Understory Research Articles

Deep learning for conifer/deciduous classification of airborne LiDAR 3D point clouds representing individual trees

The purpose of this study was to investigate the use of deep learning for coniferous/deciduous classification of individual trees segmented from airborne LiDAR data. To enable processing by a deep convolutional neural network (CNN), we designed two discrete representations using leaf-off and leaf-on LiDAR data: a digital surface model with four channels (DSM × 4) and a set of four 2D views (4 × 2D). A training dataset of tree crowns was generated via segmentation of tree crowns, followed by co-registration with field data. Potential mislabels due to GPS error or tree leaning were corrected using a statistical ensemble filtering procedure. Because the training data was heavily unbalanced (~8% conifers), we trained an ensemble of CNNs on random balanced sub-samples. Benchmarked against multiple traditional shallow learning methods using manually designed features, the CNNs improved accuracies up to 14%. The 4 × 2D representation yielded similar classification accuracies to the DSM × 4 representation (~82% coniferous and ~90% deciduous) while converging faster. Further experimentation showed that early/late fusion of the channels in the representations did not affect the accuracies in a significant way. The data augmentation that was used for the CNN training improved the classification accuracies, but more real training instances (especially coniferous) likely results in much stronger improvements. Leaf-off LiDAR data were the primary source of useful information, which is likely due to the perennial nature of coniferous foliage. LiDAR intensity values also proved to be useful, but normalization yielded no significant improvement. As we observed, large training data may compensate for the lack of a subset of important domain data. Lastly, the classification accuracies of overstory trees (~90%) were more balanced than those of understory trees (~90% deciduous and ~65% coniferous), which is likely due to the incomplete capture of understory tree crowns via airborne LiDAR. In domains like remote sensing and biomedical imaging, where the data contain a large amount of information and are not friendly to human visual system, human-designed features may become suboptimal. As exemplified by this study, automatic, objective derivation of optimal features via deep learning can improve prediction tasks in such domains.

A Multi-Threshold Segmentation for Tree-Level Parameter Extraction in a Deciduous Forest Using Small-Footprint Airborne LiDAR Data

The development of new approaches to tree-level parameter extraction for forest resource inventory and management is an important area of ongoing research, which puts forward high requirements for the capabilities of single-tree segmentation and detection methods. Conventional methods implement segmenting routine with same resolution threshold for overstory and understory, ignoring that their lidar point densities are different, which leads to over-segmentation of the understory trees. To improve the segmentation accuracy of understory trees, this paper presents a multi-threshold segmentation approach for tree-level parameter extraction using small-footprint airborne LiDAR (Light Detection And Ranging) data. First, the point clouds are pre-processed and encoded to canopy layers according to the lidar return number, and multi-threshold segmentation using DSM-based (Digital Surface Model) method is implemented for each layer; tree segments are then combined across layers by merging criteria. Finally, individual trees are delineated, and tree parameters are extracted. The novelty of this method lies in its application of multi-resolution threshold segmentation strategy according to the variation of LiDAR point density in different canopy layers. We applied this approach to 271 permanent sample plots of the University of Kentucky’s Robinson Forest, a deciduous canopy-closed forest with complex terrain and vegetation conditions. Experimental results show that a combination of multi-resolution threshold segmentation based on stratification and cross-layer tree segments merging method can provide a significant performance improvement in individual tree-level forest measurement. Compared with DSM-based method, the proposed multi-threshold segmentation approach strongly improved the average detection rate (from 52.3% to 73.4%) and average overall accuracy (from 65.2% to 76.9%) for understory trees. The overall accuracy increased from 75.1% to 82.6% for all trees, with an increase of the coefficient of determination R2 by 20 percentage points. The improvement of tree detection method brings the estimation of structural parameters for single trees up to an accuracy level: For tree height, R2 increased by 5.0 percentage points from 90% to 95%; and for tree location, the mean difference decreased by 23 cm from 105 cm to 82 cm.

Patterns in a species‐rich tropical understory plant community

AbstractUnderstory plants are an important component of the high plant species diversity characteristic of neotropical rain forests. Herbs, shrubs, understory trees, and saplings of canopy trees occupy a broadly uniform environment of abundant rainfall, low light levels, and high humidity. We asked whether this community at the La Selva Biological Station in the Caribbean lowlands of Costa Rica was structured by environmental filters such as soil origin, topographic position, and understory light availability. We used nested quadrats to assess effects of soil origin (recent alluvium, weathered alluvium, residual volcanic soil) and topographic position (ridges, mid‐slopes and flats) on species composition, density, and diversity and measured six edaphic and understory light parameters. Non‐metric multidimensional scaling ordinations were based on frequency of occurrence in 20 quadrats for 272 species in the shrub size class and 136 species in the small‐tree size class for 17 sites. Three axes were correlated with composite environmental variables produced by principal component analysis representing slope, extractable phosphorus, and light. NMS site positions also reflected soil origin, topographic position, and geographic location. The analyses illustrated a complex community structured by species responses to environmental filters at multiple, interdigitated spatial scales. We suggest that light availability affected by canopy dynamics and dispersal limitation provides additional sources of variation in species distributions, which interact with edaphic patterns in complex ways.Abstract in Spanish is available with online material.

Spatial structure characteristics of the main tree species in a mixed broadleaved Korean pine (Pinus koraiensis)forest in a mountainous area of eastern Liaoning Province, China

Maintaining forest structural diversity is generally considered as an effective way to preserve forest stability and biodiversity. The spatial structure characteristics of the dominant tree species in a climax community were investigated in a primary mixed broadleaved Korean pine (Pinus koraiensis) forest in a mountainous area of eastern Liaoning. Stand spatial structure parameters were determined based on the relationships among neighboring trees. The climax communities were used as a theoretical reference for optimizing the spatial structure of a low-quality secondary forest and monoculture plantation. The diameter distribution of the trees in the pine forest exhibited an inverse J-shape, indicating that understory regeneration was relatively good and with certain proportion of large-diameter trees. The main tree species were randomly distributed across the whole plot (=0.507) and in an intensively mixed state (=0.82). An average DBH comparison of trees in the stand indicated that they were at a intermediate status (=0.506). There was a differentiation among different dominances along the high intensity mixed dimension in the stand, indicating an optimal distribution of understory trees and the rational utilization of resources. Trees in the small diameter category were at a state of complete compression, while canopy trees were at a state of complete dominance in terms of their vertical space. Individuals of each dominant tree species were randomly scattered, with a random pattern of individuals throughout the climax community.

Low Genetic Diversity Suggests the Recent Introduction of Dogwood Powdery Mildew to North America.

Cornus florida (flowering dogwood) is a popular understory tree endemic to the eastern hardwood forests of the United States. In 1996, dogwood powdery mildew caused by Erysiphe pulchra, an obligate biotrophic fungus of large bracted dogwoods, reached epidemic levels throughout the C. florida growing region. In the late 1990s, both sexual and asexual stages of E. pulchra were regularly observed; thereafter, the sexual stage was found less frequently. We examined the genetic diversity and population structure of 167 E. pulchra samples on C. florida leaves using 15 microsatellite loci. Samples were organized into two separate collection zone data sets, separated as eight zones and two zones, for the subsequent analysis of microsatellite allele length data. Clone correction analysis reduced the sample size to 90 multilocus haplotypes. Our study indicated low genetic diversity, a lack of definitive population structure, low genetic distance among multilocus haplotypes, and significant linkage disequilibrium among zones. Evidence of a population bottleneck was also detected. The results of our study indicated a high probability that E. pulchra reproduces predominately via asexual conidia and lend support to the hypothesis that E. pulchra is an exotic pathogen to North America.[Formula: see text] Copyright © 2019 The Author(s). This is an open access article distributed under the CC BY-NC-ND 4.0 International license.

Effects of Understory Vegetation Management on Plant Communities in Oil Palm Plantations in Sumatra, Indonesia

Oil palm plantations have expanded rapidly in recent decades, and are causing substantial impacts on tropical habitats and biodiversity. However, owing to its long lifespan (25-30 years), oil palm forms a much more varied and structurally-complex habitat than many other crops. This can include abundant understory vegetation and also epiphytes on palm trunks. However, the diversity of this plantation vegetation has been poorly studied, and there has been little consideration of the impacts of common plantation vegetation management practices on plant communities. We conducted a long-term vegetation management experiment that forms part of the Biodiversity and Ecosystem Function in Tropical Agriculture (BEFTA) Programme in Riau, Indonesia. We manipulated herbicide and manual cutting regimes within mature oil palm plantations to create three different understory complexity treatments (Reduced, Normal, and Enhanced vegetation) across replicated sets of plots. Plant communities were surveyed before and after experimental understory vegetation treatments began in three different microhabitats: within the middle of the plantation block (core), on the road edge (edge) and on oil palm trunks (trunk). Part of the sampling was also conducted during a drought event. We recorded 120 plant species, which comprised a mixture of native, non-native, ‘beneficial’, and ‘problem’ species. We found substantial variation in plant communities between edge, core, and trunk microhabitats, indicating high levels of heterogeneity within the plantation. There were significant effects of varying understory treatment within both core and edge microhabitats, but no spillover of impacts into the trunk microhabitat. We also observed substantial impacts of drought on plant communities, with declines in either biomass, percentage cover, or richness seen across core, edge, and trunk microhabitats during low-rainfall periods. Our findings highlight the diversity of plant communities that can be supported within oil palm plantations, and the substantial impacts that management decisions, and also drought, can have on them. Given the role that diverse plant communities can have in supporting species in other groups, this is likely to have a significant impact on the wider plantation biodiversity. We suggest that plantation management strategies give greater consideration to within-plantation understory plant communities and choose more wildlife-friendly options where possible.

Long-term vegetation transition on man-made slopes 53 years after construction in Central Japan

Currently, the quantitative assessment of revegetation is often limited to within a few years of vegetation construction. Thus, it is imperative to gather information regarding vegetation transition such as plant species composition and identify new indicators that can be applied to evaluate long-term vegetation restoration on revegetated slopes. To achieve this, the vegetation on the oldest expressway slopes in Japan and its transitions were studied 53 years after construction using the Braun–Blanquet method, cluster analysis, and indicator species analysis (INSPAN), and two vegetation successional patterns were identified. The first pattern, a vegetation transition dominated by deciduous broad-leaved trees, was observed on orthosere slopes. Despite a previous study predicting the establishment of a Pinus densiflora-dominant community by about 50 years after construction, no adult P. densiflora trees were identified in this study, and pine wilt disease was implicated. The second pattern, a transition to a plagiosere with Pueraria lobata as an indicator species, was observed on slopes where a P. densiflora-dominant community had been established 33 years after the construction. A plagiosere with a bamboo-dominant community was newly identified in addition to the previously reported P. lobata-dominated one. All the plagiosere slopes displayed single-peak community structures, in which the highest coverage was recorded for the lower tree layer, community height was less than 10 m, and P. lobata as an indicator species. It is suggested that these common features are indicative of the vegetation shift to a plagiosere and are useful as long-term revegetation evaluation criteria.

A hybrid method for segmenting individual trees from airborne lidar data

Segmentation of individual trees from airborne lidar data uses either the point cloud directly or canopy height models (CHMs) derived from the point cloud. Point-based segmentation is able to detect understorey trees but is sensitive to the point density and often demands a high overhead cost of computing. Conversely, CHM-based segmentation can be easily implemented but it is impractical for the detection of understorey trees. To identify highly accurate treetops as well as understorey trees, this paper presents a hybrid method by modifying a CHM-based individual tree crown delineation (ITCD) algorithm and integrating it into a point-based algorithm. A multiscale local maxima (LM) algorithm is developed to improve the accuracy of LM obtained from CHMs in different spatial resolutions. The improved LM are used as seeds to segment the lidar point cloud into individual trees. For each tree, histogram analysis is applied to investigate the presence of understorey trees. Field measurements of tree heights and crown widths are used as ground truth to evaluate how well the proposed method is performing. The mean errors of tree heights and crown widths are 0.147 m and −0.004 m, respectively. The proposed method is also compared with five conventional methods of individual tree segmentation, namely ITCD, fixed window local maxima, Popescu and Wynne’s local maxima, variable area local maxima, and Li’s point-based segmentation. The comparison results indicate that the proposed hybrid method outperforms the conventional methods in terms of detection rate, omission error, commission error, mean absolute error of tree heights and root-mean-squared-error of tree heights.

Susceptibility of Small-Diameter Norway Spruce Understory Stumps to Heterobasidion Spore Infection

Heterobasidion spp. cause economically important losses in conifer forests in the Northern Hemisphere, especially in Norway spruce stands. Freshly cut stumps are the main route for Heterobasidion spp. infection. Even small stumps of spruce seedlings are known to be susceptible to spore infection, however, very little is currently known about the susceptibility of small stumps of understory spruce to Heterobasidion spore infection. To determine the frequency of spore infections among stumps of understory trees, we analyzed 756 Norway spruce stumps in eight sample plots in eastern Latvia. Understory trees 35–68 years of age, with a stump diameter of 2–10 cm, were felled 33–48 weeks before sampling. In total, 514 (68%) of the spruce stumps were infected by Heterobasidion spores. The infection frequency among the plots varied from 56% to 86%. Both infection frequency and stump surface area occupied by Heterobasidion positively correlated with stump diameter, but neither correlated with the time span elapsed between felling and sampling. Colonization of trees by a competitor fungus, Phlebiopsis gigantea (Fr.) Jülich, was observed in 30 (4%) of investigated stumps, but did not have any effect on the frequency of Heterobasidion infections. Our data show that Norway spruce stumps can be highly susceptible to Heterobasidion spore infection. In order to control airborne Heterobasidion infections, stump treatment should be considered during the sporulation period of Heterobasidion spp.

Biodiversity-ecosystem functioning relationships of overstorey versus understorey trees in an old-growth temperate forest

The results of this study, based on a large dataset with mapped tree locations, suggest that the relationship between biodiversity and ecosystem functioning (BEF) is different between the over- and understorey in the investigated forest community. Neither the functional nor the phylogenetic diversity showed a significant advantage in predicting aboveground biomass and biomass production (aboveground biomass increment) when compared with species richness. Consistent conclusions have still not been reached regarding the relative importance of the different diversity indices in predicting biomass production. Furthermore, the effects of certain forest strata on the BEF relationship remain unclear. The woody plant species in the study area were classified as two vertical strata and were referred to as the “overstorey” (trees with dbh > 10 cm) and “understorey” (trees with dbh ≤ 10 cm). The relationships between different metrics of tree diversity and biomass production were quantified using linear models. The set of the best predictors for tree biomass and productions were determined using a multi-model selection approach. The differences in diversity-ecosystem functioning relationships between the two forest vertical strata, and the effects of canopy trees on the understorey trees, were investigated using structural equation models. The relative importance of the species diversity, phylogenetic diversity, and functional diversity in predicting the aboveground biomass production (ABP) was examined to compare the potential of phylogenetic and functional diversity indices and species richness in predicting biomass productivity and compare the influence of niche complementarity, selection effects, or mass-ratio effects on aboveground biomass variables and to assess possible differences in the BEF relationships between the overstorey and understorey layers in the studied ecosystems. Also, this study investigated possible differences in the BEF relationships between the overstorey and understorey layers in the studied ecosystems. The results confirmed a positive diversity-productivity relationship in the natural coniferous and broadleaved mixed forest. However, the relationship in the overstorey stratum was statistically stronger than that in the understorey layer. The combination of diversity indices and functional traits could explain more of the variations in the biomass and productivity than when examined separately. The correlation between the richness of the overstorey and understorey species was found to be positive. However, the aboveground woody biomass of the overstorey layer had negative effects on the understorey biomass. In summary, this study found a stronger positive correlation between the woody plant diversity and productivity in the overstorey than in the understorey stratum. These findings indicate that the BEF relationships are hierarchically dependent, and supported by niche complementarity. Also, a much weaker relationship was observed between the functional diversity and biomass production than between the species diversity and biomass production due to a selection effect. The aboveground woody biomass was found to be enhanced in communities dominated by functional traits correlated to smaller leaf areas, greater maximum tree heights, and higher leaf carbon content. These findings suggest that the BEF relationships were driven by a mass-ratio hypothesis.

Responses of Phyllostomid Bats to Traditional Agriculture in Neotropical Montane Forests of Southern Mexico.

Bat communities' responses to land use change in neotropical montane forests have scarcely been studied. We hypothesized that, like in lowland forests, a montane agricultural area will have a lower species richness, abundance, diversity and species composition of understory phyllostomid bats than a native forest (montane cloud forest and pine-oak forest). Monthly surveys over the course of a year gave an overall low species richness and abundance (167 captures corresponding to nine species). We found a slight loss of species richness in agricultural areas with respect to the montane cloud forest (one species) and pine-oak forest (two species). However, differences in abundance were noteworthy: 45% and 73% fewer captures in agricultural areas than in the montane cloud forest and pine-oak forest, respectively. Species diversity was higher in the montane cloud forest than the pine-oak forest, but the diversity of agricultural areas did not differ between the types. Species and guild compositions did not differ between crops and forests. At least for the understory phyllostomid bats, and at the spatial scale studied, traditional management of agricultural areas in the study area and the surrounding matrix could explain the similarity in species richness, composition, and diversity between the agricultural area and native montane forests; however, other indicator groups should be evaluated to understand the effects of habitat loss on montane forests.

Tracking the long-term structure changes of a mature deciduous broadleaf forest stand using digital hemispherical photography

Abstract Hemispherical photography provides permanent records of forest canopy structure. We analysed digital hemispherical images taken during the period of 2007–2018 in a mature silver birch stand located in Järvselja, Estonia. The stand was thinned in 2004. Understory trees were removed in the spring of 2018. Images were processed using the LinearRatioSC method. Effective plant area index L eff during the leafless phenophase increased as a result of tree growth from 0.92 to 1.24 and understory cutting was not detectable. During the full foliage condition L eff increased from 3.6 in 2008 to 5.8 in 2017. After removal of understory trees from the stand L eff decreased, and repeated measurements in the summer of 2018 estimated the plant area index range 4.5 < L eff < 4.8. The results are in agreement with the expected changes following forest growth and demonstrate that LinearRatioSC is a suitable method for the estimation and long-term monitoring of forest canopy properties from digital hemispherical images.

Individual-Based Modeling of Amazon Forests Suggests That Climate Controls Productivity While Traits Control Demography

Climate, species composition, and soils are thought to control carbon cycling and forest structure in Amazonian forests. Here, we add a demographics scheme (tree recruitment, growth, and mortality) to a recently developed non-demographic model - the Trait-based Forest Simulator (TFS) – to explore the roles of climate and plant traits in controlling forest productivity and structure. We compared two sites with differing climates (seasonal versus aseasonal precipitation) and plant traits. Through an initial validation simulation, we assessed whether the model converges on observed forest properties (productivity, demographic and structural variables) using datasets of functional traits, structure, and climate to model the carbon cycle at the two sites. In a second set of simulations, we tested the relative importance of climate and plant traits for forest properties within the TFS framework using the climate from the two sites with hypothetical trait distributions representing two axes of functional variation (‘fast’ versus ‘slow’ leaf traits, and high versus low wood density). The adapted model with demographics reproduced observed variation in gross (GPP) and net (NPP) primary production, and respiration. However NPP and respiration at the level of plant organs (leaf, stem, and root) were poorly simulated. Mortality and recruitment rates were underestimated. The equilibrium forest structure differed from observations of stem numbers suggesting either that the forests are not currently at equilibrium or that mechanisms are missing from the model. Findings from the second set of simulations demonstrated that differences in productivity were driven by climate, rather than plant traits. Contrary to expectation, varying leaf traits had no influence on GPP. Drivers of simulated forest structure were complex, with a key role for wood density mediated by its link to tree mortality. Modelled mortality and recruitment rates were linked to plant traits alone, drought-related mortality was not accounted for. In future, model development should focus on improving allocation, mortality, organ respiration, simulation of understory trees and adding hydraulic traits. This type of model that incorporates diverse tree strategies, detailed forest structure and realistic physiology is necessary if we are to be able to simulate tropical forest responses to global change scenarios.

ENERGY WOOD STORES IN UNDERGROWTH OF FORESTS IN LATVIA

INTRODUCTION
 Forest resources are the most significant natural asset of the state of Latvia. According to data of the 2nd stage of forest resource monitoring for 2014, Latvia has 3575 thousand hectares of forest land comprising 55.3 percent of the total territory of Latvia, while the total timber stock is estimated at 668 million cubic meters (Bumanis et al., 2014).
 However, from the available data on forest resources it is only possible to theoretically and hypothetically state what proportion of these resources would be useful and economically justified as an energy supply. Each forest stand has a certain amount of undergrowth and understorey – small woody plants (shrubs) which have not been researched much until now. A precise determination of the amount of energy wood in Latvian forests would be of great benefit to the Latvian economy.
 MATERIAL AND METHODS
 Research data were collected in forests at the Jelgava Forest District “Forest Research Station”. The research was carried out in two forest subquarters of forest site type Myrtillosa mel. Eight circle-shaped sample plots were established. The area of each single plot was 25 m2.
 In the sample plots, the understorey and undergrowth trees were cut at the root neck. A sample was prepared from each tree harvested which was then sent for drying.
 The wood samples were transferred to "Forest and wood products research and development institute Ltd” for moisture determination. Total moisture content of the wood sample was determined according to standard LVS EN ISO 18134-2: 2016.
 RESULTS
 In the forest subquarter with a stand composition of 10Pine (66 years old) the sum biomass of undergrowth and understorey was 177.91 kg per sample plot. In the forest subquarter with a stand composition of 9Pine1Birch (88 years old) there was a total understorey tree mass of 180.9 kg but 16.17 kg of undergrowth per plot. This means there was more biomass in understorey than undergrowth in the site investigated.
 DISCUSSION
 When the tree stand was 10Pine (66 years old) the amount of dry matter to be extracted from all sample plots was 12.37 t ha-1 on average. In the forest subquarter with a stand composition of 9Pine1Birch (88 years old) the amount of dry mass is 10.24 t ha-1 on average.
 According to previous research, 7-20 t ha-1 of dry mass was obtained in Sweden, 7 to 12 t ha-1 in Poland, 6 to 14 t ha-1 in Germany and 8 to 12 t ha-1 in Latvia (Lazdina et al., 2010). There are also several researches papers which describe biomass from young hardwood stands on abandoned agricultural land in Canada: the values vary from 0,6 t ha-1 to 82,6 t ha-1 (Lupi et al., 2017). Consequently, a sufficient amount of biomass was obtained in the forest subquarters investigated in our research, which fits with the results of other studies carried out.
 CONCLUSION
 The volume of potential energy wood in undergrowth and understorey in Myrtillosa mel. forest site types is significant and it is advisable to use it as a raw material for energy production together with felling residues. However, it is necessary to evaluate the technical and technological capabilities from an economic perspective in each particular case.

SURVIVAL OF PINUS WASHOENSIS (PINACEAE) FOLLOWING FIRE ON BABBITT PEAK, SIERRA COUNTY, CALIFORNIA

The Cottonwood Fire consumed about 185 square kilometers in the Tahoe and Toiyabe National Forests near Loyalton and Sierraville, CA in 1994. The wildfire reached a population of one of the rarest pines in North America, Pinus washoensis H.Mason & Stockw. (Washoe pine), in the Babbitt Peak Research Natural Area (RNA). A field survey and census was conducted in 1996 to find, map, and count living and dead P. washoensis individuals, and to assess the effects of the fire on the population. Although the fire reached most of the P. washoensis stands, 763 living reproductive individuals were found and less than 5.2% mortality was observed. The fire's intensity declined as it crossed the Bald Mountain ridge and mostly consumed understory vegetation and litter in the P. washoensis stands. The fire had the potential to devastate or eliminate this rare population, but instead, a combination of stochastic, vegetative, and topographic controls spared the P. washoensis stands and appears to have had beneficial effects for this population. All age classes of P. washoensis except seedlings were found in the Bald Mountain Range. The lack of seedlings is surprising because the two years between fire and survey had high precipitation. No sign of hybridization with other yellow pine species was evident. While it is promising that the stands of this rare pine survived the Cottonwood Fire, prospects for the continued persistence of this population are not good until there is a reproductive year that provides many recruits.

The persistence of carbon in the African forest understory.

Quantifying carbon dynamics in forests is critical for understanding their role in long-term climate regulation1-4. Yet little is known about tree longevity in tropical forests3,5-8, a factor that is vital for estimating carbon persistence3,4. Here we calculate mean carbon age (the period that carbon is fixed in trees7) in different strata of African tropical forests using (1) growth-ring records with a unique timestamp accurately demarcating 66 years of growth in one site and (2) measurements of diameter increments from the African Tropical Rainforest Observation Network (23 sites). We find that in spite of their much smaller size, in understory trees mean carbon age (74 years) is greater than in sub-canopy (54 years) and canopy (57 years) trees and similar to carbon age in emergent trees (66 years). The remarkable carbon longevity in the understory results from slow and aperiodic growth as an adaptation to limited resource availability9-11. Our analysis also reveals that while the understory represents a small share (11%) of the carbon stock12,13, it contributes disproportionally to the forest carbon sink (20%). We conclude that accounting for the diversity of carbon age and carbon sequestration among different forest strata is critical for effective conservation management14-16 and for accurate modelling of carbon cycling4.

Differential effects of elevation on leaf size of overstorey and understorey species in a temperate rainforest

ABSTRACTConspicuous geographic variation in the average leaf size of plant communities suggests the physical environment must exert strong selection pressures on this trait. Leaf energy balance theory has recently been applied to explain global patterns in leaf size, through its strong influence on vulnerability to (a) overheating in dry sunny conditions and (b) chilling on cold clear nights. The same theory should also predict leaf size patterns on local elevational gradients, but there are fewer data available to examine this prediction. We collected recently-fallen leaf litter at 50 m elevational intervals in temperate rainforest on Mt. Te Aroha, New Zealand, hypothesising that average leaf size of angiosperm tree and shrub species should decrease with increasing elevation. Data from overstorey species (≥ 10 m tall) strongly supported this hypothesis (R2 = 0.36), average leaf size declining from 24.5 cm2 at 150 m to 4.3 cm2 at 850 m. Among understorey trees and shrubs (< 10 m tall) the relationship with elevation was weaker (R2 = 0.18) and the effect size smaller, average leaf size falling from 29.8 cm2 at 150 m to 9.8 cm2 at 850 m. The data are therefore consistent with limitation of leaf size of overstorey species by an elevational decrease in minimum temperatures. Leaf size of shorter tree and shrub species appears to be less constrained by elevational declines in temperature, a pattern that may be partly explained by protection from radiative cooling by overtopping taller trees. However, the occasional occurrence of large-leaved (notophyll) small trees and shrubs on open, recently-disturbed sites at high elevation requires clarification.

Does the Genetic Diversity of Macuilillo, Oreopanax xalapensis (Araliaceae), Change Along Successional Gradients of the Montane Cloud Forest?

Forest regeneration after human-induced or natural disturbances results in a sequential replacement of species, starting with colonizing of pioneer species that modifies the physical conditions of the environment, which later allows the establishment of mid- and late-successional species. Whether the forest succession implies a reduction of population size of tree species might be expected a less genetic diversity in populations inhabiting oldest succession stages. We aimed to study whether genetic diversity of Macuilillo, Oreopanax xalapensis (a understory tree), decreases in advanced age-classes and in late successional of montane cloud forest at Huitepec Ecological Reserve (Chiapas, Mexico). We counted individuals from three age classes: seedlings, saplings, and young reproductive individuals in nine plots along three successional stages (early-, middle-, and late-successional forests). Seven enzyme systems (SOD, AP, FUM, G6PDH, GOT, GPI, and G3PDH) were used to obtain common genetic diversity; Φ-statistics and a Bayesian model were used to infer population structure. At the early-successional stage, O. xalapensis had a bigger population size and significantly high numbers of seedlings. In general, among the successional stages, the genetic diversity of O. xalapensis was moderate to high ( He = 0.25–0.46); the saplings presented the highest expected genetic diversity ( He). Populations were genetically structured by age-classes but not by successional stages. The successional stages affected the demographic dynamics with a mild impact on the genetic structure across age-classes of O. xalapensis. The scenic and recreational conservation strategies implemented in this forest have effectively allowed for population development of O. xalapensis and provide some explanation of genetic diversity at early-succession stage.

A Single-Tree Processing Framework Using Terrestrial Laser Scanning Data for Detecting Forest Regeneration

Direct assessment of forest regeneration from remote sensing data is a previously little-explored problem. This is due to several factors which complicate object detection of small trees in the understory. Most existing studies are based on airborne laser scanning (ALS) data, which often has insufficient point densities in the understory forest layers. The present study uses plot-based terrestrial laser scanning (TLS) and the survey design was similar to traditional forest inventory practices. Furthermore, a framework of methods was developed to solve the difficulties of detecting understory trees for quantifying regeneration in temperate montane forest. Regeneration is of special importance in our montane study area, since large parts are declared as protection forest against alpine natural hazards. Close to nature forest structures were tackled by separating 3D tree stem detection from overall tree segmentation. In support, techniques from 3D mathematical morphology, Hough transformation and state-of-the-art machine learning were applied. The methodical framework consisted of four major steps. These were the extraction of the tree stems, the estimation of the stem diameters at breast height (DBH), the image segmentation into individual trees and finally, the separation of two groups of regeneration. All methods were fully automated and utilized volumetric 3D image information which was derived from the original point cloud. The total amount of regeneration was split into established regeneration, consisting of trees with a height &gt; 130 cm in combination with a DBH &lt; 12 cm and unestablished regeneration, consisting of trees with a height &lt; 130 cm. Validation was carried out against field-based expert estimates of percentage ground cover, differentiating seven classes that were similar to those used by forest inventory. The mean absolute error (MAE) of our method for established regeneration was 1.11 classes and for unestablished regeneration only 0.27 classes. Considering the metrical distances between the class centres, the MAE amounted 8.08% for established regeneration and 2.23% for unestablished regeneration.

Understory Vegetation in Oil Palm Plantations Benefits Soil Biodiversity and Decomposition Rates

Oil palm is the most productive vegetable oil crop per unit area and is crucial to the economy of developing countries such as Malaysia and Indonesia. However, it is also highly controversial due to the impact it has on biodiversity. Inputs of herbicides to control understory vegetation in plantations are high, which is likely to harm native biodiversity, but may be unnecessary in protecting oil palm yield. In this study we investigate the effects of understory manipulation using herbicides on soil fauna, litter decomposition rates, and soil abiotic variables: pH, soil organic carbon, soil water content, nitrogen, carbon/nitrogen ratio, potassium, and phosphorous. Understory vegetation was manipulated in three treatments: enhanced understory complexity (no herbicides, developed understory), normal understory complexity (intermediate herbicide use with some manual removal) and reduced understory complexity (heavy herbicide use, no understory vegetation). Two years after treatment, soil macrofauna diversity was higher in the enhanced than the normal, and reduced understory treatment. Furthermore, both macrofauna abundance and litter decomposition was higher in the enhanced than the reduced understory treatment. By contrast, soil fertility did not change between treatments, perhaps indicating there is little competition between oil palms, and understory vegetation. The reduction of herbicide use should be encouraged in oil palm plantations, this will not only reduce plantation costs, but improve soil biodiversity, and ecosystem functioning.