Mapping Tropical Forest Trees Using High‐Resolution Aerial Digital Photographs

The spatial arrangement of tree species is a key aspect of community ecology. Because tree species in tropical forests occur at low densities, it is logistically challenging to measure distributions across large areas. In this study, we evaluated the potential use of canopy tree crown maps, derived from high‐resolution aerial digital photographs, as a relatively simple method for measuring large‐scale tree distributions. At Barro Colorado Island, Panama, we used high‐resolution aerial digital photographs (~0.129 m/pixel) to identify tree species and map crown distributions of four target tree species. We determined crown mapping accuracy by comparing aerial and ground‐mapped distributions and tested whether the spatial characteristics of the crown maps reflect those of the ground‐mapped trees. Nearly a quarter (22%) of the common canopy species had sufficiently distinctive crowns to be good candidates for reliable mapping. The errors of commission (crowns misidentified as a target species) were relatively low, but the errors of omission (missed canopy trees of the target species) were high. Only 40 percent of canopy individuals were mapped on the air photographs. Despite failing to accurately predict exact abundances of canopy trees, crown distributions accurately reproduced the clumping patterns and spatial autocorrelation features of three of four tree species and predicted areas of high and low abundance. We discuss a range of ecological and forest management applications for which this method can be useful.     

Spatial analysis of oblique photo‐point images for quantifying spatio‐temporal changes in plant communities

Question: Can spatial analytical techniques be used to extract quantitative measurements of vegetation communities from ground‐based permanent photo‐point images? Location: Mount Aspiring National Park, south‐western South Island, New Zealand. Methods: Sets of ground‐based photographs representing two contrasting vegetation types were selected to test two spatial analytical techniques. In the grid technique, a grid was superimposed onto the photographs and the frequency of species presence in each grid‐square was calculated to estimate species abundance/cover over the defined area. In the object‐oriented technique, the photographs were segmented into meaningful objects, based on the colour of the pixels and the textural patterns of the images, and the area occupied by an object in the image was used to derive species abundance/cover over the area. Results: Both techniques allow quick and easy classification of digital elements into ecologically relevant categories of vegetation components. The grid technique appeared more robust, being quick and efficient, accommodating all image types and providing presence/absence matrices for multivariate analysis. Fewer classes were identified using the object‐oriented technique, in particular for the forest interior site and for small individual plants such as Astelia spp. Conclusions: Both techniques showed potential for the objective quantitative analysis of long‐term vegetation monitoring of cover and changes of several component species, using repeat ground‐based photographs more specifically for grassland habitats. However, both rely to various degrees on manual classification. Corrective factors and strict protocols for taking the photographs are necessary to account for variation in view angles and to compute values more representative of absolute species abundance.     

Using aerial photographs to remotely assess tree hollow availability

Tree hollows provide critical habitat for many species worldwide. The conservation of hollow-bearing trees presents a particular challenge for forest managers, partly due to difficulties in predicting their occurrence across a landscape. We trialled a novel approach for assessing relative hollow availability, by remotely estimating mature crown cover and senescence from aerial photographs in Tasmania, Australia. These estimates were tested against plot-based field assessments of actual occurrence of hollow-bearing trees. In dry forest we conducted ground-based surveys of hollows for all mature trees (>50 cm dbh) in 37 half-hectare plots. In wet forest, we conducted helicopter-based surveys of hollows for all mature trees in 45 oldgrowth plots (0.29–4.63 ha). Aerial photographs (1:10,000–1:25,000) were used to classify the senescence and cover of mature crowns in each plot. Regression analysis showed that, in dry forest, hollow-bearing tree densities were strongly related to the remote assessment of mature crown cover, with an 8% increase in variability explained if senescence was also included (R ² = 0.50). In wet forest, mature crown cover alone was the best model (R ² = 0.53 when outliers were removed). Assessing senescence was less important in dense wet forests than dry forest because trees take longer to form mature-shaped crowns and so mature-shaped crowns are more likely to have hollows. These results suggest that, with skilled photo-interpretation, aerial photographs can be useful for remotely assessing the relative density of hollow-bearing trees. This approach has the potential to greatly improve conservation planning for hollows and hollow-dependent fauna.     

Rapid expansion of lichen woodlands within the closed-crown boreal forest zone over the last 50 years caused by stand disturbances in eastern Canada

Aim Our two main goals are first to evaluate the resilience of the boreal forest according to latitude across the closed‐crown forest zone using the post‐disturbance distribution and cover of lichen woodlands and closed‐crown forests as a metric, and second to identify the disturbance factors responsible for the regeneration and degradation of the closed‐crown forest according to latitude since the 1950s. Location The study area extends between 70°00′ and 72°00′ W and throughout the closed‐crown forest zone, from its southern limit near 47°30′ N to its northern limit at the contact with the lichen woodland zone at around 52°40′ N. Methods Recent (1972–2002) and old (1954–1956) aerial photos were used to map the distribution of lichen woodlands across the closed‐crown forest zone. Forest disturbances such as fire, spruce budworm (Choristoneura fumiferana (Clemens)) outbreak, and logging were recorded on each set of aerial photos. Each lichen woodland and stand disturbance was validated by air‐borne surveys and digitized using GIS software. Results Over the last 50 years, the area occupied by lichen woodlands has increased according to latitude; that is, 9% of the area that was occupied by closed‐crown forests has shifted to lichen woodlands. Although logging activities have been concentrated in the same areas during the last 50 years, the area covered by logging has increased significantly. Outbreaks by the spruce budworm occurred predominantly in the southern (47°30′ N to 48°30′ N) and central (48°53′ N to 50°42′ N) parts of the study area, where balsam fir stands are extensive. In the northern part of the study area (51°–52°40′ N), extensive fires affected the distribution and cover of closed‐crown forests and lichen woodlands. Main conclusions Over the last 50 years, the area occupied by closed‐crown forests has decreased dramatically, and the ecological conditions that allow closed‐crown forests to establish and develop are currently less prevalent. Fire is by far the main disturbance, reducing the ability of natural closed‐crown forests to self‐regenerate whatever the latitude. Given the current biogeographical shift from dense to open forests, the northern part of the closed‐crown forest zone is in a process of dramatic change towards the dominance of northern woodlands.     

Fine‐scale forest variability and biodiversity in the boreal mixedwood forest

Local spatial variation in species distributions is driven by a mix of abiotic and biotic factors, and understanding such hierarchical variation is important for conservation of biodiversity across larger scales. We sought to understand how variation in species composition of understory vascular plants, spiders, and carabid beetles is associated with concomitant spatial variation in forest structure on a 1‐ha permanent plot in a never‐cut mixedwood forest in central Alberta (Canada). Using correlations among dendrograms produced by cluster analysis we associated data about mapped distribution of all living and dead stems > 1 cm diameter at breast height with distributions of the three focal taxa sampled from regular grids across the plot. Variation in each of these species assemblages were significantly associated with several forest structure variables at various spatial scales, but the scale of the associations varied among assemblages. Variation in species richness and abundance was explained mostly by changes in basal area of trees across the plot; however, other variables (e.g. snag density and tree density) were also important, depending on assemblage. We conclude that fine‐scale habitat variation is important in structuring spatial distribution of the species of the forest floor, even within a relatively homogeneous natural forest. Thus, assessments that ignore within‐stand heterogeneity and management that ignores its maintenance will have limited utility as conservation measures for these taxa, which are major elements of forest biodiversity.     

Crown typology and the identification of rain forest trees on large-scale aerial photographs

This paper assesses the potential of large-scale aerial photographs for the identification of rain forest trees. Colour slides at 1: 3000 scale were acquired in French Guiana, above the canopy of a 25 ha study site where trees had already been identified. Firstly, the stereoscopic analysis of photographic prints served to establish a typology of the crowns, based on seven main classes of criteria: crown size, crown status, crown contour, crown architecture, foliage cover, foliage texture and colour, completed where possible by information on phenology. The terminology chosen was based on those proposed in previous studies. Secondly, a smaller area of 5 ha was delimited in the field, on which 15 tree categories (vernacular names) represented by 5 crowns or more on the photographs were selected. For each category, a standard crown was described using terminology previously defined. Twelve tree categories, including first and second class commercial timbers, displayed specific characteristics allowing them to be identified on aerial photographs. Further analyses will be undertaken in the future to measure the success of this identification method. This method may be applied for the recognition of particular species of interest: commercial, rare, endemic or key-stone species. Phenological data, as well as information on crown architectural development, can also be retrieved from aerial photographs, bringing new prospects for a better knowledge of crowns biology and their functional role in the forest ecosystem.     

Estimation of carbon storage based on individual tree detection in <Emphasis Type="Italic">Pinus densiflora</Emphasis> stands using a fusion of aerial photography and LiDAR data

The objective of this study was to estimate the carbon storage capacity of Pinus densiflora stands using remotely sensed data by combining digital aerial photography with light detection and ranging (LiDAR) data. A digital canopy model (DCM), generated from the LiDAR data, was combined with aerial photography for segmenting crowns of individual trees. To eliminate errors in over and under-segmentation, the combined image was smoothed using a Gaussian filtering method. The processed image was then segmented into individual trees using a marker-controlled watershed segmentation method. After measuring the crown area from the segmented individual trees, the individual tree diameter at breast height (DBH) was estimated using a regression function developed from the relationship observed between the field-measured DBH and crown area. The above ground biomass of individual trees could be calculated by an image-derived DBH using a regression function developed by the Korea Forest Research Institute. The carbon storage, based on individual trees, was estimated by simple multiplication using the carbon conversion index (0.5), as suggested in guidelines from the Intergovernmental Panel on Climate Change. The mean carbon storage per individual tree was estimated and then compared with the field-measured value. This study suggested that the biomass and carbon storage in a large forest area can be effectively estimated using aerial photographs and LiDAR data.     

Spatial and temporal patterns of morel fruiting

The biotic and abiotic factors conditioning morel fruit body production are incompletely known. We examined spatial and temporal patterns of Morchella esculenta fruiting over five years in a wooded site in Missouri, USA. Fruiting onset was inversely correlated with spring air and soil temperatures, whereas abundance was positively correlated with rain events (>10 mm) during the 30 d preceding fruiting. The two years with the greatest fruiting had the shortest fruiting seasons (6–7 d). Fruiting season length was positively correlated with soil warming, suggesting that a narrow range of optimum soil temperatures favour the explosive production of fruit bodies. All woody stems of at least 1 cm diam were mapped and stem diameter and crown condition were noted. Morel fruit bodies were significantly closer to stems of Carya spp., Tilia americana and Ulmus americana than predicted by the frequencies of these woody species or their contribution to the total basal area on the site. Although intra-annual clustering of fruit bodies was often observed, inter-annual clustering was not. The spatial pattern of M. esculenta fruiting appears to be associated with vegetation pattern, whereas the onset and abundance of fruiting are determined by the interaction of spring temperatures with availability of supporting precipitation.     

INTERACTIONS BETWEEN CROWN STRUCTURE AND LIGHT ENVIRONMENT IN FIVE RAIN FOREST PIPER SPECIES

Measurements of light variation among leaves within crowns of five Piper species were compared with estimates of spatial variation in light within understory, forest edge, and clearing habitats to estimate the extent to which crown structure contributes to variation in leaf light environment. Daily photon flux density (PFD) varied greatly within and among crowns. Coefficients of variation for daily PFD among sensors within a single crown ranged from 26 to 79%. Within a single crown located in a clearing, the range in daily PFD among leaves was nearly as great as the range over the entire sample of plants. In the understory, localized sunfleck activity contributed to a high degree of spatial variation in instantaneous and total PFD among leaves within individual crowns. Much of the microsite variation in sunfleck activity, however, reflected environmental conditions within the understory habitat. Within an array of sensors placed next to Piper crowns in the understory, correlations were poor for light sensors spaced only 0.2 m apart, and only 8% of the variance in light readings was explained by measurements made 0.5 m away. In the clearing habitats, microsite heterogeneity among leaves was more strongly influenced by leaf positions within crowns and leaf angles than by spatial heterogeneity within the habitat.     

Beyond the tropics: forest structure in a temperate forest mapped plot

Question: How do the diversity, size structure, and spatial pattern of woody species in a temperate (Mediterranean climate) forest compare to temperate and tropical forests? Location: Mixed evergreen coastal forest in the Santa Cruz Mountains, California, USA. Methods: We mapped, tagged, identified, and measured all woody stems (≥1 cm diameter) in a 6‐ha forest plot, following Center for Tropical Forest Science protocols. We compared patterns to those found in 14 tropical and 12 temperate forest plots. Results: The forest is dominated by Douglas‐fir (Pseudotsuga menziesii) and three species of Fagaceae (Quercus agrifolia, Q. parvula var. shrevei, and Lithocarpus densiflorus), and includes 31 woody species and 8180 individuals. Much of the diversity was in small‐diameter shrubs, treelets, and vines that have not been included in most other temperate forest plots because stems <5‐cm diameter had been excluded from study. The density of woody stems (1363 stems ha^?1) was lower than that in all but one tropical plot. The density of large trees (diameter ≥30 cm) and basal area were higher than in any tropical plot. Stem density and basal area were similar to most other temperate plots, but were less than in low‐diversity conifer forests. Rare species were strongly aggregated, with the degree of aggregation decreasing with abundance so that the most common species were significantly more regular than random. Conclusions: The patterns raise questions about differences in structure and dynamics between tropical and temperate forests; these need to be confirmed with additional temperate zone mapped plots that include small‐diameter individuals.     

Taxonomic identification of Amazonian tree crowns from aerial photography

Question: To what extent can aerial photography be used for taxonomic identification of Amazonian tree crowns? Objective: To investigate whether a combination of dichotomous keys and a web‐based interface is a suitable approach to identify tree crowns. Location: The fieldwork was conducted at Tiputini Biodiversity Station located in the Amazon, eastern Ecuador. Methods: High‐resolution imagery was taken from an airplane flying at a low altitude (600 m) above the ground. Imagery of the observable upper layer of the tree crowns was used for the analysis. Dichotomous identification keys for different types of crowns were produced and tested. The identification keys were designed to be web‐based interactive, using Google Earth as the main online platform. The taxa analysed were Iriartea, Astrocaryum, Inga, Parkia, Cecropia, Pourouma, Guarea, Otoba, Lauraceae and Pouteria. Results: This paper demonstrates that a combination of photo‐imagery, dichotomous keys and a web‐based interface can be useful for the taxonomic identification of Amazonian trees based on their crown characteristics. The keys tested with an overall identification accuracy of over 50% for five of the ten taxa with three of them showing accuracy greater than 70% (Iriartea, Astrocaryum and Cecropia). Conclusions: The application of dichotomous keys and a web‐based interface provides a new methodological approach for taxonomic identification of various Amazonian tree crowns. Overall, the study showed that crowns with a medium‐rough texture are less reliably identified than crowns with smoother or well‐defined surfaces.     

Role of Acacia and Erythrina trees in forest regeneration by vertebrate seed dispersers in Kibale National Park,Uganda

Previous studies suggest that forest regeneration in grasslands is often slow because of grass competition and fire and that regeneration may be dependent on fire‐resistant savannah trees. To examine the potential of savannah trees in facilitating regeneration, species diversity, number and total abundance of species of woody plants were determined below and away from Acacia sieberiana and Erythrina abyssinica tree crowns. Additionally, crown size and distance from a natural forest were estimated to determine their influence on natural regeneration. Results showed that the environment under tree crowns positively influence diversity compared to that outside crowns: including for biodiversity (3.08 versus 2.82), the number of species and total abundance (P < 0.001). However, distance from the forest to trees in the grassland had no influence on these parameters. Vertebrate animals were found to be the major seed dispersers in grasslands of Kibale. We concluded that forests that establish below crowns of savannah trees will be more diverse than those in treeless areas and that crown size is more important than distance from natural forest in facilitating regeneration. Furthermore, A. sieberiana could be more suitable in facilitating natural regeneration, while animals have proved to be vital for regeneration.     

The potential for forest canopy litterfall interception by a dense fern understorey, and the consequences for litter decomposition

Most studies on litter decomposition have assumed that all falling plant litter reaches the ground where it then decomposes. In many forests a proportion of this litter may in fact be intercepted by understorey vegetation, but the ecological significance of this remains poorly understood. We performed two experiments in a temperate rainforest in southern New Zealand, in which there was a dense understorey of the crown fern Blechnum discolor. The fronds of this fern originate from a crown, and have a funnel‐like arrangement that can trap falling litter and prevent it from reaching the ground. The first experiment measured the effects of ferns on the spatial distribution of litter accumulation over one year. The ferns intercepted a substantial proportion of the total litterfall, and the fern crowns (from which the fronds originate) retained 10% of the total incoming litterfall (despite occupying only 2% of the ground area). The retained litter had a substantially higher ratio of twig to foliar litter than did the incoming litterfall. Further, much of the litter not retained on the crowns of the ferns accumulated at the base of the fern trunks. The second experiment considered litter decomposition in fern crowns versus the ground under the ferns. The litter that had accumulated in the crowns was characterized by higher microbial basal respiration and active microbial biomass than was the litter that had accumulated on the ground. The use of litterbags revealed that litter decomposition rates were significantly higher on the fern crowns than on the ground at 30 cm and 60 cm from the fern trunks. These results show that litter interception ameliorates the decomposer environment and increases the rate of litter decomposition. In total, this study provides evidence for understorey ferns greatly influencing both the spatial distribution of litterfall and the decomposition of plant litter. Although the ecological role of understorey vegetation in forested ecosystems has received little attention to date, our results point to understorey species as an important driver of forest ecosystem processes.     

Managing mature forest features: The production,accuracy and ecological relevance of a landscape‐scale map

Mature trees and forests contain structural features such as tree hollows, large coarse woody debris and large spreading crowns that provide critical habitat for a wide range of species. These features can take hundreds of years to develop and require careful management to ensure their continued availability. Managing these features requires spatial mapping layers to facilitate landscape‐scale management. This paper outlines how a map of mature forest habitat was developed for Tasmania, Australia. The map was produced using spatial data on vegetation type, mature crown density and senescence, a global layer of forest loss data derived from satellite imagery, a database on timber harvest plans and a spatial layer on the extent of fire. The relationship between mapped mature habitat availability (high, medium, low or negligible) and tree hollow availability in wet forest areas was explored, complementing a similar published study in dry forests. The number of large trees likely to have hollows significantly increased with mapped mature habitat availability, although there was considerable variation and overlap between map categories. Data from a fauna locality database and two radio‐tracking studies showed that three of the vertebrate hollow‐using species examined (Swift Parrot, Common Brushtail Possum and the Tasmanian Long‐eared Bat) and nest records of a species reliant on large tree crowns (the Wedge‐tailed Eagle) were all more likely to occur in areas of higher mapped mature habitat availability. It is concluded that this map reflects the relative availability of tree hollows, is ecologically meaningful and will be useful when managing mature forest habitat at large spatial scales, but the variable accuracy of the map at fine scales needs to be taken into account.     

Irradiance heterogeneity within crown affects photosynthetic capacity and nitrogen distribution of leaves in Cedrela sinensis

Because light conditions in the forest understory are highly heterogeneous, photosynthetic acclimation to spatially variable irradiance within a crown is important for crown‐level carbon assimilation. The effect of variation in irradiance within the crown on leaf nitrogen content and photosynthetic rate was examined for pinnate compound leaves in saplings of Cedrela sinensis, a pioneer deciduous tree. Five shading treatments, in which 0, 25, 50, 75 and 100% of leaves were shaded, were established by artificial heavy shading using shade screen umbrellas with 25% transmittance. Although the nitrogen content of leaves was constant regardless of shading treatment, ribulose 1·5‐bisphosphate carboxylase/oxygenase (Rubisco) content and light‐saturated photosynthetic capacity were lower in shade leaves within partially shaded crowns than within fully shaded crowns. Shade leaves within partially shaded crowns contained higher amount of amino acids. Most shade leaves died in partially shaded crowns, whereas more than half of shade leaves survived in totally shaded crowns. Assumptions on photosynthetic acclimation to local light conditions cannot explain why shade leaves have different photosynthetic capacities and survival rates in between partially and totally shaded crowns. Irradiance heterogeneity within the crown causes a distinct variation in photosynthetic activity between sun and shaded leaves within the crown.     

Effect of prey availability on the abundance of White‐breasted Wood‐Wrens,insectivorous birds of tropical lowland forests

Some understory insectivorous birds manage to persist in tropical forest fragments despite significant habitat loss and forest fragmentation. Their persistence has been related to arthropod biomass. In addition, forest structure has been used as a proxy to estimate prey availability for understory birds and for calculating prey abundance. We used arthropod biomass and forest structural variables (leaf area index [LAI] and aerial leaf litter biomass) to explain the abundance of White‐breasted Wood‐Wrens (Henicorhina leucosticta), tropical understory insectivorous birds, in six forests in the Caribbean lowlands of Costa Rica. To estimate bird abundance, we performed point counts (100‐m radius) in two old‐growth forests, two second‐growth forests, and two selectively logged forests. Arthropod abundance was the best predictor of wood‐wren abundance (wi = 0.75). Wood‐wren abundance increased as the number of arthropods increased, and the estimated range of bird abundance obtained from the model varied from 0.51 (0.28 – 0.93 [95%CI]) to 3.70 (1.68 – 5.20 [95%CI]) within sites. LAI was positively correlated to prey abundance (P = 0.01), and explained part of the variation in wood‐wren abundance. In forests with high LAI, arthropods have more aerial leaf litter as potential habitat so more potential prey are available for wood‐wrens. Forests with a greater abundance of aerial leaf litter arthropods were more likely to sustain higher densities of wood‐wrens in a fragmented tropical landscape.     

Spatial Pattern of Pindó Palm (Syagrus romanzoffiana) Recruitment in Argentinian Atlantic Forest: The Importance of Tapir and Effects of Defaunation

Hunting pressure, fragmentation and deforestation have caused global declines in animal abundance, and the consequences for plant communities are poorly understood. Many large‐seeded plants, for instance, depend on large and endangered vertebrates for seed dispersal. In some Semi‐deciduous Atlantic Forests, endangered tapirs (Tapirus terrestris) are major dispersers of pindó palms (Syagrus romanzoffiana). Here, we compare recruitment patterns of pindó palms between protected and disturbed (defaunated) Atlantic Forest areas in Argentina and evaluate the potential consequences of the lack of the main disperser for pindó palm regeneration. We analyzed the number and spatial pattern of pindó adults, offspring, and tapir dung piles within ten plots established in an area spanning tapir latrines inside Iguazú National Park and in a fragmented forest area outside the park where tapir is locally extinct. In both areas, we evaluated recruitment levels beneath 24 adult palms in circular plots centered on adult stems. We found lower pindó palm recruitment outside the park where offspring tended to be aggregated around adult palms. In contrast, in Iguazú National Park offspring were spatially associated with tapir dung‐piles, in which most offspring were registered. Recruitment under adults was higher outside the park suggesting a lower rate of seed removal in disturbed areas. Our results show that tapir dispersal promotes higher recruitment levels of pindó offspring and shapes their spatial pattern, breaking the spatial association with adult (presumably maternal) palms. These results are useful for predicting the impact of local tapir extinction on this palm.     

Validation of a landscape‐scale planning tool for cavity‐dependent wildlife

Tree cavities provide important habitat for wildlife. Effective landscape‐scale management of cavity‐dependent wildlife requires an understanding of where cavities occur, but tree cavities can be cryptic and difficult to survey. We assessed whether a landscape‐scale map of mature forest habitat availability, derived from aerial photographs, reflected the relative availability of mature trees and tree cavities. We assessed cavities for their suitability for use by wildlife, and whether the map reflected the availability of such cavities. There were significant differences between map categories in several characteristics of mature trees that can be used to predict cavity abundance (i.e. tree form and diameter at breast height). There were significant differences between map categories in the number of potential cavity bearing trees and potential cavities per tree. However, the index of cavity abundance based on observations made from the ground provided an overestimate of true cavity availability. By climbing a sample of mature trees we showed that only 5.1% of potential tree cavities detected from the ground were suitable for wildlife, and these were found in only 12.5% of the trees sampled. We conclude that management tools developed from remotely sensed data can be useful to guide decision‐making in the conservation management of tree cavities but stress that the errors inherent in these data limit the scale at which such tools can be applied. The rarity of tree cavities suitable for wildlife in our study highlights the need to conserve the tree cavity resource across the landscape, but also the importance of increasing the accuracy of management tools for decision‐making at different scales. Mapping mature forest habitat availability at the landscape scale is a useful first step in managing habitat for cavity‐dependent wildlife, but the potential for overestimating actual cavity abundance in a particular area highlights the need for complementary on‐ground surveys.     

Californian mixed-conifer forests under unmanaged fire regimes in the Sierra San Pedro Mártir, Baja California, Mexico

Aim This study appraises historical fire regimes for Californian mixed‐conifer forests of the Sierra San Pedro Mártir (SSPM). The SSPM represents the last remaining mixed‐conifer forest along the Pacific coast still subject to uncontrolled, periodic ground fire. Location The SSPM is a north–south trending fault bound range, centred on 31°N latitude, 100 km SE of Ensenada, Baja California. Methods We surveyed forests for composition, population structure, and historical dynamics both spatially and temporally over the past 65 years using repeat aerial photographs and ground sampling. Fire perimeter history was reconstructed based on time‐series aerial photographs dating from 1942 to 1991 and interpretable back to 1925. A total of 256 1‐ha sites randomly selected from aerial photographs were examined along a chronosequence for density and cover of canopy trees, density of snags and downed logs, and cover of non‐conifer trees and shrubs. Twenty‐four stands were sampled on‐the‐ground by a point‐centred quarter method which yielded data on tree density, basal area, frequency, importance value, and shrub and herb cover. Results Forests experience moderately intense understory fires that range in size to 6400 ha, as well as numerous smaller, low intensity burns with low cumulative spatial extent. SSPM forests average 25–45% cover and 65–145 trees per ha. Sapling densities were two to three times that of overstory trees. Size‐age distributions of trees ≥ 4 cm dbh indicate multi‐age stands with steady‐state dynamics. Stands are similar to Californian mixed conifer forests prior to the imposition of fire suppression policy. Livestock grazing does not appear to be suppressing conifer regeneration. Main conclusions Our spatially‐based reconstruction shows the open forest structure in SSPM to be a product of infrequent, intense surface fires with fire rotation periods of 52 years, rather than frequent, low intensity fires at intervals of 4–20 years proposed from California fire‐scar dendrochronology (FSD) studies. Ground fires in SSPM were intense enough to kill pole‐size trees and a significant number of overstory trees. We attribute long fire intervals to the gradual build‐up of subcontinuous shrub cover, conifer recruitment and litter accumulation. Differences from photo interpretation and FSD estimates are due to assumptions made with respect to site‐based (point) sampling of fire, and nonfractal fire intensities along fire size frequency distributions. Fire return intervals determined by FSD give undue importance to local burns which collectively use up little fuel, cover little area, and have little demographic impact on forests.     

Forest canopy structure analyzed by using aerial photographs

A method was developed using aerial photographs to analyze forest canopy structure. Digital elevation models of both the land and canopy surface in a mesh of 5 m intervals were made from aerial photographs taken in winter (without tree leaves) and summer (with leaves), respectively, in a 60 ha area of temperate deciduous forest. The difference between the two elevation values at each point was regarded as the canopy height, and a canopy height profile was constructed. The estimated canopy structure was compared with that obtained by ground observations in a 6 ha part of the study area. Large gaps (>100 m²) were adequately detected by the method, and the gap size distribution obtained was similar to the one observed on the ground. The method was found to be effective in analyzing the forest canopy structure of large areas, but it is not suitable for the detection of small gaps.