Savanna vegetation structure in the Brazilian Cerrado allows for the accurate estimation of aboveground biomass using terrestrial laser scanning

Zimbres, Bárbara; Shimbo, Julia; Bustamante, Mercedes M. C.; Levick, Shaun R.; Miranda, Sabrina do Couto de; Roitman, Isaac; Silvério, Divino; Gomes, Letícia; Fagg, Christopher William; Alencar, Ane

doi:10.1016/j.foreco.2019.117798

Cited by 36 publications

(44 citation statements)

References 60 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…One way to improve this classification is by adding structural characteristics to the spectral components of the NV, such as height. Some studies have demonstrated the ability of LiDAR (Light Detection and Ranging) technology to differentiate between Cerrado vegetation types [54,55]. However, this time series represents the best comprehensive temporal data set on NV distribution in Cerrado.…”

Section: Innovative Machine Learning Approach To Map Temporal Dynamicmentioning

confidence: 99%

Mapping Three Decades of Changes in the Brazilian Savanna Native Vegetation Using Landsat Data Processed in the Google Earth Engine Platform

et al. 2020

Self Cite

View full text Add to dashboard Cite

Widespread in the subtropics and tropics of the Southern Hemisphere, savannas are highly heterogeneous and seasonal natural vegetation types, which makes change detection (natural vs. anthropogenic) a challenging task. The Brazilian Cerrado represents the largest savanna in South America, and the most threatened biome in Brazil owing to agricultural expansion. To assess the native Cerrado vegetation (NV) areas most susceptible to natural and anthropogenic change over time, we classified 33 years (1985–2017) of Landsat imagery available in the Google Earth Engine (GEE) platform. The classification strategy used combined empirical and statistical decision trees to generate reference maps for machine learning classification and a novel annual dataset of the predominant Cerrado NV types (forest, savanna, and grassland). We obtained annual NV maps with an average overall accuracy ranging from 87% (at level 1 NV classification) to 71% over the time series, distinguishing the three main NV types. This time series was then used to generate probability maps for each NV class. The native vegetation in the Cerrado biome declined at an average rate of 0.5% per year (748,687 ha yr−1), mostly affecting forests and savannas. From 1985 to 2017, 24.7 million hectares of NV were lost, and now only 55% of the NV original distribution remains. Of the remnant NV in 2017 (112.6 million hectares), 65% has been stable over the years, while 12% changed among NV types, and 23% was converted to other land uses but is now in some level of secondary NV. Our results were fundamental in indicating areas with higher rates of change in a long time series in the Brazilian Cerrado and to highlight the challenges of mapping distinct NV types in a highly seasonal and heterogeneous savanna biome.

show abstract

Section: Innovative Machine Learning Approach To Map Temporal Dynamicmentioning

confidence: 99%

Mapping Three Decades of Changes in the Brazilian Savanna Native Vegetation Using Landsat Data Processed in the Google Earth Engine Platform

et al. 2020

Self Cite

View full text Add to dashboard Cite

show abstract

“…However, in tropical savannas, the assumption of spatial homogeneity underlying this approach is unlikely to be met, given inherent spatial variability at distances of 20 to 30 m ( Figure 6). TLS data allows us to extract geolocation along with DBH for individual trees and is increasingly used to investigate ecological processes such as random vs clumped tree distributions, structural change over time following disturbance [25,30,46] and over annual to decadal scales, or woody encroachment and thickening [47]. We show that TLS data obtained using a lower-cost scanner enables fast and accurate acquisition of both, full 3D tree measurements and geographic positioning required to investigate landscape scale heterogeneity in tree size class distribution (Figures 1 and 5).…”

Section: Heterogeneity At Landscape/stand Levelmentioning

confidence: 99%

“…LiDAR scanning has become a well-recognised and rapidly developing approach to quantify forest structure in great detail, as it enables us to produce detailed 3D reconstructions of individual trees and their structure [23,24]. Of the range of LiDAR platforms, terrestrial laser scanning (TLS) has been increasingly used for investigating heterogeneity in tree and stand structure in space and time, due to its high point densities and scan collections from beneath the canopy [25,26]. TLS data allows us to extract traditional field measured variables such as DBH, stand basal area and tree canopy height with a high degree of precision [27].…”

Section: Introductionmentioning

confidence: 99%

Exploring the Variability of Tropical Savanna Tree Structural Allometry with Terrestrial Laser Scanning

et al. 2020

View full text Add to dashboard Cite

Individual tree carbon stock estimates typically rely on allometric scaling relationships established between field-measured stem diameter (DBH) and destructively harvested biomass. The use of DBH-based allometric equations to estimate the carbon stored over larger areas therefore, assumes that tree architecture, including branching and crown structures, are consistent for a given DBH, and that minor variations cancel out at the plot scale. We aimed to explore the degree of structural variation present at the individual tree level across a range of size-classes. We used terrestrial laser scanning (TLS) to measure the 3D structure of each tree in a 1 ha savanna plot, with coincident field-inventory. We found that stem reconstructions from TLS captured both the spatial distribution pattern and the DBH of individual trees with high confidence when compared with manual measurements (R2 = 0.98, RMSE = 0.0102 m). Our exploration of the relationship between DBH, crown size and tree height revealed significant variability in savanna tree crown structure (measured as crown area). These findings question the reliability of DBH-based allometric equations for adequately representing diversity in tree architecture, and therefore carbon storage, in tropical savannas. However, adoption of TLS outside environmental research has been slow due to considerable capital cost and monitoring programs often continue to rely on sub-plot monitoring and traditional allometric equations. A central aspect of our study explores the utility of a lower-cost TLS system not generally used for vegetation surveys. We discuss the potential benefits of alternative TLS-based approaches, such as explicit modelling of tree structure or voxel-based analyses, to capture the diverse 3D structures of savanna trees. Our research highlights structural heterogeneity as a source of uncertainty in savanna tree carbon estimates and demonstrates the potential for greater inclusion of cost-effective TLS technology in national monitoring programs.

show abstract

“…Although the gallery's forests occupy less than 10% of the savanna area, they are home to an enormous diversity of flora and fauna (Oliveira-Filho and Ratter, 2002). Basic differences of the manually measured variables between formations (forested savanna, woodland savanna (rainy and dry) highlight that the gallery forest had the tallest canopy, greatest trunk volumes and biomass, even when there is a lower density of individuals (Zimbres et al, 2020). Besides, the Brazilian Forest Code regulates the protection of Riparian Zones, which are categorized as riparian permanent protection areas (RPPAs).…”

Section: Introductionmentioning

confidence: 99%

Leaflet phenotypic plasticity in three woody species in two strata of a gallery forest

Santos

Marenco

Ferreira

et al. 2021

CERNE

View full text Add to dashboard Cite

Background: Along the vertical canopy profile, natural light availability is variable and may induce leaves morphophysiological changes due to phenotypic plasticity. Morphological changes in juvenile and adult leaflet of compound leaves of plant species is still poorly understood, including in woody species of gallery forests. The present study evaluated the leaflet morphophysiological characteristics and plasticity index of three woody species in canopy and understory leaves in a gallery forest.Results: All morphophysiological characteristics varied within species (Myracrodruon urundeuva Allemão, Copaifera langsdorffii Desf. and Tapirira guianensis Aubl.). The leaflets of canopy trees showed significantly higher values of petiole length (PL) and diameter (PD), leaflet thickness (fresh -FT and dry -DT), specific leaf mass (SLM), in relation to understory leaflets. On the other hand, relative water content (RWC) did not change between adult trees and saplings, except for T. guianensis. The plasticity index of leaflet characteristics varied significantly within species. Plasticity index of three morphophysiological characteristics (PL, PD and DT) varied between species. PL and PD had the highest plasticity index values across species, whereas RWC had the lowest plasticity. Conclusion:Plasticity index of the tree species associates these leaflet morphophysiological traits (PL, PD and DT) with variations in the physical environment between strata of a gallery forest.

show abstract

Savanna vegetation structure in the Brazilian Cerrado allows for the accurate estimation of aboveground biomass using terrestrial laser scanning

Cited by 36 publications

References 60 publications

Mapping Three Decades of Changes in the Brazilian Savanna Native Vegetation Using Landsat Data Processed in the Google Earth Engine Platform

Mapping Three Decades of Changes in the Brazilian Savanna Native Vegetation Using Landsat Data Processed in the Google Earth Engine Platform

Exploring the Variability of Tropical Savanna Tree Structural Allometry with Terrestrial Laser Scanning

Leaflet phenotypic plasticity in three woody species in two strata of a gallery forest

Contact Info

Product

Resources

About