Complex Forest Structure Research Articles

Can Douglas fir (Pseudotsuga menziesii (Mirb.) Franco) sustainably grow in complex forest structures?

The sustainability of shade-intermediate tree species such as Douglas fir in complex forest structures (with more than two distinct storeys) has been controversially discussed for a long time and traditionally shade-tolerant tree species have been preferred in plenter forest management. The most critical question in this context is whether recruitment trees of a particular species can survive in the shady conditions of the understorey and whether they can contribute to a demographic continuity. Using data from the unique, replicated monitoring site Artist’s Wood in North Wales (UK) we investigated the recruitment dynamics of Douglas fir to address the question whether this species can grow in complex forest structures on a long-term basis. Then we compared the results with those from Norway spruce and silver fir plenter forests in the Swiss Jura. Demographic continuity in complex structures is ensured if all forest development stages are represented in appropriate demographic proportions. We quantified these development stages in terms of stem diameter classes and analysed the corresponding diameter growth, competition and the loss of trees due to mortality and forest management. Based on this analysis we determined demographic equilibrium conditions for Artist’s Wood. The results confirmed that Douglas fir at Artist’s Wood can indeed grow sustainably in complex forest structures at a stand density of 27.4m2 ha−1 basal area which corresponds to a standing volume of 342m3 ha−1. However, equilibrium conditions for Douglas fir can only be achieved at the expense of a lower overall stand density of approximately 85% of that of Norway spruce. Despite this the corresponding Douglas fir stand volume increment of 15.3m3 ha−1 year−1 is greater than that of Norway spruce. Thus Douglas fir plenter forests are not only possible but are also appealing considering the broad site amplitude of the species in conjunction with ongoing climate change and the potential of the species for CO2 retention.

Can gullies preserve complex forest structure in frequently burnt landscapes?

Structurally complex forest provides important habitat for a diverse array of vertebrate fauna. Frequent fire can simplify forest structure, though topographic mitigation of fire effects could potentially preserve structurally more complex habitat within certain topographic locations of fire prone landscapes. Our study assessed whether the effects of fire frequency on forest structure (tree hollows, log volume, vegetation complexity) varied with topographic position. The effect of wildfire severity and intensive logging were also examined. Frequent fire reduced vegetation cover on ridges, but not in gullies. The risk of collapse (i.e. presence of fire scars) increased for large trees on frequently burnt ridges, but remained suppressed in gullies. Crown fire reduced hollow presence in dead trees (i.e. snags), but increased hollow presence in trees with a healthy crown. The volume of extensively decomposed logs was three times greater in gullies than ridges, but was unaffected by fire frequency. Intensive logging reduced the number of hollow bearing trees and increased the volume of extensively decomposed logs, though future declines in log volume are predicted due to bottlenecks in log input. Our results suggest that gullies may play a critical role in preserving structurally complex stands within frequently burnt temperate eucalypt forests. Protecting gullies from land clearing and intensive logging is likely to be an important step in maintaining key habitat features and associated fauna in these landscapes.

Changes in ant communities along an age gradient of cocoa cultivation in the Oumé region, central Côte d'Ivoire

AbstractWe identified the extent to which ant diversity occurs despite conversion of forests into cocoa plantations by examining the communities across four age classes of plantations (classes I–IV with increasing age from 0–5 to 21–40 years) and in their original forests. An extensive sampling protocol consisting of pitfall trapping, leaf litter sampling, soil sampling and hand sampling was used to characterize ant species richness and composition in three replicates of each age class and in the remaining forest patches. A total of one hundred ant species was found in all habitats combined. While the forest was the richest habitat (73 species), species richness in the different plantation age classes varied as follows (sorted in descending order): class IV (69 species) > class III (57 species) > class I (52 species) > class II (43 species). Age gradient was thus significantly positively correlated with mean species richness and with the relative abundance of some subfamilies. The species composition differed greatly between some plantation age classes and the forest. The two youngest cocoa age classes (I and II) were most dissimilar to the forest. In contrast, forest ants were well represented in the old cocoa age classes (III and IV). Three functional guilds (generalist predators, specialist predators and territorially dominant arboreal species) were in their relative abundance significantly correlated to the age gradient. Overall, cocoa cultivations retaining a floristically diverse and structurally complex forest structure are a suitable management system for the conservation of ant species of the formerly forested habitats.

Positive Relationship between Aboveground Carbon Stocks and Structural Diversity in Spruce-Dominated Forest Stands in New Brunswick, Canada

Abstract Maintaining both the structure and functionality of forest ecosystems is a primary goal of forest management. In this study, relationships between structural diversity and aboveground stand carbon (C) stocks were examined in spruce-dominated forests in New Brunswick, Canada. Tree species, size, and height diversity indices as well as a combination of these diversity indices were used to correlate aboveground C stocks. Multiple linear regressions were subsequently used to quantify the relationships between these indices and aboveground C stocks, and partial correlation analysis was also adopted to remove the effects of other explanatory variables. Results show that stand structural diversity has a significant positive effect on aboveground C stocks even though the relationship is weak overall. Positive relationships observed between the diversity indices and aboveground C stocks support the hypothesis that increased structural diversity enhances aboveground C storage capacity. This occurs because complex forest structures allow for greater light infiltration and promote a more efficient resource use by trees, leading to an increase in biomass and C production. Mixed tolerant species composition and uneven-aged stand management in conjunction with selection or partial cutting to maintain high structural diversity is therefore recommended to preserve biodiversity and C stocks in spruce-dominated forests.

Restoration of semi-natural forest after clearcutting of conifer plantations in Japan

We reviewed recent studies on the restoration of semi-natural forests after clearcutting of conifer plantations with specific reference to the importance of setting restoration goals, general and specific factors influencing restoration, and the prediction and judgment of successful restoration. For the restoration of semi-natural forests following the clearcutting of conifer plantations, recovery levels and appropriate restoration methods cannot be considered separately from the aims of restoration to provide ecosystem services. Restoration needs to be based on the setting of goals that correspond to the type and degree of ecosystem services targeted. When we aim to restore vegetation dominated by late-seral canopy trees within as short a period after the clearcutting of conifer plantations as possible, advanced regeneration in the pre-logged plantation is the most important and reliable source of post-logging regeneration, just as it is in the recovery of disturbed non-plantation forests. Advanced regeneration is not always present in plantations. Seedling establishment by seed rain from adjacent natural forest patches after a disturbance can contribute to late-seral or canopy species; however, the range of the edge effect is generally short, and the degree of the edge effect depends on the natural forest type and is more pronounced in deciduous forests, such as those in cool-temperate climates, or in forests with a complex forest structure. Thus, advanced regeneration is a better predictor of short-term forest recovery than measures based on seed rain from adjacent seed sources. However, restoration achieved only through advanced regeneration can simplify the species composition, and the recolonization processes of gradual species accession from seeds should be considered in the long-term context of restoration.

Restoring a rainforest habitat linkage in north Queensland: Donaghy’s Corridor

Summary Donaghy’s Corridor is a 1.2 km × 100 m planting of rain forest species on the Atherton Tableland, Queensland, designed to link an isolated fragment (498 ha) to adjacent continuous forest (80 000 ha). Vegetation and fauna monitoring commenced immediately after the linkage was completed. Vegetation surveys showed 119 plant species established in the linkage in 3 years, and 35 of these were not known to occur within the extant linkage either as planted stock or as natural individuals existing prior to project commencement. There were differences between the fauna trapped within the restoration, adjacent open pasture habitats, forest interior sites and forest edge sites. Differences likely reflect variation in species habitat preferences and the habitat suitability of the planted vegetation. Now over 10 years old, Donaghy’s Corridor has developed a complex forest structure, with the tallest planted stems exceeding 20 m in height. This feature article provides information about the planning, implementation and monitoring of the linkage, and shows how restoring landscape and ecological connectivity can be a locally effective strategy to counter forest fragmentation.

Nitrogen dynamics across silvicultural canopy gaps in young forests of western Oregon

Silvicultural canopy gaps are emerging as an alternative management tool to accelerate development of complex forest structure in young, even-aged forests of the Pacific Northwest. The effect of gap creation on available nitrogen (N) is of concern to managers because N is often a limiting nutrient in Pacific Northwest forests. We investigated patterns of N availability in the forest floor and upper mineral soil (0–10 cm) across 6–8-year-old silvicultural canopy gaps in three 50–70-year-old Douglas-fir forests spanning a wide range of soil N capital in the Coast Range and Cascade Mountains of western Oregon. We used extractable ammonium (NH 4 +) and nitrate (NO 3 −) pools, net N mineralization and nitrification rates, and NH 4 + and NO 3 − ion exchange resin (IER) concentrations to quantify N availability along north-south transects run through the centers of 0.4 and 0.1 ha gaps. In addition, we measured several factors known to influence N availability, including litterfall, moisture, temperature, and decomposition rates. In general, gap-forest differences in N availability were more pronounced in the mineral soil than in the forest floor. Mineral soil extractable NH 4 + and NO 3 − pools, net N mineralization and nitrification rates, and NH 4 + and NO 3 − IER concentrations were all significantly elevated in gaps relative to adjacent forest, and in several cases exhibited significantly greater spatial variability in gaps than forest. Nitrogen availability along the edges of gaps more often resembled levels in the adjacent forest than in gap centers. For the majority of response variables, there were no significant differences between northern and southern transect positions, nor between 0.4 and 0.1 ha gaps. Forest floor and mineral soil gravimetric percent moisture and temperature showed few differences along transects, while litterfall carbon (C) inputs and litterfall C:N ratios in gaps were significantly lower than in the adjacent forest. Reciprocal transfer incubations of mineral soil samples between gap and forest positions revealed that soil originating from gaps had greater net nitrification rates than forest samples, regardless of incubation environment. Overall, our results suggest that increased N availability in 6–8-year-old silvicultural gaps in young western Oregon forests may be due more to the quality and quantity of litterfall inputs resulting from early-seral species colonizing gaps than by changes in temperature and moisture conditions caused by gap creation.

Disturbance history and stand dynamics in tall open forest and riparian rainforest in the Central Highlands of Victoria

Abstract Spatial heterogeneity in the intensity of past disturbances has directly influenced the structure and composition of present‐day forests around the world. In south‐eastern Australia infrequent, high‐intensity wildfires are a major part of the historical disturbance regime. While these fires are often assumed to produce even‐aged stands, spatial heterogeneity in fire intensity due to highly variable topography may lead to more complex forest age structures. Our study describes the influence of disturbance on the age structure and dynamics of a mosaic of tall, open eucalypt forest, cool temperate rainforest and mixed species forest surrounding Bellel Creek in the Central Highlands of Victoria using dendrochronological techniques. We were particularly interested in the impacts of the 1939 Black Friday fire and its effects on forest age structure and subsequent stand development patterns. Within our study site tall open forest displayed two distinct age cohorts: (i) trees that established immediately after the 1939 fire and accounted for the majority of individuals in the forest, and (ii) scattered groups of older trees estimated to be approximately 200–250 years old. Cool temperate rainforest and mixed forest were also dominated by the post‐1939 fire age cohort. However, a greater proportion of trees in these forest types survived the 1939 fire relative to the tall open forest. The impact of the 1939 fire on the growth of surviving trees was highly variable but generally short‐lived. In most cases growth decreased after the 1939 fire, but generally returned to prefire levels within 1–3 years. Non‐fire disturbances were limited to small‐scale branch‐ and tree‐fall events, although the extreme snowstorm of 1977 appears to have caused extensive damage to rainforest communities. Our study demonstrates the opportunities for dendroecological studies to reconstruct historical dynamics and disturbance patterns in Australian forests and provides important insights into variation in landscape‐scale fire impacts and their effect on subsequent forest development patterns.

A review of mechanistic modelling of wind damage risk to forests

Summary This paper reviews the current status of mechanistic models for wind damage risk assessment, describing model structure, applicability, validation and current limitations. We focus particularly on the hybrid mechanistic/empirical models GALES and HWIND, which have been designed for calculating wind damage risk at the stand level within uniform forests and which are the most widely adopted models within the research community. These models have been integrated with different methods for predicting the local wind climate in order to calculate the probability of wind damage in a number of different countries. We also discuss ongoing modelling work and proposals for future development in order to deal with complex forest structures and to predict the wind damage risk of individual trees within stands through the integration of mechanistic risk models with forest growth and yield models within a geographical information system framework. This kind of model integration will enable spatial representation of tree lists and damage propagation and allow managers to evaluate the effect of different harvesting and thinning scenarios on the risk of windthrow of both stands and individual trees within a stand.

A Comparative Study of Landsat TM and SPOT HRG Images for Vegetation Classification in the Brazilian Amazon

Complex forest structure and abundant tree species in the moist tropical regions often cause difficulties in classifying vegetation classes with remotely sensed data. This paper explores improvement in vegetation classification accuracies through a comparative study of different image combinations based on the integration of Landsat Thematic Mapper (TM) and SPOT High Resolution Geometric (HRG) instrument data, as well as the combination of spectral signatures and textures. A maximum likelihood classifier was used to classify the different image combinations into thematic maps. This research indicated that data fusion based on HRG multispectral and panchromatic data slightly improved vegetation classification accuracies: a 3.1 to 4.6 percent increase in the kappa coefficient compared with the classification results based on original HRG or TM multispectral images. A combination of HRG spectral signatures and two textural images improved the kappa coefficient by 6.3 percent compared with pure HRG multispectral images. The textural images based on entropy or second-moment texture measures with a window size of 9 pixels × 9 pixels played an important role in improving vegetation classification accuracy. Overall, optical remote-sensing data are still insufficient for accurate vegetation classifications in the Amazon basin.

Using air photos to interpret quality of Marbled Murrelet nesting habitat in south coastal British Columbia

Reliable habitat assessment methods are needed to ensure the adequate management of Marbled Murrelet (Brachyramphus marmoratus) habitat in British Columbia. In two south coastal study regions, the Sunshine Coast and Clayoquot Sound, we evaluated the effectiveness of a qualitative habitat classification that uses air photo-interpreted forest structural characteristics for identifying and ranking habitat quality. Using a sample of 118 nest sites and 157 random sites within forests greater than 140 years old, we found that murrelets selected nest patches non-randomly with respect to forest characteristics. While selectivity varied between study regions, generally nest patches had taller and larger trees, exhibited more complex forest structure, and were located at lower meso-slope positions near large gaps or nearby edges. In addition, these patches were more often ranked higher in terms of habitat quality. However, we found that probable breeding success was greater in habitats classified as lower quality. Thus, further research is needed to understand our findings relative to other influences on breeding productivity, such as predators and hierarchal habitat selection. In summary, while our study supports the use of the current air photo habitat classification standards to improve identification and selection of murrelet nesting habitat for management, some modifications to these standards may be needed.

Impact of wind direction on diurnal and seasonal changes in wind profiles

We explored the relationship between directional variation (changes in direction from a reference point) in vegetation and wind profiles, and propose an empirical wind profile model that may reproduce the wind profile within the canopy (such as secondary wind maximum) and reduce calculation loads. Based on the results of our observations in secondary broad-leaved forest, we clarified the variation in secondary wind maximum and aerodynamic parameters in wind direction, and assessed the influence of forest structural heterogeneity on the wind profile. Wind blew from specific directions depending on the time of day and season, and secondary wind maxima were observed in particular wind directions. Outlier estimations of aerodynamic parameters were determined for wind directions of 45°, 120–210°, and 300°, and these did not show a logarithmic distribution. Our proposed model reproduced the wind profile within and above the canopy, closely resembling K-theory and closure models, and reduced the required number of calculations and observations. The influence of forest structural heterogeneity was greatest in the upper part of the canopy (9.14 m), and reproducibility of the wind profile was improved by 0.05 m/s (3.4% on average) by changing the forest structural parameters of the model according to the wind direction. It was clear that forest structural heterogeneity did not cause critical errors in estimation of the wind profile, even at our study site, which had complex forest structure.

Preface to special section on New Approaches to Quantifying Exchanges of Carbon and Energy Across a Range of Scales

[1] Estimating carbon and energy exchanges in the soilplant-atmosphere continuum across spatial and temporal scales is a primary focus of coordinated interagency collaborations (i.e., U.S. Global Change Research Program, Climate Change Science Program, and the Carbon Cycle InterAgency Working Group) that are outlined in North American Carbon Plan [Wofsy and Harriss, 2002], Carbon Cycle Science Plan [Sarmiento and Wofsy, 1999], Intergovernmental Panel on Global Climate Change [Intergovernmental Panel on Global Climate Change, 1995] and other position papers. Eddy covariance (EC) has served well to define these exchanges over scales of 10–100 ha in relatively simple terrain under well-mixed conditions. These data have contributed to significant advances in our understanding of important ecosystem processes and quantifying the key biotic and abiotic processes that control these rates [Loescher et al., 2003, 2005; Law et al., 2002, 2003; Anthoni et al., 1999, 2002; Bowling et al., 2001; Clark et al., 1999; Goulden et al., 1996], and useful in developing regional approaches to constraining the global CO2 budget [Townsend et al., 2002; Battle et al., 2000]. However, extension to complex terrain and measurement under nonideal meteorological conditions requires new approaches. New approaches are also needed to increase our ability to scale these exchanges. The focus of this thematic issue is to quantify uncertainty in energy and carbon fluxes, and introduce new approaches for extending the measurements to larger scales. [2] This thematic issue presents a suite papers covering a wide range of topics including estimation of uncertainties in flux measurements, examination of problems associated with nighttime fluxes, extraction of physiological parameters from flux measurements, methodological issues with measurements, use of isotopic measurements, and regionalization. The section begins with Loescher et al. [2006] outlining the uncertainties associated with EC measurements (e.g., random and systematic errors, gap filling, and flows not accounted for), how these errors scale with time and space, and how these errors compare to the errors associated with traditional estimates of ecosystem respiration and net primary productivity. [3] Estimating the uncertainty in EC data can be challenging because the variability in 30-min EC estimates increase with the magnitude of the estimate, i.e., heteroscedastic quantities. Moreover, it is now a criterion to provide an estimate of uncertainty in EC estimates when reporting data from any (and all) AmeriFlux site. Hagen et al. [2006] provide a robust methodology to accomplish these goals. By using bootstrapping and an artificial neural network, these authors found by that the uncertainty in EC was 100% for 30-min estimates, but was reduced with longer temporal scales (annual estimates) to 10%. Their research site was micrometeorologically ideal (average and uniform surface roughness, flat topography), so other unaccounted flows did not likely contribute to this study. [4] The combination of thermally stratified, decoupled nocturnal boundary layers and structurally complex forest structures make nighttime EC estimates difficult to estimate and create the conditions when drainage of CO2 below the height of the measured turbulent flux can occur. Goulden et al. [2006] used remotely sensed data to examine potential nighttime CO2 drainage patterns in an Amazon forest and surrounding landscape, and identify when drainage flows develop and become significant. Juang et al. [2006] provide a promising inverse approach (Eulerian rather than Langrangian) using below-canopy air temperature to model ecosystem respiration. Lai et al. [2006] use stable carbon and oxygen isotopes of respired CO2 to partition the abiotic controls on ecosystem-level uptake and respiration from a C3/C4 grassland. [5] Using eddy covariance data to estimate annual integrals is hampered by incomplete data sets due to enviable data gaps due to calibrations, instrument failure, precipitation, and power problems. Gap filling strategies have included look-up tables, interpolations or estimates of uptake and respiration (e.g., those based on Ball-Berry and Q10 equations). Gove and Hollinger [2006] present an innovative methodology to gap fill data that relies on the heteroscedastic nature of EC-derived 30-min, and optimizes parameters used in uptake and respiration equations. [6] Wolf et al. [2006] invert a model and constrain the governing equations by measured EC data to estimate key ecosystem parameters from uncertainties associated with light use efficiency models that primarily stem from the JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 111, D14S91, doi:10.1029/2006JD007135, 2006

Groundwater use by dominant tree species in tropical remnant vegetation communities

Defining groundwater dependence and water-use requirements of terrestrial vegetation represents a significant challenge to water-resources managers. Terrestrial vegetation may exhibit complex spatial and temporal patterns of groundwater dependence. In this study we have assessed the sources of water used by dominant tree species in remnant vegetation of Pioneer Valley, Mackay, in northern Queensland. Water use by tree species was determined by sapflow techniques and the sources of water were investigated by using a combination of isotopic and water-potential measurements. Within the remnant vegetation communities of the Pioneer Valley there were complex patterns of water use and water-resource partitioning. However, all communities within the study showed some degree of groundwater use. Riparian communities that were reliant on groundwater discharge for maintenance of river baseflow exhibited high species diversity and complex forest structure and different species within these communities accessed a range of water sources including shallow soil water, river water and groundwater. In contrast, the woodlands and open forest were principally reliant on soil water. Although, species such as Corymbia clarksoniana appeared to be reliant on groundwater for their dry-season water-use requirements. This study demonstrated use of groundwater by remnant vegetation communities in the Pioneer Valley but determination of groundwater dependence requires a better understanding of the temporal patterns of water use and sources of water used by each species.

Inversion of a Lidar Waveform Model for Forest Biophysical Parameter Estimation

Due to its measurement principle, light detection and ranging (lidar) is particularly suited to estimate the horizontal as well as vertical distribution of forest structure. Quantification and characterization of forest structure is important for the understanding of the forest ecosystem functioning and, moreover, will help to assess carbon sequestration within forests. The relationship between the signal recorded by a lidar system and the canopy structure of a forest can be accurately characterized by physically based radiative transfer models (RTMs). A three-dimensional RTM is capable of representing the complex forest canopy structure as well as the involved physical processes of the lidar pulse interactions with the vegetation. Consequently, the inversion of such an RTM presents a novel concept to retrieve biophysical forest parameters that exploits the full lidar signal and underlying physical processes. A synthetic dataset and data acquired in the Swiss National Park (SNP) successfully demonstrated the feasibility and the potential of RTM inversion to retrieve forest structure from large-footprint lidar waveform data. The SNP lidar data consist of waveforms generated from the aggregation of small-footprint lidar returns. Derived forest biophysical parameters, such as fractional cover, leaf area index, maximum tree height, and the vertical crown extension, were able to describe the horizontal and vertical forest canopy structure.

Radar backscattering model for multilayer mixed-species forests

A multilayer canopy scattering model is developed for mixed-species forests. The multilayer model provides a significantly enhanced representation of actual complex forest structures compared to the conventional canopy-trunk layer models. Multilayer Michigan Microwave Canopy Scattering model (Multi-MIMICS) allows overlapping layer configuration and a tapered trunk model applicable to forests of mixed species and/or mixed growth stages. The model is the first-order solution to a set of radiative transfer equations and includes layer interactions between overlapping layers. It simulates SAR backscattering coefficients based on input dimensional, geometrical, and dielectric variables of forest canopies. The Multi-MIMICS is an efficient realization of actual forest structures and can be shaped for specific interest of forest parameters. We present the model's application and validation in the paper. The model is parameterized using data collected from a 220,000-ha area of forests in central Queensland, Australia. Fifteen 50/spl times/50 m test sites representing the general forest diversity and growth stages are chosen as ground truth. Polarimetric backscattering airborne SAR (AIRSAR) data of the same area are acquired to validate the model simulations. The model predicts SAR backscattering coefficients of the test areas. Simulation results show a good agreement with AIRSAR data at most frequencies and polarizations. The simulated backscattering coefficient from the multilayer model and the standard MIMICS are also compared and significant improvements are observed.

An analysis of spatial forest structure using neighbourhood-based variables

The study presents an analysis of forest spatial structure and diversity in the Federal State of Durango where the majority of the forests consist of pure pine stands or pine mixed with oak. Natural forests of greater diversity and of high ecological significance are found only in a few isolated localities in the Santa Bárbara valley. These forests, with rare conifers including the genera Picea, Abies and Pseudotsuga are found on particular sheltered, humid sites. For one such rare site, a detailed analysis of forest spatial structure was made, based on three one-quarter hectare plots where all the trees and their coordinates had been assessed. The objective of the study was to provide a quantitative description of the spatial structure of the plots, using new parameters of spatial diversity and to present a method for comparative analysis of the three forest sites. The analysis is using a new approach for describing complex forest structures in a straightforward manner. To evaluate the spatial attributes, it is not necessary to measure distances between trees or to establish tree coordinates. The spatial characteristics can be established merely on the basis of evaluating the immediate neighbourhood of a given number of reference trees. The variables describe the distributions of spatial mingling, size differentiation and contagion, which can be easily interpreted allowing quantitative comparisons between complex forest structures.

Stand delineation and composition estimation using semi-automated individual tree crown analysis

Stand delineation and species composition estimation are cornerstones of forest inventory mapping and key elements to forest management decision making. Improved mapping techniques are constantly being sought in terms of speed, consistency, accuracy, level of detail, and overall effectiveness. Semi-automated analysis of high-resolution imagery at the individual tree crown level may offer such benefits. Methods, however, need to be developed and tested under a variety of forest conditions. High-resolution (60 cm) multispectral airborne imagery was acquired over a predominantly young conifer forest and plantation test area on the west coast of Canada. Automated tree isolation algorithms were applied to the data in order to delineate tree crowns or clusters of crowns. An object-oriented single tree classification was conducted using a maximum likelihood classifier. Stands of similar species composition, closure, and stem density were defined through a sequence that first generated images of these parameters from the automated delineation and classification, used these as input to an unsupervised classification, and then filtered and smoothed the resulting classification clusters. Because of the dense nature of the stands and small crowns on the site, the isolation process often delineated clusters of several trees. Species classification accuracy was determined by comparing the average stand composition from the automated technique to that derived from ground transects or plots. Species classification was good, with average composition error (difference between field measured and automated composition) over all 16 test stands being 7.25%. Most errors for individual species in stands were below 20%, but a few were up to 30%. The automatically generated stand boundaries mimicked well those of known plantation and interpreted inventory boundaries. The automated technique created a few larger stands and some additional small stands in areas of complex forest structure. Overall, for the young fairly uniform stands of the site, both stand delineation and species composition estimation were of a quality suitable for operational use in inventory and forest management. Further development and testing is needed to extend results to situations covering large areas, multiple flight lines, varied topography, and different forest conditions.

Development of oak plantations established for wildlife

Extensive areas that are currently in agricultural production within the Mississippi Alluvial Valley are being restored to bottomland hardwood forests. Oaks ( Quercus sp.), sown as seeds (acorns) or planted as seedlings, are the predominant trees established on most afforested sites. To compare stand development and natural invasion on sites afforested by planting seedlings or by sowing acorns, we sampled woody vegetation on ten 14- to 18-year-old oak plantations established to provide wildlife habitat. Stem densities of about 900 oaks/ha were comparable between stands established by sowing 4000 acorns/ha and stands established by planting 900 seedlings/ha. Densities of oaks in stands established from seedlings increased 38% from densities detected when these stands were 4- to 8-year-old. Densities of oaks established from field-sown acorns increased >100% during this same 10-year span. Oaks that were planted as seedlings were larger than those established from acorns, but trees resulting from either afforestation method were larger than trees naturally colonizing these sites. Natural invasion of woody species varied greatly among afforested sites, but was greater and more diverse on sites sown with acorns. Afforested stands were dominated by planted species, whereas naturally invading species were rare among dominant canopy trees. When afforestation objectives are primarily to provide wildlife habitat, we recommend, sowing acorns rather than planting seedlings. Additionally, planting fewer seeds or seedlings, diversifying the species planted, and leaving non-planted gaps will increase diversity of woody species and promote a more complex forest structure that enhances the suitability of afforested sites for wildlife.

Approaches to quantifying forest structures

Summary For some time, structure indices ‐ quantifying spatial stand structure ‐ have been integrated into forest research and are used to provide a measure of biodiversity. In addition, correlation functions ‐ developed initially for problems outside forestry ‐ enable analysis and characterization of forest stand structures, generating more accessible information. This paper outlines a classification of structural indices measuring alpha diversity and examines typical representatives of the classification groups such as the Shannon index, the aggregation index of Clark and Evans, the contagion index, the coefficient of segregation of Pielou, the mingling index, the diameter differentiation index, the pair correlation and the mark correlation function. These can be used to measure differences between forests in time and space, to generate forest structures, to analyse the differences between observed and expected structures and to characterize modifications of forest structure resulting from selective harvesting. These algorithms are the keys for assessing complex forest structures, which can be the result of continuous cover forestry methods. Continuous cover forests with selective harvesting are being promoted in the new forest policies of Britain. Case studies have shown that from given spatial forest structures one can possibly conclude the suitability for habitats, a hypothesis which has yet to be proved by further appropriate analysis. The equations for the quantification of stand structure presented in this paper have the advantage that they are easier to survey during forest inventory than the more direct measures of ecological variety.