Characterization of Brazilian forest types utilizing canopy height profiles derived from airborne laser scanning

    loading  Checking for direct PDF access through Ovid

Abstract

Questions:

This study explores how airborne laser scanning surveys can contribute to characterizing and discriminating different Brazilian forest types based on canopy height profiles, and how sampling intensity and plot size affect the results.

Location:

Seven different vegetation types in Brazil – open ombrophilous Amazonian forest type in the state of Pará State, dense ombrophilous Atlantic forest in the state of São Paulo, semi-deciduous and rupestrian field in the state of Minas Gerais, and three fast-growing eucalyptus plantations in the state of São Paulo.

Methods:

Canopy height profiles were modelled for each forest type using a two-parameter Weibull distribution. Differences in the values of the Weibull cumulative distribution parameters were studied to evaluate their discriminative capacity. Simulation was subsequently used to determine the number of cells needed to define a reliable sampling intensity for each cell size tested. Eight grid cell sizes were tested against six sampling intensities using 1000 iterations for all seven forest types.

Results:

Each forest presented distinct characteristics regarding the canopy height profile. The higher presence of ground returns was remarkable in less layered structured forests (i.e. rupestrian field and eucalyptus plantations). Depending on the forest type, monotonic and sigmoidal shapes were observed for the cumulative distribution function of the canopy height profile. A sigmoidal cumulative distribution function was typical of a more layered structure (dense and open ombrophilous, and semi-deciduous), whereas less structured typologies (i.e. Rupestrian field and forest plantation) were associated with monotonically shaped functions. Modelling convergence for the canopy height profile was achieved in most cases with data representing at least 2% cover.

Conclusion:

Canopy height profiles generated from ALS data proved different among five different forest types: open ombrophilous, dense ombrophilous, semi-deciduous, rupestrian field and eucalyptus plantations. Furthermore, large-scale forest inventories can benefit from the results observed in this study, as it provides recommendations on grid cell size and sampling intensity levels.

This study explores how airborne laser scanning surveys can contribute to characterizing and discriminating different Brazilian forest types. Canopy height profiles were modelled for each forest type using a two-parameter Weibull distribution. Each forest presented distinct characteristics regarding the canopy height profile, resulting in specific recommendations on grid cell size and sampling intensity levels.

Related Topics

    loading  Loading Related Articles