plantation life, averaging 63 cm taller (p < 0.10) and 0.7 cm greater in dbh (p < 0.05) than the ungrazed plots by 1964. By 1985 the difference in height (122 cm) and dbh (1.0 cm) between grazed and ungrazed plantations was not statistically significant. These data suggest that although site preparation can positively influence conifer growth, total clearfelling is no better than thinning oaks. Furthermore, proper grazing can increase height and dbh growth of the conifers during and immediately after the grazing years.     

Red-tree vole habitat suitability modeling: Implications for conservation and management

In the Pacific Northwest, USA, red-tree voles (Arborimus longicaudus) are of conservation and management interest owing to their apparent association with late-seral forests and the relatively small extent of such forests, largely a function of timber harvest, fire, and conversion of forests to non-forest uses during the past century. We created and evaluated a series of red-tree vole habitat association models, and applied the best model to evaluate tree vole habitat quality within and outside of reserves throughout most of their range in Oregon and northern California. We modeled presence and absence of tree vole nests across a gradient of biotic, abiotic, and spatial features; and within and outside of reserves. The best model included spatial coordinates, percent slope, basal area of trees with diameter at breast height (dbh) between 45 and 90 cm, maximum tree dbh, and standard deviation of conifer dbh. Plots with tree vole nests contained many late-seral/old-growth forest attributes such as large diameter, older, and variably sized trees. Evaluation of the best model, including rigorous cross-validation, showed the model to be statistically robust and to have very good/excellent predictive ability. Reserves had significantly higher mean habitat quality than non-reserved lands, and reserves had much more high quality habitat than non-reserves.     

Oak/Douglas-fir/sheep: A three-crop silvopastoral system

A small scale agroforestry study which was begun in 1952 was revisited in 1985 to evaluate the long-term influence of site preparation and grazing on tree growth and survival in a Douglas-fir (Pseudotsuga menziesii)-white oak (Quercus garryana)-sheep silvopastoral system. In 1952–1953, two-year-old Douglas-fir seedlings were planted at the rate of 2500 trees/ha under three levels of site preparation: (1) no treatment, (2) oak thinned by 50%, and (3) oak clearcut. From 1954 to 1960, yearling ewes grazed one-half of each of the three thinning treatments for 3–4 weeks each spring. The conifers have been undisturbed since grazing was discontinued in 1960. Survival of planted conifers averaged 64% in 1985 and did not vary among either site preparation or grazing treatments. From 1964 to 1985, trees on the thinned and clearcut plantations grew an average of 1060 and 990 cm in height, respectively, compared to 900 cm on the unthinned plantation. Diameter at breast height (dbh) averaged 3.8 and 5.6 cm greater on thinned or clearcut plantations, respectively, than on the unthinned control by 1985 (p < 0.05). Conifers on grazed plantations had increased height and dbh growth during the first 12 years of plantation life, averaging 63 cm taller (p < 0.10) and 0.7 cm greater in dbh (p < 0.05) than the ungrazed plots by 1964. By 1985 the difference in height (122 cm) and dbh (1.0 cm) between grazed and ungrazed plantations was not statistically significant. These data suggest that although site preparation can positively influence conifer growth, total clearfelling is no better than thinning oaks. Furthermore, proper grazing can increase height and dbh growth of the conifers during and immediately after the grazing years.     

Biomass,stem basic density and expansion factor functions for five exotic conifers grown in Denmark

Adequate allometric equations are needed for estimating carbon pools of fast growing tree species in relation to international reporting of CO₂ emissions and for assessing their possible contribution to increasing forest biomass resources. We developed models for predicting biomass, stem basic density and expansion factors of stem to above-ground biomass for five fast growing conifers. Data included destructive measurements of 236 trees from 14 sites, covering a wide range of growth conditions. To ensure model efficiency, models for predicting stem, crown and total above-ground biomass for the five species were estimated simultaneously using a linear, mixed effects model that allowed contemporaneous correlations between the different tree components. Models differed among species and included dbh and tree height. The models explained more than 98% of the variation in above-ground biomass and reflected differences in the allometry between tree species. Stem density differed among species but generally declined with increasing site index and dbh. The overall model for predicting stem basic density included dbh, H₁₀₀ and site index and explained 66% of the total variation. Expansion factors decreased from 1.8–2.0 in small trees (dbh < 10 cm) to 1.1–1.2 for large trees (dbh > 25 cm), but differed among species. The overall model explained 86% of the variation and included quadratic mean diameter and dbh.     

Impacts of selective logging on above-ground forest biomass in the Monts de Cristal in Gabon

Selective logging is an important socio-economic activity in the Congo Basin but one with associated environmental costs, some of which are avoidable through the use of reduced-impact logging (RIL) practices. With increased global concerns about biodiversity losses and emissions of carbon from forest in the region, more information is needed about the effects of logging on forest structure, composition, and carbon balance. We assessed the consequences of low-intensity RIL on above-ground biomass and tree species richness in a 50 ha area in northwestern Gabon. We assessed logging impacts principally in 10 randomly located 1-ha plots in which all trees ?10 cm dbh were measured, identified to species, marked, and tagged prior to harvesting. After logging, damage to these trees was recorded as being due to felling or skidding (i.e., log yarding) and skid trails were mapped in the entire 50-ha study area. Allometric equations based on tree diameter and wood density were used to transform tree diameter into biomass.Logging was light with only 0.82 trees (8.11 m³) per hectare extracted. For each tree felled, an average of 11 trees ?10 cm dbh suffered crown, bole, or root damage. Skid trails covered 2.8% of the soil surface and skidding logs to the roadside caused damage to an average of 15.6 trees ?10 cm dbh per hectare. No effect of logging was observed on tree species richness and pre-logging above-ground forest biomass (420.4 Mg ha⁻¹) declined by only 8.1% (34.2 Mg ha⁻¹). We conclude from these data that with harvest planning, worker training in RIL techniques, and low logging intensities, substantial carbon stocks and tree species richness were retained in this selectively logged forest in Gabon.     

Effects of management, environment and landscape conditions on establishment of hardwood seedlings and saplings in central Japanese coniferous plantations

There is an increasing need to restore natural hardwood forests in landscapes dominated by monocultural conifer plantations. A convenient restoration approach is to exploit natural regeneration processes. Natural regeneration, however, is affected by diverse interacting factors, for which better understanding is required, in order to optimize restoration programs. To identify optimal management practices for improving natural regeneration of hardwood trees in coniferous plantations, we examined the effects of multiple factors on the abundance of seedlings, small saplings and large saplings (height <0.3, 0.3-1.3 and ?1.3 m, respectively) of hardwood tree and shrub species in both line thinned (LT) and unthinned (UT) plantations of sugi (Cryptomeria japonica) and hardwood forests (HF) in central Japan. The effects of management practices (number of the times of weeding and cleaning, thinning method, years after thinning and forest age), environment (slope position, slope angle and canopy openness), and landscape conditions (distance from nearest hardwood forest, altitude and landuse before planting) on the number of hardwood individuals were examined by using the data obtained from the LT plantations. We also compared hardwood density between LT and UT plantations to examine the effect of line thinning. Finally, we examined species composition of LT plantations and HF to identify hardwood forest components in the thinned plantations. The effects on hardwood regeneration of environmental conditions, landscape factors and management practices applied in the plantations varied, depending on the size class and life form of the regenerating species. The abundance of large saplings of tall tree species was affected by several management factors, especially number of the times of weeding. Landscape conditions (distance from the nearest hardwood forest and altitude) affected the abundance of small saplings and seedlings of tall tree species, but not the other classes. Seedlings and small saplings of many tall tree species that contribute to hardwood forest canopies were less abundant in the LT plantations. The results show that numerous factors affect the establishment and abundance of naturally regenerating hardwood tree species, and suggest that successful establishment during early plantation stages can have long-lasting effects on natural regeneration of tall tree species.     

Browsing damage on broadleaved trees in semi-natural temperate forest in Sweden,with a focus on oak regeneration

Absence of, or poor, oak (Quercus spp.) regeneration is a problem in uneven-aged, mixed closed-canopy broadleaved forests. Browsing by ungulates on small trees may contribute to poor oak regeneration in such forests. This possibility was investigated in 25 Swedish stands, and browsing damage was analysed in relation to landscape and stand factors. The proportion of browsed small (<20 cm tall) oak seedlings and other seedlings was low, and apparently a minor mortality factor. For saplings (20–130 cm tall), accumulated browsing damage was generally higher on oak than on five major competing tree species: Fraxinus excelsior, Corylus avellana, Tilia cordata, Acer platanoides and Sorbus aucuparia. Leaf removal was rare in late summer, except for rowan. The amount of cover (shelter) for ungulates near plots was positively correlated with oak browsing intensity; within plots, a high density of ash saplings may reduce browsing on oak saplings. In these forests, browsing probably retards growth of oak saplings relative to competing trees. Oak may persist as a minor stand component, but monitoring is needed to study future changes.     

Gap-, stand-, and landscape-scale factors contribute to poor sugar maple regeneration after timber harvest

Natural regeneration in canopy gaps is a key process affecting long-term dynamics of many forests, including northern hardwood forests. The density and composition of regenerating trees are often highly variable, reflecting sensitivity to a suite of driving factors operating at different scales (e.g., harvest gap to regional landscape), including production of seeds, physical characteristics of gaps and stands, competition with non-tree vegetation, and browsing by animals. Multivariate analyses over broad geographic areas provide insights into the relative effects of these factors and permit exploration of spatial patterns in regeneration. We examined the effects of gap-, stand-, and landscape-scale factors on densities of tree seedlings (<1 m tall) and saplings (1-2 m tall) in 59 selection-harvested northern hardwood stands located across a 4500 km² region of Michigan's Upper Peninsula. We used Bayesian multilevel modeling to account for the hierarchical structure of the data and assess uncertainty in parameter estimates. Sugar maple (Acer saccharum) saplings were absent from 61% of 154 m² plots centered in harvest gaps (n = 347) despite its high shade tolerance and overstory dominance, but densities were high in other gaps. Densities of sugar maple seedlings and/or saplings were negatively associated with a combination of greater stand-scale densities of white-tailed deer (Odocoileus virginianus), greater gap-scale cover of non-tree vegetation, and lower gap-scale light availability, with deer density having the greatest effect. Densities of unpalatable and commercially less valuable ironwood (Ostrya virginiana), the second most common regeneration species, were positively related to gap-scale seed-production potential but were unrelated to factors affecting sugar maple. Ironwood tended to replace sugar maple saplings in areas with high deer density. At the landscape scale, densities of sugar maple seedlings and saplings decreased with decreasing latitude and snow depth and increasing winter deer densities. These inverse spatial patterns suggest that deer herbivory can lead to landscape-scale variation in regeneration success. However, the spatial distribution of habitat types (a proxy for soil moisture and nutrient conditions) confound this observation, with higher densities of sugar maple generally located on stands with less nutrient-rich habitat types. Results demonstrate that combinations of factors operating at different scales, and with different relative magnitudes of impact, contribute to high variation in regeneration composition and density following timber harvest. Selection silvicultural practices, as currently applied, do not ensure regeneration of desirable species; practices might require modifications in general (e.g., increasing gap size) and to match them to regionally varying factors like deer density.     

Sustainability of wild pistachio (Pistacia atlantica Desf.) in Zagros forests,Iran

Wild pistachio (Pistacia atlantica Desf.) is the most economically important tree species in many rural areas in the west of Iran. The species produces resin used for a wide variety of traditional uses. Because the resin can be harvested non-destructively, the trees are maintained until mortality occurs from natural causes. The result is that natural, managed stands include a variety of age classes. In recent years, a lack of smaller size classes has been observed in the Qalajeh forest, which is located in the Zagros Mountain region of western Iran. We established a series of plots in an area typical of Qalajeh forest to characterize the diameter distribution of the wild pistachio component. We confirmed a deficit of stems <30 cm dbh, based in the expectation that the landscape-level diameter distribution should be characterized by a negative exponential curve. For trees ≥30 cm dbh, de Liocourt's equation closely fit the diameter distribution (r² = 0.93), translating to a q-factor of 1.34. We used this curve to estimate the deficit number of stems in diameter classes <30 cm. We estimate that this forest should have 19–24 wild pistachio trees/ha in the 5–25 cm classes, as compared to about 5 trees/ha found currently. Based on local conditions, we recommend that at least 30 seedlings/ha should be planted to allow 6–8 trees to reach to the 5 cm class.     

How much variation in tree mortality is predicted by soil and topography in Central Amazonia?

Tree mortality in Amazonia has been related to regional variation in soil, topography and climatic disturbances, but the magnitude of the effect of these factors on tree mortality at local and mesoscales remains poorly determined. We investigated tree mortality in 72 1-ha permanent plots spanning 64 km² of tropical moist forest in Reserva Ducke, Manaus, Brazil. Plots were censused three times (2000-2003, 2003-2005, and 2005-2008), resulting in two census intervals. The relationships of soil and topography to tree mortality were dependent on tree size. Small- and medium-sized trees (1 ? dbh < 30 cm) had similar relationships of mortality with soil and topography, while large trees (dbh ? 30 cm) showed different (or no) relationships. The effects of soil and topography on tree mortality also varied temporally. In the second census interval after storms, soil and topography explained about one-fourth of the spatial variation in mortality of small- and medium-sized trees (<30 cm dbh), whereas no effects were detected in the first census interval. In particular, soil fertility was the most important predictor of tree mortality in the study area. Topography alone (altitude and slope) was associated with only 12% of the spatial variation in tree mortality and the magnitude of the effect of soil and topography on tree mortality also increased after storms. In general, plots on more fertile soils, on steep slopes and sandy soils in valleys showed greater tree mortality than those on plateau with well-drained clayey soils. Therefore, disturbance history and tree size should be included when scaling up tree mortality from local to regional scales. As much variation remains unexplained, other landscape features, such as watershed morphology and wind exposure, may be necessary to make more precise predictions on patterns of tree mortality in Central Amazonia.     

Relationship between annual rainfall and tree mortality in a tropical dry forest: Results of a 19-year study at Mudumalai,southern India

Variability in rainfall is known to be a major influence on the dynamics of tropical forests, especially rates and patterns of tree mortality. In tropical dry forests a number of contributing factors to tree mortality, including dry season fire and herbivory by large herbivorous mammals, could be related to rainfall patterns, while loss of water potential in trees during the dry season or a wet season drought could also result in enhanced rates of death. While tree mortality as influenced by severe drought has been examined in tropical wet forests there is insufficient understanding of this process in tropical dry forests. We examined these causal factors in relation to inter-annual differences in rainfall in causing tree mortality within a 50-ha Forest Dynamics Plot located in the tropical dry deciduous forests of Mudumalai, southern India, that has been monitored annually since 1988. Over a 19-year period (1988–2007) mean annual mortality rate of all stems >1 cm dbh was 6.9 ± 4.6% (range = 1.5–17.5%); mortality rates broadly declined from the smaller to the larger size classes with the rates in stems >30 cm dbh being among the lowest recorded in tropical forest globally. Fire was the main agent of mortality in stems 1–5 cm dbh, elephant-herbivory in stems 5–10 cm dbh, and other natural causes in stems >10 cm dbh. Elephant-related mortality did not show any relationship to rainfall. On the other hand, fire-related mortality was significantly negatively correlated to quantity of rainfall during the preceding year. Mortality due to other causes in the larger stem sizes was significantly negatively correlated to rainfall with a 2–3-year lag, suggesting that water deficit from mild or prolonged drought enhanced the risk of death but only with a time lag that was greater than similar lags in tree mortality observed in other forest types. In this respect, tropical dry forests growing in regions of high rainfall variability may have evolved greater resistance to rainfall deficit as compared to tropical moist or temperate forests but are still vulnerable to drought-related mortality.     

Forest edge structure as a shaping factor of understorey vegetation in urban forests in Finland

We investigated the effects of edge structure (i.e. side-canopy openness based on tree, sapling and shrub characteristics, and the composition of tree species) on the understorey vegetation at mesic urban conifer-dominated forest edges in southern Finland. Forest edge structure had an effect on understorey vegetation, and on the spatial extent of the edge effect into the forests. At open edges the edge effect (in terms of the abundances of understorey vegetation) penetrated at least up to 30 m into the forest patches whereas closed edges may prevent these effects. A multilayered canopy with saplings and shrubs at the edge is important to alleviate the effects of the edge. We found that 225–250 m³ ha⁻¹ of trees (diameter at breast height (dbh) > 5 cm) is adequate to restrict the edge effect near the edge. However, the number of broad-leaved trees may be high at edges which, in turn, diminishes the abundance of mosses and favours herb species, thus changing the original natural understorey vegetation composition. Therefore we recommend that conifers be favoured at the edges of mesic conifer-dominated forest patches if the purpose is to restrict the extent of the effects of habitat edges. The appropriate proportion of conifers at these edges should be 80% or more.     

Population status and gender imbalance of the marula tree, Sclerocarya birrea subsp. caffra in northern Namibia

The population status of many wild fruit trees that support rural people in Africa remains poorly understood despite its importance for their management. Here, we establish a baseline for size class distribution and gender ratios for marula (Sclerocarya birrea subsp. caffra), a widespread but sparsely distributed species of wooded farmland, that has both traditional importance to rural communities and an emerging commercial potential. A population of marula trees around two neighbouring villages in northern Namibia was surveyed in August 2001. The stem diameter at 1.3 m height (dbh) of all individuals ge1 cm dbh in 40 fields, totalling 286 ha in area, was measured and their gender was recorded as female, male or unknown. Over 400 trees were enumerated, the largest >100 cm dbh. Despite the low threshold (1 cm dbh) for inclusion, relatively few trees (around 40%) were ¡20 cm dbh, suggesting the population may not be self-sustaining. The area was nevertheless well-stocked, with ca 1 tree ha^?1 ge20 cm dbh – an unusually high density for Sclerocarya. Tree gender becomes evident when individuals reach about 15 cm dbh. In one village, the sex ratio of larger trees (40–80 cm dbh) was significantly skewed in favour of females. This was attributed to selective elimination of individuals that have failed to fruit and was much less pronounced in the second village. The sparse distribution and gender imbalance of trees have implications in relation to management requirements for ensuring adequate pollen flow.     

Growth and survival of termite-piped Eucalyptus tetrodonta and E. miniata in northern Australia: Implications for harvest of trees for didgeridoos

The most common canopy trees in the savannas of northern Australia, Eucalyptus tetrodonta and E. miniata are also two of the most common species harvested to make didgeridoos, the traditional musical instrument of northern Australian Aboriginal peoples now experiencing high demand from international markets. Most of the trees of the area naturally have hollow cores, or pipes, due to termite activity, but little is known of the relationships of the cores to size of tree, tree growth or survival. In a wooded savanna of northern Australia, 267 individual trees with known growth and survival rates were cored to determine degree of termite-piping. Generalized linear modelling and multi-model inference showed that frequency of piping increased with diameter (dbh) tree for E. tetrodonta, but >85% of E. miniata trees were piped regardless of dbh. Growth (dbh increment) and survival (4-year) were size-dependent. Survival of both species decreased strongly with degree of piping (pipe ratio). For any given diameter, the growth rate of E. miniata trees was independent of pipe ratio, but for E. tetrodonta trees decreased strongly with pipe ratio. From modelled data, a 10-cm tree with pipe ratio of 0.60 was very vulnerable, growing at 0.0 cm year⁻¹ with 46% survival rate, whereas a 40-cm tree, even with large pipe ratios (0.80), grew 0.05 cm year⁻¹ with 98% survival rate. Traditional methods of tree harvesting remove only those smaller hollow trees that are already suffering low growth rates and are likely to die before reaching maturity, whereas current large-scale commercial methods also remove trees with higher growth and survival rates—those trees most likely to contribute to sustainable tree populations. Incorporating traditional selection and harvest methods into current commercial operations would help ensure longevity of this source of livelihood for indigenous peoples of the region.