Dataset Information

Performance of non-parametric algorithms for spatial mapping of tropical forest structure.

ABSTRACT: BACKGROUND:Mapping tropical forest structure is a critical requirement for accurate estimation of emissions and removals from land use activities. With the availability of a wide range of remote sensing imagery of vegetation characteristics from space, development of finer resolution and more accurate maps has advanced in recent years. However, the mapping accuracy relies heavily on the quality of input layers, the algorithm chosen, and the size and quality of inventory samples for calibration and validation. RESULTS:By using airborne lidar data as the "truth" and focusing on the mean canopy height (MCH) as a key structural parameter, we test two commonly-used non-parametric techniques of maximum entropy (ME) and random forest (RF) for developing maps over a study site in Central Gabon. Results of mapping show that both approaches have improved accuracy with more input layers in mapping canopy height at 100 m (1-ha) pixels. The bias-corrected spatial models further improve estimates for small and large trees across the tails of height distributions with a trade-off in increasing overall mean squared error that can be readily compensated by increasing the sample size. CONCLUSIONS:A significant improvement in tropical forest mapping can be achieved by weighting the number of inventory samples against the choice of image layers and the non-parametric algorithms. Without future satellite observations with better sensitivity to forest biomass, the maps based on existing data will remain slightly biased towards the mean of the distribution and under and over estimating the upper and lower tails of the distribution.

SUBMITTER: Xu L

PROVIDER: S-EPMC4996895 | biostudies-literature | 2016 Dec

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Performance of non-parametric algorithms for spatial mapping of tropical forest structure.

Xu Liang L Saatchi Sassan S SS Yang Yan Y Yu Yifan Y White Lee L

Carbon balance and management 20160824 1

<h4>Background</h4>Mapping tropical forest structure is a critical requirement for accurate estimation of emissions and removals from land use activities. With the availability of a wide range of remote sensing imagery of vegetation characteristics from space, development of finer resolution and more accurate maps has advanced in recent years. However, the mapping accuracy relies heavily on the quality of input layers, the algorithm chosen, and the size and quality of inventory samples for calib ...[more]

PMID: 27617029

Similar Datasets

Project description:BackgroundMapping the aboveground biomass of tropical forests is essential both for implementing conservation policy and reducing uncertainties in the global carbon cycle. Two medium resolution (500 m - 1000 m) pantropical maps of vegetation biomass have been recently published, and have been widely used by sub-national and national-level activities in relation to Reducing Emissions from Deforestation and forest Degradation (REDD+). Both maps use similar input data layers, and are driven by the same spaceborne LiDAR dataset providing systematic forest height and canopy structure estimates, but use different ground datasets for calibration and different spatial modelling methodologies. Here, we compare these two maps to each other, to the FAO's Forest Resource Assessment (FRA) 2010 country-level data, and to a high resolution (100 m) biomass map generated for a portion of the Colombian Amazon.ResultsWe find substantial differences between the two maps, in particular in central Amazonia, the Congo basin, the south of Papua New Guinea, the Miombo woodlands of Africa, and the dry forests and savannas of South America. There is little consistency in the direction of the difference. However, when the maps are aggregated to the country or biome scale there is greater agreement, with differences cancelling out to a certain extent. When comparing country level biomass stocks, the two maps agree with each other to a much greater extent than to the FRA 2010 estimates. In the Colombian Amazon, both pantropical maps estimate higher biomass than the independent high resolution map, but show a similar spatial distribution of this biomass.ConclusionsBiomass mapping has progressed enormously over the past decade, to the stage where we can produce globally consistent maps of aboveground biomass. We show that there are still large uncertainties in these maps, in particular in areas with little field data. However, when used at a regional scale, different maps appear to converge, suggesting we can provide reasonable stock estimates when aggregated over large regions. Therefore we believe the largest uncertainties for REDD+ activities relate to the spatial distribution of biomass and to the spatial pattern of forest cover change, rather than to total globally or nationally summed carbon density.

Project description:Light is a key resource for tree performance and hence, tree species partition spatial and temporal gradients in light availability. Although light distribution drives tree performance and species replacement during secondary forest succession, we yet lack understanding how light distribution changes with tropical forest development.This study aims to evaluate how changes in forest structure lead to changes in vertical and horizontal light heterogeneity during tropical forest succession.We described successional patterns in light using a chronosequence approach in which we compared 14 Mexican secondary forest stands that differ in age (8-32 years) since agricultural abandonment. For each stand, we measured vertical light profiles in 16 grid cells, and structural parameters (diameter at breast height, height and crown dimensions) for each tree.During succession, we found a rapid increase in stand size (basal area, crown area and length) and stand differentiation (i.e. a gradual leaf distribution along the forest profile), which leads to fast changes in light conditions and more light heterogeneity. The inflection points of the vertical light gradient (i.e. the absolute height at which 50% relative light intensity is attained) rapidly moved towards higher heights in the first 20 years, indicating that larger amounts of light are intercepted by canopy trees. Light attenuation rate (i.e. the rate of light extinction) decreased during succession due to slower accumulation of the crown area with height. Understorey light intensity and heterogeneity slightly decreased during succession because of an increase in crown size and a decrease in lateral gap frequency. Understorey relative light intensity was 1.56% at 32 years after abandonment.Synthesis. During succession, light conditions changed linearly, which should lead to a continuous and constant replacement of species. Especially in later successional stages, stronger vertical light gradients can limit the regeneration of light-demanding pioneer species and increase the proportion of shade-tolerant late-successional species under the canopy. These changes in light conditions were largely driven by the successional changes in forest structure, as basal area strongly determined the height where most light is absorbed, whereas crown area, and to a lesser extent crown length, determined light distribution.

Project description:Identifying the consequences of tropical forest degradation is essential to mitigate its effects upon forest fauna. Large forest-dwelling mammals are often highly sensitive to environmental perturbation through processes such as fragmentation, simplification of habitat structure, and abiotic changes including increased temperatures where the canopy is cleared. Whilst previous work has focused upon species richness and rarity in logged forest, few look at spatial and temporal behavioural responses to forest degradation. Using camera traps, we explored the relationships between diel activity, behavioural expression, habitat use and ambient temperature to understand how the wild free-ranging Bornean banteng (Bos javanicus lowi) respond to logging and regeneration. Three secondary forests in Sabah, Malaysian Borneo were studied, varying in the time since last logging (6-23 years). A combination of generalised linear mixed models and generalised linear models were constructed using >36,000 trap-nights. Temperature had no significant effect on activity, however it varied markedly between forests, with the period of intense heat shortening as forest regeneration increased over the years. Bantengs regulated activity, with a reduction during the wet season in the most degraded forest (z = -2.6, Std. Error = 0.13, p = 0.01), and reductions during midday hours in forest with limited regeneration, however after >20 years of regrowth, activity was more consistent throughout the day. Foraging and use of open canopy areas dominated the activity budget when regeneration was limited. As regeneration advanced, this was replaced by greater investment in travelling and using a closed canopy. Forest degradation modifies the ambient temperature, and positively influences flooding and habitat availability during the wet season. Retention of a mosaic of mature forest patches within commercial forests could minimise these effects and also provide refuge, which is key to heat dissipation and the prevention of thermal stress, whilst retention of degraded forest could provide forage.

Dataset Information

Performance of non-parametric algorithms for spatial mapping of tropical forest structure.

Publications

Performance of non-parametric algorithms for spatial mapping of tropical forest structure.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets